The World Philosophy Made

by Scott Soames

  Nevertheless, certain socially significant results involving voting behavior of individuals and of the actions of individual political actors have been achieved. One derives from Anthony Downs’s An Economic Theory of Democracy. There he notes that because the probability that a single vote will decide an election in today’s societies approaches zero, the rational utility of voting would seem to be tiny, and so, one might think, be easily outweighed by the inconvenience of casting a ballot. The idea is compelling, but it can’t be the whole story. A substantial number of people do vote. Why? Are they irrational, or are there other utilities involved that don’t depend on whether one’s favored candidate wins or loses? Presumably the latter. These utilities may include the sense that it is one’s moral obligation to vote, but they also may include the desirability of being able to say one has voted—and so to signal one’s virtue, to maintain one’s social relationships, and to solidify one’s standing in a group with which one identifies—without the discomfort of lying, or the fear of being found to have lied.

  Of course, accomplishing these goals by going to the polls is one thing; gathering information needed to cast an informed vote is another. Because the costs—in time and effort—of gathering and assessing the information needed to become informed are high, many votes will be cast in ignorance. This is why simplistic branding of parties and candidates, whether accurate or not, is so important in influencing low-information voters. It is also why candidates are tempted to insult and ridicule not only their political rivals, but also the supporters of their rivals, thereby reducing the agent-relative utility of identifying oneself as such a supporter.

  Problems such as these are among the many defects inherent in modern democratic government. In order to combat them one needs, among other things, an objective and open-minded press and educational system with both the ability and the will to widely disseminate accurate information. As the founder of public choice theory Nobel laureate James Buchanan (1919–2013) insisted, one also needs a constitution encouraging consensus, limiting the scope of government and constraining the powers of majorities to impose their will on minorities. By a constitution, he meant a set of fundamental decision rules, intended to last for generations, guiding the procedures by which ordinary political matters are decided—e.g., rules governing elections, rules governing the operation of different branches of government, and rules preventing governments from doing certain things without authorization by supermajorities.21

  In evaluating acceptable constitutional rules, he adopted what might be regarded as an analytical fiction. He insisted that such rules must be capable of being unanimously accepted in a hypothetical, idealized constitutional convention by rational, expected-utility-optimizing agents possessing knowledge of human nature and the world we inhabit, but not knowing what social and economic positions they would come to occupy over time. Realizing that no constitution would guarantee a political system that always maximized social utility while avoiding suppression of minorities, he believed that some constitutions could objectively be shown to be better than others. These are the ones he imagined ideally rational expected-utility-optimizing agents to be able to agree on. Historically existing constitutions were to be judged by their similarity to those that could be so adopted.

  Although Buchanan regarded the identification of defensible constitutional rules to be a matter of highly idealized continuing inquiry, it is clear that his conception of such rules was importantly similar to that of America’s founders. Thus, it is not surprising that he ended his 1986 Nobel Prize lecture with this paragraph.

  In 1987, the United States celebrates the bicentennial anniversary of the constitutional convention that provided the basic rules for the American political order. This convention was one of the very few historical examples in which political rules were deliberately chosen. The vision of politics that informed the thinking of James Madison was not dissimilar, in its essentials, from that which informed Knut Wicksell’s less comprehensive, but more focused, analysis of taxation and spending. Both [Madison and Wicksell] rejected any organic conception of the state as superior in wisdom to the individuals who are its members. Both sought to bring all available scientific analysis to bear in helping to resolve the continuing question of social order: How can we live together in peace, prosperity, and harmony, while retaining our liberties as autonomous individuals who can, and must, create our own values?22

  Having laid the foundations of a subjectivist understanding of probability and incorporated it into a logical model of rational decision and action that can be applied to individual and institutional actors in any domain, twentieth-century philosophers gave their colleagues in the social sciences one of their most powerful tools, the applications of which have only just begun.

  CHAPTER 9

  MIND, BODY, AND COGNITIVE SCIENCE

  The classic mind-body problem; new perspectives on mind, meaning, and representation, the new mind-body debate; understanding, naming, necessity and conceptual possibility; functionalism about the mind; a failed argument against identifying human pains with neuro-physiological events; where we stand; computation, cognitive psychology, and the representational theory of mind.

  THE CLASSIC MIND-BODY PROBLEM

  Descartes famously imagined a scenario in which, despite having all his everyday thoughts, feelings, and sensations, he was really dreaming, or, worse, being deceived by an evil demon. He realized that if that were so, he wouldn’t know most elementary things about himself and the world around him. He wouldn’t even know there was a world around him; nor would he know that he had a physical body, or that anything outside his own thoughts and experiences existed. He would, nevertheless, know he existed, since the very fact that he was having thoughts and experiences would show that he did. So, he reasoned, it must be possible to know that one exists, without knowing that one’s body, one’s brain, or anything physical exists. From this he concluded that he, Descartes, wasn’t identical with (i.e., the very same thing as) any body, any brain, or anything physical.

  Although the argument is not entirely without appeal, the fantastic dreaming scenario may distract from what is really going on. This is easily fixed. Imagine waking up in the dark, unable to move or to see your body or anything else nearby. Your only sensations are of a tiny point of light and a faint sound of music. Although you are able to think perfectly well, you have no idea what happened. In such situation you might know little more than that you must exist, since you have thoughts and experiences. What is it that you know? Not simply that there are thoughts and experiences (of a certain type), or that someone is having them. You know something stronger and more specific than that.

  What do you know? Consider the propositions associated with your use of the sentences “I exist” and “N exists” (Where ‘N’ is your name). These sentences predicate existence of the same person. Since there are no further predications, the propositions they express are representationally identical. Nevertheless, they are different, since you can take different cognitive attitudes toward them. In fact, your use of the two sentences puts three propositions in play—one that requires you to identify yourself in the first-person way that each agent uses to cognize him or herself, one that requires identification via the name ‘N’, and one that doesn’t require one specific way with which you are identified.1 Since the three propositions have the same truth conditions, it is impossible for one to be true without their all being true. But it is possible for some agent or other to believe or know one of the two more demanding propositions without believing or knowing the other one, and it is possible for some agent to believe or know the undemanding proposition without believing or knowing either demanding proposition.

  One more step. Let ‘B’ name your brain, your entire nervous system, or your body—anything physical that might be taken to be a candidate for being you. Keeping B in mind, we return to the scenario in which you awake in the dark, aware only of your thoughts and sensations. You know you exist, but it is h
ard to say whether you know B exists. Although you don’t know the proposition that requires identifying the putative existent via the name ‘B’, you might know the undemanding but representationally identical proposition that imposes no such requirement. Whether or not you do depends on whether or not you are B. If you are, then in knowing that you exist you predicate existence of yourself—i.e., B—and so know that B exists. If you aren’t B, this isn’t so. This observation is fatal to the Cartesian argument. Since one can’t decide whether you know a proposition predicating existence of B without first deciding whether you are B, one can’t use conclusions about what you do, or don’t, know to decide whether you are a purely physical being.

  The point can be made more general. No one would dream of arguing that Mark Twain wasn’t Samuel Clemens just because there is a sense in which some people know (i) that Mark Twain wrote The Adventures of Huckleberry Finn without knowing (ii) that Samuel Clemens wrote The Adventures of Huckleberry Finn. If these two knowledge claims are correct, then the two propositions (i) and (ii)—one known and the other not—must differ, which, in turn, means that names must sometimes contribute more than their referents to propositions expressed by uses of sentences containing them. As we saw in chapter 7, there is a puzzle here to which philosophers have offered solutions, but it is one about thought and language, rather than minds and bodies.

  This dismissal of Descartes’s argument doesn’t show that he was wrong to think that we are spiritual beings, distinct from our bodies. It only shows that his argument provides no reason to believe it. All the classical positions remain open—that we are purely physical, that we are purely nonphysical, or that we are part physical and part nonphysical.

  NEW PERSPECTIVES ON MINDS, MEANING, AND REPRESENTATION

  Although Descartes’s argument for mind/body dualism was unsuccessful, his idea that thought, or—as we might more broadly put it—representation, is the essence of the mental has been taken to heart in the philosophy of the last few decades. Minds represent things in the world, including themselves, in thought and perception, which, in turn, make goal-directed action and linguistic communication possible. As seen in chapter 8, agents act on the basis of their beliefs, trying to bring about changes they desire. As seen in chapter 7, mental concepts like knowledge, belief, and assertion are relations between agents who know, believe, or assert and the things—propositions—known, believed, or asserted.

  Conceiving of propositions as purely representational cognitive acts or operations, we see that minds are the source of both representation and meaning. Propositions—pieces of information or misinformation—represent because they are cognitive acts agents perform in representing things as being one way or another. They are also meanings, or semantic contents, of some sentences. For a sentence S of a language L to mean p (or for uses of S to have the semantic content p) is (roughly) for uses of S by speakers of L to be performances of p. Sentences are linguistic tools for performing representational acts and coordinating one’s representations with others. Those who understand the sentence ‘The earth is round’ use the name to pick out the planet and the predicate to ascribe being round to it. To do this is to perform the act type that is the meaning (semantic content) of the sentence. Since no other cognition is needed, knowing what S means in L doesn’t require having any thoughts about p or L, let alone knowing that S stands in a complicated theoretical relation R to p and L.

  Because language is a social institution, it expands our cognitive horizons. It does so by making it possible for us to believe or disbelieve propositions our cognitive access to which depends on others. For example, I believe Uranus to be a distant planet, despite never having perceived it, seen pictures of it, or had any direct contact with it. I believe the proposition because I picked up the name from others, intending to use it to designate whatever they use it to designate. I have also been told it is a planet far from Earth. Because I both understand the sentence ‘Uranus is a distant planet’ and accept it, I count as believing the proposition it expresses in English. This doesn’t require associating the name with an identifying description that could be substituted for it without changing meaning in sentences in which it occurs. In fact, the main identifying descriptions I have for the name are parasitic on uses of it by others on whom my uses depend. The name, as I use it, refers to whatever they use it to designate. Still, what I believe would have been true, even if the name ‘Uranus’ hadn’t existed. Thus, my belief that Uranus is a distant planet isn’t about the name; it simply predicates being a distant planet a certain object.

  The same story can be told for many proper names of persons, places, and things, as well as for general terms that are, in effect, common names for natural kinds—e.g., bauxite, tungsten, manatee, lemur, and sequoia.2 The propositions expressed by sentences containing such terms represent objects or kinds as having or lacking properties that many agents wouldn’t be able to bring to mind were it not for the chains of communications in which the agents stand to other users of the terms. Since understanding and accepting sentences containing these terms counts as believing the propositions they express, it often turns out that when language is involved, even the solitary thought of an individual agent has a social dimension. Since, in many such cases, we are not able to entertain the propositions sentences express prior to understanding the sentences themselves, it also turns out that understanding a sentence isn’t always a matter of searching through our stock of antecedently entertainable propositions to find which one others in our linguistic community use the sentence to express. It simply involves conforming our linguistic intentions to those of others, whose patterns of use determine the governing linguistic conventions. In recent years, the traditional philosophical mind-body debate has been reformulated in a way that makes it sensitive to these linguistic issues.3

  THE NEW MIND-BODY DEBATE

  In the 1950s and ‘60s a group of prominent philosophers including U. T. Place, J.J.C. Smart, David Lewis, and David Armstrong turned the mind-body debate in a new direction. Instead of employing purely philosophical arguments, like Descartes, to try to establish the truth of one position or another, or contenting themselves with refuting the arguments of those who did, they argued that whether mental states and processes could be identified with physical states and processes was a scientific question. Observing the great progress that had been made in physics, chemistry, and biology, they noted that surprising theoretical identities were already the order of the day.4 For example, they maintained that it had already been discovered that heat is mean molecular kinetic energy, gold is the element with atomic number 79, biological genes are sequences of DNA, lightning is an electrical discharge, the colors of objects are properties of reflecting light waves within precisely defined frequencies, and water is H2O. Because the explanatory power of these advances was a product of the well-established physicalistic framework for investigating the natural world, there was, these philosophers thought, every reason to expect that advances in understanding the mind would be made within the same framework, identifying mental states and processes with physical states and processes. These early philosophical proponents of the psychophysical identity theory thought that pains, sensations, and consciousness itself could, in one way or another, be identified with neurological states and processes. They didn’t specify which neurological states and processes were identical with pain states, sensation states, or consciousness. That was a job for neuroscientists. The philosophical job was to defeat conceptual objections to the possibility that any such identification could be correct, and to articulate the explanatory advantages of incorporating the mental into physical science.

  According to these theorists, identifying a mental type, say pain, with a neurochemical type—call it “C-fiber stimulation”—is conceptually no more problematic than identifying lightning with a type of electrical discharge, heat with mean molecular kinetic energy, or water with H2O. There was, it must be admitted, much to be said for this idea. However, there was
also a catch. These philosophers took it for granted that, like all other theoretical identities in science, statements identifying mental states and processes with physical states and processes are both contingent and knowable only on the basis of empirical evidence. Indeed, the two points were connected. To say that scientific identities are knowable only on the basis of empirical evidence is to say that they can’t be established by philosophical reasoning alone, because observation and evidence are needed to rule out possibilities in which they are false. Since this presupposes that there are such possibilities, it seemed to follow that the identities, though true, could have been false—and so are contingent rather than necessary truths. Although these ideas, which were applied to all science, seemed simple enough, they proved to be more difficult to defend than was initially expected.