¹ Miller, Eugene F., “Positivism, Historicism, and Political Inquiry,” American Political Science Review, 66 (09, 1972)CrossRefGoogle Scholar.

² Just as it is sensible for philosophers of science to be au courant in contemporary developments in the social sciences.

³ A main thrust of the essay is surely an attempt to show how what it construes as the triumph of historicism in philosophy of science has influenced “postbehavioralist historicist” movements in political science. However, by the time he has reached the antepenultimate paragraph, Miller has come to the conviction that neither the behavioralists nor the postbehavioralists represent viable approaches to the science of politics; for neither the positivism nor the historicism which respectively lie at the roots of these positions (so he claims) “can provide a satisfactory basis for political inquiry” (p. 816). He concludes with a critique of David Easton's recent discussion of behavioralist and postbehavioralist positions. I shall eschew any detailed comment on this final part of Miller's essay—his colleagues in the discipline itself may appropriately take account of his closing remarks. I will, however, simply mention what a surprising turn his final revulsion against historicism represents to anyone who, like me, had come to think of historicism as the “hero of the piece” on the basis of the earlier part of his paper. Yes, on the basis of the earlier portions of the paper, I had been reading him as a happy partisan of historicism. What we come to realize at the end, though, is that we are really being proffered a dialectical analysis. The positivistic thesis is, to be sure, superseded by the historicist antithesis. But the latter, too, offers political science no genuine basis and we must therefore pass to a (lower?) synthesis: Traditional Political Philosophy. Here is our real if somewhat neglected hero.

Of course, as might be noted, it's not always neat or easy to construe this dialectic as a “historical process.” When positivism is identified with Hume and historicism with Kant (or perhaps Neo or Post-Kantians or Hegel or Neo- or Post-Hegelians) it's easy enough to see that positivism is the thesis and historicism the antithesis. But when historicism is identified with such 19th century proponents and positivism is identified with the Vienna Circle critics of these metaphysicians, one might think, conversely, that it was historicism that was the thesis and positivism the antithesis.

Moreover, when we pause to think that Traditional Political Science, harking back at least to the Middle Ages, must surely antedate both the “thesis” and the “antithesis,” however construed, we might experience a slight twinge of puzzlement. But this puzzlement doubtless stems from a too naive view of the dialectic. The dialectic isn't bound by mere chronology. We get to know “who's who” not by a mere round-by-round chronology, but by looking at the point-totals on the referee's score card. Never mind which rounds historicism “took”; all we need to know is that it won the decision—or at least that it would have won except that it had to be disqualified—as too punchy—too non compos—to carry the crown.

⁴ It should be pointed out that Miller's essay and, perforce the present criticism of it, both fail to take adequate account of some genuinely exciting problems about objectivity which currently do concern philosophers of science. In the main, these problems pivot on the nature of language—problems of meaning and reference—of translation and of coming to understand some linguistic context. But the chief contributors to articulation and advance on these problems are not “Millerian historicists” but rather people who, like Quine and Chomsky, are as far removed from such nineteenth-century antiscientific miasmas as could easily be imagined In short, not the least misleading aspect of Miller's account of recent philosophy is that it omits consideration of just what have been among the most vital of philosophical developments on meaning and objectivity. Another, and also vital, development vis-à-vis objectivity which Miller does mention, is one that has received greatest impetus in the work of T. S. Kuhn. But the usefulness of Miller's account is diminished here, too; for, as is indicated below (see, for example, footnote 10) Miller makes an erroneous assimilation of Kuhn's views to the historicism he has been outlining.

⁵ The strategy of the following sections will be to indicate that his interpretation suffers, if not a credibility gap, at least a Ghibelline Guelph.

⁶ In this respect, many of the works cited in these footnotes may be helpful: many of the works listed themselves contain extensive and useful bibliographies. (Indeed, even a moderately careful perusal of several of the works Miller mentions in his footnotes 3, 8, 14, and 24 would supply the reader with enough information to raise doubts about the character of his construal.)

⁷ I have intentionally indicated this problem or topic under what is perhaps its broadest rubric; for I mean to include not only the concern, which such positivists as Carnap had, with the distinction between cognitive and noncognitive meaning or statements and pseudo statements, not only the Popperian “demarcation problem” of the distinction between science and metaphysics, but also the deep philosophical underpinnings of those concerns which are to be found in the development of semantics and philosophy of language from Frege through Quine. For philosophy of science—and particularly for the philosophy of social science—the searching inquiries into the nature of such central formal or semantical concepts as reference, sense, truth, satisfaction (i.e., of a condition or of a variable), extension, intension, direct and indirect discourse, the relation of logical consequence, proof theory, model theory, axiomatization (and other modes of formalization), have been inquiries of fundamental importance. The results of Frege, Russell, Tarski, Carnap, Wittgenstein, and Quine in semantics and philosophy of language kept pace with, stimulated, and helped form important developments in major aspects of philosophy of science over, approximately, the past fifty years. During the past decades the importance of such inquiries is represented most strikingly, perhaps, by the role of Quine's work—from that relatively early influential critique of empiricism (by an empiricist), “Two Dogmas of Empiricism” (see this and other of his essays in the philosophy of language reprinted in his From a Logical Point of View [Cambridge: Harvard University Press, 1953]Google Scholar), to his brilliant exploration of the nature of meaning and translation in Word and Object (New York: John Wiley and Sons, Inc., 1960)Google Scholar, especially Chs. 1 and 2, and his subsequent elaborations of those explorations in Ontological Relativity and Other Essays (New York: Columbia University Press, 1969)Google Scholar. I will advert at least briefly to the importance of some of the aspects of this work of Quine's to philosophy of social science—to problems about “relativism” and “objectivity” in social science—in the text below. For the present, I wish simply to point out that treatment of Quine and Quine's contributions to philosophy of science during the past few years, though by any informed standard a major aspect of recent developments in philosophy of science, is omitted from Miller's account of what has been going on in the field.

⁸ “By 1950 positivism was virtually dead as a philosophical movement. It had come under strong attack even in the philosophy of science. The leading theory of knowledge by this time was one, i.e., historicism, whose foundations lie in the work of Kant and Hegel or, more precisely, in the radicalization of the Hegelian tradition which occurred in the latter half of the nineteenth century” (Miller, , “Positivism, Historicism, and Political Inquiry,” p. 796Google Scholar). Incidentally, as is abundantly clear from several of the immediately succeeding sentences, Miller's reference to the “radicalization of the Hegelian tradition” is to epistemological and ontological, not to political, radicalization.

⁹ I have deliberately listed Dewey as a “participant collaborator” in the evolution of what, I am arguing, is an ongoing empiricist (i.e., nonhistoricist) tradition in philosophy of science, because the “mutual influence” of which I speak is certainly documentable in this case, and because I think it is a mistake (one of Miller's mistakes of commission) to attempt to assimilate Dewey to historicism or to a nonobjectivist or irrationalist position in epistemology. For similar reasons, I have refrained from listing here the members of a very interesting recently developing historicalist (note, not historicist) “school” of philosophers and historians of science. I am referring to people like N. R. Hanson, Thomas Kuhn, Paul Feyerabend, Margaret Masterman, and Stephen Toulmin. The analysis which Miller suggests of the specific import of work like Kuhn's is also, I believe, a mistake of commission and I shall be adverting to it below.

See, in any case, the appropriate items cited in these footnotes for works of many of those just listed. These works bear out both the nonhistoricist positions of their authors and their contributions to the mainstream of the nonhistoricist development of contemporary philosophy of science. Not that the “mainstream” neglected, long ago, to consider or even embrace and then abandon theses which Miller attributes to historicism. The collections comprised in Ayer (Ayer, A. J., Logical Positivism [Glencoe, Illinois: Free Press, 1959]Google Scholar), and Schilpp (Schlipp, P. A., ed., The Library of Living Philosophers, Vol. XI, The Philosophy of Rudolf Carnap [LaSalle, Illinois: Open Court, 1963]Google Scholar) are of special interest, for they show how early in the history of logical empiricism a “dialectical” opposition on e.g., “the given” was internally of special interest. The former gives some of the original papers showing how, for example in the arguments between Neurath and Schlick (35–40 years ago!), the problems of sense-data statements, protocol statements, confirmations, and direct and indirect observation statements, were given searching scrutiny.

The conflicting positions, and the rigor as well as the vigor with which they were pressed, had a profound influence on some of Carnap's most important works. Carnap had started from a position, in the 1920s, significantly different from both Neurath's and Schlick's (see Carnap, R., “Autobiographical Essay,” in Schilpp, , The Philosophy of Rudolf CarnapGoogle Scholar). By the time he had produced the influential “Testability and Meaning,” Philosophy of Science, 3 (1926)Google Scholar and 4 (1937), he had arrived at a still different position under the influence of both Neurath and Schlick, and of others, as well as under the influence of his critical reflections.

Many of the essays in the Schilpp collection—particularly those by Charles Morris, Robert Cohen, Philipp Frank, Feigl, Ayer, and, above all, Carnap's autobiographical essay, together with his reply to his critics—copiously document the continuing development of the “standard position” in philosophy of science through stages which saw its positivist early fashioners themselves considering, themselves adopting, themselves criticizing, and themselves abandoning, both of the types of position that Miller sees as the protagonists in a conflict between positivists and “victorious” historicists. In particular, the substance of both the positions as he sketches them (pp. 798–801) have long been abandoned by the SP (“standard position”) in favor of considerably more sophisticated views. I have in mind, moreover, the more sophisticated views about all three of the “points” of conflict he mentions (pp. 800–801).

For example, regarding the notion that “the data of sensation are present to consciousness in their original or pure form” (ms. p. 800), SP philosophers have long abandoned such “metaphysical uses” of terms like ‘sensation’ in favor of interpreting them as technical terms of psychology. I know of no SP philosopher who would deny the well-evidenced finding in psychology, anthropology, biology, and common experience concerning ways in which sensation might be affected by cultural and, indeed, by innate—that is to say by genetic, including phylogenetic, conditions. (As early as 1925 Carnap had abandoned as untenable the notion of given “simple sense-data” and had done so explicitly “under the influence or the Gestalt psychology of Wertheimer and Köhler (Schilpp, , The Philosophy of Rudolf Carnap, p. 16Google Scholar).) Of course, the conclusions of historicism concerning the metaphysical import of this for the possibility of gaining knowledge of the “real world” were likewise rejected by SP philosophers.

Similar considerations apply to the view that our beliefs in any one epoch or culture are not affected by the cultural milieu in which they are set. I know of no SP philosopher who rejects the social-scientifically, well-evidenced thesis that such conditioning of our beliefs does take place. But, again, I know of none who would accept the metaphysical position which Miller indicates as the historicist metaphysical alternative, namely, that all knowledge “arises not so much from discovery of the real character of nature as from social or individual creation” (p. 800). Indeed, few, if any, SP philosophers would accept the dualism of the quotation at all—cleaving instead to the view that if there is anything which can properly be called “the real character of nature,” then “social or individual creation” are certainly a part of it, not excrescences divorced from nature.

Finally, with respect to Miller's third point of contradiction between positivism and historicism, the position of SP philosophers surely has long been a rejection of both the positions Miller sketches on “the relativity of truth” (p. 800). They surely reject the notion that “knowledge that is true or ‘objective’” is gained in some sense “by grasping nature as it really is.” They reject the Platonic metaphor of “grasping” entirely—just as they reject the metaphor of truth as consisting of “the congruence” between “thoughts and things.” Yet, the “relativity of truth” that they do accept is certainly not that of the historicist alternative. It is precisely the relativity of the semantical conceptions, fashioned in the 'thirties and before, by Tarski and Carnap. That is to say, ‘truth’ is construed as a predicate or term applicable to accounts of the world, i.e., applicable to sentences. But the concept of sentence is language-relative—a sentence may only be a sentence of some language or other. So, too, ‘true’ (and its cognates) is language-relative, and the full concept to be defined is always ‘true in L’ where L is some specific language.

The implications of this result of Tarski's have been immensely important in philosophy of language, in formal science, and in philosophy of science generally—but they lend no support whatever to the historicist views that Miller describes.

To sum up: the main point of this extended footnote has been to challenge the plausibility of Miller's view that the current situation in philosophy of science finds historicism holding the field after mounting a successful revolution against positivism on three specific “points of contradiction.” In contrast, I have been suggesting that there has been, and continues to be, an evolution of a prevailing or standard position in the field, fashioned by the ongoing internal critique of such nonhistoricist philosophers of science as those named above.

Of course neither ‘revolution’ nor ‘evolution’ is the clearest term to have come down the ideological pike. The continuing rebuilding of the ship of philosophy of science while it is under weigh (to change metaphors and borrow the figure from Neurath which has been used so effectively by Quine) has yielded a remarkably changed vessel. So, if one wants to call the craft, now afloat, a revolutionary rather than an evolutionary outcome, why, it would be pointless to quarrel. Whatever the epithet, the one thing that is clear is that the craft is still manned by the extended crew which has been working it over for decades and that crew is not historicist.

¹⁰ This provincialism is sharply in evidence in his discussion of the “historicalist school” (which could, as well, be called “the history-of-science-school”) of philosophers of science (ms. p. 804). Here, Miller fastens upon the interesting views, critical of both the method and some of the substantive results that have been put forward by SP philosophers. The history-of-science-school is a convenient label for a group of writers, of whom Hanson, N. R. (Patterns of Discovery. Cambridge: Cambridge University Press, 1958)Google Scholar, Feyerabend, Paul (“Problems of Empiricism” in Colodny, R., ed., Beyond the Edge of Certainty, Englewood Cliffs, New Jersey: Prentice-Hall, 1965)Google Scholar, Toulmin, Stephen (The Philosophy of Science, London: Hutchinson, 1953Google Scholar; and Foresight and Understanding, London: Hutchinson, 1961Google Scholar), and Thomas Kuhn have been especially influential. Kuhn's, The Structure of Scientific Revolutions (Vol. II, No. 2 of Neurath, Otto, ed., The International Encylopedia of Unified Science. Chicago: University of Chicago PressGoogle Scholar; a second edition of Kuhn's book revised and enlarged by a new chapter entitled “Postscript 1969” published as a Phoenix Paperback, Chicago: Chicago University Press, 1970)Google Scholar, in particular, has been widely read and commented on. (The book has gained currency outside “the field”. In numbers, diversity, and perhaps acumen, its heterogeneous reading public appears to have outstripped considerably the reading public of many of the essays on it—including some containing Kuhn's own later criticisms and modifications of his theses.)

Miller's appraisal of the signifiance of this line of development and these works, particularly Kuhn's contributions, is unconvincingly narrow on at least three counts. First, Kuhn's arguments about the nature of scientific revolutions are focused on just one problem area in a field which has a large number of other significant problem areas occupying attention. Even if Miller were correct in his apparent appraisal of the cogency of Kuhn's views, no case whatever has been made (and it would have to be a strong and detailed one to carry much weight) that prevailing views or positions in these other areas had been overturned or replaced.

In the second place, Miller is unduly narrow in not taking account of the critical treatment, and the cogency of that treatment, which Kuhn's views (and the views of other of the “historicalists”) have received and which have tended to vitiate the impact of those views even within their circumscribed problem areas. A wider and deeper account of the literature (in, say, the last ten years) would surely temper the notion that in confrontation with such arguments as Kuhn has put forward, the standard position has been shattered and superseded by an historicist position. For a discussion of the weaknesses occurring at many of the pivotal junctures in Kuhn's arguments (and other historical arguments) see, for example, the discussions by Watkins, Toulmin, Popper, Lakatos, and indeed, Kuhn, himself, in Criticism and the Growth of Knowledge, ed. Lakatos, I. and Musgrave, A., (Cambridge: Cambridge University Press, 1970)Google Scholar. Shapere's, Dudley “The Structure of Scientific Revolutions,” Philosophical Review, 73 (1964), 383–394CrossRefGoogle Scholar and “Meaning and Scientific Change” in Mind and Cosmos, ed. Colodny, R. (Pittsburgh: Pittsburgh University Press, 1966)Google Scholar, cited but not expounded by Miller, provide careful assessments of Kuhn's arguments which show them to be uncompelling. The doubts Shapere raises about crucial aspects of the “historicalist” arguments have still not been resolved by them. If, though, there is the opportunity for only one detailed assessment of the ramifications of positions like Kuhn's and positions like historicism for the concept of objectivity in science, I would urge careful attention to Israel Scheffler's book, Science and Subjectivity (Indianapolis: The Bobbs Merrill Co., Inc., 1967)Google Scholar. It contains an incisive and meticulous analysis of what some of the better articulated arguments about objectivity in science come to. They are evaluated lucidly and their defects, where they occur, are exhibited in a way which enhances our understanding both of the nature of claims for and against objectivity in science, and also of some of the problems about objectivity which remain as challenges to inquirers.

A third manifestation of what I've been calling the “narrowness” or “provincialism” of Miller's perspective on the current status of philosophy of science is to be found in its simplistic assimilation of Kuhn's position to the irrationalist historicism which Miller delineates. It seems fairly clear that Kuhn himself would reject the charge as Miller formulates it. See, especially, Kuhn's discussion of the charges of irrationalism and relativism in his elegantly titled second essay, “Reflections on my Critics,” in Lakatos, and Musgrave, , pp. 259–266Google Scholar. And consider by way of sample, the following remarks by Kuhn in the essay:

“Members of a scientific community can, I am held to have claimed, believe anything they please if only they will first decide what they agree about and then enforce it both on their colleagues and on nature. The factors which determine what they do choose to believe are fundamentally irrational, matters of accident and personal taste. Neither logic nor observation nor good reason is implicated in theory-choice. Whatever scientific truth may be, it is through-and-through relativistic.

“These are all damaging misinterpretations ….” (p. 260) (emphasis added).

“To name persuasion as the scientist's recourse is not to suggest that there are not many good reasons for choosing one theory rather than another. It is emphatically not [emphasis in original] my view that ‘adoption of a new scientific theory is an intuitive or mystical affair, a matter for psychological description rather than logical or methodological codification.’ On the contrary, the chapter of my Scientific Revolutions from which the preceding quotation was abstracted explicitly denies ‘that new paradigms triumph ultimately through some mystical aesthetic,’ and the pages which precede that denial contain a preliminary codification of good reasons for theory choice. These are, furthermore, reasons of exactly the kind standard in philosophy of science: accuracy, scope, simplicity, fruitfulness, and the like” (p. 261) (emphasis added).

“… no part of the argument here or in my book implies that scientists may choose any theory they like so long as they agree in their choice and there-after enforce it. Most of the puzzles of normal science are directly presented by nature [emphasis added], and all involve nature indirectly. Though different solutions have been received as valid at different times, nature cannot be forced into an arbitrary set of conceptual boxes. On the contrary, the history of proto-science shows that normal science is possible only with very special boxes, and the history of developed science shows that nature will not indefinitely be confined in any set which scientists have constructed so far” (p. 263).

“… I have not previously and do not now understand quite what my critics mean when they employ terms like ‘irrational’ and ‘irrationality’ to characterize my views. These labels seem to me mere shibboleths, barriers to a joint enterprise whether conversation or research. My difficulties in understanding are, however, even clearer and more acute when these terms are used not to criticize my position but in its defense. Obviously there is much in the last part of Feyerabend's paper with which I agree, but to describe the argument as a defense of irrationality in science seems to me not only absurd but vaguely obscene, [emphasis added] I would describe it, together with my own, as an attempt to show that existing theories of rationality are not quite right and that we must readjust or change them to explain why science works as it does. To suppose, instead, that we possess criteria of rationality which are independent of our understanding of the essentials of the scientific process [emphasis added] is to open the door to cloud-cuckoo land” (pp. 263–264).

These Quotations merely indicate Kuhn's view of the ascription of irrationalism to his view of science—and, of course, Kuhn may be wrong about the import of his own views. Prima facie, however, they call for a stronger case than Miller has thus far provided for assimilation of Kuhn to the irrational historicism that he describes.

To sum up: it may be a too narrow perspective which leads Miller to claim, without sufficient warrant, (1) that the cogency of a critique of a standard position on one problem area would shatter the standard positions taken over the entire field of philosophy of science; (2) that the critique is cogent and has had that effect even in its “own” problem area; and (3) that the basis of the critique can be assimilated to an irrationalist historicism which he delineates.

¹¹ Paul Feyerabend comes closer, perhaps, than any other of the philosophers, referred to by Miller (and others, e.g., Shapere) as being “in revolt” against the “standard view,” to maintaining the “irrationalist-relativist” position which Miller describes as historicism. Neither the precise import of Feyerabend's remarks in this regard, nor what he intends is always clear. (Sometimes his essays have the character rather of performances than of arguments—this would by no means, incidentally, be regarded as a pejorative ascription by Feyerabend. See, e.g., his “Against Method: Outline of an Anarchistic Theory of Knowledge” in Feigl, H., Scriven, M., and Maxwell, G., general eds., Minnesota Studies in the Philosophy of Science [Minneapolis: University of Minnesota Press], Vol. 4, ed. Radner, M. & Winokur, S. [1970]Google Scholar, “Philosophy of Science, A Subject with a Great Past” in Minnesota Studies …, Vol. 5 [1970]Google Scholar, and “Consolations for the Specialist” in Lakatos, and Musgrave, , eds., Criticism and the Growth of Knowledge.Google Scholar) Feyerabend has contributed essays (some very long ones, in fact) to several volumes in the Minnesota Studies series and to other of the “establishment” series in philosophy of science. Despite his coming closer to being a “Miller-historicist” than anyone else of significant influence presently among the historicalists, it would again be a distortion to assimilate even him to the irrationalist-relativist model—it would be a mistake to equate his passionately expressed commitment to pluralism, the “principle of proliferation” (see “Against Method,” p. 26 and passim)—simplistically with irrationalism. One key to the actual complexity of Feyerabend's views is to be found in his alignment of them, explicitly, with J. S. Mill's reasoned defense of pluralism of intellectual, artistic endeavors and life styles within any society which hopes to avoid the suppression of truth or happiness. What may have evaded Miller's attention is that the kind of irrationalist, relativist historicism he is describing would be epitomized bv successive “monolithic” world outlooks—each coordinated to the epoch or society or culture to which it is relative. But such a view, I suggest, may be incompatible with (or at anv rate, quite different from) the methodological pluralism Feyerabend espouses. For the latter is to be a pluralism within any one epoch or society or culture. One hardly can see J. S. Mill as the intellectual or stylistic hero of Millerian-historicism; and Feyerabend's enthusiasm for Mill should make us suspicious of any facile categorization.

I believe that Feyerabend's most important contributions have been the additions that his astute and lively historical analyses (of concrete instances of the history of theory-change and theory-choice) have often made to the long continuing body of work constituting an internal critique (from at least Neurath forward) of the “observation-theory” distinction in philosophy of science. I do not think that close attention to his specific arguments about concrete cases—often very illuminatingly presented—would warrant the conclusion that he eschews rationality, considerations of evidence, scope, power, simplicity, openness to test, to criticism, and many other of the quite traditional touchstones of scientific appraisals of theories.

The burden then of this note, as was the burden of the two preceding ones, is to suggest that Miller's vision of a revolution in the enterprise is a dubious vision: for, among other reasons, the group of philosophers of science he fastens on, by no means uniformly, individually or collectively, hold views than can be easily equated with his historicism. Moreover, the maior views held by this group have by no means come to dominate contemporary philosophy of science or (I should say ‘and a fortiori’) philosophy at all—even if their views were to be found compelling, their thrust or comprehensiveness is by no means coextensive with the whole of philosophy of science. And finally, their views, or at least major aspects of them, and the vehicles they choose for their expression of them (e.g., note that Kuhn's, T. S.The Structure of Scientific Revolutions is Vol. II, No. 2Google Scholar of the International Encyclopedia of Unified Science) are more plausibly to be construed as evolutionary changes, or attempts which, if successful, would constitute further evolution of, rather than revolutions against, the evolving mainstream in the last 50 years.

¹² Hempel himself, in a footnote (“On the ‘Standard Conception’ of Scientific Theories,” in Minnesota Studies, Vol. 4, p. 146Google Scholar) provides a list of works which, in recent times, represents “characteristic stages in the evolution” of the “standard” construal of scientific theories, and I cite the list he gives here: Campbell, N. R., Physics: The Elements (Cambridge: Cambridge University Press, 1920)Google Scholar, chapter 5; Ramsey, F. P., “Theories” (in Ramsay, , The Foundations of Mathematics and other Logical Essays. London: Routledge and Keegan Paul, 1931)Google Scholar; Carnap, R., Foundations of Logic and Mathematics (Vol. I, No. 3 of The International Encyclopedia of Unified Science. Chicago: University of Chicago Press, 1939), especially sections 21–25Google Scholar; Braithwaite, R. B., Scientific Explanation (Cambridge: Cambridge University Press, 1953)Google Scholar, Chapters I-III; Carnap, R., “The Methodological Character of Theoretical Concepts,” Minnesota Studies, Vol. 1, ed. Feigl, H. and Scriven, M. (1956), pp. 38–76Google Scholar; Carnap, R., Philosophical Foundations of Physics, ed. Gardner, M. (New York: Basic Books, 1966)Google Scholar, part V; Nagel, E., The Structure of Science (New York: Harcourt Brace & World, 1961)Google Scholar, chapters 5 and 6.

Hempel is, however, too modest inasmuch as he has omitted two important landmarks in the evolving path of the standard construal: his Fundamentals of Concept Formation in the Empirical Sciences (Vol. II, No. 7 of The International Encyclopedia of Unified Science, 1952)Google Scholar and his “The Theoretician's Dilemma: A Study in the Loeic Theory Construction” in Minnesota Studies, Vol. 2, ed. Feigl, H. & Maxwell, G. (1958)Google Scholar. The brief sketch of the standard position's evolution I give is wholly indebted to the list just cited and numerous other works too; as indicated above; the sketch entails, perforce, very severe simplifications.

¹³ So early, in fact, as to be mythological. I know of no one in the history of logical empiricism (certainly no one in the last 40 years—and, indeed, not even Hume) who actually held the view being outlined. It seems to have been used often (as is being done above) as a didactic or expository device which (perhaps like the “contract theory” of the formation of polities) allows for brief (and dangerous) shortcuts to the illumination of more complex issues. (For one version, of such an idealized account, which bears on the social sciences see Rudner, R., Philosophy of Social Science [Englewood Cliffs: Prentice-Hall, 1966]Google Scholar, chapter 2.)

¹⁴ For an important and illuminating discussion of criteria for, and problems of, selecting primitive bases, see Goodman, N., The Structure of Appearance (Cambridge, Mass.: Harvard University Press, 1951)Google Scholar, chapter 3.

¹⁵ I am deliberately disregarding any differentiation here between ‘sensing’ and ‘observing’ or between ‘data of direct sensation’ and ‘observed data’. Such differences come to be important, however, in the ensuing historic discussions.

¹⁶ Only thus could another part of the program attributed to this mythological, primeval stage of logical positivism be carried out: the “strong verificationist” tenet to the effect that an hypothesis or a theory be itself logically implied by the set of observation consequences which verified it.

¹⁷ Carnap's views in particular were gradually modified during the 'twenties and early 'thirties under the impact of Neurath's critique and the results of his own continued deep explorations into logical problems of theory and formal language construction. See, in particular, his accounts in Schilpp, , The Philosophy of Rudolf Carnap, pp. 16–24, 863–865, 882–886Google Scholar; for the influence of philosophers, other than Neurath, with whom Carnap held discussions (including, incidentally, many philosophers such as Charles Morris, Ernest Nagel, Sidney Hook, and C. I. Lewis who were in the pragmatist tradition) see, e.g., Schilpp, pp. 29–36, 40, and 860–862. Now, as Carnap points out (Schilpp, p. 50), under the influence of such philosophers as Mach and Russell, he very early (at the time of Der Logische Aufbau der Welt (Berlin: Weltkreist-Verlag, 1928)Google Scholar, pursued the quest for certainty believing then that “the task of philosophy consists in reducing all knowledge to a basis of certainty.” However, he also goes on to add that in “the Vienna [Circle] discussions my attitude changed gradually. ….” Moreover, it is true that certain traditional empiricist doctrines influenced Carnap to believe that this “basis of certainty” was to be found in the “immediately given” data of sensation. Yet, to identify the general view itself as empiricism is again to be unduly simplistic: such an identification ignores, on the one hand, the fact that the tradition of the quest for certainty (as an attempt at “reducing all knowledge to a basis of certainty”) has its main roots in Descartes and Leibnitz (not to say Plato!); on the other hand, it ignores the strong elements of corrosive scepticism in such empiricists as Hume—and, indeed, ignores a long antecedent tradition of empirioscepticism back to classical times. In any event, that quest for certainty, along with any notion of the epistemological primacy of phenomenalist bases was early abandoned. In the 'thirties and before, even Carnap was holding that such a phenomenalist claim was “meaningless metaphysics.” Despite the early and repeated published demurrers, however, it has always been easier for superficial readers of the ongoing tradition to identify positivism with the vulnerable “childhood” position of just some of its proponents, rather than to make instead the arduous effort required to understand the technically complex ways in which those positions were evolving in even the early days. Apropos of this, there is a touch of rueful amusement in Carnap's account (Schilpp, p. 38) of a conversation with Einstein at Princeton (c. 1952).

In the light of the last sentence one is again struck by how odd will seem a contrast which has not been very severely and precisely qualified, between an absolutist positivism on the one hand and a relativist historicism on the other. In its light, also, aspects of the stance that Miller takes toward the contrast become implausible to anyone closely familiar with the development and present situation of positivism or, better, with the development of the standard view in philosophy of science.

¹⁸ The device of the reduction sentence is introduced explicitly by Carnap in 1936 in “Testability and Meaning.” Under the impact of criticisms of this “way out” by Goodman, (Fact, Fiction and Forecast, 2nd edition, New York: The Bobbs-Merrill Co., Inc., 1965)Google Scholar, Scheffler, (The Anatomy of Inquiry, New York: Alfred Knopf, 1963)Google Scholar, Hempel (“The Theoreticians Dilemma”), Pap, and others, Carnap came ultimately to modify his position on reduction—in a direction that might be fairly characterized as even greater liberalization. See his “The Methodological Character of Theoretical Terms” (1956) and “The Philosopher Replies” (in Schilpp, , The Philosophy of Rudolf Carnap, pp. 958–966)Google Scholar. Note that neither the 1936 liberalization or the subsequent ones, though of considerable importance in the evolution of the “standard view,” can properly be called “historicist changes” or “historicist-caused changes.”

¹⁹ See Feigl, , “The ‘Orthodox’ View of Theories: Remarks in Defense as well as Critique” in Minnesota Studies, Vol. 4 (1970), p. 3Google Scholar and early works by both Carnap and Schlick cited by Feigl.

²⁰ Feigl, , “The ‘Orthodox’ View of Theories …,” p. 6Google Scholar. This exposition and critique is put forward as a common basis for the discussion of the contributors to the volume who are focusing upon aspects of the standard view.

²¹ These “bridging sentences” (as Hempel now calls them) or their counterparts have been variously described in the literature as “dictionary sentences,” “operational definitions,” “coordination rules,” “correspondance rules.” One of Hempel's contributions to the field has been his perspicuous elaboration of this conception of theories over a period of decades. His formulation and discussions of this conception in Fundamentals of Concept Formation in Empirical Science, “The Theoreticians Dilemma,” and Aspects of Scientific Explanation (New York: The Free Press, 1965)Google Scholar, together with those of Braithwaite in Scientific Explanation and Nagel in The Structure of Science, are, perhaps, the most widely known and influential formulations both inside and outside the field. One of Hempel's conclusions has been that the sentences of a theory—in particular, its theoretical sentences—cannot be considered to be meaningful or true or false in isolation from other sentences of the theory—that the unit of meaningfulness or “cognitive significance” may best be construed as the theory or theoretical system as a whole. (See, e.g., “Empiricist Criteria of Cognitive Significance” in Aspects of Scientific Explanation, p. 117.Google Scholar) In this regard, Carnap appears to have continued to differ with Hempel (see Carnap's, “The Philosopher Replies” in Schilpp, , The Philosophy of Rudolf Carnap, pp. 961–966)Google Scholar; but, in any case, the position Hempel has now come to is a departure both from his older, and from Carnap's, versions of the standard conception.

²² The Quinian notions (see Quine's, W. V.Word and Object [N.Y.: John Wiley, 1960]Google Scholar; and Ontological Relativity and Other Essays [N.Y.: Columbia U. Press, 1969]Google Scholar) which support this sort of further modification by Hempel of the standard view also underlie his regarding historicalist positions like those of Feyerabend, and Kuhn, with doubt (see Minnesota Studies, Vol. 4, pp. 155–156Google Scholar, and also in the same Studies volume, Hempel's, remarks in the “Discussion at the Conference on Correspondance Rules,” pp. 223–227, 253–254Google Scholar).

²³ Hempel's proposed modifications, however important, and however clearly “departures from orthodoxy” still, it will be noted, regard theories as deductively systematized sets of statements. The proposal does not eschew all axiomatization in suggesting it be “appropriate axiomatization,” and while discarding the theoretical-observational dichotomy among terms, his proposal replaces it with a “new theoretical” vs. “antecedently available” term dichotomy. A close perusal of the essay will, in fact, further bear out that it is much more a continuation of Hempel's development of the standard view than it is a radical or revolutionary (or, for that matter, in any sense a historicist) total break with that developing tradition.

²⁴ Kuhn, himself, may have come somewhat belatedly to an increasing recognition of the bearing of Quine's results upon his (Kuhn's) arguments about the nature of what is involved in coming to accept a new theory. See especially his “Reflections on my Critics” in Lakatos, and Musgrave, , eds., Criticism and the Growth of Knowledge, pp. 266–271 and 275–277Google Scholar, for a discussion of Quine's results and of “the utility, indeed the importance of the linguistic parallel” (p. 267) to problems of theory change in the history and philosophy of science.

Incidentally, I am refraining from anything more than the reminder of the importance of Chomsky's brilliant and original work which is conveyed simply by the mention of it. For that work has had fairly wide currency outside of philosophy. Perhaps, though, it would be suitable just to remark that though Chomsky's work is, clearly, one of the developments of significance in the problem area, and though it is, in an interesting sense antiempiricist, it is by no means irrationalist or historicist; it is by no means antiscientific. Miller gives this work no consideration.

²⁵ Winch's, PeterThe Idea of a Social Science (New York: Humanities Press, 1958)Google Scholar, is an interesting exception—at least as far as his explicit claim to derive from Wittgenstein is concerned. On the other hand, it's not at all clear that that claim is justified—there are some grounds for believing Winch to have misconstrued Wittgenstein—in particular in his contention (a sort of “social-privacy” view that might well have been rejected by Wittgenstein) that speakers of a language are the only possible adequate observers of the linguistic behavior of those speakers.

²⁶ In speaking explicitly of the omission of consideration of Quine and Chomsky, I do not, of course, mean to suggest that Miller's account is guiltless of other important omissions as well. A particular case in point, again, is Goodman's important work on criteria for accuracy and adequacy of definitional, or explicational systems (see The Structure of Appearance, chapters 1–3), as well as his work bearing upon problems of choices of hypotheses (e.g., in Fact, Fiction and Forecast, chapters 3 and 4). Of great importance in this problem area, too, has been the work of such philosophers (not alluded to in Miller's account) as Wilfrid Sellars, Gustav Bergmann, J. L. Austin, Bertrand Russell, Rudolf Carnap, Alonzo Church, Alfred Tarski, Donald Davidson, Israel Scheffler and many others who have contributed to continuing work on the concepts of meaning and reference and upon the implications of these concepts for problems in philosophy of science.

²⁷ I wonder if David B. Truman, who gave a characteristically judicious (wholly unfrenzied) welcome to behavioralism in his 1955 Brookings Lecture, “The Impact on Political Science of the Revolution in the Behavioral Sciences,” would recognize the “baby” from the description we've garnered here.

²⁸ It is, in the rhetoric of damnation from early, or biblical, positivism, a pseudoquestion.

²⁹ Still we should, at least in a footnote, point out that Miller must surely in (e) intend ‘testable’, rather than ‘tested’—otherwise the view he intends to portray would be too far removed to be even called a caricature of positivism.

³⁰ I am attending here to all but the third characterization of behavioralism (3), from Surkin. Surkin seems to be berating behavioralism, or positivism, for expressing the wrong ideology or, at any rate, for giving aid of positivism. and comfort to, and generally tending to work at “preserving” the existing society. It may be the case that, as a matter of fact, behavioralists have manifested this “conservative” political cast—but they surely need not, on the basis of their methodological positions, have done so; they surely are not barred by their position from pursuing scientifically whatever problems the anti-behavioralists pursue.