BEHAVIORAL AND BRAIN SCIENCES (2009) 32, 493–561 doi:10.1017/S0140525X09990975 The evolution of misbelief Ryan T. McKay Institute for Empirical Research in Economics, University of Zurich, Zurich 8006, Switzerland; and Centre for Anthropology and Mind, University of Oxford, Oxford OX2 6PE, United Kingdom [email protected] http://homepage.mac.com/ryantmckay/ Daniel C. Dennett The Center for Cognitive Studies, Tufts University, Medford, MA 02155-7059 [email protected] http://ase.tufts.edu/cogstud/incbios/dennettd/dennettd.htm Abstract: From an evolutionary standpoint, a default presumption is that true beliefs are adaptive and misbeliefs maladaptive. But if humans are biologically engineered to appraise the world accurately and to form true beliefs, how are we to explain the routine exceptions to this rule? How can we account for mistaken beliefs, bizarre delusions, and instances of self-deception? We explore this question in some detail. We begin by articulating a distinction between two general types of misbelief: those resulting from a breakdown in the normal functioning of the belief formation system (e.g., delusions) and those arising in the normal course of that system’s operations (e.g., beliefs based on incomplete or inaccurate information). The former are instances of biological dysfunction or pathology, reflecting “culpable” limitations of evolutionary design. Although the latter category includes undesirable (but tolerable) by-products of “forgivably” limited design, our quarry is a contentious subclass of this category: misbeliefs best conceived as design features. Such misbeliefs, unlike occasional lucky falsehoods, would have been systematically adaptive in the evolutionary past. Such misbeliefs, furthermore, would not be reducible to judicious – but doxastically1 noncommittal – action policies. Finally, such misbeliefs would have been adaptive in themselves, constituting more than mere by-products of adaptively biased misbelief- producing systems. We explore a range of potential candidates for evolved misbelief, and conclude that, of those surveyed, only positive illusions meet our criteria. Keywords: adaptive; belief; delusions; design; evolution; misbelief; positive illusions; religion; self-deception 1. Introduction A misbelief is simply a false belief, or at least a belief that is RYAN T. MCKAY is a Research Fellow at the University not correct in all particulars. We can see this metaphori- of Oxford, UK. He was educated at the University of Western Australia (B.Sc. Hons. in Psychology) and at cally: If truth is a kind of target that we launch our Macquarie University (MClinPsych, Ph.D.) in beliefs at, then misbeliefs are to some extent wide of the Sydney, Australia. His research interests include cogni- mark. Of course, there is no philosophical consensus tive neuropsychiatry, evolutionary psychology, and be- about just what a belief actually is. In what follows we havioural economics. He has held previous intend to avoid this question, but we offer here the follow- postdoctoral positions in Boston (Tufts University), ing working definition of belief, general enough to cover Belfast (Queen’s University), and Zu¨ rich (University most representationalist and dispositional accounts: A of Zu¨ rich). He has also previously worked as a clinical belief is a functional state of an organism that implements neuropsychologist at the National Hospital for Neurol- or embodies that organism’s endorsement of a particular ogy and Neurosurgery in London and as a lecturer in state of affairs as actual.2 A misbelief, then, is a belief psychology at Charles Sturt University in Australia. Recently, he has been conducting experimental investi- that to some degree departs from actuality – that is, it is gations in the cognitive science of religion. a functional state endorsing a particular state of affairs that happens not to obtain. DANIEL C. DENNETT is University Professor, Fletcher A prevailing assumption is that beliefs that maximise the Professor of Philosophy, and Co-Director of the Center survival of the believer will be those that best approximate for Cognitive Studies at Tufts University. He is the reality (Dennett 1971; 1987; Fodor 1983; 1986; Millikan author of Consciousness Explained (1991), Darwin’s 1984a; 1984b; 1993). Humans are thus assumed to have Dangerous Idea (1995) and Breaking the Spell: Religion been biologically engineered to form true beliefs – by as a Natural Phenomenon (2006), as well as other books evolution. On this assumption, our beliefs about the and articles in philosophy of mind, cognitive science, world are essentially tools that enable us to act effectively and evolutionary theory. He is also the author or co- author of three target articles (1983, 1988, 1992) and in the world. Moreover, to be reliable, such tools must 32 commentaries in Behavioral and Brain Sciences be produced in us, it is assumed, by systems designed and was an Associate Editor of the journal for many (by evolution) to be truth-aiming, and hence (barring years. miracles) these systems must be designed to generate # Cambridge University Press, 2010 0140-525X/09 $40.00 493 McKay & Dennett: The evolution of misbelief grounded beliefs (a system for generating ungrounded but one (codified in warranty legislation) between “culpable mostly true beliefs would be an oracle, as impossible as a design limitations” (malfunctions) and “forgivable design perpetual motion machine). Grounded beliefs are simply limitations/features.” When a given artifact fails to beliefs that are appropriately founded on evidence and perform a particular task, this failure is always due to a existing beliefs; Bayes’ theorem (Bayes 1763) specifies limitation in the design of that artifact. The question is the optimal procedure for revising prior beliefs in the whether the design limitation concerned is – from the light of new evidence (assuming that veridical belief is designer’s perspective – a blameworthy, “culpable” limit- the goal, and given unlimited time and computational ation (a design flaw, or perhaps a flaw in the execution resources; see Gigerenzer & Goldstein 1996). Of course, of the design), or whether it is a tolerable, “forgivable” just as we can have good grounds for believing prop- limitation. Examples of the former (with respect to the ositions that turn out to be false, so can ungrounded arbitrary task of “keeping time”) include: beliefs be serendipitously true (others arguably lack 1. My $20,000 Bolex watch loses 10 seconds every day truth values). To keep our exposition manageable, we (contra the advertisement). will not consider such ungrounded beliefs to be misbeliefs, 2. My cheap Schmasio watch loses 10 minutes every although we acknowledge that false and (serendipitously) day (contra the advertisement). true ungrounded beliefs (and perhaps those lacking Examples of the latter limitation include: truth values) may well be produced in much the same 1. My toaster does not keep time at all. way – and by much the same types of mechanism (we 2. My Bolex loses a second every day (within warranted return to this issue in sect. 14). limits). If evolution has designed us to appraise the world accu- 3. My cheap Schmasio loses a minute every day (within rately and to form true beliefs, how are we to account for warranted limits). the routine exceptions to this rule – instances of misbe- 4. After putting it in a very hot oven for an hour, my lief? Most of us at times believe propositions that end up Bolex does not keep time at all. being disproved; many of us produce beliefs that others What we can see from these examples is that manufac- consider obviously false to begin with; and some of us tured artifacts either work as intended (within a tolerable form beliefs that are not just manifestly but bizarrely margin of error), or they don’t work as intended (falling false. How can this be? Are all these misbeliefs just acci- outside the tolerable margin). What’s important is not dents, instances of pathology or breakdown, or at best how well the artifacts objectively work, but how well undesirable (but tolerable) by-products? Might some of they work relative to how they were intended to work them, contra the default presumption, be adaptive in (and the intentions of the manufacturer will bear upon and of themselves?3 the advertised claims of the manufacturer). The Bolex Before we can answer that, we must develop a tentative and Schmasio examples reflect this, because the malfunc- taxonomy of misbelief. We begin with a distinction tioning Bolex still works objectively better than the prop- between two general types: those that result from some erly functioning Schmasio. kind of break in the normal functioning of the belief for- Of course, some apparent design limitations are in fact mation system and those that arise in the normal course deliberate design features. To cite a single example, con- of that system’s operations. We take this to represent the temporary consumers are frequently frustrated by DVD orthodox, albeit unarticulated, view of misbelief. Part region code restrictions. The fact that a region 1 DVD and parcel of this orthodox view is that irrespective of player (sold in North America) cannot play discs sold in whether misbeliefs arise out of the normal or abnormal Europe or Japan (region 2) is certainly, from the consumer’s operation of the belief formation system, the misbeliefs perspective at least,4 a limitation in the design of that DVD themselves are maladaptive. player – and often a frustrating limitation. In our terminol- Our aim in this target article is to evaluate this claim. We ogy, however, the limitation is forgivable because such will proceed by a process of elimination, considering and players are not designed to play DVDs from other disqualifying various candidates until we arrive at what regions, and indeed are deliberately designed not to do we argue are bona fide instances of adaptive misbelief. so. Region restrictions are, as software designers often Some candidates will prove not to be directly adaptive; say, “not a bug but a feature” of such machines, ostensibly others may be false but not demonstrably so; and still to safeguard copyright and film distribution rights.