Abstract
In this paper, I present and explore some ideas about how factive emotional states and factive perceptual states each relate to knowledge and reasons. This discussion will shed light on the so-called ‘perceptual model’ of the emotions.
Similar content being viewed by others
As a number of authors have observed, certain emotional state ascriptions are factive.Footnote 1 In particular, certain state ascriptions of the form ‘S V’s that P’ entail P (where S names an agent and V stands for an emotional verb or adjectival phrase of the form ‘is F’).Footnote 2 Some examples of factive emotional state ascriptions are as follows: Jason regrets that he danced at the party. Ann is angry that she lost the bet. Lonnie is delighted that his mother ironed his pants.Footnote 3 By contrast, non-factive emotional state ascriptions (e.g. fears-that, worries-that, hopes-that, etc.) have no such entailment.Footnote 4 Factive emotional state ascriptions form a subclass of factive mental state ascriptions that include ‘knows that’, ‘sees that’ and ‘understands that’. In this paper, I will explore some important similarities and differences between factive emotional state ascriptions and factive perceptual state ascriptions. This exercise will serve to illuminate some important structural features of factive emotional states—namely, how these states relate to reasons and knowledge—and in turn, will shed light on the so-called ‘perceptual model’ of the emotions (a view which has received considerable attention in the recent literature on emotions).
In Sect. 1, I present a brief defense of a knowledge requirement on factive emotional states.Footnote 5 In Sect. 2, I explain the source of the knowledge requirement for those states. This account will crucially depend on the connection between those states and what Grice calls ‘personal reasons’. In Sect. 3, I investigate the structural parallels and disanalogies between factive emotional state ascriptions and factive perceptual state ascriptions. In Sect. 4, I relate the previous discussion to the perceptual model of emotions.
1 The knowledge requirement
One fundamental assumption of the sections that follow is that there is a knowledge requirement on factive emotional state ascriptions:
The Knowledge Requirement: For factive emotional state ascriptions, if S V’s that P then S knows that P.
This claim is not new and is most famously defended by Unger (1975), Gordon (1987), and Williamson (2000). Of course, there are alternative hypotheses that respect the factivity of the state-ascriptions we are interested in but are weaker than the knowledge requirement. For example:
True Belief Condition: For factive emotional state ascriptions, if S V’s that P then S has a true belief that P.
Justified True Belief Condition: For factive emotional state ascriptions, if S V’s that P then S has a justified true belief that P.
However, I take it that Unger et al. are right to insist on the much stronger knowledge requirement on these states.
The most direct argument in favor of the knowledge requirement on factive emotional state ascriptions is that the result of conjoining factive emotion ascription with knowledge denial is linguistically bizarre. Consider how strange it sounds to say, ‘Sarah regrets that she punched Jason but Sarah does not know that she punched Jason’.Footnote 6
Further, a wide range of judgments can be explained by the knowledge requirement. Suppose Maria sees Sarah’s identical twin fall off a cliff and feels surprise. Here, Maria justifiably mistakes Sarah’s twin for Sarah. Meanwhile, at the same time as the twin’s accident, Sarah falls off a cliff too. In a case like this, Maria’s belief that Sarah has fallen off a cliff is true and justified yet it would be infelicitous to say that Maria is surprised that Sarah fell off a cliff. This infelicity can be explained by the fact that Maria did not know that Sarah fell off a cliff. Notice that this is a case of justified true belief without knowledge and so the infelicity cannot be explained by recourse to either the true belief requirement or the justified true belief requirement. Similar remarks apply mutatis mutandis in cases where Sarah’s identical twin prompts Maria’s delight, anger, sadness, and so on. In the relevant scenarios, it is infelicitous to assert that Maria is delighted that Sarah fell off a cliff, or sad that Sarah fell off a cliff. The knowledge requirement can elegantly explain these infelicities.Footnote 7 , Footnote 8
(Note that the knowledge requirement, as it applies to the verb ‘surprise’, concerns constructions of the form ‘S is surprised that P’. In the case just described, it would be more felicitous to say ‘Maria is surprised at the thought that Sarah fell off a cliff’ or ‘Maria is surprised because she believes that Sarah fell off a cliff’ but these are different constructions. The knowledge requirement as stated should not be conflated with the grotesque overgeneralization that surprise requires knowledge).
Thus ends my brief defense of the knowledge requirement on factive emotional state ascriptions. Adding substantially new arguments for the knowledge requirement on factive emotional states is not my task here. In the next section, I will explore the source of the knowledge requirement on such states. This will serve as a nice starting point for uncovering the structural contrasts between factive emotional states and factive perceptual states.
2 Factive emotional state ascriptions and personal reasons
In this section, I will explain why the knowledge requirement holds for factive emotional states. But first let me say some things about my use of the Gricean expression, ‘personal reasons’. In connection with the emotions, reasons talk can take a variety of forms. When we are interested in reasons that can explain an agent’s anger, we typically make use of a ‘reason why’ construction. (The reason why S is angry is that P). When we are interested in the normative reasons available to an agent for feeling anger, we typically make use of a ‘reason to be angry’ construction. (P is a reason for S to be angry). But for my purposes here, I am primarily interested the kind of reasons that are picked out by paradigmatic uses of possessive constructions of the form ‘my/his/her reason for V-ing is…’ Following Grice (2001), I use the expression, ‘personal reasons’ to refer to those reasons standardly picked out by this kind of possessive construction.Footnote 9 (Many philosophers use the expression, ‘motivating reasons’ here). As Grice notes, there are two kinds of constructions used to describe personal reasons, that-clause constructions and infinitival constructions:
-
(i)
His reason for leaving was that the house was on fire.
-
(ii)
His reason for leaving was to avoid harm.
I am interested in those personal reasons described by expressions which contain that-clauses as exemplified by (i). My use of the expression ‘personal reason’ will hereafter refer to kind (i). With the ideology of personal reasons in place, let me now put forth two conjectures:
Conjecture One: For factive emotional state ascriptions, if S V’s that P, then one of S’s personal reasons for V-ing is P.Footnote 10
Conjecture Two: If one of S’s personal reasons for V-ing is P, then S knows P.Footnote 11
These two conjectures entail the knowledge requirement and if true, offer a satisfying explanation of that requirement.
Conjecture one is extremely plausible. It would be strange to say that John is angry that his treasure has been stolen but that the proposition that his treasure has been stolen is not one of John’s reasons for being angry.
Conjecture two suggests that knowledge is a condition on having personal reasons and this claim is also very plausible. Of course, we should have no expectation that there is a knowledge condition on explanatory non-personal reasons. If an explanatory reason why Jason feels jealous is that he is chemically imbalanced, then the fact that he is chemically imbalanced isn’t necessarily available to Jason as a reason for him to feel jealous. In a case like this, it would be out of place to say that Jason’s reason for feeling jealous is that he is chemically imbalanced. Explanatory reasons that are not available to the agent (in virtue of the fact that they are unknown) pose no threat to conjecture two.
When it comes to personal reasons, the case for a knowledge requirement on factive emotional states is strong. The result of combining a personal reason attribution with knowledge denial is as linguistically bizarre as the result of combining a factive emotional state ascription with knowledge denial. It would be strange to say ‘Jason is angry that he is the shortest man in the room but Jason doesn’t know that he is the shortest man in the room’. And it would be similarly strange to say ‘Jason’s reason for being angry is that he is the shortest man in the room but Jason doesn’t know that he is shortest man in the room’. Thus, we should expect that the explanation of the knowledge requirement on factive emotional states will carry over nicely to personal reason constructions. In the earlier case where Maria sees Sarah’s identical twin fall off a cliff, it is odd to say that Maria’s reason for being delighted is that Sarah fell off a cliff. The mere observation that personal reasons (as expressed by that-clause constructions) are factive does not explain the oddity, since in the cases described, it is true that Sarah fell off a cliff. A justified true belief requirement on personal reasons fares no better since Maria’s belief that Sarah fell off a cliff is true and justified. But a knowledge requirement on personal reasons provides a satisfying explanation for the oddity.Footnote 12
Conjectures one and two thus seem true and when combined, offer a nice explanation of the knowledge requirement that they entail. In the next section, we shall see that while a knowledge requirement on factive perceptual states is also very plausible, the source of the requirement is importantly different.
3 Emotional factives and perceptual factives
With Unger and Williamson, I take it that perceptual verbs such as ‘seeing that’ and ‘perceiving that’ are not merely factive but are also knowledge entailing. I will not offer additional arguments for this view here. I will merely note that the case for the knowledge requirement on factive emotional states carries over very nicely to factive perceptual states. In the case provided in Sect. 2 where Maria sees Sarah’s identical twin fall off a cliff and forms a justified true belief (via vision) that Sarah has fallen off a cliff, it would be wrong to say that Maria saw that Sarah fell off a cliff.
On one level of abstraction, seeing that P and being angry that P relate to knowledge in the same way. Maria cannot see that Sarah is falling off a cliff unless Maria knows that Sarah is falling off a cliff. Similarly, Maria cannot be amused that she runs faster than John unless Maria knows that she runs faster than John. Nevertheless, I will argue that on close inspection, there is a marked difference in how emotional factives and perceptual factives relate to knowledge.
Let us begin by considering a vision-theoretic use of ‘sees that’. I say ‘vision-theoretic use’ because there is another use of ‘sees’ which has nothing to do with vision.Footnote 13 There are two natural settings where we might say that Bukowski sees that his horse is a failure. In one setting, Bukowski sees through vision his horse lose the race and thus comes to know that his horse is a failure via vision. In another setting, we can tell a story where Bukowski learns that his horse is a failure by some means other than vision (perhaps he learns this fact by testimony). If we ever say that Bukowski sees that his horse is a failure in the second setting, then we seem to utilize a metaphorically extended use of ‘sees’ that communicates knowledge and little more. (Note that there are similar uses of ‘sees’ in connection to mathematical insight that do not involve vision). It is the first use connected to vision that concerns me here. (Similarly there is a use of ‘perceives’ —in mathematics, for example—that has nothing to do with sensory perception. I will be interested in the straightforwardly perceptual use here.)
I have said that both ‘S sees that P’ and ‘S is angry that P’ entail that S knows P. However, the analogue for conjecture one (Sect. 2) for ‘sees’ and ‘perceives’ is hopeless. In a case where Bukowski is angry that his horse has lost, one of his reasons for being angry is that his horse has lost. But in the case where Bukowski sees that his horse has lost the race, it would be wrong to say that one of Bukowski’s reasons for seeing is that his horse has lost the race.Footnote 14 (Relatedly, the cause-effect relation between knowledge and the relevant mental state is very different in each case. When S is angry that P, S’s knowledge that P is one of the causes of S’s anger. But in a case where S sees that P, it is not that S’s knowledge that P causes S’s seeing. It is S’s use of vision which causes S to come to know P.) Thus ‘perceives that P’ and ‘sees that P’ are not constructions that are used to convey personal reasons for being in certain states. Rather, they are devices for marking how an individual comes to know that P. In its visual use, ‘S sees that P’ communicates an episode of knowing that P via vision and in its sensory-perceptual use, ‘S perceives that P’ communicates an episode of knowing that P via sensory perception. In short, ‘S sees that P’ and ‘S perceives that P’ carry information about how knowledge was acquired whereas ‘S is angry that P’ and ‘S is happy that P’ conveys an upshot of the knowledge. Knowing P makes P available to an agent to use as a personal reason. Factive emotional state ascriptions convey what emotional state occurred for what reason(s).
In sum, factive emotional verbs like ‘anger’ or ‘regret’ require knowledge via a knowledge requirement on personal reasons but ‘sees-that’ requires knowledge because it specifies the mode of coming to know.
4 The perceptual model of emotions
According to the so-called ‘perceptual model’, an emotional experience can be evidence of value in a way to how a perceptual experience of a red thing can be evidence of the presence of a thing that is red. As Michael Brady (2013) points out, the key issue for the perceptual model is whether or not there really is a deep analogy between the epistemological roles of emotion and those of perceptual experience. And here, we are looking for a deeper similarity beyond one obvious point of analogy that we can all agree upon (one that is not particularly epistemological): both emotions and perceptual experiences have phenomenal properties (a ‘what it’s like’ aspect). (Brady 2013, p. 47).Footnote 15
One central issue for the perceptual model is whether it is right to think of emotions as representing evaluative states of affairs (as opposed to merely being prone to prompt evaluative thoughts). If they do not so represent, then it seems obvious that the perceptual model is at least, prima facie, in trouble. But I will not pursue this issue here. Rather, I will press another central challenge to that model which is closely connected to the discussion thus far (though, I will now generalize beyond the case of factive emotional states): emotions and perceptual experiences seem to bear very different relations to personal reasons.
Consider a case where something looks to be red. There may be any number of explanatory reasons why the thing looks red. Some of these explanatory reasons are mundane. For example, perhaps the thing looks to be red because the lighting is normal or because one’s eyes are open and looking in the direction of the red thing. Some of the reasons that explain why the thing looks to be red are more sophisticated and would be revealed by psychophysics (e.g. that the wavelength of the light reflected by the object has a certain profile). But while there are plenty of reasons that explain the fact that something looks to be red, it is important to notice that an agent does not have personal reasons for perceptually representing a thing as red. In fact, the question, ‘What is the agent’s reason for it looking that way?’ sounds fundamentally confused. This is a point that extends to other perceptual modalities. (The questions, ‘What is the agent’s reason for feeling warm?’ or ‘What is the agent’s reason for the music sounding loud?’ seem equally confused.)
A few points of clarification. First, note that in claiming that agents lack personal reasons for perceptual experiences, I am not simply relying on the point from earlier sections—that factive perceptual state reports do not encode personal reasons for perception. After all, there is no logical incoherence in the thought that we have personal reasons for perceptual experiences but those reasons are encoded by other means. That said, even when we widen our gaze, it is not natural at all to talk about an agent as having personal reasons for a perceptual experience. That is the point I am relying on here.
Second, I do not deny that one could have a personal reason for making something look red. But in this kind of case, one has a personal reason for an action, not for the experiential state itself.
Third, I do not deny that something’s looking red might be one’s reason for feeling confident that it is red. But in this case, the experiential state (or a fact about it) is the reason for one’s confidence. It is one thing for perceptual experience (or facts about it) to serve as an agent’s personal reason for various other states of mind.Footnote 16 It is quite another thing for an agent to have personal reasons for the perceptual experience.Footnote 17
We may not have personal reasons for perceptual experiences but emotional states seem eminently fit for having personal reasons. Indeed, we have seen that factive emotional state ascriptions encode the personal reason(s) for being in some particular emotional state.Footnote 18 John cannot ask Maria (in a natural way) what her reason was for a thing looking purple to her but it would be quite natural if John asked Maria what her reasons were for feeling angry or feeling afraid. This consideration alone seems to point to a significant disanalogy between perceptual states and emotional states.
Of course, in saying that emotional states are the sorts of things for which we have personal reasons, I do not mean to suggest that we have direct voluntary control over them. I might not have direct control over my feeling of shame at being seen naked by the chimney cleaner. But my feeling of shame is still something that I have a personal reason for. (Namely, my reason for feeling ashamed is that the chimney cleaner saw me naked.) This is an important way in which we conceive of emotions as responses to what one learns about the world. The attitude of belief also illustrates the fact that being governed by personal reasons is not the same thing as having direct voluntary control. It is often noted that it seems we lack direct voluntary control over our beliefs. I cannot, simply by an act of will, come to believe that the chimney cleaner saw me naked. Nevertheless, something can be my reason for believing something else. The presence or absence of direct voluntary control does not establish whether something is the sort of thing for which one has personal reasons.
One further point of clarification: In saying that emotions are the kinds of state for which we can have personal reasons, I do not want to assume that we always have personal reasons for our emotional states. By analogy, consider the case of belief. From the fact that belief is a kind of state for which it is possible to have personal reasons, it hardly follows that we always have personal reasons for belief. Turning to the case of emotions, the above discussion indicates that cases of anger that are properly describable using expressions of the form, ‘S is angry that P’ will inevitably encode personal reasons. But I make no assumption, and indeed, I do not think that every case of anger is so describable. (In many cases where S is angry because S falsely thinks that P, no construction of the form, ‘S is angry that P’ will be apt.)
While the way we talk about personal reasons is a trouble spot for the so-called perceptual model, I do not think that the preceding considerations kill all hope for that model. In this connection, let me stress two points.
First, nothing in the above discussion precludes emotions from serving as evidence for evaluative states of affairs. A very standard rough and ready account of evidence is in terms of probability raising. A fact is evidence for P just in case it raises the probability of P. Or perhaps more carefully, a fact is evidence for a particular agent that P just in case it makes P more likely from the point of view of that agent. And the main point to bear in mind here is that there is no simple argument from the premise that emotions are the sorts of things for which we have personal reasons to the conclusion that emotions cannot be evidence for evaluative states of affairs. The evidential scope of emotions in this connection needs separate consideration. And if the main idea of the ‘perceptual model’ is merely that emotions are systematically capable of playing an evidential role, the jury is still out (as far as this paper is concerned).
Second, the above discussion has relied on observations about the ways we ordinarily talk about personal reasons. But there still remains the possibility that psychological talk for ordinary-folk is confused. Perhaps, even though we ordinarily talk as if people can have personal reasons for emotions but not perceptual experiences, there is, in fact, no psychological joint in nature that corresponds to this conceptual contrast. The question of whether our ordinary ways of talking about our psychology is apt or confused is a separate question and one which I will not worry about here.
5 Conclusion
There are striking similarities between ‘S regrets that P’ and ‘S sees that P’. Both constructions are factive and both entail that the proposition picked out by the that-clause is known. There is also a striking dissimilarity. Like other factive emotional state ascriptions, ‘S regrets that P’ encodes a personal reason for being in that particular emotional state. But like other factive perceptual state ascriptions, ‘S sees that P’ does not encode a personal reason for being in that state and instead merely encodes a way of coming to know. The contrasting relation to personal reasons appears to extend more generally to emotions and perceptual experiences. While emotions are the sorts of things for which we often have personal reasons, it seems that we never have personal reasons for perceptual representations. A proponent of the perceptual model of emotions will need to grapple with this fact, showing either that the perceptual model can survive this contrast or else do work to display the apparent contrast as misleading or illusory.
Notes
In the case of ‘happy’ there are other that-clause constructions that can be formed by combining a so-called ‘linking verb’ with ‘happy’ followed by a that-clause: felt happy that P, seemed happy that P, remained happy that P, and so on. As far as any emotional adjective (F) is concerned, I will limit my focus to ‘is F that’ (and ‘was F that’).
There are other factive environments in which emotional verbs can figure which will not be of interest to me. For example, ‘S is angry because P’ plausibly entails P (on account of the fact that ‘because’ induces a factive environment). I limit my focus to factive ‘V-that’ constructions here. I shall also not be discussing constructions which take the form of an emotional verb followed by a noun phrase. (e.g. Ann regrets the picnic) which are not of the right form to be factive.
An alternative hypothesis is that a claim of the form ‘S is angry that P’ cancellably implicates P without entailing P. This is unpromising. Consider the speech: ‘She was angry that her house had burnt down but it hadn’t, it’s just that the radio reported that it had’. That speech is very marginal but would be fine if ‘she was angry that her house burnt down’ merely generated a cancellable implicature that her house had burnt down in the same way that standard implicature cancellation cases work. But this case is very different from the standard implicature cancellation cases and the contrast is striking. ‘There is a gas station around the corner but I don’t know whether it’s open’ is fine. ‘She is angry that her money has been stolen but doesn’t know that her money has been stolen’ is markedly worse. In general, ‘angry that’ patterns in relevantly similar ways to ‘know that’ as far as factivity goes. ‘She knew that her house burnt down but it hadn’t, it’s just that the radio reported that it had’ is marginal in the same way as the ‘angry that’ example. An alternative hypothesis to the cancellable implicature hypothesis is that ‘S is angry that P’ non-cancellably presupposes P without entailing P. But given the parallels between ‘knows that’ and ‘angry that’, I see little promise in a view that concedes factivity for ‘know that’ but opts for some fancy alternative to factivity for ‘angry that’.
I will sometimes use the term ‘factive emotional states’ as shorthand for emotional states described by factive emotional state ascriptions. Note that I am not going to pursue the metaphysical question as to whether the mental state token associated with ‘she is angry that she lost’ is essentially factive. This question turns on murky metaphysical issues about events/states (such as whether it is essential to that very anger event that it occurs in a world where she lost).
In correspondence, Mark Schroeder objected. He pointed out that, ‘I believe that P but I don’t know that P’ also sounds weird but is not good evidence that belief entails knowledge. Schroeder’s objection is confused. His example relies on the use of the first-person pronoun. Described in the third-person, we find no oddity at all: ‘He believes that P but he doesn’t know that P’ is perfectly felicitous.
A commonly recycled objection to Williamson’s view about stative factives is that in fake barn cases, knowledge ascriptions are infelicitous but ascriptions of various factives such as, ‘She sees that there is a barn in front of her’, ‘She is happy that there is a barn in front of her’ etc. sound prima facie more acceptable. [This objection appears in Turri (2010), Hughes (2014) and Locke (2015)]. See Hawthorne and Magidor (forthcoming) for an extensive discussion of why this objection is unconvincing.
One reader suggested that this data could be explained by a causal requirement instead of the knowledge requirement. The causal requirement would roughly be something like, ‘V-ing that P requires a true belief that P that is caused by the fact that P’. This idea doesn’t fare well when we turn to cases where one knows P but where one’s true belief is not caused by the fact that P (the most obvious case would be one where P is a fact about the future.) The absence of a causal link does not block factive emotion ascriptions in such cases. For relevant discussion, see Dietz (2016).
Given that the possessive construction is notoriously context-sensitive, we should expect there to be other uses of the ‘my/his/her reason’ construction. As Hornsby (2008) notes, the kind of relationship expressed by ‘her reason’ in ‘her reason for going to the psychiatrist is that she thinks she is being followed’ and ‘her reason for running is that she is being followed’ is different to that in ‘her reason for running is that she thinks she is being followed’. As Hornsby would put it, the sentence following ‘that’ in the final example does not express the ‘reason for which’ one acts. It is the ‘reason for which’ use of the possessive reason construction that interests me here.
The reverse of conjecture one—if P is one of S’s reasons for V-ing, then S Vs that P—is far less plausible. One concern is has to do with certain cases where P is a fact about the agent’s psychological state. If one of John’s reasons for being angry is that John thinks that his friend has lied, it is not plausible that John is angry that he thinks that his friend has lied. (But this may involve a different use of the possessive reason construction, see footnote 9.) But troublesome cases are not limited to cases that involve psychological propositions as the complement of the reason construction. If Margot’s car has been stolen, one of Margot’s reasons for being angry might be that her car is brand new. But in this case, it doesn’t sound right to say that Margot is angry that her car is brand new. This raises the interesting question of which of the P’s that serve as personal reasons for emotional V-ing are suitable propositional complements for V-ing that. I will not present my own ideas about this issue in this paper. (Thanks to Timothy Williamson for discussion here.)
One reader suggested that something like conjecture two is also plausible for the second type of personal reasons mentioned above. According to the reader, the fact that ‘Her reason for leaving was to avoid the fire’ seems to require that there was a fire. But the reader is misled by a poor choice of example. In their example, the infinitival of choice contains ‘the’ and ‘the’ carries the presupposition of a fire. (And in any case the analogue of factivity would have to do with whether the fire was avoided not whether it existed.) There is nothing akin to a general factivity requirement on these infinitival reason constructions. Consider the following example: ‘Her reason for getting on that train, which sadly crashed, was to visit her relatives in New York’.
The idea that there is a knowledge requirement on personal reasons is not new. One will find helpful discussions of this requirement in Unger (1975), Hyman (1999, 2011, 2015), Williamson (forthcoming), Hornsby (2007, 2008), Alvarez (2010), Littlejohn (2012) and McDowell (2013), and Hawthorne and Magidor (forthcoming).
For an extensive discussion of these two uses, see French (2012).
The contrast in the text is reinforced when one considers adverbs that are connected with rational evaluation. Consider ‘foolishly’. The felicity of this adverb patterns with states that are amenable to personal reasons: ‘He foolishly left.’ ‘He foolishly believed that she was innocent.’ ‘He foolishly regretted leaving that job’. But ‘She foolishly saw that there was a tiger in front of her’ sounds terrible. (All of these examples pattern similarly when one replaces ‘foolishly’ with ‘wisely’). (Thanks to Timothy Williamson for discussion here.)
In passing, I should note that there is there is a venerable tradition—associated with James (1884) and Lange (1885)—that accepts that emotions have representational content but which holds they represent oneself as being in a certain bodily state. On this picture, fear is a representation of oneself as being in a certain bodily state, perhaps under a distinctive mode of presentation. This kind of model hardly gives comfort to the idea that emotional states represent evaluative states of affairs. Instead, it offers a very different way of developing the idea that emotions are somehow perceptual which would be very different to the ‘perceptual model’ that is discussed in recent work on emotions. It would be interesting to explore to what extent the remarks that follow make trouble for James-Langian perceptual ideas. (Thanks to Jeremy Goodman for discussion here.).
Another issue worth exploring is the relation of emotions and experience to so-called, ‘normative reasons’ paradigmatically encoded by linking ‘reason’ with infinitival expressions (as in, ‘She has reason to be angry’). Here the contrast is a little less sharp. Suppose there is an evolutionary advantage accruing to a thing’s looking a certain way. Then there is, in a natural sense, a reason for it to look that way. The main point here is that normative reasons need not encode things that are even available as personal reasons. (One familiar point to make is that if a glass contains poison and I have no hope of knowing that, then there is still a perfectly natural sense in which we would say that I have reason to refrain from drinking it. This is the so-called ‘external’ use of normative reasons talk. While ‘reason to’ constructions do not straightforwardly encode the availability of personal reasons, other normative reason-theoretic talk may be more tightly connected. For example, I am sympathetic to Unger’s (1975) claim that there is a fairly tight connection between talk of a state’s being ‘reasonable’ and having personal reasons for it. In particular, it seems that a state can be reasonable only if one has personal reasons for being in it.
Arguably, the key insight underlying historically pervasive talk of ‘perception as passive’ is that perceptual experiences are not the kinds of states for which we have personal reasons. Now certainly, some historically popular doctrines concerning passivity were overdone. For example, consider Rules for the Direction of Mind, where Descartes wrote, ‘sense-perception occurs in the same way in which wax takes on an impression from a seal’. See Descartes et al. (1984, p.40) As many philosophers and cognitive scientists since have pointed out, the contours of visual experience can be affected by various cognitive factors (including priming, conceptual repertoire and so on), and so this take on passivity is misleading. But the view that perceptual experience is in a way passive because it is devoid of personal reasons may yet be defensible. (Thanks to John Hawthorne for discussion here.).
Of course, we can also have personal reasons for emotional states that are not picked out by factive emotional verbs. For example, the fact that someone has a terminal illness might be their reason for feeling afraid. But since a proposition can be one’s personal reason only if it is true, constructions like ‘being afraid that P’ or ‘being worried that P’ are ill fit to encode personal reasons, since these constructions are not factive.
References
Alvarez, M. (2010). Kinds of reasons: An essay on the philosophy of action. Oxford: Oxford University Press.
Brady, M. S. (2013). Emotional insight: The epistemic role of emotional experience. Oxford: Oxford University Press.
Descartes, R., Cottingham, J., Stoothoff, R., & Murdoch, D. (1984). The philosophical writings of descartes. Cambridgeshire: Cambridge University Press.
Dietz, C. H. (2016). Are all reasons causes? Philosophical Studies, 173(5), 1179–1190.
French, C. (2012). Does propositional seeing entail propositional knowledge? Theoria, 78(2), 115–127.
Gordon, R. M. (1987). The structure of emotion: Investigations in cognitive philosophy. Cambridge: Cambridge University Press.
Grice, H. P. (2001). Aspects of reason. Oxford: Clarendon Press.
Hawthorne, J., Magidor, O. (forthcoming). Reasons and evidence. In D. Star (Ed.), The Oxford handbook on reasons and normativity.
Hornsby, J. (2007). Knowledge, belief, and reasons for acting. In C. Penco, M. Beaney, & M. Vignolo (Eds.), Explaining the mental (pp. 88–105). Newcastle: Cambridge Scholars Publishing.
Hornsby, J. (2008). A disjunctive conception of acting for reasons. In A. Haddock & F. MacPherson (Eds.), Disjunctivism: Perception, action, knowledge (pp. 244–261). Oxford: Oxford University Press.
Hughes, N. (2014). Is knowledge the ability to ϕ for the reason that P? Episteme, 11(4), 457–462.
Hyman, J. (1999). How knowledge works. Philosophical Quarterly, 49(197), 433–451.
Hyman, J. (2011). Acting for reasons: Reply to Dancy. Frontiers of Philosophy in China, 6(3), 358–368.
Hyman, J. (2015). Action, knowledge, and will. Oxford: Oxford University Press.
James, W. (1884). What is an emotion? Mind, 9(34), 188–205.
Lange, C. G. (1885). Om sindsbevaegelser: et psyko-fysiologisk studie. Kjbenhavn: Jacob Lunds. Reprinted in The emotions (C. G. Lange & W. James Eds., I. A. Haupt, Trans.). Baltimore: Williams & Wilkins Company 1922.
Littlejohn, C. (2012). Justification and the truth connection. Oxford: Oxford University Press.
Locke, D. (2015). Knowledge, explanation and motivating reasons. American Philosophical Quarterly, 52(3), 215–232.
McDowell, J. (2013). Acting in the light of a fact. In D. Bakhurst, B. Hooker, & M. O. Little (Eds.), Thinking about reasons: Themes from the philosophy of Jonathan Dancy (pp. 13–28). Oxford: Oxford University Press.
Turri, J. (2010). Does perceiving entail knowing? Theoria, 76(3), 197–206.
Unger, P. K. (1975). Ignorance: A case for scepticism. Oxford: Clarendon Press.
Williamson, T. (2000). Knowledge and its limits. Oxford: Oxford University Press.
Williamson, T. (forthcoming). Acting on knowledge. In J. A. Carter, E. Gordon, B. Jarvis (Eds.), Knowledge-first, Oxford: Oxford University Press.
Acknowledgements
For feedback on earlier versions of this paper, I am grateful to Bill Child, Jeremy Goodman, John Hawthorne, Clayton Littlejohn, Ofra Magidor, Ian Phillips, Mark Schroeder and Timothy Williamson.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
About this article
Cite this article
Dietz, C.H. Reasons and factive emotions. Philos Stud 175, 1681–1691 (2018). https://doi.org/10.1007/s11098-017-0929-y
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-017-0929-y