History of Science

What is the proper relationship between History of Science and Philosophy of Science?
A bibliographic survey with emphasis on Realism, Naturalism, and Evolutionary Epistemology.

Paul F. Ballantyne
Department of Psychology
York University
Fall 1993 (updated October, 2002)

The three parts of this review sift through successive largely positivist and neo-Kantian attempts to define and state the relationship between History of Science and Philosophy of Science. Each section will mention realist thinkers (including E.A. Burtt, J.H. Randall, J. Somerville) who, to my analysis, have provided the kinds of alternatives which are necessary not only to describe the varied relations between those two divergent disciplines, but also to indicate the proper relationship of both disciplines to those who actually carry out the empirical aspects of scientific inquiry.

Appendix A, (immediately below) lists various references collected in 1993 under three headings: (1) history of science; (2) philosophy science; and (3) the relationship between the two. [The list has subsequently been updated in 2002.]

After providing a realist definition of the history of science, the textual body of the survey is concerned mostly with the latter category of so-to-speak combined sources but draws upon realist trends in all three of its parts to make its point. The issues raised in part 3 in particular, which compares four divergent sources on the HS & PS relation (by Laudan, Losee, McMullin, and Gutting) will likely strike a poignant cord with 21st century historians of psychology.

Butterfield, H. (1931). Whig Interpretation of History. London : G. Bell and Sons.

Sarton, G. (1927-48). Introduction to the History of Science. Vols 1-3, Baltimore: Williams & Wilkins.

Stimson, D. (1962). Sarton the History of Science. Cambridge: Harvard University Press.

Tanton, R. (Ed.). (1957). A General History of the Sciences. Vol 3; [Ancient & Midieval; Science in the 19th Century; Beginnings of Modern Science].

Taylor, F.S. (1949). A Short History of Science and Scientific Thought. New York: Norton & Co.

Thorndike, L. (1928-1958). History of Magic and Experimental Science. New York: Columbia University Press. Vols 1-8

Whewell, W. (1857). History of the Inductive Sciences. Vols 1-3. London: Parker & Sons.

Acton, H.B. (1967). Dialectical Materialism. In P. Edwards (Ed.). The Encyclopedia of Philosophy and the Human Sciences. Kent: Croom Helm Ltd.

Anton, J.P. (Ed.). (1967). Naturalism and Historical Understanding: Essays on the philosophy of John Herman Randall. Albany: State University of New York Press.

Baldwin, J.M. (1957). Dictionary of Philosophy and Psychology. Gloucester: Macmillan.

Bitsakis, Eftichios (1987-88). For an evolutionary Epistemology. Science and Society. Vol. 51, No. 4, 389-413.

Bitsakis, Eftichios (1993). Scientific Realism. Science and Society. Vol. 57, No. 1, 160-193.

Boeselager, W.F. (1975). The Soviet Critique of Neopositivism. (T.J. Blakeley Trans.). Boston: D. Reidel Publ.

Boyd, B.; Gasper, P.; & Trout, J. (Eds.). (1991). The Philosophy of Science. London: Bradford Books.

Conforth, M. (1952). Dialectical Materialism: An introductory course. [Vol. 1: Materialism and the Dialectical Method.] London: Lawrence & Wishart.

Cunningham, F. (1973). Objectivity in Social Science. Toronto: University of Toronto Press.

Feigl, H. & May, B. (Eds.). (1953). Readings in the Philosophy of Science. New York: Appleton.

Edwards, P. (1967). The Encyclopedia of Philosophy (Vols 1-8). New York: Macmillan

Hanfling, O. (1981). Essential Readings in Logical Positivism. Oxford: Basil Blackwell.

Hodge, M.J. & Cantor, G.N. (1990). The development of philosophy of Science since 1900. pp. 838-852. In Olby, R.C.; Cantor, G.N., Christie, J.R., & Hodge, M.J. (Eds.). (1990). Companion to the History of Modern Science. New York: Routledge.

Kolakowski, L. (1972). Positivist Philosophy: From Hume to the Vienna Circle. Middlesex: Penguin.

McMullin, E. (1990). The development of Philosophy of Science, 1600-1900, pp. 816-837. In Olby, R.C.; Cantor, G.N., Christie, J.R., & Hodge, M.J. (Eds.). (1990). Companion to the History of Modern Science. New York: Routledge.

Olby, R.C.; Cantor, G.N., Christie, J.R., & Hodge, M.J. (Eds.). (1990). Companion to the History of Modern Science. New York: Routledge.

Passmore, J. (1967). Logical Positivism. In P. Edwards (Ed.). The Encyclopedia of Philosophy (Vol. 5, 52-57). New York: Macmillan.

Scheffler, I. (1982). Science and subjectivity. Indianapolis: Bobbs-Merill (original work published, 1967).

Villemaire, D.D. (2002). E.A. Burtt, historian and philosopher : a study of the author of the metaphysical foundations of modern physical science . Dordrecht : Kluwer Academic Publishers.

Anton, J.P. (Ed.). (1967). Naturalism and Historical Understanding: Essays on the philosophy of John Herman Randall. Albany: State University of New York Press.

Anton, J.P. (1999). John Herman Randall Jr. American National Biography. Vol. 18. pp. 113-4. New York: Oxford University Press.

Burtt, E.A. (1932). The Metaphysical Foundations of Modern Physical Science. (2nd rev. ed.). London: Routledge & Kegan Paul. [*Page numbers in review correspond to the 1980 soft-cover Humanities Press edition.].

Cassirer, E. (1932). The Philosophy of the Enlightenment. (F.C.A. Koeln & J.P. Pettegrove Trans., 1951). Princeton: Princeton University Press.

Randall, J.H. (1940). Making of the Modern Mind. (Rev. Ed.). Cambridge: Houghton Mifflin.

Randall, J.H. (1958). Nature and Historical Experience. New York: Columbia University Press

Randall, J.H. (1962). Career of Philosophy: From the Middle Ages to the Enlightenment. Vol. I. New York: Columbia University Press.

Randall, J.H. (1965). Career of Philosophy: From the German Enlightenment to the Age of Darwin. Vol II. New York: Columbia University Press.

Randall, J.H. (1977). Career of Philosophy: Philosophy after Darwin. Vol. III. (posthumous work, B.J. Singer, Editor). New York: Columbia University.

Laudan, L. (1990). The history of science and the Philosophy of Science, pp. 47-59. In Olby, R.C.; Cantor, G.N., Christie, J.R., & Hodge, M.J. (Eds.). (1990). Companion to the History of Modern Science. New York: Routledge.

Losee, J. (1987). Philosophy of Science and Historical Inquiry. Oxford: Clarendon Press.

McMullin, E. (1970). The History and Philosophy of Science: A taxonomy. In Minnesota Studies in the Philosophy of Science. (H. Feigl & G. Maxwell Eds.). Minneapolis: University of Minnesota Press.

Gutting, G. (1990). Continental philosophy and the History of Science. In Olby, R.C.; Cantor, G.N., Christie, J.R., & Hodge, M.J. (Eds.). (1990). Companion to the History of Modern Science. New York: Routledge.

What is the history of science? I have suggested in class that for the purposes of this paper one might define the history of science as 'that which occurred' between the woodblock depicted in figure 1 called "Christ among the elements" (1495) and the photograph taken on 18 March, 1965 shown in figure 2 -which I have named "Man outside the elements."

Figure 1: depicts Christ (as representative of man) amidst the four known medieval elements: "fire, water, air, and earth."

Figure 2: is a photographic still of the first ever "space walk" carried out by Soviet Cosmonaut Alexei Leonov.

Why pick these figures as indicative of scientific history? Firstly, because their dates lie just inside the bounds of two undeniably transformative events in human history: The "discovery" of the Americas by European civilization (1492), and the American moon landing by Apollo 11 (1969). Secondly, because the adoption of science as a means of inquiry entailed a qualitative shift in our understanding of the respective place of human kind and nature (i.e., from the center of the universe to just part of nature). In medieval thought, "man" occupied a central and determinative place in the universe, in modern scientific thought nature holds a more independent, determinative, and permanent place than human kind (Burtt, 1932).

The reader should note upfront that the above definition of the history of science refers directly to the object of study -i.e., the historical unfolding of science from an arbitrary starting point to the relative present. In that sense, I am taking what is known in philosophical circles as a thoroughly direct realist approach (outlined by J. H. Randall, 1958, and many others before and since).

Although this realist approach seems to be the obvious starting place for all inquiry, it has decidedly not been the starting point for many of the past approaches in both philosophy of science and the history of science. This fact is highly problematic when one attempts to review the literature for the purposes of working out: (1) a stand on the proper relationship between the practice of HS and PS; and (2) the relation of these two disciplines to those whose profession it is to carry out empirical scientific practice.

In class I argued that in order to be most productive, the history of science must start from an explicitly stated direct realist position on epistemology but can not rely solely upon official reports of scientific events. I used the space walk (depicted above) as an example:

Leonev's official report on the space walk described an exhilarating experience that had gone according to plans. The real event, however, was quite different. One of the initial objective indications of this (in the West) was the fact that Leonov failed to retrieve the camera from outside the craft before reentry -with this indicative fact itself being inducted from the poor quality of images provided of this important technological achievement. The other details, however, given the techno-political considerations at play could only be gathered years later after the fall of the Soviet empire by way of interviewing Leonov and the other Soyuz 2 crew members.

Just yesterday, many of the artifacts from the Soviet Union's space program were sold to the highest bidder at a New York auction house. Leonov's space suit fetched 24,000 U.S. dollars. Leonov, who was present at the auction, was quoted as saying: "All these are artifacts of history. The space program is now the US space program" (CTV News, Dec 11, 1993).

[Incidentally, the profits from the auction are reported to be earmarked for the families of former Soviet Cosmonauts instead of the Russian people who, of course, paid for the whole affair in the first place. Likewise, no Marshall Plan of economic aid has been set up and free-market forces will be allowed to run their course in this thoroughly unprepared region.]

Readers should note that the above realist approach to scientific history does not just talk about logical interpretations of evidence (logical positivism), nor refer merely to varying concepts or metaphors (neo-Kantianism), nor to mere changes in experimental technologies (operationism), nor just to a temporal description of shifting traditions of investigatory practices (constructivism) but all of these and more. The "more," here mentioned, can itself be defined loosely as the nature of the transformative development of the object under study -whatever that object might be.

Enough about philosophical realism -for now. What about the discipline called the "History of Science"? Well, that discipline is younger than you might think. It was initially promoted by the Belgian born scholar George Sarton (1884-1956) who founded the ISIS (History of Science) journal in 1912. That journal provided the first institutional tool for the eventual establishment of the History of Science (HS) discipline.

One of Sarton's students, Dorothy Stimson, in her 1962 edited collection of Sarton papers, suggests that at the turn of the century, American scholars were already shifting toward a "New History" which includes an account of culture and ideas rather than the older form of history which was an account of politics and wars (see Stimson, 1962). These historians were, therefore, receptive to the ensuing disciplinization of the history of science brought about by figures such as Sarton, Alexandre Koyre (1820-1964) and Lynn Thorndike (1882-1965). Later, however, some of the naïve forms of "progressionist" history contained in the early HS movement were reacted to by Herbert Butterfield (1900-1964) in The Whig Interpretation of History (1931) which argued that history written by the victors would be inherently self-serving. Further elaboration on this latter issue, however, will have to wait to reappear in part III.

Ernan McMullin (1970) defined philosophy of science (PS) as comprising all of those philosophical inquiries that take science as their starting point, or at least their object of concern. Two articles which address the rise of the PS discipline include McMullin (1990) "The Development of Philosophy of Science 1600-1900;" and Hodge's "The development of the [PS]…since 1900." Both are contained in the edited anthology put out by Olby, et al. (Eds.), Companion to the History of Modern Science (1990). One might supplement these historical articles with the theoretical article by Bitsakis (1988) who outlines a very consistent argument for an "Evolutionary Epistemology" while comparing it with previous philosophical bases. The latter is especially recommended for psychologists who seek the relevant counter-arguments to neo-Kantianism and should, I believe, be reprinted in any future anthology of realist thought (see also Bitsakis, 1993).

The positivist movement in philosophy of science spanned from the 1830s with the work of Comte, right up through to the mid-1940s (and later in psychology). It can be defined as a succession of normative attitudes which attempted to distinguish between "philosophical and scientific disputes" that may profitably be pursued and so-called "metaphysical" questions that have no chance of being settled. The successive manifestations of the positivist movement are typically labeled as: social positivism (Comte, Mill); evolutionary positivism (Spencer); empiriocriticism (Mach, Avenarius); and logical positivism (Schlick, Neurath, Carnap).

For a probing book-length analysis of these trends, I have found the best available sources to be Andreski (1974); Kolakowski (1972); and Boeselager (1975). Boeselager, in particular, carefully traces out successive sub-movements within positivist thinking: from a Sensualism (in Mach and Avanarius), to a Physical Formalism (in Neurath, Schlick, and Carnap), to a Lack of a Criterion for Meaning (later Carnap -who reopened the issue of the nature of basic phenomenological reports and protocol statements in an attempt to address the issue of knowledge claims).

With regard to brief article size sources covering "the" positivist movement, some of the classics are found in the Encyclopedia of Philosophy (1967) edited by Paul Edwards (see the entries on Positivism, Logical Positivism, Verification Principle, Naturalism, Neurath, Schlick, versus Acton's account of Dialectical Materialism).

The standard mid-20th century basic readings text covering the so-called late positivist consensus was by Fiegl & May (1953). It was called simply Readings in the Philosophy of Science. For a later, updated, source for essential readings in logical positivism, however, one should refer to Hanfling (1981). That work also includes brief but helpful analysis by the editor along the way.

With regard to the more recent strivings for consensus in the philosophy of science, the new anthology source is Boyd, Gasper, & Trout (1991). The editors suggest that there has been a recent consensus in PS and that it consists of the following:

1. The emergence of sophisticated realist and neo-Kantian alternatives to traditional empiricist conceptions of science.
2. The development of naturalistic definitions of natural kinds as an alternative to standard empiricist (operationalized) conceptions of such matters. [I'm not sure what that means to them but I will support the use of the word naturalism in the Baldwin's (1957) progressive sense (i.e., humans as part of the natural 'order of things'].
3. Critiques of Humean conceptions of causal relations (i.e., constant conjunction) and of the associated covering-law account of explanation.
4. A greatly increased emphasis on the relevance of history of science for work in PS and de-emphasis on the alleged distinction between the context of discovery and the context of justification.
5. The emergence of a post-positivist literature in various specialized sciences (e.g., biology, psychology, and history); including a nonreductionist account of the relation between the sciences.

While this so-called consensus appears to me as just another mixed bag (i.e., a continued split between long-held ontologically idealist versus direct realist trends in PS), many of the specific trends mentioned in Boyd et al., are also outlined in the various (divergent) articles found in the 1990, Olby, et al. volume.

The older 1930-40s era texts listed under the HS&PS heading in Appendix A, really have a lot to say to the present generation of thinkers about the relationship between HS and PS. The differences between the direct realist approach of E.A. Burtt (1892-1989) and J.H. Randall versus the neo-Kantian position of Ernst Cassirer are especially enlightening in this respect.

Edwin Arthur Burtt's Metaphysical Foundations of Modern Physical Science (1932) opens by distinguishing between medieval, mechanical, and modern evolutionary thought in terms of the respective place of man and nature. He then points out that modern philosophy (Berkeley, Hume, Kant, Fichte, Hegel, James, Bergson) has been "a series of unsuccessful protests against this new view of the relation of man to nature" (p. 25) and that the only way to sort things out again "is to plunge into the philosophy of early modern science, locating its key assumptions…. and following them out to classical formulation…" (p. 35).

In other words, for Burtt, the history of science is to be used to sort out the viability of various philosophical assumptions from which it has sprung. Having initially carried out this analysis for publication under the same title in 1924, Burtt then felt compelled to rewrite the concluding chapter of the revised 1932 edition. Here, he clearly sides with an evolutionary and explicitly direct realist approach to the history of physical science which remained largely unmatched in the HS discipline for many years to come.

In this important chapter he first points out that it has no doubt "been worth the metaphysical barbarism of a few centuries to possess modern science" (pp. 305-306). He then addresses the shift in terminology from teleological medieval to mechanical to evolutionary suggesting that the middle era's rejection of final cause (i.e., the directedness of nature) has been "overdone" (p. 309). "If so, then an adequate scientific metaphysic will not be able to manage without teleology in some form, and it becomes a question of first-rate importance what that form is to be" (p. 310).

After elaborating his position on the importance of final causality, Burtt then outlines his reasons for adopting a direct realist "correspondence," "veridical" (p. 317) theory of perception. "Surely we have somewhere run off the track of sane thinking. Is it because we have failed to distinguish between sensed qualities and characters of real objects?" (p. 316). His answer is no! He suggests instead that this distinction was itself problematic. "In practice we correct dubious perceptions by appealing to further perceptions; we never correct them by comparison with something unperceived" (p. 316) -i.e., indirect "secondary qualities" of the mind, "sensations," etc.

More specifically, in contradiction to the postulation of indirect access to spacial relations (of past views) he suggests that "the space of perception is too much like the space of real objects to reveal any essential difference from it" (p. 317). And further:

"All it needs is to be freed from illusions, private images, and other experiences lacking social objectivity, to function quite acceptably as real space. And once this point has been reached there seems no longer any excuse for maintaining the distinction between sensed qualities and the real characters to which they correspond.... A radically different theory of mind is required to construe this situation in this fashion and make the fundamental structure of scientific knowledge more than unintelligible nonsense" (p. 317).

Just to be clear on his point of veridical perception Burtt then rams his point home by explicitly stating the disastrous implications for scientific investigation if it is not adopted:

"There is simply no science possible of the realm of sensible phenomena unless the trustworthiness of our immediate perception of spatial directions and relations be taken for granted. You think yourself justified in assigning my pain to the brain because you see what happens when the nerve fibres are cut, and you rightly assume that your vision is giving you a correct picture of what is going on in that portion of space occupied by those fibres. You are more than ever confident of it when other observers confirm you. This implies that the spatial world seen is the real spatial world, not something else. But why then should you turn around and accuse me of error when I say that the pain is in my finger.... [or] in otherwise empty space after my arm has been lopped off.... Why not admit that the feeling is where I feel it.... [A] speculative apriorism that flatly contradicts the immediate testimony of sense and places its objects in spatial relations wholly different from those in which they are sensed, can only lead, if carried out to its logical conclusion, to the complete confusion and mystification of science." (pp. 318-319).

Finally, it should be mentioned that Burtt recognized that further work had to be done in establishing a more "positive" (p. 319) and "adequate" (p. 324) philosophy of mind:

"[S]uch a philosophy of mind must provide full satisfaction both for the motives of the behaviorists who wish to make mind material for experimental manipulation and exact measurement, and for the motives of idealists who wish to see the startling difference between a universe without mind and a universe organized into a living and sensitive unity through mind properly accounted for. I hope some readers of these pages will catch glimmerings how this seemingly impossible reconciliation is to be brought about" (pp. 324-325).

*[A biography of E.A. Burtt by Diane Davis Villemaire has just come out in 2002.]

The much read Ernst Cassirer represents one of the preeminent contemporary proponents of Idealist epistemology (against which Burtt was arguing). Cassirer outlines the historical elaboration of that position in his Philosophy of the Enlightenment (1932) -which was translated into English by Fritz Koelln & James Pettegrove only in 1951.

On the other hand, J.H. Randall's Making of the Modern Mind (1940) suggests a realist alternative (see also Randall, 1958; J.P. Anton, 1967; 1999). The difference between the two books is that while Randall (like Cassirer) describes the emergence of the problem of knowledge during the Newtonian era, he also goes on to cover the evolutionary solution to it that was put forward during the late-19th century.

"But the very age that has seen so impressive a growth in scientific knowledge has also been profoundly troubled by the thought that it seems very difficult to understand how in any intelligible sense such a science is possible of attainment by the human mind…. the attempts to bridge the chasm have shown that, if we start with Locke's assumptions, we are bound to end up with Kant, that whatever certainty our science may have, it does not give us any light upon the basic structure of the world; …. It was not until nineteenth-century biology gave men a quite different conception of the mind and of experience… that the difficulty seemed to lessen. If we regard man as a biological creature actively adjusting himself to an environment, and experience not as a picture in the mind but such a process of adjustment, and knowledge, not as a copy of a real world, but as a definite relation between an intelligent organism and its environment, then the problem is transformed, and, set in new terms, seems possible of solution" (Randall, 1940, pp. 267-270; emphasis added).

Randall's specific critique of Cassirer's use of history can be found in his posthumously published Philosophy after Darwin (1977). Another important reference for his "historical naturalist" views on history are found in Randall's Nature and Historical Experience (1958).

Those in psychology can learn much by contrasting Randall's (1958) account with the more problematic positivist views appearing in E.G. Boring's collection of selected papers History, Psychology, and Science (1963). Despite these differences, however, and despite the best efforts of successive neo-Kantian movements in HS, PS, and history of psychology, a progressive non-dogmatic and anti-reductive naturalist account is still very much alive in all three areas of specialty.

I will now review four relatively recent works which specifically investigate the relationship between PS and HS; and will likewise mention their implications for the history of psychology along the way. It should be noted upfront that two of the works are written by epistemological direct realists (i.e., Laudan, 1990; McMullin, 1970;) while the other two are written by explicitly neo-Kantian indirect realist types (Losee, 1987; Gutting, 1990). It will be argued that seemingly finite differences in this fundamental starting place make a world of difference with regard to where their respective evaluations end up.

We will start with a summary of Larry Laudan's (1990) paper called simply "The history of science and philosophy of science." I have augmented the following account with appropriate dates and contextual information for some of the figures mentioned. Unfortunately Laudan, who was writing for an audience of trained professionals, left much of this information out of his article.

The history of science is usually seen as emerging from the early studies of Joseph Preistley (1733-1804), Adam Smith (1723-1790) and Jean E. Montucla (1725-1799) in the late eighteenth century. Preistley wrote History of Electricity (1766) and was involved in the French revolutionary politics and religious controversies of the era. Smith wrote Wealth of Nations and requires no elaboration. In 1758, Montucla published his History of Mathematics (1758). [Note that the 1962 Stimson anthology provides coverage of all of these early figures].

Between the 1830s-1930s the two subjects (HS & PS) developed in tandem. In this era, says Laudan, it was suggested that our understanding of science needed to encompass both what science had been (hence the historical content) and what science ought to be (hence the philosophical content). Scholars such as William Whewell (author of the History of the Inductive Sciences, 1837), Auguste Comte, Ernst Mach, Pierre Duhem (1861-1916), and E.A. Burtt (1892-1964) all intermingled historical and philosophical concerns. [Incidentally, Paul Tannery (1843-1904) should be mentioned here as an exception because Sarton names him as the first de facto full-time historian of science.]

Beginning sometime during the 1930s, however, each discipline began to go its own way. Philosophers of science, smitten with logical positivism's alluring promise of rigor, began to think that the method of conceptual analysis alone was sufficient to formulate an adequate understanding of the scientific enterprise. None of the figures of logical empiricism (e.g., Moritz Schlick, Rudolf Carnap, Hans Reichenbach and Carl Hempel) saw any necessity of using history to adjudicate between rival philosophical doctrines. Formal, logical analysis was to replace historical research as the preferred mode of presentation.

Historians of science "since the 1940s," says Laudan, have generally been weary of using the historical record to address broad or general questions about the nature of scientific change. So by the late-1950s the two fields had become remote from one another.

During the 1960s Paul Feyerabend, Hanson, Toulmin and Kuhn sought to persuade philosophers to reconsider the relevance of historical research to philosophy of science. Laudan does not go into details but merely states this outright [fair enough]. He then suggests that there has been a recent return to respectability, if not dominance, of the older tradition of pursuing philosophical and historical research together. Some of the names associated with this movement have been Dudley, Shapere, Laudan, Ernan McMullin and Imre Lakatos.

By contrast, the attitudes of the historians have hardened considerably. Why? Laudan gives one reason (resistance to theory). As they see it the construction or evaluation of theories about the past is beyond the pale of legitimate historical inquiry. Herbert Butterfield and his repudiation of The Whig interpretation of history is often invoked as a rationale. In reply to that trend, Laudan states that Butterfield was a fine historian but that his essays on historiography are confused.

Be that as it may, Butterfield's term 'Whiggism' has apparently been overgeneralized to refer disparagingly to any general theory of scientific change. Laudan counter-argues by pointing out that if historians are willing to put forward such theories when discussing practical matters such as labor history, they should not be so reticent to do likewise when discussing the nature of scientific progress. One can sympathize, says Laudan, with Butterfield's concern that a tale of victory, told by the victors makes for bad history. But in denying that historians are ever justified in recognizing that certain parts of science are better than others, and in asserting that it is not part of the historian's task to explain the conditions which made them more successful, Butterfield (and those who follow him) would appear to be abandoning the program of telling a fuller story.

The book by John Losee PS and Historical Inquiry (1987) also makes a few points regarding the relationship between HS and PS. This time, however, an indirect realist Kantian position is held by the author and the analysis tends to sway toward the 'logic of concepts contained in the discourse of different positions' rather than biting the bullet and suggesting the nature of the relationship as demonstrated by changes in scientific practice. In fact, Losee constantly argues against the various direct realist positions covered in the book.

Losee starts by providing a list of various positions on the possible relationship between HS and PS:

1 PS and HS are mutually exclusive
2 PS is dependent on HS (or vice versa)
3 They are interdependent
4 They are de facto overlapping disciplines
5 PS may be subsumed under HS

The fundamental distinction which Losee makes is between prescriptive and descriptive philosophy of science. PrescriptivePS seeks to formulate and recommend criteria that ought to govern evaluative practice. It therefore sanctions a distinction between correct and incorrect evaluative practice in science.

Under descriptivePS, on the other hand, the task of the historian of science is to record the evaluative standards that are explicit within scientific practice in diverse contexts. In other words: to judge whether particular decisions of scientists conform to evaluative standards of the time. Prescriptive evaluations are thus avoided.

Losee's goal is to clearly outline the difficulties of two historiographical extremes in PS: a pure progressionist view of science on the one hand, and an equally disingenuous nonjudgmental historicism on the other. He then tries to come up with a middle position between such prescriptive and descriptive PS. With this motivation to avoid extremes we can agree, but Losee seems (to this reviewer) far too conservative in his formulation of the middle ground.

Losee's particular brand of weak descriptivePS (rather than historicism per se), while deserving of a sympathetic ear, is as untenable in the long run as the form of phenomenologically descriptive (though non-developmental) psychology suggested by Amedeo Giorgi (1985) and all the same arguments against apply. Their shared ontologically idealist starting point is highly problematic. Consequently, their descriptive ontological reference to 'historical' and 'phenomenological' regularities respectively (though carefully constructed), fall prey to their underlying epistemological indirect realism (see Ballantyne, 1991 for further elaboration of these terms).

Such historicized descriptive accounts do not proceed toward firmly grounded prescriptions because they do not talk about the nature of the development of the object (a middle-ground kind of prescriptivePS) but only about the use of technological instruments or mental techniques used to investigate them.

Put another way, for those who prefer "metaphors" and analogies over mere reference to philosophical jargon:

Purely descriptivePS is like being 'strapped into a toddler's stroller seat' (passive in influencing events in nearly every respect). Purely prescriptivePS is like being a dictatorial back-seat driver. Weak DescriptivePS is like being a Monday morning quarterback. And in contrast to all of these, the proper role of a combined HS & PS as I view it, is more like being a map-bearing road-derby race 'navigator' or aeronautic copilot.

Before moving on, perhaps one more topical analogy can be slipped in here with specific regard to Losee's (1990) work:

"Don't try to feed me TV dinners when you know I'm used to steak. We don't need no rank-beginners when it's time to shake-the-shake" (From Long John Baldry's "Don't try to lay no Boogie-Woogie on the king of Rock and Roll," It Ain't Easy, 1971).

Why the arcane reference to one of the most popular rock albums of 1971? Because a much more productive and comprehensive "taxonomy" of various PS and HS positions had already been provided by the epistemological direct realist Ernan McMullin in 1970.

McMullin's fifty-five page article (1970), was published as part of the widely available Minnesota Studies in the Philosophy of Science serial. It begins by attacking the logical positivist assumptions of Carnap who "reduces the philosophy of science to a logic of science" (p.13). The idea, says McMullin is not to seek a way of justifying scientific propositions by applying "some set of logical rules" but instead to pay attention to the "actual course that science has followed" (p. 14). The "connectives" that guide science are historical and not solely logical. After elaborating his taxonomy (covered below) McMullin returns to this point:

"What philosophers for a long time failed to see was that ontological questions necessarily involve the developmental aspects of science. They cannot be answered (or more correctly, they will be wrongly answered) if one is content with examining a temporal "slice" of scientific work. What discloses the nature of the relation between the model and the modeled is not a logical structure of here-and-now predictions and verifications, but rather a dynamic pattern visible in the way models guide inquiry" (p. 63).

The main contribution of this article, however, is that it provides a rough but very useful taxonomy of various forms of both PS and HS. In doing so, McMullin points out some of the necessary and contingent ways in which they are related. His analysis goes far deeper than that of Losee because it does not presume that an explanatory analysis of the relationship must assume any single way of describing that relationship. While Losee suggested that it all depends upon which sort of HS and PS you are comparing (hence preferring a descriptive approach), McMullin suggests that even this diversity has a traceable structure.

The initial distinction made by McMullin is between two senses of "science." Science 1 (S1) is understood as a "set" and often a "list" of scientific propositions, proposals and positions (e.g., those found in Newton's Principia or a any given Psychological Review article). S1 is a measure of achievement and is of interest primarily to the scientists of a given era. [Note that Danziger (1997) would later call this the "knowledge products" of given eras hence reintroducing their import back to historians of psychology at least].

Science 2 (S2) on the other hand, is the process by which those discoveries were made (i.e., including not only explicit statements about advances but also the various false starts, technological half-steps, implicit speculations, and rough sketches, etc.). Many aspects of S2 are often forgotten and may never have been made explicit. As such, it is of special interest to the historian of science. According to McMullin, S2 contains S1 within it and is broader in scope.

In turn McMullin distinguishes between HS1 and HS2. HS1 is of interest to scientists (e.g., the reference to prior studies which appear in the opening pages of standard empirical research reports; celebration of anniversaries of institutions or of special milestone dates, etc.). It deals with the listing of important texts, technologies, or institutions in their finished version or ultimate form.

HS2 on the other hand is of interest to the historian well versed in the general background of a given period (e.g., the sociocultural or techno-political contexts of a given scientific inquiry). Such historians will be interested in biographical details of important figures and will make much of rough drafts and unpublished manuscripts. The examples McMullin uses are Whittaker's History of Theories of Aether and Electricity for HS1 and William's Michail Faraday for HS2 but we can all, I think, come up with our own examples with respect to the history of psychology in particular.

From there, McMullin moves on to distinguish various forms of Philosophy of Science. PSExternal (PSE) looks outside science to the broader context of the times. As such, it does not necessarily rest upon the history of science (HS1) alone but it must at least provide some rational reconstruction or reference to science for its readers. For if PSE diverges too far from what scientists appear to be doing, it is likely to be ignored or challenged. The divergence of Aristotle's Posterior Analytics (which outlined his either/or formal logical laws) from his own contributions to biological knowledge is given by McMullin as an example of the first outcome. A more up to date example of the latter outcome would be the work of Paul Feyerabend which has been beaten from pillar to post for the last 30 years due, in part, for the idiosyncratic divergence of its account of science practice. [Note that in 2002, I doubt many current philosophy of science students even know (let alone care) who Feyerabend was -and this, I believe, is his just rewards.]

Referring back to its older manifestations, McMullin points out that PSE includes PSMetaphysical (PSM). Descartes Discourse on Method (which begins from a general theory of knowing -from the inside outward in this case) is mentioned here as is Carnap's Inductive Logic (in which science is thought of as a logical structure of demonstration or validation -which McMullin calls PSLogical).

Regarding the origins of the shift in PS emphasis from concern with metaphysics toward concern with methodology (i.e., toward PSInternal), McMullin informs us that discussions of the nature of science up to the seventeenth century were very external in character. The theory of inquiry was based upon a priori metaphysics or some autonomous logic. Though Bacon, Boyle and Huygens depended on their knowledge of practice in their analyses of methodology, "it was only in the nineteenth century that writers like Whewell and Mill took this new source of PS with complete seriousness" (p. 26). [Note that McMullin is in complete accordance here with not only Burtt and Randall but also Cassirer on this point].

Thus attempts toward PSInternal (PSI) rely upon a careful description of how scientists actually proceed or have in the past proceeded. History of Science 2 (which contains reference to HS1) is relevant to PSI for two different sorts of reasons: (1) because it provides completed case studies, of a kind one could not recover from mere finished product scientific reports; and (2) more fundamentally, because it allows one to study science in its all important temporal dimension (p. 29).

McMullin is then careful to point out that his above taxonomy (between S1, S2, PSL, PSM and PSI) ought not to be taken to imply that any given piece of HS or PS conforms to one and only one category of inquiry. In practice, one finds philosophers and historians of science calling on many or all of them. But it is important that the readers of such works not be mislead about what the balance between them (in a given argument) really is (p. 30). McMullin then compares the use of history by Popper, Lakatos, and Feyerabend to demonstrate his point about balance. [see McMullin for that account]

He finishes up of course with a positive assessment on 'whether one can do history and philosophy of science together' but also with some highly important statements in the naive direct realist vein regarding 'history as the clue to ontology.' [a phrase very similar, but not quite as direct as the one used by Burtt, 1932, p. 29]

"How is this striking series of [scientific] success to be accounted for?…. Since the model is the only possible mode of access we have to the world, there is no way of answering this question directly. But if we try to account for the career of the model… there seems to be no satisfactory alternative to saying that the explanatory resources of the model are due to its having revealed, however imperfectly and incomplete, an 'ontological' structure, i.e., a structure intrinsic to the world…" (McMullin, 1970, p. 66).

What does this mean for the practicing historian of science? Well, for one thing, it allows some degree of confidence in the current scientific categories. For instance, as pointed out by McMullin very early in his article, while it is inappropriate to misrepresent the views of a historical figure as if he were using presentist terminology, the assumption of realism allows the historian to investigate the respective coverage of matter right back to Aristotle and the Ancient Greeks.

Over the years, this issue of making ontological claims (and prescriptions) has been much more difficult for the neo-Kantian indirect realists (including Cassirer, Losee) to grapple with than it has for those who start from a naive or direct realist position (e.g., E.A. Burtt; J.H. Randall; I. Scheffler, F. Cunningham, E. McMullin, E. Bitsakis).

With regard to the way the issue has played itself out in the history of psychology subdiscipline, the same point can be made. Across the history of psychological investigation, different labels but the same (though historically changing) subject matter is being referred to by figures with different cultural or institutional backgrounds, biases, and interests. This having been said, the rationale for books of "basic readings" (e.g., R.I. Watson, 1979) holds up despite the adamant protests of those who can not understand, or will not concede to the more mature forms of realist and progressionist argumentation.

Given the explicitly sated bias of the present writer toward both a direct realist and direct perceptionist epistemology, one should not expect any extensive press to the Kantian group of philosophers covered in Gutting's (1990) article. I tend to view such attempts as being as noisy, and as fruitless as the spinning of wheels when one's car is stuck in the mud (i.e., only the addition of a firm layer of gravel or a wooden board to grip to will get you out again). Some of the highlights of this fourth and final article, however, do warrant brief mention.

Gutting opens with the point that the Anglo-American analytic philosophers of science have "left untouched" positivism's legacy of uncritical attitude toward the "ultimate significance of scientific knowledge" (p. 127). The most interesting part of Gutting's account therefore is his coverage of the French Network: Gaston Bachelard (1884-1962); George Canguilhem; and Michel Foucault who have addressed this uncritical attitude in their work.

Bachelard and Canguilhem focused on the physical and biological sciences, respectively, while Foucault was primarily concerned with the human sciences. Bachelard taught at the Sorbonne during Foucault's student days and Canguilhem was Foucault's thesis advisor.

Bachelard suggested that in the case of the natural sciences historians should distinguish between l'histoire perimee (the history of outdated science) and l'histoire sanctionee (the history of science judged valid by current standards). Bachelard rejects the continuity view of science while accepting its overall progress. "Science develops by a series of epistemological breaks that make it impossible to regard its history as a linear accumulation of truths within a single conceptual framework" (Gutting, p. 136).

For Canguilhem, the history of science is primarily the history of concepts and not of terms or phenomena. A history of terms which consists, for instance, of finding people who spoke of evolution before Darwin may be misled by superficial similarities in language while ignoring the question of whether two scientists had the same understanding of the given aspect of nature. In the field of psychology, the book edited by Herrenstien & Boring (1966) immediately jumps to mind as an example of that sort of history of terms (unaccompanied by appropriate commentary on the fundamental differences of conception being used). [Note that Danziger (1990) makes this point explicitly.].

Secondly, a history centered on phenomena, says Canguilhem, may downplay the fact that "the crucial factor is not what is observed but the interpretation involved" (Gutting, p. 137). Gutting uses the example of Priestley's and Lavoisier's laboratory observations and their respective interpretations regarding "dephlogisticated air and oxygen" but I was not convinced by his arguments [any more than I was later convinced by the same argument being made by Danziger's (1997) work called Naming the Mind.]

According to Gutting, the denial of terms and of phenomena as appropriate subject matter does not mean that Canguilhem denies the need to understand the influence of earlier scientific work on later work. Indeed Canguilhem sees this as an important part of understanding the genuine historicity of science and pays detailed attention to such influences in his history of the concept of the reflex. [Note that the attempted distinction between concepts and terms remains undefined by Gutting].

I personally doubt whether the points of this "French Network" are as necessary or as important as Gutting has made them out to be. For example, the recent book by Robert Richards on the history of evolutionary theories (1987) managed to avoid many of the pitfalls suggested by Canguilhem against any history of terms, theories, or phenomena. That is, the historicity and the lack of value-free status of knowledge is now widely recognized among the history of science community. It seems to me that the direct realist account is a much easier and surer road to success than the neo-Kantian byway of the French school -which itself denies the guiding importance of the phenomena under study just as the positivist school denied the guiding importance of the sociohistorical aspects on inquiry.

As for Foucault, himself, I would like to suspend judgment until further data has been collected. Gutting suggests that "contrary to some of his critics (and supporters), Foucault "does not attempt to undermine the rationality and objectivity [i.e., veridicality] of science in general or even of the human sciences taken as a whole" (p. 141). This sort of comment from an explicit neo-Kantian always makes me very cautious. The same argument has been made for years by supporters of Kuhnian paradigms. Similarly, in the history of psychology, Koch's incommensurability thesis has enjoyed the limelight in many journals despite the hedges, qualifications and logical self-contradictions that are a fundamental part of every article he has ever written.

The proper relationship between HS and PS can not be limited to any single internal/external, or descriptive/ prescriptive role on each other because they have and will continue to rely upon each other for their very existence. The proper role of both subdisciplines toward those who carry out scientific inquiry however can be stated as being one of an historically informed guidance councilor, or navigator which can help scientific practitioners avoid the pitfalls and unproductive byways of the past. This statement holds for those who study the history of psychology as much as it holds for those who study the history of physical, chemical, and biological science.

With respect to which of the available philosophical bases is best suited to performing such a guidance role, I am confident that it could be demonstrated that everything of value in the successive neo-Kantian positions was either quickly incorporated into, or already present implicitly in the works of the more competent direct realist members of the craft.

To this latter group of "historical naturalist" craftsmen I say thanks for providing the basis for "a dream that keeps my 'soul' alive" (Theme from: Midnight Express, 1979). Alternately, to contemporary followers of past neo-Kantian colleagues: "You better pull your 'thing' together, that you've been bustin' out; and if you feel like you 'just can't dig it' then you know that you 'don't know what its all about' " (from Long John Baldry's "Don't try to lay no Boogie-Woogie on the king of Rock and Roll," It Ain't Easy, 1971).

Boring, E.G. (1963). History, Psychology, and Science: Selected Papers. New York: John Wiley.

Danziger, K. (1990). Constructing the Subject: Historical origins of psychological research. Cambridge: Cambridge University Press.

Danziger, K. (1997). Naming the Mind: How psychology found its language. Thousand Oaks, CA: Sage.

Gibson, J. (1966). The Senses Considered as Perceptual Systems. Boston: Houghton Mifflin.

Gibson, J. (1979). The Ecological Approach to Visual Perception. Boston: Houghton Mifflin.

Giorgi, Amedeo, (Ed.). (1985). Phenomenology and psychological research. Pittsburgh: Duquesne University Press; Atlantic Highlands, N.J.: Distributed by Humanities Press.

Herrnstein, R.J. & Boring, E.G. (Eds.). (1966). A Sourcebook in the History of Psychology. Cambridge, Mass: Harvard University Press.

Richards, R.J. (1987). Darwin and the Emergence of Evolutionary theories of mind and behavior. Chicago: University of Chicago Press.

Watson, R.I. (1979). Basic Writings in the History of Psychology. New York: Oxford University Press.