Reasons for action, acting for reasons, and rationality

Abstract

What kind of thing is a reason for action? What is it to act for a reason? And what is the connection between acting for a reason and rationality? There is controversy about the many issues raised by these questions. In this paper I shall answer the first question with a conception of practical reasons that I call ‘Factualism’, which says that all reasons are facts. I defend this conception against its main rival, Psychologism, which says that practical reasons are mental states or mental facts, and also against a variant of Factualism that says that some practical reasons are facts and others are false beliefs. I argue that the conception of practical reasons defended here (i) provides plausible answers to the second and third questions above; and (ii) gives a more unified and satisfactory picture of practical reasons than those offered by its rivals.

1 Introduction

Throughout the paper, I shall be concerned with practical reasons: reasons for acting. Much of what I say applies also to reasons for believing, for wanting, for having emotions, etc. but I shall not discuss those reasons here.¹ In Sect. 2, I introduce and characterize two conceptions of reasons, ‘Factualism’² and ‘Psychologism’, and pose the question which of these is the correct conception of practical reasons. I focus on these because they, or variants of them, are the most plausible candidates and in fact dominate the debates mentioned above. In Sect. 3, I distinguish different kinds of practical reasons about which this question can be posed, explaining that by ‘different kinds of reasons’ I mean simply different roles that reasons play. In Sect. 4, I argue that Factualism is the better conception of practical reasons. In Sect. 5, I examine and rebut several objections to Factualism, including one that depends on claims about the relationship between acting for reasons and acting rationally. The conclusion is that there are good grounds for preferring the Factualist conception of practical reasons.

2 Factualism versus psychologism

The two conceptions of reasons I shall be assessing can be characterized as follows:

Factualism: all reasons are facts;³

and

Psychologism: reasons are mental states of agents.

Prima facie these two positions seem committed to different ontological views about reasons, for the first says that reasons are facts and the second that they are mental states, and mental states, unlike facts, are generally thought to be psychological entities.⁴ Things, however, are not so straightforward because some supporters of Psychologism hold that reasons are psychological facts. This is because Psychologism says that reasons are mental states, such as ‘Joe’s believing that he’s late’,⁵ and there’s only a small step from the claim that someone’s reason is his believing something (a mental state) to the claim that his reason is that he believes something (a psychological fact); that is, from the claim that Joe’s reason for running is his believing that he’s late (a mental state) to the claim that Joe’s reason is (the fact) that he believes that he’s late. Indeed, some writers appear to see no difference between the claim that a reason is a mental state and the claim that a reason is a psychological fact. I think there is an important difference but, for the purposes of this paper, by ‘Psychologism’ I shall mean a view that holds that reasons are either mental states or psychological facts (or both). However, for ease of exposition I shall mainly talk of Psychologism as the view that reasons are psychological facts.

Given that Psychologism includes the view that reasons are psychological facts, it is not clear that Factualism and Psychologism disagree about the ontology of reasons. For psychological facts are not in a different ontological category from other facts: as one might put it, facts are facts, whether they are biological, psychological, political, etc., and hence they all belong in the same ontological category (whatever that is), even if the things those facts are about do not. Thus, a more perspicuous way of characterizing the disagreement between the two views under discussion is the following:

Factualism: Reasons are facts about all sorts of things.

Psychologism: Reasons are psychological facts about the mental states of agents.

Before we move on to examine and assess these alternatives, another clarification is required.

Psychologism is sometimes expressed with the slogan ‘reasons are beliefs’. But, because ‘my belief’ can be used to refer to my believing something, or to what I believe, that slogan has two interpretations:

RBi: Reasons are believings: (the fact) that I believe that p.

RBii: Reasons are what is believed: that p.

The distinction is fundamental. For suppose that one says that the reason for which Othello kills Desdemona is his belief that she is unfaithful. One may mean by this that Othello’s reason is his believing that Desdemona is unfaithful, or that his reason is what he believes, namely that Desdemona is unfaithful.

These are very different claims about what Othello’s reason is, for there is a world of difference between the fact that Othello believes that Desdemona is unfaithful and what he believes, namely that she is unfaithful—which may or may not be a fact. The first is a fact about Othello’s psychology while the other is a fact, or an alleged fact, about Desdemona’s behaviour. And it is clear that it could be a fact that Othello believes that Desdemona is unfaithful without it being a fact that she is, and vice-versa. So the fact that Othello believes something, and what he believes, are quite different things, and if both are reasons, they are quite different reasons. Because of this, the claim that Othello’s reason is his belief needs to be disambiguated into either RBi or RBii—the reason for which he kills Desdemona cannot be both.

Here, I am using the term ‘Psychologism’ to express a commitment to RBi: to the view that reasons are psychological facts, rather than their contents.⁶ So, according to Psychologism practical reasons are someone’s believing something or the fact that they believe it. Factualism, by contrast, is the view that practical reasons are facts that someone believes.⁷

Which, if either, of these conceptions is right? Factualism is widely accepted for normative reasons, reasons that favour doing something. But things become more controversial when the question concerns the reasons for which we act and the reasons that explain our actions. This has led many to conclude that Factualism is right for normative reasons but not for other kinds of reasons. Before I can assess that claim, I need to say something about these different kinds of reasons.

3 Kinds of reasons

Contemporary discussions of reasons often draw a distinction between two types of practical reason: ‘normative’ and ‘motivating’.⁸ In this section I shall argue that it is more helpful to distinguish between three types: ‘normative’, ‘motivating’ and ‘explanatory’. But before doing that, I should add a note of clarification. Different kinds of reasons are sometimes said to correspond to different senses of the word ‘reason’, which is thought to be ambiguous. I do not subscribe to this view and take claims about different kinds of reasons to amount to claims about different roles that reasons can play. In itself and out of context, a reason is not a reason of any particular kind, say normative or motivating. It is only in a particular context, where the reason plays a specific role and can be cited to answer a particular question, that it can be qualified as being of this or that kind.⁹

I start with the traditional distinction between two types of reason and then offer my arguments in favour of adding a third type.

The traditional distinction says that there are normative reasons: reasons that there are for people to act—as it is often put, reasons that ‘favour’ doing something; and motivating reasons: reasons for which an agent acts, that is, the reasons that an agent takes (perhaps rightly) to favour acting as she does and for which she acts. The contrast can be seen in these examples:

• Normative: There’s a reason for Ada to go into college this Saturday, namely that there is an important lecture; and

• Motivating: Ada’s reason for going into college this Saturday is that there is an important lecture.

On this occasion, the reason that favours Ada’s going into college (normative) is the same as her reason for going (motivating). Nonetheless, these are two types of reason, since they answer different questions: whether there is any reason that favours Ada’s going into college, and what reason motivates her to go. My suggestion is that there are in fact three different questions relevant to practical reasons and that we should therefore recognise three kinds of practical reasons:

• Normative reasons: reasons that favour acting.

• Motivating reasons: a reason for which someone acts: a reason someone takes to favour acting and in the light of which they act.¹⁰

• Explanatory reasons: a reason why someone acts.

It might seem that this tripartite classification is unnecessary because a reason for which someone acts, i.e. her motivating reason, can also explain her action: we can explain why Ada is going into college this Saturday by saying that there is an important lecture.¹¹ Therefore, at least in contexts of intentional actions, we do not need to distinguish between motivating and explanatory reasons. Because of this, philosophers tend to assimilate the reasons that motivate actions and the reasons that explain them. However, as I’ve argued elsewhere,¹² I think that we should keep apart the notion of a motivating reason from that of an explanatory reason even though, on many occasions, these roles are played by the same reason. The fact that one and the same reason can be a normative and a motivating reason does not obscure the difference between those two roles that reasons can play. Equally, the fact that one and the same reason can motivate and explain an action should not obscure the difference between those roles, which is rooted in the difference between the corresponding concepts of motivation and explanation. When we are interested in what reasons motivate an agent to act, we are interested in the reasons that the agent takes to favour her action, and in the light of which she acts.¹³ When we are interested in the reasons that explain an agent’s action, we are interested in the reasons that enable us to understand her action, reasons that make her action intelligible.

To be sure, motivating and explanatory reasons are not unconnected, because knowing the reason for which someone acts is (normally) enough to understand her action.¹⁴ And this is no coincidence, since explanations of intentional actions are typically given in terms of the agent’s reason for acting, i.e. of motivating reasons. In our example, the fact that there is an important lecture is Ada’s reason for going and also a reason that explains her action. Nonetheless, the questions of motivation and explanation are distinct questions.

This is clear because the answers to the questions why someone acted and what motivated her to act need not be the same. For instance, the fact that Ada knows that there is an important lecture explains why she goes into college, but that is not the reason that motivates her to go. The reason that motivates her is that there is an important lecture.

Therefore, motivating and explanatory reasons should not be assimilated to each other because, even if the same reason can both motivate and explain, the concepts of motivation and explanation are different. We should, therefore, recognise at least three kinds of reason: normative, motivating and explanatory. This way of categorising reasons, as I shall argue below, enables us to deal with a range of cases that the binary classification cannot accommodate.

With these clarifications about kinds of reasons in place we can now return to our question.

4 Factualism or psychologism?

Our question is: What conception of reasons, Factualism or Psychologism, if either, is right? But it should now be clear that the question needs to be examined separately for normative, explanatory and motivating reasons, respectively.

4.1 As I noted at the end of Sect. 2, the consensus is that Factualism is right for normative reasons. There are examples of psychological facts that can favour acting. For instance, the fact that I believe that I’m being spied upon by light bulbs, or the fact that I want to kill my parents, may be reasons for me to see a psychiatrist.¹⁵ But it is consistent with Factualism that some normative reasons may be psychological facts.

4.2 The consensus breaks down when it comes to other types of reason. Consider explanatory reasons: reasons that explain why someone acts. It seems that explanatory reasons are often psychological facts. For instance, I can explain why I went to the hospital by saying that I did so because I knew my father had been admitted to the Intensive Care Unit.¹⁶ Similarly, we can explain why Othello kills Desdemona by saying that he believes that she has been unfaithful to him. These explanations, which cite psychological facts about what the agent knew or believed, are distinctive because they explain an action by, to use Davidson’s term, ‘rationalising’ it:

A reason rationalizes an action only if it leads us to see something the agent saw, or thought he saw, in his action—some feature, consequence, or aspect of the action the agent wanted, desired, prized, held dear, thought dutiful, beneficial, obligatory, or agreeable (Davidson 1963, p. 685).

There are, however, other ways of explaining actions by rationalising them that do not refer to psychological facts or to mental states. One such is stating the agent’s reason: for instance, I can explain why I went to the hospital by citing my reason for going, namely that my father had been admitted to the ICU—this points to something I saw in the action that made it desirable: e.g. that I could then be with my father. Another way is explanation by re-description: Anscombe’s example in Intention of saying ‘I’m lying on my bed because I’m doing yoga’ illustrates the idea. These explanations do not cite psychological facts, even though they presuppose some such facts. For instance, the first explains only on the assumption that I am aware of my father’s being in hospital; and the second presupposes that I regard lying in bed as an appropriate position to do yoga exercises.¹⁷ Nonetheless, the possibility of explaining actions in these other ways suggests that not all explanatory reasons are psychological facts.

Against that suggestion, a defender of Psychologism for explanatory reasons might urge that these explanations are elliptical and when fully spelled out their explanans contains facts about what I knew or believed, e.g., that my father had been admitted to the ICU, or that lying in bed is a good bodily position for yoga. But are these explanations really ellipses? It seems undeniable that a person cannot act for the reason that p, or on the grounds that p, unless they stand in some epistemic relation to p: they need to believe, know, accept, etc. that p. However, this does not prove that one must embrace Psychologism for explanatory reasons. These psychological facts may be part of the pragmatics of explanation without being part of the explanation itself:¹⁸ perhaps the fact that they know such things is simply a necessary condition for p to be the explanans in a reason explanation; or as Dancy suggests, their knowing may be an ‘enabling condition’ for the explanation (Dancy 2000, p. 127).

However that issue is decided, two things should be noted. First, in ‘error cases’—cases when an agent acts on the basis of a falsehood that he believes—the explanans of a true explanation must be a psychological fact. For instance, the explanation of why Othello kills Desdemona cannot be what he believes, that Desdemona has been unfaithful, but rather the fact that he believes that. This is because explanations are factive: a true explanation cannot have a falsehood as its explanans. But this is consistent with Factualism, since the latter says that explanatory reasons are facts about all sorts of things, including facts about an agent’s mental state, as in Othello’s case. The second thing to note is that, even if Psychologism is right for explanatory reasons (that is, even if all reason explanations cite psychological facts), it does not follow that Psychologism is right for motivating reasons because, as I argued in Sect. 3, these reasons need not be the same.

The view that explanations of action are always factive is widely accepted but there are dissenters. For example, in Practical Reality, Dancy rejects the idea that we need to resort to psychologised explanations when agents act on false beliefs because, he argues, we can always explain an action by specifying the reason for which it was done (Dancy 2000, p. 131ff.). And this, he holds, is true even when an agent acted on a false consideration. The problem with this view is that it commits Dancy to the conclusion that some reason explanations are non-factive. To most philosophers this is an unacceptable conclusion: surely true explanations require the truth of both the explanandum and the explanans.

In a recent paper (2014), Dancy has abandoned his earlier view that reason explanations can be non-factive but he retains his opposition to Psychologism for explanatory reasons. He still maintains that we can always explain an action by specifying the reason for which it was done, even when the ‘reason’ is some falsehood that the agent believed and in the light of which he acted. In those cases, Dancy says,

we can say that what explains the action is that it was done for the reason that p,  without committing ourselves to saying that what explains the action is that p (2014, p. 90).

He adds that in such cases the reason itself ‘need not be the case and does not make the sort of distinct contribution to the explanation that would enable us to think of it as the explanans’ (2014, p. 91). This new suggestion still seems to have rather counterintuitive results: to my ear, ‘Othello kills Desdemona for the reason that Desdemona has been unfaithful to him, although she has not been unfaithful to him’ sounds only marginally less paradoxical than ‘Othello kills Desdemona because of the fact that Desdemona has been unfaithful to him, although she has not been unfaithful to him’. It is not that one cannot give meaningful interpretations to these expressions: we may hear them as conveying what the agent him- or herself would have said if asked about their reason, adding, at the same time, that the agent was mistaken in treating their false belief as a reason to do what they did. All the same, the fact that these expressions have an air of paradox and require this special interpretation are arguably explained by the thought that both ‘the reason that’ and ‘the fact that’ are factive operators.

I shall return to this issue about ‘false reasons’ below. But whether or not I am right about the air of paradox, the point to emphasise here is that the distinction between explanatory and motivating reasons enables one to avoid the difficulties faced by Dancy’s view. For the distinction enables us to see that a reason that explains an action by rationalising it need not be the reason that motivates the agent, and so we can hold that a reason that explains why Othello kills Desdemona by rationalising his action is the psychological fact that he believes that she has been unfaithful, while resisting the conclusion that Othello’s (motivating) reason for killing Desdemona is the fact that he believes that she has been unfaithful—that is, by eschewing Psychologism about motivating reasons.

In short, if one attends to the distinction between the roles of motivation and explanation that reasons can play, as urged in Sect. 3, there should be no temptation to move from Psychologism concerning explanatory reasons in some or in all cases, to Psychologism concerning motivating reasons. To be sure, that does not show that Psychologism is not the right view for motivating reasons—only that it does not follow from Psychologism for explanatory reasons. And this brings us to the third and final question: is Psychologism right for motivating reasons?

4.3 First, since there are some normative reasons that are psychological facts about the agent, there are corresponding motivating reasons that are psychological facts. The example of the light bulbs above will serve to illustrate the point: my reason for visiting the psychiatrist may be that I believe light bulbs are spying on me, which is a psychological fact. This is consistent with Factualism, which allows that some reasons are psychological facts. Nonetheless, it is implausible that all of our reasons for acting are mental states, or psychological facts about those mental states.

Dancy (2000) and others, including myself, have given arguments against Psychologism as a view of motivating reasons. I shall mention here only what I take to be the most telling argument: when we reason about how to act, and when we offer our reasons for acting, the things that we reason from or mention are only occasionally facts about our psychology. Rather, they are (putative) facts about the world, about things and their features, and about how those features relate to each other in various ways: causal, conventional, legal, etc., so that our actions are (or seem to us to be) mandated, recommended or at least permitted by those (putative) facts. Thus, we only occasionally seem to take our reasons for acting to be our mental states or psychological facts about those mental states.

It is true that we sometimes say that our reasons are ‘our beliefs’ but, as I noted in Sect. 2, we need to tread with care here, given the ambiguity of that phrase, for the claim raises the question whether we mean what we believe, or the fact that we believe it. And if we mean the first, as the remarks in the previous paragraph suggest we do, then the claim that our reasons are our beliefs does not support Psychologism.

There is, however, a familiar difficulty for factualism, namely error cases. Consider Othello’s case again. What motivates him to kill Desdemona is his false belief (induced by Iago) that Desdemona has been unfaithful to him. But in fact she has not been unfaithful. What should we say Othello’s reason (the reason that motivates him) is, given that what he believes and is motivated by is not a fact? There seem to be three possible options.

The first option is to embrace Psychologism for error cases and say that Othello’s reason is the psychological fact that he believes that she has been unfaithful. But this seems implausible for the reasons already given: although this reason explains Othello’s action and on account of that it can be termed ‘Othello’s reason’, it is not the consideration that Othello takes to favour killing Desdemona. If it were, his despair when at the end of the play he discovers Iago’s deceit would be puzzling. But his despair is only too comprehensible because at the end he discovers that the consideration that he took to justify, indeed to demand, killing Desdemona and in the light of which he killed her is false. But what he discovers to be false is not that he believed that Desdemona has been unfaithful, but that she has been unfaithful. In other words, what he discovers to be false is precisely what motivated him, namely, the thought that Desdemona has been unfaithful.

A second option rejects both Psychologism and my version of Factualism, and says that motivating reasons are things an agent believes, regardless of whether these are true or false. In veridical cases, those reasons are facts but in error cases those reasons are false beliefs. Call this the ‘false reasons’ position. On this view, Othello’s reason is his false belief that p—note: not his (falsely) believing that p but the false thing he believes, namely that p—in Othello’s case, that Desdemona has been unfaithful. So on this view Othello does act for a reason, which is a false belief.¹⁹

This view has its attractions but I think there are also strong considerations against it. A fairly telling consideration was anticipated above: that stating these alleged reasons together with the speaker’s knowledge of the truth leads to statements with an air of paradox about them. For suppose that I say that Ellie’s reason for pushing you is that you are stepping on her toes but you tell me that you are in fact doing nothing of the kind (perhaps it’s someone else who’s being clumsy). I cannot say, without an air of paradox: ‘Ellie’s reason is that you are stepping on her toes, although you are not stepping on her toes’. As I noted above, the point is not that this form of words cannot be given a sense but rather that it requires comment or interpretation, in the way that it would not if the operator ‘her reason’ were not factive. Consider the contrast with claims about someone’s beliefs. There is no air of paradox whatsoever in the following: ‘Ellie believes that you’re stepping on her toes although you are not.’ This is because, unlike ‘her reason’, the operator ‘her belief’ is not factive.

A related consideration is that this option commits one to problematic views, such as Dancy’s claim that one’s reason for acting may be ‘a reason that is no reason’ (Dancy 2000, p. 3),²⁰ or Hornsby’s view that sometimes we might have to conclude that ‘there was no reason to do what he did, even though he did it for a reason’ (Hornsby 2008, p. 249). Comesaña and McGrath, who also endorse this option, note that their position involves denying that there is

a certain connection between there being reasons and one’s having reasons, viz., that a consideration, p, cannot be a reason one has to do something unless p is a reason there is to do that thing (2014, p. 59).

In defence of their position, however, they argue as follows:

On our view, reasons [one has] are considerations—true or false—that favor or support a person doing (believing, feeling, etc.) something. Moreover, we are inclined to think that having P as a reason to X can be factored into P’s being a reason for one to X and one’s having P (2014, pp. 76–77. Capital ‘P’ in original).

But this is a puzzling claim. For the last sentence of the passage can arguably be paraphrased as follows: ‘A has reason p to X’ can be factored into ‘p is a reason for A to X’ and ‘A has reason p’. But ‘p is a reason for A to X’ seems to imply that there is a reason, namely p, for A to X—and indeed Comesaña and McGrath explicitly accept the implication. One might think that this commits them to the ‘certain connection’ between having reasons and there being reasons (cited above) that they reject at the beginning of their paper. In response, they claim that although ‘A has a reason to X’ implies ‘there is a reason for A to X’, this should be taken to have a ‘weak’ reading of ‘there is’ that ‘merely says that there is such a thing (obtaining or not) as the consideration that supports one’s X-ing’. But ‘there is a reason for A to X’ does not merely say that there is a consideration that supports A’s X-ing: it says that that consideration is a reason. So it is hard to see that they really avoid commitment to the ‘certain connection’ between having reasons and there being reasons they disavow at the beginning of their paper.

The problematic claims just summarized are, if not straightforwardly self-contradictory, at least prima facie paradoxical, as Hornsby acknowledges.²¹ I do not take that to be a decisive argument against the views that generate them but it seems to favour considering whether there is a plausible alternative view that does not commit one to such claims. And this takes us to the third option.

The third option is to say that in error cases an agent acts on something that he takes to be a reason but which he is mistaken about and is in fact not a reason. Call this the ‘apparent reasons’ position. So, in these cases an agent acts for something that is not a reason, although he treats it as one: he acts for an apparent or putative reason.²²

An apparent reason is something that the agent took to be a fact but which was not a fact—it was rather an apparent or putative fact that the agent treated as a reason. And when an agent acts for an apparent reason, the fact that she believes what she does is the reason why she acted (an explanatory reason) and hence it can provide a rationalising explanation (in Davidson’s sense) of her action: Othello kills Desdemona because he believes that she has been unfaithful. The fact that he believes that thing explains his action by rationalising it, and in acting as he does, Othello acts motivated by an apparent reason: the (to him) apparent fact that Desdemona has been unfaithful.

I have presented a third option to the problem of error cases that is consistent with the version of Factualism I am defending, which says that all reasons are facts. The position, however, has been criticised. The fourth and final section of the paper is devoted to exploring the most significant objections to this position.²³

5 Objections to factualism

The idea that acting on a false belief is not acting for a reason has been explicitly rejected by some authors. For example, in the presentation of her disjunctive conception of reasons for acting, Hornsby explicitly rejects this proposal on the following grounds:

Someone might try to avoid saying that agents have ‘double the reason’ by denying that the agent who acts on a belief really has any reason at all for acting. Reasons for acting are all of them facts, this person says: when we know of an agent’s (B)-type reasons, we simply know reasons why they act as they do; but (B)-type reasons are not reasons for which people act. The ‘reason’ of ‘(B)-type reason’ is just the ‘reason’ of explanation.

But this is wrong. Of course (B)-type reasons do figure in reason-why explanations: an explanation is given, for instance, when it is said why Sam is going to Beech Street. And it is true that if Ann is not in the café on Beech Street, then there is actually no reason for Sam to be going there. But that doesn’t show that Sam fails to have a reason to go there. The reason Sam has for going there is known by the ignorant onlooker, who tells us what reason Sam has in telling us what Sam believes (Hornsby 2008, p. 249).

Hornsby here seems to support her claim that the false belief Sam had, namely that Ann is in the café on Beech Street, was a reason he had for going to the café (given that he and Ann had agreed to meet), with the consideration that an onlooker who knows what Sam believes but is ignorant of whether his belief is true knows Sam’s reason. But this does not show that what the thing the onlooker knows (the false belief) is a reason—it presupposes it. It is true that the onlooker knows what Sam believes (that Ann is in the café) and can therefore tell us what consideration motivated Sam to go to Beech Street. The moot point is whether that consideration is a reason Sam had for going to the café or whether, as I am claiming, it is merely an apparent reason he had for doing so. The argument given above does not show that the second suggestion is wrong. In support of her argument, Hornsby also says that

the explanation given of Sam’s going to Beech Street, when it is said what Sam believes, cannot be assimilated to any old reason-why explanation. Perhaps the reason why the bridge collapsed was that it had a structural flaw. Still, the bridge didn’t have a reason to collapse (Ibid).

This is quite right: the explanation of Sam’s going to Beech Street is a rationalization. But that is something that the form of Factualism I am defending can easily accommodate, since rationalisations are explanations that can be given only for things done for real or apparent reasons. Since bridges do not have reasons for acting, real or apparent, we cannot give rationalisations of anything they do. I therefore do not think Hornsby shows that the view under consideration is wrong.

Another objection to the third option is to say that it conflates the concept of a reason with that of a good reason for acting. Turri makes the objection in relation to reasons for believing but the point would carry to reasons for acting:

As already suggested, I assume that it is possible to believe based on a bad reason. (...) True, sometimes people say a bad reason is ‘no reason at all.’ But we understand this to mean that it is not a good reason, just as we might respond to a bad argument by saying ‘that’s not an argument,’ or characterize a mediocre fastball as ‘no fastball at all,’ or say that a defective knife is ‘not a knife.’ There are bad reasons, just as there are bad arguments, mediocre fastballs, and defective knives (2009, p. 493).

Turri is right to point out that there are particulars that are instances, albeit not very good instances of their kind, such as blunt knifes or mediocre fastballs. The question here, however, is whether a falsehood bears that relation to reasons, or whether it bears the relation that counterfeits or fakes bear to their alleged kinds: namely that of passing for a thing of a kind it does not belong to. The distinction is nicely illustrated with a recent story involving Mr Martin Lang and his ‘Chagall’ The Nude, for which he paid £100,000 in 2005 only to discover in 2014 that his painting is not what it looks like, namely, a bad Chagall. Rather, his painting is a fake Chagall, as the Paris Chagall Committee declared when he sent it for authentication. Unfortunately for him, in addition to declaring that his painting was a forgery, the Committee informed him of its intention to destroy it—a prerogative conferred on it by French law.²⁴

My claim is that a false, or in my preferred terminology, an apparent motivating reason is not merely a bad reason but no reason at all. In this respect, ‘my reason’ is like ‘my evidence’ and not like ‘my belief’: I may say my evidence was that p, but if p is false, then it’s not that I have poor or bad evidence: I have no evidence. The police may arrest someone on poor or bad evidence, such as that the suspect was observed near the scene of the crime at the time. But if it turns out that the person who was observed near the scene was in fact someone else, then what the police have against the original suspect is not merely bad evidence but no evidence at all. If the police arrested the suspect on such grounds, it would be misleading to say that they arrested him ‘on no evidence’ because that phrase is also used when the police make an arbitrary arrest. Nonetheless, it would be true: if what they took to be evidence against the suspect, e.g. that he was observed near the scene of the crime, turns out to be false, then they have no evidence against him. In such a case, they could hardly tell a prosecutor that they have some evidence against the suspect, only not very good evidence. And the same, I claim, is true of reasons: one may have a reason to do something that is a bad reason for doing it, say because it’s defeated by other reasons. But when what I have is a false belief, then I don’t have a bad reason for acting because what I have is not a reason. To be sure, if I act on this false belief I act motivated by something I have, a false belief, which, if true, might have been an excellent reason for acting as I do. Yet, when my belief is false that belief is not a reason for acting that I have and, on acting on such a belief, I do not act for a reason; just as Mr Lang had something, a bad painting and a fake Chagall, but he did not have a Chagall.

It is sometimes claimed that this is just a terminological dispute: what my opponents want to call ‘false but real reasons’ or ‘false beliefs that are reasons’, I want to call ‘apparent reasons’. But surely, the argument goes, terminology is a matter of choice and nothing of substance depends on this. It is true that there is latitude in choosing terminology when, as we often do in philosophy, we impose some regimentation on the ordinary use of words in order to make precise conceptual distinctions and clarifications. Nonetheless, some terminological choices are more apt than others because they reflect a more nuanced or precise understanding of the relevant concepts and are less likely to lead to (or to arise from) confusion and mistakes. Thus one may, if one wants to, talk about ‘false facts’, or ‘false knowledge’—but there are, strictly speaking, no such facts or knowledge, though there are things that are mistakenly taken to be facts and knowledge and for which terms such as ‘alleged fact’ or ‘presumed knowledge’ are less misleading. Likewise, one may talk of false reasons, but strictly speaking there are no such reasons, only things that appear to be reasons and for which a term such as ‘apparent reason’ is less misleading.

There is then, I contend, something substantial to the terminological choice at issue, namely whether anything other than facts are reasons in the context of the questions outlined in Sect. 3 and, relatedly, whether the notion of a reason we apply in these contexts is a unified notion.²⁵ Dancy says that ‘we use one and the same notion of a reason in answer to two distinct sorts of question, the question why someone acted and the question whether there was good reason so to act’ (Dancy 2000, pp. 98–99); and he goes on to criticize views that suggest that,

though we speak of motivating reasons and normative or good reasons, we are genuinely dealing with two distinct sorts of reason, in a way that raises the question whether the word ‘reason’ is not awkwardly ambiguous (Ibid).

I think Dancy is right to criticize those views. But the view he endorses elsewhere, that falsehoods can be motivating reasons, is susceptible to that same criticism, for those alleged motivating ‘reasons’ can never be given as an answer to a request for a normative reason, since only truths can be normative reasons. Therefore, the view that there are false motivating reasons implies that normative and motivating reasons are things of ‘distinct sorts’ and hence that the word ‘reason’ may be ‘awkwardly ambiguous’. This shows that the choice of terminology is not without importance and that there are good grounds to accept the third option: a false belief is not a motivating reason, though it can masquerade as such and thereby motivate agents to act.

I shall finish the paper by examining an objection based on the relation between acting for a reason and rationality that goes as follows. The suggestion that agents who act on merely apparent reasons do not act for reasons seems to imply that their actions are irrational. But surely agents who act on false beliefs often act rationally. And this must be because they have reasons to do what they do. Therefore agents who act on false beliefs do act for reasons: their false beliefs are reasons they have to do what they do. Comesaña and McGrath make the objection and support it with the following principle:

(RR): One does something rationally only if one has reasons that make it reasonable for one to do it and one does it on the basis of some (sub-)set of those reasons, i.e., one does it ‘for’ those reasons (2014, p. 62).

Consider a paraphrase of an example they use to illustrate and sharpen up the objection. Bernie and Ernie each order a G&T at a bar, and each is served a drink that looks like and that the bartender describes as a G&T. So each forms the belief that there is a G&T in the glass he is being served. Both reach out for their glasses but only Ernie’s contains G&T; Bernie’s contains petrol and tonic. Surely, one might say, it is equally rational for both Bernie and Ernie to reach out for their respective glasses. If so, given the above characterization of rationality, it follows that each reaches for his glass for reasons each had, that made it equally rational for him to do so: Ernie’s reason for reaching for his glass is his (true) belief that his glass contains G&T, and Bernie’s reason for reaching, the objection continues, is his (false) belief that his glass contains G&T. Therefore, Bernie acts on a false belief that is a reason.

A Factualist may respond that the claim that Ernie and Bernie are both rational in reaching out for their glasses because they both act for reasons is true because some of the considerations that motivate both, such as that each glass looks as if it has a G&T in it, or that the bartender said as much about both glasses, etc. are indeed facts and hence reasons for which they act. Arguably then, Bernie, like Ernie, acts rationally because he acts guided by those facts and those facts make it rational for him to reach out for the glass. However, Comesaña and McGrath retort, this will not do because it would follow that Ernie would act for a reason that Bernie does not act for, namely the fact that the glass contains G&T. And if so, their response concludes, it would implausibly follow that Ernie is more rational than Bernie (2014, p. 65).

One may question whether, if Ernie has a reason that Bernie does not have and acts for that reason, it follows that he is more rational than Bernie. After all, Bernie acts for some good reasons he has (the look of the glass, what the bartender says, etc.), and it is not as if he is at all irrational in acting for those reasons. But even if we concede that point, their response brings out clearly the extent to which Comesaña and McGrath’s argument depends on the characterisation of rationality in action expressed by (RR), namely that one acts rationally only if one acts for reasons one has that make it reasonable for one to so act. But must Factualists accept their characterisation of rationality? It is not obvious that they have to. Indeed, it is open to Factualists to use Comesaña and McGrath’s example to show that their (RR) principle is false. That is, according to them it is implausible to say that Bernie and Ernie differ in their rationality and, since according to Factualism Ernie acts for a reason while Bernie acts for a merely apparent reason, a Factualist may claim that it follows that rationality does not require that agents act for (real) reasons that make it reasonable to act as they do. Rather, the Factualist may say, acting rationally requires acting for real or apparent reasons that make it reasonable to act as one does.²⁶ Thus, one may agree that Ernie and Bernie are equally rational in reaching out for the glass and accept that Bernie’s apparent reason is the same as Ernie’s (real) reason, namely, ‘there is a G&T in my glass’, without accepting that they both act for (real) reasons.

So nothing in the argument so far seems to exclude a characterization of rationality according to which an agent acts rationally only when she acts on a set of her beliefs which, if true, favour acting as she does—regardless of whether those beliefs are true or false. And so nothing precludes the Factualist from saying that both Bernie and Ernie act equally rationally because they both act on the same belief which in Ernie’s case is a real reason but in Bernie’s only an apparent reason.

Before introducing their (RR) principle and developing their arguments, Comesaña and McGrath say that they use ‘rational’ to express substantive and not merely structural rationality (using Scanlon’s terminology (2003, 13). They do not explain the distinction but illustrate it with an example about beliefs:

Suppose one believes, against the evidence, that the earth is flat. And suppose one believes, further, that if the earth is flat then it is not round. In the substantive sense, the fact that one has these beliefs does not guarantee that one is rational to believe the earth is not round. The mere fact that one has beliefs the contents of which entail, and are known to entail, that p does not guarantee that one is substantively rational to believe that p. However, intuitively, one is in some sense rational to believe the earth isn’t round given that one has those other beliefs. This intuition picks up on the structural sense (2014, p. 60).

Does this undermine the response just given? I do not think it does, for the conception of rationality proposed on the Factualist’s behalf is as well suited as (RR) to capture substantive rationality. Admittedly, in order to ensure that the characterisation of rationality given above should remain substantive and not merely structural, one would have to add that the agent who acts on beliefs that provide her with real or apparent reasons should have reached those beliefs in a rational way, etc.. But the same is true of (RR), for surely an agent will act rationally only if she acts on reasons she has that it is reasonable for her to have, etc. If we accept Comesaña and McGrath’s claim that both Ernie and Bernie act rationally, and equally so, that is surely because the story tells us that they both act motivated by beliefs that it is reasonable for them to have, given the way things look and what they are told by the bartender, etc. Were it the case that they acted on beliefs that it is not reasonable for them to have, they would not act rationally, even if those beliefs themselves (e.g. that there is G&T in the glass) made it reasonable for them to act as they do. In short, nothing that Comesaña and McGrath say seems to foreclose a response from the Factualist that rejects their characterisation of what it is to act rationally in favour of a different but equally plausible one. Thus, considerations of rationality do not undermine Factualism.

There is a range of related but different questions raised by considerations about acting rationally and acting for reasons, such as whether when someone acts rationally but guided by a false belief her action is justified; and whether, regardless of whether her action is justified or not, an agent who does something that is wrong guided by false beliefs is blameworthy and if so to what degree, in each of various possible cases. These are questions for another occasion. In this paper, my aim was to show that Factualism offers the most plausible conception of reasons for acting and of what it is to act for a reason.