On being motivated

Abstract

Merleau-Ponty’s notion of being motivated or solicited to act has recently been the focus of extensive investigation, yet work on this topic has tended to take the general notion of being motivated for granted. In this paper, I shall outline an account of what it is to be motivated. In particular, I shall focus on the relation between the affective character of states of being motivated and their intentional content, i.e. how things appear to the agent. Drawing on Husserl’s discussion of perceptual awareness, I suggest that the intentional content of states of being motivated has a horizonal structure, in which both affective and perceptual features are implied. In states of being motivated, the agent becomes aware of certain possibilities for action, towards which they feel drawn. This structure is what Merleau-Ponty refers to as the “intentional arc” (1962, 136).

The concept of being motivated or solicited to act which Merleau-Ponty discussed in Phenomenology of Perception has recently been the subject of extensive investigation.¹ This work has tended to focus on specific aspects of being motivated (e.g. Carman 1999 on the body schema; Kelly 2002 and Jensen 2009 on motor intentionality) and on relating motivation to contemporary discussions (e.g. Dreyfus 2000 on dynamic systems theory; Romdenh-Romluc 2011 on embodied cognition; Smith 2012 on depression). It is generally assumed by those taking up Merleau-Ponty’s ideas that to be motivated is to feel inclined to act in a certain way (Dreyfus and Kelly 2007, 52; Dreyfus 2007b, 104 n.1), such that one can respond immediately to what one perceives (Dreyfus and Kelly 2007, 53; Romdenh-Romluc 2011, 90). It is often further assumed that this response cannot be understood as “the blind and mechanistic workings of material causality” (Wrathall 2005, 111); nor need one take that which motivates one as a reason for acting (Dreyfus 2007a, 361; Rietveld 2010, 187). Indeed, one of the most interesting features of being motivated is that it holds out the promise of a third way of explaining behaviour, which is neither mechanistically causal nor a matter of acting for reasons (Dreyfus 2005a, 56; Wrathall 2005, 112, 117–121).

However, Merleau-Ponty’s descriptions of being motivated amount to a series of suggestive vignettes rather than a systematic account. In motivation, Merleau-Ponty tells us, “One phenomenon releases another […] by the meaning which it holds out” (1962, 50). “The motive is an antecedent which acts only through its significance”, and which sets up a reciprocal relation with what is motivated (259); indeed, this is an “internal relation” (50). As a result, “the flow of motivations” bears one along (295); one’s body, “as the potentiality of this or that part of the world, surges towards objects to be grasped and perceives them” (106). These comments, and Merleau-Ponty’s discussion of individual cases,² indicate an important phenomenon but do not clarify its structure, how its different features are related. In particular, they do not account for the intentional content of states of being motivated; the meaning which releases an action and in doing so sets up a reciprocal relation with it.³ Being motivated is a matter of responding or being inclined to respond to how things appear to one. Any account of what it is to be motivated must explain the link between this inclination and how things appear.

In what follows I shall not try to provide an exegesis of Merleau-Ponty’s own views. Rather, I shall take his work (and that of Husserl) as a starting-point in clarifying what it is to be motivated. After outlining some of the characteristic features of being motivated (Section I), I discuss its normative and affective aspects (Section II). These aspects are essential to states of being motivated, but we need to understand how they are related to the intentional content of these states in order to properly understand what it is to be motivated. In Section III, I provide an account of this relation, by appealing to the horizonal structure of this intentional content. In Section IV, I shall outline what it is to feel inclined to take up a certain course of action when one is motivated, and in doing so distinguish it from various characterisations of desire.

I

The basic formulation for being motivated which I shall use is as follows: an agent, A, is motivated by x to y.⁴ When A is motivated we can speak of a three-way relation of motivation between A, understood as a conscious individual capable of action; an object, x, of which A is conscious; and a course of action, y.⁵ By ‘object’ I simply mean anything of which the agent can be conscious, including properties, property-bearing entities and events. A can be conscious of or experience x in different ways (through imagination or memory, for example). However, the clearest cases of being motivated and of motivated actions involve perception, when one reacts to what one perceives in an immediate but not simply mechanical fashion. Merleau-Ponty’s own examples are primarily drawn from such cases: sexual activity (1962, 155–156), or habitual actions by which one copes with one’s surroundings, such as taking a handkerchief from one’s pocket (1962, 103) or steering through a narrow opening without measuring its width (1962, 143). In what follows I shall treat perceptual cases as paradigmatic, even though one can be motivated by something one does not perceive.

The third relatum of the relation of being motivated is a course of action rather than a specific action. I understand specific actions to be particular datable events. Strictly speaking, the agent is not motivated to perform a particular action (which may be all but indistinguishable from many others), but to take up a course of action, one which may be executed by performing any action of a certain type.⁶ What the type of action is will vary depending on the agent’s situation. The agent may be motivated to leave the room, a course of action which might be executed in a number of different ways; or to grasp a cup without spilling the liquid in it, which can only be done by holding the cup a certain way. Any action of the appropriate type will count as the agent’s acting as they were motivated to.

This formulation contains some simplifications. I am setting aside cases where an agent can be motivated by more than one object. I am also setting aside cases where the agent is simultaneously motivated towards different or even conflicting courses of action (I shall briefly consider such cases in Section IV).

A further point to add to the basic formulation is that an agent is always motivated to act by some particular objects, situations or events. As far as I am aware, this is not stated explicitly by Merleau-Ponty, but it is certainly consistent with the examples he gives, which focus on coping with specific situations and objects (e.g. 1962, 143–146). The subsequent literature on practical coping has stressed that these activities are “directed towards the object itself in all its particularity” (Kelly 2002, 385); for instance, consider Dreyfus’s work on the essentially situation-specific nature of expert coping (Dreyfus 2002, 371–372; Dreyfus 2005a, 56).⁷ States of being motivated can be thus contrasted with other states which can be directed more generally (as when one desires a good education for one’s children, or believes that people should be kind to those less fortunate than themselves). These other states can themselves be states of being motivated, in the broad sense that an agent in such a state might thereby be moved to act. However, they are not states of being motivated in the specific sense which Merleau-Ponty discusses and with which I am concerned in this paper.

Crucially, it is from objects as experienced, as they appear to the agent, that the ‘intentional stimulus’ of motivation arises: “The object stimulates me in virtue of its experienced properties and not its physicalistic ones, of which I need know nothing” (Husserl 1989, 228). This suggests that being motivated, unlike physical states such as being caused (at least on a common understanding of the latter as merely mechanical), cannot be described in terms independent of how things appear to the agent who is motivated. For example, unlike in sentences which describe causal relations, the term ‘x’ in ‘A is motivated by x to y’ is not substitutable salva veritate by co-referential terms. As Wrathall puts it, “we only capture the motivational relationship if we describe the relationship as it exists for the agent” (2005, 120). Therefore, when I speak of the motivating object or phenomenon, I am referring to the object only insofar as it appears to the subject.

Characterising motivation in terms of a three-way relation is useful insofar as it highlights what is not involved in being motivated. An agent can be motivated to act without deliberating over what to do. Furthermore, when the agent is motivated to act they need not reach a decision or will to do anything in order to act. When one’s action is motivated, it is in this sense an immediate response to what one is conscious of (Rietveld 2008, 993; Romdenh-Romluc 2011, 89–90).

The main task of this paper is to elucidate the intentional content of a state of being motivated: the way in which x appears to A so as to motivate A to y. As Merleau-Ponty puts it, “One phenomenon releases another, not by means of some objective efficient cause, like those which link together natural events, but by the meaning which it holds out” (1962, 50). This claim raises two issues: what is involved in one phenomenon releasing another; and what meaning is ‘held out’ such that it can release a course of action. I shall address the first issue in the remainder of this section, and then turn to the second.

In addressing the first issue, it is worth considering another passage from Merleau-Ponty which comes shortly after the previous quotation:

To the degree that the motivated phenomenon comes into being, an internal relation to the motivating phenomenon appears; hence, instead of one merely succeeding the other, the motivated phenomenon makes the motivating phenomenon explicit and comprehensible (1962, 50).

It seems probable that Merleau-Ponty understands an internal relation to be one where the relata are logically interdependent; that is, they cannot be defined independently of each other.⁸ The contrast is with the kind of external relation commonly taken to hold between cause and effect; a cause is “a determining factor external to its effect” (1962, 49).

Morriston argues that motivating and motivated phenomena are not internally related in this sense. For instance, regarding Merleau-Ponty’s example of a journey being motivated because one’s presence is required (because a friend has died—1962, 259), Morriston points out that it is possible that one might recognise that one’s presence is required, but not make the journey.⁹ However, this possibility does not negate the claim that a logical link holds between one’s being motivated and one’s acting in a certain way. Specifically, we cannot understand a state of being motivated without thinking of it in relation to what the agent is thereby motivated to do (though this does not entail that the state of being motivated to act in a certain way requires that one actually do so).¹⁰ It is by this logical link that “the motivated phenomenon makes the motivating phenomenon explicit and comprehensible” (this is the case even if the motivated phenomenon does not come into existence).

However, this logical link does not establish a logical interdependence between the motivating phenomenon and the motivated course of action. We certainly seem capable of understanding the concept of a course of action without appealing to being motivated to take it up. Nevertheless, there is a close link between a course of action and a state of being motivated. In a state of being motivated, a course of action is prefigured or laid out in advance in the agent’s apprehension of the motivating phenomenon. That is to say, in being motivated the agent is aware of the object, x, and is thereby aware of and drawn towards a specific course of action, y (I shall expand on this point in Section III). It is in this sense that an action which fits the motivated course of action is released by a meaning (1962, 50). I think that when Merleau-Ponty claims that there is a reciprocal relation between the motivating phenomenon and an action (1962, 259), what he is referring to is the combination of the logical dependence of the state of being motivated on the motivated course of action and the prefiguring of the course of action in the state of being motivated.¹¹

II

The preceding discussion has clarified the sense in which the motivating phenomenon releases the motivated action. I now turn to the meaning which releases the action, the intentional content of states of being motivated which prefigures certain courses of action. A key point here is that the course of action the agent is motivated to take up is presented normatively. As Morriston points out, a situation motivates one insofar as one recognises that it imposes a certain requirement: “The situation is apprehended as one that makes a certain demand, and my act is experienced as satisfying that demand. My act and the situation which motivates it are in this sense ‘internally’ related” (1979, 565; see also Merleau-Ponty 1962, 17). The relation holds between a demand which one feels, and an action’s being felt to satisfy this demand.¹² Both the demand and the satisfaction are normative, in that one feels something is called for, and a certain response feels right (Kelly 2005, 16; Dreyfus and Kelly 2007, 53). The normativity involved is relatively minimal. It need not be ethical, nor need the agent be capable of assessing the weight of the demand or considering it relative to other demands. All that matters is that they are sensitive to the demand, in the sense of feeling drawn to respond to it.

So the intentional content of a state of being motivated is normative in that it presents an object as making a demand, and a certain course of action as an appropriate response.¹³ The question is how the intentional content does this. One possible answer would be to appeal to that which is presented in the experience: the perceived objects, insofar as the subject is aware of them. On this suggestion, the intentional content of the experience of being motivated presents certain features of the subject’s environment, and these features are such as to constitute a demand on the subject.

This suggestion is unlikely to work. The problem is that it seems difficult to specify any objects, even understood in the broad sense I am using, such that being aware of them ipso facto motivates one to act in a certain way. The most plausible candidates for such objects are affordances. I think affordances are best understood as relational properties of objects which provide the agent with certain possibilities for action (Scarantino 2003, 956). However, it seems possible to be aware that x affords the possibility of y-ing, but to feel no inclination to engage in y-ing. This point is conceded by Dreyfus, who notes that one can be open to an affordance without being drawn to do anything (2007a, 361).¹⁴ Dreyfus’ own response to this is to claim that in unreflective action, “it is the affordance’s solicitations—such as the attraction of an apple when I’m hungry—to which I am directly open” (357). This may well be correct, but it does not address the question of how the affordance is presented when it does motivate or solicit a certain response. Put another way: how is an object presented such that it motivates the agent to act?

I have described the motivation as something which the agent feels: when A is motivated to y, they feel drawn to act in that way. This suggests that states of being motivated are necessarily affective. And this seems to provide a way to explain the normativity of motivational states: the object is presented to the agent in an affectively laden way, as pleasant or unpleasant, and this feeling is what draws the agent to act. States of being motivated can thus be thought of as belonging to the class of intentionally affective states or ‘feelings towards’ objects in the world (Goldie 2002, 241). The connection between affect (hedonic valence) and inclinations to act is drawn by Slaby (“By being pleasant or painful […] affective states move us to act in specific ways”—2008, 433) and by Döring (“emotions are capable of motivating because their representational content is at the same time felt, i.e., because they are affective perceptions”—2003, 224). This fits Merleau-Ponty’s insistence that affectivity is intentional, revealing objects in a particular way and inclining us towards certain actions. For example, the libido “gives sexual value or meaning to external stimuli and outlines for each subject the use he [sic] shall make of his objective body” (1962, 156).

I agree that affect is necessary for any state of being motivated. Indeed, I think that it is the affectivity of such states which makes the difference between them and states where the agent is aware of an object or situation without having any inclination to act. However, it is important to note that the presence of affect does not, in and of itself, explain why an agent feels inclined to act in certain ways. For one thing, not all affective states count as states of being motivated in the sense we are concerned with. Some affective states such as moods or existential feelings (Ratcliffe 2005) or complex emotions (de Sousa 2004, 65) arguably do not incline one towards specific courses of action at all, although they may well have an effect on one’s inclinations.¹⁵ Another issue is that even in the case of those affective states which do incline one to act, it is not yet clear how they do so. In particular, what is needed is an account of how in a state of being motivated, its affectivity (or affective perceptual content) draws the agent towards certain possibilities for acting.

III

States of being motivated are perceptual, normative and affective.¹⁶ I have suggested that each of these aspects is relevant to the agent’s being inclined to act in certain ways; but I have also noted that affect does not by itself draw the agent towards a specific course of action. My suggestion is that the affective character and the intentional content of the state of being motivated can together explain the agent’s feeling inclined to act. What is needed is an account of how these features work together. In this section I first return to the perceptual content of states of being motivated, to outline its horizonal structure. I shall then describe how this structure relates the perceptual content and the affectivity of states of being motivated. The combination of affect and perceptual content in this horizonal structure is, I suggest, what Merleau-Ponty terms the intentional arc, which situates the agent both temporally and in terms of practical contexts.

In outlining this position, it will be useful to consider some very general comments about perception made by Husserl. I suggest that these provide a model which will allow us to understand the relation between motivating and motivated phenomena.¹⁷ Consider a standard case of visual awareness, such as seeing a coffee mug. At any moment, one will see it from one angle, and so will be directly presented with one side of it, but one will see it as having further sides, which are not directly presented at that moment. We can distinguish generally between what is directly given in perceptual experience (the features of the object one actually sees or touches or tastes at any moment) and what is merely anticipated (the further features which are not directly given, but are presented as available to be directly given should one explore further). In perceiving an object, I am

vaguely aware of further kinaesthetic possibilities open to me: eye positions that I might undertake and anticipations that moving my eyes or my head, walking around an object, or reaching out to touch it, will yield a course of specific visual or tactile appearances (Barber 2008, 93).

These anticipated further features form the (inner) horizon of the experience (Husserl 1960, 45).¹⁸ They are given as relatively but not completely indeterminate. That is, they are not given merely as ‘possible further features’ of the object; rather, what is directly given (and what has been directly given in previous perceptions of the object) prescribes or pre-delineates the features anticipated (1960, 45; 1982, 107). As Mohanty puts it, “with regard to the possibilities of new discoveries about the same object—there is a sort of Vorwissen, a pre-cognition which is not quite empty, for the general line of possible determination is certainly anticipated” (1969, 140).

This pre-delineation is not a relation of strict logical entailment; it is possible for the anticipated features to not turn out as one expects, so that “the whole perception, so to speak, explodes and splits up into ‘conflicting physical thing—apprehensions’” (Husserl 1982, 332). Nor is it the case that the perceiver interprets what is directly given as suggesting further features. Rather, the directly given features, when perceived, always suggest or indicate further features of the object. The directly presented and merely anticipated features are thus bound together within the unity of the perceptual experience. Another way to put this is that the perceptual experience is characterised by a specific sort of holism: directly presented features are always experienced within a wider awareness of the object.

Furthermore, the horizon establishes a reciprocal relation between present and future experiences of the object. At present, I directly perceive certain features of the object, while others are merely anticipated. But these merely anticipated features can themselves be directly given in a further perception of the object. So the present experience, in containing an awareness of merely anticipated features, prefigures possible further perceptions of the object. These perceptions, if they occur, can confirm or fulfil this prefiguration by matching the way it presents the object as being (that is, by conforming to what was anticipated in the previous experience). However, it is always possible that further perceptions will not conform to what was anticipated. In this case, the new perception would correct the previous grasp of the object (Husserl 1960, 57–62).¹⁹

The final point to make about perceptual content is that what is anticipated includes not just further aspects of the perceived object, but possibilities for action. I referred in Section II to affordances, relational properties which provide the agent with possible ways of acting. Affordances can be given to an agent in perception in that, in perceiving an object, one can perceive it as affording various possibilities for activity (Gibson 1979, 134; 240). Many of Merleau-Ponty’s descriptions of bodily coping concern motor intentionality, the agent’s awareness of what they can do in specific situations such as a workshop or sitting at a church organ (1962, 106; 146). What agents are aware of in these cases are (or include) various affordances. Importantly, while Merleau-Ponty’s descriptions concern cases where the agent is motivated to act, this is not required for an awareness of affordances. The content I have outlined so far in this section is what one might call purely perceptual: it does not, in and of itself, involve any inclinations to act on the part of the agent (although it pre-delineates certain possibilities for acting).

I suggest that the intentional content of states of being motivated has a similar horizonal structure to the content of perceptual states.²⁰ Each state of being motivated has a specifically perceptual content (the way x appears to A). As we saw in the previous section, it also has an affective character. The point I wish to stress is that in states of being motivated, the affect is not felt independently of the perceptual content. Rather, it is integrated into the horizons of the perceptual content. That is, various possibilities for action are pre-delineated in the perceptual content of the experience, and some among these are felt by the agent as demanded. The affective character of the state is, one might say, directed at these courses of action.

Because the affective character of states of being motivated is integrated into the horizonal structure, the intentional content of these states is not purely perceptual. States of being motivated have what can be called motivational content: the object is experienced as demanding a certain response, as drawing the agent to act in certain ways. Motivational content is normative and affective: it feels like I should do this. This content is not a matter of affect alone, nor of the one state having both affective and perceptual content, but of the affective content drawing the agent to respond in a certain way to what is perceived. Because of how the object appears the agent is aware of possibilities for action, and because of what the agent feels they are stimulated to react in a certain way, to comport themselves towards the object in one way or another (Husserl 1989, 231). It is because the affectivity is integrated into the horizonal structure of the perceptual content that the agent is thereby inclined to respond in certain ways to what they perceive.

The horizonal structure of being motivated is, I suggest, what Merleau-Ponty refers to as the ‘intentional arc’:

the life of consciousness—cognitive life, the life of desire or perceptual life—is subtended by an ‘intentional arc’ which projects round about us our past, our future, our human setting, our physical, ideological and moral situation, or rather which results from our being situated in all these respects. It is this intentional arc which brings about the unity of the senses, of intelligence, of sensibility and motility (1962, 136).

The intentional arc is frequently mentioned in the secondary literature (e.g. Dreyfus 2002, 2005b; Berendzen 2010; Smith 2012), but the notion is not developed in detail by Merleau-Ponty. One of the advantages of the account I have just offered is that it allows us to understand the intentional arc as a particular kind of horizon or set of horizons.

As the quotation above indicates, one way of understanding the intentional arc is in terms of the way or ways in which the subject can be said to be situated. The most basic form of ‘being situated’ is temporal:

At each successive instant of a movement, the preceding instant is not lost sight of. It is, as it were, dovetailed into the present, and present perception generally consists in drawing together […] the succession of previous positions, which envelope each other. But the impending position is also covered by the present, and through it all those which will occur throughout the movement” (Merleau-Ponty 1962, 140; see also 239–240).

For example, when writing a sentence one does not need to consciously check what one has written, or think about what one will write. One experiences what one is doing at that moment within the temporally extended event of writing. What I wish to stress is that this temporal situatedness can be understood in terms of the horizons of the experience, specifically an awareness of past experiences and expectations of what is to come. The horizons of one’s present experience give one a sense of the movement as a whole, as an action one is performing in a particular context.²¹ The horizons of the experience thus form a holistic structure, an awareness of what one is doing as a whole rather than a sequence of isolated occurrences. This, I suggest, is the intentional arc which is projected around our experience.

IV

I have outlined how the affective character and perceptual content of states of being motivated can together incline or stimulate the agent to take up a certain course of action. I now turn to this inclination or intentional stimulation itself.

It is difficult to describe this intentional stimulation without resorting to metaphor. Husserl describes the stimulus as an invitation (1989, 269); the motivating object “intrudes on the subject”; it “knocks on the door of consciousness” (1989, 231). Merleau-Ponty speaks of the situation posing the agent a problem “set only in the form of a vague feeling of uneasiness”; of a felt tension in which something is “imminent” (1962, 17). It may be better to approach the issue by way of familiar examples: the one I have in mind is the experience of being tempted, such as when one sees a delicious-looking slice of cake. In such an experience, the perceived features of the cake (the texture, the colour, how carefully it has been designed and arranged) not only suggest further features (what the cake would look or feel like in further experience); they pre-delineate a course of action (eating the cake) which one feels stimulated to take up. There is a clear difference between cases where one is so tempted, and cases where one simply becomes aware that one could eat the object perceived, without feeling any inclination to do so.²²

It may be suggested that cases such as temptation can be explained by appealing to the agent’s desiring the cake. It is plausible that desiring is closely linked to action in a manner similar to being motivated; if one has a desire, one is thereby inclined to act in a way one believes will fulfil that desire (Alvarez 2010, 70–71). The most familiar conception of desire is that used to explain actions by appealing to combinations of beliefs and desires (e.g. Smith 1987). A desire, on this approach, is a mental state directed towards an end or goal; the relevant beliefs concern how to achieve this goal. Desires, or pro-attitudes more generally, are often held necessary to provide motivating reasons for action; indeed, they are often regarded as constituting the motivational force which beliefs alone cannot provide (Lewis 1988, 323). Might desires not also form the basis for an account of states of being motivated in Merleau-Ponty’s sense?

One problem with this suggestion is that desires, when used to explain actions in this way, are standardly understood as propositional attitudes. But a state of being motivated need not have propositional content; non-human animals which are incapable of propositional attitudes can be motivated to act (Dreyfus 2005a, 56).²³ Another problem is that desires are characterised in belief—desire psychology by their functional role, as states that ground dispositions of their agents to do certain things (Smith 1987, 52). I think something similar can be said of states of being motivated, in that an agent in such a state is thereby inclined to act in certain ways (I shall address this point presently). However, Michael Stocker has argued forcefully that feelings cannot be adequately explained by appealing to what he terms the “explanatory–theoretical account of desire”, which is more or less the functional characterisation (1983, 11–15).²⁴ If states of being motivated are necessarily affective, then it seems that we cannot explain such states by appeal to functionally characterised desires.

It might be suggested in response to this that motivations are a particular class of what Stocker terms “full-blooded, feeling-laden desire”, states which necessarily have a “felt pull, inclination, attraction” to what is desired (1983, 16). We can make this suggestion more precise still. The desires which are closest to states of being motivated are arguably what Alvarez terms “non-rational desires”: these are desires which we do not have for reasons, and some of which can be enjoyed by humans and non-human animals alike (2010, 73–74).²⁵ As examples Alvarez lists bodily appetites such as hunger and thirst which are characterised by certain bodily sensations (2010, 75–77); and other desires such as the desire to play and the desire to avoid pain, which do not have characteristic sensations (74).

However, non-rational desires differ from states of being motivated in ways which preclude explaining the latter by appeal to the former. For instance, a state of desiring cake need not be directed at any slice of cake in particular. One can feel a pang of hunger without this feeling being directed towards any specific portion of food. But one is motivated by a specific object or objects, and by how they appear to one.

Of course, it is possible to desire a particular object. Desires can have demonstrative content, e.g. I desire to eat that slice of cake. But this demonstrative content need not be determined by how the object appears to one. If a friend informs me that there is a slice of cake in the next room, I can desire to eat that cake without having perceived it or having any idea of what it looks like. Therefore, one can desire the slice without having one’s desire for it being directed or focused by how it appears to one. Furthermore, in states of being motivated, there is evidence that one’s body is geared (or gears itself) to deal with the specific features the object appears as having. This may account for the measurable difference between how a subject scales their hand in order to grasp an object when they see the object, and when they are about to perform a pantomime action, ‘grasping’ an object which is not there (Kelly 2002, 384–385). The motivated course of action is thus tied to how things appear to the agent in a much tighter way than in cases of desiring.

Cases such as being tempted by the cake illustrate the stimulation involved in being motivated. To characterise this stimulation more systematically, I propose taking as a clue the following comment by Husserl: “the subject of the motivation can at one time yield to the stimuli and at another time resist them” (1989, 199). The suggested account is as follows: A is motivated by x to y only if, on perceiving x, A will attempt to y unless they decide to not y.²⁶ That is, when one is motivated, one will either ‘give in’ to the stimulation and act as one is motivated to, or one will refrain from taking up that course of action. One does not need to decide or will to y in order to do so; one is simply drawn to act in that way, taken up in a flow of responses (Dreyfus and Kelly 2007, 52). This is the sense in which motivated actions are immediate responses to what the agent perceives. However, one’s y-ing is still under one’s control, in that one can, by an act of will, refrain from y-ing. Giving into temptation does not require willing to do so, but one can always refrain from giving in.²⁷

This characterisation of being drawn to act is limited in two ways. First, it applies only in cases where the agent is not simultaneously motivated to take up two or more incompatible courses of action. Second, it applies only to agents who have the ability to refrain from doing what they are motivated to do, e.g. rational agents such as mature humans (Dreyfus 2007a, 355; Pollard 2005, 76). One possible response to the first limitation might be to appeal to the affective character of each motivation. It is common to understand affective states in terms of degrees of intensity, as stronger or weaker feelings. The characterisation I have offered could therefore be modified as follows: A is motivated by x to y only if, on perceiving x, A will attempt to y unless (a) they decide to not y, or (b) they are motivated more strongly to attempt to y’, where y and y’ are incompatible courses of action.²⁸ To be motivated more strongly to y’ rather than to y is to be in a state of being motivated towards y’-ing with a stronger affective character than that drawing one to y. If this response could be made to work, it could also apply in the case of non-rational agents such as non-human animals by removing option (a).²⁹

I have said that the agent who is motivated to act will, unless conditions (a) or (b) apply, attempt to act in this way. By this I mean to indicate that the agent may be motivated to act in a certain way, not be motivated to do anything else, and not refrain from doing it, but may not act in that way; they may actually do something else, or nothing at all. This might seem like an obvious point, but it is worth mentioning in light of recent comments by Sean Kelly. Kelly denies the possibility of a mismatch between what one is motivated to do and what one actually does: “genuine solicitations to act cannot go astray” in this manner (2005, 19). He accepts that things can go wrong, but “never because my body has performed some activity that does not match with the activity afforded” (2005, 19); “it is in principle impossible to be solicited to act in one way while actually doing something else” (2005, 21).³⁰

In response to these claims, note first that the mismatch which I am suggesting is possible, and which Kelly denies, is between what the agent is motivated to do and what they actually do.³¹ So I can agree with Kelly that “there is no room for slippage between what the solicitation calls for and what my body is motivated to do” (2005, 19); that is not where I am suggesting the room for ‘slippage’ is found. What I am claiming is that one can be motivated to act in one way, but in fact do something else. Kelly responds to claims of this kind as follows:

in order to do something other than what I am solicited to do, I have to step outside of my engagement with the world in order to notice what I am doing and bring about a change; once I disengage myself from my activity, however, I am no longer solicited by the affordance to act (2005, 21; see also Dreyfus 2007a, 357–358).³²

I deny both parts of this claim. First, it is false that in order to do something other than what one is motivated to do, one must disengage from one’s activity. What one is motivated to do often requires a certain level of skill to execute. Someone who has not fully mastered that skill will be prone to mistakes in executing the relevant actions. And even someone who has mastered the skill can be motivated to do one thing and end up doing something else.³³ A chess player may be motivated to ward off a threat to her bishop but succeed only in putting it in greater peril; the driver may be motivated to navigate a narrow tunnel but scrape its side; the footballer may be motivated to pick out his teammate but send the ball straight to an opposing defender.³⁴

Second, while it may be correct that one must in some sense disengage from what one is doing in order to pay attention to what motivates one, it does not follow that in disengaging one thereby ceases to be motivated. It seems to me to be a quite common occurrence that one is drawn to a certain course of action, but unsure about whether or not one should pursue it; in such a case, one will often deliberate about what to do, but one will nonetheless remain drawn to that course of action. (Think of considering whether to have another drink when you know you’ve had enough; or feeling drawn to approach an attractive stranger at a party, but feeling too self-conscious to do so.) It is true that deliberation is not required for an agent to be motivated to act, but if one is motivated then deliberating does not ipso facto remove it.³⁵

It may finally be objected to my position that in order for the agent to attempt but fail to do what they are motivated to, they must have an intention to do it; and it has been denied that motivated actions can be governed by intentions to act (Kelly 2005, 20; see also Dreyfus 2000, 2002). I do not have space to address this issue, but it is worth noting that some commentators think that while being motivated does not require an intention to act, nor does it exclude intentions (Romdenh-Romluc 2011, 91–92). My own view is that many though not all of our motivated actions are governed by intentions to act. To establish this would require a careful examination of the different ways in which intentions can be relevant to one’s actions, as well as discussing the examples considered by Kelly and Dreyfus. I hope to provide this argument in another paper.

V

The account I have provided of being motivated leaves a number of outstanding issues, of which I shall mention three. First, as just mentioned, there is the issue of whether or not motivated actions can be governed by intentions to act. Second, I have not had space to consider the relation between being motivated and motor intentionality, a non-reflective mode of practical, skilled coping with the world which is typically considered to be non-conceptual (e.g. Jensen 2009). For instance, one direction for further study would be to compare my description of the intentional arc with that offered by Dreyfus in his account of skilful coping (2002).

Third, one of the most promising uses to which motivations have been put is in making actions intelligible without explaining them either in terms of causal relations or reasons for which an agent acts. Dreyfus has suggested that in addition to the space of reasons and the realm of law distinguished by McDowell, we can make certain actions intelligible by placing them in a “space of motivations” (2005a, 56). I agree with this suggestion, but would add that to make an action intelligible by appealing to what motivated it, we must have a reasonably clear conception of what it is to be motivated. I have attempted to provide such a conception in this paper; showing that it can be used to make actions intelligible requires further work.