1
Directions of Fit, Motivation, and Instrumentalism
A Humean conception of motivation and practical reason comprises two claims. The first, motivational non-cognitivism, holds that our actions are motivated by non-cognitive states, specifically desires. The second, instrumentalism, holds that practical reasoning consists only in figuring out how to satisfy the desires one happens to have; it does not consist in figuring out what to desire in the first place. The apparent upshot is that the desires that motivate our actions and provide the inputs into our deliberations cannot themselves be evaluated in terms of either their truth or their rationality.
Direction of fit accounts of beliefs and desires are often taken to provide support for the Humean position. The idea is that these two kinds of mental state have different directions of fit: roughly, whereas our beliefs are supposed to fit the world, the world is supposed to fit our desires. My aim in this paper is to analyze the notion of direction of fit to determine the extent to which it can support the Humean position. I will argue that, given a proper analysis, a direction of fit account of mental states supports motivational non-cognitivism, but does not support instrumentalism. Even if motivating states cannot be evaluated in terms of their truth, it may still be possible to evaluate them in terms of their rationality.
1.
We may begin to get an intuitive grasp on the notion of a mental state’s direction of fit by considering an example from G. E. M. Anscombe. She considers a man with a shopping list and a detective following him, recording what he buys. The difference between them, she claims, is that:
If the list and the things the man buys do not agree, and if this and this alone constitutes a mistake, then the mistake is not in the list but in the man’s performance (if his wife were to say: ‘Look, it says butter and you have bought margarine’, he would hardly reply: ‘What a mistake! we must put that right’ and alter the word on the list to ‘margarine’); whereas if the detective’s record and what the man actually buys do not agree, then the mistake is in the record.[1]
Each list may be characterized in terms of its direction of fit. There is a ‘fit’ between the list and the world just in case the items on the list match the items the shopper has bought. But lists have opposite directions of fit: while the items purchased are supposed to match the shopper’s list, the detective’s list is supposed to match the items purchased. Adopting Searle’s terminology, we may say that the shopper’s list has a world-to-minddirection of fit, whereas the detective’s list has a mind-to-worlddirection of fit.[2] A direction of fit account of mental states claims that types of mental states, usually beliefs and desires, can be distinguished according to whether they have a world-to-mind or a mind-to-world direction of fit.
Though Anscombe’s example is suggestive, the notion of direction of fit still needs unpacking. We have seen what it means for a list and the world to fit one another, and an obvious way of extending the idea is to say that there is a ‘fit’ between a mental state and the world just in case the state’s propositional content is true. We also noted that part of what it means to say that the detective’s and the shopper’s list have opposite directions of fit is that the former, unlike the latter, is mistaken if there is no fit between it and the world; analogously, we might say that beliefs and desires have different directions of fit insofar as the former, unlike the latter, are mistaken when they are false. We could then say that having a mind-to-world direction of fit means being truth-evaluable, whereas having a world-to-mind direction of fit means not being truth-evaluable.
This account of direction of fit is attractive to Humeans insofar as it allows them to abandon Hume’s own claim that desires lack representational or intentional content altogether while retaining the idea that they are non-cognitive.[3] However, the account of direction of fit we have developed so far is insufficient for Humean purposes. For, although it explains the sense in which desires are non-cognitive, it does not support the view that they are necessary for motivation, and so it does not support motivational non-cognitivism. Moreover, it has no obvious implications for a theory of practical reasoning, and so it does not seem to support instrumentalism.
Even if we set Humean aspirations aside, we can see that the account given so far is inadequate. It explains the mind-to-world direction of fit in normative terms; to say that beliefs are supposed to fit the world is simply to say that they are supposed to be true. Yet it does not provide an account of the world-to-mind direction of fit in normative terms, and offers no explanation of the thought that the world is supposed to fit our desires. This problem is particularly important if we allow, as we should, that there are intentional states that simply don’t have any direction of fit. It is plausible, for example, to suppose that my imaginings need not fit the world, nor need the world fit them. If we wish to distinguish desiring from imagining, we need to be able to make sense of the thought that there is a norm requiring the world to fit our desires but not our imaginings. But the account we have developed so far does not help us to do this.
A number of philosophers have suggested that beliefs and desires have constitutive aims: while beliefs aim at truth, desires aim at something like their realization or satisfaction.[4] Having a direction of fit can then be understood as having a constitutive aim, and differences in direction of fit can be understood in terms of differences in constitutive aims. At first glance, this suggestion appears to offer us the solution to our problems. For example, we now have a normative account of the world-to-mind direction of fit; just as beliefs may be evaluated in terms of their truth, so desires may be evaluated in terms of their satisfaction. In contrast, while our imaginings may serve various purposes, they are not constituted by any particular aim, and so do not have a direction of fit. Moreover, the suggestion that desires aim at their own satisfaction may help make sense of the Humean claims that they are motivating and that practical reasoning consists at least in part in figuring out how to satisfy them. It is not yet clear that this suggestion is enough to support motivational non-cognitivism or instrumentalism, since it is not yet clear that only desires (and other non-cognitive states) can motivate, or that practical reasoning consists only in figuring out how to satisfy them. But it does advance the analysis of direction of fit in a Humean direction.
Unfortunately, this account of direction of fit, however initially promising it may seem, is ultimately unsatisfying. One problem is that the notions of truth and satisfaction have not yet been made sufficiently distinct. After all, a belief is true when its content is true, and a desire is satisfied when its content is true. So, to say that they aim at truth and satisfaction is simply to say that they both aim at the truth of their content. However, we can avoid this problem by saying that desires aim not at their satisfaction but at some target that is clearly distinct from truth. One possibility is to say that they aim at the good; just as we can evaluate beliefs in terms of the truth of their content, so we can evaluate desires in terms of the goodness of their content.[5]
Nevertheless, an even deeper problem remains, one that confronts any attempt to explicate the notion of direction of fit in terms of mental states’ constitutive aims. The problem is that any such account will only allow us to evaluate the mental states in question: they may be positively evaluated when they achieve their aim, and negatively when they do not. But the intuitions that led us to the notion of direction of fit suggest that a desire may be perfectly acceptable even when the world fails to fit it. To return to Anscombe’s example, a perfectly acceptable shopping list may have ‘margarine’ on it even though the shopper returns home with butter. The idea that the world is supposed to fit our desires suggests that, when it doesn’t fit them, there is a problem with the world, not with our desires. The suggestion that mental states have constitutive aims does not help us make sense of this idea.
Of course, I have not argued that mental states do not have constitutive aims. My point, rather, is that the notion of direction of fit cannot be analyzed in terms of these aims. We will need to find some other way of analyzing it.
2.
It will help to return once more to Anscombe’s example. There is a fit between a list and the world just in case the items on the list match the items the shopper has bought. A lack of fit between the shopper’s list and the purchased items constitutes a mistake in the shopper’s performance; a lack of fit between the detective’s list and the purchased items constitutes a mistake in the detective’s list. In effect, then, the contents of both lists serve as a standard of success, a standard that is met just in case there is a fit between the list and the world. The difference between them is that their contents serve as a standard for different things: the shopper’s serves as a standard of success for the shopper’s performance, while the detective’s serves as a standard of success for the list itself; i.e., it provides its own standard of success.
We might therefore explicate the notion of direct of fit as follows. There is a fit between a mental state and the world just in case the content of the state is true. The state has a direction of fit just in case the truth of its content serves as a standard of success. And different directions of fit may be distinguished according to differences in that for which the state’s content serves as a standard. In particular, a state has a mind-to-world direction of fit just in case the truth of its contents serves as a standard of its own success. Such a state will be successful if and only if it is true. In contrast, a state has a world-to-mind direction of fit just in case the truth of its content serves as a standard of success for the action or performance of the person in the state. The person’s action will be successful if and only if it makes the state’s content true.
Can we get a plausible account of belief and desire using this understanding of the notion of direction of fit? A belief’s content, on this view, provides a standard for its own success, so that it is successful just in case it is true. So this account retains the plausible idea that beliefs are truth-evaluable. This view would also hold that a desire’s content provides a standard for success for one’s actions, so that one’s actions are successful just insofar as they satisfy one’s desires. There is surely something plausible about this idea, but it is not quite right as it stands. After all, sometimes agents have desires that they want to avoid satisfying, such as a desire for a cigarette, and it would be perverse to say that their actions are unsuccessful insofar as they manage to kick the habit.
The problem here lies not in the analysis of direction of fit, however, but in its application to mental states. Even if we don’t have an adequate direction of fit account of desires, we nevertheless have the materials for an adequate direction of fit account of intentions. Such an account would hold that an action is successful just in case it carries out the intention with which it is done, i.e, makes the content of the intention true. For example, if I intend to get cigarettes at the store, my act of shopping is successful if and only if I get cigarettes. And this is so regardless of whether I want my intention to be carried out. For example, temporarily giving in to my cravings, I may go to the store intending to get cigarettes while hoping that they are out of stock. In this case, it would not be perverse to say that my action is unsuccessful in the happy event that they are out of stock; on the contrary, it would be perfectly accurate to say that I am happy precisely because my action is unsuccessful. This example shows that intentions provide a standard of success for our actions in a way that desires do not, and this fact is captured by a direction of fit account of intention.
On reflection, it is not terribly surprising that the notion of direction of fit is better suited to an account of belief and intention rather than belief and desire. After all, a belief is a commitment to the truth of a proposition, just as an intention is a commitment to an action. But a desire, however irresistible it may be, is not itself a commitment to anything. Since they are a type of commitment, intentions are a closer practical analogue to beliefs than are. Moreover, the notion of direction of fit can help explain the sense in which they are commitments. The general idea is that I make a commitment by placing certain standards or norms upon myself. For example, a promise is a commitment because in making it I place myself under a norm, the content of which is given by the content of the promise. Similarly, a belief is a commitment because in forming a belief I place certain standards of success upon my mental state, the content of which is given by the content of the state. And according to a direction of fit account, an intention is likewise a commitment because in forming an intention I place certain norms of success upon my action, the content of which is given by the content of the intention.
We are now in a position to see that this account of direction of fit supports a version of the Humean claim that actions are motivated by non-cognitive states. This is because there is a sense of motivation in which agents are motivated to do an action just in case they have committed themselves to doing it. To see this, consider the following two agents. The first intends to do act A; the second has the very same reasons for and against doing act A as the first, but is still in the process of deliberating and has not yet formed any intentions. There is a sense in which both have the same motives for doing A, since they both have the same reasons. But there is also a sense in which the first agent is motivated to A and the second agent is not, and this is due to the fact that the first, unlike the second, is committed to doing it. Thus there is a sense in which one is motivated to do an act just insofar as one has an intention to do it. But according to our direction of fit account, intentions are non-cognitive: they are not truth-evaluable in the way that beliefs are, since the truth of their contents provides a standard for evaluating the success of action, not the success of the intention themselves. Thus, there is a sense of motivation in which only non-cognitive states are motivating.
3.
We have not yet considered whether the present account of direction of fit supports instrumentalism. Presumably, it is compatible with instrumentalism, and one may even be able to use it in conjunction with other philosophical premises in an argument for instrumentalism. But the issue here is whether one can reasonably accept this account without committing oneself to instrumentalism.
One way of arguing that the notion of direction of fit supports instrumentalism is suggested by Mark Platts:
Beliefs aim at the true, and their being true is their fitting the world; falsity is a decisive failing in a belief, and false beliefs should be discarded; beliefs should be changed to fit the world, not vice versa. Desires aim at their realization, and their realization is the world fitting with them; the fact that the indicative content of a desire is not realized in the world is not yet a failing in the desire, and not yet any reason to discard the desire; the world, crudely, should be changed to fit with our desires, not vice versa.[6]
What is notable about this passage is Platts’ implicit claim that one of reason’s functions is, very roughly, to make whatever adjustments are needed to satisfy the norms that constitute mental states’ direction of fit. Thus theoretical reasoning aims at adjusting our beliefs so that they fit the world, and practical reasoning aims at adjusting the world so that it fits our desires. This in turn suggests that desires supply us with practical reasons: we have reason to perform actions that satisfy our desires.
Of course, Platts appears to employ a constitutive aim account of direction of fit, and we have already seen that this is a mistake. Nevertheless, we may adjust his argument to suit our own account. We could argue that theoretical reasoning aims at adjusting our beliefs in such a way that they meet their standard of success, and that practical reasoning likewise aims at adjusting our actions so that they meet their standard of success. But this, in effect, means that practical reasoning consists in finding ways of carrying out our intentions, and this is more or less instrumental reasoning.