23 Los Angeles Times, March 18, 2002.
24 G. Wills, "With God on His Side," New York Times Magazine, March 30, 2003.
25 M. Rees, Our Final Hour (New York: Basic Books, 2003), 61.
26 Questions of their plausibility aside, the mutual incompatibility of our religious beliefs renders them suspect in principle. As Bertrand Russell observed, even if we were to grant that one of our religions must be correct in its every particular, given the number of conflicting views on offer, every believer should expect damnation on mere probabilistic grounds.
27 Rees, Our Final Hour, has given our species no better than a 50 percent chance of surviving this century. While his prognostications are nothing more than educated guesswork, they are worth taking seriously. The man is not a crank.
Chapter 2 The Nature of Belief
1 Proof of this fact is never so eloquent as when injury to the brain destroys one facet of a person's memory while sparing the others-and indeed, it is largely upon such clinical case histories (like W. B. Scoville and B. Milner, "Loss of Recent Memory after Bilateral Hippocampal Lesions," Journal of Neurology, Neurosurgery and Psychiatry 20 (1957): 11-21) that our understanding of human memory depends. Long-term memory has since fragmented into semantic, episodic, procedural, and other forms of information processing; and short-term memory (generally called "working memory") is now subdivided into phonological, visual, spatial, conceptual, echoic, and central executive components. Our analysis of both forms of memory is surely incomplete. The distinction between semantic and episodic memory, for instance, doesn't seem to hold for topographical recall (E. A. Maguire et al., "Recalling Routes around London: Activation of the Right Hippocampus in Taxi Drivers," journal of Neuroscience 17 [1997]: 7103-10); and semantic memory seems susceptible to further division into category-specific subtypes, as in memory for living v. nonliving things (S. L. Thompson-Schill et al., "A Neural Basis for Category and Modality Specificity of Semantic Knowledge," Neuropsychologia 37 [1999]: 671-76; J. R. Hart et al., "Category-Specific Naming Deficit following Cerebral Infarction," Nature 316 [Aug. 1, 1985]: 439-40).
2 There are ways of construing the concept of "belief" that make it appear equally disjoint. If we use the term too loosely, it can seem that the entire brain is intimately involved in "belief" formation. Imagine, for instance, that a man has come to your door claiming to represent the "Publishers Clearing House Sweepstakes":
1. You see the man's face, recognize it, and therefore "believe" that you know who this person is. Activity in your fusiform cortex, especially in the right hemisphere, is crucial for such recognition to occur, and a lesion here will lead to prosopagnosia (the inability to recognize familiar faces, or indeed to see faces as faces at all). Using "belief" in this context, it is tempting to say that prosopagnosics have lost certain "beliefs" about what other people look like.
2. Having recognized the man's face, you form the "belief," based on your long-term memory for both faces and facts that he is Ed McMahon, the famous spokesman for Publishers Clearing House. Damage to your perirhinal and perihippocampal cortices would have prevented this "belief" from forming. See R. R. Davies et al., "The Human Perirhinal Cortex in Semantic Memory: An in Vivo and Postmortem Volumetric Magnetic Resonance Imaging Study in Semantic Dementia, Alzheimer's Disease and Matched Controls," Neuropathology and Applied Neurobiology 28, no. 2 (2002): 167-78 [abstract], and A. R. Giovagnoli et al., "Preserved Semantic Access in Global Amnesia and Hippocampal Damage," Clinical Neuropsychology 15 (2001): 508-15 [abstract].
3. Not yet being sure whether this is a hoax of some sort (perhaps Mr. McMahon is now working for Candid Camera) you take another moment to study the man at your door. You form the "belief," based on his tone of voice, the look in his eye, and many other factors, that he is trustworthy and therefore means what he says. Your ability to form such judgments reliably-in particular, your ability to detect untrustworthiness-requires that you have at least one functioning amygdala (R. Adolphs et al., "The Human Amygdala in Social Judgment," Nature 393 [June 4, 1998]: 470-74), a small, almond-shaped nucleus in your medial temporal lobe.
4. Mr. McMahon then informs you that you are the lucky winner of a "big jackpot." Your memory for words (requiring different processing from your memory for faces) leads you to "believe" that you have won some money, rather than a "pot" of some sort. Making sense of this phrase will require the work of your superior and middle temporal gyri, predominantly in your left hemisphere. See A. Ahmad et al., "Auditory Comprehension of Language in Young Children: Neural Networks Identified with fMRI," Neurology 60 (2003): 1598-605, and M. H. Davis and I. S. Johnsrude, "Hierarchical Processing in Spoken Language Comprehension," Journal of Neuroscience 23 (2003): 3423-
5. Ed then produces a piece of paper, which he invites you to read. He does this by pointing. Your "belief" that he wants you to read requires what has come to be called "theory of mind" processing on your part (D. Premack and G. Woodruff, "Does the Chimpanzee Have a Theory of Mind," Behavioral and Brain Sciences 1 (1978): 515-26)-if a tree limb had swayed in the direction of a piece of paper, you would not have understood it as "pointing" at all. The anatomy underlying theory of mind processing is not entirely clear at present, but it seems that the anterior cingulate cortex as well as regions of the frontal and temporal lobes enable to you to attribute mental states (including beliefs) to others. See K. Vogeley et al., "Mind Reading: Neural Mechanisms of Theory of Mind and Self-perspective," Neurolmage 14 (2001): 170-81; C. D. Frith and U. Frith, "Interacting Minds-A Biological Basis," Science's Compass 286 (1999): 1692-95; and P. C. Fletcher et al., "Other Mind in the Brain: A Functional Imaging Study of 'Theory of Mind' in Story Comprehension," Cognition 57 (1995): 109-28.
6. Scanning the paper with your eyes, you see the following symbols appended after your name: $10,000,000. Some processing relative to Arabic numerals (probably in your left parietal lobe-G. Denes and M. Signorini, "Door But Not Four and 4 a Category Specific Transcoding Deficit in a Pure Acalculic Patient," Cortex 37, no. 2 [2001]:
267-77) leads you to "believe" that this paper is actually a check for ten million dollars.
While many diverse streams of neural activity have conspired to make you believe that you have won a terrific sum of money it is this idea-explicitly represented in language-that underwrites the sweeping changes that will take place in your nervous system, and in your life. Perhaps you will startle the benevolent Mr. McMahon by shrieking; you may even burst into tears; it is only a matter of hours before you begin shopping with an unusual degree of abandon. Your belief that you have just won ten million dollars will be the author of all these actions, both voluntary and involuntary. In particular, it will dictate the following behavior: to the question "Have you just won ten million dollars?" you will-if moved by the spirit of candor-reply yes.
3 Belief, in this sense, is what philosophers generally call a "propositional attitude." We have many such attitudes, in fact, and they are usually indicated by a clause containing the word "that"; we can believe that, fear that, intend that, appreciate that, hope that, etc.
4 The formation of certain primitive beliefs may be indistinguishable from the preparation of a motor plan. See J. I. Gold and M. N. Shadlen, "Representation of a Perceptual Decision in Developing Oculomotor Commands," Nature 404 (March 23, 2000): 390-94, and "Banburismus and the Brain: Decoding the Relationship between Sensory Stimuli, Decisions, and Reward," Neuron 36, no. 2 (2002): 299-308, for a discussion of visual judgments and oculomotor response.
5 We do not have to bring the membership of Al Qaeda "to justice" merely because of what happened on Sept. n, 2001. The thousands of men, women, and children who disappeared in the rubble of the World Trade Center are beyond our help-and successful acts of retribution, however satisfying they may be to some people, will not change this fact. Our subsequent actions in Afghanistan and elsewhere are justified because of what will happen to more innocent people if members of Al Qaeda are allowed to go on living by the light of their peculiar beli
efs. The horror of Sept. 11 should motivate us, not because it provides us with a grievance that we now must avenge, but because it proves beyond any possibility of doubt that certain twenty-first-century Muslims actually believe the most dangerous and implausible tenets of their faith.
6 A consideration of the structure of our language reveals that this is not a special case, since all words and their usages lead us in circles of mutual explanation.
7 The philosopher Donald Davidson has made this insight do some very heavy lifting in his work on "radical interpretation." One interesting consequence of the relationship between belief and meaning is that any attempt to understand a language user requires that we assume him to be basically rational (this is Davidson's "principle of charity").
8 At least at the "classical" scale at which we live. That the quantum world does not behave in this way accounts for why no one can claim to "understand" it in realistic terms.
9 D. Kahneman and A. Tversky, "On the Reality of Cognitive Illusions," Psychological Review 103 (1996): 582-91; G. Gigerenzer, "On Narrow Norms and Vague Heuristics: A Reply to Kahneman and Tversky," ibid., 592-96; K. J. Holyoak and P. C. Cheng, "Pragmatic Reasoning with a Point of View," Thinking and Reasoning 1 (1995): 289-313; J. R. Anderson, "The New Theoretical Framework," in The Adaptive Character of Thought (Hillsdale, N.J.: Erlbaum, 1990); K. Peng and R. E. Nisbett, "Culture, Dialectics, and Reasoning about Contradiction," American Psychologist 54 (1999): 741-54; K. E. Stanovich and R. F. West, "Individual Differences in Rational Thought," Journal of Experimental Psychology: General 127 (1998): 161.
10 A. R. Mele, "Real Self-Deception," Behavioral and Brain Sciences 20 (1997): 91-102, "Understanding and Explaining Real Self-Deception," ibid., 127-36, and Self-Deception Unmasked (Princeton: Princeton Univ. Press, 2001); H. Fingarette, Self-Deception (Berkeley: Univ. of California Press, 2000); J. P. Dupuy, ed., Self-Deception and Paradoxes of Rationality (Stanford: CSLI Publications, 1998); D. Davidson, "Who Is Fooled?" ibid.; G. Quattrone and A. Tversky, "Self-Deception and the Voter's Illusion," in The Multiple Self, ed. J. Elster (Cambridge: Cambridge Univ. Press, 1985), 35-57.
11 This assumes that many of the beliefs have common terms, as the beliefs of human beings invariably do.
12 This example is taken from W. Poundstone, Labyrinths of Reason: Paradox, Puzzles, and the Frailty of Knowledge (New York: Anchor Press, 1988), 183-88.
13 Recently, physical theories have been advanced that predict quantum computation across an infinite number of parallel universes (D. Deutsch, The Fabric of Reality [New York: Penguin, 1997]) or the possibility that all matter will one day be organized as an "omniscient" supercomputer
(F. Tipler, The Physics of Immortality [New York: Doubleday, 1995]) availing itself of a dilation of space-time resulting from the gravitational collapse of the universe. I have excluded these and other theoretical hierophanies from the present discussion.
14 Another way of getting at these logical and semantic constraints is to say that our beliefs must be systematic, Systematicity is a property that beliefs inherit from language, logic, and the world at large. Just as most words derive their sense from the existence of other words, every belief requires many others to situate it in a person's overall representation of the world. How the loom of cognition first begins weaving is still a mystery, but there seems little doubt that we come hardwired with a variety of proto-linguistic, proto-doxastic (from the Greek doxa, "belief") capacities that enable us to begin interpreting the tumult of the senses as regularities in the environment and in ourselves. We do not learn a language by memorizing a list of unrelated phrases, and we do not form a view of the world by adopting a string of unconnected beliefs. For a discussion of the systematicity of language, see J. A. Fodor and Z. W. Pylyshyn, "Systematicity of Cognitive Representation," excerpt from "Connectionism and Cognitive Architecture," in Connections and Symbols, ed. S. Pinker and J. Mehler (Cambridge: MIT Press, 1988). A belief must be knitted together with other beliefs for it to be a belief about anything at all. (I have left aside, for the moment, whether there exist beliefs that do not rely upon any others to derive their meaning. Whether or not such atomic beliefs exist, it is clear that most of our beliefs are not of this sort.)
The systematicity of logic seems guaranteed by the following fact: if a given proposition is "true," any proposition (or chain of reasoning) that contradicts it must be "false." Such a requirement seems to mirror the disposition of objects in the world, and therefore places logical constraints upon our behavior. If a statement like "The cookies are in the cupboard" is believed, it will become a principle of action-which is to say that when I desire cookies, I will seek them in the cupboard. In the face of such a belief, a contradictory claim like "The cupboard is bare" will be seen as hostile to my forming a behavioral plan. Confident cookie-seeking behavior requires that my beliefs have a certain logical relationship.
15 S. Pinker, The Blank Slate (New York: Viking, 2002), p. 33.
16 There is a point of contact between my remarks here and the "mental models" account of reasoning developed by P. N. Johnson-Laird and R. M. J. Byrne, Deduction (Hillsdale, N.J.: Erlbaum, 1991), chaps. 5-6, I would note, however, that our mental models of objects in the world behave as they do because objects do likewise. See L. Rips, "Deduction and Cognition," in An Invitation to Cognitive Science: Thinking, ed. E. E. Smith and D. N. Osherson (Cambridge: MIT Press, 1995), 297-343, for doubts about whether a concept like AND could be learned at all.
17 Of course, we can think of examples where certain of our words run afoul of ordinary logic. For instance, one cannot put the shadow of an apple and the shadow of an orange in Jack's lunchbox, close the lid, and then expect to retrieve one or the other at the end of the day.
18 Another property of belief follows directly from the nature of language: just as there is no limit to the number of sentences a person can potentially speak (language is often said to be "productive" in this sense), there is no limit to the number of beliefs he can potentially form about the world. Because I now believe that there is no owl in my closet, I also believe that there are not two owls there, or three ... ad infinitum.
19 Most neuroscientists believe that we have somewhere on the order of io"-io12 neurons, each of which makes an average of 104 connections with its neighbors. We therefore have something like io'5 or 1016 individual synapses. It's a big number, but it's still finite.
20 Following N. Block, "The Mind as the Software of the Brain," in An Invitation to Cognitive Science: Thinking, ed. E. E. Smith and D. N. Osherson (Cambridge: MIT Press, 1995), 377-425.
21 D. J. Simons et al., "Evidence for Preserved Representations in Change Blindness," Consciousness and Cognition 11, no. 1 (2002): 78-97; M. Niemeier et al., "A Bayesian Approach to Change Blindness," Annals of the New York Academy of Sciences 956 (2002); 474-75 [abstract].
22 R. Kurzweil, The Age of Spiritual Machines (New York: Penguin, 1999).
23 Consider a mathematical belief like 2 + 2=4. Not only do most of us believe this proposition; this belief seems to be antecedently true of us in every present moment. We do not appear to construct it as the occasion warrants, rather it is by virtue of such rudimentary beliefs that we construct others. But what about a belief like 865762 + 2 = 865764? Most of us will have never considered this sum before, and we will believe it only by virtue of constructing it according to the laws of arithmetic. And yet, doing so, we can cash it out just as we do the proposition 2 + 2 = 4. Is there any difference between these two mathematical beliefs? In phenomenological terms there surely is. You will notice, for instance, that you cannot easily speak (or think) the longer sum, while two plus two equals four comes to mind almost reflexively. As far as our basic epistemic commitments are concerned, however, these beliefs are equally
"true." In fact, all of us stake our lives on the validity of far more complicated (and therefore less transparent) mathematical propositions every time we board an airplane or cross a bridge. At bottom, most of us believe that an operation li
ke addition is truth preserving, in that it can be repeated over and over, and with arbitrarily large values, and still yield a true result. But the question remains, how can we know that our belief that 2 + 2 = 4 isn't constructed anew each time we use it? How, in other words, do we know that we believe it antecedently? If we are tempted to say that this belief is always newly constructed, we must ask, constructed with what? The rules of addition? It seems doubtful that a person could know that he was successfully practicing addition unless he already believed that 2 + 2 = 4. It seems just as certain, however, that you did not wake up this morning believing that eight hundred and sixty-five thousand, seven hundred and sixty-two, plus two, equals eight hundred and sixty-five thousand, seven hundred and sixty-four. To really exist inside your brain, this belief must be constructed, in the present moment, on the basis of your prior belief that two plus two equals four. Clearly, many beliefs are like this. We may not, in fact, believe most of what we believe about the world until we say we do.
24 See D. T. Gilbert et al., "Unbelieving the Unbelievable: Some Problems in the Rejection of False Information," journal of Personality and Social Psychology 59 (1990): 601-13; D. T. Gilbert, "How Mental Systems Believe," American Psychologist 46, no. 2 (1991): 107-19.
25 This explains why beliefs that are accidentally true do not constitute knowledge, even when they are justified. As the philosopher Edmund Gettier observed long ago, we may believe something to be true (e.g., I may think the time is exactly 12:31 a.m.), we may believe it for good reasons (I am currently looking at a clock that reads 12:31 a.m.), and our belief may be true (it really is 12:31 a.m.), but we may not be in a state of knowledge about the world (because, in the present instance, the clock is broken and shows the correct time only by accident). While there are many philosophical niceties to be explored here, the basic fact is that for our beliefs to be truly representative of the world, they must stand in the right relationship to the world.