The Role of Judgment in Doxastic Agency

We take it that we can exercise doxastic agency by reasoning and by making judgments. We take it, that is, that we can actively make up our minds by reasoning and judging. On what I call the ‘Standard View’ this is so because judgment can yield belief. It is typical to take it that judgments yield beliefs by causing them. But on the resultant understanding of the Standard View, I argue, it is unclear how judgment could play its role in doxastic agency in the way we take it to. I therefore offer an alternative understanding of how judgment yields belief. Drawing on Ryle (2009) I argue that when one comes to believe by judging the event which is one’s judging is token identical to the event which is one’s coming to believe. This paves the way for version of the Standard View capable of explaining how we can actively make up our minds despite that we cannot believe or come to believe at will.


Introduction
Knowing how forgetful I am in mornings and worried about being late for a train I might set my watch half an hour fast.By doing so I bring it about that at 8 the next morning I believe that it is 8:30 a.m.I leave the house in good time as a result.What this illustrates is one way of bringing it about that I have certain beliefs.I can, that is, deliberately cause myself to have certain beliefs and thus exercise agency over them.
But we also take ourselves to have another, more direct way of exercising agency over our beliefs.We take it that we can exercise "doxastic agency" by reasoning and judging.We take it, that is, that we can actively "make up our minds" by reasoning and by making judgments (Boyle 2011).Understanding how doxastic agency is possible is a matter of considerable difficulty, however.After all, we cannot form beliefs at will (Williams 1970).We cannot believe or come to believe that p voluntarily or intentionally, nor can we decide to do so.In this way believing that p and forming the belief that p differ from raising one's arm and thinking about whether p.As Boyle notes, this observation suggests a challenge that defenders of the application of agential notions to belief must face: they must give a clear account of what other notion of agency or control is at issue here.(2009, p. 120) If we really can exercise doxastic agency, the thought is, it needs to be explained how this is so despite the fact that we can neither believe nor form particular beliefs at will.Boyle assumes that in order to do so we must appeal to some "other", special kind of agency or control-that we must appeal to agency or control which is different in kind to that in question when one, say, raises one's arm.
It is typical to think that appeal to judgment can contribute to an explanation of how it is that we can exercise doxastic agency.It is typical, that is, to endorse what I will call the 'Standard View': (The Standard View) We can exercise doxastic agency because judging that p is a way of coming to believe that p. 1 I can, for instance, conclude deliberation on whether p by judging that p.And doing so can yield a new belief that p.It is natural to think that we can exercise doxastic agency because we can make such judgments and because such judgments can yield beliefs. Saying 'I judged that p', for instance, would be a natural way for me to express that I take myself to have made up my mind that p.But the Standard View alone does not yield an adequate response to Boyle's challenge.After all, we cannot judge at will any more than we can believe or come to believe at will. 2  I shall develop a version of the Standard View that is not subject to Boyle's challenge.Understanding doxastic agency does not require any special notion of agency or control.We can see this if we place judgment in the context of making up our minds by reasoning.In contrast to judging, it seems unproblematic to say that reasoning can be a form of voluntary, intentional action.When it amounts to making up our minds, reasoning constitutively concludes with judgment, and making such a judgment just is coming to believe something.
I shall proceed as follows.In Section 2, I consider what appears to be a common understanding of the Standard View.It holds that judgment yields belief by causing it.I argue that causal versions of the Standard View cannot accommodate our pre-theoretical conception of successful deliberation.In Section 3, I therefore draw on Ryle (2009) for an alternative.According to the 'Token-identity View' I suggest, judging typically just is coming to believe.In Section 4, I come back to Boyle's challenge and explain how the Token-identity View yields a way to conceive of doxastic agency in a way that is not subject to the challenge.I finish, in Section 5, by considering Boyle's (2009Boyle's ( , 2011) ) claim that token-identity views face a fatal dilemma.

The Causal View
We can begin with some platitudes about judgment.Judgment is a conscious mental event or act (Shah and Velleman 2005, p. 503;Soteriou 2013, ch. 10).When I judge that p, this judgment might yield belief that p.But not always.For instance, judging that p does not yield belief that p when I already believe that p (Soteriou 2013, p. 237;Toribio 2011, p. 346).Many also think that even in the absence of prior belief judgement can fail to yield belief (e.g., Cassam 2010, pp. 81-2;Shah and Velleman 2005, p. 507 is correct, we can still plausibly think of judgments as conscious acts of a kind that at least typically yield corresponding beliefs (unless we already hold those beliefs).But just how they play this role is not immediately clear.This is what must be understood first if we are to see how the Standard View can accommodate doxastic agency.
Many seem to assume that judgment yields belief by causing it.More precisely, the idea seems to be this: when judgment yields belief, an event of judging that p causes formation of a belief that p, that is, an event of coming to believe that p.Call this the 'Causal View'.Toribio appears to be endorsing the Causal View when she states that "judging is an event that very often leads to the formation of a belief" (2011, p. 346) as does Cassam when he states that "judging that p normally leads one to believe that p" (2010, p. 81). 4 On the resultant understanding of the Standard View we can exercise doxastic agency because judgments typically cause corresponding beliefs (unless we already hold those beliefs).
A worry for the Causal View and thus for the resultant understanding of the Standard View concerns whether it can accommodate our pre-theoretical conception of successful deliberation.Call this the ' Aim-of-deliberation Worry'.Though not decisive, the worry is sufficiently serious to motivate looking for an alternative.
The aim of deliberation on whether p is at least typically to find out, and thus to come to know, whether p (Soteriou 2013, pp. 351-2). 5Furthermore, when one concludes deliberation with judgment one conceives of the act as amounting to the achievement of the aim of one's deliberation.Boyle expresses this very thought when he says that when I conclude deliberation with judgment "[t]he act of judgment is the completion of my project, not a step towards it" (2009, p. 129).Consider then a paradigm case of successful deliberation.I do not yet believe either p or not-p, deliberate about whether p and conclude by judging that p.I conceive of my judging that p as amounting to my achieving the aim of my deliberation.And that aim is to come to know either p or not-p.I will thus only have achieved my aim as I conceive of it if I know and thus believe that p when I judge that p.On the current picture, however, when I conclude deliberation by judging that p I do not yet believe p.I only do something which will bring it about that I believe p all being well.And whether I in fact go on to form the belief is not settled by any act or action of mine.
Of course, when we conclude deliberation by judging we may fail to achieve the aim of such deliberation in doing so.One might fail to come to know and perhaps even fail to come to believe what one judges.The Aim-of-deliberation Worry is not that the Causal View allows that in concluding deliberation by judging one might fail to achieve the aim of one's deliberation in doing so.The worry is rather that if the Causal View is correct then judging cannot amount to achieving the aim of deliberation as we conceive of it.If the Causal View is correct then whenever I conclude some typical deliberation with a judgment I have not yet achieved the aim of my deliberation.I have rather merely taken a "step towards" doing so by doing something which will, all being well, bring it about that my aim is achieved.

An alternative to the Causal View
To develop an alternative to the Causal View and thus an alternative understanding of Standard View, I shall draw on a distinction from Ryle (2009).Events can be given what 14 Ryle called "thicker" or "thinner" descriptions.A is a thicker description of an event than B iff one As by B-ing.Take Ryle's (2009, p. 494) example of someone's winking.To give a thin description of the event would be to call it a contracting of the eyelid.To give a thicker description would be to call it a winking.To give a thicker description still might be to call it a signalling conspiratorially to an accomplice.This does not mean that there are several events in question here: a contracting of the eyelid, a winking and a signalling.There is just one event in question which can be given thicker and thinner descriptions-a single event, variously described (Ryle 2009, p. 495; see also Davidson 2001, pp. 4-5).
Similarly, when I judge that p an event occurs which can be given thicker and thinner descriptions.To give a thin description of the event would be to call it an accepting of p.To give a thicker description would be to call it a judging that p.To give a thicker description still might be to call it a concluding of deliberation on whether p.
Crucially, when I judge that p to give a thick description of the event may be to call it a coming to believe that p. Again, when such a description is correct it is not the case that there are several events in question, say, a concluding of deliberation on whether p, a judging that p and a coming to believe that p.Just one event is in question which can be given thicker and thinner descriptions.A single event occurs which can be variously described as a concluding of deliberation on whether p, as a judging that p and as a coming to believe that p.
When a judgment yields a corresponding belief the judgment is thus related to the resultant belief as getting married is related to being married and parking on double yellows may be to being a lawbreaker.Getting married does not cause a couple to be married, nor does parking on double yellows cause one to be a lawbreaker.The event which is the couple's getting married just is the event which is their coming to be married.And the event which is one's parking on double yellows may just be the event which is one's becoming a lawbreaker.Similarly, when I come to believe by judging the event which is my judging just is the event which is my coming to believe.The judgment is related to the resultant belief as the event of o's coming into being relates to o for any o.
So this is the 'Token-identity View' I suggest: if I come to believe p by judging that p, my coming to believe p is simply the same event as my judging that p. Someone might object to the Token-identity View along the following lines.It is possible to judge that p without coming to believe p.It is also possible to come to believe p without judging that p.These events have distinct modal profiles and therefore cannot be identical.But making this objection would amount to confusing event types and event tokens.The event types judging that p and coming to believe p are indeed not identical, but that does not mean that a token of the former type cannot also be a token of the latter type.

The Reasoning View
According to the Standard View we can exercise doxastic agency because judging that p is a way of coming to believe that p.With the Token-identity View on the table it can now be claimed that this is so, not because we can cause beliefs by judging, but because to judge can just be to come to believe.The Aim-of-deliberation Worry does not arise for Thought 7 (2018) 12-19 this suggestion.One can achieve the intuitive aim of deliberation about whether p-to come to know either p or not-p-by judging that p, because to judge that p can just be to come to know that p. Furthermore, the suggestion makes way for a promising account of how we exercise doxastic agency.
There are in fact two ways to develop the suggestion that we should combine the Standard View and Token-identity View in order to explain how we can exercise doxastic agency.One way to go would be to say that we can exercise doxastic agency by judging because judgment is a kind of action, albeit one that we cannot perform at will (e.g., Cassam 2010;McDowell 2009;Sosa 2015).Doing so would yield the following proposal.Judgment is a kind of action and to judge can just be to come to believe what one judges to be the case.To judge can therefore just be to exercise doxastic agency.
In response to this proposal some will doubt that judgment is a kind of action (e.g., O'Shaughnessy 2008; Soteriou 2013).More crucially, to endorse this proposal is to insist that judgment is a kind of action despite that we cannot judge at will any more than we can believe or come to believe at will.It is thus to appeal to a special kind of agency in order to explain how we can exercise doxastic agency in the face of Boyle's challenge.But to do this is not to meet Boyle's challenge unless it is explained how judgment is action despite that we cannot judge at will.It is no explanation of how we can exercise doxastic agency despite the fact that we cannot believe or come to believe at will to say that we can do so by judging, whilst admitting that we cannot judge at will.How judgment is action nonetheless must also be explained.
There is a simpler route available.The Standard View and Token-identity View can be combined to yield an account of how we can exercise doxastic agency without insisting that judgment is action and thus without having to explain how judgments are actions.I will label this proposal the 'Reasoning View'.In bringing out the Reasoning View we can begin with the following observation.Even granting judgments are not actions they can be constituents of reasoning.Judgments, that is, can occur as constituents of processes of reasoning.I might, for instance, judge that q, suppose that r and finally infer p, judging that p and thus concluding that p by doing so.When I judge that p in this context my doing so is a constituent of my reasoning.Furthermore, given the Token-identity View, this judgment may just be an event of my coming to believe p. Finally, in contrast to judging, it seems unproblematic to say that reasoning is a form of voluntary, intentional action.We can reason intentionally and voluntarily, and reasoning is the kind of thing which we can decide to do.It can be maintained that reasoning, unlike judgment, is a kind of ordinary action.
The Reasoning View is then as follows.Reasoning is a kind of ordinary action.It is also a process which can have judgments as constituents.Furthermore, such judgments can just be events of coming to believe.Reasoning can thus be action where to reason can, in part, just be to come to believe.Given this, to reason can, in part, just be to exercise doxastic agency.We can actively make up our minds by reasoning because reasoning is a kind of action and because to reason can, in part, just be to make up one's mind.
There is no commitment to the claim that judgments are actions on this proposal.We can exercise doxastic agency by reasoning despite that judgment is not a kind of 16 action because reasoning nonetheless is a kind of action.On this suggestion reasoning is action despite its being the case that events which are constituents of the process are not themselves actions (c.f.Soteriou 2013, pt.2).Analogously, when I play a fast and complex phrase on the piano an action is in question.Despite this it might be denied that some finger movements which are constituent events of this action are themselves actions.When, say, I play an A with my ring finger this event might not be intentional under any description, nor something I decide to do or do voluntarily, and we might deny that it is an action accordingly.
The availability of the Reasoning View reveals that we do not need to say that judgment is a kind of action that we cannot perform at will in order to explain how we can exercise doxastic agency in accordance with the Standard View.We can exercise it by reasoning because reasoning is a kind of action and because to reason can, in part, just be to come to believe.Crucially, Boyle's challenge simple does not arise for the Reasoning View.There is no need to appeal to some special kind of agency in order to explain how we can exercise doxastic agency.We can do so by reasoning where reasoning is a kind of ordinary action.It is because reasoning is a kind of ordinary action that we can exercise doxastic agency despite that we cannot believe, come to believe or judge at will. 5 Boyle's dilemma Boyle (2009Boyle ( , §. 4, 2011, §. 3.5) , §. 3.5) argues that judgments cannot be events of coming to believe and thus against the Token-identity View.And the Token-identity View is a crucial component of the Reasoning View-it is needed if the Reasoning View is to evade the Aim-of-deliberation Worry.
Suppose I come to believe p by judging that p.According to the Token-identity View the event which is my judging that p is identical to the event which is my coming to believe p.According to Boyle a dilemma arises for this suggestion when we ask: ' At the time at which I judge that p do I believe that p?' He argues as follows: If the subject does believe p at the moment when he judges p (or throughout the duration of this event, if it takes time), then it seems that his judging is not an event of "making up his mind" or "forming" the belief that p, for even at the first moment of its taking place, he already believes the proposition in question.(2011, p. 14 my emphasis, see also 2009, p. 130) If we suppose that when I judge that p I believe p, Boyle claims, then I already believe p when I make this judgment.I therefore cannot be forming the belief by judging.Suppose instead then that I do not believe p when I judge that p.In that case the event which is my judging is not itself the event of my forming the belief that p, for no such belief exists yet (Boyle 2011, p. 14).At best, on this view, my judgment that p can cause a later belief that p.But that is just the Causal View-the view which I am trying to provide an alternative to. 6 Boyle's dilemma is an instance of a general dilemma. 7Suppose we think that event e is the event of o's coming into being.We can then ask: 'Does o exist at the time of e's occurrence?'If the answer is 'Yes' then e is not really an event of o's coming into being Suppose I am granted such authority that I can fire you simply by saying 'You're fired' .When I do this, I do not cause you to be fired.My saying 'You're fired' rather makes it the case that you are fired.The event which is my saying 'You're fired' just is the event which is your becoming jobless.A version of Boyle's dilemma arises for this suggestion.Suppose that we ask of the event of my saying 'You're fired': ' Are you jobless when I say this or not?' If the answer is 'Yes' then you are already jobless and I have not really fired you.If the answer is 'No' then, again, I have not really fired you.The event which is my saying 'You're fired' can at best cause some future event which is your becoming fired.
On inspection, it can be seen that the first horn of this alleged dilemma trades on a mistake.It can be maintained that my saying 'You're fired' does make it the case that you are fired and that you are thereby jobless when I say this.But that does not mean that you are already fired when I say 'You're fired' .You are fired when I say this, not before.For you to be already fired when I say 'You're fired' would require you to be fired at some time prior to when I say it, which is not the case.
An analogous response is available to Boyle's dilemma.Suppose I judge that p and come to believe p by doing so.The event which is my judging that p is token identical to the event which is my coming to believe p.We can then ask: 'When I judge that p do I believe that p?' The answer is 'Yes'.This does not entail that I already believe that p when I judge that p, just as the fact that you are fired when I say 'You're fired' does not entail that you are already fired when I say this.For me to already believe p when I judge that p would be for me to have a temporally prior belief that p, which I do not in the case in question.Boyle's dilemma does not arise for the Token-identity View.

Conclusion
We began with a challenge from Boyle to "give a clear account of what other notion of agency or control is at issue" when we exercise doxastic agency (2009, p. 120 my emphasis).In response I have suggested that reasoning is a kind of action.We can, after all, reason intentionally and voluntarily, and can decide to do so.Furthermore, to reason can, in part, just be to come to believe.On this suggestion, when we make up our minds by reasoning we can say that the kind of agency in question is ordinary action.There is no need to appeal to some "other notion of agency".
).3Even if that Thought 7 (2018) 12-19 since o already exists.If the answer is 'No' then e is again not an event of o's coming into being since o does not exist yet.e can at best cause o's coming into being.For any event e, e cannot be the event of o's coming into being.To see what has gone wrong it will be instructive to consider another instance of the general dilemma.