Concepts as Private and Public
Representations
Firstly an
account of concepts is offered that I call “conceptual realism”, embracing two
subtheses, namely, intentional realism and the possibility of reference
realism. Secondly, after considering Wittgenstein´s distinction between
“private” and “public”, the main thesis in this paper is analysed, videlicet,
that concepts are private (mental) representations and also public (linguistic)
representations. Thirdly the differences between internalism and externalism
are considered to propose their overcoming. Finally some remarks are made about
roles of concepts and steps in the life of concepts.
Conceptual Realism
Certainly the topic “concept” is both important in
cognitive sciences and currently in fashion. Indeed concepts can be esteemed as
the basic constituents of thougt in so far as judgments (beliefs) are composed
of concepts and inferences in turn are composed of judgments. On other hand,
nowadays the issue about concepts is broadly studied, such as it can be noticed
in Eric Margolis and Stephen Laurence (1999).
It is not my aim in this paper to discuss the several
theories about concepts, neither the recent ones (classical, prototypical) nor
the current ones (atomist, neoprototypical, theory-theory), but to put forward
my own position, based in some classic philosophical distinctions and in some
psychological findings.
I will call this position “conceptual realism” because
I claim both that concepts are real items in the mind and that concepts can
refer to the world (although I will include literary world).
Hence the first subthesis is intentional realism in
the sense that concepts are mental particulars, that is, they are really in the
mind together with other mental processes (such as perceptions, beliefs,
memories, inferences and the like) all of which are intentional since they do
refer to another item.
I think that it is crucial to maintain the Fregean
distinction between Sinn (sense) and Bedeutung (reference)
(Frege, 1892a), but updating the involved ideas. We can distinguish in all
representation, and therefore in concepts, between their contents or senses and
their references. For example, ARISTOTLE has a content or sense, constituted by
several characteristic properties (creator of logic, tutor of Alexander the
Great, etc.) and has also a reference, that is, the person who lived from 384
to 322 B. C. Likewise CAT has a sense or content comprising a (usually
variable) list of characteristic properties or features (domestic feline,
occasional hunter of rats and mice) and has also a reference, namely, the
millions of animals that existed and exist exhibiting these properties.
It should be noticed that the Fregean distinction
between sense and reference is established firstly in relation with singular
terms and secondly in relation with sentences, in such a way that it is a
distinction relative to linguistic items, not to mental items. Furthermore the
rigid contradistinction, sustained by Gottlob Frege, between concept and
object (Frege 1892b), leads him to not
to accept in any case singular concepts like ARISTOTLE.
Anyway according to Frege (whose ideas I am updating),
in relation to singular terms, the sense is a mental content whereas the
reference is an object.
Following this idea of reference we can remember the
wellknown thesis defended by Franz Brantano (1924) claiming that what is
characteristic of mental phenomena is their reference to an object, and,
therefore, we can define mental phenomena, in contrast with physical phenomena,
by saying that they are those phenomena which contain an object intentionally
within themselves.
But I think that this issue about intentionality can
be confusing in the case that we understand (precisely like Brentano) that the
intentional object is somehow contained in the mental phenomenon, and hence
object (reference) and content (sense) could be confused[1].
Instead of speaking about intentionality perhaps is
clearer speaking about the ability to represent or representationality. Then
intentional realism, which can be called “representational realism”, puts
forward that concepts, like the other mental processes, are really in the mind,
have a content and refer to an object.
The reality of concepts is clearly seen when we note
their causal powers, such as it happens in the case of other mental processes.
These causal powers mean that concepts can produce other mental processes or
can produce behavior. For example thinking CHAIR can produce in my mind
thinking COMFORT, and also usually produces the action of looking for a chair.
But there is another subthesis in conceptual realism,
namely, the possibility of reference realism.
We must realize that one thing is the fact that
concepts refer to or represent items differents from themselves, and that other
thing is whether we can esteem that these items are “real” (in some acceptable
sense). For example, FLAME and PHLOGISTON have a reference, the diverse burning
flames in one case, and an hypothetical principle of fire in the other case,
but whereas the diverse flames are real we know, after Antoine Lavoisier, that
there is no phlogistons.
An important feature of concepts is that they can have
real references. In positive cases concepts appear as really representational
mental items, and hence as very valuable tools in the knowledge of reality. For
example, I could think that the reference of TERRORIST is several groups of
fighters for the freedom of a people, and then the reference is not real, but instead
of that I can think that the reference of TERRORIST is several groups of
irregular soldiers who usually kill indiscriminately civil population, and then
the reference is (sadly) real.
We can establish some criteria about the reality of
conceptual references. The first criterion is to have a perceptual basis, that
is, the fact that perception has supported the elaboration of concepts, but
this perception must be rich enough and free of illusions. A big part of our
concepts are perception-based, although there are concepts that are otherwise
based.
An important point about perception is that it should
be divided in direct perception and reported perception. Human beings we have
an instrument, language, very powerful in such a way that we can perceive directly
diverse objects, but also we can perceive a linguistic (oral or written) report
of these objects, and even we can perceive, in linguistic stories (again oral
or written), fictional objects, commonly reported, like unicorns or Santa
Claus. In fact a lot of our concepts are based in reported perceptions; for
example ARISTOTLE, GRETA GARBO, DON QUIXOTE, WHALE, or LYNX are reported
perception-based concepts for myself.
On other hand we have inference-based concepts, like
QUARK. In these cases the reality of the reference depends upon the validity of
the inference (from perceptual data or from other concepts), and the coherence
with the rest of the conceptual network where the concept in question is
included.
Finally there are other two sources of reality for the
reference of a concept. We can admit that some concepts are innate, however
their number should not be very high, contrary to Jerry Fodor (1998). For
example, psychological research by Elizabeth Spelke (1990) suggests that OBJECT
is innate. On the other hand, human creative imagination is also a source for
the reality of the conceptual reference, such as we can notice specially in the
literary world and in mathematics.
Anyway the possible representation of “real”
things by concepts has a definite criterion, videlicet, their publicity or
“sharedness”, but this leads us to the following paragraph.
Concepts as private and public
Ludwig Wittgenstein (1969) established the distinction
between private language and public language. As it is wellknown, according to
Wittgenstein a private language is sounds which no one else understands but
which I appear to understand, whereas a public language is sounds to
communicate with one another and to influence other people (and therefore a
language understandable by a group of several persons). Wittgenstein added that
there exists no private language, because a genuine language must be public,
with public rules that make possible the communication and influence on other
people.
However, according to Wittgenstein, there exist mental
processes (although they are very inaccesible to a scientific method), that are
private in the sense that nobody knows whether other people also have this
process or something else. But if we express our mental processes we must use a
public and shared language.
Following theses ideas we can espouse the common claim
that our mental processes, and in particular our concepts, are private, in the
sense that each of us alone can observe them directly. But when we express our
mental processes, and in particular our concepts, we do it in a public language
(English, Spanish, French, etc.). But I am interested now in concepts.
Concepts are part of our “mental life”, to employ an
expression used by William James but also by George Miller (1962). But it is
important to clarify that concepts are subpropositional units, using
“proposition” as a modern term for “judgment”. Thus I can have the judgment or
proposition that Spaniards are very proud, and in this case I have also the
concepts SPANIARD and PROUD.
On other hand each concept includes as its content
several characteristic properties that refer to one thing (singular concept) or
to a group of things (universal concept). These properties can be reduced to
one and even be fixed, like PRIME NUMBER (a number that only is exactly
divisible by itself and by the unity). But in normal cases the number of
properties is superior to one and variable. This variability depends on the
knowledge of persons and also on the historical period. For example, HUMAN can
include rational animal, or simply person, or animal with intelligence and free
will, etc.
Thus a concept can be identified with a list of
characteristic properties, and this list sometimes can be considered as
constituting the definiens of the concept. However the point of view
that concepts are definitions, named the “classical view of concepts”,
currently it has simply ceased to be a serious contender in the psychology of
concepts (Murphy, 2002). But in my opinion, although usual concepts are not
definitions, it is sometimes possible to formulate them as definitions.
Furthermore Frank Keil and Nancy Batterman (1984) have found evidences that a
characteristic-to-defining shift is present in the acquisition of word meaning.
(I shall say below a few words about the relation between meaning and concept).
We can now exam the other side of concepts, that is,
their reference. Such as I have remarked the reference of a concept can
be unreal, like in the case of PHLOGISTON, but it can be “real” (in some
interesting sense), so that concepts are fundamental to our knowledge of
reality. I include the literary world as a convenient kind of reality, although
I know that this position contradicts all the positivist tradition, specially
the theses defended by Rudolf Carnap (1935) confronting poetry and science. I
do not want to erase the distinctions between the literary reality and the
biological and physical reality, but only to do justice to the fact that the
whole of stories that constitutes the literature plays an important role in our
beliefs and hence in our lifes. In fact literary criticism can be deemed as a
human science, whose basic units are also concepts, that can have a
“real” reference. For example if my concept DON QUIXOTE contains the property
of having died in Barcelona, its reference is not real, but if DON QUIXOTE
contains the property of having mistaked windmills for giants, its reference is
real.
I have claimed that concepts are part of our mental
life, but it is also very important to realize that concepts have as well a
linguistic life. I hasten to add that concepts have a linguistic life when they
are considered as meanings.
Concepts and meanings can frequently be identified,
but they should not be confused. In general a meaning is related with a sign
(in particular a linguistic sign), whereas a concept is related with a thing or
group of things. More precisely, we can have instances of concepts (related to
things) without a sign to express them; for example, when Murray Gell-Mann
discovered the quarks before giving them this name. And we can have instances
of meanings (related to signs) without a concept possessing a real reference;
for example, the meaning of the word “caloric”.
However in usual cases there is a coincidence between
concept and meaning. For example, there is a group of cats, we use the word
“cat” to designate them, and we have the meaning of the word “cat” and the
concept CAT, which are coincident.
I would like to emphasize that concepts as meanings
have a linguistic life since they are (as meanings) related with linguistic
signs, such as Spanish terms or English terms. Nevertheless concepts are not
linguistic terms, except if they are considered members of a language of
thought, like Fodor´s LOT (Fodor, 1975), but they are mental particulars.
Likewise meanings are mental items, in such a manner that, contrary to Hilary
Putnam (1975), they are in the head, exactly in the mind-brain.
Concepts are
captures of characteristic properties of one thing or of a group of things, but
these captures are not fixed, in such a way that in relation to a same thing or
a same group of things different concepts are possible, all of them
functionning (to a lesser or greater extent) as characteristic captures.
However (and this point seems crucial to me) we will
have a genuine (objective) concept, and not a mere (subjective) image, in the
case that the concept has a public character, that is, in the case that the
concept turns out to be a “shareable” representation. Thus concepts are private
items, since they are mental particulars, and too they are public items,
because they are shareable by means of their expression in a public language
(English, Spanish, French, and so on).
When concepts are expressed as linguistic terms they
turn out to be powerful tools of communication. They allow us to speak about
what happens and also about what happens to us. If concepts are shared in a
high degree the agreement is easy, if concepts are shared in a low degree the
discussion is open.
Thus we should speak about “sharedness” (the character
of being shared) of concepts, as something parallel to the publicity of the
terms expressing them.
But to say that we can share concepts must be said
carefully in order to prevent from making mistakes. I mean that we can share identical
concepts, but, contrary to Fodor (1998), we can not share the same
concept. Here we have a very important distinction between identity and
sameness (Martinez-Freire, 1979). There are not two same concepts, because
there are not two same individuals, that is, two individuals that are exactly
identical (since they would be one). But we can have two individuals (for
example, concepts) that are partly identical. Leibniz stressed that there exist
not two indiscernible or exactly identical entities. But strangely this
principle is usually called principium identitatis indiscernibilium, in
such a way that it could suggest just the contrary of what is claimed, namely,
it could suggest that there exist indiscernible entities and that they are
identical, whereas Leibniz claimed that there exist not indiscernibles entities
and that hence they are not totally identical. For that reason in
Martínez-Freire (1979) I have proposed to call this principle principium non
identitatis discernibilium.
Let us consider DOG as a concept shared by a lot of
people. This concept in my mind is indeed different from this concept in my
brother´s mind (at least because my mind is different from my brother´s mind).
However both concepts can share a basically identical list of properties, in such a way that I and my
brother coincide in saying which animals are dogs and which animals are not
dogs.
In other terms, every concept referring to dogs is a
token in some mind, but there is not a corresponding type understood as an
abstract entity. The type corresponding to the different tokens is just the
whole of tokens. I think that this point is very important to avoid Platonism.
A philosophical tradition in the theory of concepts,
from Plato and Aristotle to Aquinas (and perhaps to Frege), understands
concepts as abstract entities. In the case of Plato these abstract entities are
Forms or Ideas in a really real, intelligible world, different from our daily,
visible world. In the case of Aristotle and Aquinas these abstract entities are
strangely embeded in things.
Internalism or Externalism?
As I have already remarked, there are so to speak two
sides in relation with a concept, an internal side or content and an “external”
side or reference.
We can say that the content or sense of a concept is
internal because it is the concept itself, and hence it is a mental item.
Therefore the Thomist distinction between conceptus formalis, or that
where we understand, and conceptus obiectivus, or what we understand
(Gredt, 1951) is unnecessary. When we have the concept TABLE we have a list of
characteristic properties (for example, a piece of furniture with a board and
legs) that we usually share with other person´s mind. And this concept is this
list, although it can be accompanied of an image of a familiar table. From a
logical point of view conceptual content has been called “intension”.
On other hand, we can say that the reference of a
concept is “external” because it is outside the concept. Certainly this
external character does not mean necessarily that conceptual references are
always in the external world. For example, the reference corresponding to the
concept COMPLEX NUMBER does not is in the external world (but possibly in our
creative imagination). However the reference of CHAIR or the reference of WATER
is in the external world, that is, outside our minds. From a logical point of
view the reference of a concept has been named its “extension”.
We can wonder whether the defence of psychological
phenomena as mental processes could lead to the neglect of human body and
social relations. If we esteem that mental processes, and in particular
concepts, are so to speak closed in the mind, we would defend some sort of
internalism. And, on other hand, if we deem that mental processes, including
concepts, are described solely in terms of bodily and social relations, then we
would defend some sort of externalism.
But in so far as concepts have both an external side
(extension or reference) and an internal side (intension or content), we should
overcome internalism and externalism as well. It is preposterous to claim that
concepts are not in relation with the external world; in fact this position
means to give up representationalism and to espouse a useless solipsism (since
without representation there is no cognition). On other side, it is absurd to
claim that concepts are not in the mind but in bodily changes or in social
agreements; in the first case we would have behaviorism, that is, psychology
without psyche or mind, whereas in the second case we would forget that social
agreements are shared beliefs and that beliefs are also mental items.
It is true (and even a truism) that human agents are
embodied and embedded[2],
that is to say, human agents have a specific body and we are situated in an
ecological niche (our peculiar world). But it is also true that our mind-brain
represents our body and represents our world, being able to put up internal
stand-ins for both. (Martinez-Freire, 2005).
An issue related with internalism and externalism is
the question about the priority of intension over extension or vice versa.
Indeed we can maintain that intension determines extension, but we can also
think that extension determines intension. I claim that both theses are true,
although suitably understood.
As the concept is its intension it seems clear that,
in the dimension of its constitution, conceptual intension possesses a priority
over conceptual extension. When we consider or analyse concepts we take their
list of properties into account. But as the concept is about things and usually
has been produced by things (excluding innate concepts) it seems evident that,
in the dimensions of its origin and application, conceptual extension has a
priority over conceptual intension. Along our life we encounter dogs, we build
the concept DOG, and then we apply it to say which animals are dogs and which
animals are not dogs.
In conclusion we can establish that concepts are real
items in the mind, constituted by its content or intension, and that concepts
are usually obtained and have a real application by means of its reference or
extension.
Complementary Remarks
I would not like to finish this paper without saying a
few words about the roles of concepts and about the steps in the life of
concepts too.
We can hold that concepts carry out roles or functions
in two levels, namely, a basic or animal level and a higher or specifically
human level. Indeed animal life demands the ability to distinguish and to
recognize things and other living beings in the environment; in the case of
human animals we have concepts that make possible not only to discriminate
using one or two features, but also to categorize in a more complete and stable
way. As a consequence human animals, in general, make less mistakes than other
animals in representing and recognizing
things and living beings around them.
But there is the higher level of sciences,
technologies and arts, that are specifically humans. In this level concepts are
indispensable tools, because we need laws and generalizations that precisely
must include general concepts and categories. In particular, besides natural
kinds concepts, like WATER or ORANGE, and artificial kinds concepts, like TABLE
or SHIP, we employ typically in this superior level inference-based concepts,
like QUARK or MOLECULAR STRUCTURE, and logical and mathematical concepts, and
also in the field of arts reported perception-based concepts, as I have noted
above.
On the other hand we can distinguish three steps or
moments in the life, so to speak, of concepts.
We can have innate concepts, like OBJECT, but possibly
the number of them is not very large. Thus the first step in the life of a
concept is its acquisition. There is the temptation to think that concepts are
built up from sensory input (Prinz, 2002), but certainly perception is not the
only source of concepts, though it is very important. Perception, as basis of
concepts, can be direct, when we perceive directly the instances or the
individual that will be the reference of the concept; for example, we perceive
several cats in the street or in the houses to acquire the concept CAT. But
perception, as base of concepts, can also be reported, when we perceive by
means of oral or written stories, or by means of pictures and movies, the
instances or the individual that will be the reference of the concept; for
example, we perceive George W. Bush in the television to acquire the concept
THE CURRENT PRESIDENT OF USA.
Besides perception, inference and creative imagination
are other sources of concepts, and indeed with a great import for the
development of sciences, technologies and arts.
The second step or moment in the life of concepts is
their possession. This is a major point, since we possess concepts in complex
ways. To begin with the possession of concepts is usually unconscious, what is
crucially necessary for the organization of knowledge (since we can not pay
attention to all our concepts), but when we use them their possession is
conscious. Then often we have concepts linked to images more or less clear and
usually familiar; for example, our concept SEA is linked to an image of waves
of sea. Certainly this is the reason why British empiricists, specially George
Berkeley, were inclined to imagism in the theory of concepts, that is, the theory
that concepts are images. On other hand the possession of concepts, as it is
normal, is localized in the different places for memories (practically
throughout the brain), in such a way that concepts and certain memories could
be identified. Likewise concepts can be stored bearing relations among them, so
that they can be organized in hierarchies (Rosch, 1978); for example, in
relation with CAT the concept ANIMAL is in a superordinate level, but the
concept KIRI (my late cat) belongs to a subordinate level.
The third moment or step in the life of concepts is
their application. In this step concepts are related with perception (again)
and with judgment. Indeed perception can be the basis for a concept, but also
very often we make perceptual identifications; for example, I can see an
intense light, and then I can perceive a headlight, applying the concept
HEADLIGHT. On other hand also I can apply concepts to make judgments of
identification in diverse levels; for example, the sentence “this is a cat” or
the sentence “cats are animals” apply the concepts CAT and ANIMAL. This
application or use for identification is very important. Sharon Lee Armstrong,
Lila Gleitman and Henry Gleitman (1983) have clearly distinguished in concepts
an identification function along with a conceptual core.
References
Armstrong, S. L., Gleitman, L. R. &
Gleitman, H. (1983). What some concepts might not be. Cognition, 13,
263-308.
Brentano, Franz (1924). Psychologie vom
empirischen Standpunkt. (Ed. O. Kraus). Leipzig: Felix Meiner.
Carnap, R. (1935). Philosophy and
Logical Syntax. London: Kegan Paul, Trench, Trubner & Co.
Fodor, J. A. (1975). The Language of Thought. New York: Crowell.
Fodor, J. A. (1998). Concepts. Where Cognitive Science Went
Wrong. Oxford: Clarendon
Press.
Frege, G. (1892a). Über Sinn und Bedeutung.
Zeitschrift für Philosophie und philosophische Kritik, 100,
25-50.
Frege, Gottlob (1892b). Über Begriff und
Gegenstand. Vierteljahrsschrift für wissenschaftliche Philosophie, 16,
192-205.
Gredt, J. (1951). Elementa Philosophiae
Aristotelico-Thomisticae. Vol. I: Logica. Philosophia Naturalis. Barcelona:
Herder.
Haugeland, J. (1998). Having Thought.
Essays in the Metaphysics of Mind. Cambridge (MA): Harvard University
Press.
Keil, F., & Batterman, N. (1984). A
characteristic-to-defining shift in the development of word meaning. Journal
of Verbal Learning and Verbal Behavior, 23, 221-236.
Margolis, E., & Laurence, S. (Eds.)
(1999). Concepts. Core Readings. Cambridge, MA: The MIT Press.
Martinez-Freire, Pascual F.
(1979). Algunas observaciones sobre la identidad. Teorema, VIII/3-4,
215-227.
Martinez-Freire, P. F. (2005). Being
inside: Putting representation, body and world together again. In P. F.
Martinez-Freire (Ed.), Cognicion y representacion, Supplement 10 of Contrastes,
39-50.
Martínez-Freire, P. F. (2007). La
importancia del conocimiento. Filosofía y ciencias cognitivas (2nd ed.). A
Coruña: Netbiblo.
Miller, G. (1962). Psychology. The
Science of Mental Life. New York: Harper & Row.
Murphy, G. (2002). The Big Book of
Concepts. Cambridge (MA): The MIT Press.
Prinz, J. J. (2002). Furnishing the Mind. Concepts and their
Perceptual Basis.
Cambridge (MA): The MIT Press.
Putnam, H. (1975). The meaning of
“meaning”. In K. Gunderson (Ed.), Language, Mind and Knowledge.
Minneapolis: University of Minnesota Press.
Rosch, E. (1978). Principles of
Categorization. In E. Rosch & B. Lloyd (Eds.), Cognition and
Categorization, Hillsdale (NJ): Lawrence Erlbaum.
Spelke, E. (1990). Principles of object
perception. Cognitive Science, 14, 29-56.
Wittgenstein, L. (1969). Tractatus
logico-philosophicus. Tagebücher 1914-1916. Philosophische Untersuchungen.
Frankfurt am Main: Suhrkamp.