Broome on the Connection between Normative
Beliefs and Motivation1

---

Acerca de la conexión entre creencias normativas
y motivación en Broome

Carlos Núñez2

Summary: I. The Motivation Question. II. Enkratic Reasoning. III. Acknowledgments. IV. References.

I. The Motivation Question

John Broome says that his main goal in Rationality Through Reasoning is to give an answer to what he calls the “Motivation Question”.3 The question is the following: “When you believe you ought to do something your belief often causes you to intend to do what you believe you ought to do. How does that happen?”.4

Broome’s answer is this: at least in the philosophically interesting cases, what happens is that you reason your way from the belief that you ought to do something to the intention to do it. You bring yourself, through reasoning, to intend to do what you believe you ought to do. How exactly does this happen?

The idea, roughly, is that your belief that you ought to do something (together with the belief that doing so is up to you —I will avoid mentioning this belief from now on, but it should be taken as given) causes you to acquire the intention to do that thing, through a process whereby you operate on the marked contents of your beliefs, following a rule, to construct the marked content of your intention.

The relevant rule is the following:

From

<I ought that p; belief>

(and <It is up to me whether or not p; belief>)

to derive

<p; intention> (p. 290)

And for the process to count as a case of reasoning whereby you are following this rule, it must be true of you both that you are disposed to behave in accordance with the rule, and that doing so would seem right to you.

Broome contrasts this answer to the Motivation Question to Michael Smith’s attempts to explain why it is that normative beliefs imply motivational states,5 and says also that it is intended to remove one of the grounds for noncognitivism.6

However illuminating I find by Broome’s account, I am perplexed by this way of framing the project. It suggests to me that he is trying to address the very same question that people working in metanormativity are trying to address when they attempt to account for what has come to be known as “judgment internalism”. This, roughly, is the thesis that (certain) normative judgments necessarily imply (certain) motivations (though the specific flavor of the necessity involved might vary according to theories). The question such theorists are trying to answer is, roughly, why it would be that normative beliefs imply motivations. Call this “the question for Judgment Internalism”. Broome’s answer to the Motivation Question explains how it is that normative beliefs often lead to motivational states. They do so through reasoning. I am in broad outlines convinced by Broome’s answer to this question. But I fail to see how it could serve also as an answer to the question for Judgment Internalism. This makes me unsure about what exactly the theoretical and dialectical target of Broome’s account is. I will concentrate on this issue in the following comments.

II. Enkratic Reasoning

Broome tells us how it is that, often, when you believe you ought to do something, you intend to do it. The answer is that you reason your way to that intention, and he explains how such reasoning —“enkratic reasoning”— could be something you do.

Suppose one is convinced by Broome’s account. Still, one could wonder why it is that such a belief would lead you to do this. In other words, granting Broome that this is something you do, one could still wonder why you would do it. After all, not all beliefs are like this. In fact, many people think normative beliefs are unique in this respect: they think that it is only such kind of beliefs that, on their own and without the aid of further motivational states, would reliably lead you to reason from them to intentions. Why do they do that? What it is about these beliefs, exactly, that would reliably engage the type of reasoning for which the natural and appropriate conclusion would be an intention?

There are two —no doubt related— issues here. On the one hand, one may want to know why it is that it would be correct reasoning for you to derive that intention from that belief. On the other hand, one may want to know why your deriving that intention from that belief would be normal reasoning, a form of reasoning we could reliably expect. Let me focus on each of these issues in turn.

1. Why is Enkratic Reasoning correct reasoning?

Why would it be appropriate to reason from the belief that you ought to do something to the intention to do it? Why do these beliefs sustain such inferences?

Broome would point here to the permission of rationality that makes this kind of reasoning correct: the Enkratic Permission. So he would say (roughly) that it would be correct to derive that intention from that belief because rationality allows you to derive that intention from that belief. But why does rationality allow this?

Broome suggests that perhaps rationality allows you to reason in this way because rationality requires (roughly) that you do not both believe you ought to do something and not intend to do it —this is the requirement of Enkrasia— and by reasoning in this manner you would bring yourself to comply with this requirement.7 The story cannot exactly be as simple as this —as Broome himself explains—8 but for my purposes it doesn’t matter. Suppose something roughly like this is true. Suppose it is appropriate to reason enkratically because rationality allows you to do so, and that rationality allows you to do so because (roughly) by reasoning in this way you can bring yourself to comply with what rationality requires. The question, then, would be: Why does rationality require this? Why is it irrational not to intend to do what you believe you ought to do?

Broome in general doesn’t tell us what makes a requirement one of rationality. This, for the most part, is perfectly fine. But, depending on what exactly his metanormative ambitions in Rationality Through Reasoning are with respect to the problem of explaining the connection between normative beliefs and motivation, the lack of such an account when it comes to the specific requirement (and so, indirectly, the permission) related to Enkrasia might be a bit more problematic.

One reason is simply that this would leave a central aspect of that relation in the dark. There seems to be a necessary connection between normative beliefs and motivation; this is the source of the problem. But there are two ways of understanding the precise nature of the necessity involved: some think it is conceptual or metaphysical, others think it is rational. Those who think that the nature of the necessity is conceptual or metaphysical need not concern themselves with the question of why it would be irrational to have the relevant normative belief and lack the relevant motivation, since they may think that it is altogether impossible to be in such a state of mind. But for those who think that the nature of the necessity involved is rational —and Broome is obviously among them— this question becomes central, at least in so far as they are trying to provide a general account of the connection between normative beliefs and motivation.9

Now, obviously, Broome need not say anything about this issue. He may not be trying to provide a general account of the connection between normative beliefs and motivation. He may be a quietist about why it would be irrational to be akratic, and so not be in the business of offering a deeper explanation of why it would be rational to reason enkratically. This would leave a central aspect of that relation in the dark, but not everyone needs to tackle every question.

At the same time, however, this obscurity may make it harder to say something informative about the other central aspect of this relation, which would seem to be the focus of Broome’s account. The thought here would be that perhaps by explaining why it would make rational sense to reason enkratically one could begin to explain why it would be normal for people to do so. I turn to this issue now.

2. Why is Enkratic Reasoning normal reasoning?

Why is it that, often, when you believe you ought to do something, you reason your way to an intention to do it?

There are two natural strategies for going about trying to answer this question: one consists in saying something informative about the content of these beliefs, the other consists in saying something informative about their nature.

The most obvious way of pursuing the first strategy would be to say something about the content of the ought-belief such that it would hook up with the content of preexisting motivations one can reliably expect agents in general to possess, so that we could understand why, when an agent considers or brings to mind her ought-belief, her preexisting motivations would lead her to reason her way to an intention.

A very clear example of this strategy is pursued by contextualist theories in metanormativity. Here is one toy-version of the theory: to believe you ought to do something is to believe that doing it is required by the norms you accept.10 Since we can reliably expect agents to be motivated to act in accordance with the norms they accept (because, the idea would be, to accept a norm is in part to intend to act in accordance with it) we can reliably expect that an agent who brings to mind her belief that she ought to φ would be motivated to intend to φ. (She would be motivated to intend to φ because she believes φ-ing is required by a norm to which she intends to conform). This would explain what it is about the belief that you ought to do something that would lead you to reason in a way that concludes in an intention to do it. It would do so by saying something about the content of the belief that hooks up with motivations we could reliably expect agents to possess.11, 12

The other natural way of trying to answer this question is to say that there is something about the nature of these beliefs that would explain why they would lead an agent to reason from them to intentions.

A noncognitivist might try to do exactly this. To do so, she must not identify the so-called “belief” that you ought to do something with the intention to do it. Otherwise there would be no metaphysical space to reason from the belief to the intention. So this strategy might be closed to someone like Allan Gibbard, who seems to think that what it is to believe you ought to do something is simply to intend to do it.13 But other forms of non-cognitivism might allow for the relevant conceptual or metaphysical space. For example, suppose the idea were that to judge you ought to do something is to desire most to do it, or to desire most to intend to do it, or something along such lines. Surely, there is reasoning that moves from desires to intentions. Surely, this can be explained by the nature of desires themselves. Plausibly, part of what it is to desire is to be disposed to form intentions in certain conditions. So surely there must be some reliable connection between desiring most to do something, or between desiring most to intend to do something, and reasoning your way to an intention to do it. If so, then this might begin to explain why, often, when you believe you ought to do something, you are led to reason from that belief to the intention to do it.

Now, as far as I can tell, Broome doesn’t address the question I am now considering (and this lies at the root of my perplexity with his way of framing the nature of his project). In any case, it is not immediately obvious to me that he does pursue, or would pursue, any of the two strategies considered above. For sure, he does not think there is anything peculiar about the nature of the ought-belief that would explain why it would reliably lead an agent to reason from it to an intention. He thinks these are cognitive states like any other belief. In fact, as I’ve said, he thinks his answer to the Motivation Question removes one of the grounds for non-cognitivism.

So he would most likely say that the explanation for why the ought-belief would lead an agent to reason from it to an intention must lie in its content. At the same time, however, Broome thinks that there is no non-circular way of saying what these beliefs are about. What you believe, when you believe you ought to do something, is simply that you ought to do it. So the explanation of why these beliefs would lead you to reason from them to intentions will not go by way of a straightforward analysis or elucidation of what these beliefs are about.

Nevertheless, although he thinks there is no non-circular way of saying what these beliefs are about, he does think there are important things we can say about their content. There is one feature that is crucial for our purposes, and it is that the concept of ought that these beliefs are concerned with is the same concept that the requirement of Enkrasia applies to.14 We can put the point this way: what you believe when you believe that you ought to φ, is that for which the following holds true: that it would be irrational for you believe this, and yet not to intend to φ. In other words, what Broome says about the content of this belief is that it concerns the sense of “ought” for which it is true that rationality requires that, if you believe that you ought, in that sense, to do something, then you intend to do it.15

Could Broome use this feature of the content of the belief to explain why it would lead you to reason your way to an intention?

One idea suggests itself naturally. Since it would be irrational to believe you ought to φ and not intend to φ, and since we are rational creatures with tendencies to conform to what rationality requires, among which are certainly tendencies to conform to such requirements through reasoning, we could simply appeal to a tendency to reason so as to conform to this requirement. The explanation why, when you believe you ought to φ, you would tend to reason your way to the intention to φ, is that you are a rational creature with tendencies to conform to what rationality requires, and by reasoning in this way you would conform to one of its requirements.

Now, there is no doubt that this, or something roughly like this, is true. The question is how much weight it would pull as an explanation. By Broome’s own standards, however, it seems like this wouldn’t be a very satisfactory account. Broome says that one could answer the Motivation Question by pointing to a brute disposition —the Enkratic Disposition— to intend to do what you believe you ought to do. And he objects that this answer would be very thin: that it would leave much to be explained.16 Similarly, if one answered the present question —this time not about the connection between the ought-belief and the intention, but about the ought-belief and the reasoning that leads to the intention— by simply pointing to a disposition to reason enkratically —call this the Enkratic Reasoning Disposition— then one might worry that this would not be very informative; that it would leave much to be explained.

To be clear, this is not to deny that much would have been explained already if one had explained, as Broome has done, how you could come to intend to do what you believe you ought to do through reasoning, and how such reasoning could be something you do. It is in relation to the further question of why the ought belief would lead you to reason in the way Broome describes that a mere appeal to a rational disposition to reason enkratically might not be very informative.

There are two related aspects in which an appeal to such a disposition might leave much to be explained. On the one hand, we still don’t know why it is irrational to believe you ought to do something and yet not intend to do it. This is the point I raised in the previous section. Because of this, we still don’t know why it would make sense for you to reason your way to that intention. Perhaps if we knew why it would make sense for you to reason in this way we could begin to understand why you would do it, and so why this tendency obtains at all. But we do not.

The second aspect in which an appeal to this disposition might leave much to be explained is that the reason why it obtains could be, as far as the account goes, extraneous to the nature, or functional role, of the attitudes involved. This doesn’t necessarily mean that the disposition would be left unexplained: it would be explained, to a certain extent, by an appeal to a more general disposition to conform to what rationality requires and by the fact that it requires that you intend to do what you believe you ought to do. But —contrary to what the other broad types of account considered above achieve— it would not show how such a disposition emerges, in an expected manner, from the natural interplay of the attitudes involved, given the functional roles we attribute to them.

It is inevitable at this point to bring attention to the fact that normative beliefs seem to be unique in this respect. It just doesn’t seem to be part of the functional role of ordinary, non-normative beliefs that they would lead agents to reason from them to intend to do anything at all, at least not by themselves, in the absence of further motivational states. It is certainly part of the nature of ordinary beliefs that they sustain specific inferential patterns, and so that they would, by themselves, lead agents who attended to them to reason to other attitudes in specific ways. Plausibly, for example, to believe that everyone in the room is a conspirator and that Jones is in the room is in part to be disposed to derive the belief, if one considered the issue, that Jones is a conspirator. So there is no question that it is part of the ordinary functional role of beliefs to lead agents to reason from them to other attitudes. The issue is simply that all these patterns would lead to further beliefs, not intentions.

Obviously, there are plenty of beliefs that would reliably lead you to form intentions if you reasoned from them. Here is an example: the belief that a lion is approaching and that if you do not run it will eat you alive. Surely —when things do not happen automatically, as they most likely would in such cases— this belief would typically engage a reasoning process that would lead you to form the intention to run. But the explanation for why this happens would point to the way in which the belief would hook up with motivations we could reliably expect agents to possess —for instance, the desire not to be eaten alive— in a way that would understandably lead the agent to reason her way to the intention. Plausibly, to desire not to be eaten alive is in part to be disposed to intend to run if you believe, and call to mind, that a lion is approaching and that if you do not run it will eat you alive. This would explain why such a belief would typically lead you to form an intention if you reasoned from it. The desire not to be eaten is an essential part of this explanation, and something analogous is true of any other familiar form of reasoning that concludes with an intention.

Broome thinks that this is very evidently not true. He thinks we are familiar with plenty of cases where beliefs without the aid of other motivational attitudes would lead to intentions. He presents one of them as a counter example to the Humean Theory of Motivation. So perhaps it is to such cases we should turn for guidance.

Following Jay Wallace, he calls the idea that “a motivation could not be derived by reasoning from an attitude that does not in some way already incorporate a motivation the principle of ‘motivation-out–motivation-in’”. He thinks this principle is implausible because, as he puts it, “intentions can be caused in all sorts of ways”:

You can wake up with a new intention, and you can get a new intention by hypnosis or a knock on the head. It is easy to construct a story in which a genuine belief causes an intention through a mental process. Here is one. Your habit is to walk the dog at 11 o’clock every day. You now believe it is 11 o’clock, and this causes you to intend to fetch the leash. This process
is automatic. You need have no desire to walk the dog…17

Now, I do not doubt that intentions can be caused in all sorts of ways, but the issue is whether beliefs can reliably lead agents to reason from them to intentions, without the aid of further motivational states, and the case Broome presents doesn’t show this. This is because it relies on the presence of a state that is itself motivational: a habit. I am not sure what the right analysis of a habit is. But it is certainly a motivational state (plausibly composed itself of more basic motivational states). In any case, the habit plays an essential part in the explanation for why you would form the intention to fetch the leash. You have a habit of walking the dog at 11, which means you are motivated to do so. You form the belief that it is 11, and this belief, together with your motivation to walk the dog at 11, leads you to form the intention to fetch the leash. There is nothing in this case that would challenge the motivation-out–motivation-in principle. Take away the habit, and we are left without an explanation why the agent would form the intention at all.

Broome specifies that in this case you do not have a desire to walk the dog, and that’s o.k., as long as we understand the term “desire” to denote a particular species of motivation, one among many possible kinds, and not as a term that refers to motivational states in general, for your habit certainly constitutes a motivation.

That ordinary beliefs do not need desires, in this restricted sense of the term, to motivate an agent is something that Humeans need not deny and have never denied. Your belief that there is a spider on the wall might motivate you to form the intention to run if you have a phobia of spiders even if you do not, in the narrow sense, desire to run or desire anything that would be served by your doing so. The Humean would gladly accept this case, as she would gladly accept Broome’s own. They do not challenge her theory of motivation.

Naturally, doing anything (including forming an intention) implies that one is somehow disposed to do it. So the Humean cannot simply say that a disposition to do something is, or implies, a motivational state. This would vacate her claim of any substance, since it would imply by definition that anything one ever does (including forming an intention) implies a previous motivation to do it. The question, then, of whether a belief on its own could reliably motivate an agent to reason from it to an intention, as I understand it, is the question of whether or not the agent’s reasoning in that way can be the expected outcome of the belief’s playing its own characteristic functional role, or whether we would need to add to the psychological breeding ground that leads to the intention the causal impact provided by the normal functioning of some other state —in particular, a motivational state— that is independently identifiable and that already plays an explanatory role in our broader theory of human thought and action.

Obviously, an objector to the Humean theory might simply claim at this point that it is part of the characteristic functional role of normative beliefs that they lead agents to reason from them to intentions. But the question, precisely, is why this would be so. That it isn’t true of beliefs in general that they have it as their characteristic functional role to lead agents to reason from them to intentions in the absence of further motivational states is, I take it, uncontroversial. Why, then, would things change when beliefs take this particular kind of content?18, 19

This is the question people working in metanormativity are trying to answer, and as far as I can tell, it is a question that would remain even if one were to accept Broome’s answer to the Motivation Question. This why I am unsure about the exact target (both theoretical and dialectical) of Broome’s account.

The fact that Broome contrasts his own account to Smith’s, and that he says it is intended to remove one of the grounds for non-cognitivism, suggests to me that he means to address the question metaethicists (or “metanormativists” more generally) are asking, and that he regards his own answer as somehow better suited to the task. As far as I can tell, however, Smith, the noncognitivists, contextualists, and others, could be thoroughly convinced by Broome’s account and still think that the question they are asking remains largely intact. The question, roughly, is: what is it about normative beliefs that would explain why they imply motivations? Broome explains that it is through reasoning that the relevant normative beliefs lead to the relevant motivations. The reason you often intend to do what you believe you ought to do is that you infer such intentions from such beliefs. But what is it about these beliefs that would sustain such inferences? What is it about these beliefs that would explain why you would be led to reason from them to intentions in this way?

My guess is that people like Smith or the contextualists would offer pretty much their own story about the content of normative beliefs, and that non-cognitivists would offer exactly their own story about the nature of normative beliefs, as an answer to this question. So, although I believe Broome’s answer to the Motivation Question is genuinely informative and plausibly right, I do not see how it could be understood as providing an alternative to these other theories, or as removing any of the grounds for non-cognitivism.

III. Acknowledgments

Thanks to Daniel Fogal, Fernando Rudy, Alex Worsnip and other participants at the Broome Seminar in Mexico City in September 2017. Thanks to Juan Vega Gómez for organizing and inviting me to participate in this event. Special thanks to John Broome for his comments, both in conversation and in written form, on an earlier version of this paper.20

IV. References

Broome J, Rationality through Reasoning (Wiley Blackwell 2013).

Dreier J, “Internalism and Speaker Relativism” (1990) 101 Ethics 1, 6–26.

Finlay S, Confusion of Tongues: A Theory of Normative Language (Oxford University Press 2014).

Gibbard A, Thinking How to Live (Harvard University Press 2003).

Hussain NJZ, “Practical Reasoning and Linking Beliefs” (2015) 91 Philosophy and Phenomenological Research 1.

Silk A, “Normative Language in Context” (2017) 12 Oxford Studies in Metaethics.

Smith M, The Moral Problem (Wiley-Blackwell 1994).

———, Ethics and the A Priori (Cambridge University Press 2004).

Wedgwood R, “The Metaethicists’ Mistake” (2004) 18 Philosophical Perspectives 1, 405–26.