Collective belief and acceptance

ABSTRACT. Margaret Gilbert explores the phenomenon referred to in everyday ascrip-tions of beliefs to groups. She refers to this type of phenomenon as “collective belief” andcalls the types of groups that are the bearers of such beliefs “plural subjects”. I argue thatthe attitudes that groups adopt that Gilbert refers to as “collective beliefs” are not a speciesof belief in an important and central sense, but rather a species of acceptance. Unlike properbeliefs, a collective belief is adopted by a group as a means to realizing the group’s goals.
Unless we recognize that this phenomenon is a species of acceptance, plural subjects willseem prone to change their “beliefs” for irrelevant reasons, and thus frequently appear toact in an irrational manner.
Margaret Gilbert explores the phenomenon referred to in everyday ascrip-tions of beliefs to groups. She refers to this type of phenomenon as“collective belief“ and calls the types of communities that are the bearersof such beliefs “plural subjects”. In this paper, I argue that the attitudesthat groups adopt that Gilbert refers to as “collective beliefs” are not aspecies of belief, but rather a species of acceptance. Unlike proper beliefs,a collective belief is adopted by a group as a means to realizing the group’sgoals.
Unless we recognize that the phenomenon that concerns Gilbert is a species of acceptance, plural subjects will seem prone to change theirviews for irrelevant reasons, and thus frequently appear to act in an irra-tional manner. When we recognize that plural subjects accept views ratherthan believe them, we will realize that some considerations that would beirrelevant to determining what an individual should believe are relevant todetermining what a group should accept.
Gilbert’s account of collective belief is part of a larger project, an attackon “singularism”, “the thesis that [our vernacular collective concepts] areexplicable solely in terms of the conceptual scheme of singular agents”(Gilbert 1989, 12). According to Gilbert, “one acts as a singular agent inso far as one acts in light of one’s own goals (Gilbert 1989, 12). Gilbert, Synthese 129: 319–333, 2001.
2001 Kluwer Academic Publishers. Printed in the Netherlands.
though, does not believe that we only act in light of our own goals. Shecontends that “in principle someone could be motivated by his understand-ing of what his group’s goals are, without reference to what pleases him”(Gilbert 1989, 427). Collective beliefs, Gilbert claims, are similar. When agroup adopts a collective belief each individual member of the group mayexpress or act on a belief that is not her own.
Gilbert presents a number of examples of the sort of phenomenon asso- ciated with everyday ascriptions of collective belief. All of the followingare instances of what Gilbert calls “collective belief”.
(1) When a preferred interpretation of a particular poem emerges at the meeting of a poetry discussion group, sometimes it is the group as a unitthat is said to hold this belief about how the poem is to be interpreted(Gilbert 1989, 288–9). As Gilbert explains, at some point remarks of thefollowing sort become acceptable: ‘We are agreed that the last line is moving’; ‘We think that, etc.’; ‘In our opinion, etc.’; ‘Wedecided that, etc.’; ‘Our view at this point is that, etc.’; ‘In the opinion of our discussiongroup, etc.’; ‘The group thought that, etc.’ (Gilbert 1989, 289) And the belief in question should be taken for granted in subsequentdiscussions.
(2) When a class of students have been told something by their teacher which “has subsequently been taken for granted in class discussions andtests”, the assumption is regarded as the collective belief of the class, as aunit (Gilbert 1994, 237). For example, if a class is “told by their teacherthat modern science has a ‘masculine’ bias”, and this has been taken forgranted throughout the course, then the class collectively believes this.
(3) When a group of scientists work together in a laboratory on a project in a particular sub-field, they may collectively believe that the sub-field isa subject worthy of investigation (Gilbert 1994, 237). Were a member ofthe group to suggest that the sub-field that they are working in “isn’t goinganywhere”, it is legitimate for the others to think “how can he say that –he’s one of us!” (4) When two parents decide that their child should be home at a certain time in the evening, it is the parents as a unit that believe this, despite thefact that neither parent may individually believe that this is when their childshould be home (Gilbert 1994, 249–250).
Gilbert argues that in all of these cases a “plural subject” of belief is formed. She uses the phrase “jointly committed” to describe the rela-tionship between the individuals who constitute a plural subject (Gilbert1994, 245). According to Gilbert, the various individuals who come to bepart of a plural subject of belief aim to coordinate their actions such thatthe collective outcome is compatible with the group as a unit believing that which they agreed to believe as a plural subject. They jointly committhemselves to act collectively as if something were true. In Gilbert’s words, the behavior that results from collective belief is driven by the concept of belief and theconcept of X-ing-as-a-body. It is as if the participants ask themselves, “what do I need todo to make it the case that I and these others together believe that p as a body?” – and thenact accordingly. (Gilbert 1994, 252–3) Each individual member of a plural subject is obliged to act in such a waythat the collective behaviour of the group is consistent with the group as aunit believing that which they are jointly committed to believe.
Further, the members of a plural subject are held accountable for their actions by each other. And, given a collective belief, dissension warrantssanctions, most often in the form of a rebuke (Gilbert 1994, 240). As a res-ult, the individuals who form a plural subject can legitimately be coercedby other members to do their part in ensuring that the plural subject acts asif it believes what they agreed to. Gilbert claims that collective beliefs even“justify the infliction of pressure by one person on another in the service ofbeliefs which may be false” (Gilbert 1994, 253). A rebuke in the context ofa plural subject relationship, Gilbert notes, is “directed at the speaker . . . inlight of certain apparently nonepistemic facts about him or her” (Gilbert1994, 236). One is not rebuked because one has uttered something false orepistemically unjustified. Rather, one is rebuked because one has violatedan obligation (Gilbert 1989, 292).
Gilbert characterizes the obligation in the following manner. When one decides to “collectively believe” something with others “one commitsoneself to presuppose-together-with-the-others or jointly to presuppose thetruth of the view in question” (Gilbert 1989, 301). “One is committed tomouth the jointly accepted view and to act upon it” (Gilbert 1989, 305).
And, “if one feels bound to speak against the group view, though one isnot ready to challenge its status as the group view, one must preface one’sremarks making it clear that one is speaking in propria persona” (Gilbert1987, 195).
Further, Gilbert claims that “one who participates in joint acceptance of p thereby accepts an obligation to do what he can to bring it aboutthat any joint endeavours among the members of [the group] be conductedon the assumption that p is true” (Gilbert 1989, 306). The jointly accep-ted belief is granted the status of an assumption in subsequent reasoningand discussions with the relevant others (Gilbert 1989, 309). Thus, thejointly accepted belief gives the parties to the collective a reason for acting(Gilbert 1994, 247).
Gilbert emphasizes that one can even become part of a plural subject without entering into an explicit agreement to that effect. As she explains, “a group may surely have a belief though there has been no consciousaim to arrive at one” (Gilbert 1989, 293). “Nor need they ever explicitlyacknowledge that a certain view is the group’s view” (Gilbert 1989, 293).
According to Gilbert “what is both logically necessary and logically suffi-cient for the truth of the ascription of group belief . . . is . . . that all or mostmembers of the group have expressed willingness to let a certain view‘stand’ as the view of the group” (Gilbert 1989, 289). Elsewhere, Gilbertdevelops the idea of “letting a certain view stand as the view of the group”in terms of a “joint commitment to accept as body” (see Gilbert 1996,7–15).1 Gilbert suggests that there are at least three good reasons for individuals to agree to become part of a plural subject and adopt a collective belief.
First, often, by adopting a collective belief, a group of individuals are ableto temporarily resolve or avoid a conflict that is preventing them fromaccomplishing some task or finding some information (Gilbert 1994, 253).
Second, plural subjects are often able to accomplish tasks that indi- viduals would find difficult or perhaps even impossible to accomplish. Aparticular sort of investigation, though possibly not feasible for any in-dividual, may be manageable for a plural subject.2 Thus, sometimes inorder to get what we desire it will be necessary to become part of a pluralsubject. Certain sorts of goods are only available to those who are willingto become part of such a subject.
Third, collective beliefs provide “individuals with a sense of unity or community with others” (Gilbert 1994, 253). Gilbert implies that this senseof community that we experience in plural subject relations is not merelyvalued instrumentally, but also as an end in itself. She suggests that in someplural subject relations “the sense of being a part of something other (and‘greater’) than oneself and other than the other is to some extent present:together you make a new thing, you are, individually, the components ofsomething” (Gilbert 1989, 225).
Gilbert states emphatically that the phenomenon she seeks to explain is not a matter of each of the members of a plural subject individuallyholding the belief in question. Following Anthony Quinton, Gilbert refersto the sorts of accounts that give such an explanation of collective belief as“summative accounts” (see Quinton 1975). A summative account is “oneaccording to which for a group G to believe that p it is logically necessarythat all or most members of G believe that p” (Gilbert 1987, 186). Gil-bert presents two “summative” accounts of “collective belief” which sherejects. Collective belief is not to be equated with either (a) a belief being“widespread in a particular community”, or (b) “common knowledge inthe community” that a particular belief is widespread.3 According to Gilbert, summative accounts are unacceptable for the fol- lowing three reasons. First, neither in cases of widespread belief nor incases of common knowledge that a belief is widespread would a rebukebe warranted were a member of the community to act in a way expressingdisregard for the belief that is held by the community (Gilbert 1994, 243).
In such situations a community may try to coerce a member into believingor acting as if they believe something that is widely held to be true. But,such coercion is not warranted as it is in a plural subject arrangement. Thedissenting member made no commitment to accept the belief in question.
The notion of a warranted rebuke figures importantly in Gilbert’s account,for, in cases of everyday collective belief ascription, “one is not rebuked[merely] because one hasn’t conformed” (Gilbert 1987, 193).
Second, “participating in believing that p as a body does not require personally believing that p” (Gilbert 1994, 251). That is, our everydayascriptions of beliefs to groups does not require that all members of thegroup individually believe that which is collectively believed. In fact, noteven most of the members of a plural subject need hold the collective beliefindividually (Gilbert 1994, 251).
Third, Gilbert does not believe that summative accounts have the con- ceptual resources necessary to account for the essential characteristic ofcollective beliefs. Specifically, in summative accounts “the fact that agroup is involved does not play any obviously essential role in what isgoing on” (Gilbert 1987, 189). Summative accounts are thus unable toexplain why it is reasonable to say that two co-extensive groups do nothave the same beliefs.
In summary, Gilbert believes that she has identified a type of phe- nomenon of which an adequate understanding can only be had if onerejects singularism. In the sorts of cases that she has in mind, there is agroup belief according to our vernacular conception, and people are actingin light of it, though they may not themselves personally hold the belief inquestion.
In “Remarks on Collective Belief”, Gilbert notes that when she initially de-veloped her account of collective belief she unreflectively used the phrases“believes that p” and “accepts that p” interchangeably (Gilbert 1994, 252).
Though she recognizes that some philosophers, such as Robert Stalnakerand Bas Van Fraassen, believe that there are important distinctions to bemade between belief and acceptance, she claims to use these words in a fa-miliar way according to which they are interchangeable (Gilbert 1994, 252; 1996, 8). Thus, Gilbert does not deny that there is a distinction betweenbelief and acceptance, but she does not explore the significance it has forher analysis.4 In the remainder of the paper, I argue that the distinction between belief and acceptance is relevant to understanding the phenomenon that Gilbertcalls “collective belief”. Unlike proper beliefs, the views adopted by pluralsubjects are adopted as a means to realizing the group’s goals. Further, Iargue that the sorts of considerations required to change a plural subject’sview are different from the sorts of considerations required to changea person’s beliefs. Understanding this difference is vital to an adequateunderstanding of the phenomenon Gilbert calls “collective belief”.
“Belief” is a concept that is under much strain. Some philosophers have even recommended abandoning it altogether, arguing that it is animpediment to achieving a scientific understanding of human cognition.
Paul Churchland, for example, argues that “our common-sense conceptionof psychological phenomena [which includes talk of beliefs] constitutesa radically false theory, a theory so fundamentally defective that both theprinciples and the ontology of that theory will eventually be displaced”(Churchland 1981, 67). Given the scope of this paper and the inconclus-iveness of the arguments against folk psychology, I will not attempt toaddress these more radical criticisms. My intention in this section is toshow that we should draw a distinction between two types of cognitiveattitudes, belief and acceptance.5 I want to begin by reviewing the distinctions that various philosophers have drawn between belief and acceptance. Stalnaker and Van Fraassenconceptualize the relationship between belief and acceptance in differentways. Most significantly, Stalnaker construes belief to be a species of ac-ceptance (Stalnaker 1984, 79), whereas Van Fraassen construes acceptanceto be a cognitive attitude distinct from belief, and differing from belief indelivering us from metaphysics (Van Fraassen 1980, 69). Both, though,claim that we can accept things that we do not believe. For example,one might accept something that one does not believe for the sake of anargument.6 L. J. Cohen (1989; 1992) also provides insight into understanding the relationship and differences between belief and acceptance. Cohen notesthe following differences. Whereas “to accept that p is to have or adopta policy of deeming, positing, or postulating that p” (Cohen 1989, 367),belief does not involve having such a policy.7 Rather, “belief that p . . . is adisposition to feel it true that p” (Cohen 1989, 368), and no such feelingneeds to accompany acceptance. In fact, one can accept something even when they feel it is false. Further, whereas acceptance is voluntary, andthus “decidable at will”, belief is not (Cohen 1989, 369).
The foregoing analyses suggest that belief and acceptance differ in the 1. you can accept things you do not believe, whereas you cannot believe 2. acceptance often results from a consideration of one’s goals, and thus results from adopting a policy to pursue a particular goal; 3. belief results in a feeling, in particular, a feeling that something is true;4. and, acceptance can be voluntary, whereas belief is not.
Most important, for our purposes, is the fact that what a person or group accepts is accepted in light of a goal or set of goals. Beliefs are not like this.
They are not tailored to our purposes.
Given these differences, outlined above, it seems that the phenomena Gilbert has in mind are more aptly described as instances of acceptance,rather than belief. Indeed, I do not deny that plural subjects behave inways consistent with their believing the collectively accepted view. But,the views plural subjects adopt are frequently the consequence of a policyto deem, posit, or postulate something. And, the goals of the plural subjectdetermine what they “collectively believe”. Proper beliefs are not like this.
That is, they are not tailored to our goals. A brief review of Gilbert’s ex-amples will aid in proving that the phenomena Gilbert has in mind oughtto be regarded as involving acceptance, rather than belief.
Consider Gilbert’s parenting example. The parents are jointly commit- ted to raising their child effectively. Thus, as a unit, the parents developviews with this goal in mind. It is their collective goal that determineswhat they claim to “collectively believe”. Sometimes they will even adoptviews that conflict with their personal preferences and beliefs. Similarly,the poetry discussion group may have a policy (perhaps only a tacit policyestablished by precedent) to end their sessions by stipulating how the poemunder study should be read. This policy will lead the group to postulate aparticular reading of the poem under study as the group’s reading. Again,we see that the group is led to adopt a view as a means to realizing theirgoal.
And, the scientists working in the laboratory together first set their goal of working together on a research project. This shared goal then leads themto adopt the view that whatever sub-field they choose to work in is a subjectworthy of investigation. There are probably a number of sub-fields thata group of scientists working in a lab together could pursue research in,but a conscious choice must be made. The required equipment must bepurchased. Similarly, in the class where a certain assumption is taken for granted, the assumption is posited, and the class then reasons accordinglyfrom that assumption. And the goal of the class determines what sorts ofassumptions are posited. Neither in this case, nor in the other three, are werequired to assume that the plural subject has a disposition to feel that thecollectively held view is true. Hence, the views of plural subjects are bothexplicable without ascribing beliefs to them, and more aptly characterizedas instances of acceptance.
So far, I have argued that belief and acceptance are distinct cognitivestates, and that when groups adopt views what they do is more aptly de-scribed as “acceptance”. Now I want to explain why we should insist ondistinguishing between belief and acceptance in this context.
When we know that a plural subject accepts a particular view, we know that the view they accept is perceived to be an effective means to realizingtheir collective goal. Hence, we also know that the subject may be imper-vious to persuasion or evidence that supports an alternative view. This isso because the commitment of the group to the view is not grounded on afeeling that the view is true. And, as Stalnaker and Van Fraassen note, onecan accept something that they do not believe. Consequently, if one wishesto change the view of the subject of a “collective belief” one must aim toshow the plural subject that their view is an impediment to the realizationof the goal of the group. Thus, altering the views of a plural subject willrequire strategies of behaviour very different from the behaviour requiredto change the belief of an individual.8 Beliefs are typically changed by eroding the subject’s disposition to feel that what they believe is true. This is done, typically, by providing evidencethat suggests that one’s belief is false. The views of plural subjects, on theother hand, may persist even when it is apparent to the subject that the viewin question is false. One who seeks to change the view of a plural subjectmust show how the obligation that the group has to accept the view isinappropriate. This may involve providing evidence that suggests that thecollectively accepted view is false, but it need not. One might also appealto other concerns, practical and ethical concerns, for example. But, onemust show that the view that the plural subject collectively accepts doesnot serve the interests or goals of the plural subject.
And, just as an individual seeking to change a plural subject’s view can appeal to considerations beyond the truth and falsity of the view, aplural subject that is considering changing its view will be open to a dif-ferent range of considerations than those that bear on the truth or falsity of the view. It is in this respect that a thorough understanding of the phe-nomenon Gilbert calls “collective belief” requires us to distinguish belieffrom acceptance. The behaviour of plural subjects will frequently seemirrational otherwise. We will notice that plural subjects sometimes changetheir views on the basis of a consideration of factors that have no bearingon the truth or falsity of the view.
For example, someone might change a laboratory team’s “collective be- lief” that their chosen sub-field is worth investigating by showing them thatsignificant sources of funding for such research will no longer be available.
This does not demonstrate that the sub-field is not worth investigating. Thatis, it does not undercut the epistemic merits of such an investigation. Itmerely demonstrates that it may not be prudent to continue research in thatarea. But such a consideration may be sufficient to lead the laboratory teamto change its “collectively held belief”. That is, the laboratory team mayno longer act as if it were true, or accept it as an assumption in subsequentreasoning and discussions. Note, though, the truth of the claim has notbeen called into question.
I want to distinguish what I am arguing here from a point made by Frederick Schmitt (1994). Schmitt argues that “chartered” groups, groupsthat have a commitment to performing only a certain kind of joint action,“must employ a special standard” of epistemic justification, a standardsuited to the social role of the group (Schmitt 1994, 273). My concernis not with epistemic justification. In fact, I am arguing that, generally,the views plural subjects accept are and should be judged, not by epi-stemic standards, but by practical standards. That is why plural subjectsare and should be responsive to different reasons than the reasons indi-viduals are responsive to when they are deciding what to believe.9 Thereis an important exception, one that I will analyze in greater detail in thenext section. Some groups have epistemic goals. In particular, researchgroups and laboratory teams have epistemic goals. Consequently, they arefrequently moved by both epistemic and practical considerations, a pointthat has too often been overlooked, and is probably partially responsible fora series of misunderstandings that have given rise to the “science wars”.
Having shown that the distinction between belief and acceptance has asignificant bearing on Gilbert’s account of the phenomenon she calls “col-lective belief”, I now want to briefly demonstrate the explanatory powerof her model in its revised form developed here. In particular, I want todemonstrate its relevance to issues in science studies. Gilbert has already shown how the notion of collective belief can offer us insight into un-derstanding the complex processes underlying scientific change (Gilbert2000). My aim is to show how what I will now call “collective accept-ance”, my preferred label for the phenomenon Gilbert describes, providesphilosophers of science with the conceptual resources necessary to diffusethe concerns raised by social constructivists like Bruno Latour and StevenWoolgar. Latour’s and Woolgar’s detailed anthropological studies of labor-atory life have deepened our understanding of scientific practice. But, I aimto show that the conclusions they draw from their studies are unfounded.
Latour and Woolgar (1986, 157, 166) aim to show that scientists are not moved solely by epistemic considerations in their decisions to accept hy-potheses. They are attacking a rationalist picture of science, a picture thattries to explain scientific success solely in terms of good reasons, evidenceand the like. The evidence against the rationalists’ view is overwhelm-ing. Scientists are frequently moved by nonepistemic considerations (seeHull 1988 and Solomon 1994). Consequently, at least in its most extremeformulations, the rationalists’ picture of science is unacceptable.10 In itsplace, though, Latour and Woolgar are proposing a social constructivisttheory of science. Their sustained study of laboratory life is meant to showhow negotiation, a social process, influenced by the interests of the peopleinvolved and their various relations of power, enable scientists to trans-form statements into facts (Latour and Woolgar 1986, 151). Accordingto Latour and Woolgar, the encounters they observed in the laboratory“can be adequately accounted for using the notion of fact construction, and. . . this makes unnecessary the use of ad hoc epistemological explanations”(Latour and Woolgar 1986, 166). It is this, their proposed successor theoryof science, that philosophers of science generally find so objectionable. AsAlvin Goldman (1999, 12–17) expresses the point, Latour and Woolgar donot allow nature to play a role in determining scientists’ beliefs.
In what follows, I will show how Gilbert’s account of collective accept- ance provides us with the resources for an alternative explanation of whatgoes on in scientific laboratories, an explanation that is apt to be regardedby most philosophers as superior to Latour’s and Woolgar’s explanation.
The key is to discern the various types of considerations that influencescientists working in laboratories. On the one hand, the people Latour andWoolgar study are part of a laboratory team. In this respect, they constitutea plural subject. Their negotiations aim, in part, at determining what theplural subject’s view will be. Here they will be moved by all sorts of prac-tical considerations. We can see this in Latour’s and Woolgar’s analyses ofwhat they call “the microprocessing of facts”. For example, one of theirsubjects, Smith, is concerned that he does not have the time to perform additional experiments to confirm a claim made by English researchers inanother laboratory (Latour and Woolgar 1986, 158–159). In his capacityas a member of the lab team, Smith has a number of responsibilities, andin order to ensure that the lab team as a whole can realize their goalshe has to use his time in the laboratory effectively. Even though Smithdoes have doubts about the findings of the English researchers, he is re-luctant to waste his time conducting the type of test required to confirmtheir claim. As a member of the lab team, he collectively accepts certaingoals and views. And it is because he has this commitment that he is in-clined to suppress his own personal worries about the English researchers’reliability.
But the subjects of Latour’s and Woolgar’s study are also scientists, people with epistemic goals. Insofar as their epistemic goals move them,they are concerned with the epistemic merits of the hypotheses they areentertaining. In this respect, they will also be moved by considerations ofevidence. This is exactly what we see in the exchanges documented byLatour and Woolgar. For example, after Smith was observed expressinghis concerns about the time required to confirm the English researchers’results, his colleague, Wilson, raises a distinctively epistemic considera-tion. He suggests that the English researchers were premature in makingtheir claim about the nature of the substance in question (Latour and Wool-gar 1986, 158–9). That is, he questions whether their claim is adequatelygrounded. And Smith is well aware that the conversation is now concernedwith epistemological considerations, rather than the practical concerns ofthe lab team, for he identifies the basis for Wilson’s concern, the factthat “there is a definite variance between pig and ovine sequence [which]you cannot deduce . . . from the amino acid analysis”, as the English re-searchers did (Latour and Woolgar 1986, 158–9). The negotiation goingon between Smith and Wilson is not aimed at constructing a fact. Rather,they are trying to determine whether their own personal epistemic worrieswarrant changing the group’s direction of action. Their own worries canpersist even if they decide that the group should proceed on the assumptionthat the English researchers’ results are reliable. Thus, contrary to whatLatour and Woolgar suggest, epistemological explanations are sometimesindispensable.
There are two problem with the constructivists’ model. First, it disreg- ards the fact that scientists working in a laboratory together are moved byboth sorts of considerations, epistemological and social. Second, it disreg-ards the fact that the scientists are both individual agents and members ofa plural subject. Indeed, Latour and Woolgar admit that “a myriad of dif-ferent types of interests and preoccupations are intermeshed in scientists’ discussions” (Latour and Woolgar 1986, 166). But, an adequate explana-tion of scientists’ interactions will require an observer to distinguish bothwhich sort of consideration is salient at each moment, and in what capacityone is acting, be it as an individual or as a member of a plural subject. Tothe insiders, the technicians and scientists, the moves between these twotypes of considerations and two types of roles may be second nature. In oneinstance a scientist considers the epistemic merits of a rival hypothesis, inthe next, she is considering the laboratory team’s commitment to a partic-ular line of research, willing to sacrifice the realization of the one set ofgoals in an effort to realize the other set.
Both rationalists and social constructivists are concerned with only one type of consideration, epistemological or social, in their explanation ofscientific success. And neither recognize the agency of plural subjects.
Gilbert’s notions of collective agency and collective acceptance, though,provide a means by which both types of considerations can be recognizedas relevant to an understanding of the behaviour of scientists. In this way,Gilbert has provided the means for us to develop a more adequate modelof laboratory life.
I began this paper by explaining Gilbert’s account of the phenomenonwhich she calls “collective belief”. I have argued that there are significantdifferences between belief and acceptance, differences that are relevantto Gilbert’s account. When a plural subject adopts a view it is a mistaketo think that the group has a belief. If we fail to recognize this, we areapt to be perplexed by the behaviour of people when they are involved inplural subject relationships. The plural subject will frequently be movedto change its view for what seem to be irrelevant reasons. These reasons,though, may seem relevant, as they often will be, if we realize that they arenot intended to show that the collectively held view is false. Rather, oftentimes, a plural subject will change its view because it has been shown thatthe collectively held view is contrary to the realization of the group’s goals.
And people seeking to change the views of such groups may sometimes doso, not by showing that the collectively held view is false, but by showingthat it is contrary to the goals of the group. Knowing this will make under-standing people’s behaviour easier. In particular, I have tried to show howsuch a realization may help us produce a reasonable peace agreement inthe “science wars”.
I would like to thank Margaret Gilbert for the valuable feedback she hasprovided in the series of exchanges we have had since I first met her atthe World Congress of Philosophy in Boston in August 1998. I thank thereferees for Synthese for their insightful reports. I would also like to thankLori Nash, Bob Ware and John Nicholas for providing critical feedback onnumerous drafts. In addition, I thank each of the following for their feed-back on an earlier draft: Kathleen Okruhlik, Alison Wylie, Cheryl Misak,Bruce Freed, Martin Kreiswirth, and Sergio Sismondo. Finally, I thank myaudience at the Philosophy Department Colloquium at the University ofCalgary when I presented an earlier version of the paper there.
1 Gilbert notes that over time she has developed a preference for this last way of puttingthe point (Gilbert 1996, 7–15).
2 One is reminded of Hilary Putnam’s (1975) remarks in “The Meaning of ‘Meaning’ ”,that steamships (and languages), in contrast to hammers, are the sorts of tools that requirea great number of people to work. If Gilbert is correct about this notion of collectivebelief, then it seems that such beliefs underlie our operation of every type of tool that,like a steamship, requires a number of people to work. Similarly, John Hardwig discussesa particular article published in Physical Review Letters, “Charm Photoproduction CrossSection at 20 GeV”, that has 99 coauthors and that “approximately 50 physicists workedperhaps 50 man/years collecting the data for the experiment” (Hardwig 1985, 357). PaulThagard (1997) provides additional examples of what he calls “collaborative knowledge”,knowledge that is only made possible by the collaborative efforts of numerous scientists(Thagard 1997, 255–8).
3 Though Gilbert believes that common knowledge is not sufficient for “collective belief”,she does suggest that it is a necessary condition (Gilbert 1989, 202).
4 Both Austen Clark (1994) and Raimo Tuomela (1992) seem to disregard the distinctionbetween belief and acceptance in their accounts of group belief, though more recentlyTuomela (1995) does seem to acknowledge that the distinction does have some bearing onthe issue.
5 If the terms of folk psychology are going to figure in a scientific understanding of humancognition we will certainly need to make the sorts of distinctions that I am recommendingwe make here.
6 Conditional proofs in deductive logical systems are reconstructions of arguments of thissort.
7 Thus, Cohen, unlike Stalnaker, does not believe that acceptance is a species of belief.
8 This is true also in cases in which an individual accepts a view. If one wishes to persuadea scientist that she should not accept a particular theory, it is not enough to tell the scientistthat the theory is false. In fact, Ronald Giere has argued that theories are not even aptlydescribed as true or false (see Giere 1988). In order to persuade a scientist that she shouldnot accept a particular theory, one must show that accepting the theory is contrary to her aims. So, for example, an economist might say “I know that the theory I accept presupposesthat consumers are utility maximizers, which they are not; but the theory is predictivelysuperior to competitor theories that make no such assumption, and my goals are betterserved by a theory that is predictively superior”.
9 Cohen (1992) seems to acknowledge this in his brief discussion of “The Purposes ofOrganizations and Artifacts”.
10 Philip Kitcher’s (1993) Advancement of Science is an attempt to rehabilitate the ration-alists’ account of scientific success, giving reasons a significant role to play in resolvingdisputes in science. He does concede, though, that epistemological considerations are notthe only type of considerations that move scientists and contribute to success in science.
Clark, A.: 1994, ‘Beliefs and Desires Incorporated’, The Journal of Philosophy XCI(8),
Cohen, L. J.: 1989, ‘Belief and Acceptance’, Mind XCVIII(391), 367–389.
Cohen, L. J.: 1992, An Essay on Belief and Acceptance, Clarendon Press, Oxford.
Churchland, P.: 1981, ‘Eliminative Materialism and the Propositional Attitudes’, The
Journal of Philosophy LXXVIII(2), 67–90.
Giere, R.: 1988, Explaining Science: A Cognitive Approach, University of Chicago Press, Gilbert, M.: 1987, ‘Modelling Collective Belief’, Synthese 73(1), 185–204.
Gilbert, M.: 1989, On Social Facts, Princeton University Press, Princeton.
Gilbert, M.: 1994, ‘Remarks on Collective Belief’, in F. Schmitt (ed.), Socializing Epistem-
ology: The Social Dimensions of Knowledge, Rowman and Littlefield, Lanham, MD, pp.
111–134.
Gilbert, M.: 1996, Living Together: Rationality, Sociality, and Obligation, Rowman and Gilbert, M.: 2000, Collective Belief and Scientific Change, in Sociality and Responsibility: New Essays on Plural Subject Theory, Rowman and Littlefield, Lanham, MD.
Goldman, A.: 1999, Knowledge in a Social World, Clarendon Press, Oxford.
Hardwig, J.: 1985, ‘Epistemic Dependence’, The Journal of Philosophy 82(7), 335–349.
Hull, D.: 1988, Science as a Process, University of Chicago Press, Chicago.
Kitcher, P.: 1993, The Advancement of Science, Oxford University Press, Oxford.
Latour, B. and S. Woolgar: 1986, Laboratory Life: The Construction of Social Facts,
second edition, Princeton University Press, Princeton.
Putnam, H.: 1975, ‘The Meaning of “Meaning”’, in Mind, Language and Reality, Cambridge University Press, Cambridge, pp. 215–271.
Quinton, A.: 1975, ‘Social Objects’, Proceedings of the Aristotelian Society 76, 1–27.
Schmitt, F.: 1994, ‘The Justification of Group Belief’, in F. Schmitt (ed.), Socializing
Epistemology: The Social Dimensions of Knowledge, Rowman and Littlefield, Lanham,MD, pp. 257–287.
Solomon, M.: 1994, ‘Social Empiricism’, Nous 28(3), 325–343.
Stalnaker, R.: 1984, Inquiry, The MIT Press, Cambridge, MA.
Thagard, P.: 1997, ‘Collaborative Knowledge’, Nous 31(2), 242–261.
Tuomela, R.: 1992, ‘Group Beliefs’, Synthese 91, 285–318.
Tuomela, R.: 1995, The Importance of Us: A Philosophical Study of Basic Social Notions,
Stanford University Press, Stanford.
Van Fraassen, B.: 1980, The Scientific Image, Clarendon Press, Oxford.
Department of Philosophy,University of British Columbia,VancouverCanadaE-mail: kbwray@interchange.ubc.ca

Source: http://simonbarker.is/Collective%20Belief%20And%20Acceptance%20-%20K.%20Brad%20Wray.pdf