Husserl and Intentionality (David Woodruff Smith, Ronald McIntyre (Auth.)
Husserl and Intentionality (David Woodruff Smith, Ronald McIntyre (Auth.)
Husserl and Intentionality (David Woodruff Smith, Ronald McIntyre (Auth.)
A PALLAS PAPERBACK
~p~
\.1[1 paperbacks
DAVID WOODRUFF SMITH
Dept. of Philosophy, University of California, Irvine
and
RONALD McINTYRE
Dept. of Philosophy, California State University, Northridge
HUSSERL AND
INTENTIONALITY
A Study of Mind, Meaning, and Language
DORDRECHT/BOSTON/LANCASTER
Library of Congress Cataloging in Publication Data
ACKNOWLEDGEMENTS vii
ABBREVIA TIONS ix
PREFACE xi
INTRODUCTION xiii
BIBLIOGRAPHY 407
We thank the appropriate parties for their kind permission to quote at some
length, for the purpose of scholarly commentary, from the following works
by Edmund Husserl:
Cartesian Meditations, English translation by Dorion Cairns (Martinus
Nijhoff, The Hague, 1960);
Experience and Judgment (revised and edited by Ludwig Landgrebe),
English translation by James S. Churchill and Karl Ameriks (Northwestern
University Press, Evanston, 1973) (British Commonwealth rights licensed to
Routledge & Kegan Paul, Ltd.);
Ideen zu einer rein en Phiinomenologie und phiinomenologischen Philos-
ophie, erstes Buch, herausgegeben von Walter Biemel (Martinus Nijhoff, The
Hague, 1950), quotations being in our own English translations;
Logical Investigations, Volumes One and Two, English translation by J. N.
Findlay (Routledge & Kegan Paul, Ltd., London, 1970) (American rights
licensed to Humanities Press, Inc., Atlantic Highlands, N.J.).
ABBREVIA TIONS
ix
x ABBREVIATIONS
I-IV) were published in 1913; Vol. II, Pt. 2 (Investigation VI) was published
in 1921.) 5th printing, Tiibingen, 1968.) The flIst edition of Logische
Untersuchungen was published in 1900-1901 in Halle by Niemeyer.
PP Phenomenological Psychology. Trans!. by John Scanlon. Nijhoff, The Hague,
1977. [Phiinomenologische Psychologie. Ed. by Walter Biemel (Husserlillna
IX). Nijhoff, The Hague, 1962.) Lectures delivered by Husserl in the sum-
mer semester of 1925.
Time The Phenomenology of Internal Time-Consciousness. Ed. by Martin Heideg-
ger. Transl. by James S. Churchill. Indiana University Press, Bloomington,
Ind., 1964. [Vorlesungen zur Phiinomenologie des inneren Zeitbewusstsein.
Ed. by Martin Heidegger. Niemeyer, Halle, 1928.)
Zeit. Zur Phiinomenologie des inneren Zeitbewusstseins. Ed. by Rudolf Boehm
(Husserlillna X). Nijhoff, The Hague, 1966. In addition to Husserl's Vor-
lesungen zur Phiinomenologie des inneren Zeitbewusstsein, this volume
contains supplementary texts not translated in Time. References to these
texts will be indicated by ·Zeit.'
PREFACE
This book has roots in our respective doctoral dissertations, both completed
in 1970 at Stanford under the tutelage of Professors Dagfmn F ¢llesdal, John
D. Goheen, and Jaakko Hintikka. In the fall of 1970 we wrote a joint article
that proved to be a prolegomenon to the present work, our 'Intentionality via
Intensions', The Journal of Philosophy 68 (1971). Professor Hintikka then
suggested we write a joint book, and in the spring of 1971 we began writing
the present work. The project was to last ten years as our conception of the
project continued to grow at each stage.
Our iritellectual debts follow the history of our project. During our dis-
sertation days at Stanford, we joined with fellow doctoral candidates John
Lad and Michael Sukale and Professors F¢llesdal, Goheen, and Hintikka in
an informal seminar on phenomenology that met weekly from June of 1969
through March of 1970. During the summers of 1973 and 1974 we regrouped
in another informal seminar on phenomenology, meeting weekly at Stanford
and sometimes Berkeley, the regular participants being ourselves, Hubert
Dreyfus, Dagfmn F¢llesdal, Jane Lipsky McIntyre, Izchak Miller, and, in
1974, John Haugeland. More recently, we enjoyed discussions and presented
some of our results at the 1980 Summer Institute on Phenomenology and
Existentialism, on 'Continental and Analytic Perspectives on Intentionality'
(held at the University of California, Berkeley, directed by Hubert Dreyfus
and John Haugeland, under the auspices of The Council for Philosophical
Studies with support from the National Endowment for the Humanities). We
are grateful to all the above-mentioned philosophers for intellectual inspira-
tions of many forms. We should also like to thank our students and colleagues
over the years and our audiences at various institutions and conferences for
their responses to presentations of ideas that were taking shape for the pre-
sent book.
The book is for the most part thoroughly co-authored, with both content
and wording being the result of inextricably joint efforts at several stages of
writing. The only exceptions are as follows. Section 2.3 of Chapter II derives
from Smith, 'Meinongian Objects', Grazer Philosophische Studien 1 (1975).
Sections 3.2 and 3.3 of Chapter N derive from McIntyre, 'Intending and
Referring: Some Problems for HusseTl's Theory of Intentionality', in Husserl,
xi
xii PREFACE
NOTE
On some prior occasions we have referred to the present book under an earlier title,
Intentionality and Intensions: Husserl's Phenomenology and the Semantics of Intentional
Modalities. That title gave way to another more accurately indicating the focus of the
f"mished work.
INTRODUCTION
2. Object-Theories of Intentionality 47
2.1. Mind-Dependent Entities as Objects of Intention: An Interpretation of
Brentano's Early Theory 47
xix
xx ANALYTICAL TABLE OF CONTENTS
1. Meaning and Possible Experience: The Turn to Husserl's Notion of Horizon 227
1.1. The "Indeterminacy" in Intentions of Transcendent Objects 227
1.2. Husserl's Notions of Object-Horizon, Act-Horizon, and Manifold 229
1.3. Horizon-Analysis as a New Method of Phenomenological Analysis 233
Our purpose in this chapter is to introduce briefly some of the main topics
with which we shall be concerned throughout the rest of the book. These
topics fall into two major categories: (1) metaphysical and ontological prob-
lems concerning the intentionality of acts of consciousness and the status of
the objects toward which intentional phenomena are directed; and (2) logical
and semantic problems concerning the behavior of linguistic expressions in
intensional contexts, specifically, in sentences attributing intentional phe-
nomena to persons. Here we shall concentrate on describing in as theory-
neutral a way as we can the characteristics of mental phenomena that consti-
tute their intentionality and the characteristics of act-sentences that constitute
their intensionality, and we shall suggest ways in which intentionality and
intensionalityare related. It will be the task of subsequent chapters to develop
a theoretical framework within which these characteristics of acts and act-
sentences can be systematically explained and understood.
1.1. Intentionality
The philosophical foundations of phenomenology, for the most and better
part, are the work of one man, Edmund Husserl. And according to Husserl,
phenomenology - both as a philosophical theory of consciousness and as
applied description of specific forms of human consciousness - simply is
the theory of intentionality, pure and applied. "Intentionality", he says,
"expresses the fundamental property of consciousness; all phenomenological
problems ... are classified according to it" (Ideas, § 146, p. 357).
In Ideas, §84, HusserI'defines the intentionality of consciousness as "the
peculiarity of experiences [Erlebnissen] 'to be the consciousness of some-
thing'" (p. 204). The term 'intentionality' derives from the Latin verb 'in-
tendere', meaning "to point to"; and in the sense of being the consciousness
of something, each intentional event of consciousness can be said to "point
to" or to "be directed toward" something. Thus, intentionality is often
characterized as the "directedness" of consciousness. Husserl himself says
2 CHAPTER I
We shall avoid the term 'mental phenomenon' entirely, and shall talk instead of 'inten-
tional experiences' [intentionalen Erlebnissen] . ... The qualifying adjective 'intentional'
names the essence common to the class of experiences we wish to Rlark off, the peculiar-
ity of intention, of relating to what is objective.... As a briefer expression ... we shall
use the term 'act'. (LI, V, § 13, p. 562; with trans. changes.)
Any mental phenomena that are not intentional are thus not to be called
"acts". And where mental phenomena are intentional, Husserl excludes from
the notion of an act any "extra-experiential", or "non-phenomenological",
elements that may be connected with them. By an act Husserl means just that
component of an intentional event of consciousness that the subject himself
can discern by "reflecting" on his experience, excluding empirical facts about
the intended object and its de facto relation to the subject. Hence, an act is
just what we might call the "experiential" component of an intentional event,
"purified" (as Husserl says) of presumptions concerning its "interlacing with
nature" (cf.Ideas, §§38, 50, 51).
This point is particularly important to keep in mind when considering
perception, Husserl's own paradigm of intentionality. In ordinary usage
perceptual notions often encompass much more than ~hat Husserl wants to
4 CHAPTER I
It is a serious error to draw a real [reell] distinction between ... "intentional" objects,
on the one hand, and "transcendent", "actual" objects, which may correspond to them,
on the other.... The intentional [or intended] object of a presentation is the same
as its actual object, and, when appropriate, as its external object . ... The transcendent
object would not be the object of this presentation, if it was not its intentional object.
(Appendix to §11 and §20, pp. 595-96;with trans. changes.)
Husser! maintains that there are as many kinds of objects of acts as there
are kinds of entities, and his own ontology is notoriously rich. Acts such as
6 CHAPTER I
Smith's seeing the Blarney Stone or his remembering his high school French
teacher are directed to concrete, "natural" individuals (physical objects or
persons). Others, such as Smith's contemplating chastity, imagining the color
red, or thinking of the number seven, seem to take as their objects abstract
entities or (as Husserl calls them) essences. There are also "propositional"
acts, such as Smith's remembering that Mont Blanc is the highest of the Alps
or judging that seven plus five equals twelve, whose objects are more complex
entities, perhaps propositions or (as Husserl maintains) states of affairs
(Sachverhalten). And, according to Husserl, there are second-order acts of
reflection that take as their objects non-reflective acts and other constituents
of consciousness; special among second-order acts are acts of "phenomenolog-
ical reflection", directed to noernata (which are abstract meaning-entities) and
to noeses and hyletic data (the temporal constituents of "phenomenologically
reduced" acts).
According to Husserl, then, the objects of acts may be concrete or abstract;
particular or universal; relatively simple or complex; and either ''immanent''
mental entities or events, occurring as a part of the stream of consciousness
in which they are intended, or ''transcendent'' external entities, existing in-
dependently of their being intended. Putting aside differences of ontological
detail, our own view agrees with Husserl's here: not all acts intend objects
of the same sort; and, in general, the objects of acts are to be found among
whatever entities we should have had to recognize anyway, independently of
any special considerations of intentionality.
Our own discussion will concentrate on acts that involve natural individuals
in rather simple ways. The problems involved in characterizing and under-
standing the features that are special to intentional relations between acts
(or egos) and objects generally are all clearly evident in acts involving natural
individuals. Consequently, by focusing on this narrower class of acts we may
hope to avoid overshadowing Husserl's general theory of intentionality with
special and controversial features of his rather liberal ontology. Husserl's own
paradigm of the perception of physical objects will thus be prominent in our
discussion, although perceptual acts exhibit some peculiarities of their own
(due, primarily, to the special role of sensation, or hyle, in perception). And
where our concern is with "propositional" acts, our emphasis will be on acts
directed to concrete states of affairs involving physical objects or persons.
The knife is the object about which we judge or make a statement, when we say that
the knife is on the table; the knife is not, however, the primary or full object of the
judgement, but only the object of its subject. The full and entire object corresponding
to the whole judgement is the state of affairs [Sachllerhalt] judged.... The wish that
the knife were on the table, which coincides (in object) with the judgement, is concerned
with the knife, but we don't in it wish the knife, but that the knife should be on the
table, that this should be so. (V, § 17, pp. 579-80.)
INTENTIONALITY AND INTENSIONALITY 9
Since there are in these cases two rather different senses in which proposi-
tional acts can be said to have objects, it will be helpful to make a distinction
terminologically between what is primarily intended in a propositional act
and what is secondarily intended in such an act. The "full and entire" object
primarily intended in a propositional act we shall (at least for now) suppose
with Husser! to be a state of affairs. For propositional acts directed to con-
crete states of affairs involving natural individuals, the object secondarily
intended will be the physical object or person about which something is
judged, believed, perceived, etc. We should note, however, that not all propo-
sitional acts are "about" individuals in this way. Among propositional acts
lacking such a secondary relation to an individual are judgments directed to
"general" states of affairs (such as Smith's judging that all men are mortal)
and also certain "existential" judgments (such as Smith's judging that some
men are Cretans).
Our focus on problems involved in the intending of natural individuals
will lead us to emphasize (though perhaps no more than Husser! does) the
relation of propositional acts to their "secondary" objects. Husser! himself
seems to consider these secondary relations to individuals about which some-
thing is intended, as well as the primary relations of propositional acts to
states of affairs, to be relations of intending. In fact, he sometimes takes
secondary intendings of individuals in propositional acts to be on a par with
direct-object intendings of individuals, choosing to speak of perceiving or
thinking of an individual and judging or wishing that something be true of
it as simply different modes of intending the very same object. In Logical
Investigations, V, for example, Husser! says:
To think of an object, e.g., the Schloss at Berlin, is to be minded in this or that descrip-
tively determinate fashion. To judge about this Schloss, to delight in its architectural
beauty, or to cherish the wish that one could do so, etc., etc., are new experiences
[Erlebnisse], phenomenologically characterized in new ways. All have this in common,
that they are modes of objective intention, which cannot be otherwise expressed in
normal speech than by saying that the Schloss is perceived, imagined, pictorially pre-
sented, judged about, delighted in, wished for, etc., etc. (§ 11, pp. 559-60; with trans.
changes.)
We shall indeed find that the same general features characterizing direct-
object intendings of physical individuals also characterize these secondary
intendings of individuals in propositional acts.
10 CHAPTER I
Our discussion in the rest of this chapter assumes with Husserl, and against
theories of ''intentional objects", that in most cases the intentionality of
consciousness does relate persons to the ordinary entities to which acts seem
to be directed and not to unusual ''intentional objects". On this assumption,
what distinguishes intentional relations from others is not that the objects
of intentional relations belong to some unusual ontological category; it is
rather that intentional relations themselves exhibit characteristics that, in
comparison with ordinary, non-intentional relations, appear to be metaphysi-
cally anomalous. In particular, we shall see that intentional relations, unlike
ordinary relations, are independent of the existence of objects to which they
relate conscious subjects and are in each case dependent on a particular
conception of the intended object.
Some philosophers might say that, since intention has the peculiarities it
does, it should not be considered a relation at all. Husserl himself contrasts
intentional relations with "real" relations having "actual" existence: of per-
ception he says that with phenomenological reduction "the actual [wirkliche]
existence of the real [realen] relation between perception and perceived is
suspended; and yet a relation between perception and perceived ... is clearly
left over" (Ideas, §88, p. 220; cf. §36, p. 80). So long as we are careful to
avoid being misled by terminology, though, we can continue to speak with
Husserl of ''intentional relations", understanding these "relations" to be of a
special and unique sort.
We tum now to a closer characterization of the special nature of inten-
tional relations. As we do so we should perhaps emphasize that here we shall
only be describing the main features that distinguish these relations from
others; ultimately, of course, an explanation of how and why intention has
these features will have to be a main concern of any theory of intentionality.
Unlike the relations of being taller than, being to the left of, riding, and
kicking, intentional relations need not relate persons to existing objects.
Smith cannot ride a pink elephant or be taller than Godot, for there are no
such entities; yet Smith's acts of seeing a pink elephant and waiting for (or
anticipating) Godot are equally as intentional as those of seeing Secretariat
and waiting for the postman. After all, to see a pink elephant is not to see
nothing, nor is waiting for Godot the same as waiting for nothing at all. And
even such acts as seeing Secretariat and waiting for the postman do not de-
pend for their intentionality on the de facto existence of their objects: should
it turn out that Secretariat and the postman do not (and never did) exist,
these acts would not thereby prove non-intentional. Husser! thus continues
the passage in Logical Investigations, V:
If, however, the intended object exists, nothing becomes phenomenologically different.
What is given to consciousness is essentially the same, whether the presented object
exists, or is fictitious, or is perhaps completely absurd. I think of [the god] Jupiter as
I think of Bismarck, of the tower of Babel as I think of Cologne Cathedral. ... (§ 11,
p. 559; with trans. changes.)
"Diogenes sits in his tub" is concerned with a relation between Diogenes and his tub.
Syntactically, at least, "Diogenes looks for an honest man" is similar: Diogenes' quest
seems to relate him in a certain way to honest men. But the relations described in this
and in ... other psychological statements, if they can properly be called "relations",
are of a peculiar sort. They can hold even though one of their terms, if it can properly
be called a "term", does not exist. It may seem, therefore, that one can be "intentionally
related" to something which does not exist. 10
... One is at first tempted to interpret the situation simply: ... an act is ... unam-
biguously determined by its qualitative character and by the object it is to intend. This
seeming obviousness is, however, delusive.... Even if quality and objective direction
are both fIxed at the same time, certain variations remain possible. Two identically
qualified acts, e.g., two presentations, may appear directed ... to the same object,
without coinciding in their complete intentional essence. The presentations "equilateral
triangle" and "equiangular triangle" differ in content, though both are directed ... to
the same object. They present the· same object, although "in a different way [Weisel".
( § 20, p. 588; with trans. changes. Our emphasis.)
16 CHAPTER I
To have the same presentation means, but does not mean exactly the same as, having
a presentation of the same object. The presentation I have of Greenland's icy wastes
certainly differs from the presentation Nansen has of it, yet the object is the same ....
We have the same presentation of a thing, when we have presentations in which the
thing is not merely presented, but presented as exactly the same . ... The same holds
in regard to other species of acts. Two judgements are essentially the same judgement
when ... everything that would pertain to the judged state of affairs according to the
one judgement, and nothing else, must also pertain to it according to the other. (§ 21,
pp. 590-91; with trans. changes. Our emphasis.)
pp. 712-14; Ideas, § §41-44; CM, §28). More generally, and more to our
present point, the conception under which a transcendent object is intended
(whether "intuitively" - i.e., with evidence, as in perception - or not) will
itself be "incomplete", characterizing the object in some respects but not in
all. The properties such an object is intended as having will always be far
fewer and often less specific than the properties the object actually has; thus,
a transcendent object, as intended in a given act, is always partly "indeter-
minate" - i.e., indeterminately characterized by the conception under which
it is intended. Hussed thus says:
We must distinguish ... between the object, as it is intended [so wie er intendiert
ist], and simply the object which is intended. In each act an object is "presented"
as determined in this or that manner .... [There can be attributed] to the identical
presented object objective properties which are not at- all in the scope of the intention
of the act in question; e.g., various new presentations can arise, all claiming ... to be
presenting the same object. In all of them the object which is intended is the same,
but in each of them the intention [intention] is different, each means [meint] the
object in a different way [Le., under a different conception]. E.g., the presentation
"German Emperor" presents its object as an Emperor, and as Germany's Emperor.
The man himself is the son of the Emperor Frederick III, the grandson of Queen
Victoria, and has many other properties here neither named nor presented. With respect
to a given presentation, one can therefore quite consistently speak of the intentional
and extra-intentional content of its object; one can also find other suitable, non-tech-
nical expressions that would not lead to misunderstandings, e.g., 'what is intended of
the object' [das Intendierte vom Gegenstande]. (LI, V, §17, pp. 578-79; with trans.
changes.)
one must know in order for one's conception of its identity to count as
knowing who (or which) it is will be a relative and pragmatic matter: what
counts as individuation in one context will be only partial in that context,
and it may count scarcely at all in another.
There are other ways in which defmiteness of direction may be achieved,
even in cases in which there is virtually no individuation of an object - where
one has almost no conception of its identity. For example, an act may be
definitely directed by virtue of a subject's perceptual "acquaintance" with
a particular object. Even the simplest perception of an individual, no matter
how meager one's knowledge of it may be, will provide a minimal sort of ftx
on which individual it is. For, at the very least, an object is given in an act of
perception as this object, the one here before me now. Let us say that such
direct-object acts of perception, as well as propositional acts "about" objects
given directly in perception, are perceptually definite.
Perceptual defmiteness seems due to the fact that perception is a species
of what HusserI calls "intuition" (Anschauung), providing one a direct and
immediate acquaintance with the perceived object. HusserI takes other sorts
of acts - apparently including direct-object acts of memory and imagination,
as well as "eidetic" acts (directed to "essences") and acts of reflection - to
be also intuitional. We might accordingly broaden the notion of perceptual
defmiteness so that it would also apply to these acts and to any others that
present their objects directly and intuitionally. Let us say that all such in-
tentions, and also propositional acts "about" objects given intuitionally in
such intentions, are intuitionally definite, or definite by virtue of acquaint-
ance. IS
We should note explicitly that intuitional deftniteness and individuative
defmiteness need not go hand-in-hand. Acts that are intuitionally defmite
may be either individuativcly indefinite - as when Smith sees (or remembers)
the little man with the big cigar but has no further conception of his identity,
of who he is - or individuatively defmite - as when Smith sees (or remem-
bers) him as being his long-lost uncle from Detroit. And, of course, acts that
are not intuitional will not be intuitionalIy defmite, whether they be individ-
uatively defmite or indeftnite.
It is rather surprising to ftnd that HusserI virtually ignores indefmite in-
tentions in his theory of intentionality. He seems to take it to be a general
truth about intention that an act's conceptual content not only prescribes
properties its object is intended as having but also, in some sense, determines
which object it is - in HusserI's words, "makes the act's object count as this
one and no other" (LI, V, §20, p. 589; with trans. changes). That HusserI
INTENTIONALITY AND INTENSIONALITY 21
3.1. Intensionality
Our attention so far in this chapter has been focused on metaphysical prob-
lems concerning the nature of intentional relations and the ontological status
of their objects. Similar problems have appeared in Twentieth-Century ana-
lytic philosophy in a different guise, namely, as logical and semantic problems
concerning the reference of expressions in so-called intensional contexts, such
as 'Smith believes that ~_ _ '. Our purpose in Part 3 is to describe these
problems of intensionality and to relate them to the metaphysical problems
of intentionality that we have already described. Thus, we focus here, not on
acts of consciousness themselves, but on "act-sentences", i.e., sentences used
to describe or report acts of consciousness.
The term 'intensional', as we here apply it to linguistic constructions,
contrasts with the term 'extensional'. In semantic theory an expression is
said to be extensional if and only if its extension is a function of the exten-
sions of its semantically significant parts. 16 The "extension" of an expression
is what is sometimes called its referent or denotation, what an expression
stands for. We may assume Rudolf Carnap's now-standard definition of
'extension': the extension of a Singular term (e .g., 'Socrates') is the individual
to which the term refers, the extension of a predicate (e.g., 'wore sandals')
is the class of individuals of which the predicate is true, and the extension
of a sentence is its truth-value. 17 The sentence 'Socrates wore sandals' is an
extensional sentence: its truth-value depends only on whether the individual
to whom 'Socrates' refers is a member of the class of individuals of whom
'wore sandals' is true; hence, its extension (i.e., its truth-value) is a function
of the extensions of its components, 'Socrates' and 'wore sandals'.
22 CHAPTER I
The sentence 'It is not the case that Socrates wore sandals' is also exten-
sional. Its truth-value is determined by the truth-value of its component
sentence 'Socrates wore sandals', which is, in turn, determined by the exten-
sions of the components 'Socrates' and 'wore sandals'; thus, the extension
of the whole sentence is a function of the extensions of these smaller com-
ponents. Indeed, the result of preftxing any extensional sentence with the
expression 'It is not the case that' will be a sentence that is also extensional.
For this reason, the context 'It is not the case that _ _ ' is itself said to
be extensional. Generally, a context is an extensional context if and only if
the result of·applying it to any extensional expression 'e' is an expression
whose extension is a function of the extensions of the semantically signiftcant
parts of 'e'. Briefly, then, extensional contexts are those that preserve exten-
sionality.
Any construction that is not extensional is said to be intensional. 18 An
intensional sentence, then, is a sentence whose truth-value is not determined
by the extensions of its semantically significant parts. The sentence 'It is
necessary that Socrates wore sandals' is such an intensional sentence. Its
truth-value is not determined by the truth-value of its component sentence
'Socrates wore sandals' (it is not settled simply by determining that Socrates
is a member of the class of individuals who wore sandals), and so its truth-
value is not a function of the extensions of the smaller components 'Socrates'
and 'wore sandals'. The failure of extensionality here is due to the expression
'It is necessary that': prefixing this expression to an extensional sentence 'p'
may result in a sentence whose truth-value is not a function of the truth-value
of 'p' or of the extensions of the semantically signiftcant parts of 'p' (we say
"may result" since 'It is necessary that p' is always false when 'p' is false).
Since it is then not extensional, the context 'It is necessary that _ _ ' is
an intensional context.
The paradigmatic, and most thoroughly studied, species of intensional
construction is that involving the sentential operators or modifters 'It is
necessary that' and 'It is possible that' (or 'Necessarily' and 'Possibly').
Necessity and possibility are themselves called "modalities". These operators
are accordingly called modal operators or, sometimes, modalities, and the
contexts 'It is necessary that _ _ ' and 'It is possible that _ _ ' are called
modal contexts.
The modal operators have the effect of invalidating, in ways we shall be
discussing, two deeply entrenched rules of logical inference, the principles
of substitutivity of identity and of existential generalization. Both these prin-
ciples hold when applied to singular terms occurring in extensional contexts;
INTENTIONALITY AND INTENSIONALITY 23
(SI) From a statement of identity 'a =b' and a sentence' ... a ... ' we
may infer ' ... b .. .', where 'a' and 'b' are singular terms and
' ... b .. .' results from' ... a .. .' by substituting 'b' for 'a' in at
least one of its occurrences.
According to (SI), if two singular terms refer to the same entity, then they
can be substituted for one another in any sentence without changing the
truth-value of the sentence. Examples of valid applications of the principle
are easy to find. If the expressions 'the author of Intention' and 'the woman
Smith saw at the tobacconist's' refer to the same person, then the identity
statement
(1) The author of Intention = the woman Smith saw at the tobac-
conist's
is true. Hence, in accordance with the principle (SI), substitution of the latter
term for the former in the true sentence
value, will not affect the truth-value of the sentence in which the replacement
is made. 24
Nevertheless, the principle of substitutivity of identity may fail when
co-referential terms are interchanged within a sentence about someone's acts
of consciousness. Even if both (1) above and
(4) Smith believes that the author of Intention is a philosopher
are true, the truth of
(5) Smith believes that the woman he [Smith] saw at the tobac-
conist's is a philosopher
does not follow. For if Smith is unaware or fails to believe that the author of
Intention and the woman he saw at the tobacconist's are one and the same
person, he may also fail to believe that what is true of the author is thereby
true of the woman at the tobacconist's. Again, from 'Balzano was the source
of Husserl's most important insights' and 'Professor Grossmann believes that
Twardowski was the source of Husserl's most important insights' one is not
permitted to infer 'Professor Grossmann believes that Twardowski was
Bolzano', the principle of substitutivity of identity notwithstanding. The same
holds for other propositional act-contexts, using 'remembers that', 'sees that',
'hopes that', 'expects that', etc. And it marks all such contexts as intensional.
The failure of (SI) in act-contexts is just the logical manifestation of what
we earlier characterized as the 'conception-dependence' of intentional re-
lations. We noted (in Section 2.4) that in order to specify an intentional
relationship it is not sufficient merely to indicate the subject, the species of
intention achieved in an act, and the object intended, for not every way of
specifying the intended object will be appropriate to the particular conception
under which it is intended. If an act-sentehce is to attribute the right intention
to a subject, a descriptive singular term referring to the object of the inten-
tional relation must not only refer to the right object; it must also describe
the object in the right way, i.e., describe it "as" conceived in the act by the.
subject. How (under what description) the object is referred to, as well as
which object is referred to, is in this way relevant to the truth of an act-
sentence. Consequently, when a singular term denoting the object of an
intentional relation is replaced in an act-sentence by another expressing
different descriptive content, even if it has the same referent, the result is
an act-sentence attributing a different act to the subject. And so one of these
sentences may be false even though the other is true.
(4) attributes to Smith a propositional act of intending something "about"
INTENTIONALITY AND INTENSIONALITY 27
and
Similarly, from 'Smith's first logic professor is the man who drove Smith's
taxi' and 'Smith remembers his first logic professor' one cannot infer 'Smith
remembers the man who drove his taxi'.
In general, we conclude, wherever verbs of intention are used in the
experiential sense - i.e., as descriptive of "conception-dependent" intentional
phenomena - they create intensional contexts in which substitutivity of
identity fails as a valid rule of inference.
(EG) For any sentence ' ... a .. .' in which a singular term 'a' occurs,
from ' ... a .. .' may be inferred '(3x) ( ... x ... )'.
(yVe use '(3x)' as the existential quantifier 'There is something x such that'.)
The principle (EG) allows one to infer, for example, from 'Smith visited the
Louvre' the existential sentence '(3x) (Smith visited x)" i.e., 'Something is
such that Smith visited it' or, more colloquially, 'Smith visited something'.
According to the principle of existential generalization, any singular term
occurring in a sentence may in this way be replaced by a variable bound by
an existential quantifier standing in front of the resulting formula.
This principle, too, seems to capture a basic intuition about the use of
referring expressions: all it says is that if a predicate is true of an entity
referred to by a term, then something exists of which the predicate is true.
And it is a clearly valid principle for extensional sentences, whose truth or
falsity is dependent on the referents of any singular terms that occur in them.
Nevertheless, the principle of existential generalization fails for terms in
intentional contexts. There are, in fact, two sorts of circumstance in which
it fails.
The most obvious circumstance in which (EG) may fail for intentional
contexts is when a singular term in an act-sentence fails to refer to any
existent entity. From the true sentence
INTENTIONALITY AND INTENSIONALITY 29
(9) Ponce de Leon hoped that the fountain of youth would be found
in Florida
But (10) seems to be false: since there was (and is) no fountain of youth, it
cannot be the entity whose existence makes (10) true, and no other entity
that does exist seems to play the appropriate role in Ponce de Leon's inten-
tion.
Existential generalization also fails for singular terms in direct-object act-
contexts. Since there is no bogey man, from the true sentence
of hoping fearing, and seeing - can hold when what is intended does not exist.
Given that they can, however, the failure of (EG) follows as a consequence.
There is a second sort of circumstance in which (EG) fails for intentional
contexts. It has nothing to do with the failure of intended objects to exist,
and so is quite different from the case we have just considered. It deserves
special attention in its own right.
(15) Mother Hubbard hopes that her most intelligent child will be-
come famous
i.e., 'Someone is such that Mother Hubbard hopes that he or she will become
famous'. In order for such a sentence involving quantifying-in to be true,
there must be some particular individual of whom the quantified formula is
true. But this condition may fail to be fulfilled and, so, (16) may be false,
even though (15) is true. The problem in this case is that, although there may
exist such an individual as Mother Hubbard's most intelligent child, the act
of hoping attributed Mother Hubbard in (15) may not be a definite intention
"about" that, or any other, particular one of her children. The truth of (15)
does not presuppose that Mother Hubbard has come to have an opinion as to
which of her children is in fact the most intelligent (and if she has several and
they are sufficiently young, she quite likely will not have). If she has not - if
she has no "conception of the identity" of her most intelligent child - then
of none of her children will it be true to say that Mother Hubbard hopes that
INTENTIONALITY AND INTENSIONALITY 31
(l7a) Sherlock believes that the murderer, whoever he is, used bella-
donna
and
(l7a') Sherlock believes that (3x) (x ::: the murderer and x used bella-
donna).
NOTES
Hintikka, The Intentions of IntentioTll1lity and Other New Models for Modalities (D.
Reidel, Dordrecht, 1975), pp. 43-58.
7 There are some apparent exceptions, such as believing the President or believing in
ghosts. But these seem to be elliptical descriptions of propositional acts, e.g., believing
that what the President says is true or believing that there are ghosts. Believing in
the President or believing in God, however, are at least more complicated and may
constitute genuine counterexamples. Judging a track meet seems only superficially
direct-object, consisting in making judgments that certain propositions are true (though
here certain "pragmatic" factors are relevant: for example, the person "judging" must
have the authority to declare a winner).
8 Bertrand Russell, An Inquiry into Meaning and Truth (George Allen & Unwin, Lon-
don, 1940), p. 210.
9 Brentano (Note 1 above), p. 88.
10 Roderick M. Chisholm, Perceiving: A Philosophical Study (Cornell University Press,
Ithaca, N.Y., 1957), pp. 169-70.
11 See LI, I, § 12, where Husser! uses the terms 'the victor at Jena' and 'the vanquished
at Waterloo' to illustrate how linguistic expressions with different meanings can refer to
the same entity. This point about language parallels our point that one and the same
entity can be conceived in different ways. We shall later see that the "conception" under
which an entity is intended is explicated by Husser! in terms of an act's also having a
meaning (or "noema"), which he says is a generalization of the notion of linguistic
meaning (see Chapter IV).
12 We take Husserl to have a rather objective notion of states of affairs, as complex
transcendent entities consisting, in the simplest case, of a property and a concrete in-
dividual appropriately coupled (cf. LI, V, § 33, p. 623; VI, §44, p. 783). On this view,
the state of affairs that the vanquished at Waterloo has a certain property is the same as
the state of affairs that the victor at Jena has that property, since the vanquished at
Waterloo is the victor at Jena. On a more intensional notion of state of affairs than that
we have attributed Husserl, a state of affairs (of the relevant kind) might consist of a
property and something like an individual concept (e.g., the concept "the vanquished at
Waterloo"). On this latter view, states of affairs involving the vanquished at Waterloo
(or the corresponding concept) would be distinct from those involving the victor at Jena
(or the corresponding concept).
13 Philippa Foot convincingly argued for this case in discussion at UCLA in spring,
1976.
14 For a short discussion of the distinction see Peter Geach, Logic Matters (University
of California Press, Berkeley, 1972), pp. 148-49. The distinction we shall draw is closely
related to (and perhaps underlies) Keith Donnellan's distinction between "referential"
and "attributive" uses of definite descriptions: see his 'Reference and Definite Descrip-
tions', Philosophical Review 75 (1966), 281-304. For its relation to the distinction
between "de re" and "de dicto" sentences of propositional attitude, see Section 3.5
below.
IS Cf. Hintikka's discussion of what he calls "perceptual individuation" in 'On the Logic
of Perception" (Note 6 above) and his more generalized notion of "contextual individua-
tion" or "individuation by acquaintance" in 'Objects of Knowledge and Belief' (Note 6
above). Hintikka's account of the difference between perceptual or contextual individua-
tion and "physical" or "descriptive" individuation is closely related to, but not identical
INTENTIONALITY AND INTENSIONALITY 37
with, the account of the difference between intuitional and individuative definiteness
that we give in Chapter VIII below. Note that Hintikka also uses his distinction to
explicate the difference between what we have called direct-object and propositional
acts.
16 See Russell (Note 8 above), pp. 324-28; and Rudolf Carnap, Meaning and Necesnty,
2nd ed. (The University of Chicago Press, Chicago, 1956), pp. 46-48. The whole of
Chapters I and III of Meaning and Necessity is directly related to our current discussion.
17 Carnap (Note 16 above), pp. 19,26,40.
18 A narrower deftnition of 'intensional' that would make intensional constructions a
sub~lass of non-extensional constructions is perhaps to be preferred here. Indeed, our
actual concern will be with sentences of propositional attitude and, to a lesser extent,
with the logical modalities but not with other non-extensional constructions, such as
quotation-contexts. Within semantic theories that recognize intensions, or meanings, of
expressions as well as extensions, intensional expressions in this narrower sense might
be roughly characterized as those non-extensional expressions whose extensions are
dependent on the intensions, and not merely on the extensions, of their semantically
relevant parts. Carnap defines a notion of intensionality that is in this spirit but, since
the propositional attitudes tum out to be neither extensional nor intensional on his
definition, it is too narrow for our purposes (see Carnap (Note 16 above), pp. 48,53-
55).
19 Hintikka is perhaps the chief proponent of this approach to act-sentences. See, for
example, his 'Semantics for Propositional Attitudes', in Models for Modalities (Note
6 above), pp. 87-111, reprinted in Reference and Modality, ed. by Leonard Linsky
(Oxford University Press, London, 1971). When we call act-operators 'modalities' we
do not intend, at this point, to suggest any particular analysis of act-sentences; rather,
we intend only to call attention to the logical and semantic similarities between act-
contexts and modal contexts strictly so~ed. Later on (in Chapter VII), however, we
shall consider the same sort of "possible worlds" analysis of intentional modalities that
Hintikka does.
20 It is not quite appropriate to call these direct-object act-sentences intentional
"modalities" in our extended sense, however, since they are not constructed from
sentential operators. But even this qualification can be dropped if these direct-object
constructions can be rendered propositional in form, perhaps along the lines of Hintikka's
analysis in 'Different Constructions in Terms of the Basic Epistemological Verbs', in
The Intentions of Intentionality (Note 6 above), pp. 1-25; 'On the Logic of Perception'
(Note 6 above); and 'Objects of Knowledge and Belief' (Note 6 above).
21 Chisholm has attempted to enunciate additional logical characteristics of intentional
contexts, over and above their intensionality, that would distinguish them not only from
extensional contexts but from the non-intentional modalities as well. See his Perceiving
(Note 10 above), Ch. 11; 'Intentionality', in The Encyclopedia of Philosophy, ed. by
Paul Edwards (Macmillan & The Free Press, New York, 1967), IV, pp. 203-204;
'Brentano on Descriptive Psychology and the Intentional', in Phenomenology and
Existentialism, ed. by Edward N. Lee and Maurice Mandelbaum (The Johns Hopkins
Press, Baltimore, 1967), pp. 21-23; 'Notes on the Logic of Believing', Philosophy and
Phenomenological Research 24 (1963-64), 195-201, reprinted in Intentionality, Mind,
and Language, ed. by Ausonio Marras (University of Illinois Press, Urbana, 1972); and
'On Some Psychological Concepts and the "Logic" of Intentionality', in Intentionality,
38 CHAPTER I
Minds, and Perception, ed. by Hector-Neri Castaneda (Wayne State University Press,
Detroit, 1967), pp. 11-35. Chisholm's criteria of intentionality have been the subject
of much discussion: see, for example, the essays in Marras, Part I; also the symposium
by J. O. Urmson and L. Jonathan Cohen on 'Criteria of Intensionality', Proceedings of
The Aristotelian Society, supplementary vol. 42 (1968), 107-42; and Mohanty (Note 3
above), pp. 25-35.
22 Cf. Romane Clark's distinction between "Scribe" and "Agent" attributions of
propositional attitude, in his 'Comments' (on Hintikka's 'On the Logic of Perception'),
in Care and Grimm (Note 6 above), pp. 176-77; cf. also Charles Taylor's comments on
"intentional description" in his The Explanation of Behaviour (Routledge & Kegan Paul,
London, 1964), pp. 58-59.
23 See Carnap (Note 16 above), pp. 133-36; Russell, 'On Denoting', Mind 14 (1905),
reprinted in Russell's Logic and Knowledge, ed. by Robert C. Marsh (G. P. Putnam's
Sons, Capricorn Books, New York, 1971), pp. 47-48; and Willard Van Orman Quine,
Word and Object (The M.LT. Press, Cambridge, Mass., 1960), pp. 141-51.
24 These points are carefully made in Dagfmn F~llesdal, 'Quine on Modality', Synthese
19 (1968), 152-53; reprinted in Words and Objections, ed. by Donald Davidson and
Jaakko Hintikka (D. Reidel, Dordrecht, 1975).
25 Some philosophers seem to believe that verbs of perception are used only non-inten-
tionally in direct-object constructions. Chisholm, for example, in Theory of Knowledge
(Prentice-Hall, Englewood Cliffs, N.J., 1966), pp. 10-11, sorts the intentional (or inten-
sional) use of perceptual verbs with the propositional construction and the non-inten-
tional (or extensional) with the direct-object construction (cf. Perceiving (Note 10
above), pp. 142ff). And Armstrong (Note 4 above), p. 228, suggests that direct-object
perceptual constructions are used when one wants to indicate the object of a perceptual
experience without committing oneself about what the subject perceives it "as". Our
own examples seem to us to indicate that ordinary usage is not univocal on these mat-
ters, though faithfulness to ordinary language is not really our major concern here.
26 See, for example, Quine, 'Quantifiers and Propositional Attitudes' ,Journal of Philos-
ophy S3 (1956), 177-87, reprinted in Quine's The Ways of Paradox (Random House,
New York, 1966), also reprinted in Linsky (Note 19 above); David Kaplan, 'Quantifying
In', Synthese 19 (1968), 178-214, reprinted in Davidson and Hintikka (Note 24 above)
and also in Linsky; and Hintikka, 'Quine on Quantifying In: A Dialogue', in The Inten·
tions of Intentionality (Note 6 above), pp. 102-36.
27 Names of fictional characters (e.g., 'Sherlock Hoimes'), whether in intensional or
extensional contexts, pose special problems in the philosophy of language. These prob-
lems relate to our discussion only insofar as intentions of or about fictional characters
are illustrative of the existence-independence of intention and, hence, only where names
of such characters occur within intentional contexts. The name 'Sherlock', which occurs
in extensional position as the grammatical subject of (17) and (18), as well as the name
'Mother Hubbard' as it occurs in (15) and (16), should therefore be understood as
there naming actual persons, not the fictional characters they are sometimes used to
name. We offer the following general disclaimer: all names occurring in this book as
grammatical subjects of act-sentences are names of actual persons; any resemblance
between these nat'IiICS and the names of fictional characters, living or dead, is purely
coincidental. And we extend the disclaimer to apply also to names occurring within
act-contexts, except where the sentences in which they occur are expressly illustrative
INTENTIONALITY AND INTENSIONALITY 39
of the existence-independence of intention. (No use of the name 'Smith' in any of our
examples refers to either of the authors of this book, however.)
28 See Georg H. von Wright, An Essay in Modal Logic (North-Holland, Amsterdam,
1951), pp. 6-35; and A. N. Prior, Formal Logic, 2nd ed. (Clarendon Press, Oxford,
1962), pp. 209-15. For some of the history of the distinction, see William Kneale,
'Modality de Dicto and de Re', in Logic, Methodology, and Philosophy of Science, ed.
by Ernest Nagel et al. (Stanford University Press, Stanford, California, 1962), pp. 622-
33.
29 Quine, 'Reference and Modality', in From a Logical Point of View, 2nd ed. (Harper
& Row, New York, 1961), p. 148, reprinted in Linsky (Note 19 above).
30 For other discussions of the relations between problems of intensionality and prob-
lems of intentionality see the symposium by Kneale and Prior on 'Intentionality and
Intensionality', Proceedings of the Aristotelian Society, supplementary vol. 42 (1968),
73-106; Carr (Note 5 above), pp. 17 -36; and J. 1. Mackie, 'Problems of Intentionality',
in Pivrevic (Note 5 above), pp. 37-50. See also Prior, Objects of Thought, ed. by Peter
Geach and Anthony Kenny (Clarendon Press, Oxford, 1971), esp. Chs. 4,8, and 9.
CHAPTER II
view about the objects of acts to embody a basic datum, so that any theory
of intentionality that held otherwise would simply have to be rejected as
incompatible with the "phenomenological facts" of experience. Nonetheless,
there are rather alluring theoretical grounds for questioning this assumption
about the objects of consciousness, and some philosophers have been led to
abandon it altogether.
The difficulty with the Husserlian view is this: if we assume that the ob-
jects of acts are just the ordinary entities they seem to be, then the relations
of intending these objects must themselves be considered "relations" of an
unusual sort, different in kind from ordinary relations. For, as our considera-
tions in Chapter I have shown, an "intentional relation" can hold when the
ordinary object that seems to be intended does not exist at all, and it can
hold under certain conceptions of this object while failing to hold under
other conceptions of it. But it is not easy to make sense of such "intentional
relations", and the considerations that led us to suggest such unusual relations
can in fact be interpreted in a different way that avoids them altogether.
Many philosophers who have dealt with the problems of intentionality have
assumed that intention is an ordinary sort of relation and have taken the
existence-independence of intention, in particular, to show that it is our
realistic assumptions about the objects of consciousness that must be aban-
doned. According to these philosophers, the intentionality of an act does not
consist in a peculiar sort of "intentional relation" to an ordinary object, but
in an ordinary relation to a peculiar sort of "intentional object".
The main feature of this second, non-Husserlian approach to the problems
of intentionality is that it attempts to explain the peculiarities of intending
as peculiarities in the objects that acts intend: intentionality is taken to be
unique (and unique to consciousness) not because intentional relations are
relations of an ontologically unique sort, but because the objects to which
they relate persons are entities of a special kind, different from the entities
that normally enter into non-intentional relations. The purpose of a theory of
intentionality, according to this line of thinking, is to explicate the ontologi-
cal type and ontological status of these "intentional objects", as they are
sometimes called, so that intending can be construed as an ordinary relation
to these extraordinary objects. Because theories based on this approach take
the problems of intentionality to be primarily problems about the objects
of consciousness, we shall speak of it as the object-approach to the theory
of intentionality and we shall call the resulting theories object-theories of
intentionality.
42 CHAPTER II
Smith ought to stand in that same relation to his next-door neighbor. But
Smith may not fear his next-door neighbor. Failing to relate Smith to his
next-door neighbor, Smith's intention seems also to fail to relate him to the
phone caller; and there are no other ordinary entities that Smith, in this act,
can more appropriately be said to intend. Barring the view that it is the rela-
tion of intending itself that is onto logically peculiar, one seems forced to
conclude that the object of Smith's act all along was not an ordinary sort of
object but an intentional object.
The conception-dependence of intention also suggests a characteristic of
intentional objects that would serve to differentiate them from concrete
existents: it suggests that intentional objects are themselves, in some sense,
conception-dependent entities. We have seen that Smith's act of fearing the
man who makes the threatening phone calls must be distinguished from his
act of fearing his next-door neighbor. Now, an object-theory of intentionality
will have to make this distinction solely in terms of the objects intended. For,
if intentions are relations of the ordinary sort, then intentions are individ-
uated by specifying their subjects, the specific relations of intending involved
(and, of course, the times at which they obtain), and the objects intended
(cf. Chapter I, Section 2.4). Here, however, the subjects are the same and so
are the relations of intending (and, we may suppose, their times). So, what
would make these acts distinct - given the assumption underlying the object-
approach - would be their directedness to distinct objects. Similarly, a third
act, e.g., Smith's fearing the banker who can refuse him a loan, involving yet
another conception of the same man would not be directed to this man but
to some third intentional object. The objeCts of acts may be thought gen-
erally to vary concomitantly in this way with the differing conceptions the
subjects of acts may have: to each way of conceiving an ordinary, non-"in ten-
tional" object there may correspond a distinct act and, according to object-
theories, each of these will intend a distinct intentional object.
Intentional objects are sometimes further characterized as ''indeterminate''
or ''incomplete'' (or ''incompletely determined") objects. As we noted earlier
(in Chapter I, Section 2.5), any particular conception of an ordinary object
is quite limited in what it prescribes of the object: one's conception need not
be completely specific about some of the properties of an object (as when
one conceives of a tree as a fir, but not as a Douglas fir or as any other specific
kind) and it need not specify others at all (as when the tree is conceived
neither as having moss on it nor as not). Since on the object-approach to
intentionality different conceptions of an ordinary object serve to defme acts
directed to distinct intentional objects, one I'Ilight suppose (thOUgh perhaps
44 CHAPTER II
one need not} that each such intentional object has just those properties
prescribed by the subject's conception in the act. But if so, then every lack
of specificity in the intention of an object becomes a lack of specificity in
the intended object itself. As conception-dependent entities, intentional
objects may in this way come to be thought of as themselves indeterminate,
or incompletely propertied.
In her provocative essay, 'The Intentionality of Sensation: A Grammatical
Feature', G. E. M. Anscombe sums up these features of "intentional objects":
Supposing 'X' to be the name of a real person, the name of something real has to be put
in the blank space in 'X bit _ _ ' if the completed sentence is to have so much as a
chance of being true. Whereas in 'X worshipped _ _ ' and 'X thought of _ _ ' that
is not so .
. . . Let us not be hypnotized by the possible non-existence of the object. There are
other features too: non-substitutability of different descriptions of the object, where it
does exist; and possible indeterminacy of the object. In fact all three features are con-
nected. I can think of a man without thinking of a man of any particular height; I cannot
hit a man without hitting a man of some particular height, because there is no such thing
as a man of no particular height. And the possibility of this indeterminacy makes it
possible that when I am thinking of a particular man, not every true description of him
is one under which I am thinking of him.
I will now define an intentional verb as a verb taking an intentional object; inten-
tional objects are the sub-class of direct objects characterized by these three connected
features.!
2. OBJECT-THEORIES OF INTENTIONALITY
Every mental phenomenon is characterized by what the Scholastics of the Middle Ages
called the intentional (or mental) inexistence of an object, and what we might call,
though not wholly unambiguously, reference to a content, direction toward an object
(which is not to be understood here as meaning a thing), or immanent objectivity. Every
mental phenomenon includes something as object within itself .... 4
The passage suggests several things: that the intentionality of an act consists
in its directedness toward an object; that every act must have an object toward
which it is directed (''include something as object") in order to be intentional;
and that these objects, toward which acts are directed and without which
they would not be intentional, are not ordinary "things" but ''immanent''
objects, having ''intentional (or mental) inexistence" within the mental phe-
nomena that are directed toward them. Thus, it seems that Brentano's theory
is an object-theory and that his specific version takes the objects of conscious-
ness to be mind-dependent entities, immanent entities "existing in" the
events of consciousness in which they are intended.
The interpretation of Brentano's theory as a theory in which the objects
of intention are mind-dependent entities is strongly reinforced by a later
discussion, in which Brentano compares the ontological status of an actually
existing entity A with an ''A which is contemplated or thOUght about":
It is just as true that this A is a contemplated A [ein gedach tes A I as it is that this A is
an actual A, existing in reality. A can cease to be actual and yet continue to be thought
about - so long as the thinking person does in fact think about it. And conversely it
can cease to be thought about - if the person stops thinking about it - and yet continue
to be actual.
In contrasting the A which is contemplated or thought about with the A which is
actual, are we saying that the contemplated A is itself nothing actual or true? By no
means! The contemplated A can be something actual and true without being an actual
A.1t is an actual contemplated A .... 5
Here Brentano is saying that the entity one contemplates or thinks about -
that is, apparently, the object of one's act of contemplating or thinking - has
an ontological status in its own right. Even if a corresponding actual entity
exists in objective reality, this immanent entity that is intended is a different
entity and independent of it. But "the A which is contemplated or thOUght
about", though it is "something actual and true", is not independent of the
SOME CLASSICAL APPROACHES 49
person who contemplates or thinks about it. In the next paragraph Brentano
goes on to say:
We have no right, therefore, to believe that the objects of so-called external perception
really exist as they appear to us. Indeed, they demonstrably do not exist outside of us.
In contrast to that which really and truly exists, they are mere phenomena. (P. 10.)
the man himself. Then these acts will not intend the same object and, for that
reason, they will be different acts. Similarly, taking the conception involved
in an intention to be itself the object of the intention will provide objects
for acts that are "indefmitely directed" (cf. Chapter I, Section 2.6). We noted
earlier that Sherlock's de dicto belief that the murderer used belladonna,
where Sherlock fails to have a conception of the identity of the murderer,
does not seem to be "about" any particular actual individual (cf. Chapter I,
Sections 3.4 and 3.5). Nonetheless, if intention is again taken to be a relation
to an immanent entity, then even this act can be given an object: namely,
Sherlock's "incomplete" and ''indefinite'' conception of "the murderer,
whoever he is''.lO
It seems, then, that a theory taking mind-dependent entities as the objects
of consciousness does meet with some success - at least insofar as its goal is
to construe every instance of intentionality as a relation to an object.
... That which is thought ... is the object of our thoughts - in the one case, horse; in
another, that which is coloured; in another, the soul, and so on. But the term 'horse'
does not signify "contemplated horse", or "horse which is thought about", the term
52 CHAPTER II
'coloured' does not signify "coloured thing which is thought about", and the term 'soul'
does not signify "soul which is thought about". For otherwise one who affirmed or
accepted a horse would be afflllning or accepting, not a horse, but a "contemplated
horse", ... and this is certainly false.l 2
Whether or not there are honest men, Diogenes in his quest was looking for an actual
honest man, not for an intentionally inexisting honest man. If the doctrine of inten-
tional inexistence is true, the very fact that Diogenes was looking for an honest man
implies that he already had the immanent object; hence it could not be the object of
his quest. 13
There may be acts that are directed to mental entities, but such acts are not
among our more frequent and ordinary intentions.
Indeed, as Chisholm's example suggests, even where the requisite ordinary
existent to which an act would seem to be directed fails to exist a mental
entity does not adequately substitute for it. Smith's desire for the woman
of his dreams, even if she does not exist, is not accurately characterized as
directed toward a mental entity: Smith's desire is not for some idea or fig-
ment of his imagination, but for another human being with whom his rela-
tionship need not be purely Platonic. And suppose Smith's dreams come
true: Brentano himself says of promising what also applies to desiring and
wishing, viz., that "it is paradoxical in the extreme to say that a man prom-
ises to marry an ens rationis and fulfills his promise by marrying a real per-
son ... ".14
An appeal to mental entities as the objects of acts also fails to give a
correct account of conception-dependence and of acts that are indefmitely
directed. No matter how "incomplete" or "inadequate" Smith's conception
of the mysterious phone caller may be, it seems clear that the object of
Smith's fear is an entity all too "actual" and "real". Similarly, Sherlock's
indefmite, or de dicto, belief that the murderer (whoever he is) used bella-
donna - if properly "about" any entity at aU - is not about a mental object
but a concrete individual, no less objectively real than the poison he used.
And consider Brentano's example of a promise to marry: even if rendered in-
defmite, or de dicto, as a promise (made to one's parents) to marry someone,
SOME CLASSICAL APPROACHES 53
someday, it can be kept only by going to the altar with some particular and
actual person.
We conclude that even the cases that should be most favorable to a theory
of mental entities as objects of intention - the cases in which it is most dif-
ficult to find appropriate ordinary existents to which acts can be said to be
directed - tum out to be quite implausibly characterized as directed toward
mental entities. If invoking such entities as the objects of acts does not pro-
vide acceptable solutions to these problem cases, then there is little motiva-
tion for invoking them in more ordinary cases, such as veridical perception
and defmite, or de re, belief. What is needed, it seems, is some solution to
the problems of intentionality that will not construe all our intentions as
relations only to the contents of our own minds.
In defending his original formulation of the thesis of intentionality against
an interpretation that elicits all these difficulties, Brentano says that it was
never intended as the claim that acts are directed toward mind-dependent
entities. 'The misinterpretation that leads to that supposed claim, he says,
concerns his use of the term 'immanent'. He now tells us that by an act's
immanent (or intentionally inexisting) object he did mean the object that
is intended in the act but did not mean that this object is a mental entity
created in the act. His purpose in calling it "immanent" was only to avoid
the suggestion that what is intended must in every case exist.
When I spoke of "immanent object", I used the qualification "immanent" in order to
avoid misunderstandings, since many use the unqualified term 'object' to refer to that
which is outside the mind. But by an object of a thought I meant what it is that the
thought is about, whether or not there is anything outside the mind corresponding to
the thought.
It has never been my view that the immanent object [i.e., the object intended] is
identical with "object of thought" (rorgestelltes Objekt). What we think about is the
object or thing and not the "object of thought". If, in our thought, we contemplate a
horse, our thought has as its immanent object - not a "contemplated horse", but a
horse. And strictly speaking only the horse - not the "contemplated horse" can be
called an object.
But the object need not exist. The person thinking may have something as the object
of his thought even though that thing does not exist.! 5
much less foolishness in Brentano's early view than there is in the interpreta-
tion of it as a theory of mind-dependent intentional objects; but if so, it is
because there is simply much less to the view than one would like to find
there.
2.3. Intentional Objects as "Objects Beyond Being": Meinong's Theory of
Objects
Alexius Meinong, who was a student of Brentano's, proposes an object-theory
of intentionality that avoids the specific objections we have raised above
against theories of mind-dependent objects of intention. Meinong (apparently
following Kasimir Twardowski) makes the absolutely crucial distinction
between what is "in the mind" in intention and what is intended. (On
Twardowski's distinction, see Chapter III, Section 2.1, below.) The former
he calls the "content" (Inhalt) of an intentional experience, the latter, the
"object" .16 Thus, Meinong avoids the problem of confusing the object
intended in an act, which is in typical cases an entity "external" to the act,
with any mental phenomena that are produced in or parts of the act. But
although for Meinong the objects of consciousness are generally not mental
entities literally in the mind, they turn out nonetheless to be entities of a
rather odd sort. Their peculiarities are due to the fact that Meinong does not
abandon the object-approach to intentionality: though distinguishing content
and object, Meinong continues to account for the peculiarities of intention
in terms of the objects intended.
like Brentano, Meinong considers the intentionality of acts to consist in
their being related, in the ordinary sense, to objects. And he follows Brentano
in taking the existence-independence of intentional relations to indicate that
the objects of these relations are themselves independent of ordinary exist-
ence: an object need not exist, in the ordinary sense, in order to be intended,
to be an object of consciousness. But Meinong's explanation of this existence-
independence differs from that of the ''mental object" theory we fust at-
tributed Brentano. It is not, as that theory would have it, that the objects
of consciousness must have a sort of existence (or "being") - such as mental
''inexistence'', existence in the mind - different from the existence of
ordinary entities. It is rather, Meinong says, that the objects of consciousness
are "ausserseiend", or "beyond being"; that is, they are independent of being
of any and every sort. 17
Meinong does hold that there are different senses or species of "being".
Concrete, physical individuals may "exist", he says, while abstract individuals
and also facts or states of affairs (which he calls "objectives") may "subsist";
SOME CLASSICAL APPROACHES 55
Meinong calls all these types of entities "objects". But acts may be directed
to objects that do not have being in either of these senses (much less, we may
infer, mental "inexistence"). Pegasus, the largest prime number, and the
earth's being flat are all objects that acts may intend. Yet, these entities do
not exist, as does Secretariat, nor do they subsist, as do the number two and
the earth's being round. And Meinong argues that nothing is to be gained by
assuming that these objects have some third sort of being. 18 Rather, he
concludes, being an object of consciousness simply does not entail having
being. Nor, of course, does it entail having non-being. The realm of "objects"
includes everything to which acts of consciousness may be directed. And
every object, as a matter of fact, either has being (exists or subsists) or has
non-being (neither exists nor subsists). Its status qua object, however, is
independent of its being or non-being. Meinong says:
Those who like paradoxical modes of expression could very well say: "There are objects
of which it is true that there are no such objects". 19
The Object as such ... stands "beyond being and non-being". This may also be
expressed in the following less engaging and also less pretentious way, which is in my
opinion, however, a more appropriate one: The Object is by nature indifferent to being
[ausserseiendl, although at least one of its Objectives of being, the Object's being or
non-being, subsists. 2o
latter object has a property, that of not being snow-capped, that the former
does not. Consequently, when Smith conceives a golden mountain he is
conceiving an object that has neither the property of being snow-capped nor
the property of not being snow-capped. By repetition of the argument, one
can conclude that the object of Smith's act of conceiving a golden mountain
has only the properties of being golden and being a mountain; for every other
property, this object neither has nor does not have the property. Such objects,
Meinong says, are "incompletely determined" or "incomplete" objects.21
An incomplete object never exists or has being in its own right, Meinong
holds, although it may have a derived sort of existence by being "embedded
(implektiert) in" existing complete objects: if the properties of an incomplete
object are shared by one or more existing complete objects, then, Meinong
says, the incomplete object exists "embeddedly" in those complete objects.22
The Meinongian argument for incomplete objects is the key to a Meinong-
ian treatment of conception-dependence, though Meinong seems not to
address that topic explicitly. Meinong holds that all objects that actually
exist (or subsist) are complete objects but that, due to the finite capacities
of the human mind, we can never conceive any such complete object. Acts of
intending complete, existent objects are, strictly speaking, directed to incom-
plete objects that stand proxy for them. When Smith conceives Napoleon, for
example, he cannot conceive Napoleon in all his detail; rather, he conceives
only an incomplete object, e.g., "the vanquished at Waterloo". Smith intends
Napoleon, not in the sense that Napoleon is the object of his intention, but in
the secondary, or indirect, sense that what he does intend - viz., the incom-
plete object "the vanquished at Waterloo" - is "embedded in" the complete
object Napoleon (that is, the properties of this directly intended incomplete
object are properties shared by exactly one existent complete object, namely,
Napoleon). Similarly, what Smith conceives as "the victor at Jena" is also an
incomplete object, and although it, too, is "embedded in" Napoleon it is
distinct from both the incomplete object "the vanquished at Waterloo" and
the complete object Napoleon. According to Meinong's views, then, acts in
which Smith conceives Napoleon as the vanquished at Waterloo and acts in
which he conceives Napoleon as the victor at Jena - both of which can be
said to be acts of intending Napoleon - are in fact directed to distinct (in-
complete) objects and, for that reason, are distinct intentions. 23
There is perhaps some plausibility in the view that acts directed toward
such objects as the woman of one's dreams or a golden mountain have objects
that are non-existent and incomplete. But we have seen that Meinong's claims
do not stop with these cases. Because intention is conception-dependent, and
SOME CLASSICAL APPROACHES 57
because our conceptions are always limited in what they can include concern-
ing ordinary objects, Meinong's object-approach to intentionality forces the
conclusion that the objects ( directly) intended in even our most commonplace
acts are not ordinary, complete existents. The objects of all our intentions
are "indifferent to being", they vary with different conceptions of ordinary
objects, and they are "incomplete": they are, in short, "intentional objects"
par excellence. 24 That Meinong's "objects" deserve to be afforded any legiti-
mate ontological status is not obvious;2s and that they, and only they, are
the objects of consciousness is hardly more plausible than the claim that all
our acts are directed toward the creations of our own minds.
Instead of saying that a person is thinking about a thing, one may also say that there is
something which is the object of his thought. But this is not the strict or proper sense
of is. For the thinker may in fact deny that there is any such object as the object he is
thinking about. Moreover, one can think about what is contradictory, but nothing that
is .;:ontradictory can possibly be said to be. We said above that roundness cannot be said
to be, in the strict and proper sense of the term; that which is round, but not roundness,
may be said to be. And so too, in the present case. What there is in the strict and proper
sense is not the round thing that is thought about; what there is is the person who is
thinking about it. The thing "as object of thought" is a fiction .... What we say can be
expressed in such a way that we do refer to a being in the strict sense of the term -
namely, the thinker who has the thought. And what holds generally for that which is
thought about also holds more particularly, for that which is accepted, that which is
rejected, that which is loved, that which is hated, that which is hoped for, that which is
feared, that which is willed, and so on. 36
When we say that a man is thought of, 'man' no longer refers to a real entity and no
longer refers to anything in and of itself. What is involved is a thought-of man and if he
is affirmed, what is really being affirmed is only someone who is thinking of him .... 38
In this part of the chapter we turn once again to semantic issues and to the
semantic problems that constitute the "intensionality" of act-sentences -
sentences of propositional attitude, in particular. Our purpose is to glean
from Frege's semantics some insights that we will later see put to work in
Husserl's theory of intentionality. Of particular interest are Frege's concep-
tion of meaning, or sense, the role of meanings in his theory of reference
generally, and his appeal to these entities in explaining the problems created
by certain intentional contexts. We will be especially concerned, ultimately, to
see what place might be found for these entities in a theory of intentionality.
and (3) singular terms like 'the vanquished at Waterloo' and 'the victor at
Jena' refer to different entities despite their being co-referential in extensional
contexts. According to the principle of substititivity of identity, if 'the van-
quished at Waterloo' refers to Napoleon in (2) then any other term that also
refers to Napoleon can be substituted for it salva veritate. But since (2) and
(3) can have different truth-values, the principle of substitutivity of identity
can be preserved only if 'the vanquished at Waterloo' in (2) does not refer to
the same entity as that to which 'the victor at Jena' refers in (3) - and in
that case there is no reason to suppose that either of these terms refers to
Napoleon in these contexts. In this way one is led to conclude that (2) and
(3) are not sentences about the man Napoleon at all.
Interestingly, Frege concluded that expressions in act-contexts refer to
their customary senses (the entities that they customarily, when in exten-
sional contexts, express as their meanings) rather than to the entities that are
their customary referents. And Frege conceived the senses of expressions as
abstract entities of a special sort, sometimes called "intensional entities",
differing in ontological type and status from the entities to which words
ordinarily refer.
We turn now to a discussion of Frege's basic distinction between sense and
referent, of the role of senses in Frege's account of linguistic reference, and
of his conception of senses as intensional entities. Afterward, we discuss the
rather different role of these entities in his analysis of act-contexts and the
possible roles this analysis suggests for intensional entities in the theory of
intentionality. We shall find in later chapters that this discussion of Frege
illuminates the role that Husserl finds for senses, or meanings, in his own
account of intentionality.
from its Bedeutung. Thus, the Sinn of an expression is the sense, or meaning,
that it expresses; while the Bedeutung of a singular term is the entity to
which it refers, i.e., its referent. 44 Frege also applies the distinction between
Sinn and Bedeutung to expressions other than singular terms - notably,
predicates and sentences - that are not usually thought of as referring expres-
sions and whose Bedeutungen are not referents in the ordinary sense. Accord-
ing to Frege, the Bedeutung of a one-place predicate is a "concept", the
Bedeutung of a many-place predicate is a "relation", and the Bedeutung of a
sentence is its truth-value. (Frege's "concepts" and "relations" are functions,
whereas more recent writers - e.g., Carnap - usually take the "extensions"
of predicates to be classes. However, the function that for Frege serves as
Bedeutung of a given predicate defines, and is defmed by, the class of entities
of which the predicate is true - Le., its extension in the usual sense.)45 Frege
does not give details about the different kinds of entities that are to serve
as the Sinne of different kinds of expressions, though, as we shall see, he
explicitly takes Sinne of all kinds to be abstract entities. Since 'Sinn' and
'Bedeutung' are technical terms for Frege, we shall freely use them without
translation; but we shall also use the terms 'sense', or 'meaning', and 'referent',
understood in an appropriately technical sense, as English equivalents. (J{e
use 'reference' for the relation of an expression to its Bedeutung, or referent,
but not for the Bedeutung itself.)
Frege's distinction between sense and referent enables him to account for
features of language that elude theories that appeal to objects of reference
alone. For example, if the sense of an expression is identified with its referent,
such expressions as 'Pegasus' and 'the fountain of youth' must be declared
meaningless unless entities of some unusual sort are postulated as their refer-
ents. But according to Frege, such expressions are meaningful if they express
Sinne, even though they have no Bedeutungen at all; for an expression's being
meaningful is not dependent upon there being any entity to which it refers.
The distinction also enables Frege to explain why the use of one expression
rather than another in a sentence may make a difference in the meaning the
sentence expresses, even though both expressions refer to the very same
entity. 'The Evening Star' and 'the Morning Star', for example, have the same
referent, for both refer to the planet Venus. Even so, Frege notes, the sen-
tences 'The Evening Star is a planet with a shorter period of revolution than
the Earth' and 'The Morning Star is a planet with a shorter period of revolu-
tion than the Earth' do not express the same sense; ''for somebody who does
[not] know that the Morning Star is the Evening Star m,ight regard one as
true and the other as false".46 According to Frege, since the Sinn expressed
SOME CLASSICAL APPROACHES 65
the entity referred to from one particular aspect, in accordance with a certain
way in which that entity may be presented.
The Sinn of an expression thus plays an important role in the expression's
relation to its Bedeutung. The Sinn is not itself the Bedeutung, it is not the
entity to which the expression refers, but it is an entity by means of which
the expression stands in a given relation of reference to the entity that is its
Bedeutung. In this sense, let us say, the Sinn of an expression mediates the
reference of the expression: an expression refers to its referent in a particular
way via, or in virtue of, the sense it expresses.
The following five theses summarize what we have said about the Fregean
account of the role of sense, or Sinn, in the relation of an expression to its
referent, or Bedeutung: (1) The sense (Sinn) of an expression is always dis-
tinct from its referent (Bedeutung). (2) To each sense there may correspond
at most one referent: the referent of an expression is a function of its sense.
(3) An expression can express a sense, and thus be a meaningful expression,
even though it has no referent. (4) The sense of an expression determines the
way in which the expression refers to its referent. (5) Different senses can
determine the same referent; thus, expressions with different senses can refer
to the same entity.
Frege's semantics of meaning and reference has become the model on
which a general tradition in semantic theory, sometimes called the theory of
"intension and extension", has been based. In this tradition, the extension
of an expression is the entity to which the expression refers, or for which it
stands; the intension of an expression is the meaning expressed, rather than
the entity referred to or stood for, by the expression; and the intension of
an expression is always distinct from its extension. Expressions of different
syntactic kinds (notably, singular terms, predicates, and sentences) are taken
to stand for entities of different kinds as their extensions and also to express
entities of different kinds as their intensions. Semantic theories within this
general tradition sometimes differ in the kinds of entities they postulate as
the extensions and intensions of various expressions, but two fundamental
principles prevail: (1) that the extension of a complex expression is a function
of the extensions of its semantically significant parts, and (2) that the inten-
sion of a complex expression is a function of the intensions of its semantically
significant parts. The first of these principles, called the principle of exten-
sionality, has already played prominently in our discussion of the problems
of intensionality (cf. Chapter I, Section 3.1). An important instance of this
principle is that the extension of a sentence - according to Frege and Carnap,
its truth-value - is a function of the extensions of its (semantically significant)
SOME CLASSICAL APPROACHES 67
parts, of what its parts stand for; thus, the principle of substitutivity of
identity is a direct consequence of the principle of extensionality applied
to sentences and their constituent singular terrns (given, of course, that
'a = b' is true just in case 'a' and 'b' have the same extension, i.e., refer to
the same object), and the principle of existential generalization is also argu-
ably grounded in the principle of extensionality.
Various semantic theories, differing from Frege's in some of their details,
have been developed within this generally Fregean tradition. The most famil-
iar, and perhaps the most ontologically economical, is Carnap's.53 like Frege,
Carnap takes the extension of a sentence (e.g., 'The victor at Jena died in
exile') to be its truth-value and the extension of a singular term (e.g., 'the
victor at Jena') to be the individual object to which the term refers. But the
extension of a predicate (e.g., 'died in exile') Carnap takes to be a class,
namely, the class of entities of which the predicate is true, rather than a
function. The entities Carnap proposes as intensions also seem to be entities
rather different from Frege's Sinne (although, as we noted earlier, Frege does
not actually describe the different sorts of senses expressed by expressions of
different syntactic categories). Carnap takes the intension of a predicate to
be the property associated with it (e.g., the property of having died in exile);
the intension of a singular term to be an "individual concept", i.e., a concept
or meaning that determines an individual rather than a class (e.g., the concept
"the victor at Jena"); and the intenson of a sentence to be a complex entity,
called a "proposition", whose components are the intensions of predicates
and singular terms (e.g., the proposition that the victor at Jena died in exile).
Importantly, Carnap stresses that the notion of intension is "not to be under-
stood in a mental sense, that is, as referring to a process of imagining, think-
ing, conceiving, or the like, but rather to something objective ... ".54 We
shall see that on this point - that intensions, or meanings, are not subjective
entities or events - Carnap is in full agreement with Frege.
The referent and sense of a sign are to be distinguished from the associated idea. If the
referent of a sign is an object perceivable by the senses, my idea of it is an internal image,
arising from memories of sense impressions which I have had and acts, both internal and
external, which I have performed .... The same sense is not always connected, even in
the same man, with the same idea. The idea is subjective: one man's idea is not that of
another .... This constitutes an essential distinction between the idea and the sign's
sense, which may be the common property of many and therefore is not a part or a
mode of the individual mind. For one can hardly deny that mankind has a common store
of thoughts [Gedanken I which is transmitted from one generation to another. 55
In his essay, 'The Thought: A Logical Inquiry', Frege also makes it clear
that the ontological category to which Sinne belong is distinct both from
that of physical objects and from that of subjective ideas. Unlike physical
objects, Sinne do not exist in space and are not perceivable by the senses. And
unlike ideas, they are timeless entities: they are not produced by the psycho-
logical processes through which they are apprehended nor is their existence
dependent on their being present to any mind. Frege says of "thoughts"
(Gedanken), the senses expressed by declarative sentences:
... I can ... recognize the thought, which other people can grasp just as much as I, as
being independent of me .... We are not bearers of thoughts as we are bearers of our
ideas. We do not have a thought as we have, say, a sense-impression, but we also do not
see a thought as we see, say, a star .... In thinking we do not produce thoughts but we
apprehend them. 56
Hence: "The thought belongs neither to my inner world as an idea nor yet
to the outer world of material, perceptible things". 57
It is appropriate to say that Sinne, or senses, as conceived by Frege, are
"intentional objects" in the ontological sense we earlier defined (cf. Section
1.3 above). Frege's Sinne are distinct from the entities to which words
ordinarily refer; and, in particular, as abstract entities they are distinct in
ontological kind from such ordinary objects as trees and planets. Although
Sinne are clearly not mind-dependent for Frege, they are in at least one sense
mind-related entities: each Sinn includes a "mode of presentation" of the
referent it determines, and so determines the referent in a particular "way",
by appeal to a particular aspect or, as could also be said, under a particular
SOME CLASSICAL APPROACHES 69
and
(3) Smith believes that the victor at Jena died in exile
have the same truth-value. But, of course, that conclusion may be false. And
from (say) 'Smith believes that p' we will apparently be able to infer 'Smith
believes that q' provided only that 'q' has the same truth-value as 'p'. S9 But
this inference, too, is clearly unwarranted. The principle of extensionality
thus seems to fail in ''belief''-contexts. According to Frege, however, such
examples do not show that the principle of extensionality fails when applied
to such intentional constructions. Rather, Frege takes these apparent excep-
tions to the principle to show only that an expression need not refer to the
same entity in every context. In particular, Frege concludes, when expressions
occur in such contexts as 'Smith believes that _ _ ' they do not take their
usual referents. Expressions that ordinarily refer to the same entity (such as
'the vanquished at Waterloo' and 'the victor at Jena') may accordingly refer
to different entities when in these intentional contexts, so that the principle
of extensionality will then not sanction their interchange. 60
Given this basic idea, that the extension, or referent, of an expression
may depend on the context in which it occurs, Frege can then appeal to the
principle of extensionality to decide what the referent of an expression is in
a given context: what an expression refers to within the context of a partic-
ular linguistic construction must be that on which the extension, or referent,
of the whole construction depends. In the case of sentences of propositional
attitude, in particular, Frege says, "it is not permissible to replace one expres-
sion in the subordinate clause by another having the same customary referent,
but only by one having the same ... customary sense".61 Hence, Frege claims,
expressions in these intentional contexts refer to the senses they customarily
(i.e., when in non-intensional contexts) express, rather than to the entities
that are customarily their referents. In contexts of this sort, Frege says,
SOME CLASSICAL APPROACHES 71
words do not have their customary referent but designate what is usually their sense ....
We distinguish accordingly the customary from the indirect referent of a word; and its
customary sense from its indirect sense. The indirect referent of a word is accordingly
its customary sense. 62
(4) Ponce de Leon hoped that the fountain of youth would be found
in Florida
one cannot infer
(5) (3x) (Ponce de Leon hoped that x would be found in Florida),
since there was, and is, no such entity as the fountain of youth (and so though
(4) be true, (5) is not). Now, although Frege himself does not discuss quan-
tifying into intentional constructions, the doctrine of indirect reference does
provide an explanation of this apparent failure of existential generalization.
As we ordinarily understand (5), the fact that there is no fountain of youth
may render (5) false, because there may then be no concrete object at all of
which the formula 'Ponce de Leon hoped that x would be found in Florida'
is true. However, on Frege's analysis of (4), the non-existence of the fountain
of youth is irrelevant to the truth of (4); for in (4) 'the fountain of youth'
refers to its customary sense, and so the truth of (4) depends not on the
customary referent but on the customary sense of 'the fountain of youth'.
And so (5), as we have understood it, cannot be inferred from (4), as Frege
analyzes it, by existential generalization. (Frege's analysis of (4), however,
suggests a modification of our ordinary understanding of (5); we shall return
to this point shortly.)
In our earlier discussion we also found a second kind of failure of existen-
tial generalization on terms in act-contexts (cf. Chapter I, Section 3.4). For
example, we argued, from
some sense - viz., that expressed by the given singular term - that is an
appropriate value of the given bound variable.
These last two difficulties with the doctrine of indirect reference, as
extended above to quantified act-sentences, have been thoroughly discussed
by David Kaplan in his fruitful article 'Quantifying In'.65 Kaplan shows how
Frege's theory may be supplemented in certain ways that shore up the pre-
ceding Fregean interpretation of quantified act-sentences, reinstate the de
dicto/de re distinction, and appropriately restrict existential generalization
on terms in act-contexts. In particular, Kaplan's work emphasizes the need
to make the customary referents of terms in act-sentences relevant to the
truth-value of the sentences. We shall reach a similar conclusion in our next
section, arguing primarily from the needs of a theory of intentionality rather
than from needs of semantic theory.
other relations except for the fact that they relate persons to intensional
entities, or senses, rather than to more ordinary entities. For in's believes
that p', 's' refers to its customary referent - the person s; 'that p' has "in-
direct" reference to the "thought", or proposition, that is customarily ex-
pressed by 'p' - as we shall say, the thought that p; and 'believes' refers to
the relation that holds between s and the thought that p just in case's believes
that p' is true. 66 Let us call the relation to which 'believes' refers "H"; we can
then express the basic claim of Frege's analysis of belief as follows:
(*) s believes that p if and only if s stands in the relation H to the
thought that p.
Now, on a Fregean analysis of belief-sentences, not only sentences but also
singular terms have indirect reference when they occur in belief-contexts:
they there refer to the senses they customarily express. Hence, in's believes
that a is 1/>', the singular term 'a' refers to its customary sense, a sense that is
a constituent of the thought that a is I/> (in the sense that this thought, or
proposition, is a function of the senses customarily expressed by 'a' and 'is
1/>'). Consequently, if's believes that a is 1/>' is true, then s stands not only in
relation H to the thought that a is I/> but also in another relation - call it "H'"
- to the sense customarily expressed by 'a'. And so we can formulate a
second Fregean claim about belief:
(**) If s believes that a is 1/>, then s stands in the relation H' to the
(customary) sense of 'a'.
The two claims, (*) and (**), seem genuinely to be presupposed by Frege's
semantics for sentences of propositional attitude. But they yield only a partial
theory of the intentionality of belief, and we will have to go beyond Frege
to fill it out. Most importantly, Frege's semantics does not tell us what role
intensional entities play in belief. We do know the role of the thought that p
in Frege's semantic analysis of a sentence's believes that p': it is the entity
to which the clause 'that p' refers. And we also know that it is the "object"
to which a "subject" s stands in the relation H when s believes that p. But we
do not know, and Frege seems not to have specified, what the relation H is
to which 'believes' refers. The obvious suggestion is that H is simply the rela-
tion of believing. In that case, Frege's analysis yields an "object-theory" of
the intentionality of belief (cf. Part 1 above): thoughts are then the objects
of belief and believing itself is an ordinary sort of relation to these intensional
entities, a species of what we earlier called "intentional objects" (cf. Sections
1.3 and 3 .3 above). However, if H is not the relation of believing but is some
SOME CLASSICAL APPROACHES 77
would be equivalent to our saying that Sherlock believes that the murderer
used belladonna?
Even if a Fregean object-theory of aboutness can somehow be rendered
plausible in the case of de dicta belief, it cannot work without serious modifi-
cations for the aboutness that occurs in "definite", or de re, belief. Consider,
for example, the mayor's belief that the governor is a scoundrel, and suppose
that the mayor has known the governor well and long so that there is some
quite specific individual who has become the object of the mayor's disaffec-
tion. (There are other varieties of definite, or de re, belief, but these need not
concern us here. See Chapter VIII, Section 1.1, and also cf. Chapter I, Section
2.6.) Now, the object-version of a Fregean theory of aboutness would yield
an account of the mayor's belief that does not at all capture the sense in
which his belief is definite. For, according to the theory, this belief would
not be about the governor herself in any way, either definitely or indefinitely;
rather, it would be construed as definite with respect to an intensional entity,
namely, the sense that it is about. In this regard, however, the aboutness of
the mayor's definite belief differs not at all from that of Sherlock's indefinite
belief. And so the difference between these beliefs goes unexplained on a
Fregean object-theory of aboutness, which would simply take both beliefs
to be about senses that do not differ from each other in any relevant way.
Even more importantly, though, the mayor's belief as described is correctly
construed only as being about a concrete individual, the governor herself.
Because there was no specific person whom Sherlock's indefinite belief could
unequivocally be said to be about, the object that belief is about was a legiti-
mate matter of speculation. But the belief we are now considering is about a
specific person: if, after campaigning long and bitterly against the governor's
reelection, the mayor announces his belief that the governor is a scoundrel,
we will surely miss his intent if we do not take him to be offering an opinion
about a particular person. Hence, the mayor's belief is of the sort that in
logical notation would be unambiguously expressed by means of an explicitly
de re belief-sentence: here, by '(3x) (x = the governor and the mayor believes
that x is a scoundrel), (cf. Chapter I, Section 3.5). And so the mayor's belief
is properly understood only as a belief about the governor and not as a belief
about the sense customarily expressed by 'the governor'.
In sum, then, the object-version of a Fregean theory of belief shares both
the strengths and the weaknesses of object-theories of intentionality generally.
The chief virtue of introducing some sort of "intentional objects" into the
analysis of intention is that a theory of intentionality can then appeal to spe-
cial features of these entities in order to explain the existence-independence
80 CHAPTER II
Church does not tell us more about this relation to a "concept", or a sense;
nor does he tell us how a sense is involved in the relation that is the inten-
tional relation of having sought. But his comments do suggest that a basically
SOME CLASSICAL APPROACHES 81
NOTES
difficulties". The crucial point, though, is that even with the admission of these "com-
pleted" objects our intentions still fail to reach complete objects: Meinong holds that the
very best we can do is to intend completed (but nonetheless incomplete) objects that "do
duty" for them. See Findlay (Note 16 above), pp. 170-80. Cf. David Woodruff Smith,
'Meinongian Objects', Grazer Philo90phische Studien 1 (1975),43-71, esp. 53-55.
24 The arguments leading to this conclusion are developed in greater detail in Smith,
'Meinongian Objects' (Note 23 above).
25 For recent interesting and sympathetic reconstructions of a generally Meinongian
ontology see Terence Parsons, 'A Prolegomenon to Meinongian Semantics', Journal of
Philosophy 71 (1974), 561-80; Parsons, Nonexilitent Objectli (Yale University Press,
New Haven, Conn., 1980); and Hector-Neri Castaiieda, 'Thinking and the Structure of
the World',Phii0liophill4 (1974), 3-40.
26 Psychology (Note I, Ch. I above), p. 300 [from 'On Genuine and Fictitious Objects',
added in 1911 ed.).
27 The True and the Evident (Note 5 above), p. 84 [from a 1906 letter to Marty).
28 The True and the Evident (Note 5 above), p. 109 [from an essay dictated in 1914).
29 Psychology (Note I, Ch. I above), p. 271 [from 'Mental Reference as Distinguished
from Relation in the Strict Sense', added in 1911 ed.J.
30 See Psychology (Note I, Ch. I above), pp. 292-94 ['On Genuine and Fictitious
Objects', added in 1911 ed.) , and p. 322 ['On Objects of Thought', dictated in 1915,
added in 1924 ed.).
31 Cf. The True and the Evident (Note 5 above), p. 117 [essay of 1915) .
32 See Psychology (Note I, Ch. I above), p. 274 ['Mental Reference as Distinguished
from Relation in the Strict Sense', added in 1911 ed.).
33 The True and the Evident (Note 5 above), p. 112 [essay dictated in 1914).
34 Cf. Psychology (Note I, Ch. I above), pp. 346-47 ['On Enli Rationili', written in
1917, added in 1924 ed.).
35 Psychology (Note I, Ch. I above), p. 272 [from 'Mental Reference as Distinguished
from Relation in the Strict Sense', added in 1911 ed.).
36 Dictated in 1914 and included in Kategorienlehre, ed. by Alfred Kastil (Felix Meiner,
Leipzig, 1933), p. 8; cited by Chisholm in 'Brentano on Descriptive Psychology and the
Intentional' (Note 21, Ch. I above), p. 15.
37 See The True and the Evident (Note 5 above), pp. 67- 70 [essay of 1904); and
Psychology (Note I, Ch. I above), pp. 322-33 ['On Objects of Thought', dictated in
1915, added in 1924 ed.).
38 Psychology (Note I, Ch. I above), pp. 334-35 [from 'On the Term "Being" in its
Loose Sense, Abstract Terms, andEntill Rationili', dictated in 1917, added in 1924 ed.).
39 Cf. Chisholm, 'Brentano on Descriptive Psychology and the Intentional' (Note 21,
Ch. I above), pp. 18-19.
40 Psychology (Note I, Ch. I above), p. 272 ['Mental Reference as Distinguished from
Relation in the Strict Sense', added in 1911 ed.).
41 See Quine, Word and Object (Note 23, Ch. I above), pp. 141-56; and 'Reference and
Modality' (Note 29, Ch. I above), pp. 139-44.
42 Anscombe (Note 1 above).
43 Cf. Findlay's elaboration of the view that Brentano takes intentionality "to be a case
of a unique logical category", in his Values and Intentions (Allen & Unwin, London,
1961), pp. 35-43.
SOME CLASSICAL APPROACHES 85
44 The fundamentals of this theory are expounded by Frege in 'On Sense and Refer-
ence', in Translations from the Philosophical Writings of Gottlob Frege, trans. by Peter
Geach and Max Black (Blackwell, Oxford, 1966), pp. 56-78, esp. pp. 56-67. The essay
was first published, as 'Ober Sinn und Bedeutung', in Zeitschrift fur Philosophie und
Philosophische Kritik 100 (1892), 25-50; a more recent German edition may be found
in Gottlob Frege, Funktion, Begriff; Bedeutung: Funf logische Studien, ed. by Gunther
Patzig (Vandenhoeck & Ruprecht, Gottingen, 1962), pp. 40-65. For fuller accounts of
Frege's philosophy of language see Montgomery Furth's introduction to Frege's The
Basic Laws of Arithmetic, ed. and trans. by Montgomery Furth (University of California
Press, Berkeley, 1967), pp. v-liii; and Michael Dummett, Frege: Philosophy of Language
(Harper & Row, New York, 1973), esp. pp. 81-109, 152-203.
45 See Frege, 'Function and Concept', in Geach and Black (Note 44 above), pp. 30-32.
Cf. Furth (Note 44 above), pp. xxxvii-xliv.
46 'Function and Concept' (Note 45 above), p. 29.
47 'On Sense and Reference' (Note 44 above), p. 58. (Note that in the sentence cited
we have used 'referent' as a translation of 'Bedeutung', where Geach and Black have
'reference' .)
48 'On Sense and Reference' (Note 44 above), p. 58. (,Referent' is again substituted for
'reference'.)
49 'On Sense and Reference' (Note 44 above), p. 58. ('Referent' is substituted for 'refer-
ence'.)
50 Begriffsschrift, in Geach and Black (Note 44 above), p. 12.
51 'On Sense and Reference' (Note 44 above), p. 57.
52 'On Sense and Reference' (Note 44 above), p. 57 (cf. pp. 41, 43-44 of Patzig's
German edition, cited in Note 44).
53 Cf. Carnap, Meaning and Necessity (Note 16, Ch. I above), esp. Chapters I and III.
Other semantic systems in the generally Fregean tradition include those of Alonzo
Church and C. I. Lewis: see Church, 'A Formulation of the Logic of Sense and Denota-
tion', in Structure, Method and Meaning: Essays in Honor of Henry M. Scheffer, ed. by
Paul Henle, H. M. Kallen, and S. K. Langer (Liberal Arts Press, New York, 1951), pp. 3-
24; Church, 'The Need for Abstract Entities in Semantic Analysis', Proceedings of The
American Academy of Arts and Sciences 80 (1951), 100--112, reprinted in Contem-
porary Readings in Logical Theory, ed. by Irving M. Copi and James A. Gould (Mac-
millan, New York, 1967); and Lewis, 'The Modes of Meaning', Philosophy and Phe-
nomenological Research 4 (1944), 236-49, reprinted in Semantics and the Philosophy
of Language, ed. by Leonard Linsky (The University of Illinois Press, Urbana, 1952).
54 Carnap (Note 53 above), p. 21, cf. p. 27.
55 'On Sense and Reference' (Note 44 above), p. 59. ('Referent' is substituted for
'reference' .)
S6 Frege, 'The Thought' A Logical Inquiry', trans. by A. M. and Marcelle Quinton, Mind
6S (1956), 307, reprinted in Philosophical Logic, ed. by P. F. Strawson (Oxford Univer-
sity Press, Oxford, 1967), also reprinted in Essays on Frege, ed. by E. D. Klemke
(University of Illinois Press, Urbana, 1968).
57 'The Thought' (Note 56 above), p. 308.
58 'On Sense and Reference' (Note 44 above), p. 65. (,Referent' is substituted for
'reference' .)
S9 Cf. Frege, 'On Sense and Reference' (Note 44 above), p. 65.
86 CHAPTER II
60 Cf. Frege, 'On Sense and Reference' (Note 44 above), pp. 66-67. Cf. also David
Kaplan, 'Quantifying In' (Note 26, Ch. I above), pp. 182-84.
61 'On Sense and Reference' (Note 44 above), p. 67. (,Referent' is substituted for
'reference' .)
62 'On Sense and Reference' (Note 44 above), p. 59. (,Referent' is substituted for
'reference' .)
63 'On Sense and Reference' (Note 44 above), p. 66.
64 Carnap has similarly proposed for the logical modalities that the range of variables
of quantifying-in be restricted to intensions; see Meaning and Necessity (Note 16, Ch. I
above), pp. 178-82. For a criticism of that approach to quantified modal logic, see
Quine, 'Reference and Modality' (Note 29, Ch. I above), pp. 150-56.
65 Kaplan (Note 26, Ch. I above).
66 Cf. The Basic Laws of Arithmetic (Note 44 above), p. 37, where Frege says: "We say
that the object r stands to the object ~ in the relation \II(E, t) if [and only if) \II(r,~)
is the True". Also cf. 'Function and Concept', in Geach and Black (Note 44 above),
pp.38-39.
67 Introduction to Mathematical Logic, Vol. I (Princeton University Press, Princeton,
1956), p. 8, n. 20.
CHAPTER III
and its structure per se. This inward turn in itself characterizes a rudimentary
sort of phenomenology, which Hussed continually sought to improve and
refine. In Part 1 we discuss the phenomenological thrust of Hussed's approach
to intentionality and also take a brief look at some of the main features that
characterize a distinctively Husserlian phenomenology. Part 2 addresses the
general notion of "phenomenological content", as conceived by Hussed and
by others whose work was familiar to Husser!' The notion undergoes impor-
tant changes in Hussed's own writings, and we compare and contrast his
conception of content in Logical Investigations with his conception of noema
(and the correlative notion of "noesis") in Ideas. Then Part 3 outlines the
basic features of Husserl's theory of intentionality, focusing on his use of
the notion of noema in dealing with the traditional problems of intentionality
that we have previously introduced.
Our interpretation of Hussed's theory of intentionality differs from those
that take the noema of an act to be some sort of object of intention. We take
the noema to be, for Hussed, the ideal "content" of the act, and we take
seriously Hussed's characterization of it as a "meaning" or "sense", indeed,
the same sort of meaning that is expressed in language. The present chapter
contains only half the case for our interpretation, however. Here we argue
that the noema is a content, rather than an object, of intention and that it
is an abstract entity. These arguments can all be accepted whether or not
one takes the additional step that identifies ideal contents, or noemata, with
linguistic Sinne of a Fregean sort; and this first half of our case is sufficient
to establish our basic account of Husserl's theory of intentionality. The
additional arguments for identifying noemata with meanings (specifically,
the meanings that are expressed in language) are presented in Chapter IV.
The identification does not change the basic account but it adds significant
detail to it, detail that ultimately enables us to see some of the shortcomings
of Hussed's theory of intentionality.
any of these entities is intended, that entity itself and not some different
"intentional object" is the object of the act in question. As Husserl says in
Logical Investigations: " ... The intentional [i.e., intended] object of a
presentation is the same as its actual object, and, when appropriate, as its
external object, and . .. it is absurd to distinguish between them" (V, Appen-
dix to §11 and §20, p. 595; with trans. changes). Husserl's general view is
that the object intended in an act - if there is such an object at all - is always
something distinct from and independent of the act that intends it and is
ordinarily an entity of some standard ontological kind.
However, the objects of intentional experiences are not Husserl's main
concern. As we saw in Chapter I (see Part 2), "intentional relations" are
apparently "existence-independent" with respect to their objects and "con-
ception-dependent". But whereas these peculiarities of intention lead some
to postulate intentional objects, they lead Husserl to a different conclusion:
the object intended in an act is not what makes the act intentional. An act
may fail to relate to anything real or actual, but, he says, that does not mean
that the act is directed toward an entity with a peculiarly "intentional" mode
of being. Rather, imaginations, hallucinations, and other "non-veridical"
experiences show that an act can be intentional even if there fails to exist
any object to which the act relates. Continuing the passage we just quoted
from Logical Investigations, V, Husserl says:
If I present to myself God or an angel, ... a physical thing or a round square, etc., I
mean the transcendent object named in each case ... ; it makes no difference whether
this object exists or is imaginary or absurd. That the object is a "merely intentional" one
does not, of course, mean that it exists, though only in the intentio (as a real [reelleg)
constituent of it), or that some shadow of it exists therein. It means rather that the in-
tention, the "meaning" ["Meinen") of an object with such qualities, exists, but not that
the object does. On the other hand, if the intentional [i.e., intended) object exists then
not only does the intention, the meaning [of it), exist but the thing meant [Gemeinte)
algo exists. (P. 596; with trans. changes. Cf. Ll, V, § 11, pp. 558-59.)
is quite different: he holds that these acts have the same object but that the
de facto identity of the object to which they relate does not suffice to make
the intentions the same. In general, Husserl concludes, the specific relation
of intending achieved in an act depends on how the intended object is con-
ceived by the act's subject and so is not determined by what is in fact true
of the object in itself. (See L/, V, §20, pp. 588-89; cf. Chapter I, Sections
2.3-2.4, above.)
Husserl's treatment of these problematic features of intentionality makes
it clear that the "relation" of intending an object, as he conceives it, is not
to be thought of as a relation of the usual sort. For "intentional relations"
between persons and the objects they intend are not dependent on the de
facto status of their objects in the way ordinary, empirical relations are.
Semantically, the difference is reflected in the "intensionality" of sentences
ascribing intentional relations: expressions in intentional contexts fail to
satisfy the logical principles of extensionality, especially the principles of
existential generalization and substitutivity of identity (see Chapter I, Part 3).
Husserl himself draws a closely related conclusion, which has important
methodological consequences. An act's intentional relation to an object is
not a "real" (reell) relation "in objective reality [Wirklichkeit]", he says
(Ideas, §36, p. 80; §88, pp. 220-21; cf. Crisis, pp. 236, 238). What Husserl
means is that an act's intentional directedness is not determined by what is
empirically and contingently true in the natural world. A relation of intend-
ing cannot be reduced to purely physical relations between a person's body
and other physical objects or even to psychophysical relations between a
person's ego (taken as a psychologically real, though perhaps non-physical,
natural entity) and the physical entities (including the person's body) that
affect it. If an act of perceiving a tree is hallucinatory, Husserl notes in Ideas
(§88, pp. 220-21), there exists no perceived object and there cannot be any
"real", empirical relations (such as causal relations) between perceiver and
perceived. There is only the act of perceiving. Nonetheless, the act itself is
intentional: the perceiver is "conscious of something", and in that sense the
act retains its intentionality, irrespective of what the actual empirical situa-
tion may be. Husserl reiterates this point quite nicely in Phenomenological
Psychology, his lectures from the summer of 1925:
involve more than this.) And so Hussed's approach to the problems of inten-
tionality differs fundamentally from the "object-approach": in every case of
intentionality, the important question for Hussed is not "What sort of object
must we say is intended in this act in order to account for its intentionality?"
but rather "What is the phenomenological structure of this act by virtue of
which it is an intentional experience, directed toward a given object in a
specific way?"
Thus, Husserl's phenomenological approach to intentionality requires a
distinction that object-theories do not: the distinction between the intended
object of an act - that which the act is directed toward, or is "of" or "about"
- and the act's content - that which gives the act its directedness, and so
makes it "of" or "about" some object. In general, for Husser!, it is only this
latter entity, the act's content, that is peculiarly "intentional" in ontological
kind. The content of an act, then, is an "intentional object" in two of the
senses we earlier defmed, but not in the third: it is an entity intentional in
kind, and it is an entity whose correlation with an act accounts for the act's
being intentional, but it is not itself an object intended in the act in which
it plays that role (see Chapter II, Section 1.3). And by contrast, what is
intended in an act - the act's object - is neither intentional in kind nor
necessary for the act's having its characteristic property of intentionality.
We shall shortly be looking in detail at Husserl's own account of an act's
"phenomenological content". The ac.count is complex, but what will emerge as
most important for our concerns is an element of content that he characterizes
as a "meaning" or "sense" (Sinn). In Logical Investigations he calls this ele-
ment of content the "matter", or the "interpretive sense" (Auffassungssinn),
of an act; and, with some evolution, it is what in Ideas and subsequent writ-
ings he calls the "noema" of an act or, more specifically, the "noematic Sinn"
in the noema of an act. Hence, we shall see, it is the "meaning" -content, the
noematic Sinn, of an act that gives the act its directedness toward the object
it intends; and to explicate an act's intentionality in terms of the act's phe-
nomenological content is primarily to explicate the role of this meaning entity
in intentionality.
task for itself cannot assume or make use of the presuppositions of the natural
attitude, on pain of begging the very questions it seeks to answer. For these
philosophical purposes, then, all such presuppositions must be suspended,
or set aside: The suspension of these presuppositions is what Husserl calls
"epoch€:", or "bracketing" the thesis of the natural attitude: to "bracket"
this thesis is to refuse to make or to use the assumption that there is a real,
natural world to which our intentions relate. And bracketing this general
assumption entails making no use of the more particular beliefs that presup-
pose it; beliefs about particular objects and all the theories of natural science
are thereby bracketed as well, Husserl says. (See Ideas, § §31-32.)
The purpose of bracketing, or epoch€:, is to turn our attention away from
the objects of the natural world so that our inquiry may focus instead on the
most fundamental evidences on which our naturalistic beliefs about these
objects are based. And for Husserl, as for Descartes, this turn to evidences is
a turn toward the conscious subject and his experiences. (Cf. CM, § §5-8).
Whether the natural world exists or not, Husser! believes, it is self-evident to
the experiencing subject that he undergoes experiences, experiences that at
least purport to be of or about external objects, and that he himself exists
as the subject, or ego, having these experie~ces. Setting aside his ordinary
concern with the natural world, the subject can explicitly direct his attention
to these experiences, and to himself as their subject, in what Husserl calls acts
of "reflection" (Ideas, § §38, 77, 78; CM, § 15.) Consequently, by bracketing
the thesis of the natural attitude as it applies to the objects of our ordinary
intentions, we effect a first "reduction" of the field of philosophical inquiry:
the search for evidences now centers, not on the objects that we ordinarily
intend, but on the intentions themselves and the ego who undergoes them, as
revealed in reflection.
We noted that Husserl calls this reduction "psychological" and that it
takes place within the natural attitude. This means that the reflection in-
volved here is not something unusual or unfamiliar to us in our everyday life
(although its Cartesian motivation may be). Rather, it is simply the kind of
reflection we engage in whenever we attend to our own conscious life and so
make explicit our awareness of ourselves as thinking, experiencing conscious
beings. But it also means that, although we have bracketed the natural attitude
as it applies to the objects of our everyday intentions, we have not bracketed
this attitude as it applies to ourselves and our experiences. We continue to
affmn our belief in the existence of ourselves as natural persons, at least
insofar as we are psychologically functioning conscious egos. And our inten-
tions, likewise, we continue to treat as natl,lral events making up this ego's
HUSSERL'S THEORY OF INTENTIONALITY 97
psychological reality. (See Ideas, § §39, 53.) Thus, this first reduction is
"psychological" inasmuch as it lays before us the kind of data that would
be subject matter for a psychology of inner experiences, i.e., a natural science
whose goal is to articulate and to understand the psychological reality of a
person independent of whether that inner reality corresponds to the external
world. A psychology that proceeds from this data would be appropriately
characterized as "phenomenological" in the broad sense we earlier defined;
Husserl calls it "phenomenological psychology". (See Crisis, §69; §72, p.
263; CM, § §16, 35; also cf. PP, §4.)
In Cartesian Meditations Husserl criticizes Descartes for having failed "to
make the transcendental turn" (§ 10, p. 23). Although we can reflect on the
ego and its experiences, Husser! says,
... It must by no means be accepted as a matter of course that, with our apodictic pure
ego, we have rescued a little tag·end of the world, as the sole unquestionable part of it
for the philosophizing Ego, and that now the problem is to infer the rest of the world
by rightly conducted arguments, according to principles innate in the ego. (eM, § 10,
p.24.)
Now, it is not clear that Descartes did think of the ego as a "little tag-end of
the world", the world whose existence he had placed in question (though it
is true, as Husserl says (p. 24), that Descartes in the final analysis conceived
the ego as causally related to that world). But however that may be, Husserl's
point is simply that the psychological reduction alone is not sufficient for
Descartes' philosophical purposes. Insofar as the ego and its acts are con-
ceived in naturalistic terms, even if we think of the ego as an essentially non-
physical entity causally interacting with the physical, talk of the ego and its
experiences already presupposes the truth of at least part of the general thesis
of the natural attitude and so cannot establish the foundations of that thesis.
Consequently, Husser! says, the method of epoche, or bracketing, must be
extended even to my own ego and to its intentions. We cannot then affirm
the existence of the ego as a psychological reality - what Husser! calls the
"empirical", or "psychological", ego - nor can we affirm the existence of
our acts as constituents of this psychological reality. Nonetheless, Husserl
holds, there still remains an inner life of consciousness that can be described
independently of even these naturalistic affirmations. Our intentions so
described Husserl calls "pure", or "transcendental", acts of the ego; and the
ego that undergoes these acts he calls the "pure", or "transcendental", ego
(/ch). The epoche that brackets the empirical elements in consciousness, thus
leaving only the transcendental ego and its pure acts, is what he calls the
98 CHAPTER III
independent of the actual facts about the empirical reality of our conscious-
ness. If so, there does indeed seem to be a level of description of the ego and
its acts that makes no ontological commitments about the ultimate de facto
reality and nature of the ego. To describe the ego and its acts in this ontologi-
cally neutral way, just as they appear in reflection, is to describe just those
features of the ego that remain when we "bracket" our empirical, or psy-
chological, beliefs about the ego as an empirical reality. But this epoche,
along with the reflection that sets before us the requisite features of the ego,
is just Husserl's "transcendental turn"; and the ego so described is just the ego
in its transcendental aspect, i.e., the transcendental ego.
With this account of transcendental reduction we see that Husserl's doc-
trine of the transcendental ego is not a doctrine of a second ego, a transcen-
dental puppeteer standing behind the empirical ego and manipulating its
activities. Rather, it is the doctrine that there is a level of description of one-
self that is methodologically independent of, and indeed prior to, any further
description of one's ego, one's experiences, and their relationship to each
other and to the world. "As transcendental ego, after all, I am the same ego
that in the worldly sphere is a human ego," Husserl says (Oisis, §72,p. 264;
cf. CM, § 15, p. 37). But transcendental-phenomenological description of this
ego and its consciousness makes no commitments about its status as "human
ego" in the "worldly sphere" - no commitments as to whether the ego and
its acts reside ultimately in soul or body, in ghost or machine, in a person in a
social milieu or merely in a brain in a vat.
What, then, can we say about the ego on this transcendental level of de-
scription? Primarily, says Husserl, the ego is the subject of experiences -
indeed, the common subject of all the experiences that make up a single
stream of consciousness (Ideas, § §57, 80; CM, §31). And to describe the ego
in more specific terms, he says, is just to describe the particular experiences,
especially the intentional experiences, that the ego undergoes and how it
undergoes them .
... The experiencing ego [erlebende Ich) is nothing that could be laid hold of in itself
and made into an object of investigation in its own right. Apart from its "ways of
relating" or "ways of comporting" ["Verhaltungsweisen"), it is completely empty of
essential components, it has absolutely no explicable content, it is in itself indescribable:
pure ego and nothing further. (Ideas, § 80, p. 195.)
Thus, the properties of the ego that are captured in phenomenological de-
scription are its properties of having, or undergoing, these and those particular
experiences and, derivatively, whatever more. enduring t~aits of the ego are
100 CHAPTER III
general. Rather, it is the eidetic science of one very special domain: the
domain of consciousness and its experiences as revealed by the transcendental
reduction:"Phenomenology ... as eidetic science [is the] theory of essences
of transcendentally purified consciousness ... " (Ideas, §60, p. 142).
Eidetic reduction is part of phenomenological method for Husserl, then,
only when it is the final step in his three-part transcendental-phenomenologi-
cal reduction. First, one reflects on consciousness: whatever act is under
consideration, one ceases to be concerned with its object (whether this object
be an individual, an essence, a state of affairs, or some other kind of entity)
and turns one's attention instead to the act in which the object is intended
and to the ego as subject of this act. Second, one disregards the naturalistic
aspects of consciousness through transcendental reduction of the ego and its
acts: this reduction isolates the "pure" data of consciousness from their
presumed naturalistic environment. Third, the data that remain over after
transcendental reduction are then studied eidetically by applying to them
the method of eidetic variation. The result is phenomenology as an "eidetic
science" of transcendental consciousness, a study of those transcendental
features of the ego and its acts that are universal and necessary.
Applied to the reflecting phenomenologist's ego, the eidetic reduction
disregards what is merely idiosyncratic, focusing instead on the transcendental
features essential to it as ego and so necessarily shared by any other ego,
actual or possible. Thus, the "science" of phenomenology includes what
Husserl calls "transcendental egology": a theory of the characteristics neces-
sary to any possible ego, not as empirical ego of any presumed natural kind,
but purely as ego - whatever its actual, empirical nature might be. Similarly,
applied to the transcendentally reduced acts of the ego, eidetic reduction
yields the features of these acts that are necessarily shared with any ego's
acts of the same kind. Thus, phenomenology also includes accounts of the
general features that are necessary for the possibility of various kinds of
experiences: these accounts constitute phenomenological, or transcendental,
theories of perception, of logical and mathematical thinking, of our experi-
ences of other persons, of aesthetic experience, and so on. (See CM, §34;
Ideas, §75.)
The theme uniting phenomenology as theory of the ego and phenomen-
ology as theory of acts is intentionality (Ideas, §84; CM, §14, pp. 32-33).
The most general universal property of acts is their intentionality: to be
intentional is an essential, an eidetic, feature of any actual or possible experi-
ence qua act, any experience of the type "act". And the most general univer-
sal property of the ego is that it is a possible subject of such experiences: to
HUSSERL'S THEORY OF INTENTIONALITY 103
... We have here a transcendental idealism that is nothing more than ... an explication
of my ego as subject of every possible cognition, and indeed [explication of it 1 with
respect to every sense of what exists. ... This idealism ... is sense-explication ...
carried out as regards every type of existent ever conceivable by me, the ego .... (CM,
§41, p. 86. See also § §11-12 and Ideas, §55.)
objects are intended as transcending any given intention of them (as opposed
to merely being so in fact), the content (in some way that Hussed must
explain) presents the object as having properties that are not explicitly deter-
mined by that content itself.
B. The content of an act is a rather complex entity. That contents are
composed of various constituents is already required by the points we have
noted. But there are also structural differences in intention that must be
accounted for in terms of content. Some acts intend their objects as individual
entities having certain properties; others intend their objects as states of
affairs, in which individuals are intended only secondarily as participants in
states of affairs or not at all. Such structural differences in intention - as
manifest, for example, in seeing the elm tree in the yard versus seeing or
judging that the tree in the yard is an elm - would seem to require corre-
sponding differences in the structure of act-contents, differences in the way
the constituents of contents are arranged rather than in the constituents
themselves. Indeed, we shall see that Hussed's construal of contents as mean-
ings allows him to take just this line. Contents are complexes of meanings,
"syntactically" structured so as to present either an individual object or a
state of affairs or an essence - if those are the fundamental ontological "cate-
gories". Contents structured so as to present states of affairs Hussed in fact
characterizes as "propositions" (Siitze).
And so we turn now to the notion of content as Hussed conceives it. Since
this notion is of the very essence of Husserl's theory of intentionality, we
shall try to develop it in detail and with close reference to Hussed's own
words about it, especially in the Investigations and Ideas.
2. "PHENOMENOLOGICAL CONTENT"
and perhaps all the way back to the Stoics. Accordingly, we shall try to
illuminate the notions of noesis and noema by viewing them within the
context of this larger philosophical tradition. We shall not attempt to give
an exact or exhaustive accounting of Husserl's philosophical debts or of his
intellectual history, but we shall seek to articulate some of the basic concep-
tual connections between Husserl's views and those of others and to see the
relations between his own later and earlier views.
In comparing the act of presenting with painting, the content with the picture, and the
object with the subject matter which is put on canvas - for example, a landscape - we
have also more or less approximated the relationship between the act on the one hand
and the content and the object of the presentation on the other. For the painter, the
picture is the means by which to depict the landscape; he wants to picture, paint, a real
or merely imagined landscape, and he does so in painting a picture. He paints a landscape
in making, painting, a picture of this landscape. The landscape is the "primary" object
of his painting activity; the picture is the "secondary" object. Analogously for presenta-
tions .... In presenting to himself an object, a person presents to himself at the same
time a content which is related to this object. The presented object, that is, the object
at which the presenting activity, the act of presentation, aims, is the primary object of
the presenting. The content through which the object is presented is the secondary
object of the presenting activity. (Pp. 15-16.)
We shall say of the content that it is thought, presented, in the presentation; we shall
say of the object that it is presented through the content of the presentation .... What
is presented in a presentation is its content; what is presented through a presentation is
its object. (p. 16.)
(p. 16). Twardowski is not very precise about just how the content of an act
achieves this task, though he does articulate several important theses about
the relation of content to object. (l) He makes it clear that the analogy with
picturing is just an analogy, and that he does not think that the content "is
simply a mental picture of the object" or that "there is a kind of photographic
resemblance between content and object" (p. 64). Rather, he cites approvingly
the predominant view of his day that "the relationship between the presenta-
tion and its object is an irreducible, primary relationship" (p. 64). (2) An
act's content is a complex structure having component parts or constituents;
and the relation between an act and its object is due, at least in part, to rela-
tions obtaining between the structure and constituents of the act's content,
on the one hand, and the corresponding structure and constituents of its
object, on the other (pp. 65ft). (3) Because of its content, every act can be
said to present (or intend) an object; but the presentation of an object,
through a content, does not require that there exist any such presented object
(or, presumably, that the object have any other "mode" of being) (p. 22).
Hence, the presentation (or intention) of an object does not entail the exist-
ence of a presented object. (4) The same object can be presented through
different contents (p. 29), and so different acts can intend the same object.
(5) In general, the content of an act yields only an "inadequate" presentation
of the object; that is, the object "has constituents to which there correspond
no constituents in the content of the presentation" (p. 78). Note that with
this view Twardowski differs importantly from Meinong (cf. Chapter II,
Section 2.3, above): a presentation is "inadequate", not because it intends
an incomplete object, but because its object transcends (in Husserl's sense of
the term) what the act's content explicitly presents of it (cf. p. 82). (6) The
content of a given act, through which the act's object is presented, is not
itself intended in that act; but it can become the object of a different act of
a special sort (what Twardowski calls a "presentation of a presentation" and
Husserl calls an act of "reflection"): " ... The content of a presentation ...
can also be presented through a different act, and this in such a way that the
content of the earlier act is now the object of the new act of presentation"
(p. 60). (7) The relation of an act to its object, achieved by means of the
act's content, is analogous to the relation of a name to its referent, achieved
through the name's meaning, or sense (pp. 8-10). Indeed, Twardowski says
that the meaning of a name just is the content of a presentation underlying
the use of the name (p. 9) and that "the object of a presentation is what is
designated by the name which means the content of the presentation" (p. 91).
All these views of Twardowski's have counterparts that we shall see in
112 CHAPTER III'
Content as such is an individual, psychical datum, an existent here and now. Meaning
[Bedeutung], however, is not something individual, not something real [Reales] , never a
psychological datum. For it is identically the same "in" a limitless manifold of individ-
ually and really distinct acts .... It would be absurd to take it as a real part of the
presentation. 7
...A mere distinction between content and object of presentation [Vorstellung], like
the one recommended by Twardowski following Zimmermann, will not remotely suffice.
. . . There is not one thing which can be distinguished as 'content' from the object named
[or presented] ; there are several things which can and must be so distinguished. Above
all, we can mean by 'content', in the case, e.g., of a nominal presentation, its meaning
[Bedeutung] as an ideal unity .... To this corresponds ... the real [reellen] content
of the presentative act .... (V, §45, p. 657; with spelling of 'Zimmermann' corrected
and with our emphasis. Cf.ldeas, §129, pp. 316-17.)
In fairness to Twardowski, it is not clear from his text that he either ignored
or could not have accommodated Husserl's notion of "ideal" content. 8 But
be that as it may, the distinction between "real content" and "ideal content"
(or "intentional content", as Husserl also calls it), and the identification of
ideal contents with meanings, are key ingredients in Husserl's own conception
of content and, hence, in his theory of intentionality as well.
If, e.g., we call an experience one of 'judgement', there must be some inner determina-
tion, not some mere outwardly attached mark, that distinguishes it as a judgment from
wishes, hopes and other sorts of acts. This determination it shares with all judgements
.... (§ 22, p. 597.)
Briefly, then, the matter of an act is that component of an act's content that
determines which object is intended in the act and also how the object is
intended, i.e., what it is intended "as". (Cf. §21, pp. 591-92; §44, p. 652;
VI, §25, p. 737.)
Some acts include further elements of content in addition to their quality
and matter. In particular, perceptual acts differ from others by also including
a sensory phase, or "sensation-content" (Emp!indungsinhalt) (cf. § 14; §21,
p. 591). The sensory content of perception is epistemologically important,
for it is what gives perception its special evidential status (the sixth Investi-
gation is largely devoted to this topic). Even so, Husser! stresses that it is the
matter of a perceptual act, and not its sensory content, that gives a percep-
tion its intentional directedness. We shall return to perception and its sensory
phase in Section 2.6 below, but for now let us take the content of an act to
include just its quality and matter. It is quality and matter that are necessarily
present in the content of every intentional experience, and Husser! accord-
ingly calls the "union" of these two elements of content the "intentional
essence" of an act (§ 21).
Turning now to ontological basics, we ask, What sort of entity is the
content of an act? This question points up what becomes for Husser! the most
important ambiguity in the notion of content. According to Husser!, there are
two quite different kinds of entities that can legitimately be called "content";
and, as we have noted, he criticized Twardowski for failing to distinguish
them. Where Twardowski distinguished content and object, Husser! would
further distinguish "the real (reellen) and the intentional content of an act"
(§ 16, p. 576), both of which are distinct from an act's object.
The real content of an act, Hussed holds, consists of "real" (reellen)
constituent parts or phases of the act. "By the real phenomenological content
of an act", he says, "we mean the sum total of its concrete or abstract parts,
in other words, the sum total of the partial experiences [Teilerlebnisse] that
really constitute it" (§ 16, p. 576). (Note that by an "abstract part" Husser!
means not an abstract entity that is a part but a non-independent part, or
phase, what he calls a "Moment": cf. LI, III, § 17.) In characterizing an act's
"real" content Husser! uses the word 'reell' rather than 'real', although the
latter would be more customary German: the reason, he says, is to connote
only "real (reelle) immanence in experience" and to avoid the suggestion of
"thinglike transcendence" conveyed by 'real' (§ 16, p. 577, n. 2). An act
itself is an experience (Erlebnis), a "real" temporal event of consciousness in
the sense intended by 'reell'; and its real content, comprising phases or "parts"
that go to make up this experience, is likewise.a "real" event, occurring in
116 CHAPTER III
inner time and in the stream of consciousness. Real content, then, is some-
thing "real", i.e., itis not an abstract, or "ideal", entity or simply a theoretical
construction; but itis "real" in the temporal sense appropriate to constituents
of consciousness rather than in the spatiotemporal sense appropriate to physi-
cal objects.
Specifically, as should be evident from our previous discussion, real content
consists of two essential parts or phases of an intentional experience: the
"part" of an experience that makes it an act of a certain kind (the act's "real"
quality) and the "part" that gives it directedness toward a particular object
in a specific way (the act's "real" matter) (cf. §45, p. 657). Quality and
matter are "parts" of an experience, Husserl says, in the same sense that
direction and acceleration are "parts" of motion. They are not independent
elements that can be separated from the experience or from each other but
are distinguishable within an act as its "phases" (Momente) or "sides" (Seiten)
(§32, p. 621). Real content simply coincides with a complete act itself if
there are no other constituents of the act. But, as we have noted, some acts
(e.g., perceptions) include further phases as well.
In contrast with the real content, the intentional content of an act is an
"ideal", or abstract, entity that can occur in different acts of consciousness.
The real content of an act is necessarily unique to that particular act alone:
just as it makes no sense to speak of numerically the same thought processes
occurring in different persons' consciousnesses, so it makes no sense to speak
of the same real content occurring in different acts. Nonetheless, there is a
sense in which two persons can be said to have the same intention and their
acts to have the same content. In fact, we have ourselves already spoken in
this fashion in characterizing the quality and matter of acts: according to
what we have said, acts of the same kind have the same quality and acts that
intend the same object in the same way have the same matter. But that can-
not be true of quality and matter as components of an act's real content, as
"real" constituents of distinct streams of consciousness. Rather, what are
properly characterized in those terms are quality and matter as components
of what Husserl calls ''intentional content". The real content of an act,
Husserl believes, is in every case the realization in that particular act of an
''ideal'', or abstract, intentional structure that can also be realized in other
acts of the same phenomenological type. This ideal, shareable intentional
structure is the act's intentional content. (See § 21, pp. 590-91.)
In the Investigations Husserl simply identifies the intentional content of
an act with its phenomenological type or species. Thus, he takes intentional
contents to be ''ideal Species of experiencing [ErlebnisspeziesJ" (§ 16, p. 577)
HUSSERL'S THEORY OF INTENTIONALITY 117
and terms the intentional content of an act its "intentional essence" (p. 578).
Essences, or species, he conceives as "ideal", shareable entities in a largely
Platonistic sense: i.e., as independently existing, atemporal universals (or
"types") that can be instantiated in distinct temporal or spatiotemporal
particulars (or "tokens"). The real content of an act is then that part of the
act that literally instantiates, is a concrete instance of, the intentional content
of the act as its intentional essence. And when two people have the same
intention (e.g., they make the same wish) the distinct real contents of their
respective acts are related to their acts' common intentional content as tem-
poral tokens of the same intentional, or phenomenological, type. It should be
clear, then, that intentional content is not "in" an act in the same way that
real content is. Real content is "in" an act in a nearly literal sense: it is a
constituent part of the act and exists only as a part of the act. Intentional
content is "in" an act in a less direct way: as an ideal entity it exists inde-
pendently of the act and is no "real" constituent of it, but it is instantiated
in the real content that is a constituent of the act. And it should also be clear
that neither real nor intentional content is itself intended in an act: what is
intended in an act is the act's object, which is neither a part of the act (as
real content is) nor instantiated in the act (as intentional content is). We
should note, however, that by the time of Ideas Husserl had changed his mind
about the ontological category of intentional contents and the relation of
real to intentional contents. As we shall soon see, Husserl no longer took
intentional contents to be essences or types but a special category of ideal
entities, which are "correlates" of real contents in an appropriately different
way (see Sections 2.3 and 3.1 below). With either view, though, Husserl took
an act to be directed toward its object by virtue of its intentional content.
Despite important changes in his conception of intentional content, with
accompanying changes in terminology, Husserl has one way of characterizing
intentional content that recurs throughout his writings: the intentional con-
tent of an act is a meaning (Bedeutung) or sense (Sinn) (cf. LI, I, § § 14,30,
31; V, §§20, 21,45). For Husserl, the ideal phenomenological content of
an act is a conceptual entity, a meaning, of the same sort that we grasp when
we understand language. Thus, we may think of an act's intentional content
as the "meaning", or "sense", of the act, by virtue of which the act intends
its object - much as the sense of an expression is that by virtue of which
the expression relates to its referent (cf. Chapter IV, Part 2, below). In
particular, Husserl calls an act's ideal matter - the specific element of inten-
tional content that determines the "way" in which the act's object is con-
ceived or "apprehended" (auffasst) in the act - the act's "interpretive sense"
118 CHAPTER III
of others, Husserl calls Bolzano "one of the greatest logicians of all time"
(LJ, Prolegomena, p. 223) and acknowledges that the Investigations "have
been crucially stimulated by Bolzano" (p. 224).
Though our discussion of historical influences on Husserl is far from com-
plete, we can see that Husserl's notion of phenomenological content belongs
to a family of views about content and object of consciousness and that the
family resemblances are often striking. A fundamental distinction among act,
content, and object is found, under differing terminologies, in Twardowski,
Meinong, Bolzano, Frege, and Husserl. And a distinction between real and
intentional, or subjective and objective, content is clearly found in Bolzano,
Frege, and Husserl, who also share similar views about language and mean-
ing. There are differences of detail among these thinkers, differences it is
ultimately important to observe. But it is no less important to see that the
foundation of Husserl's theory of intentionality and hence of his developing
phenomenology lies within a tradition of theory concerning consciousness
and its relation to its objects.
also find that there is a new kind of "correlation" between real and inten-
tional content in Ideas (cf. Section 3.1 below).
Having characterized noesis as a "real" phase of experience, especially its
"sense-giving" phase, in §88 Husserl again, as in the Investigations, distin-
guishes "the real [reel/en] components of intentional experiences and their
intentional co"elates" (p. 218), and again he characterizes the latter as a kind
of sense, or Sinn. Thus is introduced the notion of noema:
not (cf. Ideas, §90, pp. 223-24; and Section 1.1 above): the noesis is the
intending phase of an experience rather than its intended object. 13 Nor is the
noema intended in the act whose noema it is, although Husserl sometimes
uses a terminology that might suggest otherwise. He sometimes calls the Sinn-
component of an act's noema "the intended as such"; more specifically, he
calls the noematic Sinn of a perception "the perceived as such", the noematic
Sinn of a memory "the remembered as such", and so on (see, e.g., §88, p.
219). If there were nothing more to go on, this terminology might lead us to
think that the noema of an act is in some sense an object of that act, some-
thing that is itself perceived in a perception, remembered in a memory, etc.
(We discuss the terminology, and the kind of interpretation sometimes drawn
from it, in Chapter IV, Sections 1.2 and 1.3, below.) But Husserl makes it
quite clear that noemata are entities of which we are not conscious except
in special acts of reflection. The object intended in an act is what is presented
to us when we "perform" (or "live through", as Husserl says) that act itself,
when we undergo the experience that is directed toward that object. But an
act's noema is not something of which we are conscious when we undergo
or live through the act; and precisely because it is not, we must explicitly
adopt the "phenomenological attitude", in which we reflect on the experi-
ence that we normally live through, in order to become aware of its noema.
To become acquainted with the noema of a perception, for example, we must
explicitly turn our attention away from that which is perceived and, in a dis-
tinct act of phenomenological reflection, redirect our attention toward the
experience of perceiving and its phenomenological structure (see esp. Ideas,
§87, p. 217; § 150, p. 369). Thus, Husserl says, "A unique kind of reflection
can at any time be directed toward this Sinn [or noema], as it is immanent
in the perception, and the phenomenological judgment ... must conform
with what is grasped in the reflection alone" (§89, p. 222; cf. FTL, §50).14
As entities open to phenomenological reflection, noemata (and noeses)
have a privileged epistemological or phenomenological status that further
distinguishes them from the objects of our ordinary intentional experiences.
Phenomenological reflection is directed toward the "transcendental" features
of an experience, i.e., those features that remain when all that is not imma-
nent in the experience itself has been "bracketed" in "phenomenological
reduction". Now, the object intended in an act is bracketed in this reduc-
tion: the existence, or reality, of the act's object is not entailed by the act
of intending it, and the phenomenological reduction therefore "suspends"
judgments concerning its existence. But the noema and the noesis of the act
survive this reduction. Husserl says of an act of perceiving a tree:
HUSSERL'S THEORY OF INTENTIONALITY 123
We must now describe what remains as phenomenological residue when we make the
reduction to the "purely immanent", and what should then count as a real [reelles)
constituent of the pure experience and what should not. And here it must be made
completely clear that the . .. noema, which is not affected by the suspension of the
reality [Wirklichkeit) of the tree itself and of the whole world, does indeed belong to
the essence of the perceptual experience in itself, but that, on the other hand, this
noema ... is as little really [reell) contained in the perception as is the tree of natural
reality [Wirklichkeit) . (§ 97, p. 242; our emphasis.)
Hence, whereas the objects of our perceptions, memories, and the like are
"transcendent" objects, i.e., objects "transcendent" of our ordinary experi-
ences, the noemata and noeses of these intentions are "transcendental"
entities, i.e., entities that are "immanent" to the experiences and essential
to their being the intentional experiences they are.
like the intentional contents of Logical Investigations, noemata are
abstract ,entities, i.e., entities that do not have a location in space and time.
Husserl characterizes noemata and their various constituents as "ideal"
(ideell) in Ideas (§99, p. 250), and he calls Sinne "ideal" (ideal) or "irreal"
in Formal and Transcendental Logic (cf. § §48-50, 57b; also cf.EJ, § §64-
65). The ideality of noemata, or Sinne, marks yet another way in which they
differ from the objects of many of our everyday intentions. In particular, it
distinguishes the noema of a perception from the object perceived, since ab-
stract entities cannot be perceived. In §89 ofldeas Husserl himself contrasts
the noema, or Sinn, of a perception with its object (again, a tree), by stressing
the abstract character of the Sinn:
The tree simpliciter, the thing in nature, is anything but this perceived tree as such,
which as perceptual Sinn belongs inseparably to the perception. The tree can burn, can
break down into its chemical elements, etc. The Sinn, however - the Sinn of this percep-
tion, which belongs by necessity to its essence - cannot burn, it has no chemical ele-
ments, no powers, no real [realen) properties. (P. 222.)
In Formal and Transcendental Logic Husserl says that "thoughts" (as opposed
to the "real psychic processes" of thinking, judging, and the like) "are not
.
real [realen] objects, not spatial objects, but irreal entities shaped by the
mind [irreale Geistesgebilde]; and their peculiar essence excludes spatial
124 CHAPTER III
extension, original locality, and mobility" (§57b, p. 155; with trans. changes).
"An irreal object", he says, "is not individuated in consequence of a tem-
porality belonging to it originally" (FTL, § 58, p. 156). In an unpublished
manuscript, 'Noema und Sinn', Husserl says: "Sinne are nonreal objects, they
are not objects that exist in time", and "A Sinn ... is related to a temporal
interval through the act in which it occurs, but it does not itself have reality
[Dasein], an individual connection with time and duration". IS Husserl is
especially clear about the abstract character of linguistic meanings, particu-
larly the propositions expressed in judgments, in Logical Investigations, I
(§ § 11, 29-35), and Experience and Judgment (§ § 64-65). These views,
too, are relevant to noemata, since Husserl holds that noemata and other
meanings are entities of the same kind: "All ... Sinne and all ... noemata,
however different they may otherwise be, are fundamentally of one unique
supreme genus," he says in Ideas (§128, p. 314; cf. Chapter IV, Part 2,
below).
In both Logical Investigations and Ideas, then, intentional contents are
abstract, or ideal, meaning entities. However, in the time between these two
works Husserl's conception of the ideality of meanings, or intentional con-
tents, underwent an important change. As we have already mentioned, this
change is reflected in the notion of noema in Ideas. In the Investigations the
intentional content of an act is the act's "intentional essence". Intentional
contents are then a kind of universals, ideal species or types of consciousness
instantiated in acts, just as redness is a property instantiated in red things.
In Ideas, however, noemata are not act-essences, or universals, but abstract
entities of a different sort. As we shall see later, Husserl's description of the
inner structure of the specific Sinn-component of the noema seems to indicate
that Sinne are a kind of abstract particulars; in particular, the Sinn of a direct-
object act is quite like the sense of a defmite description on a Fregean theory
of meaning (see Chapter IV, Sections 3.l-3.2;cf.Ideas, §§130-3l). This
change in the ontological status of intentional contents has been traced to
unpublished texts of Husserl's from the year 1908 concerning the ideality of
meanings generally, 16 and it is explicit in Husserl's later published writings. In
Formal and Transcendental Logic (I929), for example, he speaks of "the
ideality of meanings [Bedeutungen] and the different ideality of universal
essences or species ... " (§ 57b, p. 155; with trans. changes and our emphasis).
And in Experience and Judgment (posthumous), he says:
The irreality of objectivities of understanding must not be confused with generic univer-
sality. Since, in particular, any number of affirming acts ... affinn ... one and the same
HUSSERL'S THEORY OF INTENTIONALITY 125
proposition, ... it is a great temptation to think that the proposition belongs to the
various acts of which it is the Sinn as a generic universal, perhaps as the generic essence
"redness" belongs to many red things ....
But one must say in opposition to this: ... the proposition ... is not general in the
sense of generic universality, i.e., the generality of an "extension" ... ; it is, therefore,
not general in the manner of essences .... (§64d, p. 262; with trans. changes.)
... The generic universal ... has particulars under it; but the Sinn does not have
particulars under it. (P. 263; with trans. changes.)
2.4. The Structure of an Act's Noema: Its "Sinn" and "Thetic" Components
For HusserI, an act's noesis and noema are complex, or structured, entities.
And the structure of an act's noema exactly parallels the structure of the act's
noesis: "No noetic phase [Moment] without a noematic phase that belongs
specifically to it", HusserI says (Ideas, §93, p. 232; cf. §98 and especially
§ 128). This parallel structuring of noesis and noema should not be a surprise
since noesis and noema are just HusserI's refined versions of real and inten-
tional content, which the Investigations found to be structured in a similar
fashion. There HusserI distinguished two components or phases in both the
real and intentional contents of an act: one component, called the "quality"
of the act, simply differentiates the act according to generic kind (e.g., per-
ception or desire); the other component, called "matter", differentiates the
act more closely as it determines which object is intended in the act and what
this object is there intended "as" (cf. Section 2.2 above). In Ideas, both noesis
and noerna have basically this same bipartite structure, though HusserI is now
more careful to emphasize the difference between real components of con-
tent, which belong to the noesis, and their ideal correlates, which belong to
the noema (cf. Ideas, §94 and §129, pp. 316-17). Our focus, like HusserI's,
will be mainly on the structure of an act's noema.
126 CHAPTER III
What Husserl strictly calls the "Sinn" of an act, then, is only a central "nu-
cleus" or "core" (Kern) in the complete noema; the noema's further phases,
and the noema as a whole, are "Sinne" in a more extended sense. The further
phases, or components, of the noema are ideal correlates of what Husserl
calls "thetic" phases of the noesis. The most prominent thetic phase of the
noesis and its thetic correlate iIi the noema, we shall find, correspond to real
and ideal "quality" in the Investigations.
The Sinn-component of the noema receives most of Husserl's attention,
and it is the component that bears most importantly on the problems of
intentionality that concern us. Husserl has several names for it: 'Sinn', which
he also sometimes uses to refer to the whole noema; 'noematic Sinn'; 'objec-
tive Sinn', so-called because the Sinn-component is what relates the act to
its object; and other less obvious names, including 'the intended [perceived,
remembered, etc.] as such' (cf. Chapter IV, Section 1.2, below) and 'the
cogitatum qua cogitatum ' (especially used in Cartesian Meditations). By what-
ever name, however, the role of the Sinn in the noema is to account for an
act's intentional relation to its object.
Husserl makes the same point in the manuscript 'Noema und Sinn': "Same-
ness of Sinn occurs only where the object, besides being identically the same,
is meant 'in the same Sinn', that is, from the same side, with the same prop-
erties etc." 17
These comments about Sinne, as components of noemata, are of course
but echoes of what Husserl said about "matter" in the Investigations:
We have the same presentation of a thing [Sache) when we have presentations in which
the thing is not merely presented, but presented as exactly the same; i.e., ... in the same
"interpretive Sinn" ["Auffassungssinne") or on the basis of the same matter. ... Two
presentations are in essence the same when, on the basis of either, exactly the same and
nothing else can be said about the presented thing. (Ll, V, §21, p. 591; with trans.
changes.)
the intended object will be rendered technically precise in Chapter IV. But
our intuitive grasp of the notion can be sharpened now through a few exam-
ples. These examples will also begin to distinguish those features of an act
that are due to the Sinn in the noema from those that are due to other noema-
components.
(1) My act of seeing my dog and my act of seeing my cat are acts whose
noemata have different Sinn-components. In the first case the Sinn is (or
includes) my sense of a particular object (the object I see) as "my dog"; in
the second, my sense of an object as "my cat". These are different senses,
which determine different objects for the two acts.
(2) The noema of my act of remembering a certain Freddie McAlister as
the center on my high school basketball team and the noema of Smith's act
of remembering Freddie as the boy who wanted to go to embalming school
have different Sinne. The Sinn of my act's noema is my sense, or conception,
of Freddie as "the center on my high school basketball team"; the Sinn of
Smith's act's noema is his sense of Freddie as "the boy who wanted to go to
embalming school". These are different senses, but in this case they are senses
of the same object (Freddie McAlister) and so they determine my act and
Smith's as directed, through different noemata, toward the same object.
(3) The noema of my act of seeing a house from the front and the noema
of my act of seeing that same house from the rear will almost surely have
different Sinne. Seeing a house from different perspectives will very likely
reveal different features and so give me a different sense, or conception, of
what the house is like: from the front, I may see the house as having a large
bay window but be unaware of the small veranda in the rear; from the rear,
I may be unaware of the bay window but see the house as having the small
veranda. Thus, the Sinn of each act's noema will likely include senses of the
object that are not present in the Sinn of the other's. Nonetheless, if I see the
house as the same one in each case, the complex of senses making up these
different Sinne will determine the same object; the acts will then intend the
same house but from different perspectives. Only in the unlikely event that
the house looks exactly the same to me from the front and the rear may my
acts have noemata with the same Sinn. And even if the front and the rear of
the house are perfectly similar, there may still be a difference in Sinn if I am
aware of being differently oriented to the house in each case.
(4) My act of seeing a certain house from the front and my later act of
remembering that same house exactly as I saw it from that particular perspec-
tive have different noemata with the same Sinn. Although these are different
acts, and acts of different kinds, the Sinn in their noemata is the same since
HUSSERL'S THEORY OF INTENTIONALITY 129
my sense of the object is exactly the same. Of course, if I have later forgotten
some feature that I saw the house as having, or if I saw the house as having
a bay window to the left of the front door and later remember it as having
a bay window to the right of the front door, my sense of the house will have
changed just so and the noemata of my acts will have Sinne that differ in just
that respect.
(5) My act of judging that inflation will worsen in the next year and
Smith's act of fearing that inflation will worsen in the next year are also acts
of different kinds whose noemata contain the same Sinn. But there is an
important difference between these acts and the acts considered in the pre-
ceding example. My act of judging and Smith's act of fearing are propositional
acts and, according to Husser!, the object of a propositional act is a state of
affairs rather than an individual (cf. Chapter I, Sections 1.4-1.5, above). The
Sinn in the noema of these acts, accordingly, is not a sense of an individual,
not a sense that prescribes or determines an individual as the object ~ntended
in the act, but a sense of the state of affairs that is judged or feared. A sense
or Sinn of this sort Husser! calls a "proposition" (Satz) (see Ideas, §94).18
Hence, the Sinn in the noema of my judgment is the proposition that infla-
tion will worsen in the next year. The noema of Smith's fear contains that
same Sinn, and through it we intend the same state of affairs - the state of
affairs that obtains just in case inflation does worsen in the next year.
(6) My act of hoping that the center on my high school basketball team
became a wealthy man and Smith's act of believing that the boy who wanted
to go to embalming school became a wealthy man have different noemata
with different Sinne. As the previous two examples should suggest, the differ-
ence in Sinne here is not due to my act's being an act of hoping and Smith's
being an act of believing. Rather, as in example (2), the Sinne are different
because the Sinn in my act's noema includes one sense of Freddie McAlister
(the sense "the center on my high school basketball team") whereas the
Sinn in Smith's includes another ("the boy who wanted to go to embalming
school"). However, like the acts in the preceding example, my hope and
Smith's belief are propositional acts, directed toward states of affairs through
propositional Sinne. For each of these acts, then, the sense of Freddie
McAlister is not the complete Sinn in the noema but only the part of the
Sinn that determines Freddie as the object the act is about (cf. Chapter I,
Section 1.5, above). The complete Sinn of my hope is the proposition that
the center on my high school basketball team became a wealthy man, while
the complete Sinn of Smith's belief is the proposition that the boy who
wanted to go to embalming school became a wealthy man. Included in each
130 CHAPTER III
proposition is a "sense" of the object the act is about, and each proposition
is a "sense" of the state of affairs the act is primarily directed toward. Since
the constituent senses are different in each case, so are the propositions and
hence the Sinne. Nonetheless, the acts are about the same object - Freddie
McAlister - though they intend him through different propositional Sinne
in the acts' noemata.
As our examples show, the Sinn-component of a noema can itself be
complex and can have a quite defmite structure. Such complexity is only to
be expected, since Husser! sees the Sinn as the main determinant of the inten-
tion achieved in an act and intention itself exhibits various kinds of complex-
ity. Propositional acts, for example, intend their objects as structured into
states of affairs in a way that direct-object acts do not. This difference, we
have just seen, is reflected in the Sinne of acts of these different kinds: the
Sinn in a propositional act's noema has a propositional structure, while the
Sinn in a direct-object act's noema has a simpler structure that was more
appropriately characterized as the "sense of an object" - an ''individual
sense", if you will. But even in the latter case, as our third example especially
shows, the Sinn is usually a complex sense formed from a pattern of consti-
tuent senses, inasmuch as the subject's conception of the intended object is
often quite rich and detailed. Thus, when I see a house, my conception of
what I see includes far more than my sense of it as "a house": I see it as a
particular house, as a house built in a certain style, painted a certain color,
having various prominent features, perhaps even as the house in which my
friend lives, and so on. Not everything I know or could be brought to notice
about the house belongs in my present conception of it, but whatever does
characterize the house just as I now see it is reflected as a constituent sense
in the Sinn-component of my act's noema. Such complexities of structure
within the Sinn itself will receive our closer attention at later points, as we
get into finer details of Husserl's theory of intentionality (see Chapter IV,
Section 3.1, below). But let us return now to the structure of the full noema,
of which the Sinn is only one, though a centrally important, component.
We have seen that when acts of different kinds are directed toward the same
object, conceived in exactly the same way, their noemata have exactly the
same Sinn-component. Nonetheless, Husser! says, acts of different kinds have
different noemata (and different noeses). The phenomenological character
of an act that is characteristic of its kind or species is thus embodied in
HUSSERL'S THEORY OF INTENTIONALITY 131
A blossoming tree may be under consideration throughout, and this tree may appear
throughout in such a way that the accurate description of that which appears as such
necessarily proceeds with the same expressions. But the noematic correlates are yet ...
essentially different for perception, phantasy, pictorial representation, memory, etc.
That which appears is characterized at one time as an "embodied reality", the other
time as a fiction, then again as a memory representation, etc.
These are characters that we find with the perceived, phantasied, remembered, etc.,
as such - with the Sinn of perception, with the Sinn of phantasy, Sinn of memory -
as something inseparable and as necessarily belonging to them in co"eiation with the
respective kinds of noetic experiences. (Ideas, § 91, p. 227.)
Thus, in addition to its Sinn, each noema includes an ideal correlate of the
generic "way" in which the subject is conscious in the act - perceptually,
recollectively, etc. This "way" of being conscious in an act is part of what
Hussed calls the "thetic character", or the "positing character" (Setzungs-
charakter), of the act (see Ideas, § 117). Strictly speaking, the thetic charac-
ter of an act is a real phase of the act's noesis; the corresponding "character"
in the noemais then not the act's the tic character per se but its ideal correlate.
Husser! is himself clear on this point: "It is not 'ways [Weisen] of conscious-
ness' in the sense of noetic phases that are expressed therewith", he says. "As
characters of the, so to speak, 'ideal' ['Ideellen'] , they are themselves 'idea1'
['ideell'] and not real [reell] " (Ideas, §99, p. 250).
Insofar as the thetic character of an act coincides with the act's generic
kind, the thetic phase in an act's noesis corresponds to what in the Investi-
gations Husserl called the "quality" in an act's real content. Similarly, the
ideal thetic component in an act's noema, correlated with this noetic phase,
corresponds to what ~as earlier called "quality" as a component of inten-
tional content. In Ideas, however, Husserl sees an act's kind, or quality, as
but one of several act-"characters", or thetic characters, that can change from
one act to another without necessarily effecting a change in the conception
under which a given object is intended. All these act-characters, which to-
gether make up what Husserl calls the "way of givenness" (Gegebenheitsweise)
of the object, are reflected in the extra-Sinn component of the act's noema
(see Ideas, §99, p. 250). The ideal correlate of an act's quality is then only a
part - though the most prominent part - of a broader thetic component in
the noema that correlates with the entire "way of givenness" of the object
intended in the act (cf. Ideas, § 114, p. 278; § 117; § 120, p. 296).
132 CHAPTER III
If we hold fixed the Sinn, hence the "intended" ["Vermeinte"j exactly with the content
of determinations in which it is the intended, then there clearly emerges a second con-
cept of the "object in the manner [Wiej" - in the manner of its ways of given ness lim
Wie seiner Gegebenheitsweisenj. (§ 132, p. 323.)
The structuring of an act's noema into a Sinn and a thetic component (and
the parallel structuring of the noesis) mirrors the structuring of content into
matter and quality in Logical Investigations. Now, we have argued that the
notions of noesis and noema, as Hussed himself understood them, are refmed
versions of the notions of real and intentional content in the Investigations.
The results of this section provide strong confirmation of this view: not only
do noesis and noema correspond as a whole to real and ideal content; the
components that make up the noesis and the noema also correspond to the
components that make up an act's real and intentional content.
Hussed notes this correspondence at several points in Ideas, most explicitly
in § § 129 and 133 (see also §88, p. 219, n. 1; §94; Beilage XVII). In § 129
Hussed assesses his earlier distinction between matter and quality as a neces-
sary fust step toward refining Twardowski's notion of content, though he
says the distinction fell short by being primarily "noetic". He continues:
134 CHAPTER III
The one-sidedness ... is easily overcome through consideration of the noematic parallels.
We can understand the concepts noematically thus: "quality" (judgment-quality, wish-
quality, etc.) is nothing other than what we have hitherto treated as "positing" character,
"thetic" character in the broadest sense .... Obviously, "matter" ... now corresponds
to the "noematic nucleus". (P. 317.)
§ 133 makes it clear that the "noematic nucleus" that corresponds to "mat-
ter" is the Sinn in the noema. Recalling that intentional content as a whole
(quality plus matter) was called "Sinn" in the Investigations, he says:
... The thetic phases ... have a special relation to the Sinn as noematic. In the Logical
Investigations they (under the title "quality") were included from the outset in the
concept of Sinn ... , and, accordingly, within this unit the two components "matter"
(Sinn, on the present conception) and quality were distinguished. Yet it seems more
appropriate to define the term 'Sinn' only as that "matter" .... (P. 324.)
noema of an act has two main constituents: a thetic component and a Sinn,
which correspond respectively to what were earlier called quality and matter
as ideal constituents of intentional content. The thetic component of the
noema is the ideal correlate of the the tic phase of the noesis, and the Sinn in
the noema is the ideal correlate of the Sinn-giving phase of the noesis. The
Sinn determines the specific intentional relation that obtains between the
act and its object and is therefore the key notion of content for a theory of
intentionality.
PHENOMENOLOGICAL
CONTENT
NOESIS NOEMA
(Real Content) (Intentional Content)
2.6. The Content of Perception: its Sensory (or Hyletic) and Noetic Phases
The doctrine of noesis and noema is offered by Hussed as a completely
general account of an act's phenomenological content, and the distinctions
drawn within that account apply to all acts of every kind - acts of perceiv-
ing, remembering, imagining, desiring, judging, and so on. But the case of
perception complicates this account of phenomenological content. It is one
thing merely to entertain an object in thought and quite another to see it.
For seeing is a sensory experience, an experience that essentially involves
visual sensation. Thereby its object is presented with direct sensory support
HUSSERL'S THEORY OF INTENTIONALITY 137
as colors and shapes, much less physical objects such as trees or tomatoes.
For the sensory phase "in itself has nothing of intentionality" (p. 208) .
. . . Not every real phase in the concrete unity of an intentional experience [Erlebnisses)
has itself the fundamental character of intentionality, the property "consciousness of
something". This applies, e.g., to all sensotion-data, which play so large a role in percep-
tive intuitions of things .... the "exhibiting" ["darstellender") content for the appearing
whiteness of the paper ... is the bearer of an intentionality, but is not itself a conscious-
ness of something. (Ideas, § 36, p. 81.)
brown color of the penny. (ef. Ideas, §41, p. 94; PP, §29, p. 125, and
§ §30-31.) These appearings are sensory-and-intentional parts or phases of
the perception. (Where I see this round, coppery brown penny, the appearing
of its roundness to me is a partial intention that is a proper constituent of the
perception as a whole: cf.LI, § 10, p. 701.) Thus, the role of sensation in per-
ception is that of joining with appropriate noetic phases to form appearings
of sensible qualities and thereby to give perception its sensory and evidential
characters.
The sensory phase in a perception gives intuitive evidential support to
appropriate parts of the noetic phase in the perception. Where the noetic
phase "animates" the sensory phase, the sensory phase "fills" that noetic
phase. And the corresponding components of sense in the act's noematic
Sinn, Hussed says, are filled, or intuitionally (ful)filled, by the appropriate
hyletic experiences. (Their character of being filled is not part of the Sinn
itself but part of the thetic, or way-of-givenness, component of the noema:
cf. Ideas, §99.) (See LI, VI, § § 14(b), 17,21-29; Ideas, § 135, pp. 329-30,
and §136.) Perception is thereby an evident, or self-evident, experience, a
presentation carrying evidence for what it posits. The character of evidence
in a perceptual experience lies in the "intuitional fullness" of its noetic phase,
which consists in the noetic phase's being fIlled by a hyletic phase. Thus, the
sensory phase provides sensory evidence for what the noetic phase posits -
say, a certain object having certain properties. In particular, the appearings
of colors and shapes that are parts of a visual perception embody a sensory
evidence owing to its hyletic phase: the colors and shapes that are appearing
are presented with sensory evidence thanks to the hyletic phases in the
appearings. Other properties too may be supported to some degree by the
ingredient hyle, of course: in seeing a penny, the presentation of the object's
property of being a penny surely receives some evidential support from
appropriate hyletic phases in the perception. But the sensory phase never
provides complete evidence for what is posited by the noetic phase: the
noema of a perception of a physical object always includes "unfilled" com-
ponents of sense (see Ideas, §44).
We have observed Husserl's account of the hyletic content of perception
here in order to complete our presentation of his account of phenomenologi-
cal content. However, as we have noted, the intentionality of a perception -
its being of or about an object - is due entirely to its noematic Sinn, or the
corresponding Sinn-giving phase of its noesis. Intentionality is our primary
concern, and we return now to the story of Sinn and intention.
HUSSERL'S THEORY OF INTENTIONALITY 141
We have now studied in detail Hussed's doctrine of content and the many
distinctions it involves. A statement of the relations among act, content, and
object is very neadya statement of Hussed's mature theory of intentionality.
In this part of the chapter, we pull together the main themes of this theory.
We begin with an appraisal of how, on Hussed's account, intention is achieved
via noematic Sinn. Then we gather together the basic principles of the theory
and show how it deals with the traditional problems of intentionality.
that is, its having a certain intentional essence (and, if the intention is success-
ful, there being a corresponding object). (Note that there is no interesting
relation between the ideal content, or essence, and the object of an act.)
Hussed's theory of intentionality in the Investigations, so far as it goes, is
the sort of theory that today would be called an "adverbial" theory: to see
this black crow is to intend "(seeing this black crow)-ly", and so to have an
experience of a certain type, a "seeing-this-black-crow"-type of experience.
On such a theory, intentionality is a non-relational property of an act, a
complex quality or type that receives no further ontological analysis. If you
will, to say that an act is directed by virtue of its content is to say no more
than that it is directed in a certain way by virtue of its having the property,
or essence, of being directed in a certain way. For the strict adverbialist, this
is not as unilluminating as it appears; for the point of his theory is that inten-
tionality is not analyzable in relational terms. However, if in the Investigations
Husser! holds that intentionality is in some sense relational insofar as con-
sciousness is "of" or "about" something, then the analysis he has offered is
simply incomplete. Ideas, in fact, offers a further analysis.
In Ideas, we have seen, Husserl proposes a different entity to play the role
of ideal, intentional content. There is no reason to suppose he has ceased to
believe an act has an intentional essence, or type, but he no longer simply
identifies an act's intentional content with its intentional essence. The ideal
content he now calls the act's noema, and he conceives it as a kind of mean-
ing, a kind of abstract particular rather than an experience type. In §88 he
says, "Every intentional experience ... is noetic; that is to say, its essence is
to harbor in itself something like a 'Sinn''', that is, a noema (pp. 218-19).
The essence of which he speaks is clearly in effect the act's "intentional
essence" - he is saying as much. Thus, he is offering a further analysis or
explication of intentional essence: an experience is intentional, or has an
intentional essence, just insofar as it "harbors" a noema. And so my experi-
ence of seeing this black crow has a certain intentional essence or type, i.e.,
is an intentional experience, just because it has a certain noema.
The highlights of Husserl's mature theory of intentionality in Ideas we
might summarize as follows. Intentionality is analyzed in terms of an act's
real and ideal content: the real content of an act includes the act's noesis;
the ideal content is the act's noema, which centrally includes a Sinn. By
virtue of its noesis, each act bears a characteristic relation to a unique noema,
and so to the Sinn in its noema. Husserl says the noema is the "correlate" of
the noesis; and of the relation between the noesis and the Sinn he says that
the noesis "gives" the Sinn, or that the noesis "bestows" the Sinn on the act.
HUSSERL'S THEORY OF INTENTIONALITY 143
Let us say instead, using a neutral term, that the act entertains its noema, and
specifically its Sinn. Further, a Sinn bears a characteristic relation to an object
(to at most one existing object), inasmuch as it is the Sinn's intrinsic nature
to "point to", to "represent", to "present" that object; let us say a Sinn
prescribes an object. The intentional relation of act to object is then analyzed
as the composition of two relations, the relation of act, or noesis, to noematic
Sinn (the "entertaining" relation) and the relation of Sinn to object (the
"prescribing" relation):
an act intends ( is directed toward or is intentionally related to) an object
if and only if the act (or its noesis) entertains a certain noematic Sinn and
that Sinn prescribes that object.
Schematically:
act (noesis) noema (Sinn) - - -..., [object]
entertains prescribes
intends
In this way, noematic Sinn mediates intention, and thus an act is directed
toward its object "by virtue of" its Sinn.
Hussed's basic ontological analysis of intentionality, then, is in terms of
two "entities" - a noesis, which is a temporal phase of an experience, and a
noema, which is an abstract entity correlated with the experience - and two
relations - the relations of "entertaining" and "prescribing". (The object
intended in an act is not part of this ontological analysis, we know, since
Hussed holds that its existence is not necessary to an intentional relation;
hence the brackets in the schema above.) To derive further details of Husserl's
theory of intentionality, one must pursue several questions concerning this
basic analysis. If we assume Husserl's ontology, what explanations of the tradi-
tional problems of intentionality does his analysis provide? We pursue this
question in the next section below. Concerning the ontology itself, precisely
what are noeses and noemata, and just what are these relations we have called
"entertaining" and "prescribing"? Now, except for noting more explicitly the
role of noesis and noema in intentionality, we have already said (in Part 2
above) much of what Husserl tells us about these entities. In fact, we have
nothing further to add on noesis. On the noema (and especially the noematic
Sinn), however, we shall pursue further Hussed's characterization of it as
"meaning" or "sense" and we shall find that Husserl does have more to say
about the inner structure of noematic Sinne. These further discussions of noe-
mata will also allow us to give some further analysis of the relation between
Sinne and the objects they prescribe. (These are the topics of Chapter N.)
144 CHAPTER III
The relation between an act, or its noesis, and its noema - what we call
"entertaining" - remains unanalyzed by Husserl, except for his claim that
for each phase in the noesis there is a corresponding phase in the noema.
However, to guard against a possible misunderstanding of Husserl's theory of
intentionality we must say something about what this relation is not: it is not
a species of intending, not an "intentional relation". On Husserl's theory,
noematic Sinne carry out the inner work of achieving intentional relations
and in this way are "mediators" of intention. But this does not mean that
noemata, or Sinne, stand between consciousness and its objects. Husserl does
not hold that Sinne are the proper or direct objects of consciousness and
"represent" external objects somewhat as words or pictures represent things.
His theory is not a species of "representationalism" in that sense, akin to
theories holding that we are properly or directly aware only of our own
"ideas", which in tum stand for or represent external objects. Indeed, Husserl
adamantly opposes all versions of what he calls the "fundamentally perverse
image- and sign-theories" (Ideas, §52, pp. 126-27; see also §43, §90, pp.
224-25, and LI, V, Appendix to §ll and §20, pp. 593-95). On Husserl's
theory, the noema - in particular, the Sinn - of an act is in no wise an object
intended in the act (nor is the noesis): the subject consciously intends the
object but not the noematic (or the noetic) content of the act; in fact, he
becomes explicitly aware of the content only in phenomenological reflection.
Indeed, if Husserl had explained the intention of one object in terms of the
apprehension or intention of another his theory would face an infmite regress:
if intending an object required intending a Sinn that represents that object,
then intending that Sinn would require yet another Sinn that represented the
first Sinn, and so on ad infinitum. 20 Husserl himself offers this very criticism
against "image- and sign-theories" and so rejects the view that the intention
of an object consists in the intention of some other entity that in turn repre-
sents that object (Ideas, §90, pp. 224-25;LI, V, Appendix to § 11 and §20,
p. 594). Thus, whatever "entertaining" a noema or Sinn may be, to entertain
a Sinn is not to intend it in any way and does not require any explicit aware-
ness of the Sinn by the subject who so entertains it.
To admit that there are limits to Husserl's analysis of the fundamental
notions underlying his theory of intentionality is not in itself a criticism of
Husserl. Ultimately, analysis must always stop somewhere, and every theory
must at that point simply accept some notions as primitive. The important
question is whether what Husserl says about his basic notions is enough to
enable us to understand them and the way they work. We have tried in this
chapter to make it clear what these basic notions are and, in the case of noesis
HUSSERL'S THEORY OF INTENTIONALITY 145
and noema, to say as much as we can about what kinds of entities they are,
what components they comprise, and what roles they play in intentionality.
In Chapter IV we shall continue this endeavor with respect to the Sinn-
component of the noema, the component that plays most prominantly in
Husserl's theory of intentionality. The basic structure of Husserl's theory of
intention via noematic Sinn should at any rate now be clear. As with any
theory, much of its acceptability turns on its success in handling the problems
that prompted the need for a theory in the first place. With this in mind, we
turn now to the application of Husserl's theory to the fundamental problems
of intentionality that we have detailed from the beginning of the book. We
may thereby begin to see how the theory works, although our final assess-
ment must await the further developments of Chapter IV.
(2) The noesis in an act entertains exactly one noema, which consists
of a the tic part and a Sinn.
So every act, by virtue of its noesis, entertains a noema and hence a Sinn.
And this relation of entertaining is a many-one, or functional, relation:
(3) Different noeses, and hence different acts, may entertain the
same noema.
Thus, different experiences may share the same ideal content (whether ideal
content be type, as in the Investigations, or noema, as in the Ideas theory).
The anti-psychologistic status and the transcendental status of noemata
and Sinne are secured respectively by these two further principles:
In the next chapter we shall see that Hussed conceived noemata, or Sinne,
as abstract meaningo{:ntities of the sort that can in principle be expressed in
language.
We have emphasized Husserl's distinction between the content and the
object of an act, and identified noesis and noerna with real and ideal content.
Thus:
(6) The noesis and the noema (and hence the Sinn) of an act are
distinct from the object intended in the act.
Yet there is an intimate relation between an act's noema and its object,
owing to the "presentative" or "prescriptive" character of noematic Sinne.
(We shall explain the "or"-part in a moment.) Further, the relation of pre-
scribing is a many-one, or functional, relation:
For instance, the concepts "the morning star" and "the evening star" are
concepts of the same planet, Venus.
It is through the prescriptive character of an act's Sinn that the act is
intentionally related to its object:
HUSSERL'S THEORY OF INTENTIONALITY 147
(9) The object of an act is the object prescribed by the act's Sinn.
More fully:
Thus:
purely an internal feature of the act and its Sinn, and not a true relation at
all. Where an act's Sinn correlates with an existing object, that object is what
is intended in the act; but where it does not, talk about the act's "object"
is simply improper. This view was certainly attractive to Husserl. He seems to
have affirmed it in the Investigations, where his theory of intentionality had
not itself developed beyond the "adverbial" phase (see Section 3.1 above):
The sentences 'The ego presents an object to itself', 'The ego is related in a presenting
way to an object', 'The ego has something as an intentional object of its presentation'
... mean the same as 'In the phenomenological ego ... a certain experience, which in
virtue of its specific nature is said to be a "presentation of the object", is really present'.
Just so, the sentence 'The ego judges about the object' means the same as 'Such and
such an experience of judging is present in the ego', etc. (LI, V, § 12, p. 561; with trans.
changes. Cf. § 11.)
In two of his eadier writings, his 1894 essay on 'Intentional Objects' and his
1896 review of Twardowski, Hussed took the same position. There he argued
that all talk of "merely intentional objects" should be understood as "figura-
tive" or "improper" (uneigentlich), serving merely to "mark off a certain
function in all presentations, ... one that typically remains the same whether
the related existential judgment is in addition valid or not". 22 However, in
later writings Hussed sometimes appears more sympathetic to the notion
of non-actual objects. In Ideas he says: "In the broader sense an object -
'whether it is actual or not' - is 'constituted' in certain connections of con-
sciousness ..." (§ 135, p. 332). And in Experience and Judgment he says:
" 'The same' object which I just now imagine could also be given in experience
[Erfahrung, i.e., perception] : this same merely possible object ... could also
be an actual object" (Appendix I, to §§40 and 43, p. 381). In Chapter VI
below, we shall see how a view of this kind might be developed in Husserlian
terms; however, we shall conclude that Hussed's commitments in this direc-
tion are unclear and perhaps remain consonant with his attitude up to the
time of the Investigations. Accordingly, we leave Husserl's position on this
point unsettled, and we leave principle 7 in a form that takes no stand on this
point.
The conception-dependence of intention is explained by the role of Sinn
in mediating intention. From principles (9*) and (8) it follows that acts may
intend the same object through different Sinne. So intention is relative to a
particular Sinn. But a Sinn just is a "sense" or "conception" of an object,
understood in Hussed's non-psychologistic way. In Hussed's terms, a Sinn
reflects a particular aspect of the object it prescribes, a particular "way" in
which the object is "determined", or propertied - if you will, a mode of
HUSSERL'S THEORY OF INTENTIONALITY 149
presentation of the object. (We shall discuss this point more fully in Chapter
IV, Section 3.1, below.) Thus, intention is relative to a conception of the
object intended, that is, relative to a Sinn. And so different intentional rela-
tions are achieved in acts that intend the same object through different Sinne.
Husserl's theory also accommodates the fact that many objects "transcend"
the experiences in which they are intended, in that they outrun the content
of these experiences. A Sinn that prescribes a transcendent object, especially
a natural object, reflects only a limited and finite "way of being determined",
or aspect, of the object prescribed. Moreover, the content of an act intending
a natural object such as a tree entails that the object itself transcends this
limited aspect. When I see a tree, some properties or aspects of the tree are
explicitly presented in the perception ~ some with intuitional fullness (such
as the color of the leaves on the front side) and some without (say, the color
of the leaves on the back side). But my experience also presents the tree as
having other aspects or properties that are not specified at all; these might
include some visible qualities of its back side and properties such as having
taken seed or been planted at a certain time. "A thing [Ding: material thing]",
Husser! says, "can in principle be given only 'in one aspect' ['einseitig': 'one-
sidedly']" (Ideas, §44, p. 100). Moreover, this transcendence is intended in
the act, prescribed by the Sinn of the act (see Chapter IV, Section 3.1, below).
I see "this tree with green leaves on this side and so forth". "The 'and so
forth''', Husserl observes, "is an ... absolutely indispensable element in the
thing-noema" (Ideas, § 149, p. 367). Not only physical objects but also per-
sons and, of course, the natural world as a whole are transcendent of con-
sciousness (cf. Ideas, § §44, 47, 53, 149). So the principle of intended tran-
scendence will extend to this whole range of transcendent entities. This
principle will loom important and receive further explanation in our study
of Husserl's intriguing notion of "horizon" in Chapter V.
That natural objects are intended as transcendent is a vital part of their
being intended as objective. Another part is their being intended as experi-
enceable in other acts. Where I see an object as "the same object" throughout
a continuous series of perceptions, the object is given at one moment as being
the same object that I saw a moment earlier. Such "identification" Husserl
deems "the fundamental form of synthesis" (eM, § 18); through such iden-
tification, a particular object is "constituted" in one's consciousness. I may
also see an object as "the same object" you are seeing, or as an object "ex-
perienceable" by others (a main theme in eM, Fifth Meditation). Thereby,
an object is presented as intersubjective, and intersubjectivity is a large part
of objectivity (cf. eM, § §55ff.). Acts directed toward, or intending, the same
150 CHAPTER III
object we may say are co-directed. It is, of course, in virtue of their Sinne
that two acts may be co-directed. According to principle 8, different noematic
Sinne may prescribe the same object. And from this and principle 9 it follows
that acts may be co-directed: different acts - performed by the same or by
different subjects and having the same or different Sinne - may intend, or be
intentionally related to, the same object.
This principle of Husserl's theory will be a pivotal point for the last four
chapters of the book. In Chapter IV, Section 3.1, we shall study the internal
structure of a Sinn that allows for co-directedness of acts.
The preceding principles articulate Husserl's basic theory of intentionality,
a theory of intention mediated by noematic Sinn. It is worth recalling that
Husser! meant the theory to apply to both direct-object acts and propositional
acts. A direct-object act, e.g., my seeing this toad here on the ground, is
directed via an "individual" sense to an individual object. (Cf. Ideas, § §88,
103, 131.) A propositional act, e.g., my judging that this toad is sleeping, is
directed via a propositional Sinn to a state of affairs. The noematic Sinn of
a judgment, Husser! says, is what has been traditionally called "a 'judgment',
or a proposition in the sense of pure logic" (das "Urteil", bzw. der Satz im
reinlogischen Sinne) (Ideas, §94, p. 235). In this case, my judging is "about"
the toad and is "of" a state of affairs, the toad's being asleep.
There remains an important problem of intentionality that we have not
addressed in this chapter: the problem of explaining the difference between
definite and indefinite (Le., de re and de dicto) intentions. A discussion of
Husser!'s theory vis-a-vis this problem will be a major concern in Part 3 of
the next chapter, however.
NOTES
1 Izchak Miller has treated these matters quite clearly in his dissertation, 'The Phe-
nomenology of Perception: Husserl's Account of Our Temporal Awareness', UCLA,
1979, pp. 38-73.
2 Husserl's view of the ego changed radically from the Investigations to Ideas. In the
earlier work he had written: "I must frankly confess ... that I am quite unable to find
this ego .... The only thing I can take note of ... are the empirical ego and its empirical
relations to its own experiences ... " (LI, V, § 8, p. 549). But a footnote added to this
passage in the second edition of the Investigations notes the change: "I have since
managed to find it", Husserl says. Cf. Ideas, § §57, 80.
3 Kasimir Twardowski, On the Content and Object of Presentations, trans. by Reinhardt
Grossmann (Martinus Nijhoff, The Hague, 1977); first published in German as Zur Lehre
vom Inhalt und Gegenstand der Vorstellungen (Wien, 1894). Parenthetic page references
in this section are to the English translation.
HUSSERL'S THEORY OF INTENTIONALITY 151
statement "becomes objective to us in a reflex act of thought .... This logical reflection
is not an act that takes place only under exceptional, artificial conditions: it is a normal
component of logical thinking" (L/, I, § 34, p. 332). Of course, Husser! had not yet fully
developed his conception of phenomenological reflection in the Investigations. But even
long after Ideas, in Formal and Transcendental Logic, logical and phenomenological
reflection seem to coincide for Husser!, each being reflection on sense or meaning and
the former particular!y on senses of judgments.
15 Cited by F¢llesdal in Dagfinn F¢llesdal, 'Husserl's Notion of Noema', Journal of
Philosophy 66 (1969),684, reprinted in Dreyfus (Note 10 above).
16 Guido Kling cites a lecture of Husser!'s on "Bedeutungslehre" from the summer
semester of 1908 and also the manuscript 'Noema und Sinn', as well as a later letter to
Roman Ingarden that comments on the change: see Guido Kling, 'Husser! on Pictures
and Intentional Objects', Review of Metaphysics 26 (1973), 676, Note 11. Cf. 1. N.
Mohanty, 'On Husserl's Theory of Meaning', Southwestern Journal of Philosophy 5
(1974),229-33.
17 Cited by F¢llesdal (Note 15 above), p. 683.
18 A "Satz", in this usage of the term, is a noematic Sinn whose structure is proposi-
tional. Husser! also has a more special usage for the term, according to which the com-
plete noema of an act (whether the act is propositional or not) - the act's noematic
Sinn plus its thetic noema-component - is called a "Satz". See Ideas, § 133, p. 324, and
Dei/age XXV, p. 413; cf. Chapter IV, Section 2.7, below.
19 Husserl's basic analysis of perception is explained clearly in Dagfinn F¢llesdal, 'Phe-
nomenology', in Handbook of Perception. Vol. I, ed. by Edward C. Carterette and
Morton P. Friedman (Academic Press, Inc., New York, 1974), pp. 377-86.
20 Richard E. Aquila, in an otherwise illuminating paper contrasting Husserl's theory of
meaning and intention in the Investigations with that in Ideas, takes this criticism as
decisive in favor of the Investigations theory: see his 'Husser! and Frege on Meaning',
Journal of the History of Philosophy 12 (1974), 377-83.
21 See, for example, F¢llesdal's 'Phenomenology' (Note 19 above), p. 378.
22 'Besprechung von K. Twardowski' (Note 4 above), p. 353, Note *.
CHAPTER IV
a Sinn relates an act to its object. Exploiting the connections between inten-
tion and linguistic reference, we here draw in part on contemporary theories
of linguistic reference. We find, however, that some aspects of intention
depend partly on the context in which an act takes place and so seem not
to be accounted for by an act's Sinn alone. Inasmuch as contextual factors
influence intention, we see serious problems for Husserl's basic theory of
intention via Sinn and, more generally, for his commitment to a purely
phenomenological analysis of intention.
the same as Frege's notion of meaning or sense (Sinn), and that noematic
Sinne play essentially the same role in Husserl's theory of intentionality that
senses play in Frege's theory of reference. l F~l1esdal's perspective on Husser!
connects Husserl's concerns with those of Frege and others in the analytic
tradition, but we shall attempt to show that this perspective comes straight
from Husser! himself and is integral to his phenomenological approach to
the theory of intentionality. Accordingly, the focus of our discussion of the
relation between noematic Sinn and linguistic meaning (in Part 2 below) will
not be on Frege per se but on Husserl's own theory of linguistic meaning and
reference and on his account of the relation between language use and the
intentional mental activities that underlie it. Further, our interpretation
stresses the point that noemata and Sinne are phenomenological contents of
consciousness at least as heavily as it stresses F~l1esdal's main point of em·
phasis, that they are meanings.
Let us be clear, then, about the relation between our two interpretive
claims, as we see it. First, the legitimacy of the first claim is in no way de-
pendent on the second, although each is reinforced by the other. The his-
torical and textual grounds on which we have argued that Husserl takes
noemata and Sinne to be contents, not objects, of consciousness stand on
their own, no matter what the relation of these contents to language and
linguistiC meaning might be. Second, the first claim alone is sufficient to
distinguish our interpretation of Husserl's notion of noema, and its role in
intentionality, from the alternatives with which we are familiar. If noemata
and Sinne are contents, rather than objects, of consciousness, then they are
not intermediate objects that we intend in place of ordinary entities (not,
for example, sense-data or Meinongian incomplete objects) nor are they
parts or aspects or essences of the objects that we do intend. Third, a cor-
rect understanding of our second interpretive claim is very much dependent
on having accepted the first. Indeed, it is absolutely crucial to our inter-
pretation of Husserl that the claim that noemata and Sinne are meanings
be understood only in conjunction with the claim that they are contents,
not objects, of acts. We do not claim - nor does F~llesdal - that Husserl
takes linguistic meanings, or any other kinds of intensional entities, to be
in any sense the objects of first-order (Le., non-reflective) acts of conscious-
ness - whether these be propositional acts or attitudes, such as judgments
or beliefs, or direct-object acts of perception, memory, etc. Thus, we take
Husserl to disagree with those philosophers of language who have postulated
propositional senses as objects of the propositional attitudes, and it is surely
not our view or Husserl's that abstract meaning entities are the objects that
156 CHAPTER IV
"the noema of perception", he says, is "the object such, exactly such and
only such, as the perceiving subject is aware of it, as he intends it in this
concrete experienced mental state".4 Second, while the object as intended is
not the same as the object itself, Gurwitsch maintains that it is still a part or
aspect of the object. Husserl emphasizes that the material object itself is
perceived, while Gurwitsch holds that the noematic Sinn is perceived, inas-
much as it is the "perceived as such". Gurwitsch resolves the apparent conflict
by concluding that the Sinn cannot be wholly separate from the object per-
ceived. He takes it to be that aspect of the object which "appears" in the
perception, from the perceiver's perspective. 5 Third, accordingly, Gurwitsch
takes the object itself to be nothing but a system of noemata, consisting of all
the noemata that could present that same object. In particular, the object of
a perception is the system of perceptual noemata, or "appearances", of "the
same" object from different sides. Gurwitsch says: "The thing perceived also
proves to have noematic status. As a noematic system it is a noema itself, but
a noema of a higher order, so to speak."6 So the relation between the Sinn
and the object of an act is a relation of part to whole, or of a member-noema
to a system of noemata. Fourth, for Gurwitsch, phenomenological reduction
effects a change in attitude toward what is intended: "The phenomenologist
is not concerned with objects as they really are, but as they appear through
acts of consciousness and present themselves to the experiencing subject's
mind". 7 Phenomenological reduction then does not effect a change in what
is intended in the way we have described: for instance, a change from perceiv-
ing a tree to reflecting on the perception and its content, which makes it that
perception of that tree.
The theory Gurwitsch ascribes to Husserl faces some internal conceptual
difficulties, especially in regard to the relation it imputes between the noema-
tic Sinn and the object of an act. For Husserl, of course, the Sinn will exist
even if the object does not. As he says of perceiving a tree: "The tree can
burn, can break down into its chemical elements, etc. The Sinn [the perceived
tree as such], however, ... cannot burn, it has no chemical elements, no
powers, no real [realen] properties" (Ideas, §89, p. 222). Gurwitsch readily
acknowledges this point, but the theory he imputes Husserl then confronts
two difficulties. First, how can the Sinn be the object as intended if this is a
part of the object intended? Can the "perceived material thing as it presents
itself", the thing's parts or aspects that appear to the perceiver, exist even
when there is no such thing? Second, how can the object itself - which may
be a concrete object like a tree, which can burn away - be identical with a
system of noemata or Sinne, or a higher-order noema or Sinn? If a Sinn cannot
HUSSERL'S THEORY OF NOEMATIC SINN 159
burn away, how can a system of Sinne, or a higher-order Sinn, burn away?
By Hussed's own account, trees are concrete physical objects, composed of
chemical elements and having "real" properties, while Sinne are ideal or
abstract entities, having no chemical elements and no real properties; aren't
trees and noematic Sinne or systems of noematic Sinne thus entities of irre-
ducibly different ontological kinds?
A full discussion of these conceptual difficulties would involve the ques-
tion of whether and to what extent Hussed tended to idealism, and perhaps
they are not insuperable. But our interpretation of Hussed's account of the
relation between noematic Sinn and object does not encounter these prob-
lems. On our interpretation, the Sinn is in no way a part or aspect of the
object intended. And the object is neither a complex of Sinne nor a complex
or higher-order Sinn. Husserl does indeed describe a strict correlation between
any object and the set of noematic Sinne that present that same object in
different ways (cf. Ideas, § 135). But nowhere, to our knowledge, does he go
on to identify the object and the corresponding system of Sinne. The purpose
of the correlation, we hold, is not ontological reduction but a special sort of
phenomenological explication. (We study this kind of explication in Chapters
V and VI below on Hussed's notion of "horizon" or "manifold".)
In our view, though, the fundamental problem for Gurwitsch is that his
interpretation does not square with Husserl's own account of noema as ideal
content. We studied Husserl's distinction between content and object of
consciousness at length in the previous chapter. There we found that Husserl
contrasted two notions of content - real content, which he identified with
noesis, and ideal or intentional content, which he identified with noema or
Sinn - and he distinguished both from the object intended in an act. Gur-
witsch's interpretation, however, really offers one notion of content - real
content or noesis - and two notions of object - the object as intended,
which is identified with the noema or Sinn, and the object itself. For, we take
it, on Gurwitsch's understanding of the Sinn as the intended as such, the Sinn
is itself intended. The Sinn then lies on the object-side of intention, not on
the side of content in the sense in which we have studied it.
As a consequence, Gurwitsch's interpretation is difficult to square with
Husserl on other points as well. First, Husserl consistently characterizes the
Sinn, and indeed the whole noema, as a sense or meaning, which he takes to
be an abstract, or ideal, entity. So does Gurwitsch, but he must then take
meanings to be intended objects, just as they are given in acts of perception,
memory, judgment, and so on. Husserl's notion of meaning is quite different:
for him, meanings - including noematic Sinne - are primarily contents of
160 CHAPTER IV
We would note, first, that Hussed does sometimes use such expressions as
'the intended as such' and 'the object, as it is intended' in what is evidently
just Gurwitsch's way, but in contexts that support our interpretation of the
Sinn. The clearest instance is in Logical Investigations, V, § 17, where Hussed
says:
We must distinguish, in relation to the intentional content taken as object of the act,
between the object, as it is intended [der Gegenstand, so wie er intendiert ist), and
simply the object which is intended [schlechthin der Gegenstand, welcher intendiert
istJ . ... E.g., the presentation "German Emperor" presents its object as an Emperor,
and as Germany's Emperor. The man himself is the son of the Emperor Frederick III,
the grandson of Queen Victoria, and has many other properties here neither named nor
presented. (Pp. 578-79; with trans. changes.)
Gurwitsch himself sometimes refers to this very passage when explaining his
characterization of the noematic Sinn as the object "as intended". When
viewed in context, however, the passage fails to support this characterization,
and what in fact emerges is a strong argument against Gurwitsch's interpreta-
tion. Investigation V, we recall from Chapter III above, is Hussed's discussion
of content; § 17 considers, in particular, the use of the term 'content' to refer
to the object intended in an act and the ambiguity in that usage. (Some of
Hussed's contemporaries - notably Brentano - had used the terms 'content'
and 'object' interchangeably.) But this object-notion of content - whether
taken to mean the object in itself or the object as intended - is not the notion
that Hussed embraces and later develops into the doctrine of noesis and
Roema. He closes § 17 by saying: "Since such talk is so highly ambiguous, we
shall do well never to speak of an intentional content where an intentional
object is meant, but to call the latter the intentional object of the act in
question" (p. 580). § 20 then proceeds with Husserl's notion of content,
analyzing content in terms of "quality" and "matter". But "matter", which
he later correlates with the noematic Sinn (see Ideas, § 133, p. 324), is not
the object or the object as intended. Rather, it is that in an act which pre-
scribes or determines the object as it is intended: "The matter ... is the
peculiarity in the phenomenological content of the act that determines not
only that the act apprehends the object but also as what it apprehends it ... "
(§20, p. 589; with trans. changes. See Chapter Ill, Section 2.2, above). In
short, Hussed distinguishes his notion of matter from the Gurwitschean no-
tion of the object as intended. And since in Ideas he identifies noematic Sinn
with (ideal) matter - as Gurwitsch agrees he does 8 - the Sinn is distinct
from the object as intended, if the latter is taken in Gurwitsch's sense. In
fact, Hussed had already made a distinction between Sinn and "object as
162 CHAPTER IV
It is phenomenologically false to say that the difference between the consciously grasped
content in perception, and the external object perceived (or perceptually intended) in it,
is a mere difference in the mode of consideration, the same appearance being considered
at one time in a subjective nexus (in the nexus of appearances related to the ego) and at
another time in an objective nexus (in the nexus of the things themselves). We cannot
stress sharply enough the equivocation which permits 'appearance' to refer not only to
the experience in which the appearing of the object consists (e.g., the concrete percep-
tual experience, in which the object itself seems present to us) but also to the appearing
object as such. The deceptive spell of this equivocation vanishes as soon as one takes
phenomenological account of how little of the appearing object as such is really to be
found in the experience of appearing. The thing-appearance (the experience) is not the
appearing thing (that which seems to "stand before" us in embodied selfhood). As
belonging to the nexus of consciousness, appearances are lived through by us, as belong-
ing to the phenomenal world, things appear to us. Appearances themselves do not
appear, they are lived through [erlebt). (LI, V, § 2, p. 538; with trans. changes.)
Hussed says several interesting things in this passage. First, one does not grasp
the content of an act merely by conSidering the act's object in a special way
- i.e., by considering it "as intended" or "as it appears". Second, Husserl
distinguishes two notions of "appearance": the appearing object as such
(Gurwitsch's notion) and the inner experience of appearing. The former be-
longs with the object intended, in the "objective nexus"; the latter belongs
to consciousness. Thus, Hussed here identifies a phenomenological sense of
'appearance' - appearance as content - and explicitly distinguishes it from
Gurwitsch's notion. Third, as the content of a perceptual experience, an ap-
pearance is not something that appears. This point does not transfer directly
to our view of perceptual noemata or Sinne as contents, since Husserl is here
focusing on real content (the experience of perceiving - in effect, the noesis)
rather than ideal content. Nonetheless, if Husserl's later use of 'the perceived
HUSSERL'S THEORY OF NOEMATIC SINN 163
as such' in reference to the noema denotes ideal content, it should not seem
strange that the perceived as such, in that sense, is not itself something that
is perceived.
Evidently, when Husserl calls the noema or Sinn the "intended as such"
in Ideas, he is either not using the term 'intended as such' and kindred expres-
sions in the way he used them in Logical Investigations or he has radically
changed his view of content - so much so that ideal content and object have
now virtually coalesced. Now, we have already argued against the second
alternative in Chapter III. Although we discovered some important differences
between Husserl's earlier theory of content and his later theory of noesis and
noema, we argued that his basic distinction between content and object re-
mains unchanged; indeed, we saw that Husserl himself says that, except for
his failure to stress the distinction between real and ideal content in the
Investigations, noematic distinctions in Ideas correspond to content distinc-
tions in Logical Investigations. So we believe a systematic study of the theory
of content in the Investigations, coupled with the account of phenomenologi-
cal method, the noesis-noema doctrine, and the theory of intentionality in
Ideas, already tells us that the noematic Sinn is an immanent, ideal meaning-
content. Accordingly, Husserl's identification of the Sinn with the intended as
such is not the key to discovering what the Sinn is; in fact, the identification
is less informative about the Sinn than about Husserl's use of the expression
'the intended as such': 'the intended as such' denotes the noema and, hence,
the ideal content of an act.
This way of understanding Husserl's use of expressions like 'the intended
as such' in Ideas is not at all implausible. For one thing, in On the Content
and Object of Presentations Twardowski notes that such expressions had
already been used by other philosophers to denote the content as opposed
to the object of an experience. Expressions like 'the presented' or 'what is
presented' are ambiguous, Twardowski says, sometimes referring to the object
that is presented and sometimes to the content through which it is presented.
He then says, citing a publication from 1891: "Kerry tries to avoid the mis-
understandings which occur if one speaks of a 'presented' object without any
further explanation by distinguishing between the 'presented as such' and the
'presented plain and simple' ".9 Husserl's terminology is often just this. And
he would undoubtedly have been aware of its use in distinguishing content
from object, since he reviewed Twardowski's book in 1896. Twardowski's
commentary on the terminology also helps explain how Husserl could have
used the same terms for different notions in Logical Investigations and Ideas.
Since 'the presented' is already ambiguous, Twardowski says, to speak of
164 CHAPTER IV
illuminating to contrast the theory we have attributed Husserl with the theory
Gurwitsch attributed him. The natural development of the Gurwitschian read-
ing would seem to be a theory of the kind found in Meinong, for Meinong's
'incomplete' objects are objects restricted in their nature to a limited group
of properties (cf. Chapter II, Section 2.3). Indeed, the approaches of Meinong
and Husserl are the two principal approaches to intentionality we have dis-
cussed, the "object"-approach and the "content"-approach. It is important
to see both the parallels and the differences, and the principal point of com-
parison and contrast is that between Meinong's "incomplete" objects and
Husserl's "Sinne".n
For Meinong, we would recall, some objects are "complete", or "com-
pletely determined", while others are "incomplete". According to Meinong,
complete objects, including everyday physical objects, lie beyond human
grasp because we can grasp only a finite and partial aspect of any such ob-
jects. We can intend physical objects only indirectly insofar as we intend
incomplete objects that are "embedded" in them. Thus, what we directly
intend is an object with only a limited nature. This object is embedded in a
complete physical object, from which it differs only in having a finite subset
of the latter's properties. Indeed, we might say in extension of Meinong that
the incomplete object is itself the complete object "as" having just the prop-
erties or aspect we grasp of it: the incomplete object is then the complete
object "as" intended. By virtue of intending this incomplete object, and
its embedment in the complete physical object, we indirectly intend the
complete object. And so we might say that, in this sense, the physical object
is intended through the incomplete object.
There is, then, a certain structural resemblance between the Meinongian
and the Husserlian theories, inasmuch as each analyzes the intention of a
physical object in terms of two relations and an intermediary entity. For
Meinong, awareness of a physical object consists in intending an incomplete
object, which is embedded in the physical object; for Husserl, it consists in
entertaining a Sinn, which prescribes the physical object. There are important
differences, however, in the intermediary entities and in the relations involving
them.
First, an act's Sinn is the ideal content of the act and so, in an important
sense, is mental or "immanent" in consciousness, whereas an incomplete
object is in no way mental or immanent. The Sinn is, as it were, a part of the
act, while the incomplete object is a part of the intended object. Second, the
Sinn genuinely mediates an intentional relation between act (or subject) and
object, inasmuch as the Sinn prescribes the object and, so to say, "points"
HUSSERL'S THEORY OF NOEMATIC SINN 167
the act's intentional essence, its essence or type qua intentional experience.
In the Investigations, we saw, the ideal or intentional content of an act is
identified with the intentional essence of the act. But in Ideas this notion of
content as act-essence is replaced by the notion of content as noema. As an
intentional experience, an act still has an intentional essence, which now
includes the property of "entertaining" a noema as ideal content. But the
noema itself is a kind of abstract particular correlated with the act and not
an essence or property instantiated in the act. (See Chapter III, Sections 2.3
and 3.1 , above.)
That noemata are not essences of acts still leaves open the possibility that
they are, or include, essences or properties of the objects that acts intend.
After all, noemata and essences are both abstract entities for Husserl, and
both are ontologically independent of the objects to which they relate. If
I think of Pegasus, for example, the noema of my act exists and so does the
essence "flying horse", even though Pegasus himself does not. Nonetheless,
according to Husserl, the noema and the meanings or senses it comprises are
distinct from any essences of the object and from the properties that defme
these essences. In the third volume of Ideas he explicitly warns against iden-
tifying noemata with the essences of intended objects:
Noema (correlate) and essence are not to be confused. Even the noema of a clear thing-
intuition, or of a continuous harmonious connection of intuition directed upon one and
the same thing, is not and also does not contain the essence of the thing. The grasping
of the one is not that of the other, although here a change of attitude and direction of
grasp is essentially possible, through which the grasping of the noema can at any given
time change into that of the corresponding ontic essence. But we have a different kind
of intuition in the latter case than in the former .
. . . Just as the meant simpliciter [Bedeutete schlechthin) ... is something other than
the meaning [Bedeutung), so also the essence of the meant is something other than the
meaning. (Ideas, III, § 16, p. 85.)
The noema, through the constituent senses in its Sinn, prescribes the object
as having certain essences or properties, while the essences or properties them-
selves are instantiated, or exemplified, by the object. Thus, when I see a red
ball the noematic Sinn of the act includes the senses "red" and "ball", which
prescribe the object as having the properties of being red and being a ball.
But the senses and the corresponding properties are not the same, and they
do not -stand in the same relation to the object. (2) As Husserl notes in the
passage cited above, noemata and essences are grasped in different kinds of
intuition. (3) The study of noemata and the study of the essences of material
objects are distinct disciplines, according to Husserl. Noemata are subject
matter for transcendental phenomenology, while the essences of material
objects are subject matter for "eidetic ontologies". Indeed, we saw earlier,
these essences and their ontologies are "bracketed" in phenomenological re-
duction. (See Chapter III, Section 1.2.) (4) As evidenced by all three of these
points, noemata and essences are different kinds of entities for Husserl. Noe-
mata are ideal contents of intentional experiences, grasped in phenomenologi-
cal reflection, and so are immanent, transcendental entities. By contrast, the
essences or properties of physical objects are transcendent, just as physical
things themselves are transcendent. From the highest genus "Thing" all the
way down to the most specific type or kind to which a thing belongs, Husserl
holds, the species of a material thing is transcendent (Ideas, § 149). Thus,
while the noema of an act belongs to the experience as part of its transcen-
dental structure, the essences or properties of the act's object belong to the
thing intended and are in no way a part of the experience itself. (5) An es-
sence, like an individual, can be the object of an act, intended via a noematic
Sinn that prescribes it. But transcendent essences, like transcendent individ-
uals, cannot be completely grasped in a single act of consciousness. Accord-
ingly, one and the same essence can be intended in different acts through
different noemata with different Sinne, each Sinn prescribing the essence in
a somewhat different way. (Ideas, § 149; cf. EI, § 83.) And so the essence
itself cannot be identical with any of the various Sinne that prescribe it.
Finally, (6) Husserl holds that there are "contradictory" (''widersinnig'')
noemata to which no essences correspond. If one thinks of the round square,
he says, the "thought-meaning" (Denkbedeutung) "round square" exists as
an entity in the realm of noemata, although "there is no essence 'round
square'" (Ideas, III, § 16, pp. 85-86).
It is noteworthy that Husserl's sharp distinction between meanings and
properties diverges from common practice in contemporary semantic the-
ory. At least since Carnap's Meaning and Necessity (1947), the meaning, or
170 CHAPTER IV
Our purpose in this part of the chapter is to establish, and to render more
precise, our claim that Husserl identifies intentional contents, or noemata,
with the meanings that are expressed in language. 12 In the first section we
focus on his conception of linguistic meanings as "ideal", or abstract, entities.
This ideality is the same notion, and undergoes the same changes, that we
earlier found in Husserl's characterization of intentional contents: his concep-
tion of the ideality of meanings, like that of intentional contents, changes
from the notion of act-essence or type to the Frege-like notion of abstract
particular. The second section discusses Husserl's views on the relation of
linguistic meanings to referents, comparing and contrasting them with Frege's.
Importantly, Husserl's views on this semantic relation are exactly the same
as his views on the relation of noematic Sinne to objects of intention. In the
third and fourth sections we discover the basis for these parallels between
meaning and reference, on the one hand, and Sinn and intention, on the
other: linguistic meanings and noematic Sinne are in fact the very same
entities for Husserl. Specifically, Husserl holds two important principles: (1)
that language is the expression of "thought", so that linguistic meanings are
just the noematic Sinne of underlying intentional acts or attitudes; and (2)
that every noematic Sinn is in principle capable of being expressed in lan-
guage, expressed as the meaning of some appropriate linguistic expression.
The last three sections discuss some refmements and applications of this
notion of the expressibility of noematic Sinne and also consider its applic-
ability to other noematic components.
HUSSERL'S THEORY OF NOEMATIC SINN 171
A man never has somebody else's mental image, but only his own .... It is quite other-
wise for thoughts [Gedanken; propositions 1; one and the sante thought can be grasped
by many men. The constituents of the thought ... must be distinguished from the
images that accompany in some mind the act of grasping the thought - images that each
man forms of things. 14
Such arguments were also familiar to HusserI through the works of Bolzano,
who even before Frege had made much of a distinction between "subjective
ideas" and "objective ideas" in semantic theory. IS HusserI came to recognize,
as he says in the foreword to the fIrst edition of Logical Investigations, that
psychologism cannot account for the objectivity, i.e., the intersubjectivity,
of logic and mathematics or of knowledge in general. Thus, HusserI rejects
a psychologistic view of meaning in Logical Investigations and in all his sub-
sequent writings, and in the Investigations he begins to seek a better account
than psychologism can provide of "the relationship ... between the subjec-
tivity of knowing and the objectivity of the content known" (LJ, p. 42).
It is also in Logical Investigations that HusserI begins to develop phenom-
enology as a study of objective meaning entities. Logic and semantics - as
part of phenomenology - are to study propositions and other meanings
rather than, as psychologism would have it, subjective psychological pro-
cesses. Part of the signifIcance of this de-psychologizing, HusserI believes, is
a change in the status of logical and phenomenological results. Unlike empiri-
cal psychology, phenomenology and logic are to be a priori studies of non-
contingent truths about certain abstract entities associated with the processes
of consciousness. HusserI's insistence that phenomenology, as a study of
meanings, should not be confused with introspective psychology is thus of a
piece with his anti-psychologistic approach to logic and semantics.
Hussed's own account of the objective nature of linguistic meanings, in
the fIrst of the Logical Investigations, is largely an exposition of a Bolzano-
Frege line. Meanings must be intersubjective entities, he argues, because
successful linguistic communication requires that different people express
and understand the same meanings - strictly, numerically, the same. The
meaning of an expression is thus "shared" by different speakers who utter
an expression and by various hearers who understand it. Hussed says:
HUSSERL'S THEORY OF NOEMATIC SINN 173
If we or others repeat the same sentence [Satz) with like intention [Intention], each
of us has his own phenomena, his own words and his own instances of understanding
[ Verstandnismomente). Over against this unbounded multiplicity of individual experi-
ences, is an identical element expressed in them all; it is the same in the very strictest
sense of the word. Multiplication of persons and acts does not multiply sentence-meaning
[Satzbedeutung); the judgement in the ideal, logical sense remains single. (LI, I, §31,
p. 329; with trans. changes.)
The essence of meaning [Bedeutung] is seen by us, not in the meaning-lending experi-
ence [bedeutungverleihenden Erlebnis], but in its "content", the one identical inten-
tional unity set over against the ... multiplicity of ... experiences of speakers and
thinkers. The "content" of a meaning-experience, in this ideal [idealen] sense, is not at
all what psychology means by a content, i.e., any real [realer] part or side of an experi-
ence. If we understand a name ... [or) a statement ... the meaning ... is nothing
which could, in a real sense, count as part of our act of understanding. (LI, I, §30, p.
327; with trans. changes.)
... What I mean [meine] by [a] sentence ... or (when I hear it) grasp as its meaning
[Bedeutung] is the same thing, whether I think and exist or not, and whether or not
there are any thinking persons and acts. The same holds of all types of meanings [Bedeu-
tungen 1, subject-meanings, predicate-meanings, relational and combinatory meanings,
etc. (LI, I, §31, pp. 329-30.)
It is thus clear that Husserl supports the view that Frege in 'The Thought'
puts as follows:
... Thoughts [Gedanken; propositions] are neither things of the outer world nor ideas.
A third realm must be recognized. What belongs to this corresponds with ideas, in
that it cannot be perceived by the senses, but with things, in that it needs no bearer to
the contertts of whose consciousness to belong. Thus the thought, for example, which
we expressed in the Pythagorean theorem is timelessly true, true independently of
whether anyone takes it to be true. It needs no bearer. It is not true for the fIrst time
when it is discovered, but is like a planet which, already before anyone has seen it, has
been in interaction with other planets. 17
that he there adopted for ideal act-contents (cf. Chapter III, Section 2.2,
above). He emphasized the "shared" character of meanings: the acts of con-
sciousness underlying a speaker's utterance and a hearer's understanding seem
to involve a common entity as meaning. Hussed thus assumed meanings to be
a kind of "species", or "universal objects", or "essences", which are instan-
tiated by such particular acts but which - in keeping with their ideality -
exist independently of their instantiations (cf. LI, I, §3l, p. 330). On this
view, meanings are properties or types shared by speakers' and hearers' acts
of intending the same object or the same kind of object, properties charac-
terizing them as directed toward these entities. But even at this point, Hussed
was careful to distinguish meanings, taken as universals instantiated by acts,
both from the objects of those acts and from related essences or properties
of objects. The property of being red, for example, is an essence of all red
objects; but the meaning "red", on this view, is a property of acts directed to
red things. IS
By the time of Ideas (1913), when Hussed had formulated his notion of
noema and thus refined the general notion of act-meaning (Sinn), he aban-
doned the view that meanings are act-essences, properties literally instantiated
by acts. There he adopted instead the view that meanings are abstract entities
correlated with acts and expressible by words but in no sense properties or
parts of ncts. Apparently he came to think of them as sui generis, perhaps
as a special sort of abstract particulars. (For documentation of this point see
Chapter III, Section 2.3, above.)
Since Carnap, meaning entities have come to be called "intensions" or
"intensional entities". Though various philosophers in the Fregean tradition
have chosen various entities to play the role of intensions, Frege's view that
meanings are abstract entities and that they mediate reference is paradigmatic.
And this paradigm provides good reason for our saying that Hussed also
considered meanings to be "intensions". For not only does Hussed share
Frege's view that meanings are abstract, "ideal", entities; as we shall see in a
moment, his view of their role in mediating the reference of expressions is
also basically Fregean.
The term 'intensional entity' is also suggestive of one of Hussed's own uses
of the term 'intentional object'. Hussed admits to using the word 'intentional'
in two quite different senses (vide LJ, I, §30, p. 327, n.l). Sometimes he uses
it so that 'intentional object' means the intended object, i.e., the object of
an act or the referent of an expression. At other times he uses it so that
'intentional object' means a meaning entity, specifically the noema or the
noematic Sinn of an act or the meaning of an expression. By the time of Ideas
176 CHAPTER IV
(1) The meaning and the referent of an expression are always distinct.
Husserl says: "Each expression ... not only has a meaning [Bedeutung] , but
refers to certain objects . ... But the object never coincides with the meaning"
(LI, I, § 12, p. 287).
"A word like 'Socrates' can only name different things by meaning [bedeutet]
different things, i.e., by becoming equivocal. Wherever the word has one
meaning [Bedeutung] , it also names one object" (LI, I, § 12, p. 288).19
(4) Different meanings may determine the same referent; thus, ex-
pressions with different meanings may refer to the same entity.
" ... The meaning [Bedeutung] itself can change while the objective reference
[Richtung ] remains fIxed" (LI, I, § 13, p. 289). "Names offer the clearest
examples of the separation of meaning [Bedeutung] and objective reference .
. . . Two names can mean [bedeuten] different things, but name the same
thing. Thus, for example, 'the victor at Jena' - 'the vanquished at Waterloo'
..."(§12,p. 287; with trans. changes).
(5) An expression is meaningful just in case it expresses a meaning,
even if there exists no entity to which the expression refers.
"Reference to the object is constituted in the meaning [Bedeutung]. To use
an expression meaningfully [mit Sinn] , and to refer expressively to an object
(to form a presentation [vorstellen] of an object), are thus one and the same.
It makes no difference whether the object exists or is fIctitious or even im-
possible .... One generally distinguishes objectlessness from meaninglessness"
(LI, I, § 15, p. 293; with trans. changes).
These points of agreement are both central and extensive enough to justify
our characterizing Husserl's semantic views as basically Fregean. But there
also ~eem to be some important differences between Husserl's and Frege's
views. (Since Husserl did not try to develop a formal semantic system his
position on some points is either unclear or unexpressed.) As we have already
noted, in Logical Investigations Husserl took meanings to be universals,
though his conception of their kind seems later to have become more like
Frege's. (And even the earlier view agrees with Frege that meanings are ab-
stract, non-psychological entities.) Husserl apparently does differ with Frege
about the sorts of entities that are to be taken as the referents of certain
kinds of expressions. There is some evidence that, at least early on, Hussed
took the referent of a predicate to be the objects that satisfy the predicate
rather than - as Frege would have it - the function, or "concept", that
determines that class of objects.20 And Husserl rather explicitly holds that
sentences stand for states of affairs rather than truth-values: 'that' -clauses
are used to name states of affairs, he says, while sentences used in ordinary
assertion, though not names of anything, are used to assert states of affairs
(LI, I, § 12, p. 288; V, §36, pp. 631-34).
178 CHAPTER IV
All expressions in communicative speech function as indications. They serve the hearer
as signs of the "thoughts" of the speaker, i.e., of his meaning-giving [sinngebenden)
mental experiences [psychischen Erlebnisse). (LI, I, § 7, p. 277; with trans. changes.)
The meaning-animated [sinnbelebten) expression breaks up, on the one hand, into the
physical phenomenon forming the physical side of the expression, and on the other
hand, into the acts which give it meaning [Bedeutung) ... (LI, I, §9, p. 280; with trans.
changes. Cf. also Ideas, § 124, pp. 303-304.)
Now, the passage just cited also tells us what it is for a person purposefully
to "express himself about something": certain of his acts of consciousness
"confer on" or "lend" (verleiht) his words their meaning (Sinn). These acts
Husserl variously calls meaning-giving acts (sinngebenden Akte; Akte welche
Bedeutung geben) or meaning-lending (sinnverleihenden, bedeutungverlei-
henden) acts (cf. LJ, I, § § 7, 9). "In virtue of such acts", he says, "the expres-
sion is more than a merely sounded word. It means [meint] something ... "
(LJ, I, §9, p. 280).
Husserl's metaphor of "giving meaning" is to be taken quite literally: the
meaning "given" the uttered expression in a speech act is just the noematic
Sinn of the "meaning-giving" act that "underlies" the speech act. In that
underlying act - as in acts of consciousness generally - we intend a certain
object or state of affairs, and we intend it via the act's noematic Sinn. This
intended object is what receives our primary attention in the speech act:
When we normally execute an expressing as such, we do not live in the acts which con-
stitute the expression as a physical object, our "interest" does not belong to this object;
rather, we live in the meaning~iving [sinngebenden) acts, we are exclusively turned
toward the objective (Gegenstiindlichen I that appears in them, we aim at it, we mean
[meinen: intend) it in the special, pregnant sense [i.e., attentively). (L/, V, § 19, p. 584;
with trans. changes.)
... In speaking we are continuously performing an internal act of meaning [act of mean-
ing = Meinen) , which fuses with the words and, as it were, animates them. The effect of
this animation is that the words and the entire locution, as it were, embody in themselves
a meaning [Meinung), and bear it embodied in them as their sense [Sinn). [Husser! here
footnotes L/, I.)
. .. [In) this act of meaning [Meinen) ... there is constituted ... the meaning
[Meinung) - that is, the Bedeutung, the Sinn - expressed in the locution. For example,
if we utter a judgment, we have effected, in union with the words of our assertive state-
ment, a unity of judging, of inwardly "thinking" asserting. No matter what other psychic
producings may also be effected, whereby the words themselves come about, ... we
shall pay attention only to what is fused on, namely the acts of judging that function as
meaning-giving [sinngebende I acts, i.e., that bear in themselves the judgment-meaning
[Urteilsmeinung) that finds its expression in the assertoric sentence. (Pp. 22-23; with
trans. changes. Our emphasis.)
The expressibility of noematic Sinne fmally makes good the claim that
they are intensions. Where we first saw that every Bedeutung is a Sinn ex-
pressed, we now see that every Sinn is expressible and hence (at least poten-
tially) a Bedeutung. In short, we have here just one class of meaning entities
- noematic Sinne - that playa role both in language and in acts of conscious-
ness generally. The intensional entities that get expressed in language and the
noematic entities that mediate the intentionality of acts are the very same
entities. And so HusserI himself can say, in Ideas, III: " ... The noema ... is
nothing more than the generalization of the idea of meaning [Bedeutung] to
the whole field of acts" (§ 16, p. 89).
never can all the particularities of the expressed be reflected in the expression.... Whole
dimensions of variability ... do not enter at all into the expressing Bedeuten; they, or
their correlates, do not at all "express themselves": so it is with the modifications of
relative clarity and distinctness, the attentional modifications, and so forth. (Ideas,
§ 126, p. 310.)
The clarity, distinctness, and attentiveness with which an object is given are
among the more particular and subjective features of acts, HusserI seems to
think, and apparently their noematic correlates are too idiosyncratic to be
expressed.
For roughly the same reason, any "intuitional" element in an act's noema
is not expressed when the act is brought to expression. The noema of an
intuitive act, such as visually preceiving an object, and the noema of a non-
intuitive presentation of the same object (merely thinking of it, for example)
may have the very same Sinn (see Ideas, §91; LI, V, §21). This Sinn-com-
ponent, in either case, will be expressible in language. But in the intuitive
act the object is sensuously given and so experienced with what HusserI calls
"intuitional fullness" (see Ll, VI, esp. § §21-29; cf. Chapter III, Section 2.6,
above). This "fullness" is reflected in the intuitive act's noema, but it is not a
part of the noema's expressible Sinn. That which is expressed when a percep-
tual act is brought to expression, Husserl says, includes only what is common
to the noema of the perception and the noema of a non-intuitive presentation
of the object: it is "the identical meaning [Bedeutung] that the hearer can
grasp even if he is not a perceiver" (Ll, I, § 14, p. 290; with trans. changes).
Thus, since the noema of a non-intuitive act has no "fullness" -component,
the Sinn, but not the fullness-component, of a perceptual noema is what is
expressed as a Bedeutung when expression is founded on an underlying
perception (cf. also Ll, VI, §28, p. 744).
Considerations of a different sort apply to the noematic correlate of an
186 CHAPTER IV
act's "thetic character" - that which varies with the kind or species of the
act, marking it as an act of perception, or memory, or whatever (cf. Chapter
III, Section 2.4, above). In Logical Investigations, VI, §2, Husserl effectively
maintains that when an act is brought to expression the thetic component
of its noema is not part of the meaning expressed. Husserl's point there has
nothing to do with "generality", however. It is simply that, for instance, in
expressing his judgment that the murderer is in this very room, what Holmes
expresses is the Bedeutung "The murderer is in this very room"; he does not
express the Bedeutung "I judge that the murderer is in this very room". For
the latter would be the Sinn of a different act, Holmes' act of reflecting on
his original judgment and judging that he had so judged.
So, for Husserl, only the Sinn of the act underlying an expressive utterance
is expressed. Noema-components corresponding to clarity, attentiveness,
intuitional fullness, thetic character, and any other ''ways of givenness" of
the act do not enter the meaning expressed.
Nonetheless, the thetic component (and perhaps others as well) does seem
to be expressible in a more indirect way. In Logical Investigations, VI, § § 2-
3, Husserl discusses different senses of "expressing" an act. When I judge that
p and I say "p", I express the Sinn but not the thetic component of the noema
of my judgment. When I say "I judge that p", though, I express the Sinn of
another judgment about my first judgment (cf. LI, VI, §2; Ideas, § 127, p.
313). Now, although Husserl does not explicitly say so, this second Sinn
includes both the Sinn and the thetic component of the noema of my first
judgment. And so my second utterance expresses, as a component of the
more complicated Sinn of my second judgment, the thetic component of my
first judgment's noema. In Husserl's primary sense of "expressing" an act,
my first utterance expresses my first judgment: it is this judgment that lends
its Sinn to the uttered words. And my second utterance expresses, in that
sense, my second judgment. But in Hussert's second sense of "expressing" an
act, my second utterance "expresses" my first judgment; and in this second
utterance both Sinn and the tic components of my first judgment's noema
appear in the Bedeutung of the uttered. words, for both are included in my
second judgment's Sinn, the Sinn that serves as that Bedeutung. In this way
the the tic component of my first judgment's noema is "expressed" as a
Bedeutung in my second utterance. Some further extra-Sinn noema-com-
ponents - e.g., attentiveness - probably ought also to be expressible in this
indirect way. If so, they along with the Sinn and thetic component would
be capable of serving as components of some Sinn and would be in that sense
expressible as linguistic Bedeutungen.
HUSSERL'S THEORY OF NOEMATIC SINN 187
This point, though Husserl does not formulate it himself, would assure
that some noematic components other than the Sinn are intensional entities.
Indeed, there is evidence that Husserl did conceive of the noema and all its
components - and not just the noematic Sinn - as meanings or intensions.
Of the Gegebenheitsweise-components Husserl says, "As characters of the,
so to speak, 'ideal' ['Ideellen'] , they are themselves 'ideal' and not real
[reell] " (Ideas, §99, p. 250). And Hussed sometimes uses the word 'Sinn'
to describe the complete noema. When, as is the rule, he reserves 'Sinn' for
the object-oriented component of the noema (strictly, the "objective Sinn"
or "noematic Sinn"), he suggests that the word 'Satz' ('proposition') would
appropriately describe the combination of the noema's Sinn and thetic com-
ponent (Ideas, § 133, p. 324). This terminology reinforces the interpretation
of the whole noema and its components as intensions, and we shall discuss
it further below (in Section 2.7). But we need not press this more extended
interpretation unduly, since our concern is mainly with the Sinn and its role
in intentionality.
Here the expression 'that hovering black crow' expresses as its meaning
(Bedeutung) the noematic Sinn of the perception, while 'I clearly see' ex-
presses the "way-of-givenness" component of the perception's noema (insofar
as this can be expressed). And so the act-sentence as a whole expresses the
act's whole noema. Of course, the sentence expresses the noema and its parts
only if the words chosen are appropriate to the experience. In particular, the
object-phrase 'that hovering black crow' must not merely describe the object
that is perceived but describe it exactly as it is perceived. Even if that hovering
black crow is in fact the Yaqui sorcerer Don Juan, it would be inappropriate
188 CHAPTER IV
What is the "perceived as such", what essential phases does it harbor in itself as this
perception's noema? We obtain the answer in pure surrender to the essentially given;
we can describe the "appearing as such" .... Just another expression for this is: "de-
scribing the perception in noematic respect". (P. 22l.)
It is clear that all these descriptive statements, though they can sound like statements of
reality, have undergone a radical modification of sense; just as the described itself ... is
something radically different, by virtue, so to speak, of an inverting change of sign. "In"
the reduced perception (in the phenomenologically pure experience) we find, as belong-
ing irrevocably to its essence, the perceived as such [i.e., the noematic Sinn), to be
HUSSERL'S THEORY OF NOEMATIC SINN 189
expressed as "material thing", "plant", "tree", "blooming", etc. The quotation marks
are obviously significant; they express that change of sign and the corresponding radical
modification of meaning of the words. The tree simpliciter, the thing in nature, is any-
thing but this perceived tree as such, which as perceptual Sinn belongs inseparably to the
perception. (Pp. 221-22.)
Thus, says Husserl, whereas expressions such as 'tree' and 'blooming' describe
the object perceived, which is a thing in nature, the addition of quotation
marks creates new signs - ' "tree" , and' "blooming" , - that have new mean-
ings and describe something radically different, namely, the noematic Sinn of
the act of perceiving this object.
A chief virtue of meaning quotation, as it has been used in contemporary
semantic theory, is that it reveals the structure of the meaning referred to.22
An expression enclosed in meaning quotes is taken to refer to the meaning
it customarily expresses. And if the expression is complex, the constituent
expressions are also taken to refer to their customary meanings, which are
the constituents of the complex meaning referred to by the whole expression.
Thus, '''the bard is inspired" , refers to the complex propositional meaning
that results from the composition of the nominal meaning "the bard" and
the predicative meaning "is inspired". Husserl's practice seems to conform
with this principle, although he does not himself formulate any precise
principles governing his special use of quotation. A noematic description for
the act of perception he discusses in § §88 and 89 of Ideas would presumably
include, in partial description of the noema's Sinn-component:
Husserl evidently holds that the constituent terms occurring between the
quotation marks - 'tree', 'blooming', etc. - refer to their customary senses,
which are constituent senses in the Sinn that the whole expression refers to.
For what these senses prescribe are the properties of being a tree, being in
bloom, etc., and these are precisely the properties that the act, by virtue of
its Sinn, intends its object as having. In the cited passage Husserl thus allows
the quotation marks to migrate inward, so that' "tree"', ' "blooming" " etc.
refer to components of the noematic Sinn under description. Similarly, the
fuller noematic description,
would reveal the structuring of the noema as a whole into its two main con-
stituents: the sense "I see", which belongs to the noema's thetic or ''way-of-
190 CHAPTER IV
giveness" component, the component that prescribes the act as being mine
and being an act of seeing; and the sense "this apple tree blooming in the
garden", which is the Sinn-component of the noema, the component that
prescribes the act's object as this apple tree blooming in the garden .•
Husserl frequently returns to this use of "noema quotation", as we might
call it, especially when distinguishing the structural components of noemata
and Sinne in Ideas. (It is noteworthy that he retains the device long after
Ideas as well. See, e.g., PP, §37, p. 145.) Thus, in §130 he distinguishes the
Sinn from the way-of-givenness component in a noema as follows:
... A living cogito ... has in a special sense "direction" upon an objectivity. In other
words, to its noema belongs an "objectivity" - in quotation marks - with a certain
noematic composition, which unfolds in a description with determined limits, namely in
one that as description of the "intended objective just as it is intended [vermeint] "
avoids all "subjective" expressions. [The expressions used there] all have their quotation
marks, and thus have the noematic-modified sense .... For the description of this in-
tended objective as such, expressions such as "perceptually", "recollectively", "cJearly-
intuitively", "in thought", "given" are excluded - they belong to another dimension
of description, not to the object that is consciously grasped [bewusst], but to the way
[Weise] in which it is consciously grasped. (Pp. 318-19.)
proposition [Satz] , i.e., the sense [Sinn] with its thetic character" (p. 286).
(He also acknowledges that the word 'proposition', or 'Satz', can mean the
actual state of affairs intended, as indeed many philosophers since Husserl
have used the term.) So Husserl uses the term 'proposition', or 'Satz', in two
different ways. In the narrow sense of the term, a proposition is the Sinn of
a propositional act, especially a judgment. In the extended sense, a proposi-
tion is- the Sinn plus the thetic noerna-component of any act, that is, the
whole noema of an act.
We can now pose an interesting question: If a noema is a proposition in
Husserl's extended sense, is it also a special type of proposition in the tradi-
tional sense? Consider an act whose phenomenological description, or descrip-
tion in phenomenological reflection, is:
I judge that the bard is inspired.
The Sinn ofthis act is the proposition
This second act is a second-order act directed upon the first act of judging.
The Sinn of this second act is the proposition
"I judge that the bard is inspired".
Our question is whether this propositional Sinn is the same noematic entity as
the noerna of the prior act of judging, whether the full noema of the act of
judging is identical with the Sinn of this act of supposing. Husserl does not
seem to consider the possibility that a proposition in the extended sense is
a special type of proposition in the traditional sense. He seems to assume that
adding a thetic component to a Sinn produces not a special type of proposi-
tional Sinn but a "posited" Sinn, a noema, as opposed to a mere Sinn.
Let us assume that a noema is indeed a type of proposition in the tradi-
tional sense, namely, an act-proposition, a propositional Sinn prescribing an
act with a certain phenomenological structure. We know an act entertains its
noema. But if an act's noema is an act-proposition, then qua proposition it
194 CHAPTER IV
also prescribes the act itself. (Recall Chapter III, Part 3, on the relations of
entertaining and prescribing.) For example, the noema entertained by my
judging that the bard is inspired is the act-proposition "I judge that the bard
is inspired", and this proposition prescribes my act of judging. So an act's
noema both prescribes the act and is entertained by it. Now, Husserl is right
in observing a difference between noemataand propositional Sinne. But the
difference is a difference not in noematic entities but in the roles of the same
noematic entity in different acts. The noema of my judgment is indeed the
proposition "I judge that the bard is inspired", but that proposition prescribes
my judgment only insofar as it serves as the Sinn of another act directed upon
my judgment, such as an act of phenomenological reflection. And then that
proposition is joined with a further thetic component, such as that of reflec-
tion. The result is the full noema of my reflection, my reflecting that I judge
that the bard is inspired. This noema is then itself a more complex proposi-
tion, the proposition "I reflect that I judge that the bard is inspired", which
may serve as the Sinn of yet another, higher-order act. Thus, we may sayan
act's noema is a proposition, an act-proposition, but it properly functions as a
proposition, prescribing the act, only where it serves as the Sinn in another
act's noema.
The assumption that a noema is a type of proposition is helpful because we
are accustomed to working with propositional senses, especially in semantic
theory. Indeed, recalling the previous sections, we can observe that noemata
are act-propositions expressible by appropriate act-sentences such as 'I judge
that the bard is inspired'. But noemata are expressible only indirectly. Recall
that the thetic component of an act's noema is expressible only indirectly as
we express the Sinn of a higher-order act directed upon the first act. Indeed,
if I form a description of an act as given in phenomenological reflection, a
phenomenological description of the act, then I form an act-sentence whose
sense - the proposition it expresses - just is the noema of the act described
by that sentence. So the assumption that a noema is a type of proposition in
the traditional sense fits well with Husserl's other views. Husserl does not
himself make that assumption, but that he could have done so is further evi-
dence of the internal coherence of his various doctrines about noemata and
their relation to language.
3.1. Husser/'s Account of the Structure of a Noematic Sinn: the "X" and the
"Predicate-Senses"
The Sinn-component of a noema is what determines an act's intentional rela-
tion to an object. Specifically, the Sinn determines which object the act is
directed toward, and it also determines what the object is intended "as". In
§ § 128-31 of Ideas Husserl offers an interesting analysis of the internal
structure of a noematic Sinn, an analysis that gives more detail about how
intention is achieved via Sinn.
Briefly, Husserl holds that each Sinn is a complex meaning-structure that
can be factored into two fundamental components: an aggregate of predicate-
senses, which prescribe the properties an object is given as having; and a com-
ponent of sense of a different sort, called an "X" or a "determinable X",
which prescribes the object to which those properties are ascribed in the act.
By virtue of its X, each Sinn relates to a specific object and so determines
what the act is directed toward; and by virtue of its predicate-senses, the Sinn
ascribes properties to this object and so determines what it is intended as.
Accordingly, through an act's noematic Sinn, a subject intends a specific
object as having certain properties or "determinations".
Husserl's discussion of the structure of a Sinn relies heavily on his notion
of noematic description and the special use of quotation marks it employs
(see Section 2.6 above). Indeed, the analysis emerges most clearly from a cer-
tain form that Husserl seems to presuppose for phenomenolOgical descriptions
196 CHAPTER IV
where' "object"', or' "x"', denotes the X in the Sinn and '''</>''', or '''is </>" "
denotes the aggregate of predicate-senses.
As the composition of these two components of sense, the whole Sinn is
HUSSERL'S THEORY OF NOEMATIC SINN 197
The Predicate-Senses
The predicate-senses in a noematic Sinn make up what Husserl calls the
"content" of the Sinn .
. . . An entirely fixed content [Geholt) is marked off in every noema. Every conscious-
ness has its what [Was) and each intends [vermeint) "its" objective; it is evident that for
each [act of) consciousness we must in principle be able to carry out ... a noematic
description of this objective "exactly as it is intended"; through explication and concep-
tual grasping we obtain a complete aggregate of ... "predicates", and these in their
modified meaning determine the "content" ["Inhalt") of the object-nucleus of the
noema in question. (Ideas, § 130, pp. 319-20.)
What Husserl here calls the "object-nucleus" of a noema is just the Sinn-
component, i.e., the component that relates the act to its object (see Ideas,
§ §99, 129). The "predicates" in a Sinn are unfolded in a noematic descrip-
tion of the "intended objective just as it is intended" (cf. p. 319, two para-
graphs earlier). Husser! explicitly cites the modification of meaning brought
about by the quotation marks; they signify that it is noematic constituents,
or senses, that are being referred to or described. Specifically, the "predicates"
in a Sinn are the senses customarily expressed by the predicates that occur
within quotation marks in a noematic description appropriate to the act.
And so the predicative content of the Sinn does not consist of the predicates
themselves (which are linguistic entities) or of the properties they denote
(which are transcendent entities), but of predicate-senses, gained by "concep-
tual grasping" of the content in the Sinn.
The predicative content of a Sinn is itself quite complex, apparently in-
cluding everything in a subject's conception of an object that is relevant to
what the object is intended "as" in a given act. (Later we shall modify this
view somewhat, by relegating some of the relevant conceptual elements to
a background of meanings presupposed by, but not actually present in, the
Sinn of the act in question. See Chapter V, Part 3, below.) Of the expressions
that may occur in a noematic description of a Sinn Husserl says:
that the object of the act is intended as being a tree. This property of being
a tree is a generic property that can only be exemplified by an object insofar
as the object is some specific kind of tree - an apple t'ree, an elm, etc. Such
is our conception of trees; and, insofar as the sense "tree" captures this
conception, it is part of this sense that trees come in specific varieties. Thus,
while the predicate-sense "is a tree" explicitly prescribes only the property
of being a tree, the object intended as having this property is thereby also
intended as having the more specific, though as yet unspecified, property of
being a tree of some particular kind. Indeed, insofar as an object is intended
as being a material thing of any kind, Husserl says, the sense that so charac-
terizes the object implies an infinite number of properties, most of which
are not even represented in a general way by the content of the Sinn (cf.
Ideas, §149, esp. p. 367). Properties that are in this way "implied", though
not explicitly prescribed, by the predicate-senses in a Sinn are said by Husserl
to be predelineated (vorgezeichnet) by the Sinn (see Ideas, § § 142,143, 149;
eM, § § 19, 20; EJ, § 21 c). Husser! cites as examples the properties pertaining
to the back, or "hidden", side of an object given in perception: these prop-
erties are by and large not explicitly represented by corresponding senses
in the Sinn but only predelineated by such "indeterminate" senses as "is
colored" or "has a shape" (cf. EJ, §21c). Any Sinn, whether perceptual or
not, that in this way pre delineates more than it actually prescribes presents
its object as transcending what can be predicated of it on the basis of that
Sinn alone. 23
Husserl's account of the predicative content of a Sinn makes it clear that
(except in very special cases of what he calls "adequate givenness": see Ideas,
§ § 138, 142) the object of an act is not presented in the act as a mere pro-
jection of the Sinn: if it were, the object itself would then typically be in-
determinate or incomplete, in the manner that Meinong envisioned. Rather,
the Sinn may be thought of as an indeterminate conceptual "frame", into
which the object is intended as fitting in completely determinate, but as yet
undetermined, ways. The term 'frame' is in fact one that Husser! himself uses
in connection with the "indeterminacies" of the Sinn:
... The general indeterminateness has a range of free variability; what falls within it is
... implicitly included but still not positively ... predelineated. It is a member of an
open range of more precise determinations which can be accommodated to this frame
[Rahmen] but which, beyond this, are completely uncertain. (EJ, §21c, p. 98; with
trans. changes.)
The further properties that an object could have, compatible with what the
200 CHAPTER IV
Sinn does prescribe, make up what Husserl calls the "horizon" of the object
as it is presented in an act. We will be returning to this important notion in
Chapter V.
The X
The Sinn of an act prescribes an object as having certain properties, we have
seen. Whereas the properties are prescribed by predicate-senses in the Sinn,
Husserl holds that the object presented as having these properties is prescribed
by a different kind of sense - the sense "object", or the X, in the Sinn.
Husserl discusses this further component of the Sinn in Ideas, § 131, titled
The "Object", the "Determinable X in the Noematic Sinn" '. There he offers
two main reasons for distinguishing the X from the predicate-senses in the
Sinn. (1) What is intended in an act, through its Sinn, is not a mere aggregate
of unrelated properties. Rather, the properties prescribed by the Sinn are
presented in the act as properties that some object - one and the same object
- has. The Sinn, accordingly, includes not only the subject's sense of certain
properties but also his sense of the object that bears these properties and so
relates them to one another. This sense of an object as bearer and unifier of
the properties attributed in an act is what Husserl calls the X. (2) The same
object can be given in different acts whose Sinne present it as having different
properties. The subject of these acts has a sense of the object as something
that is only partially determined by the properties he intends it as having and
so as something that can be distinguished from these properties. The X in the
Sinn is this sense of the object as that which remains identical throughout
changes in the properties it is intended as having. Thus Husserl says:
The predicates are, however, predicates of "something", and this "something" likewise
belongs with them, and clearly inseparably, to the nucleus [Le., the Sinn] in question:
it is the central point-of-unity .... It is the point-of-connection or "bearer" of the
predicates, but in no way their unity in the sense in which any complex, any binding,
of predicates would be called a unity .... We say that in the continuous or synthetic
course of consciousness the intentional object [Objekt] is persistently consciously
grasped [bewusst] , but is there again and again "differently given": it may be "the
same", it may only be given in other predicates with another determination-content,
"it" may only show itself from a different side whereby the predicates left undetermined
have been more precisely determined .... [In] the noematic description of what is
intended as such at the time ... the identical intentional "object" ["Gegenstand"]
evidently separates itself from the changing and shifting "predicates". It separates itself
out as the central noematic element [Moment]: the "object" [der "Gegenstand", das
"Objekt"] , the "identical", the "determinable subject of its possible predicates" - the
pure X in abstraction from all predicates - and it separates itself off from these predi-
cates or, more accurately, from the predicate-noemata. (Pp. 320-21.)
HUSSERL'S THEORY OF NOEMATIC SINN 201
Husserl holds, then, that within any Sinn we find both an aggregate of
predicate-senses and an X. Both are essential components of a Sinn: "No
'Sinn' without the 'something', nor again without the 'determining content'"
(p. 322). Whereas the whole Sinn is a complex sense that presents an object
as being propertied in certain ways, the X is a primitive type of sense that
stands for the object "simpliciter" or "in abstraction from all predicates".
Apparently, then, the X is a fundamental and unique kind of sense that
presents an object directly, i.e., independently of any particular way of
conceiving or descriptively characterizing the object. We shall further discuss
this characterization of the X in Section 3.4 below.
The passages cited above all home in on one idea: an X is a sense that
presents an object as the bearer of certain properties, the identical subject
of various possible properties or determinations, the object itself abstracted
from any particular properties it has. It might seem that Husserl is saying that
the X presents a "bare particular", an object "bared" of all properties. But
this is not so. For one thing, an object is never presented by an X alone but
only by an X plus predicate-senses, i.e., a Sinn: the two components belong
together "inseparably", so that every Sinn must include them both, Husserl
says in the quoted passages. The presence of an X in the Sinn only requires
that objects are constituted in experience as being distinct from their prop-
erties and, in particular, not as simply "bundles" of properties: the X, he says
above, is a "point-of-connection" for the "predicates", giving them a unity
that a mere "complex of predicates" would not have. Husserl is evidently
drawing, not on a theory of bare particulars, but on his doctrine of categories
and syntactic formations, set forth in Ideas, § § 10-11. According to that
doctrine, objects fall under ontological categories such as those of individual,
quality, or state of affairs. Complex objects have a syntactic structure: for
instance, the state of affairs that General Sherman is ruthless consists of the
individual Sherman and the property of being ruthless joined together by the
syntactic formation of property-instantiation. The elements that enter into
syntactic formations are called "substrata"; objects of the lowest level, those
that have no syntactic structure, are called "ultimate substrata", and these
202 CHAPTER IV
include individuals. Husserl holds that these ontological categories and syntac-
tic forms are paralleled by semantic categories and syntactic forms, categories
and forms of meanings. Thus, when Husserl says a Sinn includes an X as
''bearer'' of various predicate-senses, evidently he is thinking of an X as a
substratum of predications on the semantic level. Hence, an X stands for an
object as substratum of syntactic formations - in particular, property-instan-
tiation - on the ontological level.
An important role of the X's in noematic Sinne is to mark the co-directed-
ness of acts, i.e., the directedness of different acts to one and the same object.
One way Husserl characterizes the X, we have seen, is as the component of
sense that presents an object as identical throughout changes in the properties
it is intended as having. Indeed, on Husserl's account, different acts of con-
sciousness are directed toward the same object if and only if their noematic
Sinne include the same X. He says:
We associate with the one object manifold ways of being conscious [Bewusstseinsweisen I,
acts, or act-noemata .... The object is consciously grasped [bewusstl as identical and
yet in a noematically different way: in a kind of way such that the characterized nucleus
[i.e., the Sinnl is changeable and the "object", the pure subject of the predicates, is
exactly identical. ... Various act-noemata have everywhere here different nuclei, yet so
that they nevertheless close together into unity-ofidentity, into a unity in which the
"something", the determinable that lies in every nucleus, is consciously grasped as iden-
tical. (§ 131, p. 321.)
. .. Not only has every Sinn its "object", but different Sinne relate to the SlIme
object just insofar as they are organized into Sinn-unities in which the determinable X's
of the united Sinne coincide [zur Deckung kommen I with one another and with the X
of the Sinn-unity's total-Sinn. (P. 322.)
(We must be careful in these passages to distinguish the "object", which is
the X-component in a noema, from the object, which is the external entity
that the X stands for.) If we ignore cases in which the intended object of an
act does not exist, Husserl is apparently saying that there is a one-to-one
correlation between X's and intended objects. For where different Sinne have
the same X, he says, they prescribe the same object. And since every Sinn
contains exactly one X, this means that for each Sinn there is at most one
object that can satisfy it. Accordingly, Husserl's claim that every Sinn includes
an X seems to suppose that every act is a definite, or de re, intention, an
intention of a defInite individual. If so, his account of Sinn-structure does
not extend to indefmite, or de dicto, intentions. That omission aside, our
further discussions of the Sinn will have to give careful heed to the fact that,
for Husserl, the X in a Sinn is considered a mark of defmiteness of intention.
(Cf. Chapter I, Section 2.6, above; Section 3.3 below.)
HUSSERL'S THEORY OF NOEMATIC SINN 203
individual whom one has never seen or even heard of. Contrary to the account
provided by the defmite-description model, though, what seems really to
happen in these cases is that the directedness of the act somehow "by-passes"
the predicative content of the act's Sinn, so that the act intends the appro-
priate object, not via this content, but despite it.
In sum, there are important and frequent cases in which an act's being
directed toward a particular object is largely independent of the descriptive
content in the act's noematic Sinn. It is certainly plausible to maintain that
some intentions work just as the definite-description account would have it,
but for a wide range of not very exceptional cases that theory is simply in-
adequate. In fact, as we shall now see, the definite-description model of
intention fails to explain the kinds of intentions that Husserl himself most
emphasized, those that we have characterized as "definite" or "de re".
of a definite description does prescribe at most one object. But we shall argue
that definiteness is a stronger notion than this, and that the stronger notion is
required by certain features of Husserl's theory of intentionality.
Whatever Husserl might have thought, there are various kinds of acts whose
Sinne do not uniquely determine an object of intention. One kind includes
such acts as my desire for a new car, although there is no particular one I
want, or my hearing someone at the door, when I have no idea who it is. The
Sinne of these acts - "a new car", "someone at the door" - can be satisfied
by any number of individuals, and the corresponding acts are for that reason
not directed toward anyone specific individual. Husserl does not completely
ignore these acts, although he apparently thought them to be too uninterest-
ing to deserve serious treatment. In one of perhaps only two mentio·ns of
them in Logical Investigations he cannot even decide whether they deserve
to be called "intentional". He says:
Desire does not always seem to require a conscious relation to what is desired, ... we are
often moved by obscure drives or pressures towards unrepresented goals .... One may
say: This'is a case of mere ... "desire~ensations" ... , i.e., of experiences really lacking
an intentional relation, and so also remote in kind from the essential character of inten-
tional desire. Alternatively one may say: Here we are dealing with intentional experi-
ences, but with such as are characterized by indeterminately [unbestimmt] directed
intention .... The idea we have when "something" stirs, when "it" rustles, when "some-
body" rings, etc .... is "indeterminately" directed; and this "indeterminateness" belongs
to the intention's essence, wherein it is determined as presenting an indeterminate
"something". (V, § 15, p. 575; with trans. changes. Cf. VI, § 10, p. 700. A possible third
mention is in V, § 20, p. 589, n. 1.)
not be indefinite in this radical way. Nonetheless, we shall see, they will still
be indefinite in a more subtle way that Husser! seems not to have considered
explicitly.
This more subtle form of indefiniteness is the kind we earlier characterized
as "de dicta" (see Chapter I, Section 2.6). Consider, for example, the shop-
keeper's act of expecting the first customer of the day (whoever that may
be). We may suppose that the Sinn of this act is the sense of a definite de-
scription, "the first customer of the day", and since the description applies
to at most one person the act is not indeterminate in the way just described.
Yet, inasmuch as the shopkeeper has no opinion as to who her first customer
will be, the Sinn of her act leaves it open that any number of individuals
could be the one who satisfies it, depending on the circumstances: if Mr.
Black is the first to enter her shop, then he satisfies the Sinn; if Ms. White is,
then she does, and so on. In each of these circumstances the Sinn would
single out a unique object; but since the Sinn itself does not determine which
circumstance will be actual, just which object it singles out remains "in-
definite" or "indeterminate". By contrast, the shopkeeper's act of expecting
Professor Anscombe to enter her shop is definite in the sense we characterized
as "de re". The only circumstances in which the Sinn of this act will be
satisfied are those in which it is Professor Anscombe who walks through the
door; thus, in any circumstance, either the Sinn will not be satisfied at all or
it will be satisfied by the very same individual, G. E. M. Anscombe. But the
Sinn that determines an intention as definite in this de re way cannot be
simply the sense of a definite description.
What we have said about intention here has its counterpart in the theory
of reference as well. The problem in either case is that there are few descrip-
tions (or none, barring essentialism) that could not be satisfied by different
individuals given appropriately different states of affairs or courses of events.
'The husband of Xanthippe', for example, is a description actually satisfied
by Socrates; but Xanthippe might have married some other smooth-talking
Athenian, and in those circumstances some person other than Socrates meets
this description. Consequently, if the reference of 'Socrates' were determined
by the descriptive sense expressed by 'the husband of Xanthippe', then
'Socrates' would refer to different individuals under different possible cir-
cumstances. Indeed, any definite description (or the sense of any defmite
description) one is likely to use to secure a relation between 'Socrates' and its
referent will relate that name to individuals other than Socrates in some other
possible circumstances. And for this reason, it has been argued, the definite-
description theory fails to provide an adequate account of the reference of
HUSSERL'S THEORY OF NOEMATIC SINN 211
names: F¢llesdal, Kripke, and others hold that, unlike definite descriptions,
names are used so as to refer to the same individual under any actual or
possible circumstances involving that individual. 28 When we say such things
as "If Socrates had taken Crito's advice, he would never have drunk the
hemlock" or "Socrates might not have married Xanthippe" we purport to
refer to Socrates and to say what would have been true of him if certain
non-actual but possible conditions had been realized. In such counter-factual
or modal contexts, the name 'Socrates' is used to single out Socrates, not
merely in the circumstances that were in fact actual, but in these other pos-
sible circumstances as well. But since definite descriptions may refer to
different individuals under different circumstances, the sense that determines
the reference of 'Socrates' cannot be merely that of a definite description. In
short, where either reference or intention is a definite, or de re, relation to
the same individual under different possible courses of events, it is a stronger
relation than that secured by the sense of a definite description.
Now, Husserl's analysis of Sinn and intention focuses on acts that are
definite in this very way. As we have seen, he holds that acts typically intend
their objects as transcending what is actually presented in the act. The Sinn
of such an act, we said earlier, may be thought of as an indeterminate con-
ceptual "frame", which the object is intended as fitting into in various pos-
sible ways compatible with what the Sinn prescribes (see Section 3.1 above).
Thus, the Sinn leaves open a "horizon" of possibilities concerning its object.
But just how much does the Sinn leave open? If the Sinn were merely the
sense of a definite description, then it would be possible for different objects
to satisfy it. The Sinn would then leave open which object is intended in the
act, and the act would accordingly be indefinite, or de dicta. There surely are
acts of that kind, we have stressed, and Husserl ought to have recognized Sinne
of a sort appropriate to them. But what he says about the Sinn and the pos-
sibilities it leaves open seems only to apply to definite, or de re, acts. For the
cases he discusses, the Sinn must single out a specific object in such a way
that only that object can satisfy the Sinn: the Sinn leaves open what further
properties the object has, but it somehow includes a sense of the identity of
the object it prescribes and so rules out the possibility that a different object
might satisfy the Sinn.
Husserl's most thorough discussion of this topic is in Cartesian Meditations,
where he takes perception as his paradigm. And here his commitment to a
de re form of definiteness, in fact unachievable by the sense of a defmite
description, is evident. When I see an object, the Sinn of my act does not
merely prescribe whatever object happens to satisfy- the Sinn's descriptive
212 CHAPTER IV
content; what I see is this object here before me. Thus, the Sinn of a percep-
tion prescribes the object, leaving open only further properties of this object
that are compatible with what one perceives of it. Husserl makes this point
in terms of possible perceptions compatible with what one perceives - per-
ceptions whose Sinne would determine in various possible ways what is left
open by the original act's Sinn. These further possibilities of perception, he
says, include only those acts that can be joined with the original act in a
"synthesis of identification" (eM, §18, pp. 41-42), whereby they are di-
rected toward one and the same object as their common "pole of identity"
(§ 19, p. 45). Indeed, "by virtue of [this "synthetic unity"] alone", Husserl
says, ''we have one intentional object, and always this definite one, continu-
ously meant" (§20, p. 47; cf. pp. 46, 48). Evidently, then, the Sinn of a
perception specifies an object in such a way that this very same object is
intended in any possible perception compatible with the Sinn. And that is to
say that under no possible circumstances left open by the Sinn can different
objects satisfy it. By Husserl's own account of perceiving a transcendent
object, we conclude, perception requires a different sort of Sinn than that
which the defmite-<iescription model of intention proposes. (We discuss the
notion of "horizon" much more fully in chapter V.)
We pointed out in Chapter I (Section 2.6) that the defmiteness of percep-
tion differs from that of ''individuative'' intentions, i.e., intentions in which
the subject knows (or has an opinion about) who or which the intended
object is. The descriptive content of the Sinn apparently plays a rather minor
role in perceptual definiteness. When I see a man, for example, the Sinn's
predicate-senses may determine very little about him other than that he is a
man with certain apparent physical characteristics; the act is then definite
only because directed toward this individual before me, not because the Sinn
specifies who he is. For individuative acts the Sinn's descriptive content seems
to playa larger role, insofar as "knowing who" is a matter of knowing prop-
erties sufficient to distinguish one individual from others. But then our dis-
cussion in this section leaves it completely unclear how individuative intention
can be achieved, for we have argued that it is always possible for different
objects to satisfy the descriptive content of a Sinn. We shall examine this
problem in considerable detail later , concluding that complete individuative
defmiteness is an ideal that actual intentions can only approximate to a
greater or lesser degree. In fact, we shall find that Husserl himself considered
the very identity of a transcendent individual to be itself transcendent, thus
exceeding what the content of any Sinn prescribes. (See Chapter VIII, espe-
cially Parts 3 and 4.)
HUSSERL'S THEORY OF NOEMATIC SINN 213
intention achieved via Sinn, via an X? We shall now try to spell out HusserI's
account of perceptual Sinn and perceptual intention via Sinn, drawing on
these doctrines in both the Investigations and Ideas. We shall conclude that
perception poses a special problem for HusserI's basic theory of intention via
Sinn, and that although Husserl showed some awareness of the source of the
problem, his analysis of perception and its Sinn really offers no solution to
that problem.
It is natural to form a phenomenological description of a perception of an
object by use of the demonstrative pronoun 'this' (or 'that'):
I see this black bird
- or, in line with Husserl's talk of an object intended "as" such-and-such.
I see this as a black bird.
(We understand this sentence as ascribing a direct-object perception, not a
propositional perception or perceptual judgment that this is a black bird.)
This form of phenomenological description fits nicely with HusserI's analysis
of a perceptual Sinn. The predicative expression 'a black bird' would here
ascribe the predicative content of the act's Sinn, the predicate-sense expres-
sible by '[is] a black bird'; and the demonstrative 'this' would ascribe the X
in the Sinn. Thus, the X would be the type of sense expressible on a given
occasion by saying 'this' in reference to an object perceived before one, and
a perception's Sinn would be a complex "demonstrative" sense expressible
by a demonstrative phrase such as 'this black bird' or perhaps 'this as a black
bird'. We might say the "this" component of the Sinn prescribes the intended
object "itself" and the predicative component prescribes the properties it is
intended "as" having. (Recall Section 2.6 above on phenomenological and
noematic description.)
The use of demonstrative pronouns to form phenomenological descriptions
of perceptions, or alternatively to express the Sinne of perceptions, has a
foundation in Husserl's brief but penetrating account of demonstratives in
Logical Investigations (I, § 26, and VI, § § 1-5). An utterance of 'this' is
expressly keyed to perception (at least in one important use of 'this'). 'This',
HusserI says, is an "essentially occasional" expression: the referent, and even
the meaning, of 'this' varies with, and so depends on, the occasion of utter-
ance (LI, I, § 26). The referent of 'this' on a particular occasion of utterance
is the object the speaker then sees (or hears or otherwise perceives). Often
he physically points at the object, but it is the perception that is the basic
determinant in the "demonstration" of reference; the pointing is only a
HUSSERL'S THEORY OF NOEMATIC SINN 215
dramatic aid to help the hearer determine more precisely what the speaker
is looking at and therewith that he is talking about. Suppose, Husserl says,
"I have just looked out into the garden and now give expression to my per-
ception in the words: 'A blackbird flies up'" (VI, §4, p. 680; with trans.
changes). In such a use:
'This' is an essentially occasional expression which only becomes fully meaningful when
we have regard to the circumstances of utterance, in this case to an actually performed
perception. The perceived object, as it is given in perception, is what the word 'this'
refers to list mit dem 'dies' gemeint]. (VI, §5, p. 682; with trans. changes.)
Essentially occasional expressions [e.g., 'this'] are ... much like proper names, insofar
as the latter function with their authentic meaning. For a proper name also names an
object "directly". It refers to [meint] it, not in the attributive way as the bearer of these
of those properties, but without such 'conceptual' mediation, as that which it "itself" is,
just as perception might set it before the eyes. (LI, VI, § 5, p. 684; with trans. changes.)
So, like Russell and many others, Husserl holds that a demonstrative refers
"directly". And, he implies, neither names nor demonstratives express the
type of senses expressible by definite descriptions. 31 But if a demonstrative
or name does not refer by appeal to properties of the referent, as a definite
description does, then how does it refer? What is it to refer "directly", '~ust
as perception might set [an object 'itself'] before the eyes"? Qua intuition,
perception is a "direct" intention, but what is it to intend an object "di-
rectly"?
It is commonly held that the referent of 'this' on an occasion of utterance
is determined by the context of utterance, by the speaker's de facto physical
relation to the referent, including perhaps his pointing at it and perhaps its
causally affecting his senses. Then demonstrative reference would be "direct"
in that the referent would be determined by the context of the utterance, and
not by a descriptive sense expressed by the speaker. Analogously, perception
would be "direct" in that the object of a perception would be determined by
the de facto physical context of the perception, in particular, by the spatio-
temporal-causal relation that consists in the object's playing an appropriate role
in the causal genesis of the sensory phase of the perception. (This would be
a strong form of the popular "causal" theory of perception.) On this account,
the perception might still have a content or Sinn such as Husserl described.
Indeed, it might have a Sinn including predicative content that prescribes
certain properties of the object perceived, properties the object is perceived
HUSSERL'S THEORY OF NOEMATIC SINN 217
(cf. Section 3.2 above). Yet, the belief on which my judgment depends is
external to the judgment itself: it is not a part of my judgment but a part
of my background belief-system, which my judgment merely presupposes.
We might say that, whereas the perception depends on its "transcendent"
context, a circumstance in the transcendent, physical world, my judgment
depends on its "immanent" context, a circumstance within my stream of
consciousness or my background of mental attitudes. (Beliefs are not occur-
rent "acts" of consciousness; but, as Husserl might say, they are "passive"
judgments or dispositions to judge. Beliefs are then an immanent part of
mental life though not events in the stream of consciousness per se.)
Husserl was aware of the need to bring some contextual factors into the
theory of intentionality, but he seems to have attended carefully only to
what we just called immanent contextual influences. As we shall see in
Chapters V and VI, Husserl allowed for the influence of background beliefs
on intention, including particular beliefs (such as that about the woman
drinking a martini) and especially those broad beliefs that form one's larger
conceptual scheme. In later works such as Cartesian Meditations, he increas-
ingly stressed the role of such background belief-structures in directing an act
toward its object. But the elaboration of these belief-structures is an important
break from Husserl's early methodology of merely laying out the structure
of an act's noema (or corresponding noesis). For these structures are not a
proper part of the act's noematic Sinn but rather are presupposed by the act
or its Sinn. Husserl himself, we shall see, finds in these analyses "methods of
a totally new kind" (CM, §20, p. 48). Yet, insofar as background beliefs are
immanent, these methods still remain within the bounds of phenomenology,
studying what remains after bracketing the transcendent world. In a sense,
Husserl has simply broadened his notion of the phenomenological content of
an act.
The transcendent circumstances of an act are a different matter entirely.
If the physical circumstance of a perception has a role in determining the
directedness of the perception, then a purely phenomenological analysis of
intention will be inadequate for perception. For a phenomenolOgical analysis
attends only to what can be grasped in phenomenological reflection, without
consideration of anything external to the act and its content. In his early
Logical Investigations, we observed, Hussed noted the "essentially occasional"
nature of reference in the case of demonstrative pronouns, and he held that
a perception is expressible by use of demonstratives. So there he had some
awareness of the "occasional" nature of perceptual intention. Moreover, he
mentions "occasional" expressions and judgments also in his later writings,
222 CHAPTER IV
which follow his celebrated "transcendental" tum (see FTL, §80, pp. 199-
200; and Crisis, §33, pp. 122, 124). Yet, as we saw in the preceding section,
HusserI seems never to have developed a systematic theory of the occasional
nature of perception and its directedness, nor does he seem ever to have
comprehended the problem it poses for his basic theory. 33
In the following chapters we shall study HusserI's own extension of his
basic theory, by way of "horizon", in ways that accommodate the influence
of background beliefs on intention. In Chapter VIII we shall return to certain
aspects of the case of perception.
NOTES
1 F¢lIesdal's interpretation dates from his 1958 dissertation, Busserl und Frege (Note
10, Ch. III above). It is stated most succinctly in his 'Husserl's Notion of Noema' (Note
15, Ch. III above) and is applied to perception in 'Phenomenology' (Note 19, Ch. III
above). Also see his 'An Introduction to Phenomenology for Analytic Philosophers'
(Note 10, Ch. III above). The interpretation has been expounded and developed in
various ways in doctoral dissertations directed by F¢lIesdal: Hubert L. Dreyfus, 'Husserl's
Phenomenology of Perception: from Transcendental to Existential Phenomenology',
(Harvard University, 1963); Ronald Mcintyre, 'Husser! and Referentiality: The Role of
the Noema as an Intensional Entity' (Stanford University, 1970); David Woodruff Smith,
'Intentionality, Noemata, and Individuation: The Role of Individuation in Husserl's
Theory of Intentionality' (Stanford University, 1970); John Francis Lad, 'On Intuition,
Evidence, and Unique Representation' (Stanford University, 1973); and Izchak Miller,
'The Phenomenology of Perception: Husserl's Account of Our Temporal Awareness'
(UCLA,1979).
2 Aron Gurwitsch, The Field of Consciousness (Duquesne University Press, Pittsburgh,
1964), p. 173. Our discussion of Gurwitsch's interpretation is based on: The Field of
Consciousness, esp. pp. 164-68, 173-84; his 'On the Intentionality of Consciousness',
in Phenomenology, ed. by Joseph J. Kockelmans (Doubleday, Garden City, N.Y., 1967),
pp. 118-37, reprinted from Philosophical Essays in Memory of Edmund Busserl, ed. by
Marvin Farber (Harvard University Press, Cambridge, Mass., 1940); and his 'Husserl's
Theory of the Intentionality of Consciousness in Historical Perspective', in Lee and
Mandelbaum (Note 21, Ch. I above), pp. 25 -57, reprinted in Dreyfus (Note 10, Ch. III
above). Also see Dreyfus's comparison of Gurwitsch's and F¢lIesdal's interpretations of
the perceptual noema, in Hubert L. Dreyfus, 'The Perceptual Noema: Gurwitsch's Crucial
Contribution', in Life-World and Consciousness, ed. by Lester Embree (Northwestern
University Press, Evanston, 111., 1972), pp. 135-70, revised and reprinted as 'Husserl's
Perceptual Noema', in Dreyfus (Note 10, Ch. III above).
3 Gurwitsch, 'Husserl's Theory .. .' (Note 2 above), p. 45; cf. p. 47.
4 Gurwitsch, 'On the Intentionality of Consciousness' (Note 2 above), p. 128; see also
p.130.
5 See Gurwitsch, 'Husser!'s Theory .. .' (Note 2 above), pp. 52-53; and The Field of
Consciousness (Note 2 above), pp. 183-84.
HUSSERL'S THEORY OF NOEMATIC SINN 223
6 Gurwitsch, 'Husserl's Theory .. .' (Note 2 above), p. 55; also see The Field of Con-
sciousness (Note 2 above), pp. 184,223.
7 Gurwitsch, The Field of Consciousness (Note 2 above), p. 182; cf. pp. 164-68.
8 See Gurwitsch, The Field of Consciousness (Note 2 above), pp. 177-78.
9 Twardowski (Note 3, Ch. III above), p. 17.
10 Twardowski (Note 3, Ch. III above), pp. 11-17. Cf. Dreyfus's commentary on
Husserl's use of the 'as intended' and 'as such' terminology, in Dreyfus, 'Husserl's Phe-
nomenology of Perception' (Note 1 above), pp. 205 -13, and 'The Perceptual Noema'
(Note 2 above), pp. 154-59.
II Reinhardt Grossmann, in his Introduction to Twardowski (Note 3, Ch. III above),
has explicitly assimilated Husserlian noemata and Meinongian incomplete objects: see
esp. p. xxvi. A well-developed object-theory of a generally Meinongian sort is that of
Hector-Neri Castaneda: see his 'Thinking and the Structure of the World' (Note 25,
Ch. II above), and 'Perception, Belief, and the Structure of Physical Objects and Con-
sciousness', Synthese 35 (1977), 285-351. Castaneda does not purport to be interpret-
ing Husserl, but his theory seems to be the kind that would naturally follow from the
Gurwitschean interpretation.
12 Sections 2.1-2.5 below constitute a revised version of our 'Husserl's Identification
of Meaning and Noema', The Monist 59 (1975), 111-32, reprinted in part in Dreyfus
(Note 10, Ch. III above).
13 Frege (Note 9, Ch. III above), p. 79.
14 Frege (Note 9, Ch. III above), p. 79.
IS Bolzano, Theory of Science (Note 11, Ch.Ill above), esp. § §48, 270, 271.
16 Husserl sometimes uses 'Wirklich' in a broader sense, characterizing anything that
may be the object of an act prior to phenomenological (especiall)' transcendental) reduc-
tion, i.e., anything that is not a "transcendental" entity. In this broader sense mathe-
matical entities and natural essences, though nontemporal and nonspatial, are "wirklich ",
but meanings and noemata are not.
17 Frege, The Thought' (Note 56, Ch. II above), p. 302.
18 See Guido Kiing, 'Husserl on Pictures and Intentional Objects' (Note 16, Ch. III
above), p. 675; J. N. Mohanty, 'On Husserl's Theory of Meaning' (Note 16, Ch. III
above), pp. 229-32. Also see LI, I, § 33.
19 Husserl does characterize general terms, such as 'horse', as expressions that refer to
different things on different occasions of use, without changing their meanings (LI, I,
§ 12, p. 288). However, this characterization need not run counter to (3). While the
particular thing of which a general term is predicated will vary from case to case, the
extension of the term - the class of entities of which the term can be truly predicated -
will remain the same so long as the meaning remains the same. Thus, (3) is true of general
terms, too, provided we understand by the "referent" of a general term its extension.
Cf. LI, VI, § 7, p. 693, where Husserl speaks of the "range of objects" "covered" by a
general term.
20 See LI, I, § 12, p. 288; VI, § 7, p. 693. Mohanty has translated and discussed a
letter from Frege to Husserl (written in 1891) commenting on this difference: see J. N.
Mohanty, 'Husserl and Frege' (Note 10, Ch. III above), pp. 57-61; 'Frege-Husserl
Correspondence' (Note 10, Ch. III above), pp. 84-85.
21 Its most explicit statement in LI is in V, §21, p. 590 (where "semantic essence" =
"matter" = "Sinn"). The point is well conftrmed by Husserl's conception of expression
224 CHAPTER IV
in LJ, VI, § §1-15, and also in sections of Formal and Transcendental Logic and Ideas
that we shall be discussing.
22 Cf. David Kaplan's use of "meaning marks" in Kaplan (Note 26, Ch. I above), pp.
185-87. The device derives from Quine's "corner quotes" as e,mployed in W. V. Quine,
Mathematical Logic (Norton, New York, 1940).
23 We use 'prescribe' and 'predelineate' as technical terms. While 'predelineate' is our
translation of Husser!'s 'vorzeichnen', 'prescribe' does not translate a specific term of
Husserl's.
24 Our. discussion in this section and the next derives from two papers by Ronald
Mcintyre: 'Intending and Referring: Some Problems for Husser!'s Theory of Inten-
tionality', in Dreyfus (Note 10, Ch. III above); and 'Husserl's Phenomenological Con-
ception of Intentionality and its Difficulties', Philosophio, forthcoming.
25 The deftnite-<lescription theory of reference is discussed in much detail by Saul
Kripke in his 'Naming and Necessity', in Semantics of Natural Language, ed. by Donald
Davidson and Gilbert Harman (D. Reidel, Dordrecht, 1972), esp. pp. 255-60, 277-
303.
26 See Kripke (Note 25 above), pp. 253-303; Keith Donnellan, 'Proper Names and
Identifying Descriptions', in Davidson and Harman (Note 25 above), pp. 356-79 ;-Hilary
Putnam, 'Meaning and Reference', Journal of Philosophy 70 (1973), 699-711, reprinted
in Naming, Necessity, and Natural Kinds, ed. by Stephen P. Schwartz (Cornell University
Press, Ithaca, N. Y., 1977); and Putnam, 'The Meaning of "Meaning"', in his Philosophical
Papers, Vol. II: Mind, Language, and Reality (Cambridge University Press, Cambridge,
1972), pp. 215-71.
27 See Donnellan, 'Reference and Defmite Descriptions' (Note 14, Ch. I above).
28 See F¢llesdal, 'Knowledge, Identity, and Existence', Thearia 33 (1967), 1-27;
Kripke (Note 25 above), esp. pp. 264-89; and Kripke, 'Identity and Necessity', in
Schwartz (Note 26 above), pp. 77-83.
29 This section of the present chapter is based on a more detailed study of Husser!'s
account of demonstratives in David Woodruff Smith, 'Husser! on Demonstrative Refer-
ence and Perception', in Dreyfus (Note 10, Ch. III above). Izchak Miller has also studied
the "demonstrative" element in perception put forth by Husser!; see Miller, 'The Phe-
nomenology of Perception' (Note 1 above). Miller stresses the singular (what we call
the definite, or de re) character of perception. John Lad also stresses this in his account
of perception as intuition; see Lad, 'On Intuition, Evidence, and Unique Representation'
(Note 1 above).
30 The theory of demonstratives and perceptual Sinn that we have presented is not
quite that of Husser! in Logical Investigations. Husser! says demonstratives serve to
"express" - to express the Sinne of - "judgments of perception", i.e., judgments
"grounded" on perception, rather than perceptions themselves (LI, VI, § § 3-4). This
is because he holds that "perception is an act that determines but does not contain
meaning" (LI, VI, §5, p. 684; with trans. changes). And that would block him from
holding that 'this' expresses the X in the Sinn of one's perception. However, Husserl's
reasons for balking at demonstratives' expressing perceptual sense per se seem to be
flawed and to involve mistakes that he should not have made if he had then clearly
articulated the doctrine of Xs put forth in Ideas. So we have presented, for our purposes
of assessing the Husserlian approach to intentionality, the emended theory we outlined,
which is not quite Husserl's. See David Woodruff Smith, 'Husser! on Demonstrative
Reference and Perception' (Note 29 above).
HUSSERL'S THEOR Y OF NOEMATIC SINN 225
31 In this respect Husserl's view of names and demonstratives is like that of the 1970's
developed in works of Donnellan, Kaplan, and Kripke. On direct reference by proper
names, see Keith Donnellan, 'Proper Names and Identifyirlg Descriptions' (Note 26
above), and Saul Kripke, 'Namirlg and Necessity' (Note 25 above). On direct reference
by demonstratives, see David Kaplan, "Dthat" and 'On the Logic of Demonstratives',
irl Contemporary Perspectives in the Philosophy of Language, ed. by Peter A. French,
Theodore E. Uehling, Jr., Howard K. Wettsteirl (University of Mirlnesota Press, Minnea-
polis, 1979), pp. 383-412. Husserl's account of demonstratives and Kaplan's coincide
up to a point: both hold that demonstratives refer directly; both recognize two levels
of meaning for demonstratives, one that varies with the occasion of utterance and one
that does not. Where Husserl developed the phenomenological and epistemological
foundations of demonstrative reference, Kaplan has developed a modern formal logic, a
model-theoretic or possible-worlds semantics or pragmatics, for demonstratives. Two
relevant studies of demonstratives that draw on both Husserl and Kaplan are David
Woodruff Smith, 'Indexical Sense and Reference', Synthese 49 (1981), and 'What's the
Meaning of "This"?', Noils 16, No.2 (1982).
32 A study of the roles of content and context in perceptual acquaintance is found in
David Woodruff Smith, 'Content and Context of Perception', Synthese (to appear irl
1983). (Cf. the articles by Smith cited irl the previous note.) A longer study of acquairlt-
ance by Smith is irl progress under the title, Acquaintance. In those studies a partial
vindication of Husserl's leading irltuition that irltention is achieved irl virtue of phe-
nomenological structure is proposed. The general idea is this. A perception presents an
object visually before the perceiver on the occasion of perception: irl phenomenological
description he may say, "I see this tomato now here before me and affectirlg my eyes".
Now, the object of perception is not a function of the content alone, for another percep-
tion on another occasion could irl principle have the very same phenomenological content
and yet have a different object. That is, there is no functional, or many-one, relation
between the confent and the object of a perception (contra Husserl). Still, it seems, the
demonstrative content of a perception - the content "this (now here before me and
affecting my eyes)" - does prescribe the object of the perception, the object appro-
priately before the perceiver and affectirlg his senses on the occasion of the perception.
However, it is not the noematic content irl itself that so prescribes the object; rather, it is
the content only irlsofar as it is embodied irl that particular perceptual experience on
that occasion - if you will, the demonstrative content-irl-the-perception prescribes, or
is satisfied by, the object of the perception, the object contextually before the perceiver.
A similar view has been developed, irldependently and from a different direction, by
John Searle. On Searle's account, the causal relation between a visual experience and
its object is part of the "conditions of satisfaction" of the irltentional content of the
experience. (Searle's view is developed irl Intentionality (Cambridge University Press,
Cambridge, 1983 forthcomirlg), which served as the basis for his lectures at the 1980
Summer Institute on Phenomenology and Existentialism, 'Continental and Ana1ytic
Perspectives on Intentionality', held at the University of California, Berkeley, under
the auspices of The Council for Philosophical Studies with support from the National
Endowment for the Humanities. Smith's view was also presented irl lecture at the same
irlstitute.) Searle's view, like Smith's, may be seen as a partial vindication of Husserl's
view of perception, and irldeed Searle's general view of irltentionality (as well as his view
of the relation between language and thought) is broadly like Husserl's.
33 It is the "transcendental" foundation of Husserl's phenomenology that is irlcompa-
226 CHAPTER IV
tible with letting the object of perception, or any other part of the external world, play
a role in perceptual intention. An "existential" phenomenology such as Maurice Merleau-
Ponty's is not limited to what is in consciousness per se. For Merleau-Ponty, aspects of
the perceiver's body playa role in perceptual apprehension of an object. These include
the affection of the senses, for he holds that the sensuous and presentational components
of perception cannot be separated as Husserl proposed. And they include bodily skills
of which the perceiver has no consciousness or representational knowledge, such as his
ability - and tendency - to move around the object for a further view or to pick it up.
For Merleau-Ponty, it is in virtue of these bodily connections with the object, as well as
the structure of perceptual awareness, that one sees the object as "bodily present". This
interpretation of Merleau-Ponty's phenomenology of perception and its consequences
for Husserl's is put forth by Dreyfus in his 'The Perceptual Noema' (Note 2 above) .•
Compare Merleau-Ponty, Phenomenology of Perception, trans. by Colin Smith (Rout-
ledge & Kegan Paul, Ltd., London, 1962; French original, first published in 1945). So
intimate is the role of the body in the phenomenology of perception that he can say in
a title on page 203, "The theory of the body is already a theory of perception". Does
such an account of perception remain purely phenomenological?
CHAPTER V
says) "determinations". (We shall see, though, that what remains open may
be constrained, not only by the predicate·senses explicitly included in the
act's Sinn, but also by those parts of the subject's background conceptual
scheme to which the Sinn appeals, say, the subject's basic theory or concep-
tion of physical objects.)
Physical objects, persons, and the natural world as a whole are transcen-
dent, Husserl holds (Ideas, §§44, 47,53), so Sinne that relate to them charac-
terize them incompletely. But we do intend these objects as transcendent, as
complete objects having further properties not specified by the Sinn through
which they are intended in a particular act. When we perceive a physical
object, for example, we perceive it as having more to it than is directly given
in sensory "intuition" at that moment. The properties that are so given, what
Husser! sometimes calls the "genuinely" (eigentlich) perceived properties of
the object, are themselves taken as only a "side" or "aspect" of the perceived
object, which is itself intended as having further properties or aspects not
"genuinely" perceived. "The 'and so forth''', Hussed says, "is an ... abso-
lutely indispensable element [Moment] in the thing-noema" (Ideas, § 149,
p. 367). Consequently, that there is more to a physical object than what is
given in an act is itself given in the act, prescribed by its Sinn. Hussed says:
... As a consciousness, every cogito is indeed ... an intending of its intended [Meinung
seines Gemeinten: literally, 'a meaning of its meant'], but ... , at any moment, this
something intended [dieses Vermeinte] is more - something intended [Vermeintes]
with something more - than what is "explicitly" intended [Gemeintes] at that moment.
... This intending-beyond-itself [Uber-sich-hinaus-meinen], which is inherent in any
consciousness, must be considered an essential element [Moment] of it. (eM, §20,
p. 46; with trans. changes.)
Suppose I see a tree some distance away. I see a fruit tree, let us say, but
I cannot see precisely what kind of fruit is hanging from it. In what respects
is the object of my perception determined by the Sinn of my perception, and
what is left open by the Sinn? The color and shape of the tree as apparent
from my particular perspective, and perhaps something of its texture, are
"genuinely" perceived; these visual properties of the front side of the tree are
thus prescribed by "filled" predicate-senses in my act's Sinn. But, of course,
I perceive the object as much more than this. In particular, let us suppose, I
perceive it as a citrus tree in full fruit, and I also perceive - though with a low
degree of attention - something of the object's environment, say, other trees
around it in an orchard. These "determinations" of the tree, along with the
more specific visual characteristics "genuinely" perceived of its front side,
are, let us say (picking up on Husserl's terminology in the passage we quoted
from Cartesian Meditations), "explicitly" intended of the object as it is given
in this act; and let us say that the predicate-senses that prescribe these prop-
erties make up the "explicit" content of the act's Sinn. The Sinn of the act
thus leaves open, it does not specify, whether the tree is or is not a grapefruit
tree, or whether it is or is not an orange tree, or whether it is either a grape-
fruit or an orange tree or is neither. It also leaves unspecified the precise
nature of the tree's back side (whether, for example, it has more or less fruit
HUSSERL'S NOTION OF HORIZON 231
on that side) and it leaves open whether the tree may harbor living creatures
among its branches. Importantly, though, not everything about these further
aspects of the tree I see goes unprescribed by the Sinn of my current percep-
tion. Husserl says, in the passage from Ideas, that the possibilities left open
are "predelineated" by the "general sense" of what is perceived. Here, for
example, I perceive the object as a citrus tree, and the sense "is a citrus tree"
"implies" much about the object. (What is "implied" here need not be strictly
a logical implication of this sense alone but may also draw upon various items
of background knowledge and belief, some of which are empirical; see Part
3 below.) The "explicit" Sinn of my act prescribes by "implication" (with
help from the appropriate background items) that the tree is a fruit tree and
not a conifer; that it has a back side containing leaves and fruit like those on
its front side, but that it will not hold a sprouting watermelon; that among
its branches may be found any of the various arboreal creatures indigenous
to Southern California, but that I shall not fmd there a stranded platypus;
and, indeed, that the tree I see has many further "determinations" not spec-
ified at all in my current perception. So the Sinn of my perception prescribes
- if not "explicitly", then by "implication" - many further properties of
the object, though these further prescriptions are for the most part quite
general, or non-specific, and leave open various possibilities for their particular
realization. In this way, Husserl says, my act of perception "intends beyond"
itself, it "points forward" to other possible perceptions whose Sinne would
characterize the object more closely and in further respects.
The possibilities that the Sinn of an act leaves open - what remains in-
determinate - about the object as intended in the act (here including, for
instance, the possibility that it is a grapefruit tree), Husserl calls the horizon
of the object as it is intended. 2 This horizon we may think of as the circum-
scribed limits of the object's further characterization and of its "constitu-
tion" in consciousness. 3 Corresponding to this horizon of the object as given
in my perception, to what the Sinn of the act leaves open or unspecified
about the object, is the set of possible perceptions that would, if they oc-
curred, tend to complete my perceptual determination of the object. Husserl
says, in the passage we quoted above, that "the indeterminacy ... points
forward to possible manifolds of perception ... in which the continuously
enduring thing ... shows again and again new 'aspects' .... " These possible
perceptions include, for instance, perceptions of the object's back side that
expand on the original perception, and perceptions in which I take a closer
look at its front side. These perceptions are co-directed with the original, and
their Sinne include predicate-senses that tend to complete the incomplete
232 CHAPTER V
implicitly" in the act. Although Husserl does not defme the notion of "im-
plicit Sinn", it seems reasonable to suppose that, in the same sense in which
he considers the acts in the horizon of a given act to be implicit in that act,
he would also consider the Sinne of the acts in this horizon to be implicit in
the Sinn of the given act. Accordingly, we shall say that the senses in the
Sinn of an act itself make up the act's "explicit Sinn", and that the senses
in the Sinne of the acts in the act's horizon make up that act's "implicit
Sinn".
Husserl's accounts of horizon (act- and object-horizon) are couched in
terms of possible entities (possible acts and possibilities regarding the object,
which itself need not be actual). As we consider in Chapter VI, Husserl's
commitment to a metaphysics of possibilia is unclear; he might well prefer
that talk of possible entities be ultimately eliminated in terms of correspond-
ing noematic entities. However, the idiom of possibilia is suggestive, and we
shall use it regularly until we offer our final appraisal of Husserl's account of
horizon-analysis.
And then a few lines further Husserl concludes: "The horizon structure be-
longing to every intentionality thus prescribes for phenomenological analysis
and description methods of a totally new kind".
Husserl's assessment of the role and importance of horizon-analysis as a
new method of phenomenological analysis is part and parcel of his increasing
HUSSERL'S NOTION OF HORIZON 235
stress on the ego's role in the constitution of objects. This stress is particularly
clear in Cartesian Meditations, §20, where Husserl observes that "intentional
analysis" leads from analysis of noematic Sinn to analysis of horizon (the
"peculiar attainment" of intentional analysis) and then to the role of the
ego who could "perform" the acts in the horizon (without horizon-analysis,
Husserl says, "the intentionality ... would remain 'anonymous' "). The role
of the ego that most interests Husserl here reflects his increasing emphasis on
"transcendental idealism". In § § 21 and 22 he effectively treats what Kant,
the godfather of Husserl's transcendental idealism, called "categories of the
understanding" and perhaps, in a way, conceived as phenomenological struc-
ture inherent in the ego or its functioning. Specifically, Hussed emphasizes
the role in intention of a priori principles governing the constitution of ob-
jects of various types, "constitutive principles" functioning in the ego so as,
in effect, to impose boundary conditions on the possible experiences that can
belong to the horizon of an act directed toward an object of the requisite
type (cf. El, §90, on "a priori 'conditions of possible experiences'''; also
Ideas, § §142---44, 149-50). For Husserl, the analysis of an act's horizon
lays out or "unfolds" the Sinne of these possible experiences and hence
exposes the "transcendental" principles presupposed by the Sinn of the act,
by virtue of which its object is constituted as belonging to a specific ontologi-
cal category.
Transcendental idealism is not a central concern of ours in this book,
however. But horizon-analysis brings the ego into intentional analysis in
another way that bears greatly on our own assessment of its importance. We
shall see that the implicit Sinne correlated with an act's horizon are not
simply analytically tied to the act's explicit Sinn; rather, these implicit Sinne
are fIXed by the explicit Sinn only in conjunction with certain background
beliefs (or their Sinne) held by the subject of the act. Some of these beliefs
will presumably be universally held "constitutive principles" of the sort
Husserl emphasizes in Cartesian Meditations; but others, as Hussed himself
argues (see, e.g., El, §83a), have their foundation in the particular experiences
of the subject of the act. As horizon-analysis unfolds an act's impliCit Sinne
it thus displays the contribution that is made to the intention by its being
present in the context of a particular subject's other experiences and beliefs.
It is this connection between horizon, Sinn, and ego that we shall stress, and
we shall argue for it in Part 3 below. Principally, then, we shall view horizon-
analysis as a particular way of explicating Sinn, or noematic meaning, and this
view will provide a direct link to our later discussion of meaning and intention
in possible-worlds theory and possible-worlds semantics (in Chapters VI and
236 CHAPTER V
Vll). But we shall also see that horizon-analysis is not "pure" meaning-analysis,
as its appeal to background presuppositions of an intention introduces a new,
"pragmatic" twist (see Chapter VI, Section 1.4, below).
... An experience [Erlebnis) that has become the object of an ego-glance [lchblickes],
and thus has the mode of the looked at [Erblickten) , has its horizon of experiences that
are not looked at; that which is grasped in the mode of "attention" [''Aufmerksam·
keit") . .. has a horizon of background inattention .... (Ideas, § 83, p. 201. A footnote
on p. 202 compares "horizon" with "background" as discussed in § 35.)
Thus, just as, on Husserl's early account, the horizon of an object of percep-
tion as perceived is that which in the perception is given inattentively as
background surrounding the object (typically) in space, so this horizon of an
act as an object of reflection - its temporal horizon of experiences - is that
which in the reflection is given inattentively as background surrounding the
act in phenomenological time, in the stream of experience.
To generalize this early notion of horizon, we might define the horizon
of an object as given in an act to be that which in the act is given inatten-
tively, "surrounding" that which is given attentively or in the focus of atten-
tion. The horizon of an object is then relative to the act in which the object
is intended and to its noematic structure, indeed, to its noematic Sinn. The
degree or mode of attention with which something is given in an act is a
special act "character" or "way of givenness" and so is reflected in the
"Gegebenheitsweise"-component of the act's noema (cf. Chapter III, Section
2.4, above; and Ideas, §92). Those components of the act's Sinn to which
the "Gegebenheitsweise"-component "attentively" is not attached make up,
then, what we might call the horizon-portion of the Sinn.
This early notion of an object's horizon is clearly quite different from that
238 CHAPTER V
... Whatever exists in reality, but is not yet actually [aktuell] experienced [erfahren] ,
can come to givenness, and ... then that means [besagt] it belongs to the indeterminate
but determinable horizon of my current actuality of experience [Erfahrungsaktualitiit].
But this horizon is the correlate of the components of indeterminacy that essentially
attach to the thing-experiences themselves, and these leave open - always essentially -
possibilities of filling out, which are by no means arbitrary, but rather predelineated
[vorgezeichnete] according to their essential type, motivated. All actual [aktuelle]
experience points beyond itself to possible experiences, which themselves again point to
new possible experiences, and so in infinitum. (Ideas, §47, p. 112.)
This is the notion of horizon that is our concern: every act has a horizon that
consists of various possible acts (in Husserl's example, perceptions) in which
the object of the given act would be intended under various further aspects
or conceptions. 8
An act's horizon in this sense can also be construed as the "horizon" of
the act itself taken as object of internal reflection: the quoted passage is
introduced a few lines earlier by way of the observation that "any cogito
... that relates to the world ... not only intends something worldly but is
itself intended in the consciousness of internal time" (p. 44).9 However, we
may, like Hussed, omit any further reference to any particular internal reflec-
tion on the act, for the horizon will include what any careful reflection would
find.
Husserl shortly goes on to say, "The horizons are 'predelineated' poten-
tialities" (p. 45). What Husserl has in mind, as we shall see in Part 3, is that an
act's horizon is fIXed by specific components of the act's Sinn (together with
240 CHAPTER V
Now, the lowest or most basic "stratum" of the sense of objects in the
242 CHAPTER V
... Every real thing [Ding) in nature is represented by all the Sinne and Satze through
which it, as so-illld-ta determined and further to be determined, is the correlate of
possible intentional experiences [Erlebnisse), thus, represented by the maIlifolds [Man-
nigfaltigkeiten) of 'Tilled nuclei", or what here signifies the same thing, all possible
"subjective ways of appearing", in which it can be noematically constituted as iden-
tical ....
. . . To every thing and ultimately to the whole thing-world ... there correspond the
manifolds of possible noetic events, the possible experiences [Erlebnisse) of single indi-
viduals and of individuals in community that relate to it, experiences that ... parallel ...
the noematic manifolds just considered .... The unity of the thing stands over against
an infinite ideal manifold of noetic experiences [ErlebniBSe) •.. ,all in agreement in that
they are consciousness of "the same". (pp. 329-30.)
... Every consciousness has, of essential necessity, its place in a particular manifold
[Mannigfaltigkeitj of consciousness that corresponds to it, a syntactic open infmity of
possible modes of consciousness of the Same - a manifold that has, so to speak, its
teleological center in possible "experience" [Erfahrungj. (App. II, § 2a, pp. 315-16;
with trans. changes.)
role of perception in the horizon (and also the manifold), we have also seen
reasons why, for Husser!, the horizon should include more than the percep-
tual. As maximally conceived, both horizon and manifold extend beyond the
perceptual to include acts of any thetic character.
On our basic definition, an act's horizon includes, then, those possible acts
(perceptions and other acts) whose Sinne are not only co-related with the
act's Sinn but are also compatible with and more determinate than it in predi-
cate-content. In contrast, the definition of manifold asks only co-relatedness
of Sinne. However, as we shall see in Section 3.4, Husser! allows that further
experience may to some extent controvert one's current perception of an
object, and for that reason he includes in a perception's horizon possible
perceptions that present the same object "as otherwise" than it is presently
perceived. This relaxes the requirement that the Sinne of acts in the horizon
must be strictly compatible with the Sinn of the act. The only remaining
difference between horizon and manifold is that the Sinne of acts in the
horizon, but not the manifold, must be more determinate in predicate-content
than the act's own Sinn. But this requirement on horizon serves only to elimi-
nate redundancy of content between an act and its horizon, by excluding
from the horizon acts whose Sinne have no predicate-content not already
included in the Sinn of the original act. We may, if we like, also relax this
requirement; then we allow into the horizon of an act the act itself as well as
acts in which the object is less completely "determined" than in it.
We may assume, then, that the maximal horizon of an act merges with its
manifold. What is important is that an act's horizon should consist of just
those acts that are "predelineated", given the "indeterminacy" in the act's
Sinn: it includes those possible acts that intend the same object as the original
act but intend it under varying aspects or conceptions, and the more signifi-
cant of these possible acts intend the same object under more determinate
conceptions. These possible acts making up an act's horizon are just the acts
that make up the manifold associated with the act.
an act, we know, do not prescribe everything about the object of the act, and
much that they do prescribe is non-specific. Still, not everything is left open
with respect to the properties of the object that are ;.mprescribed or not
specifically prescribed. The Sinn of the act establishes what Husser!, in
Experience and Judgment (§21c), calls a "frame of indeterminateness"
(Unbestimmtheitsrahmen) into which the implicit Sinne, and what they
prescribe of the object, must fit. The predicative content of the act's Sinn
thus places limits, or boundary conditions, on the possible acts that are al-
lowed into the act's horizon (and on their Sinne). The horizon is predelineated
in the act, determined by its Sinn, because the horizon comprises only those
possible acts whose Sinne are compatible with, but more determinate in
content than, the Sinn of the act. Husser! formulates this view of predelinea-
tion in Cartesian Meditations, § 19:
The horizons are "pre delineated" [vorgezeichnete) potentialities. We say also: We can
ask any horizon what "lies in it", we can explicate or unfold it, and "uncover" the
potentialities of conscious life at a particular time. Precisely thereby we uncover the
objective Sinn implicitly meant, though never with more than a certain degree of fore-
shadowing, in the actual cogito. This Sinn, the cogitatum qua cogitatum, is never present
to actual consciousness [vorstellig] as a finished datum [Gegebenes] ; it becomes "clari-
fied" only through explication of the given horizon and the new horizons continuously
awakened. The predelineation itself, to be sure, is at all times imperfect [unvollkommen);
yet, with its indeterminateness, it has a structure of determinateness. For example: the
die leaves open a great variety of things pertaining to the unseen faces; yet, it is already
"construed" in advance as a die, in particular as colored, rough, and the like, though
each of these determinations always leaves further particulars open. This leaving open,
prior to actual closer determinings (which perhaps never take place), is an element
[Moment] included in the given consciousness itself; it is precisely what makes up the
"horizon". (P. 45; with trans. changes.)
contents of one's judgements are not ... related as premisses are to a con-
clusion" (§3, p. 272). Where a motivation is objectively justified, the moti-
vating belief furnishes a contingent "ground of probability" for the belief it
motivates (§3, p. 272). But generally, motivation arises not by inference but
by association, by "association of ideas" (§4). And association is the process,
usually "passive", of assimilating, for instance, what one currently sees with
a kind of thing one has previously seen and otherwise experienced (cf. eM,
§ §39, 50, 51; EJ, § § 16, 44).
Beliefs motivated by virtue of association will presuppose general beliefs
one has developed about things of the kind being associated. By "associating"
what I now see with what I have seen in my previous perceptions of desks, for
example, I see this object as also being a desk. Given this association, what
"motivates" my belief that this desk has four legs on its unseen underside is
my general belief, based on past experience, that desks usually have four legs.
Thus, the possibility of seeing that the desk actually does have four legs is
"motivated" in that it - unlike the "empty" possibility of seeing that it has
ten legs - is compatible not only with what I currently perceive but also with
such broad beliefs about desks in general. We conclude that, for Husser!, the
requirement that the possible acts in an act's horizon be "motivated" possibi-
lities means that their Sinne must be compatible not simply with the act's
own, "explicit", Sinn but with that Sinn together with the Sinne of certain of
the subject's background beliefs, especially his broad beliefs about the kind of
object specified by that Sinn. Husser! writes in Experience and Judgment:
The object is present from the very first with a character of familiarity; it is already ap-
prehended as an object of a type more or less vaguely determined and already, in some
way, known. In this way the direction of the expectations of what closer inspection will
reveal in the way of properties is predelineated (§ 24a, p. 113; with trans. changes; cf.
also §83a).
(or their Sinne) "motivate" the possibilities making up the horizon by pre-
scribing what would and what would not count, for the subject, as further
"determination" of the object as it is given in the present act. In this way
the horizon of a particular act is relative to the subject's fundamental back-
ground beliefs about objects of the kind the intended object is given as.
Of course, one and the same object is given in an act as being of fOore
and less generic kinds: for example, what is given as a tree is also given more
generally as a physical object, and it may be given more specifically as, say,
a citrus tree or as a grapefruit tree (cf. Ideas, § §9ff; EJ, §84). Accordingly,
the background beliefs that contribute to the pre delineation of the horizon
may be more or less general, including the subject's broadest, presumably
a priori, beliefs about physical objects generally, as well as less generic, and
empirical, beliefs that the subject has acquired about objects of more specific
kinds.
The role of the subject's more general beliefs in constraining the possible
acts in an act's horizon comes out in a somewhat convoluted way in § § 21-
22 of Cartesian Meditations, where Husserl sees horizon-analysis as revealing
the "rule-structure of the transcendental ego" (§ 22, p. 53; with trans.
changes). (Cf. Section 1.4 above.) But Hussed is more explicit on this feature
of horizon in Ideas:
Every category of object ... is a general essence that ... prescribes a general rule given
with insight for every particular object of which we become conscious in the manifold of
concrete experiences [Erlebnisse) . ... It prescribes the rule for how an object subsumed
under it is to be brought to full determinacy with respect to sense [Sinn) and way of
givenness [Gegebenheitsweise) .... (§ 142, pp. 349-50.)
That the "rules" (or "rule") of which Hussed is thinking here are of the most
general and basic sort, concerning physical objects generally, becomes clear
at the end of the same paragraph in Ideas:
... The unseen determinations of a thing [Dinges) - this we know with apodictic evi-
dence - are, like thing-determinations in general, necessarily spatial: this gives a lawlike
rule for possible spatial ways of completion of the unseen sides of the appearing thing;
a rule that, fully developed, is called pure geometry. Further determinations of a thing
are the temporal and the material: to them belong new rules for possible (thus not freely
chosen) completions of sense [Sinnes) and ... for possible... appearances. What the
essential content of these may be, to which norms their matters, their noematic (or
noetic) apprehension-characters [Auffassungscharaktere), are subject, is also predeline-
ated a priori. (P. 350; cf. § § 149, 150.)
These rules, then, are part of the concept the subject has of the basic ontol-
ogical category - here, physical object - to which the object, as intended,
HUSSERL'S NOTION OF HORIZON 251
No apprehension is merely momentary and ephemeral .... This lived experience itself
... may become "forgotten"; but for all this, it in no way disappears without a trace; it
has merely become latent. ... The object, even though it has sunk into passivity, remains
constituted as the one having been determined by the determinations in question ....
This means that, even if the object has been given again originally, that is, perceptually,
and is not only realized in memory, the new cognition has a content of sense essentially
HUSSERL'S NOTION OF HORIZON 253
other than the preceding perceptions. The object is pregiven with a new content of sense;
it is present to consciousness with the horizon . .. of acquired cognitions: the precipita-
tion of the active bestowal of sense, of the preceding allotment of a determination, is
now a component of the sense of apprehension [Auffassungssinnesl inherent in the
perception, even if it is not really explicated anew. But if the explication is renewed, it
then has the character of a ... reactivation of the "knowledge" already acquired. ( § 25,
pp.122-23.)
act, while the background Sinne are only "passively" present in it (cf. EJ,
§ §23b, 25, 67b; CM, §38).lS
Let us also be clear in separating the notion of background meaning from
the notion of "implicit" meaning that Husserl uses. Implicit meaning, or
"horizon-meaning", fills in the possibilities left open by the act's explicit
meaning together with its background meaning.
then: the acts must yet be co-directed, though their respective contents can-
not both be true of the object. Those possible perceptions that conflict in
admissible ways with the original perception belong in the act's horizon
insofar as they fit into chains of progressive perceptions of the given object,
though they bring with them corrections in the acts' progressive characteriza-
tion of the object.
Every experience [Erfahrungl has its own horizon . ... This implies that every experi-
ence refers to the possibility ... of obtaining, little by little as experience continues, new
determinations of the same thing. Every experience can be extended in a continuous
chain of explicative individual experiences, united synthetically as a single experience,
open without limit, of the same ....
Thus every experience of a particular thing has its internal horizon .... (EJ, § 8, p.
32.)
... This intending-beyond [Hinausmeinen: that is, implicit or horizon intending) is
not only the anticipation of determinations which, insofar as they pertain to this object
of experience, are now expected; in another respect it is also an intending-beyond the
thing itself with all its anticipated possibilities of subsequent determinations, i.e., an
intending-beyond to other objects of which we are aware at the same time, although at
first they are merely in the background. This means that everything given in experience
has not only an internal horizon but also an infInite, open external horizon of objects
cogiven .... (EJ, § 8, p. 33; with trans. changes.)
In this passage it might seem that the mode of attention is to playa role in
defining horizon since the external horizon includes "other objects of which
we are aware ... merely in the background". But that appears not to be
Husserl's point if we note that Husserl says, further on in the second quoted
paragraph, that the "awareness" we have is by " 'induction' or anticipation",
by appeal to background knowledge or "preknowledge" (pp. 32-33). Indeed,
Husserl explicitly generalizes external horizon beyond the background of
attention in a later section:
[By) the external horizon of the object ... we had in view [Husserl is referring to § 22),
above all, its objectively copresent surroundings [Umgebung), ... these surroundings
being always cogiven by way of background as a plurality of simultaneously coaffecting
substrates ....
But it is not only what is cogiven originaliter as perceptible in the objective back-
ground which provides occasion for relational contemplation and the acquisition of rela-
tive determinations, but also the horizon of typical preacquaintance [Vorbekanntheit)
in which every object is pregiven. This typical familiarity codetermines the external
horizon as that which always contributes, even though it is not copresent, to the deter-
mination of every object of experience. It has its ground in the passive associative rela-
tions of likeness and similarity .... (EJ, § 33, pp. 149-50.)
slips back and forth between object-horizon and act-horizon, suppressing the
distinction. The elision is even more obvious where Husserl describes internal
and external horizon in The Crisis of European Sciences and Transcendental
Phenomenology :
For consciousness the individual thing is not alone; the perception of a thing is per-
ception of it within a perceptual field. And just as the individual thing in perception
has meaning only through an open horizon of "possible perceptions", insofar as what
is actually perceived "points" to a systematic manifold of all possible perceptual ex-
hibitings belonging to it harmoniously, so the thing has yet another horizon: besides
this "internal horizon" it has an "external horizon" precisely as a thing within afield of
things; and this points finally to the whole "world as perceptual world". (§47, p. 162;
with trans. changes.)
Here it is clear that the relevant act-horizon belonging to the act of perception
corresponds to the relevant object-horizon belonging to the object of the
perception. Clearly, then, the distinction between internal and external
horizon of an object of perception will be reflected in a corresponding
division within the act-horizon of the perception itself.
An act's temporal horizon includes, then, to begin with, possible future per-
ceptions that the perceiver anticipates he will (or could) have. ("Protention"
HUSSERL'S NOTION OF HORIZON 259
ultimately "the whole world". And the possible perceptions in the horizon
align into families of "possible verification chains", temporal sequences of
continuous perceptions leading from the original act into both past and
future. Each family progressively determines the object of perception, tracing
out at the limit an entire possible perceptible history of the object and its
world. For our later discussions of the role of horizon in intentionality theory,
the most interesting structural feature of an act's horizon will be this struc-
ture of verification chains.
might seem to call in the thetic character of perception. However, the special
contribution that perception makes to horizon lies within the special items
of sense that go into perceptual Sinne: items prescribing visible shape, color,
and so on.
NOTES
1 Here" we have in mind only one sort of Sinn, the Sinn related to an object intended
in a "definite", direct-object intention. This is the only sort of Sinn that Hussed de-
scribes in detail (cf. Chapter IV, Section 3.1). But where acts have Sinne of different
structure, failures of substitutivity in sentences describing acts may require an account
somewhat different in detail.
2 Cf. EJ, § 21c, pp. 96-98, on "open possibilities", and §67a on the "horizon of open
possibilities" .
3 A notion of horizon also appears in Heidegger: thus, time is the "horizon" of human
being, much as for Hussed in one of his uses of 'horizon'. The translators of Being and
Time note that in English 'horizon' may have the connotation of something we can
expand and go beyond, whereas in German 'Horizont' connotes something that sets
limits which we cannot go beyond but must remain Within. This comment would seem
to apply to Hussed's use of 'Horizont' as well as Heidegger's. See Martin Heidegger,
Being and Time, trans. by John Macquarrie and Edward Robinson (Harper & Row, New
York, 1962), p. 19, n. 4.
4 Hussed also uses the term 'manifold' in another, not directly related sense, where the
"theory of manifolds" is the mathematical theory of pure (uninterpreted) formal logical
systems (see LI,ldeas, FTL).
5 Cf. § 92 of Ideas, where modes of attention are classified as modes of "actuality"
(Aktualitiit). Translation of 'aktuell' is bound to lead to some awkwardness: 'aktuell'
ordinarily means not 'actual' (wirklich) , but 'topical' or perhaps 'focal', which would
be apt to modes of attention; yet Hussed also contrasts 'aktuell' or 'Aktualitlit' with
'possible' or 'potential' (cf.ldeas, §47).
6 This is surely also how 'aktuell' and 'Aktualitlit' are used when Husserl introduces the
notion of horizon in Cartesian Meditations, in § 19, assigning to each actual act a horizon
of possible acts. This point of usage is confusing. Perhaps in Ideas Husserl was thinking
that what is at the periphery of attention could potentially be made the center of atten-
tion. But what is not known at all should not be thought of as at the periphery of
attention, though Husserl seems to want it in the indeterminate horizon.
7 Note that where Husserl speaks of a plurality of horizons associated with an act, these
seem to be portions of one basic, overall horizon of the act. It is the latter we shall
consistently refer to as horizon.
8 Earlier notions of horizon are not completely absent from Cartesian Meditations. In
§ 18 Hussed cites the horizon of "immanent time": cf.ldeas, § 83.
9 We might prefer to say merely that an act is always potentially intended in internal
reflection. Husserl himself, though, believes that every act includes an actual, albeit
"inaktuell" (non-attentive), constituent consciousness of itself. Cf.ldeas, §4S, pp. 104-
105, and Time, Appendix XII.
HUSSERL'S NOTION OF HORIZON 265
1. HORIZON-ANALYSIS AS EXPLICATION OF
SINN AND INTENTION
identifies the meaning of a sentence with the set of possible states of affairs -
maximally, possible worlds - in which the sentence would be true. And,
again roughly, it identifies the meaning of a singular term with a set of
possible individuals, namely, the set of individuals to which the term would
refer, respectively, in different possible situations ·or possible worlds. More
precisely, the theory identifies meanings with functions on possible worlds:
the meaning of a sentence is the function that assigns Truth to each possible
world in which the sentence would be true; and the meaning of a singular
term is the function that assigns to each possible world the individual re-
ferred to by the term in that possible world. We shall develop the details
of this theory of meaning, and some of the motivations for it, in Part 2
below.
Interestingly, Husserl's correlation of Sinn with horizon aligns closely
with the Carnapian theory of meaning if Husserl is granted the range of possi-
ble entities assumed by the Carnapian theory. As we saw in Chapter V,
Section 2.3, there corresponds to the horizon of an act the horizon of the
object of the act (as the object is intended in the act). This "object-horizon"
we defined as the set of possibilities that are left open by, or that are compat-
ible with, what the act's Sinn prescribes of the object and its environment.
And this set of possibilities, where the object's environment is considered
maximally as including the entire world about the object, is the very set of
possibilia that the Carnapian theory would identify (roughly) with that Sinn.
The significant difference between the Husserlian and the Carnapian versions
of this account of meaning would be on reduction: the Husserlian theory
would correlate but not identify meaning with an array of possibilia (or with
an appropriate function on possible worlds).
In Section 3.2 below we describe in closer detail the alignment we find
between the object-horizon that Husserl correlates with an act or its Sinn
and the array of possibilia (or an appropriate function on possible worlds)
that the Carnapian identifies with that same meaning. We can better appre-
ciate the alignment at that point, after we have developed details of the
Carnapian theory. We should say, however, that since Husserl addresses only
a specialized form of Sinn (one related to individual objects) the general align-
ment we envision depends on extrapolating from the form of Sinn whose ex-
plication in horizon he describes.
Importantly, this alignment of Sinn with a set of possibilia, or with a func-
tion on possible worlds, is the basis for our later assimilation of Husserl's
theory of intentionality and modern pOSSible-worlds semantics for intentional
modalities.
270 CHAPTER VI
ground meanings are the Sinne of certain of the subject's background beliefs
about the object intended (cf. Chapter V, Sections 3.2 and 3.3, above),
including, in our earlier example of my seeing a tree, my beliefs that the tree
has a back side, that it is a maple tree, and that it, is the tree I sat under with
Cecelia. Now, so far we have thought of horizon-analysis as giving an explica-
tion of an act's own, explicit Sinn, of that particular meaning in its own right.
But insofar as background beliefs playa role in derming horizon, analysis of
an act's horizon goes beyond explication of the act's Sinn alone, of what is to
be found "in" that Sinn itself. For horizon-analysis reveals not only the role
of that Sinn in the intention achieved in the act but also the role of these
background beliefs and their Sinne.
Some of the background Sinne in an act are conceptually tied to the act's
explicit Sinn, but others are not. In our own example, propositional Sinne
that make up the subject's theory or conception of trees may be so tied to
the act's Sinn, for at least some of these Sinne are conceptually presupposed
by the sense "is a tree", which is a component of the explicit Sinn. And other
senses, e.g., "has a back side", may be entailed (in a sense appropriate to pred-
icate-senses) by these propositional Sinne together with the sense "is a tree",
and so they will be tied analytically to the act's explicit Sinn. Explication of
such background senses will not carry the horizon-analysis beyond explica-
tion of what is "contained" in the act's Sinn at least by implication. However,
the senses "is a maple tree" and "is the tree I sat under with Cecelia" are in
no way conceptually, or analytically, tied to the act's explicit Sinn: they are
only "passively" present in the act, the former by "passive" association with
past experiences, the latter by potential recollection of previous experiences.
So, insofar as these senses constrain the act's horizon, the horizon-analysis
goes beyond explication of the act's explicit Sinn and of the meanings it
includes, either as part of its content or by implication.
Insofar as an act's horizon is keyed to background meanings or beliefs
that are not conceptually tied to the act's explicit Sinn, we must emend our
view of horizon-analysis as a way of explicating the (explicit) Sinn of an act.
For horizon-analysis seems to be more than the "semantic" analysis of
meaning that we have heretofore considered. What we propose is that horizon-
analysis, as conceived by Hussed, should be seen as further including the
phenomenological counterpart of a "pragmatic" analysis of meaning and
intention.
Traditionally, semantics studies the meaning and reference of words,
while pragmatics studies the influences on meaning and reference exerted
by various features of the occasions or contexts of the use of words. s The
EXPLICATION OF MEANING 273
other systems. A maple tree in New England might well harbor an opossum
but is unlikely to harbor a platypus or a koala bear. Are a person's beliefs
concerning local animal life then relevant material for analysis of the horizon
of his act of seeing a maple tree? We should like to see the line drawn closely
enough for horizon-analysis to explicate only those background meanings or
beliefs that play a significant role in the intention considered. However, we
ourselves have no sharper criterion to offer.
The difficulty cited concerns background beliefs that influence the inten-
tion but are not analytically related to the explicit Sinn. Now, one thing we
might consider is Quine's view that there is no sharp distinction between
analytic and synthetic propositions; 8 then there would be no sharp separation
between Sinne or beliefs that are analytically related to an act's Sinn and
those that are not. The sort of case on which we have dwelt with Hussed, that
of perceiving a natural object, lends itself to such a view. For it is not clear
whether it is a matter of meaning or of empirical botanical theory that, for
example, the leaves on one side of a tree must be of the same kind as those
on another side. Further, as Hussed rightly held, the essences of physical
objects - e.g., species of trees - are "transcendent" and so are not exhaus-
tively specified by any fmite number of propositions or beliefs (cf. Ideas,
§149, p. 365; El, §83). And this view could pose problems for the analytic/
synthetic distinction: some propositions that detail "essential" features of
(say) maple trees would then nonetheless not be involved in a person's con-
ception of maple trees and thus would not be analytically tied to the Sinn
of his act of seeing a maple tree. 9 If Quine's view is right, then there should
be no sharp separation between the "conceptual" and the "pragmatic" pre-
suppositions of an act and, probably, no sharp boundary defining those of the
subject's background beliefs that horizon-analysis may reach. We shall not,
however, cast ourselves into the analytic/synthetic dispute; and we have little
direct evidence of what Husserl's view might be on the matter.
objects of consciousness. Now, the relations of a particular act to its ego and
to other of the ego's experiences or intentions are what we have called prag-
matic features of the act; and we have seen that the analysis of such features
in horizon-analysis goes beyond analysis of the act's noematic Sinn per se.
Insofar as horizon-analysis in this way goes beyond pure noematic analysis of
the act's explicit noematic structure, it brings to phenomenological analysis
"methods of a tot~y new kind". So our view of horizon as yielding a prag-
matic rather than a purely semantic, or noematic, explication of an intention
seems to be a generalization of Hussed's own view of what is distinctive about
horizon-analysis, of how it goes beyond pure noematic analysis.
Our view of horizon-analysis thus provides an interesting perspective on
Hussed's own evaluation of the importance of the notion of horizon. Yet our
stress is probably somewhat different from what Hussed himself had in mind.
In assessing the horizon of an act of, say, seeing a tree (in Chapter V, Sections
3.2 and 3.3), we saw that the background beliefs whose role in horizon most
interests Hussed are either a priori or very general empirical beliefs about
the kind of object intended (e.g., that trees are three-dimensional physical
objects and thus have back sides, or that the leaves on one side of a tree gen-
erally resemble those on the other). Our own assessment of horizon-analysis
as pragmatic analysis, however, also stresses the role in horizon of concrete
background beliefs (e.g., that this is a tree I have seen before). We saw that
Hussed does address these beliefs as well. But, since he does so less frequently
and somewhat less pointedly, it is not clear to what extent Hussed thought
them important in the consideration of horizon.
We should note that horizon-analysis leads into what Husserl calls "genetic"
phenomenological analysis. Insofar as horizon-analysis serves to explicate the
role played in a given act by the subject's passive associations and recollec-
tions, it calls into phenomenological analysis the dependence of an act on
the subject's past experience. Whereas "static" phenomenological analysis
describes an act's noema, specifically its explicit Sinn, genetic analysis de-
scribes an act's genesis in the subject's internal life-history. Thus, genetic
analysis concerns the history of the subject's experiences or acts of conscious-
ness and its influence on his present experience (cf. FTL, AppendiX II).
two correlative methods of explicating an act's Sinn. By laying out the act-
horizon associated with a given act, the horizon of possible acts or experiences
whose Sinne are compatible with the act's Sinn, we layout in the realm of
experience "what" is experienced in the act as such, the meaning, or Sinn, of
the act. Thus, if we imagine ourselves as having these possible experiences,
we effectively dramatize for ourselves, in consciousness, the Sinn of the given
act. Correlatively, by laying out the object-horizon associated with a given act,
the horizon of possibilities compatible with what is prescribed by the act's
Sinn, we layout the act's Sinn - or its projection, so to speak - in the realm
of possibilia. .
Now, there are two aspects of horizon-analysis that we find to be of spe-
cific importance and that we shall pursue through the remainder of the book.
First, what is achieved in the explication of an act's Sinn in terms of the act's
object-horizon is effectively just what would be achieved in the Carnapian
analysis or explication of that Sinn in terms of possible worlds. The case for
this claim is to be found in Parts 2 and 3 below. This claim is the basis for
the assimilation we pursue in Chapter VII of a Husserlian approach to
intentionality and a possible-worlds semantics for propositional attitudes
(such as the type of semantics laakko Hintikka has proposed). Second,
the "pragmatic" element we found in the notion of horizon (act- or object-
horizon) carries Husserl's analysis of intention beyond what can be achieved
by his basic Fregean method of Sinn-analysis. We draw on this aspect of
horizon-analysis in Chapter VIII, where we pursue the phenomenological
analysis of certain definite, or de re, intentions in ways that are pragmatic.
These two features of horizon-analysis - its connection with possible-
worlds analysis of meaning and its connection with pragmatic analysis of
meaning and intention - are logically independent. Let us bring them into
somewhat finer perspective.
In semantics, the possible-worlds theory of meaning can be seen as a
development of the basic Fregean, and Husserlian, conception of meaning:
the possible-worlds approach to meaning is basically a specification within
possible-worlds theory of what meanings must do in determining referents.
This approach to meaning has proved to be of significant heuristic value as
an especially effective way of explicating particular meanings and how they
function. In phenomenology, horizon-analysis a la Husserl can be seen as a
similar development of the basic Husserlian analysis of an act's Sinn: analysis
of an act's horizon (specifically, its object-horizon) is a specification in terms
of possible situations or possible worlds of what the act's Sinn must do in
prescribing what is intended in the act. And horizon-analysis, too, can have a
EXPLICATION OF MEANING 277
any possible situation. For such a criterion should not only determine what
an expression refers to given the actual state of affairs, given the relevant
facts that actually do obtain; it should also determine what the expression
would refer to if somewhat different facts obtained. Thus, in 'The Modes
Meaning', Lewis linked with an expression not only an extension and inten-
sion but also what he called its "comprehension":S Lewis defined the com-
prehension of a term or predicate as the set of "consistently thinkable"
individuals falling under it - i.e., the possible as well as the actual individuals
to which it applies. The intension of an expression, Lewis held, determines its
comprehension as well as its extension, and, further, its comprehension deter-
mines its intension.
This notion of comprehension, of extension among the possible as well as
the actual, contains the germ of Carnap's later analysis of intension in terms
of possible worlds. But Lewis' notion is unsatisfactory as it stands. We cannot
take terms and predicates to apply, or fail to apply, to individuals simpliciter;
they apply only relative to worlds in which those individuals reside. Whether
an expression applies to some "consistently thinkable" individual depends on
whether the criterion for applying the expression is satisfied by that individ-
ual, and whether the criterion is satisfied depends, in turn, on what is true of
that individual in the world in which it occurs. For instance, relative to the
actual course of English history, the term 'First Lord of the Admiralty in
1939' refers to Winston Churchill. But relative to a different course of events,
a different possible history or "world", it refers to, say, Neville Chamberlain.
And in some other world it may refer to a merely possible individual, one
that occurs in that world but not in this, the actual world. It is similar with
predicates. For example, in some "consistently thinkable" situations, or
possible worlds, Robin Hood satisfies the predicate 'gave to the poor' while
in others he does not. Generally, then, the notion of extension should be
relativized to worlds. Let us say that the extension of a singular term in a
world is that individual (if any) denoted by the term in that world, the ex-
tension of a predicate in a world is the set of individuals (or n-tuples) that
satisfy it in that world, and the extension of a sentence in a world is its truth-
value in that world.
In this way the notion of extension can properly be extended beyond the
realm of the actual. But the result is not Lewis' notion of comprehension as
extension among possible individuals; it is the notion of extension in possible
worlds. To refine Lewis' notion, then, let us define the comprehension of an
expression as the set of ordered pairs (w, e), where w is a possible world and e
is the extension of the expression in w.
EXPLICATION OF MEANING 281
that satisfy it in that world; and the intension of a sentence (e.g., 'Robin Hood
gave to the poor') is the function that assigns to each world the truth-value (if
any) of the sentence in that world. (Note that the defmition allows that ex-
pressions may have no extension in some worlds.) Let us call these functions
from possible worlds to appropriate extensions in worlds meaning [unctions.
Importantly, intensions on this view are to be entities ontologically in-
dependent of the expressions and the language that serve to express them.
Mention of linguistic expressions is thus not essential to Carnap's defmition.
Intensions are defined as meaning functions of various kinds depending on
the kind of extensions they assign to possible worlds, and semantic theory
then assigns to expressions of appropriate syntactic categories functions of
appropriate kinds.
Carnap's proposal amounts to identifying the intension of an expression
with its comprehension (as we have defined comprehension). Under the
standard set-theoretic defmition of functions, a functidn is defmed as, i.e.,
identified with, the set of ordered pairs of which the first member is an argu-
ment of the function and the second is the value of the function at that
argument. Thus, a meaning function from possible worlds to extensions - an
intension - is a set of ordered pairs, where the first member of each pair is a
world and the second is the extension assigned that world. And this set just
is a comprehension. Carnap's proposal is a theoretical identification, a theo-
retical posit, based on what we know of intensions in semantic theory. As
functions, intensions are abstract. And, as identified with the relevant com-
prehension, the intension of an expression both determines and is determined
by the comprehension of the expression.
We should note that there is a complication involved in identifying inten-
sion with comprehension in this way. The proposed analysis of intension may
capture the notion of meaning for certain expressions of fairly elementary
syntactic structure, with relatively simple semantic content, such as 'the house
on the corner' or 'is an animal'. However, intension cannot in general be iden-
tified with comprehension, with the meaning functions Carnap proposed.
Lewis noted, for example, that self-contradictory predicates like 'is a round
square' and 'is a triangular circle' have the same comprehension, viz., the null
set (for us, they have the same extension, the null set, in every world).17 Yet
these predicates do not have the same meaning. Similarly, all tautologies and
all logical truths in general have the same extension (viz., Truth) in all worlds,
yet not all such sentences have the same meaning. Lewis observed that the
meanings of such contradictory or tautological expressions must reflect not
only the comprehensions of the whole expressions but also their syntactic
EXPLICATION OF MEANING 283
structures. They must reflect the way the meanings of the whole expressions
are bullt up from the meanings of their syntactic parts. And Carnap similarly
suggested that synonymous expressions must exhibit the same "intensional
structure".18 Thus, while certain elementary intensions may be defmed as
functions from worlds to extensions, the intensions of more complex expres-
sions must be more complex abstract entities constructed from such elemen-
tary intensions in ways reflecting appropriate syntactic structure. Carnap's
proposal has been developed along these lines, though we shall not pursue
that development for our own purposes. 19
... For any term, its connotation [Le., intension) determines its comprehension; and
conversely, any determination of its comprehension would determine its connotation;
by determining what characters alone are common to all the things comprehended. In
point of fact, however, there is no way in which the comprehension can be precisely
specified except by reference to the connotation, since exhaustive enumeration of all
the thinkable things comprehended is never possible. 20
A similar point holds for our revised notion of comprehension. Even if in-
tensions and comprehensions stand in a one-to-one correspondence, there
remains an important difference between them. Intensions must be entities
that our fmite human minds can grasp, for otherwise meanings could not play
their appointed roles in human language. But we cannot completely grasp
comprehensions, which are infinite sets of ordered pairs. So it seems that in-
tensions cannot properly be identified with their corresponding comprehen-
sions. And, therefore, they cannot be identified with their corresponding
meaning functions, since meaning functions are just comprehensions.
Carnap's proposal that meanings, or intensions, be identified with meaning
functions is prompted by the needs of semantic theory, principally by the
requirement that intension determine extension in every possible world. But
Lewis' caveat points up the relevance of a further important part of Husserl's
theory of meaning. The notion of meaning, for Husserl, is not exclusively
or primarily a linguistic notion, for language is used to express underlying
284 CHAPTER VI
thought, the "content" (or noematic Sinn) of which is the meaning ex-
pressed. Fundamentally, on this view, meanings are the objective "contents"
of consciousness by virtue of which acts of consciousness are directed toward
their objects. (Cf. Chapters III and IV above). These claims about meanings
go beyond the demands of pure semantic theory, entering the broader arenas
of philosophy of language and theory of intentionality. But insofar as we
accept these claims, we cannot identify intensions, or meanings, with their
corresponding meaning functions, or comprehensions.
Another alternative is perhaps compatible with these further considerations
about meaning, however: one might identify intensions with meaning func-
tions and resist, instead, the identification of meaning functions with com-
prehensions. It is the set-theoretic reduction of functions to sets of ordered
pairs that identifies meaning functions with comprehensions, but that reduc-
tion can also be resisted. One might take a function to be a "rule" for assign-
ing values to arguments and so distinguish a function itself, as "function in
intension", from the corresponding set of ordered pairs it determines, as
"function in extension". One would then allow that different functions may
determine the same set of ordered pairs. The particular appeal of this view of
functions here, of course, is that it would give a more Husserlian slant to
Carnap's proposal to identify intensions with certain functions from possible
worlds to extensions. But there are two reasons why we shall not follow this
line. First, it is the "extensional" view of functions, as sets of ordered pairs,
that has been most widely used, and it is the "extensional" version of Carnap's
proposal that has been most discussed in the literature of possible-worlds
semantics. We want to draw on these more familiar and more widely discussed
notions in developing further the role of intensions in intentionality. Second,
it is the notion of comprehension - understood as certain sets of ordered
pairs, where each pair consists of a possible world and an appropriate ex-
tension - that we shall compare with Husserl's notion of object-horizon and
to which we shall thus appeal as a way of explicating meaning through hori-
zon-analysis. So we shall assume the extensional version of Carnap's proposal,
identifying meaning functions with comprehensions, and correlating - but
not identifying - intensions with their corresponding meaning functions, or
comprehensions.
Even if meanings and meaning functions are not identical, their close
alignment is of considerable importance. For one of the most effective means
of expUcating meaning, of getting a grip on a particular meaning, is to consider
the extensions it determines in a number of different possible situations or
worlds - that is, to grasp a part of its comprehension, its corresponding
EXPLICATION OF MEANING 285
By the property Black we mean something that a thing mayor may not have and that
this table actually has. Analogously, by the proposition that this table is black we mean
something that actually is the case with this table, something that is exemplified by the
fact of the table's being as it is. 22
For Husserl, on the other hand, the meanings of predicates are not properties
(or, in his terminology, "essences"): they are not themselves characteristics
that can be attributed to individuals but are, rather, noematic entities by
virtue of which we intend individuals as having such characteristics or proper-
ties (see Chapter IV, Section 1.4, above). Further, for Husserl, states of affairs
are extensions, or referents, of indicative sentences, while the meanings ex-
pressed by such sentences are abstract, or objective, contents of consciousness
by virtue of which these states of affairs are intended (cf. LI, V, §28;Ideas,
286 CHAPTER VI
§94; FTL, § §48-49). Fregean Sinne, too, seem rather different from
Carnap's intensions. Carnap suggests that, for predicates, his distinction
between intension (Le., property) and extension (Le., class) coincides with
the distinction between what Frege called a "concept" (Le., a function whose
arguments are individuals and whose values are truth-values) and its course of
values (roughly, the class of arguments for which the value of the function is
Truth).23 Now, Frege does not say precisely what sort of entity the sense
of a predicate is. But he does hold explicitly that it is not a "concept": a
"concept", he holds, is the referent, or extension, of a predicate, not its
sense, or intension. 24 And there is also evidence that Frege took "thoughts",
the senses expressed by sentences, to be objective contents of thought rather
than states of affairs. He says: "By a thought I understand not the subjective
performance of thinking but its objective content ... ";25 and he emphasizes
that thoughts can be "apprehended", "grasped", and "communicated". 26
Generally, Husserl's "meanings" (and perhaps Frege's "senses" as well) are
mediators of intention, while Carnap's "intensions" - in Meaning and Neces-
sity, at least - are entities more like those that serve as the objects of every-
day intentions. In Husserl's idiom, Husserlian noematic meanings are "tran-
scendental" entities while Carnapian intensions are "transcendent" - though
this contrast is one of epistemological status rather than one of ontological
status or kind. The difference is in fact brought out rather sharply in Lewis'
distinction of different "modes" or kinds of meaning. In addition to its
intension (or "connotation"), its extension (or "denotation"), and its "com-
prehension", Lewis held that every meaningful expression also has a "signifi-
cation". Whereas, for Lewis, the intension of an expression is a "criterion in
mind" for determining its possible extensions, its signification is that character
that any entity must have in order to qualify as an extension of the expres-
sion. He says:
We shall say that a term signifies the comprehensive character such that everything
having this character is correctly namable by the term, and whatever lacks this character
. . . is not so namable. And we shall call this comprehensive essential character the
signification of the term. 27
the connection, it is not clear that he would also accept the restriction: so far
as we know, he does not say whether the "modes of presentation" correlated
with senses include only presentations that can be made to human minds). As
criteria in mind, for example, it seems quite unlikely that intensions can
legitimately be called upon to determine extensions in every logically, or
metaphysically, possible world; for we human beings seem not to be in
possession of such criteria. Our (tacit) criteria for deciding which individuals
are to be called "persons", or "animals", or "trees", for example, are imperfect
even as applied to the actual world; and it is surely implausible to suppose
them applicable in worlds that, though logically possible, differ wildly from
the world of our actual experience - say, in worlds where the laws of logic
hold but those of physics do not. If intensions are "criteria in mind" in any
literal sense, then, it seems that intensions at best determine extensions in
those possible worlds that do not differ drastically from the actual world as
we conceive it to be - hence, in those possible worlds in which most of our
beliefs are true. And as noematic entities, meanings are entities through
which we intend objects, not as Kantian "things·in·themselves", but as objects
constituted in accord with our past experiences and our peculiarly human
conceptual schemes. (This tenet is the heart of Husserl's so-called "tran-
scendental idealism": cf. Ideas, § §47-50, 55; eM, § §40-41.) The meanings
that play a role in human belief and other intentional phenomena - and in
language, considered as an intentional phenomenon - thus seem best re-
presented by meaning functions whose domains are sets of possible worlds
composed of things conceivable and understandable by human beings, worlds
compatible at least with our basic conceptual scheme.
Where meaning functions are invoked in explication of noematic meanings
at work in particular acts of consciousness, their domain may be restricted
even further. Basically, the role of a noematic Sinn in an act is to determine
which object is intended, and how it is constituted, in the act. In accord with
Husserl's notion of horizon, however, many possibilities concerning the object
and its environment are left open by what is intended in the act; ultimately,
the act's noematic Sinn must determine the object in all these possible
situations that the act's horizon subscribes. But these possibilities, Husserl
emphasizes, are not "empty", merely logical, possibilities; they are, rather,
possibilities "pre delineated" by the act's meaning in consort with the mean-
ings of background beliefs presupposed in the act - sometimes including quite
specific and concrete empirical beliefs about the object intended (cf. Chapter
Y, Part 3). And so the possible situations - ultimately, possible worlds -
that are relevant to the explication of a particular meaning as operant in a
EXPLICATION OF MEANING 289
the extension (if any) that it determines is a single individual. The meaning
of 'the fust European in North America', for example, determines for each
possible world w that one individual (if any) in w who was (in w) the first
European person to arrive in North America, and that individual is then the
extension of the expression in w.
Individual meanings are nicely represented, or explicated, by meaning
functions of a particular sort. For example, the meaning of 'the first European
in Nor,th America' correlates with, and is explicated by, the meaning function
that assigns to each possible world w the person (if any) who in w was the
first European on North American soil. In general, the meaning functions that
represent individual meanings will be those functions that assign to each pos-
sible world at most one individual.
distinct from their referents. Let us not pursue this issue here, however; in
Chapter VIII we shall return to proper names and demonstratives and to their
relation to rigid meanings as these playa role in noematic structures.
An important class of rigid meanings are those that achieve their rigidity
by virtue of individuation; we may call them "individuating" meanings. An
individuating meaning is an individual meaning that captures the "identity",
or the "individual essence", of an individual and for that reason prescribes the
same individual in every world.
Hintikka has stressed the importance of individuating meanings or rigid
meaning functions in the semantics of intentional modalities. For purposes
of analyzing quantification into contexts of propositional attitude, he postu-
lates a set of meaning functions he calls "individuating functions", defmed
as functions that pick out the same individual (if any) from every world. 35
Hintikka considers these functions a species of individual concepts, or in-
dividual senses, which he generally identifies with functions from worlds to
individuals. And he takes the set of individuating functions to capture "the
totality of ways" in which "we recognize one and the same individual under
different circumstances and under different courses of events".36 In Chapter
VII (see esp. Sections 2.4 and 3.3) we shall discuss the role of these functions
in a possible-worlds semantics for sentences of propositional attitude.
Ona Husserlian view, individuating meanings might also be given an im-
portant role in intentionality. We saw in Chapter IV that individuatively
defmite acts, intentions of or about particular indiViduals, require noematic
Sinne that incorporate a conception of the identity of an individual and so
determine that "definite" individual as the object intended. Hence, we might
take these noematic Sinne to be, or to include, individuating senses; individ-
uating senses would then be noematic entities in virtue of which individua-
tively defmite acts achieve their directedness to particular individuals.
There is an important complication in viewing individuating senses as
noematic entities, however. Husserl holds that the complete identity, or the
"individual essence", of an object in nature is necessarily transcendent of
human consciousness, beyond our complete grasp (cf. Chapter VIII, Section
3.2, below). But an individuating meaning is to be precisely a meaning that
captures the individual essence, or identity, of the individual it prescribes.
This complication points up an important difference in the notion of individ-
uating meaning depending on whether we think of meanings as Carnap did
or as Husserl did.
Carnap's intensions are semantic posits, apparently transcendent entities
that need not be correlated in any essential w~y with human consciousness.
292 CHAPTER VI
"IP" do not in this case join into a propositional Sinn, since the perception is
properly directed toward an individual rather than a state of affairs. And so
we will want to represent this Sinn by a function that assigns an individual
(if anything at all) to every world to which it applies. But, further, the X in
the Sinn calls for definiteness of intention, and this suggests "rigidity" of
Sinn: the Sinn somehow determines the same object (if any) in every relevant
possible world. The meaning function that explicates the Sinn should, there-
fore, assign the same individual to each possible world. And, finally, the
predicate-sense "IP" requires that the object prescribed by the Sinn in any
given world be r/> in that world. Thus, we might represent the complete Sinn -
"x as r/>" - of such a perception by the meaning function that assigns to
every possible world the same individual - viz., the one that is actually before
the perceiver - provided it is r/> in that world. (See Chapter VIII, Part 2,
below.)
There is one further complication, however. The function we have defined
represents the perception's "explicit" Sinn, yet we know the perception is
conditioned by certain background beliefs that may characterize the object
further than the explicit content "r/>" does (cf. Chapter V, Sections 3.2 and
3.3). So we might more fully represent the perceptual intention, constrained
by the relevant background beliefs, by the function that picks out that same
object in any world provided that in the given world the object is not only
r/> but also propertied in accordance with the relevant background beliefs.
Alternatively, we might let the background beliefs manifest themselves as
"pragmatic" constraints on the possible worlds that are relevant to the ex-
plication of the given intention: we could then explicate the Sinn, as operant
in the intention, by the function that assigns the same object, provided it is
r/>, to every possible world compatible with the background beliefs in question
(cf. Sections 1.4 and 2.5 above).
Let us turn to propositional acts or attitudes, considering the noematic
Sinne of certain sorts of judgment. Basically, the Sinn of a judgment is a
propositional meaning, expressible by a grammatically complete sentence.
Thus, the Sinn of the judgment described by 'Smith judges that p' is the
meaning of the sentence 'p'. And this meaning can be represented by the
function that assigns Truth to each world in which it is the case that p and
Falsehood to all others (in which 'p' has a truth-value at all). But we also need
to consider further the internal structure of propositional meanings, which
is often of considerable phenomenological interest. In particular, we need to
consider the explication of the "subject" constituents of propositional Sinne,
since these constituents determine the "aboutness" of propositional acts or
attitudes (cf. Chapter I, Section 1.5).
294 CHAPTER VI
We have seen that various sorts of meanings can be explicated (or, by some
accounts, even given ontological analysis) in terms of certain structures of
possible worlds. We wish to show now, in more detail than we have heretofore
provided, that there is a strong basis in Husserl for applying the possihle-
worlds theory of meaning to Husserl's analysis of noematic Sinn and inten-
tion. We do so by showing, first, that Husserl himself is not averse to speaking
of possible objects and possible worlds (though it is not clear that he wishes
to stake out a strong ontological commitment to possibilia), and, second, that
the specific structure he describes in the horizon of an act harbors an exact
structural basis for laying out the corresponding object-horizon, and hence
for explicating the act's Sinn, in terms of possible worlds. Effectively, we
argue, the explication of Sinn in terms of horizon is equivalent to the explica-
tion of Sinn in terms of possible worlds.
(Recall from § 133 that by 'Satz' Husserl means the noematic Sinn together
EXPLICATION OF MEANING 297
with the thetic component of the noema, effectively the whole noema.) In
Experience and Judgment he says:
"The same" object which I just now imagine could also be given in experience: this same
merely possible object (and thus every possible object) could also be an actual object.
Conversely: I can say of every actual object that it need not be actual; it would then be
"mere possibility". (Appendix I, to § §40 and 43, p. 381.)
... The correlate of our factual experience [Erfahrung), called "the actual [wirkliche)
world", emerges as a special case of manifold possible worlds and non-worlds, which,
298 CHAPTER VI
for their part, are nothing other than correlates of essentially pollllible varilltions of the
idea "experiential consciousness" with more or less ordered experiential connections.
(§47, p. 111.)
... What things [Dinge: physical objects) are, the only things about which we make
statements and about whose being or non-being, being so or being otherwise, we can
dispute and rationally decide, they are as things of experience [Erfahrung). It alone is
what prescribes them their meaning [Sinn) and indeed, where factual things are con-
cerned, it is actual [aktuelle) experience.... But if we can subject the experience-forms
of [natural) experience [Erlebnisarten der Erfahrung) and in particular the fundamental
experience [Grunderlebnis) of thing-perception to an eidetic consideration, ... then the
correlate of our factual experience, called "the actual world", emerges as a special case
of manifold possible worlds and non· worlds, which, for their part, are nothing other
than co"elates of essentially possible varilltions of the idea "experiential consciousness"
with more or less ordered experiential connections ....
This holds for every conceivable kind of transcendence that may be treated as ao-
tUality [Wirklichkeit) or possibility. An object that has being in itself [an sich seiender)
is never one such that consciousness and consciousness' ego have nothing to do with it.
(Ideas, §47, pp. 111-12.)
And, Husserl says a page later, "[The] transcendent ... necessarily must be
experienceable [er[ahrbar]" (§48,' p. 113). At issue once again is Husserl's
transcendental idealism. His point is that those objects and worlds, actual or
possible, of which we can meaningfully speak are those our conception of
which is based on "experience", i.e., perception. They are objects, or worlds
composed of objects, that are perceivable or, we would presume, accessible
by inference or by "constitution" based on the perceptual. Shortly Husserl
says, strikingly:
... The formal-logical possibility of realities [Realitiiten) outside the world, the one
spatiotemporal world that isjixed through our actual [aktuelle) experience [Erfahrung),
shows itself in fact as nonsense [Widersinn). If there are worlds, real [realen) things, at
all, then the motivations of experience constituting them must be able to reach into my
experience and that of every single ego .•.. (Ideas, §48, p. 114.)
hnportantly, Husserl is not saying here that the notion of possible worlds
other than the actual is nonsense. Indeed, he is talking only about what is
real or actual: what is nonsense is the supposition that there might be real
things (e.g., Kantian "things-in-themselves") that cannot, in principle, be
encountered in that world to which experience gives us access. His claim,
then, which ties to the previous section of Ideas, is that everything that is
EXPLICATION OF MEANING 299
Here again Husserl is not rejecting the notion of possible worlds. Quite to
the contrary, he says Leibniz is right that many "groups of monads", i.e.,
worlds, are "conceivable", that is, presumably, possible. Husserl's point here
is Leibniz's, that no two worlds are compossible, simultaneously or conjointly
realizable, and that only one is actual. We could ask for no better evidence
that Husserl accepts the notion of possible worlds than his own avowed
agreement with Leibniz himself. (There is much more afoot around §60 of
Cartesian Meditations, dealing with intersubjectivity, but it remains beyond
the scope of our points.)
Possible objects and possible worlds are, then, familiar and to some degree
amenable to Husserl. Yet it is not clear that he assumes a bona fide meta-
physics of possibilia, as Leibniz did and as some modal logicians have in
recent years. Husserl appeals to possible objects and possible worlds for
phenomenological, rather than ontological, purposes, citing them as correlates
of certain structures of noemata or possible experiences. This use of the no-
tion of possibilia suggests that his talk of possible "Objects and possible worlds
300 CHAPTER VI
to red in autumn. The explicit Sinn of this act presents a particular tree as a
maple tree with red leaves (on the side I see, at least). Central to the act's
horizon are possible perceptions, for instance, those in which I see that same
tree from the back side as. also having red leaves on that side and, perhaps, as
having an opossum hanging from one branch. The possible perceptions in
the horizon are organized into temporal sequences that we called possible
"verification chains"; each of these chains is a continuous sequence of pos-
sible perceptions extending from the original act indefmitely into its past
and future (cf. Chapter Y, Section 4.3). Each chain, then, is a sequence of
perceptions I could have of the same tree were I seeing it continuously, from
appropriately varying perspectives, before and after the given perception -
for instance, the perceptions I might have if I approached the tree from a
given angle and continued walking past it so as to continue seeing it from the
resulting angles along my ensuing path. These chains of perceptions in the
horizon can be grouped into what we called "families" of chains, where each
such family collectively yields a consistent characterization of the tree as it
might be seen from various sequences of perspectives. When ideally com-
pleted, and buttressed by possible judgments or beliefs in the horizon that
bring in more theoretical properties of the object, each maximal family of
chains collectively traces out a possible total determination of the given tree,
as seen from all perspectives, and as otherwise flxed in all its properties and
relations to other objects. Of course, at most one of these families of acts can
present the tree as it is actually determined, in the actual world; but each
complete family traces the tree through all its properties, as it would be deter-
mined, in some possible world. Correlatively, the noematic Sinne of the
possible acts in each family conjoin, by "synthesis of identification", to form
a synthetic "world-Sinn" that prescribes a possible world in which the given
tree is completely determined in ways compatible with its incomplete deter-
mination in the original act (cf. Chapter Y, Section 4.4). In this way the
complete act-horizon, by virtue of its structuring into families of chains of
possible perceptions buttressed by other possible acts, projects an array of
possible worlds featuring the given tree as completely determined in all the
various ways compatible with what is presented in the original act.
The resulting array of possible worlds corresponding to the ideally com-
plete horizon of an act coincides precisely with the correlative object-horizon,
structured so as to reflect the structure of the act-horizon. In each of these
possible worlds there is realized a maximal consistent set of possibilities com-
patible with what is given in the original act; all these possibilities are included
in the horizon of the object as intended in the act - indeed, they make up
302 CHAPTER VI
that part of the object-horizon that correlates with the complete family of
acts that presents the possible world in question. And together the possible
worlds aligned with the complete act-horizon cover all the possible patterns
of further determination of the object and its environment; hence, they com-
prise all the possibilities included in the complete object-horizon correlated
with the horizon of the original act. Consequently, as the act-horizon breaks
down into various complete families of possible perceptions, buttressed by
other possible acts, the corresponding object-horizon - the possibilities
concerning the object's further characteristics, given its characteristics as
presented in the original act - itself breaks down into possible worlds featur-
ing the object as further characterized in various ways.
Here, then, is the basis in Husserl for our thesis that the explication of
Sinn and intention in terms of horizon is effectively equivalent to the explica-
tion of Sinn and intention in terms of possible worlds. The possible worlds
that are marked out by an act's horizon are just those possible worlds com-
patible with what is presented in the original act - in our example, a certain
object given as a maple tree having red leaves on a given side (and also having
whatever further properties the relevant background beliefs may ascribe).
And these are the very possible worlds that would serve to explicate the Sinn
of the act (together with the relevant background Sinne) on a possible-worlds
theory of meaning. More precisely i we now see that the object-horizon asso-
ciated with the act is effectively equivalent to a set of possible worlds such
that in each world the same tree, with the same properties attributed it in
the original act, is singled out as the object to which the Sinn relates in that
world. To explicate an act's Sinn, and the intention achieved through it, in
the way Husserl suggests - by laying out the horizon that is pre delineated
by that Sinn (together with the background Sinne) - is thus effectively
equivalent to correlating with that Sinn, and so representing it by, the func-
tion that assigns to each of these possible worlds the same object as that given
in the original act provided the object is in that world a maple tree and has
leaves turned to red on a given side (and has the further properties prescribed
by the background Sinne). And this explication accords precisely with the
possible-worlds explication of the Sinn of a direct-object perception that we
gave, in more general terms, in Section 2.8 above.
Indeed, the notion of object-horizon in Husserl's theory of intentionality
now emerges generally as the counterpart of the notion of "comprehension"
we defmed in the possible-worlds theory of meaning: the object-horizon that
explicates the Sinn of an act is effectively equivalent to the comprehension
of an expression that would express that Sinn. (Recall that the comprehension
EXPLICATION OF MEANING 303
with an act's Sinn and hence aligned with the horizon. Further, each one is
itself too much for a fmite human mind to grasp, either in the act itself or
in phenomenological reflection on the act. Individuals are "transcendent" of
human consciousness and worlds are vastly more so: the complete determina-
tion of an individual (in a world) or of a world is beyond our knowledge and
beyond our grasp. So we can grasp an individual or a world only incompletely:
it remains for us, Husserl says, an "Idea in the Kantian sense", a "limit" for
our expanding experience (cf. Ideas, §§143, 149). Husserl is explicit on
this point, which is true of actual as well as possible individuals and worlds.
Finally, we would also warn, in some cases it may even be indeterminate
whether a given possible situation, hence world, is itself compatible with the
Sinn of an act.
Consequently, the possible worlds and possible objects correlated with an
act or its Sinn constitute only a structure of ideal limits or boundaries that
demarcate the intentional content of the act, its Sinn. We should remember
that the implicit Sinne corresponding to the horizon of an act are themselves
only "implicit" in the act and that Husserl makes comparable caveats about
horizon-analysis (cf. Chapter Y, Section 3.1). He says that the Sinn (or Sinne)
meant implicitly in an act, that which corresponds to the act's horizon, "is
never present to actual consciousness as a fmished datum" (eM, §19, p. 45).
And he says (on the same page) that the pre delineation of an act's horizon is
partly indeterminate, which perhaps suggests that it may be indeterminate
whether a given possible act belongs in the horizon. Thus, we may think of
the practicing phenomenologist as starting with the explicit Sinn of an act
and working out from there, first characterizing a possible object with a
limited set of properties and in a limited possible situation, and then gradually
expanding the characterization further and further.
NOTES
1 Husser! sometimes uses the term 'explication' for a more special, but not unrelated,
notion. Cf. EJ. § § 22f, where by the "explication" of an object of perception he seems
to mean the process of determining which further properties an object actually has; this
process itself involves a consideration of the horizon of the act of perception.
2 We would agree with the claim, made by Quine among others, that the distinction
between observation and theory is not sharp but graded; cf. Quine's 'Grades of Theore-
ticity', in llxperience and Theory, ed. by Lawrence Foster and 1. W. Swanson (The
University of Massachusetts Press, Amherst, Mass., 1970), pp. 1-17. For Husser!, per-
ception - and perceptual judgment, or "observation", in particular - is always laden
with some "theory" or "interpretation" (Auffassung) in virtue of the "noetic" phase
that overlays its sensory or "hyletic" phase (Ideas, § § 85, 88); this would seem to insure
a gradation similar to that of which Quine speaks.
3 Carnap (Note 17, Ch. V above). Cf. Quine, 'Epistemology Naturalized', in his Ontolog-
ical Relativity and Other Essays (Columbia University Press, New York, 1969), pp. 69-
90, esp. pp. 74-78.
4 Castaneda has argued that knowledge of other minds is similar to theoretical knowl-
edge: see his 'Consciousness and Behavior: Their Basic Connections', in Castaneda (Note
21, Ch. I above), pp. 121-58.
306 CHAPTER VI
5 For recent discussions, see Robert Stalnaker, 'Pragmatics', Synthese 22 (1970), 272-
89; and Richard Montague, 'Pragmatics and Intensional Logic', Synthese 22 (1970),68-
94; both reprinted in Davidson and Harman (Note 25, Ch. N above).
6 Cf. Husserl's interesting and remarkably lucid discussion of "essentially occasional
expressions": LI, I, §26, VI, §5; see Chapter N, Section 3.4, above.
7 Our notion of "pragmatic presuppositions" is analogous to that characterized by
Stalnaker (Note 5 above), pp. 279ff.
8 See Quine, 'Two Dogmas of Empiricism', in From a Logical Point of View (Note 29,
Ch. I above), pp. 20-37. Cf. Morton G. White, 'The Analytic and the Synthetic: An
Untenable Dualism', in John Dewey: Philosopher of Science and Freedom, ed. by Sidney
Hook (Dial Press, New York, 1950), reprinted in Linsky (Note 53, Ch. II above), pp.
272-86.
9 Relevant here is Hilary Putnam's 'Meaning and Reference' (Note 26, Ch. N above).
10 Lewis (Note 53, Ch. II above), p. 242.
11 'On Sense and Reference' (Note 44, Ch. II above), pp. 57-58 (p. 41 of the recent
German edition cited in note 44, Ch. II); cf. Begriffsschrift (Note 50, Ch. II above),
pp.l1-12.
12 'The Modes of Meaning' (Note 53, Ch. II above), p. 247.
13 'Meaning and Synonymy in Natural Languages', Philosophical Studies 6 (1955),
reprinted in Meaning and Necessity (Note 16, Ch. I above), p. 234.
14 'Meaning and Synonymy in Natural Languages' (Note 13 above), p. 246.
15 Lewis (Note 53, Ch. II above), pp. 238ff.
16 The proposal we develop was lust made by Carnap in conversation, according to
Montague, with the difference that possible worlds were taken to be models; see
Montague (Note 5 above), p. 91. The conversations were presumably in the late 1950's.
One can also see the beginnings of the idea in Meaning and Necessity (Note 16, Ch. I
above), perhaps with some influence from Lewis' 'The Modes of Meaning' (Note 53, Ch.
II above): see § 16 and §40. (The first edition of Meaning and Necessity was published
in 1947.) Also see Carnap's 'Replies and Systematic Expositions', in The Philosophy of
Rudolf Camap, ed. by Paul Arthur Schilpp (Open Court, La Salle, Illinois, 1963), pp.
889-900.
17 'The Modes of Meaning' (Note 53, Ch. II above), p. 246.
18 Meaning and Necessity (Note 16, Ch. I above), pp. 56-64.
19 See David Lewis, 'General Semantics', Synthese 22 (1970), 18-67, reprinted in
Davidson and Harman (Note 25, Ch. IV above).
20 'The Modes of Meaning' (Note 53, Ch. II above), p. 240.
21 Meaning and Necessity (Note 16, Ch. I above), pp. 19-22,26-32,40-41.
22 Meaning and Necessity (Note 16, Ch. I above), p. 27.
23 See Meaning and Necessiry (Note 16, Ch. I above), pp. 126-27.
24 Cf. Frege, 'Function and Concept', in Geach and Black (Note 44, Ch. II above),
pp. 30-32; 'On Concept and Object', in Geach and Black (Note 44, Ch. II above), pp.
47-48. Also see Furth's introduction to Frege's The Basic Laws ofArithmetic (Note 44,
Ch. II above), pp. xxxviii ff; and Mohanty, 'Husserl and Frege' (Note 10, Ch. III above),
pp.57-62.
25 'On Sense and Reference' (Note 44, Ch. II above), p. 62, n. *.
26 Cf. 'The Thought' (Note 56, Ch. II above), esp. pp. 307-11.
27 'The Modes of Meaning' (Note 53, Ch. II above), pp. 238-39.
EXPLICATION OF MEANING 307
28 Cf. Meaning and Necessity (Note 16, Ch. I above), pp. 8-11, 23, 27.
29 Cf. Meaning and Necessity (Note 16, Ch. I above), pp. 173-77.
30 Cf. Meaning and Necessity (Note 16, Ch. I above), pp. 173-77, 179.
31 Hintikka has argued for the same point. See his 'Carnap's Semantics in Retrospect',
Synthese 2S (1973), 372-97 (esp. pp. 379-83); reprinted as 'Carnap's Heritage in
Logical Semantics', in The Intentions of Intentionality (Note 6, Ch. I above), and in
RudolfCamap, LOgical Empiricist, ed. by Hintikka (D. Reidel, Dordrecht, 1975). Also
cf. Hintikka's 'The Semantics of Modal Notions and the Indeterminacy of Ontology',
Synthese 21 (1970), 408-24 (esp. pp. 415-20), reprinted in Davidson and Harman
(Note 25, Ch. IV above).
32 Meaning and Necessity (Note 16, Ch. I above), p. 41.
33 Cf. Kripke, 'Naming and Necessity' (Note 25, Ch. IV above), pp. 269-70.
34 See Kripke, 'Naming and Necessity' (Note 25, Ch. IV above); and Kaplan, 'On the
Logic of Demonstratives' (Note 31, Ch. IV above).
35 See esp. Hintikka's 'Semantics for Propositional Attitudes', in Models for Modalities
(Note 6, Ch. I above), pp. 101-106. Our formulation, that an individuating meaning
captures the "identity" of an individual, is somewhat more specific than Hintikka's OWll
words about individuating functions. (Our formulation is intended to serve our charac-
terization of "~dividuatively definite" intentions - cf. Chapter I, Section 2.6, above;
and Chapter VIII, Section 1.1, below.) Hintikka also describes a second kind of individ-
uating function he calls "perceptually individuating" (see Note 15, Ch. I above); we shall
consider a related sort of function, tied especially to perception, in Chapter VIII, Section
2.2, below.
36 'Semantics for Propositional Attitudes', in Models for Modalities (Note 6, Ch.
above), p. 101.
37 Cf. Hintikka, 'Carnap's Semantics in Retrospect' (Note 31 above), pp. 383ff.
38 Alternatively, we might try to avoid both possible acts and noemata by describing
the horizon in terms of counterfactual conditionals. Thus, we might say:
(3x) (I see x as a tree and as having red leaves on this side & if 1 were to
look at the other side of x then I mighl see x as having red leaves on that
side & if ... ).
The difficulty with this alternative, however, is that a proper definition of horizon in
terms of acts seems to require quantification over possible acts since we cannot specify,
by description in counterfactual terms, all the acts in an act's horizon.
CHAPTER VII
all in the actual world, it is directed toward Santa Oaus in each possible
world correlated with the act's horizon. According to the possible-worlds
Husserlian theory, then, an act of expecting Santa Claus is directed to a
merely possible object, an object that exists in various possible worlds but
not in the actual world.
Relativization of intention to a possible world also yields a perspicuous
account of "indefinite" intentions, such as the shopkeeper's act of expecting
her 100000th customer. Unlike the Sinn of an act of expecting Santa Claus,
the Sinn of this act does not prescribe any specific individual as the object
of intention. Hence, the horizon of the act will include possible acts directed
toward different individuals, and the possible worlds compatible with what
the Sinn prescribes will include worlds in which different individuals satisfy
the condition of being the shopkeeper's 100000th customer. The act will be
directed accordingly toward different objects in different possible worlds -
toward, say, Mr. Olson in one possible world (where it is Mr. Olson who walks
through the door) and toward Ms. Green in another world (where it is she
who enters the shop); and, for this reason, the act will not be "definitely"
directed toward some one specific object. (Cf. Part 3 below on the de dicto/
de re distinction and its analysis in a possible-worlds approach to intention-
ality.)
While the notion of horizon provides the Husserlian basis for a possible-
worlds theory of intentionality, a further theoretical foundation for it is
the possible-worlds explication of meaning we studied in Chapter VI (see
especially Part 2). Husserl himself saw horizon-analysis as a way of explicat-
ing an act's meaning, or Sinn (see eM, §20; cf. Chapter V, Section 1.3): in
particular, analysis of the object-horizon associated with an act can be seen as
a kind of explication of the act's meaning, achieved by laying out what the
meaning prescribes in different possible circumstances. The possible-worlds
theory of meaning applied to noematic Sinn supports this view, though it
goes beyond Hussed's own work. For, on that theory, an act's Sinn can be
represented or explicated by the meaning function that assigns to each pos-
sible world what would be prescribed by the Sinn in that world. Thus, from
each world compatible with the act's Sinn, this function picks out the object
prescribed by the Sinn - and so the object toward which the act is directed -
in that world. The explication of noematic Sinn in terms of meaning func-
tions is in this way virtually equivalent to the explication of Sinn in terms of
horizon, at least insofar as these two ways of explicating an act's Sinn merge
in the possible-worlds version of Hussed's theory of intentionality. (Cf.
Chapter VI, Section 3.2.)
INTENTIONALITY AND POSSIBLE-WORLDS SEMANTICS 313
We would note, however, that some modifications are called for in the
possible-worlds theory of meaning if it is to serve as a foundation for a
Husserlian possible-worlds theory of intentionality. What we have said so far
applies directly to individual-intentions, intentions of or about individuals.
But propositional intentions will require modification: the value at a world
of a propositional meaning function is a truth-value, but the object of a pro-
positiohal act, at least according to Husserl, is a state of affairs and not a
truth-value. Also, horizon-analysis brings to light the influence of background
beliefs or presuppositions on an act's intentionality, and so the possible-worlds
explication of meaning, as applied to intention, will have to be developed in
a way that reflects this influence (cf. Chapter VI, Sections 1.4-1.6, 2.5, 2.9).
Horizon-analysis serves to explicate not simply an act's Sinn but the act's
Sinn together with the Sinne of related background beliefs or other experi-
ences presupposed in the act. Thus, on the Husserlian possible-worlds theory
of intentionality, an act is directed toward appropriate objects in different
possible worlds via its Sinn together with whatever Sinne it presupposes.
Insofar as meaning functions are to be invoked in a theory of intentionality,
then, these functions must appropriately represent the meaning complex
consisting of the act's own Sinn and its presupposed background Sinne, and
not just the act's Sinn alone. Later we shall find that a pOSSible-worlds se-
mantics for intentional sentences takes the influence of background beliefs
into account in an interesting way.
worlds, by contrast, are not "incomplete"; rather, they are - within their
respective worlds - quite ordinary and complete entities of the very same
sort that are found in the actual world. Furthermore, the possible-worlds
approach to intentionality differs from the object-approach significantly in
that it does not treat intention as a simple relation to a single (albeit unusual)
object; rather, intention is seen as a complex relation to a vast array of ordi-
nary objects in various possible worlds.
Thus, on the object-approach, intention consists in an ordinary relation
to unusual objects; and on the Husserlian approach, intention consists in a
meaning-mediated - and therewise unusual - relation to ordinary objects. But
on the possible-worlds approach, intention consists in a relation of "multiple
directedness", a pattern of directedness reaching into various possible worlds. 2
And so it seems that the possible-worlds approach might best be classified as
a distinctive type of approach, sui generis.
However, there are good grounds for seeing the possible-worlds approach
to intentionality as basically Husserlian. True, the pure possible-worlds theory
of intentionality does not explicitly assume meaning entities. Yet, we know,
meaning entities - Carnapian meaning functions, functions from possible
worlds to appropriate extensions - are but a defmition away. For the patterns
of directedness that a possible-worlds analysis must fmd in intention are just
those regimented as meaning functions that represent the noematic Sinne a
Husserlian analysis fmds in intention. In a sense, then, there is no "pure"
possible-worlds theory of intentionality, if 'pure' means ultimately free from
a life of Sinn. And insofar as we fmd meaning functions implicit in the strict
possible-worlds theory of intentionality, we should sort it ultimately with the
Husserlian approach to intentionality.
try to keep this part as informal as possible, and details of formulation may be
skimmed over by any who prefer to do so.) Our discussion here will lay the
foundation for a fuller account of the possible-worlds theory of intentionality,
which we shall offer in Part. 3.
Meanings (Sinne) playa crucial role in the Husserlian theory of intention-
ality, and it was the Carnapian possible-worlds analysis of meaning that led
us to consider a possible-worlds theory of intentionality. Accordingly, we
shall be interested in the roles meanings might play in possible-worlds seman-
tics, especially for sentences of propositional attitude.
(i) For any atomic sentence 'a is P', 'a is P' is true if and only if the
extension (the referent) of 'a' is a member of the extension of 'is
P' (the set of individuals that "are P");
(ii) For any sentences 'p' and 'q', the molecular sentence 'p and q' is
true if and only if'p' is true and 'q' is true;
(iii) For any sentence 'p', the molecular sentence 'It is not the case
that p' is true if and only if 'p' is not true.
all without assuming meaning entities (and, let us note, without assuming
possible entities}
However, even if the classical form of Tarskian truth-theoretic semantics,
which in no way uses meanings, is adequate for elementary extensional lan-
guages such as Tarski addressed, it is not adequate for modal, or intensional,
languages. Sentences ascribing necessities or intentional acts bring failures
of extensionality, failures of substitutivity of identity and of existential
generalization (cf. Chapter I, Part 3). And so these sentences resist a Tarskian
characterization of truth, for their' truth-values seem not to depend only on
the extensions of their parts. To account for such failures, a Fregean semantics
for (say) belief-sentences assumes meaning entities and holds that meanings
playa role in the truth-conditions of belief-sentences; indeed, while Frege
held that the truth-value of a belief-sentence is a function of the referents (or
extensions) of its parts, he, also maintained that expressions in belief-contexts
refer to their customary meanings (cf. Chapter II, Section 3.4). But insofar
as a semantics for intentional sentences brings meanings into the truth-con-
ditions for these sentences, as Frege's semantics does, it goes beyond the
resources of a Tarskian semantics.
An alternative to the Fregean approach to modal sentences is that of so-
called possible-worlds semantics. On this approach, expressions in intensional
contexts refer to, or draw their extensions from among, entities in different
possible worlds. The leading idea is the Leibnizian view of necessary truth
as truth in all possible worlds: hence, where 'p' is any extensional sentence,
'Necessarily p' is true if and only if 'p' is true in every possible
world.
Thus, the truth-value of 'Necessarily p' depends on the truth-value of 'p' in
different possible worlds, and so it depends on the extensions of the seman-
tically significant parts of 'p' in different possible worlds. (As we shall see in
the next section, a similar approach has been proposed for the semantics of
intentional sentences, such as 'a believes that p'.) This Leibnizian proposal
of truth-conditions for 'Necessarily p' presupposes an analysis of the truth-
conditions for 'p' in any possible world, and so we first need to see how a
possible-worlds semantics works for extensional sentences.
A possible-worlds approach to semantics extends the basic Tarskian ap-
proach to semantics by assuming a multiplicity of possible worlds and derming
extensions and truth-conditions relative to a possible world. Thus, given a
fixed (extensional) language, expressions of basic categories are each assigned
an extension in, or relative to, each possible world (cf. Chapter VI, Sections
320 CHAPTER VII
2.1-2.2): each singular term is assigned for each possible world one individual
(at most) existing in that world; each predicate is assigned for each world a
(possibly empty) set of individuals (or ordered n-tuple,s) in that world; and
each sentence is assigned a truth-value for each world (on some alternatives,
sentences are allowed to have no truth-value in some worlds). On the possible-
worlds approach, then, an interpretation for a language is given by means of
a two-argument function that relates each expression to its extension in each
possible world. Conditions of truth for any sentence in the language are
dermed recursively after the fashion of Tarski, except truth is also dermed
relative to a possible world. (Truth simpliciter is of course just truth in the
actual world.) For instance:
(i) For any atomic sentence 'a is P', 'a is P' is true in a possible world
w if and only if the extension of 'a' in w is a member of the set
which is the extension of 'is P' in w;
(ii) For any sentences 'p' and 'q', the sentence 'p and q' is true in a
possible world w if and only if both 'p' and 'q' are true in w;
and so on.
, As described so far, the possible-worlds approach to semantics does not
assume meaning entities and so represents an alternative to the basic Fregean
approach to semantics for extensional languages. However, as we saw in
Chapter VI, Part 2, meanings may be naturally dermed, or at least repre-
sented, in possible-worlds theory as functions that assign appropriate exten-
sions to possible worlds - meaning functions, as we have called them. And a
possible-worlds semantics for an extensional language is committed to pre-
cisely these functions insofar as it assigns to each semantically significant
expression in the language its extension in every possible world. So a possible-
worlds semantics does not avoid meaning entities. Indeed, meaning functions
can easily be written into possible-worlds semantics with a role in extensional
languages like that of Fregean senses.
To construct a possible-worlds semantics for an extensional language L, we
derme an interpretation function E that assigns to each meaningful expression
0: in L and each possible world w the extension of 0: in w, E(Oi., w). But we
could equivalently derme an interpretation function I that assigns to each
meaningful expression 0: in L its intension, i.e., its meaning or sense. The
intension of 0: - 1(0:), or lOt - we would then take to be itself a function, viz.,
the meaning function that assigns to each world w the extension of 0: in w;
that is, IOt(w) = E(o:, w). (Note thatIOt(w) is the value at w of lOt, which is the
meaning function assigned to 0: by the interpretation function I: evaluate I
INTENTIONALITY AND POSSIBLE-WORLDS SEMANTICS 321
at O! and you get the intension of O!; which is itself a function Ia; evaluate this
function Ia at wand you get an extension, the extension of O! at w, E(O!, w).)
Thus, the extension of O! is determined by way of the intension given O! by I.
Given these two interpretation functions, E and I, we can then formulate
the truth-conditions in a world for a sentence in either of two ways. For in-
stance, for the English sentence 'The morning star is a planet' we have either:
What I take to be the distinctive feature of all use of propositional attitudes is the
fact that in using them we are considering more than one possibility concerning the
world ....
My basic assumption ... is that an attribution of any propositional attitude to the
person in question involves a division of all the possible worlds ... into two classes: into
those possible worlds which are in accordance with the attitude in question and into
those which are incompatible with it. 7
If 'p' is 'the morning star is risen', for instance, it is true in a given world w'
just in case the referent, or extension, of 'the morning star' in w' is a member
of the extension of 'is risen' in w'. And sO'" the truth-conditions for (say)
'Smith believes that the morning star is risen' would be formulated as follows:
324 CHAPTER VII
'a believes that p' is true in a world w if and only if 'p' is true in
every world that is a member of CPB(the referent of 'a' in w, w).
(For intuitive impact, we have here characterized the set of worlds, cf>B
(Holmes, w), that are doxastic alternatives to W for Holmes as the worlds
compatible with what Holmes believes.) Now, 'the murderer wears square-
toed boots' is true in a given world w', we know, if and only if the referent
(or extension) of 'the murderer' in w' is a member of the extension of 'wears
square-toed boots' in w' - that is, if in w' the person who is the murderer
wears square-toed boots. The truth of (1) in the actual world depends, then,
not on the referent of 'the murderer' simpliciter, its de facto referent in the
actual world, but on its referents in each of various possible worlds, the
worlds compatible with what Holmes believes. Thus, according to Hintikka,
the reference of terms in belief-contexts is a matter of "multiple" reference,
reference in an appropriate set of possible wodds. ll
This account of the reference of terms in belief-contexts yields a ready
explanation of failures of substitutivity of identify in belief-contexts. From
(1) and
we cannot infer
that is, 'Someone is such that Holmes believes that he wears square-toed
boots'. This inference may fail for either of two reasons, as we considered
in Chapter I (Sections 3.3-3.4): either because there may be no murderer
INTENTIONALITY AND POSSIBLE-WORLDS SEMANTICS 327
(perhaps the "murder" was a hoax) and henc~ no (actual) individual Holmes'
belief is about; or because the belief may not be about any particular individ-
ual, say, Mr. Jefferson Hope, since Holmes merely believes that the murderer,
whoever that may be, wears square-toed boots. The inferential failure in
either case is explained quite naturally within possible-worlds semantics for
belief-sentences. The explanations stem from the possible-worlds interpreta-
tion of de re belief-sentences such as (4).
An elementary de re belief-sentence such as (4) may be interpreted in a
generally Hintikkian way according to the following condition:
'(3x) (a believes that x is cp)' is true in a world W if and only if
there is in W an individual who satisfies the predicate 'is cp' in
every world compatible with what the referent of 'a' in W believes
in w.
(Hintikka's own formulation here is a bit different, in ways we shall be not-
ing.) Thus, (4) is true in the actual world wa just in case there exists in Wa
an individual such that, in every world compatible with what Holmes believe3
in W a , that very same individual wears square-toed boots. The most important
point here is that it is the same individual who wears square-toed boots in
each of Holmes' "belief worlds": that is what shows the belief described by
(4) to be about a particular individual.
Armed with the truth-conditions we have seen for (1) and (4), we can now
explain both cases of the failure of inference from (1) to (4). The inference
may fail in the first case because (1) may be true in wa - in that 'the murderer
wears square-toed boots' is true in all worlds compatible with what Holmes
believes in Wa - even though there is no individual to whom 'the murderer'
refers in wa. In that case there would be no individual in Wa whom Holmes'
belief might be said to be about, and so (4) would not be true in Wa. Thus,
(4) does not follow from (1). The inference fails for a more interesting reason,
however, in the second case. (1) may be true in wa even though 'the murderer'
does not refer to the same individual in every world compatible with what
Holmes believes in wa: it is compatible with Holmes' belief that any of a
number of suspects is the murderer. But then, we may suppose, there would
be no one individual who wears square-toed boots in each of Holmes' belief
worlds, and (4) would then be false in wa. And so in this case, too, (4) does
not follow from (1).
The proper interpretation of "quantifying-in" constructions, as in (4), has
been one of the thorniest problems for the semantics of belief-sentences.
We saw in Chapter II (cf. Section 3.4) that Frege's idea, that expressions in
328 CHAPTER VII
(6) 'a believes that p' is true in a world w if and only if 'p' is true in
every world w' E </>B(a, w).
(Recall that </>B(a, w) is the set of worlds compatible with what a - the
referent of 'a' in w - believes in w.) Thus, the truth of (5) in w depends not
on the truth of 'p' in w but on the truth of 'p' in the worlds compatible with
what a believes in w. Now, the propositional meaning function I. p ' assigns 'p'
its truth-values in various worlds. So we can recast the conditions of truth for
(5) as follows:
(7) 'a believes that p' is true in a world w if and only if, for every
world w' E </>B(a, w), I.p'(w') = Truth.
330 CHAPTER VII
Inasmuch as the conditions (6) and (7) are virtually equivalent,17 I,p' plays
a tacit role in condition (6). For according to (7), the truth of (5) (in a world
w) depends on/,p', the meaning function associated with 'p': on the fact that
I'p' distributes Truth among worlds in accordance with what a believes (in w).
Thus, (7) effectively interprets (5) as asserting that a certain relation obtains,
in a given world, between a and the propositional meaning function I,p'. And
so, insofar as meanings can be identified with meaning functions, (7) could
be seen as a possible-worlds rendition of Frege's interpretation of (5).18 For
Frege would take the contained sentence 'p' in (5) to refer to its customary
sense, the "thought" that p, and would thus take (5) to assert that a certain
relation obtains between a and the sense of 'p'.
Suppose, now, we consider a concrete case, the belief-sentence
(8) (1) is true in a world w if and only if, for every world w' E IPB
(Holmes, w), I'the murderer wears squared-toed boots' (w') = Truth.
But now, for any world w', I'the murderer wears square-toed boots' (w') = Truth if
and only if I'the murderer'(W') E I'wears square-toed boots'(W'). Thus,
(9) (1) is true in a world W if and only if, for every world w' E IPB
(Holmes, w), I,the murderer'(w') E I,wears square-toed boots'(W').
Here the truth of (1) (in a world) is explicitly shown to depend on the mean-
ing function associated with the term 'the murderer'. Assuming the correla-
tion of meanings with meaning functions, then, (9) compares with the Fregean
view that when a term occurs within a belief-context it refers to its customary
sense.
Consider now a simple de re belief-sentence, say,
Holmes stands in the relevant relation to the thought whose "subject" com-
ponent is s and whose "predicate" component is the sense of 'wears square-
toed boots'_ But we must be more restrictive, since not every individual sense
can establish a de re belief. In particular, the sense of a definite description
cannot (cf. Chapter IV, Section 3.3), and so the truth of (1), for instance,
cannot insure the truth of (4). Apparently, we must restrict the quantifier
in (4) to "rigid" individual senses (cf. Chapter VI, Section 2.7), in order to
secure a proper Fregean interpretation of (4). (For further development of a
Fregean interpretation of quantifying-in, see Part 4 below.)19
Consistently with these remarks, Hintikka assumes for the interpretation
of de re sentences, such as (4), a special class of "individuating functions",
which he compares with "individuating" individual concepts. 20 These individ-
uating functions are functions that assign the same individual to every world
(in which it exists). Thus, they are the meaning functions that represent what
we have called rigid meanings (cf. Chapter VI, Section 2.7); we might call
them rigid meaning functions, in line with our prior terminology. Then
Hintikka's proposed truth-conditions yield for (4):
Calling in our full range of meaning functions, we can recast this proposal as
follows:
murderer himself, Mr. Jefferson Hope.22 (Cf. Chapter II, Section 3.5.) Now,
it is worth noting that withing an intension-laden possible-worlds semantics,
de re belief-sentences may yet be interpreted so that variables of quantifying-
in range over individuals rather than meanings. This is achieved for (4) as
follows:
'a believes that p' is true in a world w if and only if 'p' is true in
all the possible worlds compatible with everything a believes in w.
but his whole network of beliefs. Thus, the analysis of believing that p would
call on the subject's overall "background" belief-structure.
In the fashion of our discussion of horizon, we would cite as relevant to a's
believing that p only those of his other beliefs that are presupposed in his
believing that p. Hintikka's approach, though, would cite all of a's beliefs,
even those that are not presupposed by or directly relevant to his believing
that p. One might support this expansiveness on grounds of the holistic
character of belief, holding that a person's total system of beliefs form a web
any comer of which is sensitive to any other area, so that in some way any
belief presupposes the whole system. But let us not pursue the issue as to how
much of one's belief-system influences a given belief. We can see in any event
that a Hintikkian possible-worlds semantics for belief-sentences is sensitive to
contextual influences on a belief by background beliefs, and this sensitivity
is like that we found in horizon-analysis as explication of intention.
We are focusing exclusively on belief, but we should note some complica-
tions regarding other species of acts or attitudes. Background beliefs may be
presupposed in an act of perception, but they may be irrelevant to an act of
phantasy quite unconstrained by what one thinks of reality. The horizon
of such a phantasy would not be constrained by background beliefs, then.
Generally, we would defme the object-horizon as the set of possibilities -
ultimately, possible-worlds - compatible with the way the object is intended
in the act, assuming intention may be constrained by what is intended in
other acts or attitudes presupposed in the given act. A possible-worlds seman-
tics for sentences ascribing a given propositional act or attitude would then
proceed in terms of a horizon of possible worlds defmed in an appropriately
general way.
'Holmes believes that p' is true in a world w if and only if, for
every world w' compatible with what Holmes believes in w,
I.p'(w') = Truth.
Holmes believes that p (in a world w) if and only if, for every
world w' compatible with what Holmes believes (in w), I.p'(w') =
Truth.
But now, although this formulation of the analysis of belief explicitly assumes
propositional meaning functions, it is not committed as to what their role
might be in the intentionality of belief. For systematic and extra-semantic
reasons that have emerged in our study of intentionality, we would take them
to represent the noematic Sinne of beliefs, the contents that mediate the
directedness of the beliefs. In particular, if Husserl's account of individual-
directed consciousness generalizes to cover propositional consciousness, so
that propositional intentions are also mediated by but not directed toward
meaning entities, then the propositional meaning function associated with
a belief must be taken to be or represent not the object but the noematic
content of the belief.
What, then, would be the object of a belief, that which is believed? Ac-
cording to Husserl's theory of intentionality, the object of any act is deter-
mined by the act's noema: the object intended in an act is the entity that the
act's noema, or meaning, prescribes. Accepting this fundamental tenet of
Husserl's theory, and having correlated a belief's associated propositional
meaning function with the noema of the belief, we should take the object
of belief (in a world) to be the value of this propositional meaning function
(in that world). Then the analysis of belief that emerges from possible-worlds
semantics for belief-sentences is an instance of the "Husserlian" possible-
worlds theory of intentionality we described in section 1.2: associated with
each belief is a noematic meaning entity - represented by, or perhaps iden-
tified with, a meaning function - by virtue of which the belief is directed
toward What the meaning prescribes in various possible worlds.
However, tllere is a certain infelicity in taking as objects of belief the
values that a belief's associated meaning function takes in the various possible
worlds under consideration. A natural view, held by Husserl among others, is
that the objects of belief are states of affairs, conceived as non-noematic enti-
ties transcendent of consciousness. But the values of a propositional meaning
function, on our present analysis, are not states of affairs but truth-values.
336 CHAPTER VII
Indeed, modern semantics has largely skirted worries about states of affairs,
at least partly by assuming with Frege that the referent or extension of a
sentence is its truth-value. 24
Now, the Husserlian view could be worked into a possible-worlds theory
of belief. In fact, the possible worlds that are assumed in the possible-worlds
analysis of belief and thence in the semantics of belief-sentences are them-
selves large (maximal consistent) states of affairs. So we could say, for each
world w such that I,p'(w) = Truth, not that the belief that p is directed
toward Truth in w, but that it is directed toward the state of affairs that is
w. The analysis still does not specifically calion states of affairs smaller than
whole worlds, however; and to accommodate smaller states of affairs would
involve notable technical changes in the customary style of possible-worlds
semantics. The natural move - in line with Husserl's view of the relation
between the meaning and the object of an act - would be to defme proposi-
tional meaning functions as functions from worlds to states of affairs rather
than truth-values, and then to define truth-conditions in a world in terms of
states of affairs' obtaining in that world. We shall pursue further the notion of
states of affairs, their role as objects of belief, and their role in semantics in
Section 3.5 and in Part 4 below. But for now let us focus on beliefs' being
"about" individuals.
A belief, we know, is directed in one sense toward what is believed, a pro-
positional entity such as a state of affairs; and it is directed in a second sense
toward the individual(s) (if any) that it is about. The individual a belief is
about is sometimes also called the object of the belief. For Husserl, aboutness
is like direct-object intention of an individual (such as seeing or imagining
an individual) in that it is mediated by a noematic Sinn (cf. LI, V, § 11,
pp. 559-60; cited in Chapter I, Section 1.5, above). However, "indefinite"
and "defmite" beliefs are about individuals in quite different ways. The dif-
ference comes out sharply in a possible-worlds analysis of belief; it is reflected
in the semantic difference between de dicta and de re belief-sentences, a
major concern of possible-worlds semantics for belief-sentences. We turn now
to an account of de dicta and de re aboutness.
other propositional acts or attitudes with the same Sinn as the belief. It is
what we call "indefmite", or "de dicto", aboutness.
Now, if we assume possible worlds, we can characterize de dicto aboutness
in a perspicuous way. The basic move is to relativize aboutness to a possible
world: a propositional act or attitude is about an individual only relative to,
or in, a given possible world (cf. Section 1.2 above). Then Holmes' belief,
occurring in a world w, is about an individual in a world w' just in case in w'
that individual is the murderer. And insofar as different individuals may be
the murderer in different possible worlds, Holmes' belief may be about
different individuals relative to different worlds.
This is the natural account of de dicto, or indefinite, aboutness to draw
from Hintikka's semantics for de dicto belief-sentences such as (1). For, on
Hintikka's interpretation,
(1) is true in a world w if and only if, for every world w' com-
patible with what Holmes believes in w, the referent (extension)
of 'the murderer' in w' is a member of the extension of 'wears
square-toed boots' in w'.
(I) is true in a world w if and only if, for every world w' com-
patible with what Holmes believes in W, !<the murderer'(w' ) E
I'wears square-toed boots'(W ' ),
where I'the murderer' is the meaning function associated with 'the murderer',
the function that assigns to any world that individual (if any) who in that
world committed the murder. Then it is natural to say the belief's being
about a given individual in a given world is mediated by the meaning repre-
sented by the meaning function I'the murderer'. And the belief is then about
INTENTIONALITY AND POSSIBLE-WORLDS SEMANTICS 339
What shows that the belief described is definite is the fact that in each of
Holmes' "belief"-worlds it is the same individual who is featured and turns
up in the extension of 'wears square-toed boots'. If aboutness is relativized
to a world, as in the preceding section, then Holmes' belief is about the same
individual in each of the worlds compatible with what Holmes believes. And
so we may say his belief is about that individual, simpliciter.
The truth-conditions for (4), as stated above, leave open an important
question about de re beliefs: can there be de re, or definite, beliefs about
nonexistent individuals? As we observed in Section 2.3 above, Hintikka's
interpretation of sentence like (4) in effect treats the quantifier as ranging
over possible individuals that mayor may not be actual. And so Hintikka's
analysis separates the question of definiteness of belief from the question of
the existence or nonexistence of the individual believed about. A belief is
about a definite individual, and in that sense is de re, just in case it is about
the same individual in each of the relevant belief-worlds; while a belief is
about an existent individual just in case the individual believed about happens
also to exist in the actual world. This is an important feature of Hintikka's
semantics insofar as the semantics reflects an analysis of aboutness. For it
permits us to preserve the Husserlian view that consciousness retains its
directedness - here, its aboutness - even if that toward which it is directed
does not exist. Thus, it leaves the analysis of defmiteness of aboutness general,
covering defmite belief about either a nonexistent or an' existent individual.
340 CHAPTER VII
likes of (4) with his observation that quantifying-in presupposes the individ-
uation of the objects the quantifier ranges over. Of course, quantification
always presupposes the individuation of the objects the quantifier ranges over.
But Hintikka's point, as we would put it, is that the quantifier in (say) (4)
presupposes the individuation in Holmes' mind or consciousness of the objects
it ranges over. Thus, the quantifier ranges over obje"cts such that Holmes has a
sense or conception or perhaps knowledge of their "identity";26 or, better,
instead of restricting the range of the quantifier, we might say that the truth
of (4) depends on the given individual's being individuated in Holmes' con-
sciousness. This point of Hintikka's is an important contribution to the
analysis of a certain kind of de re belief. Though we shall study such "individ-
uatively definite" beliefs in the next chapter, it is worth noting here that our
present picture, with aboutness mediated by an "individuating" meaning that
would seem to function as a constituent of the belief's Sinn, is overly simple.
Note also that Hintikka sees in perception a second kind of "individuation"
per se; we need not address that view here, but we consider it in part in
Chapter VIII, Section 2.2.
likely include such descriptive content, even if the explicit Sinn of the inten-
tion itself does not (cf. Chapter VIII, Section 4.1). Insofar as this is so, even
de re belief is relative to a conception of the object it is about and in that
sense is conception-dependent. Nonetheless, whatever predicative content
may be relevant to Holmes' de re belief, it cannot play the same role that it
would play in a de dicta belief: it cannot determine the aboutness of the
belief. Holmes' de re belief is directed toward (about) the same fixed individ-
ual in every world under consideration. But, as we saw in Chapter IV (cf.
Section 3.3), a conception or sense that appeals to properties of an individual,
in the manner of a defmite description, will prescribe different individuals in
different worlds compatible with the intention. Unlike a de dicta belief, then,
the determination of the object of a de re belief is relatively independent of
the presentation of the properties it is intended as having. We shall study de
re belief in detail in the next chapter. But suffice it to say here that if de re
aboutness is to be determined by a meaning, a rigid meaning, such a meaning
cannot be of the sort expressible by a definite description.
The "indeterminacy", or incompleteness, in intentions of transcendent
objects is, we know, related to their conception-dependence (cf. Chapter I,
Section 2.5). This character of intention is nicely explicated on the possible-
worlds approach. Any individual is completely "determined", or propertied, in
any world in which it occurs. Yet an individual sense such as "the murderer"
specifies only a limited aspect, an incomplete "determination", of the in-
dividual it prescribes in any world. In this sense there is an indeterminacy in
the aboutness of a de dicta, or indefinite, belief such as we described. And
a de re, or definite, belief is indeterminate in the same way, insofar as it is
dependent on a conception of the object it is about. However, there is also
a somewhat different sort of indeterminacy in de re aboutness, at least of the
individuative sort. For, as we shall consider in the next chapter, the very
"identity" of a concrete individual apparently transcends human grasp and
so is never fully given in human consciousness (cf. Chapter VIII, Section 3.2).
The "primary" objects of belief are states of affairs; and where a belief is
about an individual, the individual is a "secondary" object of the belief and
a constituent of the state of affairs that is the primary object of the belief
(cf. Chapter I, Section 1.5). Now, on the possible-worlds Husserlian theory
of intentionality, an act is directed via its Sinn toward various objects in
various possible worlds. A propositional act or attitude, such as a belief; is
then directed toward various states of affairs occurring in various worlds. We
have focused on the aboutness of beliefs, which is mediated by individual
senses and terminates in individuals in various worlds. Let us now observe the
natural account of the states of affairs toward which a belief is directed in
various worlds, in particular for the cases of certain de dicta and de re forms
of belief.
Consider Holmes' de re, or defmite, belief about a certain man - Jefferson
Hope - that he wears square-toed boots. With respect to each world this belief
is about Jefferson Hope: it is about the same man in every world, and thus
de reo Accordingly, Holmes' belief is directed toward the same state of affairs
in each world (wherein the state of affairs occurs), namely, the state of affairs
that consists in Jefferson Hope's wearing square-toed boots (habitually).
In contrast, consider Holmes' prior de dicta, or indefinite, belief that the
murderer wears square-toed boots. With respect to a given world this belief is
about an individual, but it is not about the same individual in every world.
This belief is also directed toward a state of affairs in each world, the state
of affairs consisting in a certain individual's (namely, the individual the belief
is about in that world) wearing square-toed boots. With respect to each world,
then, the belief is directed toward a state of affairs, but it is not directed
toward the same state of affairs in every world. In some worlds the state of
.:!ffairs is that Jefferson Hope wears square-toed boots, in others it is that
Arthur Charpentier wears square-toed boots, and so on, depending on who in
the given world committed the murder.
These forms of belief, one de re and the other de dicta, differ in their
primary intentionality in a basic way. The de re belief is directed toward the
same state of affairs in every possible world (wherein it occurs), whereas the
de dicta belief is not. The difference is a direct consequence of the difference
in aboutness: the de re belief is about the same individual in every world
(wherein it exists), whereas the de dicta belief is not.
Notice that we have taken both forms of belief to be directed, in any
world, toward singular states of affairs, i.e., states of affairs consisting of an
individual having a property. The contents or Sinne of de re and de dicta
beliefs differ in structure, but there is no need to maintain that their objects,
INTENTIONALITY AND POSSIBLE-WORLDS SEMANTICS 345
what unusual, but in fact Frege made.just such a distinction for predicates: he
identified the referent of a predicate with a function, or "concept", and the
extension with the class of entities falling under the concept. In our case,
the distinction is required because, on a Husserlian semantics, the extension
(or truth-value) and the referent of a sentence will not be the same: for
Husserl, we should take the object, or "objective reference", of a sentence to
be the object of a speaker's underlying judgffient~in uttering the sentence,
and that object is a state of affairs, not a truth-value. (Husserl affirmed this
seman tical position, but in the case of a predicate he took the "objective
reference" to be the several objects falling under the predicate, in effect the
extension. On these points see LI, I, §12.) With a distinction between referent
and extension we shall be able to formulate a semantics embracing the usual
treatment of extension, with sets of objects assigned to predicates and truth-
values assigned to sentences, and also to utilize Husserl's distinction between
noematic sense and object.
Hence, we assume three levels of semantic assignment: sense, referent
(object of reference or of intention) in a possible world, and extension in a
possible world. We further assume that sense determines referent in a world,
and referent in a world determines extension in a world: let us say a sense
designates a referent in a world, which captures an extension in a world.
Thus, the sense of a singular term is an individual sense, its referent in a world
is the individual (if any) designated in that world by that sense, and the ex-
tension in that world is that individual ("captured" thus by the referent).
While the sense of a predicate is a predicative concept or sense, we shall say
(a fa Frege rather than Hussed) that the referent in a world is the property
(the same in any world) designated by that sense, and the extension in a
world is the set of individuals occurring in and having that property in that
world (thus captured by that property). And, fmally, the sense of a sentence
is a propositional sense or thought, the referent in a world is the state of
affairs (if any) designated in that world by that sense, and the extension in
a world is the truth-value Truth if there obtains in that world a/the designated
state of affairs - otherwise the extension is Falsehood. We may codify these
seman tical structures in the following form.
We assume a function I that assigns to each semantically significant ex-
pression a in t! a sense - the sense or intension of a, I(a). The types of senses
are familiar: an individual sense for a singular term, a predicative sense or
concept for a predicate, and a propositional sense or thought for a sentence.
We further assume a function R that assigns to each appropriate expression a
in t! and each possible world w a referent - the "objective reference" of a in
INTENTIONALITY AND POSSIBLE-WORLDS SEMANTICS 347
w, R(o:, w). For a term this will be an individual existing in w; for a predicate,
a property (presumably, instantiated in w); and for a sentence, a state of
affairs occurring or obtaining in w. We allow that R be undefined, or take no
value, at some worlds for some expressions. Finally, we assume a function £
that assigns to each appropriate expression 0: and each possible world w an
extension of the usual sort - the extension of 0: in w, £(0:, w). We assume
the sense of 0: designates the referent of 0: in w, and the latter captures the
extension of 0: in' w. Accordingly, let Do be the function that assigns to a
sense and a world the referent it designates in that world: thus Do(/(o:} w) =
R (0:, w). And let K be the function that assigns to a referent in a world the
extension it captures in that world: thus K(R(o:, w)) = £(0:, w). So £(0:, w)
= K(Do(/(o:), w)). A sentence is true in a world, then, just in case it refers to
a state of affairs obtaining in that world, or is assigned a referent by R in that
world. Truth-conditions can be laid down then in pretty much the usual
manner follOwing Tarski, working with the extensions so determined. (The
main complication concerns the treatment of quantification, as there must
be developed a treatment of variables and quantifiers vis-a-vis sense and object
or referent.)
A systematic Husserlian semantics of the preceding sort requires a sys-
tematic doctrine of states of affairs. We leave to future research the tasks of
ferreting out Husserl's detailed views on states of affairs and then developing
the appropriate seman tical details based thereon. However, we observe some
points fundamental to our concerns. We assume the state of affairs referred
to by a subject-predicate sentence is a singular state of affairs consisting of an
individual having a property. Thus, the referent of 'Sherlock is shrewd' in a
world w is the state of affairs, occurring in w, consisting of the individual
R('Sherlock', w) having the property R('is shrewd', w). And the referent of
'The murderer wore boots' in w is the state of affairs in w consisting of the
individual R ('the murderer', w) having the property R (,wore boots', w).
(Thus, defmite descriptions are treated as .referential terms and are not elimi-
nated as per Russell's theory of descriptions. Note that the senses expressed
by these two sentences may be of different sorts even though the referents
are both singular states of affairs.) Importantly, the state of affairs referred
to by a sentence of form 'The 4> is a 1/1' in one world may be different from
that referred to in another world, because the referent of the defmite descrip-
tion in one world may be different from that in another world (we assume a
qualitative predicate refers to the same property in any world). On the other
hand, if a term 0:, perhaps the name 'Sherlock', takes the same referent in
every world, then the sentence '0: is a 1/1' refers to the same state of affairs in
348 CHAPTER VII
NOTES
1 Though we shall not foray into the details of a theory or metaphysics of possible
worlds and other possibilia, we might note at least the following fundamental assump-
tions we would make in developing a possible-worlds Husserlian theory of intentionality.
Assume possible worlds and possible individuals. Exactly one world is actual. Possible
individuals are individuals that exist in some possible worlds; merely possible individuals
exist in some worlds but not in the actual world. Assume also possible states of affairs,
possible events (including possible acts of consciousness), and possible histories or
courses of events. Possible worlds (of the sort we need consider) are maximal consistent
possible histories. Assume further that the same individual may exist in different possible
worlds: trans-world identity and individuation make sense and trans-world relations of
identity do obtain. (Cf. Husserl's interesting views on trans-world identity, discussed in
INTENTIONALITY AND POSSIBLE-WORLDS SEMANTICS 351
Chapter VIII, Section 3.4.) Meanings are abstract individuals (cf. Chapter IV, Part 2),
which we might hold either to exist in every world or to exist in no worlds (worlds are
concrete histories) but to have their being outside worlds. PossibiIia, however, are not
abstract entities. Finally, we might even assume impossible objects and impossible
worlds, if these notions prove both coherent and useful; but we shall not address issues
concerning them here.
2 Similarly, Hintikka proposes to treat the reference of terms within intensional con-
texts as "multiple reference", reference to objects in different possible worlds. This
view is used throughout his work on modalities of different species; see, for instance,
his Knowledge and Belief (Cornell University Press, Ithaca, N.Y., 1962), p. 140. The
terminology traces to his early paper, 'Modality as Referential Multiplicity', Ajatus 20
(1957),49-64.
3 Alfred Tarski, 'The Concept of Truth in Formalized Languages', in his Logic, Seman-
tics, Metamathematics (Clarendon, Oxford, 1956), pp. 152-278 [a translation of 'Der
Wahrheitsbegriff in den formalisierten Sprachen', Studia Philosophica 1 (1936), 261-
405). See also his less formal essay, 'The Semantic Conception of Truth and the Founda-
tions of Semantics', Philosophy and Phenomenological Research 4 (1944), 341- 75,
reprinted in Linsky (Note 53, Ch. II above).
4 Donald Davidson, 'Truth and Meaning', Synthese 17 (1967), 304-25.
5 This is especially perspicuous in Montague's work: cf. his 'Pragmatics and Intensional
Logic' (Note 5, Ch. VI above). Cf. also Hintikka, 'Semantics for Propositional Attitudes',
in Models for Modalities (Note 6, Ch. I above), p. 93.
6 Cf. Saul Kripke, 'Semantical Considerations on Modal Logic', Acta Philosophica
Fennica 16 (1963), 83-94, reprinted in Linsky (Note 19, Ch. I above). We draw on
Hintikka's explicitly possible-worlds-semantical 'Semantics for Propositional Attitudes'
(Note 5 above). This is closely related to his Knowledge and Belief (Note 2 above) and
a number of essays in his Models for Modalities (Note 6, Ch. I above), as well as earlier
essays, which work in terms of "model sets"; model sets are maximal consistent sets of
sentences, which would describe possible worlds. See also the bibliographies in those two
books.
7 Hintikka, 'Semantics for Propositional Attitudes' (Note 5 above), pp. 90-91.
8 'Semantics for Propositional Attitudes' (Note 5 above), p. 92.
9 Hintikka, 'On the Logic of Perception', in Models for Modalities (Note 6, Ch. I above),
p. 156.
10 Cf. Montague, 'Pragmatics and Intensional Logic' (Note 5, Ch. VI above), p. 144 (cf.
p. 158).
11 The accounts of failures of existential generalization and of substitutivity of identity
in belief-contexts that we present in this section are due to Hintikka. Cf. his 'Semantics
for Propositional Attitudes' (Note 5 above), pp. 96-98. At certain indicated points,
though, we vary somewhat from some details of the semantics outlined in that essay.
12 Quine, 'Quantifiers and Propositional Attitudes' (Note 26, Ch. I above), pp. 177-78.
13 See 'Semantics for Propositional Attitudes' (Note 5 above), p. 98 (bottom).
14 If variables inside belief-contexts range over possible individuals, how are we to take
variables outside belief-contexts? We could treat them differently, with a variable ranging
over actual individuals when it occurs outside belief-contexts but ranging over possible
individuals when it occurs inside belief-contexts. But then we have the strange situation
that in (*) '(3 x) (x is sane and Holmes believes that x is deranged)' the variable ranges
352 CHAPTER VII
fust over actuals and then over possibles. An easier approach is to let variables always
range over possible individuals but hold that (say) 'x is sane' is satisfiable in a world by
a possible individual only if the individual exists in that world. Then (*) is true in a
world w if and only if there is a possible individual that satisfies 'x is sane' in wand
satisfies 'x is deranged' in every world compatible with what Holmes believes in w.
15 Cf. Hintikka, 'Carnap's Semantics in Retrospect' (Note 31, Ch. VI above).
16 This sort of approach has been 'detailed by Montague in his 'Pragmatics and Inten-
sional Logic' (Note 5, Ch. VI above), pp. 158-59.
17 (6) and (7) are not exactly equivalent. For (6) is committed, not to the full range
of the function J'p', but only to the restriction of J'p' to the union of the sets of worlds
compatible with what Holmes believes respectively in the various worlds to which our
semantic theory applies. A similar point applies to the other sorts of meaning functions
we shall write into the semantics: the basic Hintikkian semantics we begin with is strictly
committed only to such a restriction of the meaning functions we shall write into the
relevant tru th-conditions.
18 On Frege's approach 'believes' is taken as standing for a relation between a person
and a propositional meaning, a thought. (The 'that' in 'a believes that p' forms an "ab-
stract noun clause", 'that p', wherein 'p' and its constituents stand for their customary
meanings: cf. Frege, 'On Sense and Reference' (Note 44, Ch. II above), p. 66.) On the
basic Hintikkian "modal" approach 'a believes that', as a modal operator, is taken as
standing for a relation between a person and (in) a world and a set of worlds. Generalizing
(7), we write propositional meaning functions into the interpretation of belief-sentences
'a believes that p'. The resulting semantics is equivalent in a sense to a semantics that
directly interprets 'believes' as a relation between a person and a world and a proposi-
tional meaning function, where 'that p' in 'a believes that p' refers to a propositional
meaning function. Montague has argued to this effect: cf. his 'Pragmatics and Intensional
Logic' (Note 5, Ch. VI above), pp. 157-58.
19 The definitive discussion of a Fregean approach to quantifying into belief-contexts
is Kaplan's 'Quantifying-In' (Note 26, Ch. I above).
20 'Semantics for Propositional Attitudes' (Note 5 above), pp. 101-106.
21 Note that Hintikka has distinguished two different kinds of "individuating" meaning
functions: see his 'On the Logic of Perception' (Note 9 above) and 'Objects of Knowledge
and Belief' (Note 6, Ch. I above). Our notion of an individuating meaning in Chapter
VI corresponds to only one kind of individuating meaning function characterized by
Hintikka; the other we address partly in Chapter VIII, Section 2.2. Our general notion of
"rigid" meaning corresponds to Hintikka's general notion of "individuating" function.
Actually, there are some subtle differences relevant to the discussion of individuation in
Chapter VIII, but they should not concern us at this point.
22 Hintikka has shown some ambivalence as to whether variables of quantification into
belief-contexts, which officially range over individuating functions, should be thought of
as ranging over individuals or over individuating concepts. Although he has compared
individuating functions with a species of individual concepts, he has also tended to think
of individuals themselves - possible individuals, individuals occurring in different possible
worlds - as individuating functions. Behind this tendency is a view he calls "Kantian",
his view that trans-world identities are not simply objectively real but are partly due
to our own conceptualizations, registered in individuating functions. In any event, for
Hintikka, trans-world individuals - presumably as identified with individuating functions
INTENTIONALITY AND POSSIBLE-WORLDS SEMANTICS 353
There are very different kinds of acts that involve defmite intention. There
are acts of different species or "thetic" character, and there are both direct-
object and propositional acts. Consider Smith's admiring Isadora Duncan,
his expecting his (only) brother, his seeing the snail on the garden wall, his
hoping that his psychiatrist will not think ill of him, and his judging that
Hussed will be considered one of the great philosophers. All of these acts are
either "of" or "about" a particular individual in a way that indefmite, or
de dicto, acts are not (cf. Chapter I, Sections 2.6 and 3.5; and Chapter VII,
Part 3). Let us focus on two kinds of definite acts that are especiany common
and important: seeing an individual and judging definitely, or de re, about an
individual. We shall consider, then, various modes of directedness that are
involved in these two kinds of acts.
To begin with, every perception of an individual is directed definitely in
that the individual "itself" is given in direct perceptual acquaintance. Percep-
tion involves a unique mode of directedness, and we shall study it in some
depth in Part 2 below. It has a special sort of defmiteness, which we have
called perceptual defmiteness, or defmiteness of perceptual acquaintance.
The directedness that concerns us in a judgment is its being "about" some-
thing, and defmiteness in a judgment is thus a quality of its aboutness (cf.
Chapter I, Section 1.5, and Chapter VII, Part 3). Now, a judgment too may
be defmite by virtue of perceptual acquaintance, as when Smith judges of, or
about, the woman he sees on the sidewalk ahead that she is a psychiatrist.
Here one judges about an individual one currently sees. The aboutness is
determined by the subject's perception of the individual he judges about,
and so we may say the judgment is perceptually definite. A related but more
complicated mode of aboutness is that of a judgment about an individual one
saw on a previous occasion and now remembers (or, alternatively, remembers
seeing). Here the aboutness is fixed by a sort of "deferred" perceptual ac-
quaintance, yielding a modified sort of perceptual definiteness of judgment.
Yet another mode of aboutness occurs when a person knows or believes that
someone else saw an individual on some occasion and judges something about
that individual. This judgment too seems to be definite, and defmite also by
a kind of deferred perceptual acquaintance, although the deferral is social or
intersubjective. 1
Another kind of definite judgment - to which the third and fourth parts
of this chapter are devoted - is what we have called individuative, or individ-
uatively definite, judgment. An individuative judgment occurs where one
knows (or opines or conceives) who or which an individual is and judges
about that individual. That is, the subject has knowledge or an opinion or
356 CHAPTER VIII
3.2). (The judgment is not exactly the same, it seems, as the de dicto judg-
ment that the theologian named 'Berdyaev', whoever he is, must have been
influenced by Nietzsche.) We might say the judgment is definite by virtue of
"deferred" individuation.
individuatively definite (say, Smith has made a study of Ms. Duncan's career),
or it may be defmite by virtue of perceptual acquaintance (say, he saw her
dance, not knowing who she was, and his companion said, "That is Isadora
Duncan"). Generally, when a person uses a proper name (e.g., 'Isadora
Duncan'), he has a particular individual "in mind" in some mode of defmite
intention, and he uses the name to refer to that particular individual. His use
of that particular name is part of a tradition of use of that name into which
he has been appropriately initiated. Often; perhaps even typically, the in-
dividual is individuated for him, that is, he knows who it is. But individuation
of this sort is not necessary in order to use the name. All that is required is
some mode or other of definite intention, together with the knowledge (or
reason to believe) that the individual so intended is the individual the name
is used to refer to. A name, then, need not "express" any specific structure
of noematic sense, though its use on a particular occasion presupposes some
mode of defmite intention. Insofar, Smith's utterance, "Isadora Duncan was
a truly inspiring dancer", does not fully express the noematic structure of
his judgment. (In Part 4 below we consider the noematic structure of an
individuative judgment.)
Smith himself might give phenomenological description to his judgment
by uttering the sentence,
(1) I judge that Isadora Duncan was a truly inspiring dancer.
And we might describe his act by saying,
(2) Smith judges that Isadora Duncan was a truly inspiring dancer.
Generally, the truth of a sentence such as (2), in which a proper name occurs
within an intentional context, requires at least the follOwing: that the subject
of the act described be appropriately familiar with the use of that name (the
tradition of use in which the utterer of (2) uses it), that in the act he intend
a particular individual in some mode of definite intention, and that he be
disposed to use the name to refer to the individual he so intends (or, that he
hold that the individual he so intends is the individual the name is used to
refer to-). From our skeletal account above of the use of proper names, then,
we get a basic account of their use within intentional contexts. In particular,
it should be fairly easy to account on this basis for failures of substitutivity of
identity for names in intentional contexts, even where different names for the
same individual are inter-substituted.
Now, perceptually definite judgments fmd a natural expression by the use
of demonstrative pronouns such as 'this' and 'that' (cf. Chapter N, Section
360 CHAPTER VIII
3.4). When, for instance, Smith sees an object on the beach and judges it to
be a jellyfish, he may express the Sinn of his judgment by saying, "This is a
jellyfish". Here his use of 'this' is specifically keyed to his perceptual ac-
quaintance with the jellyfish: he uses 'this' specifically to refer to the object
he sees before him. Indeed, it can plausibly be held that his uttering 'this'
expresses a basic component of the Sinn of his perception of the jellyfish.
We study this view of demonstratives in Part 2 below, where we shall see that
Hussed held a view of this sort. (Cf. Chapter IV, Section 3.4.)
Accordingly, Smith may give a phenomenological description of his per-
ceptual judgment by saying,
We, however, could not capture the Sinn of his judgment by saying,
For our utterance of 'this' would (we may suppose) be keyed to our own
perception of the jellyfish rather than to his.4
Now, apparently, the most general way to describe acts of defmite inten-
tion is by quantifying into intentional contexts, as in
and
(6) (3x) (Smith sees x as a beach bum).
2. PERCEPTUAL ACQUAINTANCEs
is selected in virtue of its location in the field. The field is defined by a dis-
tribution of colors and shapes, enjoying the support of sensuous stimulation
(for Hussed, sensuous "filling" or "hyle": see Chapter III, Section 2.6.) To
the extent that the object's spatial location is determined by the distribution
of colors and shapes among the object and its environs, the acquainting sense
includes predicative content and appeals to predicate-senses in determining
the object prescribed. But the predicate-senses involved are restricted to
predicate-senses prescribing "sensory" qualities such as color and shape, and
all enjoy the support of sensuous stimulation. By contrast, HusserI's separa-
tion of the X and the predicate-senses in a perceptual Sinn seems to entail
that the object prescribed is determined independently of the predicate-senses
in the Sinn; however, Hussed does not explicitly address the independence,
and he should be open to the dependence on sensory qualities.
If the acquainting sense at the base of a perception's Sinn includes or builds
upon sensory predicate-senses prescribing colors and shapes, it is nonetheless
not a descriptive sense: its structure is not that of, say, "the red round bumpy
object". Such a descriptive sense is not a rigid sense, where an acquainting
sense is, as we shall elaborate in the next section. Nor is such a descriptive
sense a "demonstrative" sense, as is an acquainting sense, pointing out some-
thing as before the perceiver at the time of the perception. The proper internal
structure of a perceptually acquainting sense is that of an object singled out
in a perceptual field. Nothing could be more familiar. Yet we cannot here say
more exactly what that "logical" or phenomenological structure is, except to
note that it is not a descriptive structure.
HusserI did not articulate at the base of a perceptual Sinn the structure we
have called an acquainting sense. He did, however, layout a possible founda-
tion for this structure. For he held that material objects ("things", Dinge) are
"constituted" in levels, as it were, with the most basic "stratum" being that
of a spatial object with only "sensory" qualities:
IE] very appearance of a thing necessarily conceals in itself a stratum that we call the
thing-schema: it is the spatial form IRaumgestalt] filled out merely with "sensory"
qualities - lacking every determination of "substantiality" and "causality".... (Ideas,
§ 150, p. 370; cf. also § 151 and Chapter V, Section 2.4, above for comments on these
passages.)
Romane Clark has characterized it, simply the thesis that the objects of
perception are everyday physical objects. 9 Similarly, when we sayan object
is singled out in a visually given spatial field, we do not mean a "phenomenal
field", i.e., a distribution of visually given colors forming a complex of sense-
data. Rather, we mean that in perception one is presented with an object in
a spatial field; one is presented with a local region of space occupied by
various things including the given object. If the perception is veridical, the
object presented is an ordinary physical object and the spatial field presented
is a region of physical space occupied by appropriate objects including the
given object. And if the perception is not veridical, it is still the case that the
object is intended as a physical object in a region of physical space. (Bear in
mind that, all along, we have been speaking of perception in the experiential
sense, so that the object of a perception is precisely what the noematic Sinn
of the perception prescribes.)
presents an object as directly before one. This idea is reflected here in the role
of the relational property of standing in the relation R ,to Smith, its role in
generating the values of the function described. Both Smith and R remain
fIxed as the function selects its value in each world. So the value of the func-
tion (in any world) is whatever individual is before Smith (in that world).
Now, this explication of perceptual acquaintance, or of acquainting sense
in a given perception, seems just about right; but it leaves out something
important. On this explication, perceptual intention/acquaintance and ac-
quainting sense are non-rigid (cf. Chapter VI, Section 2.7): the representative
function does not select the same individual from each possible world, since
in different worlds different individuals may stand in the relation R to Smith.
But then perceptual intention, perceptual acquaintance, is not defInite - if,
as we assumed, rigidity of intention is the mark of definiteness (cf. Section
1.4 above). (However, as we shall shortly note, Hintikka himself would main-
tain that perceptual acquaintance is rigid, but in a weaker sense than we mean
here.) Further, demonstrative reference, where mediated by such a meaning,
is also not rigid: 'this', uttered on a particular occasion of perception, does
not take the same referent in each world. Yet there is a signifIcant inclination
to think that demonstrative reference is rigid; this inclination is embodied in
David Kaplan's work on demonstrativesY Briefly, Kaplan would require that
'this' (or 'that'), uttered on a particular occasion, refer to the object appro-
priately before the speaker (and appropriately demonstrated) in the actual
context of utterance. That very object itself is the referent in any possible
world (wherein the object exists), and so the reference is rigid. This suggests
a modifIcation of our first explication of perceptual acquaintance, a modifIca-
tion that leaves perceptual intention rigid.
Consider accordingly the following proposal: an instance of perceptual
acquaintance, or an acquainting sense instantiated in a given perception, may
be represented by the function that assigns to any world the individual (if
any) that is in fact located at a certain place before the perceiver at the time
of the perception, that is, in the actual context in which the perceiver is
situated in the world in which the perception takes place. The values of this
function, like those of the function we first considered, are defIned in terms
of an object's visible spatial relation to the perceiver. But they are determined
in a different way. For the case of Smith's perception, fmd Smith in the
actual world in which the perception takes place, fmd the individual that in
the actual world stands in the relation R to Smith, and that same individual is
then the value of the function at any world (wherein that individual exists).
Thus, the function so defIned selects the same individual from each relevant
368 CHAPTER VIII
possible world and so is a rigid function. Note that this second explication of
acquaintance simply strengthens the first in an interesting respect: it repre-
sents the acquainting sense inhering in a given perception as presenting an
object not merely as before oneself, the perceiver, but as actually before one,
before one in the actual context, and so in the actual world, in which the
perception actually occurs. This seems an important phenomenological feature
of perceptual acquaintance, and so we accept this second explication of
acquaintance or of acquainting sense. 12
Notice that if the perception is not veridical, if there is no object appro-
priately before the subject in the actual world, then the perceptual intention
is a smashing failure. Not only does it fail to reach an object in the actual
world, it fails to reach an object in any world: the representative meaning or
intention function fails to pick out an object not only in the actual world but
also in any other world as well, since its value in any world is to be that same
object it selects in the actual world.
We have preferred to represent perceptual intention by the function that
assigns to any world the object actually before the perceiver, i.e., in the actual
world. The advantages of this explication over the initial proposal are two: it
recognizes a sense of actuality implicit in perception; and by doing so, it
preserves the rigidity of perceptual intention. However, the rigidity of per-
ceptual acquaintance poses a problem.
As Hintikka has stressed so well, perception is (as we would say) always
perceptually defmite but usually individuatively indefmite. Thus, I see "this"
gentleman before me but I do not see "who" he is: my perception includes a
sense of a particular person actually before me, but it includes no sense of the
identity of that person. Now, on our theory, the perception is directed by
virtue of its acquainting sense toward the same individual in each world com-
patible with the perception. Yet, since the perception does not specify "who"
"this" man is, it would seem the perception is directed toward different
individuals in various worlds compatible with the perception. Recognizing
the individuative indefiniteness of perception, Hintikka allowed that a "per-
ceptUally individuating" meaning function assign different individuals to
various worlds compatible with the perception - different, that is, by prin-
ciples of "physical individuation". Seeking to maintain the de re character
of perception with a kind of rigidity, however, he held that the individuals
assigned are, by different criteria of identity, "the same" - the same by
principles of "perceptual individuation". Thus, Hintikka dealt with these
issues by positing two methods of trans-world identification. The problem
we see for Hintikka's approach is this: is there a plausible sense in which
DE RE INTENTION IN A HUSSERLIAN FRAMEWORK 369
consciousness. In this part of the· chapter we lay the groundwork for that
analysis. The groundwork includes a careful audit of the fundamental notiohs
of identity and individuation. These are matters of metaphysics, the central
issue being wherein the identity of an individual consists. A conception of the
identity of an individual is a phenomenological matter, the issue being how
the identity of an individual is grasped or "constituted" in consciousness,
how the individual is thereby "individuated in consciousness". The metaphys-
ical principles of individuation we study here are offered as principles inherent
in our working conceptual scheme, and as such they are relevant to the sub-
sequent phenomenological study of individuative consciousness. Not to leave
Husserl completely behind, we shall trace out some of his proposed doctrines
on individuation; those doctrines are not only interesting and plausible but
ostensibly indigenous to our conceptual scheme and are so offered by Husserl.
too different in some specified way. Principles such as this last one are based
on a theory of how individuals of that kind change through time.
Let us call principles of individuation any such propositions setting condi-
tions on identity for a given kind - necessary conditions, sufficient condi-
tions or both. We shall avoid the phrase 'criterion of identity' since it may
call to mind only necessary and sufficient conditions of identity. Correct
principles of individuation will presumably be necessary truths, true in virtue
of the "essence" .of the given kind.
We shall be primarily concerned with natural individuals: inanimate physi-
cal objects, living things, and, in particular, human beings. The individuation
of a natural individual is the determination, according to appropriate prin-
ciples of individuation, of the various propositi9ns of identity and distinctness
true of it. Such individuation addresses prominently relations of identity
across time as well as at a fixed time. Indeed, the individuation of a natural
individual is achieved largely by its progress through time, that is, by its
persisting through various changes in its location and extension, its qualities,
and its relations to other things. As we shall see in Section 3.3 below, Husserl
held such a view of individuation for natural objects.
Our discussion so far has been effectively confmed to identity and individ-
uation concerning individuals within the actual world. But given a metaphysics
of possible worlds, we must consider identity and individuation within other
possible worlds and, more importantly, across different possible worlds.
Trans-world identity is identity across different possible worlds, identity such
that an individual that exists in one possible world is identical with an in-
dividual that exists in another possible world. We have already seen trans-world
identity assumed in the possible-worlds analysis or explication of de re inten-
tion. And it is assumed in certain possible-worlds analyses or explications of
de re modal attributions: for instance, 'Truman might have lost the 1948
election' would be taken to mean that in another possible but non-actual
world Truman lost the 1948 election, presupposing that one and the same
individual, Harry S. Truman, is present in each of two possible worlds.
Trans-world individuation, then, is the determination of which individual
a thing is among individuals occurring in different possible worlds. It consists
in the determination of the various propositions of trans-world identity and
distinctness that are true of a given individual - that is, the determination
of the relationships of trans-world identity and distinctness into which the
individual enters. This determination presupposes the individual's individ-
uation within each world in which it exists, the determination of those
relationships of intra-world identity and distinctness into which the individual
DE RE INTENTION IN A HUSSERLIAN FRAMEWORK 373
enters, including identity and distinctness both at a fixed time and across
time.
There do not seem to be fmite sufficient conditions of trans-world identity
for natural individuals. However, being of the same kind is a plausible neces-
sary condition. So a plausible principle of trans-world individuation for
natural individuals is this: if a natural individual x exists in one world and a
natural individual y exists in another world, then x is identical with y only if
x and yare of the same natural kind (say, camels) - and/or of the same
natural substance (e.g., flesh or coal). Like intra-world individuation, trans-
world individuation is a matter of substantive theory about a given kind of
individual, where however the theory encompasses de re modal claims about
individuals of the given kind.
Trans-world identity and individuation have been considered problematic
by some philosophers, largely because it is difficult to see what makes an
individual in one world the same as an individual in another. One issue in
question is that of essentialism, the doctrine that some of an individual's
properties belong to it essentially. Assuming a metaphysics of possible worlds
and trans-world identity, a property is essential to an individual, or the in-
dividual has it essentially, if and only if it belongs to the individual in every
possible world in which the individual exists. A further issue is whether some
fmite subset of an individual's essential properties - called, collectively, the
individual essence of the individual - suffice to individuate it in every possible
world in which it exists, and so to achieve its trans-world individuation. It is
not very plausible that natural individuals have such individual essences. Nor
does a high degree of similarity between individuals in different worlds seem
sufficient to "make" them the same. 13 An interesting and perhaps plausible
approach to trans-world individuation for natural individuals is Husserl's,
which bases trans-world individuation, like intra-world individuation, on
continuity through time. We shall consider Hussed's approach in Section 3.4
below.
Let us now summarize briefly the primary notions of this section. The
relation of identity is simply that of being the same as; the relation of dis-
tinctness, that of not being the same as. The individuation of an individual is
the determination of which individual it is, whether it is or is not identical
with this individual, with that individual, and so on. Collectively, the prop-
erties of an individual that serve to individuate it, to determine which in-
dividual it is, make up its identity. Relations of identity include relations of
identity at a fixed time, identity across time, identity within a world (in par-
ticular, the actual world), and identity across possible worlds. Relationships
374 CHAPTER VIII
comprise, for each world in which the individual exists, its full range of
(non-modal and non-"subjective") properties in that world. 16 This view of a
natural individual's identity seems pessimistic at first, but it gains plausibility
as we attend to details.
Different properties of a natural individual play different roles in its
individuation and hence in its identity. The kind to which an individual
belongs is not unique to it but yet distinguishes it from every individual
of every other kind; further, an individual's kind is essential to it and so
distinguishes it from any individual in any possible world that is of a different
kind. Thus, the genera and species of a thing serve partially to individuate it
- indeed, within the actual or any possible world - and so must have a place
in its identity. Integral to a thing's kind is its "nature", which comprises
various of its dispositions to behave in certain ways. Such dispositions too
are not unique to it but help to individuate it and so belong in its identity;
and some are arguably essential to it. (A fox is essentially crafty, or at least
essentially four-legged, barring mutation or mutilation.) Other dispositions of
an individual are strongly bound to it and help to individuate it at least in the
actual (or some one) world but are not essential to it (thus, a given politician
is tricky to the core, but had he known a different childhood and a more
successful athletic experience he might not be). Such dispositional traits will
also enter a thing's identity. It is difficult to think of any properties of an
individual that are, collectively, both unique to it and also essentially unique
to it, unique to it in every world. Thus, it is difficult to see any plausibility
in the strong sort of purely qualitative individual essence we first noted. But
of course some properties of an individual are unique to it at least in the
actual (or some given) world: for instance, its being in a given location at a
given time, or something's happening to it at a given time, or its performing
a specific action at a given time. Such "historic" properties playa prominent
role in individuation (within a given world), and so in a thing's individuation;
and accordingly we often base our judgments as to an individual's identity
on knowledge of such properties. However, an object is not completely
individuated even within a given world by anyone such property. For the
fact that an individual is, say, (uniquely) located at a given place at a given
time may not - and usually does not - determine whether it is located at
another place at another time. And yet this determination is a part of the
individual's individuation. Thus, it seems that the complete individuation of
a natural object, and hence its identity, must cover all its "historic" prop-
erties, as well as any other distinguishing characteristics (such as fmgerprints,
birthmarks, or scars, for human beings). And so it seems on reflection that
378 CHAPTER VIII
the full identity of a natural object must include a very wide range of its
properties. We mayor may not conclude that the identity of a natural
individual includes everything about it (save its· modal and "subjective"
attributes), but the preceding reflections argue that an individual's identity
goes well beyond its qualitative essence (its combined essential qualitative
properties) and includes many of its concrete contingent properties in dif-
ferent worlds.
Assayed so, the identity of an individual is clearly not a property on a par
with other properties of the individual; it is an extremely complex property
that strictly sets the individual apart from all others. Further, from what
we have said, it seems evident that the identity of a natural individual is
transcendent of human consciousness: it cannot be completely known, and
it cannot be completely grasped through (the predicate~enses in) any single
noematic Sinn or any fmite set of Sinne. Reflecting this fact, perhaps, the
notion of an individual's identity grows rather lax in everyday use.
Significantly, Husserl expressly held that the "individual essence" of a
material object escapes complete, or "adequate", human apprehension. Late
in Ideas he writes:
... The essence "(Material] Thing" is originally given (in "ideation" (p. 368), by eidetic
variation (p. 365)], but this givenness cannot on principle be adequate. We can bring
the noema or thing-meaning (Ding-Sinn] to the point of adequate presentation (in
phenomenological reflection]; but the manifold thing-meanings [Dingr-Sinne], even
taken in their fullness, do not contain the regional essence "Thing" as an originally
intuitable constituent immanent in them, just as little indeed as the manifold meanings
[Sinne] relating to one and the same individual thing contain the individual essence
[/ndividualwegen] of this thing. In other words, whether it is the essence of a thing-
individual [Dingindividuumg] that concerns us or the regional essence Thing in general,
in no case does a single thing-intuition [ie., perception] or a f'mite closed continuity or
collection of thing-intuitions suffice.. to obtain [in "ideation" (p. 368)] in adequate fonn
the desired essence in the total fullness of its essential detenninations. An inadequate
insight into the essence is, however, always obtainable ....
This holds true for all levels of generality of essence, from individual essence up to
the region Thing. (§ 149, p. 365.)
holds true from the highest to the lowest level of essence belonging to a
natural individual: from its most general kind, the "region" Material Thing, to
its genus and species and even to its specific "individual essence". He says
that no finite set of noematic Sinne "contains", or adequately presents, a
thing's "individual essence". That is to say, the "individual essence" of any
material thing, and hence that of any natural object, is transcendent of
human grasp. (N.B.: we do not interpret Husserl's 'contain' literally in this
passage since essences are not properly senses but are rather properties or
universals. Cf. Ideas, Chapter 1, especially the end of § 10.)
It is not clear just what a thing's individual essence is supposed to include
for Husser!' Clearly, it is to be an essence, or universal, and it is apparently to
individuate the object (cf. £J, Appendix I; Ideas, § 12, on "eidetic singulari-
ties"). A reasonable hypothesis may be that it is to include everything about
the object, in particular, its spatiotemporal attributes. For Husserl's own
specific discussions of individuation heavily stress spatiotemporal location
and continuity (cf. Section 33 below) and seem to align with our consid-
erations in Section 42 below. Note that a thing's individual essence then
coincides with its complete "determination", which of course we know is
transcendent. Note also that our remarks on transcendence pertain only to
individuation within a single world, presumably the actual world.
Husserl's conception of the individual essence of a thing, whatever the
details, would be a specification of our broad notion of the identity of a thing
set out above. But now, since the individual essence of a material object is
transcendent, it might be thought that our intentions of material objects
could never be strictly individuative, directed to a particular object in virtue
of a sense or knowledge ofits identity. However, that would be going too far.
In the quotation above, Husser! says that although the individual essence of a
thing cannot be given adequately (completely), it can be given inadequately,
apparently with varying degrees of adequacy or inadequacy. Thus, although
individuation in consciousness is achieved in degrees and only incompletely,
for Husserl, we often attain a sufficient fix on an individual's identity that we
may intend it with individuative defmiteness.
Husserl's specific views on the individuation of material things support his
view that the individual essence of a material thing is transcendent. We tum
now to his views on individuation.
The brown of this and the brown of that piece of the duration in question are distinct
substrata, but insofar as they fill continuously one time-stretch there is one substratum,
one enduring [entity), which goes through this time-stretch and its substrata. (Zeit.,
p.265.)
The experiencing consciousness (giving [natural) individuals at fust hand) is not only
a flowing consciousness, spreading itself out in the flux of lived experiences, but a
consciousness-of, an integrating consciousness. In it, therefore, there is to be distinguished
in every phase an objective correlate, and, in each new phase, a new correlate, but only
in such a way that all the continuous momentary objects join together in the unity of
a single object, like the moment of consciousness in a single consciousness-of. (§64b,
p. 257.)
A temporal sequence has unity ... if that which succeeds shows a certain continuity of
content and has this way of continuous transition .... The unity is that of the identical
substratum for this continuum as event (Zeit., p. 264; quoted above).
We can speak here of the likeness and similarity of the components of such worlds but
never of their identity, which would have absolutely no sense .... It makes no sense,
e.g., to ask whether the Gretel of one fairy tale and the Gretel of another are the same
Gretel, whether what is imagined for the one and predicated of her agrees or does not
agree with what is imagined for the other, or, again, whether they are related to each
other, etc. I can stipulate this - and to accept it is already to stipulate it - but then both
fairy tales refer to the same world ....
. . . In the continuation, although free and open, of the unity of a complex of imagin-
ings, it is the unity of a "possible world" which is constituted with an encompassing
form of the time of imagination pertaining to it.
In what has been pointed out, the implication is that individuation and identity of
the individual, as well as the identification founded on it, is possible only within the
world of [i.e., the world constituted in] actual experience [Le., perception and percep-
tual judgment] , on the basis of absolute temporal position. (EI, §40, p. 173.)
Husserl's view here seems clear and strong: trans-world identity and individua-
tion (constituted in imagination) make no sense. However, as we shall see, his
point is not that strong.
Husserl's actual view on trans-world individuation connects with what we
have just seen of his view of intra-world individuation. Fundamentally, HusserI
seems to hold that identity of natural individuals makes sense only if they
belong to a common flow of time. Hence, identity across worlds makes sense
only if the worlds are alternative courses of events that coincide up to some
point in time, i.e., share a common partial history, and so have a temporal
384 CHAPTER VIII
... The "things", the events, the "actualities" of one world of imagination have "nothing
to do" with those of the others. Better: the fulfillments and disappointments of inten-
tions constitutive of one of these worlds can never extend to intentions which are con-
stitutive of another .... Here the unity of time plays its special role as the condition
of the possibility of a unity of the world .... (EJ. §40, p. 172.)
In this unique world [the intersubjective, natural life-world, our earth), everything
sesuous that I now originally perceive, everything that I have perceived and which I can
now remember or about which others can report to me as what they have perceived or
remembered, has its place. Everything has its unity in that it has its fIxed temporal
position in this objective world, its place in objective time.
This holds for every object of perception as such, i.e., as an intended object, as an
object alleged to actually exist. (EJ, §38, p. 163.)
We now understand the inner truth of the Kantian thesis: time is the form ofsensi-
bility, and thus it is the form of every possible world of objective experience. (P. 164.)
Every perception and every recollection as the reproduction of a perception must,
therefore, set up for their objects a temporal relation which on principle is capable of
being made intuitive. They are connected with each other as referring to objects, either
actual or intended, within one world. This connection serves as the basis for a certain
kind of relation, for relations of the temporal location of all perceived objectivities
intended in perceptions as actually existing. (P. 166.)
Note especially Husserl's claims: "Everything [perceptible] has its unity [Le.,
its individuation] in that it has its ftxed temporal position in this objective
world", and "time . .. is the form of every possible world of objective experi-
ence [Le., perception]". (Husserl is speaking of objective time, not inner
time, here: cf. p. 163.)
Following this discussion of the role of time in individuation, Husserl in
§40 concludes that "individuation and identity ... is possible only within
the world of actual experience [perception]". But he then goes on to loosen
the restriction. §41 poses "the problem of the possibility of an intuitive
unity [in particular, identity] between objects of perception and objects
of imagination of one ego". And §42 presents a solution. Objects of differ-
ent intuitions, say a perception and an imagination (cf. EJ, §42), may be
386 CHAPTER VIII
point is this: "Different individuals ... can ... form a unified intuitive objec-
tivity, only.insofar as they are encompassed by the unity of an intuitively
constituted time".
In these passages of Experience and Judgment Husserl thus allows for
identity of individuals across worlds constituted in different imaginations
or in imagination and in perception, provided those worlds share a common
time frame. And as we observed earlier, with his theory of horizon he is
already committed to identity across different worlds constituted in different
possible perceptions~ Husserl's down-to-earth view of trans-world identity is
an appealing solution to a problem that -- pretty much independently of
issues that exercised Husserl, such as the "horizon" of a perception - has
become a pivotal point in modal ontology nearly half a century later.
The study of Husserl's views of individuation through time and across
worlds allows a smooth transition to our study of individuative consciousness
in the next part of the chapter. For Husserl's discussion of metaphysical
issues of individuation was a discussion of metaphysical principles presup-
posed in the "constitution" of individuation or identity in consciousness. Let
us tum to the associated phenomenological issues.
Our concern now is the analysis of that particular type of defmite, or de re,
intention we have called individuative, or individuatively defmite, intention.
Our task is to describe the phenomenological structure of such an act, and
how it achieves individuative directedness.
Smith simply judges about Billy Joe himself, that he is the meanest man in
town. (Even the name 'Billy Joe' may not cross his mind.) It is a natural
proposal, then, that the Sinn of Smith's judgment is a propositional Sinn
whose "subject" -component is what Husserl calls an "X" - and whose
"predicate" -component is the sense "is the meanest man in town". For an X,
we proposed, is no more than an intensional token for an individual, having
no structure per se.
Earlier, we characterized an "individuating" meaning as an individual
meaning that incorporates a conception of the "identity" of an individual
and thereby picks out that individual in any possible world (cf. Chapter VI,
Section 2.7, and Chapter VII, Section 3.3). Now, an individuating meaning
might have been expected as SUbject-component of an individuative act such
as Smith's judgment about Billy Joe: the act's aboutness is determined by the
individuation of Billy Joe for Smith, and it might have been expected that
this aboutness would be detennined directly by the Sinn's subject-component.
But the above considerations indicate that for such a judgment, though the
act is individuative, no individuating meaning is likely to occur as a component
of its explicit Sinn. Specifically, the subject-component of the Sinn is an X,
and an X is not an individuating meaning. The X in these cases presupposes
individuation, however, and so an individuating meaning seems to be involved
in th~ act in a less direct way, as we may shortly see.
What makes Smith's judgment a judgment about the individual Billy Joe is
Smith's knowledge or conception of the identity of Billy Joe, of who Billy
Joe is. Yet, as we just observed, this knowledge or conception is not itself a
part of the Sinn of the judgment. How, then, does this knowledge serve to
make the act about Billy Joe?
We have argued that the intention achieved in an act often depends not
only on the explicit Sinn of the act but also on the act's "pragmatic" presup-
positions - specifically, on a network of background beliefs presupposed in
the act (cf. Chapter V, Sections 3.2 and 3.3, and Chapter VI, Sections 1.4 and
1.6). Now, it seems clear that Smith's knowledge of Billy Joe's identity is just
the sort of knowledge that will be found dispersed throughout such a system
of background beliefs, beliefs in which Billy Joe is individuated for Smith.
(In subsequent sections we consider what might be included in these beliefs.)
These beliefs are presupposed by Smith in his act of judging; and their Sinne
are part of the act's background meaning, in the sense we distinguished in
Chapter V, Section 3.3. Thus, their Sinne are in no way incorporated in the
Sinn of the judgment, but they are in a certain way presupposed by that Sinn,
in that they are presupposed in the act. Specifically, we may hold, the X in
DE RE INTENTION IN A HUSSERLIAN FRAMEWORK 389
uncovers the role of the act's background Sinne in determining the identity of
the object intended in the act.
Nonetheless, neither method of explication, in terms of horizon or of
possible worlds, automatically reveals the substantive content of the back-
ground beliefs that serve to individuate. What horizon-analysis, or possible-
worlds explication, does is to show what the act's "meaning" must achieve.
This may then lead one to reflect on the belief-structures that individuate
the object for one. But, importantly, the background beliefs or meanings
that individuate an individual for one are not easily discovered in reflection.
Many of them have receded into "obscurity", as Husserl says, in the "passive
background" of consciousness; they are but a "habituality of the ego ...
ready for a new associative awakening" (EJ, §67b). What remains readily
accessible is a "trace" of the individuation they achieve, a trace of the individ-
ual they individuate, in the form of an X.
in the Bay of Pigs affair and the Cuban missile crisis, and so on. And I know
who Freddie McAlister is: he is the fellow who sat next to me in the fifth
grade, he played center on our high school basketball team in 1961, he went
to the state university, and he is now the only orthodontist in town.
Proper names sometimes play a special role in our accounts of knowing-
who. Thus, I know who that man across the street is: he is Dr. Lauben. And
I know who the author of L 'etre et Ie neant is: he is Jean-Paul Sartre. But
such items of knowledge - that that man across the street is Dr. Lauben or
that the author of L 'etre et Ie neant is Jean-Paul Sartre - are not the ultimate
or most basic items in which knowing-who consists. It is more basic to
knowing-who that I know that Dr. Lauben is a very old acquaintance of my
family, and so forth, and that I know that Jean-Paul Sartre studied at the
Sorbonne, was the leading spokesman of extentialism, and so forth.
Most basically, we may say, to know who a is is to know that a is l/J for
various predications 'l/J' that ascribe properties that are of significance to the
individuation of the individual a and thus form some Significant part of the
identity of a. 18 These properties may include physical traits, personality or
character traits, occupation, notable achievements or deeds, time and place
of birth, lineage, and so on. Ideally, the senses of the relevant predications
would go together to form an individuating meaning (cf. Chapter VI, Section
2.7). However, this ideal may be unrealizable. If the identity of a person is
indeed transcendent, as Section 3.2 above would suggest, then our knowledge
of a person's identity, of who he or she is, may fall short of what is required
of a bona [ide individuating meaning. In particular, a knowledge of who a
is may not yield a rigid concept of a.
There is a laxity apparent in our everyday standards for knowing-who.
Indeed, precisely what counts as knowing who a is seems not to be a defmi-
tive matter but to be relative to the needs of our immediate knowledge claims
about a. The laxity and relativity we accept may well reflect an awareness
that the identity of a person is transcendent, and that consequently we know
a person's identity only "inadequately" in bits and pieces. Thus, there is
a vagueness of degree inherent in knowing-who. Further, one may fmd it
difficult to articulate the knowledge one has of a person's identity; one may
even defer to others, to the experts, to fix a person's identity beyond a
few salient items. These attitudes, too, seem to reflect an awareness of the
transcendence of a person's identity: we rest content with a certain minimal
grasp on who a person is, knowing that his identity is "out there", more
fully determinable as need be.
Knowledge is not an occurrent act of consciousness but a dispositional
DE RE INTENTION IN A HUSSERLIAN FRAMEWORK 393
for the same reason. Also individuative is Sherlock's judgment, issuing from a
series of brilliant "deductions", that the murderer lives at a certain place and
works for a certain sinister professor and has certain traits of appearance and
behavior. This judgment, however, is an act in which an object's individuation
for the subject is explicitly achieved, rather than presupposed, in the act. The
features actually given in the act individuate the murderer for Sherlock: they
give Sherlock a rudimentary grasp of the murderer's identity. And Sherlock's
judgment is thus an act of identification of the basic sort we described.
Importantly, the individuation of an individual in one's experience is
regulated by one's fundamental background beliefs concerning individuation,
beliefs that articulate principles of individuation for the kind of individual
given. (Cf. Chapter Y, Section 3.2, on fundamental. background beliefs
presupposed in an experience. And see Part 3 above for the relevant kind of
principles of individuation.)
"The murderer entered by the window", and then, noting the size of the
window, adds; "He was a small man". Holmes' reference in saying 'He' is
called anaphoric because his term 'He' refers "back" to the referent estab-
lished by the grammatical antecedent of the pronoun 'He' (the antecedent
here being Holmes' utterance of 'The murderer'). Underlying Holmes' succes-
sive linguistic assertions are successive judgments (cf. Chapter IV, Section
2.3). In the second judgment Holmes judges about the individual he judged
about in the ftrst judgment: his intention, or his act's aboutness, in the
second judgment is in this manner tied "back" to his intention, or his act's
aboutness, in the "antecedent" judgment. Thus, we may call the intention
or aboutness in the second judgment anaphoric. (Anaphoric reference, then,
is founded on anaphoric intention.) The aboutness of Smith's judgment
is similarly anaphoric, though with two differences. First, the antecedent
intention is a concurrent attitude of belief rather than a prior act of judgment
(though the belief is no doubt the "residue" of a prior judgment or group of
judgments). Second, and more important, Smith's anaphoric judgment is
defmite, or de re, where Holmes' is indefmite, or de dicto. Signillcantly,
anaphoric intention may be either defmite or indefmite.
What is the structure of the Sinn of an anaphoric judgment? Evidently,
the Sinn of Smith's judgment about Billy Joe consists simply of an X com-
bined with a predicate-sense, where the X is somehow grounded in the Sinn
of the "antecedent" background belief. Similarly, it would seem, the Sinn of
Holmes' anaphoric judgment consists of an X and a predicate-sense, where the
X is appropriately tied to the subject-component of the Sinn of the antecedent
judgment. In general, it would seem, the Sinn of an anaphoric intention
includes an X-type of sense that is tied to an appropriate element of sense in
the Sinn of the antecedent act or attitude. But now, in Chapter IV and in
Section 1.3 above, we recognized X's only in defmite intentions. Here we
would need also to recognize X's that belong to indefmite intentions. Indeed,
anaphoric intentions inherit the character of defmiteness or indefmiteness
belonging to their antecedent founding intentions. So, if anaphoric intention
takes an X-form of sense, we must recognize that this X achieves a defmite
intention only if tied back to a defmite intention, introduced perhaps in a
defmite intention such as perception (cf. Section 2.1). But let us here focus
on deftnite anaphoric intention, observing only X's bearing the character of
defmiteness.
When we take X's as this sort of anaphoric sense, and already when we
hold that the X in a perceptual Sinn is "introduced" by an acquainting sense
and so becomes available for later thinking of that same object (cf. Section
396 CHAPTER VIII
knowing-who. (Thus are we disimpaled from our chosen hom of the dilemma.)
The beliefs that constitute a case of knowing-who are not easy to unfold, for
we may acquire many and varied beliefs about an individual and thereby in
a most unsystematic way acquire a knowledge of who the individual is. As
we observed in Section 4.2, knowing-who is in fact and for good reason a
very flexible notion. In light of Part 3 and Sections 4.1-4.2, let us unpack in
a schematic way the structure of such an array of collectively individuating
background beliefs. Here we abandon the simplifying assumption we made
shortly above, that Smith's knowing who Billy Joe is forms a single individ-
uating belief.
We may naturally see Smith's knowledge of who Billy Joe is as dispersed
through a system of beliefs that attribute various properties to the same
individual. That is, the beliefs are all co-directed, or about the same individ-
ual. And they are bound together, as Hussed might say, by a "synthesis of
identification" (cf. eM, §18), a tacit judgment or consciousness that the
individuals so given are one and the same; the synthesis is governed by Smith's
background principles of individuation, that is, by those of his background
beliefs that articulate principles of individuation fundamental to his concep-
tual scheme. Thus, the individual Billy Joe is individuated for Smith precisely
insofar as certain properties are attributed that same individual in Smith's
belief-structure. We might naturally describe the noematic structure of the
system of beliefs, then, in a Husserlian way: the Sinn of each of the beliefs
incorporates an X as sUbject-component, it is the same X that appears in each
Sinn, and the predicate-components of the various Sinne are "harmonious",
i.e., collectively consistent given the presupposed principles of individuation.
Then we may say it is that common X that appears in the Sinn of Smith's
judgment about Billy Joe.
Suppose, then, 'cf>t', ... , 'cf>n' are predicates that ascribe the various prop-
erties that serve to individuate Billy Joe for Smith, and suppose 'p(cf>t, .•• ,
cf>n)' is a complex sentence articulating Smith's principles of individuation
regarding things that satisfy 'cf>t', ... , 'cf>n'. Then we might expand our
original description (9) as follows:
(12) (3x) (Smith believes that p(cf>t, .•. , cf>n) & Smith believes that
x is cf> t & ... & Smith believes that x is cf>n & Smith judges that x
is the meanest man in town).
This form of description aptly indicates the noematic structure of the judg-
ment and the background beliefs and their connection; for quantifying-in is,
we have proposed, an appropriate means for ascribing an act whose Sinn
398 CHAPTER VIII
includes an X, and the recurrence of the variable 'x' requires that it be the
same X in the Sinne of all the beliefs and of the judgment. An appropriate
alternative to (12) is the following, since Smith has acquired the name 'Billy
Joe' for the individual intended in the beliefs described:
(13) Smith believes that P(cfJl, ••• ,cfJn) &
Smith believes that Billy Joe is cfJl & ...
Smith believes that Billy Joe is cfJn &
Smith judges that he [Billy Joe] is the meanest man in town. 20
(We use 'he' rather than 'Billy Joe' in the fmal clause only to avoid suggesting
that the name 'Billy Joe' is somehow passing through Smith's consciousness
as he judges.)
It is our central assumption that the individual Billy Joe is individuated for
Smith insofar as the indicated background beliefs are all about one and the
same individual and hence co-directed. But our analysis is not complete with
the description (12) or (13), or the Husserlian description of background
Sinne featuring a common X. For these descriptions assume that the relevant
background beliefs are themselves all defmite, sharing the same X, and we
need an account of their defmiteness. We need an account of how the X they
share entered Smith's consciousness. That is a matter of "genetic" phenom-
enology.
Perceptual acquaintance is one familiar port of entry for an X into a
person's consciousness. The acquainting sense in a perception introduces an
X (cf. Section 22). Subsequent beliefs or judgments anaphorically about the
object intended in that perception will include that X in their Sinne. It is
natural to assume Smith has seen Billy Joe on many occasions and long ago
learned to recognize him on sight. Each new perception of Billy Joe and each
act or attitude intending Billy Joe now includes an X tied anaphorically to
prior perceptions of Billy Joe that are now blurred together by the haze of
time.
More generally, we might speculate that an X enters a person's noematic
repertoire only by introduction in some indexical intention, that is, an inten-
tion that presents an individual as being in some contextual relation to the
subject. Perception presents an object as immediately before the subject and
perhaps as causally affecting his senses (cf. Part 2). Thinking of someone "by
name" - as Billy Joe or as William of Sherwood - may call on the thinker's
historical relation to the intended individual in a less immediate way. Here
we may assume a version of the causal or historical-chain theory of names.
Let it be a version, however, on which a speaker's use of a name is based on
DE RE INTENTION IN A HUSSERLIAN FRAMEWORK 399
broke into the W. B. Smythe house. The X starts its life in Sherlock's con-
sciousness bearing the character of indefmiteness. But as Sherlock acquires
enough knowledge of the right sort about that someone, "X", the X acquires
the character of defmiteness, as Sherlock in effect thinks, "That's the culprit".
Now we face a problem, especially in a case like Sherlock's "leap" to
individuative consciousness. It seems this leap is a leap of faith rather than
logic, for what in his phenomenological repertoire guarantees that his inten-
tion grasps a "defmite" individual, the same in each world compatible with
his individuative background belief-system? Generally, how can individuative
intention be rigid, directed toward the same individual in every world com-
patible with the act's content or background content? The problem sterns
from the elasticity of knowing-who in response to the transcendence of
natural individuals' identities.
The X in the Sinn of Sherlock's judgment was introduced into his phe-
nomenological repertoire with his de dicta judgment that "someone" broke
into the Smythe house. The X then accumulated about itself predications in
further judgments that would collectively constitute a sense of the identity
of the burglar ("the person who ..."). Not a complete sense of his identity,
but a sense full enough for Sherlock's investigative purposes. Let us grant
that this sense, associated with the X, confers a character of defmiteness on
Sherlock's contemplations about the burglar. But how can the X or the de-
scriptive sense of identity prescribe the same individual in all worlds? If the
X were introduced into his system in a perceptual acquaintance with the
burglar, then it would retain the rigidity of the perceptual acquaintance. But
since the X was introduced with a de dicta judgment, it inherits no rigidity
from that judgment. How then can it acquire rigidity as it picks up the ele-
ments of a sense of the identity of the intended? A likely answer lies with the
concrete view of individuation through time and across worlds that we found
in Husserl: necessary to the individuation of a natural object is its path
through time, through the history of the actual world or through diverging
possible histories in various possible worlds that coincide up to a point of
time. (Cf. Sections 3.3 and 3.4 above.) It is likely then that any Significant
sense of the identity of a natural individual will include some significant
"historic" property of the individual, e.g., that "he" (the burglar) actually
lived on Chatham Lane until one week ago today. A description like 'the man
who actually lived at Number 3, Chatham Lane, London, until one week ago
today' is somewhat like the quasi-indexical deSCription 'the man who is
actually before me now' (cf. Section 2.2 above). This sort of sense would
prescribe the same obejct in any world, so Sherlock's X could acquire rigidity
DE RE INTENTION IN A HUSSERLIAN FRAMEWORK 401
NOTES
* This chapter was written by David Woodruff Smith. The version appearing here
incorporates important emendations and revisions originating in commentary and
criticism by Ronald Mcintyre.
1 This last kind of aboutness might receive a "causal" or "historica1-chain" analysis,
artalogous to recent causal or historica1-chain theories of reference for proper names
(e.g., those proposed by Donnellan and Kripke; see Note 26, Ch. IV above). However,
we have in mind a clearly intentionalized version. Aboutness, in this one important case
at any rate, is determined by the subject's knowledge of the individual's having been
perceived. However, what makes the judgment about that individual is not the chain of
events leading from the occasion of its being perceived to the occasion of the judgment
(say, someone who saw the individual was illterviewed by a newspaper reporter, the
reporter wrote a story, and eventually our subject read the story and then judged about
the individual). Rather, the aboutness is ftxed by the subject's awareness (however
imperfect) that there is such a chain. Importantly, then, we take aboutness to be an
intentional relation between the subject or the act of judgment and the object judged
about; it is not a physical or genetic relation between the act of judgment and the object
judged about. Here we echo a point made by F¢Uesdal in 'Reference and Meaning', a
402 CHAPTER VIII
object that is actually 1/>', assuming each to be uttered by the same person (say, Smith)
on the same occasion in a given world woo And, indeed, in Wo - the "home" world in
which the utterance takes place - both descriptions have the same referent, viz., the
object (if any) in Wo that is uniquely I/> in woo But let us consider their referents in an
alternative possible world, w. 'The object that is 1/>', as uttered by Smith in Wo on the
given occasion, would refer in w to that object (if any) in w that is uniquely I/> in w,
which mayor may not be the same object that is I/> in WOo 'The object that is 1/>', then,
may refer to different objects in different worlds and so is non-rigid. By contrast, the
description 'the object that is actually 1/>' is rigid, as we propose it be understood. On
our reading this description, as uttered by Smith in w o , would refer in w to that object
in w that is uniquely I/> in Wo - not to whatever in w is uniquely I/> in w, as the former
description would. Here the same object is required to exist in both wand wO ' and the
description, so uttered in wo ' takes the same referent in any world w. As we understand
it, then, the description 'the object that is actually 1/>' is not purely a defmite description:
the term 'actually' is somewhat like an indexical term, in that it always brings us back to
the "home" world whenever we evaluate the reference, in any world, of the description
as uttered in the home world. (However, we do not want to hold that actuality is merely
relative to a world in which the description is uttered; the actual world should be actual
in some absolute sense. But these are further matters we carmot go into here. Cf. David
K. Lewis, 'Anselm and Actuality', NoW; 4 (1970), 175 ·88; and Robert Merrihew
Adams, 'Theories of Actuality', Nous 8 (1974), 211-31.) And so the sense of actuality
makes a significant difference in the two proposals for explicating the acquainting
sense in perception. On both proposals, this acquainting sense is a sense that might be
expressed by 'this'. But on the fust proposal 'this' is understood as functioning like
'the object before me', while on our preferred proposal it is understood as functioning
like 'the object actually before me'. Thus, on the Hintikka-type approach perceptual
acquaintance is a non-rigid intention, whereas on our approach it is rigid. Notice that on
both accounts the perception is required to occur in the various worlds wherein the
"referent" is determined; but on our account it is the occurrence in the "home" world
that is relevant, for the object is intended as before the subject on the occasion of the
perception in that world.
13 Cf. Roderick M. Chisholm, 'Identity Through Possible Worlds', Nous 1 (1967),
1-8; and David Lewis, 'Counterpart Theory and Quantified Modal Logic', Journal of
Philosophy 65 (1968),113-26.
14 R. M. Chisholm has defmed the notion of haecceity in the present way, ascribing to
Aquinas a view of individuation based on the notion. See his 'Individuation: Some
Thomistic Questions and Answers', Grazer Philosophische Studien 1 (1975), 25·-41.
Robert Merrihew Adams has also defmed haecceity in this way and has amply demon-
strated the notion's utility in a wide-ranging discussion that is congenial in many respects
to the views on individuation we develop. See his 'Primitive Thisness and Primitive
Identity', Journal of Philosophy 76 (1979), 5-26. The terminology of "haecceitism"
is developed there, and Adams adopts the position we have taken, dubbing it "moderate
haecceitism" .
15 Cf. Benson Mates, 'Individuals and Modality in the Philosophy of Leibniz' (mimeo-
graphed; presented as a Mahlon Powell Lecture at Indiana University, 1970); and 'Leibniz
on Possible Worlds', in Logic, Methodology, and Philo!ophy of Science, III, ed. by B.
Van Rootse1aar and J. F. Staal (North-Holland, Amsterdam, 1968).
404 CHAPTER VIII
16 Thus, we might defme the identity of an individual x either as the set of ordered
pairs <w, p) where w is any world and p is the complex property that includes all the
(appropriate) properties x has in w, or as the conjunctive property comprising all of x's
world-relativized properties of being thus-andojjo in a world.
17 For an interesting and detailed study of knowing-who, see Stephen E. Boer and
William G. Lycan, 'Knowing Who', Philosophical Studies 28 (1975), 299-344. Though
there are differences between their approach and our approach here, there is some
considerable common ground. See also Hintikka, Knowledge and Belief (Note 2, Ch. VlI
above), pp.131-32, 148ff.
18 Hintikka has proposed that (i) 's knows who a is' can be analyzed, or translated into
logical symbolism, as (il) '(3x) (s knows that x = a)' (where the variable 'x' ranges over
persons): cf. Knowledge and Belief (Note 2, Ch. VII above), pp. 131-32. These two
forms do seem to be equivalent, provided the quantifying-in is tied to "individuative"
defmiteness. However, (i) - or, presumably (ii) - does not itself ascribe a specific item
of knowledge but rather asserts that s has some system of specific items of knowledge
that constitute s's knowing who a is. Our concern is with the further analysis of knowing-
who in terms of the specific items of knowledge of which it consists.
19 Classical logic for propositional-attitude sentences - such as we have studied in prior
chapters - cannot provide an explication of ordinary English sentences with the form of
(A) Smith believes that the ~ is >/I, and Smith judges that he is x.
or
(B) Smith believes that someone is >/I, and
Smith judges that he is x.
The pronoun 'he' is obviously not a pronoun of laziness, replaceable by its underscored
antecedent; for that substitution would change the phenomenological structure ascribed
to Smith. Nor can the pronoun be rendered a variable of quantification in the familiar
syntax. If the quantifier binding the variable were outside the belief-operator of the flISt
clause, as in
(A') (3 !x) (</lX & Smith believes th.at >/Ix & Smith judges that xx)
(B') (3x) (Smith believes that >/Ix & Smith judges that xx),
then both the belief and the judgment ascribed would in each case be de reo But neither
the belief nor the judgment ascribed in either (A) or (B) is de re, as we understand (A)
and (B). The syntax we would seem to require in (A) (if 'the ~' is eliminated in terms of
quantification) and in (B) places the quantifier inside the scope of the belief-operator
and the variable bound to that quantifier inside the scope of the judgment-operator,
thus:
(A") Smith believes that [(3 !x) (~x & >/Ix]
& Smith judges that [xx])
(B") Smith believes that [(3x) (>/Ix] &
Smith judges that [xx] ).
But these sentences are not well-formed in classical logic: "quantifying out", as David
Kaplan has termed it, is not allowed. And again the judgment-clause is in each case de re,
where (A) and (B) do not ascribe de re judgments.
DE RE INTENTION IN A HUSSERLIAN FRAMEWORK 405
In our discussion in the text, we assume that sentences of the forms of (A) and (B)
are meaningful English. We assume that both the ftrst and second clauses ascribe de
dicta attitudes or acts. That is in fact our ordinary understanding of such sentences, as
becomes evident on considering the following sequence of forms of English sentences:
(B) Smith believes that someone is 'ii, and
Smith judges that he is x;
(C) Smith said that someone is 'ii, and then
Smith said that he is x;
(D) Smith said "Someone is 'ii", and then
Smith said "He is x";
(E) Smith said "Someone is 'ii, and he is x".
Without going into the matter here, it should be clear that if (E) ascribes Smith an
indefmite, or de dicta, intention in saying "he", surely (D) does too. If so, surely in the
second clause (C) ascribes Smith a de dicta assertion. And if so, surely (B) in the second
clause ascribes a de dicta judgment. Sometimes, then, an anaphoric or relative pronoun
in a propositional-attitude context does not require a de re attitude, as does a proper
variable of quantifying-in. This fact must be reflected in a complete logic for proposi-
tional-attitude sentences.
The logical problem of (A) or (B) is essentially the same as that which Peter Geach
exposed for:
Hob thinks a witch has blighted Bob's mare, and Nob wonders whether she
(the same witch) killed Cob's sow.
Cf. P. T. Geach, Logic Matters (Blackwell, Oxford, 1972), pp. 147ff. Although the
Geachian sentence involves two thinkers where our (B)-(E) involve just one, a signiftcant
complication, still such Hob-Nobbing with witches is perfectly sensible. We cannot
pursue the issues here. Suffice it to say that classical possible-worlds semantics (cf.
Chapter VII) cannot interpret the sentences in question. For an insightful and extensive
study of the semantica1 issues involved and proposals for enhanced syntactic and seman-
tic machinery to deal with the problems, see Esa Saarinen, 'Intentional Identity Inter-
preted: A Case Study of the Relations Among Quantifters, Pronouns, and Propositional
Attitudes', Linguistics and Philosophy 2 (1978), 151-223. Thanks are due Professor
Saarinen for helpful discussions on the problems posed by Geach's puzzle.
20 Using the name 'Billy Joe' in (13) offers a natural and illuminating description of
Billy Joe's individuation in Smith's belief system. For a main function of the use of
proper names is to keep track of one and the same individual through changes in its
description, that is, through different assertions about it. (F~llesdal has stressed this
point about proper names in 'Reference and Meaning' (Note 1 above).) Thus, Smith
might give expression to the various beliefs we have attributed him by saing: 'Billy Joe is
1/>1" ••• , 'Billy Joe is I/>n'. The sequence of assertions he thus makes, all being about the
same individual by virtue of his recurrent use of the name 'Billy Joe\ then traces out
the individuation of the referent of 'Billy Joe' (that is, of the person he takes to be its
referent) in his belief system. And since Smith himself might express these beliefs in that
way, we may naturally describe them by the fragment of (13), 'Smith believes that Billy
Joe is 1/>1 & ... & Smith believes that Billy Joe is I/>n'.
21 See Note 1 above.
BIBLIOGRAPHY*
Included here are most of the works .cited in the text and a few additional items of
special relevance to our main discussion. Works representative of the kind of Husserl
interpretation we have espoused are marked with '*'.
[1) Robert Merrihew Adams, 'Primitive Thisness and Primitive Identity', Journal of
Philosophy 76 (1979),5-26.
(2) Robert Merrihew Adams, 'Theories of Actuality', Nous 8 (1974), 211-31.
(3) Karl Ameriks, 'Husserl's Realism', Philosophical Review 86 (1977),498-519.
[4) G. E. M. Anscombe,/ntention, Blackwell, Oxford, 1957.
(5) G. E. M. Anscombe, 'The Intentionality of Sensation: A Grammatical Feature',
in R. J. Butler (ed.), Analytical Philosophy, 2nd series, Blackwell, Oxford,
1965, pp. 158-80.
(6) Richard E. Aquila, 'Husserl and Frege on Meaning', Journal of the History of
Philosophy 12 (1974), 377-83.
(7) Richard E. Aquila, Intentionality: A Study of Mental Acts, Pennsylvania State
University Press, University Park, 1977.
[8) Stephen E. Boer and William G. Lycan, 'Knowing Who', Philosophical Studies
28 (1975), 299-344.
(9) Bernard Bo1zano, Theory of Science (ed. and trans!. by Rolf George), University
of California Press, Berkeley and Los Angeles, 1972.
(10) Franz Brentano, Psychology from an Empirical Standpoint (ed. by linda L.
McAlister, trans!. by Antos C. Rancurello, D. B. Terrell, and Linda L. McAlister),
Humanities Press, New York, 1973.
[11) Franz Brentano, The True and the Evident (trans!. by Roderick M. Chisholm,
Use Politzer, and Kurt R. Fischer, ed. by Roderick M. Chisholm), Humanities
Press, New York, 1966.
(12) Rudolf Carnap, The Logical Structure of the World (trans!. by Rolf George),
University of California Press, Berkeley, 1969.
(13) Rudolf Carnap, Meaning and Necessity, 2nd ed., The University of Chicago
Press, Chicago, 1956.
(14) Rudolf Carnap, 'Meaning and Synonymy in Natural Languages', Philosophical
Studies 6 (1955). Reprinted in [l3), pp. 233-47.
[l5) David Carr, 'Intentionality', in Edo Pivcevic (ed.), Phenomenology and Philoso-
phical Understanding, Cambridge University Press, Cambridge, 1975, pp. 17-
36.
(16) Hector-Neri Castaneda, 'Consciousness and Behavior: Their Basic Connections',
in Castaneda (ed.), Intentionality, Minds, and Perception, Wayne State Univer-
sity Press, Detroit, 1967, pp. 121-58.
[37] Daniel C. Dennett, Content and Consciousness, Routledge & Kegan Paul,
London, 1969.
[38] Keith Donnellan, 'Proper Names and Identifying Descriptions', in [35], pp.
356-79.
[39] Keith Donnellan, 'Reference and Definite Descriptions', Philosophical Review
75 (1966), 281-304. Reprinted in [145], pp. 42-65.
*[40] Hubert L. Dreyfus (ed.), Husserl, Intentionality, and Cognitive Science, The
M.l.T. Press/Bradford Books, Cambridge, Mass., 1982 forthcoming.
*(41) Hubert L. Dreyfus, 'Husserl's Phenomenology of Perception: From Transcen-
dental to Existential Phenomenology', Ph.D. dissertation, Harvard University,
1963.
*[42] Hubert L. Dreyfus, 'The Perceptual Noema: Gurwitsch's Crucial Contribution',
in Lester Embree (ed.), Life- World and Consciousness, Northwestern University
Press, Evanston, ill., 1972, pp. 135-70. Revised and reprinted as 'Husserl's
Perceptual Noema', in [40].
*[43] Hubert L. Dreyfus, 'Sinn and Intentional Object', in Robert C. Solomon (ed.),
Phenomenology and Existentialism, Harper & Row, New York, 1972, pp. 196-
210.
[44] Michael Dummett, Frege: Philosophy of Language, Harper & Row, New York,
1973.
[45] J. N. Findlay, Meinong's Theory of Objects and Values, 2nd ed., Oxford Univer-
sity Press, Oxford, 1963.
[46] Jerry A. Fodor, Representations: Philosophical Essays on the Foundations of
Cognitive Science, The M.l.T. Press/Bradford Books, Cambridge, Mass., 1981.
* [47] Dagfmn F¢llesdal, 'Brentano and Husserl on Intentional Objects and Perception',
in R. M. Chisholm (ed.), Die Philosophie Franz Brentanos, Rodolpi, Amsterdam,
1978, pp. 83-94. Reprinted in [40].
*[48] Dagfinn F¢lllesdal, 'Husserl and Heidegger on the Role of Actions in the Consti-
tution of the World', in Esa Saarinen et al. (eds.), Essays in Honour of Jaakko
Hintikka, D. Reidel, Dordrecht, 1979, pp. 365-78.
*[49] Dagfmn F¢llesdal, Husserl und Frege, l. Kommisjon Hos H. Aschehong & Co.,
W. Nygaard, Oslo, 1958.
*[50] Dagfmn F¢llesdal, 'Husserl's Notion of Noema', Journal of Philosophy 66
(1969), 680-87. Reprinted in Robert C. Solomon (ed.), Phenomenology and
Existentialism, Harper & Row, New York, 1972; and in [40].
* [51] Dagfmn F¢llesdal, Husserl's Phenomenology, forthcoming.
*[52] Dagfmn F¢llesdal, 'An Introduction to Phenomenology for Analytic Philoso-
phers', in Raymond E. Olson and Anthony M. Paul (eds.), Contemporary Phi-
losophy in Scandinavia, The Johns Hopkins Press, Baltimore, 1972, pp. 417-29.
[53] Dagfmn F¢llesdal, 'Knowledge, Identity, and Existence', Theoria 33 (1967),
1-27.
*[54] Dagfinn F¢llesdal, 'Phenomenology', in Edward C. Carterette and Morton P.
Friedman (eds.), Handbook of Perception, I. Academic Press, Inc., New York,
1974, pp. 377-86. Reprinted as 'Husserl's Theory of Perception', in [40].
[55] Dagfinn F¢llesdal, 'Quine on Modality', Synthese 19 (1968), 147-57. Re-
printed in Donald Davidson and Jaakko Hintikka (eds.), Words and Objections,
D. Reidel, Dordrecht, 1975.
410 BIBLIOGRAPHY
[56] Gottlob Frege, The Basic Laws of Arithmetic (ed. and transl. by Montgomery
Furth), University of California Press, Berkeley, 1967.
[57] Gottlob Frege, 'Review of Dr. E. Husserl's Philosophy of Arithmetic' (transl. by
E. W. Kluge), Mind 81 (1972), 321-37. Extracts also translated in [59], pp.
79-85.
[58] Gottlob Frege, 'The Thought: A Logical Inquiry' (transl. by A. M. and Marcelle
Quinton), Mind 65 (1956), 289-311. Reprinted in P. F. Strawson (ed.), Philo·
sophical Logic, Oxford University Press, Oxford, 1967, pp. 17-38.
[59] , Gottlob Frege, (Translations from) the Philosophical Writings of Gottlob
Frege (ed. by Peter Geach and Max Black), Blackwell, Oxford, 1966. See esp.
'Begriffsschrift (Chapter I)', pp. 1-20; 'Function and Concept', pp. 21-41;
'On Concept and Object', pp. 42-55; 'On Sense and Reference', pp. 56-78;
and 'Review of Husserl's Philosophie der A rithmetik' (extracts), pp. 79-85.
[60] Gottlob Frege, 'Uber Sinn und Bedeutung', in Frege, Funktion, Begrift.
Bedeutung: Funf logische Studien (ed. by Giinther Patzig), Vandenhoeck &
Ruprecht, Gottingen, 1962, pp. 40-65. English translations in [59] and in
[104].
[61] Montgomery Furth, 'Introduction', in [56].
[62] P. T. Geach, Logic Matters, University of California Press, Berkeley,1972.
[63] P. T. Geach, Mental Acts, Routledge & Kegan Paul, London, 1957.
[64] Nelson Goodman, The Structure of Appearance, Harvard University Press,
Cambridge, Mass., 1951.
[65] Aron Gurwitsch, The Field of Consciousness, Duquesne University Press, Pitts-
burgh, 1964.
(66) Aron Gurwitsch, 'Husserl's Theory of the Intentionality of Consciousness in
Historical Perspective', in Edward N. Lee and Maurice Mandelbaum (eds.),
Phenomenology and Existentialism, The Johns Hopkins Press, Baltimore, 1967,
pp. 25-57. Reprinted in [40].
[67] Aron Gurwitsch, 'On the Intentionality of Consciousness', in Marvin Farber
(ed.), Philosophical Essays in Memory of Edmund Husserl, Harvard University
Press, Cambridge, Mass., 1940, pp. 65-83. Reprinted in Joseph J. Kockelmans
(ed.), Phenomenology, Doubleday, Garden City, N.Y., 1967, pp. 118-37.
[68] Aron Gurwitsch, Studies in Phenomenology and Psychology, Northwestern
University Press, Evanston, Ill., 1966.
*[69] Harrison Hall, 'Intersubjective Phenomenology and Husserl's Cartesianism', Man
and World 12 (1979), 13-20.
*[70] Harrison Hall, 'The Philosophical Significance of Husserl's Theory of Inten-
tionality', Journal of the British Society for Phenomenology 13 (1982), 79-84.
*[71] Harrison Hall, 'Was Husserl a Realist or an Idealist?', in [40].
[72] John Haugeland (ed.), Mind Design, The M.I.T. Press/Bradford Books, Cam-
bridge, Mass., 1981.
[73] Martin Heidegger, Being and Time (transl. by John Macquarrie and Edward
Robinson), Harper & Row, New York, 1962.
[74] Martin Heidegger, Fundamental Problems of Phenomenology (transl. by Albert
Hofstadter), Indiana University Press, Bloomington, 1982.
[75] Jaakko Hintikka, 'Individuals, Possible Worlds, and Episternic Logic', Noris 1
(1967), 33-62.
BIBLIOGRAPHY 411
[76) Jaakko Hintikka, The Intentions of Intentionality and Other New Models for
Modalities, D. Reidel, Dordrecht, 1975. See esp. 'The Semantics of Modal
Notions and the Indeterminacy of Ontology', pp. 26-42 (reprinted in [35));
'Objects of Knowledge and Belief: Acquaintances and Public Figures', pp.43-
58; 'Carnap's Heritage in Logical Semantics', pp. 76-101; 'Quine on Quantify-
ing In: A Dialogue', pp. 102-36; and 'The Intentions of Intentionality', pp.
192-222.
[77) Jaakko Hintikka, Knowledge and Belief, Cornell University Press, Ithaca, N.Y.,
1962.
[78) Jaakko Hintikka, 'Modality as Referential Multiplicity', Ajatus 20 (1957),
49-64.
[79) Jaakko Hintikka, Models for Modalities, D. Reidel, Dordrecht, 1969. See esp.
'Semantics for Propositional Attitudes', pp. 87-111 (reprinted in [101) and
in [104)); and 'On the Logic of Perception', pp. 151-83.
[80) Edmund Husser! [for main citations, see pp. ix-x above).
[81) Edmund Husser!, 'Besprechung von K. Twardowski, Zur Lehre vom Inhalt und
Gegenstand der Vorstellungen, in Husser!, Aufsiitze und Rezensionen (1890-
1910) (ed. by Bernhard Rang), Husserliana XXII, Nijhoff, The Hague, 1979, pp.
349-56.
[82) Edmund Husser!, 'Intentionale Gegenstiinde', in Husser!, Aufsiitze und
Rezensionen (1890-1910) (ed. by Bernhard Rang), Husserliana XXII, Nijhoff,
The Hague, 1979, pp. 303-48.
[83) Edmund Husser!, Philosophie der Arithmetik (ed. by Lothar Eley), Husserliana
XII, Nijhoff, The Hague, 1970. [Main text fust published in 1891.)
[84) Roman Ingarden, On the Motives which Led Husserl to Transcendental Idealism
(trans!. by Arnor Hanniba!sson), Nijhoff, The Hague, 1975.
[85) David Kaplan, 'Dthat', in Peter A. French et al. (eds.), Contemporary Perspec·
tives in the Philosophy of Language, University of Minnesota Press, Minneapolis,
1979,pp.383-400.
[86] David Kaplan, 'On the Logic of Demonstratives', in Peter A. French et al.
(eds.), Contemporary Perspectives in the Philosophy of Language, University
of Minnesota Press, Minneapolis, 1979, pp. 401-12.
(87) David Kaplan, 'Quantifying In', Synthese 19 (1968), 178-214. Reprinted in
Donald Davidson and Jaakko Hintikka (eds.), Words and Objections, D. Reidel,
Dordrecht,1975;andin [101].
[88] Iso Kern, Husserl und Kant, Nijhoff, The Hague, 1964.
[89) William Kneale, 'Modality de Dicto and de Re" in Ernest Nagel et al. (eds.),
Logic, Methodology, and Philosophy of Science, Stanford University Press,
Stanford, Cal., 1962, pp. 622-33.
(90) William Kneale and A. N. Prior, Symposium on 'Intentionality and Intensional-
ity', Proceedings of the Aristotelian Society, supp. vol. 42 (1968), 73-106.
(91) Saul Kripke, ~Identity and Necessity', in Milton K. Munitz (ed.), Identity and
Individuation, New York University Press, New York, pp. 135-64. Reprinted
in (145), pp. 66-101.
(92) Saul Kripke, 'Naming and Necessity', in [35), pp. 253-355,763-69.
(93) Saul Kripke, 'Semantical Considerations on Modal Logic', Acta Philosophica
Fennica 16 (1963), 83-94. Reprinted in (101).
412 BIBLIOGRAPHY
*(94) Gl,lido Kiing, 'Husserl on Pictures and Intentional Objects', Review of Meta-
physics 26 (1973),670-80_
* (95) John Francis Lad, 'On Intuition, Evidence, and Unique Representation', Ph.D.
dissertation, Stanford University, 1973.
(96) David M. Levin, Reason and Evidence in Husserl's Phenomenology, North-
western University Press, Eval~ton, Ill., 1970.
(97) C. I. Lewis, 'The Modes of Meaning', Philosophy and Phenomenological Re-
search 4 (1944), 236-49. Reprinted in (102).
(98) David K. Lewis, 'Anselm and Actuality', Nous 4 (1970), 175-88.
(99) David K. Lewis, 'Counterpart Theory and Quantified Modal Logic', Journal of
Philosophy 65 (1968), 113-26.
(100) David K. Lewis, 'General Semantics', Synthese 22 (1970), 18-67. Reprinted
in (35).
(101) Leonard Linsky (ed.), Reference and Modality, Oxford University Press, Lon-
don, 1971.
(102) Leonard Linsky (ed.), Semantics and the Philosophy of Language, The Univer-
sity of Illinois Press, Urbana, 1952.
(103) J. L. Mackie, 'Problems of Intentionality', in Edo Pivcevic (ed.), Phenomenology
and Philosophical Understanding, Cambridge University Press, Cambridge, 1975.
(104) Ausonio Marras (ed.), Intentionality, Mind, and Language, The University of
Illinois Press, Urbana, 1972.
[105) Benson Mates, 'Leibniz on Possible Worlds', in B. van Rootselaar and J. F. Staal
(eds.), Logic, Methodology, and Philosophy of Science, III, North-Holland,
Amsterdam, 1968.
(106) Linda L. McAlister, 'Chisholm and Brentano on Intentionality', Review of
.Metaphysics 28 (1974), 328-38.
*[ 107) Ronald McIntyre, 'Husserl and Referentiality: The Role of the Noema as an
Intensional Entity', Ph.D. dissertation, Stanford University, 1970.
*[ 108) Ronald McIntyre, 'Husserl's Phenomenological Conception of Intentionality
and its Difficulties', Philosophia, forthcoming.
*[ 109) Ronald McIntyre, 'Intending and Referring', in (40).
*[ 110) Ronald McIntyre and David Woodruff Smith, 'Husserl's Identification of Mean-
ing and Noema', The Monist S9 (1975), 111-32. Reprinted in part in (40).
[Ill) Alexius Meinong, 'The Theory of Objects' (trans!. by Isaac Levi, D. B. Terrell,
and Roderick M. Chisholm), in (29) , pp. 76.,..117.
(112) Maurice Merleau-Ponty, Phenomenology of Perception (trans!. by Colin Smith),
Routledge & Kegan Paul, London, 1962.
*(113) Izchak Miller, 'Husserl's Account of Our Temporal Awareness', in (40).
*[ 114) Izchak Miller, Husserl's Theory of Perception, The M.I.T. Press, Cambridge,
Mass., forthcoming.
*[ 115) Izchak Miller, 'The Phenomenology of Perception: Husserl's Account of Our
Temporal Awareness', Ph.D. dissertation, University of California at Los Angeles,
1979.
*(116) J. N. Mohanty, The Concept of Intentionality, Warren H. Green, St. Louis,
1972.
(117) J. N. Mohanty, Edmund Husserl's Theory of Meaning, 3rd ed., Nijhoff, The
Hague, 1976.
BIBLIOGRAPHY 413
(139) Bertrand Russell, 'On Denoting', Mind 14 (1905). Reprinted in Russell, Logic
and Knowledge (ed. by Robert C. Marsh), G. P. Putnam's Sons/Capricorn Books,
New York, 1971, pp. 39-56; in Irving M. Copi and James A. Gould (eds.),
Contemporary Readings in Logical Theory, Macmillan, New York, 1967; and
in (104).
(140) Bertrand Russell, 'On Propositions: What They Are and What They Mean', in
Russell, Logic and Knowledge (ed. by Robert C. Marsh), G. P. Putnam's Sons/
Capricorn Books, New York, 1971, pp. 285-120.
(141) Esa Saarinen, 'Intentional Identity Interpreted: A Case Study of the Relations
Among Quantifiers, Pronouns, and Propositional Attitudes', Linguistics and
Philosophy 2 (1978),151-223.
(142) Jean-Paul Sartre, Being and Nothingness (trans!. by Hazel E. Barnes), Philosophi-
cal Library, New York, 1956.
(143) Jean-Paul Sartre, The Transcendence of the Ego (trans!. by Forrest Williams and
Robert Kirkpatrick), Noonday Press, New York, 1957.
(144) Paul Arthur Schilpp (ed.), The Philosophy of Rudolf Carnap, Open Court,
LaSalle, Ill., 1963.
(145) Stephen P. Schwartz (ed.), Naming, Necessity, and Natural Kinds, Cornell
University Press, Ithaca, N.Y., 1977.
(146) John R. Searle, Intentionality, Cambridge University Press, Cambridge, 1983
forthcoming.
(147) John R. Searle, 'What Is an Intentional State?', Mind 88 (1979), 74-92. Re-
printed in (40) .
(148) Wilfrid Sellars, Science and Metaphysics: Variations on Kantian Themes,
Humanities Press, New York, 1968.
(149) Wilfrid Sellars, Science, Perception, and Reality, Routledge & Kegan Paul,
London, 1963.
*(150) Barry Smith, 'An Essay in Formal Ontology', Grazer Philosophische Studien 6
(1978),39-62.
*( lSI) Barry Smith, 'Frege and Husser!: The Ontology of Reference', Journal of the
British Society for Phenomenology 9 (1978),111-25.
*( 152) Barry Smith, 'The Ontology of Reference: Studies in Logic and Phenomenology',
Dissertation, University of Manchester, 1976.
(153) David Woodruff Smith, Acquaintance, forthcoming.
(154) David Woodruff Smith, 'The Case of the Exploding Perception', Synthese 41
(1979),239-69.
(155) David Woodruff Smith, 'Content and Context of Perception', Synthese, forth-
coming.
*(156) David Woodruff Smith, 'Husser! on Demonstrative Reference and Perception',
in (40).
(157) David Woodruff Smith, 'Indexical Sense and Reference', Synthese 49 (1981),
101-28.
*(158) David Woodruff Smith, 'Intentionality, Noemata, and Individuation: The Role
of Individuation in Husser!'s Theory of Intentionality', Ph.D. dissertation,
Stanford University, 1970.
( 159) David Woodruff Smith, 'Meinongian Objects', Grazer Philosophische Studien 1
(1975),43-71.
BIBLIOGRAPHY 415
[160] David Woodruff Smith, 'What's the Meaning of "This"?', Noils (1982), forth-
coming.
* [161] David Woodruff Smith and Ronald McIntyre, 'Intentionality via Intensions',
Journal of Philosophy 68 (1971), 541-61.
[162] Robert Sokolowski, The Formation of Husserl's Concept of Constitution,
Nijhoff, The Hague, 1970.
[163] Robert Sokolowski, Husserlian Mediations, Northwestern University Press,
Evanston, Ill., 1974.
[164] Robert Stalnaker, 'Pragmatics', Synthese 22 (1970), 272-89. Reprinted in
[35].
*[165] Michael Sukale, Comparative Studies in Phenomenology, Nijhoff, The Hague,
1976.
[166] Alfred Tarski, 'The Concept of Truth in Formalized Languages', in Tarski,
Logic, Semantics, Mathematics, Clarendon Press, Oxford, 1956, pp. 152-278.
[167] Alfred Tarski, 'The Semantic Conception of Truth and the Foundations of
Semantics', Philosophy and Phenomenological Research 4 (1944), 341- 75.
Reprinted in [102].
[168] Charles Taylor, The Explanation of Behavior, Routledge & Kegan Paul, London,
1964.
[169] James E. Tomberlin, 'Identity, Intensionality, and Intentionality', Syn these ,
forthcoming.
*[170] Robert S. Tragesser, Phenomenology and Logic, Cornell University Press, Ithaca
and London, 1977.
[171] Kasimir Twardowski, On the Content and Object of Presentations (trans\. by
Reinhardt Grossmann), Nijhoff, The Hague, 1977.
[172] J. O. Urmson and 1. Jonathan Cohen, 'Criteria of Intensionality' Proceedings of
the Aristotelian Society, supp. vol. 42 (1968), 107-42.
[173] Zeno Vendler, 'Thinking of Individuals', Noils 10 (1976), 35-46.
[174] Georg H. von Wright, An Essay in Modal Logic, North-Holland, Amsterdam,
1951.
[175] Morton G. White, 'The Analytic and the Synthetic: An Untenable Dualism',
in Sidney Hook (ed.), John Dewey: Philosopher of Science and Freedom, Dial
Press, New York, 1950. Reprinted in [102], pp. 272-86.
[176] Dallas Willard, 'The Paradox of Logical Psychologism: Husserl's Way Out',
American Philosophical Quarterly 9 (1972),94-99.
INDEX OF NAMES*
Adams, Robert Merrihew, 403 52, 154-56, 211, 222, 224, 353,402,
Anscombe, G. E. M., 44, 60, 84 405
Aquila, Richard E., 152 Foot, Philippa, 36
Aquinas, Thomas, 403 Frege, Gottlob, xvi, xvii, 35,40,45,61-
Aristotle, 13 7 82, 85-86, 88, 107, 118-19, 124,
Armstrong, D. M., 35, 38 151,153-56,160,170-78,192,205-
206,213, 223, 266-67,276-79,281,
Berkeley, George, 104 285-89,306, 308,317-23,327-31,
Boer, Stephen E., 404 334, 336, 346, 352, 354
Bolzano, Bernard, xvii, 108, 118-19, 151, Freud, Sigmund, 98
156, 16~ 172-73, 19~ 223 Furth, Montgomery, 85, 306
Brentano, Franz, xvii, 2, 10, 24, 35-36,
40, 47-54, 57-61, 80, 82-83, 89, Geach, Peter T., 36,405
109-110, 161 Goodman, Nelson, 265
Grossmann, Reinhardt, 223
Carnap, Rudolf, xvii, 21, 37-38,45,64, Gurwitsch, Aron, 154, 157-67, 222-23,
66-67, 69, 85-86, 169-70, 175, 265
265-71,276,278-87,289,291-92,
305-306, 308, 313, 316-17, 321, Heidegger, Martin, 264-65
329 Hintikka, laakko, xvii, 35-38, 276, 291,
Carr, David, 35, 39 306, 322-29, 331-34, 338-41,343,
Castaneda, Hector-Neri, 84, 223, 305,402 345, 349, 351-53,360,366-68,402-
Chisholm, Roderick M., 13, 36-38, 52, 404
58-59,78,82-84,403 Hume, David, 100
Church, Alonzo, 80, 85, 317
Oark, Romane, 38, 366,402 Ingarden, Roman, 152
Cohen, L. Jonathan, 38
Kant, Immanuel, xiv, 101, 103-104, 235,
Davidson, Donald, 318, 351 277,288,298,304,385
Dennett, Daniel C., 35 Kaplan, David, 38, 75, 86, 224-25, 306,
Descartes, Rene, xiv, xv, 94-98,103-104 352,367,402,404
Donnellan, Keith, 36, 207, 220, 224-25, Kerry, Benno, 163
401 Kneale, William, 39
Dreyfus, Hubert, 222-23, 226 Kraus, Oskar, 35, 82
Dummett, Michael, 85 Kripke, Saul, 207, 211, 224-25, 290,
306,322,324,351,401
Findlay, 1. N., 83-84 Kiing, Guido, 152, 223
F ¢lllesdal, Dagfinn, xvi, xvii, 38, 147, 151-
* We thank Eddie Yeghiayan for his assistance with the Index of Names.
417
418 INDEX OF NAMES
Lad, Joh.n Francis, 222, 224 Russell, Bertrand, 8, 36-38, 216, 290,
Landgrebe, Ludwig, 265 345,347,353
Leibniz, G. W., 299, 319, 323-24, 374,
376 Saarinen, Esa, 405
Lewis, C. I., xvii, 85, 278-83, 286-87, Sartre, Jean-Paul, 100, 151
306,353 Searle, John R., 225
Lewis, David K., 306,403 Smith, David Woodruff, 84, 222,224-25,
Lycan, William G., 404 401-402
Sokolowski, Robert, 265
Mackie, J. L., 39 Stalnaker, Robert, 306
Mates, Benson, 403
McAlister, Linda L., 82 Tarski, Alfred, 317- 20, 345, 347, 351
McIntyre, Ronald, 222, 224, 401 Taylor, Charles, 38
Meinong, Alexius, xvii, 40, 47,54-58,80, Twardowski, Kasimir, xvii, 54, 109-13,
83-84, 109, 111-12, 119, 155,160, 115,118-19,133,148,150-51,163-
166-67,170,199,315 64,170,173,223
Merleau-Ponty, Maurice, 226, 265
Miller, Izchak, 35,150-51,222.224 Urmson, J. 0., 38
Mohanty, J. N., 35, 38, 151-52, 223, 306
Montague, Richard, xvii, 306, 351-52 Van Fraassen, Bas, 353
Vendler, Zeno, 402
Parsons, Terence, 84 von Wright, Georg H., 39
Peirce, C. S., 267
Prior, Arthur N., 39,46,82 White, Morton G., 306
Putnam, Hilary, 224, 306, 402
Zimmermann, J., 110, 112
Quine, Willard Van Orman, 32, 38-39,
60,84,224,274,305-306,328,351
INDEX OF TOPICS
Aboutness: defined, 8-9; as intentional Constitution, 103, 148, 149, 231, 235,
relation, 12-14,.18-19,51-53,76- 241-43
81, 129-30,293-94,336-43, 355- Content (lnhalt) (vs. object of act): xv,
58, 387-89, 391, 394-96; and inten- 54, 87-88,92-93, 108-40, 154-56;
sionality, 30-33, 72-74, 327-28, Twardowski's theory of, 109-12;
331-32,349-50, horizon of a., 263 quality and matter as constituents of,
Acquaintance (perceptual definiteness), 113-15,133-36; real (reelle) vs. ideal
20, 212-19, 225 (n. 32), 355-56, (or intentional) c., 115-25, 135-37;
358-60,362-69,398-99 ideal c. as act-type, 116-17, 141-42;
Act (of consciousness): defined, 3-5; a. ideal c. as meaning, 112, 117-18, 121,
vs. object of consciousness, 5-6, 89- 124-25, 154-56; c. as noesis and as
90; direct-object a. vs. propositional a., noema or Sinn, 119-27,131,133-36,
6-9, 23; Husserl's analysis of a., see 142-43, 154-56, 159-65; c.-theory
Content, Hyle, Noema, Noesis; "pure" of intentionality, xv, 104-108, 141-
a., see Transcendental reduction. 50; sensory c., 115, 136-40
Adumbration (Abschattung), 139
Anaphoric reference and anaphoric inten- De dicto: modalities, 31-33, 72-75,
tion, 395-401 326-27, 336-39; intentions, see In-
Appearance, 162 definite intention
Appearing, 138-40 Definite description: in act-sentences, 16,
25; d.-<1. model of reference and inten-
Background beliefs, 220-21, 246-55, tionality, 204-13
271-75,277,332-33,388-401 Definite (de re) intention: 18-21,32-33,
Bedeutung: as linguistic meaning for 79,202,208-19,261,294,339-41,
Husserl, 171-82; as linguistic referent 344-45, 354-69, 387-401; vs. in-
for Frege, 63-66 definite (de dicto) intention, 18- 21,
Bracketing (epoche), xiv, xvi, 92, 96-101, 32-33; individuatively d. i., see In-
105,122-23,160,169,234 dividuative intention; perceptually
(intuitionally) d. i., see Acquaintance
Co-directedness (of acts), 149-50, 202, Demonstratives: 204, 213-22, 290-91,
231-32,240,244-46,261,263 359-60, 362-69; and perception,
Comprehension, 279-85, 302-303 213-19, 359-60, 362-69; Husserl's
Conception-dependence (of intentional theory of, 213-19
relations): defined, 13-16; and failure Determinable X, see X
of substitutivity of identity, 26- 28, De re: modalities, 31-33, 74-75, 327-
62-63; in Husserl's theory of inten- 28, 339-41; intentions, see Definite
tionality, 90-91, 107, 148-49,206; intention
in object-theories, 41-44,50-52,56, Direct-object act, 6-8, 27 -28
77 - 78; in possible-worlds theory,
341-43,390 Ego: empirical vs. transcendental, 97-100;
419
420 INDEX OF TOPICS