Navigation – Plan du site
Dire et vouloir dire

Is Seeing Intentional? A Response to Travis

Walter Hopp


This is a response to Charles Travis's article "Is Seeing Intentional?" In it, I argue that while seeing differs from other intentional states in a variety of ways, seeing is indeed intentional, at least in the philosophically central sense of "intentional" introduced to us by Brentano and Husserl. Seeing is, quite often at least, the consciousness of something. I spend the majority of the paper discussing Travis's arguments that it is not, and providing reasons for thinking they are inconclusive. That seeings do not harbor ambitions and that "sees (NP)" is not an intensional context do not entail that seeings are not intentional. Furthermore, I argue that Travis's contention that states of seeing are relational is inconclusive, and also argue that even if that is so, the proper conclusion to draw is not that seeing is not intentional, but that some intentional states are relational.

Haut de page

Notes de la rédaction

La contribution de Walter Hopp est la discussion critique de l'article de Charles Travis, "Is seeing intentional?", disponible en ligne à l'adresse suivante : Invité à exercer son droit de réponse, Charles Travis, qui, dans un premier temps, avait accepté le principe, a finalement renoncé à le faire.

Texte intégral

  • 1 Hua III, p. 188. Edmund Husserl (1976), Ideen zu einer reinen Phanomenologie und phanomenologischen (...)

1According to Husserl, intentionality is “the own peculiarity of mental processes (Erlebnissen) ‘to be the consciousness of something’.”1 This is a very common understanding of intentionality. If this definition has a problem, it is that it is too narrow—perhaps unconscious states, and perhaps even non-mental entities, can exhibit intentionality. But whether seeing is intentional, on this understanding, seems fairly obvious. First, perceptual consciousness, as opposed to judgment or thought, is the paradigmatic sort of consciousness of something. When I tell my students, while merely thinking about Russia, that I am conscious of Russia, my remark elicits laughter—until I remind them that I am not, after all, unconscious of Russia. Seeing, moreover, is the paradigmatic kind of perceptual consciousness. It is the most revelatory of the world, and the modality least liable to be confused with mute sensation. In thinking about states that exhibit the feature of being the consciousness of something, then, it seems that seeing is the sort by whose aid we would ostensively define what we are talking about in the first place. Seeing is the consciousness of the seen, and intentionality is that “peculiarity” which it, along with many other kinds of states, exhibit.

2Philosophers, of course, say a great deal more about intentionality than these bare remarks, as they should. And I do think that many of things that philosophers have said about intentional states in general fail to be true of seeing, and that Travis’s paper makes that clear. But I wonder whether Travis has managed to cast doubt on the claim that seeing is intentional—that seeing is the consciousness of something. I do not think he has.

3First let us look at Travis’s case against the claim that seeing is intentional. As he points out, there does not seem to be any ordinary, non-technical sense of the word ‘intentional’ such that seeing is intentional. The sense in which I intentionally go to the movies is not a sense in which seeing is intentional.

4Metaphors, Travis notes, are at hand, and the most widely appealed to is that of aiming at or being directed upon an object. But aiming, ordinarily, involves intentions or, at least, harbors ambitions. But this, again, leaves seeing out, since seeing does not necessarily harbor ambitions or involve intentions. Nor, unlike judging and wanting, is seeing a matter or exposing oneself to the risk of error. In seeing, as in knowing, I don’t ask anything of the world. I simply “tak[e] up its offers” (p. 2).

“There are,” [as Travis puts it], “mental phenomena, like looking for something, which aim at something, have a goal, are directed. And there are others which, to speak loosely, aim at nothing, but rest where they have arrived—ones which are, in some sense,factive, purely relations to the way things are. Perception—seeing, for example—would seem to be a central example of that second sort of case. Knowing would be another” (p. 19).

5This line of attack is rather unconvincing. In saying that seeing or any other state, such as knowing, is “directed upon” or “aims at” an object, we need not commit ourselves to the further claim that it, like the aimings of arrows or policies or actions, either involves intentions or harbors ambitions. Rather, the relevant notion of aiming at an object is to be grasped by appealing to the phenomenon of seeing and knowing and grasping what they manifestly do have in common with other mental phenomena such as judging and wanting. In seeing the chair, it is true that I take up the world’s offers, and what it offers is a chair—that very chair, situated just so. But the world cannot offer me anything, in the sense in which it offers me the chair, without my being conscious of it, and it is through seeing that I am conscious of it. More pointedly: my seeing is the consciousness of it, my reception of the world’s offering that, in the absence of intentionality, would be wasted. The world does not offer chairs, in the relevant sense of “offer,” to tables or rocks or fires. That this property of being of or about something other than itself bears the admittedly unfortunate name “intentionality,” and that the sense in which an act of seeing is “intentional” differs from any of the colloquial senses, does not show that seeing does not bear the property or relation that it is widely thought to bear.

6The inconclusive nature of Travis’s line of reasoning can be made vivid, I think, by noting the following problem. Judging, according to Travis, is “intrinsically exposing oneself to error” (p. 2, Travis’s emphasis). It harbors ambitions, and to that extent is “intentional” in some acceptable sense of the term. Knowing, however, is not a matter of exposing oneself to error. Surely, however, I can judge what I know. I can, upon seeing a red rose, judge and believe that the rose is red. I also, barring the sorts of epistemic calamities philosophers routinely concoct, thereby know that the rose is red. And in so judging, I judge what I know. But it is difficult to see why my judging would be an exposure to risk while my knowing is not, if what I judge is what I know, and if, as is the case, I judge as I do precisely because I know what I do. And even if, somehow, the judging is an exposure to risk, my knowing that the rose is red quite obviously shares an extremely salient feature with my judging that the rose is red: each represents the rose as being red. Why isn’t this manifest feature what we mean, or part of what we mean, by “intentionality”?

7Moving on, Travis suggests that perhaps we should understand intentionality in terms of intensionality, as some philosophers have attempted to do. Intensional contexts are widely thought to share two features. First, in an intensional context, co-referential terms cannot be substituted salva veritate.

8(1) “Lois Lane believes that Superman flies”

does not entail

9(2) “Lois Lane believes that Clark Kent flies,”

even thought Superman = Clark Kent. Second, existential quantification fails in intensional contexts. (1) above, for example, is true, but

10(3) “There exists someone such that Lois Lane believes that he can fly,” is not. After all, there is no such person as Superman.

  • 2 See, for instance, David Woodruff Smith and Ronald McIntyre (1982), Husserl and Intentionality, Dor (...)

11The connection between intensionality and intentionality might be this: what explains the failure of (1) to entail either (2) or (3) is something about the manner in which mental states latch onto their objects, something, moreover, that informs our own linguistic reports of such phenomena.2 It is because Lois Lane apprehends Superman under distinct modes of presentation that she can harbor inconsistent attitudes towards him. And it is because intentional states have the remarkable ability to be of the nonexistent that her believing various things about Superman does not require there to be any existing person towards whom she holds those attitudes.

12Two problems present themselves, however. First, according to Travis, intensionality is essentially a linguistic phenomenon—a matter of words, not concepts. Suppose the concept “greasy” were intensional. Then a might satisfy it while b does not, even when a = b. But that cannot be. Nor can it be the case that both a satisfies the concept, but there be no entity x such that x is greasy. Secondly, linguistic contexts such “Sees (NP)” are not intensional contexts. If I see Venus, and Venus is Hesperus, then I see Hesperus, whether I know it or no. And if I see Venus, there is some entity such that I see it. No one, for example, can see Vulcan.

13The first point, I believe, is mistaken. Between intensionality being explicable in terms of “concepts”—and by “concept” Travis follows Frege in designating something more like properties than mental representations—and its being explained in terms of words, there is a third salient option: it is explicable in terms of entities that Frege himself helped to discover, namely Sinne—“concepts” in a much more familiar contemporary (and Kantian) sense. If I believe that silver conducts electricity, but fail to believe that the element with 47 protons in its nucleus conducts electricity, that is not just because I am ignorant of language. It is, in part, because the concept “silver” is distinct from the concept “element with 47 protons in its nucleus,” even though silver = the element with 47 protons in its nucleus, and because I have failed to master one or both of those concepts. This does not violate Leibniz’s Law. It is I, not silver, that possesses the concept “silver” but not “element with 47 protons in its nucleus.” Silver does not possess, and is not even a candidate for possessing, any concepts whatsoever (similarly, nothing that has the property of being inanimate possesses the concept “inanimate,” though each has the property of being inanimate and thereby satisfies the concept “inanimate.”) Furthermore, my ignorance is not just a matter of not knowing that the terms “silver” and “element 47” co-refer, as though the differences in the names and the sentences in which they occur somehow (how?) obscure what would otherwise be trivial tautologies. If that were so, then learning that silver is element 47 would be just like learning that “silver” and its symbol “Ag” co-refer, which it obviously is not. In the latter case, one simply learns to express the same concept by means of another linguistic sign. In the former case, but not the latter, one acquires a new concept of the very same thing.

14Travis’s second point—that “Sees (NP)” is not an intensional context—is well taken. There is no use of “sees” such that Lois Lane can be said to see Superman but fail to see Clark Kent. Nor is there any non-philosophical use of “sees” such that Lois Lane can see what isn’t there to be seen. Undoubtedly this tells us something important about seeing. But it’s not clear just what it tells us about whether seeing is intentional. Once again, it doesn’t seem to threaten the idea that seeing is the consciousness of something.

15Nor even does it undermine the idea that acts of seeing belong to the same fundamental kind as acts which somehow manage, in spite of Travis’s misgivings, to “relate” one to the non-existent. Let us grant the following:

16(4) Necessarily, if M is a case of seeing O, then if M exists, O exists.

And let us suppose that

17(5) M exists.

This does not entail

18(6) O exists.

19To secure (6), we would need the additional premise that M is necessarily an act of seeing. Similarly, while it’s true the if R is a refutation of P, then P is false, it does not follow straightforwardly that R’s existence alone secures the falsity of P. R might not essentially be a refutation; it might possibly have been or be an unsuccessful rebuttal of P. Similarly, it might be the case that while my seeing two sheets of paper entails that there are two sheets to be seen, my act of seeing could have been a hallucination of two sheets of paper. “Seeing,” that is, might not mark a natural kind, but simply the successful cases of a natural kind, just as “refutation” marks the successful cases of a common kind.

20This line of thought—that seeings belong to the same kind as hallucinations and illusions, which in turn manage to “relate” one to non-existent objects or situations, and therefore bear one of the marks of intentionality—comes under fire in Travis’s discussion of just whether we can relate, in the required way, to the non-existent in any intentional states, perceptual or otherwise. Before proceeding, however, it’s worth noting that the ability to relate one to the non-existent is not universally considered to be a “mark of intentionality,” if by that we mean (and I do not suppose that Travis does mean) that every type of intentional state is such that some of its possible members can do that. Neither Brentano nor Husserl, to cite two examples, maintains that every type of intentional state has the power to relate one to the non-existent. Inner perception, for Brentano, is not only infallible, but infallibly relates one to things which exist. For Husserl, adequate intuition is its own species of act which relates one only to things which exist. So it’s not clear why, even if seeings belong to a natural kind none of whose members relate one to the nonexistent, that would have any tendency to show that seeing isn’t intentional.

21Nor is it clear that seeings do belong to a fundamentally different natural kind than do acts which can reach out to the nonexistent, such as hallucinations. Suppose, to take Travis’s example, that Sid has a visual experience as of two sheets of paper, but without there being two sheets of paper in the relevant portion of the environment. There is, says Travis, a use of “see” such that “Sid sees two sheets” can express a truth. But this does not show that there is an intentional use of “seeing.” Similarly, he argues, if Pia is imagining herself running with the bulls, there is a use of “Pia is running with the bulls” that expresses a truth. But this hardly shows that there is an intentional use of “run with the bulls,” and even if it did, it would cast the net of intentionality too widely. In saying that Pia is running with the bulls, but without the benefit of bulls, we are describing how Pia imagines things.

“Similarly,” [writes Travis], “when, on the above account, ‘Sid sees two sheets of paper’ expresses a truth though there are no two sheets of paper, it is not being used to describe what Sid is doing, or at least not what he is doing as seeing two sheets of paper. It is being used to describe what, for him, it is as though he were doing” (p. 11).

22The two situations are, however, disanalogous in an important respect. In imagining herself to run with the bulls, the activity of running is part of the state of affairs of which Pia is conscious, as an object, while her act of imagining is not. She is not imagining herself imagining herself running with the bulls. Sid, for his part, is in a state S whose object is two sheets of paper, but is not in a state with his seeing, or seeming to see, as an intentional object. (I don’t mean to deny that we are conscious of seeing when we see, but just that it isn’t itself something seen.) Sid’s state S, that is, is more like Pia’s imagining than it is like her running. The structure of their acts, that is, is as follows:

23Pia: Imagines : running with the bulls.

24Sid: Sees (or seems to see) : two sheets of paper.

But not this:

25Pia: Imagines running : with bulls.

Or this:

26Sid: Sees (or seems to see) : seeing two sheets of paper.

27It seems to me that in order for Travis’s analogy to hold up, however, one of the latter two depictions of the structure of their intentional acts would have to be correct. So saying that there is no intentional use of “run with the bulls” does not show that there is no intentional use of “see.” Rather, what we would need to establish is that there is no intentional use of “imagine.” I don’t know whether or not there is an intentional use of “imagine.” What I do know is that imagining is an intentional state if anything is.

28In any case, let us agree in any case that Sid no more sees two sheets than Pia runs with bulls, despite the fact that the argument does not quite show that. Still, I think it’s undeniable that both states of affairs involve intentionality in a straightforward way, and, moreover, that in both cases we find Pia and Sid relating to the nonexistent. Start with Pia. No, she is not running with any bulls. But imagining oneself running with bulls, without the benefit of bulls, appears to be the type of state that “relates” one to the nonexistent. That state is not one of running, of course, but one of imagining.

29Now to Sid. In order for it to be, for Sid, as though there were two sheets before him, he must be conscious of two sheets. One might be tempted to say that the sensory experience was not the consciousness of two sheets, and place all of the burden on the state of its seeming to Sid as though there were. Even so, he is conscious, by way of a seeming, of two sheets, and there aren’t two sheets. Sid is, like Pia, related to the nonexistent. And I suspect that, in at least some cases, the error falls on the side of the experience. Sid can cease judging entirely that there are two sheets there, but his experience will persist in being of two sheets. Hallucinations do not successfully reveal weird objects that exist—mental imposters or sense-data or “appearances,” say. Rather, they are unsuccessful states with ordinary, but non-existent, objects.

30If that is right, then seeing and hallucinating—just like imagining and thinking—can have the same kinds of objects (even if not the same individual objects). That hardly establishes they belong to the same kind. But further considerations suggest that they do. States of the same kind tend to be harmonious with, or disharmonious with, one another. By this I do not mean just that their contents can be consistent or inconsistent with one another. Rather, the states which bear those contents can exist in more or less coherent experiential wholes. Two beliefs, for instance, whose contents conflict with one another cannot coexist, or at least not stably, in a single act of consciousness. Nor even can two imaginings. My imagining-consciousness is unstable if I imagine myself both to be running with the bulls and, say, sitting in a chaise lounge simultaneously. Imaginings and beliefs, however, do not compete in this way. No matter what I believe, I am free to imagine whatever I please, with no felt tension. Nor does imagination compete with perception. Who knows what Pia sees while she imagines running with the bulls. Whatever it is, it won’t prohibit her from imagining that, unless it robs her of the required attentional resources, which is a different competition on another field.

31Visual hallucinations, however, do compete with seeings. Perceptual experiences might be silent on a number of issues, but they need not be silent on whether there is one or two sheets of paper before one. If they were, they would have no bearing on what I am entitled to judge to be before me. Silence doesn’t justify. And just as a perceptual experience of two sheets of paper cannot coexist with a perceptual experience of just one sheet of paper, so a hallucination of two sheets of paper cannot coexist with a perceptual experience of just one sheet of paper. It is hard to think of any other type of state of which this is true. Even beliefs can coexist with perceptual experiences whose content undermines or conflicts with theirs. I can believe, for instance, that there is really only one sheet of paper before me, even if I genuinely see two. One plausible reason this is so is that hallucinations, just like perceptual experiences, aim at the world and purport to present, rather than merely emptily represent, it. And that is an awful lot to have in common with them.

32But are Pia and Sid really “related,” in any way, to the nonexistent? Travis, if I read him rightly, thinks not. Even in judgment and desire this does not occur.

“Things being as they are, in the normal catholic sense of ‘things’, is the extent of what there is to judge of” (p. 13).

33Of course Travis does not mean that all judgments are true—that facts are the only things we can judge of. One could judge that there is a fountain of youth in the Everglades, for instance, and one can want to find it. But what one is thereby related to in judging and wanting is not an object—not a thing in the normal catholic sense. So what is it?

“There is a way for someone to be: having found a fountain of youth. It is a perfectly good way for someone to be, even if it will never be instanced by anyone being as he is” (p. 17).

34Applied to Pia’s case, we might say: in imagining that she is running with the bulls, Pia is not first picking out some bulls, and then imagining herself running with them. What she is imagining is a way for someone to be, namely running with some bulls, and that is a perfectly possible way for things to be.

35I remain unconvinced that this avoids the problem of the nonexistent and our ability to “relate” to it—to intend it in thought and desire and, possibly, sensory experience. First, ways for things to be that are not instantiated may exist, but they are not actual. But if mental states can relate to entities which are not actual—not part of any environment—then already they exhibit a very peculiar feature not possessed by much of anything else. What else can relate to uninstantiated ways things might be?

36Secondly, I am not convinced by the analyses above. Is having found a fountain of youth a perfectly good way for someone to be? I am in no authoritative position to answer this, but I am not convinced that it is. At the very least, I am prepared to learn from biologists or physiologists or even physicists that such a thing is not possible, but am not prepared to conclude Ponce de León was not looking for one on that basis.

37Nor does the suggested analysis of Pia’s imagination hold water. Among the things that Pia imagines is a way for things to be, namely someone running with bulls. But Pia imagines more than that. She imagines herself running with bulls. And while that too is a way for things might be, it is not how things are. Her act of imagining is not relational, since she is not related to anything actual.

38Finally, not all judgment and desire successfully relates us to ways things might be, since plenty of judgments are necessarily false and plenty of ambitions are directed towards the impossible. Someone who intends to square the circle or invent a perpetual motion machine is not related to anything actual, nor even to a way things might be. Things won’t ever be like that. But people have intended to do that. Nor could things be the way a child who wants to meet Santa Claus hopes they will be. Even if, incredibly, “Santa Claus” is, like “the fountain of youth,” not a name (see p. 16), it does not denote a possible person, and so meeting that person is not a way for things to be.

39I do not wish to minimize the difficulties in affirming that intentional states manage to be of or about things which do not exist. But if it is “cheating” to affirm that all intentional states aim at objects, only to add that the objects might be nonexistent, it is equally cheating to deny that mental states aim at the nonexistent by construing the false and absurd ones as relations to non-actual, uninstantiated ways things might be. Nor is that answer satisfactory for a wide range of mental states.

40So, I remain unconvinced that at least some mental states do not have the power to “relate” one to what does not exist. More importantly, even if seeing is the type of state none of whose possible instances could do that, I fail to see how that shows that it isn’t intentional, where intentionality is, as Husserl has it, the peculiarity of conscious states to be the consciousness of something. If seeing is relational, the proper conclusion to draw, it seems to me, is not that it is not intentional, but that some intentional states are relational.

Haut de page


1 Hua III, p. 188. Edmund Husserl (1976), Ideen zu einer reinen Phanomenologie und phanomenologischen Philosophie. Erstes Buch: Allgemeine Einfuhrung in die reine Phanomenologie, Husserliana vol. III, Karl Schuhman (ed.), Den Haag, Martinus Nijhoff. Edmund Husserl (1982), Ideas pertaining to a pure phenomenology and to a phenomenological philosophy. First book: general introduction to a pure phenomenology, F. Kersten (transl.), The Hague, Martinus Nijhoff, p. 200.

2 See, for instance, David Woodruff Smith and Ronald McIntyre (1982), Husserl and Intentionality, Dordrecht, D. Reidel Publishing Company, p. 33ff.

Haut de page

Pour citer cet article

Référence électronique

Walter Hopp, « Is Seeing Intentional? A Response to Travis », Methodos [En ligne], 14 | 2014, mis en ligne le 05 mai 2014, consulté le 28 mars 2017. URL : ; DOI : 10.4000/methodos.4108

Haut de page


Walter Hopp

Boston University

Haut de page

Droits d’auteur

Licence Creative Commons
Les contenus de la revue Methodos sont mis à disposition selon les termes de la Licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo DOAJ - Directory of Open Access Journals
  • Logo UMR Savoirs, Textes, Langage
  • Logo CNRS - INSHS
  • Logo Université de Lille