Tese Hansen ImperativesAndDeonticLogic

Imperatives and Deontic Logic
On the Semantic Foundations of Deontic Logic

Der Fakultat fu
r Sozialwissenschaften und Philosophie
der Universitat Leipzig
eingereichte
DISSERTATION
zur Erlangung des akademischen Grades

doctor philosophiae
(Dr. phil.)
vorgelegt
von Herrn M. A. J
org Hansen
g e b o r e n a m 09.09.1967
in Cochem
Eingereicht am 24.01.2008
Offentlich
verteidigt am 18.11.2008
Tag der Verleihung: 25.11.2008
Abstract
Deontic logic has so far been almost exclusively interpreted by a possible
worlds semantics and ideality or preference relations between the worlds. In
distinguishing itself from attempts at a logic of norms or logic of imperatives, deontic logic has portrayed its formulas as deontic propositions, true
or false statements about what is obligatory, permitted or forbidden according to some unspecified normative system. Based on this idea, the imperatival tradition of deontic logic, an unorthodox sidearm of the mainstream,
has interpreted the formulas of deontic logic not with respect to ideal or best
worlds, but to given sets of norms or imperatives. In a series of papers standing in this tradition, I have shown that by using an explicitly represented set
of imperatives and what they command as the logical semantics by which
the truth of deontic formulas is defined, all the standard systems of monadic
and dyadic deontic logic can be reconstructed, i.e. they are sound and complete with respect to such semantics. The basic concepts are motivated here
from a position of imperativological scepticism, and the main results are
summarized.
Keywords:
deontic logic, logic of imperatives, modal logic
Contents
Introduction
1 The
1.1
1.2
1.3
1.4
1.5
1.6
1.7
Logic of Imperatives
Beginnings: Poincares Proposal . . . .
Jrgensens Dilemma . . . . . . . . . .
Dubislavs Trick and Related Theories
Explanations of Imperative Inferences .
1.4.1 Logic of Satisfaction . . . . . .
1.4.2 Logic of Existence . . . . . . . .
1.4.3 Logic of Non-Factual Existence
1.4.4 Formalistic Approaches . . . . .
Rosss Paradoxes . . . . . . . . . . . .
Ordinary Language Arguments . . . .
The Way to Go Forward . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2 Deontic Logic
2.1 Beginnings: Leibnizs Discovery . . . . . . . . .
2.2 Standard Deontic Logic . . . . . . . . . . . . . .
2.3 Andersons Reduction . . . . . . . . . . . . . .
2.4 Possible Worlds Semantics . . . . . . . . . . . .
2.5 Dyadic Deontic Logic . . . . . . . . . . . . . . .
2.6 Doubts and Questions . . . . . . . . . . . . . .
2.6.1 Doubts About Possible Worlds . . . . . .
2.6.2 The Question of Conflicts and Dilemmas
2.6.3 Interpretations of Deontic Logic . . . . .
2.7 A Fundamental Problem of Deontic Logic . . .
ii
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
3
5
8
15
16
17
19
21
25
35
46
.
.
.
.
.
.
.
.
.
.
50
50
53
55
57
60
66
66
69
72
77
3 Imperative Semantics for Deontic Logic

79
3.1 The Imperatival Tradition of Deontic Logic . . . . . . . . . . . 79
3.1.1 Andersons Reduction Revisited . . . . . . . . . . . . . 79
3.1.2 Kangers Welfare Programs . . . . . . . . . . . . . . . 81
3.1.3 Ziembas Deontic Syllogistics . . . . . . . . . . . . . . . 82
3.1.4 Stenius Logic of Normative Systems . . . . . . . . . . 83
3.1.5 Alchourron & Bulygins Expressive Conception . . . . 84
3.1.6 Van Fraassens Proposal . . . . . . . . . . . . . . . . . 85
3.1.7 Input/Output Logic . . . . . . . . . . . . . . . . . . . 87
3.2 Imperative Semantics for Deontic Logic . . . . . . . . . . . . . 89
3.3 The Reconstruction of Deontic Logic . . . . . . . . . . . . . . 94
3.3.1 Some Logics about Imperatives [2001] . . . . . . . . . . 94
3.3.2 Problems and Results for Logics about Imperatives [2004] 96
3.3.3 Conflicting Imperatives and Dyadic Deontic Logic [2005]100
3.3.4 Deontic Logics for Prioritized Imperatives [2006] . . . . 101
3.3.5 Prioritized Conditional Imperatives [2007] . . . . . . . 103
3.4 The Paradoxes of Deontic Logic Reconsidered . . . . . . . . . 107
3.4.1 Rosss Paradoxes . . . . . . . . . . . . . . . . . . . . . 107
3.4.2 Contrary-to-duty Paradoxes . . . . . . . . . . . . . . . 110
3.4.3 Paradoxes of Defeasible Reasoning . . . . . . . . . . . 113
4 Conclusion
117
Acknowledgements
120
Bibliography
121
Appendix: Research Papers

Sets, Sentences and Some Logics about Imperatives[2001] .
Problems and Results for Logics about Imperatives [2004]
Conflicting Imperatives and Dyadic Deontic Logic [2005] .
Deontic Logics for Prioritized Imperatives [2006] . . . . . .
Prioritized Conditional Imperatives [2007] . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
142
142
164
187
215
249
Anhang
274
Zusammenfassung . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
Bibliographische Beschreibung . . . . . . . . . . . . . . . . . . . . . 281
Lebenslauf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
iii
iv
Publikationen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
Vortrage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
Versicherung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
Introduction
At the biannual EON-workshop on Deontic Logic in Computer Science,
held at the University of Bologna, 810 January 1998, David Makinson drew
attention to the distinction between norms, that cannot meaningfully be
termed true or false, and deontic propositions, true or false statements about
what is obligatory, permitted or forbidden according to the norms, and that
this distinction has tended to get lost in modern deontic logic literature.
He therefore called for a reconstruction of deontic logic as a logic concerned with norms, but in accord with the philosophical position that norms
are devoid of truth values. Motivated by this call I have, in a series of research papers since then, provided a semantics for deontic logic that defines
its operators by means of an explicitly given set of imperatives. It can be
demonstrated that all of the standard systems of monadic and dyadic deontic
logic can be reconstructed with respect to such an imperative-based semantics, but also other systems are explored that consider specific issues, like
the possibility of conflicts and dilemmas, or the accommodation of priority
relations between the imperatives.
Here, the topic is explained and the results are presented that have been
achieved so far. Chapter 1 addresses the question of whether there exists a
logic of imperatives. If it does, then it would perhaps be easier to devise a
deontic logic as a logic that reflects already given logical relations between
imperatives, but the arguments make it clear that our start must be from
a position of imperativological scepticism, and no direct logical relations
between imperatives can be assumed. Chapter 2 lays out the main systems
of deontic logic that are later to be reconstructed, and explains the background of Makinsons criticism directed at traditional deontic logic. Chapter
3 first provides an overview over an already existing imperatival tradition
of deontic logic that has, instead of using standard possible worlds semantics with ideality or preference relations, interpreted deontic operators with
1
INTRODUCTION
respect to given sets of imperatives or norms. Then the imperative semantics used here to reconstruct the systems of deontic logic is explained, and
the results of the previous research are summarized. Finally it is examined
in which way previous paradoxes of deontic logic may be treated within
the new semantical framework. The conclusion explains what the results
mean for our understanding of deontic logic, and hints at possible further
developments.
Chapter 1
The Logic of Imperatives
1.1
Beginnings: Poincar
es Proposal
Can imperatives, i.e. sentences in the imperative mood, be part of logical

inferences? Henri Poincare considered this question in his 1913 essay La
Morale et la Science [172]. He begins by observing that if the premisses
are all indicatives, then so will be the conclusion, hence for an imperative
conclusion at least one premiss in the imperative mood is required, and so
science alone cannot establish standards of morality. However, just as steam
can be put to use in different machinery, science may also be used for moral
reasoning:
It [the moral sentiment] will give us the major premiss of our
inference which, as it happens, is in the imperative mood. At
its side, science will put the minor premiss which will be in the
indicative mood. From these a conclusion can be drawn that is
in the imperative mood.
Poincare seems to have in mind Aristotelian syllogisms of the following kind:
Hang all dwellers of the Nottingham Forest!
All members of Robins band dwell in the Nottingham Forest.
Therefore: Hang all members of Robins band!
Poincare then proceeds to give a second example of an inference with an
imperative conclusion:
LOGIC OF IMPERATIVES
One can imagine inferences which are of the following type: do

this, but now if one does not do that, one cannot do this, so do
that. And such reasoning is not outside the field of science.
The following is an example of the suggested inference:
Open the door!
The door cannot be opened unless it is first unlocked.
Therefore: Unlock the door!
So an order to do one thing includes orders to do all that is necessary to satisfy
the primary command. Poincares proposals raise questions: by exchanging
in his first example the syllogism barbara for camestres we obtain:
Hang all dwellers of the Nottingham Forest!
No member of Robins band is hanged.
Therefore: No dweller of the Nottingham Forest shall be a
member of Robins band!
But it did not seem as if the speaker, e.g. the Sheriff of Nottinghamshire,
was creating rules for band membership. The second type of inference is
problematic when there are no legal means to fulfill an imperative (cf. Foot
[54] p. 384):
Sustain your aged parents!
I can only sustain my aged parents if I rob somebody.
Therefore: Rob somebody!
So if commanding means also commanding all necessary acts, then even
forbidden acts may be included. To improve matters, the second clause in
Poincares scheme might be changed to this can only legally be brought about
by doing that. But this introduces a normative element into a premiss that
Poincare assumed to be established by science alone.
While all this suggests that imperative inferences might require some
additions and modifications, the most difficult question has turned out to
be what makes them inferences. The problems attached to this question go
under the name of Jrgensens Dilemma.
1.2
Jrgensens Dilemma
Logics concern is with the soundness of arguments, or inferences. These

consist of sentences that represent the premisses and usually one sentence
that forms the conclusion. The argument is then called sound, valid
or logical, if it is not possible that all of its premisses are true but the
conclusion false. The premisses are then said to entail the conclusion which
thus follows from them.1 Expressions in the imperative mood are not, in
any usual sense, true or false. Therefore they are incapable of functioning
as premisses or conclusions in logical inferences. However, people maintain
that the opposite is true and that there are inferences that have conclusions
in the imperative mood and premisses of which at least one is likewise in
the imperative mood (cf. Poincares examples above). This is a puzzling
situation, which was first pointed out by Jrgen Jrgensen in [110]:
So we have the following puzzle: According to a generally accepted definition of logical inference only sentences which are capable of being true or false can function as premisses or conclusions in an inference; nevertheless it seems evident that a conclusion in the imperative mood may be drawn from two premises
one of which or both of which are in the imperative mood. How
is this puzzle to be dealt with?
To find this puzzle, called Jrgensens Dilemma, perplexing, one must
accept that imperatives cannot be meaningfully termed true or false. This
seems to be the philosophical consensus, it can point to Aristotles definition
of an assertion as a grammatical entity that can be true or false, in distinction
to other grammatical entities like requests that are neither true or false (De
interpretatione 17 a 4). Nevertheless, a way out of the dilemma may consist
in giving up just this claim. Most prominently, Kalinowski [111], [114] has
argued that in the case of expressions of moral or legal norms, the attitude of
the ordinary, non-philosophical person is to treat these as true or false. E.g.
people say that it is true that another persons right to live must be respected,
or that slander is prohibited, and people would uphold these truths even if
particular legislators did not enact such norms, or proclaimed otherwise.
So Kalinowski concludes that legal or moral norms can be part of logical
inferences. I think that these considerations confuse truth with the notion of
1
For such textbook definition cf. Mates [147] p.5, Lemmon [134] p.1, Hodges [98] p.55.
a legal or moral norms validity: the external recognition of a norm as valid

in a certain society. For the present discussion it suffices that Kalinowski
himself restricts his view to legal and moral norms and does not claim that
imperatives in the strict sense can be said to be true or false, and in fact
writes that they are not true or false.2 But it is these that we are concerned
with.
On another view, any imperative can be equivalently replaced by a first
person expressions like I command you to ..., I order you to ..., I request of
you that you ... or I want you to ....3 But expressions like I command you
to ... cannot only be used to command, but also to assert that I do in fact
command so-and-so. Therefore Sigwart [195] claims that each imperative
includes the statement that the speaker wills the act which he commands.
While he adds that nevertheless the import of what is said by the imperative
is not the communication of a truth but a summons to do this, to leave that
undone, Ledig [129] argues that because imperatives include such assertions,
one must consequently apply the terms of truth and falsity to an imperative,
and that it will be true unless it is e.g. stated for fun. Likewise, Kamp
[118] points out that from the viewpoint of the addressee, it really makes
no difference if an expressions like I command you to ... is understood
as a command or as an assertion, since if the utterance is appropriate (i.e.
made in earnest in the appropriate conditions) then it must be true and so
the practical consequences are the same. Similarly, Opalek & Wole
nski [164]
call a normative statement qua performative true if it is effective. However,
it is hard to see how these considerations can solve Jrgensens dilemma. If
any imperative that is meant seriously is true, then (i) any imperative either
follows from any other true imperative or is not meant seriously, and (ii)
all imperatives follow from one made for fun. If imperatives are ambiguous
and include an assertion, then the fact that this assertion may be used in an
indicative inference does not mean that the indicative conclusion is likewise
ambiguous. A sign post bearing the words Frankfurt/Main conveys the
information that if I follow the indicated direction I will eventually arrive
at Frankfurt. I may also infer that there exists a place called Frankfurt and
that a geographical entity called Main, namely a river, exists in its proximity.
Knowing that sign posts tend not to include redundant information, I can
2
Cf. Kalinowski [112] p. 36 and [114] p. 107.

In the case of English grammar, it has been claimed that the exclamation mark,
or rather the characteristic intonation that it replaces in written language, is only the
remainder of such explicit performative lead-ins, cf. Harris [89] pp. 391392.
3
also conclude that there must be another place that is also called Frankfurt
(namely Frankfurt/Oder), and that this place does not lie near the Main,
because otherwise the extra information would not have been discriminating.
But the sign post does not point into the direction of this other city. Similarly
the information that is obtained by interpreting an imperative utterance may
be used to infer some other information. But this is not inferring imperatives.
The fact that imperatives are traditionally not considered to be true or
false finds its explanation in the different intentions in which imperatives
and indicatives are used. The main use of indicatives is to convey what the
speaker believes the world to be like. If it is so, then the sentence is called
true, if not, then it is called false and the recipient might point out that I
should perhaps change my beliefs. By use of an imperative I tell the addressee
what I want to be done. If the addressee does what is demanded, the action
may be qualified as right, or satisfactory with respect to the command, and
if not, then the behavior of the addressee is in some sense wrong and I will
perhaps remind the agent of his or her obligation. So truth and falsity are
the qualities of descriptions when things are or are not as they have been
described, while right or wrong are the qualities of acts that are or are not
in accordance with what has been prescribed. Descriptions and prescriptions
have a different direction of fit, and true/false are the terms used to express
the match/mismatch on the language side in case of a descriptive use of
language, and right/wrong are the terms employed for the match/mismatch
on the world side in case of a prescriptive use.4 Therefore it is a confusion
of language, and indicates a misunderstanding of the intention in which the
sentence has been uttered, if imperatives are termed true or false.
Accordingly, the most effort regarding Jrgensens dilemma has been
spent on developing alternative definitions for imperative inferences, rather
than arguing for the application of the terms of truth and falsity to imperatives unless one is already convinced by the dilemma that such things as
inferences with imperatives are at all impossible (e.g. Keene [120]).
4
This explanation of why the terms of truth and falsity are not applicable to normative
uses of language originates with Anscombe [19] 32. Independent accounts can be found
in Kenny [122] p. 68 and Peczenik [166], [167] who speaks of the norm as a qualifying
utterance. The dual terms right/wrong are used as corresponding qualifications e.g. by
Englis [52] and Kelsen [121] p. 132.
1.3
Dubislavs Trick and Related Theories
To deal with his own dilemma, Jrgensen [110] endorsed a proposal by

Walter Dubislav [50] to transfer the usual definitions of inferences between
indicatives to imperatives by analogy. Dubislav gives the following example:
Though shalt not kill.
Therefore: Cain shalt not kill Abel.
Here, he argues, the analogue of the following ordinary inference is applied:
No human being kills any other human being.
Cain and Abel are human beings.
Therefore: Cain does not kill Abel.
Dubislav observes that to each imperative belongs a descriptive sentence that
describes the state of affairs that obtains if the subjects of the imperative
realize what the commanding authority demands. The formalisms of descriptive inferences are then transferred to imperatives by what he calls a trick
(Kunstgriff ): imagine the state that the commanding authority desires realized, describe it, from this description infer some other descriptive sentence,
which is then again interpreted as describing a state the authority wants to
see realized. He then proposes the following convention (DC) on the meaning
of imperative inference (also see the next figure):
(DC) An imperative F is called derivable from an imperative
E if the descriptive sentence belonging to F is derivable with
the usual methods from the descriptive sentence belonging to E,
whereby identity of the commanding authority is assumed.
The convention is illustrated by the next figure (where I write !A for an
imperative to which the descriptive sentence A belongs):5
Dubislavs convention does not cover inferences with more than one imperative premiss, though he mentions this possibility.6 For such inferences,
(DC) can be modified as follows:
5
Mally [145] p. 12 seems to have introduced the symbolism !A, which was then employed
by Hofstadter & McKinsey [99] for the imperative that demands that A be the case.
However, Mally intended !A to be interpreted theoretically, as an assertion or assumption
that A ought to be, which we now call a deontic proposition and formalize by OA.
6
Cf. Dubislavs use of the plural when stating that an inference from demand-sentences
will now be formally facilitated by the following convention, and Dubislavs summary,
in which he stresses that no demand-sentence can be derived from premisses that do not
contain at least one demand-sentence.
Figure 1.1: Dubislavs convention (DC).

(DCM) An imperative F is called derivable from the imperatives
E1 , ..., En if the descriptive sentence belonging to F is derivable
with the usual methods from the descriptive sentences belonging
to E1 , ..., En , identity of the commanding authority presupposed.
Dubislav then proceeds to inferences in which the imperative premiss is accompanied by another one in the indicative mood, and where the conclusion
is again an imperative, for which he extends his convention:
(DEC) An imperative F is called derivable in the extended
sense from an imperative E if the descriptive sentence belonging
to F is at least jointly derivable from the descriptive sentence
belonging to E and true descriptive sentences that are consistent
with the first.
This extended convention (DEC) may again be modified to facilitate inferences with more than one imperative premiss to produce the following
modified extended convention
(DECM) An imperative F is called derivable in the extended
sense from imperatives E1 , ..., En if the descriptive sentence belonging to F is at least jointly derivable from the descriptive sentence belonging to E1 , ..., En and true descriptive sentences that
are consistent with these.
Jrgensen [110] endorsed Dubislavs proposal as one way to deal with his
dilemma and states that it seems clear to him that any imperative sentence
10
has an indicative parallel-sentence which describes the contents of the command or wish. Jrgensen suggests that an imperative consists of an imperative factor and an indicative factor, where the first indicates that something
is commanded, and the second what is commanded. The indicative factor
can then be separated from the imperative and formulated in indicative sentences describing the action, change or state of affairs which is commanded.
Applying the rules to these latter sentences we can thus indirectly apply the
rules of logic to the imperative sentences to make their entailments explicit.
Writing A for the assertion that the state of affairs A holds, this slightly
different analysis of imperative inferences is illustrated by the next figure:7
Figure 1.2: Jrgensens analysis of indirect imperative entailment.

Jrgensens concept was in turn further refined by R. M. Hare [86], according to whom an imperative sentence and an indicative sentence correspond
if they have the same descriptor, but different dictors, where what is described by the descriptor is what would be the case if the sentence is true or
the command obeyed, and the dictor is what does the saying or commanding.8 This is still not much different from Jrgensens analysis, but Hare finds
it is misleading to speak of an indirect, parallel or analogous application
of logic. Instead, in Hares view imperatives are logical in the same way as
indicatives; he argues that most inferences are inferences from descriptor to
descriptor and we could add whichever set of dictors we pleased. Since most
7
This formalization is used in Reichenbach [180] 57.

This distinction was anticipated by Ledig [130], who wrote that norms and descriptions have an isolable imaginary content (isolierbarer Vorstellungsinhalt). Hares [87] later
terminology of a neustic and phrastic mirrors his earlier distinction.
8
11
logical reasoning is done with descriptors only - this Hare calls the principle
of the dictive indifference of logic, there is no special need for a logic of
imperatives. Rather, all logic is recast as a logic of descriptors, where if the
descriptors of the premisses describe a state of affairs, then the descriptor of
the conclusion describes, at least partially, the same state of affairs. Hares
view of logic is pictured in the next figure:
Figure 1.3: Hares descriptor logic.

It is immediate that neither Jorgensens nor Hares account make a material difference for what imperative inferences should be accepted. Hares
concept of a dictor that operates on a descriptor poses problems: grammatically, it is hard to see how dictors can be removed from sentences, or
exchanged in them, so that the remainder or the new composite is a meaningful expression.9 So there may be reasons not to follow Hares analysis,
but closer to Dubislavs original concept speak of a thematically parallel
sentence in the indicative mood. However, this way or another, the idea of a
descriptive sentence that parallels an imperative or of an imperatives indicative factor has become the most successful part of the Dubislav-JrgensenHare analysis. Ross [188] calls this element the theme of demand, a state
the realization of which is requested by the demand, and proposed that to
9
Opalek [162] points out that even if the imperative is rephrased as I command that
... or it is obligatory that ..., the ...-part is a Latin ut-expression that only due to
a peculiarity of English grammar may be confused with an indicative sentence, also cf.
Opalek [163] ch. 2, Kalinowski [113] and Rodig [186] for similar criticism. On the other
extreme, Leonard [135] has argued that it is the descriptor that is called true and false,
and so imperatives share these properties with descriptive sentences.
12
any imperative corresponds an ordinary indicative sentence which contains

a description of the imperatives theme of demand. Frey [58] p. 440 uses
the term Erf
ullungsaussage for the parallel sentence that indicates if the imperative is satisfied or violated, which Rescher [184] p. 52 calls a command
termination statement and Keene [120] the actualization of the imperative.
Geach [60] states that for every imperative there is a future-tense statement
whose coming true is identical with the fulfilment of the imperative, Sosa
[200], [202] speaks of the propositional core of an imperative and Hanson [83]
of a state s the commanding agent envisages, of which then a description S
is used. Von Wright (e.g. [255] p. 269) calls the state the norm pronounces
as obligatory or allowed the content of a norm. The seemingly universal
consensus is explained by the pragmatic function of imperatives, that is, the
regulation of human behavior: if there were no imperative-correlated indicative sentences, it could not be understood what ought to be, and neither
would it be possible to determine whether the norm is satisfied or violated.10
This is why even anti-reductionist authors that oppose the idea that imperatives or imperative reasoning can be reduced to indicatives or indicative
logic, agree on the following principle:11
Principle (W) (Weinbergers Principle).
To each imperative there corresponds a descriptive sentence that is true if the
imperative is satisfied and false if it is not-satisfied (violated).
It is clear that an acceptance of (W) does not force us to also accept the
Dubislav-Jrgensen-Hare account of imperative inference, which nevertheless
has been accepted by a number of authors.12 (W) can then be used to
10
This explanation and the formulation of the principle below is most clearly expressed
in Weinberger [234] p. 172. Weinberger uses the term coordination instead of correspondence, but this suggests an onto or even one-to-one mapping.
11
Besides Weinberger cf. Hamblin [74] pp. 151-152: Take the exact words of the imperative, and transform them into indicative mood (...) Now the worlds which extensionally
satisfy the imperative are just those of which the description is true., and Moutafakis
theorem T3 ([155] p. 155), which expresses the equivalence of the statements that an
imperative is satisfied and that a description of the prescribed action as performed is true.
12
These include Simon [196], who converts commands to declarative mode by removing
the imperative operators from them, obtaining a theory in which all recipients obey the
commands, and then applies the ordinary laws of logic to derive new relations that may
be converted back into commands. According to Niiniluoto [156], an imperative !p entails
imperative !q if p entails q. Very close to (DCM) is von Wrights account in [246] pp. 71,
164, where he defines the content of a prescription as the prescribed thing, and defines
that a command is entailed by a second command or by a set of commands if the content
13
show that seemingly differing explanations of imperative inferences are in fact

equivalent to this account. Thus Rescher [184] defines command inferences
in terms of satisfaction in the following way:
A command inference is valid if there is no possible world in
which the premisses are all satisfied and the conclusion fails to
be satisfied.
which, using (W), is equivalent to
A command inference is valid if there is no possible world in which
the descriptive sentences corresponding to the premisses are all
true and the descriptive sentence corresponding to the conclusion
is false.
Using the textbook definition of an argument, this is equivalent to
A command inference is valid if the descriptive sentences corresponding to the premisses entail the descriptive sentence corresponding to the conclusion.
which in turn is the modified Dubislav convention (DCM).13
Using the idea that norms qualify the states of affairs that satisfies or
violates them as right and wrong, one can define:14
An imperative !A entails an imperative !B if and only if (iff)
every state of affairs that is qualified as wrong by !B is qualified
as wrong by !A.
of a command is a consequence of the conjunction of the content of a command with the
contents of none or one or several other commands.
13
Sosas [200], [202] definition of a directive argument is very similar to Reschers, but
additionally demands that the imperatives that function as premisses are jointly satisfiable
in order to cope with normative conflicts. Sosa adds a second condition, demanding that
if the imperative conclusion is violated, at least one imperative premiss must be violated,
in order to also cope with conditional imperatives. This equals Keuths [123] condition B1
that it must be logically impossible to violate the conclusion without violating a premiss.
Obviously, the second condition makes no difference for unconditional imperatives.
14
The definition is similar to the one used by Peczenik [166], [167] for forbidding norms
and the quality forbidden. Kamp [117] uses an analogous definition for permissions,
where one permission entails another if the second makes only such courses of actions
permissible that were already so before.
14
which can then be translated into

An imperative !A entails an imperative !B iff every state of affairs
that violates !B also violates !A.
which using (W) is equivalent to
in which B is false also makes A false.
which using classical logic is equivalent to
in which B is true also makes A true.
which using tertium non datur and modus tollens is equivalent to
in which A is true also makes B true.
which by definition of entailment equals
An imperative !A entails an imperative !B iff A entails B.
and this is again Dubislavs convention (DC).
Lemmon [133] defines an entailment relation for imperatives via a definition of inconsistency of a set of imperatives and indicatives, where such a
set is called inconsistent if it cannot be the case that all indicatives are true
and all imperatives obeyed.15 Lemmons entailment relation is then defined
as follows:
An imperative !A is entailed by a set indicatives and imperatives
if this set together with !A is inconsistent.
15
Lemmon expresses reservations regarding his definition, but only because he thinks
that it does not sufficiently restrict imperative conclusions to statements about future
actions. A definition similar to Lemmons seems to be intended by Philipps [171] p. 364
who defines: to do p is forbidden! is true iff the indicative someone does p is incompatible
with the class of valid prescriptions, where compatible means that if the indicative is true,
then at least one prescription is violated.
15
where !A is the imperative that is satisfied if and only if A is false. Using

(W) we obtain:
An imperative !A is entailed by a set indicatives and imperatives if the set of indicatives together with the set of descriptive
sentences corresponding to the imperatives together with A is
inconsistent.
which, using classical logic, is equivalent to
An imperative !A is entailed by a set indicatives and imperatives if
the set of indicatives together with the set of descriptive sentences
corresponding to the imperatives entails A.
and this is Dubislavs modified extended convention (DECM). This shows
that, with (W), Dubislavs proposal is equivalent, or at least very similar, to
a number of other proposals how imperative inferences are possible in the
face of Jrgensens dilemma.
1.4
Explanations of Imperative Inferences
Dubislavs trick provides a formal method that explains how inferences

between imperatives can be defined without having to assign them truth
values. What has not been made clear is what is achieved by the method, i.e.
why we should think that this is what it means to infer an imperative from
some other imperative or a set of imperatives and indicatives, or formally,
what is means that some imperative inference scheme
(ImpInf )
!A
!B
is valid. Dubislavs trick can easily be applied to e.g. sentences of the form
I doubt that .... Then from I doubt that he is staying at his sisters place
in San Francisco follows I doubt that he is staying in San Francisco, which,
though we can derive he is staying in San Francisco from he is staying at
this sisters place in San Francisco, seems wrong: I might not doubt that he
is staying in San Francisco, but doubt very much that he is staying with his
sister. So why should Dubislavs trick work for imperatives if it would not
for other expressions?
1.4.1
16
Logic of Satisfaction
On one interpretation, which has been called the logic of satisfaction, the
scheme (ImpInf) is understood as stating that if the imperative sentence !A
is satisfied, then it must be that some other imperative sentence !B is also
satisfied. This interpretation is usually attributed to Hofstadter & McKinsey
[99], whose formalization of the scheme (ImpInf) would be !A >!B, which
is derivable in their axiom system whenever A B is classically derivable.
It is immediate that (ImpInf) must be valid on this interpretation whenever
the ordinary argument
A
B
is valid for the descriptive sentences A and B: If A classically implies B, then
it must be that if !A is satisfied, then A is true and also B is true and hence
!B is satisfied. Thus Dubislavs trick receives its semantic justification. But
if (ImpInf) is interpreted in this way, then it seems one should also accept
the following scheme:
!A
A
According to our (informal) convention, !A represents an imperative sentence
that is satisfied iff A is true. So it must be that if !A is satisfied, then A is
true and so the above scheme is valid. But on the look of it, this scheme
seems to state that from an imperative that demands A it can be inferred
that A is the case, which is nonsense. And this misunderstanding reveals
that when we spoke of the possibility of an inference in which the premisses
and the conclusion are imperatives, it seems that we talk about inferring an
imperative from some other imperatives, and not about reasoning whether
or not the imperatives in question are satisfied. So though the inferences
of a logic of satisfaction are valid in the interpretation in which they were
intended, thus interpreted inferences seem not to be what we want from a
logic of imperatives. For these reasons, Ross [188] p. 61 and also Hare [88]
doubted that a logic of satisfaction is what one has in mind in the case of
practical inferences.16
16
Kanger [119] p. 49 and Fllesdal & Hilpinen [53] p. 7 criticize Hofstadter & McKinsey
for making !A and A equivalent, which is somewhat unfair since the intended interpretation of their formulas (in terms of satisfaction) is not presented.
1.4.2
17
Logic of Existence
When we speak of inferring one imperative from some other imperative, this
could mean that the existence of an imperative is logically deduced from the
existence of some other imperative.
What is meant by saying that an imperative exists ? First, it could be the
existence of an utterance of some sentence in the imperative mood by some
commanding agent towards some commanded subject.17 Second, one might
demand that the utterance, as a performative use of the imperative sentence,
was effective and established an imperativum, i. e. a command, demand,
request or the like. For this it may be required that the commanding agent
had the will to command (and did not use the words for fun) as well as some
authority (power to punish or reward) over the addressee.18 Third, for an
order by legal authorities in this capacity to come into legal existence it
may be required that the authority was competent to utter it according to
the legal rules of some normative system that confers such competence, and
similar for bodies that are constituted not by law but by other rules like a
firm or Robins band.19
Yet however much the concept of existence is thus refined, it seems to
require the presence of actual facts: a (still alive?) speaker, a linguistic
entity like an utterance and circumstances of speaking, a certain attitude of
the speaker towards the act of speaking, a backing of the speaker by force or
an authority conferred by existing and/or valid rules, etc. But it is difficult
to see how logic can stipulate such an existence. This is illustrated in the
following example by Aleksander Peczenik:20
17
This existence is what Frey [58], along with an additional property of justification,
infers in imperative inferences: If the imperatives that appear in the premisses exist and
are justified, then also the imperatives derived from these exist and are justified (p. 465).
Freys justification means that what is demanded is good regarding some aim of the
commanding agent, called axiological validity in Ziembi
nski [260].
18
Cf. von Wright [246] p. 120-126. This is Ziembi
nskis [260] thetic validity. Lemmon
[133] seems to have this notion of validity or existence in mind when he demands that
the entailment of imperatives must be defined in terms of what imperatives are in force
at a given time.
19
Bulygin [39] uses the term systemic validity. According to Weinberger [228], [231] p.
259, this validity takes the place of truth as the hereditary trait (Erbeigenschaft) that is
transferred from the premisses to the normative conclusion in inferences with normative
sentences (Norms
atze).
20
Quotation from a letter by A. Peczenik to R. Walter, printed in [218] p. 395
18
The premiss love your neighbour may be regarded as describing

the fact that the authority Jesus has in fact said love your
neighbour. The imperative existed because it was uttered by
Jesus. But the conclusion, for example, love Mr. X does not
describe anything which in fact has been said by Jesus.
Here, the intended argument from love your neighbour to love Mr. X
is not accepted because the commanding agent did not actually say what
appears in the conclusion, and so unlike the premiss the conclusion did never
exist as a fact. An imperative sentence that has not been expressed was not
received and cannot be understood by its addressee as a command or legal
order. Thus the required existence of the imperative, or validity of the
command seem to be the analogues not of truth of a descriptive sentence, but
of stated and asserted. Yet indicative logic does not force anyone to state
or assert anything, even if some other descriptive sentence was used in a way
that expresses ones commitment to it. It only explains what people mean
when they use an indicative sentence in order to assert some fact, by saying
what other sentences must be true if the stated sentence is true.21 Because
the imperative sentence in the conclusion may not exist as an utterance,
Hamblin [74] p.89 warns against speaking of inferences between imperatives.
Because the telling part (or attitude) of the speaker cannot be inferred, the
possibility of command inferences was denied by Sellars [193] p. 239-240, and
for the same reason, Lemmons [133] attempt to define imperative inferences
via the notion of an imperatives being in force was rejected by Sosa [202]
p. 61 who argues that such notions involve attitudes by the authority or its
subject that cannot be inferred. Von Kutschera [128] argues that a (used)
imperative is an action, actions do not follow from actions, so there is no
logic of imperatives. In Alchourron & Bulygins [5] expressive conception
of norms, the existence of a norm is seen as dependent on empirical facts
and the possibility of a logic of norms is consequently denied as there are
no logical relations between facts. For similar reasons, Philipp [168], [169]
denied both the possibility of a logic of imperatives and of norms.
21
Kenny [122] first pointed out that valid and invalid, interpreted as meaning commanded and not commanded, are not the analogues of true and false, but of stated
and not stated.
1.4.3
19
Logic of Non-Factual Existence
To get around this difficulty one may consider to interpret existence not
with respect to natural facts, but with respect to some ideal world of ought
or an assumed normative system that is closed under consequences, where
the closure operation may be understood e.g. in the sense of derivability by
use of Dubislavs convention. So if the agents use of the imperative mood
has resulted in the existence of a command in the world of ought, or, due
to the agents legal competence as e.g. a police officer, created an order
that now belongs to the normative system, then all the consequences of the
command that can be derived by an appropriate method exist in this world or
system as well.22 What thus has ideal existence is not the imperative sentence
as a spatial and temporal phenomenon or as a grammatically correct or
meaningful combination of words. Rather, it is what the use of an imperative
sentence expresses or accomplishes a command, request etc. Then it must
be that not only commands, requests etc. exist in this sense that in fact
have been expressed by a performative use of a sentence in the imperative
mood, but also some that only can be expressed. For if all that exists
in the normative system already exists as a result of a pragmatic use of
language, then there would be no need to let the normative system e.g.
be closed under a consequence operation. That what we can express by
using language (commands, requests, assertions etc.) has some existence,
22
The world of ought terminology originates with Walter [217], who is however following
Kelsen [121] p.195 in that an individual norm does not exist before the general norm
was applied by a judge, so orders that can only be deduced do not exist in Walters
world of ought. The idea to explain logical relations between norm sentences (like
imperative sentences) in terms of their existence in a system of norms that is closed
under consequences is that of Stenius [205]. In Opalek & Wolenski [165], norms are
non-linguistic entities expressed by (descriptively interpreted) deontic statements, and
normative systems consist not only of norms that have been expressed by a normative
authority, but also of the consequences of these basic obligations. In Alchourron &
Bulygins hyletic variant of a conception of norms [8], implicitly promulgated norms have
existence in a logically closed normative system, and descriptively interpreted (deontic)
norm propositions are then propositions about the existence of norms (in that system).
Holl
ander [100] promotes the idea of a deontically perfect world where norms exist that
obey logical principles, like that conflicts are excluded. Kelsen [121] pp. 187188 rejects
the idea of an ideal existence of norms because there is no ideal act of will that creates
them, and rejects the whole idea of a logic of norms.
20
possibly independent from any pragmatic origin23 , in some ideal world of

ought, is a difficult concept that possibly creates more problems than it
solves.24 But it is even difficult to see that it solves the problem of entailment
between imperatives. For to say that some ideal object created by use of an
imperative implies the existence of some other ideal object in some world of
ought or normative system, is not to say that an imperative implies another
imperative. The existence of a forest might imply the existence of a tree, but
to say that the forest implies the tree is making a categorical mistake. The
first is an indicative argument, which can be formalized in the usual way:
(1)
x : Forest(x)
y : Tree(y)
The argument is analytical when the words forest and tree have their
usual meaning and for all that understand this meaning and thus know that
there cannot be forests without trees. By starting to talk about the (ideal)
existence of commands it seems that we silently changed (ImpInf) into
(2)
the command given by to x by the use of the imperative sentence

!A
Therefore: the command given by to x by the use of the imperative sentence !B
which appears confused. This is because the argument form is not used as
it is usually used, and now we do not know what to make of it. We are used
to filling in the blanks of the argument form
(3)
with sentences. Whether also imperative sentences can be meaningfully used

to fill in the blanks is the open question. However, there is no pre-established
usage of filling in the blanks with names of objects, as in
(4)
23
a
b
Cf. Stenius [205] according to whom all normative systems include a norm that
demands a tautology.
24
Note that the topic of this discussion has not suddenly become the ontological status of notoriously difficult concepts of practical philosophy and jurisprudence, like moral
obligations, natural law, human rights, laws of custom etc. Our concern are still ordinary
sentences in the imperative mood, addressed e.g. to a husband, secretary, student, child
or dog (cf. Ziemba [259] p. 386).
21
where a means a forest and b means a tree. At most, this is a mistaken way
to try to express (1). Similarly, the scheme (2) must be corrected into (5):
(5)
There exists a command given by to x by the use of the imperative sentence !A.
Therefore: There exists a command given by to x by the use of
the imperative sentence !B.
Now we have arrived at something that resembles an argument though we

have not yet shown when such a scheme constitutes valid arguments. But
this argument is one that uses only descriptive sentences that can be true or
false. It is not a case of (ImpInf), i.e. not an argument where imperative
sentences function as premisses or conclusion.25 So by appealing to the notion
of existence we obtain at most an inference relation between sentences that
describe the existence of certain commands, but not a logic of commands.
1.4.4
Formalistic Approaches
Maybe we should have gone looking for the explanation of an entailment

relation between imperatives not in some relation between entities in a world
of ought, but in the meaning of the sentence that the speaker uses. Someone
who says that either Barack Obama or Hillary Clinton will be the 44th
president of the U.S., but it wont be Obama, thereby expresses the belief
that Hillary Clinton will be the 44th U.S. president. The speaker did not
say explicitly that it will be Clinton (these were not the words used), but
the speaker can be said to have implicitly, or tacitly, expressed this opinion.
Therefore if a person asserts something, she may be taken to implicitly
assert something else. Likewise, when a person commands something, she
may also be commanding something by implication. Consider the following
example employed by Hare [88]:
Go via Coldstream or Berwick!
Dont go via Coldstream!
Therefore: Go via Berwick!.
25
That the world of ought approach thus only provides arguments with descriptive
sentences is accepted by Walter [217], for he turns to identify imperatives with descriptions
of the world-of-ought-existence of a command that is created by the use of an imperative
consequently such sentences can be true or false and therefore part of logical inferences.
Thus imperative logic is reduced to indicative logic, where the difficult part is now the
verification of some descriptive sentences.
22
Here, an officer who must go from London to Edinburgh is ordered go via

Coldstream or Berwick, and a bit later is given the order dont go via
Coldstream. Both commands have been taken to imply that the officer is
(now) ordered to go via Berwick, and that the inference is therefore valid.
For the question what commands should be considered to have been implicitly commanded by explicitly given commands, one may then point to
Dubislavs conventions or similar rules e.g. (DCM) clearly accounts for
the above inference.26 To search for an explanation of imperative inferences
by interpreting explicitly given commands seems a much better idea than
to find it in otherworldly relations. Unfortunately, it is also a circular idea:
according to it one can infer a command from another command if by giving
the latter command one implicitly commands the former. Or: we may use
Dubislavs convention to infer imperatives from other imperatives because it
provides the imperative sentences that are implicitly commanded when the
first imperative sentences are used for commanding.
An interesting response to this reproach of circularity is to say that by
giving rules for inferring one imperative from another, one did all that is
required to explain the meaning of imperative inferences. It is this view
that seems to lie at the root of Dubislavs proposal to formally facilitate
inferences from demand-sentences through a convention (Ubereinkunft)

or
trick (Kunstgriff ). In fact, in his main work Die Definition [49], Dubislav
gives a similarly formalistic characterization of propositional logic (and later
predicate logic). There, Dubislav starts with a pure, game-like calculus that
is played with pieces and signs (, , brackets) by first arranging some
into initial positions and then replacing pieces and re-arranging pieces into
new game positions according to the rules of the game. This game then
becomes the calculus of propositional logic by interpreting its elements as
26
The term of commanding something by implication was introduced by Geach [61].

Alchourr
on [3] speaks of the consequences of what is prescribed as indirectly prescribed,
Alchourr
on & Bulygin [5] write that e.g. if teacher commands that all pupils should leave
the class-room, he also implicitly commands that John (who is one of the pupils) should
leave the class-room, even if the teacher is not aware of the fact that John is there, and
in [8] they view the deductive consequences of norms as implicitly promulgated, where
the deduction process is equivalent to the modified Dubislav convention (DCM). Hare [88]
and Rescher [184] both propose to define command inferences in terms of implicitly given
commands analogously, Reschers assertion logic [185] is concerned with assertions that
a speaker is implicitly committed to in virtue of overtly made assertions. It was shown
above that Reschers explanation of imperative inferences is equivalent to the modified
extended convention (DECM).
23
indicated: the pieces as propositions, the signs as not, or, and brackets, the initial positions as axiomatic basis, the game rules as the usual
rules of substitution and modus ponens, and the achievable game positions
as derivable formulas. This characterization of propositional logic is meant
by Dubislav as an exposition of Booles [34] idea that the validity of the
processes of analysis does not depend upon the interpretation of the symbols which are employed, but solely upon the laws of their combination. In
Dubislavs view, which he calls the formalistic theory, this description of
logic functions as a mould for all scientific theory: a theory is constituted
by a pure calculus (of formulas and rules), combined with a fixed interpretation. Observational sentences are captured in formulas that can be used
alongside axioms or derivable formulas of the system to derive other formulas
within the calculus. Then the assignment of these derived, or better: calculated formulas is reversed, i.e. they are translated back into observational
sentences. If these are regularly true, then the observational sentences are
explained by the theory. If a calculated observational sentence turns out
to be false, then the theory is erroneous. Thus it also becomes possible to
decide between competing, non-isomorphic theories.
The usefulness of the Dubislavs formalistic approach for the problem
of imperative logic is immediate. In fact, Dubislavs own proposal in [50]
satisfies all requirements in [49] for being a theory of imperative inference:
there are entities that may function as premisses and conclusions, namely
imperative sentences. There is an interpretation that assigns each imperative sentence a formula, namely that of the indicative parallel sentence in
the calculus of ordinary logic. There is a calculus, namely ordinary logic,
that tells us what formulas can be derived from the formulas assigned to the
imperative sentences that function as premisses. And finally, this assignment
is reversible to provide derived imperative sentences. Other authors taking
their cues from Tarskis [210] syntactical definition of consequence relations
and deductive systems,27 Tarskis [211] definition of truth,28 Gentzens [63]
27
Cf. Alchourr
on & Bulygin [8] who employ a formal consequence relation to explain
what norms are implicitly promulgated by a set of norms.
28
Both R
odig [187] and Yoshino [257] appeal to Tarski and argue that meaningful operations with prescriptions are made possible by supposing that normative attributes like
obligatory or punishable may be applied to actions. Rodig draws attention to the problem of objective verifiability and therefore truth of such statements. But he circumvents
the problem by assuming that meta-language truth conditions can be given, which is sufficient to handle normative attributes as normal predicates in the object language. Rodig
24
idea that to define a symbol is to give rules for its introduction and elimination,29 and Wittgensteins dictum that the meaning of a word is its use ([239]
43) have similarly argued that instead of searching in vain for analogues
of truth values, it suffices for an explanation of imperative inferences to give
formal rules for obtaining imperatives from other imperatives.
If this formalistic approach to the logic of imperatives is accepted, we are
still not finished yet. If the assignment of formulas, calculations and backtranslations of derived formulas are to be more than a game, there must be
some way to judge the adequateness of the theory, and be it only to decide
between competing proposals.30 In analogy to Dubislavs general approach,
where a theory is only an explanation of phenomena if its calculated observational sentences are regularly true, one should require of any proposed
logic of imperatives that the imperative it derives from other sentences are
normally not true of course, but accepted as implicit in other sentences
that are used as premisses. This resembles what is called the soundness of a
calculus: if the calculus allows false (unacceptable) conclusions to be drawn
from true (accepted) premisses, then it must be discarded as unsound.31
I now turn to the question of adequacy in this sense.
and Yoshino then use these predicates to formalize e.g. a norm that says that helping
in an emergency situation is obligatory as acts: In emergency(act) Helping(act)
Obligatory(act). The puzzling thing is that if this really is a prescription (norm), i.e. it
makes so far unregulated acts of helping in cases of emergency obligatory, then for no
such act the truth of the part Obligatory(act) can be established before the truth of the
whole is established. This at least differs from Tarskis compositional truth definition.
29
Cf. Alchourr
on & Martino [10] who provide a calculus with an introduction rule
for a prescriptively interpreted O-operator, where their rule corresponds to the modified
Dubislav convention (DCM) plus a requirement of joint satisfiability.
30
It seems consensus that there must be some test of adequacy. Weinberger [224] writes
that one must test a rule for the logical manipulation of norm sentences for its adequacy
for the area of normative thought, and Sosa [201] speaks of a control of commonsense
that is necessary because otherwise there would simply be no end to the possible logics.
31
The other possibility, that the calculus does not provide all the inferences from premisses that are acceptable (usually called completeness), is less harmful and can be dealt
with by e.g. refining it. For a similar definition of adequacy cf. Chellas [44] p. 4, where
however the terminology is vice versa.
1.5
25
Rosss Paradoxes
Shortly after Jrgensens dilemma and Dubislavs workaround for a logic of

imperatives had been described, Alf Ross re-considered inference schemes in
the most simple form, where a new imperative is inferred from one imperative premiss, i.e. where the scheme used is that of Dubislavs convention
(DC). The following is an instance of such a scheme:
!A
!(A B)
Here, !A means (as now usual) an imperative sentence that is satisfied if and
only if the descriptive sentence A is true, and !(A B) is an imperative that
is satisfied if and only if either A or B are true. It is immediate that the
second imperative can be inferred from the first sentence !A by Dubislavs
convention. Fine, said Ross, let !A be interpreted as the imperative post the
letter, so we can infer from the imperative post the letter the imperative
post the letter or burn it !(A B). So
(1)
Post the letter!

Therefore: Post the letter or burn it!
is a valid imperative inference. Ross himself points out that his paradox is
not paradoxical if this validity of an imperative inference is understood in
the sense of a logic of satisfaction. If the letter is posted and the imperative
!A satisfied, then the imperative !(A B) will likewise be satisfied this
is no more paradoxical than that A B can be inferred from A. But if
the meaning of imperative inference refers to anything like the validity or
existence of an imperative, then Ross claims that his inference is not only
not immediately felt to be evident, but rather evidently false.
Why does Rosss example of an imperative inference seem paradoxical?
In particular, regarding the formalistic theory of imperative inference given
in the last section, why should it be paradoxical to say that if one uses the
imperative !A for commanding, then one implicitly also commands !(AB)?
One explanation has been that that by using a disjunctive imperative, i.e.
an imperative sentence that like !(A B) is satisfied if some state of affairs or
some other state of affairs holds, the authority has left it to the subject how
to satisfy her command. Suppose Romeo hands a letter to Mercutio with
the words Post the letter or burn it, but relieve me from deciding its fate
26
and mine, would his friend not be free to do as he pleases? Analyzing this
freedom, it has been argued that giving a command entails an imperative
permission or implicitly authorizes to carry out the actions required to satisfy
the command.32 So the imperative post the letter or burn it would contain
the permission I hereby permit you to post the letter or burn it. Now
explicit disjunctive permissions are often understood in a strong sense that
grants both disjuncts: when someone says help yourself to a cup of coffee or
a cup of tea, then the guest is permitted to help herself to coffee and also
permitted to help herself to tea (though possibly not both). So one obtains
the following chain:
(2)
Post the letter!

Therefore: Post the letter or burn it!
Therefore: You may post the letter or burn the letter!
Therefore: You may burn the letter!
But it seems counterintuitive to say that by ordering a letter posted one

permitted it to be burned.33 To avoid this result, one may argue that it is
not the first step in (2) that is problematic, but the second, i.e. we should
not be allowed to infer a strong permission from an imperative. Yet nothing
seems wrong with the following piece of Mercutios reasoning:
(3)
Romeo asked me to post the letter or burn it.

Therefore: I may post the letter or burn the letter, as I wish.
Therefore: I may burn the letter.
The reason why the inference from the first line of (3) to the second line seems
not objectionable, while the similar inference from the second line in (2) to
its third line appears somehow wrong, may lie in the fact that the imperative
to post the letter or burn it that is used in the reasoning is only implicit,
i.e. derived, while Mercutios reasoning was about an imperative that was
explicitly used by Romeo. So one could modify ones view on the second
step in (2) by saying that one is only allowed to infer a strong permission
to do what is commanded if this command is not itself derived. I return to
32
Cf. Chellas [44] p. 19 for the term imperative permission and Keene [120] for the
implicit authorization.
33
The idea to explain the counterintuitive nature of Rosss paradox using the also, or
even more, counterintuitive inference to you may post the letter or burn it was von
Wrights in [249] pp. 2122, also cf. von Wright [256] pp. 121122 and Hintikka [96].
27
such a distinction between explicitly given premisses and implicitly given

imperatives in a moment. But consider the example from the last section,
where an officer was commanded to go via Coldstream or Berwick, and (a
little later) told not to go via Coldstream, where both commands were viewed
as implying the command to go via Berwick. The proposed modification
would still allow us to make the following inference:
(4)
I was commanded to go via Coldstream or Berwick.

Therefore: I may go via Coldstream or Berwick, as I wish.
Therefore: I may go via Coldstream.
So it seems the authority contradicted herself when ordering (a little later)

not to go via Coldstream, i.e. first a choice between the two routes was
granted, and later this choice was retracted, or rather: the second command
modified the original command.34 Whenever a command contradicts, cancels
or modifies another command, the conflict may be absorbed e.g. by application of the rule of lex posteriori, which says that as a rule authorities should
be considered competent to modify their own orders. But the puzzling thing
is that the example was originally presented as a smooth application of imperative logic as facilitated by Dubislavs convention (DCM). Nothing made
it appear as if there is some contradiction or modification involved and that
more is used or required than just a flat application of the rules.35
An answer to these problems could be to give up the idea of strong imperative permission altogether: without it, the agent cannot reason that
burning the letter is permitted. But while strong permission might be seen
as problematic,36 it is not clear why it should be altogether discarded. In
particular, nothing seemed wrong with assuming strong permission in the
case of Romeos request (3). But whatever view is taken on strong permission, there is another point that makes Rosss paradox seem counterintuitive
34
According to Hare [88] it is just a conversational implicature that gets canceled. But
it seems that by saying go via Berwick or Coldstream the authority really leaves it to
the agent which route she wants to take and later retracts this choice , while someone
who says e.g. the tickets are upstairs or in the car, and later adds they are not in the
car only made it seem as if the tickets could be in either location. If the order was only
given further orders pending, as Hare also argues, then the first order was not complete,
because it left the agent unable to determine how to fulfill it. It is as if the authority had
said in the middle of a sentence: hang on, Im not finished yet, Ill be right back.
35
This was the point in Williamss [238] criticism of Hares [88] scheme.
36
Cf. Stenius [206]: Free choice permission is too strong a concept to be useful.
28
without appealing to some implied permission: Imagine that, having been

given the command Go via Coldstream or Berwick, the agent finds the road
Coldstream blocked. Then the following reasoning seems logical:
(5)

I cannot go via Coldstream.
Therefore: I should go via Berwick.
It seems the kind of deliberation that one would expect of reasonable agents.
Likewise, Mercutio, having been asked by Romeo to post the letter or burn
it, might be found reasoning in the following way:
(6)

For fear of Tybalts revenge, I cannot bring myself to post
the letter.
Therefore: I should burn the letter.
One might dispute whether Mercutios fear is really on a par with a road
blocked e.g. by a landslide. But if we suppose it is, then Mercutios reasoning
seems as impeccable as that of the officer. Now return to Rosss paradox:
here the agent was ordered to post the letter. Implicit in this imperative, so
we are told by Dubislavs convention (DC), is the imperative post the letter
or burn it. Imagine that the agent is not able to post the letter for some
cause (the postal workers are on strike and the mail bins have been locked
up). So the agent could reason in the following way:
(7)
I was (implicitly) ordered to post the letter or burn it.

I cannot post the letter.
But this reasoning is absurd. Just because the agent cannot fulfill her obligation to post the letter, this does not mean that she is obliged to do something
that was never mentioned, and in fact could be anything: the words burn
the letter could be replaced e.g. by go to the zoo, kill a passer-by or love
your neighbor and the inference would be just as valid if it is valid.37
Now the agent, in reasoning in the above settings, used indicative statements about natural facts like that something cannot be done to reason
37
This is Weinbergers [225], [226] explanation of why Rosss paradox poses a problem.
29
about the imperatives go via Berwick or via Coldstream, post the letter
or post the letter or burn it. But inferences that mix imperatives and indicatives are notoriously troublesome and should perhaps be avoided. As
MacKay [139] points out, both of the following inferences
Go fly a kite!
You are going to drop dead.
Therefore: Drop dead!
You are going to fly a kite.
Drop dead!
Therefore: Go fly a kite!
are validated by Dubislavs extended convention (DEC), where both inferences seem plainly invalid, and so perhaps (DEC) should not be accepted.
Yet consider again the case of the officer. Imagine that it was not the original
authority that issued the command not to go via Coldstream, but someone
else, like the officers husband (who in the past had some bad experience on
this road). Since there is some discretion in the authoritys order, there is
no reason why the officer should not give in to her husbands request, and so
the following reasoning of the officer seems correct:
(8)

My husband asked me not to go via Coldstream.
Therefore: I should go via Berwick.
Similarly, we can imagine Mercutio to reason in the following fashion:

(9)

Tybalt threatened me not to post any of Romeos letters.
But then the following reasoning of the agent to whom Rosss imperative
post the letter was addressed must be likewise correct:
(10)
I was (implicitly) ordered to post the letter or burn it.

I have been asked not to post the letter.
There is some discretion in the (implicit) order to post the letter or burn it,
30
so why should the agent not take an additional request into account? The
only difference between the (9) and (10) is that in (9) the reasoning appeals
to an explicitly used imperative, whereas in (10) it starts by considering an
order that was only implicit in the use of some imperative. So maybe what
was wrong was that derived imperatives were used, without paying enough
attention to the fact that the derived imperatives are only part of a system, that the explicitly used imperatives have not ceased to exist, that
imperatives that are only derived do not exist on quite the same level as
explicit imperatives, or that the agent is somehow expected to make use of
the logically strongest information that is available.38 So we are back at the
proposal that a difference must be made between explicitly used imperatives, and imperatives that only derive from explicit imperatives. But to
require that reasoning with imperatives starts with explicit imperatives, and
must not start with imperatives that are only inferred, reveals an unusual,
non-classical meaning of imperative inference. For classically, logical inferences may very well be conducted by proving first that some assumptions
have some desired conclusion, and then show that the assumptions follow
from an accepted set of premisses. This is facilitated by the transitivity of
classical consequence: if A Cn(B) and B Cn(C) then A Cn(C) (consequences of the consequences are also consequences), or the monotonicity
rule: if A Cn(X) then A Cn(X Y ) (what follows from some axioms
also follows from a larger set of axioms).
Rosss paradox seems to demonstrate that given the imperative inferences
provided e.g. by Dubislavs convention, it becomes necessary to distinguish
between the imperatives that are explicitly given and the imperatives that are
inferred: agents can use the former for their reasoning, but not always the latter, or not the latter by themselves, which makes reasoning with imperatives
somehow non-classical. And so there may yet be another way to get around
the difficulties: perhaps Rosss example is not really a case of an imperative
38
R
odig ([187] p. 184185) points out that by deriving the norm to post the letter or
burn it, the original order to post the letter does not cease to exist, and that it is the
conjunction of both norms that must be satisfied. That the entailed norms do not exist
in quite the same way as explicit norms is the idea of von Wright e.g. in [255] and [256] p.
114 and p. 122. According to Stenius [206], the use of post the letter or burn it carries
the tacit information that a stronger regulation like post the letter does not belong to
the codex. For the idea that using a weaker sentence post the letter or burn it violates
a conversational presupposition cf. Hintikka [96]. Also cf. Hamblin [74] p. 88: implicit
imperatives may be different from the real thing, and we should be wary of loading them
up with the full range of imperative properties.
31
inference. Perhaps (1) is simply invalid. It is obvious that the scheme is an

application of Dubislavs convention, so (DC) must be modified. One way
to do that is to let the logic that is used for the right hand side inference in
figure 1.1 not classical logic (propositional or predicate logic), but some other
logic that does not allow one to infer A B from A. Let Dubislavs convention therefore be reinterpreted in terms of a very strong relevant deduction
developed by Weingartner & Schurz [237] and Weingartner [236], which was
tailored specifically to eliminate Rosss paradox.39 Unlike other relevant logics, their R-consequence not only blocks the inference of A B from A, but
also that of A B from (A B) A or from (A B) B. Dubislavs scheme
is then changed accordingly (cf. the next figure). Weingartner & Schurzs
Figure 1.4: Dubislavs convention with relevance.

proposal produces a very strange consequence relation: not only monotonicity fails, but also reflexivity, i.e. X Cn(X) is not valid. But as we cannot
derive A B from A any more, we also cannot derive !(A B) from !A, and
so Rosss paradox is solved.
However, in [188] Ross also presented another paradox, a variant of his
first paradox, that remains valid even on such a relevant reinterpretation of
(DC): the inference from !(A B) to !A, where A B means the sentence
that is the conjunction of A and B. I consider the paradox in the form of
Weinbergers paradox or the window paradox:40
39
R-consequence is defined by Weingartner & Schurzs [237], [236] in the following way:
a propositional formula A is a R-consequence of a set of formulas X iff (i) X classically
implies A and (ii) it is not possible to uniformly replace a proposition letter at at least
one of its occurrences in A by a random proposition letter without making the classical
inference invalid.
40
The origin of the example is unclear. The name Paradox of the Window is used e.g.
by Stranzinger [207] and Weinberger [232].
(11)
Close the window and play the piano!

Therefore: Close the window!
(12)
Close the window and play the piano!

Therefore: Play the piano!
32
Suppose that wants x to practice the piano, but neighbors have already
complained about the disturbance and even called the police on a previous
occasion. So does not want x to play the piano while the window is open.
Closing the window will reduce the noise so much that the neighbors are left
with nothing to complain about. Suppose then that sends x to play the
piano, using the words close the window and play the piano. A little bit
later, the following discussion ensues between and :
:
:
:
I told her to play the piano, but I didnt hear her doing it all
afternoon.
Well, at least she closed the window.
Why should she do that?
Here, the positive view on xs behavior by is not accepted by . Closing

the window by itself is meaningless. It might even be unwanted in general
it blocks out fresh air if it werent for the sake of piano practice. But
backed with the inference (11), can continue in the following way:
:
You ordered her to close the window, thats what she did, so
she did something right, didnt she?
Now consider the following, alternative dialogue:

:
:
:
She practised the Khachaturian with the window wide open.

What shall we tell the police this time?
It was you that told her to play the piano.
But I didnt. She was also to close the window.
s reproach for xs playing the piano is not accepted by , because playing

the piano without closing the window first was not what had asked x to
do. However, backed with the inference (12), could reply in this way:
:
You ordered her to play the piano, thats what she did, so
dont try to wiggle out of your responsibilities.
33
In these dialogues, s position seems natural, while s reaction is strange

and uncomprehensible. But given the inference schemes (11) and (12), is
right: from s command, the imperatives close the window and play the
piano can be inferred so we are told by Dubislavs convention (DC), and
Dubislavs convention restricted by relevant implication. Moreover, these
derived imperatives are used by as imperatives are meant to be used,
namely compared with reality, and reality accordingly qualified as right or
wrong. So x did something right by closing the window, as x satisfied an
(implicit) imperative, and similarly when playing the piano without closing
the window. But intuitively, closing the window by itself produces nothing
good, and playing the piano with the window open seems a clear violation
of obligations and not satisfactory in any way.41
The window paradox seems to arise whenever the states of affairs mentioned in the imperative are only conjunctively desired by an authority. That
for this reason we cannot detach conjuncts in wishes, i.e. we cannot conclude
from she wishes for a and b that she wishes for a was pointed out by Menger
[150] for the case of complementary goods, e.g. when a is a cigarette and
b is a match, for one may not wish either one of the goods by itself. Ross
[188] points out that the same difficulty arises for imperatives, e.g. when the
imperative is to write a letter and post it. Other examples have included
the imperatives take the parachute and jump, pay the bill and file it or
fill up the boiler with water and heat it.42 Goble [65] showed that even a
seemingly innocuous obligation to sing and dance at Genes party may be
planted in a setting that makes it impossible to speak of fulfilling the obligation when only one act, singing or dancing, is performed. To determine
whether an imperative is separable or inseparable, i.e. whether doing A
alone produces something right with respect to an imperative !(A B) or
not, it is necessary to examine the intentions and wishes of the authority
41
It gives the paradox a further twist if we imagine that playing the piano with the
window open is explicitly forbidden. For by Dubislavs convention (DC), the imperative
!(A B) (dont play the piano while the window is open), is derivable from the
imperative !(A B) (close the window and play the piano). But it seems that the
additional prohibition is best formalized as a conditional imperative (in Hofstadter &
McKinseys [99] formalism: B !A). Conditional imperatives pose other problems
outside the current topic. In any case, one would still have to say that playing the piano
with the window not closed was satisfactory with regard to some (derived) imperative.
42
Cf. Hare [88], Weinberger [223] and [235]. These difficulties led Weinberger to reject
the validity of an inference from !(A B) to !A in his publications since [223].
34
that used the imperative, it is not a matter of logic.43

To solve these difficulties, Kenny [122] proposes a logic of satisfactoriness. This logic uses a set of propositions to represent the wishes of the
authority under considerations. A fiat (an impersonal imperative like let
there be light) is called satisfactory if and only if whenever the fiat is satisfied then every proposition in the set of wishes true. Finally an inference of
one fiat from another fiat is defined as follows:
!B may be inferred from !A in the logic of satisfactoriness
if and only if
if !A is satisfactory then !B is satisfactory.
It is clear that the troublesome inferences (1), (11) and (12) are invalidated
by this logic of satisfactoriness: when posting the letter is satisfactory for
the wishes of the authority, then burning the letter need not be so. Likewise,
if closing the window and playing the piano is satisfactory with respect to
all wishes, then playing the piano alone does not guarantee that the wishes
of the authority are also satisfied. But Kennys approach gives rise to other
paradoxes: in the logic of satisfactoriness we can e.g. derive:44
(13)
Open the door!

Therefore: Open the door and wear a tie today!
The inference is clearly absurd and so Kennys logic does not help us to solve
the paradoxes.
In Rosss first paradox, the imperative to post the letter or burn it was
inferred from the imperative to post the letter, thus forcing one to acknowledge that some (though only inferred) imperative is satisfied by burning the
letter. In the window paradox we could infer the imperative play the piano from the imperative close the window and play the piano, thus forcing
us to acknowledge that an (inferred) imperative is satisfied when the piano
is played with the window wide open. In both cases, we would much rather
say that no imperative was satisfied by burning the letter that was meant to
be posted, and by playing the piano with the window open when it should
have been closed. This, I think, is the main cause why Rosss paradox and
the window paradox give rise to counterintuitive feelings, or are paradoxi43
44
The terminology here is that of Hamblin [74] p. 184.

A similar counterexample was given by Gombay [71], also cf. Sosa [201].
35
cal. So we should not be allowed to infer such imperatives. So Dubislavs

convention is not an apt theory to explain how an imperative may be derived
from another one. And so we are back at square one: all theories, including
the formalistic approach, have so far failed to explain what it means to infer
an imperative from some other imperative in spite of Jrgensens Dilemma.
1.6
Ordinary Language Arguments
Maybe it is not really the case that all options have run out to redefine
Dubislavs scheme in a way so that it avoids Rosss paradoxes. Maybe we
have to replace the classical logic that appears in his scheme by yet another
logic, or develop such a logic.45 But it is hard to see what kind of logic this
could be, since most logics, including other non-monotonic logics, will permit
us to either infer !(A B) from !A or !A from !(A B), and so at least one of
the two paradoxes will arise. So I think, after all these troublesome attempts
to define a logic of imperatives, it is worthwhile to take another look at
Poincares proposal that originally started the controversy.
Poincares only explicit example of an inference with an imperative conclusion has the following form:
(1)
Do this!
This cannot be done without that.
Therefore: Do that!
The following is an instance of this scheme:

(2)
Drive me to the airport!

To get to the airport, one must drive in a northerly direction.
Therefore: Drive me in a northerly direction!
In which setting could these sentences be used? Suppose I have entered a

taxi and used the above sentences. But some confusion could arise. The
driver could reply: So what do you want me to do, drive you to the airport
or just drive north? The driver needs a direction. Ordering her to go to the
airport alone is sufficient for this, and the behavior expected of a passenger
45
Cf. Keene [120]: What we wanted here is a logic of actions, in which a well-defined
concept of inclusion plays a leading role.
36
entering a taxi. Using two imperatives where each contains an instruction of

where to go is unexpected and confusing.
So suppose I have just used the sentence drive me to the airport. A little
later I realize that we seem not to be going north, and I say to my partner:
Is she hijacking us? I ordered her to go to the airport, and
the airport lies to the north. So she ought to be driving us in a
northerly direction. But she is not.
This reasoning seems flawless. Yet it only involved sentences about imperatives, and did not involve sentences in the imperative mood, and so it cannot
be an example of an imperative inference. But maybe this would be a good
time to say to the driver:
(3)
I ordered you to go to the airport.

To get to the airport, one must drive in a northerly direction.
Therefore: Drive me in a northerly direction!
But here the two sentences that function as premisses are both descriptive.
Since Poincare explained that an imperative cannot be derived from indicative premisses alone (and there is no reason not to follow him), this cannot
be an imperative inference, and there must be something more involved than
the drawing of a logical conclusion. One such other function of the therefore appearing at the front of the last sentence of (3) is not to reason, but
to motivate, as in:
(4)
The car is broken.

Therefore: Take the bus into town!
Here the speaker motivates the imperative to take the bus by explaining that
driving into town is impossible, since the car is broken. So similarly, what
seems to happen in (3) is that I motivate my (new) imperative drive me in
a northerly direction by an already given command and an assumed fact.
Consider again the proposed inference (2). Just like indicative inferences
are explained by the fact that someone who accepts (or: assents to) the
premisses must also accept the conclusion, Hare [87] has argued that an
imperative inference is one where someone who assents to all imperative
premisses must also assent to the imperative conclusion:
A sentence p entails a sentence q if and only if the fact that a
37
person assents to p but dissents from q is a sufficient criterion for

saying that he has misunderstood one or other of the sentences.
(...) A person who assented to this command [Take all the boxes
to the station], and also to the statement This is one of the
boxes and yet refused to assent to the command Take this to
the station could only do so if he had misunderstood one of these
three sentences.
But what does it mean that a person assents to a command? Suppose
Johns mother tells him John, clear the table and do the washing up, and
Johns little sister echoes: John, do the washing up. If John assents to
his mothers order, does he also have to assent to an order by his sister,
whom he might not accept as an authority? Perhaps the analysis assumes
identity in the person who uses the commands. Suppose then it was not
Johns mother but some officer who used the imperative, and John is not
obliged qua son, but as this officers orderly. The second command is also
used by the officer, maybe a little later. But suppose that John is only
obliged to the officer if the commanding is done in a certain fashion, e.g.
when the officer is standing up, or when the officer is not drunk, and that
when the second imperative was used the officer was, as a matter of fact,
not standing up or already had more than her fill. Or suppose that John is
not an orderly, but some djinni, and the officer is the person who rubbed the
lamp, but that, when the first imperative was used, this already was the last
of the three wishes that had been granted. Does John, in these cases, have
to assent to the second command? It seems that such an interpretation of
assent would have to get involved into reasoning about whether the act of
using an imperative really creates a command. But it did not seem as if
such reasoning is involved in Hares proposal.
So let the word assent be understood in its weakest possible interpretation. A person could hardly be said to assent to a command given to her if
she did not satisfy or to try to satisfy it. Returning to the situation where
I have asked the taxi driver to take me to the airport, when the taxi driver
assents to this request, she will start driving me to where she thinks the airport lies, i.e. start to satisfy, or try to satisfy, my request. If the taxi driver
agrees that the airport lies in a northerly direction, she will, in obeying my
request, eventually drive in what she thinks is a northerly direction. So she
might be said to additionally satisfy, or try to satisfy, a request to drive me
in a northerly direction, had such a request been made. But did I request
38
the taxi driver to drive me north? I might be absolutely sure that the airport is to the north, but I would still blame the taxi driver for not going
where I requested if, opposite to what I believed, the airport is in fact to
the south-west of my starting point and the driver still went north. So I did
not utter such a request, and would not even imply such a request, lest I be
charged by the driver for going there instead of the airport. All we can say
is that the taxi driver would also be satisfying, and so seemingly assenting
to, a purely hypothetical request to drive me in a northerly direction, if she
satisfies the request to drive me to the airport and the airport does in fact
lie in a northerly direction. But this is again not a logic that infers one imperative from some set of other imperatives and/or indicatives, but the logic
of satisfaction as explained in sec. 1.4.1.46
It would be nice to have real life examples, cases of ordinary reasoning
with imperative premisses and an imperative conclusion, i.e. instances of
(ImpInf)
!A
!B
where !A and !B are sentences in the imperative mood, and where the use
of the inference not the imperatives is either accepted in some ordinary
discourse, or opposed (and the person who uses it blamed for being unreasonable or illogical).
Use of indicative arguments in everyday discourse often occurs in singular
sentences, like
(5)
(6)
(7)
Unemployment is rising, so there are not enough jobs created.

She has got an A in English, so she achieved top-marks in at least
one subject area.
I have read all of Vladimir Nabokovs novels, so I have read Pnin.
Here two descriptive sentences are linked with the adverb so (similar adverbs
would be therefore or hence). (5) seems analytical if one understands
enough to be elliptical for enough to make up job-losses elsewhere. (6)
is analytical if one knows that A is a top-grade and that English is one
of several high-school subjects. (7) is made into a logical argument by the
assumed background knowledge that Pnin is a novel by Nabokov. It is
often not easy to distinguish such indicative arguments from sentences that
present reasons, motives or are otherwise explanatory, for these also use the
46
This resembles the criticism by Keene [120] of Hares proposal.
39
form of descriptive sentences that are concatenated by an adverb like so or

therefore, as in the following examples:
(8)
(9)
(10)
I couldnt get the car started, therefore I took the bus.

I wanted to make friends with her, therefore I asked her if she
would go shopping with me.
There were holes in the roof, so birds had come in and were roosting
in the rafters.
(8) explains why today the speaker used the bus. Since the bus need not
have been the only means to get into town, or the speaker may have stayed
at home, the hearer cannot just conclude the second part from the first. (9)
presents the psychological motive why the speaker asked the other person to
go shopping with her. Other people might have been motivated differently
by the desire to make friends with that person. In (10), a natural event is
explained by a certain state of affairs. Again, this is not a logical argument:
the birds could also have not flown in, or flown in but not nested in the
ceiling. Now the adverbs so andtherefore can also be used to meaningfully
link imperatives. Consider the following examples:
(11)
(12)
(13)
(14)
Stop the rise of unemployment, so see to it that more jobs are

created!
Make your guests comfortable, so introduce your guests to each
other!
Dont let vermin into you house, therefore patch up the roof!
Read all of Nabokovs novels, so read Pnin!
(11) might be encountered in some political debate. At first it appears to

be a good argument, but then doubts arise: is the speaker really appealing
to logic, or is she just complementing her first imperative by a second, more
specific one, as when we say: Go there! Go there now!? And one could also
stop the rise of unemployment by e.g. prohibiting companies to dismiss their
workers, or making it more difficult for them (maybe 5 was not so analytical
after all). Then (11) would seem to be rather a case of a motivating use
of so: the imperative to see to it that more jobs are created is motivated
by the primary aim to stop unemployment. Likewise, in (12), the advice
to introduce guests to each other is rationalized by the more general aim to
make guests comfortable. It is hard to see what could be analytical here: to
ease tensions, the host may equally encourage the guests to guess each others
40
names, or serve them plenty of alcohol, or maybe the guests are easygoing
and do not really require any effort on the hosts part to make themselves
at home. Similarly, in (13) the more readily accepted advice to keep vermin
out of the house is used as a rationale to make the addressee accept the
drudgery of having to patch up the roof. The most promising candidate
for an appeal to analyticity seems to be (14), i.e. that the imperative to
read all of Nabokovs novels includes the imperative to read Pnin, given the
background knowledge that Pnin is a novel by Nabokov. Note that when
making the background knowledge explicit, it becomes a case of Dubislavs
extended convention (DEC). Such a sentence may be used e.g. by a teacher
of a literature course when addressing her students. But again we cannot
rule out that this is just a case of complementing an imperative by a second,
more specific one, as we sometimes do to get things done.
Adherents of Dubislavs convention (DC) must also accept the following
argument:
(15)
Aim for an A in English, so aim for top-marks in at least one

subject area!
But it seems dubious what reason the speaker could have for adding the so
part. Just aiming for top-marks in some subject area is clearly not what the
speaker wants the addressee to do. More meaningful would be the converse,
(15a)
Aim for top marks in at least one subject area, so aim for an A in
English!
where the advice to aim for A in English is rationalized by the wish to have
the student achieve top-marks somewhere. But since the student could not
know from the first imperative that it was the subject of English that the
speaker wanted her to achieve top marks in, this would like (12) and (13)
rather be a motivating so, and not a use of so that appeals to a logical
capability.
Matters are further complicated by the fact that expressions of the following kind can also be meaningfully employed:
(16)
(17)
(18)
The car isnt working properly, so take the bus!

I forgot my keys, therefore leave your key under the mat!
Gill is your best friend, so invite her to your party!
41
In all three sentences, the first part is descriptive and the second is in the
imperative mood. We have already noted in the case of (3) that such arguments exist, but for anyone who agrees to Poincares thesis that imperative
conclusions do not follow from an indicative premisses it is clear that (16)
(18) cannot represent valid arguments. (16) seems again a case where the
so is used to motivate the advice that is expressed by the imperative. The
so does not express a logical relation, for sometimes it is better to use a car
that stutters than a coach that wont take one back. In (17) the indicative
gives a reason why the speaker wants her request to be followed. According to Hamblin [74], such reason-providing indicatives are often attached to
advice-expressing imperatives, yet here the imperative might also be an order
(e.g. of a parent). For the same reason the speaker might have ordered the
agent to hand over her key, and not to leave it under the mat, and so what
is expressed is again not a logical relation. (18) seems also like presenting
a motive for inviting Gill to the party (she is the addressees best friend),
but here things might be a bit more complicated the expression could be
elliptical for:
(18.a) Invite your best friends to the party, Gill is your best friend, so
invite her to your party!
This is very similar to what Dubislav considered a valid argument, namely
his inference from thou shalt not kill to Cain shall not kill Abel. But then,
(18) might also be elliptical for
(18.b) Gill is your best friend, one invites ones best friends to ones parties,
so invite her to your party!
where the second part (which is not in the imperative mood) appeals to the
existence of a rule that the speaker might consider binding, or binding for the
addressee. Then this is rather a case of reason-giving, and not of a logical
inference: the speaker motivates her imperative by asking the speaker to
conform to some preexisting rule.
To tell the uses of therefores and sos that are motivating, reasongiving or explanatory in a non-logical sense, apart from those that separate
the premisses from the conclusion in an argument that is intended to be a
logical one, we can use the following trick: instead of therefore or so, use a
clause like ... It follows logically from this that ... to separate the sentences.
42
The new phrase makes the appeal to a logical capability explicit. Where the
original adverbs so and therefore were used to indicate a (claimed) logical
inference, the new formulations
(5.a)
(6.a)
(7.a)
Unemployment rates are rising. It follows logically from this that

not enough jobs are created.
She has got an A in English. It follows logically from this that she
achieved top-marks in at least one subject area.
I have read all of Vladimir Nabokovs novels. It follows logically
from this that yes, I have read Pnin.
appear only to be changes in expression. The speaker, just as before, appeals

to a shared understanding of words, concepts and background knowledge, to
make the second sentence seem to be expressing nothing new, but only a logical consequence from what has been said before. Note that it does not matter
whether the arguments are, in fact, analytical. People sometimes think they
use valid arguments when they are not. But the rephrased sentences make
it clear that the speaker intends the sentences to be just that. And the new
formulations seem not to change the meaning of the original sentences whenever a logical use of the adverbs so and therefore was really intended. By
contrast, when the first part was used to give some background information,
a reason, explanation or motive, the rephrased expressions appear odd:
(8.a)
I couldnt get the car started. It follows logically from this that I
took the bus.
(9.a) I wanted to make friends with her. It follows logically from this
that I asked her if she would go shopping with me.
(10.a) There were holes in the roof. It follows logically from this that birds
had come in and were roosting in the rafters.
The phrase it follows logically from this makes again an appeal to some
shared understanding of used words, concepts and background. But here,
this background knowledged obviously does not allow one to conclude the
second sentence from the first. The listener could not have known from the
first sentences in these examples that the speaker took the bus, asked someone
to go out shopping or has birds nesting in the roof of her house. So claiming,
as the rephrased sentences do, that the second part can be concluded from the
first, makes the sentences seem irritating, weird and false, while the earlier
43
sentences appeared quite harmless.

Now consider what happens if such a method is used on imperatives. So
far, (14) seemed the best candidate for a sentence that appeals to logic, so
I will concentrate on this example. First note that
(14.a) Read all of Nabokovs novels. It follows logically from this that read
Pnin!
is not grammatical. Now if the grammar is difficult, this may already hint
at little usage of such statements, but I think that instead of the that we
can easily use e.g. a colon, corresponding to a pause in oral language, as in
the following expression:
(14.b) Read all of Nabokovs novels. It follows logically from this: read
Pnin!
But here, the part that follows the colon seems strangely detached. Is this a
command, i.e. is the speaker, using the expression following the colon, still
giving a command? Or is the emphasis on the part before the colon, and
so the purpose of the second sentence is merely to tell (truly or falsely) that
some consequence relation holds? The impression that this is a strange use
of words increases if we add the subject of the request:
(14.c) John, read all of Nabokovs novels. It follows logically from
this: John, read Pnin!
Here, the phrase it follows logically from this makes it appear as if the
speaker was not giving commands to John at all. It seems what the speaker
really does is talking about logical relations between sentences maybe it is
a logician presenting an example of an imperative inference. So perhaps we
should try out another phrase:
(14.d) John, read all of Nabokovs novels. We can conclude from this:
John, read Pnin!
Yet this expression also has a false ring: who is doing the commanding of the
conclusion the speaker? Or the we that is to do the concluding? Do the
speaker and the listeners all join into giving John the command? Apparently
it was wrong to use the first person plural, and so we might want to change
44
the sentence into:

(14.e) John, read all of Nabokovs novels. I conclude from this: John, read
Pnin!
But this seems to be the worst alternative so far. Is the speaker concluding
the last sentence? Or is the speaker commanding it? And if so, then why
is she saying that she is concluding it? The performative acts of concluding
and commanding seem to collide, whereas the acts of stating and concluding
seemed to go hand in hand. But we have yet another phrase to try out:
(14.f) Read all of Nabokovs novels. So you can conclude for yourself:
read Pnin!
Though this is perhaps a less common phrase to signal logical arguments, the
new sentence seems to be the most successful so far. But it appears necessary
that the you is the person to whom both commands are addressed. So let
us make the addressees explicit. Of the following sentences
(14.g) John, read all of Nabokovs novels. So John, you can conclude for
yourself: read Pnin!
(14.h) John, read all of Nabokovs novels. So Mary, you can conclude for
yourself: read Pnin!
(14.i) John, read all of Nabokovs novels. So Mary, you can conclude for
yourself: John, read Pnin!
only the first seems somehow acceptable. In (14.h) it appears as if Mary is
asked to read the book, but this can hardly be concluded from a command
not directed at Mary. (14.i) makes it seem as if Mary is asked to give a
command to John (and not just to draw a conclusion). Moreover, if the
addressee is expressly included in the inferred command, then also (14.f),
which seemed so promising at first, looks strange:
(14.j)
John, read all of Nabokovs novels. So you can conclude for yourself:
John, read Pnin!
It seems that in (14.f) and (14.g) the speaker has not just asked the addressee
of the first command to draw a conclusion, but in this process to give
himself the command expressed by the second sentence, i.e. to tell himself to
45
read Pnin. When the addressee is made explicit in the inferred command,
it looks as if the addressee is additionally asked to use his own first name
when telling himself to read Pnin which is a weird thing to ask of anybody.
And this points at another problem of (14.f) and (14.g): if the person who
commands read all of Nabokovs novels (the teacher) and the person who
commands read Pnin (John himself) are not identical, how can the second
imperative be inferred from the first?
By contrast, all of the above phrases can be employed for deontic sentences (non-imperative sentences that do not prescribe, but describe what
ought to be done) without difficulty:
(19)
(19.a)
(19.b)
(19.c)
(19.d)
(19.e)
(19.f)
You ought to read all of Nabokovs novels, therefore you ought to

read Pnin.
John ought to read all of Nabokovs novels, therefore John ought to
read Pnin.
John ought to read all of Nabokovs novels. It follows logically from
this that John ought to read Pnin.
John ought to read all of Nabokovs novels. We can conclude from
this that John ought to read Pnin.
John ought to read all of Nabokovs novels. I conclude from this
that John ought to read Pnin.
John ought to read all of Nabokovs novels. You can conclude for
yourself that John ought to read Pnin.
John ought to read all of Nabokovs novels. Mary, you can conclude
for yourself that John ought to read Pnin.
All these sentences seem grammatical, meaningful and not confusing. We

might even view the inferences they express as sound, but this is not the
question here. Yet as we have seen, all attempts to use the phrases that
link these sentences, normally used to indicate logical arguments in indicative discourses, to link imperatives to indicate imperative inferences, result
in expressions that seem somehow confused and wrong. When used to link
imperatives, they mix up the roles of commanding, command-receiving, and
drawing conclusions. And since the method to use such clauses to distinguish appeals to logic from e.g. motivating uses of therefores and sos, fails
to produce sentences that do not appear strange or confused in the case of
imperatives, perhaps it did so because these adverbs really are not used to
indicate a claimed analyticity when linking imperatives. A motivating use
46
of the adverb so suffices to explain why the sentence (14) seemed meaningful: the teacher, perhaps asked by John whether he also has to read Pnin,
motivates the more specific imperative to read this book by prefixing to it
the general requirement to read all of Nabokovs novels, thus making it clear
that Pnin is in fact one of the books that John has to read. (14.f) appears
comparatively less strange than the other reformulations because to ask John
to give himself the imperative to read Pnin may be a (roundabout) way
to make sure he actually reads it. To understand (18) we do not need to
determine whether the speaker refers to an explicit command to invite ones
friends, or a social custom to do so, because what is in any case implicit in
(18) is an appeal to a preexisting obligation to motivate the agent to do what
the speaker wants her to do. It also explains why (15) seemed so strangely
pointless: the reason for using the less specific imperative to achieve some
top marks is not sufficiently explained by prefixing to it a more specific imperative to achieve top marks in English.47 And so it seems that all of the
imperative arguments (11)(18) are really cases of reason-giving and motivation, and the sos and therefores used in these expressions that like
Poincares donc, or the alsos, daher s and deshalbs of German language,
may be used to connect both indicative and imperative sentences, provide
only reasons, explanations or motives in the case of imperatives, and do not
indicate claims of analyticity.
So I want to dare the hypothesis that there are no examples of imperative
inferences, i.e. logical conclusions in the imperative mood, drawn from at
least one premiss in the imperative mood, to be found in ordinary language
arguments. They only appear in the writings of some philosophers.
1.7
The Way to Go Forward
Other authors have noted before the conspicuous absence of imperative arguments from natural language. Wedeking [221] argued that there are no
cases in which we actually use commands in arguments, and that words like
therefore before imperative sentences are employed not to mark inferences,
but for the purpose of reason-giving, of motivating the subject. Harrison
[90] argued from a point of grammar and semantics that there is no logic
47
Note that the same strangeness does not necessarily arise for deontic logic. The dean
of one faculty may say to another: Our students are obliged to have an A in English,
so yes, ours are like yours obliged to achieve top marks in at least one subject are.
47
of imperatives; the difference between his position and mine (that there are
no imperative inferences in ordinary language, and so a logic of imperatives has no point) seems very subtle.48 Even more conspicuously, there have
been little challenges to these arguments.49 On the contrary, there is a whole
tradition of normological or imperativological skepticism, of authors who
have denied the existence of a logic of norms or imperatives.50
If there are, as a matter of fact, in ordinary language, no argument forms
that resemble imperative inferences, then there also is no place for a formal
theory for such a logic. Presenting formalizations of such a logic would be
writing about what Dubislav [49] called an Unding or chimaera: a non-thing
that exists only as a concept, but no real object falls under the concept.
So did Poincare commit a mistake? Did he confuse an important insight
by Hume [107] on the use of is and ought that facts cannot be used to
48
Harrison writes (pp. 110/111, 124/125): The expression I conclude: Shut the door
does not make sense, nor does the expression I conclude that dont. One can say: So
shut the door and Therefore shut the door and Shut the door because ... but the
function of the words so, therefore and because is not in this context to indicate that
Dont shut the door is a conclusion. They have some other function, which philosophers
have confused with that of indicating that what follows them is a conclusion. (...) The
reason, therefore, why therefore and so can precede post the letter, but I conclude can
not, is that I conclude can precede only propositions (and only they can be conclusions)
and indicate that reasons have been given for the proposition, but therefore and so can
precede either propositions or imperatives. When so and therefore precede imperatives,
however, they are not reasons for the imperative, as the unwary might suppose, but for
the action enjoined, advised, recommended or directed by it. Harrison concludes (p.
81): There is no such thing as imperative logic (...) There are indeed logical relations
between one imperative and another, but this simply supports a logic in which the premises
and conclusions are indicative statements about imperatives (p. 81, my emphasis). The
closeness of Harrisons account to mine was pointed out to me by Lou Goble.
49
Casta
neda [43], replying to Wedeking, grants that differences in the meaning of inferential words in indicative and imperative inferences may exist, but argues that they do
not prohibit a concept of imperative inferences in parallel to indicative ones. This seems to
miss the point, it echoes Dubislavs convention without explaining what such formalisms
are to formalize. I owe the reference to Wedeking and Casta
neda to Peter Vranas.
50
Such authors include G. H. von Wright who writes in [256] p. 109: And now I
too, after a long and winding itinerary have come to the same view: logical relations,
e.g. of contradiction and entailment, cannot exist between (genuine) norms.. Above
we have already noted that Hamblin [74] p.89, Sellars [193] p. 239-240, von Kutschera
[128] and Philipp [168], [169] have expressed scepticism or denied the possibility of a
logic of imperatives altogether. For imperatives also cf. Moritz [153], Williams [238],
Keene [120], Opalek & Wole
nski [164]. The term is coined from Weinbergers [230] term
normological skepticism which denies logical relations not only between imperatives, but
any prescriptive language. The main proponent of normological scepticism is Kelsen [121].
48
argue that they must be so or that other facts should be made similar to them
with a statement about grammar? Curiously, in his essay, Poincare [172]
never claimed to have discovered the logic of imperatives of which he was
heralded as the pioneer. His main argument is that findings of science can
influence moral reasoning. He just presumes that, like scientific arguments
consist of sentences in the indicative mood, moral reasoning is conducted
using sentences in the imperative mood. It is true that facts can influence
the reasoning of agents about their obligations: Hares officer, who upon
being commanded to go to Edinburgh via Coldstream or Berwick finds the
road via Coldstream blocked, acts quite reasonably by concluding that she
now ought to go via Berwick. But this is a reasoning about what obligations
she has, it is a deontic argument, and not a case of inferring imperatives. So
Poincares main argument is correct, but the assumed parallelism between
sentences in indicative and imperative mood, that they can both feature in
logical arguments, does not exist. Our language does not work that way.
There are several ways to go forward from a position of imperativological
skepticism. First, one might continue the logic of imperatives as a logic of
satisfaction. The logic of satisfaction states which imperatives must also be
satisfied if some other imperatives are satisfied, and it may also be used to
state which imperatives will be violated by satisfying other imperatives. We
can use the notion of satisfaction to distinguish imperatives that might be
seen as redundant in a set of imperatives in the sense that these will also be
satisfied if some other, different imperatives are satisfied, or identify subsets
of imperatives that cannot be all satisfied and so conflict. By providing these
concepts, the logic of satisfaction, though it may appear trivial, remains a
meaningful and correct way to talk about imperatives.51
Second, imperatives normally express the wish or desire on the part of
the person or authority using the imperative that what is commanded is
satisfied. But it seems unreasonable to wish for A to be realized, but also
for A to be realized, and in this sense two wishes may exclude another. If
imperatives express wishes of one particular person, we can then point out
to her what wishes may be unreasonable. Likewise it might be desirable to
view the norms of a particular society as if they all were the wishes of one
person, the law giver, and logic may then give advice as to which norms
must be revised so that the system is reasonable. This is the position of G.
51
Cf. C. G. Hempels [91] remark with regard to Rosss Paradox that a logic of satisfaction should not be so easily rejected.
49
H. von Wright in his late work on normative logic, cf. e.g. [255], [256].52
Finally, there is deontic logic. Deontic logic uses the modal expressions it
is obligatory that, it is permitted that, it is prohibited that etc. to describe
what ought to be, is permitted or is prohibited according to given imperatives
or norms. Deontic logic has been disparagingly called a kind of ersatz truth,
that merely mirrors logical relations that already exist between imperatives
or norms, and so we should rather look for this logic than studying a deontic
logic that only reflects it and so must result in a dull isomorphism.53 But it
has been the logic of imperatives that has kept escaping us, while sentences
that use deontic expressions can easily be used to form valid arguments. So
maybe it is not a logic of imperatives that is the proper subject of study
and makes deontic logic just an ersatz theory, but it is the other way round,54
and the idea of a logic of imperatives has been a fata morgana, leading us to
ever more futile attempts to explain inference relations between imperatives,
to find analogues of truth values, or new logics to explain Dubislavs scheme,
whereas any plausibility of this idea was just a reflection of the real, but quite
distinct possibility of a logic about imperatives, namely of deontic logic.55 If
my hypothesis holds, then it is the only logic regarding normative concepts
such as obligation that we should be concerned with. So it is to this logic
that I now turn.
52
The idea that commands can be identified with wishes, which in the above sense
relate to each other, goes back to Bentham [29] pp. 9597. Note that this does not force
one to acknowledge that there is a logic of commands. There is a difference between a
theoretical inconsistency and a practical inconsistency, or: what is inconsistent, and what
is inconsistent to say, are two entirely different things (Harrison [90] p. 95).
53
This is Hares view in [88] p. 325; also cf. Alchourron [2] pp. 264266; Kalinowski
[112] p. 134; Weinberger [230] p. 58, [232]; Wagner & Haag [216] p. 102. The idea that
deontic logic reflects the logical properties of norms is that of von Wright in [246] p. 134.
54
Cf. Alchourr
on & Bulygin [7] p. 463: This logic of norms is, so to say, a reflection
of the logic of normative propositions. It is because we regard as inconsistent a system in
which it is true that Ox p and Ox p, that we say that the norms !p and !p are incompatible.
So it is the logic of norm propositions which yields the foundations for the logic of norms..
55
Thus Hares [87] proposed inference of Take this box to the station from the imperative Take all the boxes to the station and the statement This is one of the boxes
reflects a deontic deliberation of the agent: I have been ordered to take all the boxes to
the station, this is one of the boxes, so I must take this box to the station. Casta
neda [43]
argues in favor of imperative inferences that we can certainly say: John, go home follows
from, and was inferred by Smith from, John, go home and study. While logicians use
sentences of the form: ... follows from ..., with all kinds of expressions filling the gaps,
it seems to me that the underlying natural inference is again deontic: Smith concludes
from Johns having been ordered to go home and study that John ought to go home.
Chapter 2
Deontic Logic
2.1
Beginnings: Leibnizs Discovery
Deontic logic is a branch of modal logic that is concerned with the logic of natural language expressions such as it is obligatory that ..., it is permitted
that ..., it is forbidden that .... It seems that Gottfried Wilhelm Leibniz
(16461716) was the first to explain in detail that these legal modalities
appear to be related in a similar way to each other as the alethic modalities
that describe something as necessary, possible and impossible. Leibniz, in an
unpublished manuscript dating from 1671 for his book Elementa Juris Naturalis, proposes that what is obligatory (debitum) is what is necessary for a
good man to do, what is permitted (licitum) is what is possible for a good
man to do, and what is forbidden (injustum, illicitum) is what is impossible
for a good man to do. Leibniz lets a fourth legal modality, that of indifference, correspond to contingencies as contingent is what is possible but
not necessary, so what is indifferent is what is permitted but not obligatory.
Thus Leibniz seeks to transfer all the complications, transpositions and oppositions of the modalities from Aristotles De Interpretatione to the iuris
modalia ([131] pp. 480-481).1
While it would be unfair to say that this analogy between legal concepts
and their natural or alethic counterparts became forgotten among logical
1
As Knuuttila [124] explained, the idea that a good man cannot violate the law can be
traced through the middle ages back to Aristotles Nicomachean Ethics EN 1100b 33-35,
where he states that no happy man can become miserable, for he will never do the acts
that are hateful and mean.
50
DEONTIC LOGIC
51
philosophers it did not2 the analogy was first formally explored in G. H.

von Wrights 1951 Essay on Modal Logic [243]. The legal, or deontic
modalities are studied here in comparison to other modal categories such as
the alethic, quantificational and epistemic, as illustrated by the table below:
alethic
necessary
possible
contingent
impossible
Modal Categories
epistemic
deontic
verified
obligatory
unfalsified
permitted
undecided
indifferent
falsified
forbidden
existential
universal
existing
partial
empty
So among the above modal categories, the deontic concept of obligatory

corresponds to the alethic concept of necessary or the quantificational concept of universal, while the deontic concept of permitted corresponds to
the alethic concept of possible or the quantificational concept of existing.3
For the alethic modalities, consider the modal square of oppositions from
Aristotles De Interpretatione chapters 12, 13:
It is necessary that A
=It is not possible that not A
=It is impossible that not A
contraries
It is necessary that not A

=It is not possible that A
=It is impossible that A
implies
oppositions
implies
It is possible that A
=It is not necessary that not A
=It is not impossible that A
2
It is possible that not A

subcontraries
=It is not necessary that A

=It is not impossible that not A
Watts [220] mentions the analogy several times, cf. part II ch. 2 sect. iv. As Hruschka
[106] points out, Gottfried Achenwall [1] 12, 44 uses language very similar to Leibniz
when distinguishing between the actio moraliter possibilis, actio moraliter impossibilis and
the actio moraliter necessaria, identifying the necessitas moralis with obligatio. Bentham
[29] p. 97 constructs a deontic square similar to the one below, and shortly before the
advent of axiomatized deontic logic, Grelling [73] suggests the construction of a formal
syntax of ought-sentences after the image of modal logic.
3
Cf. von Wright [242] p. 2. The changes to von Wrights original table are due to
Anderson and Moore [18] who added the quantificational concept of partial (for some
but not for all), and the epistemic concept of unfalsified (verified or undecided).
DEONTIC LOGIC
52
In its corners we find three equivalent statements, expressed in terms of the

three alethic modalities necessary, possible, impossible. Horizontally,
the statements in the left and right upper corner are contraries, meaning
that not one statement from the left and one from the right corners can
be true together. The statements in the left and right lower corner present
subcontraries, meaning that not one statement from the left and one from the
right corners can be false together. Vertically we find the sub-alternatives,
meaning that any statement from the right (left) upper corner implies any
statement from the right (left) lower corner. The statements that stand
diagonally to each other are oppositions, meaning that each statement from
one corner implies that any statement from the diagonally opposite corner is
false. Given the similarity between the two, the alethic and deontic modal
categories, this square can then be translated to a deontic square as follows
(cf. Prior [178] p. 220):
It is obligatory that A
=It is not permitted that not A
=It is forbidden that not A
contraries
It is obligatory that not A

=It is not permitted that A
=It is forbidden that A
implies
oppositions
implies
It is permitted that A
=It is not obligatory that not A
=It is not forbidden that A
subcontraries
It is permitted that not A

=It is not obligatory that A
=It is not forbidden that not A
The expressions it is obligatory that A, it is not permitted that not A

and it is forbidden that not A are taken to be equivalents. Each corner has
three such equivalents, expressed in terms of the three modalities obligatory,
permitted and forbidden. If it is obligatory that A, then A should not be
forbidden at the same time, and so the statements in the right and left upper
corners are contraries to each other. Furthermore, if it is obligatory that A
implies that A is not forbidden, this means that A is permitted, and so any of
the statements in the upper left corner implies all statements in the lower left
corner, and any of the statements in the upper left corner stand in opposition
to any of the statements in the diagonally opposite lower right corner, and
so on for all the corners. Thus the relations from Aristotelian modal logic
are transferred to the deontic modalities, just as Leibniz suggested.
DEONTIC LOGIC
2.2
53
Standard Deontic Logic
Not counting an earlier attempt by Ernst Mally [145],4 the first successful
axiomatization of the relations between deontic expressions was presented
by Georg Henrik von Wright in his 1951 article Deontic Logic [242]. Von
Wright employs three modal operators O, P and F , where OA means
that it is obligatory that A, P A means that it is permitted that A and F A
means that it is forbidden that A. The three operators are cross-defined by
letting O be primitive, and defining P A as OA and F A as OA, so
P A means that it is not obligatory that not A, and F A means that it is
obligatory that not A.5 His axiomatization, called the classical system of
deontic logic [251], can be characterized by the following axioms and rules:
(Ext) If A B is a theorem, then OA OB is a theorem.
(M)
O(A B) (OA OB)
(C)
(OA OB) O(A B)
(D)
OA P A
Here, the principle of extensionality (Ext) expresses that if A and B are
equivalent, then so are the sentences OA and OB. The principle of monotony
(M) expresses that if it is obligatory that both A and B, then it is also
obligatory that A and obligatory that B, and the principle of cumulativity
(C) expresses vice versa that if it is obligatory that A and obligatory that
B, then it is also obligatory that both A and B. Together, (M) and (C) are
equivalent to von Wrights principle of distribution (P A P B) P (A B),
meaning that if a disjunction of acts is permitted, then one of the disjuncts
is permitted, and vice versa. (D) is the characteristic deontic axiom that
expresses that if it is obligatory that A then it is permitted that A; it is the
equivalent of von Wrights principle of permission P A P A that expresses
that any act is either permitted or its negation permitted, and it is also due
to the definition of P A as OA equivalent to (OA OA). It is also
crucial for obtaining the modal relations expressed by the deontic square:6
4
Mallys axiomatization culminated in the result that it is obligatory that A implies
that A is the case, and vice versa, or, in Mallys words, the Aquivalenz
von tats
achlich
Gefordertem und Tats
achlichem (the equivalence between the factually obligatory and the
facts) cf. [145] p. 269. Mallys error lies in his missing distinction between logical and
material implication, cf. Wole
nski [241], Goble & Lokhorst [70].
5
Von Wright originally used P as primitive and equivalently defined OA as P A,
and F A as P A, but the use of O as primitive became standard.
6
Since (D) is missing from Oskar Beckers independent publication [28], von Wright is
DEONTIC LOGIC
54
OA,
P A, F A
(D)
OA,
P A, F A
(D)
(D)
(D)
P A,
OA, F A
P A,
OA, F A
If one adds to von Wrights classical system the additional axiom (N)7
(N)
O(A A)
one obtains the equivalent of the so called standard deontic logic system
SDL, which is more rigorously formalized as follows:8
Definition 2.1 (The Language of Deontic Logic) The alphabet of the
language of deontic logic LDL is like that of propositional logic LPL except
that it additionally contains the deontic operator symbol O. LDL is defined
by the same rules as the set LPL plus the following: if A LDL , then also
OA LDL . P A abbreviates OA, and F A abbreviates OA.
Definition 2.2 (The System SDL) SDL is the smallest set that contains
all LDL -instances of tautologies and the axiom schemes (M), (C), (D) and
(N), and is closed under the rules of (Ext) and modus ponens.
rightly acclaimed to be the founder of modern deontic logic.
7
(N) makes all tautologies obligatory. While von Wright [242] p. 11 rejected it, its
philosophical import seems rather small: if something is obligatory, e.g. if OB is true,
then the truth of O(A A) can be derived by means of (C) and (Ext).
8
The deontic systems studied here will be based on propositional logic PL. The alphabet of its language LPL consists of a set of proposition letters Prop = {p1 , p2 , ...}, Boolean
connectives , , , and brackets (, ). LPL is then the smallest set such that
(a) any proposition letter p1 , p2 , ... is in the set, (b) if A, B are in the set, then so are
A, (A B), (A B), (A B). Semantically, valuation functions v : Prop {1, 0} define
the truth of a sentence A LPL (written v |=PL A) as follows: (a) v |=PL pi if and only
if (iff) v(pi ) = 1, (b) v |=PL A iff it is not the case that v |=PL A, (c) v |=PL A B iff
v |=PL A and v |=PL B, (d ) v |=PL A B iff v |=PL A or v |=PL B, (e) v |=PL A B iff
v |=PL A or v |=PL B. If all valuations that make true all sentences of a set of sentences
also make true A, then implies A (written A). If A is true under all valuations v,
A is called a tautology, if it is false under all valuations, it is called a contradiction. From
the set PL of all tautologies, the notions of provability, consistency and derivability (I
write PL A) are defined as usual. denotes an arbitrary tautology, and an arbitrary
contradiction.
DEONTIC LOGIC
55
Disregarding varying axiomatizations, SDL was first introduced by Prior

[178] ch.1 6, in the nomenclatura of modal logic it is the normal modal
system KD, i.e. the modal system K to which (D) is added. It differs from
the alethic system KT only in so far as that (D) replaces the axiom
(T)
OA A
typical for such systems: for alethic modalities, (T) states that if it is necessary that A then A (is true), which should not hold in the deontic interpretation if it is obligatory that A then A (is true). It should be noted that
the move from von Wrights first proposal to standard deontic logic SDL also
brought about a change in the meaning of the expression A in the scope of the
deontic operators. Von Wright [242], [243], similar to Becker [28], intended
A to be the name of an act: P A is interpreted as the proposition that the act
named by A is permitted. By contrast, Prior lets A designate a proposition
that asserts that an act of the sort A is done. Deontic logic literature has
favored Priors interpretation, and OA is standardly read as it is obligatory
that the situation described by the descriptive sentence A is realized.9
2.3
Andersons Reduction
Independently from each other, Stig Kanger [119] and Alan Ross Anderson
[12], [14] showed in 1956/7 that the analogy between the alethic and deontic
modal concepts may not only be used to transfer the rules of the former to
the latter, but to reduce deontic logic to alethic modal logic altogether. To
this effect, a prohairetic constant Q (Kanger, Q denoting what morality
proscribes) or S (Anderson, S meaning a sanction or a bad state of affairs)
is added to any modal system from K upwards, and the deontic O-operator
is then defined in the following way, respectively:
(Def-OQ)
OA =def (Q A)
(Def-OS)
OA =def (A S)
So in Kangers interpretation, that A is obligatory means that what morality
prescribes necessitates A, and in Andersons interpretation it means that not
doing A necessitates the sanction (or the bad thing).10 If the alethic modal
9
Cf. Allen [11], Fllesdal & Hilpinen [53] p. 14.

By closer inspection, Kangers variant of this reduction of the deontic modalities to
alethic modal logic may well be attributed to Leibniz [131], when he wrote that what is
obligatory is what is necessarily done by a good man (who by nature does what morality
10
DEONTIC LOGIC
56
system is K, and the following respective axioms are added:11

(DQ)
Q
(DS)
S
then the deontic fragment of this system coincides with SDL. More formally:
Definition 2.3 (The Language of the Modal System KQ + [KS + ])
The alphabet of the language of the language LKQ+ [LKS + ] is like that of
propositional logic LPL except that it additionally contains the alethic operator symbol and the prohairetic constant Q [S]. LKQ+ [LKS + ] is defined
by the same rules as the set LPL plus the following two: the constant Q [S]
is in the set; if A is in the set, then so is A. A abbreviates A.
Definition 2.4 (The Modal System KQ + [KS + ])
The axiomatic system KQ+ [KS + ] is the smallest set of that contains all
LKQ+ [LKS + ]-instances of tautologies and the axiom-scheme
(K)
(A B) (A B)
as well as (DQ) [(DS)] and is closed under the rules of modus ponens and
(RN) If A is in the set, then so is A.
Theorem (Translation Theorem for SDL [
Aqvist [22] pp. 683-688]).
Let : LDL LKQ+ [LKS + ] be a translation from the language of deontic
logic to the language of alethic modal logic with the constant Q [S] such that
(a) (pi ) = pi ,
(c) (A B) = (A) (B),
(e) (A B) = (A) (B),
(b) (A) = (A),

(d) (A B) = (A) (B),
(f) (OA) = (Q (A)) [((A) S)].
Let the deontic fragment DF (KQ+ ) [DF (KS + )] contain all A KQ+ [KS + ]
with A = (B) for some B LDL . Then DF (KQ+ )[DF (KS + )] = (SDL).
The reduction of deontic to alethic modal logic is interesting in its own
right, it can even point to an history of its own. Kangers variant, that
something is obligatory if it is necessary for an ideal state, stands right in
prescribes). Knuuttila [124] p. 236 points out that a reverse reduction of the alethic
modalities to deontic modalities can be found in the Dialectica of Peter Abelard (1079
1142) who identified necessity with what nature demands, possibility with what nature
allows, and impossibility with what nature forbids.
11
Anderson showed that even this addition can be avoided if one chooses the modal
system KT and replaces S by the formula B B. For Kangers constant Q, the formula
B B constitutes a similar replacement.
DEONTIC LOGIC
57
the footsteps of Leibniz [131], according to whom what is obligatory is what

is necessarily done by a good man;12 Husserl, Menger and Bohnert have also
presented related ideas.13 Andersons method of defining the obligatory via
a sanction or bad state has been called the most natural interpretation of
deontic sentences (Hilpinen [92]) and was later taken up by dynamic deontic
logic.14 Nevertheless, Andersons and Kangers reductions can also be seen
as a first shift from an informal understanding of deontic expressions as
belonging to a discourse about moral or legal norms to an interpretation via
prohairetic terms like the ideal or bad.
2.4
Possible Worlds Semantics
Axiom systems characterize the logic in question by means of sets of theorems. E.g. we know that
OA P B P (A B),
meaning that if something is obligatory and something else is permitted,
then we may realize both conjunctively, is a theorem of standard deontic
logic SDL, because its theoremhood can be established by means of this
systems axioms and the rules, namely (K) and modus ponens. If A is a
theorem of e.g. SDL, we write SDL A and may call A provable in SDL. As
usual, from this notion a variety of others derive: a set of sentences LDL
is called SDL-inconsistent if and only if there are A1, ..., An in , n 1, such
that SDL (A1 ... An ), and is SDL-consistent otherwise. A sentence
12
Very similarly, Andersons sanction has been interpreted to represent simply failure
to be perfect (Prior [179] p. 146.)
13
Cf. Husserl ([108] p. 41/42, my translation): The following forms are equivalent:
A ought not to be B, and an A which is not B, is generally a bad A, or only an A,
which is not B, is a good A , an A may be B = an A, which is B, needs, for this
reason, not to be a bad A ; Menger [150]: The statement I command p (...) may
be interpreted as unless p, something unpleasant will happen ; Bohnert [33] reduces
commands to motivational declarative sentences like P B(x), where B is the behavior
desired of x and P a penalty, or E B(x), where E is some desired end. Both, Menger
and Bohnert, are aware that their definitions, phrased in terms of ordinary disjunction
and material implication, have unwanted consequences (like that all factual truths are
commanded), to avoid them Menger applies a three-valued logic, Bohnert demands that
B(x) should be causally necessary for the end E.
14
By singling out an undesirable state of affairs, an action may be defined as forbidden
if it necessarily results in this state, as obligatory if the action not- is forbidden, and as
permitted if is not forbidden (cf. Meyer [152]).
DEONTIC LOGIC
58
A LDL is SDL-derivable from a set LDL (we write SDL A) iff

{A} is SDL-inconsistent, etc.
However, axiom systems, and particularly those of deontic logic, have
often been devised ad hoc, philosophers taking their cue from intuitions as
they come along. Was it a good idea to add (N) to von Wrights system?
If one thinks that the object of human wishes and commands are neither
necessities nor impossibilities (cf. Menger [150] p. 97) should one not retract
(N) and instead add O(A A) and O(A A) to the set of axioms,
thus making sure that neither tautologies nor contradictions can ever be
obligatory? The first addition would yield OB for any B, and so incidentally
free us of all obligations: responsible is the seemingly innocuous (M), so we
would have to retract (M), but what replace it with? By contrast, the second
addition would be harmless, in fact it is a theorem of SDL.
A much better way to go forward in developing a logical system is by
means of a logical semantics. A logical semantics describes mathematical
structures or models that enable us to evaluate each sentence of the formal language as true or false with regard to the model. The sentences of
the language that are valid according to a semantics (true in all its models)
may correspond to all, and only, the theorems of an axiomatically described
logical system; this is then called sound, and complete, with respect to the semantics.15 The logical-semantical interpretation of the logical symbols (connectives, operators, predicates etc.) of the formal language may then be
compared to intuitions about the use of the ordinary language expressions
they symbolize, where the aim is to let the properties of the mathematical models mirror the aspects and regularities of ordinary language usage as
closely as possible, while on the other hand the logical semantics might help
us clarify our intuitions about that usage.16
Since 1957, deontic logic has been interpreted in terms of a possible worlds
semantics (cf. Hintikka [97], Kanger [119]).17 Its models M are triples
W, R, V , where W is a nonempty set of possible worlds, R W W is
a binary relation between worlds, where for any w, v W , wRv means that
v is a deontic alternative to w, and v : P rop W {1, 0} is a valuation
function that associates a truth value with each proposition letter in any
15
Axiomatic systems may thus correspond to a range of different semantics.

For the uses of a logical semantics and its limits cf. Stekeler-Weithofer [204] p. 141.
17
With Kripke [127], interpretations in terms of possible worlds became standard for
systems of modal logic, and the account I present here is Kripke-semantics as reapplied to
deontic logic by e.g. Hanson [82], also cf.
Aqvist [22] pp. 665-669.
16
DEONTIC LOGIC
59
world w W . Truth of LDL -sentences in any model M and world w W

is then defined recursively:
(a) M , w |= pi
iff
v(pi , w) = 1,
(b) M , w |= A
iff
M , w A,
(c) M , w |= A B
iff
M , w |= A and M , w |= B,
(d) M , w |= A B
iff
M , w |= A or M , w |= B,
(e) M , w |= A B iff
M , w |= A or M , w |= B, and
(f) M , w |= OA
iff
v W : if wRv then M , v |= A.
So the truth of sentences of propositional logic at a world w is defined as
usual, while a sentence of the form OA is true at a world w if and only if A
is true at all deontic alternatives to w. The deontic alternatives may also be
viewed as the ideal worlds that are accessible from w, so what is obligatory
is what is the case in all ideal worlds accessible from the actual world w.
(The semantics are illustrated below by a model M = W, R, V such that
M , w1 |= O(p1 p2 ); the deontic alternatives to w1 are here the worlds w2
and w3 .) A LDL -sentence A is called valid iff it is true at any world w in
Figure 2.1: Possible worlds semantics for monadic deontic logic

any model M (we write |= A). If we add the restriction that for all models
M , R is serial, i.e. for any w W there exists at least one v W that is a
deontic alternative to w, then the set of valid LDL -sentences coincides with
SDL, i.e. SDL is sound and complete with respect to this semantics.
Turning back to the axiomatization of deontic logic, we can now see how
the axiom (N) can be defended from the viewpoint of possible worlds semantics: something is obligatory if it is true in all accessible ideal worlds. But
DEONTIC LOGIC
60
at any world, for all models M , A A, being a tautology, is true, and so

O(A A) is true. It is claimed that this shows the role of possible worlds
semantics in the clarification of our intuitions concerning the behavior of deontic operators, whereas a purely axiomatic approach is not conclusive (cf.
Wole
nski [240]).
2.5
Dyadic Deontic Logic
Early in the development of deontic logic, doubts arose as to the adequacy of

standard deontic logic SDL for the formalization of conditional obligations,
and in particular that of contrary-to-duty obligations. It is possible that A
is forbidden, and additionally that if A contrary-to-duty obtains, then it
is obligatory that some act of reparation B is done. But how to express this
second consequence relation? It cannot be formalized as O(A B), for as
Prior [177] pointed out, we have
(PP)
F A O(A B),
as a theorem of deontic logic. Possible worlds semantics reveals it as an

offspring of one of the paradoxes of material implication A (A B):
F A means that at all accessible ideal worlds A is true, and so also A
B is true there, hence O(A B) is also true, for any B. Prior thought
this theorem, later called Priors Paradox, troublesome, since it can be
understood as stating that the doing of what is forbidden commits us to the
doing of anything whatsoever, e.g. the forbidden act of stealing commits us
to committing adultery.
One may think that the situation can easily be remedied by using the
formalization A OB, instead of O(A B), for conditional obligations, or
at least by doing so when contrary-to-duty obligations are involved. However,
Chisholm [46] showed that whatever combination of these formalizations is
chosen for the following sentences
(1) It ought to be that a certain man go to the assistance of his neighbors.
(2) It ought to be that if he does go he tell them he is coming.
(3) If he does not go, then he ought not to tell them he is coming.
(4) He does not go.
then any of the four sets {OA, O(A B), A OB, A}, {OA, O(A
B), O(A B), A}, {OA, A OB, O(A B), A}, and {OA, A
OB, A OB, A} will either be inconsistent in SDL, or one of its four
DEONTIC LOGIC
61
sentences will derive one of the others. But intuitively, the normative situation is not inconsistent and the sentences are independent from each other,
so their formalization should not make them otherwise. Thus this example
was termed Chisholms Paradox.
In reaction to Priors paradox, G. H. von Wright in 1956 [244] introduced a
stroke / in the scope of the operator O, where the dyadic formula O(A/B)
means that A is obligatory under the condition B. The old, monadic Ooperator is then defined as OA = O(A/ ), i.e. as an obligation that holds
in arbitrary, or default conditions. More formally, the new language is as
follows:18
Definition 2.5 (The Language of Dyadic Deontic Logic)
The alphabet of the language of dyadic deontic logic LDDL is like that of LDL ,
except that it additionally contains the stroke /. LDDL is then the smallest
set such that
(a) if A, B are in LP L , then O(A/B) is in the set,
(b) if A, B are in the set, then so are A, (A B), (A B) and (A B).
Again, P (A/B) abbreviates O(A/B) and F (A/B) abbreviates O(A/B).
OA, P A and F A abbreviate O(A/ ), P (A/ ) and F (A/ ) respectively.
With the introduction of the stroke, soon numerous new axiomatic systems of dyadic deontic deontic logic emerged. Systematically, the following
dyadic analogues of the axiom schemes of SDL appear to be good candidates
for an axiom system of dyadic deontic logic:
(Ext)
If A B is a tautology, then O(A/C) O(B/C) is a theorem.
(ExtC)
If C D is a tautology, then O(A/C) O(A/D) is a theorem.
(DM)
O(A B/C) (O(A/C) O(B/C))
(DC)
(O(A/C) O(B/C)) O(A B/C)
(DD)
O(A/C) P (A/C)
(DN)
O(A A/C)
These axioms allow reasoning with antecedents like C or D only in the very
weak form of (ExtC). But sometimes we want to use information from the antecedent for our reasoning with consequents. Consider the following example
by Powers [174]:
18
Note that the language LDDL is more restricted than LDL and e.g. does not allow
iterated or nested O-operators as in OOA or O(OA A), or mixed propositional and
deontic formulas as in B OA. I simplify matters, as an extended language would e.g.
require operators and axioms for universal necessity in order to make the system complete
with respect to the kind of semantics given below.
DEONTIC LOGIC
62
John must either not impregnate Suzy Mae or marry her. Suzy
Mae is pregnant by John. So John must marry Suzy Mae.
A rule that permits such inferences is achieved by strengthening (Ext) to a
principle of circumstantial extensionality :
(CExt)
If C (A B) is a tautology, then O(A/C) O(B/C) is a
theorem.
Using this rule, we obtain the theorem O(A B/A) O(B/A), i.e. in
a situation where A is true (Suzy is pregnant by John) the obligation to
realize A B (Either Suzy is not pregnant by John or John marries Suzy)
is equivalent to an obligation to realize B (John marries Suzy). If (Ext) is
replaced by (CExt), it is clear that either (D) or (N) must be weakened to:
(DD-R) Unless C is a contradiction, O(A/C) P (A/C) is a theorem.
(DN-R) Unless C is a contradiction, O( /C) is a theorem.
Otherwise for impossible conditions the theorem O(/) is derivable from
(DN) and (CExt), and also the contradictory theorem O(/) from (DD),
(DC) and (CExt), i.e. the set of axioms would be inconsistent.
Two other early dyadic deontic axioms or theorems were the principles of
conditionality and of rational monotony:19
(Cond)
O(A/C D) O(D A/C)
(RMon) P (D/C) (O(A/C) O(A/C D))
(Cond) expresses that if A is obligatory in the situation C-and-D, then it
is obligatory in the situation C that if one realizes D then one also realizes
A. (RMon) expresses that if one realizes something permitted, then ones
obligations do not change. Together, these axioms, with the choice between
unrestricted (DD) and restricted (DN-R), or restricted (DD-R) and unrestricted (DN), describe the two most successful systems in dyadic deontic
logic:20
19
(Cond) is Reschers [182] axiom A4, it also derives from the left-to-right direction of
von Wrights [244] axiom A2: P (A B/C) P (A/C) P (B/C A). (RMon) is derivable
from its right-to-left direction and is also Reschers [182] theorem T4.16. Given (CExt),
the conjunction of (Cond) and (RMon) is equivalent to A2.
20
As for the not so successful approaches, note first that these should not be theorems:
Up
O(A/C) O(A/C D)
Down
O(A/C) O(A/C D)
With Up or Down, any contrary-to-duty obligation, contrary to intuitions, creates a conflict of duties. E.g. consider a Chisholmian set {OA, O(B/A), O(B/A)}. With Up we
derive OB and OB from O(B/A) and O(B/A), so there is a conflict. From OA and
DEONTIC LOGIC
63
Definition 2.6 (Systems DDLH , DDLL )

DDLH and DDLL are the smallest sets that contain all LPL -instances of
tautologies and the axiom schemes (DM), (DC), (Cond), (RMon), where
DDLH additionally contains all LPL -instances of (DD-R) and (DN), and
DDLL additionally contains all LPL -instances of (DD) and (DN-R). Both
sets are closed under (CExt), (ExtC) and modus ponens.
DDLH is Hanssons [84] system DSDL3 as axiomatized by Spohn [203], and,
neglecting differences of language, DDLL is the synoptical Lewis-type system
DFL of
Aqvist in [23]. The following are typical theorems of the two systems:
(Ref)
(Pres)
(P -Pres)
(CCMon)
(Cut)
(Or)
(DR)
(FH)
(Trans)
(P -Loop)
(Loop)
O(A/A)
[ only DDLH ]
O(/C) (O(A/D) O(A C/D))
P (/C) (O(A/D) O(A C/D))
O(A D/C) O(A/C D)
[ = (P -Cond) ]
O(D/C) (O(A/C D) O(A/C))
[ = (P -RMon) ]
O(A/C) O(A/D) O(A/C D)
O(A/C D) (O(A/C) O(A/D))
P (C/D) (O(A/C D) O(A/C))
P (A/A B) P (B/B C) P (A/A C)
P (A2 /A1 ) ... P (An /An1 ) P (A1 /An ) P (An /A1 )
O(A2 /A1 ) ... O(An /An1 ) O(A1 /An ) O(An /A1 )
The names are from analogues in the study of nonmonotonic reasoning

(cf. Makinson [140]), namely reflexivity, preservation, conjunctive cautious
monotony, disjunctive reasoning and transitivity. (Ref) is typical of Hanssontype logics. (CCMon) is Reschers [182] th. 4.4, (Or) is the right-to-left
version of von Wrights [247] (B3), (DR) is Hanssons [84] th. 13, and (FH)
is Fllesdal & Hilpinens th. 77. (Trans) is transitivity of weak preference
defined as A
B =def P (A/A B) in Lewis [136] p. 54. (P-Loop) was
introduced by Spohn [203] to define the relevant equivalence classes in his
completeness proof of Hanssons DSDL3, and its O-form was rediscovered by
Kraus, Lehmann & Magidor [126].
The reason that Hanssons system DDLH and Lewis-type system like
O(B/A) we derive OB with (Cond), (DC) and (DM), and with Down we then obtain
O(B/A), which conflicts with O(B/A). Nevertheless, axiom systems by von Wright
[247], [248] and A. R. Anderson [15] derive Up, and Reschers system [182], that also
provides (Cond), (DC) and (DM), derives Down.
DEONTIC LOGIC
64
DDLL were more successful than others was not so much because they included some widely recognized theorems, but because they are also sound
and (weakly) complete with respect to a preference semantics that models
contrary-to-duty situations in an intuitively plausible way:21 Let (at least
as good as) be a reflexive, transitive and connected preference relation on
B (the set of all Boolean valuations v : P rop {1, 0}) that additionally
satisfies the Limit Assumption:
(LA)
For all A LPL , if A = then best( A ) =
Here A = {v B | v |= A} is the subset of Boolean valuations or worlds
that make A true (the A-worlds), and for any X B, best(X) = {v
X | v X : v v }) is the set of best worlds in X. The Limit Assumption tells us that for all PL-sentences that are true at some world, there are
worlds at which they are true that are at least as good as any other world
at which the sentence might be true. Finally, alongside the usual truth definitions for Boolean operators the following truth definitions for the deontic
O-operator are used, for DDLH and DDLL respectively:
DDLH :
|= O(A/C)
iff
best( C ) A
L
DDL :
|= O(A/C)
iff
v C A ) : v C A : v
v
So for DDLH , O(A/C) is defined true iff the best C-worlds are A-worlds, and
for DDLH , O(A/C) is defined true iff there is a (C A)-world such that no
(C A)-world is at least as good.
Just as the introduction of the stroke / into the scopes of the deontic
operators was motivated by the search for a representation of contrary-toduty imperatives, so was the development of preference semantics for dyadic
deontic logic. This was expressed particularly clearly by Bengt Hansson:22
The problem (...) is what happens if somebody nevertheless
performs a forbidden act. Ideal worlds are excluded. But it may
be the case that among the still achievable words some are better
than others. There should then be an obligation to make the best
out of the sad circumstances.[84]
21
A proof of (weak) completeness for DDLH was first presented by Spohn [203], it can
be straightforwardly adapted to DDLL . As the semantical details vary between Hansson
and Lewis, I use a description that fits both systems.
22
Also cf. Mott [154]: The lesson to be learned from Chisholms paradox has been the
eminently convincing, indeed obvious, one: that what we ought to do is not determined
by what is the case in some perfect world, but by what is the case in the best world we can
get to from this world. What we ought to do depends upon how we are circumstanced.
DEONTIC LOGIC
65
Hanssons and Lewis semantics capture this idea in the form of a (possibly
infinitely descending) chain of sets of ideal, subideal etc. worlds: the so-called
system of spheres (cf. fig. 1). Ideal worlds are those that are best overall:
those in best(B). The second-best worlds are those that are best in the set
of all worlds without the best worlds, i.e. those in best(B\best(B)), and so
on. For any possible situation C, the best C-worlds are those that are in the
smallest C-permitting sphere, i.e. the smallest sphere that has a non-empty
intersection with the set of C-worlds C . The Limit Assumption guarantees
that for all such C there is such a smallest C-permitting sphere. As to how
Figure 2.2: System of spheres for dyadic deontic logic

contrary-to-duty situations are modeled, consider again Chisholms Paradox.
Things are best when the man goes to his neighbors assistance and tells that
he is coming, so A B are the best worlds. It seems second-best if the
man doesnt call but still goes, and so the best (A B)-worlds are the
worlds in A B . Then come the worlds in A B where the man
does not go but also does not tell that he is coming, and the worst are
the worlds in A C where he tells his neighbors that he would go, but
doesnt. This exhausts the logical possibilities. As may be checked, the thus
constructed model makes true all of {OA, O(B/A), O(B/A)}. If, upon
learning that A is true, i.e. the man does not go, we restrict the model23
to the worlds in A , then only the obligation OC not to call remains,
23
Though such a restriction is not an option for the semantics as outlined, it is possible
in Lewis [137] semantics where the set of worlds may be a subset of B, e.g. one where it
is necessary that A. Then we have (O(B/A) A) OB.
DEONTIC LOGIC
66
which is how preference semantics solves Chisholms Paradox. Other socalled paradoxes receive similar solutions in dyadic deontic logic, like the
Paradox of the Good Samaritan [179] and that of the Gentle Murderer
[56]: in monadic deontic logic it follows from the fact that it is forbidden
to assault a man that it is also forbidden to help the assaulted man, and
an obligation not to commit murder, but otherwise murder gently, is not
expressible for the reasons given for Priors Paradox. In dyadic deontic logic
we can state without contradiction that assault and murder are forbidden,
but that in the bad situation that either happens secondary obligations to
help or do it gently obtain.
2.6
Doubts and Questions
Since its emergence as a subject of philosophical logic in 1951, deontic logic

has undergone multiple periods of refinement. While it was very much an
axiomatic theory in the 1950s, study in the 1960s turned to an application
of possible worlds semantics, and the study of the late 1960s and early 1970s
is characterized by the use of preference semantics to interpret systems of
dyadic deontic logic. The late 1970s and the 1980s have reinterpreted the
possible worlds models for monadic and dyadic deontic logic in terms of
temporal tree structures and added temporal operators and/or temporally
indexed O-operators (cf. e.g. van Eck [51],
Aqvist [20]). The 1990s have in
turn used these tree systems to interpret action-based deontic operators and
combined the theory of deontic logic with the notion of STIT (seeing-to-itthat, cf. Horty [103]). At the core of all these later developments still lies
the the theory of deontic logic as sketched in the previous section. But there
have also been doubts and questions as to the adequacy of this theory.
2.6.1
Doubts About Possible Worlds
The axiomatic basis of deontic logic has always been feeble at best. It has
been claimed that deontic logic has not a single axiom that is not incompatible with some reasonable moral position and so they all violate the principle
of logical neutrality (cf. Sayre-McCord [191]). Worse still, the main semantical tool of traditional deontic logic, possible worlds semantics, is stretched
in its task to clarify our intuitions about possible theorems of deontic logic,
as it is itself the target of philosophical criticism. Consider the definition
DEONTIC LOGIC
67
OA is true at the actual world iff A is true at every deontic

alternative.
First, to establish the truth of OA, what does it mean that there exists an
alternative world in which A is true? Are possible worlds real or imagined
entities? In an ontological sense, the only possible world is our own, the
actual world, with its future developments hidden in it like the ear in a grain
of corn (Kalinowski [115]). So second, given the past history of our world, it
cannot be deontically perfect, and so we cannot conceive of a Trans World
Airline that would take us to a deontically perfect world (Geach [62]). And
third, even if we restrict our view to possible developments from now on, as
preference semantics for dyadic deontic logic does, given the provenly fallible
nature of human beings, these developments will never be deontically perfect:
each development is one where someone sins and so someones sinning is
obligatory, which is hardly intuitive (cf. Geach [62] p. 10). Fourth, even if
all ideal alternatives to our world are e.g. such that nobody smokes, it may
be that our subideal legislators, as a matter of fact, do not forbid smoking:
what justification have we then to call smoking forbidden? maybe it ought to
be forbidden, but this is not the same (cf. Wansing [219]).24 Fifth, consider
the case of the habitual milk-drinker (Solt [199]): suppose that you have the
habit of drinking a glass of milk every day, a habit so strong that at every
deontic alternative each of your alteregos also drinks a cup of milk every
day: does that suffice for the statement it is obligatory that you drink a
cup of milk every day to be true? Solt argues that in order to distinguish
between norms and laws of nature, habit or logic, the biconditional of the
truth definition should be reduced to the part if OA is true at the actual
world then it is true at every deontic alternative. But sixth, as Kalinowski
[115] points out, this direction of the iff-clause commits us to the view that
norms only prescribe what is good, for otherwise what they demand could
not be true in an ideal world, a view that is hardly justified.25 Kalinowskis
24
Wansing [219] therefore calls possible worlds semantics inappropriate as an informal

semantical foundation of deontic logic.
25
Goble [65] pp. 193/4 points out a related problem: suppose Smith robs a bank, ought
he to be punished? what if, unpunished, he would go on to perform wonderful works of
philanthropy with his loot, making the world a much better place? Goble argues that
we ought not to consider best logically possible worlds, nor even only actually available
alternatives, but only those worlds that result through minimal change from the actual
world. I think things are much simpler if we have a way to represent norms in deontic logic:
Smith ought to be punished because there is a norm that says he ought to be punished.
DEONTIC LOGIC
68
subsequent advice not to evaluate the truth of deontic propositions in terms

of the good, better or ideal, but to return to reasoning with existing norms,
echoes a description of ordinary deontic reasoning by Ross [189] p. 17:
When a plain man fulfils a promise because he thinks he ought
to do so, it seems clear that he does so with no thought of its total
consequences, still less with any opinion that these are likely to
be the best possible. He thinks in fact much more of the past
than of the future. What makes him think it right to act in a
certain way is the fact that he has promised to do so that and,
usually, nothing more. That his act will produce the best possible
consequences is not his reason for calling it right.
Some of the criticism of deontic logics possible worlds semantics seems
too harsh e.g. when we aim to prove mathematical results about an axiomatically specified logic, its meaning is just that of a formal tool. Still, it
seems that possible worlds semantics has, at least for a number of authors,
lost its power of also mirroring and guiding our intuitions. And Rosss view
that ordinary deontic reasoning starts with norms or laws, and not with prohairetic notions like ideal or best, that these notions are secondary at best,
is even apparent in an author who expressed that
deontic logic, to put it in a nutshell, is the study of logical relations in deontically perfect worlds
insofar as the same author who in [252] turned away from possible worlds
semantics altogether, it is G. H. von Wright [254] explains that ideal worlds
are what is envisaged by a system of norms, and that any normative code
may be said to contain an implicit description of a state of affairs, viz. the
ideal state envisaged by the norm (ibid.). In particular, von Wright refers
to the idea of a Kantian Kingdom of Ends, used by Hintikka [95] p. 73 to
motivate the idea of a deontic alternative:
The Kingdom of Ends is the world such as it would if all and
sundry rational beings always honored all their obligations (duties). In this respect, a Kantian Kingdom of Ends is like a
deontic alternative to the actual world.
Not because it makes the world a better place.
DEONTIC LOGIC
69
But this seems like a bad defense of possible worlds semantics. If the motivation for deontic possible worlds semantics starts with given norms or duties,
and defines the ideal worlds as the ones where all is as is envisaged by these
norms, then why not appeal to these reasons directly and try, as Kalinowski
advises, to reconstruct deontic logic on the basis of norms?
2.6.2
The Question of Conflicts and Dilemmas
A second, more practical problem is that existing norms may conflict, or, in
unforeseen circumstances, result in dilemmas. For moral norms the question
of the existence of true conflicts has been the subject of intense philosophical
debate since E. J. Lemmon [132] started it; in positive law they are, according to Alchourron [2] and Alchourron & Bulygin [6], frequently found and
no rarity, while computer scientists seem to have come to view conflicts
between specifications as inevitable. But standard deontic logic excludes the
possibility that OA and OA are both true by its axiom (D), which is equivalent to (OA OA). Possible worlds semantics makes it impossible that
all deontic alternatives make true A and also A unless there are no deontic
alternatives, which is excluded by the seriality restraint that corresponds to
axiom (D). But to drop seriality seems not a proper way to model conflicts:
in a conflict, there are not too little, but rather too many ideal ways to act.
Similarly, for the axiomatic approach, simply removing (D) from the set of
axioms will not be sufficient: unless one accepts that a case of conflict makes
anything obligatory, one would also have to remove or replace (M) or (C),
which make it possible to derive from OA and OA that OB for any B (so
called deontic explosion).
For dyadic deontic logic there is not only the problem of conflicts, but also
that of dilemmas: for any two apparently harmless statements of obligation
OA and OB, the situation (AB) creates a dilemma, i.e. a situation where
something that is obligatory can no longer be realized (unless there is the
rather artificial assumption that either A implies B or B implies A). Just like
standard deontic logic cannot accommodate conflicts, the traditional dyadic
deontic systems DDLH or DDLL do not accept dilemmas: given that (AB)
is consistent, O(A/(A B)) derives O(A/(A B)) with (DD) or (DDR), whereas with (CExt) we derive O(A/(A B)) from O(B/(A B)).
In dyadic deontic semantics, to model any two sentences OA and OB, the
system of spheres must specify which state of affairs A or B is to be preferred
in the section ||(A B)|| of the spheres, though nothing might be specified
DEONTIC LOGIC
70
about such a situation by the sentences that one aims to model consider
the above modeling of Chisholms dilemma: we assumed that not going and
not telling is better than not going and telling, though the original sentences
were silent about this question.
An easy way out of the problem of conflicts and dilemmas is to say that
the task of deontic logic is to model the moral or legal situation after it has
been resolved by appropriate mechanisms of conflict resolution. But we then
admit that deontic logic is neither able to describe conflicts nor to analyze
the resolution mechanism:
It is of no avail to say that logic pictures the situation that arises
after the normative conflict has been resolved. For rather that
what is being resolved, the conflict of norms, is the interesting
state of affairs that must be logically characterized. (Alchourron
& Bulygin [6] p. 25)
Another way out is to give up the idea that the expressions of deontic logic
somehow relate to real sets of norms and rules, describe what ought to be
the case according to some system of norms (which can be inconsistent and
thereby falsify axiom D), but that deontic logic only describes rules for rational norm giving: a rational law giver cannot enact norms which make both
OA and also OA true, because a rational law giver wants that all the norms
are obeyed, but a subject could not obey all norms if according to the norms
both OA and OA are true. This is the route that von Wright has taken
in his work since [252], and it seems also to be the one taken by Weinberger
when he writes that normative systems which contain conflicting norms can
exist in social reality, they are logically flawed, but do not loose their function and social meaning ([234] p. 25, my translation).26 But lawyers, when
they explain what ought to be done, may be done or is forbidden to do in
a certain situation, when e.g. a law has been violated or such a violation
is imminent, do not advise the law giver. They advise their clients, which
are usually the subjects of the law. Thus a solution that views deontic logic
only as a tool for law givers, which they may, unlikely as that seems, use in
the activity of passing a law, would miss a large part of ordinary language
discourses in which the expressions you are obliged to, you may, you are
forbidden to are used that deontic logic claims to formalize.
26
Cf. Holl
ander [100] p. 98 (my translation): By concluding that a norm-giver may be
reasonable or unreasonable (...), Weinberger reaches the same conclusion as von Wright.
DEONTIC LOGIC
71
The hard way is to accept the possibility of conflicts and make changes to
accommodate them and explain resolution mechanisms. Approaches to accommodate conflicts in traditional deontic logic reach from replacing (D) by
(P) O and dropping (C), and in the semantics use minimal models (neighborhood semantics) to pick alternative sets of deontic alternatives instead of
one (minimal deontic logic),27 or to use multi-ideality or multipreference
semantics that has not one deontic alternative or preference relation but several, and thus can make OA and OA true in a model that makes A true
in all ideal or best alternatives according to one relation, and A is true in
the ideal or best alternatives according to another,28 to sequent systems that
replace (C) by a rule of consistent cumulativity, which tests consistency
of A B before cumulating OA and OB to O(A B) and which can only
be used before a use of (M), and so avoids the derivation of OB from OA
and OA,29 to axiomatic systems that leave (C) intact but exchange (M)
for a rule of permitted monotony for the same reason.30 So there are ways
in which deontic logic can be modified to accommodate conflicts, and even
keep possible worlds semantics, though this semantics may not always be
very intuitive.
As for mechanisms to resolve conflicts, a natural way to do so is by
weighing the conflicting obligations and giving preference to the ones with
the highest priority.31 Though it seems not to have been done before, this
too might be accomplished by using preference semantics and in particular
Gobles [67] multiplex preference semantics, through introducing a preference
relation over the multiple ideality or preference relations (OA then holds all
things considered if A is true in the most ideal of the ideal worlds). However,
the severe changes that would be required and the multitude of proposals how
to go forward to solve the problem of accommodating conflicts and dilemmas
shows that there are no easy solutions along the old ways of thinking about
deontic logic. In fact, the difficulties of traditional deontic logic to handle
such problems have led to doubts about the whole enterprise:
27
Cf. Chellas [45] pp. 200203, 272277 for its monadic and dyadic variants, also cf.
Nortmann [159] who likewise employs minimal models but instead of (C) drops (M).
28
Cf. Schotch & Jennings [192], and Goble [66] [67] for such solutions.
29
Cf. van der Torre & Tan [214] for such a labelled sequent system.
30
Cf. Goble [69]. In [68] Goble also presents a possible worlds semantics that is sound
and complete with respect to such axiomatic systems.
31
Cf. Isaac Watts [220] part II, ch. V, sec. III, principle 10: Where two duties seem
to stand in opposition to each other, and we cannot practise both, the less must give way
to the greater, and the omission of the less is not sinful.
DEONTIC LOGIC
72
Various notions from normative reasoning, including the notions

of prima facie obligation, contrary-to-duty obligations, and moral
dilemmas have raised serious problems for deontic logicians, problems which have defied solution within standard deontic logic and
its extensions. As a result, deontic logic has fallen into ill-repute
with those who must actually engage in normative reasoning.
(Nute and Yu [161] p. 1)
I think that these problems, too, call for a fresh approach to deontic
logic. It seems that a more natural treatment of conflicts, dilemmas and
their resolution can be achieved by directly specifying the norms that conflict. Resolution mechanisms based on priorities or rankings can then use
an ordering of the norms themselves. Also, there are other problems that
are similarly hard to solve with traditional deontic logic and its possible
worlds semantics. One is the distinction between simple morality and the
supererogatory. When norms are explicitly represented, we can distinguish
between basic rules, binding on all, and the higher flights of morality of
which saintliness and heroism are outstanding examples (Urmson [215] p.
211), where the additional satisfaction of the second type of norms will make
an act supererogatory. We can also introduce a count on the number of
obligations satisfied, and say that up from a certain count, an act that satisfies the obligations is no longer ordinarily obligatory, but a supererogatory
act. By contrast, possible worlds semantics would require us to construct
series of ideal and more ideal worlds to model this notion, a construction
that seems as superficial as talk about preferences over preferences, or ideal
ideal worlds, that might be required to resolve conflicts and dilemmas in a
possible worlds setting.
2.6.3
Interpretations of Deontic Logic
The expressions that deontic logic formalizes, like it is obligatory that, it

is permitted that or it is forbidden that, are ambiguous in everyday usage:
they may be used to describe or report that something is obligatory according
to some implicitly referred-to moral or legal code, but also to make something
obligatory, permitted or forbidden. E.g. when a school teacher says it is
forbidden for you to talk in class during the exam, this may be because
there is such a general rule at the school, but it may also be so because I,
DEONTIC LOGIC
73
the teacher, say so. The ambiguity of such expressions is well-known.32

To distinguish their descriptive from their normative use, the descriptively
used expressions have been called normative, or deontic, propositions, as
opposed to norms. When devising the first axiom system of deontic logic,
von Wright [242] makes quite clear that it is the descriptive use of deontic
expressions which he has in mind:
The system of Deontic Logic, which we are outlining in this
paper, studies propositions (and truth-functions of propositions),
about the obligatory, permitted, forbidden (...). We call the propositions which are the object of study deontic propositions.
Deontic logic, in this first and strict sense, is thus not a logic of norms,
but a logic of propositions about norms. According to this descriptive interpretation of deontic logic, deontic sentences (represented by the formulae of
deontic logic) describe what is regarded as permitted, obligatory, forbidden
etc., in some unspecified normative system (Fllesdal & Hilpinen [53]). If
a sentence it is obligatory that ... is true, then there must exist norms
with respect to which it is true, and so in this descriptive interpretation the
deontic propositions can also be seen as statements about the existence of
norms: norm propositions are propositions to the effect that certain norms
exist (von Wright [251]). In one formulation or another, this descriptive
interpretation of deontic logic has come to represent the standard view.33
32
Cf. Ayer [24] pp. 105/106, Reichenbach [180] pp. 19/20, von Kutschera [128] pp. 11
14, Hilpinen [92] pp. 299/300. Bentham was the first to succinctly point at this ambiguity:
The only mood to which grammarians have thought fit to give the name imperative is
by no means the only one which is calculated to perform the office of imperation. (...)
But in many instances it will be found that the style of imperation is altogether dropped.
He proceeds to give a variety of forms of a law against the exportation of corn, among
these the legislator speaking as it were in the person of another man who is considered
as explaining the state which things are in, in consequence of the arrangements taken by
the legislator, It is not permitted to any man to export corn. It is unlawful for any man
to export corn. No man has a right to export corn. Why? because the legislator has
forbidden it. The exportation of corn is forbidden. ([29] pp. 104-106, 154)
33
Also cf. Kalinowski [114] p. 118 (my translation): For what does an expression of
the form it is obligatory that A mean? If this is not a norm but a proposition about
norms it means that the law giver L has enacted the valid norm N according to which A
is obligatory ; Opalek & Wole
nski [164] p. 384: Deontic logic is conditioned by norms,
because deontic statements are true or false depending on the existence of respective acts
of norming.; Sinowjew [197] p. 266 (my translation): The truth values of normative
DEONTIC LOGIC
74
The interpretation that the propositions of deontic logic are true and
false descriptions about what is obligatory, permitted or forbidden, and are
not themselves norms or imperatives, has not been shared by all authors.
Casta
neda [42] viewed deontic logic as a modal logic of imperatives and
resolutives.
Aqvist [21] argued that it should be possible to interpret deontic logic atheoretically as a logic of commands, in the sense that OA
expresses a command, and not a proposition about a command; occasional
oddities like the difficulty of interpreting formulas like OA should be accepted as a small price for a logical theory of commands. Such a prescriptive
interpretation of deontic logic was also adopted by Alchourron [2] when he
identified deontic logic with a logic of norms, and set out to develop a logic
of normative propositions in parallel. Chellas [44] replaced the O-operators
of deontic logic by the symbol !, where !A is to be understood as representing a natural language imperative let it be the case that A, and presents
a logic for such formulas that is equivalent to SDL. For Bailhache [25], deontic logic is the logic of deontic norms (norms that are created by the use
of deontic expressions), where the deontic formulas are then evaluated with
respect to accessible ideal worlds as usual. More recently, Kamp [116] p.
233 has demanded that the formulas of deontic logic must be interpretable
both descriptively and prescriptively. Sometimes it is not even quite clear in
which way standard accounts of deontic logic desire its formulas to be read:
Carmo & Jones [40] write that deontic logic is a formal tool needed to design normative systems and, just like Bailhache in [26], throughout call the
sentences of deontic logic norms or deontic norms and so seem to adopt
the prescriptive interpretation, but the authors then employ possible worlds
semantics to evaluate deontic propositions (norms?) as usual. Fllesdal &
Hilpinen [53] speak of deontic propositions constituting or being implied by
a normative system (e.g. pp. 13, 29), or of deontic logic formalizing imperatives (p. 26), but portray the descriptive interpretation of deontic logic as
one that they shall often resort to (p. 8). In the face of all this confusion,34 David Makinson [141] noted that work on deontic logic goes on as if a
propositions are determined according to the usual rules. E.g. a proposition it is forbidden
to smoke is true if it really is forbidden to smoke (if there exists such a norm), false if it
is permitted to smoke, and undetermined if there is neither a prohibition nor a permission
to smoke.
34
Holl
ander [100] pp. 9899 uses Hintikkas concept of a deontically perfect world to
motivate his deontic logic as a logic of norms in the following way: the deontically perfect
world is a place where conflicting norms cannot exist, as opposed to the real world where
DEONTIC LOGIC
75
distinction between norms and normative propositions has never been heard
of.
In part, the confusion about what the formulas of deontic logic mean can
be attributed to von Wright himself, who later wrote that while deontic logic
is a theory of descriptively interpreted expressions, its laws concern logical
properties of the norms themselves, and that therefore the basis of deontic
logic is a logical theory of prescriptively interpreted O- and P -expressions
([246] p. 134). This reflection of logical properties of norms in the logical
properties of deontic propositions have led von Wright to famously claim that
logic has a wider reach than truth ([245] p. vii). Still later, von Wright [251]
reaffirmed that deontic logic is a logic of norm-propositions and denied the
possibility of a logic of norms, except in the form of advice to a rational law
giver (cf. [252] p. 132, [256] p. 110).
However, it is quite clear that OA and P A cannot be interpreted as representing imperatives or norms. I have explained in sec. 1.2 why imperatives,
and for that reason, also deontic norms, cannot be meaningfully termed
true and false. But if norms are neither true nor false, then the Boolean
operators occurring in the formulas of deontic logic such as OA OB,
P A OA, OA P B, cannot have their usual, truth functional meanings as and, or, not, if ..., then. So a logic of norms, even if one believes
it exists, cannot resemble deontic logic,35 and the meaning of (sub-) formulas
such as OA must be interpreted descriptively.36 So the descriptive interpretation of the formulas of deontic logic is the only tenable one, and is also the
one used here.
What the confusion in deontic logic literature as to the meaning of its
formulas illustrates, is a need to better explain the relation between the
formulas of deontic logic, as logic of deontic propositions, and the norms
that these propositions are statements about. Von Wright hints at this when
writing in 1951 [243]:
they can. The logical relations studied are then those that exist between norms in the
deontically perfect world. But deontic alternatives, as described by Hintikka [97], [95], are
sets of propositions, not places where norms exist. If deontic propositions are true in a
world that is ideal to ours (it is not just one, as Hollander seems to assume), this is due
to other worlds that are ideal to the ideal world. All the previously mentioned problems
and confusions about deontic logic seem to eclipse here.
35
Cf. Makinson [141] p. 30. Keuth [123] and Swirydowicz [208] therefore restrict their
logic of norms to statements of normative entailment of the form !A !B.
36
That is, unless one is prepared to leave the common ground and call norms true and
false, as Kalinowski [111], [114] and Kamp [118], [116] have done.
DEONTIC LOGIC
76
In this essay the deontic modalities are treated as absolute. (...)

Instead of dealing with propositions of the type A is permitted,
we might consider propositions of the types A is permitted according to the moral code C (...) Extensions of the system P
will not, however, be studied in this essay.37
Traditional literature on deontic logic has mostly ignored this relation. If
mentioned at all, the deontic sentences are called true or false according to
some unspecified normative system (cf. Fllesdal & Hilpinen [53] p. 8).
But if norms are not somehow explicitly represented, it becomes difficult to
discuss examples. It has e.g. been suggested that if both O(A B) and
O(B C) true, i.e. there is a conflict, then a cumulation of contents should
not be allowed, for O(AB B C) would declare something impossible as
obligatory, but that instead a disjunction of the contents should be obligatory,
i.e. O((A B) (B C)) should be true. Thus the agent would be obliged
to satisfy at least one norm (cf. Jacquette [109]). But the sentences O(AB)
and O(B C) need not be true because two norms demand exactly A B
and B C. They might be true because two norms demand AB C and
ABC but then O((AB)(BC)) would only express an obligation
to violate at least one norm. Similarly, in order to cope with conflicts, it has
been suggested to exchange (C) for a form of consistent aggregation: it is
allowed to cumulate OA and OB to O(AB) if AB is consistent (cf. Goble
[68] p. 81). But OA and OB might be true because there are two norms that
demand A C and B C. Then O(A B) is demanding too much: one
of the norms must be violated anyway, so either doing A or doing B must
achieve nothing. It is no wonder that in such discussions it is tempting to
slip into a manner of speaking as if OA expresses a deontic norm, which is
violated by A (e.g. Carmo & Jones [40] p. 317, 321,). But OA is not a
norm, it is a sentence that cannot be violated or fulfilled, but just be true
or false; it is, at best, a premiss. Ziemba [258] p. 155 is therefore right
when he claims that the lack of distinction between commands (norms) and
deontic propositions, or propositions on commands, is a source of various evil
in deontic logic.
Taking our cue from von Wrights suggestion, I think that explaining
the meaning of the formulas of deontic logic by means of not just basically
37
Much later, Casta

neda [41] indexed deontic operators by adverbial qualifiers to indicate according to which set of norms the truth of Oi A holds, where i means e.g. a previous
promise or agreement, the rules of a job, or morality.
DEONTIC LOGIC
77
unknown, but an explicitly represented sets of norms will clarify such discussions and the understanding of deontic logic.
2.7
A Fundamental Problem of Deontic Logic
In the face of the problems and doubts outlined above, and the multitude of
proposals to modify deontic logic, it has been concluded that
deontic logic has fallen into ruts. The older rut is the axiomatic
approach, with its succession of propositional and occasionally
quantified calculi. The newer one is possible worlds semantics,
with endless minor variations in the details. (Makinson [141])
In consequence, a fresh start has been called for. A representation of norms
in deontic logic could constitute such a fresh start. However, if norms are to
be represented in deontic logic, there is then the question of how to do so
without falling into the trap of Jrgensens Dilemma [110]: how can norms
be a subject of logical reasoning if norms are neither true nor false and so
cannot be part of any logical inference? This has been called the fundamental
problem of deontic logic:
It is thus a central problem we would say, a fundamental problem of deontic logic to reconstruct it in accord with the philosophical position that norms are devoid of truth values. In other
words: to explain how deontic logic is possible on a positivistic
philosophy of norms. (Makinson [141])
Jrgensens dilemma cannot be dealt with, and the fundamental problem
not solved, by making the object language of deontic logic one of norms.
As was explained above: we cannot negate norms, form conjuncts or adjuncts out of expressions representing norms, for what meaning would the
Boolean operators then have? We must let the statements of deontic logic
remain true or false statements about norms, deontic or normative propositions that describe what is obligatory, permitted or forbidden according to
the norms. However, the norms according to which the normative proposition may be true or false should be explicitly represented in the semantics
of deontic logic. Deontic logic would then, as it once aspired, not talk about
ideal states, preferred worlds or perhaps preferences over preferences, but
DEONTIC LOGIC
78
more directly about positive law, not to forget the more profane usages of
normative language like rules set up by mums and dads, employers or school
teachers, with all their curiosities and inconsistencies.
If the semantics by which we evaluate the truth of deontic formulas is
changed, there are then the questions of what truth definitions should be
used for the deontic operators, and in particular what a thus semantically
reinterpreted deontic logic will be like: will the traditional systems of deontic logic still be sound and complete, or in Makinsons terms: can they
be reconstructed with respect to such a new semantics? Or will we arrive
at completely new axiomatic systems? The next chapter describes how the
formulas of deontic logic can be evaluated with respect to a logical semantics
that explicitly represents them (I call this imperative semantics), and what
systems result.
Chapter 3
Imperative Semantics for
Deontic Logic
3.1
The Imperatival Tradition
The idea to represent norms or imperatives in the logical semantics for deontic
logic is not entirely new. While the mainstream of deontic logic adhered to
possible worlds semantics, quite a number of authors have tried, in one way
or another, to link deontic logic to norms or to imperatives we might
call this the normative or imperatival tradition of deontic logic. What is
typical for such approaches is that they have, besides the truth definitions for
deontic operators, a set of objects that are meant to be prescriptions, or that
contains the propositions that relate to prescriptions in the sense explained
by Weinbergers Principle (W): the proposition that must be true when the
imperative is satisfied, and false if it is violated. The deontic operators
are then in turn defined with respect to this set, and not a possible worlds
semantics. Any proposal that has these features is counted as belonging to
this imperatival tradition of deontic logic.
3.1.1
Andersons Reduction Revisited
As explained in sec. 2.3, Alan Ross Anderson was the first to propose to
reduce deontic logic to alethic modal logic, and define OA by means of a
constant S representing a sanction:
OA =def (A S)
79
IMPERATIVE SEMANTICS
80
So A is obligatory iff A necessitates the sanction S. However, it was soon

pointed out that the relation between a wrongdoing and its sanction is not
one of necessity, since in this wicked world, it is not always true that when
(say) a robbery is committed someone suffers a sanction (Nowell-Smith &
Lemmon [160]) and Casta
neda [42] adds: Failure to do ones duty never
entails or strictly implies that punishment will be inflicted. It is commonplace
that many criminals die unprosecuted. Hence Anderson is accused of the
naturalistic fallacy, concluding from the moral or legal fact that something
is obligatory the empirical fact that something else (an act of sanctioning)
happens with necessity in this world.
Interestingly, in answering these accusations, Anderson did not, as NowellSmith & Lemmon suggested, re-interpret his necessity operator in terms
of a moral necessity. Rather, Anderson [17] argues that it is analytic of the
notion of obligation that if an obligation is not fulfilled, then something has
gone wrong.1 S is no longer interpreted as representing a penalty or sanction,
as in Anderson [16], but as the statement that some wrong has been done
(cf. Anderson [13]), that not all obligations have been fulfilled, or that
some set of rules is violated, and consequently the letter S is exchanged
for the letter V for violation. Similarly, von Wright argues in defense of
Anderson that his construction avoids the naturalistic fallacy if the sanction
is interpreted as punishability, where punishability does not obtain if and
only if all obligations are fulfilled (von Wright [250] pp. 36-37). Andersons
definition can therefore be rephrased as follows:
OA is true iff A necessarily implies V , where V means that some
rules are violated or that not all obligations have been fulfilled.
To further refine this definition, we obviously need an explicit set of obligating norms, with respect to which the terms violation or fulfilment can
then be defined. In [75] I have employed a set of imperatives to define constants v (some imperative is violated) and S (all imperatives are satisfied),
together with their duals V (all imperatives are violated) and s (some imperative is satisfied) in order to reconstruct Andersons defense and thus his
deontic truth definition. The standard system of deontic logic SDL and variants of it can then easily be shown to correspond to such a reconstruction of
Andersons reduction of deontic to alethic modal logic.
1
Also cf. Goble [64] p. 199: There is an analytical connection between the statements
(i) it is obligatory that p and (ii) if not-p, then V, where the constant V is to represent
some violation or bad state of affairs or wrong in the world.
3.1.2
81
Kangers Welfare Programs
As explained above (sec. 2.3), Stig Kanger, independently from Anderson,

also had the idea of reducing deontic logic to alethic modal logic, but instead
of an Andersonian sanction employed a constant Q for his definition of a
deontic O-operator, which reads
OA =def (Q A).
So OA is true iff Q necessitates A. Kanger gives the informal meaning of his
constant Q as stating what morality prescribes. While this may seem an
elusive concept, Kanger explains it as follows:
Consider the universe of discourse, and suppose there are welfare programs for this universe. A welfare program consists of a
set of propositions expressing what is desirable for this universe
from the viewpoint of human welfare. (...) Now the proposition
OughtA is true in the universe of discourse if and only if A is
entailed by each (...) welfare program for this universe.
Refining this idea, let P = {P1 , ..., Pn } be a set of sets of propositions, where
each P LPL represents some welfare program for the universe. Then the
following truth definition for a deontic O-operator corresponds to Kangers
explanation:
OA is true iff P P : P |=PL A.
In a proposal endorsed by Barcan Marcus [27], T. J. Smiley [198] elaborated
on Kangers ideas and explained that in Kangers definition, Q should be
seen as expressing the content of an arbitrary moral code, in which case
the definition is tantamount to saying that something is obligatory if it is
entailed by the code, permitted if it is consistent with it, and forbidden if
it conflicts with it. Smiley identifies Kangers constant Q with a code of
obligatory propositions, i.e. Q LPL and defines
OA is true iff there are propositions p1 , ..., pn in the code such
that p1 , ..., pn necessarily imply A.
Smiley then explains that this definition yields von Wrights classical system,
given that there is an additional consistency assumption that excludes that
there are obligatory propositions p1 , ..., pn such that (p1 ... pn ) is a
tautology. This seems to be the first reconstruction of a deontic axiomatic
82
system by means of a semantics that explicitly represents norms. I examine

Smileys definition in more detail in [75], [76]. Regarding the truth definition
corresponding to Kangers explanation, it seems not to have been explored
anywhere so far, but it can easily be seen as corresponding to a normal
modal logic, and to SDL iff each welfare program P P is assumed to be
consistent.2
3.1.3
Ziembas Deontic Syllogistics
Zdzislaw Ziemba [258], [259] defines a deontic syllogistics by use of a set C

of conditional commands. A command let every A be B! is formalized by
(A!B), where this is taken to be the name of the command. A describes a
property of a certain person over a certain period of time, e.g. the property
of being in a store with inflammable materials. If the person has the property
A, she is addressed by the command (A!B). B is then another property
of this person, e.g. the property of this persons not smoking over the same
period of time. So (A!B) is the command that forbids this person to smoke
while being in the store. A person is fulfilling a set of commands C if for all
commands (A!B) of the set that address her, she is B.
Deontic statements are then defined true and false with respect to this
given set of commands C . Thus the statement x prohC A, that being A is
forbidden for x according to C , is defined true iff x is addressed by some
command in C and it is not possible for x to be fulfilling all commands in C
and be A. Being obliged to be A is then defined as being prohibited to be
A, and being permitted as not being prohibited to be A.
Ziembas main idea is that to some person, by virtue of some set of commands, a certain conduct is forbidden when this person is an addressee of
a command belonging to this set, and that if the person fulfills the conduct, she inevitably disobeys some command of the set of which she is an
addressee. Ziemba [259] p. 382 is thus distinguishing a command from
a deontic proposition stating that this particular person has an obligation,
that he is forbidden or permitted to do so and so with respect to such and
such a set of commands. These commands are not logical propositions, and
they are neither true nor false. Though I will not use the syllogistics that
Ziemba proposes, he explains the meaning of his deontic statements solely
2
Regarding preference semantics, the proposal not to use one, but several sets of norms
corresponds to Gobles [66], [67] multipreference semantics that uses not just one, but
multiple preference relations to explain OA in terms in what is best according to all.
83
by use of an explicitly given set of commands, and so his proposal belongs

to the imperatival tradition of deontic logic.
3.1.4
Stenius Logic of Normative Systems
Stenius [205] distinguishes a norm sentence (a sentence used to express a

norm) from a factual interpretation that something is obligatory according
to some (possibly unspecified) system of norms S. The first is expressed by
Op, the second by Op S (read as p is obligatory in S).
Stenius then proposes several principles for a logic of normative systems
S. The two basic principles are:3
If (p1 ... pn ) q is a tautology,
then ((Op1 S) ... (Opn S)) (Oq S) is a theorem.
and
If p is a tautology then Op S is a tautology.
If we use Op to abbreviate Op S for some fixed system of norms S, and
Pp for Op
/ S, then O behaves like a normal modal operator, i.e. the logic of
this operator is described by the axiom scheme (K) O(p q) (Op Oq)
and is closed under the rules (N) if p is a tautology then Op is a theorem and
modus ponens. Other principles are rejected by Stenius, and in particular the
deontic axiom scheme (D) Op Pp is not viewed as valid, since although
a system of norms according to which Op S and Op S are true and so
it is impossible to obey might be called unreasonable, to say that such a
system exists is not a logical contradiction.
While Stenius interprets the factual or deontic statements Op (or Op
S) as statements about a normative system, the logic of these deontic statements only mirrors what statements of the form Op are already contained
in S. The principles he lays out are primarily principles of closure for S, not
principles of a deontic logic. It is therefore crucial for his proposal that e.g.
two norms Op and Oq that already are in S can be seen as deriving another
norm O(p q), i.e. a norm that must by the first principle then also be
in S. Thus his proposal hinges on the question whether there really are such
3
Note that p and q are here proposition variables, i.e. sentences from LP L , rather than
proposition letters. In this section I try to keep to the original texts as closely as possible.
84
logical relations between norms.4 In this respect his proposal and related
proposals5 differ from other proposals discussed in this section, and also the
proposal explained in the following sections: I do not assume any theory of
logical relations between norms or imperatives, for the simple fact that it is
doubtful and heavily disputed whether such relations exist, and that I agree
with a position of imperativological scepticism that, at least for imperatives,
altogether denies them. If, contrary to this view, such relations do exist, then
nothing is lost: we can close the set of norms or imperatives by any accepted
principle or rule and still use the formulas of deontic logic to describe what
is obligatory according to this set, just as Stenius proposes. However, if such
logical relations do not exist, then the interpretation of deontic logic as one of
expressions about sets of norms or imperatives will still remain meaningful.
3.1.5
Alchourr
on & Bulygins Expressive Conception
of Norms
In Carlos E. Alchourron & Eugenio Bulygins [5], [8] expressive conception

of norms, norms are taken to be sentences used to prescribe that something ought, ought not or may be (done). The existence of such sentences
is dependent on certain empirical facts, and as there are no logical relations
between facts, there is no room for a logic of norms. However, as Alchourron
& Bulygin point out, this does not preclude the possibility of a deontic logic
as a logic of normative propositions. For the propositions that have been
commanded by some authority (called Rex by Alchourron & Bulygin) can
be used to form a set, the commanded set A. The following first definition
for what is obligatory according to this commanded set is then proposed:
p is obligatory in A is true if p has been commanded by Rex and
so is a member of the commanded set A.
Alchourron & Bulygin then proceed to expand the set A and define the
concept of a normative system as the set of all propositions that are consequences of the commanded propositions, i.e. the system corresponding to A
4
For this reason, Stenius [206] later retracted his proposal and only called a system of
norms well-formed if it is not closed under consequences.
5
In a very similar way to Stenius, Keuth [123] and Swirydowicz

[208], [209] have defined
deontic operators via a normative system closed under rules, where Swirydowicz
uses
Ziembas [258], [259] dyadic formalization of imperatives which he enriches with logical
relations between the imperatives.
85
is Cn(A) (where Cn means classical consequence). Their second, corrected

definition for the truth of normative propositions is then the following:
It is obligatory that p in A is true if and only if p is a member of the system Cn(A) i.e. if and only if p belongs to the
consequences of A.
This second definition is similar to one employed by Niiniluoto [157] p. 113:
Op is true if and only if some norm-authority x has commanded
!p or p is a logical consequence of some sentences q1 , ..., qn such
that x has commanded !q1 , ..., !qn .
In [76] I consider an O-operator O1 that is similar to Alchourron & Bulygins
first proposal, in that O1 A is true if and only if it is explicitly commanded
to bring about A. For Alchourron & Bulygins second definition and the
definition used by Niiniluoto it is easy to see that both correspond to a
normal modal system, and that they correspond to standard deontic logic
SDL if it is additionally assumed that the norms are consistent, i.e. O is
not true. SDL is thus easily reconstructed with respect to a semantics that
models norms.
3.1.6
Van Fraassens Proposal
Conflicts of obligation are difficult to accommodate in standard deontic logic,

the equivalent formula (OAOA) of the characteristic axiom (D) excludes
them. Therefore Bas van Fraassen [57] introduced a semantics that defines
the truth of deontic operators with respect to explicitly given, possibly conflicting imperatives. The basis of his semantics is a set of imperatives in force
I = {i1 , ..., in }. To every imperative there is a class i+ B of valuations
v (van Fraassen calls these possible outcomes) such that i is fulfilled in v.
His first proposal for the definition of deontic O-operator is then:
OA is true iff, for some imperative i that is in force, i+ is part of
the set of possible outcomes in which A is true.
So something is obligatory if it is necessary for one imperative to be fulfilled.
Van Fraassens definition fails to include that as obligatory what is demanded not by one, but several (consistent) imperatives collectively. To do
so, he proposes the following, second definition of a deontic O-operator:
OA is true iff there is a possible state of affairs v in A whose
score is not included in the score of any v in A .
86
Here, the score of an outcome v is defined as the subset of the imperatives

that are fulfilled at v. So it is obligatory to do A if doing A makes it possible
to satisfy a set of imperatives that cannot all be satisfied by doing A. This
can also be expressed in another way: A is obligatory if A is necessary to
satisfy some maximal non-conflicting subset of the imperatives.
With respect to van Fraassens idea, John F. Horty [101] proved that the
same definition can be provided via an application of Reiters [181] default
logic. Here, a default is a triple A : B/C such that A is the antecedent
or prerequisite that must be established to draw the conclusion C, and
B, the justification, is a formula that a conclusion set must be consistent
with in order for the conclusion to be drawn (we might also think of B as
exceptional circumstances in which drawing the conclusion C is not permitted
even if A is present). A default theory = W, D consists of a set of of
formulas W (the facts) and a set of defaults D. The set of -conclusions of
an arbitrary set of formulas (we write Concl ()) is the smallest set such
that (i) W Concl () (the facts are conclusions), (ii) Cn(Concl ()) =
Concl () (the set is logically closed), and (iii) for each default A : B/C ,
if the prerequisite A Concl () and the negation of the justification B
/
, then the conclusion C Concl (). The appropriate conclusion sets (or
extensions) of the default theory are then all sets of formulas E such that
Concl (E) = E (the fixed points of the conclusion operator). Let I be a
basic imperative structure as defined above, and define the associated default
theory = W, D to be such that W = and D = { : A/A | !A I}.
Horty shows that van Fraassens second definition is equivalent to defining
OA is true iff A E for some extension E of the default theory
associated with I
In [102], Horty then presents a variation of van Fraassens definition that
makes OA relative not to some, but all extensions of the associated default
theory:
OA is true iff A E for all extensions E of the default theory
associated with I
This definition can then be rephrased as follows: A is obligatory if it is necessary to satisfy any maximal non-conflicting subset of the imperatives. Both
definitions provide an interesting way to continue reasoning about obligations in situations when there are conflicts or dilemmas, a way that standard
deontic logic does not support. In [76] and [77] I have explored the definitions
in more detail.
3.1.7
87
Input/Output Logic
In answering his own call [141] for a reconstruction of deontic logic in accord
with the position that norms are neither true nor false, David Makinson,
together with Leon van der Torre [142], [143], develops input/output logic
as a tool to extract the essential mathematical structure behind this reconstruction (cf. [144]). In their view, input/output logic should be seen not
as a new kind of logic, but as a way of handling norms: given a set of norms,
it must e.g. be determined which of these are still operative in a situation
that already violates (or satisfies) some of them. The set of norms G is taken
as a black box or transformation device, and the task of logic is then that
of a secretarial assistant, preparing the factual information that goes into
this device (the input A) and coordinating the production of the output,
written as out(G, A).
The simplest input/output operation that Makinson & van der Torre
suggest is that of simple-minded output. Let G be a set of conditional
norms (a, x), where a and x are sentences in the language of propositional
logic LPL , a describing some condition or situation, while x represents what
the norm tells us to be obligatory in that situation. A set of sentences
A LPL serves as explicit input. The operation first expands A to its
classical closure Cn(A). This is passed into the transformer G which delivers
as immediate output the set G(Cn(A)) = {x | (a, x) G and a Cn(A)}.
Simple-minded output out1 (G, A) is then again the classical closure of this
set, i.e. out1 (G, A) = Cn(out1 (G, A)). This operation is depicted by the
following figure from [144]:
Figure 3.1: Simple-Minded Output

So simple-minded output provides the propositions that must be true if the
88
conditional norms that are triggered by the input, i.e. whose conditions
are true given the input, are to be satisfied. Makinson & van der Torre
then refine this simple-minded output and provide further output operations
that e.g. provide reasoning by cases (if there are two norms (a, x) and (b, x)
in G, then x is in out(G, {a b})), normative detachment (if there are two
norms (a, x) and (x, y) in G then y is in out(G, {a})), or permit throughput
(if there is a norm (a, x y), then y is in out(G, {a, x})). Finally, the
operation may employ restraints to e.g. ensure that only such a (maximal
sub-) set of norms is used that provides only consistent output (output such
that
/ out(G, A)).
In the above outline, the output is a set of descriptive sentences that
might be seen as describing what is obligatory given the norms in G and the
situation described by the sentences in A (deontic interpretation). Hence we
may use an input/output operation out(G, A) to define a deontic logic via
G, A |= Ox iff x out(G, A)
i.e. Ox is true given the set of norms G and the factual descriptions A iff x
is in the output out(G, A) of the used output operation. But the operation
might also be viewed as providing new norms conditional on the situation
described by a singular sentence a as input, where the new set of norms
may be defined as out(G) = {(a, x) | x out(G, {a})}, i.e. the operation
provides the norm (a, x) just if x is in the output out(G, {a}) (normative
interpretation).6 In the first interpretation the head depicted in the above
figure acts as a kind of lawyer or legal advisor, telling the agent what ought
to be done according to the norms in the situation described by A. In the
second interpretation the head acts as a kind of judge, or subsumtion machine
(Subsumtionsautomat), handing out norms for specific situations as a result
of the (general) laws known to the judge, a picture that is remindful of the
Latin proverbs iura novit curia and da mihi facta, dabo tibi ius (the court
knows the law, give me the facts and I will give you the law). Makinson & van
der Torre leave it open how the output should be interpreted, in particular no
deontic operators are defined, but if the aim is to reconstruct deontic logic,
then obviously the first interpretation should be preferred. It is easy to see
that then e.g.
6
Cf. Makinson & van der Torre [142] p. 385. Swirydowicz [208] defines a normative
consequence relation very similar to Makinson & van der Torres basic output to provide
a method to derive norms from norms.
89
(Conj)
OA OB O(A B)
(N)
O(A A)
or statements such as if G, A |= Ox then G, A {a} |= Ox (strengthening
of the input) all hold for the simple-minded output operation out1 . The
main question will then be which kind of output operation provides the most
adequate results for a correspondingly defined deontic O-operator. In [80] I
have considered truth definitions of dyadic deontic operators that resemble
the operations of input/output logic, and discussed some of their properties.
I will to return to the results of this work shortly, but will now explain my
own approach to model norms in a semantics for deontic logic.
3.2
Imperative Semantics for Deontic Logic
If deontic propositions are to be evaluated not with respect to possible worlds,

but to given sets of norms, then these must be somehow represented in the
logical semantics of deontic logic. There are many different types of norms
like individual and general, norms that empower, grant licenses, or define
terms within other norms, act as basic principles, direct procedures, rule the
creation of norms, etc., and though it should be possible to finally arrive
at a logical semantics that represents all of them, it is better to start from
most basic terms. These seem to be simple, obligating norms that require
an individual to perform some action or achieve some state of affairs, the
sort of norms that are, in ordinary language, usually created by the use of
an (unconditional) imperative expression. In fact, I will simply employ a set
of imperatives I. Of course not all imperatives create norms. Imperatives
may fail to create a normative relationship, to bind the addressee to do the
bidding of the speaker, e.g. if uttered by a child or passerby, and sometimes
the subject can choose to accept them, e.g. when we enter a game or when a
cheerleader says everybody, clap your hands. But we can simply stipulate
that I is a set of imperatives that are accepted by the subject and, at least
in the subjects view, create a binding obligation.7
7
By stipulating that the imperatives in I are binding or valid, a notion that may in
turn depend upon some acceptance by the subjects and/or a legal community, I hope to
side-step all discussions on whether normative propositions are really true or false, or
rather just acceptable or non-acceptable (cf. Mazzarese [148], Niiniluoto [158]), for from
the same viewpoint that accepts these imperatives as binding, some normative propositions
that something is obligatory, permitted or forbidden according to the imperatives must
90
To each imperative corresponds a descriptive sentence that describes the

situation that obtains if and only if the imperative is satisfied. If there were
no such sentence, then the subject could not understand what the person
using the imperative wants her to do, and therefore not create an obligation.
This is Weinbergers principle (W), and I have already explained in sec. 1.3
that the fact that there is such a correspondence between imperatives and
indicatives seems to be unilaterally accepted. So in addition to the set I there
is a function f : I LBL that associates with each imperative a formula
from the language of some (descriptive) basic logic BL. For an imperative
i for which f (i) is A, I write !A, in accordance with tradition.8 As for
the basic logic, while any logic suffices that provides a consequence relation
BL A to express that A can be derived from , I will mostly let the
basic logic be propositional logic PL. This, the couple I = I, f , is the
basis of the semantics I will use. I call I a basic imperative structure and a
semantics that uses such structures basic imperative semantics. Extensions
of basic imperative semantics are possible and straightforward. E.g. in [78],
[80] I have added a relation < on I to express priorities between imperatives.
Other functions may be employed to indicate authorities or subjects of the
imperatives.
One might be tempted to reduce the semantics and replace I by f (I),
i.e. by a set of formulas of the basic logic LBL .9 But our semantics is about
imperatives, not indicatives, and there might be different imperatives that are
fulfilled under the same condition, but are e.g. of different rank or priority,
and so it might be important to distinguish them. Also note that, unlike in
other approaches like that of Alchourron [3], it is not assumed that the set
of imperatives I is closed under any logical rule. There is no commitment
to any logic of imperatives, theory of implicit commanding, etc.
I do not provide any further analysis of the content of an unconditional
imperative than what is explained above. Weinberger later argued that (W)
should be accepted only on the condition that the corresponding indicative
is not contradictory and not tautological,10 but I will not assume generally
that such restrictions hold. Also, it is clear that imperative sentences are
compositional and regularly contain connectives like not and or or, as
then be true or false.
8
Cf. Hofstadter & McKinsey [99], Reichenbach [180].
9
I have done so for better legibility in the largely formal papers [76] and [77].
10
Cf. already Weinberger [222] p. 121, and then similarly [225] pp. 1719, [229] p. 69,
[231] p. 229231.
91
in do not eat with your fingers, buy apples and walnuts, write her a
letter, or at least a postcard, and these imperatives are then satisfied by
not eating with ones fingers, buying apples and buying walnuts, and writing
her a letter or writing her a postcard. But Weinbergers principle as well
as the analysis of Hofstadter & McKinsey [99] show that in terms of the
sentences that describe the satisfaction of an imperative, a formalization of
these connectives within imperatives would only amount to a doubling of the
usual Boolean operators: if we employ the name !(p1 p2 ) for the imperative
buy apples and walnuts, its associated descriptive sentence f (!(p1 p2 )) is
still p1 p2 , i.e. the imperative is still satisfied just in case its addressee
buys apples and buys walnuts. Also, we may want to use expressions that
are imperatives not in the grammatical, but in a functional sense in our
examples, like I must ask you to pack your gear and leave the property,
can you close the window or turn off the ventilation, please? or you ought
to go now, and then, with the aid of principle (W), we need not worry about
each expressions proper formalization too much.
I will, however, make two assumptions about the contents of the imperatives: i) that imperative sentences are not separable, and ii) that they are
independent. Sometimes even doing only part of what has been requested
or commanded is seen as something good and (partly) following the order,
while at other times failing a part means that satisfying the remainder no
longer makes sense. E.g. if I am to satisfy the imperative buy apples and
walnuts, and the walnuts are needed for biscuits and the apples for dessert,
then it makes sense and may be required to get the walnuts even if apples
are out. (We just have the biscuits for dessert.) If, however, both land in a
Waldorf salad, then it might be unwanted and a waste of money to buy the
walnuts if I cannot get the apples. As the example illustrates, there is no
logical method to distinguish one case from the other. The following may
be used as a test for separability:11
(Sep) If by use of an imperative sentence i someone is required, among
other things, to do A and B, and is still required to do A even if
B cannot be done, then i is separable.
Instead of a separable imperative, we may just use two or more inseparable
imperatives to express what is required. However, ordinary usage of two
imperative sentences does not guarantee their independence. Suppose a flight
11
This resembles Hamblins [74] p. 184 definition of separability.
92
instructor tells a student to do the following at takeoff:12

Add power slowly at first and then gradually, smoothly bring
in full power. Take a quick glance in your mirror just as you
add power on the roll. Pay attention to the gliders nose. When
we add power too quickly, the gliders nose pitches up and the
tail slams down. If you think you have added power too quickly,
check your mirror. If the gliders nose is high in the air, then go
ahead and smoothly add full power.
Imagine the mirror is broken and so the flight student cannot check the
gliders nose when going up: is she still required to add power? Or should
she, as I think, rather not start with a broken mirror? In the second case the
separate imperatives combine inseparably, i.e. it is, from the standpoint of
the authority, not required to do one thing when the other one cannot be
done.13 The following test may be used to determine dependent imperatives:
(Dep) If two imperative sentences i1 and i2 have been used, but they
are such that if what i1 demands cannot be carried out, then the
satisfaction of i2 is no longer required, then i1 and i2 are dependent
on each other (combine inseparably).
Two or more separate imperatives that combine inseparably may then be
represented by one imperative that demands all of what the separate imperatives demand.
That sentences may be independent grammatically, but should not be
formalized so, is a phenomenon that also holds for descriptive sentences. E.g.
the statement Imagine me on the panel! I will most likely make a fool out of
myself! should not be treated as an imperative sentence and an (unqualified)
indicative prediction, but rather be reformulated as a conditional If I get put
on the panel, I will most likely make a fool out of myself. This then is an
encouragement to treat such problems as problems of formalization (proper
semantical representation), rather than as problems of logic.
The assumption that the imperatives in I are inseparable and independent becomes important when treating conflicts and dilemmas (where not all
imperatives can be satisfied collectively) or contrary-to-duty circumstances
(where some imperative has been violated). E.g. having spent most of the
12
13
The quote is from the manual of the Schweizer 2-33 sailplane.

Again, the terminology is owed to Hamblin [74] p. 184.
93
money on ice cream, I might not be able to buy both the apples and the walnuts, and so find myself in a dilemma. I might even have spent so much on
ice cream that I cant buy the walnuts with the change: I am in a contraryto-duty situation. But it might still be enough to purchase the apples. Depending on how the imperative is to be understood, I might still be under
an obligation to buy the apples or the walnuts (in the first case), or just the
apples (in the second), or nothing at all (if I was shopping to help prepare a
Waldorf salad).
The above approach to modeling normative situations is very basic. As
noted above, there are all kinds of norms for which the core concept would
have to be amended and adjusted think of explicit permissions (licenses),
or of general norms: in order to model, within the basic concept, a general
norm not to lie, one would have to add to the set, for every subject and any
occasion to lie, an individual, independent imperative that obliges the subject
not to lie on this occasion. While some adjustments of the basic concepts
are treated below, a more fundamental criticism of my approach would be
that it does not seem to account for norms that are not created by use of
speech. One may think e.g. of natural law, Kants categorical imperative, or
more down-to-earth rules of courtesy, custom or tradition. Now, as long as
these can be rephrased to be inseparable and independent, and correspond
to descriptions of situations in which they are satisfied or violated in the
sense explained by Weinbergers principle (W), nothing hinders us to make
I a set of such norms. The set I is not an ontological category, though it
is explained in terms of imperatives, uttered by some authority, that are,
by some subject, believed to be binding. This seems to me the best way
to reconstruct everyday normative reasoning, when we consider individual
demands from possibly different sources and try to satisfy them as best as
we can. However, I admit that when obligations are considered that are
(only) justified by an appeal to prohairetic notions something is obligatory
because it is the best thing to do, it brings about the greatest good for the
greatest number, in all possible situations everyone is better off with it done
than not done, etc. , then the more natural model may be one in terms of
ideality or preference relations between possible worlds after all.
3.3
94
The Reconstruction of Deontic Logic
Heeding Makinsons call for a solution of the fundamental problem and

a reconstruction of deontic logic as a logic concerned with norms, but in
accord with the philosophical position that norms are devoid of truth values,
we can apply the semantics described in the previous section to give truth
definitions of monadic and dyadic deontic operators, and define systems of
deontic logic that are sound and complete with respect to such a semantics.
Among these are also the standard systems of monadic and dyadic deontic
logic as described in chapter 2, and so it is demonstrated that these can be
reconstructed with respect to a semantics that explicitly represents norms,
but not as entities that can be true or false. My project to apply imperative
semantics to solve the fundamental problem was carried out in a series of
research papers since Makinson first stated it at the EON98 workshop
in Bologna, Italy, and here the main results of the individual papers are
summarized.
3.3.1
Sets, Sentences, and Some Logics about Imperatives [2001]
The first paper in the series, written for the EON00 workshop held in
Toulouse, France, 2022 January 2000 and subsequently published in revised
form in 2001 [75], describes the basic imperative semantics and outlines how
it may be applied to solve Makinsons fundamental problem.
The first result is almost trivial: Let I = I, f be a basic imperative
structure and define
I |= OA iff {i1 , ..., in } I : f (i1 ), ..., f (in ) |=PL A
So according to this truth definition, OA is true if and only if there are imperatives in the set I such that their associated descriptive sentences classically
imply A, i.e. A is necessary to satisfy some imperatives. Letting the truth
definitions for Boolean operators be as usual, if I is assumed to be a set
of non-conflicting imperatives, i.e. f (I) PL , then von Wrights classical
axiomatic deontic system is sound and complete with respect to this semantics. If I is additionally assumed to be non-empty, i.e. there is at least one
imperative i I, then standard deontic logic SDL, restricted to a language
that does not permit mixed formulas and nested deontic operators, is sound
and complete with respect to this semantics.
95
The paper then proceeds to a reconstruction of Andersons reduction.

Taking our cue from Andersons idea that his constant means that some
violation has occurred, and Kangers idea that his constant represents the fact
that all propositions of a welfare program are satisfied, the paper presents
four different constants defined with respect to a basic imperative structure
I = I, f and valuation v B:
I, v |= S
iff i I : v |= f (i)
I, v |= s
I, v |= V iff i I : v |= f (i)
I, v |= v
So according to these definitions, S is true at a valuation v if all imperatives
are satisfied at it (this models Kangers constant), s is true if some imperative
is satisfied at v, V is true if all imperatives are violated at v, and v is true if
some imperative is satisfied at v (this is Andersons constant). The following
(Kangerian) definition of a deontic O-operator is then employed:
OA =def ( s v) (S A)
Here, the operator represents (logical) necessity, and A means that A is
true at all valuations v B. Possibility is defined as as usual. Since
imperatives that demand contradictions or tautologies to be realized are not
per se excluded, the formula s v must be used to ensure that the set
of imperatives is non-empty, i.e. there are imperatives to satisfy or violate.
So OA is true iff the set of imperatives is non-empty and A is necessary to
satisfy all imperatives. Since S is equivalent to v, an Andersonian definition
of OA
OA =def ( s v) (A v)
may be equivalently employed to define OA as true if A is necessary to avoid
a violation. It is then shown that the deontic fragment of a system that
axiomatizes these constants and operators (the axioms are immediate) equals
the classical system of deontic logic if an axiom S is employed to ensure that
the set of imperatives is again non-conflicting, and equals standard deontic
logic SDL if additionally an axiom s v is employed to ensure that the set
of imperatives is non-empty.
Finally the paper presents a first approach to the problem of reconstructing the dyadic deontic systems of deontic logic with respect to imperative
semantics. For this, a nave contraction operator n is defined that removes those sentences from a set LPL that derive a sentence A, i.e.
n A = {B | |=P L B A}. This contraction operator is then used
96
to remove all those imperative-associated sentences from f (I) that are already violated in a situation described by C, where C represents a possibly
contrary-to-duty situation. This contraction is nave as, of course, there
may be imperative-associated sentences left that each can be individually but
not collectively true together with C. But I wanted to distinguish contraryto-duty situations from conflicts and dilemmas, and preserve the possibility
of the latter and the possible truth of O in such situations. So consequently
the truth of a dyadic deontic O-operator is defined with respect to a basic
imperative structure I = I, f as follows:
I |= O(A/C) iff f (I) n C {C} |=PL A
According to this definition, O(A/C) holds if A is necessary to satisfy all
the imperatives that are not violated in the circumstances described by C.
Curiously, the dyadic deontic logic that is proven to be sound and complete
with respect to such a truth definition is just like the system DDLH described
in sec. 2.5 (i.e. like Hanssons DSDL3) except that it lacks its typical deontic
axiom (DD-R). We can also make the full set DDLH sound and complete by
assuming that dilemmas are excluded for all possible circumstances, i.e. that
if PL C then also f (I) n C PL C. However, this would commit us to
assuming that for any two imperatives !i1 , i2 I, either |=PL f (i1 ) f (i2 )
or |=PL f (i2 ) f (i1 ) otherwise the situation (f (i1 ) f (i2 )) constitutes a
dilemma. To assume that the demands of all imperatives are thus chained
rather corresponds to the nested system of spheres provided by preference
based semantics, but it seems unlikely that any real-world set of commands
will ever meet this assumption.
3.3.2
Problems and Results for Logics about Imperatives [2004]
My second paper on imperative-based semantics, written for the EON02

workshop held in London, UK, 2224 May 2002 and subsequently published
in 2004 [76], explores possible definitions for monadic deontic operators in
more detail.
In the first part the following truth definitions for deontic O-operators
with respect to a basic imperative structure I = I, f are considered:
97
I |= O1 A iff i I : f (i) = A
I |= O2 A iff i I : |=PL f (i) A
I |= O3 A iff i I : |=PL f (i) A
I |= O4 A iff i1 , ..., in I : |=PL (f (i1 ) ... f (in )) A
I |= O5 A iff f (I) |=PL A
So O1 takes obligatoriness very literal, describing A as obligatory only if there
really is an imperative that demands a state of affairs properly described by
A. O1 is the operator first proposed by Alchourron & Bulygin [5]. O2 is similar, but allows A to be replaced by any logically equivalent proposition. For
O3 A to be true it is sufficient that A is necessary to satisfy one imperative.
This is the first of the definitions for OA that van Fraassen [57] proposed.
The definition of O4 is similar to Smileys [198] definition of a deontic operator: O4 A is true iff there is non-empty set of imperatives such that A
is necessary to satisfy them. We have already seen in the first paper that
this definition, with an assumption of the imperatives being non-conflicting,
yields von Wrights classical deontic system (with the language restricted so
that O-operators do not occur iterated or nested). Finally O5 is similar to
the corrected definition used by Alchourron & Bulygin [5] in that O5 A is
true iff A is necessary to satisfy all imperatives (regardless if there are any).
This definition, together with the assumption that the imperatives do not
conflict, then yields SDL (with the language restricted as above). Sound and
complete axiomatic systems are given for all operators.
I then consider what happens if we mix the operators. Obviously we can
then express truths such as Oi A Oj A for 1 i < j 5. A maybe
unexpected finding is that though all possible mixtures of operators have
sound and (weakly) complete systems (the additionally required axioms are
forthcoming), no system but the one that only uses operators O4 and O5 is
also strongly complete: the semantics is not compact, i.e. there are infinite
sets of formulas such that each finite subset of these can be modeled using
basic imperative semantics, but not the whole sets. A similar experience
awaits us for the use of semantic restrictions. Four such restrictions are
considered:
[R-0] I =
(Non-Triviality)
[R-1] i I : PL f (i)
(Excluded Impossibility)
[R-2] f (I) PL
(Collective Satisfiability)
[R-3] i I : PL f (i)
(Excluded Necessity)
1
2
It turns out that for O and O , no axiomatic system is strongly complete
98
if the set of imperatives is restricted by [R-0], and for O3 and O4 no system

is strongly complete with respect to a set of imperatives restricted by [R3]. Logics based on O5 are strongly complete with respect to semantics
that use either [R-0] or [R-3], but not when both restrictions on the set
of imperatives are applied collectively. For all other variations there are
sound and (strongly) complete systems. The lesson to be learned here is that
though all of the above restrictions seem somehow intuitive (exclude empty
sets of imperatives, exclude imperatives that demand the impossible or
the logically necessary, exclude conflicts), one cannot arbitrarily apply them
without risking to loose mathematical properties that might be desirable for
a corresponding deontic theory.
The next part of the paper focusses on a truth definition that is similar to
van-Fraassens [57] second truth definition. Phrased with respect to a basic
imperative structure I = I, f , it reads:
I |= OF A iff I : f () PL A
Van Fraassen intended to permit conflicts, i.e. cases in which OF A and
OF A are both true, but not OF . To do so, the truth of OF A is defined
with respect to maximal non-conflicting subsets of imperatives: I A is the
set of all maximal sets of imperatives such that the sets of corresponding
descriptive sentences do not derive A i.e. I A contains all I such
that (i) f () PL A and (ii) there is no I: and f () PL A.
OF A is then defined true iff A is necessary to satisfy the imperatives in some
maximal non-conflicting subset of I.
What surprised van Fraassen himself is that such a definition corresponds
to a sound and (as I show: only weakly) complete deontic axiom system that
employs the axiom schemes (Ext), (M), (N) and (P) P for the OF -operator.
But his truth definitions also yields the following truth: if !A, !B I, and
A B is consistent, then OF (A B) is true, i.e. if two imperatives do not
conflict then what they conjunctively imply is also obligatory. So a limited
form of cumulativity is admitted, but axiomatically this is not expressed.
I show in the paper how this difficulty can be overcome by using the O2 operator as defined before.14 However I want to note here that instead also
the following OC -operator could be used:15
14
E.g. we have that if (A B), then O2 A O2 B OF (A B).

I owe this observation to Lou Goble (private correspondence). Note that OC is stronger
than Kennys [122] operator of satisfactoriness (cf. above p. 34): a definition of Kennys
operator is I |= OK A iff i I : |= A f (i); and then OC A OK A is valid. Also note
15
99
I |= OC A iff i1 , ..., in I : |=PL (f (i1 ) ... f (in )) A

So according to this operator, it is obligatory that A iff A is necessary and
sufficient to satisfy one or more imperatives. A mixed axiomatic system
based on the operators OC and OF and defined via the axiom schemes (Ext)
and (C) (for OC ), and (Ext), (M), (N) and (P) (for OF ) as well as
(F1 ) If P L A then OC A OF A is a theorem.
(F2 ) OC A OF B OF (B A) OF (B A)
is then sound and (weakly) complete with respect to this semantics. (F1 )
expresses the already noted truth that if A is what one or several imperatives
demand, and is also consistent, then it is obligatory according to the truth
definition for OF . (F2 ) is the corresponding truth, so far unnoticed in deontic
logic literature, that if A, though demanded by some imperative, cannot be
agglomerated with a sentence B for which OF B is true, i.e. OF (B A) is
true, then the fact that OF B holds must be due to a conflicting imperative
that demands B and the contrary of A, and so OF (B A) must be true.
In the paper I then introduce a sceptic operator OS first formally discussed
by John F. Horty [102],
I |= OS A iff I : f () PL A
that expresses that A is obligatory iff its truth is necessary for the satisfaction
of any maximal consistent subsets of the imperatives, and show that SDL is
sound and complete with respect to this truth definition. I also introduce
axioms for a combined system for the O2 , OF and OS -operators that is then
proved to be sound and (weakly) complete.
In the final part of the paper I address the question of nested or iterated
deontic operators. The language of deontic logic is now, like LDL defined
in sec. 2.2, completely unrestricted, i.e. we may have mixed formulas like
B OA, or formulas in which deontic operators occur iterated as in OOA
or nested as in O(A OB). In the semantics, the imperative-associated
sentences use the same language LDL , and e.g. !OA means an imperative
let it be forbidden that A which is satisfied iff it is forbidden that A. I prove
that for all of the previously defined O-operators O1 O5 (defined now using
deontic logic as basic logic), the earlier systems are still sound and complete,
and in particular the axiom schemes proposed for such a language, like
(DT) O(OA A),
(D4c ) OOA OA,
that the O-operator of S. O. Hanssons [85] preference-based deontic logic PDL behaves
similarly to OC (cf. van der Torre [213] pp. 16-17).
100
all fail: an order to see to it that parking on highways is forbidden does

not by itself make it forbidden. In this respect, deontic semantics based on
imperatives and deontic semantics based on ideality or preference part ways.
3.3.3
Conflicting Imperatives and Dyadic Deontic

Logic [2005]
My third paper on imperative-based semantics, written for the EON04

workshop held in Madeira, Portugal, 2628 May 2004 and published in 2005
[77], explores the two truth definitions proposed by van Fraassen and Horty
to cope with possible conflicts of obligations in a dyadic deontic setting.
The following truth definitions of dyadic deontic operators, relative to a
basic imperative structure I = I, f , stand in the center of the paper:
I |= OF (A/C) iff I C: f () {C} PL A
I |= OS (A/C) iff I C: f () {C} PL A
So OF (A/C) is true if A is required to satisfy some imperatives whose demands are collectively consistent with the current, possibly contrary-to-duty
situation described by C. Correspondingly, OS (A/C) is defined true if A
is required to satisfy any maximal subset of imperatives such that their
demands are collectively consistent with the situation C. Again, the OF operator allows conflicts, it may be that both OF (A/C) and OF (A/C)
are true, whereas OS cautiously only pronounces that as obligatory what
is necessary to satisfy any maximal set of imperatives whose demands are
consistent with the situation C, and so (OS (A/C) OS (A/C)) holds for
all non-contradictory C.
Similar definitions to the ones above have already been used in other contexts, and the paper proceeds to point out similarities to Kratzer and Lewiss
premise semantics [125], [138],16 Poole systems without constraints [173], the
X-logics of Siegel and Forget [194], [55], and the epistemic states-semantics
of Bochman [30], [31], [32] that is similar to the cumulative models in Kraus,
Lehmann and Magidor [126]. It is then proved that to the definitions above
correspond sound and (I prove: only weakly) complete axiomatic system,
where the axioms for OF are just like that for DDLL except that (DC) does
16
Lewis [138] (p. 233) states in a footnote that for deontic conditionals, the premises of
the premise semantics might be understood to be something that ought to hold, and so
seems the inventor of this kind of imperative semantics.
101
not hold, (DD) is replaced by (DP) P ( /C), and (CCMon) is used as additional axiom scheme, and the axioms for OS are exactly like those for DDLH ,
except that (RMon) is exchanged for (CCMon), and finally the following
mixed axiom schemes are added:
(RMonF SS )
P F (D/C) (OS (A/C) OS (A/C D))
(RMonSSF )
P S (D/C) (OS (A/C) OF (A/C D))
The OS -axioms are equivalent to those used by Kraus, Lehmann and Magidor
[126] for their system P except that the deontic axiom (DD-R) is added.
The OF -axioms are those of Bochman [30], except that his axiom (Pres) is
strengthened to (DP). The mixed schemes are again those of Bochman.
Besides the new and relatively complex completeness proof, the paper
points out a link to Gobles deontic multiplex preference semantics [66], [67]
which in terms of possible worlds ordered by several preference relations
similarly attempts to accommodate conflicting obligations. The dyadic
deontic formula O(A/C) is defined true by multiplex preference semantics
iff there is at least one preference relation such that the best C-worlds are
A-worlds (this corresponds to OF ) or if according to all preference relations
the best C-worlds are A worlds (this corresponds to OS ). I show that the
construction of a set of imperative-associated sentences described by the
earlier completeness proof, that makes true a formula of deontic logic A, can
also be used to define preference relations that make true A by use of these
definitions, and that except for a small change in the restrictions applying
to (DN) and (DD)/(DP) the above axiomatic system is then also weakly
complete with respect to such a multiple preference semantics.
3.3.4
Deontic Logics for Prioritized Imperatives [2006]
My fourth paper on imperative semantics was originally written for the working group Law and Logic at the XXII. World Congress of Philosophy of Law
and Social Philosophy (IVR 2005), held 2429 May 2005 in Granada, Spain;
a revised version was published in 2006 [78].
While the previous paper considered conflicts between imperatives and
how deontic operators, in particular OF and OS , can be defined that take
them into account and localize these conflicts, this paper studies methods to
resolve conflicts. It is commonplace that a priority relation, or ranking, between the imperatives may help to resolve conflicts. W. D. Ross [189], to that
effect, distinguished prima facie duties, possibly conflicting obligations that
102
appear to hold before we look for some ordering, provided e.g. by considering which are more pressing, ranks of authorities, gross differences in utilities
etc., from what we view as our duties sans phrase after all these things have
been considered. While e.g. the O3 -operators employed previously might be
seen as describing prima facie duties only, letting it be sufficient for OA that
A is necessary to satisfy one imperative, I now consider prioritized imperative structures I = I, f, < that additionally employ a priority relation <
between imperatives, where i1 < i2 means that i1 takes priority over i2 , and
how these might be used to resolve conflicts to arrive at a definition for an
O-operator that describes all-things-considered obligations.
There have been quite different proposals as to how priorities might be
used to resolve conflicts. I adopt an algorithm first introduced by Rescher
[183] and further developed by Brewka [35], [36] for the purpose of determining preferred subtheories in theory revision, a proposal that makes very
weak demands on < (it is irreflexive, transitive and well-founded, i.e. infinite descending chains are excluded). This then is compared with other
methods suggested in the literature, such as Hortys [104] definition of binding imperatives, Alchourron & Makinsons [9] criterion of least exposure,
Prakkens [175] hierarchical rebuttal, Sartors [190] prevailing relation, and
Alchourrons [4] use of safe contraction. These methods then turn out to be
either equivalent, or to produce counterintuitive results in examples where
Brewkas method does not.
The dyadic deontic O-operator is then defined with respect to such prioritized imperative structures I = I, f, < as follows:
I |= O(A/C) iff I C: f () {C} PL A
This is similar to the definition of the dyadic OS -operator used in the previous paper (a definition similar to that for OF could have been used just
as well), but instead of defining O in terms of what is required to satisfy
the imperatives in all maximal subsets of I such that f () does not imply
C, i.e. all sets in I C, we use all sets in a set I C of maximal
preferred subsets such that f () does not imply C. Very roughly, for any
situation C, any set in I C is defined by first adding a maximal set of
the most important imperatives such that their demands do not derive C,
then adding to it a maximal set of the second most important imperatives
that can be added without the set of corresponding demands now deriving
C, and so on. The axiomatic system that corresponds to such a definition
is then not surprisingly, because it was not assumed that all conflicts can
103
be resolved by the use of priorities still the system P of Kraus, Lehmann

& Magidor [126] with the dyadic deontic axiom (DD-R) added, as proved
before.
Finally the paper examines what properties of the priority relation must
be assumed so that it resolves all conflicts that might arise for possible situations C in a set of imperatives I. This is in line with von Wrights proposal in
[249] pp. 68, 80 that an axiological order can provide a safeguard against any
genuine predicament. It obviously suffices to let the (irreflexive, transitive,
well-founded) priority relation be total, i.e. for any two distinct imperatives
i1 , i2 , either i1 < i2 or i2 < i1 , but the paper shows that this requirement of
a total ordering may be relaxed somewhat. It is then proved that if the set
of imperatives is thus uniquely prioritized, then the axiomatic system that
is sound and complete with respect to the above truth definition is DDLH ,
i.e. Hanssons logic DSDL3. It is easy to prove that if for such uniquely
prioritized imperatives a truth definition similar to OF is used, i.e.
I |= O(A/C) iff I C: f () {C} PL A,
then the corresponding axiom system is the Lewis-type dyadic deontic logic
DDLL . This completes the reconstruction of the deontic standard systems
presented in chapter 2: the standard dyadic deontic logics DDLH and DDLL
can be reconstructed by use of an imperative semantics that assumes that
all conflicts and dilemmas are solvable by use of a priority relation.
Dyadic deontic logic has often been linked to conditional prescriptions,
but the reconstruction shows that it is not necessary to take special considerations about conditional imperatives into account: when the primary imperative is !A, a contrary-to-duty imperative may be formalized as !(A B)
or !(A B) to produce the result that O(A ) and O(B/A) are true.
That dyadic deontic logic can be viewed as being primarily concerned with
contrary-to-duty situations, and not so much with conditional prescriptions,
may be seen as an interesting point of the 2004 and 2006 research papers.
3.3.5
Prioritized Conditional Imperatives:

Problems and a New Proposal [2007]
Imperatives may be conditional. E.g. consider the instruction when the

water boils, add the fat and the flour. Using a material conditional in the
imperatives associated sentence, and e.g. representing it by !(C (A B)),
will not do: otherwise one could fulfill the instruction simply by not letting
104
the water boil. But if the water does not boil, e.g. because there is a power
failure or one decides not to continue with the recipe, one has not done
anything right or wrong: the opportunity to satisfy or violate the instruction
did not arise.17 Much of the literature on imperatives and deontic logic is
devoted to understanding the reasoning about conditional norms, which von
Wright [253] called the touchstone (Pr
ufstein) for normative logic.
In a paper written for the interdisciplinary seminar on Normative MultiAgent Systems NorMAS07 held 1823 March 2007 at Dagstuhl, Germany,
and subsequently published in revised form in [80], I have formalized conditional imperatives by introducing an additional function g : I LPL to basic
imperative semantics that associates with each imperative the description of a
condition that provides the opportunity for satisfying or violating the imperative (I write A !B for an imperative i such that g(i) = A and f (i) = B).
Given a set of factual descriptions W LPL , an imperative is called triggered if its antecedent is true given W then the opportunity for its satisfaction or violation arises.18 The set Triggered(W, I) = {i I | W PL g(i)}
is then the set of imperatives that are triggered given the facts W . A triggered imperative g(i) !f (i) is satisfied iff W |=PL g(i) f (i), i.e. if also
its consequent is true, and violated iff its consequent is false given the facts,
i.e. iff W |=PL g(i) f (i). This makes it possible that a conditional imperative g(i) !f (i) is neither satisfied nor violated. When an imperative is
neither satisfied nor violated because the antecedent is false, the imperative
is called obeyed.19 A subset I of imperatives is obeyable given the
facts W if m() W PL , where m(i) = g(i) f (i) is the materialization of i, i.e. the material conditional corresponding to the imperative. So
is obeyable given W iff it is not the case that for some {i1 , ..., in } we
have W PL (g(i1 ) f (i1 )) ... (g(in ) f (in )): otherwise we know that
whatever we do, i.e. given any maxiconsistent subset V of LPL that extends
W V , at least one imperative in must be violated.
The truth definitions of deontic operators are then defined with respect
these conditional imperatives, and in particular with respect to those impera17
This is Casta
nedas [41] distinction between those contents of an obligation statement
that work as circumstances and those that work as deontic foci.
18
Cf. Rescher [184], Sosa [200], van Fraassen [57] Also cf. Greenspan [72]: Oughts do
not arise, it seems, until it is too late to keep their conditions from being fulfilled. Horty
[104] seems to have introduced the term triggered.
19
The terminology differs here: Downing [48] uses the term complied by instead of
obeyed, while Vranas [?] calls the imperative avoided.
105
tives in the set of imperatives I that are triggered. Several options arise, and
in parallel to input/output logic as developed by Makinson & van der Torre
[142], [143], I discuss the following definitions for O-operators with regard to
thus extended conditional imperative structures I = I, f, g :
(td -cd1) I |= O(A/C) iff f (Triggered({C}, I)) PL A
(td -cd2) I |= O(A/C) iff V LP L C : f (Triggered(V, I)) PL A
(td -cd3) I |= O(A/C) iff f (Triggered ({C}, I)) PL A
(td -cd4) I |= O(A/C) iff V LP L C : f (Triggered (V, I)) PL A
Truth definition (td -cd1) corresponds to Makinson & van der Torres simpleminded output out1 . It pronounces A as obligatory if A is necessary to satisfy
all imperatives that are triggered by C. Definition (td -cd2), like basic output out2 , allows for reasoning by cases, it e.g. makes O(A/C D) true
in case there are two imperatives C !A and D !A and the situation is
described by C D (in the definition, each V LP L C is a maximal
subset of the language that is consistent with C). Definition (td -cd3), corresponding to reusable output out3 , allows normative detachment and makes
e.g. O(A/C) true if there are two imperatives C !B and B !A and the
situation is described by C (in the definition, the set Triggered (W, ) of iteratively triggered imperatives means the smallest subset of I such that
for all i , if f (Triggered (W, )) W PL g(i) then i Triggered (W, )).
Definition (td -cd4), corresponding to Makinson & van der Torres reusable
basic output out4 , then permits both, reasoning by cases and normative detachment.
It seems important to be able to use circumstantial reasoning, i.e. employ the information about the situation not only to determine if an imperative is triggered, but also for reasoning with its consequent. E.g. if the set of
imperatives is {A!(B C)}, with its imperative interpreted as expressing
if you have a cold, either stay in bed or wear a scarf, one would like to
obtain O(C/A B), expressing that given that I have a cold and dont stay
in bed, I ought to wear a scarf. To allow for such circumstantial reasoning
for all the above definitions, and corresponding to the throughput versions
out+ of input/output logic, the above definitions may be changed into
(td -cd1+ ) I |= O(A/C) iff f (Triggered({C}, I)) {C} PL A
(td -cd2+ ) I |= O(A/C) iff V LP L C: f (Triggered(V, I)) V PL A
(td -cd3+ ) I |= O(A/C) iff f (Triggered ({C}, I)) {C} PL A
(td -cd4+ ) I |= O(A/C) iff V LP L C : f (Triggered (V, I)) V PL A
Though all these modifications of the original simple-minded definition (td -
106
cd1) may be justified by corresponding intuitions, we are, unfortunately, not

completely free in combining them: both (td -cd2+ ) and (td -cd4+ ) collapse
into treating an imperative A !B equally to one that demands a material
conditional !(A B), and (td -cd4) permits ghostly contrapositions that I
argue do not seem adequate for a logic about conditional imperatives. So
reconstructing deontic logic with respect to conditional imperatives presents
some difficult choices.
The paper then focusses on how these definitions may be changed to
accommodate possible conflicts of imperatives, but where (similar to the
previous paper) a priority relation holds between the (now conditional) imperatives that can be used to resolve some of them. Instead of conditional
imperative structures I = I, f, g we now use prioritized conditional imperative structures I = I, f, g, < to evaluate the truth of a formula O(A/C).
The question is then how the definitions must be changed to reconstruct
real-life reasoning used to resolve conflicts between conditional imperatives
by means of priorities. In this respect, a challenge is posed by the so-called
order puzzle. Consider the imperatives in the following drinking and driving example, ranked in descending order:
(1) Your mother says: if you drink anything, then dont drive.
(2) Your best friend says: if you go to the party, then you do the driving.
(3) Some acquaintance says: if you go to the party, then have a drink
with me.
and assume that the situation is such that you go to the party. Intuitively,
you should satisfy the second-highest ranking imperative used by your best
friend, i.e. you should do the driving, but not satisfy your acquaintances
desire to have a drink with her, as this would then violate your mothers order
which ranks highest. So you should drive, but it is not true that you should
drink. Formalizing the ordered imperatives by A !B < C !B < C !A
and the situation by C, it turns out that, curiously, existing approaches to
formalize reasoning with prioritized conditionals, proposed in Horty [105],
Marek & Truszczy
nski [146], Brewka [37] and Brewka & Eiter [38], yield
either O(A B/C) (so you ought to drink and drive) or at least O(B/C) (so
you should drink).
Several methods are discussed to resolve the order puzzle in accord with
intuitions. I propose that regarding the truth definitions above, the relevant
subsets of the possibly conflicting imperatives in I are those that are maximally obeyable in the sense explained above: to help determine what the
107
agent ought to do, the method should at least not hand out sets of imperatives of which it is already known beforehand that at least one imperative
must be violated. In order to respect priorities and use them to solve some
or possibly all of the conflicts, Brewkas original algorithm from [35], [36]
(cf. above sec. 3.3.4) is then employed to construct the preferred maximally obeyable subsets by testing at each step whether adding a next-lower
imperative leaves the constructed set obeyable. Instead of I, these preferred
maximally obeyable subsets are then used in the above definitions to describe
as obligatory what is, for all these sets, necessary to satisfy all the imperatives in the set that are triggered, triggered etc. I show that all of the above
definitions produce the intuitively right result for the order puzzle, and so
the proposed method is also neutral with respect to the question which of
the above definitions should be chosen.
3.4
The Paradoxes of Deontic Logic Reconsidered
For a long time deontic logic has been plagued by so-called paradoxes,
seemingly counterintuitive formulas that are nevertheless valid in the main
systems of deontic logic. It might be questioned whether these are at all
solvable, or do not rather stem from different uses, in quite different contexts
of normative reasoning, of the natural language phrases it is obligatory that,
it ought to be that, it is required that, it is morally necessary that, it
has been commanded that that seemingly all are to be encompassed by the
one O-operator of deontic logic. A reaction to this situation may be, as
was proposed above, to define not one, but several O-operators that talk
differently about the norms. The paradoxes of deontic logic then serve as
benchmarks how well one particular definition of an O-operator reflects the
intended usage of the natural phrases in these benchmark examples.
3.4.1
Rosss Paradoxes
The first paradox, Rosss paradox, was originally levelled not at deontic logic,
but at the logic of imperatives as designed by Jrgensen [110] in accord with
Dubislavs proposal in [50] (cf. above sec. 1.5). Ross [188] showed that if
such imperative inferences are accepted, then it is possible to derive from
108
the imperative post the letter the imperative to post the letter or burn it,
which seems counterintuitive.
In deontic logic, both in von Wrights classical system and in the standard
system SDL, all instances of the scheme OA O(A B) are theorems. If p1
is meant to formalize the sentence John posts the letter, and p2 the sentence
John burns the letter, then Op1 O(p1 p2 ) is an instance of this scheme
and formalizes the natural language expression if John ought to post the
letter, then John ought to post the letter or burn it, which might be seen as
similarly problematic as the original imperative inference.20 In sec. 3.3.2 we
have seen that definitions of O-operators of the type O4 and O5 may be used
to reconstruct the classical and the standard system of deontic logic, and
so the formula Op1 O(p1 p2 ) is valid for such readings of the O-operator.
However, it should be noted that using the definitions of these operators,
Rosss paradox appears much less problematic than it did for imperative
inferences. Though both O4 and O5 describe the normative situation, the
truth of Op1 O(p1 p2 ) for both operators cannot be used to claim that
by burning the letter some, if only derived, imperative is satisfied. A good
model of the normative situation in case of Rosss paradox is to let the
set of imperatives be {!p1 }, i.e. there is (only) the imperative to John to
post the letter in the relevant set of imperatives. The definitions of O4
and O5 then make true Op1 , so John ought to post the letter, and, since
both operators define OA as true if and only if A is necessary to satisfy
some (or all) imperatives, also O(p1 p2 ) is true, for any situation that is
satisfactory with regard to the only imperative !p1 will also make p1 p2
true. However, we do not have that bringing about p1 p2 is also sufficient
to satisfy some imperatives: in particular, there is no imperative in the set
{!p1 } that will get satisfied by bringing about p2 , i.e. burning the letter. This
was different for Rosss original setting, where burning the letter satisfies the
derived imperative !(p1 p2 ), and so seemingly forces us to acknowledge that
in this case, John did at least something right, did something that satisfied
an imperative etc. Since the set of imperatives is not closed by any logic
of imperatives (there is none), there is also no imperative in the set that
gets satisfied by burning the letter. So I think that in this way, imperative
semantics really takes the edge off Rosss paradox.
20
Cf. Danielsson [47] for a recent claim that such a theorem is problematic and should
be avoided. However, his axiomatic proposal to eliminate the paradox does not work, as
shown in [79].
109
It seems to me that defendants of the view that Op1 O(p1 p2 ) should

not be true, use a different interpretation of ought than that which was used
to define the operators O4 and O5 : a reading of OA for which it does not
suffice that bringing about A is necessary for satisfying the imperatives, but
in that A must be also sufficient for satisfying at least one imperative. In
this reading, the truth of O(p1 p2 ) is unacceptable: just posting the letter
or burning it will not suffice for John to do as he was commanded to do. For
such strong readings of ought, the operators O1 , O2 and OC may be used
that were also defined in sec. 3.3.2: for these operators OA is defined true
only if A is necessary and sufficient to satisfy at least one of the imperatives.
Of course this is not only the way a deontic operator can be semantically
defined to avoid the truth of the formula Op1 O(p1 p2 ): we can e.g.
define an operator OR in the following way:
I |= OR A iff f (I) R A
Here R means Weingartner and Schurzs relevant R-consequence that was
expressly tailored to avoid Rosss paradox (see above p. 31). If one thinks
that this really is how the reasoning about obligations works in natural
language, then such a definition would be appropriate. The set of proposals
for how to define O-operators in imperative semantics is not closed. But I
think that a reading of ought in the sense that bringing about A is required
to satisfy the imperatives (but not necessarily sufficient to satisfy them) is
used at least sometimes in natural discourses and that the paradox only
occurs when we confuse this weak sense of ought with its strong sense. So
with its O-operator, standard deontic logic may catch at least sometimes the
natural meaning of ought.
Rosss second paradox, also called Weingartners Paradox or the window
paradox in sec. 1.5, receives a similar solution. Here it is seen as paradoxical that from an imperative close the window and play the piano a single
imperative to close the window or to play the piano can be inferred. The
situation may be modeled by a set {!(p3 p4 )}, where its singular imperative
demands that John closes the window (formalized by p3 ) and plays the piano (formalized by p4 ). As explained before, the imperative !(p3 p4 ) is, as
it should be, modeled as inseparable: nothing right is done by closing the
window or playing the piano alone. Although e.g. the operators O4 and O5
used to reconstruct the classical and the standard system of deontic logic
make true all of O(p3 p4 ), Op3 and Op4 (all of the sentences in the scope of
the O-operator describe what is necessary to satisfy the imperatives), neither
110
p3 nor p4 alone are also sufficient to satisfy an imperative. So again I think

this takes the edge off the paradox. Someone who views it as paradoxical
that we may conclude John ought to play the piano from John ought to
close the window and play the piano seems to be using a stronger ought
than expressed by these operators, and for such a strong ought one of the
operators O1 , O2 , or OC may be used, as they only make OA true in the
model for formulas A that are (at least) equivalent to p3 p4 . But consider
again the setting of the window paradox (cf. sec. 1.5). Suppose that ,
after issued the command close the window and play the piano to x, and
not hearing anything for a while, enters xs room, saying Did you not listen
to ? You ought to play the piano. x replies: It is true that I could play
the piano, and I could even first close the window in order to not disturb the
neighbors. I heard what said and I understand it. I also accept that I have
to do what I am ordered to do by . But it is not true that I ought to play
the piano. I think that this would be a very strange reaction to what and
have said. So I think that at least in some interpretation, given the order
to x close the window and play the piano, the statements are true that x
ought to play the piano, x is required to play the piano, x was ordered
to play the piano etc. So logics that formalize this reading should not be
discarded. It suffices that the semantics provides methods for all parties to
define their reading of ought in this paradox.
3.4.2
Contrary-to-duty Paradoxes
Priors paradox of commitment [177] as well as Chisholms paradox [46] were

expressly directed at the first systems of deontic logic. However, in discussing
the paradoxes it might be helpful to keep apart the prescriptions that underly
the respective settings from the deontic propositions that are true then.
Priors paradox pointed at the difficulty of defining conditional obligation
in (monadic) deontic logic: from a statement Op1 we derive O(p1 p2 ),
which may be interpreted as, given that p1 describes something forbidden,
like stealing, the realization of p1 commits the agent to p2 , where p2 may describe anything (e.g. committing adultery, etc.). Let {!p1 } be the relevant
set of imperatives, it contains just one imperative to Jane not to steal some
certain goods. The operators O4 and O5 used to reconstruct the classical
and the standard system of deontic logic both make O(p1 p2 ) true, which
expresses that it ought to be that Jane either does not steal or commits adultery. However, note that there is no secondary, contrary-to-duty imperative
111
in the set of imperatives. In this sense, it is not true that Jane is under a
conditional obligation to commit adultery. Now consider what happens if
the set does in fact contain such a conditional, contrary-to-duty imperative,
e.g. one that in this case commands Jane to return the stolen goods. Let
p1 !p3 be this contrary-to-duty imperative, i.e. the set of imperatives is
now {!p1 , p1 !p3 }. The first thing to note here is that the first imperative
does not derive the second imperative there is no such inference relation
between imperatives. For modeling obligations that arise from conditional
imperatives in possibly contrary-to-duty circumstances, a number of definitions were proposed in [80]. They also take into account possible priority
relations, but we need not worry about these here. When we are not interested in how obligations in one situation may influence obligations in some
other situation, we can also draw the situation description outside the scope
of the dyadic operator, i.e. make the situation part of the model and not
the operator. The simplest definition of OA with respect to a conditional
imperative structure I = I, f, g (cf. sec. 3.3.5) and a set of facts W is then
the following:
()
I, W |= OA
iff
W : f (Triggered(W, ))
PL
The set I m W is the set of maximally obeyable subsets of the set of

imperatives I regarding the situation described by W , which is defined as
the set of all maximal sets I such that {g(i) f (i) | i } W PL .
The set Triggered(W, ) = {i | W PL g(i)} contains all conditional
imperatives that are triggered by the facts, i.e. their antecedent made true.
OA is then defined true if the truth of A is required for the satisfaction of
the triggered imperatives in all maximal subsets of conditional imperatives
that in this sense do not conflict with the facts. Applying this to the set
of conditional imperatives {!p1 , p1 !p3 }, if there is no special information
about the situation, i.e. the sets of facts W is empty, then OA is true just
for propositions A that are necessary for the truth of p1 , i.e. Jane must
not steal, or do anything that includes stealing. If we obtain the additional
fact that Jane has violated the imperative to steal the goods, then the set
of facts is changed to W = {p1 }, and then the obligations also change:
the imperative not to steal is violated and therefore removed from the set
of relevant imperatives used to deliberate what is obligatory from now on,
and the imperative to return the goods has been triggered. The result is
that OA is now true only for such A that are necessary for the truth of p3 ,
i.e. for returning the stolen goods. We do not have Op2 in the modeled
112
contrary-to-duty situation, i.e. it is not true that Jane ought to commit

adultery given that she has stolen the goods. O(p1 p2 ) is still true in the
initial situation, but only because not-stealing-or-committing-adultery is a
necessary condition of not stealing. It does not express a commitment to
bring about p2 . If one is unhappy with such a weak notion of ought, and
prefers a stronger version that makes only obligatory what is necessary and
sufficient to satisfy one or more imperatives, then akin to the operator OC
defined before the following definition may be used,
()
I, W |= OA iff I m W : i1 , ..., in Triggered(W, ) :
PL (f (i1 ), ..., f (in )) A
and then in the initial situation OA holds for just those A that are equivalent
to p1 (i.e. not stealing), and in the contrary-to-duty-situation that W =
{p1 } then OA holds for just the A that are equivalent to p3 , i.e. to returning
the stolen goods.
Chisholms paradox receives a similar resolution. Chisholms [46] paradox
consists of the following sentences:
(1) It ought to be that a certain man go to the assistance of his neighbors.
(2) It ought to be that if he does go he tell them he is coming.
(3) If he does not go, then he ought not to tell them he is coming.
(4) He does not go.
The sentences were originally seen as paradoxical because whatever formalization is chosen in standard deontic logic, they will either be inconsistent
in SDL, or one of its four sentences will derive one of the others. For its
treatment in imperative semantics, let the set of relevant imperatives be
{!p1 , p1 !p2 , p1 !p2 }. It is easy to see that in a the situation without
the fact described by (4), i.e. with W = , definition () makes true OA
just for those A that are necessary for the truth of p1 , i.e. for the mans
going. Among these is Op1 , but also O(p1 p2 ). If one does not like the
last statement to be true and prefers a stronger ought-operator, one can
switch to definition (), and then OA is true just for those A that are either
equivalent to p1 , i.e. the man is just under an obligation to go and assist
his neighbors. It is disputed whether already the obligation to tell that he
is coming should be considered as triggered in the initial situation: though
the man has an obligation to go and thus to make true the antecedent of
the conditional imperative to call the neighbors, we do not know yet for sure
if the man will in fact go, and then telling and not going might be considered worse then not telling and not going. But for those who accept such a
113
deontic detachment, () may be changed into

()
I, W |= OA iff I m W : f (Triggered (W, )) PL A,
where the definition of Triggered considers also those imperatives as triggered whose antecedents must be true if the other triggered imperatives are
to be satisfied (cf. sec. 3.3.5 for the formal definition). Definition () then
makes OA true for all propositions A that are necessary for the truth of
p1 p2 , i.e. for going and calling, among these O(p1 p2 ), Op1 , Op2 , but
also O(p1 p2 ) and O(p1 p2 ). Again, if the last two truths are not
accepted, and a stronger version of ought is preferred, then the definition
may be changed again into
()
I, W |= OA
iff
I m W : i1 , ..., in Triggered (W, ) :

PL (f (i1 ), ..., f (in )) A,
and then OA holds only for those A that are equivalent to either p1 p2 , p1
or p2 , i.e. only going, calling and going-and-calling is obligatory. Now change
the set of facts into W = {p1 }: Chisholms sentences tell us that the man is
not going to go. The set of relevant imperatives shrinks to {p1 !p2 }: it
is the only imperative triggered and still possible to satisfy given our updated
knowledge. Both definitions () and () make OA true just for those A that
are necessary for the truth of p2 , i.e. for not telling the neighbors that he
is coming, and the stronger definitions () and () define OA true only for
those A that are equivalent to p2 . So not calling his neighbors is the only
obligation left. If we add the definition
I, W |= A iff W PL A,
for all non-deontic, propositional formulas A LPL , then in this situation
also p1 is described as true, which is what also might be demanded regarding
Chisholms paradox. Thus I think Chisholms paradox is adequately treated
in imperative semantics.
3.4.3
Paradoxes of Defeasible Reasoning
Finally, there have appeared paradoxes for deontic logic that also address
contrary-to-duty imperatives, but in these scenarios the contrary-to-duty situation gives rise to a special obligation to violate a primary rule. One of these
paradoxes, termed the paradox of the considerate assassin describes the following situation: the set of imperatives is {!k, !c, k !c}. The imperative
!k is intended to mean that you should not kill the witness, !c means
114
that you should not offer anyone a cigarette (including the witness), and
k !c means that if you kill the witness, you should offer her a cigarette
first. Prakken & Sergot [176], who first introduced this example (though in a
deontic setting, not a setting with explicitly represented imperatives), argue
that the solution should make Oc true for the situation k, as this applies
the imperative that is more specific for the given circumstances. However,
it is easy to see that e.g. truth definition (), with the set of imperatives as
described and the set of facts W = {k}, does not make Oc true: the set of
maximally obeyable subsets of I in this situation is {{!c}, {k !c}}, i.e.
the assassin has the choice to either keep to the rule of not offering cigarettes,
or offering, in this special case, a cigarette. Both commands apply to the
situation, and it has not been explicitly specified that the second should take
priority over the first. Hence only O(cc) is true, and thus only tautologies
are described as being obligatory.
A similar idea underlies the second example, also first proposed by Prakken
& Sergot [176] and slightly modified by Makinson [141]. Here the set of imperatives is {!f, f !(f w), d !(f w)}: There is a general prohibition
of fences !f except if there already is one in that case it should be white,
i.e. f !(f w) or if the owner has a dog, in which case the owner
should have a white fence, i.e. d !(f w). Again, the more specific imperative is intended to be applied in the situation where there already is a
fence or there is a dog, and so for the situation f the sentence O(f w)
should be true, and likewise for the situation d. For the situation f , definition () produces the right result: the set of maximally obeyable subsets
in this situation is {{f !(f w), d !(f w)}}, the only triggered imperative in this set is f !(f w), and so Of w) is defined as true.
However, things are different for the situation where there is (only) a dog,
i.e. W = {d}: the imperative that there is to be no fence is not yet violated
and so still considered relevant, and so the set of maximally obeyable subsets is {{!f, f !(f w)}, {f !(f w), d !(f w)}}. In each of the
two subsets there is just one imperative that is triggered in the situation d,
namely !f and d !(f w) respectively, and so only O(f (f w)), and
what else is necessary to make f (f w)) true, is described as obligatory.
In particular, O(f w) is false.
While it is true that neither definition (), nor any other definition proposed for imperative semantics, includes a specificity test, I am not convinced that this is a defect. For default rules that describe a knowledge base,
it might be argued that facts learned for more specific circumstances should
115
override possibly incomplete more general knowledge, and so the rule that
birds fly should not be applied to penguins once we have learned that these
do not fly, but it still should be applied to other birds. For imperatives there
does not appear to be such a general rule of overriding specificity. A soldier who is ordered before a mission to, under no circumstances, use lethal
force, and is then ordered by another officer to, if meeting a certain enemy,
kill that enemy, might, upon meeting this enemy, wonder what to do even
if both commands stem from equally ranking officers. In law there is the
legal principle lex specialis derogat legi generali, but this is not universally
applicable to all sets of norms, particularly not if they stem from from different parts of the legal system, and when it competes with other principles
like lex posterior, lex superiori, or standard argument forms like teleological
interpretation. But it seems that if there is a case where a more specific
imperative should take priority, we can use a priority ordering to express
such relations between the imperatives. So, for the first example, let the priority ordering be !k < k !c < !c, and let the priority ordering include
d !(f w) < !f in the case of the second example (where i < j means
that the imperative i takes priority over the imperative j). For prioritized
conditional imperative structures I = I, f, g, < the definition () may be
redefined as follows:21
()
I, W |= OA iff PIo (W, I) : f (Triggered(W, )) PL A
Here, the set P o (W, I) is the set of preferred maximally obeyable subsets as
defined in [80] (cf. above sec. 3.3.5). For total priority orderings like the ones
in the above examples (we ignore the additional imperative f !(f w) in
the second example, the paradox pertaining to this imperative was already
solved) it suffices that for this case the (only) set in PIo (W, I) is constructed
by adding the highest-ranking imperative i1 to i1 if W {g(i1 ) f (i1 )}
is not inconsistent, adding the second highest-ranking to i2 if i1 W
{g(i2 ) f (i2 )} is not inconsistent, and so on, and letting be the union
of all sets i such that i I. So in the first example, where W = {k},
I = {!k, !c, k !c}, and the prioritization is !k < k !c < !c, the
set of relevant (preferred) imperatives is {k !c}: in the first step, !k is
not added since it is already violated in the situation k, in the second step
k !c is added since it is consistent with k, and in the third step nothing
is added since {k c, c} is not consistent with k. Because k !c is also
21
This is similar to definition (td -pcd 1-8) that was defined with respect to preferred
maximally obeyable subsets in [80].
116
triggered given the facts W = {k}, definition ( ) then makes true Oc. So if
the assassin is going to kill the witness, and the order to offer the witness
a cigarette first takes priority over the general rule not to offer cigarettes,
then the assassin ought to offer the witness a cigarette first. The solution
for the second example, and the situation that there is a dog, is constructed
similarly. Thus both paradoxes receive a fair solution.
I think that these paradoxes also show the usefulness of imperative semantics for deontic logic. Here, priority orderings pertain to given imperatives,
not to statements of obligation that are true in relation to them. If we only
had these statements, as in traditional deontic logic, a reasoning with priorities would be much harder: we would have to ask e.g. whether a formula
OA that derives from a formula O(A B) of known priority also inherits
its rank, and if it is exactly the same rank or some lower one. Likewise it
would be hard to determine the preferred subsets that are consistent with
the situation: we do not know whether the truth of OA holds because there
is an obligation to realize A, or perhaps an obligation to realize A B which
might no longer be relevant since B is true, or which might already be
overridden by a higher-ranking obligation to realize B but canceling the
obligation from which OA derives would force us to cancel also OA, unless it
also derives from some other obligation, but in that case its rank must now
perhaps be changed, etc. But statements of obligation are not true by themselves, they are true because there exist certain norms or imperatives, and it
seems much more natural to say that it is these that get canceled if they can
no longer be satisfied (or are already satisfied), or are overridden by other
norms. Imperative semantics makes these relations easier to understand and
to model the deontic propositions that are true with respect to the norms.
Chapter 4
Conclusion
In 1998, David Makinson pointed out a fundamental problem of deontic
logic: if deontic logic is supposed to be concerned with norms, how can it
be reconstructed in accord with the philosophical position that norms are
devoid of truth values? In reaction to Makinsons call I argue that norms, or
simpler: imperatives, should be explicitly represented in the logical semantics
of deontic logic. The approach proposed here, and the imperatival tradition
of deontic logic it follows, thus depicts deontic logic as a logic about norms,
not of norms.
If there is a logic of imperatives or norms, then perhaps this approach
would not be much worth. If, in the face of Jrgensens Dilemma, we can
e.g. use Dubislavs convention to derive an imperative !A from a set of
imperative !B1 , ..., !Bn if the set {B1 , ..., Bn } classically implies A, then an
assumption of closure of the set of norms, perhaps together with an assumption of normative consistency (exclusion of conflicts), yields a deontic logic
(a logic of propositions about what is obligatory according to the norms) that
is isomorphic to such a logic of norms. The logic of normative systems proposed by Stenius serves as a typical example: here deontic logic is restricted
to stating what prescriptions exist in a normative system that is closed under given rules, i.e. the deontic proposition OA is true if and only if the
prescription OA exists in the system of of norms S. From such a viewpoint,
deontic logic must then appear as a dull isomorphism or ersatz theory.
However, Rosss paradox, in which from the imperative post the letter the
imperative post the letter or burn it is derived, and Weinbergers variant,
the window paradox, in which from the imperative close the window and
play the piano the imperative play the piano is derived, have drawn into
117
CONCLUSION
118
doubt the adequacy of Dubislavs convention: in both cases we are forced to

acknowledge that the (possibly forbidden) burning of the letter or the (possibly forbidden) playing the piano without closing the window first, satisfies
an imperative, though only a derived one. Since Poincare first suggested the
possibility of imperative inferences, various versions of normological skepticism have denied the very concept of a logic of imperatives or norms. In the
first chapter I have laid out my own argument why the search for a logic of
imperatives is futile and should not be continued: there appear to be no logical ordinary language arguments in which a premiss and the conclusion are
in the imperative mood. So imperative logic turns out to be what Dubislav
called an Unding or chimaera: a non-thing that exists only as a concept, but
no real objects (ordinary language arguments) fall under the concept.
By contrast, ordinary language discourses abound about what one ought
to do, may do, should not do in the face of requests, demands or orders
that are explicitly given, known and accepted, but are sometimes difficult to
follow and occasionally conflict with each other. Deontic logic may clarify
what agents should or need not do, first by examining the demands, see
if they are dependent or independent, separable or inseparable, conditional
or unconditional, and then by telling the agents what is, according to the
interpretation used, necessary to satisfy the demands, as far as they still
can be in given circumstances. When there are conflicts, logic may aid in
determining underlying rankings or priorities, which can then be used to solve
the conflict. Even if a conflict is not resolvable, logic may formalize methods
to localize it and tell the agent what other obligations remain. Deontic logic
seen in this way is a logic at work, it goes beyond merely explaining its modal
expressions, and also it is not primarily a tool to construct normative systems
or give advice to rational law givers, but rather it provides and reconstructs
methods which are and can be used by rational subjects drawing conclusions
about what they ought to do when faced with multiple demands, in situations
that the authorities might or might not have foreseen, and even when the
demands are such that a single, rational authority would not have made
them. It is, in this sense, a positivistic deontic logic. Such a logic should not
be seen as an an ersatz theory for some as yet undiscovered logic of norms.
Rather, by employing a logical semantics for deontic logic that interprets its
operators in terms of explicitly given norms, it overcomes the weaknesses
of both, proposed logics of norms (that formalize norms, but cannot agree
on logical principles) and of traditional deontic logic (that provides logical
principles, but neither formalizes norms nor represents them semantically).
CONCLUSION
119
The reconstructions demonstrate that all the work put into the construction of existing deontic logic systems has not been in vain. Underlying intuitions have been captured in axiomatic systems that are still sound and
complete with respect to a semantics that explicitly represents norms. In
particular, SDL is sound and complete for a semantics that defines a deontic
O-operator in terms of what is necessary to satisfy a set of imperatives, provided that they do not conflict, and Hansson- or Lewis-type standard dyadic
deontic systems are sound and complete with respect to a semantics that
defines a dyadic deontic O-operator in terms of what is necessary to satisfy
all those imperatives that still can be satisfied in possibly contrary-to-duty
circumstances, provided that all conflicts and dilemmas are taken care of by
a total priority relation. But aside of these standard systems, new systems
have appeared that allow us to e.g. meaningfully talk about obligations even
if a priority relation does not exist or is not total and therefore conflicts
must be taken into account, and new definitions are possible for operators
that e.g. single out obligations pertaining to particular imperatives, be they
overridden by stronger ones or not, or define what is necessary and sufficient
to satisfy at least one imperative. This multitude of possible definitions does
not, in my view, mean that deontic logic has become a logic à la carte. It
rather responds to the different circumstances in which deliberation is required, and to different ways we talk about obligations, e.g. when we treat
them as being prima facie or all things considered.
Inevitably, with the reconstruction of existing deontic logic systems the
work has just begun. It started with a semantics that represents unconditional imperatives, but as I demonstrated in [80], the representation of
conditional imperatives presents difficult choices for the definition of an appropriate O-operator. Permission is a neglected concept in deontic logic, we
must find a way to represent explicitly given licenses which override prohibitions that may apply otherwise (it appears senseless to explicitly permit something that has not previously been, and never will be, prohibited).
Norms about the creation of new imperatives or licenses must be taken into
account. Some of these problems and other challenges have been outlined in
Hansen & Pigozzi & van der Torre [81]. However, I think that the reconstructions have shown that modeling natural language imperatives and the
reasoning of the subjects they address is a fruitful approach to deontic logic.
So I am confident that the imperatival tradition of deontic logic, to which my
work contributes, will continue to be useful in developing intuitive solutions
to these problems.
Acknowledgements
My work on imperative-based semantics for deontic logic would not have been
possible without the fruitful discussions at the biannual EON workshops
where the small deontic logic community meets. In particular I want to
thank Lennart
Aqvist, Lou Goble, Jeff Horty, David Makinson and Leon van
der Torre for their constant encouragement, helpful suggestions and always
well-placed criticism. All errors that remain are mine.
120
Bibliography
[1] Achenwall, G., Prolegomena Iuris Naturalis, In Usum Auditorum, 3rd
ed., Gottingen: Sumtibus Victorini Bossigeli, 1767.
[2] Alchourron, C. E., Logic of Norms and Logic of Normative Propositions, Logique & Analyse, 12, 1969, 242268.
[3] Alchourron, C. E., The Intuitive Background of Normative Legal Discourse and its Formalization, Journal of Philosophicl Logic, 1, 1972,
447463.
[4] Alchourron, C. E., Conditionality and the Representation of Legal
Norms, in: Martino, A. A. and Socci Natali, F. (eds.), Automated
Analysis of Legal Texts, Amsterdam: North Holland, 1986, 173186.
[5] Alchourron, C. E. and Bulygin, E., The Expressive Conception of
Norms, in [94] 95124.
[6] Alchourron, C. E. and Bulygin, E., Unvollstandigkeit, Widerspr
uchlichkeit und Unbestimmtheit der Normenordnungen, in: Conte,
A. G., Hilpinen, R. and von Wright, G. H. (eds.), Deontische Logik und
Semantik, Wiesbaden: Athenaion, 1977, 2032.
[7] Alchourron, C. E. and Bulygin, E., Pragmatic Foundations for a Logic
of Norms, Rechtstheorie, 15, 1984, 453464.
[8] Alchourron, C. E. and Bulygin, E., On the Logic of Normative Systems, in: Stachowiak, H. (ed.), Handbuch pragmatischen Denkens,
Hamburg: Meiner, 1993, 273293.
[9] Alchourron, C. E. and Makinson, D., Hierarchies of Regulations and
Their Logic, in [94] 125148.
121
BIBLIOGRAPHY
122
[10] Alchourron, C. E. and Martino, A. A., Logic Without Truth, Ratio

Juris, 3, 1990, 4667.
[11] Allen, L. E., Deontic Logic, Modern Uses of Logic in Law, 2, 1960,
1327.
[12] Anderson, A. R., The Formal Analysis of Normative Concepts, Yale
University, Technical Report, 1956, reprinted in Rescher, N. (ed.), The
Logic of Decision and Action, Pittsburgh: University Press, 1967, 147213.
[13] Anderson, A. R., The Logic of Norms, Logique & Analyse, 2, 1958,
8491.
[14] Anderson, A. R., A Reduction of Deontic Logic to Alethic Modal
Logic, Mind, 67, 1958, 100103.
[15] Anderson, A. R., On the Logic of Commitment, Philosophical Studies, 19, 1959, 2327.
[16] Anderson, A. R., The Formal Analysis of Normative Concepts, in:
Rescher, N. (ed.), The Logic of Decision and Action, Pittsburgh: University Press, 1967, 147213, first appeared 1956.
[17] Anderson, A. R., Some Nasty Problems in the Formal Logic of Ethics,
Nous, 1, 1967, 345360.
[18] Anderson, A. R. and Moore, O. K., The Formal Analysis of Normative
Concepts, American Sociological Review, 11, 1957, 917.
[19] Anscombe, G. E. M., Intention, Oxford: Blackwell, 1957.
qvist, L., Some Theorems about a Tree System of Deontic Tense
[20] A
Logic, in [94] 187221.
qvist, L., Interpretations of Deontic Logic, Mind, 73, 1964, 246
[21] A
253.
[22]
Aqvist, L., Deontic Logic, in: Gabbay, D. and Guenthner, F. (eds.),
Handbook of Philosophical Logic. Volume II: Extensions of Classical
Logic, Dordrecht: D. Reidel, 1984, 605714, reprinted in [59] pp. 147
264.
BIBLIOGRAPHY
123
[23]
Aqvist, L., Some Results on Dyadic Deontic Logic and the Logic of
Preference, Synthese, 66, 1986, 95110.
[24] Ayer, A. J., Language, Truth and Logic, New York: Dover Publ., 1938.
[25] Bailhache, P., Several Possible Systems of Deontic Weak and Strong
Norms, Logique & Analyse, 21, 1980, 89100.
[26] Bailhache, P., Essai de logique deontique,
Philosophique J. Vrin, 1991.
Paris:
Librairie
[27] Barcan Marcus, R., Iterated Deontic Modalities, Mind, 75, 1966,
580582.
[28] Becker, O., Untersuchungen u
ul, Meisenheim/Glan:
ber den Modalkalk
Anton Hain, 1952.
[29] Bentham, J., Of Laws in General, London: Athlone Press, 1970, edited
by H. L. A. Hart. First appeared 1782.
[30] Bochman, A., Credulous Nonmonotonic Inference, in: Dean, T.
(ed.), Proceedings of the 16th International Joint Conference on Artificial Intelligence, IJCAI 99, Stockholm, August 1999, San Francisco:
Morgan Kaufmann, 1999, 30 35.
[31] Bochman, A., A Logical Theory of Nonmonotonic Inference and Belief
Change, Berlin: Springer, 2001.
[32] Bochman, A., Brave Nonmonotonic Inference and Its Kinds, Annals
of Mathematics and Artificial Intelligence, 39, 2003, 101121.
[33] Bohnert, H. G., The Semiotic Status of Commands, Philosophy of
Science, 12, 1945, 302315.
[34] Boole, G., The Mathematical Analysis of Logic, Cambridge: Macmillan, Barclay, and Machmillan, 1847.
[35] Brewka, G., Preferred Subtheories: An Extended Logical Framework
for Default Reasoning, in: Sridharan, N. S. (ed.), Proceedings of
the Eleventh International Joint Conference on Artificial Intelligence
IJCAI-89, Detroit, Michigan, USA, August 20-25, 1989, San Mateo,
Calif.: Kaufmann, 1989, 10431048.
BIBLIOGRAPHY
124
[36] Brewka, G., Bevorzugte Teiltheorien: Wissensrevision in einem

Ansatz zum Default-Schlieen, Kognitionswissenschaft, 1, 1990, 27
35.
[37] Brewka, G., Reasoning about Priorities in Default Logic, in: HayesRoth, B. and Korf, R. E. (eds.), Proceedings of the 12th National Conference on Artificial Intelligence, Seattle, WA, July 31st - August 4th,
1994, vol. 2, Menlo Park: AAAI Press, 1994, 940945.
[38] Brewka, G. and Eiter, T., Preferred Answer Sets for Extended Logic
Programs, Artificial Intelligence, 109, 1999, 297356.
[39] Bulygin, E., Existence of Norms, 1999, in [149] 237-244.
[40] Carmo, J. and Jones, A. J. I., Deontic Logic and Contrary-to-duties,
in [59] 265-343.
[41] Casta
neda, H.-N., The Paradoxes of Deontic Logic: The Simplest
Solution to All of Them in One Fell Swoop, in [94] 3785.
[42] Casta
neda, H.-N., Obligation and Modal Logic, Logique & Analyse,
3, 1960, 4049.
[43] Casta
neda, H.-N., There Are Command sh-Inferences, Analysis, 32,
1971, 1319.
[44] Chellas, B. F., The Logical Form of Imperatives, Stanford: Perry Lane
Press, 1969.
[45] Chellas, B. F., Modal Logic, Cambridge/New York: Cambridge University Press, 1980.
[46] Chisholm, R. M., Contrary-to-Duty Imperatives and Deontic Logic,
Analysis, 24, 1963, 3336.
[47] Danielsson, S., Taking Rosss Paradox Seriously. A Note on the Original Problems of Deontic Logic, Theoria, 71, 2005, 2028.
[48] Downing, P., Opposite Conditionals and Deontic Logic, Mind, 63,
1959, 491502.
[49] Dubislav, W., Die Definition, 3rd ed., Leipzig: Meiner, 1931.
BIBLIOGRAPHY
125
[50] Dubislav, W., Zur Unbegr

undbarkeit der Forderungssatze, Theoria,
3, 1938, 330342.
[51] van Eck, J., A System of Temporally Relative Modal and Deontic
Predicate Logic and Its Philosophical Applications, Logique & Analyse, 25, 1982, 249290, 339381.
[52] Englis, K., Die Norm ist kein Urteil, Archiv f
ur Rechts- und Sozialphilosophie, 50, 1964, 305316.
[53] Fllesdal, D. and Hilpinen, R., Deontic Logic: An Introduction, in
[93] 135.
[54] Foot, P., Moral Realism and Moral Dilemma, Journal of Philosophy,
80, 1983, 379389.
[55] Forget, L., Risch, V. and Siegel, P., Preferential Logics are X-logics,
Journal of Logic and Computation, 11, 2001, 7183.
[56] Forrester, J. W., Gentle Murder, or the Adverbial Samaritan, Journal of Philosophy, 81, 1984, 193197.
[57] van Fraassen, B., Values and the Hearts Command, Journal of Philosophy, 70, 1973, 519.
[58] Frey, G., Idee einer Wissenschaftslogik. Grundz
uge einer Logik imperativer Satze, Philosophia Naturalis, 4, 1957, 434491.
[59] Gabbay, D. M. and Guenthner, F. (eds.), Handbook of Philosophical
Logic, vol. 8, 2nd ed., Dordrecht: Kluwer, 2002.
[60] Geach, P. T., Imperative and Deontic Logic, Analysis, 18, 1958,
4956.
[61] Geach, P. T., Imperative Inference, Analysis, 23, 1963, 3742.
[62] Geach, P. T., Whatever Happened to Deontic Logic?, Philosophia
(Israel), 11, 1982, 113.
[63] Gentzen, G., Untersuchungen u
ber das logische Schlieen, Mathematische Zeitschrift, 39, 1934, 176210; 405431.
BIBLIOGRAPHY
126
[64] Goble, L., The Iteration of Deontic Modalities, Logique & Analyse,
9, 1966, 197209.
[65] Goble, L., A Logic of Good, Should, and Would, Journal of Philosophical Logic, 19, 1990, 169199.
[66] Goble, L., Preference Semantics for Deontic Logics: Part I Simple
Models, Logique & Analyse, 2003, 383418.
[67] Goble, L., Preference Semantics for Deontic Logics: Part II Multiplex Models, Logique & Analyse, 2004, 335363.
[68] Goble, L., A Proposal for Dealing with Deontic Dilemmas, in: Lomuscio, A. and Nute, D. (eds.), Deontic Logic in Computer Science
(Proceedings of the 7th International Workshop on Deontic Logic in
Computer Science, DEON 2004, Madeira, May 2004), LNAI 3065,
Berlin: Springer, 2004, 75 113.
[69] Goble, L., A Logic for Deontic Dilemmas, Journal of Applied Logic,
3, 2005, 461483.
[70] Goble, L. and Lokhorst, G.-J. C., Mallys Deontik, Grazer
Philosophische Studien, 67, 2004, 3757.
[71] Gombay, A., What is Imperative Inference?, Analysis, 27, 1967, 145
152.
[72] Greenspan, P., Conditional Oughts and Hypothetical Imperatives,
Journal of Philosophy, 72, 1975, 259276.
[73] Grelling, K., Zur Logik der Sollsatze, Unity of Science Forum, 1939,
4447.
[74] Hamblin, C. L., Imperatives, Oxford: Blackwell, 1987.
[75] Hansen, J., Sets, Sentences, and Some Logics about Imperatives,
Fundamenta Informaticae, 48, 2001, 205226.
[76] Hansen, J., Problems and Results for Logics about Imperatives,
Journal of Applied Logic, 2, 2004, 3961.
BIBLIOGRAPHY
127
[77] Hansen, J., Conflicting Imperatives and Dyadic Deontic Logic, Journal of Applied Logic, 3, 2005, 484511.
[78] Hansen, J., Deontic Logics for Prioritized Imperatives, Artificial Intelligence and Law, 14, 2006, 134.
[79] Hansen, J., The Paradoxes of Deontic Logic: Alive and Kicking,
Theoria, 72, 2006, 221232.
[80] Hansen, J., Prioritized Conditional Imperatives: Problems and a New
Proposal, Autonomous Agents and Multi-Agent Systems, 17, 2008,
1135.
[81] Hansen, J., Pigozzi, G. and van der Torre, L., Ten Philosophical Problems in Deontic Logic, in: Boella, G., van der Torre, L. and Verhagen,
H. (eds.), Normative Multi-Agent Systems, no. 07122 in Dagstuhl Seminar Proceedings, Internationales Begegnungs- und Forschungszentrum
f
ur Informatik (IBFI), Schloss Dagstuhl, Germany, 2007.
[82] Hanson, W. H., Semantics for Deontic Logic, Logique & Analyse, 31,
1965, 177190.
[83] Hanson, W. H., A Logic of Commands, Logique & Analyse, 9, 1966,
329343.
[84] Hansson, B., An Analysis of Some Deontic Logics, Nous, 3, 1969,
373398, reprinted in [93] 121147.
[85] Hansson, S. O., Preference-based Deontic Logic (PDL), Journal of
Philosophical Logic, 19, 1990, 7593.
[86] Hare, R. M., Imperative Sentences, Mind, 58, 1949, 2139.
[87] Hare, R. M., The Language of Morals, Oxford: Oxford University
Press, 1952.
[88] Hare, R. M., Some Alleged Differences Between Imperatives and Indicatives, Mind, 76, 1967, 309326.
[89] Harris, Z. S., Transformational Theory, Language, 41, 1965, 363
401.
BIBLIOGRAPHY
128
[90] Harrison, J., Deontic Logic and Imperative Logic, in: Geach, P. T.
(ed.), Logic and Ethics, Dordrecht: Kluwer, 1991, 79129.
[91] Hempel, C. G., Review of Alf Ross: Imperatives and Logic, Journal
of Symbolic Logic, 6, 1941, 105106.
[92] Hilpinen, R., On Deontic Logic, Pragmatics and Modality, in: Stachowiak, H. (ed.), Pragmatik: Handbuch des pragmatischen Denkens,
vol. 4, Hamburg: Felix Meiner.
[93] Hilpinen, R. (ed.), Deontic Logic: Introductory and Systematic Readings, Dordrecht: Reidel, 1971.
[94] Hilpinen, R. (ed.), New Studies in Deontic Logic, Dordrecht: Reidel,
1981.
[95] Hintikka, J., Some Main Problems of Deontic Logic, in [93] 59104.
[96] Hintikka, J., The Ross Paradox as Evidence for the Reality of Semantical Games, in: Saarinen, E. (ed.), Game-Theoretical Semantics,
Dordrecht: Reidel, 1977, 329345.
[97] Hintikka, J. K., Quantifiers in Deontic Logic, Helsingfors: Societas
Scientiarum Fennica, Commentationes Humanarum Litterarum 23:4,
1957.
[98] Hodges, W., Logic, Harmondsworth: Penguin, 1977.
[99] Hofstadter, A. and McKinsey, J. C. C., On the Logic of Imperatives,
Philosophy of Science, 6, 1938, 446457.
[100] Hollander, P., Rechtsnorm, Logik und Wahrheitswerte. Versuch
einer kritischen Losung des Jorgensenschen Dilemmas, Baden-Baden:
Nomos, 1993.
[101] Horty, J. F., Moral Dilemmas and Nonmonotonic Logic, Journal of
Philosophical Logic, 23, 1994, 3565.
[102] Horty, J. F., Nonmonotonic Foundations for Deontic Logic, in: Nute,
D. (ed.), Defeasible Deontic Logic, Dordrecht: Kluwer, 1997, 1744.
BIBLIOGRAPHY
129
[103] Horty, J. F., Agency and Deontic Logic, Oxford: Oxford University
Press, 2000.
[104] Horty, J. F., Reasoning with Moral Conflicts, No
us, 37, 2003, 557
605.
[105] Horty, J. F., Defaults with Priorities, Journal of Philosophical Logic,
36, 2007, 367413.
[106] Hruschka, J., Das deontologische Sechseck bei Gottfried Achenwall im
Jahre 1767, Gottingen: Vandenhoek & Ruprecht, 1986.
[107] Hume, D., A Treatise of Human Nature, Oxford: Oxford University
Press, 1888, edited by L. A. Selby-Bigge. First appeared 1739.
[108] Husserl, E., Logische Untersuchungen, vol. 1, Halle: Max Niemeyer,
1900.
[109] Jacquette, D., Moral Dilemmas, Disjunctive Obligations, and Kants
Principle that Ought implies Can, Synthese, 88, 1991, 4355.
[110] Jrgensen, J., Imperatives and Logic, Erkenntnis, 7, 1938, 288296.
[111] Kalinowski, G., Le Problème de la Verite en Morale et en Droit, Lyon:
Emmanuel Vitte, 1967.
[112] Kalinowski, G., Einf
uhrung in die Normenlogik, Frankfurt/M.:
Athenaum, 1973.
[113] Kalinowski, G., Uber

die deontischen Funktoren, in: Lenk, H. (ed.),
Normenlogik. Grundprobleme der deontischen Logik, Pullach: Dokumentation, 1974, 3963.
[114] Kalinowski, G., Uber

die Bedeutung der Deontik f
ur Ethik und Rechtsphilosophie, in: Conte, A. G., Hilpinen, R. and von Wright, G. H.
(eds.), Deontische Logik und Semantik, Wiesbaden: Athenaion, 1977,
101129.
[115] Kalinowski, G., Zur Semantik der Rechtssprache, in: Krawietz, W.,
Opalek, K., Peczenik, A. and Schramm, A. (eds.), Argumentation und
Hermeneutik in der Jurisprudenz, vol. 1 of Rechtstheorie, Beihefte,
Berlin: Duncker & Humblot, 1979, 239252.
BIBLIOGRAPHY
130
[116] Kamp, G., Logik und Deontik, Paderborn: Mentis, 2001.

[117] Kamp, H., Free Choice Permission, Proceedings of the Aristotelian
Society, 74, 1973/74, 5774.
[118] Kamp, H., Semantics versus Pragmatics, in: Guenthner, F. and
Schmidt, S. J. (eds.), Formal Semantics and Pragmatics for Natural
Languages, Dordrecht: Reidel, 1978, 255287.
[119] Kanger, S., New Foundations for Ethical Theory: Part 1, duplic., 42
p., 1957, reprinted in [93] 3658.
[120] Keene, G. B., Can Commands Have Logical Consequences?, American Philosophical Quarterly, 3, 1966, 5763.
[121] Kelsen, H., Allgemeine Theorie der Normen, Wien: Manz, 1979.
[122] Kenny, A. J., Practical Inference, Analysis, 26, 1966, 6575.
[123] Keuth, H., Deontische Logik und Logik der Normen, in: Lenk, H.
(ed.), Normenlogik. Grundprobleme der deontischen Logik, Pullach:
Dokumentation, 1974, 6486.
[124] Knuuttila, S., The Emergence of Deontic Logic in the Fourteenth
Century, in [94] 225248.
[125] Kratzer, A., Conditional Necessity and Possibility, in: Bauerle, R.,
Egli, U. and von Stechow, A. (eds.), Semantics from Different Points
of View, Berlin: Springer, 1979, 116147.
[126] Kraus, S., Lehmann, D. and Magidor, M., Nonmonotonic Reasoning,
Preferential Models and Cumulative Logics, Artificial Intelligence, 44,
1990, 167207.
[127] Kripke, S. A., Semantical Analysis of Modal Logic I: Normal Modal
Propositional Calculi, Zeitschrift f
ur mathematische Logik und Grundlagen der Mathematik, 9, 1963, 6796.
[128] von Kutschera, F., Einf
uhrung in die Logik der Normen, Werte und
Entscheidungen, Freiburg/M
unchen: Karl Alber, 1973.
BIBLIOGRAPHY
131
[129] Ledig, G., Zur Klarung einiger Grundbegriffe. Imperativ, Rat, Bitte,
Beschlu, Versprechen, Revue Internationale de la Theorie du Droit,
3, 1928/29, 260270.
[130] Ledig, G., Zur Logik des Sollens, Der Gerichtssaal, 100, 1931, 368
385.
[131] Leibniz, G. W., Elementa Juris Naturalis [2nd half 1671, holographic
concept B (Hannover)], in: Preussische Akademie der Wissenschaften
(ed.), Gottfried Wilhelm Leibniz: Samtliche Schriften und Briefe, vol. 1
of 6. Reihe: Philosophische Schriften, Darmstadt: Otto Reichl, 1930,
465485.
[132] Lemmon, E. J., Moral Dilemmas, Philosophical Review, 71, 1962,
139158.
[133] Lemmon, E. J., Deontic Logic and the Logic of Imperatives, Logique
& Analyse, 8, 1965, 3971.
[134] Lemmon, E. J., Beginning Logic, 2nd ed., London: Chapman and Hall,
1987.
[135] Leonard, H. S., Interrogatives, Imperatives, Truth, Falsity and Lies,
Philosophy of Science, 26, 1959, 172186.
[136] Lewis, D., Counterfactuals, Oxford: Basil Blackwell, 1973.
[137] Lewis, D., Semantic Analyses for Dyadic Deontic Logic, in: Stenlund,
S. (ed.), Logical Theory and Semantic Analysis, Dordrecht: Reidel,
1974, 1 14.
[138] Lewis, D., Ordering Semantics and Premise Semantics for Counterfactuals, Journal of Philosophical Logic, 10, 1981, 217234.
[139] MacKay, A. F., Inferential Validity and Imperative Inference Rules,
Analysis, 29, 1969, 145156.
[140] Makinson, D., General Patterns in Nonmonotonic Reasoning, in:
Gabbay, D. M., Hogger, C. and Robinson, J. (eds.), Handbook of Logic
in Artificial Intelligence and Logic Programming, vol. 3: Nonmonotonic
Reasoning and Uncertain Reasoning, Oxford: Clarendon Press, 1994,
35 110.
BIBLIOGRAPHY
132
[141] Makinson, D., On a Fundamental Problem of Deontic Logic, in: McNamara, P. and Prakken, H. (eds.), Norms, Logics and Information
Systems, Amsterdam: IOS, 1999, 2953.
[142] Makinson, D. and van der Torre, L., Input/Output Logics, Journal
of Philosophical Logic, 29, 2000, 383408.
[143] Makinson, D. and van der Torre, L., Constraints for Input/Output
Logics, Journal of Philosophical Logic, 30, 2001, 155185.
[144] Makinson, D. and van der Torre, L., What is Input/Output Logic,
in: Lowe, B., Malzkom, W. and Rasch, T. (eds.), Foundations of the
Formal Sciences II : Applications of Mathematical Logic in Philosophy
and Linguistics (Papers of a conference held in Bonn, November 10-13,
2000), Trends in Logic, vol. 17, Dordrecht: Kluwer, 2003, 163174.
[145] Mally, E., Grundgesetze des Sollens. Elemente der Logik des Willens,
Graz: Leuschner & Lubensky, 1926.
[146] Marek, V. W. and Truszczy
nski, M., Nonmonotonic Logic. ContextDependent Reasoning, Berlin: Springer, 1993.
[147] Mates, B., Elementary Logic, 2nd ed., Oxford: Oxford University Press,
1972.
[148] Mazzarese, T., Norm Propositions:
Queries, Rechtstheorie, 22, 1991, 3970.
Epistemic and Semantic
[149] Meggle, G. (ed.), Actions, Norms, Values. Discussions with Georg Henrik von Wright, Berlin: de Gruyter, 1999.
[150] Menger, K., A Logic of the Doubtful. On Optative and Imperative Logic, Reports of a Mathematical Colloquium, 2, 1939, 5364,
reprinted in [151] 91102.
[151] Menger, K., Selected Papers in Logic and Foundations, Didactics, Economics, Dordrecht: Reidel, 1979.
[152] Meyer, J.-J. C., A Different Approach to Deontic Logic: Deontic Logic
Viewed as a Variant of Dynamic Logic, Notre Dame Journal of Formal
Logic, 29, 1988, 109136.
BIBLIOGRAPHY
133
[153] Moritz, M., Der praktische Syllogismus und das juridische Denken,
Theoria, 20, 1954, 78127.
[154] Mott, P. L., On Chishoms Paradox, Journal of Philosophical Logic,
2, 1973, 197211.
[155] Moutafakis, N. J., Imperatives and their Logic, New Delhi: Sterling,
1975.
[156] Niiniluoto, I., Truth and Legal Norms, Archiv f
ur Rechts- und Sozialphilosophie Beiheft, 25, 1985, 168190.
[157] Niiniluoto, I., Hypothetical Imperatives and Conditional Obligations, Synthese, 66, 1986, 111133.
[158] Niiniluoto, I., Norm Propositions Defended, Ratio Juris, 4, 1991,
367373.
[159] Nortmann, U., Deontische Logik ohne Paradoxien. Semantik und Logik
des Normativen, M
unchen: Philosophia Verlag, 1989.
[160] Nowell-Smith, P. H. and Lemmon, E. J., Escapism: the Logical Basis
of Ethics, Mind, 69, 1960, 289300.
[161] Nute, D. and Yu, X., Defeasible Deontic Logic. Introduction, in:
Nute, D. (ed.), Defeasible Deontic Logic, Dordrecht: Kluwer Academic
Publishers, 1997, 116.
[162] Opalek, K., On the Logical-Semantic Structure of Directives, Logique
& Analyse, 13, 1970, 169196.
[163] Opalek, K., Theorie der Direktiven und Normen, Wien: Springer, 1986.
[164] Opalek, K. and Wole
nski, J., Is, Ought, and Logic, Archiv f
ur Rechtsund Sozialphilosophie, 73, 1987, 373385.
[165] Opalek, K. and Wole
nski, J., Normative Systems, Permission and
Deontic Logic, Ratio Juris, 4, 1991, 334348.
[166] Peczenik, A., Doctrinal Study of Law and Science, Osterreichische

Zeitschrift f
ur offentliches Recht, 17, 1967, 128141.
BIBLIOGRAPHY
134
[167] Peczenik, A., Norms and Reality, Theoria, 34, 1968, 117133.
[168] Philipp, P., Logik deskriptiver normativer Begriffe, 1989/90, In [170]
pp. 241290. First published as reports of the Karl-Marx-Universitat
Leipzig, Untersuchungen zur Logik und zur Methodologie 6:6588, 1989,
and 7:5182, 1990.
[169] Philipp, P., Normative Logic Without Norms, 1991, In [170] pp. 291
301. Edited version of a presentation to the Meeting of the Central
Division of the American Philosophical Association, Chicago, 1991.
[170] Philipp, P., Logisch-philosophische Untersuchungen. Edited by Ingolf
Max and Richard Raatzsch., Berlin: de Gruyter, 1998.
[171] Philipps, L., Braucht die Rechtswissenschaft eine deontische Logik?,
in: G., J. and Maihofer, W. (eds.), Rechtstheorie. Beitrage zur Grundlagendiskkussion, Frankfurt: Klostermann, 1971, 352368.
[172] Poincare, H., Dernières Pensees, Paris: Ernest Flammarion, 1913.
[173] Poole, D., A Logical Framework for Default Reasoning, Artificial
Intelligence, 36, 1988, 2747.
[174] Powers, L., Some Deontic Logicians, Nous, 1, 1967, 380400.
[175] Prakken, H., Logical Tools for Modelling Legal Argument, Dordrecht:
Kluwer, 1997.
[176] Prakken, H. and Sergot, M., Contrary-to-duty obligations., Studia
Logica, 52, 1996, 91115.
[177] Prior, A. N., The Paradoxes of Derived Obligation, Mind, 63, 1954,
6465.
[178] Prior, A. N., Formal Logic, Oxford: Clarendon Press, 1955.
[179] Prior, A. N., Escapism: the Logical Basis of Ethics, in: Melden, A. I.
(ed.), Essays in Moral Philosophy, Seattle: University of Washington
Press, 1966, 135146.
[180] Reichenbach, H., Elements of Symbolic Logic, New York: CollierMacmillan, 1947.
BIBLIOGRAPHY
135
[181] Reiter, R., A Logic for Default Reasoning, Artificial Intelligence, 13,
1980, 81132.
[182] Rescher, N., An Axiom System for Deontic Logic, Philosophical Studies, 9, 1958, 2430.
[183] Rescher, N., Hypothetical Reasoning, Amsterdam: North-Holland,
1964.
[184] Rescher, N., The Logic of Commands, London: Routledge & Kegan
Paul, 1966.
[185] Rescher, N., Assertion Logic, in: Rescher, N. (ed.), Topics in Philosophical Logic, chap. XIV, Dordrecht: Reidel, 1968.
[186] Rodig, J., Kritik des Normlogischen Schlieens, Theory and Decision,
2, 1971, 7993.
[187] Rodig, J., Uber

die Notwendigkeit einer besonderen Logik der Normen, Jahrbuch f
ur Rechtssoziologie und Rechtstheorie, 2, 1972, 163
185.
[188] Ross, A., Imperatives and Logic, Theoria, 7, 1941, 5371, Reprinted
with only editorial changes in Philosophy of Science 11:3046, 1944.
[189] Ross, W. D., The Right and the Good, Oxford: Clarendon Press, 1930.
[190] Sartor, G., Inconsistency and Legal Reasoning, in: Breuker, J. A.,
De Mulder, R. V. and Hage, J. C. (eds.), Legal Knowledge Based Systems: Model-based Legal Reasoning (Proceedings of the Fourth International Conference on Legal Knowledge Based Systems JURIX 1991,
Lelystad, December 1991), Lelystad: Koninklijke Vermande, 1991, 92
112.
[191] Sayre-McCord, G., Deontic Logic and the Priority of Moral Theory,
Nous, 20, 1986, 179197.
[192] Schotch, P. K. and Jennings, R. E., Non-Kripkean Deontic Logic, in
[94] 149162.
[193] Sellars, W., Imperatives, Intentions and the Logic of Ought,
Methodos, 8, 1956, 227268.
BIBLIOGRAPHY
136
[194] Siegel, P. and Forget, L., A Representation Theorem for Preferential

Logics, in: Aiello, L. C., Doyle, J. and Shapiro, S. C. (eds.), Principles
of Knowledge Representation and Reasoning (Proceedings of the 5th
International Conference, KR 96), San Francisco: Morgan Kaufmann,
1996, 453460.
[195] Sigwart, C., Logic, vol. I. The Judgment, Concept, and Inference, 2nd
ed., London: Swan Sonnenschein & Co., 1895.
[196] Simon, H. A., The Logic of Rational Decision, The British Journal
for the Philosophy of Science, 16, 1965, 169186.
[197] Sinowjew, A. A., Komplexe Logik, Berlin: VEB Deutscher Verlag der
Wissenschaften, 1970.
[198] Smiley, T. J., The Logical Basis of Ethics, Acta Philosophica Fennica,
16, 1963, 237246.
[199] Solt, K., Deontic Alternative Worlds and the Truth-Value of OA,
Logique & Analyse, 27, 1984, 349351.
[200] Sosa, E., The Logic of Imperatives, Theoria, 32, 1966, 224235.
[201] Sosa, E., On Practical Inference and the Logic of Imperatives, Theoria, 32, 1966, 211223.
[202] Sosa, E., The Semantics of Imperatives, American Philosophical
Quarterly, 4, 1967, 5764.
[203] Spohn, W., An Analysis of Hanssons Dyadic Deontic Logic, Journal
of Philosophical Logic, 4, 1975, 237252.
[204] Stekeler-Weithofer, P., Grundprobleme der Logik, Berlin/New York:
Walter de Gruyter, 1986.
[205] Stenius, E., The Principles of a Logic of Normative Systems, Acta
Philosophica Fennica, 16, 1963, 247260.
[206] Stenius, E., Ross Paradox and Well-Formed Codices, Theoria, 48,
1982, 4977.
BIBLIOGRAPHY
137
[207] Stranzinger, R., Ein paradoxienfreies deontisches System, in: Tammelo, I. and Schreiner, H. (eds.), Strukturierungen und Entscheidungen
im Rechtsdenken, Wien: Springer, 1978, 183193.
[208] Swirydowicz,
K., Normative Consequence Relation and Consequence
Operations on the Language of Dyadic Deontic Logic, Theoria, 60,
1994, 2747.
[209] Swirydowicz,
K., A New Approach to Dyadic Deontic Logic and the
Normative Consequence Relation, in: Murawski, R. and Pogonowski,
J. (eds.), Euphony and Logos. Essays in Honour of Maria Steffen-Batog
and Tadeusz Batog, Amsterdam/Atlanta: Rodopi, 1997, 283317.
[210] Tarski, A., Uber

einige fundamentale Begriffe der Metamathematik,
Comptes Rendus des Seances de la Societe des Sciences et des Lettres
de Varsovie, 23, 1930, 2229, published in English under the title On
Some Fundamental Concepts of Metamathematics in [212] pp. 3037.
[211] Tarski, A., Der Wahrheitsbegriff in den formalisierten Sprachen, Studia Philosophica, I, 1935, 261405.
[212] Tarski, A., Logic, Semantics, Metamathematics, Oxford: Oxford University Press, 1956.
[213] van der Torre, L., Reasoning About Obligations, Amsterdam: Thesis
Publ., 1997.
[214] van der Torre, L. and Tan, Y.-H., Two-Phase Deontic Logic, Logique
& Analyse, 43, 2000, 411456.
[215] Urmson, J. O., Saints and Heroes, in: Melden, A. I. (ed.), Essays
in Moral Philosophy, Seattle: University of Washington Press, 1958,
198216.
[216] Wagner, H. and Haag, K., Die moderne Logik in der Rechtswissenschaft, Bad Homburg: Gehlen, 1970.
[217] Walter, R., Jorgensens Dilemma and How to Face It, Ratio Juris,
9, 1996, 168171.
BIBLIOGRAPHY
138
[218] Walter, R., Some Thoughts on Peczeniks Replies to Jorgensens

Dilemma and How to Face It (with Two Letters by A. Peczenik),
Ratio Juris, 10, 1997, 392396.
[219] Wansing, H., Deontic Modalities: Another View of Parking on Highways, Erkenntnis, 49, 1998, 185199.
[220] Watts, I., Logick: or, the Right Use of Reason in the Inquiry after
Truth, with a Variety of Rules to Guard against Error in the Affairs
of Religion and Human Life, as well as in the Sciences, London: John
Clark, Richard Hett et al., 1725, republished at Whitefish, Montana:
Kessinger Publishing, 2004.
[221] Wedeking, G. A., Are There Command Arguments, Analysis, 30,
1970, 161166.
[222] Weinberger, O., Uber

die Negation von Sollsatzen, Theoria, 23, 1957,
102132.
[223] Weinberger, O., Die Sollsatzproblematik in der modernen Logik:
Konnen Sollsatze (Imperative) als wahr bezeichnet werden?, Prague:
Nakladatelstv Ceskoslovenske Akademie Ved, 1958, page numbers refer to the reprint in [227] pp. 59186.
[224] Weinberger, O., Bemerkungen zur Grundlegung der Theorie des juristischen Denkens, Jahrbuch f
ur Rechtssoziologie und Rechtstheorie,
2, 1972, 134161.
[225] Weinberger, O., Der Begriff der Nicht-Erf
ullung und die Normenlogik, Ratio, 14, 1972, 1532.
[226] Weinberger, O., Ideen zur logischen Normensemantik, in: Haller,
R. (ed.), Jenseits von Sein und Nichtsein. Beitrage zur MeinongForschung, Graz: Akademische Druck- und Verlagsanstalt, 1974, 295
311, Reprinted in [227] pp. 259277.
[227] Weinberger, O., Studien zur Normlogik und Rechtsinformatik, Berlin:
Schweitzer, 1974.
[228] Weinberger, O., Ex falso quodlibet in der deskriptiven und praskriptiven Sprache, Rechtstheorie, 6, 1975, 1732.
BIBLIOGRAPHY
139
[229] Weinberger, O., Normentheorie als Grundlage der Jurisprudenz und

Ethik, Berlin: Duncker&Humblot, 1981.
[230] Weinberger, O., Der normenlogische Skeptizismus, Rechtstheorie, 17,
1986, 1381, reprinted in [233] pp. 431499.
[231] Weinberger, O., Rechtslogik, 2nd ed., Berlin: Duncker&Humblot, 1989.
[232] Weinberger, O., The Logic of Norms Founded on Descriptive Language, Ratio Juris, 4, 1991, 284307.
[233] Weinberger, O., Moral und Vernunft, Wien: Bohlau, 1992.
[234] Weinberger, O., Alternative Handlungstheorie, Wien: Bohlau, 1996.
[235] Weinberger, O., Logical Analysis in the Realm of Law, 1999, in [149]
291-304.
[236] Weingartner, P., Reasons from Science for Limiting Classical Logic,
in: Weingartner, P. (ed.), Springer, Berlin, 2003, 233248.
[237] Weingartner, P. and Schurz, G., Paradoxes Solved by Simple Relevance Criteria, Logique & Analyse, 29, 1986, 340.
[238] Williams, B. A. O., Imperative Inference, Analysis, 23, 1963, 3036.
[239] Wittgenstein, L., Philosophical Investigations, Oxford:
1953.
Blackwell,
[240] Wole
nski, J., Deontic Logic and Possible Worlds Semantics: A Historical Sketch, Studia Logica, 49, 1990, 273282.
[241] Wole
nski, J., Remarks on Mallys Deontik, in: Hieke, A. (ed.), Ernst
Mally Versuch einer Neubewertung, St. Augustin: Akademia, 1998,
7380.
[242] von Wright, G. H., Deontic Logic, Mind, 60, 1951, 115.
[243] von Wright, G. H., An Essay in Modal Logic, Amsterdam: North Holland, 1951.
[244] von Wright, G. H., A Note on Deontic Logic and Derived Obligation,
Mind, 65, 1956, 507509.
BIBLIOGRAPHY
140
[245] von Wright, G. H., Logical Studies, London: Routledge and Kegan
Paul, 1957.
[246] von Wright, G. H., Norm and Action, London: Routledge & Kegan
Paul, 1963.
[247] von Wright, G. H., A New System of Deontic Logic, Danish Yearbook
of Philosophy, 1, 1964, 173182, reprinted in [93] 105115.
[248] von Wright, G. H., A Correction to a New System of Deontic Logic,
Danish Yearbook of Philosophy, 2, 1965, 103107, reprinted in [93] 115
120.
[249] von Wright, G. H., An Essay in Deontic Logic and the General Theory
of Action, Amsterdam: North Holland, 1968.
[250] von Wright, G. H., Normenlogik, in: Lenk, H. (ed.), Normenlogik.
Grundprobleme der deontischen Logik, Pullach b. M
unchen: Verlag
Dokumentation, 1974, 25 38.
[251] von Wright, G. H., Problems and Prospects of Deontic Logic, in:
Agazzi, E. (ed.), Modern Logic: A Survey, Dordrecht: D. Reidel, 1980,
399423.
[252] von Wright, G. H., Norms, Truth and Logic, in: von Wright, G. H.
(ed.), Practical Reason: Philosophical Papers vol. I, Oxford: Blackwell,
1983, 130209.
[253] von Wright, G. H., Bedingungsnormen, ein Pr
ufstein f
ur die Normenlogik, in: Krawietz, W., Schelsky, H., Weinberger, O. and Winkler,
G. (eds.), Theorie der Normen, Berlin: Duncker & Humblot, 1984,
447456.
[254] von Wright, G. H., Is and Ought, in: Bulygin, E., Gardies, J.-L. and
Niiniluoto, I. (eds.), Man, Law and Modern Forms of Life, Dordrecht:
D. Reidel, 1985, 263281.
[255] von Wright, G. H., Is there a Logic of Norms?, Ratio Juris, 4, 1991,
265283.
[256] von Wright, G. H., A Pilgrims Progress, in: von Wright, G. H. (ed.),
The Tree of Knowledge and Other Essays, Leiden: Brill, 1993, 103113.
BIBLIOGRAPHY
141
[257] Yoshino, H., Uber

die Notwendigkeit einer besonderen Normenlogik
als Methode der juristischen Logik, in: Klug, U. (ed.), Gesetzgebungstheorie, Juristische Logik, Zivil- und Prozessrecht. Gedachtnisschrift
f
ur J
urgen Rodig, Berlin: Springer, 1978, 141161.
[258] Ziemba, Z., Deontic Syllogistics, Studia Logica, 28, 1971, 139159.
[259] Ziemba, Z., Deontic Logic, 1976, Appendix in [260], pp. 360430.
[260] Ziembi
nski, Z., Practical Logic, Dordrecht: Reidel, 1976.
Appendix: Research Papers

Sets, Sentences, and Some Logics about Imperatives, Fundamenta Informaticae, 48, 2001, 205226.
Problems and Results for Logics about Imperatives, Journal of Applied
Logic, 2, 2004, 39-61.
Conflicting Imperatives and Dyadic Deontic Logic, Journal of Applied
Logic, 3, 2005, 484511.
Deontic Logics for Prioritized Imperatives, Artificial Intelligence and Law,
14, 2006, 134.
Prioritized Conditional Imperatives: Problems and a New Proposal, Autonomous Agents and Multi-Agent Systems, 2007, demnachst. Online verf
ugbar unter http://www.springerlink.com/content/p253015562l43212.
142
APPENDIX: RESEARCH PAPERS
143
Fundamenta Informaticae 48 (2001) 205226
205
IOS Press
Sets, Sentences, and Some Logics about Imperatives

Jorg Hansen
Institut fur Philosophie

Universitat Leipzig
Leipzig, Germany
jhansen@uni-leipzig.de
Abstract. Though deontic logic is regarded as the logic of normative reasoning, norms as entities
lacking truth values are usually represented neither in its language nor its semantics. Limiting
ourselves to unconditional imperatives, we propose a concept for their semantic representation and
show that existing systems of monadic and dyadic deontic logic can be reconstructed accordingly.
Keywords: deontic logic, logic of imperatives
1. Norms and Deontic Logic

Deontic logic, so the definition runs, is the logical study of the normative use of language. Its subject
matter is normative concepts, notably those of obligation, prohibition, and permission. To this effect,
the systems of deontic logic employ modal operators , , and , where is read as it ought to be
that or it is obligatory that is done, depending on language and the nature of . While normative
concepts are its subject, it is quite difficult to see how deontic logic relates to norms. Imperative or
permissive expressions such as John, leave the room! and Mary, you may enter now do not describe,
but demand or allow a behaviour on the part of of John and Mary. Being non-descriptive, they cannot
meaningfully be termed true or false. Lacking truth values, these expressions cannot in the usual sense
be premise or conclusion in an inference, be termed consistent or contradictory, or be compounded by
The ideas of this paper were motivated by discussions at the EON 98 conference. The analysis of dyadic deontic logics
considerably gained from Wlodek Rabinowicz critical remarks on an earlier version of this paper, and the paper entirely
profited from Leon van der Torres constant encouragement and well-placed criticism. I am obliged to an anonymous referee
for many helpful comments.
Address for correspondence: Institut fur Philosophie, Universitat Leipzig, Leipzig, Germany

206
144
J. Hansen / Some Logics about Imperatives
truth-functional operators. Hence it seems there cannot be a logic of norms this is Jrgensens dilemma
[28].
Though norms are neither true nor false, one may state that according to the norms, something ought
to be (be done) or is permitted: the statements John ought to leave the room, Mary is permitted to enter, are then true or false descriptions of the normative situation. Such statements are sometimes called
normative statements, as distinguished from norms. To express principles such as

(Conj)
with Boolean operators having truth-functional meaning at all places, deontic logic has resorted to interpreting its formulas , , as representing such normative statements. 1 A possible logic of
normative statements may then reflect logical properties of underlying norms thus logic may have a
wider reach than truth, as von Wright [54] famously stated.
Norms, for the reasons given, are absent in the language of deontic logic. But since the truth of
normative statements depends on a normative situation, it seems that norms may be represented in a
logical semantics that models such truth or falsity. Standard deontic semantics evaluates deontic formulas
with respect to sets of worlds, in which some are ideal or better than others. Deviating from this
standard approach, a semantics that seeks to represent norms has been proposed by several authors,
namely Erik Stenius, Stig Kanger, Carlos Alchourron and Eugenio Bulygin, and Bas van Fraassen.
Stenius [47] uses a set of prescriptively interpreted norm sentences, each norm sentence having the
form , and being a Boolean formula. The following principle restricts :
, then
is then defined as
is true if and only if (iff)

(r-St)
if
and
The truth of descriptively interpreted sentences
(td-St)
Obviously, is a normal modal operator. Note, however, that Stenius closure principle yields Ross
paradox [44], where the imperative Post the letter or burn it! is derived from the imperative Post
the letter!: even the forbidden will satisfy a norm sentence in any set that conforms to (r-St). Also,
it is quite a matter of dispute whether norms contain indicatives 2 , and Stenius does in fact define as
sentence radical, so the formulation of his principle can be criticized, cf. Makinson [34]. Kanger
[29] and Alchourron/Bulygin [2] instead use a set of descriptive sentences, where Kanger thinks of these
sentences as representing a welfare program for the universe of discourse, and to Alchourron/Bulygin
the sentences in their set are the sentences commanded by an authority. Their truth definition for
then reads
is true iff
(td-AB)
is the consequential
closure of . Again, is normal. Ross paradox can be argued not to
arise: yields for any but does not necessarily succeed in satisfying a norm
since is not closed under consequence. Yet Kanger and Alchourron/Bulygin make it appear as if
norms can somehow be replaced by descriptive sentences in normative reasoning an approach termed
reductionist by some philosophers (cf. Moutafakis [37], Hamblin [17]). But sentences are true or false,
1
Cf.Fllesdal/Hilpinen [16]. This already holds for Mallys deontic logic [35], cf. p. 12, where he writes that to interpret !
as the theoretical statement it is a fact that ought to be is the standpoint of (his) theory that describes obligating norms and
their laws.
2
Some believe imperatives contain indicatives as their propositional core (Sosa [45], also cf. Alchourron/Bulygin [3]),
whereas others deny any straight-forward grammatical relation and more cautiously speak of a thematically parallel sentence
(Opaek [39] p. 35).
145
207
norms are not, so semantics that claims to be about norms should describe the gap, and explain how it
can be bridged. And certainly norms have their special properties: the implication that any feature one
may want to semantically model time of promulgation, authority, subject, hierarchical position, etc.
must relate to a sentence which is but one of these features, not the norm itself, seems hardly acceptable.
Van Fraassen [50] has a set of imperatives in force, where to each imperative there corresponds
a class of possible outcomes

in which the imperative is satisfied. The truth of
is evaluated with
respect to single imperatives, where at first the truth definition reads ( is the set of outcomes where
is true, and the set of imperatives in force):
(td-vF1)
is true iff there is an such that

Van Fraassen then introduces the notion of a score relative to an outcome and the set of imperatives
, where the score of the outcome is the subset of imperatives it satisfies. His final truth definition reads
(td-vF2)
is true iff there is a such that for all
where is the score of the outcome among the set of imperatives . Van Fraassens suggestive
paper extends this analysis to conditional imperatives, where these relate to sentences that describe the
opportunity for satisfying the imperative. The idea of a score has been further examined by Horty [26],
also cf. [25] p. 284, and related research on the subject of conditional imperatives was carried out by
Makinson [34].
According to Makinson [34], the fundamental problem of deontic logic is to reconstruct it such
that it is in accord with the position that norms are devoid of truth values. I propose a semantic platform, related to the above semantics, on which such reconstructions can be carried out. The paper is
confined to unconditional obligating norms (imperatives, commands), and will not treat explicit permissions, derogations, general norms, or other types. The basic semantics is given in section 2. In section
3 we provide reconstructions for systems of monadic deontic logic. In section 4 we consider the reduction of deontic logic to alethic modal logic by Anderson [7] and Kanger [29], reconstructing it such that
the Andersonian-Kangerian constants have definite meanings. Since the late Fifties, deontic logics have
employed dyadic operators to formalize conditional obligations. In section 5 we propose reconstructions
for well-known systems of dyadic deontic logics. The results may be viewed as justifying and explaining
the view that deontic logic is a logic of reasoning about, if not of, norms.
2. Imperative Structures
2.1. The Basic Concept
What is needed to model imperatives in a logical semantics? Imperatives have a source, and a recipient,
and some authority must be assumed on the part of the source, of which the recipient is a subject (cf.
Rescher [43] ch. 2). The decisive component of an imperative is held to be its theme of demand: a
state the authority wants to see realized, or an action the subject ought to perform. If this state is realized
or the action performed, then the imperative is satisfied. We can identify this state or action by using a
descriptive sentence which corresponds to the imperative and is true just when the imperative is satisfied.
This sentence will usually be rather complex, e.g. it must spell out whether chance fulfilment or an
attempt suffices, and special attention must be paid to seemingly ordinary connectives and quantifiers.
Hence transforming the exact words of the imperative into indicative mood is just an initial approach.
That there nevertheless is such a correspondence between imperatives and indicatives is even accepted

208
146
by the anti-reductionists (cf. Weinberger [60] p. 69, Moutafakis [37] T3 pp. 155-156, Hamblin [17]
pp. 112, 139/140, 152-153), so we may formulate the following postulate:
(W) To each imperative there corresponds a descriptive sentence that is true if the imperative is satisfied,
and false if it is not-satisfied (violated).
Let an imperative structure therefore be a pair , where is a (possibly empty) set of objects;
they are meant to be imperatives. We make no assumptions about the nature of : it may be the set
of imperatives from a specific source, directed towards a certain subject, belonging to a system, being
valid or in force at a certain time, or derivable by some here unspecified procedure from some other
set, e.g. as individual obligations from more general norms. The function
associates a
sentence in the language
of a basic logic BL with each imperative in . These are meant to be the
descriptive sentences that correspond to imperatives. In the reconstructions, BL will be a propositional
logic defined below. But BL may be any logic that accommodates concepts of truth, semantic validity
of a sentence (we write
), and satisfiability, so we could assume BL to be a logic of act
propositions (cf. Alchourron/Bulygin [3]) or a logic that represents time and intentionality. This is our
basic platform to develop and reconstruct specific deontic logics. However, this platform is open to, and
requires, extensions and restrictions to model further qualities of imperatives.
2.2. Semantic Restrictions

The norm-theoretical assumptions of our basic concept are very poor: all we suppose is that it makes
sense to treat imperatives as members of a set, to which sentences that describe the circumstances in
which an imperative is satisfied can be related. Further assumptions can be made explicit by adding
semantic restrictions to the basic concept. Consider e.g.:
(r-1)
For all :
(r-2)
is -satisfiable
(r-3)
For all :
(r-4)
If
then there exists an such that
Someone taking the view that imperatives express a will, and that a will cannot be directed towards
things that are logically impossible, and so imperatives that demand the impossible cannot exist (cf.
Kelsen [31] p. 171), may want to restrict the semantics of imperative structures by (r-1). Kelsen
[30] p. 136 argued that conflicting norms cannot coexist within the same system. Then (r-2) might be
adequate, demanding that all imperatives must be satisfiable collectively. 3 It has often been assumed
that the nature of imperatives is to direct, to serve as a means to change present facts or prevent unwanted
changes taking place (cf. Jrgensen [28]). (r-3) might then serve to exclude the case that an imperative
demands what is true on logical grounds alone (cf. Weinberger [59] p. 121, Alchourron/Bulygin [4] p.
284). Alchourron/Bulygin [4] also believe that one cannot command without implicitly commanding
all logical consequences of the commands contents, and then (r-4) might be accepted. We have already
noted that Ross paradox is connected with such closure conditions.
The effect of employing semantic restrictions is that imperative structures that do not conform to
these are excluded from analysis. For a general exclusion the grounds must be very good indeed, and
is the set of values of to any member of .
147
209
though the above may appear reasonable, we hesitate to integrate them into the basic concept. Still,
restrictions might be useful if the behaviour of specific imperative structures that satisfy these is to be
examined. In this sense, restrictions are a primary tool to reconstruct specific deontic logics on the basis
of our concept.
2.3. Truth Definitions

We do not predetermine how deontic operators should be defined in a deontic logic that makes use of the
,
as giving information about a specific set of
above semantics. If we consider deontic sentences
imperatives (or norms), then the truth definitions will depend on the questions we want to ask about this
set. For
consider e.g.:
such that
such that
such that
According to (td-1) is true iff is the descriptive sentence correlated to some imperative. An such
that is defined true by (td-1) might be considered an answer to the question What sentence (literally,
grammatically) corresponds to a given imperative?. 4 If the primary interest is not in the wording,
but the
-logical meaning we assume of the imperative-correlated sentences, we may instead employ
(td-2). Each of a sentence defined true by this definition may then be considered an answer to the
question What describes a state logically necessary and sufficient to satisfy one imperative?. (td-3) is
similar to van Fraassens truth definition in [50]. Each of a sentence defined true by this definition
(td-1)
(td-2)
(td-3)
(td-4)
is true iff there is an

is true iff
answers the question What is necessary to satisfy one imperative?. A subject may be interested not
so much in single imperatives, but what must be achieved to have all imperatives satisfied. Then, (td4), similar to the truth definition of Alchourron/Bulygin in [2], defines each of a sentence
to be
an answer to the question What is logically necessary to satisfy all imperatives?. We are at liberty to
mix
the above,
and
any other
deontic operators
we choose to define, and can then state truths such as
2.4. Logical Relations Between Imperatives?

The expression it ought to be that ... is ambiguous: it may be used to describe what ought to be according
to some norm, but also to prescribe, to create an obligation to this effect. Jrgensens dilemma withstanding, authors have wondered whether logical relations can be constructed not just between descriptive, but
also between prescriptive oughts (cf. Aqvist

[10], Alchourron/Martino [5], Alchourron/Bulygin [4]).
In a strict sense, a logic of imperatives is not possible here: To write

or
is
meaningless if the operators or are to be truth-functional, since we have not assumed truth
values of these objects this is our tribute to Jrgensens dilemma. But while the truth of a sentence
depends on the existence of some imperative in the set , the other side of the coin is that
can
be made true by putting new imperatives into this set. This property may be used to define relations
Except for imperatives with identical corresponding sentences, the set of (td-1)-true -sentences will mirror the set of imperatives. So (td-1) comes close to an operator that allows us to directly talk about imperatives (cf. van Fraassen [50] p. 18 for the
need for such operators).

210
148
be any imperative structure,

between imperatives on top of a logic about imperatives: 5 Let
and let
be shorthand for
so
is the imperative structure we obtain
from adding a new imperative to and expanding such that
. Let
be the name of
this new imperative.

might be called an imperative-candidate with respect to . Suppose we have a
logic about imperatives that defines deontic operators with respect to an imperative structure . Since
the semantics of may employ semantic restrictions, we call an imperative structure that conforms with
these a -structure. We now define (where

, is a -structure, and
denotes the set of
-sentences defined true with respect to ):

is a -structure, and
is -covered by
iff

is not a -structure
is -rejected by
iff

is a -structure, and
is -independent from
iff

is -covered by , then from the point of someone using as a tool for the analysis of a set of
If
imperatives it does not matter whether this imperative-candidate is added to the set or not, since the set
of -truths does not change.6 On the other hand, someone who believes that imperative structures should
conform to the semantic restrictions of could not tolerate the addition of an imperative-candidate
if
it is -rejected by , since thereby these restrictions would be violated. Finally if

is -independent
from , then whether it should be added to or not seems not disputable on -logical grounds alone.
We cannot further explore this topic here. But it seems important that the above semantics may not
only be used to construct descriptively interpreted deontic operators, but also to explain (meta-)logical
relations between prescriptions.
Our basic logic will be propositional logic given in the usual fashion: The alphabet has a set Prop of
proposition letters , logical signs , , , , , and brackets , . All proposition
letters are PL-sentences, and if and are PL-sentences, so are
(negation),
(conjunction),
(disjunction),
(conditional), and
(biconditional), and nothing else. Outer
brackets and corner quotes for mixed expressions are mostly omitted. The language
is the set of
PL-sentences. A valuation
assigns a truth value to each
. B is the set of
all such valuations. Truth of a PL-sentence at a valuation (we write
is defined recursively,
starting with
iff
iff
is not true
and as usual for the remaining connectives.

is the set of all
at which is true. Validity of
a sentence (we write
, tautologies, and contradictions are defined as usual ( designates an
arbitrary tautology and an arbitrary contradiction).
is contingent iff it is neither a tautology nor a
contradiction. A set of PL-sentences
is PL-satisfiable iff there is a such that for any
,
, and -implies a sentence
iff for every that satisfies ,

(we then
The idea of constructing normative logics as built on top of logics of normative statements appears in Alchourron/Bulygin
[3] p. 463, whereas in Alchourron/Bulygin [4] p. 285 the relation is vice versa. The idea of defining the normative status of
a considered norm by testing what happens to a system if this norm is added is the basis of von Wrights logic of normative
entailment (cf. [57], [58]).
6
The term coverage is borrowed from Rescher [43] ch. 6. The notion is akin to that of acceptance used in default logic, cf.
van der Torre [48].
149
211
write
). We suppose a sound and complete axiomatic system for PL.
provable in this system.
means that
is
3. Monadic Deontic Logic

To reconstruct monadic deontic logics, we employ two operators
PL. We intend to give the following meaning to the operator :
is true iff
and
on sentences of our basic logic
is necessary to satisfy the imperatives.
Since we have not introduced permissive norms, to give

the usual reading as it is permitted that
is somewhat ambiguous. What we intend to express is that the state described by does not oppose the
satisfaction of any imperative, and so define:
is true iff the imperatives and
are co-satisfiable.
Both definitions refer to a satisfaction of imperatives; in the second the terms of ordinary satisfiability
and satisfaction of an imperative even appear mixed. This seems more that a semantic coincidence if
we accept postulate (W), for then we know a set of imperatives to be satisfiable iff the set of correlated
sentences is satisfiable, and can rephrase the definitions accordingly. Both definitions also refer to
the imperatives. However, the question is what to do if there are none. Instead of rejoicing as a
normal person would, most deontic logicians churlishly exclude this possibility by demanding that at
least something must be obligatory. We want to cover the case where the imperative set is empty, and
then nothing naturally appears to be necessary to satisfy the imperatives or as opposing an imperatives
false, and
true, for all sentences . 7
satisfaction. So the truth definitions then have
, except that additionally
The language of analysis will be , where the alphabet is like that of
we have and . Then is the smallest set such that
then and
then so are and
The truth definitions are relative to an imperative structure

and are just like those for
PL-sentences, except that their first clause is replaced by the following two clauses:
iff
and
iff
or is -satisfiable
We use the following restrictions:
is -satisfiable
(R1)
(R2)

(a)
(b)
If
If
(R1) excludes that the imperatives are, for reasons of the logic of their correlated sentences, not or not
all satisfiable. Otherwise we may have the result that everything is obligatory everything is necessary
to satisfy the imperatives though nothing can and nothing is permitted but it is not particular states
but the imperatives themselves that oppose their satisfaction. (R2) may then be used to rule out the case
that the imperative set is empty.
Consider the following axiom-schemes and rules:
7
Beyond intuition, to have such a definition may be important in a more dynamic setting where we successively remove
imperatives invariably satisfied or violated from a given set. With
we then have a formula true when an agent is done.

212
150
(Def)

(Conj)

(D)
(N)

(Ext)
If then
Let be any of the names in the left column of table 1:
TABLE 1
Name
System
(Def), (Conj), (Ext)
(Def), (Conj), (Ext), (D)
(Def), (Conj), (Ext), (N)
(Def), (Conj), (Ext), (D), (N)
Semantic Restrictions
None
(R1)
(R2)
(R1), (R2)
The axiomatic system is the smallest set of DL-sentences that contains all -instances of PLtheorems as well as all PL-instances of the axiom-schemes indicated to the right of the system name in
the center column and is closed under modus ponens and any indicated rule. If
, we write
and call provable and refutable in . If and satisfies the restrictions for indicated in
the right column of Table 1 and in the same row as the system name, we call
-satisfiable. If
for all such , we call

-valid and write
.
DL is von Wrights original deontic logic in [52], DL is DL without the claim that obligations must
be consistent, SDL is the so-called standard deontic logic, and SDL the system that demands something
be obligatory, but not that obligations are consistent (the name is adopted from al-Hibri [1]). In all
systems the following derived rule holds:
(Mon)
If
then
Theorem 3.1. All systems DL , DL, SDL , and SDL are sound and complete.
Proof: Soundness is trivial with respect to truth definitions and restrictions. To prove (weak) complete
ness, for any -sentence such that

we construct an -structure such that : We
assume
so is not refutable in . We form a disjunctive normal form of and eliminate
all negation signs in front of deontic operators using (Def). The result is a disjunction of conjunctions,
where each conjunct is a sentence or . One disjunct must then be not refutable in . Let
be that disjunct. Let be the set of -sentences such that appears in , and be the set of
-sentences such that appears in . Let
if for all
otherwise
is -satisfiable
and
be an identity function on . This is our desired . We show that

demonstrating that any conjunct of is true: Let be a conjunction member of . Then
by
151
213
so
and since
,
, so
is true. Let
be a conjunction member of
. Now suppose
is false, i.e.
and
is not -satisfiable. So
is not
-satisfiable, and there are
such that
. By (Mon), (Conj)
and (Def) we have
. So , containing the conjuncts
,
is contrary to our assumption refutable in .
will be -satisfiable (R1) if contains (D): If is not -satisfiable, then there are
such that
and therefore we have
by
(D), (Def), (Ext) and (Conj), so is refutable in again. will be non-empty (R2) if contains (N):
Suppose
. Then
and for one
,
is not -satisfiable, so
. But then
by (N), (Def) and (Ext), so again is refutable in .
Remark 3.2. Suppose that in the truth definitions we drop the clause regarding empty :
iff
iff

is
-satisfiable
Then SDL is sound and complete with respect to semantics that employs these definitions, and SDL is
sound and complete for such semantics if restricted by (R1). (Validity of
is trivial with respect to the
new definitions. To adjust the above completeness proof, define
.)
4. Andersons and Kangers Reduction

We are at liberty to sort given imperatives, according to some quality of rank or importance, and also
according to what the world is like, whether it leaves them satisfied or violated. The semantics of PL
makes use of a valuation to logically model truth and falsity of sentences describing the world. We
define the satisfaction set
to be the set of all imperatives of of which the sentences correlated
by are true at , and likewise the violation set
as the set of all imperatives of of which the
sentences correlated by are false at :
If the imperative structure and the valuation are clear from the context, we will just write and . Since
accepts the principle of bivalence we have
and
.
Andersons [7] and Kangers [29] idea was to define deontic operators by a combination of alethic
operators and one constant. Unhappily, Anderson [7] and Anderson/Moore [9] explained their constant
S as representing a sanction or bad thing, and defined the statement
as meaning that not doing
necessitates the sanction. Lemmon and Nowell-Smith [32] objected, since in this wicked world it is not
always true that when (say) a robbery is committed someone suffers a sanction and accused Anderson
of the naturalistic fallacy. Interestingly, in answering these accusations Anderson [8] argued that if one
fails to fulfill ones obligations, that is the bad thing, no sanction being presupposed. But then we can
give Andersons constant a definite meaning: the bad thing is that
, since then an imperative has
been violated. Kanger [29] introduced a constant Q standing for what morality prescribes. Though
this seems just as dubious at first, Kanger likened morality to a welfare program for the universe of
discourse, represented by a set of propositions, and defined Q to be true iff all propositions of this
program are true.
is then defined true iff the truth of Q necessitates the truth of . Now we have a set

214
152
of sentences that might be understood as a welfare program: the set of sentences that correspond
to the imperatives. We therefore interpret Kangers constant as meaning that all sentences in are
true, hence by (W) every imperative is satisfied, so
.
Here we will have four constants, of which only two can be defined. The constants of Anderson
and Kanger neither alone nor combined suffice for logics properly parallel to
and , or axioms
expressing the principle of bivalence which the notions of satisfaction and violation inherited from our
basic logic. The language of analysis will be , where the alphabet is like that of
, except that
additionally we have operators and and constants , , , and . The definition of the set
of sentences differs from that of
only in the respect that now also , , , and are sentences, and
if is an -sentence, then also
and
are. Note that will represent Andersons constant, and
is our interpretation
of Kangers Q. The truth definitions are relative to a valuation and an imperative
structure
, and are as for -sentences, except for these additional clauses:
iff
for all :
(logical necessity)
iff
for some :
(logical possibility)
iff
(every imperative is satisfied)

iff
(some imperative is satisfied)

iff
(every imperative is violated)

iff
(some imperative is violated)
(Df1)
(Df2)
(Biv1)

(Biv2)

(AK-D)
(AK-N)

(Nec)
If
then
TABLE 2
Name
System
(Df1), (Df2), (Biv1), (Biv2), (Nec)
(Df1), (Df2), (Biv1), (Biv2), (Nec), (AK-D)
(Df1), (Df2), (Biv1), (Biv2), (Nec), (AK-N)
(Df1), (Df2), (Biv1), (Biv2), (Nec), (AK-D), (AK-N)
Semantic Restrictions
None
(R1)
(R2)
(R1), (R2)
We define the axiomatic system

to be the smallest set of AK-sentences that contains all AKinstances of PL-theorems and the modal logic S5 as well as all axioms indicated to the right of the
system name in the center column, and is closed under modus ponens and any indicated rule. If
we write
and call provable in . -satisfiability and -validity of AK-sentences are then
defined as before.

216
153
5. Dyadic Deontic Logic

The aim
of dyadic deontic logic, that formalizes If is the case, it ought to be that is the case
as
, was to cope with deficiencies of monadic deontic logic with respect to the representation
of conditional obligations. First, Priors paradox of derived obligation [41] showed
to be
inadequate for contrary-to-duty imperatives
expressing
a
secondary
duty
in
case
a
primary
norm is
violated: From we derive

with (Mon) for any (be hanged for a sheep as for a
lamb). Then, Chisholm [14]
showed that if we formalize contrary-to-duty obligations as ,
the set
, expressing a primary obligation, a contrary-to-duty imperative, an ordinary conditional imperative, and the fact that the primary obligation was violated, is
inconsistent in standard deontic logic, and otherwise formalized mutually dependent, though intuitively
they should be neither.
To solve these problems, von Wright [53] introduced dyadic deontic operators, meaning to reflect a
conditional structure of the operators scope. To reconstruct dyadic deontic logics, we add an auxiliary
sign / to the alphabet and define the language of dyadic deontic logic to be like except
that in the definition of sentences clause (a) now reads
(a)
If , then and
Semantics for dyadic deontic logics is usually based on a preference relation (at least as good as)
on the set of all Boolean valuations B (Hansson [19]) or a set of possible worlds W (Lewis [33], Aqvist
is true
iff
the
best
-worlds
are
-worlds,
[12]). According to the Hansson-type definition,
whereas the Lewis-type truth definition demands that there is a -world such that no 9
world is at least as good. Hansson motivated his semantics by the observation that while one should
always try to make the real world an ideal world, in circumstances in which somebody nevertheless
performs a forbidden act ideal worlds may be excluded, but some achievable worlds may still be better
than others. One should then make the best out of sad circumstances, i.e. realize the best worlds left. In
our semantic framework, the best state obviously is that all imperatives are satisfied. However, suppose
that in circumstances some imperatives cannot be satisfied. It then seems reasonable to demand that
all other imperatives still satisfiable in circumstances be satisfied, and to permit
only what is cosatisfiable in with the imperatives left. The truth of dyadic-deontic sentences
and
will be defined with respect
to
the
subset
of
imperatives
in
not
yet
violated
in
situation each
sentence such that is true answers the question: What is necessary to satisfy all imperatives
that are still satisfiable, given circumstances ? Note that we do not presuppose any special conditional
structure of the imperatives.10
For the semantics, let be a function that removes those sentences from a set which (individually)
imply the truth of a sentence :
Let be any imperative structure. is then the set of just those sentences in
that are individually satisfiable with . We give the following truth conditions:
9
Lewis-type truth definitions have become more standardly employed with Horty ([25], [27]) to define deontic operators with
regard to (general, dominance, maximin) orderings in stit-models.
10
So in the present setting, the solution to Chisholms paradox is to let

In the original
paradox, the case that the man can still go ( ), but has already told he wont ( ), is not accounted for. If the priority is to go,
we can change the above set into

154
217

is -satisfiable
Before we take a closer look at note that the truth definitions are strict parallels of those in section
3. Concerning their first clause, I believe that when all imperative correlated sentences are rejected by
, i.e. all imperatives are violated in circumstances , then there is nothing left to be done. To
iff
iff
and
or
call something obligatory or not permitted in such a situation seems as pointless as when there are no
imperatives at all: E.g. if Im to see to it that the two witnesses appear at the wedding, and one wont
come, I might just as well tell the other one its off. However, we shall also consider truth conditions
where this clause is dropped:
*
*
iff
iff
is
-satisfiable
We will then see that whereas conditions (a) and (b) produce Lewis-type dyadic logics, semantics that
makes use of conditions (a*) and (b*) will relate to Hansson-type systems.
Obviously, is a contraction function, and the truth of

depends on whether is implied
by the corresponding revision of
with . Contraction functions serve to remove sentences from
theory bases if a certain sentence is desired not to be in the theory (cf. [6], [20]). The problem is
that the process is ambiguous: Consider a basis that contains the individual sentences ,
Then for a
contraction with
it suffices to remove just one of the two, so there is a choice. Absolutely safe
full meet contraction removes both, while partial meet contraction employs a choice function to pick the
sentences to be removed, leaving us with several possibly resulting contracted theory bases. Now is
naive in the sense that it avoids ambiguity, but at the price of producing no contraction in the classical
sense: If we -contract with a non-tautological sentence, this sentence may still be a consequence of
the contraction (see the above case). So naive contraction is really the opposite of absolutely safe full
meet contraction. Why use this weird function instead of some other available concept?
My position is that we should not exclude dilemmas created by imperatives that
conflict for
some
situation. The logical evidence of there being a dilemma is the truth of both
,
. A
situation where
was demanded initially, but is now true, is a state of violation, not a dilemma.
Such demands are disregarded by
In a dilemma, we also know that invariably some demands
will be violated, but which ones is still up to us. Then saying You should do and also You should
may reflect a tenable position (this van Fraassen [50] argued for in detail, also cf. von Wright
do
[55], [56], Hamblin [17] ch. 5). So some logic should be able to state the dilemma, and we should not
beforehand smooth out the situation by removing one of the still satisfiable demands, or change them into
a demand for, say,
. Also, in a dyadic setting the existence of a dilemma for a situation is not
entirely devastating, even
if
is constructed as a regular operator (supporting agglomeration of contents)
so subsequently
is true: There might neither be a dilemma prior to the truth of , nor might
a dilemma ever arise again once one of the demands is violated, as it must be. So dilemma-permitting
logics with regular operators may still be of interest. 11
We use the following restrictions on , which parallel (R1), (R2) employed above (p. 211):
To avoid the truth of
while accepting the possible truth of both
, van Fraassen [50] proposed

defining deontic operators with regard to maximal sets of collectively satisfiable imperatives. This idea was explored in detail
by Horty [26], and implies giving up agglomeration for some cases, though not all.
11

218
155
(DR1)
(DR2)
If
If
then
then
is -satisfiable
(no predicaments)
(no outlaws)
(DR1) expresses the assumption that all possible conflicts (dilemmas, predicaments) are resolved in
. If the circumstances can be realized at all, the imperatives that are then left must be co-satisfiable
with them:
Though in possible circumstances e.g. two imperatives
with
and
might be satisfiable with individually, they cannot both be satisfied any longer
and therefore are not both
Note that if (DR1) applies, then for all
allowed
in the set of imperatives.
either or .12
(DR2) demands that in all possibly realized circumstances there are still some unviolated imperatives left. I call this restriction No outlaws, since it might be understood as stating that we cannot get
into a situation, however despicable, that would completely rid us of all duties. (DR2) may be motivated
by the assumption that contains an infinite supply of imperatives for any occasion, though of course
one tautological imperative would suffice.
then

if
if
then
if then

if then
(DDef)
(DConj)
(CExt)
(ExtC)
(Up)
(Down)
(DD )
(DN )
(DDR )
(DNR )
(DDef) and (DConj) are the analogues

of monadic (Def), (Conj). (CExt) is a contextual extensionality
rule for the consequent in

13 and (ExtC) an extensionality rule for circumstances. (Up) transfers obligations conditionally from stronger to weaker circumstances, (Down) allows for a corresponding
transfer of obligations from weaker to stronger, yet not contrary-to-duty circumstances. 14 Concerning
(DD ), (DN ), it is obvious that axiomatic systems that contain both and (DDef), (CExt) will be inconsistent. So we have to choose between (DD ), (DN ), and the analogues of the monadic (D) and (N)
axioms must then be restricted as in (DD R ), (DNR ).15
12
I owe this observation to Leon van der Torre. So the contents of all norms must be chained: The norm giver must, implicitly
or explicitly (e.g. by use of unless-clauses), indicate which demands should be given priority in case each is still satisfiable
individually, but not all collectively. This does not mean that in a finite code all but the strongest norms are redundant, since
some norms may be contrary-to-duty.
13
Lewis [33], van Fraassen [49] employ non-contextual extensionality and an axiom to the same effect.
14
(Up) and (Down) are derivable in the systems CU of Lewis [33], Spohn [46], and Aqvist
[11], [12], [13]. (Up) was first
employed by Rescher [42], and (Down) was first proposed by Fllesdal/Hilpinen [16], p. 31.
15
Hansson [19], Spohn [46], and Aqvist

[11] give priority to the analogue of monadic (N) and restrict the analogue of monadic
(D) to noncontradictory (possible) antecedents, whereas van Fraassen [49] and Lewis ([33] cf. p. 5) give priority to the analogue
of (D), Lewis restricting his analogue of (N) similarly.
156
219
TABLE 3
Name
DDL
DDL
DSDL
DSDL
System
Restrictions
(DDef), (DConj), (CExt), (ExtC), (Up), (Down), (DD )
None
(DDef), (DConj), (CExt), (ExtC), (Up), (Down), (DD ), (DD )
(DR1)
(DDef), (DConj), (CExt), (ExtC), (Up), (Down), (DD ), (DNR )
(DR2)
(DDef), (DConj), (CExt), (ExtC), (Up), (Down), (DD ), (DD ), (DN )
(DR1), (DR2)
(DDef), (DConj), (CExt), (ExtC), (Up), (Down), (DN ), (DNR )
DSDL3
DSDL3
None
(DDef), (DConj), (CExt), (ExtC), (Up), (Down), (DN ), (DNR ), (DDR )
(DR1)
We define the axiomatic system to be the smallest set of DDL-sentences that contains all DDLinstances of PL-theorems as well as all PL-instances of the axioms indicated to the right of the system
name in the center column, and is closed under modus ponens and any indicated rule. If
we write
and call provable in . For systems DDL , DDL, DSDL , DSDL, validity and satisfiability are
defined with respect to truth conditions (a) and (b) and the indicated restrictions, and similar for systems
DSDL3 , DSDL3 with respect to truth conditions (a*) and (b*) (the horizontal line indicates the change
in truth definitions).
Neglecting differences of language, DSDL is the synoptical system DFL of Aqvist

[11], [12], [13],16
and DSDL3 is the system of von Kutschera [51]17 . DSDL3 is the system of Hansson [19] as axiomatized
by Spohn [46]. Analogues of the remaining systems seem to be missing in the literature. We give some
theorems common to all:
then
(CMon)
if
(T1)
(T2)
(T3)
(T4)
(T5)
(T1) is the down axiom employed e.g. by Rescher [42] and Aqvist
[12],18 and (T2) is an equivalent of
Lewis [33] down axiom (A8). In the presence of (Up) and (CExt), (T1) and (T2) are derivable from
(Down), while for a derivation of (Down) from (T1) the use of (DN R ), and for a derivation of (Down)
from (T2) the use of (DDR ) seems crucial. (T3) is Lewis [33] upward axiom (A6), derivable here from
(Up) and (CExt). (T4) and (T5) both derive from (Up) and will be useful in the proofs below.
Theorem 5.1. The systems
are sound.
Proof: The validity of (DDef), (DConj), (CExt), and (ExtC) is trivial, and so is (DD ) since
. For the validity of (Up), first observe that
Assume the truth
of
so
and also
We have
16
DFL contains the systems of Danielsson [15], van Fraassen [49], and Lewis [33].
We add (ExtC), which von Kutschera erroneously believed to be derivable.
18
Von Wrights axiom (A2) in [53] is equivalent to the conjunction of (T1) and (Up), and so are von Kutscheras (A17) in [51],
and Spohns (P5). For a discussion of versions of (Up) and (Down) cf. [40].
17

220
157
and since ,
from monotony of we obtain

: Again observe first
For (Down),
consider
its
equivalent
that
Suppose and are
true. Either
Suppose
is -satisfiable, so
is satisfiable with ,
for all
but
we
assumed
the
contrary
for
all
So
for
to
must
but then
be true,
isbe empty,
so holds
trivially.
Or
-satisfiable.
If
then and again trivially
is -satisfiable. Suppose
-satisfiable, so for all
sois once
is satisfiable
with
so
more
and
is -satisfiable. So in all three possible cases in which

so
are true, is true.

and
For (DDR ), if then due to (DR1)
is satisfiable, so the second
disjunct
definition holds for
. Concerning (DN R), if then due to (DR2)
of the truth
, so the first conjunct of the truth definition holds for
, and the second holds
trivially since
is always a consequence.
Theorem 5.2. The systems DDL , DDL, DSDL , and DSDL are complete.
for any
Proof: To prove (weak) completeness, we have to show that
-sentence
and any system
We assume so is not refutable in .

We build a disjunctive normal form of
and eliminate all negation signs in front of deontic operators
by using (DDef). The result is a disjunction of conjunctions, where each conjunct is a sentence
or
. One disjunct must then be not refutable in . Let
be that
be the
disjunct. Let
-sentences that contain only proposition
letters
occurring
in
.
Let
be
a
set
of
2
mutually
non-equivalent representatives of
, where is the number
. By writing PL of proposition letters in
sentences we now mean their unique representatives in

. We construct a set with the following
restrictions:
(a)
Any (with CExt, ExtC from
equivalently replaced) conjunct of is in ,
(b)
for all

either
or
,
(c)
is -consistent.
It then suffices to prove that there is an imperative structure for that satisfies all
. The proof
follows Hansson [19] and Spohn [46], i.e. we first identify the deontic basis of a sentence , identify
supersets
such that the deontic basis of is implied by that of (which reflects partly ordinary,
not contrary-to-duty, conditionality of ), and then see to it that the deontic basis of
is implied by
(so the same holds for ). However, unlike Hansson

and Spohn, we must allow for
sentences
with an empty deontic basis (in case
), or no deontic basis at all (in case
). So let be the (finite) set of sentences such that

, and be the
such that
. Observe that

and
corresponding set of sentences
For all such that

we define the deontic basis as the conjunction of all
sentences in : We have:
iff

iff
iff
iff
is not defined
158
For all with a defined basis,

by (DConj), and
(CMon) and (b), (c), for all such we have:
221
by (CExt). Furthermore, from

For each sentence
, the sentences such that may be ordered by extensional inclusion.

We define to be the set of all sentences maximal within that order, and prove some observations
regarding :

and for all : if then

such that
(L3) If
then there is a
and

Proof : If then there must be a such that
and if then likewise for some sentence

. There is only a finite number of
sentences in
so
for
some
such
there
cannot
be a such that
,
and So and

(L4) For all sentences

.
Proof: Apparent from (L3) and the fact that

(L5) For any such that
for some , we have
, so we obtain

Proof : Suppose
Then

from (Down) , and so .

(L6) Let
for some Then either or
. Then from and (Down) we have

Proof : Suppose
,
so

(L7) If
, then there is a sentence such that and
Proof : Suppose
We have
and
by (Down) also

. So by definition
and

of
We
have
by
(CMon)

, so
But then
with (Down) we obtain

(L8) Let
then
.
for some . Then and if
Proof :
, so with
(CMon), we have

so
from
Suppose
of , so byuse

(Down) and by definition
Then
of (Down) we obtain
contrary to the assumption

(L9) For all sentences such that
and :
Proof : From
and we have by use of (Up).
So

and
From

we have
by use of (T2).
So
and since also
we
obtain
(L1)
(L2)
iff
iff

222
159
, let

for some
Definition 5.3. For the desired imperative structure
and
,
be some proposition letter not in
is the set given by the following clauses, and is again an identity function on :
,
and

and

(i) For each

(ii)
19
Lemma 5.4. (Coincidence

Lemma)

For all

iff
Proof: For each , either

and
, or
.
or
Suppose
. So
. We have from (L6) and (L8). From (L8) it
follows that logically implies for each
, and also that logically implies for each
.
Concerning the sentences from clause (ii), by the same
So no sentence from clause (i) can be in
reasoning logically implies and also the negation of the consequent. So neither are these
two
, so and both sides of the iff-clause prove false for any .
sentences in

Suppose
and
Due to (L7) there is a such that logically implies ,
and
for otherwise
from (L9) and
contrary to our assumption. It
and so so from
follows that
. Now for any other , we have that if or are in

(L9)
then : Assume Then since

so
so
from (L6), and since (L9) gives us
we have

Assume
So , and since we have

from (L6). If also
then , and from
, so

and (Down) we have and

from (L6), so but we
assumed otherwise. So
so again

So for all sentences
deriving from clause (i) we have and for the sentences from clause (ii) this
. So with (L9) and (L1) we have iff
holds trivially since

iff
Suppose
Then
We have for otherwise

and
by
sentences from clause (ii) are in

(T5), but we assumed
So both So

Assume Then for all ,
so
Assume that there are For each such
for otherwise
For any

due to (T4). So
so
for
each
such

, either
then due to (L3) there is a
such that
or So
and again So both sides of the iff-clause prove true for

any
.
By these clauses is constructed inconsistent: The sentences from clause (ii) jointly imply that some
is true, which is
denied by the sum of all sentences from clause (i). It is the job of the contraction function to remove sentences to make
consistent, if so required.
19
160
Lemma 5.5. (Verification Lemma)

If contains (DDR ), then (DR1) holds for . If
223
contains (DN R ), then (DR2) holds for .
there is a such that and for all

iff For (DR1) suppose Then We have
from (DDR), so If then , so also
and we
assumed to be satisfiable. If then
iff
so is satisfiable.
so also and, since
so
, is satisfiable.
(DR2) suppose
so We
from (DNR), so For
again
have
If
then
and since
If
then
so
both imperatives from clause (ii) are in and at least one must also be in
Proof: For each
This completes the proof of Theorem 5.2.
Remark 5.6. If is DDL or DSDL, we can

more easily construct by use of the system of spheres
characteristic of these logics. Let
be the smallest set such that
then
if and
then

is then defined as
. The obvious fact here that for all , either
R
or
, thus reflects a property of deontic bases resulting from (DD ) that, when presented in
if
(i)
(ii)
the usual possible worlds style, goes through without comment.
Theorem 5.7. The systems DSDL3 , DSDL3 are sound and complete.
Proof: Regarding soundness, the validity of (DN ), (DNR ) is now trivial with respect to the changed
truth definitions, and so is the validity of (DD R ) for semantics restricted by (DR1). (Up) and (Down) are
proved as before (for Down there is now just one case to consider). To prove (weak) completeness, the
above proof may be adapted in the following way: First observe that for all
is now well defined,
since
due to (DNR ), (DN ). Now possibly
(just where in the previous construction
), so (L4) only holds for such that

and (L8) cannot be stated since (L7) does
not define
any more. All other observations still hold. In the definition of , clause (ii) now has
. In the proof of the coincidence lemma, case

is now excluded
and case
done as before. In case
if
then
holds
trivially, so suppose
Then both sentences of clause (ii) are in
since
neither
excludes nor . Assume first

then for all
we have
so
, and
The case
is then proved as before. In the proof
of the verification lemma, for (DR1) case
is redundant and the remainder proved identically.
Note that also (DR2) holds trivially since any nonempty
has a nonempty intersection with or
so one of the sentences from clause (ii) is in

.
6. Future Study
We motivated a semantics that models imperatives, and demonstrated that existing systems of deontic logic reconstruct nicely with respect to this semantics. In particular we could observe which truth

224
161
definitions and restrictions on imperative structures are necessary to obtain these reconstructions, while
standard possible worlds semantics tends to conceal that and how its restrictions on relations between
worlds are in fact restrictions on the presupposed normative system. We can only hint at how other
concepts may be accomodated:
We may define some subset of I as minimal requirements. We can then define operators expressing what is necessary to satisfy the minimal requirements, but also operators of supererogation as
meaning that while something is not minimally required, it still satisfies an imperative, or operators of offense as meaning that some not minimally required imperative is violated (cf. [36] for
a possible worlds setting of supererogation).
We may have hierarchies of imperative structures, and define operators in relation to satisfaction
sets that fulfil e.g. the following criterion: Any imperative of the highest authority is satisfied, any
imperative of the second highest authority is satisfied that can be if the imperatives of the highest
authority are, etc. (cf. [18] for such command hierarchies).
We might use a second function to associate (numerical) values with each imperative, and define
as true iff is necessary for achieving that the sum of all values of imperatives in the satisfaction set is the sum of all values of imperatives in , the highest achievable sum, or surpasses a
fixed value. It may then be obligatory to violate one imperative when it serves to satisfy others of
higher value.
Regarding conditional imperatives, we may follow van Fraassens proposal in [50] and employ
a function that associates a second -sentence with each imperative, describing the opportunity for its satisfaction or violation. For all valuations we then have the set
of actualized
imperatives
such that
. We can define a monadic operator
as true at iff
, and a dyadic operator

as true iff
is true in all
. This operator
will be indefeasible, i.e. allow strengthening of the antecedent. We may then combine such dyadic
operators with contrary-to-duty operators as defined above (cf. [48] for a similar enterprise).
Some puzzling detail remains: In section 2, we have separated imperatives and their propositional
content, linking them by a function that associates a descriptive sentence with each imperative. But in
the reconstructions this separation was not used the restrictions and truth definitions could equally well
be given with respect to a set of sentences instead. This does not mean that imperatives are reducible to
indicatives. It merely shows that the reconstructed logics do not require the modelling of other qualities of
imperatives. The real puzzlement comes from the other side of the equation: Quite a number of deontic
logics may be characterized as being not about specifically normative notions at all, but as reasoning
about sets of sentences and their underlying classical (meta-)logic. To modify von Wrights famous
statement: It may have turned out that deontic logics have a wider reach than normative concepts.
References
[1] al-Hibri, A.: Deontic Logic, Washington: University Press, 1978.
[2] Alchourron, C. E. and Bulygin, E.: The Expressive Conception of Norms, in [24], 95123.

162
225
[3] Alchourron, C. E. and Bulygin, E.: Pragmatic Foundations for a Logic of Norms, Rechtstheorie, 15, 1984,
453464.
[4] Alchourron, C. E. and Bulygin, E.: On the Logic of Normative Systems, in Stachowiak, H. (ed.): Handbuch
pragmatischen Denkens, vol. IV, Hamburg: Meiner, 1993, 273294.
[5] Alchourron, C. E. and Martino, A. A.: Logic Without Truth, Ratio Juris, 3, 1990, 4667.
[6] Alchourron, C. E., Gardenfors, P., and Makinson, D.: On the Logic of Theory Change: Partial Meet Contraction and Revision Functions, Journal of Symbolic Logic, 50, 1985, 510530.
[7] Anderson, A. R.: The Formal Analysis of Normative Concepts, in Rescher, N. (ed.): The Logic of Decision
and Action, Pittsburgh: University Press, 1967, 147213.
[8] Anderson, A. R.: Some Nasty Problems in the Formal Logic of Ethics, No u s, 1, 1967, 345359.
[9] Anderson, A. R. and Moore, O. K.: The Formal Analysis of Normative Concepts, American Sociological
Review, 22, 1957, 917.
[10] Aqvist,
L.: Interpretations of Deontic Logic, Mind, 73, 1964, 246253.
[11] Aqvist,
L.: Deontic Logic, in Gabbay, D. and Guenthner, F. (eds.): Handbook of Philosophical Logic, vol.
II, Dordrecht: Reidel, 1984, 605714.
[12] Aqvist,
L.: Some Results on Dyadic Deontic Logic and the Logic of Preference, Synthese, 66, 1986,
95110.
[13] Aqvist,
L.: Introduction to Deontic Logic and the Theory of Normative Systems, Naples: Bibliopolis, 1987.
[14] Chisholm, R. M.: Contrary-to-Duty Imperatives and Deontic Logic, Analysis, 24, 1963, 3336.
[15] Danielsson, S.: Preference and Obligation, Uppsala: Filosofiska foreningen, 1968.
[16] Fllesdal, D. and Hilpinen, R.: Deontic Logic: An Introduction, in [23], 135.
[17] Hamblin, C. L.: Imperatives, Oxford: Blackwell, 1987.
[18] Hanson, W. H.: A Logic of Commands, Logique et Analyse, 9, 1966, 329343.
[19] Hansson, B.: An Analysis of Some Deontic Logics, No u s, 3, 1969, 373398. Reprinted in [23], 121147.
[20] Hansson, S. O.: New Operators for Theory Change, Theoria, 55, 1989, 114131.
[21] Hare, R. M.: Imperative Sentences, Mind, 58, 1949, 2139.
[22] Hare, R. M.: The Language of Morals, 2nd ed., Oxford: University Press, 1964.
[23] Hilpinen, R. (ed.): Deontic Logic: Introductary and Systematic Readings, Dordrecht: Reidel, 1971.
[24] Hilpinen, R. (ed.): New Studies in Deontic Logic, Dordrecht: Reidel, 1981.
[25] Horty, J. F.: Agency and Obligation, Synthese, 108, 1996, 269307.
[26] Horty, J. F.: Nonmonotonic Foundations for Deontic Logic, in [38], 1744.
[27] Horty, J. F.: Agency and Deontic Logic, New York: Oxford University Press, 2001.
[28] Jrgensen, J.: Imperatives and Logic, Erkenntnis, 7, 1937/38, 288296.
[29] Kanger, S.: New Foundations for Ethical Theory, in [23], 3658.
[30] Kelsen, H.: Reine Rechtslehre, Leipzig/Wien: Deuticke, 1934.
[31] Kelsen, H.: Allgemeine Theorie der Normen, Wien: Manz, 1979.

226
163
[32] Lemmon, E. J. and Nowell-Smith, P. H.: Escapism: The Logical Basis of Ethics, Mind, 69, 1960, 289300.
[33] Lewis, D.: Semantic Analyses for Dyadic Deontic Logic, in Stenlund, S. (ed.): Logical Theory and Semantic Analysis, Dordrecht: Reidel, 1974, 114.
[34] Makinson, D.: On a Fundamental Problem of Deontic Logic, in McNamara, P. and Prakken, H. (eds.):
Norms, Logics and Information Systems, Amsterdam: IOS, 1999, 2953.
[35] Mally, E.: Grundgesetze des Sollens, Graz: Leuschner und Lubensky, 1926.
[36] McNamara, P.: Doing Well Enough: Toward a Logic for Common Sense Morality, Studia Logica, 57,
1996, 167192.
[37] Moutafakis, N. J., Imperatives and their Logics, New Delhi: Sterling, 1975.
[38] Nute, D.: Defeasible Deontic Logic, Dordrecht: Kluwer, 1997.
[39] Opaek, K.: Theorie der Direktiven und Normen, Wien: Springer, 1986.
[40] Prakken, H. and Sergot, M.: Dyadic Deontic Logic and Contrary-to-Duty Obligations, in [38], 223262.
[41] Prior, A. N.: The Paradoxes of Derived Obligation, Mind, 63, 1954, 6465.
[42] Rescher, N.: An Axiom System for Deontic Logic, Philosophical Studies, 9, 1958, 2430.
[43] Rescher, N.: The Logic of Commands, London: Routledge & Kegan, 1966.
[44] Ross, A.: Imperatives and Logic, Theoria, 7, 5371.
[45] Sosa, E.: The Logic of Imperatives, Theoria, 32, 1966, 224235.
[46] Spohn, W.: An Analysis of Hanssons Dyadic Deontic Logic, Journal of Philosophical Logic, 3, 1975,
237252.
[47] Stenius, E.: The Principles of a Logic of Normative Systems, Acta Philosophica Fennica, 16, 1963, 247
260.
[48] van der Torre, L.: Reasoning about Obligations, Amsterdam: Thesis Publishers, 1997.
[49] van Fraassen, B.: The Logic of Conditional Obligation, Journal of Philosophical Logic, 1, 1972, 417438.
[50] van Fraassen, B.: Values and the Hearts Command, Journal of Philosophy, 70, 1973, 519.
[51] von Kutschera, F.: Normative Praferenzen und bedingte Gebote, in Lenk, H. (ed.): Normenlogik, Pullach:
UTB, 1974, 137165.
[52] von Wright, G. H.: Deontic Logic, Mind, 60, 1951, 115.
[53] von Wright, G. H.: A Note on Deontic Logic and Derived Obligation, Mind, 65, 1956, 507509.
[54] von Wright, G.H.: Logical Studies, London: Routledge and Kegan, 1957.
[55] von Wright, G.H.: A New System of Deontic Logic, Danish Yearbook of Philosophy, 1, 1964, 173182.
Reprinted in [23], 105120.
[56] von Wright, G.H.: A Correction to a New System of Deontic Logic, Danish Yearbook of Philosophy, 2,
1965, 103107. Reprinted in [23], 105120.
[57] von Wright, G. H.: Norms, Truth, and Logic, in von Wright, G. H.: Practical Reason, Oxford: Blackwell,
1983, 130209.
[58] von Wright, G. H.: Is There a Logic of Norms?, Ratio Juris, 4, 1991, 265283.
[59] Weinberger, O.: Uber

die Negation von Sollsatzen, Theoria, 23, 1957, 103132.
[60] Weinberger, O.: Normentheorie als Grundlage der Jurisprudenz und Ethik, Berlin: Duncker & Humblot,
1981.
164
Journal of Applied Logic 2 (2004) 3961

www.elsevier.com/locate/jal
Problems and results for logics about imperatives

Jrg Hansen
Institut fr Philosophie, Universitt Leipzig, Leipzig, Germany
Abstract
Deviating from standard possible-worlds semantics, authors belonging to what might be called
the imperative tradition of deontic logic have proposed a semantics that directly represents norms
(or imperatives). The paper examines possible definitions of (monadic) deontic operators in such a
semantics and some properties of the resulting logical systems.
2004 Elsevier B.V. All rights reserved.
Keywords: Logic of imperatives; Deontic logic
1. Imperative semantics and basic operators

Deontic logic, i.e., the logical analysis of obligation, permission and prohibition, is usually modelled by a possible-worlds semantics, in which among the set of worlds a deontic
alternative or preference relation selects some as ideal or better compared to others.
However, when there is an explicit set of given imperatives, a code of norms, or a number
of specified tasks to complete for the agent, the appeal to some prohairetic notion for the
meaning of ought seems out of place. Instead, what is obligatory then depends on whether
it serves to satisfy the imperatives, avoid norm violations, or complete all the tasks.
In what might be termed the imperative tradition of deontic logic, a number of authors
have deviated from the standard approach by giving semantics that relates the meaning
of deontic operators to an explicitly given set of norms or imperatives.1 The general idea
E-mail address: jhansen@uni-leipzig.de.
1 In [8] I mention semantics of Kanger [14], Stenius [20], van Fraassen [21], Alchourrn and Bulygin [1].
A more comprehensive list should also include Smiley [19], and Niiniluoto [17]. Smiley, in a concept later endorsed by Ruth Barcan Marcus [15] and motivated by an Andersonian [2] definition of a constant representing
the satisfaction of all normative demands, uses a normative code consisting of propositions to define OA as
meaning that there is a finite number of propositions p1 , . . . , pn in this code such that (p1 pn A) is a
logical truth. Niiniluoto represents commands by a tuple containing a proposition p, where the truth of OA then
depends on whether A is logically implied by one such p.
1570-8683/$ see front matter 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jal.2004.01.003

40
J. Hansen / Journal of Applied Logic 2 (2004) 3961
behind imperative semantics is that to each command or imperative there is a descriptive

sentence that describes what must hold iff this imperative is satisfied. If a set of imperatives is under consideration, the set of corresponding descriptive sentences is then used of
in the definition of deontic operators. The proper representation of this set is controversial:
Directly representing imperatives by a set of descriptive sentences, as Kanger [14] and Alchourrn and Bulygin [1] have done, makes it appear as if norms can somehow be reduced
to factual statements. Others like van Fraassen [21], Niiniluoto [17] and myself [8], have
more cautiously represented imperatives by a set of objects that refer to states of affairs or
propositions, thereby following the doctrine that norms bear no truth values. For reasons of
readability I skip this intermediate step here, and use a set I L[BL] of descriptive sentences thatin the language of a basic logic BLmean the sentences associated with some
imperatives in the above sense. Deontic operators are then considered as giving information
about the properties of this set of imperative-associated descriptive sentences. The following definitions of operators O 15 , differing in strength, will be employed throughout:
I
I
I
I
I
|= O 1 A
|= O 2 A
|= O 3 A
|= O 4 A
|= O 5 A
iff
iff
iff
iff
iff
A I,
B I :|=BL B A,
B I :|=BL B A,
B1 , . . . , Bn I :|=BL (B1 Bn ) A,
I |=BL A.
O 1 A is true iff A is one of the sentences in I , i.e., A (literally) corresponds to what an imperative demands. O 2 A is true iff the truth of A is BL-logically a necessary and sufficient
condition for the fulfillment of what one imperative demands.2 Similar to Niiniluotos [17]
operator, O 3 A is true iff A is BL-implied by what some imperative demands. Similar to
Smileys [19] operator, O 4 A is true iff a finite subset of imperative-associated sentences
BL-implies A. Finally and similar to AlchourrnBulygins [1] operator, O 5 A is true iff A
is BL-implied by the set of all such sentences. Note that while all of O 1 O 4 presuppose
the existence of some sentence in I , O 5 A can hold for (BL-valid) sentences A even when
I contains no object.
Discussing the general framework, I showed in [8] how a number of well-known systems of monadic and dyadic deontic logic can be reconstructed with respect to imperative
semantics. The present paper examines some details and problems connected with monadic
deontic logics facilitated by the above semantics. Section 2 gives logical systems for each
of the above operators. Section 3 then examines systems that include several of these operators. In Section 4 we take a look at possible semantic restrictions on the content of
imperatives, and see what changes must be applied to logical systems to correspond to
them. Section 5 examines more closely the definitions van Fraassen [21] proposed for
O-operators, and provides a monotonic system to resemble his reasoning, and Section 6
addresses a sceptical definition of ought proposed by Horty [12] to deal with normative
conflicts. Finally, in Section 7 I give systems and semantics for extended languages that
permit iterated and nested deontic operators of the types presented above.
2 In the terminology of Brown [4] O 2 is a type 2 operator, whereas operators O 35 would be type 1, their
scope being a necessary condition for norm satisfaction only.
165
166
41
2. Logics DL15
for deontic operators O15
1
Let our basic logic be propositional logic PL. The alphabet of the language L[PL]
has a set of proposition letters Prop = p1 , p2 , . . . , truth-functional operators , , ,
, and brackets (, ). The set of sentences is defined as usual. ,
in front
of sets of sentences means their conjunction and disjunction, and, e.g., ni=1 Ai further
abbreviates {Ai , . . . , An }. In the semantics, valuation functions v : Prop {1, 0} define
the truth of sentences A L[PL] as usual, A meaning the set of valuations v that make
A true. means an arbitrary tautology, and an arbitrary contradiction. We suppose a
sound and complete axiomatic system for PL.
In this section, we examine deontic logics DLi1 for operators O i , 1 i 5, defined
above. The upper index in the system name (and the names of most axiom schemes) indicates the type of operator used, and the subscript indicates that the scope of O i is L[PL],
not a deontic language (in Section 7 we give up this restriction). The alphabet of languages
L[DLi1 ], 1 i 5, is like L[PL] except for one additional operator O i . L[DLi1 ] is then
the smallest set such that
(a) if A L[PL] then A L[DLi1 ],
(b) if A L[PL] then O i A L[DLi1 ], and
(c) if A, B L[DLi1 ], so are A, (A B), (A B), (A B), and (A B).
Note that we permit mixed expressions like p1 O 1 p2 . The axiomatic systems DLi1
are then defined by the following clauses (we write DLi A for A DLi1 ):
1
(a) All L[DLi1 ]-instances of PL-tautologies are in DLi1 .

(b) DLi1 is closed under modus ponens and the following rule:
(Exti1 ) If PL A B then DLi O i A O i B, 2 i 5.
1
(c) For all A, B L[PL] ( being an arbitrary tautology):
(Mi1 ) DLi O i (A B) (O i A O i B), 3 i 5,
1
(Ci1 ) DLi (O i A O i B) O i (A B), 4

(Ni1 )
DLi
O i ,
5,
i = 5.
As usual, a set of sentences L[DLi1 ] is DLi1 -inconsistent iff there are A1 , . . . , An

in , n 1, such that DLi (A1 An ) , and is DLi1 -consistent otherwise.
1
A sentence A L[DLi1 ] is DLi1 -derivable from a set L[DLi1 ] (we write DLi A) iff
1
{A} is DLi1 -inconsistent. If A is derivable from the empty set, A is called refutable.
For the semantics, let v be as before, and I be a set of PL-sentences. The truthdefinitions of DLi1 -sentences are relative to v and I , where the truth of a proposition letter
is defined as before, and the truth of O i A is defined with respect to I as given in Section 1,
truth-definitions for Boolean operators being as usual. VerDLi (I, v) denotes the subset of
1
DLi1 -sentences defined true for I , v, and we write I , v |= A for A VerDLi (I, v). A is
1

42
167
DLi1 -valid (we write |=DLi A) iff A L[DLi1 ] is true for arbitrary I , v, and L[DLi1 ]
1
DLi1 -entails A (we write |=DLi A) iff I , v |= A for all I , v such that I , v |= B for all
1
B . If there is a pair I , v such that I , v |= B for all B L[DLi1 ], we call

DLi1 -satisfiable. Note that Boolean valuations v are introduced just to give non-deontic
formulae their usual interpretation, and the set of sentences I is not made relative to one
such valuation. If imperatives are meant to arise in specific circumstances only, this is a
matter of proper representation of (conditional) imperatives, for which here no means are
provided. So for example p1 O 1 p2 states that either p1 is true, or that p2 is what
some imperative demands, regardless of the situation.
Theorem 2.1. Each system DLi1 , 1
5, is sound and complete.
Proof. Soundness is trivial with respect to the truth definitions employed. For completeness, we prove equivalently that if is DLi1 -consistent then is DLi1 -satisfiable:
Let A1 , A2 , . . . be a fixed enumeration of L[DLi1 ]. Let = n n , where 0 = ,
and
n+1 =
n {An+1 },
if this is consistent,
n {An+1 }, otherwise.
Clearly each n must be DLi1 -consistent with either An+1 or An+1 , hence (i) is DLi1 consistent, (ii) for all A L[DLi1 ] either A or A . Now we define:
v=
1, if p ,
0, otherwise,
I = {A | O i A }.
We prove I, v |= A iff A by induction on A. We only give the case for A = O i B, the

others are trivial: Suppose O i B . Then B I and for all operators O i , I, v |= O i B.
/ , so O i B :
Suppose O i B
i = 1:
i = 2:
i = 3:
Assume I, v |= O 1 B, so B I, so O 1 B , but then is DL11 -inconsistent.

Assume I, v |= O 2 B, so there is a C I such that |=PL B C. So O 2 C .
But from (Ext21 ) and PL-completeness we have DL2 O 2 B O 2 C, so is DL21 1
inconsistent.
Assume I, v |= O 3 B, so there is a C I such that |=PL C B, so O 3 C .
But from (Ext31 ), (M31 ), and PL-completeness we have DL3 O 3 C O 3 B, so
1
i = 4:
is DL31 -inconsistent.
Assume I, v |= O 4 B, so there is a non-empty finite set {C1 , . . . , Cn } I such that
|=PL (C1 Cn ) B. So {O 4 C1 , . . . , O 4 Cn } . But from (Ext41 ), (C41 ),
(M41 ), and PL-completeness we obtain DL4 (O 4 C1 O 4 Cn ) O 4 B, so
1
i = 5:
Assume I, v |= O 5 B, so I |=PL B. If I = , then by strong completeness of PL
there is a non-empty finite set {C1 , . . . , Cn } I such that |=PL (C1 Cn )
B, so the r.a.a. is done as in case i = 4. If I = then DL5 O 5 B by (Ext51 ), (N41 ),
so is DL51 -inconsistent.
168
43
Op
3. Combined logics DL1 for operators O15

The definitions of operators O 15 all refer to a set I . Instead of having languages
L[DLi1 ] that make use of just one of these operators, it makes sense to permit mixed statements such as O 4 p1 O 3 p1 which expresses that though p1 is not implied by what a
single imperative demands, it is implied by a finite subset of all imperative demands. So let
Op
Op be a subset of operator indices {1, 2, 3, 4, 5}, card(Op) > 1, and let L[DL1 ] be like
i
i
L[DL1 ] except that the alphabet contains all of O , i Op, and clause b) in the definition
of sentences now reads
Op
b) if A L[PL] then O i A L[DL1 ], for all i Op.

Op
Let the description of systems DL1 follow that of DLi1 , where the clause a) is now
Op
relative to the language L[DL1 ], and clauses b), c) hold for all i Op satisfying the
additional requirements. We have the following additional axiom schemes:
(Oi Oj ) DLOp O i A O j A, {i, j } Op, i < j ,
1
(O3 O2 ) DLOp O 3 O 2 , {2, 3} Op,

1
(O4 O3 ) DLOp O 4 O 3 , {3, 4} Op,

1
(O5 O3 ) if
PL
A then DLOp O 5 A O 3 , {3, 5} Op,
(O5 O4 ) if
PL
A then DLOp O 5 A O 4 A, {4, 5} Op.
1
1
Op
Theorem 3.1. All systems DL1
are sound and weakly complete.
Proof. Soundness is again trivial with respect to O i -truth definitions. For weak completeness, we have to prove that if DLOp A then DLOp A. We assume DLF A so A is not
Op
refutable in DL1 . One disjunct in a disjunctive normal form of A is then not refutable
Op
in DL1 . From the non-deontic conjuncts of we obtain a valuation v that satisfies them.
As for the deontic conjuncts, for all i Op let Oi = {B L[PL] | O i B is a conjunct of
}, and Qi = {B L[PL] | O i B is a conjunct of }. Let 1 , 2 be proposition letters that
do not occur in . We define
1 Op (a)
I = O1
2
2 Op (b)
(B (1 1 )) | B O
(B 1 ) | B O3
3 Op (c)
2 , (2 (B 1 )) | B O4
4 Op (d)
5
2 , (2 (B 1 )) | B O and PL B
5 Op (e).
Obviously, for all B Oi , I, v
O i B. We prove that for each B Qi , I, v
O i B:
/ O1 by PL-consistency of . But all other B I contain propoB Q1 : If B I then B

sition letter 1 not occurring in any B Q1 .

44
B Q2 : For all C I by clause (a) or (b), PL (B C) by PL-consistency of and

axiom scheme (O1 O2 ). All other C I derive some contingent formula 2 or
(2 1 ). Since B contains no occurrence of 1 , 2 , if PL B C then PL
B. The C I generated by clauses (d), (e) all are contingently true or false,
so if there is a C I such that PL C it must be due to clause (c). But then
Op
Op
O 3 CDL1 -derives O 2 C by (O3 O2 ) and O 2 C DL1 -derives O 2 B by (Ext21 ), so
Op
is DL1 -refutable.
B Q3 : For all C I by clause (a), (b), or (c), PL (C B) by PL-consistency of and
axiom schemes (O1 O3 ), (O2 O3 ), (M31 ), (Ext31 ). So only by virtue of clauses (d),
(e) can there be a C I such that PL (C B). But no single C generated by
these clauses PL-implies B unless PL B. But then there is some conjunct O 4 C
Op
in , or some conjunct O 5 D with non-tautological D, from which DL1 -derives
Op
4
O 3 C by use of (O5 O3 ) or (O4 O3 ), (M41 ), (Ext1 ). So is DL1 -refutable.
B Q4 : If O 4 B is true, then I = , Oi = for some i Op. By the construction
I B iff PL ( 4i=1 Oi {C O5 | PL C}) B. For any C in
PL
4
i
5
4
i 4
5
i=1 O {C O | PL C} we derive O C by (O O ), i Op, and O B using
Op
4
4
(M1 ), (Ext1 ), so is DL1 -refutable.
Op
B Q5 : If PL B then is DL1 -refutable by (N51 ), so again I = , Oi = for some
i Op. The proof relies on axiom schemes (Oi O5 ), i Op, and is done as in
case Q4 .
Op
Theorem 3.2. No DL1 , Op = {4, 5}, is strongly complete.

Op
Proof. The semantics of DL1 for Op = {4, 5} is not compact, i.e., there are finitely
Op
satisfiable sets L[DL1 ] such that is not satisfiable: Suppose {i, j } Op. We give
non-satisfiable sets and an If for each finite f such that If , v satisfies f , v being
arbitrary.
i = 1, j = 2: Let = {O 1 A ||=PL A p1 } {O 2 p1 }. For any finite subset f of
there is a proposition letter that does not occur in any A f . Each set f is
satisfied by If = {(( ) p1 )}.
i = 1, 2, j = 3: Let = {O i A ||=PL A p1 } {O 3 p1 }. Let f be any finite subset,
and be as before. Each set f is satisfied by If = {( p1 )}.
i = 1, 2, 3, j = 4, 5: Let = {O i A PL A} {O j p1 }. Let f be any finite subset, and
be as before. Each set f is satisfied by If = {( p1 ), }.
{4,5}
Theorem 3.3. DL1
is (strongly) complete.
Proof. We adapt the completeness proof of DL41 in Theorem 2.1: is now defined with
{4,5}
{4,5}
respect to an enumeration of DL1 -sentences and DL1 -consistency. I , v is constructed
4
4
as for DL1 , in particular I = {A | O A }. The proof that I , v |= O 4 A iff O 4 A is as
before. To prove that I, v |= O 5 A iff O 5 A , assume first O 5 A . Either O 4 A ,
169
170
45
but then A I , and I |=PL A is trivial. Or O 4 A

/ , then O 4 A , but then A must
5
4
be tautological, for otherwise by use of (O O ) the finite subset {O 4 A, O 5 A} is
{4,5}
DL1 -inconsistent. So |=PL A and I, v |= O 5 A is true. For the other direction assume
5
O A
/ , so O 5 A . For r.a.a. assume I, v |= O 5 A, so I |=PL A. If I = then I, v |=
O 4 A, so by completeness of PL O 4 A , but then due to (O4 O5 ) the set {O 5 A, O 4 A}
{4,5}
is DL1 -inconsistent. If I = then A is tautological, so by using (N51 ), (Ext51 ){O 5 A}
{4,5}
4. DLi1 -logics and semantic restrictions
So far, a set of sentences has been used to model what a set of imperatives demands,
there being neither restrictions on the size of I , nor restrictions on the logical type of
sentences in I . But concerning the size of I , it may be argued that a deontic logic should
be applied only if there is at least one demand to be considered, since reasoning about what
one ought to do when nothing is explicitly obligatory seems a borderline case. Concerning
the elements of I , it may be argued that imperatives that demand the logically impossible
or logically necessary are not proper, and one may also want to use a rationality restraint
to the effect that what the imperatives demand should be (jointly) satisfiable. Consider the
following restrictions:
[R-0]
[R-1]
[R-2]
[R-3]
I = (Non-Triviality),
A I : BL A (Excluded Impossibility),
I BL (Collective Satisfiability),
A I : BL A (Excluded Necessity).
Consider the logics DLi1 defined in Section 2 and the following additional axiom
schemes:
(XIi1 ) if PL A then DLi O i A,
1
(Di1 ) if PL (A1 An ) then DLi (O i A1 O i An ),

1
(XNi1 ) if PL A then DLi O i A.

1
Warranted by the following theorems, the table below lists, for languages L[DLi1 ],
1 i 5, axiom systems sound and complete with respect to DLi1 -semantics conforming
to any of the above restrictions individually (square brackets indicate weak completeness
only):
L[DLi1 ] i = 1
[R-0]
[R-1]
[R-2]
[R-3]
i=2
i =3 i =4 i =5
DLi1 + (Ni1 ) DL51

i
i
DL1 + (XI1 )
DLi1
[DLi1 ]
DLi1 + (Di1 )
DLi1 + (XNi1 )
[DLi1 ]
DL51

46
Theorem 4.1. The following systems are sound and (strongly) complete:
1.
2.
3.
4.
5.
6.
7.
DLi1 + (Ni1 ) for DLi1 -semantics restricted by [R-0], i = 3, 4.

DL51 for DL51 -semantics restricted by [R-0].
DLi1 + (XIi1 ) for DLi1 -semantics restricted by [R-1], i = 1, 2, 3.
DLi1 for DLi1 -semantics restricted by [R-1], i = 4, 5.
DLi1 + (Di1 ) for DLi1 -semantics restricted by [R-2], i = 1, 2, 3, 4, 5.
DLi1 + (XNi1 ) for DLi1 -semantics restricted by [R-3], i = 1, 2.
DL51 for DL51 -semantics restricted by [R-3].
Proof. Soundness is again obvious. The completeness proofs are forthcoming adaptations
of that for the DLi1 -systems. When there is a specific axiom, it obviously ensures that the
constructed I conforms to the given restriction. Consider cases 2, 4, 7 where there are no
specific axioms:
Case 2: O 5 by (N51 ), so by the construction I = .
Case 4: We construct as before, but now define I = {A | O i A and PL A}. Suppose O i A and A
/ I , so |=PL A. Then |=PL A p1 , |=PL A p1 ,
so by (Exti1 ) we have O i p, O i p1 , so {p1 , p1 } I, so I |=PL A. The
remainder is as before.
Case 7: Let be as before, but now I = {A | O 5 A and PL A}. Suppose O 5 A
and A
/ I , so |=PL A. But then trivially I |=PL A, and the remainder of the proof
is as before.
Theorem 4.2. DLi1 -semantics, i = 1, 2, is not compact if restricted by [R-0]. DLi1 semantics, i = 3, 4, is not compact if restricted by [R-3].
Proof. To show that DLi1 -semantics, i = 1, 2, is not compact if restricted by [R-0], let
= {O i A | A L[DLi1 ]}. To satisfy all of , I must be empty, which is excluded by
[R-0], so is not satisfiable by [R-0]-restricted semantics. But any finite subset f
is satisfied by If , v, where If = {}, and is a proposition letter that does not occur in
any A f , v being arbitrary.
For the proof that DLi1 -semantics, i = 3, 4, is not compact if restricted by [R-3], let
= {O i A PL A} {O i }. To satisfy all of , I must contain tautologies only, which
is excluded by [R-3], yet it also cannot be empty due to the expected truth of O i , so
is not satisfiable by [R-3]-restricted semantics. But any f is again satisfied by If , v, as
described above.
Theorem 4.3. The following systems are sound and weakly complete:
1. DLi1 for DLi1 -semantics restricted by [R-0], i = 1, 2.
2. DLi1 for DLi1 -semantics restricted by [R-3], i = 3, 4.
171

172
47
Proof. Soundness is again obvious. For weak completeness, suppose {A} is DLi1 consistent. To demonstrate that there are I, v such that I, v A, we do a construction as in
the proof of Theorem 3.1, i.e., form a disjunctive normal form of A, from the non-deontic
conjuncts of a non-refutable disjunct obtain a valuation v that satisfies these, and from the
deontic conjuncts construct sets Oi and Qi as before. Again, let be a proposition letter
not occurring in A.
Case 1: Let
I=
Oi , if Oi = ,
{}, otherwise.
The construction ensures that I = (R-0), and for all B Oi , I, v O i B. If i = 1 and

B Q1 , then B
/ O1 by PL-consistency of {A}, and B = since does not occur in A,
so B
/ I and I, v O 1 B. If i = 2 and B Q2 , then no C O2 is PL-equivalent to B due
to (Ext21 ) and PL-consistency of {A}, and no formula not containing is PL-equivalent
to . So there is no C I s.t. PL (B C), and I, v O 2 B.
Case 2: Let
Oi {A L[PL] | PL A}, if this is non-empty,

I = {}, otherwise and if Qi {A L[PL] | PL A} = ,
, otherwise.
Obviously I contains no tautological element (R-3). Suppose B Oi . If there is a nontautological C Oi , for all B Oi , tautological or non-tautological, I, v O i B. If there
are tautological B Oi (only), there cannot be a tautological C Qi due to (Exti1 ) and
PL-consistency of {A}, so I and again I, v O i B. For each B Qi , I, v O i B:
B Q3 : If there is a tautological C Q3 , O3 = due to (Ext31 ), (M31 ) and PL-consistency
of {A}, and I = by the above construction, so I, v O 3 B. Otherwise, no
C O3 PL-implies B again due to (Ext31 ), (M31 ) and PL-consistency of {A}, and
if does then B is tautological, but this was excluded.
B Q4 : As before, if C Q4 is tautological then O4 = and I = , so I, v O 4 B.
Otherwise, B is neither implied by any {C1 , . . . , Cn } O4 due to (Ext41 ), (M41 ),
(C41 ) and PL-consistency of {A}, nor by I = {} since a tautological B was excluded.
If the semantics employs more than one of the above restrictions, then generally there is
a sound and (strongly) complete axiomatic system iff each of the semantics with just one
of these restrictions has such a system, and the axiomatic system is obtained by including
all axioms and rules of the systems for the singularly restricted semantics. However, this
rule fails in one case:
Theorem 4.4. DL51 -semantics is not compact if restricted by [R-0], [R-3].
Proof. To show incompactness, let = {O 5 A PL A}. To satisfy all of , I cannot
contain anything but tautologies, which is excluded by [R-3], yet it also cannot be empty
due to [R-0], so is not satisfiable by [R-0][R-3]-restricted semantics. But any finite

48
subset f is again satisfied by If , v, where If = {}, being as before, and v being

arbitrary.
Theorem 4.5. DL51 is sound and weakly complete for [R-0][R-3]-restricted DL51 -semantics.
Proof. Soundness is again obvious. For completeness we assume that {A} is DL51 consistent and construct an I, v such that I, v A as in the proof of Theorem 4.3, now
letting
I=
Oi {A L[PL] |
{},
PL
A}, if this is non-empty,

otherwise.
Obviously I satisfies [R-0] and [R-3]. For all B O5 , if B is non-tautological then B I

and I, v O 5 B, and trivially so if B is tautological. Suppose B Q5 : B cannot be tautological due to (N41 ) and PL-consistency of {A}, so also B cannot be PL-implied by {},
so if I PL B by completeness of PL there must be some {C1 , . . . , Cn } O5 such that
PL ((C1 Cn ) B), but this is excluded by (Ext41 ), (M41 ), (C41 ) and PL-consistency
of {A}.
5. Van Fraassens imperative logic

Van Fraassen [21] discusses three truth definitions of O-operators with respect to given
imperatives. For all these operators, axiom scheme (C) does not hold, for van Fraassen is
concerned with possibly conflicting imperatives, and argues that when there is a demand
for A, and a demand for A, we should admit the truth of OA OA, but the derivation
of O(A A) should be blocked. In van Fraassens semantics, to each imperative i there
is a class of possible outcomes i + in which i is fulfilled. His first definition reads:
[Df-F1] OA is true iff, for some imperative i that is in force, i + is part of the set of
possible outcomes in which A is true.
In the terms used here, van Fraassens definition may be reformulated as
[Df-F1 ] I |= OA iff B I :|=PL B A.
Since van Fraassen also presupposes that no single imperative may be impossible to
fulfill, van Fraassens first logic coincides with DL31 + (XI31 ) above. However, van Fraassen
thinks definition [Df-F1] too simple minded, for two reasons: For one, imperatives are
conditional, and according to van Fraassen a conditional imperative can be fulfilled or
violated only if its condition is the case. His second logic therefore introduces conditional
imperatives and a corresponding dyadic O-operator. I leave this definition aside, since we
are not concerned with conditional imperatives here. The other argument against [Df-F1]
is that it should be allowed to draw conclusions from imperatives that do not conflict,
e.g., from two imperatives that demand p1 p2 and p2 respectively conclude that p1
173
174
49
is obligatory. But since there is no single imperative that demands p1 , [Df-F1] does not
allow the derivation of Op1 . Van Fraassen therefore introduces the notion of a score of an
outcome v, i.e., the set of imperatives that v fulfills, and defines:
[Df-F3] OA is true iff there is a possible state of affairs v in A whose score is not
included in the score of any v in A .
Again, this definition may be reformulated in the terms used here:
[Df-F3 ] I |= O F A iff v A : v A : {B I | v |= B}
{B I | v |= B}.
Let the set of maximally PL-consistent subsets I of I be the set of subsets I I such
that (i) I , and (ii) there is no I such that I I I and I . It can then be
proven that [Df-F3 ] is equivalent to [Df-F3 ] (cf. Horty [12, p. 30, Theorem 2]):
[Df-F3 ] I |= O F A iff I I : I PL A.
Let the language L[DLF1 ] be like any of the L[DLi1 ], except that O F replaces O i in
all DLi1 -sentences. Let the axiomatic system DLF1 be like DL31 + (XI31 ) + (N31 ), except
that O F replaces O 3 . Let DLF1 -semantics be defined like DL31 -semantics, except that truth
definition [Df-F3 ] replaces that for O 3 . We then have the following results:
Theorem 5.1. DLF1 is sound and weakly complete.
Proof. Concerning soundness, the validity of (Ext1F1 ), (MF1 ), (XIF1 ) and (NF1 ) is immediate. Concerning (weak) completeness, we have to prove that if DLF A then DLF A. We
1
assume DLF A so A is not refutable in DLF1 . As before (cf. Theorem 3.1), let be a
1
non-refutable disjunct in a disjunctive normal form of A. From its non-deontic conjuncts
we again obtain a valuation v that satisfies these. Concerning the deontic conjuncts of ,
let OF be the set of PL-sentences B such that O F B is a conjunct of , and QF be the set
of PL-sentences B such that O F B is a conjunct of . For any set L[PL] let
minPL = {A | B : if PL B A then PL B A}.
Let n = card(minPL OF ). Let {1 , . . . , n } Prop be an arbitrary set of n proposition
letters not occurring in , and be a function that maps minPL OF onto the set {1 , 1
2 , . . . , 1 n1 n } of n mutually exclusive PL-sentences. We then define:
I = B (B) | B minPL OF .
Note that I = {{i} | i I }, since (B) is PL-inconsistent with any (C), B, C
minPL OF and B = C. No B OF is a contradiction due to (XIF1 ), and for each consistent
B OF the definition of minPL OF ensures that there is an I I such that I PL B,
so O F B is true. Suppose B QF but I, v O F B, so there is an I I : I PL B.
Then there is a C minPL OF such that I = {C (C)} and PL C (C) B. Since

50
(C) contains no sentence letters occurring in either B or C, PL C B. But then O F C

DLF1 -derives from O F B by (MF1 ), (ExtF1 ), so is DLF1 -inconsistent.
Theorem 5.2. There is no strongly complete axiomatic system DLF1 .
Proof. Let
= O F p1 , O F p2 , O F (p1 p2 ), O F p1 , O F (p1 p2 ),
O F (p1 p2 p3 ), . . . .
is finitely satisfiable: Let be a proposition letter that does not occur in some sentence of
a finite f . Then I = {p1 , p2 } satisfies f . However, is not satisfiable:
Suppose there is a set I that satisfies . O F p1 , O F p2 , so there are PL-consistent
finite sets 1 , 2 I that derive p1 and p2 respectively. 1 2 is inconsistent, since
otherwise there is some I I such that 1 2 I and I p1 p2 , so I would
not satisfy O F (p1 p2 ). Now PL ( 1 p1 pn ) for any n 1, for
otherwise 1 PL (p1 pn ) would contradict O F (p1 pn ). The same
holds for 2 . Let n be the highest index of proposition letters occurring in 2 . Since 2
{p1 , p2 , . . . , pn } is PL-consistent, i.e., (p1 pn ) is the first disjunct in the disjunctive
normal form of 2 , we have PL ( 2 p1 pn ) (p1 pn ). So we
obtain PL ( 1 2 p1 pn ) . But 1 2 was PL-inconsistent.
After explaining that semantics based on [Df-F3] satisfies the basic criteria (the axiom
schemes of DL31 + (XI31 ), van Fraassen notes that in the semantics we have the following
additional truth: If there are two imperatives i1 , i2 in force, and if there is a v that belongs
to i1+ and also to i2+ , then if [all such] v A , O F A is true. He then asks:
But can this happy circumstance be reflected in the logic of the ought-statements alone?
Or can it be expressed only in a language in which we can talk directly about the imperatives as well? This is an important question because it is the question whether the
inferential structure of the ought language game can be stated in so simple a manner
that it can be grasped in and by itself.
If what van Fraassen means by the logic of the ought-statements alone is a monotonic
axiomatic system, then van Fraassens first question deserves a negative answer: There is
a logic of ought-statements DLF1 that is weakly complete, but the only additional truth if
compared to [Df-F1] is O F .3 In particular, DLF1 does not provide the inferences van
Fraassen desires. And Theorem 5.2 shows that there is no strongly complete axiomatic
3 Van Fraassen seems to have overlooked the additional truth of O F . To avoid its truth, the O 5 -type definition [Df-F3 ] may be changed into a more O 4 -type definition:
I |= O F A
iff
I I : B1 , . . . , Bn I |=BL B1 Bn A.
It can then be shown that DL31 + (XI13 ) is sound and weakly complete with respect to the changed definition, and
the counterexample in Theorem 5.2 again disproves compactness.
175
176
51
system in terms of the O-operator used. But, regarding his additional question, it seems to
me that the ought language game extends to operators like O 1 or O 2 that more directly
talk about imperatives. So perhaps a more powerful language can provide the missing
inferences. It is immediate that the counterexample used in Theorem 3.2 to disprove com{2,4}
pactness of DL1
also disproves compactness of any semantics that employs operators
1
2
of type O or O in addition to van Fraassens operator O F . So for an improved characterization of van Fraassens third semantics it must suffice to give a weakly complete
system:
{2,F }
{2,3}
Let L[DL1 ] be the language that is like L[DL1 ] except that O F replaces O 3 in
{2,3}
all DL1 -sentences. For the semantics, the truth of O 2 , O F is defined with respect to a
set I as above, and for proposition letters and Boolean connectives it is defined as usual.
{2,F }
{2,F }
Let the axiomatic system DL1
contain all instances of L[DL1 ] in PL-theorems, be
2
F
closed under modus ponens and (Ext1 ), (Ext1 ), contain all L[PL]-instances in the axiom
schemes (MF1 ), (XIF1 ), (NF1 ) of DLF1 , and in the following schemes:
(F-1) If
PL
(F-2) DL{2,F } (
n
n
2
i=1 Ai then DL{2,F }
i=1 O Ai
1
n
2
F
F
i=1 O Ai O B O (B
n
i=1 Ai ,
n
F
i=1 Ai )) O (B
OF
n
i=1 Ai ).
According to (F-1), agglomeration of contents of O 2 -obligations is admissible if these

are jointly PL-consistent.4 (F-2) then states that if O 2 -contents may not be added to the
content of an O F -obligation, then there is an obligation to the contrary that goes with that
content. So (F-1) and (F-2) now properly define the simple cases, where agglomeration
of contents is admissible.
{2,F }
Theorem 5.3. DL1
is sound and weakly complete.
Proof. For soundness of the new axiom schemes, validity of (F-1) is immediate from the
fact that if there are B1 , . . . , Bn I equivalent to A1 , . . . , An respectively, and A1 , . . . , An
are PL-consistent, then some I I contains {B1 , . . . , Bn }, so I PL A1 An .
For the validity of (F-2), suppose O 2 A1 , . . . , O 2 An . If O F (B ni=1 Ai ), then no I
I may derive B and all Ai , so if O F B is true and there is an I that derives B, it must
be inconsistent with ni=1 Ai and consequently derive B ni=1 Ai .
{2,F }
{2,F }
For (weak) completeness, we have to prove that if A is DL1 -valid, then A DL1 .
{2,F }
{2,F }
We assume A
/ DL1 , so A is not refutable in DL1 . As now usual, let be a nonrefutable disjunct in a disjunctive normal form of A. From its non-deontic conjuncts we
4 A difficulty in axiomatizing van Fraassens semantics was pointed out by Horty [11, p. 50]: If background
imperatives are coded into ought-statements, as they are here using operator O 2 , then the set {O 2 A | A L[BL]}
must derive O F B for any BL-consistent B, though for some basic logics like first order logic these are not recur{2,F }
{2,F }
-axioms is not decidable and DL1
not recursively
sively enumerable. But then due to (F-1) the set of DL1
enumerable, so for such basic logics no axiomatization in the usual sense is provided here.

52
again obtain a valuation v that satisfies them. As for the deontic conjuncts of , first let
O2 = B L[PL] | O 2 B is a conjunct of ,
Q2 = B L[PL] | O 2 B is a conjunct of .
Let L[PL] be the set of PL-sentences that contain only proposition letters occurring in
n
some deontic conjunct of . Let r(L[PL] ) be a set of 22 mutually non-equivalent representatives of L[PL] , n being the number of proposition letters occurring in the deontic
conjuncts of . By writing PL-sentences we now mean their unique representatives in
L[PL] . We construct a finite set with the following properties:
(a) If O F B, O F C are conjuncts of , then O F B, O F C are in .
(b) For all B r(L[PL] ), either O F B or O F B .
{2,F }
(c) {} is DL1 -consistent.
In (a), we may have to replace B, C by r(L[P L] )-equivalents using (ExtF1 ).
Let:
OF = B L[PL] | O F B ,
QF = B L[PL] | O F B .
Lemma 5.1. For all B O2 : (a) if PL B then there is a C minPL OF such that PL
C B, and (b) for all C minPL OF , if {B, C} is PL-consistent, then PL C B.
Proof. (a) is immediate from (F-1) and the definition of minPL OF (cf. the proof of Theorem 5.1). Concerning (b), either O F (B C) or O F (B C) (clause (b) in the
construction of ). If O F (B C) then PL C (B C) by definition of minPL OF ,
{2,F }
so PL C B. If O F (B C) then O 2 B, O F C, O F (B C) DL1 -derives
O F (B C) by use of (F-2), so O F (B C) by clause (c) in the construction
of . But then PL C (B C) by definition of minPL OF , so {B, C} is not PLconsistent.
Let : minPL OF L[PL] be the function defined in the proof of Theorem 5.1. Let
I = O2 B (B) | B minPL OF .
Lemma 5.2. For all I I , B L[PL] : if I = and I PL B then there is a
C minPL OF such that PL C B.
Proof. Immediate from Lemma 5.1, the construction of I , and the fact that (C) contains
no proposition letters occurring in B or C.
It remains to prove that I satisfies all deontic formulas in and :
177

O2 :
Q2 :
OF :
QF :
178
53
For all B O2 we have B I , so O 2 B is true.

For all B Q2 , if C I and PL B C, then C
/ O2 , for otherwise is
refutable by (Ext21 ). All other C I derive some i not occurring in , but no
B Q2 does.
{2,F }
Unless B OF is a contradiction, which is excluded by (XIF1 ) and DL1 F
consistency of , by definition of minPL O and the construction of I there is
a I I such that I PL B, so O F B is true.
For all B QF , there is no I I such that I PL B: For otherwise there
is a C minPL OF such that PL C B (Lemma 5.2). Then O F C , and
{2,F }
O F C due to (ExtF1 ), (MF1 ), so is DL1 -inconsistent.
6. Saving a twin without guiltthe sceptical ought O S

In the long-standing philosophical dispute on the existence of moral dilemma and
conflicting obligations, Barcan Marcus [16] provided the following Buridans ass type example: Identical twins are in danger of being crushed to death by a rock. They are pinned
down such that only one can be pulled free at a time. If nothing is done, the rock will
soon kill them both, but if either twin is removed this will cause the rock to slide and
kill the other. A mountain guide is liable for the lives of both twins. What are her obligations?
As the example is set up, it suggests the co-existence of two conflicting obligations,
i.e., the guide has to save one twin (T1 ), and also save the other (T2 ), T1 contradicting T2 . In logics DL41 and DL51 we have O 4/5T1 and O 4/5 T2 , and also O 4/5t (T1 T2 ).
But then a contradiction is obligatory, so we cannot also accept the principle ought im4/5
plies can expressed by (XI1 ). In van Fraassens logic we maintain O F T1 and O F T2 ,
and derivation of O F (T1 T2 ) is blocked by the absence of (CF1 ) and inconsistency of
T1 T2 (axiom F-1). However, in addition we have O F T1 and O F T2 : by saving one
twin the mountain guide will violate her duties towards the other. So why not simply walk
away?
Conee [5], Donagan [6], and Brink [3] have argued that morally there is no conflict:
according to Conee, either act is permitted and neither absolutely obligatory, while for
Donagan and Brink there only exists an obligation to save one twin or the other, but not two
conflicting obligations to save either. Deontically, the dilemma was examined by Jacquette
[13] and Horty [12]. Jaquette points out that van Fraassens approach is unsatisfactory,
since it does not hold that an obligatory act is also permitted: O F A P F A is not a truth
in DLF1 , with P F A defined as O F A. So according to van Fraassens notion of ought,
either twin must be saved, but can only be saved at the price of guilt (for doing something
forbidden). As an alternative to van Fraassens logic, Horty, in accord with Donagans and
Brinks disjunctive proposal, considers a sceptical theory of obligations, where something
can be obligatoryS only if there is no consistent set of norms that demand the contrary. With
respect to our semantics, Hortys definition reads:
[Df-S] I |= O S A iff I I : I PL A.

54
179
So A is obligatory iff all maximally consistent subsets of imperative-associated sentences

derive A.5 Let L[DLS1 ] be like any of the L[DLi1 ], except that operator O S replaces O i .
Let the truth of DLS1 -sentences be defined with respect to [Df-S] and the usual definitions
for proposition letters and Boolean connectives. Let DLS1 be like DL51 + (XI51 ). We then
obtain:
Theorem 6.1. DLS1 is sound and (strongly) complete.
Proof. Immediate. For completeness, is constructed just as in the proof of Theorem 2.1.
The result is welcome at first, for it seems that standard deontic logic, including agglomeration (CS1 ) and ought implies can (XIS1 ), is the proper tool to represent the sceptic
notion of ought. However, the problem is that now all the problems have disappeared:
Conflicting imperatives and their impact on what is sceptically obligatory have become
imperceivable in DLS1 . To see what goes on behind the curtain, and illustrate the relation of
sceptically defined oughts with van Fraassens logic, we must again admit the use of more
than one O-operator:
{2,F,S}
{2,F }
Let the language L[DL1
] be like L[DL1 ] except that we have the additional
{2,F,S}
{2,F }
S
operator O . The truth of DL1
-sentences is defined as for DL1 -sentences with the
{2,F,S}
additional truth definition [Df-S]. Let the system DL1
contain all axiom-schemes and
{2,F }
rules of DL1
and DLS1 , and additionally the following:
(FS-1) (O F A O S B) O F (A B);
(FS-2) ( ni=1 O 2 Ai O F ( 1
(FS-3) (
(FS-4) (
n
2
i=1 O Ai
n
2
i=1 O Ai
O F
O F
k
j =1
k
j =2
j )) O S (
k
j =1
) OS
j
n
A
)
OS
i=1 i
k
j =2
j );
j ;
n
i=1 Ai .
In FS-2 and FS-3, j is a non-empty subset of {A1 , . . . , An }. Note that (FS-3), (FS-4) are
special cases of (FS-2). (FS-3) derives (O 2 A O 2 B O F (A B)) O S (A B),
which is a syntactic version of the disjunctive solution to Marcuss dilemma: the disjunction (A B) of two possibly conflicting imperative demands A, B is obligatory, unless
there is a consistent set of imperatives that demands even A B. Hence one twin or
5 It might seem awkward that if I contains contradictory imperative demands, e.g., if I = {p , p , p } then
1
1 2
with [Df-S] the set of truths is the same as for I = {p2 }, i.e., the same as for an I without these demands. If
the fulfillment of one of the demands is to remain obligatory, an alternative would be to have a disjunction of
oughts instead of the ought of a disjunction, and define:
[Df-DS] VerDS (I, v) := {A L[DL51 ] | I I : A VerDL5 (I , v)}.

1
In the above example, (O 5 p1 O 5 p1 ) is now true under the new definition, O 5 (p1 p1 ) remains valid, and
O 5 remains false. Ignoring operator indices, we have VerDLS (I, v) VerDS (I, v) VerDL5 (I, v).
1
180
55
the other must be saved. From (FS-1) and (XIF1 ) we derive O F A O S A, so what is
obligatory according to van Fraassens ought is at least permitted in the sceptical sense. If
guilt is a notion that attaches to a violation of a sceptical ought only, then one may save a
twin without guilt, but walking away remains forbidden. From the results of Section 5 it is
{2,F,S}
immediate that DL1
cannot be strongly complete. However we obtain the following
result:
{2,F,S}
Theorem 6.2. DL1
is sound and weakly complete.
Proof. Concerning soundness, for the validity of (FS-1) assume O F A, so some I I

derives A. If O S B is true, then all I I derive B, so I derives A B and O F (A B)
is true. For the validity of (FS-3) and (FS-4) suppose O 2 Ai is true for all i, 1 i n.
Concerning (FS-3), assume O F kj =1 j , so all I I are P L-consistent with
at least one disjunct, so all I contain some j , 1 j k, so all I derive kj =1 j .
Concerning (FS-4), assume O S ni=1 Ai , so some I I is P L-consistent with
n
i=1 Ai , so there is some I I that contains each Ai . For the validity of (FS-2),
assume again O 2 Ai true for all i, 1 i n, and assume O F ( 1 kj =2 j ).
Those I I that contain a set j also derive 1 kj =2 j . Consider an I
that does not contain any j : Since I is not consistent with all Ai j , it must derive
j for each j , so it derives kj =2 j . If I is consistent with 1 , it derives 1
k
j =2
j , but then O F (
k
j =2
j ) is false. So I must be inconsistent
with 1 , so it derives 1 and hence also 1 kj =2 j .

For (weak) completeness we use the construction employed in the proof of Theorem 5.3
{2,F,S}
with the following modifications: is now defined with respect to DL1
-consistency
S
(clause c in the construction of ), and additionally we have sets O and QS :
OS = B r L[PL] | {} DL{2,F,S} O S B ,
1
Q = B r L[PL]
| {} DL{2,F,S} O S B .
1
I is constructed as in the proof of Theorem 5.3 and additionally contains the following
elements: Let QS = {B1 , . . . , Bm }. For each Bj we add CI CII CIII to I , where:
CI = OS ,
CII = { | O2 and {CI } PL Bj },
CIII = 1 n n+1 n+j .
In CIII , 1 , . . . , n are the same proposition letters used to define : minPL OF L[PL]
(cf. the proofs of Theorems 5.1 and 5.3), and n+1 , . . . , n+m are some other letters not
occurring in . We have to prove that the modified I still satisfies all conjuncts of . For
the non-deontic conjuncts and conjuncts of the form O 2 B, O 2 B, O F B, the proof is as
before. Consider conjuncts of the form O F B, O S B, O S B, and the sets QF , OS , QS :

56
QF :
OS :
QS :
For all B QF , there is no I I such that I PL B: Due to Lemma 5.1

no I I can consist of elements of OS alone. If I contains a sentence of
the form C (C), C minPL OF , I cannot contain any of the new elements
in I , and the proof is as before. Suppose I contains a sentence of the form CI
CII CIII and I PL B. Let C QS be the sentence used in the construction of
CI CII CIII . First suppose CII is vacuous, i.e., O2 contains no non-empty subset
such that {CI } PL C. If CI CII CIII is the only element of I then PL
CI B, so O S B is derivable by the use of (MS1 ) and (ExtS1 ), so O F B is derivable
by the use of (NF1 ), (FS-1), so is DL2,F,S
-inconsistent. If I contains a non1
2
empty subset O and PL CI B, then O F ( CI ) due
to (MF1 ) and (ExtF1 ), and O F ( ) due to (FS-1), but then O S is
derivable by the use of (FS-4), so is inconsistent with CI CII CIII and hence
cannot be in I . Now suppose CII is not vacuous. Then O S CII is derivable with
(CS1 ), (MS1 ) and (ExtS1 ), and O F CII is derivable by the use of (FS-3), furthermore
due to (FS-1) we have O F (CI CII ) , so again if CI CII CIII is the only
-inconsistent. If OS
element of I and PL CI CII B then is DL2,F,S
1
is a non-empty subset of I such that {CI CII } PL B then O F (

CII ) due to (FS-1), so O F ( CII ) , so with (FS-2) we derive
O S ( CII ), but then is inconsistent with CI CII and not in I .
Suppose B OS . If I I contains CI CII CIII then I PL B by virtue
of conjunct CI . Otherwise I contains a sentence of the form C (C), C
minPL OF (Lemma 5.1). From O S B, O F C we obtain O F B C by the use of
(FS-1), so O F B C , and by the construction of minPL OF we have PL
C (B C), so again I PL B.
If B QS then the sentence CI CII CIII constructed for B is in I . The conjunction CI CII CIII is not PL-inconsistent: Suppose it is PL-inconsistent, and
assume O2 = , so CII is vacuous and PL CI , but this is excluded by (CS1 )
and (XIS1 ). Or O2 = , but then PL CI CII , and since PL CI CII B
by the construction of CII , we have PL CI B, so O S B is DL{2,F,S} -derivable
by (CS1 ), (MS1 ), and (ExtS1 ), and refutable. Likewise CI CII CIII does not
derive B, for otherwise PL CI CII B, PL CI CII B, and again
PL CI B. So some I I contains CI CII CIII which does not by
itself derive B. Elements of the form C (C), C minPL OF are excluded from
I by CII . If there is some non-empty O2 such that {CI CII } PL B
then since also PL CI CII B we have {CI } PL B, but then consistency of with CI CII CIII is ruled out by CII . So one I I does not
PL-derive B.
7. Logics for extended languages L[DLi ]

Let us return to the logics DLi1 treated in Section 2. In the languages L[DLi1 ], the
scope of deontic operators is restricted to sentences of propositional logicformulas like
O 4 O 4 p1 , O 5 (p2 O 5 p3 ), in which deontic operators appear iterated or nested, are not
181
182
57
well-formed. But semantically, this exclusion is artificial: I is a set of descriptive sentences,

and sentences like O 4 p1 are descriptive, stating that according to what the imperatives
demand, p1 must hold. Natural language gives evidence that imperative expressions may
grammatically well-formed and meaningfullyrefer to the existence of other demands or
permissions. Consider, e.g.:
(1) Applicants shall be asked to produce statements in duplicate at their own costs.
(2) Students over 15y may be permitted to leave the grounds during breaks.
To express a possible appearance of deontic notions within the scope of deontic operators, let the extended languages L[DLi ], 1 i 5, be the smallest sets such that
(a) if A L[PL] then A L[DLi ],
(b) if A, B L[DLi ], so are O i A, A, (A B), (A B), (A B), and (A B).
Semantically, we again employ a pair I, v, where I L[DLi ]. The truth of proposition
letters is as usual, and for the deontic modalities we like to use the truth definitions given
in Section 1. Yet there is a snag: Since any A L[DLi ] may appear in the scope of deontic
operators, the basic logic BL appearing in the definitions of operators O 25 cannot be
propositional logic. We would like to make |=BL the entailment relation of the thus defined
DLi -semantics itself, but at the present point this creates a circle. We must therefore take
a detour: First we define the deontic degree (A) of any A L[DLi ] ( is any two-place
Boolean operator):
(i)
(p) = 0,
(ii)
(A) = (A),
(iii)
(A B) =
(iv)
(O i A) = (A) + 1.
(A),
(B),
if (A) (B),
otherwise,
For each i, 1 i 5, we define sublanguages L[DLin ] = {A L[DLi ] | (A) n},

0. Obviously, L[DLi0 ] = L[PL], L[DLin ] L[DLin+1 ], and L[DLi ] = n L[DLin ].
For DLin -semantics, the truth of each A L[DLin ] is then defined as usual with respect
to a valuation v, and (for n 1) with respect to a set I L[DLin1 ]. In the DLin -truth
definitions of operators O 25 , the entailment relation |=BL is that of DLin1 . The axiomatic systems DLin , n 1, are defined like DLi1 , except that we allow all DLin -instances
of PL-tautologies and DLin1 -instances of the relevant axiom schemes (Min ), (Cin ), or
(Nin ). (Exti1 ) now reads:
n
(Extin ) If DLi
n1
A B then DLi O i A O i B, 2
Theorem 7.1. All systems DLin , 1

DLin -semantics.
5, n
5.
1, are sound and complete with respect to

58
183
Proof. The proof proceeds just like for DLi1 . Note that strong completeness is proved
inductively for DLin , using strong completeness of DLin1 .
Let the axiomatic systems DLi be like DLin , except we now allow all DLi -instances in
PL-tautologies and relevant axiom schemes (Mi ), (Ci ), or (Ni ), and (Exti ) now reads:
(Exti ) If DLi A B then DLi O i A O i B, 2
5.
Theorem 7.2. For all A L[DLi ]: DLi A iff DLi A, n = (A).

n
Proof. For the left-to-right direction we do an induction on the construction of DLi : Suppose DLi A. Induction basis: Assume A DLi by instantiation of sentences B1 , . . . , Bk
into a PL-theorem. Then {B1 , . . . , Bk } L[DLin ], so there is the same instance for DLin .
Assume A DLi by instantiation of sentences B1 , B2 into any of the DLi -axiom-schemes
(Mi ), (Ci ), (Ni ) or (Exti ). Then B1 , B2 L[DLin1 ], so there is the same instance for
DLin . The induction step for modus ponens is trivial. For the right-to-left direction do a
double induction on the construction of DLin : Suppose DLi A and suppose n = 1. Inducn
tion basis: Assume A DLi1 by instantiation of sentences B1 , . . . , Bk into a PL-theorem.

Then {B1 , . . . , Bk } L[DLi1 ] L[DLi ], so there is the same instance for DLi . Assume
A DLi1 by instantiation of sentences B1 , B2 into any of the DLi1 -axiom-schemes (Mi1 ),
(Ci1 ), or (Ni1 ). Then B1 , B2 L[PL] L[DLi ], so there is the same instance for DLi . Assume A DLi1 by an application of (Exti1 ). So (Exti1 ) is applied on a PL-theorem, which
is also in DLi . The induction step for modus ponens is again trivial. Suppose n 1. The
only non-trivial case is that A DLin by an application of (Extin ): Then (Extin ) is applied
to a DLin1 -theorem, which by the induction hypothesis is in DLi .
Corollary 7.1. For all 1
5, n
1: DLin DLin+1 and DLi =
i
n DLn .
Proof. DLi = n DLin is immediate from Theorem 7.2, and DLin DLin+1 is proved by
induction on the construction of DLin just as in the right-to-left proof of Theorem 7.2.
We are now able to give proper truth definitions for the deontic modalities of L[DLi ]:
I , v |= p
I , v |= O 1 A
I , v |= O 2 A
I , v |= O 3 A
I , v |= O 4 A
I , v |= O 5 A
iff
iff
iff
iff
iff
iff
v(p) = 1;
A I;
n N: B I L[DL2n ]: |=DL2 B A;
n
n N: B I L[DL3n ]: |=DL3 B A;
n
n N: B1 , . . . , Bk I L[DL4n ]: |=DL4 (B1 Bk ) A;
n
n N: I L[DL5n ] |=DL5 A.
n
Theorem 7.3. DLi is sound and complete with respect to DLi -semantics.
184
59
Proof. For soundness, the validity of DLi -instances in PL-theorems is trivial, and so is
the validity of (Mi ). The validity of (Ci ) is easily established using Corollary 7.1 as well
as the soundness and completeness of each DLin . For the DL4 -validity of (Ext4 ), suppose
DL4 A B, so by Theorem 7.2 DL4 A B for n = (A B). So |=DL4 A B
n
by completeness of DL4n . Assume I, v |= O 4 A. So there are B1 , . . . , Bk I s.t. |=DLi

m
B1 Bk A for some m N. Suppose m n. Then we have DL4 A B by
m
Corollary 7.1, so |=DLi B1 Bk B by the truth definitions for Boolean operators.
m
So also I, v |= O 4 B. Suppose n
m. Then |=DLi B1 Bk A by Corollary 7.1, so

n
we also have |=DLi B1 Bk B and I, v |= O 4 B. The assumption I, v |= O 4 B is

n
done likewise, so I, v |= O 4 A O 4 B, and |=DL4 O 4 A O 4 B since I, v were arbitrary.

The proof of DL5 -validity of (Ext5 ) is done similarly.
The completeness of DLi is immediate for i = 1, 2, 3 due to Theorems 7.1, 7.2, so
consider case i = 4. As in the completeness proof for DL41 , we construct a maximally
consistent set such that for all A L[DL4 ] either O 4 A or O 4 A . In order to prove this set satisfiable, we let I = {A L[DL4 ] | O 4 A }, and define v as
before (proof of Theorem 2.1). We have to show O 4 A iff I, v |= O 4 A. The leftto-right direction is again trivial. For the right-to-left direction suppose O 4 A
/ , so
O 4 A , and I, v |= O 4 A. Then there are B1 , . . . , Bk I s.t. |=DL4 B1 Bk A
n
for some n N. By completeness of DL4n we have DL4 B1 Bk A, so by

the use of (Ext4n+1 ) we obtain DL4
n+1
O 4 B1 O 4 Bk O 4 A and from Theo-
rem 7.2 we likewise have DL4 O 4 B1 O 4 Bk O 4 A. Since B1 , . . . , Bk I we

have {O 4 B1 , . . . , O 4 Bk , O 4 A} , and DL4 (O 4 B1 O 4 Bk O 4 A) . So
the r.a.a. makes inconsistent. Case i = 5 is done similarly, using strong completeness
of DL5n .
Corollary 7.2. For all A L[DLi ]: |=DLi A iff |=DLi A, n = (A).
n
Proof. Immediate from Theorems 7.2 and 7.3.

Corollary 7.3. The following equivalences hold:
I, v |= O 1 A
I, v |= O 2 A
I, v |= O 3 A
I, v |= O 4 A
I, v |= O 5 A
iff
iff
iff
iff
iff
A I,
B I : |=DL2 B A,
B I : |=DL3 B A,
B1 , . . . , Bk I : |=DL4 (B1 Bk ) A,
I |=DL5 A.
Proof. Immediate from the truth definitions for O i -operators and Corollary 7.2. So the
truth definitions of O i -operators are equivalent to taking DLi as basic logic.
Consider the following DLi -sentences:

60
(DT) O i (O i A A);
(D4) O i A O i O i A;
(D4c ) O i O i A O i A;
(DB) O i (A O i O i A);
(DE) O i A O i O i A.
None is a theorem of any DLi . To refute (D4c ) in DL5 , consider the set I = {O 5 p1 }. We
have I |=DL5 O 5 p1 , but not I |=DL5 p1 , so I, v |= O 5 O 5 p1 is true, and I, v |= O 5 p1 false
for any v. So I, v |= O 5 O 5 p1 O 5 p1 , so DL5 O 5 O 5 p1 O 5 p1 and DL5 O 5 O 5 p1
O 5 p1 by completeness of DL5 . This also refutes (DT) since O 5 (O 5 p1 p1 ) DL5 derives O 5 O 5 p1 O 5 p1 by the use of (C5 ), (Ext5 ), and (M5 ). The same set I refutes
theoremhood of (DB), I = {p1 } refutes (D4), and, e.g., an empty set I refutes (DE).
Now (DT is widely accepted in deontic logics modelled prohairetically. In semantics
employing a deontic ideality relation R, (DT) is valid if R is semi-reflexive, i.e., any
world ideal from some standpoint is ideal with respect to itself. In Hanssons analysis of
monadic deontic logic, where a set of deontic formulas is modelled by a non-empty basis of
ideal among possible worlds, (DT)if admitted as well-formedis immediate from the
universality of ideality (cf. [9, pp. 381382]). The same is true for Andersonian subsets of
sanction-free worlds within a set of worlds linked by an alethic accessibility relation, where
(DT) is derivable as soon as alethic necessities are made to hold at the actual world (cf.
[2, p. 187, Theorem OM45]). Fllesdal and Hilpinen [7] have called (DT) a plausiblelooking candidate for logical truth. And should not, as Prior [18, pp. 255256] has
claimed, the acceptance of (DT) be trivial, reading it as It ought to be the case that what
ought to be the case is the case?
The non-derivability of (DT) in any system DLi brings to the fore the distinction between what Barcan Marcus [15] has called the descriptive and evaluative uses of ought.
Our definitions of O i are descriptive, not evaluative: something is obligatory only if it is
(at least) derivable from what has been (factually) commanded. Thus O 3 O 3 p1 means that
there is an imperative that can be satisfied only if O 3 p1 is true. But having no analytical
content, a violation of this imperative is not excluded, which means that O 3 p1 is false, i.e.,
that there is, as a matter of fact, no imperative that demandsamong other thingsp1 . To
use Marcuss example: that parking on highways ought to be forbidden does not entail, in
the semantics employed here, that parking on highways is forbidden.
For similar reasons, factual statements about the set of imperative demands cannot be
meaningfully employed for conditional commands. Consider the sentence O 3 (O 3 A
A). Does it entail O 3 A? Let I = {O 3 p1 p1 }, so I, v |= O 3 (O 3 p1 p1 ) is true,
whatever v is. However, DL3 (O 3 p1 p1 ) O 3 p1 , as can be seen from I = ,
v (p1 ) = 1. So I, v O 3 p1 . Possibly one may want to admit commands like If it is not
(already) obligatory to do , then do (by virtue of this command)!, and the addressee
should be able to conclude that (now) she has an obligation to do . However, such conditional expressions are not expressible in the language and semantics presented so far.
Exploring how imperative semantics can be extended to cover such and other conditional
expressions, and how to sail around the maelstrom of self-referentiality lurking behind the
above example, must be left to further study.
185

186
61
8. Conclusion
When the question is what ought to be, according to some existing norms or imperatives, it seems natural to employ a semantics that models the meaning of deontic operators
with respect to these imperatives, rather than some notion of absolute or relative betterness. Existing modal systems relate rather simply to ought-operators of different strength,
defined with respect to a set of sentences that describe what the imperatives demand. To
mix such operators is, however, generally not an option if the resulting semantics is to be
represented by a strongly complete axiomatic system, and the same holds for the use of restrictions on the number and kind of imperatives to be considered. To cope with conflicting
imperatives, we may define a van-Fraassen-type operator that prohibits the agglomeration
of contradicting demands, and a sceptical ought-operator according to which something
is obligatory only if there is no conflicting demand; both are best represented in systems
that also contain operators which relate more directly to imperatives. In the light of the
fact that natural languages do not prohibit the use of deontic modalities within imperative
expressions, it is a welcome result that imperative semantics is possible, and sound and
completely axiomatizable for logical languages that permit iterated and nested deontic operators. Here the roads of ought defined evaluatively and ought defined with respect to
given norms part, as demonstrated by the fact that statements like it ought to be that what
ought to be is the case remain well formed, but are true accidentally only.
References
[1] C.E. Alchourrn, The expressive conception of norms, in: R. Hilpinen (Ed.), New Studies in Deontic Logic,
Reidel, Dordrecht, 1981, pp. 95123.
[2] A.R. Anderson, The formal analysis of normative concepts, in: N. Rescher (Ed.), The Logic of Decision and
Action, University Press, Pittsburgh, 1967, pp. 147213.
[3] D.O. Brink, Moral conflict and its structure, Philos. Rev. 103 (1994) 215247.
[4] M.A. Brown, Conditional and unconditional obligations for agents in time, in: M. Zakharyaschev, et al.
(Eds.), Advances in Modal Logic, CSLI, Stanford, 2001, pp. 121153.
[5] E. Conee, Against moral dilemmas, Philos. Rev. 91 (1982) 8797.
[6] A. Donagan, Consistency in rationalist moral systems, J. Philos. 81 (1984) 291309.
[7] D. Fllesdal, R. Hilpinen, Deontic logic: an introduction, in [10], pp. 135.
[8] J. Hansen, Sets, sentences, and some logics about imperatives, Fundamenta Informaticae 48 (2001) 205226.
[9] B. Hansson, An analysis of some deontic logics, Nos 3 (1969) 373398. Reprinted in [10], pp. 121147.
[10] R. Hilpinen (Ed.), Deontic Logic: Introductory and Systematic Readings, Reidel, Dordrecht, 1971.
[11] J.F. Horty, Moral dilemmas and nonmonotonic logic, J. Philos. Logic 23 (1994) 3565.
[12] J.F. Horty, Nonmonotonic foundations for deontic logic, in: D. Nute (Ed.), Defeasible Deontic Logic,
Kluwer, Dordrecht, 1997, pp. 1744.
[13] D. Jacquette, Moral dilemmas, disjunctive obligations, and Kants principle that ought implies can, Synthese 88 (1991) 4355.
[14] S. Kanger, New foundations for ethical theory, in [10], pp. 3658.
[15] Marcus, R. Barcan, Iterated deontic modalities, Mind 75 (1966) 580582.
[16] Marcus, R. Barcan, Moral dilemmas and consistency, J. Philos. 77 (1980) 121136.
[17] I. Niiniluoto, Hypothetical imperatives and conditional obligation, Synthese 66 (1986) 111133.
[18] A.N. Prior, Formal Logic, Clarendon Press, Oxford, 1955.
[19] T.J. Smiley, The logical basis of ethics, Acta Philos. Fennica 16 (1963) 237246.
[20] E. Stenius, The principles of a logic of normative systems, Acta Philos. Fennica 16 (1963) 247260.
[21] B. van Fraassen, Values and the hearts command, J. Philos. 70 (1973) 519.
187
Journal of Applied Logic 3 (2005) 484511

www.elsevier.com/locate/jal
Conflicting imperatives
and dyadic deontic logic
Jrg Hansen
Institut fr Philosophie, Universitt Leipzig, Beethovenstrae 15, D-04107 Leipzig, Germany
Available online 2 June 2005
Editors: A. Lomuscio and D. Nute
Abstract
Often a set of imperatives or norms seems satisfiable from the outset, but conflicts arise when
ways to fulfill all are ruled out by unfortunate circumstances. Semantic methods to handle normative
conflicts were devised by B. van Fraassen and J.F. Horty, but these are not sensitive to circumstances.
The present paper extends these resolution mechanisms to circumstantial inputs, defines dyadic deontic operators accordingly, and provides a sound and (weakly) complete axiomatic system for such
deontic semantics.
2005 Elsevier B.V. All rights reserved.
Keywords: Logic of imperatives; Deontic logic; Conflict of norms
1. The question of normative conflicts

Do moral conflicts exist? The orthodox belief in the 1950s was that such conflicts only
exist at first glancethe seemingly conflicting obligations arising from the application of
merely incomplete principles. Instead, what is actually obligatory must be determined by
a careful moral deliberation that involves considering and weighing all relevant facts and
reasons, and cannot produce conflicting outcomes. Among the first that came to reject this
view were E.J. Lemmon [27] and B. Williams [43]: Lemmon observed that in cases of true
moral dilemma, one does not know the very facts needed to determine which obligation
might outweigh the other. Williams argued in reductio that if, in case of conflicting oughts,
E-mail address: jhansen@uni-leipzig.de (J. Hansen).
1570-8683/$ see front matter 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jal.2005.04.005
188
485
there is just one thing one actually ought to do, then feelings of regret about having not
acted as one should have are out of place and one should not mind getting into similar
situations again. To avoid the derivation of the ought of a contradiction from two oughts
of equal weight but with contradictory contents, Williams argued that deontic logic should
give up the agglomeration principle
(C)
OA OB O(A B).
Lemmon had no such qualms: he advocated dropping the Kantian Principle ought implies
can
(KP) OA A,
thus allowing for obligations to bring about the impossible, and concluded:
I should like to see a proper discussion of the arguments that go to resolve moral
dilemmas, because I do not believe that this is an area of total irrationality, though I do
not believe that a traditional logical approach (the logic of imperatives, deontic logic,
and whatnot) will do either.
Regarding commands and legal norms, G.H. von Wright [45, Chapter 7], like H. Kelsen
[23, p. 211] at the time, excluded the coexistence of conflicting norms from the same
source: The giving of two conflicting norms is the expression of an irrational will; it is a
performative self-contradiction and as such a pure fact that fails to create a norm. E. Stenius
[40] and later C.E. Alchourrn and E. Bulygin [1] rejected this view: A system of norms
that is impossible to obey might be unreasonable and its norm-giver blameworthy, but its
existence does not constitute a logical contradictionconflicts are ubiquitous in systems of
positive law and logic cannot deny this fact. In his later theory, von Wright [49] concedes
that existing normative systems may or may not be contradiction-free, and reformulates
deontic principles as meta-norms for consistent norm-giving. Kelsen [24] later came to
view logic as inapplicable to law.
2. Van Fraassens proposal and Hortys variation
2.1. Van Fraassens operator O F
Not taking sides, pro or contra the existence of genuine normative conflicts, but arguing
that the view in favor seems at least tenable, B. van Fraassen [11] took up the burden of
finding plausible logical semantics that could accommodate conflicting obligations. The
intended semantics should accept the possible truth of two deontic sentences OA, OA
without committing the norm-subject to the absurd by making O(A A) true, for van
Fraassen wanted to keep the Kantian Principle. Given the existence of certain imperatives
in force, i.e., imperatives that are left as valid, relevant, not overridden, etc. by some unspecified deliberation process, van Fraassens idea was to make these imperatives part of
the logical model, and to describe something as obligatory if it serves to satisfy some, not
necessarily all, imperatives. Formally, let I be the set of imperatives in force, B be the set
of possible states of affairs, and i + B be the possible states of affairs where the imperative i I is considered fulfilled. Let A B be the set of possible states of affairs where
486
189
the indicative sentence A is considered true. Finally, let score(v) be the set of all imperatives that are fulfilled in the state of affairs v: score(v) = {i I | v i + }. Van Fraassen
then defines:
[Df-F]
O F A is true iff
v A : v A : score(v)
score(v ).
So A is obligatory if and only if (iff) there is some score that can be achieved when A is
true, which is not included in any score that could be achieved when A is true. In other
words, A is obligatory iff there are imperatives that can only be (collectively) satisfied
when A is true, but not when A is false.
By slightly changing the viewpoint, van Fraassens proposal might also be described
in the following way: let I be a set not of imperatives, but of indicative sentences in the
language LBL of some basic logic BL. The motivation is that I contains one sentence A
for each imperative i in force that is true in exactly those states of affairs in which the
imperative is fulfilled, i.e., A = i + . BL is assumed to be compact and the turnstile in
BL A means a classical consequence relation that characterizes BL, LBL , A
LBL . Let the remainder set A be the set of all maximal subsets that do not derive A,
i.e., of all such that (i) BL A, and (ii) there is no such that
and A. Then Df-F is equivalent to Df-F* ( means an arbitrary tautology):1
[Df-F ]
O F A is true
iff
I I : I BL A
So A is obligatory iff it is derivable from a maximally consistent subset of I . So something

is obligatory if it is required for doing the most: if it is necessitated by a strategy to fulfill
so many imperatives that no one who satisfies these as well could satisfy more. While a
parallel operation for belief change is known as credulous reasoning, to call it orthodox
might seem more appropriate: the agent is not released from any of her obligations as long
as they are fulfillable, even if this fulfillment is at the expense of violating other norms.
To see how van Fraassens semantics work, first let I = {A, B}, where A and B are
supposedly contingent and independent. There are no conflicts, I is consistent and O F A,
O F B and O F (A B) are all true since I derives A, B and A B. Thus agglomeration of
contents is permitted so long as the underlying imperatives do not conflict. For the case of
conflict, change I into {A C, B C}; O F A and O F B are true since A and B derive
from the maximally consistent sets {A C} and {B C}, but O F (C C) is false since
no consistent subset derives C C. The same is true for O F (A B) though {A B} is
consistent: the truth of A B is not necessary for maximal norm satisfaction.
An axiomatic system DLF that is (weakly) complete with regard to van Fraassens semantics is defined by the following axiom-schemes, in addition to BL-instances and modus
ponens (cf. [17, Section 5], means an arbitrary contradiction, and the index F here and
below indicates that the deontic operators occurring in the axiom scheme are thus indexed):
(MF )
(PF )
(NF )
(ExtF )
O F (A B) (O F A O F B)
O F
OF
If BL A B then DL O F A O F B
1 Cf. Hortys [20, Theorem 2].
190
487
To van Fraassens own puzzlement, the cases where agglomeration remains permissible do
not seem axiomatizable: object language does not reveal whether particular A and B of
some O F A and O F B are derived from the demands of imperatives that do not conflict and
so O F (A B) should be supported.2
2.2. The skeptical operator O S
The invalidation of the agglomeration principle by van Fraassens semantics did not
make them popular (cf. Donagan [8, p. 298]). Moreover, let a P F -operator expressing
permission be defined in the usual way as P F A =def O F A, and consider again I =
{A C, B C}: then O F (A C) is true, and so is O F C. Applying (MF ) and (ExtF ),
O F (A C) must be true, hence P F (A C) is false. So not even the obligatory is always
permitted, which seems strange (cf. Jacquette [22]).
In reaction to the dismissal of the agglomeration principle, Donagan [8] and Brink [7]
have claimed that even if there could be a normative demand for A and a conflicting demand for B, with BL A B, it need not follow that the norm-subject has an obligation
to realize A and an obligation to realize B. Rather, there should just be a disjunctive obligation to realize A or B. Given competing normative standards of equal weight, the strategy
of this reasoning is not to trust a single standard, but to consider obligatory only what all
standards demand. Let I be as before. Varying van Fraassens truth definition, Horty [20]
has formalized this skeptical ought as follows:3
[Df-S]
O S A is true iff I I : I BL A
So O S A is true iff A is derivable from all maximally consistent subsets of I . skeptical

is the term used in the epistemic-oriented literature, yet legalist also seems fitting, since
a norm violation is never pronounced as obligatory even if it is inevitable. This does not
let the agent off the hook: by doing what is obligatory in this sense, a maximum of norms
will, by necessity, get satisfied.
Let again I = {A C, B C}. O S A and O S B are false and O S (A B) is true: just
A B, but neither A nor B are derived by both of the two consistent subsets {A C} and
{B C}. P S (A C) is also true: A, C were assumed to be contingent and independent,
so the maximally consistent subset {A C} I does not derive (A C). So what is
O F -obligatory is at least P S -permitted.
A complete axiomatic system DLS is defined by the axiom-schemes (MS ), (CS ), (PS ),
S
(N ), and (ExtS ), together with BL-instances and modus ponens. Since the truth definitions
for O F A and O S A merely depend on a set I and BL, mixed expressions such as O F A
O S A are meaningful and may be accepted as well-formed. Then
(CSF )
O S A O F B O F (A B)
2 So agglomeration requires some consistency check of the underlying imperatives contents. Van der Torre and
Tan [41,42] proposed a two-phase deontic logic, where consistent aggregation must take place before weakening. For the present imperative semantics, I suggested a bimodal approach in [17] with an operator O 2 that more
directly talks about the imperatives. For comparisons and a new proposal cf. [15].
3 More in parallel to van Fraassens original definition, one may equivalently define:
[Df-S ] O S A is true
iff
v A : v A : score(v) score(v )
488
is valid, and the mixed system DL{F,S} containing the axiom schemes for DLF , DLS ,
the axiom scheme (CSF ), all instances of BL-theorems and modus ponensis sound and
(weakly) complete (cf. [17, Section 6]).
3. Predicaments and dyadic deontic logic

Arguing for the possibility of moral conflicts, R. Barcan Marcus [33] explained:
Under the single principle of promise keeping, I might make two promises in all good
faith and reason that they will not conflict, but then they do, as a result of circumstances
that were unpredictable and beyond my control.
Note that there is no conflict at the outset: any dilemma could have been averted by not
promising anything. Moreover, there might have been some point in time at which keeping
both promises was possible: having 500 $ with me and another 1000 $ in the office, I
promise Raoul and Johnny 500 $ each on Saturday with every intention of paying them on
Monday, only to find out that the office had been burglarized over the weekend. Donagan
[8] argues that this is not a genuine conflict, because three resolving principles apply:
(i) one must not make promises one cannot or must not keep, (ii) all promises are made with
the implicit condition that they are void if they cannot or must not be kept, (iii) one must
not make promises when one does not believe that the other party has fully understood (ii).
But as I well knew beforehand, neither Raoul nor Johnny are going to let me off the hook,
regardless of what may happen at the office. According to (iii), I was wrong to make the
promises. So am I entitled to break them (both)?We have here what G.H. von Wright
[48] terms a predicament: a situation from which there is no permitted way out, but to
which there also is no permitted inlet. The normative order is consistent, it is only through
ones own fault that one finds oneself in a predicament.4 Von Wright then asks:
The man in a predicament will, of necessity, react in some way or other, either do
something or remain passive. Even though every reaction of his will be a sin, is it not
reasonable to think that there is yet something he ought to do rather than anything else?
To deny this would be to admit that it makes, deontically, no difference what he does.
But is this reasonable? (. . .) If all our choices are between forbidden things, our duty is
to choose the least bad thing. [48, p. 80]
4 That predicaments only arise from an agents own faults, and not through misfortune or the wrongdoings of
others, is a view von Wright and Donagan ascribe to Thomas Aquinas, but this does not seem quite correct: in
the discussion of oaths (Summa Theologica II.II Qu. 89 art. 7 ad 2), Thomas considers the objection that it would
sometimes be contrary to virtue, or an obstacle to it, if one were to fulfill what one has sworn to doso oaths
need not always be binding. In answering, Thomas distinguishes oaths that are unlawful from the outset, where a
man sinned in swearing, and oaths that could be lawful at the outset but lead to an evil result through some new
and unforeseen emergency: fulfillment of such oaths is unlawful.
191
192
489
Sub-ideal demands are usually represented by a dyadic deontic sentence O(A/C), meaning that it ought to be that A given C is true. By accepting all instances of (DD-)
O(A/) P (A/) as a logical truth in [48], von Wright dismisses an inconsistent normative system as conceptual absurdity: if A is obligatory on tautological conditions (i.e.,
unconditionally obligatory), then there cannot be a likewise unconditional obligation to the
contrary. Although von Wright originally used the stronger (DD) O(A/C) P (A/C) for
arbitrary C (axiom A1 of the old system in [44], and axiom B1 of the new system in
[46]), he later turned against it, arguing that while morality makes no conflicting claims,
it is not a logical impossibility that conflicting promises can give rise to predicaments.5
Dyadic operators seem essential for even making this distinction.6
Turning object language oughts into a special sort of conditionals does not necessarily
imply a change in the formalization of the background imperatives: consider the set I =
{(C A), (C A)}, corresponding to background imperatives in the usual way. I is
also its single maximally consistent subset, which derives C, so O F C and O S C are
both true. But a single standard is no longer available once C becomes true: the imperatives
have not all been fulfilled (otherwise one would not be in condition C), and any maximal
set of imperatives that is consistent with the given circumstances cannot contain all. So the
proposal is to call A obligatory in case C iff A is necessary for doing the most that can
be achieved, given the truth of C. Formally:
[Df-DF]
O F (A/C)
iff
I I C: I {C} BL A
So O F (A/C) is true iff there is some set, among the maximal subsets of I consistent
with C, that together with C derives A. This is obviously a conservative extension of the
definition given for the unconditional case, so we may define O F A =def O F (A/).
If a cautious, disjunctive approach were appropriate for cases of conflict, then it would
be hard to see why predicaments should be treated differently: that conflicts must be accounted for at the outset, but analogues of Buridans ass cannot be brought about by fate
or unpredictable human nature, would hardly be plausible. Distrusting any single standard,
such an approach would accept, given the circumstances C, only what is necessary by any
standard that could still be metno crying over spilled milk. Formally:
[Df-DS]
O S (A/C)
iff
I I C: I {C} BL A
So O S (A/C) is true iff all the maximal subsets of I consistent with C derive A, given the
truth of C. This is again a conservative extension of the unconditional case, so one may
define O S A =def O S (A/).
After a comparison of the above definitions with similar approaches namely in the study
of nonmonotonic reasoning, I will give an axiomatic dyadic deontic system DDL{F,S} ,
which I prove to be sound and (only) weakly complete with respect to the above semantics.
5 Cf. [47], [48, pp. 36, 81, 89].
6 Below, I extend the treatment of conflicts to the area of predicaments, and do not follow von Wright in ruling
out conflicts. However, this can easily be done by axiomatically adding (DD-) to the system presented below.
490
4. Comparisons
Though the truth definitions introduced in the preceding section naturally extend the
proposals of van Fraassen and Horty for dealing with normative conflicts to the dyadic
context and the related problem of predicaments, and though their resolution mechanisms
are not exactly new (cf. below), there has not been much discussion of these concepts
in the deontic logic literature. Notably, Hortys own dyadic operator in [21] is defined
with respect to (simply) maximally non-conflicting sets of prima facie oughts, and it is
disregarded that their joint demands may now be inconsistent with the situation. But the
more general literature on nonmonotonic reasoning offers a range of parallel concepts.
Regarding O S , the most obvious parallel is Kratzer and Lewiss premise semantics in
[25] and [30] which has a set of formulas H (the premises) to define counterfactuals in
much the same way as the set I is used here in the definition of deontic conditionals.
Considering Kratzers definition, and setting aside the world-relativity of H , let SH,C =
{H H | H BL C} be the set of all subsets of H that are, according to some basic
logic BL, consistent with C. A counterfactual conditional is defined in the following
way:
H |= C A iff
H SH,C : H SH,C : H H , H {C} BL A
In other words, C A is true iff each set in SH,C has a superset in SH,C that implies
C A.7 The truth definition is tailored for a basic logic that may fail compactness and so
accommodates sets H with ever-larger C-consistent subsets, but no maximal ones. Here,
BL was assumed to be compact, and we obtain:
Observation 1 (Relation to premise semantics). For any set I LBL :
I |= O S (A/C)
iff
I |= C A
Proof. Left-to-right. Suppose that C A is false, so there is some I SI,C : I

SI,C : if I I then I {C} BL A. I BL C, so by definition there is some I
I C such that I I . So I {C} BL A, so I I C: I {C} BL A and so
O S (A/C) is false.
Right-to-left. Suppose O S (A/C) is false, so I I C: I {C} BL A. Then I
SI,C , and by definition of I C there is no other I SI,C : I I , so I SI,C : if
I I then I {C} BL A, and C A is false.
Then, the definition of O S parallels that of a consequence relation associated to a Poole
system without constraints [35]: This has two sets , of formulas, the facts and the
defaults. A scenario is a set such that and BL . A maximal
7 Lewiss [30] variation requires this property only of non-empty H S
H,C . This corresponds to replacing
I {C} BL A in the truth definition for O S with B1 , . . . , Bn I : BL (B1 Bn C) A to produce
a regular instead of a normal operator. Lewis notes (p. 233) that for deontic conditionals, the premises of the
premise semantics might be understood to be something that ought to hold, so he is to be credited for the
imperative semantics employed here.
193
194
491
scenario is one where E( ), E( ) being the set of all such that { } BL

, and for all , if then { } PL . Cn being BL-consequence, a
prediction A from the facts and the defaults is then defined as:
Cn( )
|skept() A =def A
E( )
So A is predicted from and if all maximal scenarios derive A. Likewise, a credulous

prediction can be defined as
Cn( )
|cred() A =def A
E( )
(cf. Brass [5] for the analogy and notation). The following is then immediate:
Observation 2 (Relation to Poole systems). For any set I LBL :
I |= O S (A/C)
iff
{C} |skept(I ) A
iff
{C} |cred(I ) A
I |= O (A/C)
Regarding the O F -operator, it is perhaps not quite as obvious that its corresponding P F operator is closely related to the X-logics of Siegel and Forget [10,38]: The consequence
relation |X of these logics holds between a set of formulas and a formula A modulo a
set X of formulas, where the definition is
|X A iff
Cn {A} X = Cn( ) X
As Makinson [31] pointed out, X can be understood as a set of bad propositions that one
is to avoid. So |X A is true iff A can be realized together with without increasing
the set of bad proposition above those that were already true given . Here we have a
set I of desired propositions, so a statement seems bad if it asserts that some desired
proposition be false, e.g., A is true for some A I , or that at least one A1 , . . . , An I is
false, i.e., (A1 An ) is true. Let I = { {A1 , . . . , An } | {A1 , . . . , An } I, 1
n card(I )} be the bad set corresponding to I . We then obtain:
Observation 3 (Relation to X-logics). For any set I LPL , X = I :
I |= P F (A/C)
iff {C} |X A
Proof. Right-to-left. Suppose {C} |X A, so Cn(C A) I = Cn(C) I , so by

/ Cn(C) I . By
monotony of Cn there is a B Cn(C A) I such that B
definition, B = (B1 Bn ) for some B1 , . . . , Bn I . The first fact provides
{C A} BL (B1 Bn ) and by contraposition {B1 , . . . , Bn } BL C A. From the
second fact we obtain {C} BL (B1 Bn ), so by contraposition {B1 , . . . , Bn } BL
C, so {b1 , . . . , Bn } I for some maxi-consistent I I C, and so there is some
I I C: I {C} BL A, so I |= O F (A/C) and by definition I P F (A/C).
Left-to-right. Suppose I P F (A/C), so there is a I I C: I {C} BL A.
By compactness of BL there are {B1 , . . . , Bn } I with {B1 , . . . , Bn } BL C A,
492
and {B1 , . . . , Bn } BL C since they are in I . By definition there is a B I such

that B = (B1 Bn ). Hence {C A} BL B , so B Cn(C A) I , and
/ Cn(C) I . So Cn(C A) I = Cn(C) I , and {C} |X A.
{C} BL B , so B
A unified treatment of both, skeptical and credulous consequence can be found in
Bochmans epistemic states-semantics in [24]. Epistemic states, equivalent to the cumulative models in [26], are triples E = S, , , where S is a set of objects (belief states),
some asymmetric preference relation on S, and a labeling function that assigns each
state s S a deductively closed theory. min S = {s S | t S , t = s: t s} is the set
/ (s)} is the set of belief states consistent
of minimal states in S S. A = {s S | A
with A. For each A, A must be -smooth, i.e., for any s A , either s min A or there
is some t min A with t s. With BL as basic logic, Bochmans definitions for skeptical
and credulous consequence relations | and | are:
A |E B
iff
A |E B
iff
s min A : (s) BL A B
A = or s min A : (s) BL A B
Observation 4 (Relation to Bochmans epistemic states). Let I LPL , and let the corresponding epistemic state EI = S, , be such that (i) S = P(I ), (ii) (s) = Cn(s),
and (iii) s t iff t s. Then
I |= O S (A/C)
iff
iff
I |= O (A/C)
C |EI A
PL
C and C |EI A
Proof. I prove first (a) I min A iff I I A, (b) EI is an epistemic state, (c) A =
iff BL A: For (a), by definition A = {I I | I BL A}, so A is the set of subsets
of I consistent with A. I min A means that for any I min A , I = I : I I .
By definition for any I A , I = I : I I . This means there is no I I consistent
with A such that I I , which means I I A. For (b), if I I is in A , i.e., it is
/ I A then by definition of I A there is some I I A
consistent with A, and I
such that I I , so there is some I min A with I I . So EI is smooth, hence
it is an epistemic state. For (c), A = iff {I P(I ) | I BL A} = iff BL A
holds by monotony of BL. Putting together, we get: I |= O S (A/C) iff I I C: I
{C} BL A iff I min C : I BL C A iff C |EI A. Likewise: I |= O F (A/C) iff
I I C: I {C} BL A iff I min C : I BL C A iff C = and [ C =
or I min C : I BL C A] iff BL C and C |EI A.
A final parallel brings us back to deontic logic, namely the multiplex preference semantics of Goble in [1214], where a multitude of preference relations enables definitions like
all-best (universally preferred) and some-best (existentially preferred), which are then
used in definitions of deontic operators. That, in the finite case, such semantics corresponds
closely to the present account will be explicated in Section 6. Regarding meta-theory, for
a somewhat more general semantic setting the skeptical consequence relation was axiomatized by Kraus, Lehmann and Magidor [26], and the credulous consequence relation by
195
196
493
Bochman [2]. However, a completeness proof for a system that includes both8 seems to be
missing so far and this is what I shall now turn to.
5. The dyadic deontic logic DDL{F,S}
Let the basic logic be propositional logic PL: The alphabet has proposition letters
Prop = {p1 , p2 , . . .}, operators , , , , and parentheses (, ). The language LPL is defined as usual. , in front of a set of sentences means their conjunction
and disjunction, and, e.g., ni=1 Ai further abbreviates {Ai , . . . , An }. Semantically, valuation functions v : Prop {1, 0} define the truth of sentences A LPL as usual (written
v |= A), B is the set of all such valuations, and A is the extension {v B | v |= A} of A.
PL is a sound and complete axiomatic system, and PL A means that A is provable in PL.
The alphabet of the language LDDL{F,S} additionally has the operators O F , O S , and
the auxiliary /. DDL{F,S} is then the smallest set such that
{F,S}
(a) for all A, C LPL , O F (A/C) and O S (A/C) LDDL ,

{F,S}
(b) if A, B LDDL , so are A, (A B), (A B), (A B), (A B).
Outer parentheses will be mostly omitted. We define P (A/C) =def O (A/C), where
* is F or S. For simplification we do not permit mixed expressions and nested deontic
operators like p1 O S (p2 /p1 ), P S (O F (p2 /p2 )/p1 ).
For DDL{F,S} -semantics, the truth of DDL{F,S} -sentences is defined with respect to a
set I LPL (Boolean operators being as usual):
I |= O F (A/C)
iff
I I C: I {C} PL A
I |= O S (A/C)
iff
I I C: I {C} PL A
DDL{F,S} -satisfiable,
If I |= A, A is called
and DDL{F,S} -valid if I |= A for all I LPL
(we write |=DDL{F,S} A).
Consider the following axiom-schemes (* is the uniform index F or S):
(CExt )
(ExtC )
(DM )
(DCS )
(DCSF )
(DNS )
(DPF )
(Cond )
(CCMon )
(RMonF )
If PL C (A B) then DDL{F,S} O (A/C) O (B/C)

If P L C D then DDL{F,S} O (A/C) O (A/D)
O (A B/C) (O (A/C) O (B/C))
O S (A/C) O S (B/C) O S (A B/C)
O S (A/C) O F (B/C) O F (A B/C)
(DN-R ) If P L C then DDL{F,S} O (/C)
O S (/C)
F
(DP-R ) If P L C then DDL{F,S} P (/C)
P (/C)
O (A/C D) O (D A/C)
O (A D/C) O (A/C D)
P F (D/C) (O F (A/C) O F (A/C D))
8 One might add a third (monadic) deontic modality O 2 that more directly talks about the imperatives to
axiomatize consistent agglomeration, but I must leave the details to future study (cf. [17, Section 6] for the
resulting monadic system DL{2,F,S} ).
494
(RMonFSS ) P F (D/C) (O S (A/C) O S (A/C D))

(RMonSSF ) P S (D/C) (O S (A/C) O F (A/C D))
The system DDL{F,S} is then the set such that (i) all LDDL{F,S} -instances of PL-tautologies
are in DDL{F,S} , (ii) all LPL -instances of the above axiom schemes are in DDL{F,S} ,
and (iii) DDL{F,S} is closed under modus ponens. If A DDL{F,S} we write DDL{F,S} A
and call A provable in DDL{F,S} . LDDL{F,S} is DDL{F,S} -inconsistent iff there are
A1, . . . , An in , n 1, with DDL{F,S} (A1 An ) , otherwise is DDL{F,S} consistent. A LDDL{F,S} is DDL{F,S} -derivable from LDDL{F,S} (written DDL{F,S}
A) iff {A} is DDL{F,S} -inconsistent.
Theorem 1. The following are DDL{F,S} -derivable (* is a uniform index or as indicated):
(RefS )
(RW )
(Pres )
(CMon )
(Cut )
(Or )
(DR )
(FH )
(FH+ )
(Trans )
(P -LoopF )
(LoopS )
O S (A/A) (Ref-R ) If PL A then DDL{F,S} O (A/A)

If PL A B then DDL{F,S} O (A/C) O (B/C)
O (/C) (O (A/D) O (A C/D))
SSS, SFF, FSF
O (D/C) (O (A/C) O (A/C D))
SSS, SFF, FSF
O (D/C) (O (A/C D) O (A/C))
SSS, SFF, FSF
(O (A/C) O (A/D)) O (A/C D)
FFF, SFS, SSF
O (A/C D) (O (A/C) O (A/D))
FFF, FSS, SSF
P (C/D) (O (A/C D) O (A/C))
P (A C/C D) (O (A/C D) O (A/C)) FFF, FSS, SSF
FFF, FSS, SFS
P (A/A B) P (B/B C) P (A/A C)
P F (A2 /A1 ) P F (An /An1 ) P F (A1 /An ) P F (An /A1 )
O S (A2 /A1 ) O S (An /An1 ) O S (A1 /An ) O S (An /A1 )
Proof. All easy and left to the reader.

Regarding axioms and theorems, (CExt ) is a contextual extensionality rule for consequents, and (ExtC ) an extensionality rule for antecedents. (DM ) and (DC ) are dyadic
versions of their monadic analogues. The O S -axioms are like those of Kraus, Lehmann
and Magidor [26], but (CondS ) and (CCMonS ) equivalently replace (OrS ) and (CMonS ),
and (DP-RS ) is added. The O F -axioms are those of Bochman [2], where his (PresF ) is
strengthened to (DPF ). Instead of (CondF ), (CCMonF ) and (RMonF ), Goble [14] more
elegantly employs (TransF ) and (DRF ), which is equivalent given (DP-RF ). The mixed
schemes are again Bochmans. Instead of (DCSF ), (RMonFSS ), and (RMonSSF ), Goble has
(DKSF )
O S (A B/C) (O F (A/C) O F (B/C))
(TransFSS ) and (TransSFS ), which is again equivalent. The names are from the study of nonmonotonic logics, namely reflexivity, right weakening, preservation, (conjunctive) cautious
monotony, conditionalization and disjunctive reasoning. (Ref ) is Hanssons [18, Theorem 2], (CCMon ) Reschers [36, Theorem 4.4], (Or ) is the right-to-left version of von
Wrights (B3) in [46], and (DR ) Hanssons Theorem 13. Fllesdal and Hilpinen [9] introduced the strong version (FH ) of (RMon ) (their Theorem 77). (FH+ ) is even stronger:
its displayed versions could replace (CCMon ) and (RMon ), =F, FSS, SSF. (Trans )
is transitivity of weak preference given by A B =def P (A/A B) (Lewis [28, p. 54]).
197
198
495
Spohn [39] introduced (P -LoopF ) to define the relevant equivalence classes in his completeness proof of Hanssons DSDL3, and its O-form was rediscovered by Kraus, Lehmann
and Magidor [26] who put it to the same use. Note that (CCMon ) is (P -Cond ), and (Cut )
is (P -RMon ), where the deontic operators are swapped in the P -versions.
Theorem 2. DDL{F,S} is sound.
Proof. The validity of (DM ), (DCS ), (DCSF ), (CExt ), and (ExtC ) is immediate. (DNS ),
(DPF ) are valid since any subset of LPL derives , and any maximally consistent subset
is consistent. If PL C then at least is in I C, hence I C = and then both
(DN-RF ) and (DP-RS ) hold likewise.
(CondF ) Assume O F (A/C D), so there is an I I (C D) such that I {C
D} PL A and I {C} PL D A. Since I PL (C D), also I PL C,
so by maximality there is an I I C such that I I , so there is an I
I C: I {C} PL D A, so O F (D A/C).
S
(Cond ) Assume O S (A/C D). So for all I I (C D) : I {C D} PL A.
If there is an I I C: I {C} PL D A then I {C} PL D and
I PL (C D). By maximality I I (C D): I I . Since I PL
(C D), I PL C, so there is an I I C: I I . Then I I and
by maximality of I I C, I = I and hence I = I . So I is in I (C
D) and I {C D} PL A, but this violates the assumption. So for all I
I C: I {C} PL D A, and O S (D A/C).
(CCMonF ) Assume O F (A D/C), so I I C: I {C} PL A D. Then I
{C} PL D, for otherwise I PL C which is excluded by the definition of
I C. So I PL (C D), by maximality I I (C D): I I and
I {C} PL A, I {C D} PL A. Hence O F (A/C D).
(CCMonS ) Assume O S (A D/C), so for all I I C: I {C} PL A D, and so
I {C} PL D, for otherwise I PL C contrary to the definition of I C,
and so for all I I C: I PL (C D). Suppose I I (C D), so
I PL (C D) and I PL C. By maximality I I C such that I I .
In turn I PL (C D) as just proved, so I I (C D) such that I I .
But then I = I by maximality of I I (C D), so I = I I C and
I {C} PL A as assumed. So I {C D} PL A for any I I (C D).
So O S (A/C D) is true.
F
(RMon ) Assume O F (A/C), so I I C: I {C} PL A. If P F (D/C) then
I I C: I {C} PL D. So I PL (C D). So by maximality I
I (C D): I I , so I {C} PL A, by monotony I {C D} PL A, so
O F (A/C D) is true.
FSS
(RMon ) Assume O S (A/C), so I I C: I {C} PL A, and P F (D/C), so
I I C: I {C} PL D, so I PL (C D). Suppose I I (C D),
so also I PL C and by maximality I I C : I I . We have I
{C} PL A, so B1 , . . . , Bn I : {B1 , . . . , Bn } {C} PL A by PL-compactness.
If I {C} PL A then {B1 , . . . , Bn } I , by maximality I {B1 , . . . , Bn } PL
(C D), but I {B1 , . . . , Bn } I , so I {C} PL (C D) contrary to the
496
assumption. So I {C} PL A, I {C D} PL A for any I I (C D).

So O S (A/C D) is true.
(RMonSSF ) Assume O S (A/C), so I I C: I {C} PL A, and P S (D/C), so
I I C: I {C} PL D. So I PL (C D). So by maximality
I I (C D): I I , so I {C} PL A, so I {C D} PL A, so
O F (A/C D) is true.
Theorem 3. DDL{F,S} -semantics are not compact.
Proof. In [17], I provided a counterexample to the compactness of semantics that only
employ the monadic deontic operator O F . Since O F A can be defined as O F (A/), this
also refutes the compactness of DDL{F,S} and of the subsystem containing only the dyadic
operator O F . The following counterexample is expressed in terms of the dyadic operators O S only, which also refutes the compactness of the subsystem containing only this
operator: let
= O S (p2 /)
P S (p2 /p1 )
O S (pi /p1 )
i=3
P S (p2 /p1 )
O S (pi /p1 )
i=3
P S (p2 /p1 p2 )
O S (pi /p1 p2 )
i=3
P S (p2 /p1 p2 )
O S (pi /p1 p2 )
i=3
is finitely DDL{F,S} -satisfiable: let n be the greatest index of any proposition letter occurring in some finite f . Then
If = pn+1 (p1 p2 ), pn+1 (p1 p2 ), p2 , p3 , . . . , pn
satisfies f . For easy verification, I list the relevant sets of maximal subsets:
If =
{pn+1 (p1 p2 ), p2 , p3 , . . . , pn )},

{pn+1 (p1 p2 ), p2 , p3 , . . . , pn )}
{pn+1 (p1 p2 ), p3 , . . . , pn )},

If p1
=
{pn+1 (p1 p2 ), p2 , p3 , . . . , pn )}
If (p1 p2 )
If p1
{ pn+1 (p1 p2 ), p2 , p3 , . . . , pn )},
=
{ pn+1 (p1 p2 ), p3 , . . . , pn )}
If (p1 p2 )
However, is not DDL{F,S} -satisfiable: suppose I LPL satisfies , and let A
{p1 , p1 , p1 p2 , p1 p2 }. Observe that
199
200
(i) There are I1 , I2 I such that I1 PL p1 pi , I2 PL p1 pi , i
497
2.
Proof. From O S (p2 /), P S (p2 /p1 ) and the validity of (RMonFSS ), follows O F (p1 /), i.e., there is an I1 I : I1 PL p1 . Likewise from O S (p2 /)
and P S (p2 /p1 ) , it follows that there is an I2 I : I2 PL p1 . To
satisfy O S (p2 /) it is necessary that for all I I : I PL p2 , and from
O S (pi /p1 ), O S (pi /p1 ) , and the validity of (OrS ), it is obtained that for all
I I : I PL pi , i 3.
(ii) For each A, there is an IA I A : IA {A} PL p2 .
Proof. Let A {p1 , p1 p2 }. Then by observation (i) I1 I A. Since I1 PL
p2 , to satisfy P S (p2 /A) there is an IA I A such that IA I1 PL A.
So IA {A} PL (p1 p2 pn ) for some n. If n 3, then IA {A} PL
(p1 p2 pn1 ), since IA {A} PL pn is necessary for O S (pn /A) .
So IA {A} PL (p1 p2 ), so IA {A} PL p2 . Likewise, the proof for A
{p1 , p1 p2 } is obtained from I2 I A.
(iii) If A {p1 , p1 p2 } then IA {p1 , p2 , p3 , p4 , . . .} PL .
If A {p1 , p1 p2 } then IA {p1 , p2 , p3 , p4 , . . .} PL .
Proof. Suppose A {p1 , p1 p2 } and IA {p1 , p2 , p3 , p4 , . . .} PL . Then
IA {A, p2 , p3 , p4 , . . .} PL . So IA {A} PL (p2 p3 p4 pn )
for some n. But also IA {A} PL p2 p3 p4 pn by observation (ii)
and from the fact that I satisfies O S (pi /A) , 3 i n. So IA PL A, but
this contradicts IA I A. The proof for A {p1 , p1 p2 } and the set IA
{p1 , p2 , p3 , p4 , . . .} is done likewise.
It follows that Ip1 I(p1 p2 ) PL and Ip1 I(p1 p2 ) PL . This is most easily
seen by appealing to PL-semantics: some v B satisfies {p1 , p2 , p3 , p4 , . . .} and by (iii)
all elements of Ip1 as well as all of I(p1 p2 ) , so their union is satisfiable and therefore
consistent (likewise for {p1 , p2 , p3 , p4 , . . .} and Ip1 I(p1 p2 ) ). From (ii) it follows
that
Ip1 I(p1 p2 ) PL (p1 p2 ) (p1 p2 ) p2
Ip1 I(p1 p2 ) PL (p1 p2 ) (p1 p2 ) p2
But the conclusions are tautologically equivalent to p2 , so there are consistent subsets of
I that derive p2 , and I O S (p2 /), although O S (p2 /) .
Theorem 4. DDL{F,S} is weakly complete.
Proof. The proof follows the completeness proof of Spohn [39] for B. Hanssons [18]
preference-based dyadic deontic logic DSDL3. Since parts of this proof will be reused in
the next section for logics that might not include unrestricted (DN ) or (DP ), I will avoid
their use up to the last step of this proof.
498
A. Preliminaries. We must prove that if |=DDL{F,S} A then DDL{F,S} A for any A LPL .
We assume DDL{F,S} A so A is DDL{F,S} -consistent. We build a disjunctive normal
form of A and obtain a disjunction of conjunctions, where each conjunct is O (B/C)
or O (B/C). One disjunct must be DDL{F,S} -consistent. Let be that disjunct. Let the
be the PL-sentences that contain only proposition letters occur-restricted language LPL
n
, where n is
ring in . Let r(LPL ) be 22 mutually non-equivalent representatives of LPL
the number of proposition letters in . By writing PL-sentences (including and ), we
). We construct a set as follows:
now mean their unique representatives in r(LPL
(a) Any conjunct of is in .
):
(b) For all B, C r(LPL
F
either P (B/C) or O F (B/C) , and
either P S (B/C) or O S (B/C) .
(c) is DDL{F,S} -consistent.
It then suffices to find a set I LPL that makes true all B .
B. Identifying the deontic bases. We identify syntactically what Hansson called the deontic basis in an extension C (Spohn [39] writes C). Monadic deontic logic has just one
basis, dyadic deontic logic usually has one basis for any C, and here there may be several
bases, which expresses some conflict or predicament in case C.
), let
Definition 1. For any C = , C r(LPL
) | O S (A/C) },
OCS = {A r(LPL
F
) | O F (A/C) }
OC = min{A r(LPL
where min = {A | B , if PL B A then PL B A}, LPL .

From (DCS ), (RW ), and DDL{F,S} -consistency of we obtain, for any C = :
(B1) O S (A/C) iff PL OCS A,
(B2) O F (A/C) iff O OFC : PL O A.
C. Identifying the relevant class of domains. We identify the most general circumstances
CA where A is P F -permitted. To the same effect, Spohn employs equivalence classes [A]
) with P F (B /A),
defined using (P -LoopF ): A B iff B is in some {B1 , . . . , Bn } r(LPL
1
P F (B2 /B1 ), . . . , P F (Bn /Bn1 ), P F (A/Bn ) . The set of all such classes is then {[C] |
C C}.
), let
Definition 2. For all A r(LPL
CA = max C r(LPL
) | P F (A/C)
CA
C=
)
Ar(LPL
201
202
499
where max = {A | B : if PL A B, then PL B A}.

(C1) If P F (A/D) , then there is a C CA such that PL D C.
).
Proof. Immediate from definition of CA and finitude of r(LPL
(C2) For all C C: CC = {C}.

), so by definition P F (A/C) ,
Proof. By definition C CA for some A r(LPL
F
F
and P (C/C) due to (CExt ). Suppose C CC : then by definition P F (C/C ) .

With (FHF ) we get P F (A/C C ), P F (C/C C ) , so C = (C C ) = C follows
from maximality of C, C .
(C3) For all C C, if P F (C/D) , then PL D C.

), so by definition P F (A/C) .
Proof. By definition C CA for some A r(LPL
F
F
If P (C/D) , then we get P (A/C D) with (FHF ). So C = (C D) by
maximality, hence PL D C.
(C4) For all A = : CA = {CA } for some CA CA and PL A CA .

Proof. If A = then P F (A/A) due to (DP-RF ) and (CExtF ), so there is some
C CA such that PL A C by (C1). Assume C CA : By definition P F (A/C),
P F (A/C ) , so we get P F (C/C ) with (RWF ), and P F (A/C C ) with
(FHF ), so C = (C C ) = C by maximality. So C is the desired CA .
S (A O S ).
(C5) For all A = : PL OA
CA
S /A) , so with (CondS ) we
Proof. PL A CA due to (C4), and by (B2) O S (OA
S
S
obtain O (A OA /CA ) and thus the right-to-left direction PL (A OCSA )
S
S . For the opposite,
S
OA
PL OA A follows from (CExt ), by definitions and (C4)
P F (A/CA ), O S (OCSA /CA ) , so we get O S (OCSA /A) with (RMonFSS ). So
S (A O S ).
PL OA
CA
(C6) For all A = , OA OFA : OCA OFCA : PL OA (A OCA ).

Proof. Let OA OFA , so O F (OA /A) . PL A CA and (CondF ) derive
O F (A OA /CA ) , so OCA OFCA with PL (A OCA ) OA . If P F (OCA /
A) , then from O F (OCA /CA ) and (RMonF ) we get O F (A/CA ) , but
by definition P F (A/CA ) . So O F (OCA /A) , and O F (A OCA /A) by
(CExt). Since PL (A OCA ) OA we obtain P L OA (A OCA ) by minimality
of OA .
500
203
(C7) For all C = , C C: If {C OC } {D}
, then O F (C OC /D) .
Proof. {C OC } {D} . If O F (C/D) , then the conclusion is trivial. Otherwise P F (C/D) , so PL D C by (C3). For r.a.a. suppose P F (OC /D) .
With O F (OC /C) we obtain O F (OC D/C) by (FH+F ), and PL OC
D by minimality of OC . But then PL D (C OC ), which refutes the assumption. Hence O F (OC /D) and O F (C OC /D) by use of (CExtF ).
D. Identifying the multiple system of spheres. If this were ordinary dyadic deontic logic
with agglomeration and so just one basis OC for any C, we would be almost done: like
Spohn [39] orders his equivalence classes [C] by a relation before, C could be ordered
into C1 , . . . , Cn with C1 = , and Ci+1 = Ci OCi until this equals . S1 , . . . , Sn with
Si = (Ci Ci+1 ), 1 i < n, is then the system of spheres. Here this method fails since
no C C is guaranteed to have a single basis. But as it turns out, C has the structure of a
multiple system of spheres that is similarly identified.
(D1) {} C.
Proof. P F (/) by (DP-RF ), and PL C for any P F (/C) , so
C , C.
(D2) For all C C, O OFC : If C O = , then C O C.
Proof. If C O = then P F (/C O) by (DP-RF ), CC O CF . We
prove CC O = C O: PL (C O) CC O is immediate from (CExtF )
and (C1). If PL CC O (C O) then {C O} PL CC O , so O F (C
O/CC O ) follows from (C7). P F (C O/CC O ) by definition, so is
DDL{F,S} -inconsistent, but we assumed otherwise.
(D3) For all C C: If PL C D, C = D, then O OFD : PL C (D O).
Proof. Either P F (C/D) , so PL D C, C = D (C3). Or O F (C/D) , so
PL O C for some O OFD and PL C (D O).
), O OF {O S }: If D = , then D = (D O).
(D4) For all D r(LPL
D
D
Proof. If D = (D O), then PL D O. But also PL O D due to (CExt ),

so O = and O (/D) by (B1-2). So D = by (DP-R ).
(D5) Let D be such that (i) D, and (ii) if D D, O OFD and (D O) = , then
(D O) D. Then D = C\{}.
Proof. D C is immediate from (D1), (D2). As for C D, for each C C, C = ,
there is some D D such that (a) PL C D, and (b) for no O OFD : PL C
204
501
(D O). (a) is guaranteed by D, and (b) follows from (D3), (D4) and finiteness
). So C = D by (D3).
of r(LPL
E. Canonical construction and coincidence lemma.
)
Definition 3 (Canonical construction). For all C C {}, D r(LPL
F-Succ(D) = {D O | O OFD },
F-Chain(C) be the set of sequences D1 , . . . , Dn , 1 n, where D1 = , Di+1
F-Succ(Di ), Di = Di+1 for any 1 i < n, and Dn = C,
S-Chain(C, D) be the set of sequences D1 , . . . , Dk , Dk+1 , . . . , Dn , 1 k < n, where
S , D =D
i < n, and
D1 , . . . , Dk F-Chain(C), Di+1 = Di OD
i
i+1 for any k
i
Dn = D.
For any C C\{}, C F-Succ(C), let
] be a function that associates a unique proposition letter not
: C {} [Prop\LPL
occurring in with each element of C {},
(C, C ) = (C ) {(C ) | C F-Succ(C), C = C },
(C) = {(C ) | C F-Succ(C)}.
For any C C {} \ {}, C1 , . . . , Cn F-Chain(C), let

i F [ C1 , . . . , Cn ] = C
n1
i=1 (Ci , Ci+1 ).
For any C C, D1 , . . . , Dk , Dk+1 , . . . , Dn S-Chain(C, D), Dk = C, let

k1
i=1 (Ci , Ci+1 )
i S [ D1 , . . . , Dk , Dk+1 , . . . , Dn ] = D (C)
(C)
if C = ,
otherwise.
Let I F be the set of all such i F [ C1 , . . . , Cn ], and likewise I S be the set of all such
i S [ D1 , . . . , Dk , Dk+1 , . . . , Dn ]. Then finally I = I F I S .
The definition provides the construction of the canonical set I to make all of true.
F-Succ(C) is the set of immediate contrary-to-duty successors C of C, i.e., O OFC with
C = C O. (D2) showed each C C to be such a successor of (a successor of . . .) ,
and F-Chain(C) is the set of all such chains beginning with and ending with C. is used
to make any two i F [ch(C )], i F [ch(C )], C = C being successors of (successors of. . .)
C, inconsistent with each other and with any i S [ch(C, D)] via . Since C is finite, so is
the number of proposition letters introduced by , I F , I S and I . Regarding the sequences
used to construct I , I use ch(C) for D1 , . . . , Dn F-Chain(C) with C C {} \ {},
), and ch, ch , etc. for
ch(C, D) for D1 , . . . , Dn S-Chain(C, D) with C C, D r(LPL
any sequence for which either holds. We obtain:
(E1) For all ch = D1 , . . . , Dn , PL Di+1 Di and
PL
Di Di+1 , 1
i < n.
502
(E2) If {i F [ch(C)], i F [ch(C )]} PL , then ch(C) is a segment of ch(C ) or vice versa.
(E3) If {i S [ch(C, D)], i S [ch(C , D )]} PL , then C = C and ch(C, D) is a segment of
ch(C , D ) or vice versa.
(E4) If {i F [ch(C)], i S [ch(C , D)]} PL , then ch(C) = C1 , . . . , Ci is a segment of
ch(C , D) = D1 , . . . , Dk , Dk+1 , . . . , Dn , where Dk = C and 1 i k < n.
(E5) No i F [ch(C)] or i S [ch(C, D)] I is a contradiction.
Proof. (E1) is immediate from (D4). (E24) are immediate from the definitions of and
)-conjunct and a [L \L ] . For (E5), first note that each i I consists of a r(LPL
PL
PL
conjunct. Since no proposition letter occurring in one occurs in the other, if i is a con )-conjunct, for any
tradiction, then so must be one of its conjuncts. Regarding the r(LPL
F
i [ch(C)] it is C which must be consistent since C = is excluded. For any i S [ch(C, D)]
)-conjunct is D, and ch(C, D) = D , . . . , D
the r(LPL
1
n with D1 = , Dn = D and
]-conjunct of any i F [ch(C)],
n = 1, so by (E1) D = is excluded. For the [LPL \LPL
n1
ch(C) = C1 , . . . , Cn F-Chain(C), it is i=1 (Ci , Ci+1 ), a conjunction of conjunctions
of non-negated and negated proposition letters, which cannot be a contradiction:
No conjunct (Ci , Ci+1 ), 1 i < n, is a contradiction: For any C = C C, (C ) =
(C ), and no (C ) occurs negated and non-negated in (Ci , Ci+1 ).
If (C ) occurs non-negated in (Ci , Ci+1 ) and negated in (Cj , Cj +1 ), i < j ,
then C = Ci+1 and C F-Succ(Cj ). So there is a ch(C ) F-Chain(C ), ch(C ) =
C1 , . . . , Ci , C , . . . , Cj , C , which violates (E1).
If (C ) occurs negated in (Ci , Ci+1 ) and non-negated in (Cj , Cj +1 ), i < j , then
C F-Succ(Ci ) and C = Cj +1 . PL Cj +1 Ci+1 , so PL C Ci+1 . So there are
O1 , O2 OFCi with C = Ci O1 , Ci+1 = Ci O2 , and PL (Ci O1 ) (Ci
O2 ). Then PL O2 (Ci O1 ), and with (CExtF ) PL O2 O1 . By minimality
O2 = O1 and C = Ci+1 , but (Ci , Ci+1 ) left (Ci+1 ) non-negated.
]-conjunct of i S [ch(C, D)], the case that (C ) occurs non-negated in
For the [LPL \LPL
(Ci , Ci+1 ) and negated in (C) is done like the second case above.
), I I B, I = :
For any B r(LPL
(E6) there is some designated i x I , i x = i F [ch] or i x = i S [ch], such that for all i F [ch ],
i S [ch ] in I , ch is a segment of ch ,
)-conjunct of i x PL-derives the r(L )-conjunct of any i I ,
(E7) the r(LPL
PL
(E8) i x is i F [ch] or i S [ch] with ch = D1 , . . . , Dn such that PL B Dn1 .
)-conjunct
Proof. (E6) is immediate from (E2-4) and finiteness of I . For (E7), the r(LPL
of i is Dn , where Dn is the last member of some ch = D1 , . . . , Dn such that i = i F [ch]
)-conjunct of i x is D for some 1
or i = i S [ch]. By (E6) the r(LPL
k n, and by
k
(E1) PL Dk Dn . For (E8), note that Dn1 exists as ch = is excluded by the
construction. If Dn1 = , then PL B Dn1 is trivial. Otherwise there must be some
ch = D1 , . . . , Dn1 such that i (ch ) I , * being F or S. By (E6), i (ch ) cannot be in
205
206
503
)-conjunct D
x
I , though the r(LPL
n1 derives the r(LPL )-conjunct Dn of i by (E1)
]-conjunct is derived
and hence that of any other i I due to (E7), while its [LPL \LPL
by that of i x . So it must be that {Dn1 } {B} PL , and PL B Dn1 .
):
Lemma 1 (Coincidence lemma). For all A, B r(LPL
I |= O F (A/B)
iff O F (A/B)
I |= O S (A/B)
iff O F (A/B)
Proof. Coincidence for O F .

Right-to-left. Assume O F (A/B) , so some OB OFB derives A. By (C6) PL ((CB
OCB ) B) OB for some CB C, OCB OFCB . By (D2), CB OCB C
{}, so i F [ch] I for some ch F-Chain(CB OCB ). If {i F [ch]} PL-derives
)-conjunct C O
B, only its r(LPL

B
CB can be relevant, since the [LPL \LPL ]conjunct is consistent (E5) and has no proposition letter in common with B.
If {CB OCB , B} PL , then OB = which contradicts P F (/B) by
(DPF ) and DDL{F,S} -consistency of . So {i F [ch]} PL B, so for some I
I B : I {B} PL CB OCB , hence I {B} PL A.
Left-to-right. Assume P F (A/B) and for r.a.a. suppose I I B: I {B} PL
A. Suppose I = , so let i x be the designated member of I and D its
)-conjunct. Then {D} {B}
r(LPL
PL A as the r(LPL )-conjuncts of any
i I are PL-derived by D (E7), and the [LPL \LPL ]-conjuncts are not rel ). i x is i F [ch] or i S [ch] for some ch =
evant for a derivation of A r(LPL
S
}. So
D1 , . . . , Dn with D = Dn = (Dn1 O), O OFDn1 {OD
n1
F
S
{O} PL B A, so O (B A/Dn1 ) or O (B A/Dn1 ) by
(B1), (B2). From P F (A/B) we get P F ((B A)/B) with (CExtF ).
By (E8) PL B Dn1 , so with (FH+F ) or (FH+SSF ) we obtain O F ((B
A) B/Dn1 ) or O S ((B A) B/Dn1 ) respectively. But
then O B follows from minimality of O. So {B} PL Dn1 O, i.e.,
/ I . So by (E6) I = . Then
{B} PL D, and since {i x } PL D we get i x
F
F
{B} PL A. With P (A/B) and (CExt ) we get P F (/B), so by (DN-RF )
B = and I B = , completing the r.a.a.
Coincidence for O S .
Right-to-left. Assume O S (A/B) and for r.a.a. suppose that there is some I
I B: I {B} PL A. Assume I = , so let i x be the designated member of I ,
)-conjunct. i x is i F [ch] or i S [ch] for some ch = D , . . . , D
and D its r(LPL
1
n
S
with D = Dn = (Dn1 O). Either O OFDn1 , then PL O OD
n1
S
. By (E8) PL B Dn1 , so
follows from (DCSF ), or trivially if O = OD
n1
S
S
S
by (Cond ) PL ODn1 (B OB ), and also I {B} PL O. Chaining the
results, we get I {B} PL OBS , and I {B} PL A by definition of OBS ,
), B = , we
contrary to what was assumed. So I = . For any B r(LPL
504
207
have i S [ch(CB , (CB OCSB ))] I S by (C4) and definition of I S . If I = ,

then {i S [ch(CB , (CB OCSB ))]} {B} PL , and {CB OCSB } {B} PL
)-conjunct is relevant. By (C4) we have
since only the r(LPL
PL B CB , so
{B OCSB } PL , and by (C5) OBS = . From (DP-RS ) we get B = , so
I B = , which completes the r.a.a.
/ . B = due to (DNS ) and (CExtS ), so i S [ch(CB ,
Left-to-right. Assume O S (A/B)
S
(CB OCB ))] I for some I I B (otherwise again B = ). If

i F [ch(C )] I , then ch(C ) is a segment of ch(CB , (CB OCSB )) by (E4), so
/ I
PL C CB and as PL B CB also PL C B. So i F [ch(C )]
and I I F = . If i S [ch(C , D )] I , then ch(CB , (CB OCSB )) is a seg )-conjunct D is derived by that
ment of ch(C , D ) by (E3), so its r(LPL
]-conjuncts are not relevant, so if
of i S [ch(CB , (CB OCSB ))]. The [LPL \LPL
S
I B PL A, then {CB OCB } {B} PL A. Since PL B CB , by (C4) then

{B OCSB } PL A, and by (C5) PL OBS A, so O S (A/B) by (B1), contrary
to the assumption. So I B PL A, so not for all I I B: I {B} PL A,
so I |= O S (A/B) is false.
6. A link to multiplex preference semantics

In the preceding section, the completeness theorem was proved by identifying a multiple
system of spheres. This multiple system of spheres can just as well be used to construct a
multitude of preference relations, whichas originated with Goble [13,14]can then in
turn be used to define the deontic operators: let P be a non-empty set of preference relations
P B B such that each P is transitive, connected, and satisfies the limit assumption:
(LA) If A = then bestP ( A ) =
where bestP ( A ) = {v A | v A : vP v }. For Hansson-type operators, let
+
LDDL{F + ,S} be like LDDL{F,S} except that O F replaces O F , and let the truth definitions
for the deontic operators read:
+
P |= O F (A/C)
S
P |= O (A/C)
iff
iff
P P: bestP ( C ) A
P P: bestP ( C ) A
Likewise, for Lewis-type operators, let LDDL{F,S } be like LDDL{F,S} except that O S replaces O S , and the truth definitions now read:
P |= O F (A/C)
P |= O S (A/C)
iff
iff
P P: v C A : v C A : not v P v
P P: v C A : v C A : not v P v
+
The axiomatic system DDL{F ,S} is like DDL{F,S} except that (DNF ) replaces (DN-RF )and
(DP-RF ) replaces (DPF ). Similarly, DDL{F,S } is like DDL{F,S} except that (DN-RS )
208
505
replaces (DNS ) and (DPS ) replaces (DP-RS ). So all systems only differ on the mindboggling [29] question whether everything or nothing is obligatory in impossible cir+
cumstances. DDL{F ,S} and DDL{F,S } are sound (cf. [14], also Arrows axiom: if
bestP ( C D ) C = , then bestP ( C ) = bestP ( C D ) C , is helpful). I have
no counterexample to compactness, so the semantics might just be compact. Weak completeness is easily obtained from the previous constructions, but seems not to have been
stated before, so I shall give the proof in full.
+
Theorem 5 (Completeness of DDL{F ,S} and DDL{F,S } ). The systems DDL{F ,S} and
DDL{F,S } are weakly complete with respect to the above multiplex preference semantics.
Proof. In proving DDL{F,S} -completeness, up till the coincidence lemma no use was made
+
use of unrestricted (DPF ) and (DNS ) missing in DDL{F ,S} and DDL{F,S } respectively.
So we can reuse and continue that construction with all pertaining lemmas in the canon+
ical construction for DDL{F ,S} and DDL{F,S } , with the implicit understanding that for
+ ,S}
{F
+
the index meant is F rather than F , and for DDL{F,S } the index meant is S
DDL
instead of S.
Let F-Chain(), S-Chain(C, ), be defined as before, C C. We only consider ch =
ch = D D
D1 , . . . , Dn that are in such a set. Let OD
i < n. Note that
i
i+1 , for any 1
i
S }, and O D by (CExt ), so
ch
F
Di+1 = Di O for some O ODi {OD
i
PL ODi
i
O. For any 1 i < j n:
(S1) Dj Di
ch = (and O ch O ch = with CExt , j = n)
(S2) Dj OD
Dj
Di
i
(S3) n =
ch O ch
(S4) OD
Dn1 = B
1
Proof. (S1) and (S2) are immediate from (E1) and the definition of ch. n is finite (S3)
) is finite and repetitions in ch are excluded. For (S4),
since r(LPL
ch
ch
ch
OD
= Dn1 (Dn1 Dn )
= Dn1 OD
B\ OD
n
1
n1
and Dn = by definition.
For any ch = D1 , . . . , Dn , v, v B, let Pch be such that
vPch v
iff
ch
v OD
,
i
ch
v OD
,
j
j <n
By (S2) and (S4), each v must belong to exactly one sphere. The index of each Ci is
transitive and connected, so Pch is as well. LA holds due to (S3) and (S4).
Let ch = D1 , . . . , Dn be as described above. Let Di be its smallest A-permitting
ch A = and j , 1
ch A = (we
sphere, i.e., a Di with OD
j < i < n: OD
i
j
write DA for Di ). We then obtain, for any ch and A = :
(S5) There is a DA ch,
506
209
(S6) A DA , and
ch A .
(S7) bestPch ( A ) = OD
A
Proof. (S5) is immediate from (S1), (S4) and D1 = . For (S6) let DA = Di ch, 1 i <
n: If A
DA , then A DA = , so by (S5) there is a DADA = Dj ch. If DA =
ch = O ch
ch
ch
DADA , then OD
DADA , by construction PL ODA DA , so ODADA
A
A DA = contrary to the definition of DADA . If i < j , then DADA DA ,
ch
but then DADA A DA = and again OD
A DA = . So
A DA
ch
ch
j < i, but then ODA A DA = implies ODA A = , so DA was not the
ch and P .
smallest A-permitting sphere. (S7) then follows from the definitions of OD
ch
A
Finally, let P = {Pch | ch F-Chain() C C S-Chain(C, )}. Note that P = : by
S , then , F-Chain(),
(D1) C, so even if O = for all O OF O
, S-Chain(, ), hence P , = B B P. So P is as required.
), A = and P P and saves us from having
The next lemma holds for all A r(LPL
to do separate proofs for the two systems:
(S8) bestP ( A ) B iff v A B : v A B : not v P v.
Proof. Assume bestP ( A ) B : A = , so bestP ( A ) = due to (LA). So v
bestP ( A ) s.t. v A B . Suppose v P v for some v A B : so v A and v
bestP ( A ) by transitivity of P and definition of best. But then bestP ( A ) B = ,
contradicting the assumption. Assume v A B : v A B : not v P v, and
for r.a.a. suppose that bestP ( A ) B = : So v bestP ( A ) B . Then v
A B by definition of best, and not v P v, as assumed. But v A , so v P v by
definition of best, which completes the r.a.a.
):
Lemma 2 (Coincidence lemma). For all A, B r(LPL
P |= O F (A/B)
iff O F (A/B)
P |= O S (A/B)
iff O F (A/B)
Proof. Coincidence for O F .

+
Case B = . If B = , then in the case of DDL{F ,S} , O F + (/) holds due to

+
+
+
(DNF ), so O F (A/) due to (CExtF ). Also, if B = , then for any
P P, bestP ( B ) = . P = , so P P: bestP ( B ) A holds for any
A, and both sides of the iff-clause are true, and so is the iff-clause. In the case of
DDL{F,S } , if B = , then P F (/) due to (DPF ), so P F (A/)

due to (CExtF ), and so by definition of , O F (A/)
/ . Also, if B = ,
then A B = , so for any P there is no v A B such that v
A B : not v P v. So both sides of the iff-clause are false, and the iff-clause
true.
210
507
Case B = . Right-to-left. O F (A/B) , so OB OFB : PL OB A, so by (C6)

OCB OFCB : OCB B = OB . Since CB C, there is a ch F-Chain() such
that ch = D1 , . . . , Di , Di+1 , . . . , Dn , 1 i < n, D1 = , Di = CB , Di+1 =
ch = O . For any 1
ch
j < i, OD
C OCB , and Dn = . By definition OD
CB
i
j
ch B = , and O ch B = , for
CB = due to (S2), so also OD
Di
j
otherwise OCB B = = OB , by (B2) O F (/B) , and by (DP-RF ) B = ,

ch B =
contrary to what was assumed. So Di = DB , and bestPch ( B ) = OD
B
OCB B = OB , and so bestPch ( B ) A and by (S8) P |= O F (A/B).
Left-to-right. Suppose P |= O F (A/B), so P P: bestP ( B ) A by (S8).
By construction there is some ch = D1 , . . . , Dn such that P = Pch , ch
F-Chain() or ch S-Chain(C, ) for some C C. As B = , by (S5) there is
ch B
some smallest B-permitting sphere DB in ch, with bestP ( B ) = OD
B
ch
ch
ch }
F
and ODB B = . Either DB C and ODB ODB : then {DB OD
PL
B
ch
ch
B, so O F (DB ODB /B) by (C7), PL B DB by (S6), so O F (OD
B
ch = O S , so
B/B) with (CExtF ) and O F (A/B) by (RWF ). Or OD
D
B
B
ch /D ) , and O S (B A/D ) by (RWS ). Assume O F (A/B)
/
O S (OD
B
B
B
, then P F (A/B) by definition of , so P F ((B A)/B) by
(CExtF ), and so with (FH+SSF ) we obtain O S ((B A) B/DB ) , so
S B, but then O ch B = , contrary to what was
by definition PL OD
DA
A
F
assumed. So O (A/B) .
Coincidence for O S .
+
Case B = . If B = , then in the case of DDL{F ,S} , O S (/) holds due to

(DNS ), so O S (A/) due to (CExtS ). Also, if B = , then for any P P,
bestP ( B ) = , so for all P P: bestP ( B ) A holds for any A, and both
sides of the iff-clause are true, as is the iff-clause. In the case of DDL{F,S } , if
B = , then P S (/) due to (DPS ), so P S (A/) due to (CExtS ),
/ . Also, if B = , then A B = , and
and by definition of , O S (A/)
since P = there is some P for which it is false that v A B : v
A B : not v P v, so it is not true for all P . So both sides of the iff-clause
are false, and the clause true.
Case B = . Right-to-left. Assume O S (A/B) , and for r.a.a. P O S (A/B), so by
(S8) P P: bestP ( B ) A = . By construction there is some ch =
D1 , . . . , Dn such that P = Pch , ch F-Chain() or ch S-Chain(C, ) for
some C C. Since B = , by (S5) there is some smallest B-permitting sphere
ch
ch
B and OD
B = . Either
DB in ch, with bestP ( B ) = OD
B
B
S
ch
ch
F
DB C and ODB ODB , or ODB = ODB : In both cases, since PL B DB
by (S6), we have O S (B A/DB ) from O S (A/B) and (CondS ), and
SF
ch }
ch
F
{OD
PL B A either by (DC ) and minimality of ODB ODB , or by definB
S . So O ch B = best ( B ) A , which completes the r.a.a.
ition of OD
P
DB
B
/ , so P S (A/B) .
Left-to-right. Suppose P |= O S (A/B), O S (A/B)
508
B = , so CB C by (C4), so let ch = D1 , . . . , Dn be in S-Chain(CB , ).

Then DB = CB : Suppose DB = Di and CB = Dj for j < i. By construction
ch = O S , and O S B = for otherwise O S = by (C6), which
of ch, OD
B
CB
CB
j
with (DP-RS ) derives B = , but this was excluded. So OCSB B = ,

so DB is not the smallest B-permitting sphere. Suppose DB = Di and CB =
ch
Dj for j > i, then CB OD
= by (S2), as PL B CB by (C4F ),
B
ch
B ODB = , and again DB is not the smallest B-permitting sphere.
ch = O S . Since P S (A/B) ,
So DB = CB and by construction of ch, OD
CB
B
S
PL OB A follows from (B1) and construction of . With (C5), (S7) we get
ch B = best
A
OBS = OCSB B = OD
Pch ( B ). So bestPch ( B )
B
S
and by (S8) P O (A/B), contradicting the assumption.
7. The puzzle is still incomplete

From a complete picture of dyadic deontic reasoning about conflicting imperatives, at
least two pieces are still missing. The first is that an imperative itself may be conditional
in a way irreducible to a material implication in its content: e.g., if Im to throw rice as
the wedding party leaves the church, but Huey, Dewey and Louie have stolen the bag,
blocking the doors wont garner me any praise. It has been argued that such conditional
imperatives have two associated propositions, the antecedent and the consequent, and that
obligations are only triggered, if the antecedents hold, thus providing the opportunity for
norm satisfaction or violation. Secondly, even though weighing out the relevant factors
may not always produce an unequivocal result, imperatives can be ordered by rank of the
issuing authority or normative weight: e.g., finding the victim of an accident on my way
to a crucial appointment, it seems clear what my obligations are and to not be the time for
skeptical or credulous reasoning.
In an attempt to tackle these complexities, Horty [21] proposed the following definition
of the imperatives binding in some circumstances A:
Binding(I ,<) (A) =def i I | (1) i TriggeredI (A),
(2) there is no j TriggeredI (A) such that
(a) i < j, and
(b) {consequent(i), consequent(j )} BL
where TriggeredI (A) =def {i I | A BL antecedent(i)}, and < is some strict partial order on I , i < j meaning that i ranks higher than j . The truth of the (skeptical or credulous)
dyadic deontic formula O (B/A) is then defined with respect to the set of consequents of
the imperatives in Binding(I ,<) (A).
Yet Hortys proposal is problematic for several reasons. First, the triggering condition does not capture all senses in which antecedents may hold. Consider the situation
(C D) and let I = C ! A, D ! A, !(A B) , where for short A ! B means
the imperative with antecedent(A) and consequent(B), !(A B) is the unconditional
211
212
509
imperative ! (A B), and the sequence represents the ordering <. Though we
do not know which imperative overrides the weakest imperative !(A B), we know
for sure that it is overridden in these circumstances and so should not be included in
Binding(I ,<) (C D), but with Hortys definition it is. I suggest that, for a better definition of the set TriggeredI (C D), we need an operation like Makinson and van der
Torres [32] basic output, which is expressly tailored to process such disjunctive inputs
(triggering conditions) intelligibly.
Secondly, the inconsistency check seems both too rigid and not rigid enough. For the
latter, let I1 = !(A B), !(B C) and I2 = !((A B) D), !D, !(B C) : in
both cases more important imperatives are in conflict with the weakest, but it is rejected
only in the first. For the former, let I = C ! D, C ! (B D) and the situation be
(C D). C ! (B D) is not in Binding(I ,<) (C D), its consequent contradicting
that of a more important imperative. But this has become unfulfillable, which intuitively
clears the way for obligatoriness of B. For a solution, I propose to leave inconsistency
checks entirely to the (credulous or skeptical) reasoning strategy defined via sets consistent
with the circumstances C: let each of these include a maximally C-consistent subset of
the most important triggered imperatives consequents, a maximal subset of the second
most important triggered imperatives consequents that can be C-consistently added to the
former, etc. This is the incremental maximizing employed for belief revision by Brewka
[6] and Nebel [34] (to work, < must be well-founded). Drawing on a parallel result by
Rott [37, Theorem 7], as long as conflicts between incomparable or equally important
imperatives are allowed, the logic for accordingly defined deontic operators should still be
DDL{F,S} .
Acknowledgement
I thank David Makinson for inspiring remarks, and Lou Goble and Leon van der Torre
for comments on an earlier version presented at EON 2004. The paper is part of a project
begun in [16] to relate deontic logic to reasoning about imperatives.
References
[1] C.E. Alchourrn, E. Bulygin, Unvollstndigkeit, Widersprchlichkeit und Unbestimmtheit der Normenordnungen, in: A.G. Conte, R. Hilpinen, G.H. von Wright (Eds.), Deontische Logik und Semantik, Athenaion,
Wiesbaden, 1977, pp. 2032.
[2] A. Bochman, Credulous nonmonotonic inference, in: T. Dean (Ed.), Proceedings of the 16th International
Joint Conference on Artificial Intelligence, IJCAI 99, Stockholm, Sweden, 1999, Morgan Kaufmann, San
Francisco, 1999, pp. 3035.
[3] A. Bochman, A Logical Theory of Nonmonotonic Inference and Belief Change, Springer, Berlin, 2001.
[4] A. Bochman, Brave nonmonotonic inference and its kinds, Ann. Math. Artificial Intelligence 39 (2003)
101121.
[5] S. Brass, On the semantics of supernormal defaults, in: R. Bajcsy (Ed.), Proceedings of the 13th International
Joint Conference on Artificial Intelligence, IJCAI 93, Chambery, France, Morgan Kaufmann, New York,
1993, pp. 578583.
510
[6] G. Brewka, Belief revision in a framework for default reasoning, in: A. Fuhrmann, M. Morreau (Eds.), The
Logic of Theory Change, Workshop, Konstanz, 1989, in: Lecture Notes in Artificial Intelligence, vol. 465,
Springer, Berlin, 1991, pp. 206222.
[7] D.O. Brink, Moral conflict and its structure, Philos. Rev. 103 (1994) 215247.
[8] A. Donagan, Consistency in rationalist moral systems, J. Philos. 81 (1984) 291309.
[9] D. Fllesdal, R. Hilpinen, Deontic logic: An introduction, in [19], pp. 135.
[10] L. Forget, V. Risch, P. Siegel, Preferential logics are X-logics, J. Logic Comput. 11 (2001) 7183.
[11] B. van Fraassen, Values and the Hearts command, J. Philos. 70 (1973) 519.
[12] L. Goble, Multiplex semantics for deontic logics, Nordic J. Philos. Logic 5 (2000) 113134.
[13] L. Goble, Preference semantics for deontic logics: Part Isimple models, Logique Anal. 46 (2003) 383418.
[14] L. Goble, Preference semantics for deontic logics: Part IImultiplex models, Logique Anal. 47 (2004).
[15] L. Goble, A logic for deontic dilemmas, J. Appl. Logic 3 (2005).
[16] J. Hansen, Sets, sentences, and some logics about imperatives, Fund. Inform. 48 (2001) 205226.
[17] J. Hansen, Problems and results for logics about imperatives, J. Appl. Logic 2 (2004) 3961.
[18] B. Hansson, An analysis of some deontic logics, Nos 3 (1969) 373398, reprinted in [19], pp. 121147.
[19] R. Hilpinen, Deontic Logic: Introductory and Systematic Readings, Reidel, Dordrecht, 1971.
[20] J.F. Horty, Nonmonotonic foundations for deontic logic, in: D. Nute (Ed.), Defeasible Deontic Logic,
Kluwer, Dordrecht, 1997, pp. 1744.
[21] J.F. Horty, Reasoning with moral conflicts, Nos 37 (2003) 557605.
[22] D. Jacquette, Moral dilemmas, disjunctive obligations, and Kants principle that Ought implies Can,
Synthese 88 (1991) 4355.
[23] H. Kelsen, Reine Rechtslehre, second ed., Deuticke, Wien, 1960.
[24] H. Kelsen, Allgemeine Theorie der Normen, Manz, Wien, 1979.
[25] A. Kratzer, Conditional necessity and possibility, in: R. Buerle (Ed.), Semantics from Different Points of
View, Springer, Berlin, 1979, pp. 116147.
[26] S. Kraus, D. Lehmann, M. Magidor, Nonmonotonic reasoning, preferential models and cumulative logics,
Artificial Intelligence 44 (1990) 167207.
[27] E.J. Lemmon, Moral dilemmas, Philos. Rev. 71 (1962) 139158.
[28] D. Lewis, Counterfactuals, Basil Blackwell, Oxford, 1973.
[29] D. Lewis, Semantic analyses for dyadic deontic logic, in: S. Stenlund (Ed.), Logical Theory and Semantic
Analysis, Reidel, Dordrecht, 1974, pp. 114.
[30] D. Lewis, Ordering semantics and premise semantics for counterfactuals, J. Philos. Logic 10 (1981) 217
234.
[31] D. Makinson, Bridges between classical and nonmonotonic logic, Logic J. IGPL 11 (2003) 6996.
[32] D. Makinson, L. van der Torre, Input/output logics, J. Philos. Logic 29 (2000) 383408.
[33] R.B. Marcus, Moral dilemmas and consistency, J. Philos. 77 (1980) 121136.
[34] B. Nebel, Syntax-based approaches to belief revision, in: P. Grdenfors (Ed.), Belief Revision, Cambridge
University Press, Cambridge, 1992, pp. 5288.
[35] D. Poole, A logical framework for default reasoning, Artificial Intelligence 36 (1988) 2747.
[36] N. Rescher, An axiom system for deontic logic, Philos. Stud. 9 (1958) 2430.
[37] H. Rott, Belief contraction in the context of the general theory of rational choice, J. Symbolic Logic 58
(1993) 14261450.
[38] P. Siegel, L. Forget, A representation theorem for preferential logics, in: L.C. Aiello, J. Doyle, S.C. Shapiro
(Eds.), Principles of Knowledge Representation and Reasoning (Proceedings of the 5th International Conference, KR 96), Morgan Kaufmann, San Francisco, CA, 1996, pp. 453460.
[39] W. Spohn, An analysis of Hanssons dyadic deontic logic, J. Philos. Logic 4 (1975) 237252.
[40] E. Stenius, The principles of a logic of normative systems, Acta Philos. Fennica 16 (1963) 247260.
[41] L. van der Torre, Reasoning About Obligations, Thesis Publishers, Amsterdam, 1997.
[42] L. van der Torre, Y.-H. Tan, Two-phase deontic logic, Logique Anal. 43 (2000) 411456.
[43] B. Williams, Ethical consistency, Proc. Aristotelian Soc. (supp.) 39 (1965) 103124.
[44] G.H. von Wright, A note on deontic logic and derived obligation, Mind 65 (1956) 507509.
[45] G.H. von Wright, Norm and Action, Routledge & Kegan Paul, London, 1963.
[46] G.H. von Wright, A new system of deontic logic, Danish Yearbook Philos. 1 (1964) 173182, reprinted in
[19], pp. 105115.
213
214
511
[47] G.H. von Wright, A correction to a new system of deontic logic, Danish Yearbook Philos. 2 (1965) 103107,
reprinted in [19], pp. 115120.
[48] G.H. von Wright, An Essay in Deontic Logic and the General Theory of Action, North-Holland, Amsterdam,
1968.
[49] G.H. von Wright, Deontic logicas I see it, in: P. McNamara, H. Prakken (Eds.), Norms, Logics and Information Systems, IOS, Amsterdam, 1999, pp. 1525.

Articial Intelligence and Law (2006) 14: 134
DOI 10.1007/s10506-005-5081-x
215
Springer 2006
Deontic logics for prioritized imperatives

JORG HANSEN
Institut fur Philosophie, Universitat Leipzig, Beethovenstrae 15, D-04107, Leipzig, Germany
e-mail: jhansen@uni-leipzig.de
Abstract. When a conict of duties arises, a resolution is often sought by use of an ordering of
priority or importance. This paper examines how such a conict resolution works, compares
mechanisms that have been proposed in the literature, and gives preference to one developed by
Brewka and Nebel. I distinguish between two cases that some conicts may remain unresolved,
and that a priority ordering can be determined that resolves all and provide semantics and
axiomatic systems for accordingly dened dyadic deontic operators.
Keywords: deontic logic, logic of imperatives, priorities
1. Introduction
W. D. Ross (1930) argued that whenever there appears to be a conict of
duties, through careful study of all aspects of the situation one will arrive
at the conclusion or rather: the considered opinion that one of these
duties is more pressing than others, and this duty is then ones duty sans
phrase, whereas the others were prima facie only. Ross gives the following
example:
EXAMPLE (The road accident). If I have promised to meet a friend at

a particular time for some trivial purpose, I should certainly think myself justied in breaking my engagement if by doing so I could prevent a serious
accident or bring relief to the victims of one.
There are two conicting obligations: to keep the promise, and to prevent the
accident or help its victims. The second takes priority: it is in these circumstances more of a duty than keeping the appointment.
While in the example the determination of the priority ordering seems to
rely on a comparison of the outcomes of satisfying or violating the conicting
duties under considerations of utility and possible harm, in the case of legal

2
216
JORG HANSEN
obligations or individual imperatives the ordering can often be directly

obtained from the norms position in a normative hierarchy or the rank of
the issuing authority. These factors may also relate to each other, e.g. when
the decision of a commander in the eld overrules that of her superior due to
some unforeseen danger or opportunity. I will leave aside the question of
how a particular ordering is determined, and also not address Rosss
notoriously problematic distinction between a prima facie duty and a duty
sans phrase. What interests me here is rather how a conict resolution
based on (established) priorities works, i.e. what the resolution mechanism
looks like, or should look like, when an ordering of priority or importance
of possibly conicting norms is assumed. Section 2 introduces the formal
framework and explains how it is used to dene deontic operators. After
pointing out counterintuitive results of a conict resolution based on a
method by Horty (2003), I show in Section 3 that a method developed for
the resolution of inconsistencies in prioritized theory bases by Brewka
(1989, 1991) and Nebel (1991, 1992) fares better (Section 3). A broader
comparison includes ordering based mechanisms by Alchourron and
Makinson (1981), Gardenfors (1984), Alchourron (1986), and variants
(Section 4). Section 5 explores what a priority ordering must be like to
resolve all possible conicts, and provides a sound and weakly complete
axiomatic system (which readers might nd familiar) for a corresponding
dyadic deontic operator. All formal proofs are delegated to the Appendix.
Section 6 concludes.
2. Imperative semantics and deontic logic

When a conict between norms is resolved by an appeal to some priority
ordering, I assume that what is thus conceived as ordered are the norms
themselves, though their ordering may reect a ranking of their sources, or an
axiological order of the states realized when fullling the norms. So for a
logical analysis, some formal representation of norms is required. I only
consider unconditional imperatives,1 like Invite Jones to dinner!, and I is a
set of such imperatives. To each imperative corresponds a descriptive sentence
like You invite Jones to dinner, which grammatically similar, but in the
indicative, not the imperative mood describes what must be the case if and
only if (i) the imperative is satised. Any such descriptive sentence is assumed to have a formalization in the language of a basic logic, which I let be
propositional logic PL.2 A function f : I ! LPL assigns every imperative in I
the LPL -formalization of its corresponding descriptive sentence, and the tuple
hI; f i is called a basic imperative structure. I write !A for an i in I with f (i)=A,
and use the superscripted i f ; C f instead of fi; f C for better readability. In
analogue to the usual concept of remainders, let IfA be the maximal sets of

DEONTIC LOGICS FOR PRIORITIZED IMPERATIVES
217
3
imperatives such that the sets of corresponding descriptive sentences do not

derive A (I also call these A-remainders of I), i.e. IfA contains all C I
such that (i) Cf 0PL A and (ii) there is no D I : C D and D f 0PL A.
In the imperativist tradition of deontic logic, authors used such
semantics to interpret deontic formulas, rather than employing the usual
possible worlds semantics.3 Let deontic formulas be those of a language
LDL , based on an alphabet like the one for LPL , except that it additionally
contains the operator symbol O, whereby OA formalizes the (true or false)
statement that what A describes is obligatory. LDL is then the smallest set
such that
(a) for all A 2 LPL ; OA 2 LDL ,
(b) if A; B 2 LDL , so are :A; A ^ B; A _ B; A ! B; A $ B.
Interpretations of Boolean operators being as usual, the truth denition
td-1
hI; f i OA
iff
I f PL A:
denes a normal modal logic, i.e. the set of LDL -sentences dened as true for
all tuples hI; f i equals the axiomatically dened set that contains all LDL instances into tautologies, furthermore all LPL -instances into
Ext
If PL A $ B; then OA $ OB is in the set:
M
C
OA ^ B ! OA ^ OB
OA ^ OB ! OA ^ B
O>
and is closed under modus ponens. Furthermore, the above truth denition
denes standard deontic logic SDL, which adds the deontic scheme (D):
D OA ! PA
i hI; fi is required to be such that I f is consistent (as usual, PA abbreviates :O:A). Requiring I f to be consistent excludes conicts between
imperatives and is thus a severe and in this case unwanted restriction, for to
show how conicts are resolved they must rst be semantically modeled.
But if e.g. two imperatives !p1 and !( p2 ^ :p1 ) can both be in I, then not
only does (D) fail, but also (td-1) is not very useful, making OA true for
any A 2 LPL . Instead, the following denition for a disjunctive ought
4
operator was put forward:
td-2
hI; f i OA
iff
8C 2 If ? : Cf PL A:
So OA is true when all maximally consistent subsets of what the imperatives

demand derive A. It is apparent that the denition tolerates conicts and e.g.
if !p1 and !(p2 ^ :p1 ) are in I, then O(p1 p2), but not O^ is true. Moreover,
218
JORG HANSEN
this solution is easily adapted to the dyadic case and the related problem of
dilemmas. Dyadic deontic logic uses a language LDDL that employs the
additional auxiliary sign / and is like LDL , except that clause (a) now reads
(a) for all A; C 2 LPL ; OA=C 2 LDDL ,
where O(A/C) is read as A is obligatory in the circumstances characterized
by C. The truth denitions for dyadic deontic formulas should allow some
inuence of the circumstances, so e.g. if John must either not impregnate
Suzy Mae or marry her, and she is in fact pregnant by him, then marrying her
seems to be what he must do. But simply putting
td-3
hI; f i OA=C
iff
I f [ fCg PL A:
will not suce if subjects can get (themselves) into dilemmas, i.e. situations
where the norms are collectively satisable at the outset, but due to misfortune or failure they cannot all be satised anymore. To handle such situations, and to e.g. prevent the derivation of O(^/p1) when I contains
!:p1 _ p2 and !:p2 ^ p3 , the truth denition for a disjunctive dyadic
ought operator can be given as:
td-4
hI; f i OA=C iff
8C 2 If:C : Cf [ fCg PL A:
So A is obligatory in the situation described by C if A is what the imperatives

in any :C-remainder demand, given C. With usual truth conditions for
Boolean operators, this semantics has a sound and (weakly) complete axiom
system PD dened as containing all LDDL -instances into tautologies, all
LPL -instances into
CExt If PL C ! A $ B
ExtC
If PL C $ D
then PD OA=C $ OB=C
then PD OA=C $ OA=D
DM OA ^ B=C ! OA=C ^ OB=C

DC OA=C ^ OB=C ! OA ^ B=C
DN O>=C
DD-R If 0PL :C then PD OA=C ! PA=C
Cond OA=C ^ D ! OD ! A=C
CCMon OA ^ D=C ! OA=C ^ D
and closed under modus ponens.5 PD resembles the system P dened by

Kraus et al. (1990) with the (restricted) dyadic deontic scheme (DD-R)
added, hence the name.
For the present purposes, I dene a prioritized imperative structure to be a
tuple hI; f; <i that is like a basic imperative structure, except that it additionally includes an ordering relation < on I, where the formal properties of
this relation are for the moment left open. Unfortunately, authors disagree

219
5
on the direction in which i1 < i2 is to be read, if it means that i1 takes

priority over i2 (reading < like a preference relation), or that i1 is less
important than i2 (reading < like a utility function). I assume the former,
and adapt diering denitions to this convention, so e.g. a tuple
i1 < i2 < i3. . . is read like a list that starts with what is most important. For
any ordering < on some set C; I dene min< C fi 2 C j 8i0 2 C : if i0 6 i,
then i0hig, so min< I is the set of the highest ranking, most important, etc.
imperatives, and max< C fi 2 C j 8i0 2 C : if i0 6 i, then ihi0 g, so max< I are
the imperatives that come last, are least important, rank lowest, etc.
3. Reasoning with prioritized imperatives

As explained by Ross, in the case of a normative conict one proceeds by
examining the situation for clues to an ordering of the obligations involved,
e.g. by considering the rank of the issuing authority, notions of urgency or a
gross dierence in the utilities of the outcomes. The example of the road
accident illustrates that the disjunctive ought operator dened in the previous
section, which pays no attention to priorities, produces inadequate results:
EXAMPLE (The road accident: disjunctive reasoning). Let A be

helping the accident victims, B keeping the promise, and T a conjunction of
actual necessities, including the agents present physical and psychical capabilities (I write for :T ). An imperative interpretation produces I={!A, !B}
so
and PL T ! A ! :B as helping causes me to miss the meeting.
(td-1) makes O true and the impossible obligatory, so it is not very useful.
so (td-4) makes O(AB/T) true but O(A/T) false, as
fB; Tg0PL A. So there is only a disjunctive obligation to help or proceed to the
meeting. But intuitively, helping takes priority over anything else.
The situation looks like a conict: there exist requirements which cannot
all be satised. But the conict is avoided by (so far intuitively) giving
priority to the norm of greater weight. Note that if symmetrical or incomparable obligations are not ruled out, then a demand that takes priority can
not only dissolve a dilemma, but also create a conict for an otherwise
conict-free situation:
EXAMPLE (The road accident II). It is Tuesday afternoon, and like on

all Tuesdays, Mirjam must fetch her grandmother from the day care center
before it closes at 6:30. Today, Mirjam was also asked by her boss to bring the
oce mail to the post oce after hours, which also closes at 6:30, but lies in the
opposite direction. However, when she told of her other duty, she was allowed
to leave early. Driving at 5:30 in the direction of the post oce, Mirjam

6
220
JORG HANSEN
becomes involved in a trac accident. The law requires her to stay at the
accident site until the police have recorded it, which wont happen before 6:00.
Then she can only get to one place, the post oce or the day care center, on
time. The law takes priority over her other duties, but a ranking of these is not
obvious; in particular it is dicult to say which violation could have worse
consequences, and Mirjam will have a hard time making up her mind.
To formalize the reasoning about priorities when faced with conicting
demands, Horty (2003) proposed that the priority ordering is used to rst
determine a set of binding imperatives in the set of all imperatives:
DEFINITION 1 (Binding imperatives). Let hI; f; <i be a prioritized

imperative structure. Then
So an imperative is binding if there is no higher ranking imperative with a
materially inconsistent demand (cf. Horty 2003, p. 560). If it is also higherranking than any such imperative, Horty calls it overriding. Binding,
instead of I, is then used to dene a disjunctive dyadic ought operator:
td-5
hI; f; <i OA=C
iff
8C 2 Binding f:C : Cf [ fCg PL A
I examine how this denition6 copes with the examples and variants:
EXAMPLE (The road accident: Hortys solution). A is helping the

accident victims, and B keeping the promise. So I={!A, !B} models the logical
situation, where PL T ! A ! :B, as the situation excludes both helping at
the accident site and meeting my friend. The ordering is !A < !B, so !A overrides !B and Binding = {!A}. (td-5) makes OA=T; O:B=T and P:B=T
true, so helping is obligatory, keeping the appointment forbidden, and not
keeping the appointment permitted, which is as it should be.
EXAMPLE (The road accident II: Hortys solution). Let A be

Mirjams staying at the accident site, B taking her bosss mail to the post
oce, and C fetching her grandmother from the day care center. I={!A, !B, !C}
is the logical model of the normative situation. Mirjam left early enough
to get to both the post oce and the day care center in time, so
0PL T ! B ! :C, but PL T ! A ! :B ^ C as waiting excludes her
accomplishing both. The legal obligation takes priority, so !A < !B and
!A < !C, while the ranking between !B and !C unclear. Binding = I, as the
truth of A only excludes satisfying both !B and !C, but getting to one place
remains possible. Binding f ff!A; !Bg; f!A; !Cg; f!B; !Cgg, (td-5) making
T

221
7
OA=T false as fT; B; Cg0PL A, and only OA ^ B _ C _ B ^ C=T

true, so it seems Mirjam can choose to wait or go on driving to her destinations,
which is counterintuitive.
EXAMPLE (The road accident III, with Hortys solution).

Things are as in variant II, but suppose some grave danger arises from Mirjams
not being at the day care center before it closes, as the disturbed old lady will
wander o on her own and may fall or get lost. Hence, fetching her is much
more urgent than posting the letters, of which the important ones were very
likely faxed beforehand. Fetching her grandma may be even more important
than waiting for the police, but Mirjam is not sure about that and can do both
anyway. So beside !A < !B we have !C < !B, with the ranking between !A and
!C unclear. Intuitively, Mirjam must stay until the police are nished and then
fetch her grandma. But as it is two higher-ranking imperatives that exclude, if
satised, the satisfaction of the lower-ranking one, still Binding = {!A, !B,
!C}=I, making OA=T false and even PB ^ :C=T true even though !C ranks
higher than !B.
EXAMPLE (The road accident IV, with Hortys solution).

Mirjam did not dare ask her boss for permission to leave early, sneaking out at 5:45
instead, but that was too late to get to both places in time, i.e. PL T ! C ! :B.
Again, Mirjam gets involved in a trac accident and is required to wait for the
police. Fetching her grandmother takes priority over posting the mail, so I={!A,
!B, !C}, !C < !B and !A < !B. Imagine the accident left her car a wreck,
making it impossible to get to the day care center in time, but when the police
nish around 6:15 she can still get to the post oce. Let S be this situation
(including T), so PL S ! :C. Binding = {!A, !C}, as !B is not reinstated
when satisfying !C is excluded, so Binding f:S ff!Agg, making P:B=S
true. But it is hard to see why Mirjam should not have to post the letters.
EXAMPLE (The road accident V, with Hortys solution). As in

variant IV, Mirjam left too late to make it to both the post oce and the day
care center on time, so PL T ! C ! :B. Again, fetching grandma takes
priority over posting the letters, i.e. !C< !B. Suppose it is not the damage, but
the time required by the police that makes it impossible to get to the day care
center on time (it is too far from the accident site, while the post oce is just a
block away), so PL T ! A ! :C. Making up her mind, Mirjam decides
that the legal obligation to wait for the police probably takes priority over her
familial duty, i.e. !A < !C < !B. Both !B and !C are overridden by higher
ranking imperatives and so are not in Binding f:S ff!Agg, making
P:B=S true. But again it is hard to see why Mirjam is relieved from posting
the letters.
222
JORG HANSEN
Thus Hortys set Binding solves simple cases, but is not adequate for
complex hierarchies where more than two imperatives may be in conict, and it
makes life too easy when conicting higher ranking imperatives become unfulllable or are themselves overridden.7 To overcome these diculties when
formalizing that disregard for a lower ranking imperative can (only) be excused
by obedience to higher-ranking ones (and not vice versa), I suggest that neither
remainder sets of I, nor of a xed subset Binding, but an incremental maximizing strategy should be used. For a situation C, the relevant sets are constructed by rst adding a maximal set of the most important imperatives such
that their demands do not derive :C, then adding a maximal subset of the
second most important imperatives that can be added without the corresponding demands now deriving :C, etc. Introduced by Rescher (1964, p. 50),
such incremental maximizing was more rigorously dened and employed for
the purpose of theory revision by Brewka (1989, 1991) and Nebel (1991, 1992).
Both employ a strict partial order, i.e. < is irreexive and transitive. Nebel
additionally assumes < to be the asymmetric part of a complete preorder ,
i.e. obtained from a reexive, transitive and connected ordering via dening
i<j i i j and j 6 i. Both agree that < must be well-founded, i.e. innite
descending chains are excluded.8 For any <, Brewka denes a full prioritization to be any (strict) well-order on the given set that preserves <, i.e. for all i,
j: if i< j then i j. Clearly:
THEOREM 1 (Existence of full prioritizations). For every wellfounded strict partial order < on a set C there is a full prioritization , i.e. a
strict well-order that is order-preserving with respect to <.
Brewka then denes subsets of the set as preferred subtheories. Calling
them preferred remainders (they are not theories here), his denition translates thus:9
DEFINITION 2 (Brewkas preferred remainders). Let hI; f; <i be a

prioritized imperative structure, where < is a well-founded strict partial order
on I. Then C I belongs to the preferred remainder set I + A iff (i) Cf 0PL A,
and (ii) C is obtained from a full prioritization by defining
SA
#i
(S
S
A
f
f
SA
#j [ fig if ji S#j [ fi g0PL A, and
S
A
otherwise,
ji S#j
ji
for any i 2 I, and letting SA

i2I
A
SA
#i and C S .
(i) bans the empty set from I + A for tautological A, and (ii) recursively
denes S#i to include all elements of some such set for a prior element j,
adding i if possible without the corresponding set deriving A. C is the union

223
9
of all such sets. I drop superscripts if the meaning is clear. The following is
almost immediate:
THEOREM 2 (Preferred remainders are remainders). Let hI; f; <i

be a prioritized imperative structure, where < is a well-founded strict partial
order on I. Then I + A IfA.
As noted, Nebels (1992) approach denes < as the asymmetric part of a
complete, well-founded preorder . For each i 2 I, the priority class
[i ]={j 2 I | i j and j i} contains all j 2 I of the same -priority as i. A
preference-ordering N between all subsets D; C of I is then dened by letting
D N C
iff
9i 2 I : 8j < i : C \ j D \ j and C \ i D \ i
i.e. by preferring D over C i both agree for all priority classes up to some [i],
of which D contains all elements of C plus more. Then choosing a maximally
N-preferred set among all C I with Cf 0PL A equals choosing from I + A:
THEOREM 3 (Nebels prioritized removals). Let hI; f; <i be a

prioritized imperative structure, where < is the asymmetric part of a complete,
well-founded preorder on I. Then I + A equals
fC I j Cf 0PL A and 8D I : if D N C then D f PL Ag:
There is an alternative, non-constructive denition of Brewkas preferred remainders, attributed to Ryan (1992) by Rintanen (1994) and also
appearing in Sakama and Inoue (1996): Let hI; f; <i be a prioritized imperative structure, where < is a well-founded strict partial order on I, and dene
p IfA to be the set
fC 2 IfA j 9 : 8D 2 IfA n fCg : 9i 2 C n D : 8j 2 D n C : i jg:
So some A-remainder C is in p IfA i for some full prioritization ; C

contains for any other A-remainder D some exclusive element that -ranks
higher than any element exclusively in D. The following holds.
THEOREM 4 (Preferred remainders, after Ryan and Sakama &

Inoue). Let hI; f; <i be a prioritized imperative structure, < a well-founded
strict partial order on I and I + A and pIfA be as defined. Then
I + A pIfA.
Proposing use of Brewkas and Nebels concept of preferred remainders,
based on some strict partial, well-founded ordering <, as the resolution
mechanism for conicts between imperatives or dilemmas that arise in certain
situations, a disjunctive ought operator can be dened parallel to (td-4) as
follows:
224
10
JORG HANSEN
td-6
I OA=C iff 8C 2 I + :C : Cf [ fCg PL A:
This truth definition fares better in dealing with the above examples:
EXAMPLE (The road accident: the Brewka/Nebel solution). A is

helping the accident victim, B keeping the promise, I={!A, !B} and
PL T ! A ! :B. The ordering is !A < !B, being its only full prioritization.
The construction of S< includes !A but rejects !B, so I + ={{!A}} and (td-6)
makes true OA=T; O:B=T and P:B=T, which is as it should be.
T
EXAMPLE (The road accident II: the Brewka/Nebel solution).

I f!A; !B; !Cg: PL T ! A ! :B ^ C, as waiting allows Mirjam to get to
one place, the post office or the day care center in time, but not both. The
ordering is !A < !B and !A < !C, and for !B, !C unclear. Its two full prioritizations are !A !B !C or !A !C !B, producing I + ={{!A, !B}, {!A,
!C}}. Then (td-6) makes true O(A (B C)/T), so waiting and then going to
one place, the post office or the day care center, is obligatory as it should be.
T
EXAMPLE (The road accident III: the Brewka/Nebel solution).

Still PL T ! A ! :B ^ C, i.e. waiting excludes getting to both places.
Fetching her grandma now takes priority over going to the post office, so
!C < !B and !A < !B, this time the ranking between !A, !C being unclear. The
two full prioritizations are !A !C !B and !C !A !B, so I + ={{!A,
!C}}, and (td-6) makes O(A C/T ) true. So Mirjam must stay at the site until
the police are finished with her and then go to fetch her grandmother, as it
should be.
T
EXAMPLE (The road accident IV: the Brewka/Nebel solution).

Getting to both the post office and the day care center on time was never
possible, so PL T ! C ! :B. The car being wrecked, the assumed situation
S excludes getting to the day care center on time, so PL S ! :C. Again
!C < !B and !A < !B, with the relation between !A and !C unclear, so
!A !C !B and !C !A !B are the full prioritizations, producing
I + :S ff!A; !Bgg. Hence O(A B/S), i.e. Mirjam must wait and then hurry
to the post office, which is as it should be.
EXAMPLE (The road accident V: the Brewka/Nebel solution).

Again, Mirjam cannot get to both the post office and the day care center on
time, so PL T ! C ! :B. Waiting for the police excludes getting to the day
care center on time, i.e. PL T ! A ! :C. Mirjam decides that the law

225
11
overrides her familial duty, so !A < !C < !B, which, being its only full
prioritization, yields I + ={{!A, !B}}. So Mirjam must wait and post the
letters, as it should be.
T
Let a semantics be called a prioritized imperative semantics i it denes the

truth of LDDL -sentences using (td-6), with respect to arbitrary prioritized
imperative structures hI; f; <i. Then it may be surprising though Rott
(1993, Theorem 7) already proved a similar result that the logical properties
of such a semantics are not dierent from that dening the deontic operator
using (td-4), i.e. with respect to basic imperative structures and simple
remainders, since the system PD remains sound and (only) weakly complete:
THEOREM 5 (Soundness, completeness of PD). PD is sound and

(only) weakly complete with respect to prioritized imperative semantics.
4. Alternative resolution mechanisms
4.1.
LEAST EXPOSURE AND ITS VARIANTS
Alchourron and Makinson (1981) seem to have been the rst to logically
examine the idea of resolving contradictions in a body of norms, or contradictions that arise from such a body together with some set of true
empirical facts, by imposing an order upon that body. The object of their
study is a set of regulations that is partially ordered by a relation , which
does not necessarily stand for an ordering by priority or importance. Rather,
i j means that j is as much exposed, or more exposed, to the risk of
legislative derogation as i. If a conict occurs between two parts of the code,
or between the code and some empirical facts, the aim is to nd a (possibly
maximal) non-conicting subset that is most secure from the changes which
the law-giver will presumably enact upon learning of this situation. Their
denition translates to the present framework as follows:
DEFINITION 3 (Alchourro n and Makinsons strict exposure).

Let hI; f; <i be a prioritized imperative structure, where < is a strict partial
order on I (like the asymmetric part of a partial order ). Then for all
C; D I :
C AM D
iff
D 6 [ and 8i 2 C : 9j 2 D : i < j:
So a subset is strictly less exposed than some other if for any member
of the rst there is a member of the second which is strictly more exposed.
To see how this approach compares to Brewka and Nebels, consider three
cases:

12
226
JORG HANSEN
Case 1. Let hI; f; <i be !p1 < !p2 ^ :p3 <!p3 , the imperative in the
middle conicting with the lowest. Then we have
If ? ff!p1 ; !p2 ^ :p3 g; f!p1 ; !p3 gg and
I +? ff!p1 ; !p2 ^ :p3 gg.
The exposure criterion yields f!p1 ; !p2 ^ :p3 g AM f!p1 ; !p3 g: for each left
member, a right member is strictly more exposed, namely !p3.
Case 2. Let hI; f; <i be !p1 < !:p1 ^ p2 < !p3 , so the middle now
conicts with the higher-ranking imperative. Then we have
If ? ff!p1 ; !p3 g; f!:p1 ^ p2 ; !p3 gg and
I +? ff!p1 ; !p3 gg.
But f!p1 ; !p3 g 6AM f!:p1 ^ p2 ; !p3 g: from the left set, !p3 is not less exposed
than !:p1 ^ p2 from the right. The authors recognize that a conict between
higher-ranking norms excludes lower-ranking norms from a least exposed set
and propose to use relevant logic for determining conicts as a cure (Alchourron and Makinson 1981, p. 139).
Case 3. Let hI; f; <i be !p1 < !:p1 ^ p2 ^ :p3 < !p3 , the middle now in
conict with both ends of the hierarchy (by whatever logic). Then we have
If ? ff!p1 ; !p3 g; f!:p1 ^ p2 ^ :p3 gg and
I +? ff!p1 ; !p3 gg.
Yet f!:p1 ^ p2 ^ :p3 g AM f!p1 ; !p3 g: from the right set, !p3 is more exposed than any left member. Mediocrity rules! But even if !p3 is more exposed
to legislative change, if that change came about and removed !p3, the right set
would still contain a member that ranks higher than any in the left.
Prakken, pursuing an argumentative approach, wants to employ Alchourron and Makinsons criterion at the heart of his rebuttal mechanism
used to determine justied arguments (derivations from facts and defaults.
But the criterion he presents (Prakken 1997, p. 192) translates dierently:
DEFINITION 4 (Prakkens criterion for hierarchical rebuttal).

Let hI; f; <i be a prioritized imperative structure, where < is a strict partial
order on I (like the asymmetric part of a partial order ). Then for all C; D I:
C P D
iff
9j 2 D : 8i 2 C : i<j:
So D can be improved by exchanging some member with any member of

C. Giving the rationale here and in Prakken and Sartor (1997, p. 36)10, the
change in the order of the quantiers seems intentional yet it makes a
dierence: let I be f!p1 ; !p2 ; !p3 ^ :p1 ; !p4 ^ :p2 g and !p1 < !p3 ^ :p1 and

227
13
!p2 < !p4 ^ :p2 . E.g. p1, p2 may be primary targets and p3 ^ :p1 ; p4 ^ :p2
respective secondary ones, where reaching the secondary target includes
failing to reach the (better) primary one. Reaching both primary targets
seems best, and in fact f!p1 ; !p2 g AM f!p3 ^ :p1 ; !p4 ^ :p2 g: for every
member in the left set there is a lower-ranking one in the right. Also
I + ? ff!p1 ; p2 gg, since all four full prioritizations yield this preferred
remainder. But f!p1 ; !p2 g 6P f!p3 ^ :p1 ; !p4 ^ :p2 g as no member in the
right set ranks lower than all in the left. So Prakkens criterion appears even
less suited to our task than Alchourron & Makinsons.11
Sartor (1991) used Alchourron and Makinsons criterion for a
prevailing relation between subsets modulo a rejected sentence A,12 as
follows:
DEFINITION 5 (Sartors prevailing relation). Let hI; f; <i be a

prioritized imperative structure, where < is a strict partial order on I, and
AM be as defined above. Then for all C; D I :
C A
S D
iff C [ D f PL A; and 8D0 D such that C [ D0 f PL A :

9C0 C : C0 [ D0 f PL A and C0 AM D0 :
Finally pref I f A minAS I f A.

To see how his denition works, consider rst the mediocrity rules
example: hI; f; <i is !p1 <!:p1 ^ p2 ^ :p3 <!p3 . Then f!p1 ; !p3 g ?
S f!:p1
^ p2 ^ :p3 g, as the only subset of the right set conicting with the left set is
the right set itself, and for this set some conicting subset of the left set,
namely {!p1}, is strictly less exposed than the right set. So now the result is as
it intuitively should be. Sartors relation also handles the example against
Prakkens criterion well: here I f!p1 ; !p2 ; !p3 ^ :p1 ; !p4 ^ :p2 g, with
!p1 <!p3 ^ :p1 and !p2 <!p4 ^ :p2 . Then f!p1 ; !p2 g ?
S f!p3 ^ :p1 ;
!p4 ^ :p2 g, as the three subsets of the right set conicting with the left
set are f!p3 ^ :p1 g; f!p4 ^ :p2 g and the right set itself, with which
the following respective subsets of the left set are both in conict and strictly
less exposed: {!p1}, {!p2}, and the left set itself. So reaching the primary
targets is best, as it should be. In fact, it can be proved that preferred remainders are always prevailing remainders, but the converse does
not hold:
EXAMPLE (Counterexample to pref IfA I + A. Let hI; f; <i

be such that I consists of
i1 : !p1 ^ p2 _ p3 ! q
i3 : !p2
i2 : !:p1 ^ p2 ^ p3 ! :q i4 : !p3 ^ p1 ! :p2

14
228
JORG HANSEN
and the ordering i1 < i2 < [i3, i4]. Intuitively, to satisfy i1 takes priority, and
then the only choice is the one between the equally ranking i3 and i4, and indeed,
I + ? ffi1 ; i3 g; fi1 ; i4 gg. But the remainder {i2, i3, i4}, missing the most
important imperative i1, is also in prefI f ?, as
a fi1 ; i3 g 6 ?
S fi2 ; i3 ; i4 g;
b fi1 ; i4 g 6?
S fi2 ; i3 ; i4 g:
For (a), consider {i4}, which is a subset of the right hand set: it conflicts with the
left hand set, so for the relation to hold, a strictly less exposed subset of the left
set must also conflict with {i4}. The only such subset is the left set itself, but since
for its member i3 there is no strictly more exposed member in {i4}, it is not
strictly less exposed. The refutation of (b) works similarly using {i3}.
So Sartors denition also produces counterintuitive results where Brewka
and Nebels approach does not.13
4.2.
UTILITY-REFLECTING PRIORITIES
Regarding the neighboring realm of epistemic logic, and the related problem
of revising belief sets in the face of conicting information, such information
often nds the reasoner less willing to give up some beliefs than others. In an
attempt to allocate this ordering of epistemic importance a role in determining which of the contradictory beliefs should be given up, Gardenfors
(1984) proposed the following: Let K be a belief set (set of descriptive sentences) that is the logical closure of some nite basis, and a relation (of
epistemic importance) that is a complete preorder on this set, which additionally ranks logically equivalent beliefs equally. For any remainder
C 2 K ? A there is then a spanning sentence SC in C that derives any
element in C. Then for any C; D 2 K ? A:
C G D
iff SC <SD
So a remainder is preferred to some other i its spanning sentence is epistemically at least as important as that of the other. It is essential for the
construction that < is a complete ordering on K, which due to logical closure includes the spanning sentence that is the sum of a remainder. But, the
logical philosophers not being kings, a set of imperative-contents is rarely
logically closed, which precludes a direct parallel. Yet, choosing subsets that
in sum are the most important has an analogue if the ordering of the
imperatives reects not so much their importance or rank of the source, but a
measure of goodness or utility of the outcome when satisfying the
imperative. For this, let the (well-founded, strict partial) order <u correspond to a function u : X ! R, with I f X LPL , that assigns a real
229
15
number to (at least) what the imperatives demand in a manner conversely

respecting <u, i.e. if i <u j then u(i f) > u( j f ). Then let
I u A fC IjCf 6PL A and 8D I : if
uD f
uCf then D f PL Ag:
So an A-consistent subset is preferred to another if the good brought about by

satisfying all of its demands sums up to a higher value than by doing so for the
other set. I u A includes the maximally preferred among such sets. Obviously
I u A IfA if u assigns just positive values. The O-operator is then dened by
td-7
hI; f; <u i OA=C iff
8C 2 I u :C : Cf [ fCg PL A:
If a set of imperatives that requires A to be true in the situation C constitutes

a reason to call A obligatory in this situation, then O(A/C) is true if the
reasons for obligatoriness A have more collective weight (sum up to a higher
value) than any such reasons for :A.14 Well-foundedness of <, with the
condition that u respects <, excludes innitely increasing utilities and thus
corresponds to the limit assumption in preference semantics, which avoids
counterintuitive results but may be criticized as supercial (cf. Fehige 1994).
Still, innitely lower and lower ranking imperatives, whose satisfaction
nevertheless produces some good, are not excluded, and thus not the scenario
of McNamara (1995) where a bad act like killing your mum will eventually
become permitted by (only then) piling up good deeds of small value. To
protect imperatives from getting overruled by inferiors in this manner, one
could assign to an imperatives satisfaction a utility that is absolutely higher
than the sum of the utilities assigned to the satisfaction of any number of
lower-ranking imperatives (think of the sequence 1, 0.5, 0.25, 0.125,
P ...). It is
u({ j 2 I | i
immediate that if <u is thus protected, i.e. for all i 2 I, u(if)
<u j}f), and u assigns just positive values, then I u A I + A.
4.3.
SAFE CONTRACTION AND A MODIFICATION
What characterizes Brewka and Nebels approaches is that they try to

maximize the number of higher-ranking imperatives in a set that avoids a
conict. This intuition has a counterpart formulated by Alchourron (1986):15
It is logical to believe that the reasonable way of overcoming a
conflict of obligations is to leave aside the less important norms
contributing to its creation.
Alchourrons proposal is then to remove from the set of norms all those that
are least-ranking in a minimal conicting subset, thus removing the conict
as well. For the formal description of this safe contraction, let IgA be the
minimal sets of imperatives such that the corresponding sentences derive
A (the A-kernels of I), i.e. the set of all C I such that (i) Cf PL A and

16
230
JORG HANSEN
(ii) for all D I, if D C then D f 0PL A. < being a strict partial ordering on
I, the following denes the set of all the least important imperatives of
A-kernels of I:
rIgA def fmax< C j C 2 IgAg:
Due to PL-compactness, any C 2 I gA is nite and so max< C 6 if

C 6 . Then the set I/A of elements of I that are safe with respect to A is
dened by
I=A
def
InrI gA:
If C characterizes the situation, the dyadic deontic operator can then be

dened as in (td-3), but using I=:C instead of I:16
td-8
hI; f; <i OA=C
iff I=:C f [ fCg PL A:
To see how this works, consider again the example of the road accident:
EXAMPLE (The road accident: safe contraction). I={!A, !B},

with !A < !B, as helping the victims of the accident is more important than
proceeding to the appointment. PL T ! A ! :B, because staying excludes
meeting my friend.
hence
= {!A}. Hence I must stay and help, as it should be.
EXAMPLE (The road accident V: safe contraction). I={!A, !B, !C}.

Mirjam left too late for both the post office and the day care center, so
PL T ! C ! :B. Waiting makes her too late for the day care center, so
PL T ! A ! :C. The legal requirement to stay and her duty to fetch her
grandma are both more important than posting the letters, and Mirjam decides
that the law also overrides her familiar duty, so !A<!C<!B. Intuitively
Mirjam must wait and then hurry to the post office. But
f!A; !Cg; f!B; !Cgg; max< f!A:!Cg f!Cg; max< f!B; !Cg f!Bg: So
and
= {!A}. Hence a solution by safe contraction only
requires Mirjam to wait for the police, though she could still post the letters when
the police are finished.
The last example illustrates that safe contraction removes elements too
liberally; even when it has already removed some element of a kernel, further
elements get removed as well even though under the denition of a kernel
removing one suces: !C was removed due to its conict with !A, so there
was no need to also remove !B to avoid the conict with !C. This makes life
231
17
too easy for the norm subjects, and some moderation appears necessary.
Following the idea that the removal mechanism should somehow be adjusted
to the sets shrinking it brings about, a moderated version of safe contraction
can be dened as follows:
DEFINITION 6 (Moderated safe contraction). Let hI; f; <i be a

prioritized imperative structure, where < is a well-founded strict partial order
on I. For each A 2 LPL , let M<;a I; 0 a card(I), be
MA
<;a def
b<a
MA
<;b [ min< rIn
b<a
MA
<;b gA:
ScardI A
Finally MA
M<;a and
.
< def
a1
So if X are the <-minimal elements in rIgA, then the moderated mechanism rst puts X in the set M< of elements to be removed, then the
<-minimal elements in rIn X gA, etc. Thus elements get removed in
each step until there is no A-kernel left in I minus the last version of M< , which
also means that the cardinality of I suces for the indices (I omit superscripts if
the meaning is clear). To see how this works, consider again the above example:
EXAMPLE (The road accident V: moderated safe contraction).

!C is the minimal element in
Removing !C is unavoidable:
the only other element !A in the -kernel in which !C is maximal ranks
before !C, so if !A was maximal in some -kernel, !C would not be in
So !C is in M<;1 f!Cg, equalling M< since no -kernel is left
, so both obligations to wait and post the
in Inf!Cg. Hence
letters remain.
If < is not a well-order, the moderated method still removes too much:
if it removes anything from a kernel, it removes all <-maximal members,
but by denition of a kernel, one is enough. Instead, one might again consider
the full prioritizations that preserve <, rather than <. The relation between
moderated safe contraction and Brewkas method is then the following:
THEOREM 6 (Moderated safe contractions and preferred

remainders). Let hI; f; <i be a prioritized imperative structure, where < is
a well-founded strict partial order on I. Then
prioritization of <}.
is some full
5. Uniquely prioritized imperatives

As demonstrated, the method proposed by Brewka and Nebel seems adequate
for a resolution of normative conicts by use of priority orderings of the

18
232
JORG HANSEN
underlying norms. Whether all conicts and dilemmas are thus avoidable is a
matter of dispute. W. D. Ross may be understood as claiming that conicts
are merely apparent and that by weighing all relevant facts and reasons, it can
be decided which of the conicting prima facie duties are really our duties (cf.
Ross 1930, Searle 1980, p. 242). G. H. von Wright stated that an axiological
order can provide a safeguard against any genuine predicament (cf. von
Wright 1968, p. 68, 80). And Hares description of critical moral thinking,
that lets principles override other, less important principles, suggests that this
process can overcome all moral conicts (Hare 1981, p. 43, 50). On the other
hand, Barcan Marcus (1980) and Horty (2003, p. 564) have argued that if it is
the presence of certain facts that determines the ordering, i.e. this is not an
arbitrary hacking through the Gordian knot, then situations might be
incomparable (if all such facts are missing), or be completely symmetrical (e.g.
identical obligations towards identical twins), so conicts remain possible.
Rather than take sides in this controversy, I will examine what is required if
the method of Brewka and Nebel is to resolve all conicts and dilemmas. For
all possible17 situations C; I + :C must then be a singleton: otherwise there are
C; D 2 I + :C such that Cf [ Df PL :C, and there is a dilemma. So I dene:
DEFINITION 7 (Uniquely prioritized imperatives). Let hI; f; <i

be a prioritized imperative structure, Then hI; f; <i is called uniquely prioritized iff for all C 2 LPL such that, 0PL :C, cardI + :C 1.
The question of a resolution of all conicts can now be rephrased to ask
what an imperative structure hI; f; <i must be like to be uniquely prioritized.
The most obvious way to avoid all conicts for any situation C is to let the
strict partial order < be total, i.e. for any two i, j 2 I, either i < j or j < i.
Then there is just one full prioritization that preserves <, namely < itself.
This result was also noted by Nebel (1992, Proposition 11), who constructs < as the strict part of a complete preorder ) then requiring < to
be total makes each equivalence class [i] a singleton. Gardenfors (1984,
p. 146), who also constructs < as the strict part of some complete preorder
, points out that it is enough if the only choice left is between equivalents,
i.e. here either i < j or j < i for any i, j 2 I with 0PL i f $ j f . But this still
requires too much. It suces that the demands of elements in each [i] are
chained, so for all j1 ; j2 2 i :PL j f1 ! j f2 or PL jf2 ! j f1 then it does not
matter in which order j1 and j2 appear in a full prioritization of <.18 And
one can be even more lax, as demonstrated by the following cases, which also
are of a sort in which no ambiguity ever arises:
Case 1: Let hI; f; <i be !p1 ; !p2 <!p1 ^ p2 , so the demands of the two
higher-ranking imperatives are doubled by a lower one. If both, !p1, !p2 are
in S 2 I + :C, then adding !p1 ^ p2 adds nothing. Otherwise
S#!p1 ^p2 [ f!p1 ^ p2 g f PL :C, so !:p1 ^ p2 cannot be in S .

233
19
Case 2: Let hI; f; <i be !p1 ; !p2 <!:p1 ^ p2 , so the demands of the two
higher-ranking imperatives run contrary to the lower ones. Either both
!p1,!p2 are in S 2 I + :C, so !:p1 ^ p2 cannot be consistently added, or
S#!:p1 ^p2 f [ fCg PL :p1 ^ p2 , so adding !:p1 ^ p2 adds nothing.
Case 3: Let hI; f; <i be !p1 ^ p2 <!p1 ^ :p2 <!p1 ; !p2 . If p1 ^ p2
is consistent with C, then I + :C is ff!p1 ^ p2 ; !p1 ; !p2 gg. Otherwise,
fCg PL :p1 ^ p2 . If p1 ^ :p2 is consistent with C, then I + :C is
ff!p1 ^ :p2 ; !p1 gg. Otherwise, also fCg PL ; :p1 ^ :p2 . Hence fCg PL :p1 .
Then any set in I + :C contains at most !p2, depending on whether C is consistent
with p2.
Unable to make out a necessary and sucient requirement for
cardI + :C 1, without reference to particular C, I can only rephrase its
denition as follows:
THEOREM 7 (Property of uniquely prioritized imperatives).

Any hI; f; <i, where < is the strict part of some complete preorder , is uniquely
prioritized iff for all consistent C 2 LPL ; C 2 I + :C; i 2 I and j1, j2 2 [i],
fi0 2 Cji0 <igf [ fCg PL j1f ! j2f or fi0 2 Cji0 <igf [ fCg PL j2f ! j 1f :
Let a semantics be called a uniquely prioritized imperative semantics i it

denes the truth of LDDL -sentences using (td-6), but only considers uniquely
prioritized imperative structures. It validates the additional axiom scheme:
RMon
PD=C ! OA=C ! OA=C ^ D:
Consider the system PD that was sound and (weakly) complete with respect
to prioritized imperative semantics. The system that results when (RMon) is
added is Hanssons (1969) DSDL3 as axiomatized by Spohn (1975):
THEOREM 8 (PD+RMon equals DSDL3). Let PD+(RMon) be

like PD, except that (RMon) is added as art axiom scheme. Then
PD+(RMon) = DSDL3, which is the smallest set that contains all LDDL instances into tautologies as well as all LPL -instances of the following schemes:
A0 OA=A
A1 If 0PL A then DSDL3 :O? =A
A2 OB ^ C=A $ OB=A ^ OC=A
A3 If PL A $ A0 and PL B $ B0 then DSDL3 OB=A $ OB0 =A0
A4 PB=A ! OC=A ^ B $ OB ! C=A
and is closed under modus ponens.

20
234
JORG HANSEN
Hanssons DSDL3, which is also the core of Aqivists (1986) system G,

usually characterizes a preference-based dyadic deontic semantics, i.e. an
interpretation of deontic formulas using a betterness relation between
valuations or possible worlds, and not using explicitly given norms as it is
here. However, this extreme interpretational change did not result in a
changed logical behavior of the deontic operators, i.e. DSDL3 can be
reconstructed quite naturally using an imperative semantics with an axiological order in the spirit of von Wright (1968):
THEOREM 9 (Soundness, completeness of DSDL3). DSDL3 is

sound and (only) weakly complete for uniquely prioritized imperative
semantics.
The construction used to prove the completeness theorem also exhibits
the following relation between priorities and contrary-to-duty norms: suppose there is a nite, or nitely based, set of deontic truths D LDDL , and
hI; f; <i is a uniquely prioritized imperative structure that makes true all of
D. The construction used to prove DSDL3-completeness provides a uniquely prioritized imperative structure hI0 ; f0 ; <0 i that also makes true all of
D, but the demands of these imperatives are now chained and so the priority relation can remain empty or let all imperatives rank equally (cf.
Theorem 7). E.g. if hI; f; <i is !p1 < !p2, then hI0 ; f0 ; <0 i is
!p1 ^ p2 ; !:p1 ^ p2 ! p1 ; !:p1 ^ p2 ^ :p1 ! p2 . These can be
viewed as contrary-to-duty norms, where the primary obligation is: to make
p1 ^ p2 true, if that is not possible, to make p1 true, and if that is also not
possible, to make p2 true. So instead of using ranks and priorities to avoid
conicts, contrary-to-duty formalizations can be employed to produce the
same eect. While there may be pragmatic reasons to attach higher priority
to the commands of the king than the wishes of his jester, from the
standpoint of logic, exception clauses like if the king did not say otherwise suce.
6. Conclusion
Describing how a resolution of normative conicts using priorities works is
surprisingly dicult. A method based on a proposal by Horty could not solve
complex cases where more than two norms conict or overriding norms are
no longer satisable or are themselves overridden. A method developed for
theory revision by Brewka and Nebel, which creates maximally non-conicting sets by starting with a maximal set of what is most important and

235
21
incrementally adding maximally to it, is able to resolve these diculties.

Alternative mechanisms discussed in normative theory, namely Alchourron
and Makinsons exposure criterion and variants by Prakken and Sartor, have
counterintuitive results in cases that Brewka and Nebels method adequately
solves. The same holds for Alchourrons safe contractions, but the intuition underlying his construction, that in a conict the least important norms
should be set aside, is captured in a moderated version that is equivalent to
Brewkas. I explain how a semantics that models explicitly given imperatives
can be used to dene deontic operators. When the preferred remainders
from Brewka and Nebels method are thus used for the denition of a disjunctive (skeptical) dyadic deontic operator, then such a semantics is characterized by the axiom system PD, which resembles Kraus, Lehmann and
Magidors system P with the dyadic D-axiom added. Whether all conicts
can be resolved using priorities is left to philosophical dispute, but conditions
are discussed that guarantee priority orderings, which do just that. For a
semantics that denes its dyadic deontic operators with respect to such
uniquely prioritized imperatives, Hanssons axiom system DSDL3 is
proved to be sound and complete. The proofs construction also exhibits the
fact that priorities are dispensable and that contrary-to-duty constructions
can take their place.
Most of the approaches discussed here include conditional entities,
which pose dierent problems like the following (rephrased from Rintanen
1994):
(1)
(2)
(3)
(4)
a says: if you drink anything, then dont drive.

b says: if you go to the party, then you do the driving.
c says: if you go the party, then have a drink with me.
You go the party.
Suppose that b does not mind if you have one drink with c, and c does not
care that you may be driving, and let the three imperatives be ranked in descending order. One may be tempted to reason as follows: consider rst the
imperative in line (1), but it has not yet been triggered as you have not yet
drunk anything, so it is set aside. Regarding (2), its condition is true, so you
must do the driving. Still, only (3) is triggered, so you should have a drink with
c. But satisfying (2) and (3) both triggers and violates the highest-ranking
imperative. Is it not more prudent to violate one of the lower-ranking imperatives instead of the higher-ranking one? For a solution, we need an adequate
denition of triggering (that can handle e.g. disjunctive inputs, like Makinson
and van der Torres basic output (Makinson and van der Torre 2000, 2001),
and to nd a maximizing strategy that is consistent with the above intuition. It
is clear that the present discussion has not provided the tools to properly
address such problems, so these must be left to further study.

22
236
JORG HANSEN
Acknowledgments
I thank David Makinson for valuable help with an earlier version, and Lou Goble,
John F. Horty and Henry Prakken for helpful discussions and clarifying correspondence. An earlier version of the paper was presented to the working group Law
and Logic of the XXII World Congress of Philosophy of Law and Social Philosophy
(IVR 2005) 2429: May 2005, Granada.
Notes
1
Though some discussed approaches cover conditional imperatives, or entities that can be
interpreted as such, these cause problems that are best considered separately.
2
PL is based on a language LPL , dened from a set of proposition letters Prop ={p1,
p2,...}, Boolean connectives :, ^, _, , M and brackets (,) as usual,
The truth of a LPL -sentence A is dened recursively using a valuation function v :
Prop ! f1; 0g (I write v A), starting with v p i v(p)=1 and continuing as usual. If
A 2 LPL is true for all valuations it is called a tautology. PL is the set of all tautologies,
and this set is used to dene provability, consistency and derivability (I write C PL A) as
usual. > is an arbitrary tautology, and ? is :>.
3
E.g. (td-1) most closely resembles denitions of Kanger (1957) and Alchourron and
Bulygin (1981). For authors belonging to this tradition cf. Hansen (2001), Section 1 and
Hansen (2004), fn. 1, in addition to which Ziemba (1971) must be mentioned.
4
Cf. Horty (1997). The disjunctive ought is more commonly referred to as skeptical
non-monotonic inference. Horty (2003) attributes the proposal to Brink (1994), yet the idea
to use such a denition for (dyadic) deontic logic already appears in Lewis (1981). For
alternatives in the deontic-logical treatment of normative conicts cf. Goble (2005).
5
Cf. Kraus et al. (1990), where, however, the proofs are done in a more general setting. Also
cf. my (Hansen 2005) for constructive proofs in the manner of Spohn (1975) as well as more
comparisons and truth denitions and axioms for an alternative credulous O-operator.
6
Hortys denition only employs circumstances to derive consequents from a set of conditional imperatives, but this has no eect on the solution of the examples.
7
Horty is preparing a rened version of Binding that solves all of the examples (private
correspondence).
8
Brewka (1991) and for Nebel cf. Rott (1993, fn. 9). For the rationale, let the ordered I
be hi0 ; . . . ; i0:125 ; i0:25 ; i0:5 ; i1 i, with i1 !p; i0:5 !:p; i0:25 !p; i0:125 !:p, etc., and i0=!q.
We cannot tell whether p or :p is obligatory, but this is not a case of conict either, all
imperatives !p being overridden by ones demanding :p, and vice versa.
9
I use notation from both, Brewka (1989, 1991), Brewka and Eiter (1999), and Nebel
(1991, 1992).
10
Also cf. Sartor (2005, p. 734): preference must be given to the argument such that its
weakest defeasible subreasons are better than the weakest defeasible subreasons in the other.
11
Prakken could argue that he only compares minimal conict pairs (subarguments), which f!p1 ; !p2 g; f!p3 ^ :p1 ; !p4 ^ :p2 g is not, while f!p1 g P f!p3 ^ :p1 g and
f!p2 g P f!p4 ^ :p2 g hold. But let I f!p1 ; !p2 ; !p3 ^ p4 ! :p1 ^ p2 ; !p4 ^ p3 ! :
p1 ^ p2 g, with !p1 < !p3 ^ p4 ! :p1 ^ p2 and !p2 < !p4 ^ p3 ! :p1 ^ p2 , so
with the primary targets one bonus secondary target is reachable. Still
f!p1 ; !p2 g 6P f!p3 ^ p4 ! :p1 ^ p2 ; !p4 ^ p3 ! :p1 ^ p2 g, and this is a minimal
conict pair. Yet intuitively, the argument for p1 p2 should win over any for :p1 ^ p2 .

237
23
12
Sartor (1991) simply rejects contradictions, yet the adjustment to any A is immediate.
The indicative version of the example also shows that Prakken cannot avoid counterintuitive results by replacing, in his denition of a rebuttal in (1997), his own relation P
by the relation AM of Alchouron and Makinson for the comparison of minimal conict
pairs: then all arguments for q are rebutted by arguments for :q as demonstrated, but
intuitively any consistent argument including i1 cannot be defeated and so the argument
for q should win.
14
For resolving legal arguments by summing up weights of reasons cf. Hage (1991, 1996).
15
Also cf. Iwin (1972, p. 486): When we are in a situation compelled to satisfy two
obligations requiring contradictory actions, then the most natural way out of this diculty consists in comparing the two obligations and not satisfying the less important
one.
16
I=:C is the notation in Alchourron and Makinson (1985), whereas the notation in
Alchourron (1986) would be I/C. Alchourrons own truth denition for deontic operators
employs a deontic logic as basic logic and conditional imperatives that are not treated
here.
17
If C is a contradiction, then by denition I + :C [.
18
I owe this insight to Leon van der Torre (private correspondence).
13
Appendix: Proofs
THEOREM 1 (Existence of full prioritizations). For every wellfounded strict partial order < on a set C there is a full prioritization , i.e. a
strict well-order that is order-preserving with respect to <.
Proof. Let < be a well-founded strict partial order on the set C. Let each
x 2 C be assigned an ordinal ax in the following way: ax=0 for the elements
in min< C, and for any other x 2 C; ax supfay jy<xg 1, where sup denotes
the supremum of a set of ordinals. Transnite induction available for wellfounded partial ordered sets tells us that ax is well-dened for any x 2 C. Let
the equivalence class x fy 2 Cjay ax g. Finally is an arbitrary strict
well-order on elements of the same equivalence class, and x y if ax is
smaller than ay. Clearly is a strict well-order on C. To prove x y if x<y,
for any x; y 2 C, suppose x<y. Then ax is in {az|z<y}, so ay=sup
{az|z<y}+1 is at least ax+1. Hence x y.
THEOREM 2 (Preferred remainders are remainders). Let hI; f; <i

be a prioritized imperative structure, where < is a well-founded strict partial
order on I. Then I + A IfA.
Proof. Suppose S 2 I + A. Since S f 0A there is some C 2 IfA such
that S C. Suppose C 6 S . Let i be some
element such that i 2 C, but
S
iS62 S . Then by the construction of S ; ji S#j f [ fif g PL A, and since
f
f
f
ji S#j S , also S [ fi g PL A. But then C PL A, which is excluded by C 2 IfA. So C S and S 2 IfA.

24
238
JORG HANSEN
THEOREM 3 (Nebels prioritized removals). Let hI; f; <i be a

prioritized imperative structure, where < is the asymmetric part of a complete,
well-founded preorder on I. Then I + A equals
fC I j Cf 0PL A and 8D I : if D N C then Df PL Ag:
Proof. Left-to-right: Suppose C 2 I + A. Suppose for some

D I : D N C. Then there is a priority class [i] such that for all
j < i; C \ j D \ j and C \ i D \ i. Let be the full prioritization
used to construct C S and i be theS-rst element of [i] with i 62 C, but
iS2 D. By construction of S ; ji S#j f [ fi f g PL A, and also
ji S#j fj 2 S j j ig. Since is order-preserving, for all j 2 S , if
j i, then j< i or j 2 [i]. Then j 2 D, by denition of N or by choice of i.
Hence Df PL A.
Right-to-left: Suppose C I; Cf 0PL A and 8D I. If D N C,
then Df PL A. Let be a well ordering on I respecting < which puts i j for
any i 2 C and j 2 inC, i.e. an ordering that positions the elements of C before
their -equals, and D S . We prove 8i 2 I : D \ i C \ i by induction
on <:
For the induction basis, let i be some -least element. By denition of
the elements of C \ i are positioned -before any other elements of D,
and by assumption Cf \ i0PL A, so C \ i D due to the construction of
S D. Suppose there is some j 2 D \ i such that j 62 C \ i. Then
C \ i D \ i, and since trivially 8j < i : C \ j D \ j as i was -least and
so no such j exists, D N C holds. Then it must be that Df PL A, but by
denition of S D this is excluded. So D \ i C \ i and, hence,
D \ i C \ i.
For the induction step, let i be some arbitrary element of I. By denition of
, the elements of C \ i are positioned -before any other elements of
D \ i. The induction hypothesis guarantees that for all j 2 I with
j<i : D \ j C \ j, so fj 2 D j j ig fj 2 C j j ig. As fjS
2 C j j igf
f
[C \ i 0PL A (otherwise C PL A, for any j 2 C \ i : kj S#k f
[fj f g0PL A, so j 2 D and hence C \ i D. Suppose there is some j 2 D \ i
with j 62 C \ i. Then C \ i D \ i, and since also 8j<i : C \ i D \ i
by the induction hypothesis, D N C holds. Then it must then be that
Df PL A, but by denition of S D this is excluded. So D \ i C \ i and
hence D \ i C \ i.
Since 8i 2 I : D \ i C \ i we have C D S and so C 2 I + A.
THEOREM 4 (Preferred remainders, after Ryan and Sakama &

Inoue). Let hI; f; <i be a prioritized imperative structure, < a well-founded
strict partial order on I and I + A and pIfA be as defined. Then
I + A pIfA.

239
25
Proof. Left-to-right: Suppose S 2 I + A. Assume for r.a.a. that there is a

C 6 S in IfA such that for all i 2 S nC there is a j 2 CnS with j i. Let i
be the -least element in S nC, which is not empty, because otherwise
S C, but this is excluded by maximality
S of S , so existence of i is guaranteed by well-foundedness of . Then ki S<#k C: otherwise there is
Since
some k 2 S with k i but k 62 C and so i would not be the -least.
S
j 62 S it must be that S<#j f [ fj f g PL A. But j i, so S<#j ki S<#k ,
and chaining results we get Cf PL A, which contradicts C 2 IfA.
Right-to-left: Suppose C 2 pIfA. Let be a full prioritization for which
the statement in the definition of pIfA is true. By definition, S is in I + A
6 C. Then there is some
and so also in IfA. Assume for r.a.a. that S S
i 2 CnS such that for all j 2 S nC; i j. Suppose ji S#j 6 fj 2 C j j ig.
Then there is a j 2 S n C : j i and so i 6 j by the antisymmetry of , and
ihj
since preserves <, but we assumed otherwise. So
S
S
fj
#j
ji
S 2 C j j ig. Since i 2 CnS it must be that i 62 S , so by
definition ji S#j f [ fi f g PL A. So Cf PL A, which contradicts
C 2 IfA.
THEOREM 5 (Soundness, completeness of PD). PD is sound and

(only) weakly complete with respect to prioritized imperative semantics.
Proof. (Sketch) We must prove that PD is (a) sound with respect to
prioritized imperative semantics, (b) (weakly) complete with respect to prioritized imperative semantics and (c) only weakly complete, i.e. not compact.
(a) Soundness. The validity of (CExt), (ExtC), (DM) and (DC) is
immediate. (DN) is unrestrictedly valid since 8C 2 I + :C : Cf PL > is
trivial. Theorem 1 guarantees the existence of a generating , so if C is
not a contradiction (then I + :C [ by definition) at least [ is in I + :C,
so I + :C 6 [ and (DD-R) must hold. Leaving details to the reader, for
(Cond), suppose OA=C ^ D, so 8C 2 I + :C ^ D : Cf [ fC ^ Dg PL A
and assume for r.a.a. that O(D fi A/C) is false, which yields
9D 2 I + :C : D f [ fCg0PL D ! A and also includes D f 0PL :C ^ D.
Likewise,
for
(CCMon)
suppose
OA ^ D=C,
which
yields
8D 2 I + :C : D f [ fCg PL A ^ D and also includes D f 0PL :C ^ D. D is
obtained from some full prioritization via Definition 2, i.e. D S:C
.
Appealing to the -first element on which they might differ, prove that
:C^D
.
S:C
S
(b) Weak completeness. We must prove that for any A 2 LDDL such
that :A 62 PD there is a prioritized imperative structure hI; f; <i that
models A. I proved PD to be weakly complete for basic imperative
semantics and (td-4) in Hansen (2005), so take the basic imperative structure hI; f i that models A and define < [. Then I + :C If:C for any
C. Hence hI; f; <i also models A using (td-6).

26
240
JORG HANSEN
(c) Non-compactness. Leaving details to the reader, I gave a counterexample to compactness of basic imperative semantics in Hansen (2005, Theorem 3), i.e. a non-satisable set C LDDL of which all nite subsets are
satisable, which applies here as well. The tricky part is to show that the sets
Ip1 [ Ip1 $:p2 and I:p1 [ Ip1 $p2 are in I +?, which works using the complete
preorder induced by ordinal labels from the proof of Theorem 1 (for each
union choose a full prioritization that puts its elements -rst in any
equivalence class and prove that S in I +? constructed from equals the
respective union).
THEOREM 6 (Moderated safe contractions and preferred

remainders). Let hI; f; <i be a prioritized imperative structure, where < is a
well-founded strict partial order on I. Then
prioritization of <}.
is some full
Proof. I rst give a construction next(i), containing the compared to i

next-larger element in M , and a helpful lemma:
LEMMA 1. Let be a full prioritization of <; A 2 LPL and M be

accordingly defined (cf. Definition 8). For any i 2 I, let
nexti df min fj 2 M j i j or i jg
contain i if i 2 M or else the element in M that comes -next to i. Let

Mi
min fM;a j nexti M;a g if nexti 6 [,

otherwise
M
be the first set in the construction of M that includes next(i) if that is nonempty and otherwise the whole of M . Then
Mi n nexti M \ fj 2 I j j ig:
Proof. For next(i), note that due to well-orderliness, the -minimum of a

non-empty set is a singleton, and so is next(i), unless there is no j in M with
i=j or i j, in which case it is [. For Mi, note that since M;a increases with
each step a, 0 a < card(I), the inclusion-minimum is well dened. To prove
the lemma, if neither i 2 M nor a -next member in M exists and so
Mi M , then the equivalence is trivial. Otherwise Mi M;a for some a,
and we must prove that (i) if j 2 I is in M;a n nexti then j i, and (ii) that
if j 2 M is not in M;a n nexti then j
= i.
ad i): Let j 2 M;a . We assumed next(i) to be non-empty, so let next(i)={i0 }.
By denition, i 0 is the element of M;a that was added at step a and so if i0 =j,
then j is not in M;a A n nexti. Otherwise, by the construction of M;a ; j is in
some M;b with b < a. Let b be the smallest of these. To be in the increase of
241
27
M;a ; i0 must be -maximal in some A-kernel

S X in I n c<aSM;c . Since the
process Sis incremental, we have I n c<a M;c I n c<b M;c . So
X I n c<b M;c . Hence, if i0 j,S then j is not -minimal among the maximal members of A-kernels in I n c<b M;c , but this contradicts that j is the
element by which M;b was increased at step b. So i0
= j, we supposed i0 j, and
0
so by -connectedness j i . Also, if i=j or i j, then i0 =j or i0 j by the
definition of next(i) and j 2 M . By the previous result and -transitivity, we
obtain j j, contradicting -irreflexivity. So i
= j; i
6 j and hence by connectedness, j i.
ad ii): Assume j 2 M is not in M;a n nexti. Again let next(i)={i0 }. So
0
i is the element of M;a , which is added at step a. If j=i, then i=j or i j
and in both cases j
= i. Otherwise j must be in some M;b with a<b. Let b be
the smallest of these. ToSbe in the increase of M;b ; j must be -maximal in
some
S A-kernel X in ISn c<b M;c . Since theSprocess is incremental, we have
I n c<b M;c I n c<a M;c . So X I n c<a M;c . Suppose j i0 , then
i0 is
S not -minimal among the -maximal members of A-kernels in
I n c<a M;c . But this contradicts that i0 is the element by which M;a was
increased at step a. So j
= i0 , and we supposed j i0 . Thus by -connect0
edness, i j and hence i j by -transitivity and definition of next(i).
For the theorem, it now suffices to prove that for each full prioritization
; A 2 LPL and accordingly constructed sets S (Definition 2) and M
(Definition 8), I n M S ; which is done by induction over :
Induction basis: Let i0 be the -least element in I. Suppose i0 is in
I n M : I n M contains no A-kernels, so fi0 g0PL A and so i0 is in S . Suppose i0 is not in I n M . Then i0 is -maximal in some A-kernel of I. So for all
other j in this A-kernel, j i, but also i j because i0 is -least. So by antisymmetry this A-kernel equals {i0}, so fif0 g PL A and by definition
i0 62 S . Hence i0 2 I n M iff i0 2 S .
Induction step: Right-to-left: Suppose i 62 I n M , so iS2 M . Let a be
the step that has i in its increase, i.e. M;a b<a M;b [ fig.
Since i 2 M
S ; nexti i, so M;a is Mi as defined above. So
by
Lemma
1.
Hence
Mi nSfig b<a M;b M \ fj 2 I j j < ig
I n b<a M;b \ fj 2 I j j ig I n M \ fj 2 I j j ig \ fj 2 I j j ig
hypothesis equals
I n M \ fj 2 I j j ig which by the induction
S
S \ fj 2 I j j ig, and by definition this equals ji S#j . To be
Sin the increase
of M;aS
; i must be -maximal in some A-kernel X in I n b<a M;b , so
\ fj 2 I j j ig [ fig. So by definition of an A-kernel and
X I n b<a M;b S
the above equation, ji S#j f [ fif g PL A. Hence S
i 62 S#i and i 62 S .
f
f
.
So
by
definition
Left-to-right:
Suppose
i
2
6
S

ji S#j [
S
Sfi g PL A
f
and ji S#j 0PL A. By the induction hypothesis
ji S#j
S \ fj 2 I j j ig I n M \ fj 2 I j j ig. By the above lemma, M \
fj 2 I j j ig Mi n nexti and so I n M \ fj 2 I j j ig I n M \
fj 2 I j j ig \ fj 2 I j j ig I n Mi n nexti \ fj 2 I j j ig, which,

28
242
JORG HANSEN
since nexti I, equals I n Mi [ nexti \ fj 2 I j j ig, which equals

I n Mi \ fj 2 I j j ig since nexti \ fj 2 I j j ig [ by definition of next(i). So we obtain the results that I n Mi \ fj 2 I j j ig f [ fif g PL A and
I n Mi \ fj 2 I j j ig f 0PL A. First observe nexti 6 [ otherwise
Mi M by definition, so I n M f [fif g PL A, but by definition
I n M f 0PL A, so i 62 I n M and, hence, i is in M , but then
nexti fig 6 [. So Mi M;a for some a. Next assume i 62 M . Then
i 2 I n Mi . By the first result obtained above, i is -maximal in some Akernel in I n Mi . Suppose i is not -minimal among the -maximal members of A-kernels in I n Mi : then some A-kernel must lie completely
in I n Mi S\ fj 2 I j j ig, but the other result excluded this. So
min rI n b<a M;b gA fig, so i is in the increase at step a, i.e. i 2 Mi,
so i 2 M , which contradicts its assumed negation and so is true. So
i 62 I n M .
THEOREM 7 (Property of uniquely prioritized imperatives).

Any hI; f; <i, where < is the strict part of some complete preorder , is uniquely prioritized iff for all consistent C 2 LPL ; C 2 I + :C; i 2 I and
j1 ; j2 2 i;
fi0 2 C j i0 <igf [ fCg PL j f1 ! j f2 or fi0 2 Cj i0 <igf [ fCg PL j f2 ! j f1 :
Proof. For the left-to-right direction (in contraposition), suppose there are
j1, j2 2 [i], a consistent D 2 LPL and C 2 I + :D with neither fi0 2 C j i0 <igf
[fDg PL j f1 ! j f2 nor fi0 2 C j i0 <igf [ fDg PL j f2 ! j f1 . Consider C
D ^ :j f1 ^ j f2 . If fi0 2 C j i0 <igf PL :C, then fi0 2 C j i0 <igf PL :D_
j f1 ^ j f2 , which includes fi0 2 C j i0 <igf [ fDg PL j f1 ! j f2 , contrary to what
we supposed. Let 1 be a full prioritization of < that for all i <i is like C,
i.e. it contains fi0 2 C j i0 <ig, and then puts j1 1 -first among the members of
[i]. Likewise let 2 be a full prioritization of < that contains fi0 2 C j i0 <ig
and puts j2 2 -first among the members of [i]. fi0 2 C j i0 <igfSPL :C is excluded, so regarding the constructions S1 ; S2 2 I + :C; k1 j1 S1 #k
S
S
fi0 2 C j i0 <ig k2 j2 S2 #k . If k1 j1 S1 #k f [ fj f1 g PL :C, then fi0 2 C j
i0 <igf [ fj f1 g PL :D _ j f1 ^ j f2
or
equivalently
fi0 2 C j i0 < igf [
S
fDg PL j f1 ! j f2 , but that was excluded. So k1 j1 S1 #k f [ fj f1 g0PL :C and
so j1 2 S1 , and likewise j2 2 S2 is proved. But then S1 6 S2 otherwise
S1 f PL j f1 ^ j f2 and so S1 f PL :C, which is excluded. So card
I + :C 6 1 for some non-contradictory C 2 LPL .
For the right-to-left direction, assume for r.a.a. that cardI + :C 6 1 for
some consistent C 2 LPL , so there are two full prioritizations 1 and 2

243
29
such that S1 6 S2 for the according S1 ; S2 2 I + :C. Then there is a i 2 I
such that S1 \ i 6 S2 \ i: let [i] be the <-first such equivalence class. So
there is a j1 2 [i] with either j1 2 S1 and j1 62 S2 or vice versa. This being
S
equal, assume the former, so k2 j1 S2 #k f [ fj f1 g PL :C. So there is a
S
(j f1 ! :C)-kernel X in k2 j1 S2#k , and X \ i 6 [, else X S1 by choice
of [i] and since j1 2 S1 we obtain S1 f PL :C, which is excluded. We get
fj 2 S2 j j < igf [ fCg PL j f2 ! j f1 or fj 2 S2 j j < igf [ fCg PL j f1 ! j f2
from the right side of the iff-clause for any j2 2 X S
\ i. Suppose the first case
holds for a j2 2 X \ i. Since j2 2 X k2 j1 S2#k we obtain
S
k2 j1 S2#k f PL C ! j f1 , and since the ( j f1 ! :C)-kernel X is in
S
S
f
k2 j1 S2#k , this yields k2 j1 S2#k PL :C, which is excluded. So for all
f
f
j2 2 X \ i: fj 2 S2 j j<igf [ fCg
S PL j 1 ! j 2 . For all other k 2 X, i.e.
k 62 i, we have k 2 i by X k2 j1 S2#k and so k 2 S1 since j1 2 [i] and
due to the choice of i. So since j1 is in S , and also fj 2 S2 j j<ig S by the
choice of i, for any k 2 X : S f [ fCg PL kf . Hence, S f PL :C, which
the construction of the antecedent excludes. This completes the r.a.a.
THEOREM 8 (PD + RMon equals DSDL3). Let PD+(RMon) be

like PD, except that (RMon) is added as an axiom scheme. Then
PD+(RMon) = DSDL3.
Proof. Easy and left as exercise to the reader.
THEOREM 9 (Soundness, completeness of DSDL3). DSDL3 is

sound and (only) weakly complete for uniquely prioritized imperative semantics.
Proof. We must prove that DSDL3 is (a) sound with respect to prioritized
imperative semantics, (b) (weakly) complete with respect to prioritized
imperative semantics, and (c) only weakly complete, i.e. not compact.
(a) Soundness. Only the validity of (RMon) needs to be proved, as Theorem 5 proved validity of the PD-axioms with respect to all, not necessarily
uniquely, prioritized imperative structures. The proof is like the one for
(CCMon) in Theorem 5, i.e. we assume O(A/C ) and obtain that for all
D 2 I + :C : D f [ fCg PL A and obtain the additionally included fact that
for all such D; D f [ fCg0PL :D by assuming P(D/C) and from card
I + :C 1 in case C is consistent (otherwise DN and CExt make the
conclusion OA=C ^ D trivially true).
(b) Weak completeness. The proof is a condensed version of its more
widely applicable variants in Spohn (1975) and my (2001, 2005), to which I
refer for further features of the construction. We must prove that for
any DSDL3-consistent. A 2 LDDL , i.e. :A 62 DSDL3, there is a uniquely

30
244
JORG HANSEN
prioritized imperative structure hI; f; <i that models A. We build a disjunctive normal form of A and obtain a disjunction of conjunctions, where each
conjunct is O(B/D) or :OB=D. One disjunct must then be DSDL3-consistent. Let d be that disjunct. Let the d-restricted language LdPL be the PLsentences that contain only proposition letters occurring in d. Let rLdPL be
n
22 mutually non-equivalent representatives of LdPL , n being the number of
proposition letters in d. LPL -sentences now mean their representatives in
rLdPL . Construct a set D & LDDL such that:
(a) Any conjunct of d is in D.
(b) For all B; D 2 rLdPL : either P(B/D) or O:B=D 2 D.
(c) D is DSDL3-consistent.
It then suces to nd a uniquely prioritized imperative structure that
makes true all of D. We identify what Hansson (1969) called the deontic
V basis
~ by letting OC fA 2
in an extension kCk (Spohn 1975 writes C)
rLdPL j OA=C 2 Dg be the sum of everything demanded in the situation
C. From this denition and (b), (c) it follows immediately that OOC =C 2 D.
Furthermore, observe
(O1)
OA=C 2 D iff PL OC ! A:
System of spheres: Let S hC1 ; . . . ; Cn i be dened recursively by letting

(i) C1 >,
(ii) if Ci 2 S and Ci 6? then Ci1 Ci ^ :OCi (otherwise i=n).
Observe that for all Ci, Cj, 1 i < j n:
(O2)
PL Cj ! Ci
(O3) Cj ^ OCi ?
Proof. Immediate (due to A0 and O1, PL OCj ! Cj so also OCj ^ OCi ?).
(O4) Ci 2 rLdPL ; 1
(O5) The sequence is finite, i.e. n 6 1

(O6) OC1 _ . . . _ OCn >
Proofs. (O4) is immediate from the denitions. Regarding (O5), rLdPL is

nite, so if S is innite, then Ci=Cj for some 1 i < j n. Then
Ci ^ OCi ? due to (O3), but since PL OCi ! Ci due to (A0) and (O1), then
OCi ? and so due to (A1) Ci ?. But then the sequence ends with i. For
(O6), if OC1 _ . . . _ OCn 6 > then > ^ :O> ^ . . . ^ :OCn1 ^ :OCn 6?. The
left side equals Cn ^ OCn , so Cn is not last in the sequence, contrary to what
was assumed.
Smallest A-permitting sphere: Let CA be the rst Ci with
0PL A ^ OCi !?, i.e. for all j < i : A ^ OCj ?. Then furthermore:

245
31
(O7) For any rLdPL -sentence A 6? there is some CA 2 S:

(O8) PL A ! CA
(O9) IfA 6?; then PA=CA 2 D:
(O10)
PL OA $ OCA ^ A
Proofs. (O7) is immediate from (O6). Regarding (O8) let CA=Ci, so for all
j < i : OCj ^ A ?, so PL A ! > ^ :O> ^ . . . ^ :OCi1 and thus PL A
! Ci . For (O9), if CA exists, then by the construction of D either
PA=CA 2 D or O:A=CA 2 D, but in the second case OCA ^ A ? due to
(O1), which is excluded by the definition of CA. Regarding (O10): due to (O8)
and derivability of (Cond) we obtain OA ! OA =CA 2 D from
OOA =A 2 D, so PL OCA ^ A ! OA , which is the right-to-left direction. If
A ?, then OA ? since PL OA ! A due to (A0) and (O1) and then the leftto-right direction is trivial. Otherwise PA=CA 2 D due to (O9) and
OOCA =A 2 D is obtained from OOCA =CA , (O8) and derivability of
(RMon). Thus PL OA ! OCA and due to (A0) and (O1) also PL OA ! A.
Both include the left-to-right version.
Canonical construction of hI; f; <i: Let I fC ! OC j C 2 Sg; f be identity and < be the strict part of =I I (or [, al gusto). Verification: due to
the constructions of S and I it is immediate that for all i; j 2 I : PL i f ! j f or
PL j f ! i f , so the demands of the imperatives are chained and I is uniquely
prioritized (Theorem 7). All imperatives are equally ranked, so I + A IfA.
Coincidence: We must prove that for all B; D 2 rLdPL , hI; f; <i models
O(B/D) iff OB=D 2 D. Right-to-left: Assume OB=D 2 D. If D =^, then
by definition I + :D If:D [, so (td-6) makes O(B/D) trivially true.
Otherwise, by (O7) there is some CD 2 S. By its definition, D ^ OCD is consistent, so D ^ CD ! OCD is also consistent and so CD ! OCD must be in
(the) C 2 If:D. Since PL D ! CD by (O8) and PL OCD ^ D ! OD by
(O10) we have PL CD ! OCD ^ D ! OD , so C [ fDg PL OD and due to
(O1), C [ fDg PL B and so (td-6) makes O(B/D) true. Left-to-right: Assume
that for C 2 I + :D If:D we have C [ fDg PL B. If D=^, then
OB=D 2 D follows trivially from (A0), (A2) and (A3). Otherwise, by (O7)
there is a CD 2 S: D ^ OCD is by definition consistent, so D ^ CD ! OCD is
consistent and CD ! OCD must be in (the) set C. CD=Ci for some i in the
construction of S. For all j < i; PL D ! Cj due to (O8) and (O2), and
PL D ! :OC j by definition of CD, so Cj ! OC j 62 C. For all
j > i; PL OC D ! :Cj by (O3), so with (O8) we obtain fCD ! OCD g[
fDg PL Cj ! OC j for any Cj ! OC j 2 C. So if C [ fDg PL B, then
fCD ! OCD g [ fDg PL B. Hence PL OCD ^ D ! B by (O8), PL OD ! B
by (O10), so OB=D 2 D by (O1).
246
32
JORG HANSEN
(c) Non-Compactness. Compactness is disproved by any infinite set D of

LDDL -sentences, of which any finite subset Df is satisfiable, but not D.
Consider
D fOp1 =>g [ fP:p1 =:p1 ^ . . . ^ pn j n > 1g
[ fPA ^ :pn =:p1 ^ . . . ^ pn j n > 1 and fAg0PL pn g
Let Df D be a finite subset of D. Let n be the highest index such that pn

occurs in some formula in Df . Then hI; f; <i with I f!p1 ^ . . . ^ pn g,
< being [, satisfies Df : I +? fIg ff!p1 ^ . . . ^ pn gg. So for all sets C in
I + ? : Cf PL p1 , so Op1 => is true. The only C 2 I + p1 ^ . . . ^ pn , n > 1,
is the empty set and that plus :p1 ^ . . . ^ pn is consistent with :p1
and A ^ :pn if fAg0PL pn . So P:p1 =:p1 ^ . . . ^ pn and PA ^ :pn =:
p1 ^ . . . ^ pn are true for any n > 1. Hence D is finitely satisfiable.
Suppose D is satisfiable, i.e. there is a uniquely prioritized imperative structure hI; f; <i that makes all D true. The sets I +? and
I + p1 ^ . . . ^ pn must be singletons (Definition 9). To make Op1 =>
f
true, there must be some full prioritization such that S?
PL p1 ,
?
where S 2 I + ? is as described in Definition 2. Due to PL-compactness
f
there must be some i 2 I such that S?
#i PL p1 . Let i be the -first such
element. Its existence is guaranteed by well-foundedness of since 0PL p1 . So
S
f
f
?
ji S?
#j 0PL p1 and i 2 S . Let n be the smallest index such that fi g0PL pn .
i f must be consistent since i 2 S?
finite length of if alone guarantees the
, so the S
f
existence of such an n. We have ji S?
#j 0PL p1 ^ . . . ^ pn and so
S
p1 ^...^pn
p ^...^pn
?
, where S 1
2 I + p1 ^ . . . ^ pn is as described
ji S#j S
p ^...^pn
1
in Definition 2. S?
#i 6 S
, for otherwise P:p1 =:p1 ^ . . . ^ pn
f
could not be true, so it must be that S?
#i PL p1 ^ . . . ^ pn . But then
p ^...^p
p ^...^p
n
n
S 1
f PL i f ! p1 ^ . . . ^ pn , which includes S 1
f PL i f ! pn .
f
Hence Pi ^ :pn =:p1 ^ . . . ^ pn cannot be true. But PA ^ :pn =
:p1 ^ . . . ^ pn must be true for all A such that fAg0PL pn and if meets the
condition. So D is not satisfiable.
References
Alchourron, C. E. (1986). Conditionality and the Representation of Legal Norms, in Martino,
A. A. and Socci Natali, F. (eds.), Automated Analysis of Legal Texts: Edited Versions of
selected papers from the Second International Conference on Logic, Informatics, Law,
Florence, Italy, September 1985. North Holland, Amsterdam, 173186.
Alchourron, C. E. and Bulygin, E. The Expressive Conception of Norms, in Hilpinen, R.
(1981). New Studies in Deontic Logic. Reidel: Dordrecht, 95124.
Alchourron, C. E. and Makinson, D. Hierarchies of Regulations and Their Logic, in Hilpinen,
R. (1981). New Studies in Deontic Logic. Reidel: Dordrecht 95124, 125148.

247
33
Alchourron, C. E. and Makinson, D. (1985). On the Logic of Theory Change: Safe Contraction. Studia Logica 44: 405422.
Aqvist, L. (1986). Some Results on Dyadic Deontic Logic and the Logic of Preference.
Synthese 66: 95110.
Brewka, G. (1989). Preferred Subtheories: An Extended Logical Framework for Default
Reasoninig, in Sridharan, N. S. (ed.), Proceedings of the Eleventh International Joint
Conference on Articial Intelligence IJCA189, Detroit, Michigan, USA, August 2025,
1989, San Mateo, Calif.: Kaufmann, 10431048.
Brewka, G. (1991). Belief Revision in a Framework for Default Reasoning, in Fuhrmann, A.
and Morreau, M. (eds.), The Logic of Theory Change, Workshop, Konstanz, Germany,
October 1315, 1989. Springer, Berlin, 206222.
Brewka, G. and Eiter, T. (1999). Preferred Answer Sets for Extended Logic Programs. Articial Intelligence 109: 297356.
Brink, D. O. (1994). Moral Conict and Its Structure. Philosophical Review 103: 215247.
Fehige, C. (1994). The Limit Assumption in Deontic (and Prohairetic) Logic, in Meggle, G.
and Wessels, U. (eds.), Analyomen 1: Proceedings of the 1st Conference Perspectives in
Analytical Philosophy, Saarbrucken 1991. de Gruyter, Berlin, 4256.
Gardenfors, P. (1984). Epistemic Importance and Minimal Changes of Belief. Australasian
Journal of Philosophy 62: 136157.
Goble, L. (2005). A Logic for Deontic Dilemmas, Journal of Applied Logic: 961983.
Hage, J. C. (1991). Monological Reason Based Reasoning, in Breuker, J. A., De Mulder, R. V.,
and Hage, J. C., Legal Knowledge Based Systems: Model-based Legal Reasoning (Proceedings of the Fourth International Conference on Legal Knowledge Based Systems JURIX
1991, Lelystad, December 1991), 7791.
Hage, J. C. (1996). A Theory of Legal Reasoning and a Logic to Match. Articial Intelligence
and Law 4: 199273.
Hansen, J. (2001). Sets, Sentences, and Some Logics about Imperatives. Fundamenta Informaticae 48: 205226.
Hansen, J. (2004) Problems and Results for Logics about Imperatives, Journal of Applied
logic, 3661.
Hansen, J. (2005) Conicting Imperatives and Dyadic Deontic Logic, Journal of Applied
Logic, 484511.
Hansson, B. (1969). An Analysis of Some Deontic Logics. Nous 3: 373398, reprinted in
Hilpinen, R. (1971). Deontic Logic: Introductory and Systematic Readings. Reidel:
Dordrecht, 121147.
Hare, R. M. (1981). Moral Thinking. Clarendon Press: Oxford.
Horty, J. F. (1997). Nonmonotonic Foundations for Deontic Logic, in Nute, D. (eds.),
Defeasible Deontic Logic. Kluwer, Dordrecht, 1744.
Horty, J. F. (2003). Reasoning with Moral Conicts. Nous 37: 557605.
Iwin, A. A. (1972). Grundprobleme der deontischen Logik: in Wessel, H. (eds.), Quantoren
ModalitatenParadoxien. VEB Deutscher Verlag der Wissenschaften, Berlin, 402522.
Kanger, S. (1957). New Foundations for Ethical Theory: Part 1, duplic., 42 p., reprinted in
Hilpinen, R. (1971). Deontic Logic: Introductory and Systematic Readings. Reidel:
Dordrecht, 3658.
Kraus, S., Lehmann, D. and Magidor, M. (1990). Nonmonotonic Reasoning, Preferential
Models and Cumulative Logics. Articial Intelligence 44: 167207.
Lewis, D. (1981). Ordering Semantics and Premise Semantics for Counterfactuals. Journal of
Philosophical Logic 10: 217234.
Makinson, D. and van der Torre, L. (2000). Input/Output Logics. Journal of Philosophical
Logic 29: 383408.

34
248
JORG HANSEN
Makinson, D. and van der Torre, L. (2001). Constraints for Input/Output Logics. Journal of
Philosophical Logic 30: 155185.
Marcus, R. B. (1980). Moral Dilemmas and Consistency. Journal of Philosophy 77: 121136.
McNamara, P. (1995). The Connement Problem: How to Terminate Your Mom With Her
Trust. Analysis 55: 310313.
Nebel, B. (1991). Belief Revision and Default Reasoning: Syntax-Based Approaches, in Allen,
J. A., Fikes, R. and Sandewall, E. (eds.), Principles of Knowledge Representation and
Reasoning: Proceedings of the Second International Conference, KR 91, Cambridge, MA,
April 1991. Morgan Kaufmann, San Mateo, 417428.
Nebel, B. (1992). Syntax-Based Approaches to Belief Revision, in Gardenfors, P. (eds.), Belief
Revision. University Press, Cambridge, 5288.
Prakken, H. (1997). Logical Tools for Modelling Legal Argument. Kluwer: Dordrecht.
Prakken, H. and Sartor, G. (1997). Argument-based Logic Programming with Defeasible
Priorities. Journal of Applied Non-classical Logics 7: 2575.
Rescher, N. (1964). Hypothetical Reasoning. North-Holland: Amsterdam.
Rintanen, J. (1994). Prioritized Autoepistemic Logic, in MacNish, C., Pearce, D. and Pereira,
L. M. (eds.), Logics in Articial Intelligence, European Workshop, JELIA 94, York, September 1994, Proceedings. Springer, Berlin, 232246.
Ross, W. D. (1930). The Right and the Good. Clarendon Press: Oxford.
Rott, H. (1993). Belief Contraction in the Context of the General Theory of Rational Choice.
Journal of Symbolic Logic 58: 14261450.
Ryan, M. (1992). Representing Defaults as Sentences with Reduced Priority, in Nebel, B.,
Rich, C. and Swartout, W. (eds.), Principles of Knowledge Representation and Reasoning:
Proceedings of the Third International Conference, KR 92, Cambridge, MA, October 1992.
Morgan Kaufmann, San Mateo, 649660.
Sakama, C. and Inoue, K. (1996). Representing Priorities in Logic Programs, in Maher, M.
(eds.), Joint International Conference and Syposium on Logic Programming JICSLP 1996,
Bonn, September 1996. MIT Press, Cambridge, 8296.
Sartor, G. (1991). Inconsistency and Legal Reasoning, in Breuker, J. A., De Mulder, R. V.,
and Hage, J. C., Legal Knowledge Based Systems: Model-based Legal Reasoning
(Proceedings of the Fourth International Conference on Legal Knowledge Based Systems
JURIX 1991, Lelystad, December 1991), 92112.
Sartor, G. (2005) Legal Reasoning. A Cognitive Approach to the Law, vol. 5 of A Treatise of
Legal Philosophy and General Jurisprudence, Springer: Dordrecht.
Searle, J. (1980). Prima-facie Obligations, in Straaten, Z. v. (ed.), Philosophical Subjects:
Essays presented to P. F. Strawson. Clarendon Press, Oxford, 238259.
Spohn, W. (1975). An Analysis of Hanssons Dyadic Deontic Logic. Journal of Philosophical
Logic 4: 237252.
von Wright, G. H. (1968). An Essay in Deontic Logic and the General Theory of Action. North
Holland: Amsterdam.
Ziemba, Z. (1971). Deontic Syllogistics. Studia Logica 28: 139159.
249
Auton Agent Multi-Agent Syst

DOI 10.1007/s10458-007-9016-7
Prioritized conditional imperatives: problems

and a new proposal
Jrg Hansen
Springer Science+Business Media, LLC 2007
Abstract The sentences of deontic logic may be understood as describing what an agent
ought to do when faced with a given set of norms. If these norms come into conflict, the best
the agent can be expected to do is to follow a maximal subset of the norms. Intuitively, a
priority ordering of the norms can be helpful in determining the relevant sets and resolve conflicts, but a formal resolution mechanism has been difficult to provide. In particular, reasoning
about prioritized conditional imperatives is overshadowed by problems such as the order
puzzle that are not satisfactorily resolved by existing approaches. The paper provides a new
proposal as to how these problems may be overcome.
Keywords
Deontic logic Default logic Priorities Logic of imperatives
1 Drinking and driving

Imagine you have been invited to a party. Before the event, you become the recipient of
various imperative sentences:
(3) Some acquaintance says: if you go to the party, then have a drink with me.
Suppose that as a rule you do what your mother tells youafter all, she is the most important
person in your life. Also, the last time you went to a party your best friend did the driving, so
it really is your turn now. You can enjoy yourself without a drink, though it would be nice to
have a drink with your acquaintanceyour best friend would not mind if you had one drink,
and your acquaintance does not care that you may be drivingbut your mother would not
approve of such a behavior. Making up your mind,
J. Hansen (B)
Institut fr Philosophie, University of Leipzig, Beethovenstrae 15, 04107 Leipzig, Germany
e-mail: jhansen@uni-leipzig.de
123
250
(4) You go to the party.

I think that intuitively it is quite clear what you must do: obey your mother and your best
friend, and hence do the driving and not accept your acquaintances invitation. However, it
is not so clear what formal algorithm could explain this reasoning.
An example of a similar form was first employed in epistemic logic,1 and has been termed
the order puzzle (cf. Horty [21]). For the epistemic version, consider the following sentences:
(5) You remember from physics: if you are in a car, lightning wont strike you.
(6) The coroner tells you: he was struck by lightning.
(7) Your neighbor says: he must have been drinking and driving.
Suppose that driving includes being in a car, that you firmly believe in what you remember from physics, that you believe that information by medical officers is normally based
on competent investigation, and that you usually dont question your neighbors observations, but think that sometimes she is just speculating. It seems quite clear what happens:
you keep believing what you remember from school, and dont doubt what the coroner told
you, but question your neighbors information, maybe answering: This cant be true, as the
authorities found he was struck by lightning, and you cant be struck by lightning in a car.
In both cases, the problem as to how the underlying reasoning can be formally reconstructed seems so far unsolved. Both involve a priority ordering of the sentences involved.
While the paper discusses the imperative side of things from the angle of philosophical (deontic) logic, its solution seems also relevant for the similarly structured problems of conditional
beliefs and desires and the modeling of agent reasoning in the face of such conditionals.
Sections 2 and 3 present the formal framework for the discussion of conditional imperatives
and resulting obligations. Section 4 examines various proposals as to how a priority ordering
may be used to resolve conflicts, it turns out that all of these do not solve the order puzzle.
A postulate at the beginning of Sect. 5 summarizes our intuitions in this matter, but also
delegates the solution to the problem to a proper definition of what it means that conditional
imperatives conflict in a given situation. Three such definitions are studied, of which the
third seems to solve the problems. Section 6 gives theorems of a corresponding deontic logic
and Sect. 7 points at remaining problems for the representation of conditional imperatives.
Section 8 concludes.
2 Formal preliminaries
To formally discuss problems such as the one presented above, I shall use a simple framework: let I be a set of objects, they are meant to be (conditional) imperatives. Two functions
g and f associate with each imperative an antecedent and a consequentthese are sentences
from the language of a basic logic that here will be the language L P L of propositional
logic.2 g(i) may be thought of as describing the grounds, or circumstances in which the
consequent of i is to hold, and f (i) as associating the sentence that describes what must be
1 Cf. Rintanen [34] p. 234, who in turn credits Gerhard Brewka with its invention.
2 PL is based on a language L
P L , defined from a set of proposition letters Prop = { p1 , p2 , ...}, Boolean
connectives , , , , and brackets (, ) as usual. The truth of a L P L -sentence (I use upper case letters
A, B, C, ...) is defined recursively using valuations v : Prop {1, 0} (I write v | A), starting with v | p
iff v( p) = 1 and continuing as usual. If A L P L is true for all valuations it is called a tautology. PL is the
set of all tautologies, and used to define provability, consistency and derivability (I write P L A) as usual.
is an arbitrary tautology, and is .
123
251
the case if the imperative i is satisfied, its deontic focus or demand.3 In accordance with
tradition (cf. Hofstadter and McKinsey [19]), I write A !B for an i I with g(i) = A and
f (i) = B, and !A means an unconditional imperative !A. Note that A !B is just the
name for a conditional imperative that demands B to be made true in a situation where A is
trueit is not an object that is assigned truth values. A useful construction is the materialization m(i) of an imperative i, which is the material implication g(i) f (i) that may be
thought of as corresponding to a conditional imperative. For any i I and I , instead of
f (i), g(i), m(i), f ( ), g( ) and m( ), I may use the superscripted i f , i g, i m, f , g and
m for better readability.
Let I be a tuple I, f, g , let W L P L be a set of sentences, representing real world
facts, and I be a subset of the imperatives: then we define
Triggered I (W, ) = {i
|W P L g(i)}.
So an imperative i is triggered if its antecedent is true given W . Tradition wants it that

a conditional imperative can only be fulfilled or violated if its condition is the case.4 So I
define:
Satisfied I (W, ) = {i
Violated I (W, ) = {i
|W P L i g i f },
|W P L i g i f },
An imperative in Satisfied I (W, ) [Violated I (W, )] is called satisfied [violated] given the
facts W . It is of course possible that an imperative is neither satisfied nor violated given the
facts W . If an imperative is triggered, but not violated, we call the imperative satisfiable:
SatisfiableI (W, ) = {i Triggered I (W, )|W
PL
i f }.
Moreover, the following definition will play a major rle in what follows:
ObeyableI (W, ) = {
| m W
PL
}.
So a subset of is obeyable given W iff it is not the case that for some {i 1 , . . . , i n }
g
f
g
f
we have W P L (i 1 i 1 ) . . . (i n i n ): otherwise we know that whatever we do, i.e.
given any maxiconsistent subset V of L P L that extends W V , at least one imperative in
is violated.5 We speak of a conflict of imperatives when the triggered imperatives cannot all
be satisfied given the facts W , i.e. when Triggered I (W, ) f W P L . More generally
speaking I will also call imperatives conflicting if they are not obeyable in the given situation.
As prioritized conditional imperatives are our concern here, we let all imperatives in I be
ordered by some priority relation < I I . The relation < is assumed to be a strict partial
order on I , i.e. < is irreflexive and transitive, and additionally we assume < to be wellfounded, i.e. infinite descending chains are excluded. For any i 1 , i 2 I, i 1 < i 2 means that
i 1 takes priority over i 2 (ranks higher than i 2 , is more important than i 2 , etc.). A tuple I, f, g
will be called a conditional imperative structure, and I, f, g, < a prioritized conditional
imperative structure. If all imperatives in I are unconditional, we may drop any reference to
the relation g in the tuples and call these basic imperative structures and prioritized imperative structures respectively.
3 In analogy to Reiters default logic one might add a third function e that describes exceptional circumstances
in which the imperative is not to be applied. I will not address this additional complexity here.
4 Cf. Rescher [33], Sosa [38], van Fraassen [9]. Also cf. Greenspan [11]: Oughts do not arise, it seems, until
it is too late to keep their conditions from being fulfilled.
5 Terms differ here, e.g. Downing [8] uses the term compliable instead of obeyable.
123
252
3 Deontic concepts
Given a set of imperatives, one may truly or falsely state that their addressee must, or must
not, perform some act or achieve some state of affairs according to what the addressee was
ordered to do. Regarding the drinking and driving example, I think it is true that the agent
ought to do the driving, as this is what the second-ranking imperative, uttered by the agents
best friend, requires her to do, but that it would be false to say that the agent ought to drink
and drive. Statements that something ought to be done or achieved are called normative or
deontic statements, and the ultimate goal of deontic logic is to find a logical semantics that
models the situation and defines the deontic concepts in such a way that the formal results
coincide with our natural inclinations in the matter.
3.1 Deontic operators for unconditional imperatives
For unconditional imperatives, such definitions are straightforward. Given a basic imperative
structure I = I, f , a monadic deontic O-operator, that formalizes it ought to be that A (is
realized) by O A, is defined by
(td-m1) I | O A if and only if (iff) I f P L A.
So obligation is defined in terms of what the satisfaction of all imperatives logically implies.6
With the usual truth definitions for Boolean operators, it can easily be seen that such a definition produces a normal modal operator, i.e. one that is defined by the following axiom
schemes plus modus ponens:
(Ext) If P L A B, then O A O B is a theorem.
(M) O(A B) (O A O B)
(C) (O A O B) O(A B)
(N) O
Furthermore, (td-m1) defines standard deontic logic SDL, which adds

(D) O A OA
iff the imperatives are assumed to be non-conflicting and so I f is PL-consistent, i.e. I f P L .
It is immediate that in the case of conflicts, (td-m1) pronounces everything as obligatory, and
in particular defines O true, thus making the impossible obligatory. If conflicts are not
excluded, a solution is to only consider (maximal) subsets of the imperatives whose demands
are consistent and define the O-operator with respect to these (I write I C for the set of all
C-remainders, i.e. maximal subsets of I such that f P L C):
(td-m2) I | O A iff I
: f P L A
Quite similarly, a dyadic deontic operator O(A/C), meaning that A ought to be true given
that C is true, can be defined with respect to the maximal subsets of imperatives that do not
conflict in these circumstances:
(td-d1) I | O(A/C) iff I
C : f P L A
6 Such a definition of obligation was proposed e.g. by Alchourrn and Bulygin [1].
123
253
So A is obligatory given that C is true if A is what the imperatives in any C-remainder

demand.7 In the case of conflicts, this definition produces a disjunctive solution: e.g. if
there are two imperatives !A and !B with P L C (A B), then neither O(A/C) nor
O(B/C) but O(A B/C) is true.8
Often, we want to use the information that we have about the circumstances also for
reasoning about the obligations in these circumstances. E.g. if the set of imperatives is {!( p1
p2 )}, ordering me to either send you a card or phone you, and I cannot send you a card, i.e.
p1 is true, I should be able to conclude that I should phone you, and so O( p2 / p1 ) should
be true. Such circumstantial reasoning is achieved by the following change to the truth
definition:
(td-d1+ ) I | O(A/C) iff I
C : f {C} P L A
So A is obligatory given C is (invariably) true iff all maximal subsets of the imperatives
demands (the imperatives associated descriptive sentences) that are consistent with the circumstances C, plus C, derive A. With the usual truth conditions for Boolean operators, a
semantics that employs (td-d1+ ) has a sound and (weakly) complete axiom system PD that
extends the system P of Kraus et al. [22], defined by these axiom schemes
(DExt)
If P L A B then O(A/C) O(B/C) is a theorem.
(DM)
O(A B/C) (O(A/C) O(B/C))
(DC)
O(A/C) O(B/C) O(A B/C)
(DN)
O(/C)
(ExtC)
If P L C D then O(A/C) O(A/D) is a theorem.
(CCMon)
(CExt)
(Or)
O(A D/C) O(A/C D)

If P L C (A B) then O(A/C) O(B/C) is a theorem.
O(A/C) O(A/D) O(A/C D)
with the additional (restricted, dyadic) deontic axiom scheme

(DD-R) If
PL
C then PD O(A/C) O(A/C)
(sometimes called preservation of classical consistency), hence the name PD.9

3.2 Deontic operators for conditional imperatives
Unlike their unconditional counterparts, conditional imperatives have been found hard to
reason about. von Wright [41] called conditional norms the touchstone of normative logic,
and van Fraassen [9] wrote with regard to logics for conditional imperatives: There may
7 Though statements like O(C/C) are syntactically well-formed, they are thus defined false for any pos-
sible situation Cthis is the same for any dyadic deontic logic since Hansson [15] and Lewis [23] (for the
motivation cf. [26] pp. 158159). Similarly, imperatives like C !C are treated as violated as soon as they
are triggered by the facts. There exist meaningful natural-language imperatives like close the window if it
is open, but I think that in these the proposition in the antecedent is different from the negation of the one
corresponding to the consequent, in that the second refers to a different point of time (see to it that the window
is closed some time in the near future if it is open now), so they should not be represented by C !C.
8 For alternative solutions to the problem of conflicts cf. Goble [10] and Hansen [12], [13].
9 For proofs, and an additional credulous ought that defines O(A/C) true if the truth of A is required to
satisfy all imperatives in some C-remainder, cf. Hansen [13].
123
254
be systematic relations governing this moral dynamics, but I can only profess ignorance of
them.
Representing a conditional imperative as an unconditional imperative that demands a material conditional to be made true yields undesired results. Most notorious is the problem of
contraposition: consider a set I with the only imperative !( p1 p2 ), meaning e.g. if it rains,
take an umbrella. (td-d1) makes true O( p2 / p1 ), but also O( p1 / p2 ), so if you cannot
take your umbrella (your wife took it) you must see to it that it does not rain, which is hardly
what the speaker meant you to do. One may think that such problems arise from the fact that
antecedents of conditional imperatives often describe states of the affairs that the agent is not
supposed to, and often cannot, control. But consider the set {!( p1 p2 ), !( p1 p3 )},
it yields O( p2 / p3 ) with (td-d1). Here, p2 is what the consequent of some imperative
demands, so it supposedly describes something the agent can control. Now let the imperatives be interpreted as ordering me to wear a rain coat if it rains, and my best suit if it does
not: it is clear nonsense that I am obliged to wear a raincoat given that I cant wear my best
suit (e.g. it is in the laundry). Such problems are the reason why we use special models for
conditional imperatives that separate antecedents and consequents (conditional imperative
structures), and write p1 ! p2 instead of !( p1 p2 ). But this only delegates the problem
from the level of representation to that of semantics, where now new truth definitions must
be found.
Let I = I, f, g be a conditional imperative structure, and let us ignore for the moment
the further complication of possible conflicts between imperatives. Then the following seems
a natural way to define what ought to be the case in circumstances where C is assumed to be
true:
(td-cd1) I | O(A/C) iff [Triggered I ({C}, I )] f P L A
So dyadic obligation is defined in terms what is necessary to satisfy all imperatives that are
triggered in the assumed circumstances. E.g. if I = { p1 ! p2 }, with its only imperative
interpreted as if you have a cold, stay in bed, then O( p2 / p1 ) truly states that I must stay
in bed given that I have a cold.
Like in the unconditional case, it seems important to be able to use circumstantial reasoning, i.e. employ the information about the situation not only to determine if an imperative
is triggered, but also for reasoning with its consequent. E.g. if the set of imperatives is
{ p1 !( p2 p3 )}, with its imperative interpreted as expressing if you have a cold, either
stay in bed or wear a scarf, one would like to obtain O( p3 / p1 p2 ), expressing that given
that I have a cold and dont stay in bed, I must wear a scarf. So (td-cd1) may be changed
into
(td-cd1+ ) I | O(A/C) iff [Triggered I ({C}, I )] f {C} P L A.
Though the step from (td-cd1) to (td-cd1+ ) seems quite reasonable, such definitions have
also been criticized for defining the assumed circumstances as obligatory. In the above example, (td-cd1+ ) also makes true O( p1 / p1 p2 ), so given that you have a cold it is true that
you ought to have it. The criticism loses much of its edge in the present setting, where one
can point to the distinction between imperatives (there is no imperative that demands p1 )
and ought sentences that describe what must be true given the facts and the satisfaction of all
triggered imperatives: then the truth of O( p1 / p1 ) seems no more paradoxical than the truth
of O that is accepted in most systems of deontic logic.
123
255
3.3 Further modifications

In Makinson and van der Torres [25] more general theory of input/output logic, (td-cd1)
is termed simple-minded output, and (td-cd1+ ) is its throughput version.10 As the names
suggests, the authors also discuss more refined operations, which again might be considered for reasoning about conditional imperatives. One modification addresses the possibility
of reasoning by cases that e.g. makes true O( p2 p4 / p1 p3 ) for a set of imperatives
I = { p1! p2 , p3! p4 }. This may be achieved by the following definition, where L P L C
is the set of all maximal subsets of the language L P L that are consistent with C:11
(td-cd2) I | O(A/C) iff V L P L C : [Triggered I (V, I )] f P L A
In the example, each set V L P L that is maximally consistent with p1 p3 either contains
p1 , then p1! p2 is triggered and so p2 and also p2 p4 is implied by [Triggered I (V, I )] f ,
or it contains p1 , but then it cannot also contain p3 and so must contain p3 , so p3! p4
is triggered and therefore p4 and also p2 p4 implied, so for all sets V, p2 p4 is implied
and so O( p2 p4 / p1 p3 ) made true.
In order to add circumstantial reasoning to (td-cd3)or, in Makinson and van der Torres
terms, for its throughput versionone might, in the vein of (td-d1+ ) and (td-cd1+ ), try
this definition:
(td-cd2 ) I | O(A/C) iff V L P L C : [Triggered I (V, I )] f {C} P L A
But the definition seems too weak. Consider the set { p1 !( p2 p4 ), p3 ! p4 } and the
situation ( p1 p2 ) p3 . We would expect a reasoning as follows: in this situation, either
p1 p2 is true, so the first imperative is triggered but we cannot satisfy it by bringing about
p2 , and so must bring about p4 . Or p3 is true, then the second imperative is triggered and
we must again bring about p4 . So we must bring about p4 in the given situation. But the
definition fails to make true O( p4 /( p1 p2 ) p3 ). Like Makinson and van der Torre [25],
I therefore combine reasoning by cases with a stronger version of throughput:
(td-cd2+ ) I | O(A/C) iff V L P L C : [Triggered I (V, I )] f V P L A
As is easy to see, this resolves the difficulty: for { p1!( p2 p4 ), p3! p4 }, O( p4 /( p1
p2 ) p3 ) is now true, as desired. But this modification has an unwanted consequence: it
makes reasoning about conditional imperatives collapse into reasoning about consequences
of their materializations (cf. [26] p. 156):
Observation 1 By (td cd2+ ), I | O(A/C) iff m(I ) {C} P L A.
Proof For the right-to-left direction, for any imperative i I and any set V L P L C,
either V includes g(i), so i Triggered I (V, I ) and therefore [Triggered I (V, I )] f P L
g(i) f (i), or it does not include g(i), but then it includes g(i) by maximality, hence
V P L g(i) f (i). So [Triggered I (V, I )] f V P L g(i) f (i). For the left-toright direction, if m(I ) {C} P L A then m(I ) {C} {A} is consistent, so there is
a V L P L C such that m(I ) {C} {A} V . It is immediate that for each i
10 If I resembles the generating set G of input/output logic, then O(A/C) means that A is an output given
the input C (Makinson and van der Torre write A out (G, {C})). Though these authors liken their generating
set G to a body of conditional norms, it should be noted that they do not themselves introduce dyadic deontic
operators.
11 Makinson and van der Torres [25] call the resulting operator basic output, of which a syntactical version
was first presented by Swirydowicz

[39] p. 32.
123
256
Triggered I (V, I ), m(I ) V P L f (i), so if [Triggered I (V, I )] f V P L A then m(I )

V P L A and since m(I ) V also V P L A. Since V was consistent and included A, it
cannot also derive A, and so by contraposition [Triggered I (V, I )] f V P L A.
But such an equivalence makes all the problems of identifying conditional imperatives with
unconditional ones that demand their materializations reappear, in particular the problem of
contraposition.12 So it seems we must choose between reasoning by cases and circumstantial reasoning. Another modification that these authors consider is that of reusable output:
when an imperative is triggered that demands A, and A is the trigger for some imperative
A !B, then we also ought to do B. Such a modification can easily be incorporated into a
truth definition and its throughput version:
(td-cd3) I | O(A/C) iff [Triggered I ({C}, I )] f P L A
(td-cd3+ ) I | O(A/C) iff [Triggered I ({C}, I )] f {C} P L A
where Triggered I (W, ) means the smallest subset of I such that for all i , if
[Triggered I (W, )] f W P L g(i) then i Triggered I (W, ). Moreover, the two modifications of reasoning by cases and reusable output can be combined to produce the
following definition and its throughput variant:
(td-cd4) I | O(A/C) iff V L P L C : [Triggered I (V, I )] f P L A
(td-cd4+ ) I | O(A/C) iff V L P L C : [Triggered I (V, I )] f V P L A

The topic of reusable output is discussed under the name of deontic detachment in the
deontic logic literature, and there is no agreement whether such a procedure is admissible (Makinson [24] p. 43 argues in favor, whereas Hansson [16] p. 155 disagrees). E.g. let
I = {! p1 , p1 ! p2 }, and for its interpretation assume that it is imperative for the proper
execution of your job that you develop novel methods, which make you eligible for a bonus,
and that if you develop such novel methods you owe it to yourself to apply for the bonus.
Truth definitions that accept deontic detachment make true O( p2 /), and so tell us that
you ought to apply for the bonus, which seems weird since it may be that you never invent
anything. However, proponents of deontic detachment may argue that in such a situation,
O( p1 p2 /) should hold, i.e. you ought to invent new methods and apply for the bonus,
and that the reluctance to also accept O( p2 /) islike the inference from you ought to put
on your parachute and jump to you ought to jumpjust a variant of Ross Paradox that
is usually considered harmless.
For (td-cd4) we once again obtain O( p2 / p3 ) for I = { p1 ! p2 , p1 ! p3 }: for any
V L P L p3 , p3 V , furthermore either p1 V and so p1 ! p2 Triggered I (V, I ),
or p1 V , then p1 ! p3 Triggered I (V, I ), and since { p3 } { p3 } P L p1 , again
p1! p2 is in Triggered I (V, I ), hence [Triggered I (V, I )] f P L p2 for all V L P L p3 .
But as we saw above, this result seems counterintuitive.13 Note that (td-cd4+ ) is again equivalent to I | O(A/C) iff m(I ) {C} P L A and thus to (td-cd2+ ) (cf. [25] observation 16;
[26], p. 156).
Observation 2 By (td-cd4+ ), I | O(A/C) iff m(I ) {C} P L A.

Proof Similar to the proof of observation 1. For the left-to-right direction, use that for each
i Triggered I (V, I ), m(I ) V P L f (i), which is immediate.
12 (td-cd2 ) does not fare much better: though it does not include contraposition, it again makes O( p / p )
2
3
true for I = { p1! p2 , p1! p3 }, which is counterintuitive.
13 With respect to their out -operation that corresponds to (td-cd4), Makinson and van der Torre [25] speak
4
of a ghostly contraposition.
123
257
3.4 Operators for prioritized conditional imperatives

This paper focuses on prioritized conditional imperatives, and for these there is a further hurdle to finding the proper truth definitions for deontic concepts. Priorities are only required
if the imperatives cannot all be obeyedotherwise there is no reason not to obey all, and
the priority ordering is not used. So the truth definitions must be able to deliver meaningful
results for possibly conflicting imperatives. The intuitive idea is to use the information in the
ordering to choose subsets of the set of imperatives under consideration that contain only
the more important imperatives and leave out less important, conflicting ones, so that the
resulting preferred subset (or rather, subsets, since the choice may not always be uniquely
determined by the ordering) only contains imperatives that do not conflict in the given situation. More generally, let I be a prioritized conditional imperative structure I, g, f, < , and
let be a subset of I . Then PI (W, ) contains just the subsets of that are thus preferred
given the world facts W . The above truth definitions can then be adapted such that they now
describe something as obligatory iff it is so with respect to all the preferred subsets of the
imperatives, i.e. they take on the following forms:
(td- pcd1)
(td- pcd1+ )
(td- pcd2)
(td- pcd2+ )
(td- pcd3)
(td- pcd3+ )
(td- pcd4)
(td- pcd4+ )
I|
O(A/C) iff PI ({C}, I ) :

[Triggered I ({C}, )] f P L
[Triggered I ({C}, )] f {C} P L
V L P L C : [Triggered I (V, )] f P L
V L P L C : [Triggered I (V, )] f V P L
[Triggered I ({C}, )] f P L
[Triggered I ({C}, )] f {C} P L
V L P L C : [Triggered I (V, )] f P L
V L P L C : [Triggered I (V, )] f V P L
A,
A,
A,
A,
A,
A,
A,
A.
So e.g. (td pcd1) defines A as obligatory if the truth of A is required to satisfy the triggered imperatives in any preferred subset. Of course, the crucial and as yet missing element
is the decision procedure that determines the set PI ({C}, I ) of preferred subsets. The next
section discusses several proposals to define such subsets; a new proposal is presented in the
section that follows it.
4 Identifying the preferred subsets

4.1 Brewkas preferred subtheories
The idea that normative conflicts can be overcome by use of a priority ordering of the norms
involved dates back at least to Ross [35] and is also most prominent in von Wrights work (cf.
[40] p. 68, 80).14 However, it has turned out to be difficult to determine the exact mechanism
by which such a resolution of conflicts can be achieved. This is true even when only unconditional imperatives are considered. Discussing various proposals for resolution of conflicts
between unconditional imperatives, I have argued in [14] that an incremental definition
should be used for determining the relevant subsets. Based on earlier methods by Rescher
[32], such a definition was first introduced by Brewka [3] for reasoning with prioritized
14 But cf. already Watts [42] part II, ch. V, sec. III, principle 10: Where two duties seem to stand in opposition
to each other, and we cannot practise both, the less must give way to the greater, and the omission of the less
is not sinful.
123
258
defaults. For any priority relation <, the idea is to consider all the full prioritizations of
< (strict well orders that preserve <), and then work ones way from the top of the strict
order downwards by adding the -next-higher imperative to the thus constructed preferred
subtheory if its demand is consistent with the given facts and the demands of the imperatives
that were added before. For the present setting, the definition can be given as follows:
Definition 1 (Brewkas preferred subtheories) Let I = I, f, g, < be a prioritized conditional imperative structure, be a subset of I , and W L P L be a set of PL-sentences.
Then PIB (W, ) iff (i) W P L , and (ii) is obtained from a full prioritization by
defining
f
i =
for any i
j {i} if W
otherwise,
j i j
j i
, and letting =
j i
j {i}
PL
, and
i .
Clause (i) ensures that for an inconsistent set of assumed facts, no set is preferred. Somewhat roundabout, owed to the possibility of infinite ascending subchains, clause (ii) then
recursively defines a set PIB (W, ) for each full prioritization : take the -first i (the
exclusion of infinite descending subchains guarantees that it exists) and if W {i f } P L
then let i = {i}; otherwise i is left empty.15 Similarly, any -later i is tested for possible
addition to the set j i j of elements that were added in the step for a j that occurs
-prior to i. is then the union of all these sets.
To see how this definition works, consider the set I = {!( p1 p2 ), ! p1 , ! p2 }, with the
ranking !( p1 p2 ) < ! p2 and ! p1 < ! p2 . For an interpretation, let !( p1 p2 )) be your
mothers request that you buy cucumbers or spinach for dinner, ! p1 be your fathers wish
that no cucumbers are bought, and ! p2 your sisters desire that you dont buy any spinach.
We have two full prioritizations !( p1 p2 ) < ! p1 < ! p2 and ! p1 < !( p1 p2 ) < ! p2
let these be termed 1 and 2 , respectively. The construction for 1 adds the imperative
!( p1 p2 ) in the first step and, since no conflict with the situation arises, ! p1 in the second
step. In the third and last step, nothing is added since ! p2 conflicts with the demands of
the already added imperatives. For 2 the only difference is that the first two imperatives
are added in inverse order. Thus PIB (W, I ) = {{!( p1 p2 ), ! p1 }}. Using (td pcd1) we
obtain O( p1 p2 /), which means that you have to buy spinach and not cucumbers, thus
fulfilling your parents requests but not your sisters, which seems reasonable.
As I showed in [14], Brewkas method is extremely successful for dealing with unconditional imperatives. It is provably equivalent for such imperatives to methods proposed
by Ryan [36] and Sakama and Inoue [37], and it avoids problems of other approaches by
Alchourrn and Makinson [2], Prakken [30] and Prakken and Sartor [31]. Moreover, an
equally intuitive maximization method proposed by Nebel [28], [29], that adds first a maximal set of the highest-ranking imperatives, then a maximal set of second-ranking imperatives,
etc., but for its construction requires the ordering to be based on a complete preorder, can
be shown to be embedded in Brewkas approach for such orderings. So my aim will be to
retain Brewkas method for the unconditional casein fact, all proposals that follow meet
this criterion. However, when it is applied without change to conditional imperatives, the
15 As usual, the union of an empty set of sets is taken to be the empty set.
123
259
algorithm may lead to incorrect results. E.g. consider a set I with two equally ranking imperatives { p1 ! p2 , p1 ! p2 }, meaning e.g. if you go out, wear your boots and if you
dont go out, dont wear your boots: since the consequents contradict each other, an unmodified application of Brewkas method produces PIB ({ p1 }, I ) = {{ p1! p2 }, { p1! p2 }},
which fails to make true O( p2 / p1 ) by any truth definition of Sect. 3.4: the right set contains
no imperatives that are triggered by p1 . So we cannot derive that you ought to wear your
boots, given that you are going out. But intuitively there is no conflict, since the obligations
arise in mutually exclusive circumstances only.
4.2 A nave approach
A straightforward way to adopt Brewkas method to the case of conditional imperatives is to
use not all imperatives for the construction, but only those that are triggered by the facts W ,
i.e. to use Triggered I (W, ) instead of :
Definition 2 (The nave approach) Let I = I, f, g, < be a prioritized conditional imperative structure,
be a subset of I , and W L P L be a set of PL-sentences. Then
PIn (W, ) iff PIB (W, Triggered I (W, )).
The change resolves our earlier problems with Brewkas method: consider again the set of
imperatives { p1! p2 , p1! p2 }, where the imperatives were interpreted as ordering me
to wear my boots when I go out, and not wear my boots when I dont. The new definition produces PIn ({ p1 }, I ) = {{ p1 ! p2 }}, its only preferred subset containing just the
one imperative that is triggered given the facts { p1 }. By all truth definitions of Sect. 3.4,
O( p2 / p1 ) is now true, so given that you go out, you ought to wear your boots, which is as
it should be.
The nave approach is similar to Hortys proposal in [20] in that conflicts are only removed
between imperatives that are triggered (though the exact mechanism differs from Hortys).
When I nevertheless call it nave, this is because there are conceivable counterexamples to
this method. Consider the set of prioritized imperatives ! p1 < p1! p2 < ! p2 , and for an
interpretation suppose that your job requires you to go outside p1 , that your mother, who is
concerned for your health, told you to wear a scarf p2 if you go outside, and that your friends
dont want you to wear a scarf, whether you go outside or not. In the default situation
only the first imperative and the third are triggered, i.e. Triggered I ({}, I ) = {! p1 , ! p2 }.
Since their demands are consistent with each other, we obtain PIn ({}, I ) = {{! p1 , ! p2 }},
for which all truth definitions of Sect. 3.4 make O( p1 p2 /) true. So you ought to go
out and not wear a scarf, thus satisfying the first and the third imperative, but violating the
second-ranking imperative. But arguably, if an imperative is to be violated, it should not be
the second-ranking p1! p2 , but the lowest ranking ! p2 instead.
4.3 The stepwise approach
To avoid the difficulties of the nave approach, it seems we must not just take into account
the imperatives that are triggered, but also those that become triggered when higher ranking
imperatives are satisfied. To this effect, the following modification may seem
reasonable:
Definition 3 (The stepwise approach) Let I = I, f, g, < be a prioritized conditional
imperative structure, be a subset of I , and W L P L be a set of PL -sentences. Then
123
260
P s (W, ) iff (i) W

defining
PL
, and (ii) is obtained from a full prioritization by

f
i =
for any i
j {i} if i SatisfiableI (W
otherwise,
j i j
j i
, and letting =
j i
, ), and
i .
So at each step one considers what happens if the imperatives that were included so far are
satisfied, and adds the current imperative only if it is satisfiable given W and the satisfaction
of these previous imperatives. Note that satisfiability of an imperative, like its satisfaction and
violation, presupposes that the imperative is triggered. The new definition not only includes,
at each step, those imperatives that are triggered and can be satisfied given the facts and the
supposed satisfaction of the previously added imperatives: it also includes those that become
triggered when a previously added imperative is satisfied.
This modification avoids the previous difficulty: consider again the set of prioritized
imperatives ! p1 < p1 ! p2 <! p2 . There is just one full prioritization, which for W = {}
yields in the first step the set {! p1 }, and in the second step {! p1 , p1! p2 }, since p1! p2 is
triggered when the previously added imperative ! p1 is assumed to be satisfied. In the third
step, nothing is added: though the imperative ! p2 is triggered, it cannot be satisfied together
with the previously added imperatives. So we obtain PIs ({}, I ) = {{! p1 , p1 ! p2 }}, and
hence O( p1 /), but not O( p1 p2 /), is defined true by all truth definitions of Sect. 3.4.
Operators that accept deontic detachment (td- pcd3(+) , 4(+) ) make true O( p1 p2 /), so
you must go out and wear a scarf, which is as it should be.
However, a small change in the ordering shows that this definition does not suffice: let
the imperatives now be ranked p1 ! p2 < ! p1 < ! p2 . (For the interpretation, assume
that the conditional imperative to wear a scarf when leaving the house was uttered by some
high-ranking authority, e.g. a doctor.) Then again PIs ({}, I ) = {{! p1 , ! p2 }}: in the first
step, nothing is added since p1 ! p2 is neither triggered by the facts nor by the assumed
satisfaction of previously added imperatives (there are none). In the next two steps, ! p1 and
! p2 are added, as each is consistent with the facts and the satisfaction of the previously
added imperatives. So again all truth definitions of Sect. 3.4 make true O( p1 p2 /), i.e.
you ought to go out and not wear a scarf, satisfying the second and third ranking imperatives
at the expense of the highest ranking one. But surely, if one must violate an imperative, it
should be one of the lower-ranking ones instead.
4.4 The reconsidering approach
The merits of the stepwise approach were that it did not only consider the imperatives that are
triggered, but also those that become triggered when already added imperatives are satisfied.
Such considerations applied to those imperatives that follow in the procedure. Yet the satisfaction of already added imperatives might also trigger higher-ranking imperatives, which by
this method are not considered again. So it seems necessary, at each step, to reconsider also
the higher-ranking imperatives. An algorithm that does that was first introduced for default
theory by Marek and Truszczynski [27] p. 72, and later employed by Brewka in [4]; it can
be reformulated for the present setting as follows:
Definition 4 (The reconsidering approach) Let I = I, f, g, < be a prioritized conditional
imperative structure, be a subset of I , and W L P L be a set of PL-sentences. Then
123
261
PIr (W, ) iff (i) W

defining
PL
, and (ii) is obtained from a full prioritization by

f
i =
for i
j i
j min SatisfiableI W
, and letting =
j i
j i
i .16
The definition reconsiders at each step the whole ordering, and adds the -first17 imperative (due to the definition of there is just one) that has not been added previously and is
satisfiable given both the facts W and the consequents of the previously added imperatives.
To see how the definition works, consider again the example which the stepwise approach
failed, i.e. the set of prioritized imperatives p1! p2 < ! p1 < ! p2 . We are interested in the
preferred sets for the default circumstances , i.e. the sets in PIr ({}, I ). I is already fully
prioritized, so there is just one such set. Applying the algorithm, we find the minimal (highest
ranking) element in SatisfiableI ({}, I ) is ! p1 , so this element is added in the first step. In
the second step, we look for the minimal element in SatisfiableI ({} {! p1 } f , I ), other
than the previously added ! p1 . It is p1! p2 , since the assumed satisfaction of all previously
added imperatives triggers it, and its consequent can be true together with {} { p1 }. So
p1 ! p2 is added in this step. In the remaining third step, nothing is added: ! p2 is not in
SatisfiableI ({}{! p1 , p1! p2 } f , I ), and all other elements in this set have been previously
added. So PIr ({}, I ) = {{! p1 , p1! p2 }}. Now all truth definitions of Sect. 3.4 make true
O( p1 /), but not O( p1 ! p2 /), and operators that accept deontic detachment make
true O( p1 p2 /). So, in the given interpretation, you must go out (as your job requires)
and wear a scarf (as the doctor ordered you to do in case you go out), which is as it should
be.
However, again problems remain. Let the imperatives now be prioritized in the order
p1 ! p2 < ! p2 < ! p1 . Let p1 ! p2 stand for the doctors order to wear a scarf when
going outside, let ! p2 stand for your friends expectation that you dont wear a scarf, and let
! p1 represent your sisters wish that you leave the house. Construct the set in PIr ({}, I )
since I remains fully prioritized, there is again just one such set. The minimal element
in SatisfiableI ({}, I ) is ! p2 , and so is added in the first step. The minimal element in
SatisfiableI ({} {! p2 } f , I ), other than ! p2 , is ! p1 which therefore gets added in the
second step. Nothing is added in the remaining step: ! p2 and ! p1 have already been added,
and p1 ! p2 is not in SatisfiableI ({} {! p2 , ! p1 } f , I ): though it is triggered by the
assumed satisfaction of ! p1 , its consequent is contradicted by the assumed satisfaction of
! p2 . So PIr ({}, I ) = {{! p1 , ! p2 }}. Hence all truth definitions of Sect. 3.4 again make
true O( p1 p2 /), so you ought to go out without a scarf, again satisfying the second and
third ranking imperatives at the expense of the first, which seems the wrong solution.
4.5 A fixpoint approach
To eliminate cases in which the reconsidering approach still adds imperatives that can
only be satisfied at the expense of violating a higher-ranking one, a fixpoint approach was
first proposed for default reasoning by Brewka and Eiter [5]. It tests each set that may be
considered as preferred to see if it really includes all the elements that should be included:
16 Note that in , i is used just as an indexit does not mean that i is considered for addition at this step,
i
and may be added at an earlier or later step (or not at all).

17 For any ordering < on some set , min = {i |i : if i = i, then i i}, and max = {i
<
<
|i : if i = i, then i i }, as usual.
123
262
imperatives that are triggered given the facts and the assumed satisfaction of all imperatives
in the set, and would be added by Brewkas [3] original method that adds the higher ranking
imperatives first. The procedure translates as follows:
Definition 5 (The fixpoint approach)
Let I = I, f, g, < be a prioritized conditional
imperative structure, be a subset of I , and W L P L be a set of PL-sentences. Then
f
PI (W, ) iff PIB (W, Triggered I (W f , )).

To see how this definition works, consider the above set of prioritized imperatives p1
! p2 < ! p2 < ! p1 . It is immediate that the set {! p1 , ! p2 } cannot be in PIf ({}, I ):
if we assume all imperatives in this set to be satisfied, then all imperatives are triggered,
i.e. Triggered I ({} {! p1 , ! p2 } f , I ) = I . By Brewkas original method, PIB (W, I ) =
{{ p1 ! p2 , ! p1 }}: < is already fully prioritized, and for this full prioritization the method
adds p1! p2 in the first step, ! p2 cannot be added in the second step since its consequent
contradicts the consequent of the previously added p1 ! p2 , and in the third step ! p1 is
added. So since the considered set {! p1 , ! p2 } is not in PIB (W, I ), it is not a fixpoint.
Rather, as may be checked, the only fixpoint in PIf ({}, I ) is { p1 ! p2 , ! p1 }. Then all
truth definitions of Sect. 3.4 make true O( p1 /), but no longer O( p1 p2 /). Moreover,
truth definitions that allow deontic detachment make true O( p1 p2 /). In the given
interpretation this means that you must leave the house at your sisters request and wear a
scarf, as the doctor ordered you to do in case you go out.
Though the construction now no longer makes true O( p1 p2 /), its solution for
the example, that determines the set { p1 ! p2 , ! p1 } as the fixpoint of the set of prioritized
imperatives p1 ! p2 < ! p2 < ! p1 , seems questionable. Though this now includes the
doctors order, you now have no obligation anymore to satisfy the imperative that is second
ranking, i.e. your friends request that you dont wear a scarf; truth definitions that accept
deontic detachment even oblige you to violate it by wearing a scarf. Now consider the
situation without the third ranking imperative ! p1 : it can easily be verified that for a set
I = { p1 ! p2 , ! p2 } the only fixpoint in PIf ({}, I ) is {! p2 }. So for the reduced set,
(td pcd1) makes true O( p2 /), i.e. you ought to obey your friends wish. That the satisfaction of this higher ranking imperative ! p2 should no longer be obligatory when a lower
ranking imperative ! p1 is added, seems hard to explain. If the ranking is taken seriously, I
think one should still satisfy the higher ranking imperatives, regardless of what lower ranking
imperatives are added.
For another problem consider the set of prioritized imperatives p1 ! p2 <!( p1 p2 ) <
! p3 . For an interpretation, let the first imperative be again the doctors order to wear a scarf
in case you go out, the second one be your friends request to go out and not wear a scarf,
and the third be the wish of your aunt that you write her a letter. It is easily proved that the set
has no fixpoint, and so there is also none that contains ! p3 , hence all truth definitions make
O( p3 /) false, so you do not even have to write to your aunt. But even if the presence of
a higher ranking conditional imperative and a lower ranking imperative to violate it poses a
problem (why should it? after all, the lower ranking imperative is outranked), it is hard to see
why the subject should be left off the hook for all other, completely unrelated obligations.18
18 The lack of fixpoints is a well-known problem of such definitions (cf. e.g. Caminada and Sakama [7]).
Another approach to conditional imperatives by Makinson in [24] has trouble resolving the same example:
for the default circumstances it produces the set {!( p1 p2 ), ! p3 }. p1 ! p2 is not considered, since its
only label (roughly: a conjunction of the circumstances, the imperatives antecedents that would trigger it,
and its consequent) is inconsistent (it is ( p1 p2 ) p2 ). But why should the agent not be free to obey
p1! p2 , and not violate it by satisfying !( p1 p2 )?
123
263
4.6 Discussion
For a discussion of our results so far, let us return to the drinking and driving example from
the introduction. Let the three imperatives:
(3) Some acquaintance says: if you go to the party, then have a drink with me.
be represented by the set of prioritized imperatives p1 p2 < p3 p2 < p3 p1 . Let the
set of facts be { p3 }, i.e. you go to the party. Brewkas original method is not tailored to
be directly employed on conditional imperatives, as it ignores the antecedents altogether.
The next three approaches, the nave, the stepwise and the reconsidering ones, produce
PIn ({ p3 }, I ) = PIs ({ p3 }, I ) = PIr ({ p3 }, I ) = {{ p3 ! p2 , p3 ! p1 }}, which by all truth
definitions of Sect. 3.4 makes true O( p1 p2 / p3 ), so you ought to drink and drive. The
fixpoint approach produces PIf ({ p3 }, I ) = {{ p1! p2 , p3! p1 }}, so all truth definitions
make true O( p1 / p3 ), which means you ought to drink. Truth definitions with deontic detachment additionally make true O( p1 p2 / p3 ), so you ought to drink and not drive. But
being obliged to drink runs counter to our intuitions for the drinking and driving example.
So we have to look for a different solution.
Before we do that, I will, however, question again our intuition in this matter. Horty [21]
has recently used a structurally identical example to argue for just the opposite, that the solution by the fixpoint approach is correct. His example is that of three commands, uttered by
a colonel, a major and a captain to a soldier, Corporal OReilly. The Colonel, who does not
like it too warm in the cabin, orders OReilly to open the window whenever the heat is turned
on. The Major, who is a conservationist, wants OReilly to keep the window closed during
the winter. And the Captain, who does not like it to be cold, orders OReilly to turn the heat
on during the winter. OReilly is trying to figure out what to do. The intended representation
is again the prioritized conditional imperative structure employed above for the drinking
and driving example, where p1 now means that the heat is turned on, p2 means that the
window is closed, and p3 means that it is winter. We saw that the fixpoint approach yields
the preferred subset { p1 ! p2 , p3 ! p1 }, making true O( p1 / p3 ) for all truth definitions,
and O( p1 p2 / p3 ) for truth definitions that accept deontic detachment, so OReilly must
turn on the heat and then open the window, and thus violate the Majors order. Horty argues
as follows in support of this choice:
OReillys job is to obey the orders he has been given exactly as they have been
issued. If he fails to obey an order issued by an officer without an acceptable excuse,
he will be court-martialed. And, let us suppose, there is only one acceptable excuse for
failing to obey such an order: that obeying the order would, in the situation, involve
disobeying an order issued by an officer of equal or higher rank. (...) So given the set
of commands that OReilly has been issued, can he, in fact, avoid court-martial? Yes
he can, by (...) obeying the orders issued by the Captain and the Colonel (...). OReilly
fails to obey the Majors order, but he has an excuse: obeying the Majors order would
involve disobeying an order issued by the Colonel.
Hortys principle seems quite acceptable: for each order issued to the agent, the agent may
ask herself if obeying the order would involve disobeying an order of a higher ranking officer
(then she is excused), and otherwise follow it. The result is a set of imperatives where each
imperative is either obeyed, or disobeyed but the disobedience excused. When I nevertheless
think the argument is not correct, it is because I think it confuses the status quo and the status
123
264
quo posterior. Obeying the Majors order does not, in the initial situation, involve disobeying
the Colonels order. Only once OReilly follows the Captains order and turns on the heat, it
is true that he must obey the Colonel, open the window and thus violate the Majors order. But
this does not mean that he should follow the Captains order in the first placeas by doing
so he brings about a situation in which he is forced, by a higher ranking order, to violate a
command from another higher ranking officer. Quite to the contrary, I think that being forced
to violate a higher ranking order when obeying a lower ranking one is a case where following
the lower one involves such a violation, and so the only order the agent is excused from
obeying is the lowest ranking command.
Consider finally this variant: suppose that if I am attacked by a man, I must fight him (to
defend my life, my family etc.). Furthermore, suppose I have pacifist ideals which include
that I must not fight the man. Now you tell me to provoke him, which in the given situation
means that he will attack me. Let self-defense rank higher than my ideals, which in turn rank
higher than your request. Should I do as you request? By the reasoning advocated by Horty,
there is nothing wrong with it: I satisfy your request, defend myself as I must, and though I
violate my ideals, I can point out to myself that the requirement to fight back took priority.
But I think if I really do follow your advice, I would feel bad. I think this would not just be
some irrational regret for having to violate, as I must, my ideals, but true guilt for having
been tempted into doing something I should not have done, namely provoking the man: it
caused the situation that made me violate my ideals. So I think our intuitions in the drinking
and driving example and the other cases have been correct.
5 New strategies and a new proposal

In the face of the difficulties encountered so far, it seems necessary to address the issue of
finding an appropriate mechanism for a resolution of conflicts between prioritized conditional
imperatives in a more systematic manner. So far intuition has guided us mainly as to what
imperatives should be included in a subset of preferred imperatives. I think the following
postulate sums up the intuitions that have so far influenced the proposal and rejection of
solutions:
Postulate Any imperative should be considered relevant (included) as long as it is not
violated or, in the given situation, conflicts with other imperatives that are also considered
relevant (included) and do not rank lower.
The postulate makes clear that the need to satisfy a lower ranking imperative cannot serve
as an excuse to violate an imperative of higher priority. However, this postulate delegates
the answer of how the set of preferred imperatives should be constructed to the answer of
another question: when do conditional imperatives conflict in a given situation? There appear
to be several possible answers to this question, which lead to different solutions.
5.1 Subsets with consistent extensions
For the definition of conflicts between conditional imperatives one might recur to a definition
of conflicts between desires as proposed by Broersen et al. [6]. Their idea, translated to our
setting, is that for any set of facts W and set of conditional imperatives there is a smallest set
E(W, ) L P L such that (i) W E(W, ) and (ii) for any i , if E(W, ) P L g(i)
then f (i) E(W, ). The set of imperatives is then defined as conflicting given the facts
W if the thus constructed extension E(W, ) of the facts is inconsistent, i.e. E(W, ) P L
123
265
(cf. [6] def. 5.5). This test for conflicts may then be employed within a variant of Brewkas
original method.
Definition 6 (Preferred Subsets with Consistent Extensions) Let I = I, f, g, < be a prioritized conditional imperative structure,
be a subset of I , and W L P L be a set of
PL-sentences, and let E(W, ) be defined as described above. Then PIe (W, ) iff (i)
W P L , and (ii) is obtained from a full prioritization by defining
i =
for any i
j {i} if E(W,
otherwise,
j i j
j i
, and letting =
j i
j {i})
PL
, and
i .
To see how this definition works, consider the drinking and driving example, where the set
of prioritized imperatives is p1 ! p2 < p3 ! p2 < p3 ! p1 and the situation W = { p3 }.
There is only one full prioritization which is identical with <. In the first step, p1 ! p2 is
added, since E({ p3 }, { p1 ! p2 }) = { p3 } which is consistent. In the second step, p3 ! p2
is added, as E({ p3 }, { p1 ! p2 , p3 ! p2 }) = { p3 , p2 }, which is again consistent. In
the third step, p3 ! p1 is rejected, since E({ p3 }, { p1 ! p2 , p3 ! p2 , p3 ! p1 }) =
{ p3 , p2 , p1 , p2 }, which is inconsistent. So we have PIe ({ p3 }, I ) = {{ p1 ! p2 , p3
! p2 }}, making true O( p2 / p3 ) for all truth definitions of Sect. 3.4, so given that I go to the
party I must do the driving, which is as it should be.
However, there is a problem for the test using consistent extensions, as for some truth definitions it delivers sets of imperatives that are clearly conflicting, in the sense that they make
O(/C) true for a consistent fact C: Consider the set of facts W = { p1 p2 } and the set of
imperatives I = { p1 ! p3 , p2 ! p3 , ! p3 }. We have E(W, I ) = { p1 p2 , p3 }, which
is consistent, and so all imperatives in I are added to the preferred subset, regardless of their
ordering. But for any truth definition that allows for reasoning by cases (td-pcd2(+), 4(+) )
we then have both O( p3 / p1 p2 ) and O( p3 / p1 p2 ), and so also O(/ p1 p2 ).
This is simply because the construction of extensions does not take care of reasoning by
cases, i.e. it does not add p3 to the extension in case we both have p1 p2 in W and
{ p1 ! p3 , p2 ! p3 } I . Perhaps the definition of consistent extensions can be amended,
but to avoid delivering preferred sets of imperatives that make true O(/C) for some truth
definition and some consistent fact C suggests a different solution that is explored in the next
section.
5.2 Deontically tailored preferred subsets
In the unconditional case, the reason to move from definition (td-m1) to (td-m2) was that
when there are conflicts between imperatives, the former makes true the monadic deontic
formula O, i.e. the agent ought to do the logically impossible. This result was avoided by
considering only maximal sets of imperatives with demands that are collectively consistent,
i.e. sets that do not make O true. When faced with the question what dyadic deontic formula should not be true when conflicts are resolved for arbitrary situations C, the formula
O(C/C) appears to be the dyadic equivalent: a mechanism for conflict resolution should
not result in telling the agent to change the supposed, unalterable facts.19 So to define the set
PI ({C}, I ) required by the truth definitions (td- pcd1(+) 4+ ), we can modify Brewkas
19 This test is identical to the one used by Makinson and van der Torre [26] p. 158/159 to determine consistency of output (cf. also for arguments why O(C/C) should be used, i.e. for their setting, the output should
be consistent with the input, rather than the formula O(/C) and thus consistency of output simpliciter).
123
266
original method in such a way that it tests, at each step, for each of the constructed subsets, if
the corresponding truth-definition (td-cd1(+) 4+ ) does not make O(C/C) true for this
set.20 Formally:
Definition 7 (Deontically Tailored Preferred Subsets) Let I = I, f, g, < be a prioritized
conditional imperative structure, and C LPL describe the given situation. Let (tdpcd) be
any of the truth definitions (td- pcd1(+) 4(+) ). Then is in the set PI ({C}, I ) employed
by this truth definition iff (i) {C} P L , and (ii) is obtained from a full prioritization
by defining
i =
j {i} if
j i j {i}, f, g
otherwise,
j i j
j i
for any i I , and letting =
iI
O(C/C)by(td-cd),
i .
By this construction, each of the preferred subsets contains a maximal number of the imperatives such that they do not make true O(C/C) for the situation C and the truth definition
that is employed, and so the resulting truth definition likewise avoids this truth. Such a construction of the preferred subsets might be considered tailored to the truth definition in
question, and any remaining deficiencies might be seen as stemming from the employed
truth definition. But this being so, the method reveals a strong bias towards truth definitions
that accept deontic detachment, and in particular truth definitions (td-pcd2+, 3(+), 4(+) ):
Consider the set of imperatives I = {! p1 , p1 ! p2 , ! p2 } with the ranking ! p1 <
p1 ! p2 < ! p2 , that was used to refute the nave approach. As can be easily checked,
PI ({}, I ) = {I } for all truth definitions (td-pcd1, 1+, 2). So by all these truth definitions,
O( p1 p2 /) is true. So they commit us to violating the second-ranking imperative,
whereas intuitively, the third-ranking imperative should be violated instead. By contrast,
all truth definitions (td-pcd3(+), 4(+) ), that employ reusable output, and of course likewise
(td-pcd2+ ) that is equivalent to (td-pcd4+ ), handle all given examples exactly as it was suggested they should. In particular, consider the ordered imperatives p1! p2 < ! p2 < ! p1 ,
that were used to refute both the reconsidering and the fixpoint approaches: for =
2+, 3(+) , 4(+), PI ({}, I ) is {{ p1 ! p2 , ! p2 }}, making O( p2 /) true by all these truth
definitions, which thus commit us to satisfying the second ranking imperative, and not to
violating it in favor of satisfying the third ranking imperative as these approaches did. The
drinking and driving example is also handled correctly: the set of prioritized imperatives
p1 ! p2 < p3 ! p2 < p3 ! p1 produces, for the situation p3 , the set PI ({ p3 }, I ) =
{{ p1! p2 , p3! p2 }}. So the third ranking imperative, that commits the agent to drinking
and thus, by observation of the highest ranking imperative, prevents the agent from driving,
is disregarded. Instead, the truth definitions make true O( p2 / p3 ), so the agent must do the
driving if she goes to the party, as her best friend asked her to.
Is this the solution, then? Some uneasiness remains as to the quick way with which definitions (td-pcd1, 1+ , 2) are discharged as insufficient. Why should it not be possible to
maintain, as these definitions do, that conditional imperatives only produce an obligation
if they are factually triggered, while at the same time maintaining that the above examples
should not be resolved the way they are? The purpose of a truth definition for the deontic
O-operator is to find a formal notion of ought that reflects ordinary reasoning, and our
intuitions on that matter may differ from our ideas as to what may constitute a good choice
from a possibly conflicting set of prioritized conditional imperatives. I will now make a new
proposal how to construct the preferable subsets, that keeps the positive results without
20 The preferred subsets are thus a choice from the maxfamilies defined in [26].
123
267
committing us to prefer any of the truth definitions of Sect. 3.4 by virtue of their handling of
prioritized imperatives alone.
5.3 Preferred maximally obeyable subsets
What made Brewkas approach so successful is that it maximizes the number of higher ranking imperatives in the preferred subsets of a given set of unconditional imperatives: for each
rank, a maximal number of imperatives are added that can be without making the sets
demands inconsistent in the given situation. As was shown, Brewkas approach cannot be
directly applied to conditional imperatives, since it makes no sense to test the demands of
imperatives for inconsistencies if these imperatives may not be triggered in the same circumstances. Just considering triggered imperatives is also not enough, as was demonstrated for
the nave approach. But if the maximization method is to include imperatives that are not
(yet) triggered, then we must look for something else than inconsistency of demands to take
the role of a threshold criterion for the maximization process.
To do so we should ask ourselves why, for the unconditional case, the aim was to find a
maximal set of imperatives with demands that are collectively consistent with the situation.
I think that by doing so we intend to give the agent directives that can be safely followed.
While in the unconditional case this means that the agent can satisfy all the chosen imperatives, the situation is different for conditional imperatives: here an agent can also obey
imperatives without necessarily satisfying their demands. If you tell me to visit you in case I
go to Luxembourg next month, I can safely arrange to spend all of next month at home and
still do nothing wrong. If we think not so much of imperatives, but of legal regulations, then
I can obviously be a law-abiding citizen by simply failing to trigger any legal norm (even
though this might imply living alone on an island): whether I do that or boldly trigger some
of the regulations antecedents and then satisfy those I have triggered seems not a question of
logic, but of individual choice. So I think the threshold criterion to be used should be that of
obeyability: we should maximize the set of imperatives the agent can obey, and only disregard
an imperative if its addition to the set means that at least one of the added imperatives must
(now) be violated, given the facts.21
For a given set of conditional imperatives and a set of factual truths W , the subsets of
imperatives that can be obeyed are described by ObeyableI (W, ), i.e. they are those subsets
such that W m P L . To maximize not by collective consistency of demands, but
by collective obeyability, Brewkas original approach can therefore be changed as follows:
Definition 8 (Preferred Maximally Obeyable Subsets) Let I = I, f, g, < be a prioritized
conditional imperative structure, be a subset of I , and W L P L be a set of PL-sentences.
Then PIo (W, ) iff (i) W P L , and (ii) is obtained from a full prioritization by
defining
i =
for any i
j {i} if j i j {i} ObeyableI (W, ), and
otherwise,
j i j
j i
, and letting =
i .
The change from Brewkas original definition is only minute: we test not the demands of the
imperatives for consistency, but their materializations. Note that this is a conservative extension of Brewkas method, since for any unconditional imperative i we have P L f (i)
21 While Hansson, in [16] p. 200, also advocates a move from consistency to obeyability, what is meant
there is rather the step from (td-m2) to (td-d1).
123
268
m(i). As can easily be seen, the new construction solves all of the previously considered
difficulties, regardless of the chosen truth definition for the deontic O-operator:
To refute a direct application of Brewkas original method, we used the set I = { p1 p2 ,
p1 p2 } with no ranking imposed. m(I ) is consistent and so PIo ({ p1 }, I ) = {I },
making O( p2 / p1 ) true for all definitions of Sect. 3.4 So you ought to wear your boots in
case you go out, as it should be.
To refute the nave approach, we used the set of prioritized imperatives p1 < p1 p2 <
p2 . Since < is already fully prioritized, the construction produces just one maximally
obeyable subset, which is {! p1 , p1! p2 }, as its two imperatives get added in the first two
steps, and nothing is added in the third since m(I ) is inconsistent. All truth definitions
make true O( p1 /), none makes true the non-intuitive formula O( p1 p2 /), and
definitions that accept deontic detachment make true O( p1 p2 /). So you must go
out and wear a scarf, which is as it should be.
To refute the stepwise approach the ordering of the imperatives was changed into p1 p2 <
p1 < p2 . Still PIo (}, I ) = {{! p1 , p1 ! p2 }}, so the sentences made true by the truth
definitions of Sect. 3.4 likewise do not change, and in particular the non-intuitive formula
O( p1 p2 /) is still false, and definitions that accept deontic detachment make true
O( p1 p2 /), so again you must go out and wear a scarf, which is as it should be.
To refute the reconsidering and the fixpoint approaches the ordering of the imperatives
was again changed into p1 ! p2 < ! p2 < ! p1 . Now PIo (}, I ) = {{ p1 ! p2 , ! p2 }}.
All truth definitions make true O( p2 /) but not O( p1 /) so the agent must satisfy the
second ranking imperative, but not the third ranking imperative, which otherwise would
include violating the highest ranking imperative, which is as it should be.
Troublesome for the fixpoint approach was also the set of prioritized imperatives p1
! p2 < !( p1 p2 ) < ! p3 : no fixpoint could be made out and so the approach produced no
preferred subset, making everything obligatory. The preferred maximally obeyable subset
is { p1! p2 , ! p3 }, eliminating the second ranking imperative that demands a violation of
the first, and making O( p3 /) true for all truth definitions, which again is as it should be.
Finally, consider the drinking and driving example: the set of prioritized imperatives
p1! p2 < p3! p2 < p3! p1 produces, for the situation p3 , the set of preferred maximally obeyable subsets PIo ({ p3 }, I ) = {{ p1! p2 , p3! p2 }}, making true O( p2 / p3 )
for all truth definitions of Sect. 3.4, so given that I go to the party I must do the driving,
which is as it should be.
As could be seen, all truth definitions now produce the right results in the examples
used. Moreover, since all truth definitions refer to the same preferred subsets PIo ({C}, I ), it
is possible to index the O-operators according to the truth definition employed, and e.g. state
truths like O 1 (A/C) O 3 (B/C) O 4 (A B/C), meaning that if, for any maximal set of
imperatives that I can obey in the situation C, imperatives are triggered that demand A, and
that if I satisfy all such triggered imperatives, I will have to do B, then obeying a maximal
number of imperatives includes having to do A B. It may well be that natural language
ought-statements are ambiguous in the face of conditional demands, the discussion in Sect.
3 suggested this. If maximal obeyability is accepted as the threshold criterion that limits what
norms an agent can be expected to conform to in a given situation, then definition 8 leaves
the philosophical logician with maximal freedom as to what deontic operator is chosen.
123
269
6 Theorems
Truth definitions (td-pcd1(+) 4(+) ) define when a sentence of the form O(A/C) is true
or false with respect to a prioritized conditional structure I and a situation C. So I briefly
consider what sentences are theorems, i.e. hold for all such structures, given the usual truth
definitions for Boolean operators. It is immediate that for all truth definitions, (DExt), (DM),
(DC), (DN) and (DD-R) are theorems (cf. Sect. 3.1). (DD-R) states that there cannot be both
an obligation to bring about A and one to bring about A unless the situation C is logically
impossible, so our truth definitions succeed in eliminating conflicts. All these theorems are
monadic in the sense that C is kept fixed; in fact, they are the C-relative equivalents of
standard deontic logic SDL. More interesting are theorems (known from the study of nonmonotonic reasoning) that describe relations between obligations in different circumstances.
Obviously we have
(ExtC)
If P L C D then O(A/C) O(A/D) is a theorem
for all truth definitions, i.e. for equivalent situations C, the obligations do not change. As
long as truth definitions are not sensitive to conflicts, e.g. for (td cd (+) 4(+) ), we have
strengthening of the antecedent, i.e. for these definitions
(SA) O(A/C) O(A/C D)
holds. When only maximally obeyable subsets are considered, i.e. for truth definitions (tdpcd1(+) 4(+) ), both (SA) and the weaker rational monotonicity theorem
(RM) O(D/C) O(A/C) O(A/C D)
are refuted e.g. by a set I = {!( p1 p2 ), !( p1 p2 ), p2 p1 } of equally ranking imperatives: though O( p1 /) is true and O( p2 /) false, O( p1 / p2 ) is false. However, for all
definitions of Sect. 3.4, (conjunctive) cautious monotonicity
(CCMon) O(A D/C) O(A/C D)
holds, which states that if you should to two things and you do one of them, you still have
the other one left.22 Moreover, truth definitions (td-pcd1+ , 2+ , 3+ , 4+ ) validate the circumstantial extensionality rule
(CExt)
If P L C (A B) then O(A/C) O(B/C) is a theorem
that corresponds to circumstantial reasoning. All definitions that accept reasoning by

cases, i.e. (td-pcd2, 2+ , 4, 4+ ), make
(Or ) O(A/C) O(A/D) O(A/C D)
a theorem. Note that (CExt) and (Or) derive
(Cond) O(A/C D) O(D A/C),
which in turn derives (Or) in the presence of (DC), and that by adding (CExt) and (Or) we
obtain again the system PD (cf. Sect. 3). Finally, all definitions with deontic detachment,
i.e. (td-pcd3, 3+ , 4, 4+ ), make
(Cut) O(A/C D) O(D/C) O(A/C)
22 This is Hanssons [15] theorem (19).
123
270
a theorem. (Cut) is derivable given (Cond) (use Cond on the first conjunct O(A/C D) to
obtain O(D A/C), agglomerate with O(D/C), and from O(D (D A)/C) derive
O(A/C)), which syntactically mirrors the semantic equivalence of definitions (td-pcd2+ )
and (td-pcd4+ ). Theoremhood of all of the above for semantics that employ the respective
truth definitions is easily proved and left to the reader (cf. Hensen [13] and [14] as well as
Makinson and van der Torre [25] for the general outline). Makinson and van der Torres results
also suggest that these theorems axiomatically define complete systems of deontic logic with
respect to semantics that employ the respective truth definitions (td-pcd1(+) 4(+) ), but this
remains a conjecture that further study must corroborate.23
7 Back to the beginning: questions of representation

One might wonder if it is always adequate to represent a natural language conditional imperative if then bring about that ww by use of a set I containing an imperative i with a
g(i) that formalizes and a f (i) that formalizes ww. This is because there is a second
possibility: represent the natural language conditional imperative by an unconditional imperative !(g(i) f (i)). We saw in Sect. 3 that this is not generally adequate. But that does
not mean that such a representation is not sometimes what is required. Consider the crucial
imperatives in the previous examples: perhaps what your mother meant was simply dont
drink and drive; perhaps what the doctor meant was dont go out without a scarf; perhaps
the Colonel meant to tell OReilly not to do both, turn the heat on and keep the window
closed; perhaps self-defense required me to see to it that I am not attacked without fighting
back. These interpretations seem not wholly unreasonable, and if they are adequate, then the
best representation would be by an imperative !(g(i) f (i)) instead of g(i) ! f (i).
What then are the conditions that make a representation by an unconditional imperative
adequate? One test may be to ask: Would bringing about the absence of the antecedent
condition count as satisfaction of the imperative? Would not drinking, not going out, not
turning on the heat, making the man not attack, count as properly reacting to the imperatives
in question? It should be if what the imperatives demand is a material conditional, since then
the conditional imperatives in question are equivalent to telling the agent either dont drink
or dont drive, its your decision, either dont go out, or wear a scarf, either dont turn on
the heat, or open the window, etc. Another test would be to examine if contraposition is
acceptable. Can we say that your mother wanted you not to drink if you are going to drive,
that the doctor wanted you to stay inside if you are not going to wear a scarf, that the Colonel
wanted OReilly to turn off the heat if the window is closed, that self-defense requires you
to make the man not attack if you are not going to fight back? If the proper representation is
by imperatives that demand a material conditional, then the answers should be affirmative. I
do not think these are easy questions, however, and leave them to the reader to discuss and
answer at his or her own discretion. But it is easy to see that, had we chosen to represent the
crucial imperatives in the above examples (including the drinking and driving problem) by
unconditional imperatives that demand a material implication to be realized, then all of the
discussed methods would have resolved these examples.
23 For (td-pcd2+, 4+ ), completeness of PD is immediate from the results in [13], [14].
123
271
8 Conclusion
Reasoning about obligations when faced with different and possibly conflicting imperatives
is a part of everyday life. To avoid conflicts, these imperatives may be ordered by priority
and then observed according to their respective ranks. The drinking and driving case in the
introduction presented an example of such natural reasoning. Providing a formal account is,
however, additionally complicated by the fact that there are various and mutually exclusive
intuitions about what belongs to the right definition of an obligation in the face of conditional
imperatives, i.e. the definition of a deontic O-operator. Based on similar definitions of operators by Makinson and van der Torre [25], [26] for their input/output logic, but leaving
the choice of the right operator to the reader, I presented several proposals in Sect. 3 for
definitions of a dyadic O-operator, namely (td-pcd1(+) 4(+) ). These were dependent on
a choice of preferred subsets among a given set of prioritized conditional imperatives. A
particularly successful method to identify such subsets, but applying to unconditional imperatives only, was Brewkas [3] definition of preferred subtheories within a theory. In Sect. 4
I discussed various approaches that extend this method to conditional imperatives, but these
failed to produce satisfactory results for a number of given examples. In Sect. 5 I proposed
that the maximality criterion used to construct the preferred subsets should be the avoidance
of conflict. A recent approach to this task by consistent extensions was found to be biased
towards definitions of obligation that do not accept reasoning by cases; another, that aims
to avoid the truth of O(C/C) for possible circumstances C, produced satisfactory results
only for truth definitions that accept deontic detachment. I then argued that the solution is
to adapt Brewkas method in such a way that it constructs, instead of maximal subsets of
imperatives that are collectively satisfiable by an agent, maximally obeyable subsets of the
imperatives, in the sense that the facts do not derive that some imperative of the set must be
violated. I showed that this new proposal provides adequate solutions to all of the examples,
and in particular the drinking and driving example is resolved in a satisfactory fashion for
all of the discussed deontic operators.
Acknowledgments I am grateful to Lou Goble, John F. Horty, David Makinson and Leon van der Torre
for helpful comments and discussions. Also, I thank four anonymous referees for pointed remarks and useful
suggestions. A preliminary version of this paper was presented to the participants in the seminar Normative
Multi-agent Systems (NorMAS07) held in Dagstuhl, Germany, from 18 to 23 March 2007.
References
1. Alchourrn, C. E., & Bulygin, E. (1981). The expressive conception of norms. In [18], 95124.
2. Alchourrn, C. E., & Makinson, D. (1981). Hierarchies of regulations and their logic. In R. Hilpinen
(Ed.), New Studies in Deontic Logic (pp. 125148). Dordrecht: Reidel.
3. Brewka, G. (1989). Preferred subtheories: An extended logical framework for default reasoning. In
N. S. Sridharan (Ed.), Proceedings of the eleventh international joint conference on artificial intelligence
IJCAI-89, Detroit, Michigan, USA, August 2025, 1989 (pp. 10431048). San Mateo, Calif.: Kaufmann.
4. Brewka, G. (1994). Reasoning about priorities in default logic. In B. Hayes-Roth & R. E. Korf (Eds.),
Proceedings of the 12th national conference on artificial intelligence, Seattle, WA, July 31stAugust 4th,
1994 (Vol. 2 pp. 940945). Menlo Park: AAAI Press.
5. Brewka, G., & Eiter, T. (1999). Preferred answer sets for extended logic programs. Artificial Intelligence,
109, 297356.
6. Broersen, J., Dastani, M., & van der Torre, L. (2002). Realistic desires. Journal of Applied Non-Classical
Logics, 12, 287308.
7. Caminada, M., & Sakama, C. (2006). On the existence of answer sets in normal extended logic programs.
In G. Brewka, S. Coradeschi, A. Perini, & P. Traverso (Eds.), Proceedings of the 17th European con-
123
272
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
ference on artificial intelligence (ECAI 2006), August 29September 1, 2006, Riva del Garda, Italy (pp.
741742). Amsterdam: IOS Press.
Downing, P. (1959). Opposite conditionals and deontic logic. Mind, 63, 491502.
van Fraassen, B. (1973). Values and the hearts command. Journal of Philosophy, 70, 519.
Goble, L. (2005). A logic for deontic dilemmas. Journal of Applied Logic, 3, 461483.
Greenspan, P. (1975). Conditional oughts and hypothetical imperatives. Journal of Philosophy, 72, 259
276.
Hansen, J. (2004). Problems and results for logics about imperatives. Journal of Applied Logic, 2, 3961.
Hansen, J. (2005). Conflicting imperatives and dyadic deontic logic. Journal of Applied Logic, 3, 484511.
Hansen, J. (2006). Deontic logics for prioritized imperatives. Artificial Intelligence and Law, 14, 134.
Hansson, B. (1969). An analysis of some deontic logics. Nus, 3, 373398. Reprinted in [17], 121147.
Hansson, S. O. (2001). The structure of values and norms. Cambridge: Cambridge University Press.
Hilpinen, R. (Ed.) (1971). Deontic logic: Introductory and systematic readings. Dordrecht: Reidel.
Hilpinen, R. (Ed.) (1981). New studies in deontic logic. Dordrecht: Reidel.
Hofstadter, A., & McKinsey, J. C. C. (1938). On the logic of imperatives. Philosophy of Science, 6,
446457.
Horty, J. F. (2003). Reasoning with moral conflicts. Nos, 37, 557605.
Horty, J. F. (2007). Defaults with priorities. Journal of Philosophical Logic, 36, 367413.
Kraus, S., Lehmann, D., & Magidor, M. (1990). Nonmonotonic reasoning, preferential models and cumulative logics. Artificial Intelligence, 44, 167207.
Lewis, D. (1974). Semantic analysis for dyadic deontic logics. In S. Stenlund (Ed.), Logical theory and
semantic analysis (pp. 114). Dordrecht: Reidel.
Makinson, D. (1999). On a fundamental problem of deontic logic. In P. McNamara & H. Prakken (Eds.),
Norms, logics and information systems (pp. 2953). Amsterdam: IOS.
Makinson, D., & van der Torre, L. (2000). Input/output logics. Journal of Philosophical Logic, 29, 383
408.
Makinson, D., & van der Torre, L. (2001). Constraints for Input/output logics. Journal of Philosophical
Logic, 30, 155185.
Marek, V. W., & Truszczynski, M. (1993). Nonmonotonic logic context-dependent reasoning. Berlin:
Springer.
Nebel, B. (1991). Belief revision and default reasoning: Syntax-based approaches. In J. A. Allen, R.
Fikes & E. Sandewall (Eds.), Principles of knowledge representation and reasoning: Proceedings of the
second international conference, KR 91, Cambridge, MA, April 1991 (pp. 417428). San Mateo: Morgan
Kaufmann.
Nebel, B. (1992). Syntax-based approaches to belief revision. In: P. Grdenfors (Ed.), Belief revision (pp.
5288). Cambridge: Cambridge University Press.
Prakken, H. (1997). Logical tools for modelling legal argument. Dordrecht: Kluwer.
Prakken, H., & Sartor, G. (1997). Argument-based logic programming with defeasible priorities. Journal
of Applied Non-classical Logics, 7, 2575.
Rescher, N. (1964). Hypothetical reasoning. Amsterdam: North-Holland.
Rescher, N. (1966). The Logic of commands. London: Routledge & Kegan Paul.
Rintanen, J. (1994). Prioritized autoepistemic logic. In: C. MacNish, D. Pearce, & L. M. Pereira (Eds.),
Logics in artificial intelligence, European Workshop, JELIA 94, York, September 1994, Proceedings (pp.
232246). Berlin: Springer.
Ross, W. D. (1930). The right and the good. Oxford: Clarendon Press.
Ryan, M. (1992). Representing defaults as sentences with reduced priority. In B. Nebel, C. Rich, &
W. Swartout (Eds.), Principles of knowledge representation and reasoning: Proceedings of the third
international conference, KR 92, Cambridge, MA, October 1992. (pp. 649660). San Mateo: Morgan:
Kaufmann.
Sakama, C., & Inoue, K. (1996). Representing priorities in logic programs. In: M. Maher (Ed.), Joint
international conference and symposium on logic programming JICSLP 1996, Bonn, September 1996
(pp. 8296). Cambridge: MIT Press.
Sosa, E. (1996). The logic of imperatives. Theoria, 32, 224235.
Swirydowicz,
K. (1994). Normative consequence relation and consequence operations on the language
of dyadic deontic logic. Theoria, 60, 2747.
von Wright, G. H. (1968). An essay in deontic logic and the general theory of action. Amsterdam: North
Holland.
von Wright, G. H. (1984). Bedingungsnormen, ein Prfstein fr die Normenlogik. In W. Krawietz,
H. Schelsky, O. Weinberger, & G. Winkler (Eds.), Theorie der Normen (pp. 447456). Berlin: Duncker & Humblot.
123
273

42. Watts, I. (1725). Logick: or, the right use of reason in the inquiry after truth, with a variety of rules to
guard against error in the affairs of religion and human life, as well as in the sciences. London: for John
Clark, Richard Hett et al.
123
Zusammenfassung
zur Dissertation
Imperatives and Deontic Logic:
On the Semantic Foundations of Deontic Logic
eingereicht an der Fakultat f
ur Sozialwissenschaften und Philosophie der Universitat Leipzig
von Herrn M. A. Jorg Hansen
angefertigt am Institut f
ur Philosophie
offentlich verteidigt am 18. November 2008
Im Jahr 1671 wies Gottfried Wilhelm Leibniz auf die Ahnlichkeit

der juristi
schen (deontischen) Modalitaten des Gebotenen, Erlaubten und Verbotenen
mit den nat
urlichen (alethischen) Modalitaten des Notwendigen, Mogli
chen und Unmoglichen hin und stellte die These auf, da alle bekannten
Komplikationen, Transpositionen und Oppositionen zwischen diesen auf
jene u
bertragen werden konnen. Mit seinem Aufsatz Deontic Logic stellte
Georg Henrik von Wright 1951 den ersten Modalkalk

ul vor, der dieser Behauptung Rechnung tragt. Sein zum Standardsystem der deontischen Logik
weiterentwickeltes Axiomensystem ist ein normaler Modalkalk
ul, in welchem
freilich das zusatzliche Axiom (T) A A der alethischen Modallogik nicht
gilt, sondern stattdessen das Axiom (D) OA P A: Was geboten ist, ist
274
ANHANG: ZUSAMMENFASSUNG
275
anders als das Notwendige zwar nicht zwingend auch der Fall, aber zumindest ist das, was geboten ist, zugleich auch erlaubt. (D) ist in diesem System
aquivalent zu (OA OA): Wenn A geboten ist, kann nicht zugleich non-A
geboten sein. Eine interessante weitere Beziehung zwischen der deontischen
Logik und der alethischen Modallogik wurde 1957 von Alan Ross Anderson
aufgezeigt: Die deontische Logik kann auf die alethische Modallogik reduziert
werden, wenn OA als (A S) definiert wird A ist geboten, wenn non-A
notwendig S nach sich zieht, wobei die Konstante S f
ur das Eintreten einer
Sanktion steht. Unterstellt man, da S moglich ist, also das Eintreten einer
Sanktion vermieden werden kann, so wird das Standardsystem der deontischen Logik mittels einer solchen Definition zu einem Fragment des (um die
Konstante S bereicherten) alethischen Modalkalk
uls.
Wie die Ausdr
ucke der alethischen Modallogik werden auch die Ausdr
ucke der deontischen Logik u
berwiegend mithilfe einer Mogliche-Welten
Semantik interpretiert: OA, der mit A beschriebene Zustand ist geboten,
ist wahr, wenn A in allen Idealwelten zu unserer aktuellen Welt wahr ist. Mithilfe einer logischen Semantik, deren Modelle eine Menge moglicher Welten,
eine zwischen ihnen bestehenden Idealitatsrelation, sowie eine Wahrheitsfunktion angeben, die den Zustand in jeder Welt beschreibt, kann sodann
die G
ultigkeit aller Satze und die Vollstandigkeit des Standardsystem der
deontischen Logik und seiner Abwandlungen bewiesen werden.
Die deontische Logik wird bis heute gelegentlich als eine Logik der Nor
men charakterisiert. Diese Darstellung ist freilich falsch, denn Normen konnen weder wahr noch falsch sein, sie wollen die Welt nicht beschreiben, sondern etwas in ihr bewirken. Nur wahre oder falsche Satze konnen aber Einsetzunginstanzen f
ur OA und P A etwa in dem obigen Schema (D) sein, denn
die durch den Pfeil ausgedr
uckte wenn ... dann Beziehung ist wahr
heitsfunktional: Der Satz ist falsch genau dann wenn A geboten ist, aber A
gleichwohl nicht erlaubt ist (weil z.B. A ebenfalls geboten ist). Wenn OA
deshalb keine Norm sein kann, sondern ein beschreibender Satz ist, so gelten
die normalen Regeln: OA, also A ist geboten, ist genau dann wahr wenn
A geboten ist, wenn es also eine Norm gibt, die A gebietet. Die deontische
Logik versteht daher ihre Ausdr
ucke u
blicherweise als wahre oder falsche Behauptungen dar
uber, was gema einem (meist nicht naher beschriebenen)
Normsystem geboten, verboten oder erlaubt ist.
Wenn sich aber die Wahrheit der Satze der deontischen Logik nach dem
richtet, was gema tatsachlich bestehender Normen als geboten, verboten
oder erlaubt anzusehen ist, mu sie die Eigenheiten bestehender Normsy-
276
steme abbilden konnen. Hierzu gehort zunachst die Begleitung von Primarnormen durch Sekundarnormen (oder contrary-to-duty imperatives), die
Pflichten f
ur solche Falle statuieren, in denen die Primarnormen verletzt
werden. Solche Sekundarpflichten sind etwa: sich entschuldigen, den Schaden wiedergutmachen, eine Strafe zahlen usw. Hierzu gehort aber auch die
Moglichkeit eines Normenkonflikts: Es mag, durch einen Fehler des Gesetzgebers oder den Irrtum eines Richters, zwei oder mehr Normen geben, die
Gegensatzliches verlangen. Haufig gibt es aber in Fallen von Normenkonflikten zugleich Losungen mithilfe von Prioritatsregelungen: die Norm des Verfassungsrechts hat Vorrang vor der des normalen Gesetzesrecht, die spatere
Norm vor der fr
uheren, die speziellere vor der allgemeinen, diejenige, die bedeutende Rechtsg
uter sch
utzt, hat Vorrang vor bloen Ordnungsregeln, usw.
Sowohl axiomatische wie semantische Ansatze der traditionellen deontischen Logik geraten bei der Bewaltigung dieser Besonderheiten in Schwierigkeiten. Das Standardsystem der deontischen Logik hat sich zur Beschreibung
von Sekundarnormen als unbrauchbar erwiesen und wurde durch Systeme
mit zweistelligen deontischen Operatoren ersetzt: In der sogenannten dyadischen deontischen Logik bedeutet O(A/B): A ist geboten in der Situation
B. In der logischen Semantik wurden die Idealwelten durch eine Praferenzrelation zwischen moglichen Welten ersetzt: O(A/B) ist nun wahr, wenn A
in den besten aller moglichen B-Welten der Fall ist. Noch schwieriger ist
die Darstellung von Normenkonflikten: wenn es moglich ist, da sowohl OA
wie auch OA wahr sind, ist (D) als Axiom widerlegt. Aber auch die u
brigen Axiome des Standardsystems m
ussen modifiziert werden, damit aus OA
und OA nicht OB folgt (sogenannte deontische Explosion). Semantisch
reichen die Abhilfevorschlage vom Einsatz minimaler Modelle (sog. Nachbarschaftsemantiken, bei denen nicht eine, sonderen mehrere Mengen von
Welten gegen
uber einer anderen ideal sind) bis zu Multipraferenzseman
tiken, bei denen nicht eine, sondern mehrere Praferenzrelationen die jeweils
besten aus einer gegebenen Menge von Welten auswahlen. Noch schwieriger ist die Rekonstruktion von Konfliktlosungsstrategien mittels Prioritaten:
Hier m
uten konsequenterweise Praferenzrelationen zwischen Mengen idealer
Welten ( beste Nachbarschaften) oder zwischen anderen Praferenzen zum
Einsatz kommen, selbst wenn dies meines Wissens noch nirgendwo versucht
worden ist. Alle diese Semantiken haben jedenfalls den Nachteil, da sie nicht
sonderlich intuitiv sind.
In der Behandlung von Beispielen hat es sich als besonders unvorteilhaft
erwiesen, da es in der Sprache der deontischen Logik unmoglich ist, unmittel-
277
bar auf bestimmte Normen Bezug zu nehmen. So wurde etwa vorgeschlagen,

in dem Fall, da sowohl O(A B) als auch O(B C) wahr sind, zwar keine
Kumulierung der Gebotsinhalte zuzulassen, denn O(A B B C) ware
unerf
ullbar, wohl aber die disjunktive Erf
ullung beider Gebote zu fordern,
also O((A B) (B C)). Damit w
urde der Normadressat im Falle eines
Konflikts zur Erf
ullung wenigstens eines Gebots angehalten. Doch die Satze
O(A B) und O(B C) m
ussen nicht deshalb wahr sein, weil zwei bestimmte Normen genau A B und B C fordern: Vielleicht sind die Satze
deshalb wahr, weil zwei Normen A B C und A B C fordern. Dann
stellt O((A B) (B C)) allein das widersinnige Gebot dar, wenigstens
eine Norm zu verletzen. Ebenso ist angedacht worden, unter Ber
ucksichtigung von Normenkonflikten eine Kumulation zweier Gebote OA und OB zu
O(A B) gleichwohl dann zuzulassen, wenn A B konsistent ist (sogenannte
konsistente Agglomeration). Doch OA und OB mogen deshalb wahr sein,
weil zwei Normen AC und B C fordern es macht dann keinen Sinn, von
den Normadressaten sowohl die Realisierung von A als auch B zu fordern,
wenn doch nur eine dieser beiden Normen erf
ullt werden kann: Das Tun von
entweder A oder B ware u
ussig.
berfl
In dieser Situation ist es kein Wunder, wenn sogar in Standardwerken zur
deontischen Logik aus Gebots- und Erlaubnissatzen klammheimlich Imperative oder deontische Normen werden, etwa gesagt wird, da A die deon
tische Norm OA verletze. Da dies unzulassig ist, habe ich oben festgestellt.
Und so war auch meine eigene Ausdrucksweise im vorigen Absatz irref
uhrend,
denn OA ist kein Gebot, sondern ein Satz, da etwas Bestimmtes geboten
ist, er kann nicht verletzt oder erf
ullt, sondern nur wahr oder falsch sein;
er ist bestenfalls eine Pramisse. David Makinson sah sich deshalb wahrend
des Workshops EON98 in Bologna 1998 zu der Feststellung veranlat, die
Arbeiten zur deontischen Logik erweckten vielerorts den Eindruck, als habe
der Autor noch nie von einer Unterscheidung zwischen Normen und deontischen Satzen u
ber Normen gehort. Die deontische Logik sei in Einzelteile
zerfallen: in zahlreiche Axiomatisierungen, und in noch mehr unterschiedliche Mogliche-Welten-Semantiken mit ihren unzahligen Detailabweichungen.
Er forderte dazu auf, die deontische Logik zu rekonstruieren : als eine Lo
gik, die mit Normen befat ist, aber respektiert, da Normen weder wahr
noch falsch sein konnen.
Diesem Ruf folgend habe ich seither in einer Reihe von Forschungsarbeiten
untersucht, wie sich mithilfe einer imperativistischen Semantik die deonti
schen Operatoren neu definieren lassen, und welche alten oder neuen Systeme
278
der deontischen Logik hierzu vollstandig alle g

ultigen Satze enthalten. Das
Grundkonzept ist denkbar einfach: Sei I eine Menge (von Imperativen) und
sei f eine Funktion, die jedem Objekt dieser Menge (jedem Imperativ) einen
Satz aus einer formalisierten Sprache (etwa der Aussagenlogik) zuordnet. Die
Vorstellung ist, da dieser Satz beschreibt, was der Fall ist, wenn der Imperativ erf
ullt ist, und was nicht der Fall ist, wenn der Imperativ verletzt ist. Da
es f
ur jeden Imperativ einen solchen Satz geben mu, ist erkennbar unstreitig,
denn anderenfalls konnte der Adressat des Imperativs nicht verstehen, was
von ihm verlangt wird. Wir konnen nun z.B. Operatoren des Typs OA definieren, die wahr sind, wenn A genau einen Imperativ erf
ullt, wenn A mehrere
Imperative erf
ullt, wenn A notwendig zur Erf
ullung eines Imperativs, mehrerer Imperative oder aller Imperative ist. Wir konnen in einer bestimmten
Situation dasjenige als geboten beschreiben, was zur Erf
ullung aller Imperative erforderlich ist, die in dieser Situation noch erf
ullt werden konnen; wir
konnen Konflikte zwischen den Imperativen zulassen und fragen, was notwendig ist, um eine oder alle maximalen Mengen von nicht miteinander in
Konflikt stehenden Imperative zu erf
ullen; wir konnen Prioritatsbeziehungen zwischen den Imperativen darstellen und untersuchen, auf welche Weise
diese zur Losung mancher oder aller Konflikte beitragen konnen. Schlielich
konnen wir bedingte Imperative abbilden, indem wir etwa eine zweite Funktion g jedem Imperativ einen Ausloser zuordnen lassen, einen Satz, der
beschreibt, unter welcher Bedingung der Imperativ zu erf

ullen ist (und verletzt werden kann). O(A/B) mag dann etwa in einer bestimmten Situation B
dasjenige als geboten beschreiben, was zur Erf
ullung aller in dieser Situation
ausgeloster Imperative erforderlich ist.
Es erweist sich sodann, da das Standardsystem der deontischen Logik

einer Definition entspricht, welche dasjenige als geboten beschreibt, was zur
Erf
ullung eines oder mehrerer Imperative erforderlich ist, wenn zugleich unterstellt wird, da die Menge der Imperative nicht leer ist und die Imperative
nicht miteinander in Konflikt stehen. Die beiden Hauptsysteme der dyadischen deontischen Logik entsprechen solchen Definitionen, die dasjenige in
einer bestimmten Situation als geboten beschreiben, was zur Erf
ullung aller
verbliebenen, in dieser Situation noch nicht verletzten Imperative erforderlich ist, wenn wir unterstellen, da eine totale Prioritatsrelation etwa nach
einer von Gerhard Brewka entwickelten Methode alle f
ur diese Situation entstehenden Normenkonflikte auflost. Auch die Andersonsche Reduktion der
deontischen Logik auf die alethische Modallogik lat sich mittels der neuen
Semantik rekonstruieren: Seiner Sanktionskonstante S entspricht dabei die
279
Feststellung, da mindestens ein Imperativ verletzt ist. Das Projekt einer

Rekonstruktion der traditionellen deontischen Logik ist damit abgeschlossen. Zugleich treten neben die traditionellen Systeme jedoch weitere. Hierzu
gehoren etwa solche, deren zugehorige Semantik nicht die Auflosung aller
etwaigen Normenkonflikte verlangt, diese entsprechen im wesentlichen nichtmonotonen Logiken, wie sie etwa von Kraus, Lehmann und Magidor oder
von Alexander Bochman entworfen wurden.
Meine Arbeit beschaftigt sich zunachst mit der Frage, ob es nicht eine
Logik der Imperative gibt. Gabe es sie, w
urde eine Rekonstruktion der
deontischen Logik als Logik von Satzen dar

uber, was gema einer Menge von
angenommenen Imperativen geboten und erlaubt ist, denkbar einfach sein:
Wir konnten eine solche Menge von Imperativen nach deren eigenen logischen
Gesetzmaigkeiten abschlieen, und die Satze der deontischen Logik w
urden
dann nur noch spiegelbildlich beschreiben, welche Imperative in einem derart
abgeschlossenen System existieren oder nicht existieren. Die Diskussion in
Kapitel 1 ergibt jedoch, da die Existenz unmittelbarer logischer Beziehungen zwischen Imperativen nicht unterstellt werden kann. Hierf
ur liefere ich ein
Argument in der Tradition der Ordinary Language Philosophy, und damit
ist zugleich klargestellt, da mein Ausgangspunkt die Position eines Impe

rativologischen Skeptizismus ist. Im Kapitel 2 erlautere ich die spater zu rekonstruierenden traditionellen Systeme der deontischen Logik und beschreibe dann die Schwierigkeiten, die zu Makinsons Aufforderung zu ihrer normenbezogenen Rekonstruktion gef
uhrt haben. Kapitel 3 wendet sich sodann
der imperativistischen oder normativen Tradition der deontischen Logik
zu, denn einige Autoren haben bereits in Abweichung von der herkommlichen Mogliche-Welten-Semantik versucht, die deontischen Operatoren mittels explizit angegebener Normsysteme oder Imperativmengen zu definieren.
Sodann erlautere ich die zur Rekonstruktion der traditionellen deontischen
Systeme verwendete Semantik und beschreibe zwei wichtige Grundvoraussetzungen: die Unabhangigkeit und die Untrennbarkeit der in der logischen
Semantik modellierten Imperative. Die oben bereits kurz erlauterten Ergebnisse meiner Forschungsarbeiten zu dieser Semantik werden anschlieend zusammengefat. Abschlieend wende ich mich den sogenannten Paradoxien
der deontischen Logik zu und untersuche, wie der neue semantische Ansatz
mit diesen Paradoxien umgehen kann.
In gewisser Weise bedeutet eine imperativistische Semantik einen Pa
radigmenwechsel f
ur die deontische Logik. Verstand sich die deontische Logik
zuvor, etwa im Spatwerk Georg Henrik von Wrights, als Ratgeber eines ra-
280
tionalen Gesetzgebers, so steht im Vordergrund der vorgeschlagenen Semantik f

ur die deontische Logik das rationale Normsubjekt: Wie lat sich mit
Normen umgehen, wenn diese Normenkonflikte enthalten konnen, wenn wir
in vom Gesetzgeber unvorhergesehene dilemmatische Situationen geraten,
wenn das schrittweise entwickelte Normsystem Vorschriften enthalt, die ein
einziger rationaler Gesetzgeber moglicherweise nicht erlassen hatte? Konnen
Konflikte isoliert werden, lat sich sagen, was gleichwohl noch geboten bleibt,
und wie sehen die Mechanismen aus, mit denen rationale Normsubjekte Normenkonflikte mittels Prioritatsbeziehungen tatsachlich losen? Wenn man so
will, kann man dies auch als positivistische deontische Logik bezeichnen:
Eine Rationalitat des Gesetzgebers wird nicht vorausgesetzt, wohl aber ein
u
berlegendes Normsubjekt. Neben solchen grundsatzlichen Erwagungen zu
unserem Projekt wird im Schluteil auf fortbestehende Schwierigkeiten hingewiesen. Gebotsausdr
ucke lassen sich in Bezug auf konditionale Imperative in
verschiedener Weise definieren, konnen z.B. die sogenannte Wahlfeststellung
zulassen (wenn A B der Fall ist, und A und B beide Imperative auslosen,
die nur zu erf

ullen sind, wenn C wahr ist, ist dann bereits C geboten?), oder
einen normativen Kettenschlu (wenn A der Fall ist, und einen Imperativ
auslost, der seinerseits zu seiner Erf
ullung die Wahrheit von B erfordert,
ist dann auch geboten, was ein Imperativ verlangt, der durch B ausgelost
wird?), und schlielich umstandebezogenes Argumentieren zulassen (wenn

A B der Fall ist, und A einen Imperativ auslost, zu dessen Erf
ullung
B C erforderlich ist, ist dann C geboten?). Diese Argumentationsweisen konnen jedoch nur in bestimmten Kombinationen zugelassen werden,
und f
uhren in anderen Kombinationen zu Paradoxien: Welches ist also die
richtige Gebotsdefinition f
ur konditionale Imperative? Erlaubende Normen
sind bislang vernachlassigt worden: Sie sollten sinnvollerweise nicht einfach

neben Imperative gestellt werden, sondern das bislang Gebotene so modi
fizieren, da etwas moglicherweise vormals Verbotenes nun als erlaubt gilt.
Dies und anderes, wie etwa die Reprasentation genereller Normen und von
Normen, die die Erzeugung von Normen regeln, stellen weitere Herausforderungen f
ur die Entwicklung einer imperativ- bzw. normbezogenen Semantik
f
ur die deontische Logik dar.
Bibliographische Beschreibung
Hansen, Jorg
Imperatives and Deontic Logic: On the Semantic Foundations of
Deontic Logic
Universitat Leipzig, Diss., 120 S., 260 Lit.
Referat:
Die Satze der deontischen Logik werden traditionell mithilfe der Mogliche
Welten-Semantik interpretiert. Andererseits versteht die deontische Logik,
in Abgrenzung von einer Normenlogik, ihre Satze als wahre oder falsche
Aussagen dar
uber, was gema einem meist nicht naher beschriebenen System von Normen geboten, verboten oder erlaubt ist. Dieser Erklarung ihres Gegenstands folgend haben einige Autoren versucht, die Wahrheit oder
Falschheit der Satze der deontischen Logik nicht mit R
ucksicht auf mogliche
Welten zu interpretieren, sondern mittels einer explizit angegebenen Normmenge oder Menge von Imperativen, und was zu ihrer Erf
ullung erforderlich
ist. Es kann dann gezeigt werden, da sich im Sinne dieser imperativisti
schen Tradition der deontischen Logik die traditionellen axiomatischen Systeme der monadischen und dyadischen Logik rekonstruieren lassen. Zu
gleich ermoglicht dieser logisch-semantische Ansatz die Konstruktion anderer Systeme, die etwa Konflikte zwischen Imperativen zulassen oder solche
Konflikte mittels Vorrangbeziehungen zwischen den Imperativen ganz oder
teilweise losen. Die Arbeit erlautert die Grundidee dieser imperativistischen
Semantik, ihre Anwendungen sowie die bereits erzielten Resultate.
281
Lebenslauf
09.09.1967
01/1985 12/1985
21.06.1988
10/1988 03/1991
05/1989 03/1991
04/1991 06/1997
05/1991 09/1993
08/1992 10/1999
05.08.1994
04/1995 09/1995
10/1995 03/1996
02/1997 06/1999
02.06.1997
04.06.1999
14.09.1999
09/1999
10/1999
Geburt in Cochem an der Mosel

Bowen State High School, Bowen, N.Q., Australien
Abitur
Studium an der Universitat Bielefeld: Rechtswissenschaft (Staatsexamen), Philosophie (Magister).
Hilfskraft am Lehrst. f. Offentliches

Recht (Dr. Werner Heun) u. f. Europarecht (Prof. Dr. Meinhard Hilf)
Studium an der Universitat Hamburg: Rechtswissenschaft (Staatsexamen), Philosophie (Magister).
Hilfskraft am Seminar f
ur Europarecht (Prof. Dr.
Meinhard Hilf)
Mitarbeiter der Verbraucher-Zentrale Hamburg e.V.
Erste Juristische Staatspr
ufung
Tutor f
ur Logik am Philosophischen Seminar (Dr. Ali
Behboud)
Tutor f
ur Logik am Philosophischen Seminar (Prof.
Dr. Werner Diederich)
Rechtsreferendar am HansOLG
Magister Artium der Universitat Hamburg, Fachbereich Philosophie und Sozialwissenschaften
Zweite Juristische Staatspr
ufung
Zulassung als Rechtsanwalt
Rechtsanwalt, Kanzlei Hansen & Varwig, Eisenach
Promotionsstudium an der Universitat Leipzig
282
ANHANG: LEBENSLAUF
10/2000 03/2001
283
Hilfskraft am Institut f
ur Logik und Wissenschaftstheorie (Dr. Peter Steinacker)
07/2001 12/2001 Geschaftsf
uhrer Bachhaus Eisenach gGmbH
12/2005
Geschaftsf. u. Direktor Bachhaus Eisenach gGmbH
21.11.2007
Fachanwalt f. Arbeitsrecht
Publikationen
Paradoxes of Commitment, in: Meggle, G. (Hrsg.), Actions, Norms, Values.
Discussions with Georg Henrik von Wright, De Gruyter: Berlin, New York,
1999, 255263.
On Relations between
Aqvists Deontic System G and Van Ecks Deontic
Temporal Logic, in: McNamara, P. und Prakken, P. (Hrsg.), Norms, Logics

and Information Systems, IOS Press: Amsterdam, 1999, 127144.
Sets, Sentences, and Some Logics about Imperatives,Fundamenta Informa
ticae, 48, 2001, 205226.
Problems and Results for Logics about Imperatives, Journal of Applied
Logic, 2, 2004, 39-61.

Conflicting Imperatives and Dyadic Deontic Logic, Journal of Applied Lo
gic, 3, 2005, 484511.
Deontic Logics for Prioritized Imperatives, Artificial Intelligence and Law,
14, 2006, 134.

The Paradoxes of Deontic Logic: Alive and Kicking, Theoria, 72, 2006,
221232.
Prioritized Conditional Imperatives: Problems and a New Proposal, Auto
nomous Agents and Multi-Agent Systems, 2007, 17, 2008, 1135.
Mit Gabriella Pigozzi und Leon van der Torre:
Ten Philosophical Problems in Deontic Logic, in: Boello, G., van der Torre,
L. und Verhagen, H., Normative Multi-Agent Systems, Dagstuhl Seminar

Proceedings 07122, Internationales Begegnungs- und Forschungszentrum f
ur
Informatik (IBFI), Schloss Dagstuhl, 2007. Online verf
ugbar unter
http://drops.dagstuhl.de/opus/volltexte/2007/941.
284
Vortr
age
10.01.1998
On Relations between
Aqvists Deontic System G and Van Ecks
Deontic Temporal Logic (Workshop EON98 in Bologna, Italien)
22.01.2000
Sets, Sentences, and Some Logics about Imperatives (Workshop
EON00 in Toulouse, Frankreich)
22.05.2002
Problems and Results for Logics about Imperatives (Workshop
EON02 in London, England)
28.05.2004
Conflicting Imperatives and Dyadic Deontic Logic (Workshop
EON04 in Madeira, Portugal)
27.05.2005
Deontic Logics for Prioritized Imperatives (Arbeitsgruppe Law
and Logic i.R.d. Weltkongresses der Internationalen Vereinigung

f
ur Rechts- und Sozialphilosophie IVR 2005, Granada, Spanien)
20.03.2007
Prioritized Conditional Imperatives: Problems and a New Propo

sal (Seminar NorMAS07, Internationales Begegnungs- und Forschungszentrum f
ur Informatik IBFI, Schlo Dagstuhl)
20.03.2007
Ten Philosophical Problems in Deontic Logic (mit Gabriella Pi

gozzi und Leon van der Torre, Seminar NorMAS07, Internationales
Begegnungs- und Forschungszentrum f
ur Informatik IBFI, Schlo
Dagstuhl)
05.08.2007
Conditional Imperatives and Dyadic Deontic Logic (Arbeitsgrup

pe Legal Argumentation and Legal Logic i.R.d. Weltkongresses
der Internationalen Vereinigung f

ur Rechts- und Sozialphilosophie
IVR 2007, Krakau, Polen)
285

ANHANG: VORTRAGE
286
28.04.2008
Ten Problems in Deontic Logic (mit Gabriella Pigozzi, Institut
dhistoire et de philosophie des sciences et des techniques, Sorbonne,

Universite de Paris)
04.-15.08.2008
Deontic Logic in Computer Science (Seminar mit Leon van der
Torre, ESSLI 2008, 20th European Summer School in Logic, Language and Information, Hamburg)
Versicherung
Hiermit versichere ich, da ich die vorliegende Arbeit ohne unzulassige Hilfe
Dritter und ohne Benutzung anderer als der angegebenen Hilfsmittel angefertigt habe; die aus fremden Quellen direkt oder indirekt u
bernommenen
Gedanken sind als solche kenntlich gemacht.
Bei der Auswahl und Auswertung des Materials sowie bei der Herstellung
des Manuskriptes habe ich Unterst
utzungsleistungen von folgenden Personen
erhalten: dem englischsprachigen Korrekturleser Herrn Jason F. Ortmann.
Weitere Personen waren an der geistigen Herstellung der Arbeit nicht beteiligt. Insbesondere habe ich nicht die Hilfe eines Promotionsberaters in
Anspruch genommen. Dritte haben von mir weder unmittelbar noch mittelbar geldwerte Leistungen f
ur Arbeiten erhalten, die im Zusammenhang mit
dem Inhalt der vorgelegten Dissertation stehen.
Die Arbeit wurde bisher weder im In- noch im Ausland in gleicher oder
ahnlicher Form einer anderen Pr
ufungsbehorde vorgelegt und ist auch noch
nicht veroffentlicht worden.
287

Tese Hansen ImperativesAndDeonticLogic

Transféré par

Informations du document

Description originale:

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Tese Hansen ImperativesAndDeonticLogic

Transféré par

Droits d'auteur :

Formats disponibles

Imperatives and Deontic Logic

On the Semantic Foundations of Deontic Logic

zur Erlangung des akademischen Grades

3 Imperative Semantics for Deontic Logic

Appendix: Research Papers

Can imperatives, i.e. sentences in the imperative mood, be part of logical

One can imagine inferences which are of the following type: do

Logics concern is with the soundness of arguments, or inferences. These

a legal or moral norms validity: the external recognition of a norm as valid

Cf. Kalinowski [112] p. 36 and [114] p. 107.

Dubislavs Trick and Related Theories

To deal with his own dilemma, Jrgensen [110] endorsed a proposal by

Figure 1.1: Dubislavs convention (DC).

Figure 1.2: Jrgensens analysis of indirect imperative entailment.

This formalization is used in Reichenbach [180] 57.

Figure 1.3: Hares descriptor logic.

any imperative corresponds an ordinary indicative sentence which contains

show that seemingly differing explanations of imperative inferences are in fact

which can then be translated into

where !A is the imperative that is satisfied if and only if A is false. Using

Explanations of Imperative Inferences

Dubislavs trick provides a formal method that explains how inferences

The premiss love your neighbour may be regarded as describing

Logic of Non-Factual Existence

possibly independent from any pragmatic origin23 , in some ideal world of

the command given by to x by the use of the imperative sentence

with sentences. Whether also imperative sentences can be meaningfully used

Now we have arrived at something that resembles an argument though we

Maybe we should have gone looking for the explanation of an entailment

Here, an officer who must go from London to Edinburgh is ordered go via

inferences from demand-sentences through a convention (Ubereinkunft)

The term of commanding something by implication was introduced by Geach [61].

Shortly after Jrgensens dilemma and Dubislavs workaround for a logic of

Post the letter!

Post the letter!

But it seems counterintuitive to say that by ordering a letter posted one

Romeo asked me to post the letter or burn it.

such a distinction between explicitly given premisses and implicitly given

I was commanded to go via Coldstream or Berwick.

So it seems the authority contradicted herself when ordering (a little later)

without appealing to some implied permission: Imagine that, having been

I was commanded to go via Coldstream or Berwick.

Romeo asked me to post the letter or burn it.

I was (implicitly) ordered to post the letter or burn it.

I was commanded to go via Coldstream or Berwick.

Similarly, we can imagine Mercutio to reason in the following fashion:

Romeo asked me to post the letter or burn it.

I was (implicitly) ordered to post the letter or burn it.

inference. Perhaps (1) is simply invalid. It is obvious that the scheme is an

Figure 1.4: Dubislavs convention with relevance.

Close the window and play the piano!

Close the window and play the piano!

Here, the positive view on xs behavior by is not accepted by . Closing

Now consider the following, alternative dialogue:

She practised the Khachaturian with the window wide open.

s reproach for xs playing the piano is not accepted by , because playing

In these dialogues, s position seems natural, while s reaction is strange

that used the imperative, it is not a matter of logic.43

Open the door!

The terminology here is that of Hamblin [74] p. 184.

cal. So we should not be allowed to infer such imperatives. So Dubislavs

Ordinary Language Arguments

The following is an instance of this scheme: