Husserl and Phenomenology
Sean D. Kelly
Last modified: 9/27/01
To appear in: Robert C. Solomon and David Sherman (eds.), Blackwell Guide to Continental Philosophy, (London: Blackwell, 2002).
Phenomenology is a movement in French and German philosophy that flourished during the first half of the 20th century (roughly 1900-1950). It continues to be practiced in modified form today both in the United States and elsewhere. Phenomenology provides a foothold into philosophical problems of various sorts – from problems in philosophical logic, ontology, and metaphysics to problems about the nature of mind and the content of perception. But to a first approximation at least, phenomenology takes its start in the fundamental problem of describing accurately and completely the essential features of our everyday lived experience.
Phenomenology stands at the foundation of a wide range of 20th century philosophy as it was practiced on the European Continent. European philosophers as diverse as Derrida, Habermas, Foucault, Gadamer, Levinas, de Beauvoir, Marcel and Sartre all worked, at some point in their career, either within or in relation to the phenomenological tradition. But perhaps the three most important and influential phenomenologists, the philosophers who did most to define and develop the method and substance of phenomenology, were Edmund Husserl (1859-1938), Martin Heidegger (1889-1976), and Maurice Merleau-Ponty (1908-1961).
Phenomenology is often contrasted with the form of philosophy generally practiced in English speaking parts of the world in the 20th century. This contrast is neither very informative nor very accurate. It is true that Anglo-American philosophers such as Russell, Ayer, C.I. Lewis, Strawson, Evans, and Davidson were not influenced by – nor in many cases even familiar with – work in the phenomenological tradition. Nevertheless, the founding concerns of phenomenology are less alien to those of the Anglo-American tradition than the typical caracature would suggest. It is no concern of the present essay to defend this claim, but its truth should become apparent to those in a position to judge.
What, then, is the concern of the present essay? I cannot hope, within the constraints of the essay form, to give a full discussion of even the major contributors to the phenomenological tradition. In place of completeness, however, I offer unity. The goal of this essay is to begin an interpretation of the phenomenological tradition that identifies both the founding contribution that Husserl made to it and the way in which this contribution was appropriated, refined, and finally in some ways rejected by his successors.
I say I will begin the interpretation; what I have to offer is incomplete in at least three respects. First, I do not discuss Heidegger’s important contribution to phenomenology in any substantive way at all. The interpretation has a place for Heidegger’s contribution, and indeed it has been strongly influenced by the way I see Heidegger’s role in the phenomenological tradition. But the story I tell here will focus almost exclusively on the work of Husserl and Merleau-Ponty.
Second, in presenting the work of Husserl and Merleau-Ponty, I will not attempt to do justice to its development over time. This certainly goes against the main tendency of the secondary literature. Commentators, for instance, often identify three distinct periods in Husserl’s phenomenological work, and it is common to differentiate between at least an early and a late Merleau-Ponty. In contrast, I will emphasize a unified strain of thought that I believe characterizes the general outlook of each of these philosophers over the course of his career. No doubt there are particular passages from various periods that go against features of this general interpretation. But I will not attempt to defend it against them here.
Finally, a more comprehensive interpretation of the phenomenological tradition would attempt to place it in the larger context of 20th century philosophical work. In particular, it would discuss in some detail the relation between the concerns of the phenomenologists and those of the early 20th century Anglo-American philosophers. There is much to be gained from such a discussion, but it goes beyond the scope of my present ambition.
There is an important sense, then, in which my project is stiffly constrained. Despite its apparent modesty, however, I believe it is a real achievement to give even the start of a unified interpretation to the phenomenological tradition. Indeed, it is an achievement that the figures within the tradition notoriously failed to attain. Heidegger’s efforts to distance himself philosophically from his mentor Husserl, for instance, are well-documented. Their only attempt to work together – in producing an article on “Phenomenology” for the Encyclopaedia Brittanica in 1928 – ended in a rupture that lasted the remainder of Husserl’s life. And although Merleau-Ponty, in contrast, attempted to align himself strictly with the work of Husserl, this effort provides one of the more notable failures in his corpus. Nevertheless, it is not by chance that it was Husserl in relation to whom Heidegger and Merleau-Ponty chose to work. What advance Husserl made over his predecessors, then, is one of the guiding questions for this essay. What advance his successors made over him, of course, is the other.
I have said already that phenomenology takes its start in the fundamental problem of describing accurately and completely the essential features of everyday lived experience. By “everyday lived experience” I mean the kind of active, engaged experience we have of the world throughout the course of our everyday life: hearing the toll of a campus bell, seeing the smile of a friendly face, grasping a coffee mug by the handle and bringing it to one’s mouth to sip. These experiences present the world to us; they do not – at least not in the first instance – present our experience of it. By the “essential features” of everyday lived experience I mean those features that are necessary and sufficient for them to be experiences, and in particular for them to be the very experiences that they are. Phenomenology’s most basic premise is that it is more difficult to capture the essence of everyday lived experience completely and accurately than one might have thought.
William James, a contemporary of Husserl’s whom he admired very much, characterized this difficulty well. James considered the example of hearing a bell toll. It sometimes happens, he said, that I realize all of a sudden both that the bell has been ringing for some time and that I’ve been counting the rings. Perhaps when this realization dawns on me I’ve already counted the first four rings and am in the process of counting the fifth. James asked the difficult question, In what sense were first four rings present to me in experience? What, in other words, were the essential features of my experience of them?
The question is difficult because the two obvious possibilities – full conscious awareness and complete unawareness – are ruled out of court. We can’t say that I was aware of the first four rings in the same way I was aware, fully consciously, of the fifth; for if we did there would be no substance to the sudden realization that distinguished the one from the others. But we can’t say I was completely unaware of the first four rings either, for otherwise I wouldn’t have been able to keep track of them as distinct entities in a series. Counting is precisely the kind of conscious activity that seems to require awareness of entities as such. To account for the details of this experience, then, we need a new category that lies somewhere between full conscious awareness and complete lack of awareness. But how to characterize this kind of dim awareness is not immediately clear. Husserl thought that the experiences of everyday life are replete with various kinds of dim awareness of this sort, and that it is the most basic task of phenomenology to characterize them.
Husserl came upon this project in a roundabout way. A mathematician by training, his first work in philosophy focussed on the philosophy of arithmetic. After a critical review of this work by Gottlob Frege,  however, in 1894, Husserl turned his attention more generally to foundational problems in philosophical logic. His goal was to develop a philosophical approach to logic that not only accounted properly for the formal relations allowable between propositions, but also for the content found within them.
In order to get “philosophically clear” about the contents of propositions, Husserl believed, one must think about the mental states that typically invoke them. In the first place these are linguistic utterances. So Husserl began his phenomenological inquiry by asking how linguistic utterances come to be the kinds of intentional structures that they are; how they come to be, in other words, mental states that are characteristically of, about, or directed towards objects and states of affairs in the world.
The key to answering this question, according to Husserl, lies in an analysis of the experiences that, in the most basic cases, make our linguistic utterances about the world possible. This analysis reveals two important facts. In the first place, everyday experiences, like the linguistic utterances they make possible, are intentional: we hear the toll of the campus bell, we see the smile of a friendly face, we grasp the coffee mug by the handle. Of course we can sometimes have episodes of conscious awareness – like hallucinations or dreams, for instance – that aren’t directed toward actually existing objects. Perhaps we can even imagine a free play of conscious awareness – a manifold sensation of color, shape, and texture, for instance, without any awareness of these as the things they are. But perception is in the most basic cases directed toward objects and properties as such, and these other cases are the exception instead of the norm.
The second important fact about experiences, according to Husserl, is that they always reveal their object from a perspective. This perspectivism is natural for bodily perceivers like us who are restricted to spatio-temporal points of view on the world, though of course it would not apply to omniscient knowers who are capable of taking a so-called “view from nowhere.” That we are not such omniscient knowers is a phenomenological insight that bears much fruit.
When we combine the perspectivism of experience with its intentionality we come upon a phenomenologically fascinating problem. For although experience can only reveal its object from a perspective, it is the full three-dimensional object toward which we are intentionally directed when we have an experience of it. To capture this fact Husserl says that objects are presented in experience as transcending – or “going beyond” – the experience we have of them. But how can experience be essentially perspectival and at the same time present objects to us as transcending the perspective we have on them? Phenomenology’s founding problem is to account for this possibility.
Husserl’s understanding of intentionality in terms of transcendence, and especially his understanding of transcendence in the context of perception, was a crucial breakthrough peculiar to phenomenology. Both Heidegger and Merleau-Ponty took their start from the basic orientation that it provided. To characterize the breakthrough in Heideggerean terms, we can say that Husserl had learned to ask (though not necessarily to answer) the question of the being of entities. He had learned, in other words, to puzzle over the way in which entities are present to us in our experience of them. His was the first move beyond the blind Cartesian dogma that objects, even in our experience of them, can be no more than extended things.
If Husserl’s characterization of objects was a revelation, however, his understanding of intentional states generally, and of experiences in particular, was hampered by a much more traditional commitment. Intentional states, according to Husserl, are immanent instead of transcendent. Instead of going beyond what we experience of them, in other words, our experiences, when we reflect upon them, are present to us all at once in their entirety. The immanence of intentional states, according to Husserl, stands in stark contrast with the transcendence of the objects toward which they are directed.
Both Heidegger and Merleau-Ponty reject Husserl’s claim that intentional states are immanent in this sense. In Heidegger’s language, Husserl’s commitment to this idea reflects his inability to ask the question of the being of intentional consciousness. It reflects his inability, in other words, to puzzle over the way in which experiences themselves are presented to us. By simply assuming that experiences are presented to us all at once in their entirety, Husserl leaves out the possibility that we could discover facts about an experience that we weren’t aware of explicitly when we were in the midst of it. He leaves out the Jamesian possibility, for instance, that there was more to our experience of the first four rings of the bell than what we noticed about it at the time. And perhaps there are other kinds of experience that are ruled out as well.
Merleau-Ponty, indeed, thought that many genuine phenomenological features of experience are excluded if we accept Husserl’s assumption that intentional states are immanent. He pursued his work on this topic from an interdisciplinary perspective, both as Professor of Child Psychology at the Sorbonne and later as Chair in Philosophy at the Collége de France. This interdisciplinary perspective gave him a wide array of physiological, psycho-physical, phenomenological, and philosophical data on the basis of which to evaluate Husserl’s claim. In the end, though, Merleau-Ponty based his rejection of the immanence of intentional states largely on his analysis of the phenomenology of unreflective bodily experiences such grasping and other skilled visuo-motor activities. As we will see, he argued that these kinds of bodily activity represent the world in a way that goes beyond what we can capture of them upon reflection. This is so, according to Merleau-Ponty, because the way of representing the world that is manifest in our bodily activity depends intimately upon the situation in which that activity occurs. Once we step out of that situation to reflect upon the activity itself, we change the content of the representation that was manifest in it.
Merleau-Ponty rejects, therefore, the Husserlian principle that intentional states are immanent, because it leads to descriptively inaccurate claims about the nature of bodily experience. Nevertheless, Merleau-Ponty’s rejection of this Husserlian thesis takes place against the background of the phenomenological advances that Husserl made over those who came before him. In the sections that follow I will attempt to defend this claim in greater detail.
Husserl’s phenomenological account of intentionality is based on, but supercedes, the work of two influential predecessors. From the account of perception developed by the British Empiricists Husserl takes two ideas. The first is that perception is, in some sense, the most fundamental, and therefore the paradigmatic, mental state. The second is that perception is in essence perspectival. From his teacher Franz Brentano, on the other hand, Husserl inherits the idea that mental states in general are characterized by their intentionality. Combining these two views, Husserl attempts to develop a phenomenological account of intentionality that takes perception (rather than belief or judgment) as the paradigmatic intentional state.
As we have seen, the central feature of Husserl’s account is that objects are experienced as transcending the intentional states directed toward them. In this lies the principle advance of Husserl’s theory. First, it encourages him to emphasize (against the empiricists) that perceptions are more than mere perspectival images of their objects. Second, it encourages him to emphasize (against Brentano) that intentional states, nevertheless, present their objects from a perspective. The goal of this section is first to characterize these advances, and then to show how Husserl’s commitment to the immanence of intentional states constrains his phenomenological account of them.
2.1 Perception as the paradigmatic intentional state for Husserl
Husserl believes that experiences make it possible for our thoughts to be about the world. He inherits this belief from the empiricist tradition of Locke, Berkeley, and Hume. In particular, Husserl believes both that experience is the “ultimate source” of thought, and that the perception of physical things is the paradigmatic kind of experience. That experience plays such a crucial role in the development of Husserl’s phenomenology is not often stressed.  It is nevertheless, I believe, the key to understanding Husserl’s central contributions.
Husserl focuses on perception because he believes, as he says in Ideas I, that perception is that “primal experience from which all other experiencing acts derive a major part of their grounding force,” and because he believes experience generally conceived is the “ultimate source” of intentional life. Further, he focuses on the perception of spatio-temporal objects in particular because he believes that “it is sufficient to treat perception of the physical thing as the representative of all other perceptions (of qualities, processes, and the like).” These concerns with perception, and in particular with the perception of spatio-temporal objects, are apparent from the beginning of Husserl’s phenomenological work.
It is a commonly agreed that at the end of his career Husserl emphasizes the grounding role that perception plays in intentionality. For instance, in Experience and Judgment, an unfinished text from the last period his life, Husserl claims that one of the central goals of phenomenology is to characterize the relation between judgments and the underlying “pre-linguistic” experiences that make them possible. Likewise, the passages quoted in the paragraph above show that in Ideas I, the canonical middle-period text published in 1913, Husserl considers experience, and indeed the perception of spatio-temporal objects, to be a central concern. But it is interesting to notice that already in 1901, in the Logical Investigations, Husserl clearly states his belief in the importance of experience to the possibility of intentional life. He writes there:
If we imagine a consciousness prior to all experience, it may very well have the same sensations as we have. But it will intuit no things, and no events pertaining to things, it will perceive no trees and no houses, no flight of birds nor any barking of dogs.
In short, without experience our mental states are not directed toward objects in the world.
Furthermore, it is clear that the perception of physical objects in particular is central to Husserl’s understanding of intentional objects generally. This is true even at the time of Husserl’s early phenomenological work between the Logical Investigations and the Thing and Space Lectures of 1907. As we will see, the perception of physical objects provides a paradigmatic model for Husserl in his development of the idea that intentional states are directed toward objects that are transcendent to them. This notion of transcendence is central to Husserl’s account of intentionality, and it is a feature of intentionality that is clearest in the perceptual case. Although I will not be able to discuss it much here, I believe that this makes Husserl’s treatment of the problems of intentionality importantly different from the treatment they are typically given by Anglo-American philosophers working in the tradition of Frege, a tradition that famously emphasizes the methodological priority of language over experience.
2.2 Husserl’s argument against the empiricist image theory of perception
The empiricists believed, like Husserl, that perception is perspectival (using the word very generally for the time being). Something like this idea is already pre-figured in the work of the Rennaissance painters on the laws of perspective. Indeed, the empiricist image theory of perceptual representation builds on the work of the Rennaissance painters, since it is based on the idea that what we directly perceive is internal, perspectivally rendered images of objects. Perception is intentionally directed toward physical objects in the world, according to this view, in virtue of the similarity that obtains between the internal image and the physical thing it’s an image of. As Bishop Berkeley’s character Philonous puts the point, aping the position of his antagonist Hylas in the Three Dialogues:
It seems then, you will have our ideas, which alone are immediately perceived, to be pictures of external things: and that these also are perceived by sense, inasmuch as they have a conformity or resemblance to our ideas.
The empiricist idea that we immediately perceive pictures or images, instead of full three-dimensional objects, emphasizes the perspectival nature of perception.
Husserl emphasizes his version of the perspectival nature of perception in the following passage from the Sixth Logical Investigation. In perception, he says,
The object is not actually given, it is not given wholly and entirely as that which it itself is. It is only given ‘from the front,’ only ‘perspectivally foreshortened and projected,’ etc. ... [T]he elements of the invisible rear side, the interior, and so on, ... are not themselves part of the intuitive ... content of the percept.
Husserl goes beyond the empiricist theory, however, when he insists that the experience of an object is not properly characterized as the mere experience of a perspectival image of it. There is a distinction, Husserl insists, between the presentation of a visible side of an object, which is what the empiricist offers, and the presentation of the object from a side, which is what we get in genuine perception. As Husserl says in the Fifth Logical Investigation, it is an essential feature of perception that it
enables us to go beyond the “image” which alone [according to the empiricists] is present in consciousness, and to relate to it as an image to a certain extraconscious object. ... [R]elation to its [transcendent] object is part and parcel of the phenomenological essence of consciousness.
Perhaps an example will make this claim clear. Suppose I see something that I take to be a coffee mug. Necessarily I see it from some point of view. But that doesn’t mean that my perception of what I take to be a coffee mug is the same as my perception of what I take to be the relevant coffee mug façade. It is not. It is true, of course, ex hypothesi, that the very same pattern of colors is projected onto my retina in both cases; for the empiricist the very same image is perceived. Nevertheless, in the first case I experience much more than a mere façade. I see a coffee mug presented from one perspective, and I see it as something that transcends the perspective I have on it. I see it as a full-fledged three-dimensional object, in other words, a thing that has various sides not now visible to me, and whose various hidden sides each has its own color, shape, size, texture, and so on. This is part of experiencing something as a coffee mug, and it distinguishes that from experiencing it as a coffee mug façade.
This phenomenological distinction, Husserl argues, is something the empiricists cannot account for. As a result, their image theory of perception is wrong. The argument for this claim reappears later in Merleau-Ponty, but Husserl had the gist of it.  It goes like this: In order to account for my experiencing something as a coffee mug, the empiricist would have to argue that the image projected by the object I’m experiencing resembles a coffee mug more than it does a coffee mug façade. Since the mug and the mug façade present exactly the same image, however, it cannot by itself resemble either object more than the other; it is the image of both equally. For the empiricist, therefore, there can be no distinction between experiencing something as a coffee mug and experiencing it as a coffee mug façade. Since by hypothesis there is a distinction between these experiences, the image theory of perception is wrong.
Husserl’s improvement over the empiricists is to insist that we don’t get raw, uninterpreted images in consciousness, but data that are already interpreted as images of some object or another. In order to characterize this distinction Husserl says that in perception we are presented not with mere images of the visible side of an object, but with adumbrations (Abschattungen) of the object itself. The adumbration of the object that is presented in perception is the visible side interpreted as a side of the transcendent object that goes beyond it. We will see later what this interpretation consists in for Husserl. What’s important for the moment is to notice that the motivation for this view is Husserl’s desire to account for the phenomenological fact that normal perception is intentionally directed toward objects rather than just toward perspectivally rendered images of them.
2.3 Husserl’s advance over Brentano
Husser’s phenomenological account of intentionality represents an advance over Brentano as well. Brentano believed, like Husserl, that mental states are intentionally directed toward objects.  Indeed, for Brentano intentionality – directedness toward an object – is the defining feature of the mental. In order to make sense of the possibility of intentionality, however, Brentano employed the medieval doctrine of mental in-existence – the doctrine that every mental state contains its object completely within itself. The intentional object, in other words, is immanent to the mental state.
A major motivation for this account of intentionality – as Brentano later makes clear – is that the object the thought is about need not exist in reality for the thought to be intentionally directed toward an object. For instance, suppose I deny the existence of a certain golden mountain, and suppose that this denial is justified – no such mountain exists. Nevertheless, my thought has an intentional object – if it didn’t it would not be a thought at all. The intentional object must therefore not be an object in the physical world. As Brentano puts it:
If someone thinks of something, the one who is thinking must certainly exist, but the object of his thinking [i.e., what the thinker has as his object] need not exist at all. In fact, if he is denying something, the existence of the object is precisely what is excluded whenever his denial is correct. So the only thing which is required by mental reference is the person thinking. The terminus of the so-called relation does not need to exist in reality at all.
Though the terminus of the relation need not exist in reality, however, it must have some kind of existence, according to Brentano, or else the mental state will not be directed toward anything; i.e., it will not be a mental state. Brentano’s suggestion, therefore, is that the intentional object exists immanently within the mental state directed toward it.
This said, Husserl’s advance over Brentano should be clear. Although Brentano insists, like Husserl, that mental states are intentionally directed toward their objects, he has no room for the distinction between the extramental object presented and the perspectival presentation of it in experience. If intentional objects are immanent to mental states – presented to them completely and all at once – then they cannot also go beyond what is presented in the mental states. If Brentano’s conception of intentionality were applied to the perceptual case, it would have the effect that we experience something like a Cubist presentation of all sides of the object simultaneously. Clearly this is not a phenomenologically adequate account of experience. As Husserl says,
A three-dimensional intuition ... one that would bring to presentation all at once the full content of the thing in each of its constitutive parts and moments, outer and inner, front and back, is impossible.
This impossibility is what Brentano is incapable of accounting for. His linguistically-focussed motivation may be what misleads him in this regard.
The empiricists and Brentano, therefore, make complementary mistakes. The empiricists fail to notice that objects, rather than visible sides of objects, are presented in experience. Since images are nothing more than perspectival renderings of the visible sides of an object, and since there is no way to account for the possibility that these renderings should actually be directed toward the physical object that exists outside the mind, the empiricist image theory of perception is incapable of accounting for intentionality at all (with respect to physical objects). Brentano’s weakness is the complementary one. Although he saw the directedness of mental states as their defining feature, he failed to notice that perception does not grasp its object as a whole and all at once, but always comes at it from one side or another. His doctrine of the mental in-existence of objects is by definition a rejection of the extramental transcendence of objects to intentional states. In a word, then, the empiricists couldn’t make sense of the idea that it is objects we’re directed toward, while Brentano couldn’t make sense of the idea that the extramental objects that we’re directed toward transcend our experience of them. The great advance of Husserl’s phenomenology is that it is predicated on making sense of both of these ideas at once.
2.4 On the immanence of intentional states
Husserl’s distinction between the presentation of an object and the object presented, we have seen, is a central feature of the phenomenology of perception. Since neither the empiricist account of perception nor Brentano’s account of intentionality preserves this distinction, the question naturally arises, “What account of our mental states can make sense of the fact that objects are presented to us as going beyond our experience of them?” In the next section I will examine Husserl’s answer to this question. In this section, however, I will discuss Husserl’s idea that intentional states are themselves immanent instead of transcendent. Husserl’s commitment to this idea ultimately constrains the answer he can give to the question of object transcendence.
Although the idea that objects transcend intentional states is an important breakthrough for Husserl, the idea that mental states are themselves immanent is much more traditional. In Husserl’s mature work this idea supports four related features of an intentional state: the indubitability of its existence, the incorrigibility of the subject’s knowledge of its qualities, its metaphysically basic nature, and its structure as an essence. Only the first two of these will be important here.
Both Descartes and the empiricists agreed that our knowledge of our own mental states is indubitable. Although I can doubt whether the thing my thought is about exists, they believed, I cannot doubt whether the thought about it does. This kind of indubitability, Husserl also believes, is one of the defining features of our mental states. His view on this issue is perhaps most clearly expressed in Ideas I. Suppose I find, in reflecting upon it, that I now take myself to be perceiving a table. “It would be a countersense,” Husserl claims, “to believe it possible that a mental process given in that manner does not in truth exist.”
More than that, however, the qualities I take the experience to have, in reflecting upon it, are certain to characterize the experience as it really is. My knowledge of them, in other words, is incorrigible. This is perhaps the most important aspect of Husserl’s claim that intentional states, and especially perceptions, are immanent. Because they do not present themselves perspectivally, like physical objects do, there is nothing to any given perception beyond what I see in it:
Everything which we have worked out about the givenness of the physical thing loses its sense here, and one must make that fully clear to oneself in detail. A mental process ... is not adumbrated. If I look at it, I have something absolute; it has no sides that could be presented sometimes in one mode and sometimes in another. ... [W]hat I see when I look at it is there, with its qualities, its intensity, etc., absolutely.
In this way all intentional states are, for Husserl, a bit like qualia, at least as they are understood by some recent writers in contemporary philosophy of mind: if I take them to exist they do, and as I take them to be they are. The interesting thing about qualia, however, is that they are not typically taken by themselves to have intentional features. This is because there is something uncomfortable about the combination of intentionality and incorrigibility, and, as we will see, Husserl’s blind commitment to the latter directs and in some ways invalidates his treatment of the former.
Many criticisms of Husserl have focussed on his commitment to the immanence of intentional states, or on features of his phenomenology that arise out of it. It is this principle, for instance, that leads to Husserl’s famous, and famously controversial, transcendental reduction. The transcendental reduction proceeds by “bracketing existence,” that is to say, by looking at the features of our pure mental states independent of the things in the world toward which they are intentionally directed. Such a procedure makes sense only on the assumption that intentional states form a purely independent realm, an assumption that is justified by the claim of immanence. Eventually the claim of immanence leads Husserl to argue for the ontological priority of transcendental subjectivity, and indeed for a kind of transcendental idealism. In my discussion of Merleau-Ponty I will say more about criticisms that focus on this aspect of Husserl’s work.
The central point that I wish to emphasize here, however, is that Husserl’s commitment to the immanence of perceptual states – and especially to the incorrigibility of the subject’s knowledge about them – strongly constrains any account he can give of how perceptual states represent their objects as transcending them. For if a subject’s knowledge about his perception is incorrigible, then there cannot be any further question about the features he take its object to have. Those features, in other words, cannot but be presented to him as completely determined. Husserl’s trick, as we will see, is to allow for the possibility that the subject sees an object to have a certain determinate kind of feature without now being presented the determinate feature itself. But every aspect of a subject’s perception, according to Husserl, must be either a determinate presentation or the kind of thing that could later become a determinate presentation. As we will see in the Merleau-Ponty section, this is a metaphysical constraint that is not justified by the phenomenological facts.
2.5 Husserl’s answer to the question how intentional states can refer beyond themselves
What account of our mental states does Husserl give, then, that explains the phenomenological fact that objects are presented to us as going beyond our experience of them? The central feature of Husserl’s account is that the raw data of sensation (what Husserl calls the “hylé”) are not experienced as such in their raw, uninterpreted state. Rather, they are always interpreted as data that present some object or other. Further, the object they are interpreted as presenting is, in my perception of it as that object, understood to have features that are not presented determinately by the hylé themselves. In this section I will attempt to clarify just what form this interpreted presentation takes.
The hylé, under a given interpretation of them, are central to what Husserl comes to call, by the time of his middle-period works, a noema (or more particularly, a noematic Sinn).  Versions of the noema, though not under that name, can be found as early as the Logical Investigations. One basic job of the noema is to categorize the hylé as falling under a concept, or fitting, as Husserl says, into some conceptual “frame.” The outline of the idea is essentially Kantian. For instance, if I intend the object as a coffee mug, then the hylé for the front side of the coffee mug are interpreted as fitting into the conceptual frame for coffee mug. The frame for a coffee mug says what a typical mug is like: it has a front side and a back side, each of these has a color, a shape, a size, a texture, and so on. It may further say things about holding coffee and being made of ceramic and having a handle – in general it lists all the features of the prototypical mug. Think of the frame as a list of feature-slots that any given coffee mug is assumed to fill in some determinate way. The hylé, then, fill in some of these slots. They fill in, for instance, the slots for the color, shape, size, and texture of the front side of the mug. These features of the object, when they are presented in good light at the right distance (and so on) are “determinate” in my experience of them.
But not every feature of an object is clearly and determinately presented to me in every experience. Accordingly, there are some feature slots that are incompletely filled or, as Husserl says, “indeterminate”.
If I see a house in sunlight, when the air is clear, then the color of the side turned toward me appears in its determinateness. If I see the house in the dark or in fog, then its color appears more or less indeterminately.
At the extreme there are features of the house that I know it has, but for which I have no sensuous presentation at all. For instance, the slots for the color, shape, size, and texture of the back side of the house are completely unfilled. Since I see the thing to be a house, I see it as having a back side that has a determinate color, shape, size, and texture. But how these features are manifested in this particular house is indeterminate in my current experience of it. For this reason, Husserl insists that,
Indeterminateness is never absolute or complete. Complete indeterminateness is nonsense; the indeterminateness is always delimited in this or that way. I may not know exactly what sort of form the back side has, yet it precisely has some form; the body is a body. I may not know how matters stand with the color, the roughness or smothness, the warmth or coldness, yet it pertains to the very sense of the apprehension of a thing that the thing possess a certain color, a certain surface determination, etc.
For Husserl, then, there are, at the extreme, two very different kinds of features that make up my experience of an object: there are the features that are determinately presented by the hylé – like the features of the front side of the mug - and there are the features that I take the object to have in virtue of having interpreted it as some particular object, but which are not themselves presented to me at all – like the features of the back side of the mug. Husserl calls these features respectively the proper and improper features of the perceived object, or sometimes the full and empty intentions of it. It is important to emphasize that the improper features of the object in my experience of it are in no way presented to me. As Husserl says in Thing and Space,
The clear result of these considerations is therefore that improperly appearing moments of the object are in no way presented. Perception is, as I also express it, a complex of full and empty intentions ... The full intentions ... are the properly presentational ones; the empty are precisely empty of any presentational material.
Although the improper features of the object are in no way presented to me, however, they are nevertheless an essential part of my experience of the thing as an object. To emphasize this Husserl sometimes says that the improper features are co-apprehended, as opposed to presented sensibly. By “co-apprehended,” here, he means something like “seen in virtue of the interpretation given, but not in virtue of any sensible presence”:
The improperly appearing objective determinations are co-apprehended, but they are not “sensibilized,” not presented through what is sensible, i.e., through the material of sensation. It is evident [however] that they are co-apprehended, for otherwise we would have no objects at all before our eyes, not even a side, since this can indeed be a side only through the object.
The improper features of the perceived object, therefore, are what account for the possibility that I can see the object as transcending my experience of it. In taking the hylé to represent an object of a particular sort, I see the object as having features that are not now determinately presented to me. These improper features of the experienced object are, as Husserl says, indeterminate in my experience of it.
But having said this, a question arises immediately. In light of the immanence of mental states, as we discussed it in the last section, we might ask how it is possible for there to be “indeterminate” features of experience. Recall that to say that perception is immanent is to say, at least in part, that my knowledge of the features of my perceptual state is incorrigible – there cannot be anything in the state that I do not have certain and complete knowledge of. But it seems, at least on the face of it, that there is a tension between an object’s presenting itself indeterminately to experience, and the experience itself being available to me completely and with certainty. For example, suppose my experience of the color of my coffee mug is indeterminate because a thick fog surrounds the mug. Naturally I may not be able to say, in this circumstance, what the color of the mug is. But it may be that in addition the experience itself is so foreign and unrecognizable that I cannot say what it is like either. Perhaps the problem is exacerbated when I try to think of how to characterize the experience I have of the even more indeterminate back side of the mug. At any rate, once we have insisted, as Husserl does, that I have an experience of the back side, it seems less clear that the features of my experience are all available to me completely and with certainty.
Whether it is possible for there to be “indeterminate” features of experience in this sense depends, not surprisingly, on what the meaning of “indeterminate” is. Husserl’s commitment to the incorrigibility of my knowledge of my mental states forces him to understand the indeterminate features of experience in a very special way. To say that the improper features of the perceived object are indeterminate, for Husserl, is to say that they are, to coin a phrase, hypothesized but sensibly absent in my current experience of the object. They are hypothesized in the following sense. It is in virtue of my having interpreted the hylé as of a certain kind of object – as having hypothesized, for instance, that the object is a mug – that my experience represents its object as the kind that has a handle on it. Assuming I don’t see the handle on the front side of the mug, the experience will represent its object as a mug that has a handle on the back. But this improper feature of the experienced mug is sensibly absent. That means that the feature slot for the handle in my concept of the object is unfilled. In short, I experience the mug determinately as having some handle, but the experience of that feature is indeterminate in the sense that it prefers no one handle over any other. As Husserl puts the point:
[I]n the case of an appearing physical thing-Object, it would again fall in the bounds of the description to say: a “front side” is thus and so determined with respect to color, shape, etc., its “rear side” has “a color” but a “not further determined’ one; the appearing physical thing-Object is, in these and those respects, altogether “undetermined” as to whether it is thus or so.
This is the trick, then. In order to intend the object as going beyond the hylé I’m now presented with, the noema interprets these sense data as being directed toward some object that I know has specific kinds of further properties that have not been explicitly determined yet in my experience of it. The indeterminate features in my experience of the object, according to Husserl, are not indeterminate in any substantive metaphysical sense; they are not the kind of thing that in their nature resist a complete and determinate characterization. Rather, the indeterminate features in my experience of the object are just those that I take the object to have, but that I have not yet had any determinate sensuous presentation of. These kinds of features pose no problem at all for the incorrigibility thesis, since they are just the kinds of features that I could attribute completely and with certainty to my experience.
The phenomenological question, taken up by the critics of Husserl, is whether this characterization of our experience is accurate; whether there is phenomenological evidence, in other words, for the claim that the indeterminate features of my experience of an object are not presented to me in any way at all. Merleau-Ponty, for one, claims that the phenomenological evidence points rather in the direction of a positive presentation of the indeterminate. And it is a central project of his Phenomenology of Perception, as he says right at the beginning of that book, to recognize the indeterminate as a positive phenomenon. The role of the body in presenting objects to us, and in particular the role of the strange category that Merleau-Ponty calls “motor or bodily intentionality,” is essential, as we shall see, to the completion of this project.
Merleau-Ponty gets from Husserl both the idea that we perceive objects as transcending what we determinately see about them, and also the idea that the project of phenomenology is to describe the details of this experience. He moves beyond Husserl, however, in his characterization of the features we experience as indeterminate. For Merleau-Ponty, the essentially bodily motor intentional relation to an object gives our experience some of its essentially indeterminate features. This focus on the role of the body in perception makes Merleau-Ponty’s account of object transcendence importantly different from the one proposed by Husserl. It also opens up the anti-Husserlian possibility, which Merleau-Ponty endorses, that perceptual states present the world to us in a way that transcends our capacity to reflect upon them.
3.1 Making the indeterminate a positive phenomenon
We have seen that on Husserl’s account the indeterminate features of a perceived object are hypothesized by the perceiver but sensibly absent in his experience of them. According to Merleau-Ponty, however, “we must recognize the indeterminate as a positive phenomenon.” The indeterminate features of the object are not merely features that I have no current sensuous experience of at all. As he says, “... the perceived contains gaps which are not mere ‘failures to perceive’.” Rather, the indeterminate features are those that I am now experiencing, though not as determinate features of the object:
The project, for Merleau-Ponty, is to say what this positive but indeterminate experience is.
In many cases the indeterminate features of our experience are present to us in our bodily engagement with the thing toward which we’re directed. The phenomenon of size constancy in perception provides a helpful example of this. Size constancy is the phenomenon according to which we experience a given object to be a constant size throughout a wide variety of perceptual contexts. For instance, as I move closer to and further away from an object, it looks to me to be a constant size throughout. Merleau-Ponty’s phenomenological analysis of size constancy invokes an essentially bodily relation to the object seen.
Many empiricist philosophers and psychologists have found it difficult to explain the phenomenon of size constancy. They tend to think about the problem somewhat as follows. When I move in relation to an object, the size of the image that it projects onto my retina varies accordingly. As I move closer, the object projects a larger image onto my retina; as I move further away the size of its image decreases. This change in the retinal stimulus is what makes the phenomenon of size constancy so puzzling. For it is natural to think, according to these empricists, that there must be a constant correlation between the properties of the retinal stimulus and the features of the perceptual experience that the subject has of the object. This assumption is sometimes called the “constancy hypothesis;” it is a modern descendant of assumptions that drive the older empiricist image theory of perception that Husserl criticized. If the constancy hypothesis is correct, then the change in the retinal stimulus that occurs as I move in relation to the object should be accompanied by a correlative change in my perceptual experience of the object. As I move away from the object, and the size of its retinal image decreases, I should experience the object to be getting smaller; and conversely as I move toward it. The phenomenon of size constancy suggests, however, that no such correlative change occurs. For this reason the phenomenon is hard to explain.
The empiricists, nevertheless, have a characteristic account of the mechanism of the phenomenon of size constancy. The details of their account, and of Merleau-Ponty’s trenchant assessment of it, provide an interesting example of phenomenological criticism; but I will not go into them here. Instead, I note only the lesson that Merleau-Ponty draws from this case. According to him, we must conclude against the constancy hypothesis that “the sensible [experience] cannot be defined as the immediate effect of an external stimulus.” We must give up on the ’constancy hypothesis,’ according to Merleau-Ponty, for the most basic of phenomenological reasons: it “conflicts with the data of consciousness.”
Still, the question remains how to account for the phenomenon of size constancy. One natural thing to say is that the subject, in his experience of the size of the object, is somehow “taking into account” the distance from which the object is being perceived. If that’s right, then every experience of the size of an object involves in some way an experience of the distance to it as well. But what kind of experience of the distance to the object does the subject have? One option is that the subject experiences the distance as a determinate amount – 20 feet, for instance. If this is right, then the theorist can attribute to the subject (or perhaps to the subject’s brain) a simple geometric algorithm by means of which he can calculate the constant size of the object given the size of the retinal image it casts and the determinate distance to it. This kind of cognitivist view, which reduces perceptual experience to rational algorithmic performance, has become the orthodoxy in perceptual psychology, and has been championed in particular by the late Irvin Rock. It is the approach that Husserl prefers as well. 
But Merleau-Ponty argues against this kind of cognitivism too. It is right, he believes, that every experience of an object involves in some way an experience of the distance to it (as well as the experience of many other contextual features). But we do not experience the distance to an object as a determinate value. Anyone who has ever rented an apartment understands this already. It is one thing to know that the living room is 18 feet long; it is quite another to stand in it and see its size. Merleau-Ponty says, therefore, that it is wrong to think, of distance and other contextual features, “that they can be treated as variables or measurable sizes, and therefore that they are already determinate.” Rather, we experience the distance to an object in an essentially indeterminate way.
The indeterminate experience of distance to an object, according to Merleau-Ponty, is present to us in our bodily engagement with the thing:
If I draw the object closer to me or turn it round in my fingers in order ‘to see it better’, this is because each attitude of my body is for me, immediately, the power of achieving a certain spectacle, and because each spectacle is what it is for me in a certain kinaesthetic situation.
Further, this bodily engagement with the thing manifests an essentially normative relation to it:
For each object, as for each picture in an art gallery, there is an optimum distance from which it requires to be seen, a direction viewed from which it vouchsafes most of itself: at a shorter or greater distance we have merely a perception blurred through excess or deficiency. We therefore tend towards the maximum of visibility, and seek a better focus as with a microscope. ... The distance from me to the object is not a size which increases or decreases, but a tension which fluctuates round a norm.
This normative feature of the experience of distance, which is manifest in our bodily engagement with things, is precisely what cannot be captured by the cognitivist account. For the cognitivist, 18 feet is a fixed, determinate value; it is the same no matter what the context. But for genuine perception, according to Merleau-Ponty, 18 feet may be perfect for viewing one thing, awful for another. This perfection and awfulness, this sense of the appropriateness of the distance to the thing seen, is an essential part of the way I experience distance to an object. And it is presented to me in what can only be called an immediate bodily way. To spell this out more clearly, we must look at the relation between our body and the experience we have of space. Merleau-Ponty’s work on this topic is groundbreaking. As we will see, it has received support, too, from recent research in cognitive neuroscience.
3.2 Body and space
Merleau-Ponty often procedes, in the Phenomenology of Perception, by considering cases of visual pathology. In these pathological cases, he believes, the subject has explicitly available to him features of experience that are hidden from normal perceivers in everyday life. By studying these pathological cases therefore, we can more easily make explicit to ourselves those features of experience that are normally hidden from us. To this end, Merleau-Ponty describes a patient named Schneider, whose visual pathology stems from a traumatic injury to the brain incurred during trench warfare in the First World War. Schneider’s case of morbid motility, according to Merleau-Ponty, “clearly shows the fundamental relations between the body and space.” The following somewhat lengthy passage occurs near the beginning of Merleau-Ponty’s discussion of Schneider:
In the ... patient ... one notices a dissociation of the act of pointing from reactions of taking or grasping: the same subject who is unable to point to order to a part of his body, quickly moves his hand to the point where a mosquito is stinging him. ... [A]sked to point to some part of his body, his nose for example, [he] can only manage to do so if he is allowed to take hold of it. If the patient is set the task of interrupting the movement before its completion ... the action becomes impossible. It must therefore be concluded that ‘grasping’... is different from ‘pointing.’ From the outset the grasping movement is magically at its completion; it can begin only by anticipating its end, since to disallow taking hold is sufficient to inhibit the action. And it has to be admitted that [even in the case of a normal subject] a point on my body can be present to me as one to be taken hold of without being given in this anticipated grasp as a point to be indicated. But how is this possible? If I know where my nose is when it is a question of holding it, how can I not know where it is when it is a matter of pointing to it?
“It is probably because,” Merleau-Ponty concludes, “knowledge of where something is can be understood in a number of ways.”
The general point of Merleau-Ponty’s discussion is that the understanding of space that informs my skillful, unreflective bodily activity – activity such as unreflectively grasping the coffee mug in order to drink from it, skillfully typing at the keyboard, or automatically walking closer to an object to see it better - is not the same as, nor can it be explained in terms of, the understanding of space that informs my reflective, cognitive or intellectual acts – acts such as pointing at the coffee mug in order to identify it. As Merleau-Ponty says, in skillful, unreflective bodily activity
my body appears to me as an attitude directed towards a certain existing or possible task. And indeed its spatiality is not ... a spatiality of position, but a spatiality of situation.
To give a name to intentional activities that essentially involve our bodily, situational understanding of space and spatial features, Merleau-Ponty coins the phrase “motor intentionality.” Grasping is the canonical motor-intentional activity.
As recently as 1992, perceptual psychologists were loathe to distinguish between the kind of spatial information available to the visual system for visuo-motor activities such as grasping and the kind available for perceptual judgments about location implicit in acts of pointing. In a forward thinking paper of the day one psychologist writes:
We often do not differentiate between grasping and pointing when we generalize about how vision is used when generating limb movements. It is possible, that how individuals use vision may vary as a function of whether they are generating pointing or grasping movements, and that some principles of how vision is used during reaching and pointing is (sic) not generalizable to grasping.
This was a maverick view in 1992. Since that time, however, the important work of neuroscientists A. David Milner and Melvyn Goodale has opened the way for acceptance of this basic Merleau-Pontean distinction – the distinction between essentially bodily understandings of space and spatial features, on the one hand, and essentially cognitive or reflective understandings of these on the other. Much of Milner and Goodale’s work comes from an analysis of D.F., a patient who suffered carbon monoxide poisoning that resulted in a visual pathology strikingly similar to Schneider’s. Milner and Goodale describe her situation as follows:
D.F.’s ability to recognize or discriminate between even simple geometric forms is grossly impaired. ... [Her] pattern of visual deficits [however] ... is largely restricted to deficits in form perception. D.F. ... recovered, within weeks, the ability to reach out and grasp everyday objects with remarkable accuracy. We have discovered recently that she is very good at catching a ball or even a short wooden stick thrown towards her. ... She negotiates obstacles in her path with ease .... These various skills suggest that although D.F. is poor at perceptual report of object qualities such as size and orientation, she is much better at using those same qualities to guide her actions.
In particular, Milner and Goodale report, D.F. is capable of responding differentially to spatial features of an object like its size, shape, and orientation even in cases in which she is incapable of visually identifying those very features. One test of this involved the identification of the orientation of a slot. Quoting again from Milner and Goodale:
[We] used a vertically mounted disc in which a [rectangular] slot ... was cut: on different test trials, the slot was randomly set at 0, 45, 90, or 135º. We found that D.F.’s attempts to make a perceptual report of the orientation of the slot showed little relationship to its actual orientation, and this was true whether her reports were made verbally or by manually setting a comparison slot. [Further examination revealed a large variety of other reporting methods for which her performance was equally bad.] Remarkably, however, when she was asked to insert her hand or a hand-held card into the slot from a starting position an arm’s length away, she showed no particular difficulty, moving her hand (or the card) towards the slot in the correct orientation and inserting it quite accurately. Video recordings showed that her hand began to rotate in the appropriate direction as soon as it left the start position. [One is reminded here, by the way, of Merleau-Ponty’s claim that “from the outset the grasping action is magically at its completion.”] In short, although she could not report the orientation of the slot, she could ‘post’ her hand or a card into it without difficulty.
Milner and Goodale go on to suggest a neurophysiological basis for the dissociation between pointing and grasping. They claim that there are two different streams of visual information flow in the brain, one of which is geared to perceptual judgment, the other of which is geared directly to action. D.F.’s case is one of the principle pieces of evidence that there is not one common understanding of orientation on the basis of which both judgment and action occur, but rather two different ways of understanding spatial qualities like orientation. Indeed, D.F.’s understanding of the orientation of the slot, unlike the more familiar cognitive understanding, is essentially in terms of her bodily capacities and dispositions to act with respect to it. In the terminology of Merleau-Ponty, she has a motor intentional understanding of orientation.
3.3 Motor intentionality and the positive indeterminate
I said that motor intentionality provides us with an essentially bodily relation to the object. Let me try to spell this out a bit more clearly. After I do so, I will show that this essentially bodily relation to the object is just what we need to make sense of Merleau-Ponty’s notion of a “positive indeterminate” understanding of the back side of our coffee mug.
It is clear that motor intentional activities – like unreflectively grasping the coffee mug to drink from it – succeed at least partly in virtue of facts about the object toward which they are directed. For instance, I will change my way of grasping the mug if someone moves it; I’ll reach over there instead of over here. But I’ll also conform my way of grasping an object to the kind of object it is. For instance, even if it’s in the very same spot, I’ll grasp the mug differently when it’s full than when it’s empty, or when the handle is broken than when it’s not. The differences in my bodily relation to the object are pervasive. My grip forms itself differently, my hand opening scales itself differently, my entire body may even prepare itself differently if the object is perceived to be, for instance, very heavy instead of very light. The upshot is that in identifying an object motor intentionally my body typically prepares itself to deal with the entire object, not just with some independently specifiable set of spatial features of it.
When I say that my body prepares itself to deal with the entire object, I mean also that my body prepares itself to deal with the actual existing object, not with some representation of it. Indeed, the perceived existence of the object is so important to the grasping act that without it the action is measurably distinct. This is clear from another interesting empirical result, this one reported by Goodale, Jakobson, and Keillor. These authors have shown that there are measurable qualitative differences between natural grasping movements directed toward an actual object and "pantomimed" movements directed toward a remembered object. When an actual object is present to be grasped, the subjects typically scale their hand opening for object size and form their grip to correspond to the shape of the object. In pantomimed actions, on the other hand, when there is no object present, although the subjects continue to scale their hand opening, their grip formation differs significantly from that seen in normal target directed actions. It seems that the actual perceived presence of a thing, and not just some independent representation of it (like a memory), is necessary for the motor intentional activity directed toward it. This is why Merleau-Ponty insists that motor intentional activity is directed toward the object itself in all its particularity. As he says,
In the action of the hand which is raised towards an object is contained a reference to the object, not as an object represented, but as that highly specific thing towards which we project ourselves, near which we are, in anticipation, and which we haunt.
This is not merely the kind of direct realism that is sometimes found in the philosophical literature nowadays; it is not just the rejection of representational intermediaries. It is in addition an embrace of the positive notion of a whole bodily understanding of the object.
The understanding of the entire object that I have when I am grasping it is not an understanding I can have independent of my bodily activity with respect to it. My bodily activity with respect to the object just is my way of understanding it. We saw this already in the case of D.F. – the understanding of the orientation of the slot that she has in posting a card through it is not an understanding she can have independent of the posting activity. In particular hers is not the kind of understanding of orientation that she can report in any way other than by actually posting the card through the oriented slot. But this kind of bodily understanding of the world is familiar to normal subjects as well. Merleau-Ponty gives the example of a typist’s bodily understanding of the keyboard:
To know how to type is not, then, to know the place of each letter among the keys, nor even to have acquired a conditioned reflex for each one, which is set in motion by the letter as it comes before our eye. If [bodily skill] is neither a form of knowledge nor an involuntary action, what then is it? It is knowledge in the hands, which is forthcoming only when bodily effort is made, and cannot be formulated in detachment from that effort.
That there is a peculiarly bodily type of understanding of objects is the central point of Merleau-Ponty’s category of motor intentionality: motor intentional activity is a way of being directed toward objects that essentially involves a motor or behavioral component. As Merleau-Ponty says in introducing the phrase:
... we are brought to the recognition of something between [reflex] movement as a third person process and thought as a representation of movement - something which is an anticipation of, or arrival at, the objective and is ensured by the body itself as a motor power, a 'motor project' (Bewegungsentwurf), a 'motor intentionality' ...
In motor intentional activity, in other words, there is not an independent way we have of understanding the object on the basis of which we act differentially with respect to it. Rather our bodily activity is itself a kind of understanding of the object. I believe that this kind of essentially bodily engagement with the world is substantially different from every other kind of intentional state.
The main difference between motor intentionality and cognitive intentionality lies in their logical structure. Every cognitive intentional state – states like believing, hoping, desiring, and so on – has two separable parts: the intentional content of the state and its propositional attitude. For instance, when I have a belief that the sun is rising, the content of the belief is that the sun is rising, and the attitude I have towards that content is belief. I might have hoped that the sun is rising instead of having believed it; likewise, I might have believed that the sun is not rising instead of that it is. The content and the attitude of (cognitive) intentional states are logically separable from one another.
The logical structure of motor intentional activities is different. For motor intentional activities, there is no independently specifiable content toward which the subject can have an attitude. This is because motor intentional activity identifies its object in such a highly specific and context-sensitive way that any attempt to take up that specification of the object as such changes it into something other than it was at the time it was had. An example should make this clear.
When she is posting the block through a slot oriented at 45º, D.F. is motor intentionally engaged with the orientation of the slot. But what is the content – the representation of the way the slot is – that is manifest in this motor intentional activity? To specify this content, we typically use concepts – roughly words – that D.F. already possesses and can apply in a variety of contexts. The problem is that there seems to be no concept that D.F. possesses in virtue of which she is capable of performing the posting activity. If there were such a concept, it would have to apply to objects in the world that are oriented at 45º, and only to those objects. But D.F. doesn’t seem to be able to apply any such concept to the oriented slot. Remember, she can’t say of it that it is oriented at 45º, but she also can’t draw the slope of the slot on a piece of paper or even rotate her hand into the correct orientation without at the same time moving it toward the slot. She seems, in other words, not to be able to represent the orientation of the slot at all except by means of posting the card through it.
But still, why can’t we think of this activity itself as a way of understanding the orientation of the slot toward which she can have the attitude of belief? Why can’t she say, in other words, “I believe the slot is oriented this way [said while posting the card through the slot]”? Well, she can say such a thing of course – she can utter the words – but the question is whether in doing so she is invoking the representation of the orientation of the slot that constituted the understanding she had of it when she was posting the card through the slot. We can easily see she is not.
The reason is that when she tries to use the posting action to refer to the orientation it identifies, the thought she has seems to be not about the orientation of the slot but rather about whatever orientation her hand happens to be in. If you change the orientation of the slot after she stops moving her hand, for instance, and you don’t let her begin the posting activity again, she will continue to say that the orientation of the slot is whatever orientation her hand ended up in. What is revealed in the posting activity, however, is the actual orientation of the slot – it’s that orientation itself that the activity is sensitive to. So even if she can have an attitude toward the activity that manifests an understanding of the orientation, this is not the same as having an attitude toward the understanding of the orientation that the activity manifests.
The understanding of the orientation of the slot that D.F. manifests in her motor intentional activity, therefore, is of a peculiar sort. It contains no independently specifiable content toward which she can have an attitude. Instead of “representing” the orientation of the slot, therefore, we might say that her motor intentional activity “discloses” the orientation to her directly, and cannot be captured in the process of doing so. This coheres with Schneider’s report of his own experience, for he says,
I experience the movements as being a result of the situation, of the sequence of events themselves; myself and my movements are, so to speak, merely a link in the whole process and I am scarcely aware of any voluntary initiative. … It all happens independently of me.
Because motor-intentional activity is called forth by the situation in this way, and is therefore to some degree independent of the autonomous will of the subject, it does not have at its heart the kind of autonomous representational content that a subject could have an attitude toward.
This account of motor intentionality gives us the tools we need to understand Merleau-Ponty’s notion of the “positive indeterminate” in experience. Recall that Merleau-Ponty’s goal was to explain how our experience of the back side of the mug does not merely hypothesize the existence of a handle, but is somehow positively, though indeterminately, aware of the actual thing. Now we can see that the bodily relation to the mug gives us a positive awareness of it as a whole, including an awareness of the “hidden” features like its back side. This positive awareness of the mug is manifest in our bodily set, by means of which we are prepared to deal with it as a highly specific thing. The preparation to deal with the handle on the backside manifests itself in various features of my motor intentional activity: my grip forms itself in such a way as to take account of the shape of the handle, for instance, and my hand opening scales itself to account for its size. My body may get these features of the handle wrong, of course, and if it does my motor intentional activity will reveal itself to have understood the object as something other than what it is. This has the real repercussion that I’ll probably knock the mug over or drop it. But when things are going smoothly, the whole mug – back side and all – will be positively revealed to me in my motor intentional activity towards it.
Motor intentional engagement with an object is indeterminate in a clear sense as well. For the way of understanding an object that is manifest in our motor intentional activity is not specifiable as a definite representational content. There are two reasons for this, as we have seen. First, motor intentional activity depends on the perceived existence of the actual object, not just on some representation of it; motor intentional activity, in other words, discloses the world. Second, the motor intentional understanding of the object is not specifiable independent of the motor intentional activity itself; it is an essentially bodily engagement with the world. These two features of motor intentional activity give it its essential indeterminateness. Therefore, any attempt to specify the motor intentional understanding of the object as a determinate representational content is self-defeating: the very attempt to characterize the content determinately turns it into something other than the essentially indeterminate thing it is.
Finally, when we understand the positive indeterminate aspects of experience in the way that Merleau-Ponty does, we are forced to deny Husserl’s claim that intentional states are immanent. Recall that incorrigibility is one of the four features of immanence for Husserl. To say that the subject’s knowledge of his intentional state is incorrigible is to say that the qualities he takes the state to have, in reflecting upon it, are certain to characterize it as it really is. But if Merleau-Ponty is right, then our understanding of the essential features of motor intentional activity is very far from being incorrigible. Indeed, to say that motor intentional activity is indeterminate, in Merleau-Ponty’s sense, is to say that the disclosive understanding of the world that it manifests is not specifiable as such. If the subject cannot even specify the content of her motor intentional activity, indeed cannot have any attitude toward it at all, then her knowledge of the features of that intentional state is certainly not incorrigible. Rather, motor intentional activity gives us a relation to, or a bodily understanding of, the world that goes beyond the subject’s capacity to characterize it. Motor intentionality is transcendent in precisely this sense.
Husserl’s development of phenomenology made a genuine advance over his predecessors. By taking perception as the paradigmatic intentional state, Husserl was able to emphasize both the perspectival aspect of intentionality (against Brentano) and the intentional aspect of perception (against the empiricists). In doing so, he happened upon perhaps the most basic problem of phenomenology: to characterize in a descriptively accurate manner how perception represents its object as transcending what is presented in the perception of it. He tried to guarantee the possibility of descriptive accuracy, however, by claiming that the essential features of experience are available to us incorrigibly; that experience, like all intentional states, is immanent. He hoped that phenomenology would be the study of the pure realm of immanent content that is produced by the transcendental reduction. But this commitment forced him to give a peculiar answer to the phenomenological question. For according to Husserl, the hidden aspects of an object – those that transcend our experience of them – are hypothesized but sensuously absent.
Merleau-Ponty took up Husserl’s phenomenological problem, but argued that we must recognize the indeterminate in experience as a positive phenomenon. Merleau-Ponty’s emphasis on our bodily engagement with the world gave him the tools to make this claim precise. Our motor intentional relation to the hidden aspects of an object – like the handle on the back side of a coffee mug – is positive but indeterminate. It is positive in the sense that it gives me a bodily preparation for a very particular handle; I’ll be very surprised by, and will probably deal ineptly with, the mug if the handle turns out to be something else. This preparation for a very particular handle is very unlike what we find in Husserl’s account, according to which our experience prefers no handle over any other.
Motor intentional activity, according to Merleau-Ponty, is indeterminate as well. That’s because our bodily engagement with the mug isn’t specifiable as a determinate representation toward which the subject could have an attitude. Any attempt to specify the content of motor intentional activity in this way is self-defeating, since the very process of doing so turns the motor intentional relation to the object into something else. This is also unlike Husserl, since it defines a kind of intentionality that transcends my capacity to reflect upon it. Merleau-Ponty, therefore, rejects Husserl’s claim that perception is immanent, and he rejects it on the very grounds that descriptive accuracy requires us to do so. This is closely tied to his re-evaluation of Husserl’s phenomenological method, and in particular of the transcendental reduction to a pure realm of immanent content. As Merleau-Ponty says in the preface to the Phenomenology of Perception, “the most important lesson that the transcendental reduction teaches us is the impossibility of a complete reduction.”
 I would like to thank Bob Solomon and Bert Dreyfus for helpful comments and discussion.
 The drafts for the Encyclopaedia Brittanica article, and the correspondence between Husserl and Heidegger about it, appear in Psychological and transcendental phenomenology and the confrontation with Heidegger: the Encyclopaedia Britannica article, the Amsterdam lectures "Phenomenology and anthropology," and Husserl's marginal notes in Being and time, and Kant and the problem of metaphysics, tr. by Thomas Sheehan and Richard E. Palmer, (Boston : Kluwer Academic Publishers, 1997).
 See, for instance, the passages on the phenomenological reduction in the preface to Merleau-Ponty’s Phenomenology of Perception. It is not surprising that a left-wing French academic writing in 1945 should choose to align himself with Husserl (a persecuted Jew) instead of Heidegger (a Nazi). But attempt depends upon an interpretation of the phenomenological reduction that bears little resemblance to Husserl’s understanding of that technique.
 I think this example occurs in the Principles of Psychology, but I haven’t been able to track it down.
 He did his doctoral work on the calculus of variations, working for a while under the famous German mathematician Karl Weierstrauss.
 His Philosophie der Arithmetik was published in 1891.
 For more on their relation see the letters between Husserl and Frege published in Gabriel, Gottfried et. al., Gottlob Frege: Philosophical and Mathematical Correspondence, tr. by Hans Kaal, (Chicago: University of Chicago Press, 1980).
 Edmund Husserl, Logical Investigations, tr. J.N. Findlay, (London: Routledge and Kegan Paul, 1970) p. 269. This is §1 of of Investigation I. Hereafter I will abbreviate references to this text as LI, followed by the Investigation number and the page in the edition listed here.
 LI I, p. 248.
 Heidegger writes that in bracketing the entity in order to focus on our experience of it, “what really is at issue now is the determination of the being of the very entity.” See History of the Concept of Time, tr. Theodore Kisiel, (Bloomington: Indiana University Press, 1992), p. 99. Hereafter I will refer to this text as HCT.
 Perhaps this is not fair, since Kant’s system is obviously concerned to think about not just objects, but objects as they are understood through the pure concepts of understanding. But Husserl’s project was not the same as Kant’s. In particular, Kant’s transcendental method is completely anathema to the descriptive procedures of phenomenology.
 More accurately, our experiences present themselves as being present all at once in their entirety. But I will be sloppy about this point in the text.
 See HCT, §11.
 as present
 The embodied nature of all types of perceptual experience, however, was vital to Merleau-Ponty’s phenomenological work.
Brentano, following the Scholastics, developed his account of intentionality primarily for the cases of belief and judgment.
 See Locke’s Essay Concerning Human Understanding, II.1.2: “Let us then suppose the Mind to be, as we say, white Paper, void of all Characters, without any Ideas; How comes it to be furnished? Whence comes it by that vast store, which the busy and boundless Fancy of Man has painted on it, with an almost endless variety? Whence has it all the materials of Reason and Knowledge? To this I answer, in one word, From Experience.” Similarly, we read in Hume’s Enquiry Concerning Human Understanding, §II: “Or, to express myself in philosophical language, all our ideas or more feeble perceptions are copies of our impressions or more lively ones.”
 David Bell’s book Husserl (London: Routledge, 1990) is typical in this respect. By emphasizing the idea that “Husserl’s theory of intentionality is entirely general and, to a large extent, purely formal,” (p. 115) Bell fails to do justice to the way perception motivates Husserl’s broader understanding of intentionality. Aron Gurwitsch, in The Field of Consciousness (Pittsburgh, PA: Duquesne University Press, 1964), places perception at the center of his discussion of Husserl’s phenomenology. But Dreyfus 1982 argues convincingly that Gurwitsch’s interpretation of the perceptual noema as a percept, rather than a concept, has little support in Husserl’s texts. Dreyfus, therefore, follows Føllesdal 1982, who shows definitively that Husserl’s noema is a generalization of the Fregean Sinn to all mental states. Føllesdal’s important and influential work, however, has given rise to a school of interpretation that places the emphasis on the logical, rather than the properly phenomenological, aspects of Husserl’s work; it focuses attention, in other words, on the linguistic rather than the pre-linguistic cases of intentionality. Although this emphasis is not apparent in Føllesdal’s own work, selective attention to the Fregean features of his Husserl has given rise not only to this linguistic school of interpretation, but also to such misguided criticisms of Husserl as those found in Dummett 1981.
There is some illuminating discussion of Husserl’s account of perception in Smith and Mcintyre’s book Husserl and Intentionality (Dordrecht, Holland: D. Reidel Publishing Company, 1982), as well as in Kevin Mulligan’s article “Perception,” in The Cambridge Companion to Husserl, ed. Barry Smith and David Woodruff Smith (Cambridge: Cambridge University Press, 1995). In neither case, however, is perception presented as the paradigmatic intentional state. By contrast, I will argue that the perceptual case is central for Husserl, since it is the paradigm of an intentional relation to a transcendent object. That Husserl described this intentional relation in terms that were too Cartesian and cognitivist is what generates the phenomenological responses to Husserl by Heidegger and Merleau-Ponty. That he had the radical idea to characterize intentionality on the model of the perceptual, rather than the linguistic, relation to a transcendent object, however, is what made his studies in phenomenology relevant to these thinkers in the first place. It is also what distinguishes the phenomenological approach from the traditional analytic approach to intentionality that is grounded in the seminal work of Frege.
 Edmund Husserl, Ideas Pertaining to a Pure Phenomenology and to a Phenomenological Philosophy, First Book, tr. F. Kersten, (Dordrecht: Kluwer Academic Publishers, 1982), pp. 82-3/70. Hereafter abbreviated in the text as Ideas I.
 Ideas I, p. 82/70.
 Ideas I, p. 83/71. We might disagree with Husserl about either of these claims. For instance, we might attempt to show that perception of a physical thing is not representative of perception generally by arguing that the perception of an event, like the presentation by Lincoln of his Second Inaugural Address, is not relevantly akin to the perception of an object, like a house. Similarly, we might attempt to show that experience is not the ultimate source of intentional life by arguing that zombies, who by definition have no experiential life at all, are nevertheless quite capable of intentionality. Whether these arguments have any merit is not a question that I will pursue here. Husserl himself seems not to have wondered whether these empiricist claims about experience are justified. By this I mean that he accepted without question the empiricist emphasis on perception as the ground of thought, though he did not, of course, accept the empiricist account of perception. (See the next section.) But that these kinds of experiences are the primary model for his phenomenological account of intentionality is crucial.
 See, e.g., Edmund Husserl, Experience and Judgment: Investigations in a genealogy of logic, tr. James Churchill and Karl Ameriks, (Evanston, IL: Northwestern University Press, 1973), pp. 50-51.
 Logical Investigations, Investigation I, §23, p. 309. By a “consciousness prior to all experience,” Husserl seems to mean a being who consciously senses colors, sounds, shapes, textures, and so on, but does not experience them as the red of an apple, the cry of a baby, the roundness of a billiard ball, the rough of a carpet. Roughly this is a being who has sense-data but no interpretation of them. See the discussion in footnote 30 of Philonous’ passage in Berkeley’s Three Dialogues of Hylas and Philonous.
 Locke is the empiricist to whom this view is most often attributed; it is sometimes labelled the “picture-original theory of perception” in the secondary literature on him. See, for instance, Mackie, J.L., “Locke and representative perception,” in Vera Chappell (ed.) Locke (Oxford: Oxford University Press, 1998), pp. 60-8. Whether Locke himself, or indeed any of the British Empiricists, actually held such a view is not particularly important to my point. The common attribution of the view to them is sufficient to account for its influence on Husserl. Perhaps it is more fair to attribute the image theory to the early sense-datum theorists, but I know of no evidence that Husserl was familiar with their work.
 George Berkeley, Three Dialogues between Hylas and Philonous (USA: Hackett Publishing Company, 1979), p. 39. Berkeley, of course, sides with Philonous against the image theory on the grounds that something that is sensible (like an idea) couldn’t possibly be similar in any relevant sense to something that is insensible (like a material object). See, e.g., Philonous’ comment on p. 41 of the Three Dialogues: “In a word, can anything be like a sensation or idea, but another sensation or idea?” This argument against the image theory, of course, is different from the one against it that Husserl gives. Interestingly, however, Heidegger and Frege both use this kind of Berkeleian argument to inveigh against the correspondence theory of truth (see my Relelevance of Phenomenology to the Philosophy of Language and Mind, (New York: Garland Publishing Inc., 2001), chapter 1), though both attempt to avoid the radical idealism to which Berkeley himself was led by the move.
 LI, pp. 712-13.
 LI, pp. 593-4. The complete passage in the original reads: “Woran liegt es also, daß wir über das im Bewußtsein allein gegebene “Bild” hinauskommen und es als Bild auf ein gewisses bewußtseinsfremdes Objekt zu beziehen vermögen? ... [I]m phänomenologischen Wesen des Bewusstseins in sich selbst alle Beziehung auf seine Gegenständlichkeit beschlossen ist.” (pp. 436-7 in the Husserliana edition, vol. XIX/I).
 Husserl’s critique of the image theory of perception is presented in the Appendix to §§11 and 20 of the Fifth Logical Investigation (pp. 593-596). Also important to this critique is the section on perceptual adumbration in the Sixth Logical Investigation. This is §14, especially §14b, pp. 712-715. It can be difficult to divine Husserl’s intent here by reading only the Findlay translation of Logical Investigations, since the crucial German term, Abschattung, is translated in a variety of ways in this section. Among the translations are “shadowing forth,” “aspect,” and “projection.” Noticeably absent is the English word “adumbration,” which is the preferred translation of the term in the Kersten translation of Ideas I.
 The empiricist could attempt to add a story about the importance of associations and memories in distinguishing these experiences. (Thanks to Casey O’Callaghan for pushing this point.) Perhaps I see something to be a barn, the empiricist could argue, in virtue of the memories I have for dealing with barns and the associations I have with them; seeing it to be a barn façade would involve a separate set of associations and memories. It is in virtue of the difference in associations and memories, on such an account, that there is a difference in my experience of something as a barn and my experience of it as a barn façade. The problem with this account, Husserl would say, is that there is no reason for one set of associations to get triggered over the other. Since the image of the barn is exactly the same as the image of the barn façade, the associations and memories triggered by the two images must be exactly the same as well. This follows from the empiricist assumption that the only thing that I actually see is the image. I would have different associations and memories with different images, of course. But if two images are themselves exactly the same, then there’s no reason to suppose that they could by themselves trigger distinct associations and memories. Husserl gets around this by assuming that the “image” is always already taken under an interpretation in my experience of the thing.
 The empiricist view is again clearly stated in Berkeley’s Three Dialogues, this time by Philonous: “For instance, when I hear a coach drive along the streets, immediately I perceive only the sound; but from the experience I have had that such a sound is connected with a coach, I am said to hear the coach. It is nevertheless evident, that in truth and strictness, nothing can be heard but sound: and the coach is not then properly perceived by sense...” (p. 39). This view is very much like the early sense-datum theory of Bertrand Russell.
 The most important work of Brentano’s for Husserl is the first edition of Psychology from an Empirical Standpoint, tr. A. Rancurello et. al. (London: Routledge, 1973). The German original was published in 1874. Husserl studied under Brentano in Vienna from 1884-1886.
 Cf. Psychology from and Empirical Standpoint, pp. 88-94. Brentano’s committment to this doctrine may have changed throughout his career, as is argued by many commentators. If so, there was good reason for it, since the doctrine has some strange consequences. Cf., Dagfinn Føllesdal, “Brentano and Husserl,” in Dreyfus and Hall, Husserl, Intentionality, and Cogntive Science (Cambridge, MA: MIT Press, 1982), pp. 31-41. But both the doctrine and Brentano’s committment to it are clear in the 1874 edition of Psychology. Cf. Barry Smith, Austrian Philosophy: The Legacy of Franz Brentano (Chicago: Open Court), pp. 41-45.
 From the Supplementary Remarks prepared for the 1911 edition of Psychology from an Empirical Standpoint, p. 272. My italics.
 See especially Husserl’s treatment of Brentano in both the Logical Investigations and in Ideas I. The relevant passages are at LI V, §11, pp. 557-560, and §23, esp. p. 598, as well as in Ideas I, §85, esp. p. 206/174-5. For a related treatment of Husserl’s advance over Brentano, cf. Theodore de Boer, The Development of Husserl’s Thought, tr. Theodore Plantinga (The Hague: Martinus Nijhoff, 1978), pp. 133-141. De Boer emphasizes Husserl’s rejection of the Brentanian thesis that perceived content is immanent, but fails to appreciate the important relation between the transcendence of the perceived object and the necessarily perspectival presentation of it to perception.
 Cf. Føllesdal’s treatment of this issue in “Brentano and Husserl,” esp. p. 35. Føllesdal locates the distinction between Husserl and Brentano in Husserl’s introduction of a noema – that by means of which mental phenomena are directed toward their objects. This seems exactly right, albeit a bit underspecified. What I will add below is a story about what it means to say that the noema is that by means of which mental phenomena are directed toward their objects.
 Thing and Space: Lectures of 1907, tr. Richard Rojcewicz (Dordrecht: Kluwer Academic Publishers, 1997), p. 44/52.
 I cannot spell out this thought completely here, but the basic idea is this. Even if we stick with Brentano’s concern about mental states directed toward non-existent objects (like golden mountains), it is not at all clear that the intuition that motivates him in the linguistic case is preserved in the perceptual case. His intuition was that the thought about the golden mountain is completely indifferent to the question whether the mountain itself exists. Whether the mountain exists or not, the thought about it is exactly the same. Even if this is right for the linguistic case, it’s much less clear in the case of perception. Because perceptions seem to be so closely connected with their objects, it may be a conceptual truth that veridical perception – which actually lands on an object - is a different kind of thing than hallucination – which has no object at all. This “disjunctivist” view, that hallucination has nothing at all in common with veridical perception, has recently been defended by McDowell and others. Cf. John McDowell, Mind and World (Cambridge, MA: Harvard University Press, 1994). Husserl – and Merleau-Ponty as well – is trying to give an account of the related phenomenological fact that perception normally takes itself to be directed toward objects.
 See Ideas I, §§44-50. See also §11 of Heidegger’s History of the Concept of Time. Notice that what is indubitable is the existence of the intentional state, and what is incorrigible is the subject’s knowledge of the features of the intentional state. Naturally it is an essential feature of intentional states that they present objects as having certain features. But of course neither the existence of the object nor of the features it is represented to have is guaranteed.
 Ideas I, p. 100/85.
 Here, it seems, Husserl agrees with Descartes but not with the empiricists. About Locke, for instance, Michael Ayers writes, “Another important difference from Descartes lies in Locke’s conception of our awareness of the ‘operations of our minds,’ which he calles ‘reflection.’ Traditionally, in Aristotelian as well as Cartesian philosophy, the mind’s reflexive awareness of its own activity ... is a function of intellect, not sense. ... For Locke, in contrast, ‘reflection’ is simply a part of ‘experience’ ... An important implication is that thought is not, as Descartes had held, transparent to itself. Just as the senses give us only superficial, coarse knowledge of external objects, so ‘reflection’ keeps us aware of our thinking, but not of the ultimate nature of thought.” See Ayers, Michael, Locke, (New York: Routledge, 1999), p. 7.
 Ideas I, p. 96/81.
 For important historical discussions of the transcendental reduction see Jean-Paul Sartre, The Transcendence of the Ego: an existentialist theory of consciousness, tr. Forrest Williams and Robert Kirkpatrick, (New York: Hill and Wang, 1989) and Roman Ingarden, On the motives which led Husserl to Transcendental Idealism, tr. A. Hannibalsson, (The Hague: Hijhoff, 1975), among many others.
 Husserl’s noema has famously been interpreted as a generalized version of Frege’s Sinn – the public entity in virtue of which my linguistic utterances come to be about the world. See Dagfinn Føllesdal’s important and influential paper “Husserl’s notion of noema,” in Hubert L. Dreyfus and Harrison Hall (eds.) Husserl, Intentionality, and Cognitive Science, (Cambridge, MA: MIT Press, 1982) for the best characterization of this view. Føllesdal is certainly right that the noema is that in virtue of which mental states come to be about the world. But because the noema finds its paradigmatic application in the context of perception, the problems it has to solve – and the ways in which it goes about solving them – are somewhat distinct. Failure to appreciate this difference has led some appropriators of Føllesdal’s work to see in it an implicit condemnation of the Husserlian project. See especially Michael Dummett’s off-the-mark criticisms of Husserl in his Origins of Analytical Philosophy (London: Duckworth, 1993).
 See Experience and Judgment §21c for discussion of the frame (translated as “framework” in the Churchill and Ameriks translation). The following owes much to the development of this material by David Woodruff Smith and Ronald McIntyre in their book Husserl and Intentionality (Dordrecht: Reidel, 1982).
 Thing and Space, §18, p. 49/58.
 Thing and Space, §18, pp. 49-50/59.
 Thing and Space, §18, p. 48/57.
 Thing and Space, §17, p. 46/55.
 Ideas I, §130, p. 312/270. Cf. Thing and Space, §18, p. 50/59.
 They are not, for instance, metaphysically vague and in that sense uncharacterizable.
 Maurice Merleau-Ponty, Phenomenology of Perception, tr. Colin Smith (London: Routledge and Kegan Paul, 1962), p. 6.
 Phenomenology of Perception, p. 11.
 Notice that the Smith tranlation of this phrase, “a vision of something or other,” precisely covers up the difference between MP and Husserl. According to MP it is not a vision of some thing or another which is itself determinate but which I have not yet determined. It is, rather, a positive presentation of something indeterminate, of an I don’t know what. It’s a scope difference. Husserl thinks that there exists some determinate thing with which I have not yet been presented. MP think that there exists a presentation of something that is itself not a determinate entity (for me). Spell out more clearly.
 Phenomenology of Perception, p. 6.
 See, for instance, pp. 299-300 in Phenomenology of Perception.
 Phenomenology of Perception, p. 8.
 Phenomenology of Perception, p. 7.
 See, e.g., Irvin Rock, Indirect Perception (Cambridge, MA: MIT Press, 1997).
 See (Mulligan 1995), esp. §6.1 for some discussion of Husserl on the various phenomena of perceptual constancy.
 Phenomenology of Perception, p. 301.
 Phenomenology of Percpetion, p. 303.
 Phenomenology of Perception, p. 302.
 The following two sections draw substantially on Kelly, Sean D. 2002. The logic of motor intentional activity. Forthcoming in Ratio.
 Phenomenology of Perception, p. 103.
 Phenomenology of Perception, pp. 103-4.
 PP, p. 100.
Carnahan, Heather, "Eye, head and hand coordination during manual aiming", in Proteau, L. and Elliott, D. (eds.), Advances in Psychology 85: Vision and Motor Control., Elsevier Science Publishers B.V., 1992, p. 188.
 The Visual Brain in Action, pp. 126-128.
 VBA, p. 128.
Goodale, Jakobson, Keillor, "Differences in the visual control of pantomimed and natural grasping movements.", Neuropsychologia, Oct., 1994, v. 32 (n. 10), pp. 1159 - 1178.
Merleau-Ponty, Phenomenology of Perception, p. 138, italics in the original.
 Phenomenology of Perception, p. 144.
ibid., p. 110.
 For a more detailed account of the argument here, see Kelly, Sean D. 2002. The logic of motor intentional activity. Forthcoming in Ratio.
 See, for instance, Searle, John 1983. Intentionality (Cambridge: Cambridge University Press).
 A helpful analogy exists in Frege’s account of concepts. Frege says that concepts are unsaturated in the sense that they need to have an object added to them in order to be specifiable entities at all. Because of this, any attempt to refer to them as such turns them into something other than what they originally were.
 If we specify the content using concepts she doesn’t possess, then there’s little sense in saying we’ve characterize her understanding of the orientation. That she must be able to apply these concepts in a variety of contexts is what Gareth Evans calls the “generality constraint” on concept possession. See Gareth Evans, The Varieties of Reference, (Oxford: Oxford University Press, 1982), chapter 4.
 This is not to say that she doesn’t possess the perfectly good concept [oriented at 45º]. She may very well possess that concept. But that is not the concept manifest in her motor intentional activity, as we shall see.
 Personal communication with Melvyn Goodale.
 Phenomenology of Perception, p. 105.
 No doubt Husserl’s desire to develop phenomenology as a “rigorous science” encouraged him in this respect.
 Phenomenology of Perception, p. xiv.