Reply INFORMAL LOGIC XVI.l, Winter 1994 Sizing Up Science: A Reply to Fuller MIRIAM SOLOMON Temple University 1. Introduction Fuller claims that epistemic properties of scientific activity are size-sensitive. Size is measured in terms of numbers of scien- tists and wealth of their material resources (e.g. funding, instrumentation, access to information). He states that when science gets much bigger, as it has this century, the epistemology of science changes: the "classical search for knowledge" ends (p. 8) and we move into a "post-epistemic so- ciety" (p. 10). Fuller claims that such "Big Science" (p. 4) is epistemologically at risk, and may fail to be science altogether. This provocative suggestion is worth consider- ing, for the reasons he gives and for other reasons also. Naturalistic considerations support the claim that human strategies for investigat- ing the world may change as social and material conditions change. Scientific methods are not a priori. Different meth- ods and policies will prove successful in different social and psychological situa- tions, and for different problem domains. Availability of instruments and other ex- perimental resources should also influence choices of method and direction of research. I shall agree with Fuller that size be- longs as a variable in epistemological ac- counts of science. I offer reasons to doubt that it is as pivotal a variable as he claims. I also respond to Fuller's suggestions for preventing Big Science from becoming the end of all science with some suggestions of my own. 2. Epistemic Maladies and Epistemic Myths Much scientific work is large in scale and thus, in the present social conditions, dependent on support from the govern- ment, military, and industry. Grant applica- tions are tailored to suit the ends of these institutions as well as, or instead of, so- called "internal" goals of science. Further- more, most scientific work is fully accessible only to a relatively small group of scientists in a particular subspecialty; other scientists, as well as the general pub- lic, are regarded as "unqualified" to make evaluations. Scientific work was not always this way. There was a time when many scien- tists with modest fortunes, generous friends or university affiliation could sup- port their own experimental research (Harvey, Galileo, Newton, Priestley, Faraday, Darwin, Lyell) and relatively little technical expertise and background knowledge was needed to understand sci- entific controversies (e.g. the path of blood flow in the body, the Scientific Revolution, the Chemical Revolution, debates over the evolution of species and changes in the earth's crust). Fuller writes that the increased need for major external funding, and the frag- mentation of the scientific community into units with particular specialized knowl- edge are new maladies of science that are likely to inhibit scientific progress. He is joined by many working scientists weary 56 Miriam Solomon of the constraints and pressures of grant administration or struggling with the de- mands of contributing and gaining recog- nition in a particular field. It is tempting to look back to a time when there was no Big Science, and think that things used to be epistemologically better. This is a myth, akin to many myths that weary commuters and consumers hold about preindustrialized societies. There never was an "epistemic world" where scientific work was free from inter- ests and biases that can be disruptive to the scientific enterprise. When scientists needed support for experimental work or simply living expenses, they had to negotiate the whims of royalty, aristocracy and university hierarchy. Their ideas were received and criticized in communities with entrenched ideologies. Personal factors such as pride, competitive spirit and responsiveness to peer pressure generally influence scientific decision making. Cognitive biases such as confirmation bias, overvaluation of salient and available information and belief perse- verance also have pervasive effects on scientific thinking. Much new work in so- ciology and psychology of science docu- ments the role of interests and biases in historical case studies of scientific change'. Inasmuch as human beings have done sci- ence in a social world, there cannot have been the "classical search for knowledge" (p. 8) that Fuller retrodicts. Judging from his other writings, Fuller is quite aware of this work in history, soci- ology and psychology of science. It must be inferred that he has a special concern for the epistemological character of Big Science, that does not simply register the influence of "external" interests and the limited role of criticism. One possibility is that Fuller is claiming that Big Science is more influenced by biases and interests than Little Science, and that the biases and interests of Big Science exceed some threshold of epistemic toleration. If this is his view, it lacks evidence. It may be that increased specialization within science has raised the qualifications for participation in criticism, and thereby excludes many scientists and most of the public from sci- entific debate. It is also the case, however, that there are many more scientists, and perhaps as many are involved in each par- ticular dispute as in the past (although not proportionately as many of the scientific community). Within subspecialties, scien- tists can get both criticism and recognition for their work. Peer review for journals and academic conferences for exchange of in- formation and results are relatively new fo- rums for criticism associated with the rise of Big Science. Standards of professional ethics-both academic freedom and aca- demic honesty-are now part of the insti- tution of science. For every epistemic step back that Big Science makes, it appears that there is also a (different) step forward. There is no reason to think-without more evidence-that Big Science is any less likely to "represent the structure of reality" (p. 6)2 than Little Science. 3. Public Accountability In my view, Fuller's central concern is not the degree to which Big Science is biased and interested but the particular moral and political values it instantiates. When Fuller deplores the lack of "account- ability" of Big Science he is especially concerned about the lack of public ac- countability and less concerned about the lack of criticism across the scientifiC com- munity. Big Science fails to be accountable to the public because its interests are de- veloped around the hands that feed it- government, the military, industry-rather than around the hands that need it for health care, technology and the environ- ment, and are morally entitled to it. It also fails to be accountable to the public be- cause increased specialization has brought with it an attitude of professional elitism that eschews criticism from outsiders, who are regarded as unqualified to judge. Fuller makes specific proposals for im- proving the public accountability of sci- ence. The incentive structure of science could be changed by having scientists wager with one another, in a public compe- tition, about the results of a proposed plan of research. This would cut some of the strings to large funding institutions. The public could be involved in scientific dis- putes, through media coverage of staged debates between scientists. Just as audi- ences for sporting events engage in criti- cism of particular performances, they could make judgements of scientific merit. Note that these are proposals for im- proving public accountability of science, and not obviously proposals for improving science. Fuller conflates the two, deliber- ately. It is his view that the moral and the epistemic cannot be separated (p. 7). Even if this view is true, however" these morals and the various goals of the scientific enterprise can be separated. Groups do not always correct for bias- es made by individuals; indeed, in some cases, they amplify bias.4 Furthermore, groups of non-experts are criticizing from a position of relative ignorance. Their gen- eral mechanisms for thinking may be, as Fuller says, the same as scientists' (p. 2), but it does not follow that their reasoning will be as good, because their background assumptions are less reliable. Scientific work mayor may not improve as a result of replacing the interests of large politicized institutions with the interests of individual humans; it is an empirical ques- tion, and one whose answer is relative to particular social situations. The marine re- search crucial to the development of plate tectonic theory was dependent on the re- sources ofthe U.S. Navy. The future of the supercollider project is dependent on the support of the NSF. It is difficult to see how either of these projects could be organized as a village industry around the human in- terests of most people. Other worthwhile scientific projects, however-perhaps less costly projects with more direct technolog- Sizing Up Science: A Reply to Fuller 57 ical applications-might do better when governed by public interests. Fuller claims that scientists can always devise and then choose Little Science projects to answer scientific questions (p. 6). No doubt this is true in some cases and should be encour- aged for reasons of economy, but it would be surprising if it were true in general. It is just as much a choice to eschew large projects as it is to embrace them. Re- stricting the community to Little Science projects is no more desirable for the scien- tific enterprise than over-emphasizing Big Science. One possibility for reorganization is to centralize resources (e.g. through tax- ation) without centralizing power over dis- tribution of these resources. Distribution could be decided upon by elected experts, or simply randomized. These suggestions (as Fuller's suggestions) are worth investi- gating for feasibility and results; the im- pact on the scientific enterprise cannot be known in advance. Fuller's view that scientific disputes can be, and should be, made accessible to non-experts who can then influence out- comes is a refreshing and welcome chal- lenge to the authority of scientists. Again, the feasibility and results of doing this can- not be known in advance or in general. As mentioned above, the worth of non- expert criticism of science is not known. Spectator sports are hardly without prob- lems caused by media coverage and public involvement; they are hardly the role mod- els for science that Fuller suggests. A sug- gestion that Fuller does not consider in this paper is the contribution that expertise from humanistic fields can make to the sci- entific enterprise. For example, if scientific education included history, politics, ethics, philosophy and psychology of science (perhaps as a matter of public demand rather than governmental interference) scientific work might be more creative, tolerant of differences, critical of both trendiness and conservativeness and careful with issues of academic honesty. Most sci- entists have archaic views about scientific 58 Miriam Solomon progress and scientific method, and it would be interesting to see whether and how their practices might change with edu- cation. Even this proposal needs to be test- ed in social context, and cannot be decided a priori. 4. Conclusions Size is just one of the many variables that describe the scientific enterprise. Its epistemic import depends on the back- ground social and political context, the internal organization of science into subgroups and the particular scientific problem under investigation. Big Science seems no less in danger of failing than Little Science; they court epistemically equivalent although different dangers. The problems of each need to be separately investigated and addressed through assessments of the relative effectiveness of various policies and kinds of reasoning. Fuller's criticisms of Big Science tum out to be more moral than epistemic. He objects to the control that large politicized institutions-especially the military- have on the distribution of scientific re- sources and thereby on the direction of sci- entific progress. I have argued that the alternatives to this that Fuller presents- and I supplement-are worth investigat- ing, but there is no telling a priori what the impact on science will be. Only morals can be settled in advance. Notes I See, for example, the work of Kuhn and more recent work in the Kuhnian tradition such as Harry Collins, Changing Order: Replication and Induction in Scientific Practice (Chicago: University of Chicago Press 1992 [1985]); Bruno Latour and Steve Wool gar, Laboratory Life: The Construction of Scientific Facts (Princeton, NJ: Princeton University Press 1986 [1979]); Steven Shapin and Simon Schaffer, Leviathan and the Air-Pump: Hobbes, Boyle and the Experimental Life (Princeton, NJ: Princeton University Press, 1985). The bibliography to Fuller's paper contains further examples. 2 This is Fuller's stated view of the goal of science, not mine. The same epistemological points can be made for other views of the goals of science, including my own view which is that empirical success (broadly construed) is the goal of scientific work. 3 Robert McCauley, "Epistemology in an Age of Cognitive Science," Philosophical Psychology 1:2 (I988) pp. 143-151 argues thatthe goals of science must be articulated in a general discus- sion of values. Although I do not agree with this view, it is widely held, and I grant it for the purposes of discussion here. 4 See, for example, G. Hill, "Group vs. Individ- ual Performance: Are N+I heads better than One?" Psychological Bulletin 91 (1982) pp. 517-539. MIRIAM SOLOMON DEPARTMENT OF PHILOSOPHY TEMPLE UNIVERSITY PHILADELpHIA, R-1 19122