the scientific dimensions of social knowledge and their distant echoes in 20th-century american...

E-mail address: pmirow

0039-3681/$ - see front ma

doi:10.1016/j.shpsa.2003.1

[email protected] (P. Mirowski).

tter # 2004 Elsevier Ltd. All rights reserved.

1.002

Stud. Hist. Phil. Sci. 35 (2004) 283–326

www.elsevier.com/locate/shpsa

The scientific dimensions of social knowledgeand their distant echoes in 20th-century

American philosophy of science

Philip Mirowski

400 Decio Hall, University of Notre Dame, Notre Dame, IN 46556, USA

Received 14 August 2003; received in revised form 19 November 2003

Abstract

The widespread impression that recent philosophy of science has pioneered exploration ofthe ‘‘social dimensions of scientific knowledge’’ is shown to be in error, partly due to a lackof appreciation of historical precedent, and partly due to a misunderstanding of how thesocial sciences and philosophy have been intertwined over the last century. This paperargues that the referents of ‘‘democracy’’ are an important key in the American context, andthat orthodoxies in the philosophy of science tend to be molded by the actual regimes ofscience organization within which they are embedded. These theses are illustrated by con-sideration of three representative philosophers of science: John Dewey, Hans Reichenbach,and Philip Kitcher.# 2004 Elsevier Ltd. All rights reserved.

Keywords: Social dimensions of science; Logical positivism; Democracy; Context of discovery/justifi-

cation; Goals of science

1. Introduction: what do philosophers of science know about the ‘‘social dimensions

of scientific knowledge’’?

In The Stanford encyclopedia of philosophy, the entry for ‘The social dimensions

of scientific knowledge’ makes the following observation: ‘Since 1980, interestin developing philosophical accounts of scientific knowledge that incorporatethe social dimensions of scientific practice has been on the increase’. Many

P. Mirowski / Stud. Hist. Phil. Sci. 35 (2004) 283–326284

philosophers of science regard themselves as participating in the ‘growing recog-nition of the social character of scientific inquiry’.1 While this impression is appar-ently quite widespread amongst the philosophy of science community, I shall arguein this paper that it is factually misleading, and furthermore, symptomatic of someof the ways in which the American philosophy of science community in the 20thcentury has revealed itself either unwilling or incapable of regarding both the scien-tists they describe and themselves as integrated into a larger set of social structuresof inquiry. (One would find it more difficult, I suspect, to make a similar case forcontinental European philosophy of science.) There are at least four problems withthis self-congratulatory opinion:1) It is historically inaccurate. A broader perspective will reveal that the philo-

sophy of science is sometimes better regarded from the perspective of a subset ofsocial science, especially in the 20th-century American context, in the sense that ithas always participated in larger trends in social theory, making conscious orunconscious use of their doctrines. As such, it has always subscribed to someaccount or another of the ‘‘social dimensions of science’’.2) It takes ‘‘society’’ as an unproblematic entity. But the account of ‘‘society’’

which has informed dominant schools of the philosophy of science has notremained constant or persisted as a fixed premise within their larger conceptualframeworks. Indeed, nagging suspicions about the legitimacy of the ontologicalentity ‘‘society’’ constitutes a major subtext of our story, especially in Americanphilosophy of science. Hence we can identify at least two ‘‘ruptures’’ or profoundrevisions in the ways philosophers would see themselves as forging accommoda-tions with the ‘‘social’’ in the 20th century. One occurs from the later 1930s to theimmediate post-World War II period, and another in the 1980s.3) It represses the fact that philosophy of science as an academic formation has

always been pitted in competition with other academic disciplines which also seekto make various pronouncements upon the social significance of science, the epis-temological efficacy of scientific method, and the preferred modes of funding andorganization of scientific communities. The ‘‘science policy’’ community wasperhaps the most significant rival throughout the 20th century, but sociologists ofscience have also presented themselves as candidates, as one cannot help but noticefrom the surfeit of books and articles by philosophers seeking to reprimand them.2

One might suspect that interpretative charity should ideally begin at home, withphilosophers of science themselves coming to acknowledge that the conditions oftheir own academic professionalization as a discipline had more than a little to dowith some major shifts in the structure and conduct of their own inquiries.

1 The first quote (Longino, 2002c) may be found at http://plato.stanford.edu/archives/sum2002/

entries/scientific-knowledge-social/. The second quotation is from Longino (2002a, p. 1), but could just

as easily have come from Alvin Goldman, Joseph Rouse, Philip Kitcher, Miriam Solomon, or, indeed,

any of a wide range of contemporary philosophers. See Schmidt (1994).2 The science policy literature may be sampled in the flagship journal Research Policy. For a sample of

philosophers’ reprimands of sociologists of science, see Brown (2001); Kitcher (1993, 2000, 2001);

Koertge (1998), and Wray (2001).

http://plato.stanford.edu/archives/sum2002/entries/scientific-knowledge-social/

http://plato.stanford.edu/archives/sum2002/entries/scientific-knowledge-social/

285P. Mirowski / Stud. Hist. Phil. Sci. 35 (2004) 283–326

4) It is neglectful nearly to the point of oblivion about the conditions which

prompt the existence of watersheds in the organization of science such as the more

recent experience in the 1980s. A relevant congenital tic of the American philo-

sophy profession (although, it must be conceded, not its alone) is a demonstrated

unwillingness to regard science as an historically changing entity, not just in the

realm of epistemic ‘‘values’’ but also in terms of actual social structures. Science

takes place in very specific settings, and is paid for in very specific ways; and scien-

tific findings are disseminated in highly structured and regimented formats. None

of these has remained constant throughout the 20th century, and almost none has

been subjected to sustained scrutiny by philosophers of science. Of course, there

have been notable exceptions to this generalization; but in those instances, I believe

it is most instructive to see how philosophers allowing for generational changes in

scientific structures have been treated as pariahs by their peers. I am thinking here

especially of Paul Feyerabend, who was vilified in his lifetime as a dangerous

enemy of science (Preston, Munevar, & Lamb, 2000).This paper is dedicated to the memory of Paul Feyerabend; whereas one might

reject his remedies for what he believed ailed science, he at least knew how to

confront the issue of the relationship of science to society head-on as the rightful

province of an engaged philosophy of science. Here again we find the self-

congratulation of modern philosophers is unwarranted. He understood, as they

apparently do not, that communal considerations were not a mere accessory, a

‘‘social dimension to scientific knowledge’’, but instead the scientific dimension of

social knowledge that deserved scrutiny and critique. Skeptical of the contributions

of contemporary academic ‘‘social science’’ to the discussion, he nevertheless

situated his books squarely within the province of explicit politics, and then

worked outwards to inquire how it was that various beliefs about the nature of

science served certain interests and frustrated others. He was not only an advocate

but practised his own prescription that ‘the history of ideas is an essential part of

the scientific method’ (Feyerabend, 1999, p. 214). He understood that one could

not and should not rest satisfied with talk of ‘‘society’’ as a whole, but rather had

to look hard at which intellectual constituencies were aligned with particular

groups, and then evaluate their transcendentalist claims. Writing in the midst of

the Cold War, he suggested ‘Church and State are now carefully separated. State

and Science, however, are clearly tethered’ (Feyerabend, 1978, p. 74). Had he but

lived into the 21st century, he would have confirmed his initial hypothesis that no

scientific alliances endure forever (this being the real message of Against method),

and that now, if anything, Church and State are much less partitioned off than

they once were in America, but that State and Science, if not exactly divorced from

one another, are certainly exploring a trial separation. The omitted fourth party in

the menage a quatre, throughout his own work as well as in the philosophy of

science generally, was the private corporation. This shifting pattern of alliances

over the century is the Rosetta Stone which will allow us to recover the longer-

term interplay of social knowledge and the philosophy of science.


2. The bigger picture: regime changes

Lately there seems to be a belated attempt to provide the consideration of the‘‘social dimensions of science’’ within the philosophy of science with a genealogy,but it has assumed a most curious configuration.3 For some reason, everybodytends to start out with John Stuart Mill’s essay On liberty. This, in itself, mightconstitute a harmless custom, in that all points of departure are to some extentarbitrary, except for the fact that most everyone (with the possible exception ofNancy Cartwright) tends to ignore his actual writings on what would now becalled the philosophy of science, and show almost no interest in his political andeconomic doctrines, which Mill himself thought part and parcel of his philosophi-cal position. One wonders whether these Anglo-American searchers after theirroots are more attracted to the author of Utilitarianism and the advocate of a‘‘negative conception of freedom’’, or alternatively drawn, as Isaiah Berlin put it,to the prophet of an ‘almost Hegelian distrust of simple models and cut-and-driedformulae to cover complex, contradictory, and changing situations . . . [He courted]Fallibility, the right to err, as a corollary of the capacity for self-improvement;distrust[ed] symmetry and finality as enemies of freedom’ (Berlin, 1969, pp. 193,192). We may never fully grasp the exact character of the motives behind thisparticular point of departure, because the bulk of those who cite Mill pass over hisideas post haste, possibly because they may harbor a suspicion that his capacioussense of the openness of inquiry may not in fact have much correspondence to thepractices of modern science.4

Once past Mill, this genealogy tends to tip a hat to Charles Sanders Peirce, butnever lingers there, since all and sundry confess themselves perplexed at the luxur-ious indiscipline of his massive oeuvre. Then, the narrative displays a rather yawn-ing gap, usually absent any comment, leading up to contemporary figures thatphilosophers of science deem to be dealing with social epistemology in a seriousand responsible manner.5 The impression imparted by these genealogies is thatonce professional philosophers stumbled directly upon the social dimension of

3 This is based upon comments in the aforementioned entry in the Stanford encyclopedia, combined

with Longino (2002a, pp. 3–7), and Kitcher (2001). An even more recent concern to reinstate the logical

positivist claim to have a rich tradition of consideration of ‘‘social dimensions’’ (Richardson, 2000, 2002;

Uebel, 2000b) is covered in later sections of this paper.4 For example, see Kitcher (2001, pp. 106–107). By contrast, Feyerabend counted himself a true

disciple of Mill, by starting from the premise that the contemporary organization of scientific research

was not ideal. See Feyerabend (1999, Ch. 11; and Feyerabend, 1987). Nevertheless, even such an ardent

philosopher of freedom of inquiry as Mill ultimately had to resort to social distinctions to help quaran-

tine off those who might be considered illegitimate interlocutors, or as he put it, ‘Despotism is a legit-

imate mode of government when dealing with barbarians’. This repressed aspect of his politics of

epistemology is discussed with great perspicuity by White (1989).5 Sometimes various authors attempt to render the gap less daunting and embarrassing by inserting

Karl Popper into the breach. An interesting argument by a Popperian that Popper did not in fact

discuss actual social structures of science, even though that would have been the most logical impli-

cation of his position, can be found in Ian Jarvie (2001a,b). For the purposes of limiting our exposition

to the American context, we shall regretfully have to pass this issue by in the present paper.


science—which they had previously unaccountably overlooked—they judiciouslydevoted some of their scarce attention to it and made their peace with it.More recently, a few specialists in the history of philosophy of science have

pointed out that this sketchy account is really untenable.6 Rather, a more sensitivereading of the literature reveals that the ‘‘social dimension’’ was never altogetherabsent. Confining ourselves only to 20th-century Anglo-American texts (itselfintroducing a substantial bias), it has been observed that the American Pragmatistshad quite a bit to say about the relationship of science to society in the first thirdof the century; John Dewey was regarded as a major public spokesman for theinterplay of science and social values, and this is an area we shall explore in somedetail below. In Britain, the ‘Social Planning of Science’ movement associated withJ. D. Bernal and Patrick Blackett during the Depression argued for a reorientationof science towards greater consideration of public welfare; this provoked a majorresponse on the part of such figures as Michael Polanyi and Friedrich Hayek(McGuckin, 1984). Philosophers of science ignore these writings at their peril, aswe shall argue below; the presence of a number of Nobel Prize winners shouldsignal that something significant was going on here. Researchers such as ThomasUebel, Alan Richardson, George Reisch, and Don Howard have of late been reco-vering the ‘‘social’’ content of the logical positivist movement; and this will requirerather more attention than has been devoted to the phenomenon so far. The dis-placement of the Pragmatists by the logical positivists in the American scene inter-sects with the question of the professionalization of the field of the philosophy ofscience after World War II, and coinciding with the Cold War reorganization ofAmerican science, an issue raised by John McCumber in his Time in the ditch(McCumber, 2001). Contemporaneously with these events, science policy managerssuch as Vannevar Bush, Warren Weaver, and James Conant weighed in with theirown narratives about the place of science in society, and far from being restrictedto bureaucratic contexts, their prognostications were also felt back in the philo-sophical discussion, as we shall observe below. Perhaps less familiar is a spate ofwritings on the social prerequisites of technical change and their relationship toscience (Hounshell, 2000). Jerry Ravetz (1971) proposed greater scrutiny of socialcontrol of social structures of science. Philosophers know that Kuhn, Lakatos, andFeyerabend belong somewhere in here, in the period of the 1960s–1980s, but none-theless seem to flinch from acknowledging their social doctrines, perhaps feelingsomething subsequently went awry with the proposed alliance of the history andthe philosophy of science. The congealing out of a separate sociology of sciencealso occurs in this era, although philosophers regarded it more as a source of irri-tation than edification. Philosophers such as Joseph Rouse (1987) attempted tocall attention to the political aspects of epistemology. Finally, we might point tothe impact of Steve Fuller in his quest to broaden the conception of a ‘‘social

6 Pride of place goes here to Don Howard (2003a,b) and Thomas Uebel (2000a). Struan Jacobs (2002)

raises the issue of the initial appearance of the discourse of ‘‘scientific communities’’ in philosophical dis-

cussion. See also Mirowski (1998).


epistemology’’ within the philosophy profession, dating back to the 1980s (Fuller,1988, 2002b); he has also done yeoman work in trying to bridge the divide betweenphilosophers, science policy advocates, and the science studies community.So, as it turns out, there did exist a surplus of meditations upon the ‘‘social

dimensions of science’’ throughout the 20th century, most of it fairly high-profile,some of it performed by bona fide philosophers, and quite a bit of it directly con-cerned with the interplay of science organization and human welfare. So whencecame the impression that philosophers of science have only ‘‘recently’’ turned theirattention to serious consideration of the social dimensions of scientific inquiry?How is it that, prior to this conversion, many of the elect seemed to be held in thethrall of what has been characterized as a ‘Legend’, as Philip Kitcher puts it, wheresaid ‘Legend conceives of scientific method as determinate. There are no optionsfor the good scientist’ (1993, p. 68)? Since the ‘Legend’ asserted that there was onlyone uniquely correct form of rationality, and if scientists were behaving accordingto their telos, then for the purposes of the philosophy of science they were effec-tively all alike, and it followed that there was no need to consider the ‘‘society’’ ofsuch rational agents. Furthermore, how did it come to pass that the question of therelationship of science to society was deemed moot, if only because ‘science is thepinnacle of human achievement not because of its actual successes but in virtue ofthe fact that its practice, both in attaining truth and in lapsing into error, is thor-oughly informed by reason’ (Kitcher, 1993, p. 4). However did science come to beportrayed as ‘‘value-free’’ because it was lauded as transcending secular values?The status of such an exemplary state of grace putatively did not need to be con-sidered in its relationship to the larger society; and indeed, many thought Societyshould feel honored and blessed that it found itself subsisting (along side of?embedded within? enfolded around?) such a noble entity. Herein lies the origin ofthe strange polarity which contrasts a ‘‘rational’’ Science with an ‘‘irrational’’Society. As I have argued elsewhere (Mirowski & Sent, 2002), the postwar credothat scientists somehow enjoyed a dispensation from having an ‘‘economics’’ hadmuch to do with this philosophical construction.The coexistence of this bizarre conviction that science at its best was somehow

a-social, both in its internal operation and in its external relations with the culturewithin which it was situated, juxtaposed with the demonstrable fact that very spe-cific policies concerning the viability of particular social structures of science werebeing deployed in its day-to-day defense, constitutes the primary topic of thepresent paper. Philosophers are perhaps fed up with sociologists chiming in withthe mantra that ‘‘science is social’’; here it will be conceded that there is no needfor further repetition of that rallying cry at this late date. Rather, we are concernedto explore how it was that science came to be portrayed by philosophers asa-social, autarkic, and value-free in America in the middle of the 20th century, andthen, how it has subsequently become possible for the philosophy profession tocome to the conclusion that this portrayal could be easily reversed without derang-ing their entire project. It so happens that this narrative is intimately intertwinedwith another strand of discourse, one that tracks attitudes towards the health of‘‘democracy’’ in America. Of necessity, this must be an historical inquiry, and thus


we find that a few historians have cleared a path for exploration of this issue.7 Butwe should prefer this to be regarded as an exercise in the social epistemology of therelationship of science to society. If we were to proceed to treat the ‘‘social dimen-sions of science’’ solely as a set of propositions conforming to (or not) a uniquelogic of societal organization, then we should be recapitulating the error which hasjust been diagnosed with regard to the ‘Legend’ of an earlier vintage of the philo-sophy of science. Rather, it will be argued that certain configurations of scienceorganization, in conjunction with certain widely accepted images of society, havegiven rise to very specific orthodoxies in the philosophy of science: these representthe scientific dimensions of social knowledge.There have been (at least) three distinct positions regarding the relationship of

science to society dominant in the American philosophy profession in the 20thcentury. For the purposes of exposition in this paper we shall identify each with arepresentative agent, viz., John Dewey, Hans Reichenbach, and Philip Kitcher.This personalization of categories undoubtedly constitutes a misrepresentation ofcommunally shared beliefs in their respective eras, but that should be regardedmerely as a provisional point of departure in what would eventually be a largerresearch project. While we shall be concerned with providing a faithful thumbnailcharacterization of their respective accounts of the ‘‘social dimensions of science’’,our real innovation will be to supplement them with parallel accounts of the‘‘scientific dimensions of the social’’, or the conditions that permitted theseaccounts to thrive in their respective eras. Consequently the dominant philosophi-cal stance of the time should not be directly attributed to the actual charismaticprotagonists named Dewey, Reichenbach, and Kitcher; rather, they should be firstcharacterizations of a set of working beliefs related to the actual social structuresof science dominant within those epochs. In particular, in this paper we shall seekto relate their philosophical pronouncements to the types of environments in whichscientific research was being prosecuted, the types of social theories widely preva-lent in their eras, and last but not least, the conundrum of the role of the philos-opher in those respective regimes. The variables that we shall touch upon aresummarized below in Table 1.The correlation of regimes of science organization with contrasting philosophical

accounts of the relationship of science to society is intended to raise the troublingquestion: just how effectively have philosophers been able to train their analyticalskills upon the complex of problems called ‘‘social dimensions of science’’? Werestructural obstacles hampering their comprehension of the interplay of such vexingbinaries as the connection of science to democracy, the mutual conditioning ofscience and the economy, the contrast of the natural to the social, the impact ofeducational formats upon the flourishing of science, and the impact of science onthe constitution of communal aspirations of its clients? This will bring us to withinhailing distance of a most unsettling question: To what extent does the modern

7 Here we would especially mention Hollinger (1996), Purcell (1973), and Westbrook (1991) as having

provided important clues.


orthodoxy in the philosophy of science even make contact with the contemporarystate of science as found on the ground in the new millennium?

3. Dewey’s philosophy for the masses

The notion of a complete separation of science from the social environment is afallacy which encourages irresponsibility on the part of scientists regarding thesocial consequences of their work. (Dewey, 1985, p. 483)

For most people, ‘‘science is a mystery in the hands of initiates who havebecome adepts in virtue of following ritualistic ceremonies from which the pro-fane are excluded’’. (Dewey, 1927, p. 164)

Our contemporaries may not generally regard John Dewey as a philosopher ofscience in good standing, but nevertheless, that was perhaps the most important facetof his reputation in the first three decades of the 20th century. Dewey was not only thebest known Pragmatist philosopher of his era, but he was also a public intellectual,seeking to make philosophy speak to the most insistent and pressing problems of thenation. Louis Mumford perceptively summed up his crusade as opposition to leisure-class notions of thinking, struggling to replace them with flexible, nondogmatic, anddemocratic modes of thought. In this quest, his characterization of ‘‘science’’ played apivotal role. In the early 20th century, it was a trite commonplace to assert that sciencehad liberated humanity from earlier metaphysical, primarily theological, fetters; whatmade Dewey distinct from run-of-the-mill science idolaters was that he realized thatcontemporary practices of science had rendered any such emancipatory promise notonly feeble but deeply implausible, and that if left unchecked, the corruption of thisparticular ideal would spread to other, even more important cultural values.

Table 1

Three regimes8 and their philosophies

Time period

8 The concept of regimes

Mirowski & Esther-Mirjam S

WWI to 1940 1

of science organization i

ent (2002) and Mirowski

940–1980 1

n America is discussed

(2003, 2004).

980–present

Science organization

managers:

foundations’ program

officers

C

o

old War military

fficers

g

c

lobal privatization

orporate officers

Prime location:
corporate lab r esearch universities i ndustry hybrids
Philosophical orthodoxy:
pragmatism l ogical empiricism s ocial epistemology
Exemplary philosopher:
John Dewey H ans Reichenbach P hilip Kitcher
Social theory:
institutionalism d ecision theory g ame theory
historicism o
perations research n eoclassical economics
Science exists for:
communal welfare u nimpeachable truth v aluable information
Society is:
a democratic nation r ational individual
multiplied

a
marketplace of ideas
Fundamental challenge:
national inferiority m ilitary dominance i ndustrial control
The enemy:
ignorance e rror i nefficiency
at some length in Philip


To understand Dewey’s reasoning, it is important to first appreciate that scienceas pursued in pre-Depression America was a very different phenomenon from whatwe now conceive of as the conventional process and settings of scientific research.9

Very little support for scientific research was provided by government funds a cen-tury ago, and most colleges were not set up to promote scientific research.American higher education was largely patterned upon the liberal education model,providing a generalist moral curriculum to a small proportion of the population,not oriented towards vocational training. Indeed, most individuals who soughtadvanced academic training in the natural sciences had to go abroad, primarily tothe German universities, then deemed the best in the world. Outside of a handful ofuniversities, most scientific research was funded by and prosecuted under the aus-pices of large corporations. Behemoths such as General Electric, DuPont, Amer-ican Telephone and Telegraph, and Eastman Kodak employed the vast majority ofscientifically trained personnel in the USA, and even supported a few Nobel Prizewinners. The motives for this configuration of privatized science had more to dowith the fin de siecle merger wave, antitrust policies and the need for routine in-house testing capabilities in the newer science-based industries, than it had to dowith far-sighted innovation policies or benevolent intentions towards the generalwelfare. Even the funding of the miniscule sphere of academic research into sciencewas conducted on terms dictated by the industrial behemoths. The vast wealthamassed by families such as the Carnegies and the Rockefellers was partially diver-ted into eponymous foundations in order to ‘‘give back’’ something to the nationthat had made them wealthy. Although the entire raison d’etre of the foundationswas to fund research which was not dictated by the exigencies of the pursuit ofprofit, these nominally eleemosynary institutions were still run by men whose back-grounds were in corporate bureaucracies, and therefore the very criteria of researchfunding did not escape the stamp of corporate imperative. For instance, a fewfavored private (not state-sponsored) universities were encouraged to nurture therole of academic entrepreneur, mimicking the captains of industry who had pro-vided their seed capital. Grants became patterned upon the business instrument ofcontracts rather than the previous template of handouts for poor relief. Applica-tions forms, progress reports, bureaucratic peer evaluation, and the other trappingsof the hierarchical M-form corporation were inserted into research protocols.10

The corporate sway over science bore many other consequences for public atti-tudes towards science, and scientists’ attitudes towards the public. For instance, wenow tend to forget that the first formal initiative aimed at shaping the ‘‘publicunderstanding of science’’ dated from the 1920s, with the Scripps NewspaperService instituting its Science Service in 1921 (Tobey, 1971, p. 67). The fact that thebulk of scientifically trained personnel was employed by corporations was also

9 Descriptions of early 20th-century science are taken from Reingold (1991), Shinn (2003), McGrath

(2002), and sources cited in Mirowski & Sent (2002). The failures of early 20th-century American

attempts to garnish government support for science are recounted in Tobey (1971).10 The corporate character of early foundations and their effect on scientific research in the early 20th

century is discussed in Kohler (1991) and Kay (1997).


a major conditioning factor behind the cultural fascination with the newly

professionalized engineers as a putatively progressive political force in the 1920s. Inthe popular press, scientific theories were being persistently dragooned into serviceto justify various forms of elitism, from Social Darwinism to Technocracy to reha-shes of the variational principles found in mechanics as grand theories of Natural

Efficiency. The ambitions of many scientists/engineers to assume credit for the pro-gress of America provoked a reaction in the form of questioning the baleful influ-ence of scientists on their social surroundings.11 The supposed meritocraticcharacter of science was often parlayed into anti-egalitarian precepts and corporateboosterism; for instance, Robert Millikan told a Chicago audience in 1919, ‘It is

probable that the total possibilities of improvement of conditions through distri-bution are very limited, while possibilities of improvement through increases ofproduction are incalculable’ (quoted in Tobey, 1971, pp. 182–183).This pervasive corporate character of early 20th-century American science was

the looming backdrop to Dewey’s distress over the relationship of science tosociety. Science as then practised constituted a problem because of the way it had

been imperfectly integrated into the social fabric. As he wrote, ‘the concept thatnatural science somehow sets a limit to freedom, subjecting men to fixed necessi-ties, is not an intrinsic product of science . . . [but] a reflex of the social conditionsunder which science is applied so as to reach only a pecuniary function’ (Dewey,1984, p. 105). Dewey suspected that that the bureaucratic/industrial location of the

scientist was a prime reason for the encapsulation of the scientific method withinthe cult of the expert, erecting an artificial barrier between science and society. Hebemoaned the phenomenon that ‘the idea of experts is substituted for that of philo-sophers, since philosophy has become something of a joke, while the . . . expert inoperation is rendered familiar and congenial by the rise of the physical sciencesand by the conduct of industry’ (Dewey, 1927, p. 205). Philosophers bore some ofthe responsibility for this sorry state of affairs, because ‘The philosopher’s idea of acomplete separation of the mind and the body is realized in thousands of industrialworkers’ (Dewey, 1984, p. 104). How one thought about the process of thinking

was shaped by the social structures which enabled the activity of thought. As hewrote in Experience and nature:

the ulterior problem of thought is to make thought prevail in experience, notjust the results of thought by imposing them upon others, but the active processof thinking. The ultimate contradiction in the classic and genteel tradition is

11 See, for instance, Clarence Ayres (1927, p. 275): ‘Scientists frequently argue that apart from making

science one’s profession, the chief advantage to be obtained from the study of science is the scientific

mind: freedom from dogma, hospitality to unexpected truth, the experimental attitude . . . Theoretically,

all may gain these insights. But actually, only a few of us have ever done so. Theoretically, we might

become a scientific people; but we have not, and are not likely to, except in the sense in which we are

now a Christian people’. In a curious incident in 1936, Franklin Delano Roosevelt wrote a letter to

Science in which he suggested that scientists and engineers were not doing enough to help society

‘absorb the shocks of the impact of science’ (in McGrath, 2002, p. 50).


that while it made thought universal and necessary and the culminating good ofnature, it was content to leave its distribution among men a thing of accident,dependent upon birth, economic and civil status. Consistent as well as humanethought will be aware of the hateful irony of a philosophy which is indifferent tothe conditions that determine the occurrence of reason while it asserts the ulti-macy and universality of reason. (Dewey, 1981, p. 99)

In retrospect, we can appreciate that Dewey was casting about for a role for thephilosopher within the Jazz Age American system of science. Dewey’s alternativeto the prevailing cult of the expert was to imagine a different sort of science, ascience dedicated to the promotion of communal intelligence, a generic ‘‘experi-mental’’ method that would be made available to all members of the community asa part of their birthright. Pragmatic knowledge of nature would shade impercep-tibly into useful knowledge concerning regularities of communal behavior. In orderto counter the corporatist and elitist connotations of science rife in his era, Deweymade a conceptual move that will reverberate down through the remainder of ournarrative. Throughout his later career, Dewey concertedly and repeatedly blurredthe definitions of ‘‘democracy’’ and ‘‘science’’ prevalent in his lifetime, so that hecould conflate the two and provide an counterweight to the forces dragging scienceaway from its liberal and liberationist potential (Dewey, 1939). As he insisted,‘democracy is not an alternative to other principles of associated life. It is the ideaof community life itself’ (Dewey, 1927, p. 148). And from the obverse side, he sawhimself ‘raising the question of what science can do in making a different sort ofworld and society. Such a science would be the opposite pole to science conceivedmerely as a means to special industrial ends’ (Dewey, 1984, p. 107).The yoking together of science and democracy was not such an obvious winning

combination in the early 20th-century context; it had yet to attain its subsequentunassailable American status commensurate with Mom and apple pie. Indeed,Dewey’s book The public and its problems was a response to an intellectual currentwhich framed the duo as incompatible in structure and content.12 As we haveobserved, a common form of ‘‘naturalism’’ was regularly being used for nativist,racist, anti-egalitarian, and conservative ends. But perhaps more disturbing, thespread of empiricist protocols to the newly established academic social scienceswere producing observations that suggested conventional understandings ofdemocracy were a sham. Political scientists were demonstrating that the UnitedStates government was not at all run by the people for the commonweal, but by asmall handful of insiders for their own power and enrichment. Legal realists weredocumenting that judicial decisions were neither impartial nor logical, but ratherthe product of powerful interests. Psychologists were demonstrating that the voterswere largely irrational and easily swayed by those who controlled the corporatemedia, particularly newspapers and the new-fangled radio. The sum total of thisresearch portrayed a populace so easily manipulated and exploited that an expan-

12 This case has been made with great insight by Purcell (1973) and Westbrook (1991).


ded franchise and enhanced participation in the political process was widelyregarded as dangerous, if not foolhardy; the democratic election of fascist partiesin Europe only reinforced that impression. Not only was science perceived asintrinsically undemocratic; science when applied to society was uncovering the darkside of democracy.Another Jazz Age trend that is oftentimes forgotten today is that there existed a

fair degree of academic opposition to the idea that there was or could be a generic‘‘science’’ which would apply equally to Nature and Society. High-profile figuressuch as Frank Knight in economics and Pitrim Sorokin in sociology were arguingthat the natural sciences (and especially physics) provided misleading paradigmsfor theory in the social sciences, and were citing German philosophical theses thatnothing like the ‘‘laws’’ of physics could be discovered when the subject wassociety. A general inclination towards evolutionary arguments was being deployedas explanations why there were no absolutes in human experience, and to argue infavor of the essential plasticity of human nature. But this opened the door tothe cynical manipulation of the masses by experts. The major opponents to this‘‘relativist’’ threat in social science were theological, and in particular Catholic,academics who sought to reassert the centrality of values through reimposition oftheological absolutes (Purcell, 1973, Ch. 10). These were not the sort of peopleDewey could see himself forging alliances with, and therefore, he was driven tofind a third way to relegitimize science and democracy.Dewey’s pathway out of the impasse was to insist that science would cease to

undermine liberal democracy and that the corporate sway over science would beprogressively diminished if and only if we came to regard science and democracy asinseparable parts of the same communal activity; that is, a) the practice of democ-racy would come to resemble science at its best, which was procedurally non-dogmatic and experimental; and b) more science would be reorganized and con-ducted in the communal democratic interest. It will prove important for us to getthe subtleties of Dewey’s equation of science and democracy correct, because itwould very rapidly become corrupted into something very different in World WarII, especially under the auspices of Robert Merton, Michael Polanyi, and JamesConant (Fuller, 2000; Mirowski, 1998; Wang, 1999), something which Dewey per-sonally would have regarded as pernicious, and something which is sometimes mis-takenly attributed to Dewey. In World War II, under the imperatives of wartimemobilization of science, the separate, autarkic and self-governing scientific com-munity began to be held up as the icon of what a democratic community couldaspire to be, in the guise of an ideal ‘‘republic of science’’. In this construction, sci-entists did lay claim to an esoteric expertise in generic rationality inaccessible to (orat least rare for) the common layperson. Dewey could never have been a pro-ponent of this position for a number of reasons, but primarily because the corpor-ate organization of science then dominant could never have been plausiblyportrayed as self-governing in that era; nor, for that matter, could scientific ration-ality plausibly have been pictured as politically ‘‘free’’ from corporate imperatives.The separate constitution of the scientific community as a social formation was notyet a conceptual possibility. That could only become conceivable after the war.


Dewey’s reconciliation of science and democracy did not come as a bolt from

the blue, but was built up from resources available to him in the 1920s. Philosophy

of science was just one component of theories of the social in that era, and not

something notionally apart or distinct from them. Economics and psychology were

other key components of the conflationist scheme. In Dewey’s case, he left numer-

ous clues that he had made an implicit pact with the Institutionalist school of

economics and the ‘‘habit’’ school of social psychology, both very active traditions

in the contemporary American context.13 His position bore a number of striking

resemblances in particular with the writings of the Institutionalist Thorstein

Veblen, one of the very few authors whom he regularly cited (Dewey, 1984, p. 102;

Westerbrook, 1991, p. 310). The pediment of their shared themes was that meta-

physics has operated in the past to reinforce existing class relationships in society;

hence Dewey united in opposition with Veblen to ‘‘leisure class thinking’’, first

described in the Theory of the leisure class. Another was that the ‘‘individual’’ self

is constantly under reconstruction in modern society, and therefore the older ‘‘indi-

vidualist’’ orientation of both epistemology and social science stood as a major

obstacle to the constitution of his ideal science. The text Individualism old and new

is primarily one long argument against the classical economic elevation of the

‘natural’ individual as the basis for understanding society:

[T]he chief obstacle to the creation of a type of individual . . . in whom socia-

bility is one with cooperation in all regular human associations is the persistence

of that feature of the earlier individualism which defines industry and commerce

by ideas of private pecuniary profit. (Dewey, 1984, p. 84)

For Dewey, it was not that the average citizen was woefully ‘‘irrational’’; rather,

it was the received portrait of rationality that had led the social scientists astray. In

Dewey’s system this implied the rejection of classical Utilitarianism, and conse-

quently, of neoclassical economics:

The idea that there is something inherently ‘natural’ and answerable to ‘natural

law’ in the working of economic forces, in contrast with the man-made artifici-

ality of political institutions. The idea of a natural individual in his isolation

possessed of fully-fledged wants . . . and of a ready-made faculty of foresight

and prudent calculation is as much a fiction in psychology as the picture of the

individual in possession of antecedent political rights is in politics. (Ibid.,

p. 102)

Veblen famously asked ‘Why is economics not an evolutionary science?’; Dewey

wanted to equate an evolutionary approach with what he considered to be the

13 On the history of Institutionalist economics, consult Mirowski (1988); Rutherford (1994); Yonay

(1998). On habit psychology, see Camic (1986); Dalton (2002); Westerbrook (1991), pp. 286–293; Ayres

(1927), Ch. 22. His periodic discussions at Columbia with another of the leading Institutionalist figures,

Wesley Clair Mitchell, is documented in Mitchell (1969), p. 450.


‘‘experimental method’’. Yet this did not mean mimicking the actual quotidian

procedures of the physical scientists, as he repeatedly insisted:

When we say that thinking and beliefs should be experimental, not absolutistic,

we have then in mind a certain logic of method, not, primarily, the carrying on

of experimentation like that of laboratories. (Dewey, 1927, p. 202)

What purports to be experiment in the social field is very different from experi-

ment in natural science. (Dewey, 1939, p. 65)

Thus, by seeking to equate ‘‘good science’’ with democracy, Dewey was definitely

not appealing to any theories of natural science to underpin theories of democracy:

he was no friend of any social physics. His version of pragmatism led him to

deflect attention from the ends and content of science, and towards the means

through which science was purportedly conducted. Thus he leaned heavily upon

the Institutionalist rejection of ‘‘natural law’’ theories of the nature of humanity in

order to prevent misunderstanding of what he intended by a scientific democracy.

Dewey’s commitments to certain particular social theories were substantively more

fervent than have been noticed by later commentators. This is significant, because

incipient revolutions in social theory would have direct implications for the plausi-

bility of Dewey’s philosophical theses for his later audiences.Even writers deeply sympathetic to Dewey’s project have been forced to admit

that his matchmaking activities between science and democracy subsequently

proved practically barren, a most unhappy prognosis for a pragmatist philos-

opher. Long passages of clotted prose never led to any practical political pro-

grams. ‘He appeared to have given little thought to the problems and possibilities

of participatory government . . . Dewey had surprisingly little to say about demo-cratic citizenship’ (Westbrook, 1991, p. 317). Politics, one of his trademark con-

cerns, ended up a lingering embarrassment, rather like a cynical guest at a

patriotic gathering. More germane to our present concerns, he also made no sug-

gestions as to how scientists could be unshackled from their lab benches in major

corporations, much less the imperative of the profit motive; there was no serious

consideration of how science was to be paid for at all. One got the impression

those questions would be in the nature of details to be worked out in the distant

future when, believe it or not, the captains of industry would grow bored and

would relinquish their ownership of ‘‘machine production’’ to socialized entities

‘so that they may devote their energies to affairs which involve more novelty,

variation and opportunities for gain’ (Dewey, 1927, p. 61). And yet, Dewey also

seemed to argue that America had the best chance of realizing his vague ideal of

democracy as science, raising the bar for human excellence relative to the older,

richer, but metaphysically lumbered cultures of Europe, producing a moral and

intellectual advance in the career of mankind as a whole. This pie-in-the-sky

aspect of Dewey’s thought illustrates the fact that his popularity was frequently

more directly attributable to his quest for a non-Marxian lowest common

denominator of politics and economics which could attract the allegiance of the


broadest array of American left-leaning intellectuals (Purcell, 1973, p. 202), than

it was to any specific reform agenda.14

Yet however toothless and nonspecific his reform program turned out to be in

practice, it nonetheless proved to be untenable from World War II onwards, andfurthermore, it unintentionally provided the major resource for certain more poin-ted and virulent doctrines on the social relations of science developed in Americain the postwar period. In a subsequent degradation of his program for the con-flation of science and democracy, university-accredited scientists came to be said to

conduct their activities within an ideal democracy—namely, Merton’s ‘‘norms’’ ofuniversalism, communism, disinterestedness, and organized skepticism. No longerwas democracy-in-the-making deemed a legitimate aspiration for the scientificcommunity at large; instead, it became ossified as the esoteric virtue of the adeptwithin the cloistered monastery of the newly re-engineered American research uni-versity. Rationality was increasingly deployed as a logical abstraction, the dog-

matic preserve of the expert. These cloistered elect were explicitly absolved of allresponsibility for the state of the larger society and its political aspirations. Indeed,this is what Conant and Weaver meant by their championing of ‘‘Free Science’’.And, in a twisted non sequitur, since America was putatively a democracy, it nat-urally acceded to the position of premier bastion of legitimate science, in compari-

son with a decadent and totalitarian Europe. Dewey’s pragmatic stress on meansover ends became transmuted into the elevation of pure science over application;Dewey’s critique of corporate science became twisted into a denial of any relevanceof economic support to the conduct of science; Dewey’s advocacy of a pragmaticlogic of inquiry became the reification of a language of science of almost mysticalpowers, namely, formal logic and mathematical axiomatization (Reichenbach,

1938, p. 49).We should not be understood as claiming that Robert Merton or anyone else

intended this defeasance; our argument is different. Rather, in World War II andafter, the social organization of science had been re-engineered from the groundup, as had the universities and, to a lesser extent, the government. The science/industry partnership so inveighed against by Dewey was being replaced with ascience/military partnership: Science had become betrothed to the State, but not in

any way Dewey had imagined. Further, philosophers had learned in the interim adifferent way to make themselves useful. The world that had been the referencepoint of Dewey’s philosophy no longer existed, so it was a foregone conclusionthat Dewey’s account of the social relations of science could no longer resonatewith either the scientists themselves, a newer generation of philosophers, or withthe broader public.

14 This may have some relevance for coming to comprehension of Richard Rorty’s quest to revive

Dewey for the modern reader. Rorty distances himself from Dewey’s ‘habit of announcing a bold new

positive program when all he offers, and all he needs to offer, is criticism of the tradition’ (quoted in

Cahn (1977), p. 56). A very interesting exercise that situates Rorty in his social context, as we have

attempted to do here for Dewey, is Gross (2003).


4. Reichenbach’s philosophy for the operations researcher

No human being is completely rational, i.e., makes decisions strictly accordingto the balance of probabilities and valuations. The decision may depend onemotional factors, which vary from moment to moment, such as an ‘itch tofight’, or inversely, an indolence to the challenge of reason. If a decision is madenot by one person but by a body of persons, such as a government, there arefurther random factors to be envisaged . . . the tiredness or sickness of leaders;the inertia of bureaucracy; the pleading of voices of people frightened by thehorrors of war, constitute factors which may deter from a decision for war. Allthese factors are random factors; their effects are unpredictable and may extendin a positive or negative direction. (Reichenbach, 1949, p. 9)

The philosopher James Robert Brown has written: ‘One of the travesties of cur-rent science studies debates is the branding of the positivists as political reaction-aries . . . Dealing with political issues meant as much or more to Neurath, Carnap,Frank and Hahn as coming to grips with science for its own sake’ (Brown, 2001, p.54). There is no doubt that politics mattered for the logical empiricist movement;but the rather imprecise characterizations of the actual history which are exempli-fied by comments such as the above are yet another symptom of the weaknesses ofcontemporary philosophers in their quest to pronounce upon the ‘‘social dimen-sions of science’’. In this particular instance, the choice of relevant representativesof the positivist movement is more than a little misleading: with the exception ofCarnap, none played much of a role in the professionalization of academic philo-sophy of science in the American context.15 There is also the unfortunate tendencyto think that, just because someone has identified himself as some species of‘‘socialist’’, then that immediately absolves him from any accusations of reaction-ary political activity.16 For the reasons we shall outline below, it will prove rathertoo hasty to hew to the conviction that ‘Logical positivism was a casualty of thecold war, not one of its villains’ (Reisch, 2002, p. 391).The predominant fact about American science at mid-century was the assump-

tion of its funding and organization by the military during World War II, estab-lishing a novel regime of science management which lasted well into the 1980s.17

As one can readily appreciate, this opened up a Pandora’s Box of problems for the

15 On this point, see Howard (2003b) and Hardcastle & Richardson (2003). Ron Giere has reminded us

that ‘The European origins of logical empiricism are not intellectually continuous with its later develop-

ment in North America . . . It is with this fact that any future history of logical empiricism in North

America must begin’ (1996, p. 336).16 That appears to be the thrust of comments such as the following: ‘A socialist and an internationalist,

Carnap nevertheless lived through situations which demanded defense, retreat, self-criticism, stubborn

decency, maximal intelligence about minimal possibilities . . . In his passionate way, he was a man of theresistance’ (Cohen, 1971, p. xlii). Carnap’s ambivalence towards democracy has only begun to be

noticed: see Notturno (1999).17 This is a well established fact in the history of American science, such that we should not need to

document it here. See, for instance, Kevles (1995); Kragh (1999), Ch. 20; Dennis (1997). The following

three paragraphs are a summary of an extensive case made in Mirowski (2002), Ch. 4.


relationship of science to society in general and democracy in particular, problemsthat the various constituencies identified in this narrative were anxious to address.Scientists recruited to the war effort were quick to realize that their newfoundlargesse came freighted with dangers and responsibilities. One response in Britainwas the previously mentioned ‘Social Planning of Science’ movement of Blackettand Bernal, which provoked a hostile response by Michael Polanyi, Friedrich vonHayek, and others (McGuckin, 1984). The latter conceived of their crusade asbroadly ‘‘anti-positivist’’, in that Otto Neurath was a prime target of their disdain(Uebel, 2000a). This dispute was closely related to another, which also tended todraw upon Viennese intellectual resources, namely, the ‘‘socialist calculationcontroversy’’, of which Otto Neurath was again a major protagonist. While somemodern philosophers are quick to point to Neurath as someone attuned to our finde siecle conundrums, it should be remembered that he bore little relevance forthose weathering the seachange in science regimes in postwar America. Denuncia-tions of ambitions for the planning of science for social ends, such as those mootedby Neurath, were the stock in trade of Vannevar Bush, James Conant, and WarrenWeaver; but ironically, these were the primary protagonists in the wartime mobili-zation of science in America, the very people assigning the research tasks andcutting the checks. While public figures such as John Dewey had been talkingabout the social planning of science in a vague way before the war, the prospect ofgovernment direction of scientific research had abruptly become a much more tan-gible prospect in the US with the advent of the MIT Radiation Lab, theManhattan Project, and a host of lesser mobilizations. The programs of sciencemobilization in Soviet Russia and Nazi Germany also had to be somehow dis-tinguished from what was going on in America, if only to soothe anxieties aboutthe global corruption of science and maintenance of the real source of Americanexceptionalism. It was the very palpability of a model for ‘‘socialist’’ planning pro-vided by military mobilization—an important theme in Neurath’s writings(Cartwright et al., 1996, pp. 14–18)—which called forth a response to the perceiveddangers of the reorganization of the social relations of science then in progress.And then, we must never forget, after 1945 there lurked the vexed issue of therelationship of a democracy to the production and use of the Bomb. Philosopherswere a little slow off the mark to respond to these controversies; but they soontook their cue from some natural and social scientists who had to come to rapidaccommodation with the gales of change buffeting them from all sides.It was these wartime-induced anxieties that prompted the first real appearance of

the conceptual innovation of a free-floating ‘‘scientific community’’ which purport-edly subsisted autonomously from the larger culture in which it was situated. Somescholars have argued that such a notion was not seriously entertained prior to thewritings of Michael Polanyi (Jacobs, 2002); others credit Merton (Wang, 1999).This, as we have already hinted in the previous section, was the key innovation thatutterly transformed the terms of debate concerning the relationship of science tosociety. Yet the watershed was not simply conceptual; it was anchored in unpre-cedented innovations of social identities for scientists in wartime. When scientistswere recruited into the military during World War II, they were very much con-


cerned to differentiate themselves from the mass of conscripted manpower, but alsoexercise a fair amount of discretion in the deployment of their expertise. Whatbegan as scientists serving as glorified troubleshooters for the creation and deploy-ment of new weapons systems such as radar and the atomic bomb was transformedby the scientists themselves into a never-before permitted institutional adjunct to themilitary, namely, as cadres of experts in a generic ‘‘scientific method’’ whose remitwas to assist in the generation of rational strategy and tactics for battle. These scien-tists were frequently contemptuous of tradition or ‘‘conventional’’ military doctrinethat they encountered, and sought to replace it with what they regarded as ‘‘empiri-cal’’ methods. In practice, much of this methodological innovation consisted in theimposition of physical models, especially those adapted from thermodynamics andmechanics, upon abstract agglomerations of men and machines. Because the quan-tum physics and nuclear cross-sections so bound up with the radar and atomicweapons that occupied their attentions were so dependent upon stochastic formal-isms, facility with probability theory tended to be a hallmark of their models. This‘‘profession’’, which did not even exist before World War II, grew rapidly duringthe war, and came to be known in the USA as ‘‘Operations Research’’ or OR.While it would be impossible to adequately summarize the scope and content of

OR in the space of this paper, it will be indispensable to appreciate that OR was apractical response to the problems and paradoxes of the military planning andorganization of science in the mid-century.18 Physicists such as Patrick Blackett,Philip Morse, George Gamow, and Ivan Getting, as well as mathematicians suchas John von Neumann, Richard Bellman and their comrades, wanted to enjoy themilitary largesse but maintain a fair amount of latitude in evading direct control bythe military. Indeed, they believed that they were far smarter than your averagelieutenant colonel, and should be allowed to run things as they saw fit; in effect,they wanted to exist simultaneously within but apart from the military chain ofcommand. They wanted to be paid by the military but not really be in the military;as physicists, they wanted to do social research for the military but not becomeconfused with social scientists; they wanted to tell others what to do, but not beheld responsible for the commands given. After the war, they wanted to return totheir university posts without having to relinquish their lucrative military ties. Tobe granted these extraordinary dispensations, they had to innovate new roles thatembodied this delicate amalgam of engagement and aloofness. The construct of the‘‘operations researcher’’ was the professional device which fostered the reconcili-ation of these conflicting demands; significantly for our present concerns, it alsobecame the empirical template for the idea of a free-floating ‘‘scientific com-munity’’, distinguished by its possession of a special expertise rooted in a generic‘‘scientific method’’, subsisting with a fair degree of autonomy within but apartfrom a larger social community. OR was the anvil upon which the postwar

18 There is no good history of Operations Research that can be recommended to the general reader.

Existing histories of OR are summarized in Mirowski (2002), Ch. 4. Kirby (2000) is a guide to the alter-

native trajectories of OR in the British and American contexts.


relationship between scientists and the American state was hammered out; oncesuccessful, the blade was then turned to carve out a new model of society thatcould be amenable to the rapprochement of science and the military. OR providedmuch of the metallic durability and intellectual firepower for postwar Americansocial sciences such as decision theory, organization theory, management theory,and neoclassical economics; but it also provided the framework for Cold Warphilosophy of science.I do not intend this thesis as to be regarded as trafficking in vague ‘‘influences’’

and murky insinuations about science tainted by military dollars. That ilk of argu-mentation is rightly rejected by philosophers. I am instead pointing out that theprofessionalization of American philosophy of science in the immediate postwarera grew directly out of the soil of Operations Research; that major figures of thelogical empiricist movement in America served their country in dual capacities asoperations researchers; that the editors of the flagship journal Philosophy of sciencein the critical transition period, C. West Churchman (1951–1959) and RichardRudner (1960–1975), were better known as operations researchers; that philoso-phers of science were employed at major OR research centers such as RAND(a fact discussed below). Much of the content of so-called ‘‘analytical philosophy’’in that period, ranging from its disdainful attitudes towards surrounding disciplines(not to mention its own history) to its preferred mathematical formalisms, is easilyrecognizable in its family resemblance to the contemporary subsets of OR knownas decision theory and formal computational logic. The fascination with physics asthe first science among equals reprised the historical fact that OR was first insti-tuted by physicists for the protection and promotion of physics; the presumptionthat one could proceed to formalize the behavior of rational empirical man inno-cent of any familiarity with, or acknowledgment of, the social sciences which sup-posedly were already concerned with them was also an echo of the credo of theoperations researcher.19 The conviction that prior practitioners simply did notsufficiently appreciate ‘‘the facts’’ was the complaint most frequently launched byoperations researchers against their military patrons.Rather than try to document these generalizations for an entire generation, we

shall here be forced to settle for a brief account of a representative figure of thismovement, Hans Reichenbach. Reichenbach’s career began in Stuttgart and thenthe Berlin Circle for empirical philosophy, where he had garnished a reputationas a socialist radical. The rise of the National Socialists and the Race Laws of1933 forced him out of Berlin and into exile in Turkey for five years, after whichhe managed to get an appointment at UCLA in 1938. Reichenbach’s tribulationsdid not end there, for he had to suffer the indignity of house arrest as an‘‘enemy alien’’ during a portion of World War II. Nevertheless, Reichenbach didultimately flourish in Southern California until his death in 1953. Along with

19 Has anyone ever noticed that the Library of Congress call numbers that categorize the philosophy of

science in American libraries also designate books in OR of a postwar vintage? Clues are quite abundant

if one just permits oneself to perceive them.


Rudolf Carnap’s appointment at the University of Chicago in 1936, and Herbert

Feigl’s move to Iowa in 1931, his professorship was conceded to be one of the

pivotal outposts for the promulgation of logical positivism throughout the Amer-

ican landscape (Giere, 1996).Reichenbach’s positivism, as is well known, began with detailed examinations of

recent theoretical developments in physics for the purpose of drawing out epis-

temological lessons. While he sought to derive anti-Kantian morals from the

theory of relativity and uncover anomalies for standard ideas of causality from

contemporary interpretations of quantum mechanics, he was also an advocate of

some larger themes which bore relevance for the ‘‘defense’’ of science against those

he deemed its detractors. For instance, he thought that one task of the philosophy

of science was to provide criteria for the distinction between propositions that

might be deemed conventional and those which were incorrigibly empirical in the

language of physics. In Experience and prediction, he divided knowledge up into

those propositions which were governed by the ideal of truth, which were the prov-

ince of logic, and those which were motivated by ‘‘volitional resolutions’’, which he

called conventions. Conventions often masqueraded as logical statements, but

in fact were nothing more than bald ‘‘decisions’’ in Reichenbach’s lexicon.

‘‘Decisions’’ were neither true nor false, but nevertheless the philosopher could

offer advice concerning their efficacy by, among other things, pointing out whether

they were consistent with one another (Reichenbach, 1938, pp. 9–13). Those fam-

iliar with ‘‘decision theory’’ in OR will recognize their discipline in embryo here in

1930s logical positivism.20

Reichenbach never passed up an opportunity to vent his distaste for traditional

wisdom, which he associated with the conventional character of certain epistemo-

logical precepts; and was unapologetic about his contempt for history. Another

task he undertook was the demonstration of the ‘‘fact’’ that physics had managed

to solve the problem of induction, largely by means of formal models of statistical

inference. Perusal of his later popular works, such as the Rise of a scientific philo-

sophy, shows that these projects melded imperceptibly into a defense of science as

having stood stalwart and incorruptible, arrayed against Marxism (e.g., Reich-

enbach, 1953, pp. 71–72) and Fascism; not such an unusual conviction for one who

had endured the types of persecution which he had suffered. It is noteworthy that

the way in which Reichenbach went about this project was to posit the existence of

a formal logic not readily accessible to the general public, conceived as a ‘language’

that served to inoculate the scientific elect against the irrationality that plagued

many nations in the 1930s and 1940s. Science, he maintained, was situated beyond

dispute by the layman; and the only people who could really appreciate this

immunity were those who had taken the trouble to have immersed themselves in

20 On pp. 27–31 of Experience and prediction one will also discover the ‘physical symbol systems

hypothesis’ of early artificial intelligence avant la lettre, itself soon to also spring forth from its origins in

OR.


the technicalities of physics and mathematical logic: ‘technicalities, not dialectic, isthe instrument of modern philosophy’ (Reichenbach, 1978, p. 253).The retreat to technicalities did not altogether banish any consideration of older

themes of the relationship of science to democracy, those troublesome holdoversfrom the earlier Pragmatist tradition. In Rise, Reichenbach portrays the scientist asgoverned by an algorithmic logic, but by contrast the general populace as governedby ‘‘volitions’’, which he then proceeded to equate with ‘‘preferences’’.21 Althoughhe does not make much of it, this move signals his implicit endorsement of the typeof social theory that starts with given individual preferences of convenient proper-ties, such as that found in neoclassical economics.22 Logic is subject to codificationby the philosopher, but ‘it is therefore irrelevant where volitions come from, andwe do not ask . . . whether we are conditioned to our volitions by the milieu inwhich we grew up’ (Reichenbach, 1953, p. 282). We are then informed that ‘moraldirectives . . . express volitional decisions on the part of the speaker’, and from thishe concludes that ‘Science tells us what is, but not what should be’ (ibid., p. 291).Curiously, the first time that democracy is mentioned by Reichenbach is with

respect to the precept that, ‘everybody is entitled to set up his own moral impera-tives and demand that everyone else follow these imperatives’ (ibid., p. 295), whichconstitutes his working definition of democracy. Conflict is to be expected amongstthe laity, because all they do is blindly try to thrust their inexplicable and incom-patible desires and volitions upon one another. Scientists qua scientists had mana-ged to avoid all that (and corruption by Fascists, and so on) by conforming to thedictates of a logic of empirical evidence, and therefore were located in a space situ-ated outside of the democratic sphere. Note well that science is not portrayed as anarena for the democratic hashing out of workable ethics and values, becausescience is not conceived as a part of society at all. As Reichenbach put it bluntly,‘Science is its own master and recognizes no authority beyond its confines’ (1953,p. 214). Dewey had effectively been banished.Reichenbach was not afraid to upbraid Dewey directly for not maintaining

adequate separation of science from society in his famous essay in the Schlipp vol-ume on Dewey’s philosophy. Although as to be expected Reichenbach beratedDewey for his lack of ‘‘technique’’ (‘philosophic analysis of modern science cannotbe achieved without a profound study of mathematical methods’ (Reichenbach,1989, p. 191), it is important to observe that he also backhandedly acknowledgeda major source of their difference came in how they approached the social

21 Here I differ from an interpretation in Richardson (2000, 2002) that portrays Reichenbach as arguing

for an ineliminable role for volitions in science. It is true that Reichenbach thought they would never be

altogether banished in practice, in the same sense he expected that sociology of science could never be

banished in principle; but the role of the philosopher was to partition them off from the true account of

the operation of rational science. One observes this in the RAND paper quoted at the beginning of this

section. The effect was to ratify the scientist/layperson divide.22 He was undoubtedly familiar with this school, since he had attempted when younger to try and

mathematize economics in conjunction with his friend Carl Landauer. On this, see Reichenbach (1978),

p. 29.


implications of their respective philosophies. In a most dubious rhetorical move,Reichenbach tried to make it appear that he, and not Dewey, was the standardbearer of a solid intellectual basis for the advocacy of socialism:

There are ethical systems which for instance consider the idea that private propertyis sacrosanct as a demonstrable truth in the same sense it is demonstrable that priv-ate property is destructible by fire. It is the danger of pragmatism that its theory ofreality is made to order for ethical theories of this type, although the pragmatiststhemselves may not intend these implications. (Reichenbach, 1989, p. 180)

The Popperians were among the first to flag this sort of argumentation as illegiti-mately stacking the deck against democracy and free inquiry. For them, ‘democ-racy requires a view of rationality that permits dissent, and inductive logic forbidsit’ (Agassi, 1995, p. 158). The situation was even more indefensible once one finallyrealized that the positivists could never really settle upon the unique inductive for-malism which supposedly governed the scientists’ activities: in that situation, thepositivists had failed in providing anything approaching a logical criterion whichwould demarcate the scientist from any other political actor:

Although [the positivists] have not found the right inductive rule,23 they knowthat it should justify established scientific opinion, and so they demand thateveryone follow this still-not-known rule of induction and endorse receivedscientific opinion under pain of being branded irrational. (Agassi, 1995, p. 158)

The other place in which Reichenbach drove home his separation of sciencefrom society was in his infamous distinction between a context of discovery and acontext of justification, which first appeared in his Experience and nature (1938,pp. 6–7). He wrote there that ‘epistemology is only occupied in constructing thecontext of justification’, but curiously enough for a topic relegated to the categoryof the descriptive tasks of epistemology, the philosopher would not recover theactual social modes of argumentation and reason, but rather a ‘rational reconstruc-tion’ of what should have been said according to the canons of the philosopher.Not only was a wedge being driven between philosophy and sociology, but anotherwas driven between the actual prosecution of science and its supposed logical struc-ture (ibid., p. 381). As has frequently been noted, it was a convenient immuni-zation stratagem for the logical positivists to prevent their theory of the scientificmethod from ever being falsified with data from the history of actual scientificinquiry: ‘all this is a logical reconstruction. It was never intended to be an accountof the origin and development of scientific theories’ (Feigl, 1969, p. 17). But it wasalso much more than that. In Rise of scientific philosophy, Reichenbach tied adher-ence to the distinction to the very legitimacy of a logic of inductive inference.

23 Many positivists were admitting this by the late 1960s. Witness Herbert Feigl: ‘the so-called problem

of induction . . . cannot be solved either by the logical or the statistical concept of probability’ (1969,

p. 11). By the 1970s, it was clear that the positivists had completely misrepresented a monolithic charac-

ter of probability theory: see Fine (1973).


There he admits that induction could never help the scientist find a new theory (apoint made much more cogently by the pragmatist C. S. Peirce), but could onlyused to evaluate it after the fact: ‘The act of discovery escapes logical analysis . . .But it is not the logician’s task to account for scientific discoveries . . . logic is con-cerned only with the context of justification’ (Reichenbach, 1953, p. 231). Develop-ments categorized as falling within the context of discovery were placed upon thesame epistemic plane as ‘‘volitions’’: one did not ask where they came from. Thispolicy of ‘‘don’t ask, don’t tell’’ proved extremely awkward for his philosophicalsystem, since it threatened to situate a major component of what everyone elsethought of as science, namely, its capacity to generate novel explanations ofphenomena, stranded outside of the cloistered ideal community of scientists, andleave it mired in the irrational bog of ‘‘society’’. Nevertheless, it would prove indis-pensable for his political program. It was almost as though the later Reichenbachwas so driven to erect impermeable floodgates between science and society that hewas willing to risk the interior of his citadel providing shelter for the diminishedscientific status quo as containing a few ‘‘reputable’’ physicists and hardly anyoneelse. The role of the philosopher had thus been rehabilitated as mounting the‘‘defense’’ of the upper reaches of the orthodoxy of the physicists in the worldaccording to the philosopher; whether the physicists felt they needed this assistancewas a moot point.Perhaps it commits the Whig fallacy, but it now seems a bit difficult to under-

stand why anyone would find this version of ‘‘positivism’’ compelling on its face.24

The answer tendered in this paper is that the explanation must be sought in thepact which postwar American philosophy of science made with OperationsResearch. Put bluntly, it was no accident that so very much of Reichenbach’s laterphilosophy resembled the attitudes and content characteristic of postwar OR,because for a short while, they were actually one and the same. The idolization ofphysics, the contempt for tradition, the insistence upon a generic ‘‘scientificmethod’’ based on logic and probability and indifferently portable to any subjector discipline, the crusading stance as rooting out ‘‘error’’ which besets the masses,the role of the philosopher as a consultant therapist for decision theory, the postu-lation of a closed corporate priesthood which possessed sole control of theseesoteric methods, the suggestion that the problem of induction had been dissolvedby statistical algorithms, the conflation of mathematical prowess with intellectualvirtue, and the assertion that the advice provided was pitched somewhere beyondethics or morality, were all hallmarks of American OR. Operations Research hadbegun as a negotiation to create a separate social identity for scientists operating ina semi-autonomous capacity within the military; postwar logical positivism endedup as an ideology of science in general as a virtual community separate and auton-omous from the social system as a whole.

24 As Ron Giere asks, ‘How did a naturalistic pragmatism incorporating an empirical theory of inquiry

get replaced by a philosophy that regarded induction as a formal relationship between evidence and

hypothesis?’ (1996, p. 347).


The symbiotic pact between OR and logical positivist philosophy of science

would not have flourished if it did not take root in fertile ground. There were a

number of circumstances specific to the postwar American context which were pro-

pitious for the graft. The first and most important was the shift in science funding

and management from industry dominance to military dominance. The various

protocols worked out during World War II—viz., that scientists would conduct

research under contracts that were issued by the military but managed through

their universities, that they would be subject to indirect controls through security

clearances and classification of secrets, that downstream development would be the

province of the military science managers, that bureaucratic evaluation would be

deployed through ‘‘peer review’’, that grant overheads could buy off the principal

investigator’s academic obligations to his home institution—all militated in favor

of treating the scientist as though she were a member of a community apart from

the general run of intellectual life. This reification of a separate and unequal

science reflected a Cold War truth in a manner it never could have done under the

previous industrial regime of science organization. The notion of the scientist as

sequestered in an ivory tower was encouraged by the military, especially once a few

of the elect physicists had suffered crises of conscience about ‘‘knowing sin’’ after

the dropping of the Bomb. The notion that scientists somehow were members of a

commonwealth apart from society became smoothly integrated with the oft-

intoned refrain that they could not be held responsible for how their discoveries in

‘‘pure science’’ were put to use by ‘‘others’’. Of course, the very plausibility of the

notion of a free-floating ‘‘pure science’’ detached from its prospective utility and

retrospective funding was itself the product of a fair amount of legal and economic

construction initiated by the military.25

The relationship of military-organized science to democracy was perhaps the

issue most fraught with controversy in the immediate postwar period. The military

and the Operations Researchers shared a jaundiced view of democracy when it

came to prosecuting wars, and the suppression of democratic debate over the use

of the Bomb was viewed by many in the political classes as a betrayal of funda-

mental principles which the Bomb was conceived nominally to protect. Dewey’s

blind faith in democracy had, therefore, to be revised in the Cold War era. OR the-

orists responded to the call, and went to work describing various ways in which

democratic decision procedures were ‘‘irrational’’ when it came to such momentous

choices.26 The most famous of these doctrines produced at RAND was the ‘Arrow

Impossibility Theorem’, based directly upon the assumptions of neoclassical econ-

omic theory:

25 This case has been lately made by Asner (2002), who describes many of the legal, accountancy, and

economic devices invented in order to give some solidity to the category ‘‘pure science’’.26 The complex interplay of attitudes towards democracy, OR, and decision theory in this period are

covered at length in Amadae (2003). Pp. 128–132 especially explain how decision theory was a negation

and repudiation of Dewey’s conception of science and democracy.


If we exclude the possibility of interpersonal comparisons of utility, then theonly methods of passing from individual tastes to social preferences which willbe satisfactory . . . are either imposed or dictatorial . . . the doctrine of voters’sovereignty is incompatible with that of collective rationality. (Arrow, 1951,pp. 59–60)

The upshot of this claim was that market expression of citizen preferences was afaithful and dependable representation of their desires, whereas standard majorityvoting procedures were not. This was an extremely felicitous Cold War doctrinefrom the military viewpoint, since it suggested that the military was legitimatelydefending the welfare of the citizenry by allowing it free choice in its purchaseswhile simultaneously conducting the national defense without the need for itsexplicit political acquiescence. This ‘‘double truth’’ doctrine had its exact parallel inReichenbach’s separation between science and society: scientists were furtheringthe welfare of the citizenry by allowing them free choice in the ‘‘products’’ of theirendeavors in the marketplace while conducting their fundamental research withoutthe need for prior accommodation or any explicit political acquiescence.27 Demo-cratic procedures were best kept well clear from such activities.The Cold War also constituted a watershed in the history of the American social

sciences. The earlier orthodoxies of Institutionalist economics and habit psy-chology had suffered precipitous declines during the war and after (Yonay, 1998;Mirowski, 2002). Hence Dewey’s program to ‘‘naturalize’’ philosophical inquiry,insofar as it bore its commitments to these particular social science doctrines on itssleeve, was increasingly seen as backing the wrong horse. Not only would Europeanemigres find these commitments utterly unintelligible, but indigenous philosopherswould also be aware of more concertedly ‘‘scientific’’ trends in economics and psy-chology (often jump-started in America by other European emigres). Psychologygrew more behaviorist and individualistic under wartime exigencies, and neoclassi-cal economics rose to dominance in this time period. The gambling metaphor wassuffused throughout postwar psychology (Goldstein & Hogarth, 1997, p. 10), andas Reichenbach himself put it, ‘the scientist resembles a gambler more than a pro-phet’ (1953, p. 248). It is again no accident that both these movements in econom-ics and psychology were themselves linked to OR, especially beholden to it fortheir trademark mathematical models and statistical protocols; any philosophywilling to posit the motivations of the larger public as volitions/preferences andthe rationality of the scientist as algorithmic optimization would find that theirdoctrines would resonate with a new postwar cross-disciplinary orthodoxy of thesciences of man. Thus OR reshaped both the social sciences and the philosophy ofscience in America. Philosophy as the logical ratiocination of the isolated individ-ual, cocooned with his identical brethren in research universities away from thetemptations of the crass world of the public, was born.

27 Here Reichenbach’s autonomy of the scientist dovetails very nicely with the ‘‘linear model’’ of

research and development popularized by Vannevar Bush in the same era.


The exact character of the role of the market in OR and logical positivism wasalso rooted in the politics of the immediate postwar era. How could the logicalpositivists have been such pliant Cold Warriors when they identified themselves associalists? Here is where left/right distinctions should not be taken for granted asdictating political affiliations, but need to be translated into the specific spatio-temporal context. The fact of the matter was that in the 1950s, the OR professionwas itself a veritable hotbed of self-identified socialists, both in Britain and in theUS. The reason for this incongruous fact was that it was staffed by people whoregarded themselves as applying scientific methods of command, planning, andcontrol to improve the efficiency of social action undertaken by groups such asarmies and governments. It should be remembered that the military constituted thelargest planned economy in any Western nation, and that the safest way to wardoff perilous accusations of being ‘‘soft on communism’’ and ‘‘un-American’’ (aheightened concern for those having recently emigrated) was to sport a militaryclearance. A certain type of non-Marxian socialist found the broad church ofOperations Research a most congenial shelter from the storms of the McCarthyera; Reichenbach fit right in.28 An American socialist (who was also a Europeanrefugee?) in the Cold War would have enthusiastically embraced the notion thatscience existed in a world apart from the world they were forced to live in.Hans Reichenbach did not just resemble an Operations Researcher in the

immediate postwar period; he became an Operations Researcher. As his studentNorman Dalkey (himself a RAND employee) informs us, Reichenbach signed onas a consultant at RAND in 1948, that is, very soon after it broke away fromDouglas Aircraft and reconstituted itself as a free-standing non-profit corpor-ation.29 So what would a logical positivist philosopher be doing in a military think-tank dedicated to ‘‘thinking the unthinkable’’ about nuclear war? The clarificationbegins with comments by another Operations Researcher who came to RAND in1951, Albert Wohlstetter:

I had known some of the first people who were on the RAND staff because theywere also mathematical logicians: J. C. C. McKinsey, and Olaf Helmer, and alsoM. A. Girschick . . . [they told me about] one of the most important contribu-tions to metamathematics by Alfred Tarski, which had never been published,which showed that it was possible to get a decision method, an entschi-dungsverfahren, for a rather large and rich section of classical geometry, corre-sponding to high school geometry . . . Well, that was very interesting to

28 Although we shall not explore the issue here, this fact helps clear up the noxious fog that has

surrounded Alan Sokal’s quest to ‘‘save’’ the American Left from the postmodern Left and its tendency

to see scientists as co-opted by the Right. The specifically American politics of the Science Wars has

been thoroughly misconstrued by Europeans generally, and is something that is left opaque by commen-

tators such as Brown (2001) and Kitcher (2000).29 In Reichenbach (1978), p. 51. The history of RAND has recently attracted the interest of a number

of historians, even though access to their archives is still severely restricted to outside researchers. Some

good sources are: Hounshell (1997); Collins (1998); Jardini (1996); Hughes & Hughes (2000); Amadae

(2003); Mirowski (2002, Ch. 4); Rescher (1997b).


mathematicians. I was surprised, however, to find that RAND, this organizationthat worked for the Air Force, was publishing a research memorandum byTarski which presented this result, because it didn’t seem to me [back then] tohave much directly to do with strategic bombing or anything of that sort . . . wejust ran into Abe Girschick, Olaf Helmer and Chen McKinsey on the street, andthey were overjoyed to see us. Mathematical logic was a very, very small world.There were only a little over a dozen mathematical logicians before the war inthe US, and two jobs in mathematical logic. For the rest, you had to teach cal-culus, as Chen McKinsey did, or philosophy of something of that sort.30

Prior to the Cold War, career options in mathematical logic and formalist philo-sophy of science were miniscule to non-existent, as this comment acknowledges.When OR units at RAND, Stanford and elsewhere began to hire mathematicallogicians at ‘unprecedentedly remunerative’ rates with military funding (Quine,1985, p. 217), this fostered the basis for the professionalization of these disciplinesin postwar America. During the war many logicians had worked on cryptography,command and control, and the development of the electronic computer, so theywere predisposed to become integrated into the burgeoning world of the defenseintellectual after the war. For these reasons, RAND stood out as the locus of thedensest concentration of mathematical logicians in the USA in the 1950s. Further-more, RAND was predisposed to hire a particular type of formalist philosopher: asanother RAND consultant, Willard von Orman Quine put it, ‘young philosophersof the Carnap persuasion’ (1985, p. 217). RAND did not employ this specializedclass of philosophers because they were especially interested in the elimination ofmetaphysics through the logical analysis of formal language or anything else ofthat sort, but rather because the logical positivist program could readily be sub-ordinated to the objectives of the Operations Researchers, as we have enumeratedabove.The fact that the logical empiricist program in the US owed its good fortune to

OR has not previously been the subject of much commentary, at least in partbecause the linkage has been suppressed by the participants themselves. We mustremember that a precondition of joining the defense establishment was a willing-ness to submit to its stringent requirements concerning classification, secrecy andthe ‘‘double truth’’ doctrine. In my experience, the vitas and bibliographies of thefigures in question omit their military papers; their archives have been ‘‘sanitized’’of most of their military records; their collected works have blank pages. In theirretrospective accounts, if they mention their ties at all, they tend to treat themas ‘boondoggles’ (Quine, 1985, p. 217), ‘diversions’ (Davidson, 1999, p. 32) andother trivial pursuits. Even figures who are a bit more open about their RAND

30 Albert Wohlstetter interview with Martin Collins, 27 July 1987, pp. 1–2. RAND Oral History Pro-

ject. Transcript from: National Air and Space Museum, Washington, D.C. Wohlstetter had studied with

Tarski as an undergraduate at CCNY, and had done graduate work in logic and philosophy of science

at Columbia. He later was infamous as one of the hawkish operations researchers that defined RAND’s

systems analyses of nuclear war.


experience, such as Nicholas Rescher (1997a, Ch. 8), are still noticeably reticentabout discussing exactly what it was they did under those auspices. The RANDarchives themselves, long after the Fall of the Wall, are still effectively closed tomany outside researchers. The dense web of interconnections which tied the fledg-ling OR community to the following important figures in postwar philosophy ofscience still awaits its historian: Olaf Helmer, Carl Hempel, Paul Oppenheim,Alfred Tarski, Willard Quine, John Kemeny, J. C. C. McKinsey, Patrick Suppes,Donald Davidson, Nicholas Rescher, Leonard Savage, and Rudolf Carnap. How-ever, we can give some idea of the shape that influence took in the work of HansReichenbach.Norman Dalkey (in Reichenbach, 1978, p. 52) informs us that Reichenbach

wrote at least three papers for RAND while a consultant there. Two of them, bothdated 1949, were ‘General form of the probability for war’, RAND D-515, and‘Rational reconstruction of the decision for war’, RAND D-539. The objectives ofthese papers, consonant with the general orientation of OR at that time, was‘Based on a rational reconstruction of Russia’s doctrines, [to] supply the math-ematical form of the probability that Russia will go to war’ (1949, p. 23). Whileone rather doubts that any American war planner ever even entertainedReichenbach’s master equations #35 and #36, much less found them useful, theirsignificance from our present perspective is the extent to which they resemble hisformalization of the inductive problems of the scientist, and the way in which theydraw a distinction between the ‘‘scientific’’ approach to war and the less-than-rational ‘‘political’’ approach presumed to hold sway over other sorts ofsocial formations. Reichenbach starts out by distinguishing between an empiricistapproach—that is, actual intelligence gathering—and a ‘‘rational reconstruction’’of the decision to go to war, an ‘‘imagining ourselves in the enemy’s place’’, whichwas the nature of his own exercise. His free flight of imagination is disciplined bythe postulation of a ‘‘valuation function’’ resembling the utility function of neo-classical economics and Reichenbach’s own ‘‘volitional’’ preferences discussedabove, as well as construction of a formal model of inductive inference. While thedecision to go to war sounds a whole lot like the decision whether or not evidenceconfirms a scientific theory in his account, the problem with understanding theenemy (as Reichenbach conceives it) is that he is susceptible to all sorts ofirrational motivations (see the quote heading this section). Reichenbach’s solutionis to treat these irrational elements as random, and adjust the model with aparameter j to allow for the ‘degree of rationality’ of the enemy (1949, p. 13). Hethen discusses the possibility of the empirical estimation of j, but worries that it isobservationally indistinguishable from another variable he calls ‘‘the scale ofdecision ratings’’; to thwart the threatening underdetermination problem, he thenposits a ‘‘coordinative definition’’, a theme borrowed from his earlier work on thephilosophy of space and time.This paper helps reveal the ways in which OR did not give rise to the philosophy

of logical positivism, but it most certainly transformed it in the American contextas a prelude to launching it as the characteristically professionalized philosophy ofscience that came to dominate the American landscape. Many of the representative


themes are present there: the sharp separation between scientific rationality andsocial structures (as if the Russians did not have their own Operations Research-ers!), the suspicion of democracy, the contempt for history and cultural considera-tions, the conflation of social groups with individual agency, the confusion ofstatistical inference with scientific method, the portrayal of science as algorithmic,the veiled endorsement of Western economics through recourse to a behaviorist‘‘decision theory’’, the posturing concerning agnosticism about values, the under-estimation of the difficulties of empiricism (especially with regard to the under-determination of theory by evidence), the conflation of mathematical logic withhuman rationality, the absence of any actual empirical evidence, the psychologicalmetaphor of gambling, the claim to esoteric expertise. Some holdouts from the pre-vious Pragmatist school may not have wanted to subscribe to much of this, butthen they had less and less to say about it, as they were left out in the cold by thepostwar reorganization of academic science by the military.

5. Kitcher’s philosophy for the technology transfer bureaucrat

When we confront an intricate causal explanation of how particular features ofour lives are brought about through the interaction of molecules, will we be ableto sustain our sense of the special value of our lives? . . . Grand rhetoric abouthuman freedom seduces us into thinking that we must, quite literally, make our-selves . . . the conception was always incoherent. If the self that allegedly makesitself is already fully formed, then it does not, after all, make itself; if it is notfully formed, then it does not make itself. To find our freedom, we have to startacknowledging that we are the people we are because of events that are beyondour control, even beyond our understanding. (Kitcher, 1996, pp. 272, 282)

If one trembles while making broad generalizations about the Cold War scienceand its social milieu because they still might be considered controversial, then justtry and propose a few simple observations about what is happening to science rightnow. The writing of contemporary history has always been fraught with traps forthe unwary, ranging from descent into journalism to ascent into the most rarifiedforms of wishful thinking. As we shall see, it is not just geriatric generals who aresusceptible to the tendency to remount the last campaign in the name of a clearand present danger. Nevertheless, it will be necessary to give a brief characteriza-tion of the current situation of science, in order to make some arguments about thestatus of contemporary philosophy of science in parallel to those broached in theprior two sections.It has now become commonplace in the science policy and science studies litera-

tures to concede that something rather dramatic has happened to the organizationof science and the university in the quarter-century or so since 1980.31 The chal-

31 Consult the following for documentation: Mirowski & Sent (2002); Bok (2003); Biagioli &

Galison (2003); Nowotny, Scott & Gibbons (2001); Noll (1998); Fuller (2002a); Dreyfus, Zimmerman, &

First (2001); Mirowski (2003, 2004).


lenge is to characterize exactly what that something consists of, since we are all stillcast adrift in the midst of it. The way many contemporary observers have talkedabout this interconnected set of phenomena is by asserting that the military domi-nance of science funding and organization in America has more or less waned,only to be replaced by the progressive extension of corporate commercial controlof science. The first pitfall to be skirted in making such a generalization is that thistransformation constitutes some sort of ‘‘return’’ to the corporate situation whichreigned supreme prior to 1940: nothing could be further from the truth. Indeed,one of the curiosa which has confounded perceptions of the current situation isthat the very largest corporations have spun off or otherwise divested themselves oftheir in-house corporate labs over this period, essentially gutting what had beensome of the most productive and hallowed research organizations in Americanhistory: AT&T spun off Bell Labs, IBM has downsized Yorktown Heights, Xeroxhas divested itself of PARC, and so forth. The subjection of science to the modernmarketplace has been prosecuted through a whole range of unprecedented socialinnovations, most of which take place in the fine structure of research practicesand protocols, never even remotely entertained in the Jazz Age. While we cannoteven begin to address the question of why this has happened in the space of thispaper,32 we can however briefly make an inventory of those changes, as a necessaryprelude to discussing the modern predicament of the philosophy of science.Most observers agree that the transformation began in the 1980s when a wave of

concern over the loss of American ‘‘competitiveness’’ swept politicians and corpor-ate leaders into a frenzy. This meant a diversity of things to many different con-stituencies, but was cashed out in science policy to mean that unquestionedAmerican superiority in pure science was not being adequately transmuted intodevelopment and innovation in the commercial sphere, at least relative to commer-cial rivals. Politicians began to respond to this perception with a whole sequence oflegislation, generally conceded to have begun with the Bayh–Dole Act of 1980.This legislation changed the entire relationship of universities to governmentfunding of scientific research, all subordinate to the aim of rendering ‘‘technologytransfer’’ from the academic lab to the corporate setting more smooth and rapid.The proportion of federal funding of science in America in general fell, while thatof industry rose sharply. The second pitfall in surveying modern science is toregard this as solely an American phenomenon. What had commenced as anationalistic initiative was rapidly co-opted by vanguard elements of the corporatesector, however, who were not themselves all that avidly concerned with Americannational superiority, but did seek to bend developments to their own ends. Forinstance, Ryan (1998) and Sell (2003) reveal how the Uruguay Round of GATTtrade liberalization negotiations was captured by pharmaceutical, computer, andentertainment interests in order to endow the movement to extend American inno-vations in intellectual property law to other countries with real political muscle. Ina nutshell, the aims of activist segments of the corporate sector were to cut their

32 But see Mirowski & Sent (2002); Mirowski and Van Horn (forthcoming); Mirowski (2003); Sell (2003).


R&D expenses by outsourcing their research activities in much the same way theyhad previously outsourced their labor-intensive manufacturing processes, and yet,by the same token, to exercise greater control over the ‘‘intellectual property’’which might be generated extramurally. In international venues, ideas and objects(genomes, algorithms, bits) which had never before been treated as commoditieswere now subject to an enclosure of the scientific commons, hedged round withthickets of legal restrictions. The vanguard of this movement was to be found inthose industries which were not committed to classical heavy manufacturing, suchas pharmaceuticals, biotechnology, computer software, and entertainment. The factthat these industries were perceived as dynamically progressive and the high-techwave of the future only lent their crusade to reconfigure science a patina ofcredibility.The consequences of these legal, financial, and social innovations have yet to be

completely digested in the modern context. Universities have more recently sufferedthe brunt of this revolution in science regimes. To put it bluntly, few would claimwith confidence anymore that the university primarily serves the interests of state-building, which was, of course, the genesis of the original German academic model.The purpose of university education today is no longer socialization, but insteadthe vocational stocking up on ‘‘human capital’’. Universities don’t produce‘‘citizens’’ anymore; indeed, the majority of those receiving advanced training inAmerica in the natural sciences are foreigners, so the problem becomes how to jus-tify what remains of public subsidy to train the manpower for tomorrow’s rivals. If‘‘information’’ is a commodity, as it surely has become, then the question confront-ing universities is how to tame their faculty to produce the commodity efficientlyand cheaply, and consequently to capture more completely the profits of infor-mation production. Indeed, aren’t many of the perquisites of the faculty justbloated costs that need to be swept away by gales of creative destruction, withresearch perhaps sited in leaner and meaner structures such as the firms located innearby ‘‘research parks’’, or better yet, the newly burgeoning ‘‘contract researchorganizations’’?33 This profound change of mindset has recast the working imageof the universe of research as one big ‘‘marketplace of ideas’’; hence the questionof the relationship of science to society has become rephrased to one of efficientbalancing of costs and benefits.The fallout for scientists has been profound. First and foremost, the Cold War

obsession with the freedom of the individual scientist has melted into thin air; afterall, in this brave new world it is only the entrepreneur who truly relishes freedom.The indiscriminate extension of patents to all sorts of entities threatens to renderwhole areas of scientific research legal minefields, with (among other truly bizarredevelopments) the owners of prior intellectual property claiming downstream

33 The rise of contract research organizations in the 1980s, mostly though not exclusively in the area of

pharmaceuticals, is one of the great blind spots of modern science studies. An attempt to gauge the

extent to which they have the capacity to displace university research on a larger scale is the subject of

Mirowski & Van Horn (forthcoming).


ownership rights over the output of your research because you made use of theirconcept/device as research input (Eisenberg, 2001). As footsoldiers of the NewOrder, academic technology transfer officers tie researchers in knots in the name ofcapturing the pecuniary benefits of discoveries for the university. The honorific of‘‘authorship’’ itself becomes decoupled from those doing the actual research workonce it becomes, like everything else, subject to being bought and sold.34 The struc-tures of scientific publication have also become intimately influenced by the privati-zation of the journals (Scheiding, forthcoming). Bureaucratic evaluation structuresabound, with scientific productivity ‘‘rated’’ on a periodic basis. Graduate studentsand postdocs become the reserve army of casual labor in an atmosphere ofperpetual cost-cutting, with ‘‘educational’’ preliminaries to the ever-receding mir-age of a tenured academic position causing the standard career path of the mediannatural scientist to be stretched out interminably. In the never-ending search forcheaper labor, the site of research itself is increasingly shifted to poorer countries,especially if it involves human subjects.I would argue that, just as in the 1940s/1950s, emerging alterations in the

organization of science are not uniformly apparent across the board, nor initiallywell understood, and that philosophers of science have clocked in a little late and afew dollars short of the realization that their world has been irrevocably turnedupside-down. In particular, taking the components of the current regime changeinto account, the impression that philosophers have finally come to grips with the‘‘social dimensions of science’’ with which this paper began would seem little morethan an inadequate reflex to a nagging feeling that something has drastically chan-ged since the 1980s, and that philosophers must diversify their portfolio if they areto have anything at all to say about it. They have acknowledged that physics nolonger stands poised as the pivot around which all other sciences are thought torevolve, and have even begun to pay serious attention to biology, but without ask-ing why one science comes to displace another as the cultural icon of progressiveresearch and fruitful knowledge. (Hint: consult the above list of ‘‘progressive’’industries.) They have unmasked the ‘Legend’, and now acknowledge ‘‘error’’ ismuch more context sensitive. They have even begun to pay serious attention to theactual history of their own discipline. Yet instead of dealing with the root cause oftheir discomfort, which is the fact that the globalized privatization regime of scien-tific research has left many of their traditional concerns (unimpeachable truth,empirical adequacy, realism as permanence) languishing in irrelevance, they havesought to argue that their new-found version of ‘‘social epistemology’’reassures usthat social considerations do not unduly derange the portrayal of science handeddown from their predecessors, the logical positivists. Science, they reassure us, stilltranscends the sound and fury of hoi polloi in the marketplace. Some rambunctiousgroups, and maybe some feminists, might complain about this or that aspect of

34 Biagioli & Galison (2003) collect some articles that examine the problems of attribution of ‘‘author-

ship’’ in modern science; Mirowski & Van Horn (forthcoming) discuss the ‘‘guest-ghost author’’ in bio-

medicine.


science—perhaps it has intermittently been put to nefarious purposes, or been twis-ted to ‘‘undemocratic’’ ends in the past—but philosophers can nonetheless feel con-fident they can enlighten the populace in ways to offset this, or counsel them howto live with it. Moreover, they can accomplish their task by building little gametheoretic models of science! It should not escape the reader that game theory wasthe jewel in the crown of Operations Research in the 1950s, and indeed constitutesthe core doctrine of modern neoclassical microeconomics today.35

Philip Kitcher will serve as our exemplary protagonist in this modern drama,although as we have mentioned before, many other contemporary philosophersalso partake of a similar approach to ‘‘social epistemology’’. Kitcher’s intellectualtrajectory can perhaps serve to provoke us into rethinking how the scientificdimensions of social knowledge play themselves out in modern philosophy. In hiscase, what began as an attempt to banish the twin specters of Kuhnian relativismand the skeptical prognostications of sociologists of science in his Advancement ofscience (1993) became transformed into a defense of the Human Genome Project inThe lives to come (1996), then an ill-starred call to reinstate a Mertonian sociologyof science (2000), and then, finally, an attempt to reconcile the demands of demo-cratic principles with the structures of science in Science, truth and democracy(2001). We should not, however, infer that because the wheel has turned roundonce more to a corporate dominance over the practice of science, that somehowthis also betokens a return by philosophers to the concerns and prognostications ofDewey. It would seem that disciplinary exigencies demand the appearance of conti-nuity with previous philosophy of science, but the continuity that is conjured iswith the positivists, and not Dewey’s brand of philosophical engagement. What isironic is that although each of our protagonists in this paper considered himselfliberal-to-socialist on the political spectrum, yet each only dimly perceived the out-lines of the New Economic Order that had selected them as representative of theemerging account of science in action.What is the problem to which the bulk of Philip Kitcher’s intellectual pursuits

are devoted? Kitcher, it seems, bears a visceral wariness of the looming irrelevanceof modern philosophy of science, and wants to open a breach in the wall betweenscience and society that Reichenbach and his comrades had erected; but he alsobelieves that this is a perilous project, since it threatens to let the barbarians cometumbling into the breach. Sometimes he writes as though the scientific communitywere composed of strong enough stuff that it could withstand most unruly incur-sions; but the fact that he repeatedly keeps revising his account of the vinculumwhich binds them together suggests that there persists the nagging worry that atruly ‘‘socialized’’ science is vulnerable to corruption, and that the duty of the phil-osopher is to diagnose any debilitating social disease. It seems that Kitcher wantsto convince us that scientists still really do effectively constitute a society apart, atleast for epistemic purposes. Here the contrast with Reichenbach is instructive.Due to his capacity as operations researcher, Reichenbach was aware that there

35 For a history of game theory, see Mirowski (2002), Chs. 3 & 6.


were important social actors who had good reasons to maintain at least the pre-tense of the existence of a separate scientific community, and a social sciencetradition which might buttress it; whereas Kitcher seems not to appreciate that thenew paymasters have little interest in underwriting elaborate justifications for sci-entists as a breed apart, or share his concerns over their vulnerability to hoi polloi.Perhaps the corporate research officers want their scientists to mix with the barbar-ians a bit more. Furthermore, contemporary social sciences are probably moreattuned to follow the money in this respect. Yet Kitcher has come to represent thecentrist position in philosophy of science due to the fact that he seems to lend legit-imacy to the modern privatization regime.In Kitcher’s case, the entire sequence of his writing is dominated by an overriding

fear of attempts to encroach upon the self-sufficient scientific community by thosewhom he would regard as afflicted with misguided enthusiasms. Earlier in his careerthe barbarians were identified as creationists, sociobiologists, and (perhaps) Bour-bakists; more recently, Latourian sociologists of science have assumed the mantle.In his recent book, he dismisses the following positions as ‘‘Marxist’’ (Kitcher, 2001,p. 172): that one vice of the natural sciences is elevating the efficiency of means overends, that natural scientific models applied to society distort and dehumanize ourunderstanding of ourselves and others, and that ‘science turns out to be the scullionof capitalism’. Dewey, of course, would be offended. All in all, one doubts whetherParisian sociologists of science comprise the vanguard of third columnists seeking toundermine the integrity of the scientific community from within. Curiously enoughin a book nominally concerned with citizen intervention in science, there is not a sin-gle reference to the science studies scholars and activists who analyze and promotesuch activities: Sheila Jasanoff, Brian Wynne, Richard Sclove, Yaron Ezrahi, and soforth. But then, just as Reichenbach and Feigl insisted, perhaps all Kitcher is pro-posing is a ‘‘logical reconstruction,’’ and not an empirical account of the politics ofscience and democracy. But the repressed problem persists whether a coherent‘‘scientific community’’ makes much sense in the present regime.Probably in the interests of maintaining continuity with previous philosophy of

science, Kitcher reprises many of Reichenbach’s analytical moves, such asReichenbach’s terminology of ‘posits’:

Instead of thinking about the virtues and vices of whole theories . . . we shouldseparate those parts of past doctrine that are put to work in prediction andintervention, the ‘working posits’ . . . The practice of inferring truth from successshould be understood as one in which we support belief in the working posits ofour theories. (Kitcher, 2001, p. 18)

If there is any concrete divergence from Reichenbach, it is that Kitcher hasrepeatedly restated his position as acknowledging that the larger social structurehas something to do with the way scientific research is carried out, but that in themajority of cases, the criteria for the success of science are effectively independentand separable from the value criteria by which individuals and societies gauge theconformity of science to their values and aspirations (ibid., p. 124). This is nothing


more than an ‘‘invisible hand’’ argument and, as many have noted, there is astrong undercurrent of the treatment of society as a ‘‘marketplace of ideas’’ inKitcher;36 science purportedly serves transcendent ends which are not and need notbe the ends of the individuals engaged in their idiosyncratic self-interested pursuits.Furthermore, these ends are said to be intrinsic goods in and of themselves, decou-pled from the system of incentives and restrictions imposed by the paymasters ofscience.37 It is significant that Kitcher has never been able to definitively settleupon the exact identity of these transcendental goods over the course of his career:they started out looking like formal notions of Tarskiesque ‘‘truth’’, and then weremutated to the identification of putative ‘‘natural kinds’’ that later became trans-mogrified into the ‘‘unification of explanation’’; and as for the present installment,he seems to have retreated to the logical empiricist default position of providingaccurate predictions. But no matter: Kitcher’s very real conundrum is that almostnobody today believes any longer that these abstract philosophical criteria of thesuccess of science are sufficient to underwrite a separate, self-governing andindependent cloister of scientists supported by unlimited public funds: in otherwords, Cold War philosophy of science is dead.Here is where the themes broached in our previous sections come back into play.

During the Cold War, when most science was primarily funded and organized bythe military and the academic system was expanding exponentially, stories likethose told by Polanyi and Reichenbach were eminently useful. They asserted thatthe public should simply acquiesce in the kinds of science it got in return for itsinvestment, because scientists were the only legitimate judges of their own activityand, in any event, the public could never fully come to comprehend how scienceworked. As Reichenbach said, Science should be its own master. This impliedScience constituted a coherent autonomous social structure: scientists should notset out to challenge research programs in their own areas of specialization, butrather forge their identities and chart their personal progress within parameterspreset by their peers. Scientists were ushered into the mysteries of apprenticeship,not freely joining a debating society or exercising their political rights to freeassembly. The myriad ways in which this doctrine dovetailed with such outsidestructural impositions as security classifications, open vs. closed publication outlets,activist science managers, handpicked peer review panels, and so forth should beobvious. Now, in the modern regime of globalized privatization and flexibleresearch specialization, all of that insistence upon the independent pristine consti-tution of the scientific sphere seems more than a little threadbare and outdated.Short employment spells and bowling alone are no longer just the prerogative ofthe blue-collar worker: most scientists are just employees. The old Reichenbach

36 See Longino (2000b); Mirowski & Sent (2002); Porter (2003); Turner (2003).37 See Kitcher in Callebaut (1993), p. 216: ‘there are special epistemic goals that in a certain sense we

all tacitly know; and that the philosopher’s task is to make those kind of things explicit’. Kitcher’s

account of the science policy literature is misleading in this regard: ‘Despite its elitism, Bush’s report . . .

was plainly permeated by democratic values’ (2001, p. 141). Bush was one of the architects of the Cold

War regime, as we have insisted above.


catechism has lost much of its conviction. Indeed, Kitcher’s pressing problem hasbecome to argue that the modern corporate research structures and their academicoffshoots which are increasingly results-oriented in the shorter term, stress bureau-cratic accountability to outsiders, are infested with lawyers, and less infatuated withthe disciplinary power of professionalization, are nonetheless insignificant in theirimpact upon the true transcendental goals of science. Be it a CRO, Craig Ventner’sCelera, or the British Research Assessment Exercise, never fear, Science will find away. What is required is a new apologetics for the technology transfer bureaucrat.The way Kitcher has sought to try and square this circle is to resort to the

language and mathematical models of the neoclassical economists, or as he puts it,‘inviting a collaboration between the philosophy of science and the social sciences toinvestigate how we might make up for the deficiencies’ of modern scientific organi-zation (Kitcher, 2002a,b, p. 569). But is this a substantive collaboration, or is it justdressing up the older Ivory Tower myth in clothes you can wear to the corporatecocktail party? Recall that science was not deemed to even have an ‘‘economics’’until very recently (Mirowski & Sent, 2002). Emulating the language of neoclassicaleconomists, Kitcher avows science is driven by ‘‘interests’’ (and who would want todeny it in the present atmosphere of corporate science?), but these interests are con-strued as merely individual biases, which must be modeled as ‘‘preferences’’.38 Pre-ferences are not systematic or structural (Reichenbach counseled ‘Don’t ask’); andmoreover, they merely serve to focus our attention. Hence interests just affect wherewe look for our answers, not how we conduct research or what it is that we findthere. Thus every question about possible pathologies of modern scientific researchboils down in one way or another to questions about ‘‘preferences’’ and what to doabout them. While this certainly is the way analyses are conventionally framed inneoclassical economics, it is not so very clear that they help solve conundrums in thephilosophy of science. Here Kitcher becomes involved in awkward contradictions,exacerbated by the fact he believes he can incorporate political, moral, and econ-omic theory into Cold War philosophy of science without actually taking intoaccount the former’s dauntingly voluminous literatures.39

Let us briefly assay just one example. As one would suspect, ‘‘democracy’’ meansmany things to many people, and has been the subject of much dispute downthrough the ages. These disputes could be arrayed along a myriad of axes, but

38 ‘We do better to deploy the notion of aims in its most natural home, referring to the aims of individ-

ual agents rather than to some abstraction’ (2001, p. 87). It seems symptomatic of the strange disengage-

ment from social theory that a philosopher can so blithely pronounce ex cathedra that the ‘‘individual’’

is not an abstraction, or that some homes are innately ‘‘natural’’.39 Recall that Operations Researchers wanted to do social research without actually becoming social

scientists. One of Kitcher’s habits is to present himself as advocate of a Golden Mean between histori-

cists and analytical philosophers, but without attending to the long history of arguments over the

relationship of science to democracy (Hollinger, 1996; Purcell, 1973) or indeed, the extended and vexed

relationship between the history of utilitarian models of science and the philosophy of science. Many

reviewers have sought to situate Kitcher’s project within a Rawlesian discourse of ‘‘justice’’, but this

neglects the extent to which rational choice theory has come to dominate most American political dis-

course; on this, see Amadae (2003).


the one salient to our present concerns is the distinction between freedom as

action upon present desires/preferences (Hobbes, Nozick) and freedom as a devel-opmental notion requiring participation and self-transformation (Mill, Dewey,Toqueville). This argument was framed in a Cold War fashion as privileging the

former over the latter by Isaiah Berlin (1969) as the distinction between negativeand positive liberty. Political philosophers have noted that to opt for neoclassicalmodels of preference optimization generally implies subscribing to the former

framework to the exclusion of the latter (Cunningham, 2002). Kitcher wants toargue that science can contribute to a developmental conception of general demo-cratic welfare, but that this developmental conception does not and should notapply to the actual conduct of scientific research itself: in other words, for Kitcher,

science still exists as a distinct social formation with its own dependable socialdynamic. To repress this troublesome asymmetry, he proposes to model all agentsas neoclassical rational choosers, and to downplay the significance of freedom as a

goal in its own right. In the process, all temporally specific social structures ofresearch disappear, and instead the problem is said to reside within the largersociety and its flawed political structures. The proposed solution to the problem of

science and democracy is to impose all sorts of restrictions on the decision proce-dures (which he misleadingly dubs ‘‘well-ordered science’’) of the citizenry, and noton the scientists. Those who detect echoes of Reichenbach’s original distinctionbetween discovery/justification, his mistrust of history, as well as his crusade

against metaphysics, would not be mistaken.40

But a polity consisting of neoclassical rational agents bears its own conundrums.Initially, Kitcher tries to equate collective social welfare with the aggregation ofindividual preferences under a democratic regime: surely he has some reason todiscount the infamous Arrow impossibility results (1951) cited above which suggest

that the two are incompatible?41 Kitcher also engages in talk about ‘‘freedom’’ and‘‘morality’’, but (like Polanyi, like Hayek, like Kuhn, like Reichenbach) he betraysa fear of what he calls ‘‘vulgar democracy’’, and perhaps religion as well.42 His sol-

ution to this tension is to seek to alter the ‘‘preferences’’ of those individuals whomhe deems untrustworthy, converting (brainwashing?) them into possession of whathe calls ‘‘tutored preferences’’. So much for the shibboleth of freedom of thought—

although this is not such a problem for Kitcher, since he takes a decidedly jaun-diced view of freedom, as evidenced by the quote which heads this section.43 In

40 ‘The modest realism I defend is supposed to strip away metaphysical excrescences . . . to allow issues

about ethical, social and political values to enter, not under the rubric of some confused metaphysical or

epistemological proposal’ (Kitcher, 2002b, p. 571).41 The Arrow proofs suggest voting procedures, such as those advocated by Kitcher, cannot lead to

correct aggregate preferences. For a modern summary of the various voting paradoxes, see Saari (2001).42 ‘Behind the often evangelical rhetoric about the value of knowledge stands a serious theology, an

unexamined faith that pursuit of inquiry will be good for us’ (Kitcher, 2001, p. 166).43 Longino (2002b), p. 564, also comments upon this. However, few realize that ‘‘freedom’’ no longer is

taken to characterize the quintessence of the scientist under the modern globalized commercialization

regime.


case the carefully tutored citizens still do not qualify as sufficiently pliable for thescientists’ plans, Kitcher then proceeds to invent an elaborate Shangri-La, for-tuitously fully equipped with an agora of ideal forms of deliberation concerningprospective scientific research programs, voting on which to enjoy support (withwhat nature of franchise?) and a Supreme Court of experts to sort things out whenthey get a little testy. Curiously in one normally extolled as a clear writer, Kitcherthen writes as though piling on neoclassical economic jargon renders the imaginarystate more plausible:

[W]ith respect to each budgetary level, one identifies the set of possible distribu-tions of resources among scientific projects compatible with the moral constraintson which the ideal deliberators agree, and picks from this set the option yieldingmaximum expected utility, where the utilities are generated from the collectivewish list and the probabilities obtained from the experts. (Kitcher, 2001, p. 121)

If this sounds akin to style of analysis found at RAND in the 1960s concerningthe optimal choice of weapons systems by the Pentagon, then one would not be ledtoo far astray concerning its intellectual lineage. Indeed, decision theoretic termsapplied to technoscientific options was one of the prime specialties of OperationsResearch in that era (Hounshell, 2000) and designedly so: it draped a patina ofpolitical accountability over a process which was clearly spinning out of controland rife with conflicting interests, Machiavellian manipulation and backroom poli-tics. But more to the point, it continues the tradition of seeking political inspi-ration for the philosophy of science in the same anti-democratic doctrines whichwe outlined in the previous section. The RAND ‘‘model’’ was never intended tointegrate the citizenry into the process of research; it merely served to reinforce theunbridgeable gap which safely quarantines them off from it. Kitcher’s predicamentis that his advocacy of an economic model of the agent with its disdain for historyrenders him unable to realize it is equally true of his own work. His decision theor-etic exercises evade all the usual empirical criticisms of rational choice models: ‘themodels Goldman and I develop aren’t intended as literal descriptions of people’(Kitcher, 2002a, p. 559). Kitcher, for all his vaunted ambition to be au courant,innovating new raison d’etre and fresh topics for the philosophy of science, ends upbeing a commercialized and game-theoretic Reichenbach.44 If there is any substan-tial difference, it is that we have traded the rhetoric of timeless truth for that of aquest after value-added information. As he writes, ‘I have attempted a systematicsurvey of all the possibilities for showing that ‘‘truth is better than profit’’ andhave come up empty’ (Kitcher, 2001, p. 166).The tragedy of Kitcher’s version of social epistemology is that it ends up being

both irrelevant and tautologous. It is irrelevant in that his ideal world of‘‘well-ordered science’’ has nothing whatsoever to do with the present predicaments

44 ‘The fundamentals of human reasoning are pretty much everywhere the same . . . So the ways in

which arguments are justified (and discovered) in the human sciences should be no different than those

employed in the natural sciences’ (Kitcher, 2001, p. 175).


of science—the runaway revisions of intellectual property, frequent resort to liti-gation to solve scientific problems, the crisis of scientific publication, the disen-franchisement of university faculty from academic governance, the bankruptcy ofritual mechanized induction, the over-emphasis on arcane mathematical expertiseto validate relevance, the rationalization of teaching and lab work into tempMcJobs, numerous popular crises of confidence in scientists as political actors—allthis and more might as well be happening in another galaxy far, far away fromKitcher’s ideal republic. Helen Longino (2002b, p. 567) recently launched this criti-cism against Kitcher; his response was most disconcerting: ‘If one is firmly commit-ted to well-ordered science as an ideal . . . then the privatization of research lookslike a trend that we should want to resist strenuously . . . I take it as a strength ofthe ideal of well-ordered science that it treats the pitching of scientific research intothe marketplace as a disaster’ (Kitcher, 2002b, p. 570). Not only is it unlikely thatany such prognosis be drawn from Kitcher’s analysis, as I have argued above, butit is disturbing that he is seemingly unaware that his own defense of the HumanGenome Project (a public project in effect captured by Celera and the pharmaceu-tical companies) and reification of science as a marketplace of ideas has playeddirectly into the hands of the globalized privateers. Indeed, spinning stirring fairy-tales about democratic plebiscites over government controlled and funded researchprograms while ignoring the dominant corporate control of the vast bulk of scien-tific research and intellectual property would qualify as a perfect example of whatKarl Mannheim used to call ‘‘ideological discourse’’.Doubly vexed, Kitcher’s social epistemology is also tautologous because of its

circular notion of what it means to be a ‘‘tutored’’ legitimate agent participating indeliberations about science. If ‘‘having a legitimate voice’’ means getting a gold-plated Ph.D. in the relevant discipline, then we are back to Polanyi’s and Kuhn’soriginal prescription that only holders of valid union cards (and then, only thosewho publicly submit to the authority of the master) could get to cast their vote forwhether we get to enjoy the Human Genome Project or the Texas Supercollider ora fortified Social Security system. Of course, by then your ‘‘tutored’’ vote is moreor less a foregone conclusion. If, per contra, getting ‘‘tutored’’ only means youneed to know something about the sciences in question, then the prescription isunavailing, since there can be no settled prerequisite manifest of everything oneneeds to know in order to participate. If Kitcher had consulted the philosophicalliterature on utilitarianism, he would have discovered that this idea of ‘‘cleansed’’or suitably informed preferences is the Achilles heel of left-liberal neoclassicaleconomics (Cowen, 1993). The fundamental commitment of neoclassical models isto treat ‘‘preferences’’ as the independent unmoved movers of changes in marketbundles. The mid-20th-century revision of the economic agent towards greaterresemblance to an information processor (Mirowski, 2002) raised the disturbingspecter that if agents took action to change their own preferences through experi-ence, learning, and interpersonal interaction, then the vaunted determinacy ofeconomic equilibrium would evaporate. A whole sequence of attempts to stem thetide, from Nash equilibrium to schemes of mechanical induction to so-called‘‘rational expectations’’, have sought to endow the economic agent with some


semblance of invariance, but Kitcher’s ideal Republic of Tutees doesn’t even beginto address the problem. Not only do we lack any guarantee that ex ante intentionsneed not map into ex post evaluations, but the very notion of ‘‘democracy’’ as giv-ing the people what they want loses all cogency. As Cowen writes, ‘Fully informedpreferences do not offer an Archimedean point for value theory in a world ofimperfect information’ (1993, p. 266).Dewey endorsed a vision of social theory that was conceptually aligned with his

political objectives, even though he faced implacable opposition from the corporatemasters who funded and organized American science. Reichenbach adopted asocial theory that underwrote some of his philosophical objectives, and resonatedwith the plans of the military funders of science, in part because they employedhim to help produce it. Kitcher has sought to deploy a version of social theory thatneither resonates with his own avowed political objectives nor fully complies withthe current plans of the corporate paymasters of globalized science. One expectsthe next philosophical protagonist in this sequence will concoct some ‘evolutionary’arguments that market incentives naturally and most efficiently result in all thefavored epistemic virtues, combined with the benefit that they more fully integratescience into the preferences of the general citizenry. Perhaps she will even occupyan endowed chair in a private think-tank.

6. Conclusion

What do philosophers of science know about the ‘‘scientific dimensions of socialknowledge’’? While one can only commend their ambition to ‘‘do something forsociety’’, the answer must still be nowhere near enough. There was a time whenphilosophers of science would light upon some specific recent theory in physics andinflate it up to represent some grand generalization about the nature of epistem-ology. The temptation seems to be currently to do something similar with recenteconomics, proclaiming a new and fruitful alliance with the social sciences. Theonly antidote to this short-sighted practice is historical appreciation of theconvoluted ways in which philosophy of science, social theory, and the socialstructures of science have always and everywhere been intercalated and implicatedin each other’s orthodoxies and, further, the way in which the dominant structuresof the funding and organization of scientific research in a particular culture selectthose doctrines which give the appearance of rationalizing their practices. Thisphenomenon might itself be a fruitful topic for a unified philosophy of a social/natural science; but there seems to be little enthusiasm for it in our globalizedprivatized regime.

Acknowledgements

I would like to thank the participants at the Bielefeld–Notre Dame Conferenceon Science and Values, July 2003, and attendees at the CEPA seminar at the New


School, September 2003, for their searching questions. I would especially like tothank James Robert Brown, Roger Strand, S. M. Amadae, and Tom Uebel forcomments. This paper is dedicated to the memory of Paul Feyerabend.

References

Agassi, J. (1995). Contemporary philosophy of science as thinly masked antidemocratic apologetics. In

K. Gavroglu, J. Stachel, & M. W. Wartofsky (Eds.), Physics, philosophy and the scientific community:

Essays in the philosophy and history of the natural sciences and mathematics: In honor of Robert S.

Cohen (pp. 153–170). Dordrecht: Kluwer.

Amadae, S. M. (2003). Rationalizing capitalist democracy: The Cold War origins of rational choice liber-

alism. Chicago: University of Chicago Press.

Arrow, K. (1951). Social choice and individual values. New York: Wiley.

Asner, G. (2002). Corporate fad or government policy? The linear model, the Department of Defense,

and the golden age of industrial research. Paper presented to the Nobel Symposium, November,

Stockholm, Sweden.

Ayres, C. E. (1927). Science: The false messiah. Indianapolis: Bobbs-Merrill.

Berlin, I. (1969). Four essays on liberty. New York: Oxford University Press.

Biagioli, M., & Galison, P. (Eds.). (2003). Scientific authorship: Credit and intellectual property in science.

New York: Routledge.

Bok, D. (2003). Universities in the marketplace. Princeton: Princeton University Press.

Brown, J. R. (2001). Who rules in science? Cambridge, MA: Harvard University Press.

Cahn, S. (Ed.). (1977). New studies in the philosophy of John Dewey. Hanover, NH: University Presses of

New England.

Callebaut, W. (1993). Taking the naturalistic turn, or, how real philosophy of science is done: Conversa-

tions with William Bechtel . . . [et al.]. Chicago: University of Chicago Press.Camic, C. (1986). The matter of habit. American Journal of Sociology, 91, 1039–1087.

Cartwright, N., Cat, J., Fleck, L., & Uebel, T. (1996). Otto Neurath: Philosophy between science and poli-

tics. Cambridge: Cambridge University Press.

Cohen, R. (1971). Homage to Rudolf Carnap. In R. C. Buck, & R. S. Cohen (Eds.), PSA 1970: In

memory of Rudolf Carnap (pp. XLII–XLIII). Boston Studies in the Philosophy of Science, 8.

Dordrecht: Reidel.

Collins, M. (1998). Planning for modern war: RAND and the Air Force, 1945–50. Ph.D. thesis, University

of Maryland.

Cowen, T. (1993). The scope and limits of preference sovereignty. Economics and Philosophy, 9, 253–269.

Cunningham, F. (2002). Theories of democracy. London: Routledge.

Dalton, T. (2002). Becoming John Dewey. Bloomington: Indiana University Press.

Davidson, D. (1999). Intellectual autobiography of Donald Davidson. In L. Hahn, (Ed.), The philosophy

of Donald Davidson (pp. 1–70). Chicago: Open Court.

Dennis, M. (1997). Historiography of sciene: An American perspective. In J. Krige, & D. Pestre (Eds.),

Science in the twentieth century (pp. 1–26). Amsterdam: Harwood.

Dewey, J. (1927). The public and its problems. New York: Holt.

Dewey, J. (1939). Freedom and culture. New York: Putnam.

Dewey, J. (1981). The later works, 1925–1953 (17 vols.) (Vol. 1, 1924: Experience and nature). Carbon-

dale: Southern Illinois University Press.

Dewey, J. (1984). The later works, 1925–1953 (17 vols.) (Vol. 5, 1929–1930). Carbondale: Southern Illi-

nois University Press.

Dewey, J. (1985). The later works, 1925–1953 (17 vols.) (Vol. 12, 1938). Carbondale: Southern Illinois

University Press.

Dreyfus, R., Zimmerman, D., & First, H. (Eds.). (2001). Expanding the boundaries of intellectual

property. New York: Oxford University Press.


Eisenberg, R. (2001). Bargaining over the transfer of proprietary research tools. In R. Dreyfus,

D. Zimmerman, & H. First (Eds.), Expanding the boundaries of intellectual property (pp. 223–250).

New York: Oxford University Press.

Feigl, H. (1969). The origin and spirit of logical positivism. In P. Achinstein, & S. F. Baker (Eds.), The

legacy of logical positivism (pp. 3–24). Baltimore: Johns Hopkins Press.

Feyerabend, P. (1978). Science in a free society. London: New Left Books.

Feyerabend, P. (1987). Farewell to reason. London: Verso.

Feyerabend, P. (1999). Philosophical papers (3 vols.) (Vol. 3, Knowledge, science and relativism). New York:

Cambridge University Press.

Fine, T. (1973). Theories of probability. New York: Academic Press.

Fuller, S. (1988). Social epistemology. Bloomington: Indiana University Press.

Fuller, S. (2000). Thomas Kuhn: A philosophical history for our time. Chicago: University of Chicago

Press.

Fuller, S. (2002a). Knowledge management foundations. Oxford: Butterworth–Heinemann.

Fuller, S. (2002b). Prolegomena to a sociology of philosophy in the 20th century English-speaking

world. Philosophy of the Social Sciences, 32, 51–177.

Giere, R. N. (1996). From Wissenschaftliche Philosophie to philosophy of science. In R. N. Giere, &

A. W. Richardson (Eds.), Origins of logical empiricism (pp. 335–354). Minnesota Studies in the

Philosophy of Science, XVI. Minneapolis: University of Minnesota Press.

Goldstein, W., & Hogarth, R. (Eds.). (1997). Research on judgment and decision making. New York:


Gross, N. (2003). Richard Rorty’s pragmatism: A case study in the sociology of ideas. Theory and

Society, 32, 93–148.

Hardcastle, G., & Richardson, A. (Eds.). (2003). Logical empiricism in North America. Minneapolis:

Univeristy of Minnesota Press.

Hollinger, D. (1996). Science, Jews and culture. Princeton: Princeton University Press.

Hounshell, D. (1997). The Cold War, RAND and the generalization of knowledge. Historical Studies in

the Physical and Biological Sciences, 27, 237–267.

Hounshell, D. (2000). The medium is the message, or how context matters: The RAND Corporation

builds an economics of innovation, 1946–1962. In A. C. Hughes, & T. P. Hughes (Eds.), Systems,

experts and computers: The systems approach in management and engineering, World War II and after

(pp. 255–310). Cambridge, MA: MIT Press.

Howard, D. (2003a). Marxist themes in early logical positivism. Presentation to the Bielefeld–Notre

Dame Conference on Science and Values.

Howard, D. (2003b). Two left turns make a right: On the curious political career of North American

philosophy of science at mid-century. In G. Hardcastle, & A. Richardson (Eds.), Logical empiricism

in North America. Minneapolis: University of Minnesota Press.

Hughes, A. C., & Hughes, T. P. (Eds.). (2000). Systems, experts and computers: The systems approach in

management and engineering, World War II and after. Cambridge: MIT Press.

Jacobs, S. (2002). The genesis of ‘scientific community’. Social Epistemology, 16, 157–168.

Jardini, D. (1996). Out of the blue yonder. Ph.D. thesis, Carnegie–Mellon University.

Jarvie, I. C. (2001a). Science in a democratic republic. Philosophy of Science, 68, 545–564.

Jarvie, I. C. (2001b). The republic of science. Amsterdam: Rodopi.

Kay, L. (1997). Rethinking institutions: Philanthropy as a problem of knowledge and power. Minerva,

35, 283–293.

Kevles, D. (1995). The physicists. Cambridge, MA: Harvard University Press.

Kirby, M. (2000). Operations research trajectories: Anglo-American experience, 1940–1990. Operations

Research, 48, 661–670.

Kitcher, P. (1993). The advancement of science. New York: Oxford University Press.

Kitcher, P. (1996). The lives to come. New York: Simon & Schuster.

Kitcher, P. (2000). Reviving the sociology of science. Philosophy of Science, 67(3) (Suppl., PSA 98, Pt.

II, Symposia papers), S33–S44.

Kitcher, P. (2001). Science, truth and democracy. New York: Oxford University Press.


Kitcher, P. (2002a). The third way: Reflections on Helen Longino’s The fate of knowledge. Philosophy

of Science, 69, 549–559.

Kitcher, P. (2002b). Reply to Helen Longino. Philosophy of Science, 69, 569–572.

Koertge, N. (Ed.). (1998). House built on sand. New York: Oxford University Press.

Kohler, R. (1991). Partners in science. Chicago: University of Chicago Press.

Kragh, H. (1999). Quantum generations. Princeton: Princeton University Press.

Longino, H. (2002a). The fate of knowledge. Princeton: Princeton University Press.

Longino, H. (2002b). Science and the common good: Thoughts on Philip Kitcher. Philosophy of Science,

69, 560–568.

Longino, H. (2002c). The social dimensions of scientific knowledge. In E. N. Zalta (Ed.), The Stanford

encyclopedia of philosophy (Summer 2002 ed.). http://plato.stanford.edu/archives/sum2002/entries/

scientific-knowledge-social/

McCumber, J. (2001). Time in the ditch: American philosophy and the McCarthy era. Evanston: North-

western University Press.

McGrath, P. (2002). Scientists, business and the state, 1890–1960. Chapel Hill: University of North

Carolina Press.

McGuckin, W. (1984). Scientists, society and the state. Columbus: Ohio State University Press.

Mirowski, P. (1988). Against mechanism. Totawa: Rowman & Littlefield.

Mirowski, P. (1998). On playing the economics trump card in the philosophy of science: Why it didn’t

work for Michael Polanyi. Philosophy of Science, 64(4) (Suppl., PSA 97, Pt. II, Symposia papers),

S127–S138.

Mirowski, P. (2002). Machine dreams: Economics becomes a cyborg science. New York: Cambridge

University Press.

Mirowski, P. (2003). The commercialization of science and the future of the university in the 21st cen-

tury. Lecture. Multimedia presentation. www.nd.edu/~pmirowsk/

Mirowski, P. (2004). Effortless economy of science?. Durham: Duke University Press.

Mirowski, P., & Sent, E. -M. (Eds.). (2002). Science bought and sold: Essays in the economics of science.

Chicago: University of Chicago Press.

Mirowski, P., & Van Horn, R. (forthcoming). The contract research organization and the commerciali-

zation of scientific research. Social Studies of Science.

Mitchell, W. C. (1969). Types of economic theory (2 vols.) (Vol. 2). New York: Kelley. (First published

1949)

Noll, R. (Ed.). (1998). Challenges to research universities. Washington: Brookings.

Notturno, M. (1999). Popper’s critique of scientific socialism. Philosophy of the Social Sciences, 29(1),

32–61.

Nowotny, H., Scott, P., & Gibbons, M. (2001). Re-thinking science. Cambridge: Polity.

Porter, T. (2003). Review of P. Mirowski & E.-M. Sent (Eds.), Science bought and sold: Essays in the

economics of science. Chicago: University of Chicago Press, 2002. British Journal for the History of

Science, 36, 381–382.

Preston, J., Munevar, G., & Lamb, D. (Eds.). (2000). The worst enemy of science? New York: Oxford

University Press.

Purcell, E. (1973). The crisis of democratic theory: Scientific naturalism and the problem of value.

Lexington: University of Kentucky Press.

Quine, W. V. O. (1985). Time of my life. Cambridge, MA: MIT Press.

Ravetz, J. (1971). Scientific knowledge and its social problems. Oxford: Oxford University Press.

Reichenbach, H. (1938). Experience and prediction. Chicago: University of Chicago Press.

Reichenbach, H. (1949). Rational reconstruction of the decision for war. Santa Monica: RAND D-539.

Reichenbach, H. (1953). The rise of scientific philosophy. Berkeley: University of California Press.

Reichenbach, H. (1978). Selected writings (2 vols.) (M. Reichenbach, & R. S. Cohen, Eds.). Dor-

drecht Reidel.

Reichenbach, H. (1989). Dewey’s theory of science. In P. Schlipp, & L. Hahn (Eds.), The philosophy of

John Dewey (pp. 157–192). Chicago: Open Court.

Reingold, N. (1991). Science American style. New Brunswick: Rutgers University Press.


Reisch, G. A. (2002). Review of J. McCumber, Time in the ditch: American philosophy and the McCarthy

era. Evanston: Northwestern University Press, 2001. Philosophy of Science, 69, 389–392.

Rescher, N. (1997a). Instructive journey. Lanham: University Presses of America.

Rescher, N. (1997b). H2O: Hempel–Helmer–Oppenheim. Philosophy of Science, 64, 334–360.

Richardson, A. (2000). Science as will and representation: Carnap, Reichenbach, and the sociology of

science. Philosophy of Science, 67(3) (Suppl., PSA 98, Pt. II, Symposia papers), S151–S162.

Richardson, A. (2002). Engineering philosophy of science: American pragmatism and logical empiricism

in the 1930s. Philosophy of Science, 69(3) (Suppl., PSA 2000, Pt. II, Symposia papers), S36–S47.

Rouse, J. (1987). Knowledge and power: Toward a political philosophy of science. Ithaca: Cornell Univer-

sity Press.

Rutherford, M. (1994). Institutions in economics. New York: Cambridge University Press.

Ryan, M. (1998). Knowledge diplomacy. Washington: Brookings Institution.

Saari, D. (2001). Decisions and elections. New York: Cambridge University Press.

Scheiding, T. (forthcoming). Publish and perish: Electronic publication and the serials crisis. Ph.D. thesis,

University of Notre Dame.

Schmidt, F. (Ed.). (1994). Socializing epistemology. Lanham: Rowman & Littlefield.

Sell, S. (2003). Private power, public law: The globalization of intellectual property rights. New York:


Shinn, T. (2003). The industry, research, and education nexus. In M. J. Nye (Ed.), The Cambridge his-

tory of science (Vol. 5, The modern physical and mathematical sciences) (pp. 133–153). New York:


Tobey, R. (1971). The American ideology of national science, 1919–30. Pittsburgh: University of

Pittsburgh Press.

Turner, S. (2003). The third science war. Social Studies of Science, 33, 581–611.

Uebel, T. E. (2000a). Some scientism, some historicism, some critics: Hayek’s and Popper’s critiques

revisited. In M. Stone, & J. Wolff (Eds.), The proper ambition of science (pp. 151–173). London:

Routledge.

Uebel, T. E. (2000b). Logical empiricism and the sociology of knowledge: The case of Neurath and

Frank. Philosophy of Science, 67(3) (Suppl., PSA 98, Pt. II, Symposia papers), S139–S150.

Wang, J. (1999). Merton’s shadow: Perspectives on science and democracy since 1940. Historical Studies

in the Physical Sciences, 30(1), 279–306.

Westbrook, R. (1991). John Dewey and American democracy. Ithaca: Cornell University Press.

White, M. (1989). The politics of epistemology. Ethics, 100, 77–92.

Wray, K. Brad (2001). Science, biases, and the threat of global pessimism. Philosophy of Science, 68(3)

(Suppl., PSA 2000, Pt. I, Contributed papers), S467–S478.

Yonay, Y. (1998). The struggle over the soul of economics. Princeton: Princeton University Press.

the scientific dimensions of social knowledge and their distant echoes in 20th-century american...

Documents