Académique Documents
Professionnel Documents
Culture Documents
DOI 10.1007/s11229-007-9177-6
Received: 15 September 2006 / Accepted: 17 April 2007 / Published online: 26 May 2007
Springer Science+Business Media B.V. 2007
1 Introduction
In Hilberts famous Paris list of problems, the fifth challenge was to formulate Lies concept
of a continuous group of transformations without the assumption of the differentiability
of the functions defining the group. To solve the problem Hilbert wrote a short memoir
on the foundations of geometry ber die Grundlagen der Geometrie during the winter
The author wishes to thank Academy of Finland for financial assistance that enabled her to work on this
article.
M. H. Hartimo (B)
Department of Mathematics, Statistics, and Philosophy,
University of Tampere,
Pinni A, Tampere 33014, Finland
e-mail: mirja.hartimo@uta.fi
123
226
123
227
similar to Gausss. Grassmanns Theory of Extension is a coordinate-free geometrical calculus, a theory of n-dimensional vector spaces (Torretti 1978, p. 109).
123
228
From the present-day standpoint Husserls criticism appears silly: he does not seem to
understand Gausss empirical and dynamic approach at all. However, similar criticism has
been expressed by D. M. Johnson in 1979 from the point of view of the development of
topology. According to Johnson, [o]n a very profound level there is a serious problem related
to Riemanns entire approach. Ultimately he links his basic topological objects with numbers
and coordinate systems. In other words, a Riemannian manifold is always reducible to a
number-manifold. Consequently, when trying to construct a framework for nonmetrical
analysis situs, he nonetheless seems to fall back on concepts of measurement and ordinary
analytic geometry. In this way he appears to complete a logical circle (1979, p. 126). The
ultimate solution to this problem requires the concept of a topological mapping (Johnson
1979, p. 127). Thus Husserls criticism could be understood to express a worry that fueled
the subsequent development of topology. Husserl wanted the analysis situs to be entirely
non-metrical.
Husserls final point is that [t]he truth is that the concept of surface is not mathematically
but only logically definable, and that mathematics cannot do anything but presuppose this
concept, and with help of its determinations declare mathematical characteristics, through
which one can express the position of a point in such surface manifold. Not the concept of
surface manifold, but the position of a point in a surface manifold is to be mathematically
defined. (Husserl 1983, pp. 345346). Husserl also complains that in Riemannian geometry,
Euclidean geometry is just an arbitrary case of geometry in general (Husserl 1983, p. 347).
Hence, Husserl thinks that only mathematical, not philosophical, value can be attributed to
Riemanns theory (Husserl 1983, p. 347). Bertrand Russell has formulated the issue more
poetically as follows: For mathematics, where quantity reigns supreme, Riemanns conception has proved itself abundantly fruitful; for philosophy, on the contrary, where quantity
appears rather as a cloak to conceal the qualities it abstracts from, the conception seems to
me more productive of error and confusion than of sound doctrine. (Russell 1897, p. 69).
Later, the Norwegian mathematician Sophus Lie followed in the steps of Riemann and
Helmholtz in solving Helmholtzs problem of space purely mathematically by use of transformation groups (Torretti 1978, p. 154, 172). No detailed notes on Lies work can be found
in Husserls published writings. This is not surprising because Husserl writes in a letter to
Natorp that he engaged in an intense investigation in geometry, arithmetic, and the theories
of manifolds in the years from 1888 to 1893 (Husserl 1983, p. 396), while Lie published
his first papers on the foundations of geometry in 1890, but the third volume of his Theory
of Transformation Groups, in which Lie gave the most expanded solution to Helmholtzs
problem of space, appeared in 1893 (Torretti 1978, p. 154). Curiously, however, Husserl
mentions Lies transformation groups in Prolegomena as an example of a theory of theories.
However, Husserls list of potentially fruitful theories is long and contradictory. Besides Lies
transformation groups, Husserl explicitly mentions Cantors, Grassmanns, and Hamiltons
approaches as well as RiemannianHelmholtzian theory (Husserl 1975, p. 252).
In an 1892 letter to Brentano, Husserl claims that he has changed his view about Riemann
Helmholtz view. However, though he now thinks that the RiemannHelmholtz theory has
a valuable core, he himself has followed a much deeper and unvergleichlich ebener und
leichter way in his investigations (Husserl 1994b, p. 11). Around 1893 Husserl drafted an
outline for how one should proceed in solving the problem of space. Husserls to-do list is
the following:
the origin and content of the geometrically basic concepts have to be studied
the origin and content of the axioms, and thereby are also clarified
the origin and content of the geometrical space. (Husserl 1983, p. 286)
123
229
This list suggests that Husserl favored a synthetic approach to geometry. However,
Husserl did not want to compromise the method of calculation either and accordingly favored
Hermann Grassmanns approach. Here, Husserls criticism of Ernst Schrders Vorlesungen
is analogous to his criticism of analytic geometry. Husserl complains about the externality
of calculations that were used in place of genuine deductions (Husserl 1983, p. 8, 1994a,
p. 56) as well as of the arbitrariness of Schrders stipulative definitions that did not capture
the natural processes of thought (Husserl 1983, p. 33, 1994a, p. 81). In general Husserl
does not object to calculations as such, but only that they were not founded on genuine
thought. This attitude is demonstrated in his remarks on geometry as well. For a long time
Husserl thought that Grassmanns Ausdehnungslehre provided the approach that combined
the logical analysis of our space intuition and the analytic methods purely, without presupposing Cartesian coordinates. In 1897, Husserl still thought that the only satisfactory general
framework for Euclidean manifolds could be found in Grassmanns Ausdehnungslehre of
1862 (In a letter to Natorp in March 29, 1897. Husserl 1994c, pp. 6061).
However, by the turn of the century Husserl is in approximate agreement with Hilbert,
who views the axiomatics as the way of proper analysis of our perception of space. In 1901
Husserl gave two lectures to a Gttingen mathematical society. Husserls view of Definitheit in his Double Lectures has been discussed in detail in Authors (2007) paper, so I
will not digress now except to mention that in his lectures Husserl shares Hilberts view
of completeness or definiteness of an axiom system as categoricity. Thus Husserls development in the 1890s takes him from the justification of imaginaries through Grassmannian
approach to axiomatics. Contrary to Hilbert, who was influenced by projective geometry,
Husserl followed the route via justification of the complex numbers. In Husserls writings
the search for intrinsic methods is clearly combined with the desire to justify the usage of
complex numbers. Accordingly Husserl explains in the Formal and Transcendental Logic
that his initial motivation to study definite systems came from questions such as: Under what
conditions can one operate freely, in a formally defined deductive system with concepts
that, according to the definition of the system, are imaginary? (Husserl 1974, p. 101.
English translation 1969, p. 97). Likewise in the foreword to the Philosophie der Arithmetik,
Husserl held that in the second volume he is going to discuss a new philosophical theory of
Euclidean geometry based on Gausss work Anzeige der Theoria residuorum biquandraticorum, Commentatio secunda from 1831 (Husserl 1970, p. 8). Husserl was inspired by Gausss
treatment of complex numbers in his 1831 work and for the same reasons Husserl seems to
have been interested in Grassmanns Ausdehnungslehre (Husserl 1983, pp. 396397).
Thus Husserls statement in the Formal and Transcendental Logic (1929) summarizes his
and Hilberts development:
Throughout the present exposition I have used the expression complete system of
axioms, which was not mine originally but derives from Hilbert. Without being guided
by the philosophico-logical considerations that determined my studies, Hilbert arrived
at his concept of completeness (naturally quite independently of my still-unpublished
investigations); he attempts, in particular, to complete a system of axioms by adding a
separate axiom of completeness. (Husserl 1974, 1969, 31)
123
230
was added to the French translation of the Grundlagen (1900), and then to its second edition.
Both directions were guided by similar criticisms of analytic geometry. But in contrast to
the Festschrift, Hilberts Memoir is a contribution to the HelmholtzLie problem of space.
Hilberts approach is group-theoretical and it uses the notion of Jordan curve as well as
Cantors set theory. In the words of Weyl, Hilbert in the Memoir does get rid of [Lies
assumptions of differentiability] as far as Helmholtzs problem in the plane is concerned.
The proof is difficult and laborious; naturally continuity is now the foundation, and not the
keystone of the building as it had been in his Grundlagen book (Weyl 1968, p. 156). In his
notes, Husserl lists Hilberts axioms, briefly describes the main idea of the proof, and then
makes some remarks about the theorems. The first axiom is the definition of a plane, the
second the definition of a motion, the third states that the motions form a group. The fourth is
a definition of a rotation, and the fifth one states that the system of rotations is closed. Hilbert
also defines the true circle and the true line, which are isomorphic to the number circle
and number line, respectively. The assertion that Hilbert proves is the following: A plane
geometry in which Axioms IIII are satisfied is either the Euclidean plane geometry or the
BolyaiLobachevskian geometry. (Hilbert 1990, p. 155). In other words Hilbert manages
to show that by means of his first three axioms one can characterize either the group of
Euclidean motions or the group of BolyaiLobachevski motions, i.e., that the three axioms
alone characterize Bolyais absolute geometry of plane (cf. Torretti 1978, p. 187). Thus
Hilbert gives a solution to the HelmholtzLie problem of space.
The editors of the Husserliana date Husserls notes to around the turn of the century
(wohl um 1900). Hilbert first presented the introductory part of the Memoir in a meeting of Gesellschaft der Wissenschaften zu Gttingen on November 8, 1901 (Hilbert 2004,
pp. 8, 537). This paper was added to the English translation of the Festschrift of 1902 as a
summary of the forthcoming Memoir. According to the editors of (Hilbert 2004), the paper
lacks the detailed working-out of the proofs of the latter part of the Memoir. The finished
paper, published in the Mathematische Annalen, has been signed on May 10, 1902, and it
appeared in the Mathematische Annalen, Volume 56 (3) in 1903.
Husserl moved to Gttingen in the fall of 1901; he started lecturing at the end of October
1901 (Schuhmann 1977, p. 67); attended Hilberts lecture Abgeschlossenheit von Axiomensystemen in the Mathematical Society on November 5 (Schuhmann 1977, p. 68); and he gave
the Doppelvortrag in Gttingens mathematical society in November 26, and December 10,
1901 (Schumann and Schumann 2001, p. 97). In his notes, Husserl summarizes the Memoir
without giving a detailed working out of the proofs. Husserl makes references to Hilberts
saying something: Mit dem Raum habe ich mich nicht befasst, sagt Hilbert (Husserl 1983,
p. 413), hence we can assume that Husserls notes are based on Hilberts oral explanation
of the Memoir, either in a presentation or privately. Husserl might have attended Hilberts
lecture on November 8, 1901, or the two might have discussed the matter privately as well:
at some point Hilbert had also shown Husserl Freges letter of December 27, 1899 in which
Frege claims that the consistency follows from the truth of the axioms and not the other
way around. Husserls remark to this is that Frege does not understand the sense of Hilberts
axiomatic grounding of geometry (Husserl 1970, p. 448, English translation Husserl 2003,
p. 469).
There are some differences between the published Memoir and Husserls notes. For
example, in his notes on Hilberts memoir Husserl first distinguishes between two directions in geometry: one is the development from Euclid to Gauss, Lobatchefsky, and others,
the direction in which the points, lines, etc, are equally justified, and where one studies
relationships between them seeking for the simplest possible system with which to capture
all of geometry (Husserl 1983, p. 412). This is the approach of the Festschrift.
123
231
The other direction is the analytic tradition of Riemann, Helmholtz, and Lie. According to
the notes Lies approach was the first mathematically rigorous formulation. But, his approach
is complicated and non-geometrical (Husserl 1983, p. 412). In his Paris problems, as well as in
the Memoir, Hilbert similarly claimed that Lies solution is rather forced and complicated
(Browder 1976, p. 13). The problem is the artificial assumption of the differentiability of the
functions defining the group.
Instead of distinguishing between the two approaches Hilberts Memoir starts directly with
a discussion of Lies approach. However, in the end of the Memoir, there are some remarks that
should have interested Husserl too, but are missing from Husserls notes. Hilbert concludes
the memoir by pointing out the characteristic difference between it and the Festschrift. The
difference is in the arrangement of the axioms. In the Festschrift Hilbert listed them so that the
axioms of continuity were listed last, so that the question as to what extent geometry could
be developed without them would arise in a natural way. In the Memoir, on the contrary,
continuity is required first among the axioms by the definition of the plane and a motion so
that here the most important task has been rather to determine the least number of conditions
from which to obtain by the most extensive use of continuity the elementary figures of
geometry (circle and line) and their properties necessary for the construction of geometry.
(Hilbert 1990, p. 189). It is likely that Hilbert did not explain this remark in a conversation with
Husserl for Husserl would have mentioned it in his notes, but that it was added to the Memoir
afterwards. This addition may prompt one to speculate on Husserls possible influence on
Hilbert. In his Double Lectures Husserl criticized Hilberts treatment of completeness as an
axiom, saying that, to him, it should never be an axiom but a theorem (Husserl 1970, p. 102).
This suggests that Husserl and Hilbert might have discussed continuity and completeness
and Hilberts remark in the published Memoir might have been inspired by these discussions.
sofern er ein theoretisches Wesen ist, also ein Wesen, das die Fhigkeit besitzt zu denken, zu erkennen,
und somit Wissenschaften zu bauen bzw. wissenschaftliche Theorien entdecken, welche die verschiedenen
Sachgebiete zur systematischen Darstellung bringen. Das Vernnftige an der Vernungtttigkeit des Menschen
liegt in der Form der Wissenschaft, die also das ausmacht, was die Idee der Vernunft eigentlich konstituiert
(Husserl 2001a, p. 305).
123
232
the ideas of logic, the logical concepts and laws, to epistemological clarity and definiteness. Here phenomenological analysis must begin (Husserl 1984, p. 9; English translation,
Husserl 2001b, p. 168, italics in the original). Respectively, in the Logical Investigations
Husserl develops the notion of categorial intuition aiming to give epistemological justification for Hilbertian axiomatics (more on this, see Hartimo 2007).
In his Memoir Hilbert gives alternative, group-theoretical foundations for geometry. Thus
he gives an alternative solution to Husserls earlier concern about analytic geometry and
Riemann and Helmholtzs approach to the problem of space. The Memoirs non-metrical,
topological nature must have been pleasing to Husserl. But otherwise it must have raised a
question of how to reconcile with it the view that Hilbertian axiomatics gives the form of
reason. Hilberts Memoir suggests that the proper theory form could take a group-theoretical
form instead of the form of the Festschrift. In addition, shortly afterwards Minkowskis work
on special relativity was surely another source suggesting the group-theoretical form for the
a priori.
It seems likely that the two alternative ways of characterizing the a priori inspired Husserl
to distinguish between the formal and the material eidetic sciences. Thus, Husserl in the
Ideen I describes geometry at times as a formal discipline and at times as material eidetics
(for example, Husserl 1977, p. 20 vs. p. 135). This is not a sign of indecision on
Husserls part; rather, depending on the occasion, he refers to either the group-theoretical or to
the axiomatic approach. To undertake a phenomenological analysis of the group-theoretical
approach, Husserl is led to analyze the givenness of groups. Given that Richard Tieszen
(2005) has shown, quite plausibly, that the group-theoretical approach to modern geometry
can be seen as a realization of eidetic intuition, it is no wonder that Husserl soon started to
develop his view of eidetic intuition.
References
Bernays, P. (1967). Hilbert, David. Entry in the Encyclopaedia of Philosophy. Vol. 3. Edited Paul Edwards.
New York: Macmillan & Free Press, London: Collier Macmillan, pp. 496505.
Browder, F. E. (Ed.) (1976). Mathematical developments arising from Hilbert problems. Proceedings of symposia in pure mathematics, Vol. XXVIII, Part 1. Providence, Rhode Island: American Mathematical
Society.
Dorier, J.-L. (1995). A general outline of the genesis of vector space theory. Historia Mathematica, 22,
227261.
Grassmann, H. (2000). Extension theory. Lloyd C. Kannenberg (transl.) American Mathematical Society,
London Mathematical Society [1862].
Hartimo, M. (2006). Mathematical roots of phenomenology: Husserl and the concept of number. History and
Philosophy of Logic, 27(4), 319337.
Hartimo, M. (2007) Towards completeness: Husserl on the theories of manifolds 18901901. Synthese 156(2),
281310.
Hilbert, D. (1903). Ueber die Grundlagen der Geometrie. Mathematische Annalen, 56, 381422. English
translation in Hilbert (1990), pp. 150190.
Hilbert, D. (1990). Foundations of geometry. Translated from the 10th edition by Leo Unger. La Salle, Illinois:
Open Court [1971].
Hilbert, D. (2004). David Hilberts lectures on the foundations of geometry, 18911902. Edited by Michael
Hallett and Ulrich Majer. David Hilberts lectures on the foundations of mathematics and physics 1891
1933, Vol. 1. Berlin: Springer.
Husserl, E. (1969). Formal and transcendental logic. Translated by Dorion Cairns. The Hague: Martinus
Nijhoff.
Husserl, E. (1970). Philosophie der Arithmetik. Husserliana XII. The Hague: Martinus Nijhoff.
Husserl, E. (1974). Formale und transzendentale Logik. Husserliana XVII. The Hague: Martinus Nijhoff.
Husserl, E. (1975). Logische Untersuchungen. Erster Band. Prolegomena zur reinen Logik. Husserliana XVIII.
The Hague: Martinus Nijhoff.
123
233
Husserl, E. (1977). Ideen zu einer reinen Phnomenologie und phnomenologischen Philosophie. Erstes Buch:
Allgemeine Einfhrungin die reine Phnomenologie 1. Halbband: Text der 1.-3. Auflage - Nachdruck.
[Ideas pertaining to a pure phenomenology and to a phenomenological philosophy, first book: general
introduction to a pure phenomenology. First half binding. Text of the 13 editions. Reprint.] Edited by
Karl Schuhmann. The Hague, Netherlands: Martinus Nijhoff.
Husserl, E. (1983). Studien zur Arithmetik und Geometrie. Texte aus dem Nachlass 18861901. Edited by
Ingeborg Strohmeyer. Husserliana XXI. The Hague, Boston, Lancaster: Martinus Nijhoff Publishers.
Husserl, E. (1984). Logische Untersuchungen. Zweiter Band, Erster Teil. Husserliana XIX/1. The Hague,
Boston, Lancaster: Martinus Nijhoff.
Husserl, E. (1994a). Early writings in the philosophy of logic and mathematics. Collected Works, Vol. V.
Translated by Dallas Willard. Dordrecht, Boston, London: Kluwer.
Husserl, E. (1994b). Briefwechsel. Band I. Die Brentanoschule. Dordrecht, Boston, London: Kluwer.
Husserl, E. (1994c). Briefwechsel. Band V. Die Neukantianer. Dordrecht, Boston, London: Kluwer.
Husserl, E. (2001a). Logik. Vorlesung 1902/03. Husserliana Materialien Vol. 2. Edited by Elisabeth
Schuhmann. Dordrecht, Boston, London: Kluwer.
Husserl, E. (2001b). Logical investigations. Vol. 1. Translated by J. N. Findlay. London, New York: Routledge
[1970].
Husserl, E. (2003). Philosophy of arithmetic, psychological and logicl investigations with supplementary texts
from 18871901. Translated by Dallas Willard. Collected Works X. Dordrecht, Boston, London: Kluwer.
Johnson, D. M. (1979). The problem of the invariance of dimension in the growth of modern topology, Part I.
Archive for History of Exact Sciences, 20(2), 97188.
Klein, F. (1979). Development of mathematics in the 19th century. Translated by M. Ackerman. Massachusetts:
Math Sci Press. Translation of Vorlesungen ber die Entwicklung der Mathematik im 19 Jahrhundert.
Teil I, Berlin [1928].
Russell, B. (1897). An essay on the foundations of geometry. London, Glasgow, Leipzig, New York, Bombay:
Cambridge University Press.
Schuhmann, K. (1977). Husserl-Chronik. Denk- und Lebensweg Edmund Husserls. Husserliana Dokumente
Vol. 1. Den Haag: Martinus Nijhoff.
Schuhmann, E., & Schuhmann, K. (2001). Husserls Manuskripte zu seinem Gttinger Doppelvortag von 1901.
Husserl Studies, 17, 87123.
Tappenden, J. (1995). Geometry and generality in Freges philosophy of arithmetic. Synthese, 102, 319361.
Tappenden, J. (2006). The Riemannian background to Freges philosophy. In Jeremy Gray & Jose Ferreiros
(Eds.), The architecture of modern mathematics (pp. 107150). Oxford University Press.
Tieszen, R. (2005). Free variation and the intuition of geometric essences: Some reflections on phenomenology
and modern geometry. Philosophy and Phenomenological Research, LXX(1), 153173.
Torretti, R. (1978). Philosophy of geometry from Riemann to Poincar. Dordrecht, Boston, Lancaster:
D. Reidel.
Weyl, H. (1968). Gesammelte Abhandlungen. Band IV. Berlin, Heidelberg, New York: Springer-Verlag.
123