INFO KAMAR OPERASI

Engagement for progress: applied philosophy of science in context
Author(s): Heather Douglas
Source: Synthese, Vol. 177, No. 3, MAKING PHILOSOPHY OF SCIENCE MORE SOCIALLYRELEVANT (December 2010), pp. 317-335
Published by: Springer
Stable URL: http://www.jstor.org/stable/40985707
Accessed: 27-01-2017 04:55 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusteddigital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information aboutJSTOR, please contact support@jstor.org. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available athttp://about.jstor.org/terms
Springer is collaborating with JSTOR to digitize, preserve and extend access to Synthese
This content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTC
All use subject to http://about.jstor.org/terms
Synthese (2010) 177:317-335 DOI 10.1007/s 11229-010-9787-2 Engagement for progress: applied philosophy of science in context Heather Douglas Received: 23 March 2009 / Accepted: 23 July 2010 / Published online: 30 September 2010 © Springer Science+Business Media B.V. 2010 Abstract Philosophy of science was once a much more socially engaged endeavor, and can be so again. After a look back at philosophy of science in the 1930s- 1950s, I turn to discuss the current potential for returning to a more engaged philosophy of science. Although philosophers of science have much to offer scientists and the pub- lic, I am skeptical that much can be gained by philosophers importing off-the-shelf discussions from philosophy of science to science and society. Such efforts will likely look like efforts to do applied ethics by merely applying ethical theories to particular contexts and problems. While some insight can be gained by these kinds of endeavors, the most interesting and pressing problems for the actual practitioners and users of science are rarely addressed. Instead, I recommend that philosophers of science engage seriously and regularly with scientists and/or the users of science in order to gain an understanding of the conceptual issues on the ground. From such engagement, flaws in the traditional philosophical frameworks, and how such flaws can be remedied, become apparent. Serious engagement with the contexts of science thus provides the most fruit for philosophy of science per se and for the practitioners whom the philos- ophers aim to assist. And if one focuses on contexts where science has its most social relevance, these efforts can help to provide the thing that philosophy of science now lacks: a full-bodied philosophy of science in society. Keywords Socially relevant philosophy of science • Values in science • Explanation • Prediction • Weight of evidence H. Douglas (El) Department of Philosophy, University of Tennessee, 801 McClung Tower, Knoxville, TN 37996-0480, USA e-mail: hdouglas@utk.edu o SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
3 1 8 Synthese (20 1 0) 1 77:3 1 7-335 1 Introduction In this paper, I will lay out a general approach for doing socially relevant philosophy of science derived from my own experience in attempting to do such work. I have found it to be a very productive approach, not only for providing potentially useful insights to science practitioners, but also for thinking about and challenging standard positions within philosophy of science. Contrary to typical views of applied work, doing this kind of work has not involved a straightforward application of ready-made philoso- phy of science to a socially relevant context. Standard philosophical positions have not proven adequate for the contexts of use in which I am interested. Instead, it has involved understanding the conceptual terrain in a socially relevant context ("on the ground," as it were) and thinking through the norms involved, employing distinctly philosophical techniques of conceptual analysis. Because the normative considerations that arise from the context require refinements or revisions of philosophical views, the concep- tual work in the context of interest generates the particularly productive character of this approach. Two caveats are needed before I begin. The first is about what I mean by socially relevant work, and the second about what my arguments imply for the trajectory of philosophy of science. By socially relevant work, I mean to focus attention on those aspects of science that are of particular relevance to society. I am, of course, not the first to do this. I will note below some of the growing body of work with this character. Doing socially relevant work can be distinct from taking a political approach to philosophy of sci- ence. In the approach I describe here, one need not have a particular political agenda driving socially relevant work. The only political commitment underlying the work I describe below is to general democratic principles, a minimal stance. Additional particular political commitments have driven some philosophers of science, such as feminist concerns in feminist philosophy of science or socialism in the work of Otto Neurath. So political concerns can drive socially relevant philosophical work, but it need not. Social relevancy can occur through looking at issues that arise in socially relevant science, and thus that may cut across particular political agendas. Because of the importance of public discourse and accountability in democracies, a commitment to democracy generates an impulse for as much clarity and transparency as one can muster, an impulse in line with most philosophers' instincts. But such clarity can serve a range of political commitments when at work on particular cases, intentionally or not. Thus, the socially relevant work I wish to promote need not be politically charged or directed. And while it may have interesting political implications, it need not be done with expressly political motivations. In describing how socially relevant philosophy of science can be done and how it can be both productive for the field of philosophy of science and helpful to the practitioners and users of science, I do not mean to suggest that this is the only type of philosophy of science one should pursue. Such a view would impoverish the field, reducing the kind of disciplinary conversation that provides coherency. However, room should be made for socially relevant work, without prejudice or penalty. And as issues arise that fall outside the traditional boundaries of philosophy of science (which as we shall see is not a very long tradition), such work should challenge the philosophy of science to 4y SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
Synthese (2010) 177:317-335 319 construe itself more broadly. Any serious considered part of philosophy of science, moral, or social aspects of science. And of the field should be taken seriously make problematic standard positions Before I begin to lay out and illustrate 1 have found useful, I will take a look in its self-conception. Through historical necessary about the current construal 2 Philosophy of science: a short disciplinary Since at least the 1960s, the disciplinary topics concerning the epistemic and have most gripped the field, debates entific change, laws of nature, the relationship tific models, the unity or disunity of claims made in science, the structure support for those claims. In other words, branch of metaphysics and epistemology, those who work on specific scientific questions particular to that science. concerns largely for their own sake, or not scientists, or even more distantly in these issues has been beside the practice of science (e.g. the ethical public role of the scientist, or the optimal even be such a thing) that are clearly carved away as not being part of philosophy While epistemological issues in science ophy of science, the field had a broader journal Philosophy of Science and its Association (PSA), were founded in that was well aware of its potential William Malisoff, noted in an editorial "It started publication in 1934 when was beginning its systematic abuse to carry on in its pages a fight against unfriendly appeasers." (Malisoff 1944, This is not surprising given the political 1930s philosophy of science. As recent cism has noted, it was a belief in the 1 For a similar assessment, see Frodeman and set of essays that address one aspect of how Ö SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
320 Synthese (2010) 177:317-335 concern over the obscurantist metaphysics that drove philosophers such as Rudolph standing of science (e.g. Friedman 1996). science project was not about reductionism, disparate fields of science so that science humanity and the solving of social problems with whom the logical empiricists found Dewey) agreed with many aspects of In characterizing pre-WWII philosophy losophers of science, qua philosophers, implications of their work and saw implications into the social and political Even as the forces unleashed by science about the inherently beneficial nature continued to be broadly construed. 1946, the fledgling society was dedicated of the subject of philosophy of science, of practical consequences which may losophers in particular and to men 1948, vol. 15, p. 176). It seemed obvious be such "practical consequences ... of This broad venture was reflected in phy of science. Philosophy of science probability theory, or scientific explanation. between science and democracy, the place of social values within the practice reflected this range. For example, not the ethos of science appear in Philosophy WWII work on science-society relations (Butts 1948) and "Science as Morality" essays on the role of science in policy Merton's essay is part of a symposium tion"), on the sociology of knowledge and on the production of science in It also carried on a fairly robust discussion ethics (e.g. Hartman 1950, 1958; Gottshalk However, in the 1950s, arguments should not be taken to be philosophy be included in the field be substantially gaged philosophy of science. Consider, anthology Readings in the Philosophy looks familiar to today's philosopher declares that studying the social context or any other ethical questions related of science (1953, pp. 3-4). The philosophy to explicate what scientists mean by â SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
Synthese (2010) 177:317-335 321 science, or an attempt to reflect on scientific knowledge. Thus, the philosopher the nature of causality, of scientific laws, in science, of probability - what now (1953) anthology reflects this view of (1954) presents a similar program for By the late 1950s, these arguments typical of the journal from 1934-1959 this narrowing of the acceptable topics greater sense of disciplinarity. In reflection was pared down to the development broader practical consequences eschewed: the furthering of studies and free discussion philosophy of science, and the publishing field." (Philosophy of Science, 1959, welcomed, philosophy of science was issues, and agenda; whether it was of tant. Howard laments of the narrowing: place of science in society had all but philosophy of science pursued by a new specialists whose inability to think carefully as disdain for irrelevant, non-technical Recent scholarship on this period helps narrowing the scope of philosophy of so focused (Reisch 2005, 2009; Uebel bly pressures to depoliticize academic science could focus exclusively on the potential political entanglements of Drives toward professionalism also seem as was the growing institutionalization rowing of the field that we see the founding Boston) for philosophy of science. I will not attempt here to develop a the philosophy of science was narrowed sophical understanding of science in of science has altered its self-conception the possibilities, can do so again. 2 It must be noted that Feigl did not appear to (1956), he includes some of the sociological concerns a legitimate part of philosophy of science. 3 Howard (2003) blames Richard Rudner for this It was an increasingly active PSA governing board of articles in the journal, under the direction Washington University, St. Louis). 4Q SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
322 Synthese (2010) 177:317-335 3 Doing philosophy of science in context: Today there are pressing issues concerning philosophers of science could provide role for science? As funding streams p. 79), how will this affect science and mercialize the products of science affected Has science become too politicized (separate science and policy? If so, how? its research efforts? Should some research powerful abilities such research could can benefit from such knowledge and What are the obligations of the scientist public?4 Should scientists be advocates, this fairly lengthy list, there is plenty of science to do. Pursuing these conceptual issues will of science, expanding back to the early ical issues relevant to science, not this, a key step for philosophers of which practitioners face the issues vant work, philosophers of science of science in society is apparent. Examples addressing them) include the use of the making of policy with respect entific communities (Longino 2002; work in close contact with the public start, much more can and should be In developing this engagement, the losophy of science and the broader interest, will not be one of mere application has become apparent in the field of rarely provides either the philosophical ing into a complex context such as Kantian approach) and attempting to assistance or illumination (Beauchamp a straightforward application, applied one off-the-shelf theory is in grappling theoretical development.5 4 Elliott (2006) and Pielke (2007) present an 5 My experience teaching environmental ethics apply traditional ethical frameworks to the duties because we care about humans) seem dressed (e.g. inter vs. intra generational equity ethical extensionism (attempts to extend traditional or ecosystems). Sober (1986) provides a nice â SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
Synthese (2010) 177:317-335 323 Similarly, it is rare that philosophy shelf approach. The debates that have of science since 1960 have little bearing provide inappropriate guidance (as I relevant philosophy of science in this such work, the possibility for critical short, there are issues that arise within ical attention, and such attention could relevant publics) and the philosophy The best argument one can give for is a demonstration of its effectiveness have also used this approach (e.g. I will illustrate this approach with examples and thus attempt to show how doing relevance can be productive for both from philosophical insight. Of course, to convince a skeptic. Once this approach mine whether in fact it proves its merit. whether others will use it with fruitful gathered. Is such an effort worth it? One might worry whether and why practitioners and users of science will welcome philosophers of science to debates over how to use science in policy, the proper public role for scientists, and how to direct science funding. I think philosophers have much to offer practitioners - philosophers of science bring into these contexts a valuable skill set, namely capabilities to think clearly about sci- ence, about the relationships between evidence and theory, and about the assumptions and inferences made in science. Philosophers can use these skills to pinpoint, articu- late, and address difficult conceptual issues. Even if the standard philosophical views provide little direct assistance, utilizing these skills in various contexts can provide an invaluable service to scientists and the public. Engaging in these messy and complex contexts can also provide invaluable insights for philosophy of science. By paying close attention to the particular difficulties a sci- entific community and/or its public faces, rather than presuming the issues on the ground will fit into standard philosophy of science topics, philosophers of science can gain new perspectives on traditional debates. It is in the examination of and engage- ment with scientific practice in society that philosophers of science can test their well developed theories, see their pitfalls, and develop new approaches to old problems. The crucible of application forces clarity and adds needed complexity. Indeed, many philosophers of science whose work focuses on a particular science (neuroscience, physics, evolutionary biology, chemistry, ecology) already know how useful it is to examine actual scientific practice. By engaging with these issues on the ground, not just in the special disciplines but also in the contexts of science engaged with society, and with an openness to challenging standard philosophy of science views, fruitful interaction can flourish. Ô SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
324 Synthese (2010) 177:317-335 4 Science in policy and the value-free So onto the illustrative examples. My the interface between science and policy-opment of science for and use of science place of expertise in society goes back the use of science in policy-making, regarding science in society that result, As many commentators have noted, science and government in the United Kevles 1995; Smith 1990). Some have ence in the post- WWII era. Others in advising the government. Underlying and policy was the "linear model," the science will produce goods (information, idea was that one needed first basic research, which could then produce tain the "store" of basic research upon would eventually suffer. Thus, the famously in Vannevar Bush's Science: of the relationship between science mid-1960s, and was not seriously disrupted In addition to providing a reason to relationship between science and policy, value-free ideal for science (Pielke free ideal, which is the ideal that scientists to assist with the assessment of theories, (Douglas 2009a, Chap. 3). By the mid-that some values are needed in science and evidence. As no amount of evidence are always needed to assess whether (Rudner 1953; Hempel 1965). If scientists needed to complete the argument for range of values, including social and gested), philosophers of science came to "canons of scientific inference" simplicity, scope, explanatory power, Although not expressly mentioned acceptance of the linear model probably ability for the value-free ideal. The 6 One might think that debates over underdetermination But those debates are usually about the fate of is the only logical choice given a complete such luxury, involving what Howard has called we know right now, incomplete as it is, and thus â SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
Synthese (2010) 177:317-335 325 enter into the picture only after scientific mous scientific community produced stock of intellectual resources for became relevant when deciding on tions to pursue. Thus, the linear model proceed wholly separate from societal and ethical values, would be relevant However, by the 1980s, the linear prosperity had faltered despite rising to doubt whether scientists left to their benefits to society (Sarewitz 1996, ject (an ill-conceived attempt to drill (Greenberg 1999, Chap. 9). Whether internal affairs with even basic integrity made headline news (Guston 2000, applied to product is seen as either grossly tion of actual practice (Pielke 2007, of how to conceive of science in society replacement for the linear model. Just as the linear model was once ideal seems central to the protection wide range of philosophers across multiple Shrader-Frechette 1994, p. 53; Shrader-underlying concern usually raised in values to influence science in any role ence, leaving it open to overt political particularly social or political, could the very value of science. Yet if one talks to practitioners at the creates problems for them. For scientists plete) that will be used to shape policy to decide which interpretation to accept of no assistance, as it provides faulty situation where one must make a theory context. Asking whether a theory to broad scope, is beside the point. Several claimed to be the simpler or the more obvious relation to whether the interpretation making. In the policy context, what to interpret available data provide a reliable 7 Although Shrader-Frechette also undermines Frechette 1991, Chap. 9; Shrader-Frechette 1994, 8 This is something I have had the opportunity for Risk Analysis (SRA). I am grateful for all SRA and who have responded to my ideas presented â SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
326 Synthese (2010) 177:317-335 the interpretation is used, the predictions out. Such reliability is central to the interventions. As Larry Laudan has broad scope, or even apparently explanatory true (Laudan 2004). Such "cognitive is likely to be fruitful, not about whether accurate. Laudan distinguishes epistemic nal consistency (which are needed to the cognitive concerns focused on fruitfulness. the minimal standards of empirical should not be considered a live option. the preferences of the scientists, according tists at the policy interface to consider indicative of predictive accuracy makes The value-free ideal also blocks crucial considerations of how much evidence should be considered sufficient. In deciding this, scientists should be concerned with the amount of uncertainty that is acceptable for any given decision, which can and should depend on the consequences of error. Those consequences include the potential construction of faulty policy based on the possibly erroneous scientific claims, and thus include social and ethical consequences (Douglas 2000, 2009a, Chap. 5). These are considerations that scientists working on policy-relevant science should be weigh- ing openly. Thus, in addition to attempting to select the most reliable interpretation of available data, there are important considerations pertaining to the consequences of choosing a model incorrectly - which sorts of errors are more acceptable, false positives or false negatives? Answering such a question in the science policy context requires the consideration of social and ethical values, the very values vanquished by the value-free ideal. The value-free ideal is of no help to such scientists, demanding that they focus on values of little relevance and that they ignore values of central importance. In examining the science-policy context more closely, and listening to the practi- tioners grapple with their concerns, the flaws of the value- free ideal become apparent. There is no rational reason why social and ethical values should be ignored when deciding how much evidence is sufficient for a claim, and there are moral responsibil- ity reasons for why they should be considered (Douglas 2003, 2007, 2009a, Chap. 4). Rejecting the value-free ideal should allow for more open discussion of how burdens of proof and levels of sufficient evidence should be set for science in policy debates. It also opens the door to greater democratization of the science advising process (Douglas 2005; Elliott 2008; Guston 2004). If social and ethical values are needed to weigh potential consequences of error, the affected publics can and should have input on the values scientists should use in this weighing, be it through mechanisms such as consensus conferences which allow for informed deliberation on value ques- tions or participatory research models where members of the public work closely with scientists to develop needed knowledge. However, some of the philosophical worries about the possibility of values running roughshod over evidence remain. There is a need to define a new ideal, one that both <ö SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
Synthese (2010) 177:317-335 327 protects the integrity of science and society. This is clearly work for philosophers Taking up this challenge, I have developed in scientific reasoning. In doing so, I reasoning rather than the kinds of value roles can be articulated, a direct role are a primary reason for a choice. They role values play in shaping many of values are a reason not to proceed. If a reason to proceed. At some locations problematic, and indeed may be required. or which methodologies are ethically values direct our choices. However, at role is deeply problematic. When deciding should be a reason in itself for such a choice. Nor should values direct our choices in how to characterize evidence. Values in the direct role at these locations instantiate the concern about wishing making it so, or about values having the same (or even more) weight as the available evidence. The pernicious uses of values in science can be traced to the use of values (whether cognitive or ethical) in a direct role when assessing theories. But the direct role is not the only role values can play in reasoning. The indirect role is also important, although values constrained to this role are not as determinative of the decision being made. In the indirect role, values only have import for assessing the sufficiency of evidence, for assessing how serious lingering uncertainties are. As there is always an inductive gap between evidence and theory, there is always some uncertainty to weigh. And in weighing this uncertainty both cognitive and ethical val- ues can be useful, depending on the context. Ethical values can help us assess our concerns over the consequences of error, of making an incorrect interpretative choice. Cognitive values can help us assess what the likelihood is that an error will be discov- ered sooner rather than later. Recall that theories which instantiate cognitive values like simplicity, scope, and explanatory power are at their core reflecting the potential fruitfulness of the theory - whether it is easier to work with, to develop, and to test further. As a result, the needed refinements (or wholesale rejections) of such theories are more likely to be apparent sooner. Thus, in theory assessment, the indirect role is the proper avenue for both cognitive and ethical values. And it is a weaker role than a direct role, protective of scientific integrity. In the indirect role, the values (ethical or cognitive) only have import through uncertainty, not through the direct assessment of the theory. The more one is able to reduce uncertainty, the stronger the evidential case, the less important the values are. Because values are constrained to the indirect role only at the heart of science, we could call this the constrained- value ideal. With this understanding of values in science, we can both clarify what the role for values in science should be (a central philosophical question) and clarify how scien- tists should proceed in their work at the science-policy interface. Values can saturate science, but still should not compete with evidence. Scientists can legitimately use social and ethical values to assess how much evidence should be considered sufficient for a claim, but with the value-free ideal replaced with the constrained-value ideal ^ SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
328 Synthese (20 1 0) 1 77:3 1 7-335 described above, scientists should feel free what they take to be sufficient evidence. should be as open and as clear as possible What does this discussion show about the phy of science? Note how an off-the-shelf key difficulties of the practitioners at the a direct engagement with a particular context have gained the most. Through such engagement, to provide relevant normative guidance and constrained- value ideal will prove a viable remains to be seen. But the general approach robust than one example. In actual contexts guidance, developing the implications for of science gains greater insight into the science become more socially relevant, but as a result. 5 Explanation, prediction, and weight of The potential productivity of such engagement the science-in-policy context. A few years approached me at a Society for Risk Analysis entists doing work relevant to policy-making, I might have some insight. The problem from multiple disciplines, often called the Krimsky 2005; Linkov et al. 2009). While in medical arenas for several decades 2005), such hierarchies are of little use in humans would be immoral. In the cases utterly unacceptable to conduct controlled with insights from various sources, from to animal studies, to epidemiological studies. yet all provide pieces of the puzzle important ture. Rhomberg wanted to know if philosophers such pictures were constructed and how reliability. The pitfalls of each evidential source are well known to the scientists. Epidemiolog- ical studies must study humans in real contexts with no ready-made controls to manage confounding factors. Epidemiologists must attempt to construct control groups, aware that confounders are nearly impossible to eliminate completely. In addition, actual exposure assessments usually must be indirect, relying upon historical accounts, per- sonal memory, or other surrogates rather than direct measurement. Biochemical stud- ies, while capable of providing great insight, are rarely complete. Cellular chemistry is so complex that a full accounting of a chemical's impact on cellular processes is not currently within our grasp. We can assess whether a chemical is likely to cause Ô SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
Synthese (2010) 177:317-335 329 mutations or to interact with key receptors, a chemical can do in a cell. Animal studies to a chemical and confounding factors, for humans, particularly given the uniformities the differences among strains and species, So no one source of evidence is going for studies of environmental contaminants. the evidence from these disparate sources. be weighed? As we discussed the problem, it was plex set of evidence were often developed, different explanations of the available invoked.) Scientists create explanatory evidence as possible, discounting countervailing as possible. For example, consider a X. Suppose one has some evidence can cause cancer in humans, but there exposures to other substances that are also mixed, with chemical X causing of rodents, but with less clear results statistical significance or disputes about studies. And suppose biochemical studies some evidence of ability to interact in the cell. This kind of complexity account could interpret this set of evidence X to cause cancer through a possible metabolism, a potential that shows up Another explanatory account could X does not cause cancer, because the studies present have a greater ability particularly susceptible to metabolic so susceptible. What scientists in a policy-making account is better? And by better, the more fruitful or interesting explanation, as a basis for policy-making, for developing science can be of some use, and in the questions. When philosophers of science first began seriously tackling the issue of scientific explanation, the relationship between prediction and explanation was tight (Hempel and Oppenheim 1948). Explanations, under the traditional deductive-nomological theory, were deductive arguments from scientific laws and specific initial condi- tions. If one could deduce the outcome under these circumstances, one had both an 9 For more details about these sources of uncertainty in the case of dioxin, see Douglas (1998). Ô SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
330 Synthese (20 1 0) 1 77:3 1 7-335 explanation of the outcome and, from a of the outcome. As critiques of the deductive-fler 1957; Rescher 1958; Hanson 1959; Scriven prediction and explanation was undermined, aspect of the discussions on scientific By the 1970s, various competing theories including unificationism and causal explanation Salmon 1989). By 2005, the literature had various strengths and weaknesses of the logical, causal, unificationism, mechanistic), of the competitors. Each seemed to be a in different cases, and each seemed to scientific explanations that did not fall viding a completely satisfying view of accept such plurality (e.g. Lipton 2004). philosophical examinations of explanation. Throughout this period (post- 1970) of tion, prediction did not appear to be conceptually by the weight of evidence problem, I have reconnect explanation to prediction (Douglas are central to science in part because they way of organizing empirical information ily forthcoming. I have given an account each could serve the cognitive goal of generating can then serve to test the reliability of the prediction and explanation that one can an understanding of the important function To see how explanations serve as cognitive ing example. Causal explanations rely upon ena. For these kinds of explanations, one causation to predict phenomena that would if one has a causal story about how a substance altering something in the bloodstream, ations by exposing animal subjects to the relevant components in their blood that are the appearance of the alterations, and then Depending on the type of explanation, utilized to generate a new prediction. For presence of a causal factor to generate predictions. nations, the deduction of effects from a prediction (if completed prior to testing). unified include boundary areas that now account - and thus new clear predictions tic explanations, the understanding of the it, and thus new predictions. For different erated in different ways, but for all explanations, £} SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
Synthese (2010) 177:317-335 331 Having this understanding of how with the weight of evidence problem. weight of evidence accounts is that atic evidence (from the perspective While consistency in the reasons for further testing of the explanatory can be used to ensure that the explanation of the evidence. By providing a further of that explanation can be better ensured, Of course, such a predictive test does that it will always prove to be reliable. alistic such expectations would be. approach to weight of evidence, linked framework for the weight of evidence addition, it has provided useful insight philosophers of science - the relationship oping the theory I have laid out above will demonstrate whether the theory refine my views on explanation and diction and explanation is but one part which must also include approaches exclusion of evidence, and the comparison further fodder for developing a socially 6 Conclusion: broadening the picture As I hope these two examples show, socially relevant contexts can produce broader society. In doing so, philosophers The second example above, the weight a largely epistemic problem, and the insights on standard epistemic approaches. the context of science in policy-making, into account. Working with actual scientific practice is a not new technique for philosophers of science. What is new here is the engagement with less "traditional" (if tradition is traced back only to the early 1960s) contexts of scientific practice. Concern with, for example, the use of science in policy-making, the development of policy for science, the understanding of science in education, the public role of scientists, or the proper ethical restrictions on scientific practice, will not only broaden philosophy of science, but improve its conceptualization of science. These kinds of activity are an integral part of the scientific enterprise - even a cursory look at the amount of money dedicated to "applied science" or the journal space devoted to science relevant to public policy indicates what a profound part of the scientific world such practices compose. As such, science is not just an epistemic enterprise. It is also a moral, social, and political £) SpringerThis content downloaded from 175.111.91.124 on Fri, 27 Jan 2017 04:55:48 UTCAll use subject to http://about.jstor.org/terms
332 Synthese (2010) 177:317-335 enterprise. A philosophy of science is impoverished and anemic. Despite their general neglect of these losophers of science have useful insights contexts, particularly in examining contexts where science and society meet, Working on these issues will broaden potentially alter our theories of science. account changes the scope of the philosophical difficult to grapple with the greater for philosophy of science to do so, allowing refinements. Thus, the best way to make philosophy apply standard views to problems of philosophy of science should be a two-assist with thorny problems and those science. In reflecting on embedded science, of social relevance. Acknowledgments My thanks to Kevin Elliott, Glenn Graber, Ted Richards, Jacob Stegenga, the History and Philosophy of Science and Technology Reading Group at the University of Tennessee, and two anonymous reviewers for assistance with and feedback on this essay. Also thanks to Lorenz Rhomberg for bringing the weight of evidence problem to my attention and for collaborating with me on its solution. A special thanks goes to Katie Plaisance and Carla Fehr for organizing the APA Pacific mini-conference in 2008 on "Making Philosophy of Science More Socially Relevant," where the bare bones of this paper were first presented, for critical feedback on this paper, and for shepherding the resulting volume through to publication. References