For Lack of Intentionality

From Phenomenology 2005, Vol. 5, Selected Essays from North America, ed. Lester Embree and Thomas Nenon (Bucharest: Zeta Books, 2007). A collection of essays from the International Organization of Phenomenological Organizations. Available in printed as well as electronic form at


Much of contemporary Philosophy of Language has attempted to explain the relationship between language and the objects referred to by it without recourse to the intentionality of acts of consciousness, as Husserl and other Phenomenologists have understood it. This essay takes one author from the "Analytic" tradition, David Wiggins, and points out the inadequacies in his recent attempt to explain how "natural kind terms" connect up with the objects to which they apply. It traces the failure to build an intelligible bridge between the terms and their extension to failure to incorporate intentionality into the analysis of meaning.


In this essay I would like to exhibit some important ways in which a phenomenological treatment of the relationship between consciousness or language and its objects fundamentally differs from the treatment given by means of linguistic analysis. Since the latter treatment exhibits wide variations, though all within a basic orientation, I will focus upon the account given in David Wiggins’ Sameness and Substance Renewed (2001).1 The account worked out in this book claims close affinity with the works of Frege, Kripke and Putnam, and the difference between it and the accounts of these other writers does not really matter for the issues I shall be raising.

On the phenomenological side I shall limit myself to views of Edmund Husserl. Wiggins invites comparison to Husserl by utilizing a lengthy passage from the latter’s Formal and Transcendental Logic as an epigraph for his Chapter Five, "Conceptualism and Realism." This passage reads as follows:

The relation between my consciousness and a world is not a mere matter of contingency imposed on me by a God who happened to decide the matter this way rather than that, or imposed on me by a world accidentally preexisting and a mere causal regularity belonging to it. It is the a priori of the judging subject which has precedence over the being of God and the world and each and everything in the world. Even God is for me what he is in consequence of my own productivity or consciousness. Fear of blasphemy must not distract us here from the problem. [But] here too, as we found in the case of other minds, the productivity of consciousness does not itself signify that I invent or fabricate this transcendency, let alone this highest of transcendencies. 

(Husserl, 1969, § 99)

Wiggins’ concern in using this passage is to bring out a major point of agreement in outcome between his analyses and those of Husserl: namely, the existence and nature of reality as independent of the consciousness or language that represents it—or that even apprehends it in knowledge. Act and object are suited by their natures to each other, but are existentially independent. The same point is made by a quote from Simone Weil just above the one by Husserl: "When we come to exhibit the birth of thought we shall find that it is born into a universe that is already ordered." The act does not give its object its nature, does not make it what it is. A very elaborate account of concepts and of how consciousness (language) represents or grasps its "objects" is not inconsistent with the reality grasped existing and having its nature on its own. Thus Wiggins’ chapter title, "Conceptualism and Realism." Such a position is strongly opposed to another position, most familiar in a line of thinkers running from Kant to Donald Davidson and Richard Rorty. So the point at issue is a major one indeed, affecting the overwhelmingly "constructionist" tendency of Modern thought up to the present, and with roots that reach far back beyond Kant, at least to John Locke and the rise of Empiricism.

Still, the analysis of consciousness (or language) in relation to objects differs profoundly between Husserl and Wiggins. It is not clear how they can come to a shared outcome, when their accounts differ so widely. But we shall not argue this point. Our strategy here will be as follows: First we will sketch the structure which Husserl discovers between consciousness and its world. Second, we will do the same for Wiggins. Then we will point out some significant issues for Wiggins’ account.


The act of consciousness. The basic unit of analysis for Husserl is the "act" of consciousness. Some things Wiggins says might suggest that the same is true for him, but his overall account makes it clear that this cannot be so. For Husserl, the "act" of consciousness is not really an activity, but is a peculiar state or ‘fact’ of mind. (Husserl 1970, pp. 533-534, 552-556, 563n) It is the act-ualization of a set of properties that define the state. The most salient feature of an act of consciousness is its "intentionality," the specific ‘ofness’ or ‘aboutness’ of that particular state of mind. Following a long tradition revived by Husserl’s teacher, Franz Brentano, the act of consciousness is individuated by its object. But that is dangerous and misleading language, for the ‘object’ may or may not exist. In the case where it does not exist, it is still the ‘object’ of the given act. It is still what the act is of or about. So it might really be better—though a stretch on the English language—to say that the act of consciousness is individuated by its specific ‘ofness’ or ‘aboutness.’ Acts with different ‘objects’—different specific ‘ofnesses’ or ‘aboutnesses’—cannot be identical.

But the specific ofness or aboutness of an act—which we will from here on simply refer to as its "intentionality" ("selectivity" could be another term for it)—involves much more than just which object or entity it is about. Two or more acts can be about the same object, but about it in different ways. The object may be presented under different properties: as the victor of Jena, the vanquished of Waterloo, Josephine’s husband, and so forth—Napoleon Bonaparte in each case. And the object may also be presented with less or greater clarity or fullness of detail. The object may be represented in combination with different "propositional attitudes," as we now (rather misleadingly) call them. Three additional ways in which acts of consciousness, with their intentionalities, may differ is with respect to their inner complexity and composition, their foundations (what—especially what other acts—are presupposed in their obtaining), and their "margins" or "horizons" (what other acts, with other objects, they may pass over into in a coherent and continuous manner). These are some of the major ways in which "acts" of consciousness may differ from one another, and there are yet others to be discussed for any exhaustive account of consciousness and its acts.

Thus, intentionality is not, for Husserl, a simple "beam" that falls now on this object and now that, in the manner of a flashlight, as one might suspect from what some other traditions of thought have to say—especially Empiricism and its offshoots. "Intentionality" refers to a vast field of inquiry around which the Phenomenology that Husserl developed organized itself. (Husserl 1931, pp. 404 & 12-13) The method of Phenomenology is description through reflective analysis, comparison and abstraction (aided by variation in imagination), directed upon acts of consciousness with reference to their objects and their components. It is possible, on Husserl’s account, to observe or view the bearing of acts of consciousness, including those involving language, upon their objects of various kinds.

Now the basic nature of the "act" being of or about its particular object is regarded by Husserl as a descriptive ultimate, something you come to know what it is by "viewing" it or bringing it before the mind in intuition. (Husserl 1970, p. 400) This is very much like what Locke had to say about "ideas of reflection." (Essay Concerning Human Understanding, Bk. II, Chap. IX) However, Locke never manages to get beyond particular species of mental acts to intentionality itself. Once the amazingly rich ‘inner’ complications of acts of consciousness come into phenomenological "view," however, there is much to say about how particular types of objects, including realities (ordinary "things"), come before the mind, and even, in many cases, what knowledge of those objects consists in and how it develops and is to be obtained. One of Husserl’s earliest and most enduring interests in all of this was the role of language and other symbolisms in the representation and knowledge of the subject matter of Mathematics—and then, very soon, of the subject matters of the other sciences and of the common sense matters of the world in which we live. The theory of "constitution" was, for him, an account of how objects or objectivities of all kinds come before the mind in the various ways—including knowledge—that they may do so. Major parts of the theory of constitution dealt with sensation, concepts, propositions, time-consciousness, language and linguistic "meanings," and the roles of culture and history, but all organized around how objects of various kinds come to mind.

However, it can never be too early to say that what is "constituted," according to Husserl, is never created or modified by the "constituting." That is the main point of the passage quoted above by Wiggins, and of many similar passages in Husserl. What is created or modified is the act or acts in which the objects come before the mind and sometimes are grasped in knowledge.

Let us now sketch the overall structure of the bridge between the mind or acts of consciousness and the objects represented or known by them. To keep it as simple as possible, we will consider the representation of familiar physical objects such as those Wiggins concentrates on: horses, trees and people.

We have said that the unit of analysis for Husserl is the act of consciousness, and that these are individuated by their "object"—by their being of a certain thing. On his approach we do not start out toward objects from what is "within" the mind or within the act. That we come to later. Philosophical analysis of the mind and its acts works its way "in," starting from what the acts are of or about. We do not start with something such as sensations or "ideas" and try to work toward an object. The essence of "intentionality" is that we are, from the beginning, open to the world of objects and it to us. The mind is not an enclosure. We start with a thought or other way of being conscious of an object, and what the object of the act is is either clear from the outset, or clear enough to lead toward increased or (in a few cases) total clarity by further acts prescribed. (Husserl 1970, p. 412-413) I am aware of something across a field, perhaps, but I can’t tell if it is a lamb or a pig. But the experience I am having opens up into or indicates others which will settle the question. Moving closer I see it is a pig, but not which pig it is. (In the progression of this series of acts of perception, what they each are of remains the same—the "determinable X" of §§131-135 of Ideas I.) Upon still closer examination perhaps I identify or "individuate" the "X" in question. I recognize it as the very pig I bought from farmer Brown. I now know or at least represent which pig it is. I grasp its identity. My conscious acts have moved from being of "some animal over there" to being of that pig, remaining all the while of the same object or "determinable X." The modification of the intentionalities in the stream of distinct but essentially inter-related acts directed upon the same object eventuates in a consciousness of the identity of the object, which encompasses the kind of thing it is and which thing of that kind it is.2

One further step in the progression of acts directed upon the same object is possible in some cases. We might call this the step of verification. Husserl himself, after describing the above type of progression around a "determinable X," asks: "But is it actually the same? And is the object itself ‘actual’? Can it not be non-actual while the manifoldly harmonious and even intuitionally fulfilled posita—posita of any essence-content whatever—still flow off in the way peculiar to consciousness?" (§135 of Ideas I, Husserl 1982, p. 324) Husserl believes, beyond question, that such questions can be answered in most cases, one way or the other, with adequate if not apodictic Evidenz. (§§ 137-138 of Ideas I) The area of phenomenological analysis that deals with this he calls "The Phenomenology of Reason," which elucidates the relationships between "Reason and Reality," the title of the "Fourth Section" of Ideas I. In general, verification involves the bringing of the respective object of a sequence of acts to increasing intuitive fullness through that sequence. What this means will vary greatly depending on the kind of object involved. Here it might mean examining identification marks on the pig and looking at my records of transaction, or, in an extreme case DNA or other testing. This is the process of "fulfillment," so called because in it "empty" or merely signitive components of the intentionality directed upon the same object are "filled full" of intuitive presence of "the object itself." At the extreme upper levels of "fulfillment" the acts of consciousness pass over into knowing, where the object is found to be as it was represented or thought to be.3 The"Phenomenology of Reason" is the descriptive analysis of how, in general, this goes for the various main categories of objects. In the case of knowing, the object is included within a whole consisting of it and the appropriate, very complex, acts of consciousness. In this case "intentional inexistence," the possibility of the object not existing, does not apply, as it does to mere intentionality.4 It should be emphasized for the purposes of this essay that, according to Husserl, the major connections within and between acts in which an objectivity is represented or known are to be viewed or observed in reflective analysis, not inferred or theorized. "On our view," Husserl says in 1913, "theory of knowledge, properly described, is no theory…. It neither constructs deductive theories nor falls under any…. Its aim is not to explain knowledge…, but to shed light on the Idea of knowledge in its constitutive elements and laws." (Husserl, 1970, pp. 264-265; see also § 75 of Ideas I) We should also note that Husserl painstakingly elaborated a (supposedly) descriptive set of structures and processes whereby his analysis of the act of consciousness carries rather directly over to language in use and in life. (Husserl 1970, pp. 269-333 and § 124 of Ideas I)


Wiggins’ link to the extension. It will help us find our way if we say at the outset that there is nothing in Wiggins’ scheme of analysis that corresponds to "intentionality" (basic "ofness" or "aboutness") in Husserl’s.

Wiggins states that "that which individuates—in the one sense in which the word will be used in this book5—is in the first instance a thinker. (Derivatively, but only derivatively, one may find oneself saying that a substantive or predicate individuates.)" (Wiggins 2001, p. 6) On page 110, discussing A. J. Ayer’s theory of referring, Wiggins remarks, in agreement with Ayer, that "Where a thinker A conceives of an individual x as Φ (and by moving from ‘refer’ to ‘conceive of ‘ we can only make Ayer’s claim more foundational), Φ cannot assume just any value. There are restrictions on Φ and they depend on which entity in particular the entity x is." This statement expresses the fundament principle of Wiggins’ exposition of de re necessity, but for now we only note that "conceiving of " (something a thinker does) is held, with no explanation, to be "more foundational" than "referring" (something language does, or a thinker does with language).

Wiggins’ analysis in this book, however, is carried out purely in terms of the referring or "picking out" done by language, in particular by "sortal predicates" and especially by "natural kind words." The reason for this is pretty obvious. There is a highly regarded tradition of analysis of names and terms, running from Frege to Kripke and Putnam, which Wiggins can avail himself of and extend to solve his problems. There is, in Wiggins’ thought environment, nothing similar for the analysis of thought or "conceiving of." His acknowledgement that "conceiving of" is "more foundational" than "referring to," together with the necessity (as he sees it) of carrying out his analyses only in terms of language (sortal predicates), creates an unresolved tension that blurs Wiggins’ entire line of argument as well as its conclusion. But for now we turn to his account of how sortal predicates manage to "pick out" or individuate entities in the real world. That is, how they manage to unite or come together with them in the manner peculiar to them. (Obviously the word as flatus vocis does not "come together" with the object.) What are the connecting links that bridge the gap?

A few passages from Wiggins must serve to give the elements of the bridge from sortal predicate in action to the members of the corresponding extension:

A sortal predicate has a sense, a reference or designation, and an extension; and there pertain to it both a concept (which may have marks) and a conception. (p. 79n)

Like other predicables, a sortal predicate expresses a sense and, by virtue of expressing this sense, it stands for a concept. Under this concept individual things may fall. See the diagram in Frege’s <1894>letter to Husserl. To understand a predicate and know what concept it stands for is to grasp a rule that associates things that answer to it with the True and things that don’t answer to it with the False. (The extension of the concept is therefore the inverse image of the True under the function determined by this rule.) To grasp the rule is to grasp how or what a thing must be (or what a thing must do) in order to satisfy the predicate. To grasp this last is itself to grasp the Fregian concept. Thus ‘horse’ stands for that which Victor is and Arkle is, for instance—just as, outside the sortal category, the verb-phrase ‘runs swiftly’ stands for that which Arkle does. When I declare that to grasp this rule is to come to understand what horse is or run swiftly is, someone may insist that, in that case, the concept so spoken of, horse or run swiftly or whatever it may be, is a property. I shall not demur, but simply insist in my turn that the notion of a rule of correlation to which I appeal is pretheoretical. 

(pp. 9-10)

Immediately following this, Wiggins tries to clear up his crucial distinction between concept and conception:

Seen in this way, as something with instances, the concept belongs on the level of reference (reference in general being thing of which naming is one special case). But there is another use of the word ‘concept’ which is equally common, if not more common, and this belongs on the level of sense. It is this rival use of the word ‘concept’ that we find in discussions that are influenced directly or indirectly by Kant. In those discussions, talk of things falling under a concept, or of concepts having extensions, may be less felicitous…. Perhaps everything will fall into place, however, and the connexion will be visible between the two uses of the word, if we try to reserve the word ‘concept’ for the Fregean use and we prefer the word ‘conception’ to cover the Kantian use (seeing a Fregean sense as a very special case of a conception). The connexion that there is between the two may then be understood as follows:

Thinker T has an adequate conception of the concept horse (an adequate conception, as one says in English, of a horse) if and only if T can subsume things under horse, knows what it would take for a thing to count as a horse, and has some sufficiency of information about what horses are like.


Some confusion on Wiggins’ part is perhaps indicated by his statement here that the conception is in no way the same as the concept, whereas he previously allowed (see above) that the Fregian sense (concept) could be regarded "as a very special case of a conception." Certainly if the conception is a set of beliefs it is hard to see how a Fregian sense could ever be a special case of that.

But before beginning critique, let’s just try to lay out the pieces of the bridge from sortal predicate to members of its extension. We seem to have here a number of items:

  1. The word as mark or sound (flatus vocis).
  2. The sortal predicate, the word in use (the sound or mark plus...?).
  3. The sense as "conception." (A certain set of beliefs.)
  4. The "rule" associating the sortal predicate with members of its extension.
  5. The concept (Fregian sense)
  6. The properties which something must have in order to "satisfy" the predicate.
  7. The things that "fall under" the concept.


It is likely that (5) and (6) are identical, on Wiggins’ view, but all of these must play a part in constituting the bridge. Now, how do they all hang together? On this point one must say that not only Wiggins, but others who have used a similar scheme, prove to be remarkably unhelpful when closely examined.6

Speaking in terms of the thinker Wiggins remarks: "…For a thinker to single out or individuate a substance, there needs to be something about what he does, something about his rapport with x or his relation state toward x and his practical sensibility in relation to x which (regardless of whether he articulately knows this or not—for all he needs is clear indistinct knowledge…—and regardless of whether it is a singling out as) sufficiently approximates to this: the thinker’s singling x out as x and as a thing of a kind f such that membership in f entails some correct answer to the question ‘What is x?’" (p. 7, see pp. 82-83 and note #6 on p. 83) Indeed, what is the rapport, the required "something about" what the thinker does? We shall stay with the account in terms of language, "sortal predicates," for in fact Wiggins simply has nothing at all to say about the bridge between the individuating thinker, his "individuating," and what he "singles out." (He does not take up the question of whether the thinker can do it without language, nor, indeed, how he would do it with language.)

So what are we to make of the links between the elements of the bridge? What are the issues concerning them? Here we will briefly comment on:

(A). The link between the word and the (Kantian) ‘sense’ or conception.

(B). The link between the conception or ‘sense’ and the Fregian sense or concept.

(C). The link between the concept and the members of the extension.


And first, what are we to make of the link between the word "horse," for example, and the "sense" or "conception" that somehow carries it toward the entities (horses) which "satisfy" the term or which it picks out or applies to? How is the word related to this (Kantian) "sense"? Wiggins says above that to "understand a predicate…is to grasp a rule that associates things that answer to it with the True…. to grasp how or what a thing must be…in order to satisfy the predicate." But the notion of "a rule that associates" the predicate (word) with what a thing must be in order to satisfy that predicate surely just packs into a phrase all that we need to be given some account of if we are to understand how the word is, finally, to be united to the members of its extension. It simply presupposes what is to be explained. It also threatens to use the "thinker" or "user" as part of the bridge, which will again surely presuppose what needs to be explained, if the user must be able to "pick out" the members of the extension in order for him to associate the predicate word with them. I imagine that what Wiggins’s has in mind, to unite the word with its sense, is something that just "goes on" with people using a language in their common world. The terminology he uses suggests this to me: terminology such as "a rule that associates" words and things and such as a "function." For this "rule" certainly is not going to be, in the usual run of things, an explicit, conscious assignment of word to thing, but will be an assignment that "happens" in the atmosphere of language use under social constraints. But it also certainly is not going to be a matter of a law-governed regularity with no input from consciousness and choice. The language user will not be a robot. So what kind of thing would a rule have to be to "associate" words with things? What exactly does this association amount to? And to speak of a "function" perhaps indicates something about the "association," but makes nothing clear about its nature or why it is there.

The obscurity only deepens when we turn to the link (B), between the conception and the concept. The "conception" (a set of beliefs?) is said by Wiggins to be of the (Fregian) concept, its reference. So the conception refers to, designates, the concept. He holds that the conception may (within limits) not actually grasp the concept rightly, but still have it, and that the concept and thereby the extension is accessed to the sense or conception, and then to the word, in virtue of "clear and indistinct" applications to a few paradigmatic exemplars, which may exhibit marks or necessary conditions of the concept’s presence at once (1) in the exemplars and (2) "to" the Kantian "sense" which "refers" to it. The "marks" apparently serve to inexplicitly guide the sense and the word home to the concept (or essence) in cases where we do not yet know what the "concept" or essence is. (The familiar case of water before we knew it was H2O, for example.) Thus it appears that the Kantian "sense" may be of the concept, its "reference," even in cases where the one applying the relevant term does not know what the concept is. One can still, quite mysteriously, have an "adequate" conception of the concept.

Turning to the last link, (C), what Wiggins seems to be saying here is at first rather clear. The concept consists of those properties of members of the extension which constitute the essence of those members. It is what a horse, for example, is. It is "something with instances," and those instances are things like Sea Biscuit and War Admiral, in the case of the sortal concept horse. "It is something general or, better, universal;…" (p. 10) The relationship between the concept and members of its extension is, we presume, predication (instantiation, exemplification): the relation of properties to the things that have them. This is philosophically familiar, at least, if not wholly uncontentious. But a familiar manner of speaking, which Wiggins himself uses, has the extension also "falling under" the concept. And "falling under a concept" has not been generally understood as the same relation as instancing a property. Also, the characteristics of concepts seem to be broadly different from—though essentially related to—the characteristics of the properties of the things that fall under the concepts.7

The "other side" of the concept as described by Wiggins, toward the "conception" whose reference it supposedly is, is not only unfamiliar, but seems to me almost totally obscure, given what Wiggins says of it. He gives this side a familiar name, "reference," but it is certainly hard to see how the concept (as above) could really be the reference of the conception, even roughly as Tom is the reference of "Tom," which Wiggins suggests. And in any case, being a "reference," in this sense, is being an object of a referring. The relationship between the "reference" and what refers to it (in this case a set of beliefs) is the referring, and of this "referring" Wiggins has nothing at all to say. That returns us to the questions about link (B), which to my mind is the fatal rupture in the proposed bridge from the sortal predicate to the extension.

Finally by way of "issues," Wiggins—as indeed others in his tradition of analysis—has simply nothing to say of how one comes to know that the bridge between the sortal predicate and its extension consists of the elements cited and that they are linked up in the way suggested. The most likely suggestion might be that they are giving a transcendental argument of sorts. We know that sortal predicates do apply to a range of things and not to others. Perhaps we even want to say that this is an observable fact, though that is far from clear. But how words, being what they are, "reach" to a determinate range of entities and not to others is not an observable fact by any usual sense of "observable." So the bridge and its links are perhaps proposed as an explanation of how they do that. That might be proposed simply as a hypothesis, but I doubt it could ever be presented as one that is empirically testable. Nor as one which is based upon a priori insight. So the alternative might be to present the bridge proposed as "the conditions of the possibility" of individuation or picking out by sortal predicates. In that case we might avoid the task of saying how the links work For that, as is well known, is one of the advantages of "transcendental" arguments, in Kant and later thinkers.


Is there an omission? What I hope is clear, at this point, is that Wiggins’ account of how sortal predicates come together with members of their extension, of how they "pick them out" or individuate them, is severely incomplete and unclear. Let us restate some main points. The relation of the "concept, in his Fregian sense, to the extension of the predicate appears to be a familiar one, predication, and one which, once the word gets to the "concept," easily explains how it extends itself onward to members of its extension. All of that makes pretty good sense.

Not so, I believe, for the rest of the "links" in the bridge. The link from the (Kantian) "sense" to the (Fregian) "concept" is simply not made out by Wiggins. And it is here, I think, that the lack of intentionality weighs most heavily on his account. If, indeed, the sense is a set of beliefs, they would be good candidates for bearers of intentionality directed upon his "concept." But that would have to be a matter of concepts which the beliefs contain, not concepts the beliefs are of. It is, precisely, the "of-the-concept" that must be made intelligible in application to the "conception." Indeed, I suspect that finding himself forced to two senses of "sense" goes much deeper into the problematic of "picking out" than Wiggins himself realizes. The "ofness" of the beliefs certainly cannot consist in the properties necessary to the objects picked out. The problem is, precisely, to get to those properties. And here Wiggins simply lacks any account. I believe a similar lack afflicts the other writers who take an approach along similar lines. (Willard 1994)

Other gaps open up in the Wiggins bridge, not least the problem with the "rule of association" mentioned above. That association must surely go through or depend upon the connection of the word with the sense (and then the sense with the concept). But Wiggins simply has nothing illuminating to say about this, much less about the relation of the word as flatus vocis (mere sound or mark) to the word "in action" (the sortal predicate). Further questions might be raised about the kind and how it fits into the bridge, if it does, but we have said enough for now.


Where the difference lies. I have not intended in this essay to make any judgment about whether Wiggins or Husserl has the superior account overall of the union between sortal terms or thoughts and members of their extension: of how they are "together" with each other, as they undoubtedly are. Obviously Husserl’s account assumes a radically different ontology and theory of knowledge than Wiggins.’ In some general sense, I believe, Wiggins and "Linguistic Analysis"—though certainly not Frege—wants to be within an empirical tradition, while Husserl explicitly treats that tradition as a hopeless cause. (Husserl 1970, pp. 115-117 and Husserl 1931, §§ 19-20) Which side triumphs in the analysis of "picking out" will obviously depend, in part, upon resolutions of broader issues.



The resources on this website are freely available here because of the generosity of our supporters.

Become a supporter


Related Resources

You may also like...