Read (as "The Concept of Referent") at the New Jersey Regional Philosophical Association, Douglas Campus, Rutgers University, November 3, 1973 (with corrections and additions).

I should like to begin my paper on THE "ONTOLOGY" OF REFERENT with an explication of Quine's treatment of the problem of reference.  His is typical of a delimiting approach with which I find fault.
Quine's major concern regarding the existence of real things is concentrated on the development of genuine referral expressions, which, for him, necessarily involves bound variables.  He was apparently led to this solution out of the necessity to solve the problem of speaking meaningfully of non-existents.  In doing so he appears to have ignored other complex functions of referral expressions despite his admitted acceptance of a behavioristic theory of meaning and language.
If I interpret Quine correctly, natural language is inoperable as an instrument for testing ontological commitments.  This is so because it is permeated with the subjectivity of sensory data, sloppy usage, indeterminacy of translation and inscrutability of reference all of which are the outcome of behavioral response and socialistic evolution.
The validity of ontological commitment to the physical and the abstract depends upon circumventing the four items above.  Hence, the formal logic of bound variables is, for Quine, the most useful instrument, to date, for circumvention of such linguistic disorder, or as he puts it "ontological slums."
We have ontological slums in the mainstream of philosophic (and non-philosophic) discourse in the form of outlandish ontological commitments.  For example, existentialists speak of a Substantive nothingness and many of us speak of a Santa Claus as if these things exist.  I believe these are the kinds of concepts upon which Quine wishes to bring to bear the principles of utility and efficacy in ontological commitment, i.e., principles similar to those of science.  To quote Quine:

Our acceptance of an ontology is similar in principle to our acceptance of a scientific theory. . . .  We adopt. . .the simplest conceptual scheme into which the disordered fragments of raw experience can be fitted and arranged. (16, Logical Point of View, 1961)

Quine is concerned to distinguish, for the purposes of ontological commitment, between referral expressions that are authentic and those that are not.  From this we may draw the conclusion that such legitimate referral expressions point to ontological commitments as the closest thing to referents possible and that expressions which cannot be converted to a categorical system of bound variables do not.  Nevertheless, Quine is quite clear that he is talking about real things also. "Semantic ascent, as I speak of it, applies anywhere," he says. (p. 271f Word and Object, 1960)  He wishes to include theory, but offers an example of semantic ascent as " . . .the shift from talk of miles to talk of 'miles.'  Quine wants, in other words, to insist that semantic ascent is the linguistic device for dealing with what is and that philosophy deals with what is as well as with the analysis of language.  The use of semantic ascent, then, is not to deny the sensibility of discussing real objects but rather to help the various expert authorities to deal with the different categories of their disciplines.  Quine says:

What distinguishes between [an] ontological philosopher's concern and all this is only breath of categories.  Given physical objects in general, the natural scientist is the man to decide about wombats and unicorns.  Given classes, or whatever other broad realm of objects the mathematician needs, it is for the mathematician to say whether in particular there are any even cubic numbers.  On the other hand it is scrutiny of this uncritical acceptance of the realm of physical objects itself, or of classes, etc., that devolves upon ontology.  Here is the task of making explicit what had been tacit, and precise what had been vague; of exposing and resolving paradoxes, smoothing kinks, lopping off vestigial growths, clearing ontological slums. (p. 275, Ibid.)

Quine, then, is concerned with devising the means through semantic ascent whereby the confusions of subjective sensory data, constructs of fantasy, willy nilly conceptual whims, illusion of understanding and knowing, hypostatization of words and concepts, etc., can be successfully sifted out through a systematic, utilitarian efficacy somewhat analogous to the successful theoretical and predictive constructs of science.  Thus is derived an acceptable ontological commitment as opposed to the illusions of one.
I am quite sympathetic to Quine's felt need.  From my point of view, however, his solution to reference is insufficient and constitutes a much more limited approach to the problem of reference than I am willing to accept.  It restricts (authentic referral expressions to ontological commitments via a strict system of bound variables.  I do not see how Quine's restrictive use of the term 'reference' will find universal usage -- not to mention understanding.  Nor can I foresee, considering the complex functions of language, the wide-spread practical value comparable to that which ensues from scientific principles.  There is a need to account for those referral expressions which do not commit us to the existence of anything.  Natural, (i.e., behavioral) language contains genuine referral expressions that do not translate into the canonical form of bound variables.  Quine's referral expression, viz., bound variables commit us to an ontology whatever its form.  There are, however, referral expressions which do not commit us to an ontology; for example, the ceremonial function of language).
Language has multiple functions most of which are not at all related to ontological commitment but rather to inter-behavioral factors, such as personal, community, or social relationships.  Frequently, referral expressions even appear to refer to some form of ontology when in fact they function otherwise; that is to say, their referents are not ontological.  The term 'referent' may or may not be an expression of ontological commitment.  And a referent may or may not possess ontological status dependent upon the nature and the meaning of the referral expression and the linguistic covenant in use.  Had Quine made use of the term' referent,' he might have been led to such directions of thought.  To follow him, sentences such as "Socrates is mortal," and "The fire is burning," etc., cannot be said to be genuine referral expressions.  Hence, in those forms they would have no referents.  They might, however, be translated into bound variable expressions.)
No one would deny that an ontic commitment (whatever can be meant by the term 'commitment') must not be confused with ontology per se.  However, ontic commitment is a performative act.  By so committing oneself, referents, for the referral expression, come into being, so to speak.  But, in popular jargon (ordinary language) the term 'ontology' takes on many forms, viz.: physical, conceptual, perceptual, hallucinatory, illusory, imaginative, linguistic, etc.  It is, for instance, difficult to deny the ontology of the idea (whatever may be the nature, vagueness, or ambiguity of an idea) of Pegasus as Quine seems to admit.  Surely the term 'Pegasus' refers to something.  That is to say, someone uttering the term means something by it even in "Pegasus does not exist."  The task then would be to determine what is the referent rather than whether there is an existent "object" or "similarity" of phenomenon that goes by the name.  On a metalanguage level, even the sound of a word can have for its referent at least its written form and vice verse.  I propose, therefore, to take a radically different approach to the problem.
I would posit the following:  The terms 'referent,' referential,' 'reference,' and 'referring' should not be allied with formalized logic or the concept of ontology in the realist or idealist senses of the latter -- physical or abstract.  There is no referential quality with referential ontological status inherent in any given substance or "entity" to which linguistic attention is drawn.   Neither the term 'referent' nor the "object" called a referent possesses a generic quality which distinguishes a "referent" from, for instance, a "symbol."  That is to say, there are no referents or symbols as such because there are no things whose principle functions are to be a referent or a symbol.  Rather, 'referent' is merely a term ranging over an ever shifting domain of focal points of attention. 'Referent' and 'symbol' are merely noun forms for acts of mind, i.e., the mental acts of referring and symbolizing.  Likewise, one formulates referral expressions utilizing entities and 'entities' (be they physical objects, ideas, concepts, words, sounds, etc.), as focal points for acts of mind (however we may define "mind").  And in cases of awareness of awareness, one even formulates referral expressions about referral expressions -- which Quine has gone to great lengths to do.  In all such cases what the referral expressions refer to are themselves, in any strict sense, not referents, symbols, or inferences.  They are merely said (thought) to be so only when designated (expressed or not) by some linguistic covenant.  They are no more than the focal points of attention which, by convention, we "substantize" in noun form 'referent,' 'symbol,' and 'inference' for the purpose of smoothing our attempts to communicate with each other.  Simplistically put, there were no referents, symbols, or inferences before intelligence or language existed in the universe.  Symbols, referents, and inferences are all different kinds of acts of meaning, i.e., intentional acts.  They have no ontological status as such.
For the moment, however, let us colloquially use the term 'ontology' to mean "exist" and the term 'mean' to mean "some intentional reference to an 'object' of linguistic referral."  For instance:

The (physical) book exists.  Ideas exist.  Numbers exist.  Geometric shapes exist. Democracy exists.  By 'chair,' I mean "that object used to sit on."  By 'idea,' I mean some "thought," "concept," "belief."  By, 'God' I mean "a supernatural entity," and on and on as used by "everyman."

When we use the terms 'exist' (ontology) and 'mean' this loosely, each is an on-going process, and as such experientially and ontologically some "referents" cease to exist as soon as they come into being, like Heraclitus' "river."  Quine, of course, insists that the general term 'river' should not be confused with the stages of a river and that there really is no difficulty in stepping into the same river twice.  Likewise, he would insist that in speaking of a Mr. Smith, for instance, we are referring to the same person even if at various time intervals.  It is only the stage of the person that is different.  Assuming he is correct, it does not solve the problem of reference.  Rather, it seems each stage becomes a referent, that is, if we are going to hypostatize 'referent,'  And, under certain linguistic covenants, we should.  If my grandson speaks of Mr. Smith as the same teacher I studied under forty years ago, there is little doubt that there will be a radical difference between our separate understandings (referents) of who Mr. Smith is in terms of his appearance, present personality, etc.  Surely, also, he was not Mr. Smith, the teacher, when he was 15 years old.
I am, in turn, opting for a concept of 'referent' which will, from the point of view of understanding, eliminate the concept of ontology and circumvent arbitrary and conventional linguistic covenants for the term, while at the same time, permitting the implication of the first and the socio-culturally necessary use of the second.  There is, of course, little doubt that the inscrutability, untranslatability, and indeterminacy of a shifting, changing ontology -- whether physical, phenomenological, conceptual, or what have you pose difficulties.  This is particularly the case in uses of language which demand a uniqueness (spatio-temporal or other) of reference for communicative purposes.  It is herein, however, that it might be said that 'referent' means many things dependent upon the level of language or the linguistic context in use (or in Quinean terms, background language.)  My position is that Quine's is one of many fulfilling a special purpose, and that neither he nor anyone else will solve the problem via bound variables if ontological commitment is taken as a test of ontology, or referent is the object of specific referral expressions.  Certainly a referent can be said to "hold" its character discursively as the enduring thread (term -- like Quine's "river") upon which is woven the fabric of the psychological or phenomenological referential data.
The real issue is to identify the use of the term 'referent' with its specific "ontological" range or process, if you will: that is, to identify its existence as idea, as memory, as perception, as conception, or as something physical, imaginative, illusory, hallucinatory, or sensory, etc., each a process, each a focal point of referral attention nounized according to the linguistic context (i.e., ordinary language, scientific language, historical language, etc., required.  Thereby clarity of thought may ensue dependent upon the degree of clarity needed as is the case with the need for varying degrees of explanation.
On the extensional level of language all "referents" are complex "entities" of activity, "held together" for discussion by an enduring term.  Not only are such so-called "referents" complexes of  beginning, on-going, or ending, processes, but given the reality of the fact that language refers to our perceptions, not to a physical reality, each perception is a construct of an ď'infinity' of dates" -- if Einsteinís theory of relativity has merit.  This verifiable fact alone creates the need to invoke pragmatic ordinances of linguistic usage.  As Einstein points out, "The universe of ideas [the giving of meaning, i.e., language], is just as little independent of the nature of our experiences as clothes are of the form of the human body."  However, the individual governing agents are to be found intentionally. They are acts of mind which lay down the determinants permitting such designations as "referent," "symbol," "inference," etc., according to whether various particular minds, at a particular time, use the same object of attention as a focus for a referring act of mind, a symbolizing act of mind, or an inferring act of mind, etc.

Strictly speaking symbols, referents, and inferences do not exist outside the mind, however much we colloquially refer to them as "things," "words," etc.  Any thing (process) can serve as a focus for those acts of mind which we call "symbolizing," "referring," "inferring."  When something is not so serving because no mind is present, it cannot validly be called a "symbol," a "referent," or an "inference."  Furthermore, in the presence of different minds, what is a symbol to one often is a referent to another, and an inference to a third*

*P. S. Schievella, Critical Analysis: Language and Its Functions (New York: Humanities Press, Inc., 1968), p.37, footnote.  See also: -------- "Symbols, Referents, and Communication in the Human Use of Language," The Journal of Critical Analysis, Vol. I, July 1969, No. 2, p.82.

Certain acts of mind are acts of referring -- sometimes manifested in referring expressions, sometimes not.  Others are acts of symbolizing, reflecting, criticizing, synthesizing, analyzing, and many more.  Each of these, of course, is in turn manifested in reflective expressions, critical expressions, analytical expressions, etc.  Each act is different in character from every other as are various kinds of referents different -- one from another.  Strictly and ultimately, there is no entity that can be called a criticism, a synthesis, an analysis, etc., separate from those acts.  In kind there is no referent (strictly speaking) that is the substantive object of the referral expression inherently possessing the quality of referent such that one special set of "objects" (not to be construed here as merely physical) can be relegated to the category of "referent" while others cannot.  Any "object" can be made to fit that category.  In short, much of the confusion arising out of the problem of trying to determine the character and nature of referent lies in identifying acts of referring with "objects."  It is the act of referring that induces the pragmatic conversion into noun form, the designation, and individuation of an "object" which is thereby "objectified" as a referent.  What needs to be emphasized is that what refers in every instance of referral is some "entitizing" process in the mind.  Some of these processes are stimulated into "existence" by some presumed, acknowledged, or inferred "entity" outside the mind.  Some are stimulated by various processes of thought (like conceptualizing).
Pursuit of a basic definition that will serve as the complete answer to what is the "meaning" of a referral expression (or a word) comes close to pursuit of an ideal language -- a will-o-the-wisp -- primarily because "referents" (as are "meanings") for the "symbols" constituting the expression are so diverse and multi-faceted.  The exception, of course, is a linguistic or cononical covenant -- a Quinean bound variable, for example -- directed to a particular need and to a particular linguistic orientation that will fit only particular human needs, which in the final analysis is analytical in character and cannot apply to the many diverse functions of language.


Multiple functions of referral expressions other than that of ontological commitment have been ignored or de-emphasized.  Consequently, the concept of "referent" has been shortchanged.

There is no inherent referential quality with ontological status.  When some "thing" is designated (or individuated) as a referent, (or an ontological commitment), the designation "referent" is but a social covenant of referral (finding expression in noun form, i.e., 'referent') which is in fact but a linguistically functional term (as opposed to ontologically functional entity) designating an act of mind -- often expressed in verb form, i.e., "refers."

Acts of referring manifested (but not always) in referral expressions often do not commit us to the existence of anything.  Yet. they are by linguistic covenant (without ontic referral) accorded referents.  My thesis is that there are no referents ontologically and that use of the term 'referent' is only a linguistic convenience of speech act. (The same object is accorded the term 'symbol' at the same moment by a different mind.)

Once this is recognized, we can begin to distinguish the kinds of "ontology" (sensa, phenomena, concept, percept, memory, abstraction, operation, etc.) to which the term 'referent' alludes and referral expressions refer.

There are no such things as referents as such because there is no such thing whose principle function is to be a referent.

© 1997 by Pasqual S. Schievella