Practical Perception John Bengson University of Wisconsin–Madison Abstract. Perceiving things to be a certain way may in some cases lead directly to action that is intelligent (e.g., skillful, wise, clever, astute). This phenomenon has not often been discussed, though it is of broad philosophical interest. It also raises a difficult question: how can perception produce intelligent action? After clarifying the question—which I call the question of ‘practical perception’—and explaining what is required for an adequate answer, I critically examine two candidate answers drawn from work on related topics: the first, inspired by Hubert Dreyfus’s phenomenological analysis of absorbed coping (and of a piece with James Gibson’s theory of affordances), focuses on awareness of situational features; the other, suggested by Gilbert Ryle’s classic treatment of knowledge-how, focuses on possession of behavioral dispositions. I argue that neither approach is adequate. Subsequently, I develop and defend an alternative answer that emphasizes the agent’s conceptual understanding.

1. Introduction The topic of this paper is perception’s role as producer of intelligent action. The phenomenon is, I think, a familiar one. Stated abstractly, perceiving things to be a certain way may in some cases lead directly to action of a certain kind—specifically, to action that is not blind but displays the agent’s quality of mind: for example, her wisdom, skill, or sensibility or, perhaps instead, her foolishness, incompetence, or dullness. I say that perception leads directly to action in the sense that there are no intervening conscious steps or transitions. One perceives, and, straightaway, one acts— not mindlessly, but intelligently. But how is it possible for intelligent action to arise directly from perception? That is a central question about intelligent action. A central question about perception is how intelligent action can arise directly from it.1 The question can be regarded as expressive of two (related) puzzles, centering on a pair of putative obstacles to such a connection. The first is prompted by the thought that the intelligence of action owes to some mental activity, such as deliberation, which precedes and somehow guides subsequent action, together with the tendency to view perception as fundamentally receptive, a passive event in which the world simply impresses itself upon the mind via the senses. How can something so passive produce anything bearing the marks of intelligence? The second puzzle arises from the dominance of what, coopting a label used by G. E. M. Anscombe in another context, we might call “an incorrigibly contemplative conception” of perception, according to which genuine perception “must be something that is judged as such [i.e., evaluated as a successful instance of its kind] by being in accordance with the facts”.2 Such a conception does not square easily with the possibility that the function of genuine perception is not merely to reflect facts but sometimes to make things happen, in its role as a source of intelligent

                                                                                                                1

My intention is not to suggest that intelligent action always involves perception as its source, or that perception always has intelligent action as its product; nor is it to deny these things. Rather, the point is to emphasize that there is at least sometimes this connection, and to ask how it is explained. This initial presentation of the question is inspired by Stampe’s (1987, 335) introduction of his query about desire. 2 Anscombe (1957, 57).

 

1

  action.3 Although these puzzles are not identical to our question, they may help to indicate some of the pressures attaching to the effort to answer it. In fact, as will emerge below, they are among the primary obstacles standing between us and a stable picture of our phenomenon, which I will call ‘practical perception’. The question of practical perception has not to my knowledge been explicitly discussed, and bringing it into view is one of the principal goals of this paper (§2). I also hope to convey the question’s interest and difficulty, by exploring how contemporary work at the intersection of philosophy of mind and action that might be thought to address it falls short (§3). Subsequently, I begin to sketch my own proposal (in §§4-6). At first glance, there might seem to be an easy answer to our question, which makes further inquiry unnecessary. Invoking the functionalist formula: in some cases, agents enjoy perceptual states of a type whose function it is to produce intelligent actions under certain conditions. This proposal is simple, and might even be correct at some level of abstraction. But even if so, it is not informative, at least insofar as it does not help to illuminate perception’s role—the function in question—in the production of intelligent action. As we shall see, adequately answering the question on which I wish to focus requires more than a functional description (definition, analysis, etc.) of a type of state in terms of its connection to intelligent action. It requires laying bare how the state, a perceptual state, manages to do what it does, or what it is its function to do, viz., to produce intelligent action. Eventually I hope to provide reason to think that there is a philosophically interesting account of how perception sometimes produces intelligent action, centering on the agent’s conceptual understanding. Schematically, the proposal will be that there exists a type of perceptual state such that a capacity to act in certain ways is partly constitutive of it. In this sense, there is a type of perception that is essentially practical. Further, it is conceptual: it is also partly constituted by a capacity for conceptualization. So, I will suggest, there is a type of perceptual state that has two interlocking dimensions, a conceptual dimension and a practical dimension; and—here is the crux—the practical dimension is grounded in the conceptual dimension. The possibility of practical perception is thus owed to conceptual understanding. The thesis that conceptual capacities play a central role in an adequate philosophical account of perception is sometimes traced to Kant and Hegel, and has featured prominently in recent controversies in epistemology and philosophy of mind.4 One of the upshots of the present discussion, supposing my proposal is on the right track, is that debates concerning the role of conceptual capacities in perception has far broader significance than has yet been appreciated, having implications also for philosophy of action, moral psychology, ethics, metaethics, and other areas in which perception might be recognized as playing a practical and not merely contemplative role. For example, an answer to our question may shed light on the motivational efficacy of perceptual experience and introduce resources to which moral realists and other cognitivists can appeal when engaging objections arising from the practicality of moral judgment.5 More boldly, perhaps, it may provide a framework within which to develop, and perhaps to vindicate, Iris Murdoch’s famous

                                                                                                                3

The contrasting features to which I am drawing attention are often expressed in the language of “direction of fit”: it is said that some mental states aim to fit or accord with the world, whereas others aim to change or make it. Practical perception appears to satisfy both characterizations, although this combination might initially seem impossible (cf. Smith 1994, §4.6)—hence, the puzzle. 4 See, e.g., Evans (1982), McDowell (1994, 2007a, 2007b), Brewer (1999, 2005), Heck (2000), Peacocke (2001), Dreyfus (2002, 2006), Byrne (2004), Tye (2006), Wu (2008), Bengson et al. (2011), Schear (2012), and Orlandi (2014). 5 See, e.g., Smith (1994, 60ff.), Blackburn (1998, 70), and Gibbard (2003, 8-17).

 

2

  suggestion that “true vision occasions right conduct”.6 Relatedly, it may illuminate an important dimension of Aristotelian phronêsis and point the way to a satisfactory philosophical treatment of “flow” and related automatic actions (sometimes commended by virtue ethicists, among others7)— although it should be emphasized up front that these compose a motley class and will, as a result, require diverse explanations.8 The next section begins to home in on our target here. 2. The Target Phenomenon 2.1 The Notion of Intelligence In chapter two of The Concept of Mind, Gilbert Ryle targeted “activities which directly display qualities of mind, yet are [not] themselves intellectual operations” (e.g., thinking or reasoning), but rather involve overt physical behavior, activities such as fishing, soldiering, or pruning trees. 9 To call attention to the relevant “qualities of mind”, or “states of intellect and character”, he isolated a class of concepts “which belong to that family of concepts ordinarily surnamed ‘intelligence’,” as expressed by adjectives such as: ‘clever’, ‘sensible’, ‘careful’, ‘methodical’, ‘inventive’, ‘prudent’, ‘acute’, ‘logical’, ‘witty’, ‘observant’, ‘critical’, ‘experimental’, ‘quick-witted’, ‘cunning’, ‘wise’, ‘judicious’ and ‘scrupulous’. When a person is deficient in intelligence he is described as ‘stupid’ or else by more determinate epithets such as ‘dull’, ‘silly’, ‘careless’, ‘unmethodical’, ‘uninventive’, ‘rash’, ‘dense’, ‘illogical’, ‘humourless’, ‘unobservant’, ‘uncritical’, ‘unexperimental’, ‘slow’, ‘simple’, ‘unwise’ and ‘injudicious’.10 As Ryle was aware, the indicated “epithets” (as he called them) can, with equal felicity, designate features of agents, as when we say that an agent is judicious, or of the actions performed by those agents, as when we say that an agent’s act was judicious, or was done judiciously. Among the latter designata are both act-types (e.g., choosing to eat healthy foods is sensible) and particular actions by particular agents—throughout, I focus on the latter.11 Although some of Ryle’s epithets have a positive valence (e.g., ‘intelligent’) whereas others have a negative valence (e.g., ‘stupid’), all belong to a common “family”. There is a sense in which deficiency in intelligence does not entail the complete absence of intelligence. To be stupid or to act

                                                                                                                6

Murdoch (1970, 64). A recent example is Annas (2011, 71ff.). 8 First, not all putative nondeliberative, automatic actions are produced by perception, as in, e.g., Arpaly and Schroeder’s (2012) case of associative thinking leading to discovery of a relevant consideration. Nor, second, are they all intelligent in the relevant sense, as in, e.g., Clark’s (2001, §2) central cases of postural adjustments and “well-oriented precision grips”; similar cases are explored by Gregory (1970), Milner and Goodale (1995), and Hohwy (2013, ch. 4). Both conditions are explained in detail in §2. 9 Ryle (1949, 26). Ryle presented these activities as those we know how to do (Ibid., 27-8), and thus the “qualities of mind” of which Ryle here speaks become crucial to his eventual treatment of knowledge how as distinct from knowledge that (and to his treatment of other notions, including moral conscience; see Ryle 1940). I have examined this further aspect of Ryle’s discussion in other work (Bengson and Moffett 2011a) and will not engage it here. 10 Ryle (1949, 25). This list is not exhaustive; Ryle provides additional examples elsewhere (see esp. Ryle 1945, ch. 2). 11 The relations between the features of agents and the features of the actions performed by those agents can be understood as follows: a particular action ϕ (e.g., pruning trees) by a particular agent x displays a state of intellect and character (e.g., displays skill) iff x exercises a state of intellect and character in ϕ-ing (e.g., x exercises skill in pruning trees); in such a case, ϕ is an exercise of a state of intellect and character (e.g., ϕ is an exercise of skill), and ϕ has a quality of intellect and character (e.g., ϕ is skillful or done skillfully). 7

 

3

  stupidly involves intelligence in a broad or inclusive sense of ‘intelligence’ that is concerned, as Ryle makes clear, not merely with “smarts”, but with a whole style of evaluation that includes all states of intellect and character, even stupidity, idiocy, foolishness, and the like. Hereafter, I will reserve ‘intelligence’ (lower case ‘i’) for intelligence in the narrow sense, namely, what is intelligent but not stupid, idiotic, and the like; I will use ‘Intelligence’ (capital ‘I’) as an umbrella term covering all states of intellect and character, including stupidity and idiocy. And I will use the adjective ‘Intelligent’ and adverb ‘Intelligently’ (both with capital ‘I’) in a similarly broad or inclusive way. It is in this sense that I will speak of action that is Intelligent, or done Intelligently.12 With this broad notion in mind, we distinguish between cases in which perception gives rise to an action or response that is Intelligent from cases in which the response is not. Consider: Subway. You are sitting on a crowded subway. You look up and see, standing before you, a heavily pregnant woman who looks weary. Straightaway, upon perceiving the woman before you, you stand up and offer your seat.13 Hunting. You are hunting deer. You hear the sound of a buck scratching his antlers and, straightaway, you stop walking, mid-step, frozen in place, your breath flattening as you listen for some further indication of its location. Additional examples can be drawn from philosophical work on tool use and automatic responsiveness. Here, for instance, is John Campbell describing the use of a hammer: Of course it can happen that you say to yourself, “this nail isn’t all that strong, better not whack it too freely,” but there are more basic cases in which you simply size up the hammer and nail and act. Your perception of the standing properties of the hammer and nail will modulate the way you use them, but not because you have engaged in any reflection on the point. In these basic cases of intelligent tool use, you size them up and then simply act.14 In each of these three examples perception gives rise to a response that is Intelligent. In Subway, offering to give up your seat is sensitive, perhaps even sagacious. In Hunting, your reaction to the sound while hunting is astute. When you hammer in the way Campbell describes, your use of the tool is skillful. In each case, your response is Intelligent. By contrast, in some cases your response is not Intelligent. For example:

                                                                                                                12

This notion of Intelligence is not purely technical but has a place in ordinary thought and discourse. For example, familiar talk of the “search for intelligent life” presumably would be deemed successful upon discovery of even very foolish, moronic, or uncritical beings: what is sought is Intelligent life. Correlatively, we can locate a broad or inclusive notion of Stupidity (capital ‘S’), which covers the complete absence of Intelligence—rather than its mere deficiency (as with, e.g., ‘dull’, ‘dumb’, or ‘stupid’ with a lower-case ‘s’). Such a notion seems to be operative when saying, “It’s just a Stupid machine.” If pressed, one might add, to make such use clear: “It’s not even a moron—it’s that kind of Stupid.” 13 A similar case is described by Blum (1991). One important difference is that Blum’s characters do not act— his perceivers are spectators, remaining sitting throughout—which means that his case does not provide an example of our target. 14 Campbell (2011, 179 emphasis added). Intelligent conversation offers another—indeed, widespread— example. Campbell does not dwell on how your intelligent tool use occurs in light of your perception, emphasizing instead the negative point that “there may be no mediation by further reasoning or reflection” and in this sense involves (what he calls) “direct wiring”. We will discuss this aspect of practical perception in the next subsection.

 

4

 

Flinch. You are sitting on a bench. Your son, intending to tease you, taps you on the shoulder. You look up and see something (your son’s hand) heading towards your face. Although it comes to a stop before making contact, you flinch. Manchurian Candidate. You are the Manchurian candidate, unwittingly conditioned in a nefarious assassination plot. Later you are shown the Queen of Diamonds and proceed directly in a trance-like state—your mind blank, without awareness of what you are doing— to take steps to kill the President. Avalanche. You are a poorly sighted, uninformed hiker caught in an avalanche. You incorrectly perceive the falling snow to be water, and begin, straightaway, to make swimming motions. Unbeknownst to you, this is the correct strategy for escaping an avalanche, and as a result, you escape the avalanche.15 In Flinch, your flinching (a reflexive reaction) is not Intelligent: you do it, but you do not do it cleverly, sensibly, dully, idiotically, sagaciously, or astutely (etc.). While it may be a sign of your exceptional constitution that you flinched as and when you did, your flinching in Flinch is not properly evaluated using an Intelligence-epithet. Likewise for your reaction to the Queen of Diamonds (a conditioned response) in Manchurian Candidate, and your escaping the avalanche (a non-voluntary action) in Avalanche.16 To be sure, in these cases your responses are evaluable in some respects. For instance, it may have been appropriate that you flinched, bad that you took steps to kill the President, and fortuitous that you escaped the avalanche. There also may be some sense in which each of these things was done well or poorly. But it does not follow that they were Intelligent, or done Intelligently. That is a dimension of evaluation that does not apply in these particular cases.17 2.2 Three Theses about Practical Perception We are now in a position to formulate three theses that jointly characterize practical perception. The first is:

                                                                                                                15

This is based on an example due to Hawley (2003), originally employed in a different context. It might be claimed that some reflexive reactions, conditioned responses, and non-voluntary actions involve immense skill and can, as a result, be evaluated using Intelligence-epithets. Perhaps this is so; though they are certainly not central cases. That noted, since I am focusing on particular actions (rather than act-types), all that is required for the present point is the relatively modest claim that at least some reflexive reactions, conditioned responses, and non-voluntary actions do not qualify as Intelligent in the relevant sense. 17 As this indicates, Intelligence-epithets comprise a proper subset of evaluative and normative epithets more generally (cf. ‘well’, ‘poor’, ‘good’, ‘bad’, ‘appropriate’, ‘successful’, ‘valid’, ‘correct’, ‘fortuitous’). Similarly, Intelligent actions comprise a proper subset of actions more generally (recall Avalanche). Further, while many actions can be done both intentionally and Intelligently, the two categories do not merely invoke distinct ideologies but arguably may come apart in particular cases. For example, the physical action shifting weight from left foot to right foot and the mental action entertaining the proposition that the number of stars in the universe is even may each be done intentionally, without being done Intelligently (e.g., cleverly or dully). Conversely, one may act unintentionally in a way that displays one’s acumen or skill—hence, Intelligence— regarding the very activity in question (as in the claim at the outset of the previous note). What explains this divergence? This is a substantive question, which I cannot fully address here. One hypothesis is that what is crucial to intentional action is belief or intention, whereas what is crucial to Intelligent action is a “state of intellect and character” (a variety of which are designated by Ryle’s epithets)—which sometimes but perhaps not always involves belief or intention. 16

 

5

  The Basic Thesis: Action follows upon perception. This thesis is not self-standing, since it calls for further specification of the action and the connection—which it describes simply as “following upon”—between the action and the perception. Such specification is provided by the remaining two theses: The Intelligence Thesis: The action is Intelligent (i.e., one acts Intelligently). The Perception Thesis: The perception is that in virtue of which one acts Intelligently.18 While the Intelligence Thesis introduces a restriction on the action, the Perception Thesis describes the perception’s connection to what follows upon it. The previous subsection explained the restriction in the former thesis, and depicted a few concrete examples involving the connection in the latter. The next subsection seeks to further explicate this connection. 2.3 The Connection between Perception and Action Recall Subway, in which you see the pregnant woman who is standing and weary, and straightaway you stand up and offer your seat. It is useful to contrast this with another scenario, in which, upon seeing the woman, you (still sitting) think to yourself: “This woman is weary and pregnant, yet standing. One who is sitting should stand up and offer one’s seat to a weary, pregnant, standing woman.”—at which point you form the intention, upon which you act, to stand up and offer your seat. Perhaps in this more reflective scenario, the thought or reasoning that is induced by your perception occurs so quickly that it is scarcely noticed, and perhaps an outside observer, admiring the promptness with which you respond, would think it did not occur at all. For all that, it occurs nonetheless, and produces the Intelligent action that you subsequently undertake. By contrast, no such thought or reasoning produces your standing in Subway: rather, upon perceiving things to be as you perceive them to be, you respond Intelligently thereby. We can also contrast Subway with a case in which one’s perception does not induce reflection, as in the scenario described in the preceding paragraph, but is rather accompanied by it. For example, you are sitting on the subway and thinking what to do. Although your eyes are closed, you are aware that you are surrounded by many who are standing. You are also aware of being in excellent health, and realize that it is likely that some who are standing are far more likely than you to need a seat. Consequently, you form the belief that you ought to stand up—at which point you open your eyes to check that you have enough space, and, in the split second that elapses before you arise, you see the woman. Once again, although an outside observer might think you are responding just as in Subway, they would be mistaken, for in this variation it is not your perception of a weary, heavily pregnant woman in need of a seat, but rather your reflection on your condition that moves you to action. These contrasts highlight both the immediacy and independence of the connection between perception and action in cases like Subway: there are no conscious processes of reasoning, deliberation, or reflection, nor occurrent acts of choice or decision, that operate alongside your perception (this is the independence) or mediate between it and your Intelligent response (this is the immediacy). In cases of practical perception, it is more like you simply find yourself acting Intelligently in response to what you perceive. Of course, one might go further, describing the connection as “automatic”, “spontaneous”, “unreflective”, or even “unthinking”. We need not object to any of these descriptions, so long as they are not misinterpreted as implying that the subsequent

                                                                                                                18

 

Unless otherwise noted, I use ‘in virtue of’ to designate not partial but full grounding.

6

  action is mindless, blind, or Stupid.19 Although your acting Intelligently is not a result of conscious processes of reasoning, reflection, or deliberation, nor an occurrent act of decision or choice, it is nevertheless Intelligent. As the foregoing already suggests, the contrasts are also meant to highlight the productivity of your perception in cases like Subway. In both of the variants described above, your occurrent intention or reflection moves you to act Intelligently: the latter occurs in virtue of the former. By contrast, in cases of practical perception, your acting Intelligently is not the product of such intention or reflection but rather of your perception: your perception is that in virtue of which you act. This is the Perception Thesis.20 Although the three features I have highlighted—immediacy, independence, and productivity—are distinct, they are intimately connected. Suppose, as per the Basic Thesis and Intelligence Thesis, that Intelligent action follows upon perception. Suppose, further, that the former’s following upon the latter is both immediate and independent (as motivated by the contrasts described above), from which it follows that the source of Intelligent action is not any conscious process of reasoning, deliberation, or reflection, nor an occurrent act of choice or decision. Abductively, the conclusion that the perception is productive—i.e., is that in virtue of which the agent acts Intelligently, as stated by the Perception Thesis—then enjoys substantial plausibility. Let me clarify two aspects of the Perception Thesis, which will play important roles in what follows. First, the Perception Thesis tells us that the perception itself—something in its nature, character, or structure—makes it the case that one acts Intelligently. One performs an Intelligent action because one enjoys the perceptual state or event that one does. This is a metaphysical thesis about the source or ground of the action’s Intelligence, and thus it requires viewing the perceptual state as playing more than a non-trivial causal role in the generation of the agent’s response.21 Second, that noted, the Perception Thesis is compatible with there being other psychological states—e.g., unconscious representations or computational states at the subpersonal level—which

                                                                                                                19

On ‘Stupid’ with a capital ‘S’, see note 12. To be clear, the Perception Thesis does not oppose the idea that your perception is an instance or expression of a choice or decision. I do not wish to deny that perception can sometimes be viewed as manifesting practical thought, but rather to guard against injecting instances of practical thought in addition to perception (in the relevant cases). 21 It should be noted that the Perception Thesis does not imply that we must hold the perception to rationalize or to provide a rational explanation for the agent’s action. Canonically, such an explanation cites the agent’s reason for acting, that is, the reason for which she did what she did (e.g., as in statements of the form ┌x’s reason for ϕ-ing was that p┐ or certain instances of ┌x ϕ-ed because p┐). As should be clear, the role of perception in the production of Intelligent action is not as one’s reason for that action, and explaining that role is not a matter of citing one’s reason, whatever it might be. Nor is it a matter of citing one’s motivation for acting. This is not just because ‘motivation’ is said in many ways; although distinguishing among these ways does help reveal why the Perception Thesis is best formulated in terms other than motivation. On one usage, ‘motivation’ designates considerations that (are taken to) favor or disfavor a particular course of action, which dovetails with the notion of a reason for action just filed as distinct from production. On a second usage, the term covers a proper subset of psychological states that play a causal role in the production of action, for example, those which push, drive, or rouse one to act. We might call these the “springs of action”, or telic states, which may include desire, inclination, impulse, craving, longing, urge, fear, or—in some cases—perception. (For recent defenses of the idea that perception is motivational in something like the telic sense, see Siegel (2014) on “soliciting affordances” and Watzl (2014) on “perceptual guidance”; both cite Kelly (2010).) However, I am interested in cases in which one is not merely roused to act but genuinely acts, and does so Intelligently. Consequently, motivation in the telic sense is also distinct from production and thus is not what is at play in the Perception Thesis. 20

 

7

  precede (and play causal roles in generating) that response.22 It is possible to acknowledge such states and their place in the overall story, while recognizing that their presence does not render the perception itself inert. Explaining the agent’s response requires explaining her performance of an action that, per the Intelligence Thesis, has a robust normative feature, being Intelligent. The Perception Thesis tells us that, in cases such as Subway, the explanation of this, the agent’s acting Intelligently, proceeds through the agent’s perception—how she perceives her environment. But again, the thesis does not thereby insist that we view her perception as the sole item that is relevant, causally or otherwise, to what she does. (We will return to both of these points below.) 2.4 Summary The point of the foregoing discussion has been to clarify our question about practical perception, which we are now in a position to see as asking how all three of the following theses could be true: The Basic Thesis: Action follows upon perception. The Intelligence Thesis: The action is Intelligent (i.e., one acts Intelligently). The Perception Thesis: The perception is that in virtue of which one acts Intelligently. I have sought to explain these theses and, to the extent possible, reveal their plausibility. First and foremost, however, they are tools for delineating our explanandum, namely, perception giving rise directly to Intelligent action. As I shall understand the problematic henceforth, if you reject one or more of the theses, then you reject our question—how can perception give rise directly to intelligent action?—and the examples that prompt it. In light of the intuitive intelligibility of both the question and the examples, however, this is not the response I recommend. Still, we should be sensitive to the possibility of compelling, non-question-begging considerations that would challenge the conjunction of the three theses without presupposing general skepticism about perception, Intelligent action, or production. Are there such considerations? Suppose there were undefeated reasons to think (i) that perception necessarily has certain characteristics, such as those putatively inactive qualities, receptivity and contemplativity, featured in our initial puzzles (from §1); and (ii) any state with those characteristics could not produce Intelligent action. These reasons would entail the impossibility of the conjunction of the three theses. To defeat such reasons, it must be shown either that perception does not have those characteristics, or that a state with them could in fact give rise directly to Intelligent action. One way to achieve this is by sketching an account of perception and its connection to Intelligent action that shows that the former is a kind of state that could give rise directly to the latter. Such an account would make room for practical perception by describing how it happens. That is what I explore in the remainder of this paper. The next section takes a first step in this direction by critically examining two candidate accounts, drawn from work on related topics: the first focuses on awareness of situational features, the second on possession of behavioral dispositions. While these approaches do not in my view provide fully adequate answers to our question, each contains an important truth that I hope to incorporate in my positive proposal, explained subsequently.

                                                                                                                22

This is as it should be: acknowledging the target phenomenon is neutral with respect to a range of theories in cognitive science about perception, which often posit various types and levels of information processing.

 

8

  3. Situations and Dispositions In a suggestive passage in The Phenomenology of Moral Experience, Maurice Mandelbaum’s distinctive phenomenological investigation of ethical life and action, Mandelbaum writes: I sense the embarrassment of a person, and turn the conversation aside; I see a child in danger and catch hold of its hand; I hear a crash and become alert to help. …[I]n such cases I am reacting directly and spontaneously to what confronts me …[I]t is appropriate to speak of “reactions” and “responses,” for in them no sense of initiative or feeling of responsibility is present. …[W]e can only say that we acted as we did because the situation extorted that action from us.23 A similar idea is found in Hubert Dreyfus’s influential treatment of the phenomenology of skilled action, which he takes to support the claim that an expert “immediately responds to the current concrete situation”:24 [I]f one is expert at the game, things are going well, and one is absorbed in the game, what one experiences is more like one’s arm going up and its being drawn to the appropriate position, the racket forming the optimal angle with the court—an angle one need not even be aware of—all this so as to complete the gestalt made up of the court, one’s running opponent, and the oncoming ball. One feels that one’s comportment was caused by the perceived conditions… [T]he situation is experienced as drawing the appropriate action out of me.25 I take it that the actions in the examples described by Mandelbaum and Dreyfus can properly be evaluated using one or more of Ryle’s ‘Intelligence’-epithets, for example, ‘quick-witted’, ‘observant’, or ‘skillful’: they are Intelligent actions. They are also produced by perception (whether sight, audition, or perhaps some complex combination of perceptual modalities). Consequently, Mandelbaum’s and Dreyfus’s discussions can be seen to speak straightforwardly to the question at hand: how does perception produce Intelligent action? Their discussions can likewise be seen as proposing an answer. Phenomenologically, from the first-person perspective, according to Mandelbaum and Dreyfus, the “situation” is experienced as “extorting” or “drawing the action out of” the agent. Insofar as they aim fully to preserve the phenomenology, how things are experienced by the agent, Mandelbaum and Dreyfus can thus be read as introducing the following account of the target phenomenon, an account that I will call extortionism: Extortionism: x’s perception gives rise directly to x’s Intelligently ϕ-ing because: x’s perception is perception of a situation and its properties that are such that, if they are perceived by an agent, then the agent thereby ϕ-s Intelligently. Such an account is compatible with the claim that some sophistication is required to perceive the

                                                                                                                23

Mandelbaum (1955, 48-49). Dreyfus (2006, 47). 25 Dreyfus (2002, 379-380). The above remarks by Mandelbaum are elsewhere quoted approvingly by Dreyfus (1992). As I understand them, Csikzentmihalyi (1990), Varela (1999), Todes (2001), Taylor (2002), Wright (2007), and Rietveld (2008, 2010, 2013) also display sympathy with the perspective articulated by Mandelbaum and Dreyfus, the latter of whom claims to be inspired by Merleau-Ponty (1945/1962), presumably owing to passages such as this: the phenomenal forces at work in my visual field elicit from me, without calculation on my part, the motor reactions which establish the most effective balance between them…” (1945/1962, 122). 24

 

9

  relevant situation and its properties. (Dreyfus, in particular, explicitly endorses this requirement; and he has offered a hypothesis about the developmental process by which a perceiver may come to achieve the requisite sophistication.26) On the intended reading of the account, however, the need for such sophistication does not diminish the significance of the situation and its properties, for it is they that make it the case that the agent subsequently acts as she does: again, when perceived, they prompt or “extort” such action. It seems incontrovertible that situational factors are relevant to practical perception. Part of what perception does is to make various elements of a perceiver’s external surroundings available to her, and thereby to enable them to affect what she does. And there is no doubt that in one way or another they do, often unreflectively. Whether a perceiver acts skillfully or clumsily, cunningly or naïvely, wisely or foolishly, she does not in any case do so in a vacuum. However, extortionism holds that external, situational factors are not just relevant, affecting what one does, but determinative, making it the case that one does what one does, viz., act Intelligently. And that is difficult to comprehend. The problem is not, or not simply, that extortionism posits environmental properties that are fairly exceptional, insofar as they are not just actionconducing—a profile emphasized by Maurice Merleau-Ponty and James Gibson—or even actionprescribing—a profile which J. L. Mackie famously alleged to be “queer”—but action-extorting.27 Rather, the worry is that extortionism is not yet a theory of the target phenomenon. In two ways, at least, it leaves out precisely what we want to know. First, extortionism does not tell us how the properties of the agent’s surroundings manage to connect the agent’s perception of them to the agent’s action in such a way as to make the action Intelligent, as in cases of Intelligence, rather than a reflexive, conditioned, or non-voluntary response to perceptual stimuli, as in cases of non-Intelligence. This states the objection abstractly, though it can also be made concrete by recalling Flinch, in which you see something coming towards your face and react accordingly. Your reaction is the result of perceiving a certain situation and its properties; but it—your flinching—is not Intelligent. Similarly for other cases of non-Intelligence, which impart the lesson that responding to perceived situational properties is not by itself sufficient for Intelligent action. Second, extortionism does not tell us how the perception manages to serve as the medium by which the properties do their work, prompting the agent’s action. The properties that extort action are said to be perceived, but it remains an open question how, if at all, they figure into the contents of perception or otherwise affect the perception’s character, nature, or structure—and, moreover, how they do so as to produce the subsequent action. In these two ways, extortionism does not take us much beyond the idea that somehow an agent’s perceiving things to be a certain way gives rise directly to Intelligent action, which is precisely what we seek to understand. One way to try to connect the dots is through a kind of disposition, which Ryle invoked in his classic treatment of Intelligent action as the manifestation of knowledge-how.28 Although Ryle did not consider our question, concerning the role of perception in the production of Intelligent action, his appeal to dispositions points us to a second possible approach to the target phenomenon, which I will call dispositionalism: Dispositionalism: x’s perception gives rise directly to x’s Intelligently ϕ-ing because: x

                                                                                                                26

See, e.g., Dreyfus and Dreyfus (1991). Merleau-Ponty’s (1945/1962) phenomenological descriptions emphasized perceiving things as, e.g., ‘to be cut up’ or ‘to be sewed’, and Gibson’s (1977) theory of affordances invoked properties such as being conducive to cutting or affording sewing; Dreyfus cites both approvingly (see esp. 2002 and 2007). Cf. Nanay (2010) on “Qable properties” and Siegel (2014, §1) on varieties of affordances. 28 Dreyfus sometimes expresses sympathy for a Rylean view of know-how or skills, as when he equates the latter with “dispositions to respond to the solicitations of situations in the world” (2002, 367). 27

 

10

  possesses a reliable disposition to ϕ, whose manifestation is triggered by x’s perception. Here, again, such an account is compatible with the claim that some sophistication is required to enjoy the perception that triggers the relevant disposition, a standing state of the agent. On the intended reading of the account, however, the need for such sophistication does not diminish the significance of the standing state, for it is the latter that makes it the case that the agent subsequently acts as she does: one sees, and this triggers a disposition; consequently, the resulting action is an actualization of the agent’s disposition to so act.29 While I propose to accept the idea that standing states of some sort are required for practical perception, there are several reasons to be unsatisfied with an answer to our question that hangs its hat on a simple appeal to perceptually-triggered dispositions to act as one did. The first is that a disposition, even a reliable disposition, can be operative even when the subsequent action is not Intelligent, but rather reflexive, conditioned, or non-voluntary. Recall the Manchurian candidate, who acts on the basis of a reliable disposition. When he is shown the card, his mind goes blank, he lacks all awareness of what he is doing, and the resulting behavior is as it were spat out by some subpersonal mechanisms, which performance is decidedly not Intelligent. Similarly, in Avalanche, your action—escaping the avalanche—results from the manifestation of a disposition triggered by perception, though it is not Intelligent. Acting on a disposition is insufficient for Intelligent action. A second challenge to dispositionalism focuses not on the Intelligence of the action that follows upon perception, but rather on the role of the perception itself. Dispositionalism locates the connection to action in something, a disposition, that is independent of any genuine perception that takes place on the occasion of action. The view thus renders perception in an important sense incidental to the Intelligence of the action that follows. Although the disposition might be thought to be intimately related to the agent’s perception, it is not itself identical to, or part of, the perception. There is the seeing, there is a separate disposition, and the former triggers the latter, which in turn gives rise to an Intelligent action. The disposition, not the perception, moves one to action. Thus the connection to action that dispositionalism is concerned to describe is not one that attaches to the perceptual state itself, but in something additional to it, namely, a disposition.30 Here one might be tempted to invoke a behavioristic dispositionalism, which holds that the perception just is, or somehow involves, a disposition. However, we cannot simply squeeze together a perception and a disposition (a standing state, presumably also with some generality) into a complex

                                                                                                                29

Cf. Papineau’s (2013, 191-2) explanation of skilled batting and other sporting behavior, which invokes “a raft of conditional dispositions”, triggered by a batter’s perception of the oncoming ball, resulting in an expert shot that is “automatic and unthinking.” Notably, however, Papineau insists that the relevant dispositions are “set” in advance through prior “deliberation and intention-formation”: “a competent batter will assess the situation and form a view about how to bat—a conscious intention to adopt a certain strategy.” This does not plausibly describe our cases of practical perception (e.g., in Subway, your perception is not preceded by any assessment of the situation and the formation of “a conscious intention to adopt a certain strategy”). The most plausible version of dispositionalism will allow the relevant dispositions to have diverse sources. 30 I think that this objection applies to McDowell’s position in “Virtue and Reason”: “A kind person has a reliable sensitivity to a certain sort of requirement which situations impose on behaviour. …The sensitivity is, we might say, a sort of perceptual capacity. …[T]he deliverances of his sensitivity constitute, one by one, complete explanations of the actions” (1979, 331-2); and so “occasion by occasion, one knows what to do, if one does, not by applying universal principles but by being a certain kind of person: one who sees situations in a certain distinctive way” (347). McDowell’s “reliable sensitivity” is a disposition that underlies and explains the agent’s perception; this is why McDowell labels it a ‘perceptual capacity’. Importantly, however, the disposition is not itself identical to, or part of, the perception.

 

11

  state and call it a day. Jonathan Dancy’s complaint, although issued in a different context, applies equally here: An adequate account must be one that gives us a recognizable notion of…perception. [It is a problem if the account leaves open that] we might be dealing with…feeling or sentiment rather than anything worth calling perception.31 It is fair to say that it is not at all clear how a behavioristic approach might work up a satisfactory answer to this complaint.32 In its most plausible form, dispositionalism will of course find some place in its story for a state that is recognizably perceptual, no less than extortionism does. One difference is that whereas extortionism regards the agent’s perceptual state as a medium for situational properties, dispositionalism views the perceptual state as a trigger for the agent’s disposition. Yet, there is also a similarity: dispositionalism, like extortionism, does not tell us much about the perception and, most significantly, how the perception manages to do what it does, but instead focuses attention on another component of the interaction (whether situational properties or behavioral dispositions). In effect, neither theory answers our question, for neither engages the matter of the perception itself, and of how it produces Intelligent action. Simply invoking an external situation or an internal disposition and tacking it on to a wholly separate perceptual state does not offer a satisfactory resolution of our question. From this perspective, the problem is not so much that extortionism and dispositionalism are false (though I believe that they are that as well), but that they do not illuminate the role of perception in the production of Intelligent action—which is what we here seek to understand. These criticisms can be summed up by reference to the three theses (from §2) that constrain an adequate account of practical perception. Neither extortionism nor dispositionalism adequately explains how subsequent action could be Intelligent, nor how perception could lead directly to it. That is, while both views accept the Basic Thesis, they seriously flout the Intelligence and Perception Theses. As a result, neither provides an adequate account of practical perception. 4. Perception, Concepts, and Action Answering our question—hence, satisfying our three theses—requires a description of the perceptual state that underwrites intelligent action, and in particular of the way in which it already expresses an activity of mind, in virtue of which the perception has the capacity to produce action that is Intelligent. On the view I wish to explore next, the relevant activity expressed by the perceptual state is conceptual. Although I lack the space here for a comprehensive defense of this approach, I will sketch its two main elements: the central features of the perceptual state, and the ‘active normative essence’ of the concepts it deploys. 4.1 Conceptual Perception In each of the cases of Intelligence, your perception is not, or at least not merely, a thoughtless confrontation with a neutral qualitative array of colors, shapes, and sounds, but the experiential presentation of a world to some extent already organized and made intelligible in the perception itself, in virtue of your conceptual sophistication. You visually experience the woman as weary, pregnant, and standing; you have an auditory experience regarding where the buck is, and

                                                                                                                31

Dancy (2010, 101). Dancy is discussing Audi’s (2010) account of moral perception. A similar point is made in earlier work by Wright (2007, §2) against McDowell (1979) and Jacobson (2006). 32 Behavioristic dispositionalism also does not resolve the initial worry that acting on a disposition is not sufficient for Intelligent action, which objection indicates that an adequate answer to our question must invoke something more or different than a disposition.

 

12

  what it is doing. You see or hear things under the indicated descriptions. The way that you experience things to be is in some sense already “conceptualized”. It will be useful to have a clear, explicit articulation of this notion, that is, the notion of a perceptual experience that is conceptual, and hence involves the exercise of conceptual capacities. There are two thoughts here. The first is that the relevant state is a perceptual experience. The second is that it is conceptual. I will offer explications of each thought in turn. One way to explicate the first is by recognizing two conditions on the relevant state, which I will refer to as ‘sensory presentation’: first, it is individuated, at least in part, by its sensory phenomenology (i.e., what it is like, sensorily, for its subject in having or undergoing it); second, its phenomenology involves the presentation—a felt presence to mind—of how things are. The latter goes beyond the negative point that perceptual experience is more than bare qualia to the positive affirmation that perceptual experience is, or at least seems to be, a “window onto the world”: it presents the world as being a certain way (even if one does not believe the world to be that way, as in cases of known illusion).33 Conceiving of perceptual experiences as such sensory presentations has substantial intuitive appeal. It also has theoretical merit: for example, it makes sense of the idea that perceptual experiences enjoy a kind of receptivity while distinguishing them from simple sensations, as well as from beliefs and various other mental states, such as imaginings, intellectual intuitions, and feelings or sentiments. Accordingly, I submit that the conditions provide “a recognizable notion of perception” (and thus avoids Dancy’s complaint, quoted in §3), and in the present context that is just what we need. Turning now to the second thought, there are many different theses that might be (and have been) taken to explicate the notion of a perceptual experience that is conceptual. Some theses focus on the idea that perceptual experience has conceptual content,34 understood as content of a kind that is or can be the content of judgment and belief, while others focus on the idea that perceptual experience requires concept possession, where a concept is understood as the constituent of a proposition, for example a Fregean sense, and not a mental representation or ability. 35 I shall use the term conceptualism to express a thesis that covers both ideas: Conceptualism: There is a class of perceptual experiences—i.e., sensory presentations—such that for each member σ of this class, (i) σ is identical to, or constituted by, a relation to a conceptual content, and (ii) an agent who has σ must possess the concepts that are constituents of that content.36

                                                                                                                33

According to Price (1946, 3, emphasis in original), “what we mean by ‘perception’ is a form of cognition in which something is presented to us.” I discuss the relevant notion of contentful presentation at length in my (2013, §§2-3 and 2015, §3); see also Chudnoff (2012) and the citations therein for contemporary authors who highlight this feature. 34 I will employ a broad use of the term ‘content’, on which a state σ has a content C iff C specifies σ’s truth (accuracy, correctness, veridicality, satisfaction, etc.) conditions. The claim that a type of perceptual state has content is not yet the claim that its contents are propositions; we might call the latter claim ‘propositionalism’. The former claim is compatible with propositionalism, but it does not require it. 35 I myself do not identify concepts with properties or attributes (e.g., they have different individuation conditions: see Burge 2012 and Schellenberg 2012 for recent discussion of Frege’s criterion). The characterizations of conceptual content and concept provided in the text are characterizations, not accounts: they are neutral between many substantive theories. Still, they are substantive enough for our purposes here. 36 The distinction between (i) and (ii) can be traced to Heck (2000, 484-485); cf. Speaks (2006) and Byrne (2009). Wu (2008, 1005) endorses (ii) plus an “involvement condition”, according to which the concepts “must also play some role in the state.” I take this condition to follow from the conjunction of (i) and (ii).

 

13

 

I shall call a mental state conceptual if and only if it satisfies (i) and (ii). In short, then, conceptualism is the thesis that there is a class of perceptual experiences—sensory presentations—whose members are conceptual: as I shall say, there is ‘conceptual perception’. Various theses closely related to but distinct from what I am calling ‘conceptualism’ have received this label. The above thesis is less controversial than many of the others, since it speaks of one class of perceptual states, leaving open that there are other classes with different profiles. Perhaps not all perceptual states are conceptual; but it is immensely plausible that some are.37 Conceptualism is the first step towards a positive answer to our question: its principal contribution, beyond clarifying the notion of perception as a sensory presentation, is to introduce concepts, and the agent’s possession thereof, to our picture of the Intelligent agent’s position or standpoint. The second step, discussed next, enriches this picture with a link between those concepts and corresponding actions that holds in virtue of the agent’s understanding of those concepts. 5.2 Actionable Concepts A person might possess a concept only insofar as she has a single attitude with a conceptual content with that concept as a constituent. Consider, for instance, a child who comes to naïvely believe an isolated claim about prime numbers simply on the basis of her teacher’s testimony. If she genuinely believes what her teacher tells her, then she possesses the concept prime; nevertheless, there is a good sense in which the child does not really understand this concept, at least not in the way that a mathematically competent adult (like her teacher), let alone an expert mathematician, does. In this way, concept possession comes in degrees. At one extreme, one has perfect mastery of a given concept, harboring no confusion whatsoever regarding its application (or conditional application). At another extreme, one merely possesses the concept, harboring substantial confusions regarding its application, but still having at least one attitude with a conceptual content involving that concept, as in the case of the child. In between these two extremes is reasonable mastery of (or, if you prefer, competence with) a concept, wherein one may harbor modest confusions regarding its application. I will use ‘conceptual understanding’ (or simply, ‘understanding’) as a gradable term that covers both reasonable mastery and perfect mastery; understanding excludes mere possession and nonpossession.38 Some concepts may be related to action, via understanding, in interesting ways.39 To illustrate

                                                                                                                37

For the view that all perceptual experiences are conceptual, see, e.g., McDowell (1994), Brewer (1999, 2005), and Bengson et al. (2011); cf. Chuard (2006) and Byrne (2004, §4). All that is needed for my purpose is that some class of perceptual experiences, such as instances of perceiving-as and -wh, are conceptual, and there being conceptual is not due to the presence of belief or judgment with respect to the experience’s content (as evidenced by cases of known illusion: I can see—visually perceive—one line as longer than another but not believe or judge it to be longer, because I know better). 38 Elsewhere (in Bengson forthcoming) I use the term ‘grasp’. My usage has a shadow in ordinary speech. For example, English allows us to say that one has a given concept but does not really understand or grasp it (as with the child), and it allows us to describe one as understanding a given concept more or less well (e.g., better, worse, poorly, partially, incompletely, fully, etc.). For theoretical treatment of the observations in this paragraph, see especially Burge’s (1990) discussion of Frege on conceptual understanding. The observation that it is possible to misunderstand or only partially (incompletely) understand a given concept that one possesses does not rely on, though it is compatible with, various forms of externalism (e.g., Burge 1979). See also Bealer (1998, pt. 2) for a careful investigation of these issues. Cf. Setiya (2013) on Murdoch’s “Platonic theory of concept possession”, which in my view is not best wedded to the pragmatic theory of concepts to which Murdoch attaches it (cf. Crary 2007, §1.3.ii). 39 For illuminating and essential discussion of the ideas that follow, I am indebted to Marc Moffett. We first explored the relation between concepts and action in our (2007, §3), where we discussed ‘ability-based

 

 

14

  the basic idea, consider the controversial Socratic view that to know the good is to do the good, but modify it so that the connection between knowledge and performance is replaced with a connection between conceptual understanding and poise, as follows: to understand the concept goodness is to be poised to do the good. The modification sets aside the possibility of knowers who lack the requisite conceptual understanding.40 It also allows for the possibility that one may never actually do the good, despite one’s understanding, perhaps because one does not have an appropriate or stable set of conative or affective attitudes (e.g., one lacks an all-things-considered desire to do the good). Even so, the modified Socratic view, if true, would imply that one is nevertheless poised to do the good, in virtue of one’s conceptual understanding. It is not necessary to endorse the modified Socratic view in order to appreciate its point. Still, it may be useful to consider another example, this time involving the concept my being in immediate danger. It is fair to say that one who understands this concept will be in a position to perform certain actions (e.g., fleeing or fighting) in a way that one who fails to understand this concept is not. Suppose that one merely possesses the concept, harboring substantial confusions about its application: for example, one takes immediate danger to be exhausted by assault from jack-booted thugs. As a result, one will not be poised to act in the ways that one who has reasonable or perfect mastery of this concept is. Because one fails to understand the concept, in a range of cases one will behave quite differently from one who does understand the concept, even if otherwise both possess the same set of conative or affective attitudes. What these illustrations illustrate is that certain concepts might be actionable: A concept C is actionable iffdef: one understands C only if one is thereby poised to perform certain actions in certain conditions. I will refer to the relevant conditions in the above definition as enactment conditions, whose specification includes specification of impediments to action that are either external (e.g., an uncooperative environment) or internal (e.g., an uncooperative set of standing states, such as adverse conative or affective attitudes).41 While enactment conditions might sometimes require awareness of various entities, including awareness of oneself, it is important to emphasize that one need not in general be aware of the enactment conditions, nor be aware that they obtain. What is required is simply that the enactment conditions, whatever they are, obtain. My definition of an actionable concept invokes the notion of poise—being poised to perform an action. As I shall understand it, being poised is or involves a type of capacity to act. But it is not the same as merely having a disposition to behavior. It is, rather, a state in which one is prepared to act insofar as one has and is aware (possibly implicitly) of a certain kind of option, or range of options, for action—even if, due to the presence of internal or external impediments, one does not actually act in those ways on any given occasion.42 This general characterization can be buttressed by examples that illustrate the relevant notion of poise and show how it is distinct from other notions, including the notion of a disposition. First, consider the familiar character, Satan, who rejects every option to do the good and instead always

                                                                                                                                                                                                                                                                                                                                                      concepts’ (which is related to but not identical to the notion of an actionable concept). Cf. Pavese (2015). 40 Cf. Greenberg (2009). 41 The enactment conditions for a given actionable concept may be determined on a concept-by-concept basis. Such conditions go by various names. See, e.g., Noë (2005) on “enabling conditions” and Sosa (2010) on the difference between the “seat” of one’s competence and one’s “situation” and “condition”, the latter two of which approximately correspond to what I will refer to as ‘external impediments’ and ‘internal impediments’. 42 This cognitive notion of poise should not be confused with the body-based notion of poise employed by some phenomenologists (e.g., Todes 2001, 65 et passim).

 

15

  embraces the option that is evil (or most evil). Nevertheless, Satan is prepared to do the good insofar as he has and is aware of that option: thus, in my terminology, he is poised to do the good. But Satan is not at all disposed to do the good; on the contrary, he is indisposed to it. (This is why the character is so evil: he has and is aware of the option to do the good, but he still does not do it.) As a second example, consider once more a Manchurian candidate, who is unwittingly conditioned in a nefarious assassination plot to take steps to kill the President, directly upon seeing the Queen of Diamonds. The candidate is disposed to take the relevant steps: retrieve the gun from its hiding place, advance to a predetermined location from which to shoot, and so forth. But it is not the case that the candidate has and is aware of any such option: in my terminology, he is not poised to take the relevant steps.43 But, again, he is disposed to take them. It should be emphasized that there are often many options that one has and is aware of when one understands a given actionable concept. Furthermore, two agents may be equally poised to act (i.e., they have and are aware of the same range of options), owed to their identical understandings of one and the same actionable concept, although they might not take one and the same option to act. For example, suppose you and your counterpart equally understand the concept my being in immediate danger and are, as a result, poised equally. However, if your external circumstances differ, you are likely to act differently: you head south, as the northern door is locked, whereas your counterpart heads north, as the southern door is locked. Differences in internal circumstances may also affect which action you take: being risk-adverse, you opt for the stairs; being lazy, your counterpart opts for the elevator. I defined the notion of an actionable concept and elucidated its key terms. A further question asks why actionable concepts are actionable: in virtue of what do certain concepts poise one for action? Here it is natural to invoke the essences of these concepts. In general, it is part of the essence of a given actionable concept C that C is satisfied by all and only what or who—a situation or person—is governed by some norm dictating that such-and-such actions are called for or merited with respect to that situation or person in such-and-such conditions. Call this an active normative essence. It is independently plausible that when one understands a concept, one grasps or knows its essence, at least to some extent.44 So, when one understands a concept with an active normative essence, one grasps a relevant norm, and thereby has and is aware of a certain option, or range of options, for action, namely, those that are called for or merited with respect to that to which the concept applies. Since actionable concepts have active normative essences, it follows that actionable concepts poise one for action. I have characterized a certain kind of perceptual state and a certain kind of concept. With these characterizations in hand, we are now in a position to formulate an answer to our question about how perception gives rise directly to Intelligent action. 4.3. The Positive Account We began with the possibility that, in some cases, when one perceives things a certain way, the perception leads one directly to act in a certain way. The question is how this happens. The answer we are now ready to consider is that in the relevant cases, one enjoys a sensory presentation involving an actionable concept one understands, and, as a result, given the absence of internal and external impediments, one acts as one does. The presentation involves conceptual understanding, in a fundamentally action-oriented way—that is how the perception gives rise directly to Intelligent

                                                                                                                43

One might insist that the candidate’s subsequent behavior must be explained by reference to subpersonal representations of the relevant steps. Even supposing this is correct, such representations do not amount to awareness, not even implicit awareness, of an option for action. 44 One possible explanation for this would relate essence to definition; see Fine (1994). Like Fine, I do not regard claims of essence as equivalent to claims of metaphysical necessity.

 

16

  action. We can call this proposal conceptualism about practical perception: Conceptualism about Practical Perception: x’s perception gives rise directly to x’s Intelligently ϕ-ing because: x’s perception, a sensory presentation, is a conceptual perception involving a relevant actionable concept C, understood by x, in the enactment conditions for C (where C is relevant just in case understanding it poises one to ϕ). On this account, there are certain conceptual perceptual states, or sensory presentations, involving actionable concepts, such that understanding of those concepts ensures that in certain conditions one will act in certain ways—Intelligently—on the basis of one’s conceptual perception. One thus enacts one’s poise, manifesting the understanding present in, and shaping, one’s perception. Conceptualism about practical perception preserves what I take to be the two truths in the extortionist and dispositionalist approaches discussed earlier: in cases of Intelligence, we must (i) find a place for situational factors (the extortionist truth) and (ii) acknowledge the relevance of standing states of the agent (the dispositionalist truth). Both figure into the enactment conditions for the relevant understood actionable concepts: (i) is handled by the requirement that there be no external impediments (e.g., an uncooperative environment), and (ii) by the requirement that there be no internal impediments (e.g., adverse conative or affective attitudes). Moreover, conceptualism preserves the truths in extortionism and dispositionalism without falling prey to the problems with those accounts. The reason is that conceptualism finds a place for situational factors and standing states not as producers but rather as potential enablers of the perception and subsequent Intelligent action, while articulating how the former gives rise directly to the latter: the perception, a sensory presentation, is a conceptual perception involving an understood actionable concept, and that is how it can give rise directly to Intelligent action. This means that situational factors and standing states of the agent need not be relied upon to do the central work of producing Intelligent action (per the arguments of §3). Instead, a conceptualist account of practical perception locates the connection to action in an aspect of the perception itself, and in something that can secure Intelligence: its conceptual character.45 As the preceding remarks suggest, I believe that this account preserves the three theses that characterize practical perception (from §2). According to conceptualism about practical perception, action follows upon perception—in tune with the Basic Thesis. Further, the link to action is due to the agent’s enjoying a sensory presentation whose conceptual content is constituted by an understood actionable concept. The fact that the agent acted as she did is thus due to her perceptual state’s being the perceptual state that it is, viz., a sensory presentation that is a conceptual perception involving an understood actionable concept. Hence, the connection to subsequent action is located in the perception itself—in tune with the Perception Thesis. Moreover, the subsequent action is at bottom the manifestation of a particular feature of the agent’s intellect, namely, her understanding-infusedperception. This is a kind of cognitive achievement, which constitutes the agent’s vantage point on her immediate environment and informs her action within it. As a result, her action is not blind, or

                                                                                                                45

Here we find a familiar philosophical structure: in a certain case or set of cases, α and β are both necessary for its being the case that there is an A that is F; α alone makes it the case that there is an A that is F (i.e., α is that in virtue of which this is so); β enables α to make it the case that there is an A that is F. In the present context, A is an action and F is the property being Intelligent. What is distinctive of the present account is its contention that in cases of Intelligence α is a conceptual perception involving an understood actionable concept; the account acknowledges situational factors and standing states of the agent, but—contra extortionism and dispositionalism—only in the role of β. Cf. Dancy (2004, §3.3) on varieties of relevance.

 

17

  Stupid, but an expression of her achievement. Hence, the account locates something that plausibly has the clout to make her action Intelligent—in tune with the Intelligence Thesis. While a conceptualist account of practical perception might be interpreted as telling us that in some cases perception “drives out deliberation”, reflection, and decision, it bears emphasizing that it does not imply that perception drives out agency. 46 To the contrary, conceptualism places the action—an Intelligent action, owing to one’s conceptual perception involving an understood actionable concept—squarely within the horizon of agential evaluation, preserving its basic links to freedom and responsibility (e.g., one is open to praise or blame for the action, which manifests one’s understanding). I have articulated the conceptualist account in broadly stative terms, invoking perceptual states and states of understanding. But there is another mode of articulation, which enables us to view the proposal through an alternative lens, centered on capacities—in particular, perceptual capacities. There is a tendency in contemporary philosophy of perception to focus almost exclusively on states (or attitudes, events, and to a lesser extent processes), which has the effect of missing or obscuring the significance of the capacities that underwrite the former’s possibility. Although the notion of a perceptual capacity has not been front and center at every point thus far, from a certain perspective it is actually at the core of the proposal. For, if conceptualism about practical perception is correct, then there is a type of perceptual state such that part of what it is for an agent to be in that state is to have a certain kind of capacity—what I have called ‘poise’—to act Intelligently. The account says, further, that she has this capacity in virtue of a conceptual capacity that she exercises in the very occurrence of the perception. The capacities are not incidental to the perception, but together form part of its ontological profile. The result is a type of perception that is constitutively tied to two distinct but interlocking capacities: one conceptual, one practical. Accordingly, the capacity for such perception—practical perception—is tantamount to a capacity to conceptualize in a certain sensory mode, and to act Intelligently on that basis.47 A conceptualist account answers our question about how perception can give rise directly to Intelligent action. It also dissolves the two puzzles mentioned at the outset, by describing how it is possible for perception to be neither wholly receptive nor incorrigibly contemplative, but practical. Indeed, to the extent that the account explains how some perceptual states are realized in production (i.e., they are such that one exercises the capacities constitutive of them only insofar as one acts in certain ways on their basis), it earns us the right—and does not merely reassert our wish—to view perception in this way.48 These are among its principal virtues.

                                                                                                                46

The quoted expression is from Thompson (2008, 108), who cites Aristotle’s Nicomachean Ethics (1112a34– b12). As it happens, Thompson himself thinks that “skill drives out deliberation”, and it is in deliberation that concepts enter, à la Kant; so, there is no role for concepts in any explanation of skilled action—a paradigm of Intelligent action. But we have seen that conceptual perception operates independently of deliberation, so we are free to accept Thompson’s remark (and, more generally, his “naïve action theory”) while also paying due respect to the importance of concepts to perception’s role in Intelligent action. We can thereby effect a reconciliation between Aristotle and Kant, or at least between some important strands in their thinking. 47 This account obviously has a McDowellian flavor. But it is not clear that it can be attributed to McDowell, for the reason explained earlier (see note 30); in addition, while he has come to emphasize the role of conceptual capacities in perception and action (1994, 2007a, 2007b), in the case of “practical intelligence” he says simply that the agent is “realizing a concept of a thing to do” in the sense that “her doing, under a specification that captures the content of the practical concept that she is realizing, comes within the scope of her practical rationality—even if only in that, if asked why she [did what she did], she would answer ‘No particular reason; I just felt like it’” (2007b, 368-9). This is a conceptualist answer to the question, not of how perception gives rise directly to Intelligent action (our topic), but of how Intelligent action expresses an agent’s rationality. 48 The account may thus help to answer what Watzl (2014) labels the “integration challenge”, concerning the relation between what he calls “informing” and “guiding” aspects of perception.

 

18

 

5. Anchoring the Account A conceptualist account of practical perception delivers a plausible real-world treatment of the two types of case with which we began, and comports with our basic self-understanding as perceivers and Intelligent agents. Let me explain. Ordinary perceptual reports are sometimes plausibly viewed as reports of conceptual perception. For example, one might say such things as: Sitter saw the weariness on the pregnant woman’s face. Hunter heard where the buck was, and what it was doing. Hammerer saw the nail as protruding from the board at such-and-such angle. Various tests indicate that these statements can be used to report sensory presentations. For example, in the envisaged contexts the clauses ‘although your eyes were shut tight’ or ‘although your ears were plugged’ cannot be added. Indeed, in each case we can easily imagine you to have a rich sensory phenomenology, and to have genuinely discovered something via your eyes or ears (etc.), such that each report gestures at how the situation is presented to you, visually or auditorily (or perhaps multimodally), in that case. No doubt the above reports are incomplete: in general, the contents of our perceptual states are far richer or more detailed than the contents of such reports.49 But we can elaborate on such ‘minimal reports’ in a way that, hopefully, brings out some of the relevant components of the contents of the perceptual experiences reported, as well as their connections to action. To focus attention, I will examine Subway, in which you see the pregnant woman who is manifestly weary and standing, and this perception directly leads to your offering your seat. Simplifying considerably, I will try to spell out in three stages how, on the present approach, practical perception works in this case. I will offer an intuitive gloss on the content of your perceptual state (stage one). This will allow us to see, I think, that in at least some instances of conceptual perception, the concepts involved are plausibly viewed as actionable and as understood by the agent (stage two). When combined with the observation that the relevant enactment conditions obtain, this places us in a position to see how the perception gives rise directly to Intelligent action (stage three). Stage One. Let us begin with the content of your perception, attempting to go beyond the minimal report given above. In the envisioned scenario, you see the situation in a certain way—it is presented to you that there is a woman, who is standing, heavily pregnant, manifestly weary, and here now, or nearby, in a specific relation to you yourself, who are sitting. In these and other ways, you do not simply register the presence of something or someone—as you might have were you to have been, for example, absorbed in a book or staring absently—but have a visual experience of the woman that takes in what her situation is (heavily pregnant, standing, and manifestly weary), as well as where she is (near enough to you, or nearby). And you do so in a particular context, replete with various social and cultural conventions, which contribute to making it the context it is. In full acknowledgment that these observations, left unsupplemented, would grossly underdescribe your perception in the envisioned scenario, it is fair to say that the foregoing serves up at least a few important elements of how thing are conceived or conceptualized when you look up: your perceptual state brings the scene before your eyes to mind under the concept there being a nearby, standing, manifestly weary, heavily pregnant woman (in this context). In point of fact, it may be most appropriate to invoke robust demonstrative, context-sensitive concepts: perhaps the best we can say is that your perceptual state

                                                                                                                49

Contra Evans (1982, 289), however, the “richness” or detail of experience does not provide the material for an effective argument against conceptualism. See Bengson et al. (2011, §3).

 

19

  brings the scene to mind under the informationally rich, situation-specific concept being thus. Because the non-demonstrative concept is more explicit, I will, however, continue to employ it. Stage Two. Let us suppose that you understand this concept, which features in the content of your perceptual experience. We can then consider whether the understood concept is actionable, and thus whether you are poised to act. We arrive at a prima facie positive verdict in three intuitive steps: (i) Intuitively, there being a nearby, standing, manifestly weary, heavily pregnant woman (in this particular context) merits a certain kind of response. If you understand the concept (in that context), then you grasp this. (ii) If you are in a position to see this, then you are in a certain kind of state: you enjoy a type of preparedness in which you have and are aware of a certain option—viz., the relevant kind of response. (iii) In this sense, you are poised to act in certain ways in certain conditions (even if you do not actually act in those ways on any given occasion).50 This three-step heuristic is not decisive, of course, but it does provide some reason to think that the concept is actionable. Further, this conclusion—that the understood concept is actionable—can be explained in the essentialist manner suggested above: it is part of the active normative essence of that concept that it is satisfied by all and only what or who is governed by a particular norm, roughly a norm to the effect that one who can easily relieve such a person—for example, by giving up a seat—ought to do so (in the relevant context). That is why your understanding of this concept makes it so that you are poised to relieve her—to give up your seat—in certain conditions.51 Let me make two comments about the first two stages before proceeding to the third. The first comment is that the supposition, with which Stage Two began, that you understand the concept in question is not arbitrary or gratuitous. To see this, consider a contrast case in which you have the indicated conceptual perception—your perceptual state brings the scene to mind under the concept there being a nearby, standing, extremely pregnant, manifestly weary woman—although you fail to understand this actionable concept. You are confused, so you do not grasp that things being as your perceptual state presents them as being merits any particular response; or, you simply take it to call for a big smile. In such a case (i.e., when step (i) does not apply), you do not enjoy the same

                                                                                                                50

This will of course incite questions about amoralists (or immoralists) who possess all of the relevant concepts while failing to be or feel motivated—in the dispositional sense—to act accordingly. There is a vast literature exploring such questions, which I cannot fully engage here (see, e.g., Brink 1986, Smith 1994, §3.3, Svavarsdóttir 1999, and Shafer-Landau 2003, ch. 7). For present purposes, it suffices to note that a conceptualist account of practical perception offers two plausible explanations of an agent who enjoys a conceptual perception involving the relevant actionable concept without being or feeling disposed to act accordingly. To see this, note that an amoralist either has or lacks reasonable mastery of this concept. In the former case, she will be poised to act—she has and is aware of the option in question—though she may still be or feel indisposed to take that option (as in the Satan example in §4.2), owing to some other aspect of her psychology (e.g., her conative or affective attitudes), in virtue of which the relevant enactment conditions will remain unsatisfied. If, however, she lacks reasonable mastery of the concept, then she is like the confused individual discussed in the next paragraph. 51 I will consider an objection focused on the significance and variability of context in a moment. A distinct objection is that the concept in question could not be actionable, because its constituent concepts—standing, weary, pregnant, etc.—are not actionable. However, even supposing that its constituents are not actionable, it is well known that complex concepts sometimes possess features not possessed by the concepts of which they are composed. A relevant illustration is provided by our earlier example of the actionable concept my being in immediate danger.

 

20

  preparedness for the relevant treatment: you would not be poised to act in the relevant ways (i.e., step (ii) would not apply). So, you would keep your seat. This ‘contrast argument’ lends support to the thesis that practical perception requires conceptual understanding, per the present proposal.52 Now for the second comment. Stages One and Two jointly imply that the content of your perceptual state includes a concept that is essentially connected to a context-sensitive norm that specifies what is to be done. Notice, however, that it does not follow that this norm itself is in the content of your perceptual state; nor that its content includes deontic concepts (e.g., ought) or has the forms there is a reason to A or A-ing is to be done. In fact, it does not even follow that the content makes explicit reference to actions you do or could perform (whether this is the determinate action you subsequently perform or a determinable action of which the determinate is an instance). In general, it is not the case that a mental state that has content ϕ also has as its content everything to which ϕ is somehow connected, even essentially so. This is not to insist that there is no room to pursue the idea that your perception has normative content, or that its content explicitly references action. I am open to both possibilities.53 The point is simply that the present account does not render either idea—each of which presupposes controversial (and, to my mind, difficult to establish and properly discipline) views in the philosophy of perception—obligatory. Granted, it avoids these presuppositions by committing to a substantial thesis in the philosophy of concepts. But I think that this thesis was given adequate support, and received a plausible explanation, in §4.2, and its application in the case of Subway is prima facie justified by the three-step verdict in Stage Two. I am inclined, therefore, to think that it is a virtue of the present approach that it focuses on concepts, and in particular that it can make sense of the connection between your perception and action, without requiring normativity or explicit reference to action in the content of the former. Stages One and Two spell out the content of your conceptual perception and indicate why we should regard that content as involving an actionable concept (again, without thereby being forced to commit to such content being normative or making explicit reference to action). Your understanding of this concept poises you to act in certain ways—to provide relief, by giving up your seat—in certain conditions. I have referred to these as ‘enactment conditions’. The third and final stage concerns the obtaining of these conditions. Stage Three. The enactment conditions specify the conditions under which you enact your poise. For you to stand up and offer your seat, for example, the circumstances must be just right. If the circumstances were different, either internally or externally, then you would act differently. To a

                                                                                                                52

It also supplies a response to the objection that a conceptualist approach over-intellectualizes practical perception. Over-intellectualization is a popular charge against conceptualist theories in the philosophies of mind and action. Typically, its primary motivation is the idea that conceptualist theories fail to account for “simple-minded” creatures who lack concepts (see, e.g., Peacocke 2001, §5). However, creatures so simple that they lack concepts—the sophistication required to categorize the world, to bring it to mind as being thus and so—are not relevant in the present context, since they do not provide instances of the target phenomenon: perception leading directly to Intelligent action. While simple-minded creatures may behave and perhaps even perform actions, it is far less clear that they perform normatively evaluable actions, and they plausibly do not perform actions Intelligently, at least not in the relevant sense of this term—that is, cleverly, sensibly, wisely, idiotically, foolishly, and so forth. 53 Cf. Blum (1991), Cuneo (2003), and Audi (2010) on “moral perception” (i.e., perceptual-experiential representation of moral properties). Siegel (2014) explores perceptual-experiential representation of “mandates”. Doring (2007), reminiscent of Nussbaum (1990), has proposed that some affective states have normative content; but her proposal does not in my view give us a recognizable notion of perceptual experience. Dancy’s (2014, §3) notion of a “practical seeming”, or presentation of a reason, comes closer; nevertheless, it is arguably better aligned with intuition and emotion, as Dancy acknowledges. Cf. Korsgaard (1996, 147-50; 2009, ch. 6).

 

21

  large extent this should be obvious. If a very large person falls on your lap, or you are handcuffed to the seat, or both of your legs have fallen asleep, then there are external impediments to standing, and you will do something else (or perhaps nothing at all). Similarly, if you are extremely timid, or unaccommodating, or deeply sexist, then there are internal impediments to standing, and once again you will do something else (or perhaps nothing at all). The absence of both sorts of impediments is required for the transition from perception to action. In the envisioned scenario, none of these impediments are present. The internal and external circumstances are precisely those specified by the enactment conditions, whose obtaining enables you, who are poised to stand up and offer your seat in virtue of your conceptual perception involving a relevant actionable concept, which you understand, to enact your poise to perform this action. To summarize: You enjoy a conceptual perception involving an actionable concept. You understand this concept, so you grasp that a particular kind of action, such as being offered a seat, is merited or called for. So you are poised to offer your seat in certain conditions. Those conditions obtain (i.e., there are no internal or external impediments). Consequently, you act: you get up and offer your seat. One might object to Stage Two, and in particular to its apparently cavalier treatment of the all-important context, pointing to the contingency of various conventions that might have in one way or another contributed to you taking the option that you did. After all, another culture might regard it as a matter of personal honor for a pregnant woman to do palpably burdensome things, like remain standing in such a situation. But surely an agent who inhabits this alternative culture could understand the complex concept in question, even though the agent would not be poised to stand and offer his or her seat. There is, I think, an important truth in this line of thought: two agents in different contexts may be poised differently in virtue of their understandings of the actionable concepts in their perceptions. In some cases, this can be chalked up to their perceptions’ involving diverse actionable concepts. I have emphasized that actionable concepts are often highly context-sensitive. Insofar as the context to which an actionable concept is sensitive includes salient contingent conventions, then we can allow that a different agent inhabiting a different context—one lacking that convention, for example—may be poised differently in virtue of her understanding of a different actionable concept. As indicated above, however, there is no incoherence in the idea that two agents in diverse contexts will be poised identically (i.e., enjoy states of preparedness in which they have and are aware of the same range of options for action), even though they act differently, as a result of other differences between their internal or external circumstances (e.g., they might possess diverse conative and affective states, due to participation and immersion in diverse cultures). Just as you and your counterpart may in one and the same situation react to a perceived threat differently, due to your diverse ambitions or tolerance for risk, despite equally understanding the actionable concept oneself being in immediate danger and enjoying identical poise thereby (recall the example in §4.2), you and your honor-sensitive counterpart may in one and the same situation react to the woman on the subway differently, due to your diverse sensibilities, despite equally understanding the actionable concept there being a nearby, standing, heavily pregnant, manifestly weary woman and enjoying identical poise thereby. In general, in those cases in which we find our phenomenon, unlike in those cases in which we do not (e.g., Flinch, The Manchurian Candidate, and Avalanche), I submit that we find in perceivers satisfaction of the following a-b-c triad: a. conceptual perception, b. an understanding of actionable concepts, and c. action that manifests such understanding.

 

22

  If this is correct, then reflection on concrete examples squares with, and supports, a conceptualist account of practical perception. 6. Conclusion I began with a question, concerning the role of perception in the production of Intelligent (capital ‘I’) action, which has not to my knowledge been explicitly discussed. I have tried to clarify the question, and in so doing to make fully explicit what is required for its answer. I argued that extant approaches to related topics are inadequate to the task, and proposed an answer that avoids the difficulties facing those approaches. The proposal runs counter to the common tendency to view perceptual experience as a wholly receptive or contemplative affair. Although the account remains programmatic, I hope to have shown that it—and with it, the question of practical perception itself— is worth serious consideration.54 References Annas, J. 2011. Intelligent Virtue. Oxford: Oxford University Press. Aristotle. 2000. Nicomachean Ethics. R. Crisp, trans. Cambridge: Cambridge University Press. Arpaly, N. and T. Schroeder. 2012. “Deliberation and Acting for Reasons.” Philosophical Review, 121: 209-39. Audi, R. 2010. “Moral Perception and Moral Knowledge.” Proceedings of the Aristotelian Society, 84: 79-97. Bealer, G. 1998. “A Theory of Concepts and Concept Possession.” Philosophical Issues, 9: 261-301. Bengson, J. 2013. “Presentation and Content.” Noûs, 47: 795-807. Bengson, J. 2015. “The Intellectual Given.” Mind, 124: 707-60. Bengson, J. Forthcoming. “The Unity of Understanding.” In S. Grimm, ed. Making Sense of the World: New Essays on the Philosophy of Understanding. Oxford: Oxford University Press. Bengson, J., E. Grube, and D. Korman. 2011. “A New Framework for Conceptualism.” Noûs, 45: 167-89. Bengson, J. and M. Moffett. 2007. “Know-how and Concept Possession.” Philosophical Studies, 136: 31-57. Bengson, J. and M. Moffett. 2011a. “Two Conceptions of Mind and Action: Knowing How and the Philosophical Theory of Intelligence.” In J. Bengson and M. Moffett, eds. Knowing How: Essays on Knowledge, Mind, and Action. Oxford: Oxford University Press. Blackburn, S. 1998. Ruling Passions: A Theory of Practical Reasoning. Oxford: Oxford University Press. Blum, Lawrence. 1991. “Moral Perception and Particularity.” Ethics, 101: 701-25. Brewer, B. 1999. Perception and Reason. Oxford: Clarendon Press. Brewer, B. 2005. “Do Sense Experiential States have Conceptual Content?” In E. Sosa and M. Steup, eds., Contemporary Debates in Epistemology. Oxford: Blackwell. Brink, D. 1986. “Externalist Moral Realism.” Southern Journal of Philosophy, 24: 23-40. Burge, T. 1979. “Individualism and the Mental.” Midwest Studies in Philosophy, 4: 73-122. Burge, T. 1990. “Frege on Sense and Linguistic Meaning.” In D. Bell and N. Cooper, eds., The

                                                                                                                54

I wish to thank participants in the ANU Workshop on Perceptual Capacities, Barnard-Columbia Perception Workshop, CSMN Workshop on Imperatival Aspects of Experience, a meeting of the Practical Philosophy Workshop at the University of Chicago, and conferees at a workshop at Vrjie University in Amsterdam. I am particularly grateful to Avery Archer, Tomás Bogardus, Terence Cuneo, Enrico Grube, Alex Grzankowski, Dan Hausman, Jennifer Hornsby, Marc Moffett, Elliot Paul, Sarah Paul, Susanna Schellenberg, Russ Shafer-Landau, Susanna Siegel, Jen Wright, and above all Anat Schechtman for extremely valuable questions, insights, and criticisms.

 

23

  Analytic Tradition. Oxford: Blackwell. Burge, T. 2012. “Living Wages of Sinn.” The Journal of Philosophy, 109: 40-84. Byrne, A. 2004. “Perception and Conceptual Content.” In E. Sosa and M. Steup, eds., Contemporary Debates in Epistemology. Oxford: Blackwell. Campbell, J. 2011. “Why do Language Use and Tool Use both Count as Manifestations of Intelligence?” In T. McCormack, C. Hoerl, and S. Butterfill, eds. Tool Use and Causal Cognition. Oxford: Oxford University Press, 169-82. Chuard, P. 2006. “Demonstrative Concepts without Re-identification.” Philosophical Studies, 130: 153-201. Chudnoff, E. 2012. “Presentational Phenomenology.” In S. Miguens and G. Preyer, eds., Consciousness and Subjectivity. Heusenstamm, Germany: Ontos Verlag. Clark, A. 2001. “Visual Experience and Motor Action: Are the Bonds too Tight?” Philosophical Review, 110: 495-519. Crary, A. 2007. Beyond moral judgment. Cambridge, MA: Harvard University Press. Csikzentmihalyi, Mihaly. 1990. Flow: The psychology of optimal experience. New York: Harper Publishers, Inc. Cuneo, T. 2003. “Reidian Moral Perception.” Canadian Journal of Philosophy, 33: 229-58. Dancy, J. 2004. Ethics Without Principles. Oxford: Oxford University Press. Dancy, J. 2010. “Moral Perception.” Proceedings of the Aristotelian Society, 84: 99-117. Dancy, J. 2014. “Intuition and Emotion.” Ethics, 124: 787-812. Doring, S. 2007. “Seeing What to Do: Affective Perception and Rational Motivation.” Dialectica, 61: 363-94. Dreyfus, H. 1992. “What is Moral Maturity?” In J. Ogilvy, ed. Revisioning Philosophy. Albany: State University of New York. Dreyfus, H. 2002. “Intelligence without Representation—Merleau-Ponty’s Critique of Mental Representation: The Relevance of Phenomenology to Scientific Explanation.” Phenomenology and the Cognitive Sciences, 1: 367-83. Dreyfus, H. 2006. “Overcoming the Myth of the Mental.” Topoi, 26: 43-9. Dreyfus, H. 2007. “Return of the Myth of the Mental.” Inquiry, 50: 352-65. Dreyfus, H. and S. Dreyfus. 1991. “Towards a Phenomenology of Moral Expertise.” Human Studies, 14: 229-50. Evans, G. 1982. The Varieties of Reference. Oxford: Oxford University Press. Fine, K. 1994. “Essence and Modality.” Philosophical Perspectives, 8: 1-16. Gibbard, A. 2003. Thinking How to Live. Cambridge, MA: Harvard University Press. Gibson, J. 1977. “The Theory of Affordances.” In R. Shaw and J. Bransford, eds., Perceiving, Acting and Knowing. Hillsdale, NJ: Erlbaum. Gregory, R.L. 1970. The Intelligent Eye. London: Weidenfeld & Nicolson. Greenberg, M. 2009. “Moral Concepts and Motivation.” Philosophical Perspectives, 23: 137-64. Hawley, K. 2003. “Success and Knowledge How.” American Philosophical Quarterly, 40: 19-31. Heck, R. 2000. “Nonconceptual Content and the ‘Space of Reasons’.” Philosophical Review, 109: 483-524. Hohwy, J. 2013. The Predictive Mind. Oxford: Oxford University Press. Jacobson, D. 2006. “Seeing by Feeling: Virtues, Skills, and Moral Perception.” Ethical Theory and Moral Practice, 8: 387-409. Kelly, S. D. 2010. “The Normative Nature of Perceptual Experience.” In B. Nanay, ed., Perceiving the World. Oxford: Oxford University Press. Korsgaard, C. 1996. Sources of Normativity. Cambridge: Cambridge University Press. Korsgaard, C. 2009. Self-constitution: Agency, Identity, and Integrity. Oxford: Oxford University Press.

 

24

  Mandelbaum, M. 1955. The Phenomenology of Moral Experience. Glencoe, IL: The Free Press. McDowell, J. 1979. “Virtue and Reason.” Monist, 62: 331-50. McDowell, J. 1994. Mind and World. Cambridge, MA: Harvard University Press. McDowell, J. 2007a. “What Myth?” Inquiry, 50: 338-51. McDowell, J. 2007b. “Response to Dreyfus.” Inquiry, 50: 366-70. Merleau-Ponty, M. 1945/1962. The Phenomenology of Perception. C. Smith, trans. New York: Routledge and Kegan Paul. Milner, D. and M. Goodale. 1995. The Visual Brain in Action. Oxford: Oxford University Press. Murdoch, I. 1970. The Sovereignty of Good. London: Routledge. Nanay, B. 2010. “Action-oriented Perception.” European Journal of Philosophy, 20: 430-46. Noë, A. 2004. Action in Perception. Cambridge, MA: MIT Press. Nussbaum, M. 1990. “The Discernment in Perception.” In Love’s knowledge. New York: Oxford University Press. Orlandi, N. 2014. The Innocent Eye. Oxford: Oxford University Press. Papineau, D. 2013. “In the Zone.” Royal Institute of Philosophy Supplement, 73: 175-96. Pavese, C. 2015. “Practical Senses.” Philosophers’ Imprint, 15: 1-25. Peacocke, C. 1992. A Study of Concepts. Cambridge, MA: MIT Press. Peacocke, C. 2001. “Does Perception have a Non-conceptual Content?” The Journal of Philosophy, 98: 239-64. Rietveld, E. 2008. “Situated Normativity: The Normative Aspect of Embodied Cognition in Unreflective Action.” Mind, 117: 973-1001. Rietveld, E. 2010. “McDowell and Dreyfus on Unreflective Action.” Inquiry, 53: 183-207. Rietveld, E. 2013. “Affordances and Unreflective Freedom.” In D. Moran and R. Thybo Jensen, eds., Embodied Subjectivity. New York: Springer. Ryle, G. 1940. “Conscience and Moral Convictions.” Analysis, 7: 31-9. Ryle, G. 1945-1946. “Knowing How and Knowing That.” Proceedings of the Aristotelian Society, 46: 1-16. Ryle, G. 1949. The Concept of Mind. Chicago: Chicago University Press. Schear, J. 2013. Mind, Reason, and Being-in-the-world: The McDowell-Dreyfus Debate. New York: Routledge. Schellenberg, S. 2012. “Sameness of Fregean Sense.” Synthese, 189: 163-75. Setiya, K. 2013. “Murdoch on the Sovereignty of Good.” Philosophers’ Imprint, 1-21. Shafer-Landau, R. 2003. Moral Realism: A Defense. Oxford: Oxford University Press. Siegel, S. 2014. “Affordances and the Contents of Perception.” In B. Brogaard, ed., Does Perception have Content? Oxford: Oxford University Press. Smith, M. 1994. The Moral Problem. Oxford: Blackwell. Sosa, E. 2010. “How Competence Matters in Epistemology.” Philosophical Perspectives, 24: 465-75. Stampe, D. 1987. “The Authority of Desire.” Philosophical Review, 96: 335-81. Svavarsdóttir, S. 1999. “Moral Cognitivism and Motivation.” Philosophical Review, 108: 161-219. Todes, S. 2001. Body and World. Cambridge, MA: MIT Press. Tye, M. 2006. “Nonconceptual Content, Richness, and Fineness of Grain.” In T. S. Gendler and J. Hawthorne, eds., Perceptual Experience. Oxford: Clarendon Press. Varela, F. 1999. Ethical Know-how: Action, Wisdom, and Cognition. Stanford, CA: Stanford University Press. Wright, J. 2007. “The Role of Perception in Mature Moral Agency.” In J. Winewski, ed., Moral Perception. Cambridge, MA: Cambridge Scholars Publishing. Wu, W. 2008. “Visual Attention, Conceptual Content, and Doing it Right.” Mind, 117: 1003-33.

 

25