The Will: Willpower and Freedom PDF

12 The Will Willpower and Freedom The Problem of Free Will and Determinism Debates over free will are among the most vexing debates in philosophy. 12.1 Here’s the central question: Is...

12 The Will Willpower and Freedom The Problem of Free Will and Determinism Debates over free will are among the most vexing debates in philosophy. 12.1 Here’s the central question: Is the existence of free will compatible with determinism—the thesis that every event, including every human action, is determined, in the sense of being predetermined? Think of some choice that strikes you as chosen freely. Suppose, for 12.2 example, that a person is presented with a choice of two different flavors of ice cream for dessert—chocolate and vanilla. Suppose that this person has enjoyed each flavor in the past, but tonight they decide to order two scoops of vanilla. Now, with this ice cream scenario in mind, let’s do a couple of thought experiments. Thought experiment #1:â ‡ Let us suppose that this person’s brain is under the control of a mind-influencing ray operated by some evil scientist. The scientist causes the person to utter the following words to a waiter at a restaurant: “Please give me two scoops of vanilla ice cream.” Suppose further that the way the scientist causes the person to utter these words is by causing the person to prefer vanilla over chocolate at that moment. Now, here’s the crucial question of the thought experiment: Is the person who is under the control of the evil scientist choosing vanilla of their own free will? Thought experiment #2:â ‡ This second thought experiment is a variation of the first. Let us remove the evil scientist and instead have all of the mental This Is Philosophy of Mind: An Introduction, First Edition. Pete Mandik. © 2014 John Wiley & Sons, Inc. Published 2014 by John Wiley & Sons, Inc.  172â …â … The Will: Willpower and Freedom and physical states of the person ordering ice cream fully predetermined by the decisions of actual ancient Greek gods. Suppose that Zeus and his fellow gods are really real and have made it so that our ice cream order- ing subject was fated (since the beginning of time!) to order vanilla ice cream on this particular day. Again, ask yourself the crucial question: Is the person choosing vanilla of their own free will? 12.3 Many people who contemplate such thought experiments are inclined to answer “no” to the crucial questions posed. Perhaps you are one such person. But let us now consider a third thought experiment. Thought experiment #3:â ‡ This third thought experiment is a variation of the second, but with one major difference. Instead of the person’s mental states and actions being fated by the gods, their mental states and actions are predetermined because of the laws of physics. Suppose that physical- ism is true and that everything is physical. Suppose further, as we dis- cussed in chapter 9, that that every caused event has a physical cause. Again, ask yourself the crucial question: Is the person choosing vanilla of their own free will? 12.4 Some people who contemplate this third thought experiment will answer “no” to the key question. They hold that the kind of determinism described is incompatible with free will. They hold a position known, for obvious reasons, as incompatibilism. Other people who contemplate this third thought experiment will answer “yes.” Such people see no problem in sup- posing both that determinism is true and that free will exists. They hold a position known as compatibilism. 12.5 The philosophical debate between compatibilists and incompatibilists is especially pressing because of the way it connects with the notion of moral responsibility. The idea that people can be morally responsible for their actions plays a central role in our practical, ethical, and legal deliberations. And, arguably, whether someone is morally responsible for something seems to depend on whether they chose that thing of their free will. 12.6 Consider a case in which someone causes the death of a human being. Judgments about whether the killer deserves a punishment, and how severe the punishment should be, depend largely on judgments about whether they freely chose to kill. The killer may receive a less severe punishment or no punishment at all if it’s discovered that their actions were compelled. Perhaps the killing was compelled either by circumstances, such as a gun  The Will: Willpower and Freedomâ …â … 173 to their head, or by internal factors, such as a brain tumor or chemical imbalance. According to many thinkers, free will is necessary for moral responsibility. However, as we’ve already glimpsed in the three thought experiments above, the question of whether any one has free will is argu- ably threatened by the proposition that their actions are determined in the sense of being predetermined. Further, there are powerful arguments in favor of the view that every event is determined in such a way. Sources of Determinism General remarks There are two key components of the idea of determinism. The first is the 12.7 notion of one thing making another thing happen. A typical example involves two events related as cause and effect. The idea of causes coming before effects is closely related to conceptions of determination as prede- termination. The second key component of determinism is the idea that, for any event that happens, it had to happen and could not have been otherwise. The two components of determinism are not wholly independent of 12.8 each other. If it is indeed the case that one event made another happen, then it would seem to follow that given the first event, the second event had to happen and could not have been otherwise. Consider, for argument’s sake, the following contrary supposition: Suppose event e1 happens, and even though e2a happens next, it was still entirely possible that some other event, e2b had happened instead. It seems clear, then, that in such a case, e1 didn’t make e2a happen. Perhaps e2a happened after e1 as a matter of random chance. Or perhaps what really made e2a happen was not e1 by itself but instead e1 in conjunction with some other event. To sum up, then, if an early event makes a later event happen, then, given the first event, the second event had to happen and could not have been otherwise. Thus, arguably, the two key ideas of determinism are not wholly independent of each other. Despite this dependence between the two components of determinism, 12.9 the two components figure in separate conflicts between compatibilists and incompatibilists. The component that has to do with earlier events deter- mining later events arguably conflicts with conceptions of free will in which the agent is the ultimate source of their actions. The component that has  174â …â … The Will: Willpower and Freedom to do with happenings that “could not have been otherwise” arguably con- flicts with conceptions of free will that require free agents to able to do or choose otherwise. 12.10 In the rest of this section we will examine five lines of thought in favor of determinism. These five lines of thought mostly agree on their conclusions. They mainly differ on the kinds of reasoning that lead to their deterministic conclusions. The five lines of thought can be sorted into two groups, groups that differ about how many events are determined. The first group concerns global determinism, the view that all events are determined. The second group concerns local determinism, the view that some restricted class of events is determined. In particular, the local determinisms we will examine focus on those events that can be classified as human actions. Global deter- minisms include physical determinism, theological determinism, and logical determinism. Local determinisms include ethical determinism and psychological determinism. Let’s look at these five lines of deterministic thought in a bit more detail. Physical determinism 12.11 The thesis of physical (or causal or nomological) determinism fits very closely with the kinds of physicalism that we studied in previous chapters, especially the physicalism of chapter 6. If we think of physicalism as the view that everything is either identical to or determined by arrangements of physical particles, and add to it the idea that every event is caused in a way that is fully governed by the laws of physics, then we are very close to the thesis of physical determinism. 12.12 To get a further grasp on the thesis of physical determinism, consider the following way of looking at the unfolding of events in the universe since its very beginning during the Big Bang. Consider the total state of the universe at the present moment. Everything that is happening at the present moment is the causal consequence of what was happening at a previous moment. We can sum this up by saying that the global state of the universe at time t is fully caused or fully causally determined by the global state of the universe at time t-minus-1. One way of summing up the key idea of physical determinism is by saying that any given state of the universe is determined by the previous state of the universe plus the natural laws. Another way of summing up this key idea is to say that every event is determined by a previous event plus the laws of nature.  The Will: Willpower and Freedomâ …â … 175 Most of the empirical support for physical determinism comes from the 12.13 physical sciences and thus makes it a good fit with physicalism. However, strictly speaking physicalism neither entails nor is entailed by physical determinism. It is possible to embrace physicalism (the view that every- thing is physical) while also embracing indeterminism (the view that some events are not determined by prior events and the natural laws). It is also possible to embrace dualism while also holding that mental events are fully determined by previous events and special laws governing the mental. Physical determinism is the variety of determinism that we will mostly 12.14 be concerned with in the present chapter. However, before leaving this section, let’s take a brief look at the other kinds of determinism. Theological determinism According to many who believe in the existence of God, God is both the 12.15 creator of everything and omniscient (all knowing). An omniscient creator knows every fact. One assumption about what it means to be the creator is that God was present at the beginning of the universe. One assumption about what it means to know every fact is that even facts pertaining to the future are known. If God didn’t know every fact about the future, then there would be at least one fact that he didn’t know—he would not be omniscient. For any future event, then, it seems that there is only one way for it to turn out. This includes events concerning human action. If you turn left instead of right while walking in the park, God knew at the begin- ning of time that you were going to turn left instead of right. If you had turned right instead of left, this would have contradicted God’s prior belief about what you were going to do. Given, then, that God knew billions of years ago what you are going to do at any given moment, then at each moment, there’s only one thing you can do. For each thing that you do, you could not have done otherwise, for to do so would contradict God’s perfect knowledge of the future. Logical determinism Logical determinism arises out of a puzzle discussed by Aristotle in chapter 12.16 9 of his De Interpretatione. Consider the sentence “There will be a sea battle tomorrow.” According to many logicians and philosophers, it is a law of logic that every sentence is either true or false, and not both (and not neither). This would seem to entail, then, that the sentence “There will be  176â …â … The Will: Willpower and Freedom a sea battle tomorrow” is either true right now or false right now. If it is true right now, then there must be a sea battle tomorrow. If it is false right now, then there must not be a sea battle tomorrow. But whatever happens tomorrow, it cannot contradict whatever the truth value of the sentence is today. If there is a sea battle tomorrow, then, when it happens, it will be true that it can’t have been otherwise. Similarly, if there isn’t a sea battle tomorrow, then it will be true that there couldn’t have been a sea battle then. Ethical determinism 12.17 The ancient Greek philosophers Socrates and Plato held that a person always chooses what they think of as good. This doesn’t mean that people always choose what actually is good, since there is room for a difference between thinking that something is good and its actually being good. Perhaps the thing they think of as good really isn’t good. Nonetheless, according to Socrates and Plato, whatever a person chooses is something that the person thinks of as good. If they didn’t think it was good, then why did they choose it? Doesn’t the mere fact that they choose it mean that they want it? And is there really any difference between wanting a thing and thinking of that thing that it is good? On this view, then, a person’s choices are determined by what they think. And they are thus determined. Given what they think, there’s only one thing that they’ll choose—they won’t choose otherwise. A person’s choices are thus determined by their prior psychological states—their thoughts about what is good. Psychological determinism 12.18 Psychological determinism is similar in an important respect to ethical determinism. Psychological determinism is the view that a person always chooses what they most desire. Like ethical determinism, a person’s choice is determined by their prior mental state. Psychological determinism differs from ethical determinism if it turns out that there is a difference between desiring something and thinking that it is good. 12.19 We’ve reviewed these five kinds of deterministic thought to convey the idea that there are many ways to arrive at deterministic conclusions. Of course, each of these arguments have been criticized in various ways by various philosophers. However, coming up with a coherent argument  The Will: Willpower and Freedomâ …â … 177 against all forms of determinism is a highly difficult task. Determinism will not go away quietly! We will now shift our attention to the question of the nature of free will, 12.20 and whether it is the sort of thing that is compatible with determinism. Compatibilism Compatibilists hold that the existence of free will is compatible with deter- 12.21 minism. (For simplicity’s sake, we will primarily have physical determinism in mind for the ensuing discussion.) Recall the two key aspects of determin- ism that incompatibilists allege to be threats to the existence of free will. (1) According to determinism, your preferences and actions are determined by events that occur prior to them, including events that occurred prior to your birth. (2) Given the current state of the universe and the natural laws, there is only one possible future state of the universe, and thus, for whatever you actually do, it is false that you could have done otherwise. Compatibilists disagree with incompatibilists about whether (1) and (2) are threats to free will. One line of thought in favor of compatibilism originates with the phi- 12.22 losopher Harry Frankfurt. Frankfurt’s argument can be seen as aiming to show the compatibility of free will and (2). According to Frankfurt’s line of thought, it is not a requirement of having free will that one could have done otherwise. Central to Frankfurt’s argument are certain hypothetical scenarios or 12.23 thought experiments, hypothetical scenarios that have come to be called “Frankfurt cases.” Frankfurt cases are designed to show that having free will does not require that one be able to do otherwise. To get a feel for Frankfurt cases, let us imagine a future in which brain- 12.24 control technology has been perfected. In this future, certain people have access to microchips that can be implanted in other people’s brains. These brain implants allow one person to remotely control another. Now imagine that there are two roommates, Alicia and Beyoncé, who have agreed to take turns vacuuming their carpets on alternate Mondays. The upcoming Monday, it will be Beyoncé’s turn to vacuum, but Alicia is worried that Beyoncé won’t do it. Alicia is expecting some very special guests and she wants the apartment to be very clean when they arrive. Alicia expects to be very busy doing other things in preparation for the guests, and will not be able (or willing) to do the vacuuming herself.  178â …â … The Will: Willpower and Freedom 12.25 While Beyoncé is sleeping, Alicia implants a brain-controlling microchip into Beyoncé’s brain. The way that this chip is designed to work is to monitor Beyoncé’s brain, checking to see whether Beyoncé decides to vacuum on the upcoming Monday. If Beyoncé does not decide to vacuum, then the chip will switch from its monitoring mode to its control- ling mode and make Beyoncé decide to vacuum. If, however, Beyoncé does decide to do the vacuuming, then the chip will remain in its monitoring mode and not do anything that effects Beyoncé’s decisions or actions. 12.26 Suppose that Monday rolls around and Beyoncé does decide to do the vacuuming. She decides on her own, without any intervention from the chip. It seems intuitively plausible both that Beyoncé could not have done otherwise and that she is nonetheless morally responsible for the completed vacuuming. She could not have done otherwise because the microchip would have prevented her from doing otherwise. And she’s morally respon- sible for doing the vacuuming, since her decision to do it and thus uphold her end of the agreement both resulted in a clean carpet and is morally praiseworthy. 12.27 On the assumption that free will is a requirement of moral responsibility, the following seems to be true of Beyoncé: Since Beyoncé was morally responsible for the resulting vacuuming, Beyoncé decided of her own free will to do the vacuuming. Her free will is thus compatible with her not having been able to do otherwise. The moral that many compatibilists draw from Frankfurt cases is this: Free will, at least the kind that is required for moral responsibility, is compatible with determinism. Incompatibilism 12.28 Suppose, at least for a moment, that we define compatibilism as the view that affirms both (1) the existence of free will and (2) the truth of deter- minism. Given such a definition, there are thus three ways of denying compatibilism and affirming incompatibilism. The first way is to embrace a form of incompatibilism known as hard determinism, the view that denies the existence of free will and affirms the truth of determinism. The second way is to embrace a form of incompatibilism known as libertarianism, the view free will exists and that determinism is false. The third way denies both that free will exists and that determinism is true. This third way is not usually what people have in mind when they discuss incompatibilism, and there is not a widely agreed upon name for it.  The Will: Willpower and Freedomâ …â … 179 In this chapter we will examine two arguments for incompatibilism. The 12.29 first is the origination argument. The second is the consequence argument. The arguments don’t take a stand on whether free will exists. Neither do they take a stand on whether determinism is true. The point of these argu- ments is to support a conditional, an “if–then,” statement. The aim of these arguments is to show that if determinism is true, then free will does not exist. A logically equivalent aim is to show that if free will does exist, then determinism is false. These two arguments can be distinguished in terms of two different 12.30 aspects of what incompatibilists think free will consists in. The first aspect of incompatibilist free will is the idea of the ultimate source of free choice residing in a person. On this conception, a person or some mental act of a person is the true and ultimate source of his or her free actions. This aspect of incompatibilist free will plays a key role in the origination argument. The second aspect of incompatibilist free will is the idea that in having 12.31 free will one has genuine alternate possibilities to choose from. This is the idea that there are genuine ways things could have been. These are pos- sible events that would have been actual if only the person had chosen them. In considering some future course of action there are genuine mul- tiple alternate futures that one has some power to bring into actuality. This aspect of incompatibilist free will plays a key role in the consequence argument. The origination or causal chain argument The origination argument (or causal chain argument) hinges on the aspect 12.32 of determinism that involves past events determining later events. This aspect of determinism conflicts with a certain conception of what free will involves, namely that if one were to have free will then one would be the ultimate source of the decisions that one makes. How does the idea of past events determining later events conflict with 12.33 one’s being the ultimate source of one’s decisions? The key here is best seen by contemplating events that happened before you make your decisions. If it is true that each event that happens is determined by some prior event, then this applies as well to each of your decisions, since each of them is an event, a thing that happens. If determinism is true, then a decision of yours is a link in a causal chain. Prior links make the decisions happen and the decisions themselves make your actions happen. But if this causal chain picture of determinism is correct, then neither you nor things happening  180â …â … The Will: Willpower and Freedom inside of you are the ultimate sources of your decisions, choices, and actions. Whatever the ultimate source is, it is something that happened prior to your decisions. Perhaps the ultimate source is something happen- ing an incredibly long time ago, like the formation of the physical universe during the Big Bang. 12.34 Assume that having free will requires being the ultimate source of your choices and actions. If determinism is true, then neither you nor any aspect of you is the ultimate source. Therefore, if determinism is true, then you don’t have free will. The consequence argument 12.35 At the heart of the consequence argument is a conflict over the nature of time. The conflict concerns, on the one hand, what time needs to be in order for free will to exist and, on the other hand, what it needs to be in order for determinism be true. One way to put these conflicting positions is to say that they conflict over whether the future is open in a way that the past is not. 12.36 Many people hold that there’s a big difference between the past and the future. The past is set. It is something that we no longer have any control over. We no longer have any say in how it turned out. What’s past is over and done and there’s nothing that can be done about it now. In contrast, the future is open. 12.37 Suppose that you are deciding what clothes to wear for your big job interview tomorrow. The choices are open—you can wear your blue outfit or your red outfit. We can put this point in terms of possible futures. One future has you going to your job interview in blue, another future has you going to your interview in red. It is up to you to choose one or the other. At least, that’s what you would think if you thought you had a certain kind of free will. 12.38 But now let us consider what the nature of the future would be if deter- minism is true. On the hypothesis of determinism, it looks like there are not multiple branches ahead of you for you to choose from. If, as determin- ism says, any given moment is made to happen by some previous moment in conjunction with the laws of nature, then the present moment (plus the laws) determines only a single future. On this view of determinism, it looks like the future is just as set or fixed as the past. Free will requires that the future be open. But determinism entails that the future is closed. Determinism looks to be incompatible with the existence of free will.  The Will: Willpower and Freedomâ …â … 181 What Might Free Will Be, If There Were Any Such Thing? Let us turn now to the question, “What might free will be, if there were any 12.39 such a thing?” One way we can approach this question is by first addressing the question, “Freedom aside for the moment, what is the will?” Freedom aside for the moment, what is the will? While it is not wholly uncontroversial in philosophy whether there is such 12.40 a thing or aspect of mind that is worth calling “the will,” we will not spend much time examining the controversy over whether the will exists. Instead, we will sketch a brief account that says some general things about what the will might be if there is such a thing. That is, we’ll give a quick sketch of an answer to the question of what sort of thing or things the phrase “the will” might refer to. The will as willings: Mental events that are common to both intentional action 12.41 and trying to perform actionsâ … On one view, the term “the will” serves to pick out a certain kind of mental state or mental event. These are states or events that we can call “willings” or “acts of will.” One way of understanding willings is by comparing and contrasting 12.42 them to sensations. If we think of sensory perception as the interface at which the external world influences the mind, then sensations are the first mental link in a causal chain leading from the external object to the mental event of perceiving it. In contrast to perception, which is closely linked to the inputs to the mind, action is closely linked to the outputs of the mind. Intentional action can be thought of as the interface at which the mind influences the external world. Willings, then, can be thought of as the last mental link in a causal chain leading from a plan or desire to a bodily movement or action. Recall our discussion of perception in chapter 11. There we faced the 12.43 problem of illusion. Recall that one motive for postulating things such as sense data or sensations was to account for an apparent commonality between accurately perceiving an object and hallucinating it. On one account of perception, what both situations have in common is the pres- ence of a sense datum or sensation. An analogous kind of line of thought can be used in favor of willings. Here, the apparent commonality arises  182â …â … The Will: Willpower and Freedom between, on the one hand, intentionally and successfully performing some action and, on the other hand, trying but failing due to no fault of one’s own. To illustrate, compare (1) successfully and intentionally throwing a paper ball into a wastebasket and (2) trying to do so but failing because, at the last moment, your muscles gave out or a breeze through an opened window blew the ball off course. In comparing these two situations, we notice a mental or psychological commonality between the successful doing and the mere trying. These are willings or acts of will. In both the successful and the failed trying there is the willing. There is an act of will to toss the paper ball, an act of will that results in the ball actually getting into the basket in (1) but not in (2). 12.44 The will as a source of power and weakness: Willpower, akrasia, and weakness of willâ … Another way of thinking about the will (and this is consistent with the previous way of thinking about it) is that it is something that varies along a spectrum of power versus weakness. Such variation has to do with willpower. Low degrees of willpower are associated with weakness of will— what philosophers sometimes call akrasia. 12.45 The idea of weakness of will is a commonsense idea. Many of us have experienced difficulty in doing something that we thought of as the right or preferable thing to do. We want to do a thing, but fail due to some failure inside of us. One such example might be trying to resist highly tempting junk food while trying to stick to a new and healthier diet. 12.46 Despite seeming to be rooted in common sense, the idea that there is such a thing as weakness of will has raised various philosophical problems. There is no clear consensus on how to solve the problems. We will here give a quick sketch of the heart of the problem. 12.47 Imagine that George resolves to quit drinking alcohol. George has certain mental states pertaining to alcohol. George believes that alcohol is damag- ing to his liver and he desires to no longer drink things that are damaging to his liver. In short, George both believes that alcohol is to be avoided and desires to avoid it. 12.48 George’s beliefs and desires constitute his reasons for action. On any given occasion in which George performs an intentional action, we cite such mental states as his beliefs and desires to explain his actions. We cite his reasons for acting in answering questions of the form “Why did George do that?” 12.49 Acting for reasons distinguishes intentional action from involuntary behavior. When George does something on purpose, it is something  The Will: Willpower and Freedomâ …â … 183 brought about by his beliefs and desires. When George behaves in an invol- untary way, as when his leg reflexively jerks upon being hit on his knee, the explanation of the leg’s jerk involves something other than beliefs and desires. Suppose George is at a party and he refuses a beer offered to him. Why 12.50 did he refuse the beer? He refused it because he believes that it would be bad for his liver, and he desires to avoid damaging his liver. Those are his reasons for the refusal. Now consider a case that looks like George exhibiting weakness of will. 12.51 Suppose that there is an occasion in which, despite believing that alcohol is to be avoided and desiring to avoid it, George “caves in” and has an alcoholic drink. Later at the party, he is again offered a beer. He says, “Well, I really shouldn’t,” but then goes and drinks it anyway. Why did George do that? It looks like this is an occasion in which he 12.52 acted despite what he believed and desired. George acted in a way that went against his “better reason” or “better judgment.” Nonetheless, in taking the drink, he acts intentionally. It is not like the drinking of the beer was invol- untary, as in reflexive movements or sleepwalking. So, what’s going on here? One sort of answer would be to say that we 12.53 must have been mistaken about what George really desired. Maybe in that moment he actually did desire alcohol more than he desired to avoid it. A different sort of answer appeals to a diminishment of willpower, a tempo- rary weakness of the will. On this view, George’s desires (and beliefs) about alcohol remained constant, but there was a fluctuation in his resolve, his strength of will. What might the freedom of the will consist in? Let us return now to the question of the freedom of will. What would it 12.54 mean for the will to have the sort of freedom that matters for moral respon- sibility, the sort of freedom at the heart of debates between compatibilists and incompatibilists? It should come as no surprise by now that there is a lot of controversy 12.55 about how to answer such a question. This is what much of the controversy between compatibilists and incompatibilists boils down to. Here we will present just two of the many models of what freedom of will consists in. The first is one that fits more closely with compatibilism, and the second fits more closely with incompatibilism.  184â …â … The Will: Willpower and Freedom 12.56 The hierarchy of desires model of freedom of will: A compatibilist account of free willâ … This model of free will originates with philosopher Harry Frankfurt. There are two key notions in the model. The first is a distinction between first order and higher order (second order, third order, etc.) mental states, especially mental states such as desires. The second key notion is that of a desire becoming a volition. 12.57 First, let us consider the relevant notion of order. A second order desire is a desire about a first order desire. So, what’s a first order desire? It’s a desire about something that itself isn’t a desire. Suppose George desires a beer. This is a first order desire. Suppose, despite having a desire for beer, George desires to be the sort of person who doesn’t desire beer. George has a second order desire, then—a desire to not desire beer. 12.58 Next, let us consider what it means for a desire to become a volition, or, in other words, for a desire to become one’s will or a willing. One might have a desire, but for some reason or other, fail to act on it. One desires to own a blue coat, but all of the blue coats for sale are too expensive, and so one buys a red one. In this case, the desire did not become what one willed, it did not become a volition. But, if on the following day one lucked into some prize money, or the prices of blue coats came down, then the desire for a blue coat could become a volition. 12.59 Combining the idea of second order desires with the idea of volitions gives rise to the idea of second order volitions. On this model, having free will consists in having second order volitions. On some particular occasion in which a person acts, they acted freely if the action is in accordance with their second order volition. All of this is compatible with all of the states of the person being determined by some prior state of the universe plus the physical laws. Therefore, the resultant view is a compatibilist view. 12.60 The ultimate origination model of freedom of will: An incompatibilist account of free willâ … On this model of free will, a person or their will must be the ultimate originator of their choices or actions—if the will is itself caused or determined, then it is not free. One version of the origination model of free will is the agent causation model. Central to this model is the idea of agent causation, a kind of causation distinct from event causation. Usually, we think of causation as a relation between events, as when a bomb’s exploding causes a building to collapse. The event of the bomb exploding is the cause and the event of the building collapsing is the effect. In contrast to event causation, in agent causation, the cause of an event can be an agent, a person.  The Will: Willpower and Freedomâ …â … 185 On the agent causation version of the ultimate origination model of free 12.61 will, freedom of the will requires a special kind of causation, one that is not a relation between two events but instead between an agent and an event. Many reject this model as incoherent, for it is quite difficult to see how an agent himself or herself can be a cause. According to critics of the idea of agent causation, when an agent is involved in causation, some change of state or condition of the agent must be the cause, that is, some event involving the agent and not simply the agent himself or herself must be the cause. One of the main criticisms, then, of the agent causation model of free will, is that the very idea of agent causation, which is sup- posed to be distinct from event causation, doesn’t make any sense. Conclusion Perhaps more than any other aspect of the mind, the will is especially sig- 12.62 nificant in assessments of moral responsibility. How can someone be morally responsible for an act unless they freely choose it? Despite this commonsense connection between freedom and responsibility, there are many severe challenges to the idea that free will exists. The main challenges have to do with determinism. Some philosophers view the challenges as so severe that they conclude that free will doesn’t exist. Others remain opti- mistic about the prospects of a viable compatibilism whereby free will can exist in a deterministic universe. Annotated Bibliography Dennett, Daniel (1984) Elbow Room: The Varieties of Free Will Worth Wanting (Cambridge, MA: MIT Press). Dennett defends a version of compatibilism. Frankfurt, Harry (1969) “Alternate Possibilities and Moral Responsibility,” Journal of Philosophy, 66, 829–839. Frankfurt argues against the view that being morally responsible requires that one could have done otherwise. Frankfurt, Harry (1971) “Freedom of the Will and the Concept of a Person,” Journal of Philosophy, 68(1), 5–20. Frankfurt spells out a compatibilist account of free will in terms of higher order desires—desires about desires. Hume, David (1748) An Enquiry Concerning Human Understanding, full text available at http://www.earlymoderntexts.com/he.html, accessed February 11, 2013. See especially chapter 8, where Hume defends his version of compatibilism. 186â …â … The Will: Willpower and Freedom Reid, Thomas (1788) Essays on the Active Powers of Man, full text available at http://www.earlymoderntexts.com/reac.html, accessed February 11, 2013. Reid defends agent causation. See especially his chapter 4. Strawson, Galen (2003) “On Free Will,” Richmond Journal of Philosophy, 4, full text available at http://www.bookofparagon.com/Robots/FreeWill.pdf, accessed February 11, 2013. A highly accessible discussion of the philosophy of free will. Stroud, Sarah (2008) “Weakness of Will,” in Edward Zalta (ed.), The Stanford Encyclopedia of Philosophy, full text available at http://plato.stanford.edu/ archives/fall2008/entries/weakness-will/, accessed February 11, 2013. Nice overview of the philosophical difficulties involved with the idea of a lack of willpower. Wegner, Daniel (2002) The Illusion of Conscious Will (Cambridge, MA: MIT Press). Wegner is one of several scientists defending the view that neuroscience shows that free will does not exist.  13 Intentionality and Mental Representation Introducing Intentionality Many mental states have what philosophers call “aboutness” or intentional- 13.1 ity. The belief that the sky is blue is about the sky. It is a belief that is directed at the sky or at the sky’s being blue. “Intentionality” used in this context is a technical, philosophical term, and should not be confused with the more commonsensical notion of doing something intentionally, in the sense of doing it on purpose. Despite this difference between the technical term and the commonsense term, they do have a common origin having to do with pointing—“index,” as in “index finger,” has a similar origin. The core idea of the aboutness or intentionality of mental states is that of their directed- ness. Intentionality is the directedness of a mental state toward its contents. Intentionality is the source of many difficult problems in the philosophy 13.2 of mind. One such problem has to do with the way in which intentionality seems both to be relational and to involve things that do not exist. What’s the problem here? Plausibly, you can think about or have beliefs about things that do not exist. I might have the mistaken belief that there is a man upstairs singing a song when there is no one upstairs at all. If intentionality is relational, then it is tempting to say that my belief in this example is something that relates me to some other man—the one that I think is singing. However, if it turns out that no such man exists, then there is no one that my thought thereby relates me to. Another puzzling aspect about intentionality concerns the way in which 13.3 we can have mental states directed at things that are so far away that we This Is Philosophy of Mind: An Introduction, First Edition. Pete Mandik. © 2014 John Wiley & Sons, Inc. Published 2014 by John Wiley & Sons, Inc.  188â …â … Intentionality and Mental Representation couldn’t possibly have had any causal interaction with them. Suppose I believe that there exists at least one galaxy that has exactly one trillion stars in it (not more and not less). Suppose further that there is such a galaxy. That would seem to make my belief true. However, the galaxy might be so far away that no causal interaction can transpire between its current state and mine. All causation in the universe must be mediated by exchanges of energy, and energy cannot be moved at a rate that exceeds the speed of light. That galaxy, let us suppose, is so far away that it would take a billion years for light from it to reach me. Nonetheless, there seems to be some relation between my belief and the current state of that galaxy. 13.4 One puzzle about intentionality concerns how something so mysterious can be consistent with a naturalistic or scientific view of reality. For example, if some version of physicalism is the most scientifically respectable view of the mind, how can intentionality possibly exist? The project of naturalizing intentionality is the project of showing how the existence of intentionality is consistent with the understanding of reality we have via the natural sciences. 13.5 In this chapter, we will examine some attempts to construct a naturalistic account of intentionality. But first we will further explore the puzzling aspects of intentionality. The Inconsistent Triad of Intentionality 13.6 The heart of what is so puzzling about intentionality can be formulated as an inconsistent triad of propositions concerning intentionality, existence, and relations. The three propositions are: Proposition 1:â ‡ We can think about things that do not exist. Proposition 2:â ‡ Thinking about something is a relation between the thinker and the thing thought about. Proposition 3:â ‡ There can only be a relation between two things if both of the things exist. Each proposition in the triad seems plausible when considered in isolation. However, when we consider all three propositions in conjunction, it is hard to see how they are mutually consistent. It becomes clear that at least one of them must be false. But which one? 13.7 Let’s take a closer look at each individual proposition and see what makes each one separately plausible.  Intentionality and Mental Representationâ …â … 189 Defending each individual proposition Proposition 1: We can think about things that do not exist. Do you believe 13.8 that Rudolph the red-nosed reindeer, a flying reindeer with a glowing nose, really exists? Do you believe in the real existence of Zeus, the father of the gods of Olympus, who throws lightning bolts down to Earth? Like many people you are likely to answer “no” to both questions. Also, like many people, you understand both questions perfectly well. You can demonstrate such understanding by answering the following question: In answering “no” to each question, who are you denying the existence of? Here the answers seem clear, you are denying the existence of Rudolph the red-nosed reindeer in the first case, and denying the existence of Zeus, the father of the gods of Olympus, in the second case. When you think that Zeus doesn’t exist, who are you thinking about? Again, the answer seems clear. You are thinking about Zeus. And since Zeus doesn’t exist, in thinking about Zeus you are thinking about something that doesn’t exist. Proposition 2: Thinking about something is a relation between the thinker 13.9 and the thing thought about. Consider the height of the youngest person you know. Chances are, you are taller than that person. Being taller than someone is a relation. So is being older than a person. Part of what makes being taller or being older a relation is that it involves two things, one of which is taller or older than the other. Is thinking about something a relation borne toward that thing? Think 13.10 about the piece of furniture that is nearest to you right now. This seems to be a relation as much as being near that piece of furniture is a relation. The case of thinking about the piece of furniture looks like it involves two things—there is you, the thinker, and then there is the nearby piece of furniture, the thing that you are thinking about. Proposition 3. There can only be a relation between two things if both things 13.11 exist. One way to talk about relations, such as the relation of being taller than something, is to use variables like x and y, or blank spaces, in the fol- lowing manner: “x is taller than y,” “_ is taller than _.” Consider the relation, being filled with a liquid, that holds between a coffee cup and the coffee inside of it. We can call this the “â ›‘x is filled by y’-relation” or the “â ›‘_ is filled by _’-relation.” Suppose there is some coffee cup that has absolutely nothing in it. Is it filled by something? Is there some quantity of liquid or gas that fills it? If, as we have stipulated, there is nothing in it, then there is nothing that exists that fills it. Another way of putting the point is to say that there exists no y such that the coffee cup is filled by it. Another way of put this is to say that the coffee cup, being absolutely empty, does not bear the  190â …â … Intentionality and Mental Representation “_is filled by _”-relation to anything. The relation of being filled requires two things. When no filler exists the container contains nothing. If this sort of thing is true of relations generally, then we can sum up by saying that relations can only take place between things that exist. Spelling out the inconsistency 13.12 To help spell out the inconsistency of the triad, we can select a pair of propositions from the triad, assume for purposes of discussion that each member of the pair is true, and see that the remaining member of the triad must therefore be false. 13.13 If 1 and 2 are true then 3 is false. Suppose that we can think about things that do not exist and that thinking about something is a relation borne to the thing thought about. This would seem to entail that there is at least one relation that can be borne to nonexistent things, namely the thinking- about relation. But this contradicts proposition 3, which says that relations can only be borne toward existing things. 13.14 If 1 and 3 are true then 2 is false. Suppose that we can think about things that do not exist and that relations can only be borne toward existing things. This would seem to entail that thinking about something is not a relation between a thinker and a thing thought about. But that would contradict proposition 2, which says that thinking about something is a kind of relation borne to that thing. 13.15 If 2 and 3 are true then 1 is false. Suppose that thinking about something is a relation borne to that thing and that there can only be relations between existing things. This would seem to entail, then, that we cannot think about things that do not exist, thus contradicting proposition 1. Internalism versus Externalism 13.16 One of the key ideas involved in the inconsistent triad of intentionality is the idea that intentionality might be a sort of relation. One line of thought against the idea that intentionality is a relation is the thought that I might have all the same ideas as I do now even though my mind or brain is the only thing that exists. The idea that intentionality does not depend on relations that your mind or brain bears to items in the external world is the philosophical position internalism. In opposition is the posi- tion externalism.  Intentionality and Mental Representationâ …â … 191 One way to think about the debate between internalists and externalists 13.17 is in terms of supervenience. “Supervenience” is a technical term referring to a kind of dependence between properties. Many philosophers of mind hold that mental properties depend on physical properties and they state their claims about that dependence in terms of supervenience. They say that mental properties “supervene” on physical properties. One sort of supervenience claim says that two individuals cannot differ in their mental properties without differing in their physical properties and that a single individual cannot change his or her mental properties without changing his or her physical properties. An individual has two kinds of physical properties. The first kind is those 13.18 physical properties that are intrinsic or internal to the individual. The second kind is those physical properties that involve relations to entities external to the individual. An example of the first kind is the property of having more than a billion neurons in your brain. An example of the second kind is the property of being six miles away from the nearest sample of uranium. The debate between internalists and externalists in the philosophy 13.19 of mind is a debate over which of an individual’s physical properties their mental properties supervene on. According to internalists, an indi- vidual’s mental properties supervene on only the intrinsic physical proper- ties of an individual. One version of internalism holds that if the number and arrangements of particles in two people’s nervous systems were exactly similar, then their mental states would be exactly similar. Internalists hold that external differences between two individuals are strictly irrelevant to what mental properties they each have. In contrast, externalists hold that individuals who have intrinsically 13.20 similar brains can nonetheless differ in their mental properties if there are certain differences in the relations that the individuals bear to their respec- tive physical environments. Where internalists say that mental properties supervene on physical properties that are internal to an individual, exter- nalists say that mental properties supervene on external as well as internal physical properties. Where internalists say that mental properties have a “narrow” supervenience base, externalists say that mental properties have a “wide” supervenience base. One of the most discussed versions of the internalism versus external 13.21 ism debate focuses on intentionality. The opponents disagree about the answer to the following question: Could two intrinsically similar brains nonetheless differ in what they are thinking about in virtue of being related  192â …â … Intentionality and Mental Representation to different environments? Externalists, unlike internalists, say “yes.” Externalists, unlike internalists, say things like “intentionality just ain’t in the head.” For externalism: The Twin Earth thought experiment 13.22 One famous argument for externalism originates with the philosopher Hilary Putnam and features his famous Twin Earth thought experiment. To conduct the thought experiment yourself, you need to imagine the fol- lowing situations. First, consider a stage in the history of the planet Earth wherein humans had not yet developed an understanding of chemistry that was sophisticated enough for them to know that the chemical composition of water is two parts hydrogen to one part oxygen. These are people living before anyone had discovered that water is H2O. Imagine also that these people use the English word “water” to say things like “I am very thirsty and would enjoy a glass of water.” They use the word “water” to refer to a liquid that, unbeknownst to them, happens to be the chemical H2O. 13.23 Consider one such person. Call him “Oscar.” Plausibly, since Oscar’s “water” utterances refer to what happens to be H2O, and since such utter- ances are expressions of thoughts, the thoughts thereby expressed are thoughts about H2O. Despite being ignorant of the chemical composition of water, Oscar ends up thinking about H2O in thinking the thoughts he expresses with his “water” utterances. 13.24 For the next stage of the thought experiment, imagine a planet that is extremely physically similar to Oscar’s Earth, a planet that we can call “Twin Earth.” Twin Earth is populated by beings similar to the beings populating Earth, including a counterpart to Oscar who we can call Twin Oscar. Just like Oscar, Twin Oscar uses the word “water” to say things like “I am very thirsty and would enjoy a glass of water.” The main difference between Earth and Twin Earth is the chemical composition of the clear liquid referred to as “water.” Instead of being H2O, Twin Water is a wholly distinct chemical that we, in conducting the thought experiment, can just call “XYZ.” Now, just like Oscar, Twin Oscar lives in a society wholly ignorant of the chemical composition of the stuff that they refer to as “water.” However, despite their ignorance, and in virtue of having grown up on Twin Earth, the Twin Earthlings’ “water” utterances refer to what happens to be XYZ, and such utterances are expressions of thoughts about XYZ. 13.25 The externalistic conclusion of the Twin Earth thought experiment is that Oscar’s and Twin Oscar’s thoughts each have a wide supervenience base.  Intentionality and Mental Representationâ …â … 193 In virtue of having been raised in different environments—an H2O envi- ronment and an XYZ environment, respectively—Oscar and Twin Oscar have thoughts with distinct intentional contents. Oscar and Twin Oscar are different mentally in virtue of their external physical differences. Against externalism: Swampman and the brain in the vat Not all philosophers of mind have been convinced by the Twin Earth argu- 13.26 ment. Many philosophers have been swayed by a very strong intuition in favor of internalism. This is the intuition that what matters for mentality is wholly contained within the body of a person, and that perhaps all that really matters physically for mentality are physical facts about the structure and functioning of the central nervous system. One way to probe such intuitions is by contemplating certain thought experiments. Here we will take a look at the thought experiment of swampman and the thought experiment of the brain in the vat. Imagine that there is a swamp thousands of miles away from me and 13.27 that in this swamp there is a log that gets struck by lightning. This results, by amazing coincidence, in a being that is molecule-for-molecule a perfect physical copy of me. Down to the last particle, it has a microscopic physical arrangement exactly similar to mine. This swampman is a being we can call, for the purposes of the thought experiment, Swamp Pete, although if you asked him his name, he would simply reply, “Pete.” Why would he say that? Because he is exactly internally physically similar to me, and that is precisely what I would utter in response to such a question. Many philosophers express their allegiance to internalism by urging that 13.28 my swamp counterpart would have precisely the same mental properties that I have. If I think that ice cream tastes great, then so does Swamp Pete, even though he has never actually tasted any ice cream yet—he was just formed in the swamp a few seconds ago and has not yet come into contact with any ice cream. Serving a purpose similar to the swampman thought experiment is the 13.29 brain in a vat thought experiment. Many philosophers have a very strong intuition that, for all they know, they could very well just be a brain in a vat, a brain that is not in a real body with real sense organs, but is instead hooked up to a virtual reality computer that simulates an external world and body, thus fooling the brain. In extreme versions of the brain in a vat thought experiment, the brain and virtual reality computer first formed in a swamp moments ago by a lightning strike. For all you know, according  194â …â … Intentionality and Mental Representation to this thought experiment, you are a brain in a vat that just formed swamp-style only moments ago. If you and the swamp-formed brain in a vat think all the same things—if you both have thoughts with all the same intentionality or intentional content—then internalism must be true. 13.30 One question that arises in connection with the internalism versus exter- nalism debates is the question of content determination. How does a state of the brain or mind come to represent what it does? What determines its intentionality or content? Further, how can this content be determined in a way that is consistent with a physicalist or naturalist world view? It is to such questions that we now turn. Theories of Content Determination 13.31 Some philosophers have held that intentionality is one of the main obsta- cles to physicalism. In other words, because mental states exhibit intention- ality, mental states cannot be wholly physical or wholly reducible to the sorts of properties studied by natural sciences such as physics, chemistry, and biology. 13.32 Other philosophers have resisted such a conclusion and have striven to construct theories consistent with physicalism or naturalism of how mental states can have intentional content. They have striven to naturalize inten- tionality. In the remainder of this chapter we will examine several of the main attempts to supply naturalistic accounts of what it would mean for a wholly physical system to have states with intentional content. These are attempts at spelling out a fully naturalistic theory of content. 13.33 Many such theories are spelled out in terms of a notion of mental rep- resentation. Mental representations are posits that are supposed to be the main bearers of intentionality or aboutness. In thinking a thought about something, one thereby has a mental representation that represents that thing. Many of the theories of intentional content that we will examine can be equally well described as theories of representational content or as theo- ries of what representing consists in. Resemblance theory 13.34 According to the resemblance theory of representational content, represen- tation is a relation of resemblance—mental representations represent things in virtue of resembling those things. While the resemblance theory had  Intentionality and Mental Representationâ …â … 195 many historical proponents (for example, both Aristotle and Berkeley sub- scribed to the view) contemporary thinkers tend not to be especially enthused about it. One problem with the resemblance theory is that there are many things 13.35 we are able to think about without there being any plausible sense in which a state of our mind resembles the thing thought about. This is especially so on the physicalist assumption that mental states are brain states. I can think about something’s being 60 feet tall and bright green without my brain state being either 60 feet tall or bright green. Another key objection to the resemblance theory is that the resemblance 13.36 relation is symmetrical whereas the representation relation (if representa- tion is a relation) is not. If George resembles Larry, it follows that Larry resembles George. However, if a state of George’s mind represents Larry, it does not follow that Larry represents a state of George’s mind. Other objections to the resemblance theory include points we already 13.37 discussed in chapter 11 in connection with arguments against picture- like representations (resemblance-based representations) serving to ground our general capacities for thinking. One such point is that resemblance seems ill suited to ground our ability to mentally represent negative states of affairs: There’s a difference between thinking that there are no dogs in the room and that there are no cats in the room, but a representation such as a picture may not resemble one state of affairs any differently than the other. Another such point is that resemblance is ill suited to ground our ability to mentally represent abstract properties: The abstract property of being a triangle can be represented in thought, but anything resembling a triangle by being triangular is also a particular kind of triangle, for example, an equilateral triangle. Given that an equilateral triangle resembles other equilateral triangles more than nonequilateral triangles, what can suffice to make it a representation of triangles in general—an abstract representation that equally represents all kinds of triangles? Appeals to resemblance alone are insufficient to answer this question. Interpretational semantics The interpretational approach to mental representational content posits 13.38 that what makes a mental representation have its content is that it is inter- preted as having that content. The interpretational approach may be read as saying that mental representations and nonmental representations are similar in this regard. An example of a nonmental representation is the  196â …â … Intentionality and Mental Representation printed English sentence “Dogs are furry.” It is a nonmental representation because it, the string of marks on a page or screen, is not a state of anyone’s mind. Plausibly, what makes this string of marks represent dogs being furry as opposed to fish being slimy is that we, the speakers of the English lan- guage, interpret it that way. The meaning or content of those marks is thus interpretation dependent. 13.39 The philosopher Daniel Dennett has offered that there is an important analogy to be drawn between nonmental and mental representations. On his intentional stance theory of intentionality, a physical system comes to have states with certain representational contents just in case it is useful in predicting and explaining the behaviors of such a system to adopt the intentional stance toward it. The intentional stance is a point of view toward things wherein one interprets those things as having intentionality. 13.40 Many philosophers have rejected this intentional stance approach, and interpretational approaches generally, on the grounds that they lead to either circular or infinitely regressive accounts of intentional content. At the heart of the alleged problem is that interpretation is itself a kind of representation. So, if I interpret some string of marks or someone’s brain state as representing x, then I myself am representing that string of marks or that brain state. So, in trying to explain representation in terms of inter- pretation, what the interpretational approach does is to try to explain representation in terms of representation. But this would seem to either lead to a circle or to an infinite regress. Conceptual role semantics 13.41 The conceptual role approach tries to define a mental state’s representa- tional content in terms of relations it enters into with other mental states of the same mind. One way of viewing the conceptual role approach is as an application of functionalism to intentionality. Recall the functionalist idea, from chapter 8, that what makes a state a belief as opposed to a sensory impression or a desire is the set of causal relations it bears to the other states in the causal economy of an entire cognitive system. Analogously, the conceptual role approach says that what makes the belief a belief that grass is green as opposed to the belief that the moon is round is a set of relations to other cognitive states. 13.42 For example, on this proposal, the key relations to other mental states that help define a mental state as being a representation of grass are rela-  Intentionality and Mental Representationâ …â … 197 tions to mental states that represent plants, mental states that represent slender leaves, and mental states that represent the color green. One of the main inspirations for the conceptual role account is a highly 13.43 plausible account of the source of the meaning of logical connective terms such as “and” and “or.” Plausibly, in a logical system such as a logical algebra or a computer language, what defines the meaning of a symbol such as “and” is something like the following: If symbols “P” and “Q” are both true, then the symbol string “P and Q” is true. In such a symbol system, the meaning of “and” is constituted by the role that it plays in a system of symbols. A computer contains a symbol that means “and” if it is disposed to use that symbol to output “P and Q” upon receiving both of the inputs “P” and “Q.” Such an approach is highly plausible for the intentionality or content of 13.44 concepts like “and.” This is because it is totally obscure what else the meaning of “and” might consist in. It is not like there is any plausible object or set of objects that can be regarded as the referent of “and” in the way that there are entities that “dog” refers to. One of the advantages of conceptual role semantics is that it is plausibly 13.45 naturalizable. The sorts of relations posited between system states seem like the sorts of relations between states that a suitably programmed computer can realize. If a computer can do it, then there is little doubt that the thing done is something mechanical and fully consistent with a physicalistic and naturalistic view of reality. However, even though conceptual role semantics can be seen as having 13.46 an advantage in accounting for the meaning of logical connective concepts like “and,” many philosophers regard it as highly implausible as an account of the concepts that refer to real world entities the way that “dog” does. One oft discussed complaint against conceptual role semantics is that it leads to a kind of meaning holism that entails that distinct thinkers cannot have thoughts with shared contents. Suppose that Xavier and Yvonne differ with respect to the things that 13.47 they believe about dogs. Suppose that Xavier believes that he was once bitten by a dog, but that Yvonne believes no such thing. It follows then that Xavier’s dog concept is related to a different set of beliefs in his cognitive economy than Yvonne’s dog concept is in hers. Conceptual role semantics seems to entail that Xavier and Yvonne are not thinking the same thing when they think the thoughts that each would express by uttering “the dog is on the mat.” However, it is allegedly a truism of common sense that  198â …â … Intentionality and Mental Representation distinct thinkers with distinct sets of beliefs about dogs can nonetheless think thoughts about dogs with the same content. Distinct thinkers can each equally well think that the dog is on the mat. If so, then conceptual role semantics has a serious problem accounting for this. 13.48 However, maybe this objection is incorrect. Perhaps no two thinkers ever think exactly the same things. Instead, they think similar (but not identical) things—they have thoughts with similar contents. If so, then perhaps there’s hope for conceptual role semantics after all. Causal or informational theory 13.49 Motivated in part to avoid the sorts of problems seen to arise for conceptual role semantics, many philosophers have been attracted to views of inten- tionality whereby it is a kind of direct relation between a thinker and an object. On such a view, two thinkers can both think about one and the same dog despite holding wildly divergent beliefs about that dog or even wildly divergent beliefs about dogs in general. One attempt to ground intentional- ity as a kind of direct relation to real world objects is an approach that sees intentionality as grounded in a kind of causal relation between thinkers and the real world objects that they think about. 13.50 If intentionality is a causal interaction between a thinker and an object, what kind of causal relation is it? One sort of proposal is that it is, or is grounded in, perception—the relation between perceivers and perceived objects. That intentionality is grounded in perception is a philosophical view closely associated with the history of empiricism (as discussed in chapter 11). 13.51 One source of inspiration for the causal approach is a phenomenon occurring in the natural world that some philosophers have called “natural meaning.” Natural meaning occurs when there is a causal relation between two phenomena such that one of them can be taken as a sign or signal indicating the presence of the other. Examples of uses of the words “mean” and “means” along the lines of this idea of natural meaning include “smoke means fire,” “red spots on the face mean measles,” and “a hundred tree rings in the cross-section of a tree trunk means that the tree was a hundred years old.” The basic idea of natural meaning is that some naturally occur- ring phenomena carry information about other naturally occurring phenomena. 13.52 One way of applying this idea to the intentionality of mental states is to say that mental states represent the things that they are caused by or carry  Intentionality and Mental Representationâ …â … 199 information about. Such a proposal has a certain amount of plausibility when applied to perceptual states. So, for instance, when you and I look at two different but highly similar umbrellas, what makes my perception a perception of one umbrella and not the other is that my perception is caused by the one umbrella and not the other. Despite this plausibility when applied to perceptual states, the causal 13.53 approach seems to run into problems when applied to other mental states, especially such mental states as thoughts about future events and nonexist- ent entities. Future events can’t be the causes of current thoughts, since (unless time travel is possible) things in the future cannot have effects on things in the present. Nonexistent entities can’t have effects on actual thoughts, because.€.€. they don’t exist! The problem of nonexistence is a very serious problem for the causal or informational theory of content. Teleological evolutionary theory At the heart of the teleological theory of content is the idea of teleology— 13.54 the idea of having an aim or a purpose. One way in which things have a purpose is if we assign a purpose to them. A wadded-up newspaper can become a door stop if I decide that that is its purpose. Another way in which things have a purpose, especially if those things are the traits and behaviors of living organisms, is if those things have evolved, that is, if those things have been selected for in the process of Darwinian evolution. The spots on an animal’s fur have the purpose of acting as camouflage because this has conferred a survival advantage to that animal’s ancestors. What it means, on an evolutionary teleological view, to say that the eyes 13.55 are for seeing and that the heart is for pumping blood, is that seeing and pumping blood explains how those body parts got selected for in the process of evolution by natural selection. Extending this view to account for the intentionality of mental states involves spelling out what mental states are naturally selected for. The teleological theory of content can be interpreted as an elaboration 13.56 of the causal theory of content. The elaboration is to combine the idea of natural meaning with the idea of evolutionary purpose. Mental states rep- resent what they have been naturally selected to carry information about. Such a combination shows promise of solving the problem of nonexist- 13.57 ence. Instead of saying, as the causal theory does, that a mental representa- tion represents whatever causes it, the teleological theory says that a mental representation represents whatever it was selected for through evolution to  200â …â … Intentionality and Mental Representation be caused by. To illustrate, suppose that there is a state of a frog’s brain that is activated whenever small dark things move in the frog’s visual field. Plausibly, the evolutionary purpose of such a brain state is to indicate the presence of flies, since being able to detect flies (a source of food for a frog) conferred a survival advantage on the frog’s ancestors. If a small ball bearing were dangled on a string in front of the frog’s face, this would activate the frog’s fly detectors. The states of these detectors would represent the frog’s visual field as containing a fly, even if there doesn’t currently exist any fly in the vicinity of the frog. This illustrates a way in which the tele- ological theory can be interpreted as offering a solution to the problem of nonexistence. Conclusion 13.58 Intentionality is a pervasive aspect of our mental lives. It is so pervasive that some philosophers hold intentionality to be the mark of the mental—a state is a mental state if and only if it exhibits intentionality. However, despite the centrality of intentionality, it is a deeply weird phenomenon. And philosophers have struggled to show how the existence of intentional- ity can be consistent with a naturalistic worldview. Annotated Bibliography Cummins, Robert (1989) Meaning and Mental Representation (Cambridge, MA: MIT Press). An excellent overview of the main theories of content determina- tion with an eye toward the issues that matter for cognitive science. Frances, Bryan (2002) “Twin Earth,” in Marco Nani and Massimo Marraffa (eds.), A Field Guide to the Philosophy of Mind, full text available at http://host. uniroma3.it/progetti/kant/field/tea.htm, accessed February 11, 2013. Fun overview of the discussions of externalism that grow out of Putnam’s (1975) Twin Earth thought experiment. Freeman, Walter (2007) “Intentionality,” Scholarpedia, 2(2), 1337, full text available at http://www.scholarpedia.org/article/Intentionality, accessed February 11, 2013. A scientist’s interesting though idiosyncratic overview of intentionality. Haugeland, John (1990) “The Intentionality All-Stars,” Philosophical Perspectives, 4, Action Theory and Philosophy of Mind, 383–427. Entertaining overview of some of the main positions concerning intentionality.  Intentionality and Mental Representationâ …â … 201 Jacob, Pierre (2010) “Intentionality,” in Edward Zalta (ed.), The Stanford Encyclopedia of Philosophy, full text available at http://plato.stanford.edu/archives/fall2010/ entries/intentionality/, accessed February 11, 2013. An absolutely top-notch overview of this difficult subject. Pitt, David (2008) “Mental Representation,” in Edward Zalta (ed.), The Stanford Encyclopedia of Philosophy, full text available at (http://plato.stanford.edu/ archives/fall2008/entries/mental-representation/, accessed February 11, 2013. Terrific overview of the relevance of mental representation for different parts of philosophy of mind (e.g., consciousness, imagery, eliminativism). Putnam, Hilary (1975) “The Meaning of ‘Meaning’,” Minnesota Studies in the Philosophy of Science, 7, 131–193, full text available at http://mcps.umn.edu/ assets/pdf/7.3_Putnam.pdf, accessed February 11, 2013. Classic discussion of Twin Earth and externalism.  14 Consciousness and Qualia Optimism about Explaining Consciousness Consciousness and related topics such as qualia have figured quite a bit in 14.1 the philosophical discussions covered in this book. So far, the main theme of such discussions has been a debate between pessimists and optimists concerning whether consciousness and qualia can ever be explained, espe- cially whether they can be explained in physicalistic terms. Dualists and panpsychists have occupied the pessimistic position. Pessimism about ever explaining consciousness has been central to their explanatory gap argu- ments against physicalism. Varieties of physicalism such as the mind–brain identity theory and functionalism represent an optimistic stance on the question of whether consciousness can ever be explained. What we haven’t seen yet, but will in the present chapter, are any real details on how the optimists might propose to explain consciousness. In the present chapter we will look at two such proposals. Both proposals reflect the general trend in cognitive science-inspired philosophy of mind of being functionalistic and representational. That is, they spell out explanations of consciousness in terms of the roles that certain states play, roles that may very well be played equally well by computers and biological systems. Further, there is a significant role in each theory played by mental representations. Insofar as these representational theories of consciousness are optimistic about naturalizing consciousness, they are also optimistic about naturalizing rep- resentation, a major topic of the previous chapter. This Is Philosophy of Mind: An Introduction, First Edition. Pete Mandik. © 2014 John Wiley & Sons, Inc. Published 2014 by John Wiley & Sons, Inc.  204â …â … Consciousness and Qualia 14.2 Before turning to the theories, we will take a look at some basic ideas concerning what it is that needs explaining in offering an explanation of consciousness. The theories that we will examine strive to explain one or more of these various phenomena of consciousness. Further, they strive to explain the relations between these phenomena. One useful way to get a handle on consciousness is by examining different uses of the word “con- scious” and related words. Focusing on Several Different Uses of the Word “Conscious” 14.3 It is useful to investigate uses of such words as “conscious” and “conscious- ness,” as well as such related words as “awareness.” In this chapter, we will focus on four of the main uses of “conscious” that have attracted the interest of contemporary philosophers of mind. Each of these four uses of the word “conscious” arguably labels four distinct phenomena or four distinct kinds of consciousness (or, alternately, four distinct senses of the word “con- scious”) that a satisfying theory of consciousness should seek to explain. These four phenomena are (1) creature consciousness, (2) transitive con- sciousness, (3) state consciousness, and (4) phenomenal consciousness. Creature consciousness 14.4 This first use of the word “conscious” is used to describe an entire organism, as opposed to a state of an organism, as being conscious. Near synonyms are “awake” and “aware.” When we describe a person or a nonhuman animal as being unconscious—as when they are asleep or anesthetized—we are describing them as lacking creature consciousness. The main thing it seems to mean to distinguish a conscious creature from an unconscious creature is to say that the conscious creature is awake and responsive to stimuli whereas an unconscious creature is neither awake nor responsive. Transitive consciousness 14.5 The transitive sense of the word “conscious” is marked by uses wherein we say that someone is conscious of something. In seeing something you become conscious of it. Hearing something is another way of being con- scious of something. Arguably, another way of being conscious of some-  Consciousness and Qualiaâ …â … 205 thing is by thinking about it. Unlike the creature use of the word “conscious,” which is an adjective applied to a creature, the transitive use of the word “conscious” is used to refer to a relation between a creature and the thing that the creature is conscious of or aware of. It should be noted that the relation of transitive consciousness may 14.6 simply be a relation in the grammatical sense without being a relation in the metaphysical sense. If thinking about something is not really a relation to a thing thought about (a hypothesis we discussed in the chapter 13), then in thinking about something one need not really, metaphysically, be related to an actual thing. State consciousness Like the creature use of the word “conscious,” the state use of the word is 14.7 adjectival. However, instead of being used as an adjective that applies to an entire creature, in this use it is an adjective applied only to a state of a creature, in particular, a mental state of a creature. If you are familiar with the basic idea of the Freudian unconscious, then 14.8 you may already be familiar with the basic idea of state consciousness. Sigmund Freud hypothesized that many of our behaviors were due to unconscious beliefs and unconscious desires. Since a belief or a desire is a mental state, in describing a belief or a desire as being conscious or uncon- scious, we are employing a concept of state consciousness. The difference between conscious mental states and unconscious mental 14.9 states is important in understanding certain phenomena such as subliminal perception. In examples of subliminal perception, a stimulus is presented to a subject in such a manner that the subject does not report having seen it. However, there is still evidence that the subject did indeed see it. For example, if forced to guess about what they saw, they are better than chance in guessing the right answer. In such cases, then, it seems that the subjects have perceptual states that are unconscious perceptual states. Ordinarily, when we perceive something, we are able to report that we’ve perceived it. Arguably, in cases in which the perceptual states can be reported by the subject who has them, the perceptual states are conscious perceptual states. Phenomenal consciousness The notion of phenomenal consciousness is largely a technical notion that 14.10 has more to do with certain debates in philosophy than with an established  206â …â … Consciousness and Qualia usage of “conscious” in ordinary speech. The idea of phenomenal con- sciousness is an idea that encompasses such philosophical ideas as the idea of the phenomenal character of experiences, the idea of qualia, and the idea of what it is like to be certain beings or what it is like to be in certain states. 14.11 Arguably, the phrase “phenomenal consciousness” is either a technical way of talking about creature consciousness or a technical way of talking about state consciousness. The key to seeing this point has to do with the way that phenomenal consciousness is connected in the literature with the “what it is like” phrase. One use of the “what it is like” phrase has to do with creatures, as in Nagel’s famous question about whether objective science can ever explain what it is like to be a bat. Another use of the “what it is like” phrase has to do with states of people, as in the question central to Jackson’s knowledge argument, “Can someone who has never had a red experience know what it is like to have a red experience?” The question clearly concerns a kind of mental state, the state of having an experience of redness. 14.12 Despite being a technical usage without a clear connection to common- sense usages of the term “conscious,” “phenomenal consciousness” is often used by philosophers to pick out what they consider to be the most philo- sophically important and vexing aspect of consciousness. The oft discussed “hard problem of consciousness” (to use the phrase widely associated with the philosopher David Chalmers) is a problem concerning whether any explanation can be given of phenomenal consciousness, especially explana- tions in terms of physical or functional properties. 14.13 We turn now to examine two theories of consciousness. The first is a version of a higher order representational theory of consciousness, David Rosenthal’s higher order thought theory of consciousness. The second is the first order representational theory of consciousness, especially the version originating with the philosopher Michael Tye—his “PANIC” theory of consciousness. Rosenthal’s Higher Order Thought Theory of Consciousness 14.14 The gist of the higher order thought theory of consciousness, hereafter the HOT theory of consciousness, is that a state is conscious when one has a certain kind of thought, a higher order thought, about it. Recall our discus-  Consciousness and Qualiaâ …â … 207 sion of higher order mental states from chapter 13. What it means to say of a thought that it is “higher order” is that it is a thought about a mental state. A thought about something that is not a mental state, for example, a thought that there is a cat on the mat, is a first order thought. If you think to yourself, “I am thinking about a cat on a mat,” then, in thinking that, you thereby have a higher order thought. In particular, you have a second order thought. A third order thought is a thought about a second order thought. Of the four kinds of consciousness discussed earlier in this chapter— 14.15 creature consciousness, transitive consciousness, state consciousness, and phenomenal conscious—the kind that the HOT theory primarily seeks to explain is state consciousness. HOT theory seeks to explain what makes some mental states conscious, and what differentiates conscious mental states from unconscious mental states. One of the main ideas that the HOT theory relies on is the commonsen- 14.16 sical idea that when you have a conscious mental state, you are conscious of that mental state. Recall that when someone is consciousness of some- thing, this is transitive consciousness. Thus, in seeking to explain state con- sciousness in terms of being conscious of a mental state, HOT theory is seeking to explain state consciousness in terms of transitive consciousness. According to HOT theory, the crucial way in which we are conscious of our own mental states is by thinking about them. So, since a thought about another mental state is a higher order thought, the HOT theory is explain- ing state consciousness by reference to higher order thoughts. It might seem, at least on the surface, that HOT theory is explaining 14.17 consciousness in terms of consciousness, and is thus offering a circular explanation. However, according to HOT theory, the explanation is not circular because it is explaining one kind of consciousness, state conscious- ness, in terms of a different kind of consciousness, transitive consciousness. Since it is not explaining one phenomenon in terms of itself, it is not a circular explanation. One of the key features of the HOT theory that makes it noncircular can 14.18 be put this way: A state in virtue of which one is conscious of something need not itself be a conscious state. Take, for instance, a first order thought or perception that the cat is on the mat. Suppose that this first order state is unaccompanied by any higher order thought. According to HOT theory, in such a situation one would be transitively conscious of the cat on the mat without that state, the thought that the cat is on the mat, itself being a conscious state.  208â …â … Consciousness and Qualia 14.19 According to the HOT theory, you can be unconsciously conscious of things. For instance, you can be unconsciously conscious of the cat on the mat. Another way of putting this point is to say that, in cases of subliminal perception where one has a percept without being conscious of the percept, one is nonetheless conscious of whatever it is that one perceives. In such a case one is conscious of something without being conscious of being con- scious of the fact that one is conscious of it. 14.20 So far we have seen how the main thing that HOT theory is trying to explain is state consciousness. But what about the three other kinds of consciousness that we mentioned—phenomenal consciousness, transi- tive consciousness, and creature consciousness? How are these kinds of consciousness to be explained? Though these have not been the main focus of HOT theory, HOT theory can nonetheless be interpreted as having something to say about the remaining three phenomena of consciousness. 14.21 Phenomenal consciousness is closely associated with the notion of what it is like and the notion of there being something it’s like to be such-and-such a creature or to be in such-and-such a mental state. Arguably, what it is like is a matter of the subjective appearance of your mental states—that is, it is a matter of how your mental states appear to you. Also arguably, what it is like is fully determined by the content of a higher order thought. So, if I have a higher order thought the content of which is I am seeing a red square, then, what it is like to be me at that moment is, from my own point of view, just as if I was seeing a red square. Of course, I may not actually be seeing a red square—perhaps there is no red square to see. But if I think that I am seeing a red square, then this is how my mental life will appear to me. This higher order thought, then, determines what it is like to be me at that moment. 14.22 What about transitive consciousness? Implicit in the way we’ve described HOT theory so far is that transitive consciousness is explained in terms of certain kinds of representational states, such as thoughts and perceptions. So, if you have a thought or a perception of a cat being on a mat, you thereby are transitively conscious of the cat being on the mat. In short, transitive consciousness and intentionality are pretty much the same thing. 14.23 Regarding creature consciousness, if being creature conscious is simply a matter of being responsive to stimuli, and being responsive to stimuli just is a kind of disposition to be conscious of stimuli, then we have, at least implicitly, an explanation of creature consciousness in terms of transitive consciousness. However, as already mentioned, explaining aspects of con-  Consciousness and Qualiaâ …â … 209 sciousness aside from state consciousness is not really the main concern of HOT theory. An objection to the HOT theory: Introspectively implausible One objection to the HOT theory of consciousness is that it is introspec- 14.24 tively implausible. It is far more plausible, says the objection, that we often, if not most of the time, have conscious states without being aware of those states. Another way of putting this point is to say that (1) the core of the HOT theory is the claim that all conscious states are occasions in which we are self-conscious and (2) it is introspectively implausible that we are self- conscious each and every time that we have a conscious state. The HOT theory has a convincing response to this objection. According 14.25 to this response, the reason that we find the objection plausible is that we seldom notice any higher order thoughts accompanying our conscious states. Further, the higher order thought theory has an explanation of why we seldom notice such higher order thoughts. In order to notice one of your own second order thoughts, the noticing itself would constitute a third order thought, a thought about the second order thought. However, says this response, it is rare that we ever have thoughts of a higher order than the second order. Therefore, it is to be expected on the higher order theory that we seldom notice the presence of higher order thoughts. That we seldom notice such thoughts, then, does not provide a strong objection against the HOT theory. Let’s turn to some other objections. The next two objections hinge on versions of the claim that the HOT 14.26 theory is too intellectual—it focuses too much on the intellectual part of the mind and, in so doing leaves out aspects of consciousness that are less intellectually oriented and have more to do with, for example, nonintel- lectual, sensory consciousness. Another objection to the HOT theory: Too intellectual One version of the “too intellectual” objection focuses on babies and non- 14.27 human animals. Both of these kinds of creatures plausibly have conscious states. However, goes the objection, it is implausible that these creatures are cognitively sophisticated enough to be able to have higher order thoughts. Arguably, in order to think some thought you must grasp certain concepts— concepts that need to be learned. For instance, in order to think that a cat  210â …â … Consciousness and Qualia is on a mat, you must grasp a concept of cats as well as a concept of mats. Further, such concepts are implausibly concepts anyone is born grasping and thus must be learned. Even further, in the case of animals there are many concepts that they simply will not be able to learn. For example, a dog is unlikely to ever be able to learn the concept of an MP3 or grasp the concept of what Wednesdays are. 14.28 This objection continues by saying that, in order to have higher order thoughts, one must have a concept of mental states (how else could one think of mental states?). However, it is implausible that either babies or nonhuman animals grasp such a concept. 14.29 This version of the “too intellectual” objection says that (1) the HOT theory imposes concept possession requirements on babies and nonhu- mans that they can’t satisfy even though (2) many babies and nonhumans clearly have conscious states. Stating the objection this way makes clear that the HOT theorist has two general ways to respond to the objection. The first way is to argue against (1), perhaps by arguing that babies and non- humans can satisfy the conceptual requirements on having higher order thoughts. The second way is to argue against (2), perhaps by presenting reasons for thinking that b