Rev.Phil.Psych. (2010) 1:483–498 DOI 10.1007/s13164-009-0020-5

Folk Epistemology as Normative Social Cognition Benoit Hardy-Vallée & Benoît Dubreuil

Published online: 15 December 2009 # Springer Science+Business Media B.V. 2009

Abstract Research on folk epistemology usually takes place within one of two different paradigms. The first is centered on epistemic theories or, in other words, the way people think about knowledge. The second is centered on epistemic intuitions, that is, the way people intuitively distinguish knowledge from belief. In this paper, we argue that insufficient attention has been paid to the connection between the two paradigms, as well as to the mechanisms that underlie the use of both epistemic intuitions and theories. We contend that research on folk epistemology must examine the use of both intuitions and theories in the pragmatic context of the game of giving and asking for reasons and, more generally, understand how these practices take place within the broader context of normative social cognition.

1 Folk Epistemology: Theories and Intuitions Epistemology tout court is a philosophical inquiry into the nature of knowledge (as a product) and knowing (as a process that produces knowledge). One of the epistemologist’s goals is to separate the epistemic wheat from the doxastic chaff, knowledge from mere beliefs; another is to describe the very nature of knowledge: its features, conditions, sources, justification and limits. Folk epistemology, by contrast, refers to our ordinary, commonsensical, everyday, naïve and intuitive conceptions of knowledge. As Kitchener puts it, folk epistemology consists of “our ‘untutored’ views about the nature of knowledge” (R. F. Kitchener 2002, p. 89).

B. Hardy-Vallée Decision Support Services, SBR Global, Toronto, Canada e-mail: [email protected] B. Dubreuil (*) Department of philosophy, Université du Québec à Montréal, Case postale 8888, succursale Centre-ville, Montréal, QC H3C 3P8, Canada e-mail: [email protected]

484

B. Hardy-Vallée, B. Dubreuil

Research on folk epistemology falls into two broad, sometimes overlapping, paradigms. One concerns what we might call epistemic theories and the other epistemic intuitions. Research on the former seeks to elucidate how people think, reason and represent knowledge (a field often referred to as “personal epistemology”) (Hofer and Pintrich 2002). Subjects are asked to explicate their beliefs about knowledge, its source or its justification. By contrast, research on the latter seeks to probe folk intuitions in particular cases. Instead of being asked about their beliefs as to what knowledge is in general, subjects are asked to decide whether a character in a scenario knows or merely believes something (Nagel 2007). In this paper, we argue that research on folk epistemology must take place within the broader context of research on normative social cognition. By this, we mean that folk epistemology must be conceived as a phenomenon that is produced by the cognitive machinery that underlies the more general capacities to understand intentional norm following, as well as to follow norms of action and reasoning in the context of everyday social interactions. Section 1 presents the two main research paradigms on folk epistemology, the first focused on epistemic intuitions and the second on epistemic theories. Section 2 suggests that insufficient attention has been paid to the connection between the two paradigms, as well as to the mechanisms that underlie the use of both epistemic intuitions and theories. In Section 3, we contend that research on folk epistemology must examine the use of both intuitions and theories in the pragmatic context of the game of giving and asking for reasons and, more generally, understand these practices as resulting from our normative and social cognitive abilities. Finally, in Section 4, we argue that models that have been developed to describe folkpsychological explanations of intentional action can be applied to the practice of folk epistemology as well, further strengthening the case for studying folk epistemology within the broader context of normative social cognition. 1.1 Epistemic Theories Folk epistemic theories have been studied mostly by cognitive, developmental, and educational psychologists (Bendixen and Rule 2004; Hofer 2001; Hofer and Pintrich 2002; Kuhn 2001; Schommer-Aikins 2004). Starting with Piaget (1950) and Perry (1970), this line of research has focused on epistemological folk beliefs and theories, and especially on their developmental progression and individual variation. For instance, King and Kitchener suggested that epistemological development goes through seven stages divided into three levels: prereflective thinking (stages 1–3), quasi-reflective thinking (stages 4–5), and reflective thinking (stages 6–7). Knowledge is first seen as certain, definite and transmitted from authority figures; then as something constructed and uncertain; and finally construed as testable and refutable, fallible but based on evidence (King and Kitchener 2004). Again, vocabulary and definitions are not standardized across the literature, but the development of folk epistemology is generally believed to follow stages analogous to Piaget’s logical and Kohlberg’s moral stages. Beyond development, other psychologists are interested in epistemological styles: Deanna Kuhn, for instance, showed how Absolutists, Multiplists and Evaluativists— categories roughly similar to King and Kitchener’s stages, but in a synchronic

Folk Epistemology as Normative Social Cognition

485

fashion—give different answers to a question about experts. Discussing the reasons why prisoners become repeat offenders, Kuhn asked her subjects: “Do experts know for sure what the cause is?” Typical replies were: Absolutist: If they’re experts, they know. Multiplist: I don’t think anybody knows for sure really, because there really isn’t one right answer... Evaluative: Well, I think they’re close. I mean, nothing’s for sure, but I’m sure they have good ideas about why people fail. (Kuhn 1992, p. 169) Other studies addressed how education or academic specialization impact on folk epistemologies. Students in “hard” sciences (engineering and natural science) and “soft” sciences (social sciences and humanities) differ in their beliefs about whether knowledge is a collection of facts, whether it is something certain and unchanging and whether experts and teachers are reliable sources of knowledge. Social sciences and humanities students are more inclined toward a conception of knowledge as atomistic, uncertain, and changing and of experts and teachers as reliable sources of knowledge, while students in engineering and the natural sciences are more inclined toward the opposite (Palmer and Marra 2004). In the same vein, other studies examine variations related to gender and culture (Belenky 1986; Karabenick and Moosa 2005; Unger 1992). Finally, research indicates that individuals’ stated epistemic theories match their way of managing conflicting evidence. The more subjects believe that knowledge is complex, tentative and organized (a more-or-less pragmatist conception), the more easily they adopt multiple perspectives in everyday controversies as well as in academic contexts, revise their convictions and withhold their judgment until more information is available (Schommer-Aikins and Hutter 2002; Whitmire 2004). 1.2 Epistemic Intuitions Epistemic intuitions have been studied by cognitive psychologists and experimental philosophers (Alexander and Weinberg 2007; Nagel 2007; Weinberg et al. 2001, 2003). Examining a whole range of variations in perception, cognition, decisionmaking and judgment between Easterners and Westerners, Nisbett and his collaborators suggested that “tacit epistemology” (another name for epistemic intuitions) varies with culture (Nisbett et al. 2001). For instance, Easterners accommodate contradiction, change, and pluralities of points of view more easily than Westerners, who are more prone to categorical and essentialist judgments. Following this lead, Weinberg, Nichols and Stich tested whether culture influences philosophical intuitions about “Gettier cases” (Weinberg et al. 2001). In a wellknown essay, Edmund Gettier (1963) presented apparent counter-examples to the traditional philosophical definition of knowledge as “justified true belief.” He devised examples that suggested that someone may be justified in believing that P, where P is a true belief, and yet not really know that P. Weinberg, Nichols and Stich used a Gettier-like example to see how Westerners and Easterners construe the belief/knowledge distinction. Bob has a friend, Jill, who has driven a Buick for many years. Bob therefore thinks that Jill drives an American car. He is not aware, however, that her

486

B. Hardy-Vallée, B. Dubreuil

Buick has recently been stolen, and he is also not aware that Jill has replaced it with a Pontiac, which is a different kind of American car. Does Bob really know that Jill drives an American car, or does he only believe it? Bob is justified in believing that “Jill drives an American car,” and he holds the true belief that “Jill drives an American car,” but it seems to many philosophers that Bob does not really know it. Studies found that this standard philosophical answer was common among Westerners, while a majority of Easterners held that Bob really knows. Other experiments by Weinberg, Nichols and Stich reveal that socioeconomic status also influences folk epistemological judgment. The only scenario in which a broad agreement was found is the following: Dave likes to play a game with flipping a coin. He sometimes gets a “special feeling” that the next flip will come out heads. When he gets this “special feeling”, he is right about half the time, and wrong about half the time. Just before the next flip, Dave gets that “special feeling”, and the feeling leads him to believe that the coin will land heads. He flips the coin, and it does land heads. Did Dave really know that the coin was going to land heads, or did he only believe it? Regardless of their socio-economic status or cultural background, a striking majority of subjects responded that Dave only believed it. Weinberg, Nichols and Stich speculate that such intuitions might represent a universal core of folk epistemology. In the next section, we highlight some theoretical blind spots in the study of folk psychology, that is, questions that have been neglected by both paradigms. Our argument is not intended as a criticism of earlier studies, but rather as a constructive contribution to the development of a sound model of folk epistemology. 2 Epistemic Blind Spots People use epistemic theories and intuitions to justify or assess the truth of claims such as “X knows that P”. But the distinction between epistemic intuitions and theories, although widely accepted, raises important questions that we would like to address in this section. First, what is the relationship between epistemic intuitions and epistemic theories? Second, what kind of cognitive mechanisms do folk epistemological capacities require? Third, what is someone doing when she uses her folk epistemological capacities, that is, what is folk epistemology about? Tentative answers to these questions will set the stage for our characterization of folk epistemological capacities in the next section. 2.1 The Relationship Between Epistemic Intuitions and Epistemic Theories The two lines of research on epistemic theories and intuitions developed rather independently and were motivated by different goals. Empirical research into epistemic theories was motivated primarily by pedagogical concerns: how do individual conceptions of knowledge develop and how can they be influenced by teaching and education? By contrast, empirical research on epistemic intuitions—

Folk Epistemology as Normative Social Cognition

487

promoted mainly by experimental philosophers—aimed at testing philosophical intuitions in order to identify universal features of folk epistemology. Since both paradigms address the nature and functioning of folk epistemology, there must be some connections between them, just as our moral intuitions and our moral theories must be somehow related. On the one hand, the study of epistemic theories generally relies on epistemic intuitions. Indeed, to determine how people think about knowledge, we generally need to know what they think that they (or another person) know or believe in the first place. Similarly, students of moral judgment in the Kohlbergian line typically ask subjects whether a particular behavior is morally good (i.e., they appeal to intuitions) before asking why it is so (i.e., before asking for theories). On the other hand, the study of folk epistemology might reveal interesting disjunctions between epistemic theories and epistemic intuitions. In the domain of moral cognition, it has been observed that people sometimes hold strong feelings about the rightness or wrongness of certain acts, while they are unable to explain exactly why they feel that way—a phenomenon known as “moral dumbfounding” (Haidt and Hersh 2001). Similarly, people might be confounded if asked to explain why a person saying “I have a headache” knows that she has a headache (Wright 2000). In this case as in others, we should not expect epistemic intuitions to be predictive of epistemic theories. To make sense of the connection between epistemic intuitions and theories, it might be useful to think of folk epistemology as having many dimensions (Schommer 1994). Three are particularly relevant here, because they figure in many “non-folk” (i.e., philosophical) theories of knowledge. Epistemic theories and conceptions of knowledge can be distinguished by their positions regarding: 1- The source of knowledge: God, experts, authority, science, society, language, perception, understanding, reason, intuition, introspection, etc., can all be seen (or not) as reliable sources of knowledge. 2- The robustness of knowledge: Knowledge can be (at one extreme) eternal, absolute and universal or (at the other extreme) changing, local, and relative to a perspective. 3- The organization of knowledge: Knowledge can be compartmentalized (atomistic) or integrated (holistic). We can use these three dimensions to construct an epistemological space in which epistemic theories can be plotted. This conceptual space encompasses many different folk epistemological theories, from the children who see knowledge as absolute and transmitted by teachers to the social science university student who sees knowledge as subjective and constructed by ideologies. Moreover, one can see Descartes, Nietzsche, or Quine as thinkers who explored, secured, or optimized certain regions of the epistemic space: Quine put forth the holistic structure of justification, Nietzsche advocated the context-relativity of knowledge while Descartes made a strong case for introspection. In standard cases, we first recognize intuitively certain information as knowledge and then our intuitions license the articulation of theories within an appropriate region of the epistemological space. Research on epistemic intuitions has mostly addressed the issue of how subjects distinguish belief from knowledge, and can thus provide an indication as to when or how epistemic theories are solicited. To be sure,

488

B. Hardy-Vallée, B. Dubreuil

epistemic theories are also likely to intervene in a top-down process to revise or influence our intuitions: What one takes as a reasonable standard for accepting that something is true should affect when and whether a new assertion is accepted and hence the likelihood of belief revision and conceptual change. (Kuhn 2001, p. 1) If you believe that science is a reliable source of knowledge, the mere fact of hearing “a scientific study says that P” might be enough to trigger the intuition that P is knowledge. By contrast, the same assertion might trigger different intuitions in a person with a complete distrust of science. 2.2 The Mechanisms of Folk Epistemological Capacities Psychologists and experimental philosophers conducted instructive research on how people represent knowledge. Nevertheless, most of what we know about folk epistemology, at least in the paradigms of epistemic theories and intuitions, is still restricted to people’s beliefs. Studies on epistemic intuitions deal with people’s beliefs about what people know (“Did Dave really know or only believe?”), and studies on epistemic theories with their beliefs about knowledge itself (“Do you think experts really know? Why?”). Therefore, to draw an analogy with research on folk psychology, research is mainly focused on external, rather than internal, folk epistemology (Stich and Ravenscroft 1994). Epistemic intuitions and theories are external accounts of folk epistemology because they simply systematize folk conceptions and do not propose mechanisms of epistemic categorization or inference. Stich and Ravenscroft (1994, p. 465) define external accounts of folk psychology as: (1) the “set of folk psychological ‘platitudes’ that people readily recognize and assent to” and (2) “a theory that systematizes the folk psychological platitudes in a perspicuous way.” Folk epistemology in our view analyses the same kind of platitudes. Studies examine (1) how people assent to epistemic intuitions or (2) how epistemic theories systematize intuitions in a perspicuous way. Incidentally, this external account provides a sort of unity to the two paradigms, because epistemic intuitions and theories are both examine in the same language-game. External accounts of folk knowledge, however, do not specify the mechanisms underlying folk epistemology. In cognitive science, there have been many discussions concerning whether folk psychology is implemented by an inferential-theoretical mechanism or by a simulation mechanism, and whether folk-psychological mechanisms are modular or integrated, domain-specific or domain-general (see Stueber 2006, chapter 3). These problems do not concern common generalizations, intuitions or platitudes, but rather the processes whose outputs are folk-psychological platitudes. The quest for mechanisms, by contrast, has not been at the center of research on folk epistemology. Schommer-Aikins (2004) has provided an account of the relationship between theoretical beliefs in her “embedded systemic model” of epistemological beliefs (Fig. 1).

Folk Epistemology as Normative Social Cognition

489

Fig. 1 From Schommer-Aikins 2004

There is a sense in which the model represents a mechanism, because it fleshes out the interactions between beliefs systems, learning, and so on. Nevertheless, it does not count as a cognitive mechanism, since it does not specify how epistemic theories and intuitions result from cognitive operations. 2.3 Folk Epistemology as Metacognition? A plausible suggestion is that folk epistemic theories and intuitions result from metacognitive processes (Hofer 2004; K. S. Kitchener 1983; Nagel 2007; Pintrich et al. 2000). The problem with the view, however, is that there are many ways to conceive and define such processes. Generally speaking, metacognition refers to any cognition about cognition (Flavell 1979), but more precise accounts have currency. A standard account of metacognition defines it as a cognitive process realized through two interacting subsystems: monitoring and control (Koriat et al. 2006; Nelson and Narens 1990, 1994; Proust 2007). In this view, the monitoring system tracks cognitive states and processes. The stereotypical example of such metacognitive monitoring is the tip-of-the-tongue phenomenon: one can remember what those black flat-bottomed boats used on Venetian canals look like, know that one usually knows their name, but remain unable to recall the word “gondola.” The tipof-the-tongue feeling is a variety of the “feeling of knowing” (Koriat 2000): the intuitive feeling that we know that P before recalling P. Other metacognitive monitoring processes track our estimation of learning or memory: how confident we are in our mastering of a skill, our capacity to find information, etc. The metacognitive process initiates, maintains or ends epistemic actions, such as memory search: once we feel that we know something, we allocate our attention to cues and past memories to find it. Thus some epistemic intuitions, at least those about our knowledge of something, may be generated through metacognitive mechanisms. It is not clear, however, how exactly metacognition, at least in this sense, can instruct mechanistic analyses of folk epistemology besides the assessment of our own epistemic process. Proust (2007, p. 291), for instance, presents metacognition as a self-directed

490

B. Hardy-Vallée, B. Dubreuil

evaluation of one’s own mental processes: its function is to “predict, retrodict and evaluate one’s own mental dispositions, states and properties.” It fixes intuitive beliefs about whether we know (or could remember, etc.) that P, but it is hard to explain how these epistemic feelings are invoked when we wonder if Bob really knows that Jill drives an American car or only believes it. A first possibility is that we rely on someone else’s metacognitive feeling, the “feeling of another’s knowing” (Brennan and Williams 1995). This concept, nonetheless, would imply adopting a different definition of metacognition. In the view defended by Proust (2007), metacognition involves a causal contiguity: monitoring and control processes interact through a closed causal loop. Control processes use monitoring to evaluate their commands, while monitoring informs control processes about ongoing cognitive activity (for hypotheses about neural implementation, see MacDonald et al. 2000). The feedback from the monitoring to the control process is internal, not external; in other words, there is no need to look at the world in order to have the feeling of knowing: the feeling is arrived at through simulation or imagination. Second, metacognition requires procedural reflexivity: the feeling of knowing, for instance, results form the presence of a process that tracks another process, the “tracking” being a direct, non-inferential relation. Hence while cognition is directed toward the external world, metacognition is directed toward cognition itself: Metacognition, when it is present, draws on a kind of information that is not delivered by the problem situation, but by the subject’s own procedural selfknowledge. For that reason, metacognition can deal with novel decisions, while well-practiced routines remain within the scope of cognition (where external cues can be used as predictors). (Proust 2007, p. 284) The “feeling of another’s knowing” is therefore a red herring. The term refers to situations where prosody or facial expression indicates that another person is confident or not in her claim. These feelings do not result from the access to this person’s procedural self-knowledge, but merely from an inference. We cannot monitor, nonmetaphorically, the feelings of uncertainty of another person in the same way that we monitor our own because the feeling is directed at external, environmental cues. Moreover, it is also hard to see how we could use this feeling in Gettier cases: we don’t have a cue that indicates whether Bob displays signs of confidence. One solution might be that we simulate his perspective, “put ourselves in his shoes” and use the output of this process to conclude whether Bob knows or only believes. Such simulations may involve metacognition in the sense presented above. But would it only involve metacognition? Wouldn’t it involve something else? The answer to this question depends here again on how one defines metacognition. In our view, a useful distinction must be drawn between metacognition and metarepresentation (see Proust 2007 for a detailed analysis; we follow broadly her account). Metarepresentations are representations of mental representations (e.g. Peter believes that the Earth is flat), linguistic representations (e.g. John said that the Earth is flat) or more abstract representations (“The hypothesis that the Earth is flat is absurd”). As mentioned above, metacognition involves causal contiguity and procedural reflexivity. Metarepresentation does not: my believing that John said something is not oriented toward my own cognitive processes and procedural knowledge. Metacognition is

Folk Epistemology as Normative Social Cognition

491

inherently self-predicative and self-evaluative. Metarepresentation is not. Finally, we can have recursive metarepresentations (I believe that she believes that we believe, etc.), but not recursive metacognition: there is no such thing as feeling that I feel that I feel that I know something. I either feel it (to different degrees) or not. An alternative view, defended among others by Carruthers (2009), equates metacognition with first-person metarepresentations of one’s own mental states. There is nothing inherently wrong with this terminological choice, except that it leaves us with no concept to refer to the feeling of knowing that plays a central role in our epistemic intuitions but that does not involve a representation of our own mental states. If we accept Proust’s (2007) distinction between metarepresentation and metacognition, it becomes clear however that the later can hardly elucidate all our epistemic intuitions, and especially intuitions that result form reasoning about mental states and abstract representations. What about epistemic theories? In studies on folk epistemology, some authors conflate metacognition with metarepresentations and other types of mental processes. For instance, Hofer (2004) proposes a metacognitive account of epistemic cognition but neglects metacognitive monitoring and focuses instead on beliefs and metacognitive control processes. In her model of epistemic cognition, she distinguishes epistemic metacognitive knowledge (beliefs about the nature of knowledge: e.g. certainty), metacognitive judgments and monitoring (beliefs about the nature of knowing: e.g. justification) and metacognitive self-regulation and control of cognition (the regulation of cognition during knowledge construction, e.g. deciding to learn more about something). In this model, there is no “feeling of knowing” or other metacognitive phenomena, but mostly reflective beliefs. In the terminology proposed here, a belief about knowledge, such as “knowledge is absolute,” involves metarepresentation, not metacognition. There is a genuine metacognitive process in self-regulation and control, but this only indicates that once folk epistemological judgment is effective, metacognitive processes are recruited for the implementation of a solution to a learning problem. The distinction between metarepresentation and metacognition might already shed some light on the cognitive mechanisms underlying folk epistemology. Nevertheless, we are still far from a clear internal account of folk epistemology. In order to articulate one and before doing so, we suggest that we to take a step back and ask: what is the domain of folk epistemology? What is one doing when she uses her folk epistemological capacities? It is assumed that folk biology is about living beings, folk psychology about intentional agents, folk physics about physical objects. What is folk epistemology about?

3 Folk Epistemology and the Pragmatic Turn A recurrent feature of studies on folk epistemology is their focus on concepts and beliefs. Experimenters ask their subjects to make decisions that reveal their beliefs or ask them directly about their beliefs. Consequently, students of folk epistemology routinely use questionnaires, surveys, interviews, and “think aloud” protocols. They elicit folk epistemological faculties in explicit and conscious problem solving. But

492

B. Hardy-Vallée, B. Dubreuil

how, why and when, outside the lab do we use our epistemic intuitions and theories? Here are three examples of how they intervene in everyday contexts: (1) You argue with your colleague whether the evidence was sufficient in O.J. Simpson’s trial: “Even if the glove did not fit, it does not mean that he is not guilty.” (2) At the grocery store, the clerk says that the white chocolate cookies you were looking for have been out of stock for several days. “How could it be, you say, I saw someone walking out of here with a box?” The clerk replies that you have confused the donut box that the other customer bought with the white chocolate cookies box; they have the same appearance. (3) At home, you spouse, partner or roommate says that it is your turn to wash the dishes. “No, my dear. Today is Monday, and on Mondays it is your turn” In each case, there is a question as to whether something is true or justified. We take it to be the case that what we call folk epistemology is at play in these examples. As Dewey argues, epistemological concerns arise only when we have to evaluate something: Truth and falsity present themselves as significant facts only in situations in which specific meanings and their already experienced fulfillments and nonfulfillments are intentionally compared and contrasted with reference to the question of the worth, as to reliability of meaning, of the given meaning or class of meanings (Dewey 1910, p. 95). But Dewey neglects to mention that intentional comparisons and contrasts are primarily situated in social contexts. We primarily assess the truth and validity of assertions and we do it in the context of what Sellars (1956) and Brandom (1994) called the “Game of Giving and Asking for Reasons” (GGAR): the dialogical, linguistic interactions in which humans ask and reply to the question “why”. When someone says something, we may not only agree or disagree; we may also cite reasons as to why we agree or disagree and require our interlocutor to provide us with more reasons. Many of our greatest institutions are formalized version of the GGAR: a tribunal, a scientific conference and a parliament are structured and formalized GGARs. We make claims, justify them by reasons (proof, evidence, studies, report, commissions, etc.) and expect others to provide reasons for their claims in turn. Note that we do not advocate all of Brandom’s account of meaning or normativity: we only acknowledge, as he does, the central role in human interactions of the GGAR. What we are suggesting here is a reversal in the order of explanation. The Platonist order of explanation, as Brandom (1994: 22) calls it, starts with concepts and then explains their use in terms of their content. Traditionally, epistemology first figures out what the concept of knowledge is, then determine when someone applies it correctly. Studies on folk epistemology allow us to adopt another methodology: to start from people’s intuitions (hence judgments) and theories (an inferentially articulated network of judgments) to understand their concept of KNOWLEDGE. What we put forth here is a pragmatic order of explanation, a naturalistic version of Brandom’s: starting with the use to explain the content. We should not first explain the content of the concept of KNOWLEDGE and then determine its use, but rather start with explaining the use of the KNOWLEDGE concept and then infer its content. Everyday epistemological problems that arise in

Folk Epistemology as Normative Social Cognition

493

various professional contexts (courts, universities, politics, administration) concern the justification of assertions: how can one say that X is guilty, that experiment Y proves that P, that legislation X decreases crime, or that policy Z facilitates interdepartmental communication, without having any reasons to support one’s claim? In making a claim, we commit ourselves to the consequences of the claim. If we say “the watch is red”, we are also committed to “the watch is colored”. A competent English speaker will expect us to implicitly hold the second claim as true, and to hold “the watch is blue” as false. This social-inferentialist perspective neglects, of course, the epistemology of perception that played a prominent role in the history of epistemology. There are situations where we wonder if our claims are true or justified when we perceive something: if my foot looks bigger than the sun, does it follow that it is so? (Aristotle 1986, pp. 428b 426–427) If all the perceptual qualities of a piece of wax change after being heated, is it still a piece of wax? How would I know it? (Descartes 1998, Meditation 2). But the social-inferentialist perspective correctly emphasizes the social context in which most of our epistemic judgments take place. We depend upon each other for our education and security, we live in groups (from hunter-gatherer bands to modern states), and we learn about everything we have to learn from others either directly (through linguistic communication) or indirectly (through reading, books, tv, radio). Even the epistemology of perception has a social dimension: outside of philosophical circles, people rarely wonder whether their perception is reliable when they are alone: the epistemological problems raised by perception usually take place in a GGAR. It is when we argue over the justification of an assertion that we discuss the validity of perceptions (or, for philosophers, it is when we teach or write a book, which are also social-communicative contexts). Thus epistemology, as Goldman writes, “must come to grips with the social interactions that both brighten and threaten the prospect for knowledge” (Goldman 1999, p. vii). It should also, we would like to add, come to grips with the cognitive underpinnings of these social interactions. The suggestion we would like to put forth is simply this: folk epistemology is part of broader normative social cognition. In other words, it is part of the multiple behaviors and activities that result from the cognitive machinery that underlies compliance—and expectations about others’ compliance—with norms of action and reasoning. 3.1 Folk Epistemology as Normative Social Cognition The ideas advanced in the last section suggest that folk epistemology is about assertions. Whatever the actual mechanisms are, it would be productive to see them as primarily directed at assertions, and primarily involved in evaluating the justifications for these assertions. In other words, if internal folk psychology is a set of mechanisms that produces attributions of mental states, internal folk epistemology is a set of mechanisms that evaluates assertions. It is employed for detecting unjustified claims in the GGAR. Its natural domain is thus the production of declarative sentences, at least if we see folk epistemology as a product of the evolutionary history of our species. As Sperber (2001), Cosmides and Tooby (2000) and Sterelny (2006) argued, humans are hyper-social animals that are prone to learn from their conspecifics. This is often useful, but it comes with the risk of sometimes being cheated. There are

494

B. Hardy-Vallée, B. Dubreuil

good reasons to think that our abilities for evaluating assertions are part of an adaptive response to the cost of misinformation and deception in human social environment. In communication, there is an advantage for the communicator to be trusted, and for the addressee to have true information. Much of our institutions (political, scientific, etc.) make sure that communication takes place in a truthful/ trusting environment. This is because, without such external constraints, these advantages do not necessarily accrue. The communicator might benefit from being trusted even if she is untruthful (in which case it is disadvantageous for the addressee). Words are cheap, and credulous addressees risk a lot: an agent believing anything could be easily exploited. Linguistic communication shows no reliable signs of truth and is costless to produce. Thus, trust can be adaptive in social settings, but without a capacity to suspend your judgment and ask (to yourself or to someone else) if an assertion is justified, it can be harmful (see also on this point Sperber 2001 and Mercier in this issue). This capacity, and all the processes that support it, is folk epistemology. At the affective and cognitive level, our capacity to enter the GGAR will depend on the mechanisms that allow us to follow social norms (Heath 2008). These high and low-level mechanisms underlie the processing of gaze direction, intentional action, and other social events. At the neural level, they are realized thanks to widely distributed networks that include subcortical structures such as the amygdala and the striatum, as well as several areas of the prefrontal cortex and the temporal and parietal cortices (Gobbini et al. 2007; Nummenmaa and Calder. 2009). Folk epistemological judgments also rely on the executive functions of the brain, without which we would be unable to hold in mind and compare multiple hypotheses (or counterfactuals). Relevant executive functions include the capacity to recognize rules, to detect conflicts, to inhibit prepotent responses and to direct attention to alternatives. They are realized in the prefrontal cortex, the anterior cingulated cortex and the posterior temporal cortex and inferior parietal cortex (Barbey et al. 2009; Bunge and Wright 2007; Davidson et al. 2006). They develop late in infancy, along with children’s capacity to learn more complex patterns of thought and to overcome automatic or habitual patterns, abilities that are essential to engage in the game of giving and asking for reasons.

4 The Epistemology of Action We said above that folk epistemology is about assertions, and that folk epistemological mechanisms evaluate the justification of claims. Put this way, it sounds like these mechanisms manipulate essentially linguistic representations. Sperber (2001), for instance, suggest that we evaluate claims by coherence checking: we see whether an assertion is internally coherent or if it is coherent with other assertions. Brandom’s (1994: 141–198) “deontic scorekeeping” is a similar process whereby one sees what an assertion implies, and what is implied by an assertion. Thus, the evaluation of assertions is done by checking inferential relations between propositions. Cosmides and Tooby (2000) posit also a set of inferential procedures that tag information structures as true or false, or assess various aspects of internally represented propositions (source, scope, etc). Although all these suggestions are on

Folk Epistemology as Normative Social Cognition

495

the right track, we think that they miss an important feature of the evaluation of assertions: since an assertion is an action, evaluating an assertion is roughly equivalent to evaluating an action. Hence our last suggestion to further the understanding of folk epistemology is to organize our knowledge of folk epistemological capacities around the inferential mechanisms deployed in the folk theory of intentional actions. Bertram F. Malle and other social psychologists who investigate the folk concept of intentionality have produced a unified model of the representation of intentional action. Their studies suggest that the conceptual framework is relatively similar from one individual to another (Knobe 2006; Malle 2001, 2007). Almost everybody agrees whether an action is intentional or not: subjects rely preferentially on causes to explain the unintentional actions and reasons to explain intentional ones. About 70% of intentional actions are explained by primary reasons: beliefs and desires, mainly, but also values (e.g. “she got home late because she liked the show”). When primary reasons are not evoked, subjects use either a causal history of reasons explanation or an enabling-factor explanation. The first explains why a person decided to do X not by reference to her beliefs/desires, but to factors that bring about reasons to act: for instance, “she comes from a culture in which respect is highly valued.” The enabling-factor explanations cite the condition that made its performance possible without referring to the agent’s intentions or motivations (e.g. “she had two weeks to prepare her talk”). In sum, the folk concept of intentional action is a system of inferences based on reasons, causal reason histories and enabling factors organized as in the following graph. If we accept that assertions are also actions, we can include them in Malle’s (2007) schema of possible explanations of intentional behavior (Fig. 2). This hypothesis is not incompatible with coherence-checking or deontic scorekeeping: it is rather a refinement of these ideas. The suggestion, in terms of cognitive processing, is that folk epistemological mechanisms recruit those social cognitive mechanisms involved in the evaluation of actions. From this viewpoint, epistemic evaluations are positive (an assertion is seen as justified) when the whole causal chain from the history of reasons to the production of assertions is seen as correct, and negative (an assertion is seen as unjustified) when the causal chain is incorrect in at least one respect. Thus, an assertion can be seen as false or unjustified for many reasons: 1. Causal history. The speaker might have been exposed to false or unreliable information earlier in her life. 2. Reasons. The speaker might entertain unwarranted or incoherent claims, or be plainly irrational.

Fig. 2 Adapted from Malle 2007

496

B. Hardy-Vallée, B. Dubreuil

3. Intentions. The speaker might be dishonest, deceptive or manipulative. 4. Enabling factors. The speaker might have defective cognitive (inferential, mnemonic or others) capacities. All epistemic virtue (such as truthfulness, honesty, rationality, reliability or logical soundness of assertions) refers therefore to the normal operation of the causal chain. This model of internal folk epistemology might represent the basic features of epistemic evaluations. A plausible hypothesis is that the model can be redeployed in other domains where the assertion in not an utterance (e.g. writings, email), or even when the evaluated person does not explicitly make an assertion but can be interpreted as being disposed to assert something. Since spoken language predates written language, it is plausible that this schema is redeployed in recent evolutionary contexts.

5 Conclusion The model can further help to bridge the gap between the two research paradigms in folk epistemology: the first dedicated to epistemic theories and the second to epistemic intuitions. The schema represents our intuitive categorization of epistemic and non-epistemic states: we perceive an appropriate causal chain in the first case, an inappropriate one in the second. Westerners see Gettier cases as inappropriate causal chains to acquire a disposition to say “I know that P.” Easterners and Westerners both agree that one does not really know the results of one’s flipping a coin because of one’s “special feeling,” because about every aspect of the causal chain is broken (for a related argument, see also Spicer, this issue). Second, the schema is refined and analyzed when we attempt to make explicit what is knowledge and where it comes from. Different theories of knowledge present different settings of the schema: in Kuhn’s experiment (see Section 2), the Absolutist’s conviction that experts have absolute knowledge is a representation of expertise as an appropriate (and non-negotiable) causal history and/or enabling factor. By constrast, the Multiplist’s convictions that there is no right answer stems from her representation of expertise as an insufficient causal history and/or enabling factor. Of course, this model does not propose a complete picture of our folk epistemological capacities. A more complete explanation of folk epistemology could include, among other things, a description of the mechanisms thanks to which knowledge is assessed evolved in the human lineage and how they are realized in the brain. In this sense, a satisfying account of folk epistemology must take place within the more general context of research on the natural foundations of normative social cognition. Acknowledgements BHV’s work was supported by a grant from the Social Science and Humanities Research Council of Canada (SSHRC) and BD’s work by the Fonds québécois de recherche sur la société et la culture (FQRSC). We thank Joseph Heath for having us realize the importance of understanding epistemology within the context of pragmatics and normativity, as well as Christophe Heintz, Chad Horne, and an anonymous reviewer for comments on a previous version of the article.

Folk Epistemology as Normative Social Cognition

497

References Alexander, J., and J.M. Weinberg. 2007. Analytic epistemology and experimental philosophy. Philosophy Compass 2(1): 56–80. Aristotle. 1986. On the soul (H. Lawson-Tancred, Trans.). Harmondsworth, Middlesex: Penguin Books. Barbey, A.K., F. Krueger, & J. Grafman. 2009. An evolutionarily adaptive neural architecture for social reasoning. Trends in Neurosciences 32(12): 603–610. Belenky, M.F. 1986. Women’s ways of knowing: The development of self, voice, and mind. New York: Basic Books. Bendixen, L.D., and D.C. Rule. 2004. An integrative approach to personal epistemology: a guiding model. Educational Psychologist 39(1): 69–80. Brandom, R. 1994. Making it explicit: Reasoning, representing, and discursive commitment. Cambridge: Harvard University Press. Brennan, S.E., and M. Williams. 1995. The feeling of another’s knowing: prosody and filled pauses as cues to listeners about the metacognitive states of speakers. Journal of Memory and Language 34(3): 383–398. Bunge, S.A., and S.B. Wright. 2007. Neurodevelopmental changes in working memory and cognitive control. Current Opinion in Neurobiology 17: 1–8. Carruthers, P. 2009. How we know our own minds: the relationship between mindreading and metacognition. Behavioral and Brain Sciences 32: 121–182. Cosmides, L., and J. Tooby. 2000. Consider the source: The evolution of adaptations for decoupling and metarepresentation. In Metarepresentations: A multidisciplinary perspective, ed. D. Sperber, 53–115. Oxford: Oxford University Press. Davidson, M.C., D. Amso, L.C. Anderson, and A. Diamond. 2006. Development of cognitive control and executive functions from 4 to 13 years: evidence from manipulations of memory, inhibition, and task switching. Neuropsychologia 44(11): 2037–2078. Descartes, R. 1998. Discourse on method; and, meditations on first philosophy (D.A. Cress, Trans. 4th ed.). Indianapolis: Hackett Pub. Dewey, J. 1910. The influence of Darwin on philosophy, and other essays in contemporary thought. New York: H. Holt and company. Flavell, J.H. 1979. Metacognition and cognitive monitoring: a new area of cognitive-developmental inquiry. American Psychologist 34: 906–911. Gettier, E.L. 1963. Is justified true belief knowledge? Analysis 23(6): 121–123. Gobbini, M.I., A.C. Koralek, R.E. Bryan, K.J. Montgomery, and J.V. Haxby. 2007. Two takes on the social brain: a comparison of theory of mind tasks. Journal of Cognitive Neuroscience 19(11): 1803– 1814. Goldman, A.I. 1999. Knowledge in a social world. Oxford: Oxford University Press. Haidt, J., and M.A. Hersh. 2001. Sexual morality: the cultures and emotions of conservatives and liberals. Journal of Applied Social Psychology 31(1): 191–221. Heath, J. 2008. Following the rules: Practical reasoning and deontic constraint. Oxford: Oxford University Press. Hofer, B.K. 2001. Personal epistemology research: implications for learning and teaching. Educational Psychology Review 13(4): 353–383. Hofer, B.K. 2004. Epistemological understanding as a metacognitive process: thinking aloud during online searching. Educational Psychologist 39(1): 43–55. Hofer, B.K., and P.R. Pintrich. 2002. Personal epistemology: The psychology of beliefs about knowledge and knowing. Mahwah: Erlbaum. Karabenick, S., and S. Moosa. 2005. Culture and personal epistemology: U.S. and Middle Eastern students beliefs about scientific knowledge and knowing. Social Psychology of Education 8: 375– 393. King, P.M., and K.S. Kitchener. 2004. Reflective judgment: theory and research on the development of epistemic assumptions through adulthood. Educational Psychologist 39(1): 5–18. Kitchener, K.S. 1983. Cognition, metacognition, and epistemic cognition. A three-level model of cognitive processing. Human Development 26(4): 222–232. Kitchener, R.F. 2002. Folk epistemology: an introduction. New Ideas in Psychology 20(2–3): 89–105. Knobe, J. 2006. The concept of intentional action: a case study in the uses of folk psychology. Philosophical Studies 130(2): 203–231.

498

B. Hardy-Vallée, B. Dubreuil

Koriat, A. 2000. The feeling of knowing: some metatheoretical implications for consciousness and control. Consciousness and Cognition 9(2): 149–171. Koriat, A., H. Maayan, and R. Nussinson. 2006. The intricate relationships between monitoring and control in metacognition: lessons for the cause-and-effect relation between subjective experience and behavior. Journal of Experimental Psychology: General 135(1): 36–69. Kuhn, D. 1992. Thinking as argument. Harvard Educational Review 62(2): 155–178. Kuhn, D. 2001. How do people know? Psychological Science 12(1): 1–8. MacDonald, A.W., J.D. Cohen, V.A. Stenger, and C.S. Carter. 2000. Dissociating the role of the dorsolateral prefrontal and anterior cingulate cortex in cognitive control. Science 288(5472): 1835– 1838. Malle, B.F. 2001. Folk explanations of intentional action. In Intentions and intentionality: Foundations of social cognition, ed. B.F. Malle, L.J. Moses, and D.A. Baldwin, 265–286. Cambridge: MIT. Malle, B.F. 2007. Attributions as behavior explanations: Toward a new theory. In Current themes and perspectives in social psychology, ed. D. Chadee, and J. Hunter, 3–26. St. Augustine: SOCS, The University of the West Indies. Nagel, J. 2007. Epistemic intuitions. Philosophy Compass 2(6): 792–819. Nelson, T.O., and L. Narens. 1990. Metamemory: A theoretical framework and new findings. In The psychology of learning and motivation, vol. 26, ed. G. Bower, 125–141. New York: Academic. Nelson, T.O., & L. Narens. (1994). Why investigate metacognition? In Meta-cognition: Knowing about knowing, eds. J. Metcalfe and A. Shimamura, 1–25. Nisbett, R.E., K. Peng, I. Choi, and A. Norenzayan. 2001. Culture and systems of thought: holistic versus analytic cognition. Psychological Review 108(2): 291–310. Nummenmaa, L., and A.J. Calder. 2009. Neural mechanisms of social attention. Trends in Cognitive Sciences 13(3): 135–143. Palmer, B., and R.M. Marra. 2004. College student epistemological perspectives across knowledge domains: a proposed grounded theory. Higher Education 47(3): 311–335. Perry, W.G. 1970. Forms of intellectual and ethical development in the college years: A scheme. New York: Holt. Piaget, J. 1950. Introduction à l’épistémologie génétique. Paris: Presses universitaires de France. Pintrich, P.R., C. Wolters, and G. Baxter. 2000. Assessing metacognition and self-regulated learning. In Issues in the measurement of metacognition, ed. G. Schraw, and J.C. Impara, 43–97. Lincoln: Buros Institute of Mental Measurements. Proust, J. 2007. Metacognition and metarepresentation: is a self-directed theory of mind a precondition for metacognition? Synthese 159(2): 271–295. Schommer, M. 1994. Synthesizing epistemological belief research: tentative understandings and provocative confusions. Educational Psychology Review 6(4): 293–319. Schommer-Aikins, M. 2004. Explaining the epistemological belief system: introducing the embedded systemic model and coordinated research approach. Educational Psychologist 39(1): 19–29. Schommer-Aikins, M., and R. Hutter. 2002. Epistemological beliefs and thinking about everyday controversial issues. Journal of Psychology 136(1): 5. Sellars, W. 1956. Empiricism and the philosophy of mind. Cambridge: Harvard University Press. 1997. Sperber, D. 2001. An evolutionary perspective on testimony and argumentation. Philosophical Topics 29: 401–413. Sterelny, K. 2006. Folk logic and animal rationality. In Rational animals? ed. S. Hurley, and M. Nudds, 293–312. Oxford: Oxford University Press. Stich, S.P., and I. Ravenscroft. 1994. What is folk psychology? Cognition 50: 447–468. Stueber, K.R. 2006. Rediscovering empathy: Agency, folk psychology, and the human sciences. Cambridge: MIT. Unger, R.K. 1992. Will the real sex difference please stand up? Feminism Psychology 2(2): 231–238. Weinberg, J.M., S. Nichols, and S.P. Stich. 2001. Normativity and epistemic intuitions. Philosophical Topics 29(1and2): 429–460. Weinberg, J.M., S. Nichols, and S.P. Stich. 2003. Metaskepticism: Meditations in ethno-epistemology. In The skeptics: Contemporary essays, ed. S. Luper, 227–247. Aldershot: Ashgate. Whitmire, E. 2004. The relationship between undergraduates’ epistemological beliefs, reflective judgment, and their information-seeking behavior. Information Processing and Management 40(1): 97–111. Wright, C. 2000. Self-knowledge: The Wittgensteinian legacy. In Knowing our own minds, ed. C. Wright, B. Smith, and C. Macdonald. Oxford: Oxford University Press.