This article was downloaded by: [McGill University Library]

On: 20 November 2014, At: 00:05

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

British Journal for the History of

Philosophy
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/rbjh20

A. C. Crombie styles of scientific

thinking in the European tradition
a
A. Rupert Hall
a
Tackley, Oxon
Published online: 03 Jun 2008.

To cite this article: A. Rupert Hall (1995) A. C. Crombie styles of scientific thinking in
the European tradition, British Journal for the History of Philosophy, 3:2, 409-419, DOI:
10.1080/09608789508570924

To link to this article: http://dx.doi.org/10.1080/09608789508570924

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information
(the “Content”) contained in the publications on our platform. However, Taylor
& Francis, our agents, and our licensors make no representations or warranties
whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and
views of the authors, and are not the views of or endorsed by Taylor & Francis. The
accuracy of the Content should not be relied upon and should be independently
verified with primary sources of information. Taylor and Francis shall not be liable
for any losses, actions, claims, proceedings, demands, costs, expenses, damages,
and other liabilities whatsoever or howsoever caused arising directly or indirectly in
connection with, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes.
Any substantial or systematic reproduction, redistribution, reselling, loan, sub-
licensing, systematic supply, or distribution in any form to anyone is expressly
 forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
Downloaded by [McGill University Library] at 00:05 20 November 2014
 JHP— (Review Article Vol. 3/No.2

CROMBIE
STYLES OF SCIENTIFIC THINKING IN
THE EUROPEAN TRADITION
Downloaded by [McGill University Library] at 00:05 20 November 2014

The history of argument and explanation especially in the

mathematical and biomedical sciences and arts. 3 vols.
2456 pp., ills. Duckworth, London, 1994, £180
ISBN 0-7156-2439-3
A. Rupert Hall

As a massive work on the history of science published since 1950

Joseph Needham's Science and Civilisation in China is unique, far
exceeding in scale such monumental enterprises as George Sarton's
Introduction to the History of Science (1927-47) or Lynn
Thorndike's History of Magic and Experimental Science (1923-
58). Though Alistair Crombie's book does not match any of these
in bulk, it is a majestic piece of scholarship for our times. It
contains 1555 pages of text, more than three-quarter of a million
words, and clearly the provision of apparatus (900 pp.) is very
generous too. In time it extends from Aristotle to Darwin. As this
pair of august names indicates the focus of the book is firmly in the
West; with exceptions to be noted later Crombie's vast range
largely excludes Islam and the Slavic world. For, he writes: 'When
we speak today of natural science we mean a specific vision,
created within Western culture, at once of knowledge and of the
object of that knowledge, a vision at once of natural science and of
nature, a vision explored and controlled by argument' (p. 4). And
again: 'The main concern of this study is with the historical process
by which the analytical sciences, with medicine and technology,
acquired their identity in the intellectual culture of Europe' (p. 88).
Crombie's plan is to make an historical study of episodes, phases
and indeed individuals in order to show how they exemplify types
 410 A. Rupert Hall

or styles of scientific reasoning or tactics. To speak of a 'style' of

science is (as I understand it) to employ a concept looser and more
general than that of 'method', with its crank-handle overtones. A
style of science does not generate mechanically either a discovery
or a demonstration; it is not easily definable, for it may be (with
Crombie) a 'Galilean style' or (with I. Bernard Cohen) a
'Newtonian style', or a 'seventeenth century style', a methodologi-
cally defined style (as 'experimental') or a style defined by a
problem as with classification. Of these multifarious broad entities
Downloaded by [McGill University Library] at 00:05 20 November 2014

Crombie chooses six for discussion and detailed exemplification:

(I) postulational or axiomatic; (II) experimental; (III) model-
making; (IV) taxonomical; (V) probabilistic and statistical;
(VI) historical or genetic. To each style, in this order, is devoted
one section of the book, these being of very diverse lengths. It is
obvious, for example, that taxonomy is a less complex topic than
experiment, and this section is not illuminated by any detailed
historical investigation of the author's, as other sections are.
Alistair Crombie is a veteran of many years teaching and
writing, a founder-member of the British Societies for both the
History and the Philosophy of Science and founder-editor of the
BJPS. His first book achieved scholarly distinction more than forty
years ago, and (in revised form) the thesis he formulated then is
represented in one section of the present study. Another milestone
was a monograph of 1967 on the critical changes in the
interpretation of the eye taking place about 1600, again fully
reflected here in Section III on model-making. A third large portion
of the text is indebted to work of many years ago (partly in
collaboration with Adriano Carugo). It is hardly surprising
therefore that as the reader draws in the flavour of this vast work
he finds it essentially conservative. Its own classical 'style' in the
historiography of science was established in the first half of this
century, though here expressed with a range and depth of
scholarship unattainable them. We find Crombie on more than
one occasion characterizing 'scientific relativism' as a recent and
dubious phenomenon. Of course, a consciousness of cultural
relativism is essential: 'When we read a text we must ask to what
questions the author was giving an answer', but 'we need to
control relativity both by objective scientific truth and by the
objective continuity of scientific and scholarly tradition' (pp. 5, 7).
And again: "We can recognize . . . in the solution of stable
 Styles of Scientific Thinking in the European Tradition 411

problems of nature and of thought, an objective continuity which

survives the relativity of culture.' Nature stays put; that is a first
condition for the objectivity of science, and 'objective scientific
validity' has accumulated through time. The historic European
tradition of natural science, which has developed criteria for
qualifying theories and inquiries as true or false, sterile or
profitable, cannot be much 'illuminated by an unsophisticated
sociological relativism which fails to distinguish the specific history
of science, as a problem-solving activity, from a general history of
Downloaded by [McGill University Library] at 00:05 20 November 2014

ideas lacking its identifying modes of self-correction and criteria of

acceptability' (p. 15).
One notes also that themes singled out for particular enhance-
ment in recent decades receive little attention from Crombie
(recall, however, that he is not writing a comprehensive history of
scientific thought). Such authors as Marsilio Ficino and Pico della
Mirándola and such great themes as the Neoplatonic revival are
discussed only in relation to the early Renaissance. Crombie is far
more interested in the continuing legacy of Plato's own writings,
especially Timaeus; he feels the force of the criticism that Kepler
and Mersenne directed against such occultists as Robert Fludd.
'Harmony in its threefold Platonic sense of proportion, fitness and
sympathy' was indeed to remain 'an enduring guide of the Western
search for a sufficient basis of scientific explanation . . .' but
'Fludd saw the inquirer into nature as a neoplatonic Hermetic
magus, a magician; Kepler saw him as a scientific virtuoso, a
rational artist', that is, a type to become dominantly successful in
the future (pp. 117, 515). In Crombie's view, scientific proposi-
tions in the past as in the present have been established by rational
scientific argument, not by the light of inner illumination. The
word 'Rosicrucean' is absent from the index. Paracelsus in another
potent influence who scarcely figures in this book; perhaps it might
be possible to write of a 'chemical style' in the history of modern
science, but Alistair Crombie leaves that task to others.
This minimization of influences and modes of thought that some
scholars have recently seen as being determinative (in certain
respects) of the course of modern science - trends salutary in the
eyes of veterans accustomed to a distinction between science and
pseudoscience currently regarded as pernicious - may in part be
attributed to the book's concern with the logic of demonstration
rather than the logic of discovery. Though this distinction is not
 412 A. Rupert Hall

made by the author it seems obvious that his attention is given less
to the origins of innovations than to their maturation. Thus for
him (as for Alexandre Koyré long ago) Galileo was notable for
'brilliantly conceived thought experiments' (p. 308); Galileo's
'method of argument was to eliminate rival proposals by means of
. . . rules of inference, and then to try to demonstrate the truth of
his own favoured proposal by agreement with phenomena'
(p. 576). No hint here of Galileo's significant role (as others have
Downloaded by [McGill University Library] at 00:05 20 November 2014

seen it) in exploring nature by experiments. Rather,

. . . he was at one with Aristotle in promoting a rational philosophy of
science and of nature against on one side scepticism and on the other
Neoplatonic and Hermetic magic. Mersenne, Gassendi and Descartes
promoted the same sort of rational philosophy against the same sorts of
opponents. . . . An interesting difference is that Mersenne and Gassendi
were sufficiently sceptical to disbelieve that certainty was possible in the
search for causes in natural science. Mersenne therefore insisted upon
experimental precision. Galileo and Descartes still on this question stayed
with Aristotle. This could considerably affect the relative weight given to
experimental measurement as distinct from mathematical or logical
demonstration in scientific inquiry (p. 588)
All this comment seems more appropriate to a discussion of
Galileo's demonstration of propositions, than to his manner of
arriving at them. Since divergent opinions and interpretations are
never debated in the text (though, in rare instances, they are
recorded in the notes) it is impossible to guess at Crombie's reasons
for rejecting them. Certainly he is as convinced as Koyré that
Galileo was a poor experimenter.
To close this little circle a further quotation (with which
Stillman Drake would surely have agreed) may be fitting: 'The on-
going physical argument through all [Galileo's] major writings on
natural philosophy aimed to dispute and reject the Aristotelian
conception of physical causes and to establish in its place the truly
certified conception which in the end he saw as uniformly
mechanical' (p. 588).
To some of us the perspective of past time is convergent in that
the personages of antiquity appear smaller and more remote than
those of the sixteenth century, and these in turn less than our
nearer predecessors such as Faraday or Darwin; but Alistair
Crombie is an historian shaking hands, as it were, with Aristotle
 Styles of Scientific Thinking in the European Tradition 413

and Ptolemy, Galen and Dioscorides, as familiarly as with Leibniz

or Kant. At least within the epochs of these volumes, he sees the
giants of Hellas and Christendom always shaping the thoughts of
their successors. As H. Floris Cohen has pointed out, Crombie
believes in the essential continuity of natural science, at least from
its pre-Socratic origins in Hellas to the nineteenth century (The
Scientific Revolution: A Historiographical Enquiry, Chicago
1944, pp. 105-7). Now of course this does not mean that Crombie
is imperceptive of the internal displacements that have upset the far
Downloaded by [McGill University Library] at 00:05 20 November 2014

from smooth and uniform evolution of science - the kind of

displacement effected when Copernicus argued for the physical
reality of the conclusions reached by axiomatic mathematical
science or when Galileo and Newton substituted for the rational
apodeictic certainty of Aristotle (and, indeed, Descartes and
Leibniz) the conclusions reached by mathematical analysis as
applied to planned experimentation - for a large part of the present
work is concerned with precisely such displacements, either within
a historically-evolving 'style' or in creating a new 'style'. Not the
theories nor the techniques of science are continuous in Crombie's
view, but rather a broader 'vision explored and controlled by
argument'. The word 'argument' is crucial in Crombie's vocabul-
ary. The truth or falsity of a proposition in natural science is tested
not by appeal to authority or any other extraneous arbiter but by
reasoned discussion based on evidence, logic and probability. It is
in this court that Galen and Harvey, Aristotle and Darwin, can
debate as equals. It is in determining the procedures for this court
that echoes of the Posterior Analytics or of Pappus on analysis and
synthesis ring down the ages (and into these pages) and that such
'precedents' have weight as those traced by Arthur Lovejoy in The
Great Chain of Being. It may be added that in a lesser sense too
Crombie is a 'continuist', holding that the men of the sixteenth and
seventeenth centuries were neither ignorant nor contemptuous of
medieval science, and that their knowledge of antiquity was owing
in large part to the Middle Ages. Nor had those ages lacked
innovation: Galileo (Crombie writes): 'focused the medieval
definitions and techniques of uniform, accelerated and instanta-
neous velocity, momentum, and the Merton mean speed theorem
onto the particular motions of falling bodies and projectiles as they
occurred in the actual world' (p. 423). One of the major new
elements in his treatment of medieval science in this book is a long
 414 A. Rupert Hall
essay on Alhazen, ranking 'with Ptolemy and Kepler as an architect
of scientific optics' (pp. 352-74); on the other hand, as in
Grosseteste (1953) cosmology and astronomy are given less
attention. There is - also in the same area of science - an excellent
discussion of Newton's new theory of light and colour, little on the
development of celestial mechanics. If, as one might suppose, the
'style' of axiomatic mathematical science was fully established by
Newton's time, did he not give it brilliant new facets of
validification?
Downloaded by [McGill University Library] at 00:05 20 November 2014

From all the above it must be evident that Alistair Crombie is an

intellectual or idealist historian of science, as markedly now as
forty years ago. Moreover he has not a whit weakened his thesis
that the science man creates is determined on the one hand by the
problems he chooses to tackle, on the other by his mental
equipment for tackling these problems, which in turn is derived,
directly or indirectly, from those who have addressed these
problems philosophically. A scientist is not made by nature, nor
even by the study of nature, but by philosophy. But the didactic
and the experiential are combined. Scientific innovation has been
built upon both antecedent inquiry into the possible ways of
understanding nature, and the actual exemplification of these
ways.
This is an opinion which he has not modified since Grosseteste
when it was remarked (and rejected) by Alexandre Koyré:

. . . si M. Crombie affírme que la 'revolution méthodologique du

genérale, la méthodologie était le moteur et le factuer determinant du
progres scientifique, je ne crois pas qu'il l'ait prouvé non plus. Une fois de
plus, il me semble que les résultats méme de ses recherches sapent sa these
(Diogéne, no. 16, Oct. 1956, p. 190).

I do not mean to revive this dispute, only to note that the same
thesis is here developed and illustrated on a far grander scale than
in the earlier book. Here the long-lasting methodological influence
of the Greek applied mathematicians in their diverse inquiries
(Archimedes, Hero, Ptolemy) and of Aristotle's logical writings is
made fully evident. One may conjoin to Koyré's criticism of the
earlier book Crombie's recent judgement 'of the style of late
medieval intellectual life [,] that natural philosophers formed a
systematic conception of experimental scientific argument and that
 Styles of Scientific Thinking in the European Tradition 415

they carried out some well organized experimental investigations.

Here again close attention must be paid to the relation of language
to performance . . . [magnetism and biology afford examples].
Again optics, belonging at once to the mathematical scientiae
mediae and to natural magic, provides a clear illustration of the
late medieval theory and practice of experimental analysis and
synthesis' (pp. 349-50).
So much for generalities. It is time to give an idea of the structure
Downloaded by [McGill University Library] at 00:05 20 November 2014

of this complex book. A long introduction considers the historio-

graphy of science as 'a kind of comparative intellectual
anthropology'. The first substantive Section describes the
'postulational or axiomatic' style of science, and the Greek
exploration (by Aristotle in particular) of the philosophy of the
investigation of nature which was carried forward into the Middle
Ages and Renaissance (pp. 93-309). The following Section, on
experiment, is the longest in the book (pp. 313-1077), probably
the most important and original too. Here there is a brief, revised
account of the 'medieval logic of experiment' from Grosseteste to
Ockham, Buridan and Oresme; there follows a review of fifteenth-
century humanist ideals, of geographical exploration 'the most
dramatic outcome of the mathematical rational arts of the 15th
century' (p. 434) and especially of the fertile conjunction of science
and art that then occurred:
The modem history not only of biology, geology, cartography and
technology but of central areas of the whole scientific movement is
scarcely imaginable without these visual techniques [of the printed book]
(p. 462).
Next, Neoplatonism, Hermeticism, alchemy and other post-
classical themes are introduced, principal attention being given to
Kepler and Mersenne versus Fludd. In the discussion of astronomy
Christoph Clavius figures more prominently than Copernicus. So
to Kepler and Galileo (pp. 543-626). Like Stillman Drake,
Crombie relates Galileo firmly to his father Vincenzio and to
music. I noted the following sentence on method: 'The paradox is
that he continued to use with evident conviction the apodeictic
language appropriate to a conception of a completed and closed
system of knowledge, while he was adding all the time with
supreme virtuosity to the content of a science that was bursting
 416 A. Rupert Hall

from its traditional logical categories' (p. 553). There follows a

more rapid treatment of a series of 'experimentalists' - Bacon (with
'one method for the cultivation, another for the invention of
knowledge'), Gilbert, Harvey, the Royal Society. Volume II,
continuing the theme of experimental inquiry, brings in an unusual
theme: music. The theory of consonance from Zarlino (p. 787)
onwards gives scope for explaining the less well known researches
of many familiar figures, leading to Mersenne, from whom
Crombie quotes eleven translated pages on Mersenne's 'philosophy
Downloaded by [McGill University Library] at 00:05 20 November 2014

of scientific research' and fourteen more on sensation, especially

hearing. (I mention this to make the point that the work has
something of the character of an anthology.) Mersenne is for
Alistair Crombie a figure of the highest stature from whom (he
tentatively suggests) Galileo may have plagiarized both the theory
of the pendulum and that of vibrating strings (pp. 882-888). The
Section on experiment closes with 'Exemplary Thinkers', opening
a little oddly with Descartes ('Please understand', he wrote to
Mersenne, 'that I have demonstrated the refractions geometrically
and a priori and I am astonished that you should doubt it'
(p. 904) ).
The most general lesson of this lengthy discussion is, perhaps,
that there is no such thing as a single experimental method
affording proof, but rather a variety of experimental 'styles'
affording argument:
Bacon permanently affected the style of experimental argument in
Britain, just as Descartes did that in France and the Netherlands, but in
each the style developed with scientific experience and influenced the
other (p. 987).
Crombie quotes with approval (seemingly) Fontenelle's compar-
ison of Newton with Descartes, including the sentence: 'Both
founded their physics on a geometry which they developed almost
single-handed'; whatever one thinks of this statement, it is evident
to all that Cartesian and Newtonian physics could never be called
either 'experimental' or 'mathematical' in the same senses of the
words.
Section III, on model-making, moves from the clock-universe to
the camera-obscura eye, and an exceptional treatment of hearing.
Its last chapter 'Knowing is Making' takes a more philosophical
look at 'nature conceived through its conformity with art, from
 Styles of Scientific Thinking in the European Tradition 417

early biological mechanics ('taking to pieces this automaton' as Dr

Thomas Browne wrote (p. 1262) ) to universal languages.' Both of
the following short Sections (IV, Taxonomy; V. Probability and
Statistics) dwell extensively on the Greek foundations of these
'styles' of science; in the latter the contributions from non-
mathematical sources to 'the new logic of the uncertain and the
contingent' are made clear. Pascal, Arnauld, Leibniz, Graunt
introduce the (now perhaps) better-known researches of the
mathematicians preceding Condorcet and Laplace. From statistics
Downloaded by [McGill University Library] at 00:05 20 November 2014

to Maupertuis's rescuing 'design in the history of nature' and then

to Malthus and Darwin are relatively short steps.
The final Section progresses from human history to biological
evolution. This alone might stimulate a long essay on the
treatment of time in the European consciousness. That the 'search
for origins' in human history generally, or in specific instances such
as the history of science - here J. E. Montucla is rightly praised
(pp. 1599-1600) - served as a model for ideas of biological
evolution is not claimed here, and would hardly be plausible, but
they were juxtaposed by Herder (c. 1790), for whom 'the
predominant law of creation' was that 'the general co-operation of
active powers in their determinate individuality, govern all the
events that happen among mankind, as well as all the occurrences
in nature' (p. 1620). The enigmatic status of man in relation to the
animals at the end of the 18th century (La Mettrie, Monboddo
etc.) prepares the way for the 'History of Nature', that is, historical
geology (Steno, Hooke, Burnet). With Buffon and his contempor-
aries the problem of the origins of organic diversity presented itself
urgently. Through Buffon 'the long Western tradition of the
genetic method, the characteristic Western search for explanation
through causes producing development through time, was to pass
(when serious) into a more scientifically factual future' (p. 1695).
The important distinction drawn between the Leibnizian view of
organic evolution through progressive amelioration and the view
(essentially French, before Erasmus Darwin) of organic evolution
among the plastic descendants from an original 'germ' amid an
environment changing through time introduces the final chapter
on Darwinian natural selection. Did Darwin appreciate (at the
first) that his was a statistical hypothesis? Others, like Cournot,
saw how it might be; so 'Darwin's fundamental insight . . . was
into the application of the statistical model of net advantage to the
 418 A. Rupert Hall

biology of populations, just as Descartes had applied the

mechanistic model of the automaton to the biology of individual
organisms' (p. 1741). One may therefore see in Darwinian
evolutionary theory the union of four styles - the historical, the
statistical, model-making and the taxonomic, the foundation of
Darwin's system (pp. 1746-7).
One can read this prodigious work in a variety of ways: as an
historiographical essay on the 'taxonomy of scientific styles'
providing 'an illuminating insight into very deep characteristics of
Downloaded by [McGill University Library] at 00:05 20 November 2014

our intellectual culture' (p. 1762); as a philosophical essay on the

interaction between epistemology and the actual solution of
scientific problems; as a set of analyses of episodes or characteristic
epochs in the history of science; as an idealist apologia countering
the relativist and sociological interpretations of the development of
science; and not least as a bibliographical resource. In this last
guise the book is truly phenomenal; Sources (by author) include, at
a rough estimate, nearly 3,000 titles; Studies (also by author) some
13,000. Is this the work of one man or ten? To pursue the
thousands of references, many reduced to name-and-date form,
would require Job's patience: note 13 (p. 1802) on Darwin
includes a name-date list covering three pages! Ordinary mortals
can only marvel at the proof-correction, let alone the compilation,
of information on this scale.
Alistair Crombie has brought to fruition a holistic view of
science possessing extraordinary richness, range and diversity, that
is founded on a belief in the twin studies of human civilization and
of nature as the necessary foundation of our moral choices:
immensely serious, immensely thorough. It is a work founded
upon the assurance that knowledge is purposefully and consciously
framed in the human mind, ever active in its dual role as careful
observer and active theorist. 'The antecedent intellectual explora-
tion or dissection of nature in the head, before approaching its
problems experimentally with the hands, was a habit learned from
mathematics and philosophy which established the general logical
structure of medieval and early modern science' (p. 308).
Crombie's point of view is consistently more lofty than that of
most historians. For him scientific thought is essentially auto-
geneous though influenced always by the environment of the
thinker. At least within the range of these studies, he finds
reflection upon the past far more significant than rejection of the
 Styles of Scientific Thinking in the European Tradition 419

past. In conclusion, one cannot but endorse the words of

Alexandre Koyré in the essay already quoted:
. . . les demeures du royaume de Dieu sont nombreuses. Et Ton peut
traiter l'histoire de bien des facons. Disons done que dans le royaume de
Phistoire, M. Crombie a édifíé une tres belle demeure.
Neritas filia temporis.
Downloaded by [McGill University Library] at 00:05 20 November 2014

Tackley, Oxon