CHAPTER II

HISTORICAL MANIFESTATIONS

The purpose of this chapter is to give a brief sketch of some of the manifestations that the models of representation and objectivity presented in the previous chapter have had in the history of epistemology. My aim will not be to show that any of the philosophers I consider held any of the views caricatured above. Great philosophers do not hold stripped-down over simplified views such as those presented in the first chapter. Rather, I will attempt to show how certain tendencies in the development of epistemology can be seen as manifestations of something like the models presented above.

The physical-visual model of representation seems to have originated with the Greeks with their tendency towards visual metaphors for knowledge. In both Plato and Aristotle one can detect strong tendencies that could be characterized as exemplifying this model. Chief among these tendencies is the principle that like can only be known by like. Representations were seen as representing in virtue of a similarity to their object, and only objects that could have a similarity to our representations could be objects of knowledge. Aristotle says, "As we have said, what has the power of sensation is potentially like what the perceived object is actually... ." (De Anima II, 5, 418a4; J.A. Smith trans.)[1]

Plato and Aristotle assume that we are capable of knowledge of the world and proceed to describe what the object of knowledge must be like in order for this to be possible. The physical-visual model of representation and the principle that only like can know like play an important role in determining both of their ontologies.

The general representations or concepts through which we know the world cannot be similar to particular objects as given through the senses. A different world is known with the mind than is known with the senses. In Plato, of course, the world known by the mind literally is a different world than the world known by the senses; it is the realm of forms. A realm of general, abstract objects is necessary as the object of our general concepts and words according to the like represents like principle.[2]

In Aristotle there are not literally two worlds, but each object is analyzed into matter and form. In Aristotle, there are different types of objects for each type of representation, for if representation is a matter of similarity then the type of representation will determine the type of object it can represent. Thus Aristotle has a different type of object for each sensory mode, the special sensibles; objects that can be perceived by more than one sensory mode, the common sensibles; and the form of the object itself, which can only be perceived by the mind not through any of the particular senses.[3] Aristotle says, "... what actual sensation apprehends is individuals, while what knowledge apprehends is universals,and these are in a sense within the soul." (De Anima II, 5, 417b23; J.A. Smith trans.) This ontology allows the physical-visual model of representation to work quite well. It is possible, on this view, for us to actually have the object of knowledge in our mind. Here the similarity of representation to object is carried to the extreme. In knowledge the representation is identical to its object. We are actually able to get the form of the object into our soul and use it to represent the object. (De Anima, III, 5, 430a20)

Thus, the attempt to make representations come alive and the necessity that the representation be similar to its object which follows form this attempt shape the nature of the ontologies given by Plato and Aristotle. This is the source of subject-predicate or substance-attribute metaphysics; if the representation is to be similar to the world, then the structure of the world must mirror the structure of our representations.

As we saw, both Plato and Aristotle hold what seems to be perspectivist models of objectivity. In fact, it seems that the perspectivist model originates with Plato (although it has roots in his predecessors Parmenides and the Pythagoreans). The problems with the physical-visual model of representation that lead to a perspectivist model begin with Plato's distinction between knowledge and opinion and his insight that knowledge is true justified belief.[4] Particular representations such as sensations are tied to particular perspectives. Knowledge from particular perspectives is vulnerable to the vagaries of the fleeting and everchanging phenomenal world. It can turn out to be limited only to that perspective and found to be false when applied to other realms. It can be found to be mere appearance or opinion. Since representations are seen as entities that stand between us and the world there is no way to tell if a representation from a particular perspective is objective, if it reflects the object and not the peculiarities of the perspective or the medium of representation, or if it is a mere appearance due only to that particular perspective or medium.

Objective representation requires justification. That is, it requires stepping out of the current perspective comparing it with other perspectives and forming a more general representation that encompasses all of these perspectives.[5] Both Plato and Aristotle have what I called a dialectical model of objectivity. For both, objective knowledge is knowledge of universals, and one gains access to the universals through the comparison and combination of different perspectives. One begins with particular representations and through a dialectical process arrives at more and more general and, hence, more objective representations.

In Plato this process is pictured in the divided line. (Republic, Book Six) At first we have images and appearances, representations from a single perspective. Through the comparison and combination of sensible perspectives we can arrive at sensory experience of stable objects. In mathematics we can arrive ate representations that are not tied to any particular sensory perspective, but which encompass all possible sensory perspectives. With this move the object of our representations is no longer a sensible object, but a form. Finally through the dialectical examination of the definitions involved one can make one's representations independent of assumptions that tie it to some set of perspectives. Here one moves into the realm of the higher forms and achieves unconditioned or completely objective knowledge. This process terminates, of course, with the form of the Good, which encompasses all possible perspectives. Here one has a representation that is not sullied by any distorting influences of medium or perspective.

In Aristotle the process is similar, but Aristotle does not have two separate realms of sensible and pure objects. The universals which we have knowledge of are present in the sensible objects we perceive. Therefore, Aristotle must find a way of making the dialectical process operate within sense perception. His theory of induction or epagoge provides this way.

Induction, for Aristotle, is the source of all the first principles or axioms from which scientific knowledge proceeds. In induction, particular sensory representations of an object from particular perspectives are retained in the mind and somehow compared and fused into a single representation which is identical to the form of the object. Aristotle describes the process well:

But though sense-perception is innate in all animals, in some the sense-impression comes to persist.... and when such persistence is frequently repeated a further distinction at once arises between those which out of the persistence of such sense-impressions develop a power of systematizing them.... So out of sense-perception comes to be what we call memory, and out of frequently repeated memories of the same thing develops experience; for a number of memories constitute a single experience. From experience again - i.e. from the universal now stabilized in its entirety within the soul, the one beside the many which is a single identity within them all - originate the skill of the craftsman and the knowledge of the man of science... . (Posterior Analytics, II, 19, 99b36-100a9; Mure trans.)

Aristotle is not clear about how this happens, but he does provide a vivid image:

It is like a rout in battle stopped first by one man making a stand and then another, until the original formation has been restored. The soul is so constituted as to be capable of this process. (Posterior Analytics, II, 19, 100a11-14; Mure trans.)

Aristotle manages to fit Plato's entire dialectical process, including its termination in a form that is a one that includes the many particular perspectives, into sense experience. The idea is still the same, however. Only representations that encompass many perspectives can be objective. This is due to the physical-visual model of representation which makes justification within perspectives impossible.

Descartes began to see the problems involved with the model of knowledge as representation seen on the visual model inherited from the Greeks and Aristotle in particular. He saw that there was nothing about the representations themselves that guaranteed any correspondence with external objects. The representations as modes of mental substance were separate from and independent of the material objects they were to represent. There could be no reception of the form of the object into the mind as in Aristotle. The soul was a different type of substance than the objects it had to represent. It turns out that, for Descartes, the soul isn't so constituted as to make Aristotle's inductive process possible.

But on seeing that no independent access to the object was available, Descartes was interested in attempting to find other ways of ascertaining that there was indeed a similarity between the representation and the external object. His was the foundational approach to the perspectivist problem. He thought, for example, that using certain methods of arriving at our representations would guarantee that they corresponded to objects.

Early modern philosophy is the attempt to find representations that are self-justifying to serve as foundations for a general system of knowledge. Descartes tried to arrive at these through the use of a particular method. Locke attempted to show how some of our representations could be taken as justified in virtue of their causal origin in reality. The foundationalist move that began modern philosophy is a natural consequence of the scientific revolution. A new method of paying attention to particular empirical experiences and facts and then building upon them to arrive at general laws was providing general representations of the world by deductively and mathematically building upon empirical foundations. It would be natural to assume that these foundational representations have their correspondence with the world guaranteed by some method or by their causal origin, even though it was impossible to justify these foundational representations in terms of other representations. The rest of modern philosophy up to Kant is concerned with showing how the substance dualism of Descartes and the medievals makes it impossible for there to be a similarity between mental representations and physical objects. The identification of similarity with representation is never questioned. The main concern is with finding better and better ways of making representations come alive and represent in virtue of their own properties.

With Berkeley and Hume the problems involved in the physical-visual model of representation and the perspectivist model of objectivity began to become clear. This begins with Berkeley's devastating arguments that nothing can be like an idea but an idea. The possibility of the similarity between mental representations and physical objects that is required by the physical-visual model of representation began to be questioned, although the model itself went unquestioned. In fact, the model was (and is) so firmly in place that Berkeley (like Putnam, as we shall see in Chapter Eight) uses his argument to show that the objects that we represent are themselves ideas or internal to our system of representation. Berkeley argues:

But, say you, though the ideas themselves do not exist without the mind, yet there may be things like them , whereof they are copies or resemblances, which things exist without the mind in an unthinking substance. I answer an idea can be like nothing but an idea; a colour or figure can be like nothing but another colour or figure. ... Again, I ask whether those supposed originals or external things, of which our ideas are the pictures or representations, be themselves perceivable or no? If they are, then they are ideas and we have gained our point; but if you say they are not, I appeal to anyone whether it be sense to assert a colour is like something which is invisible; hard or soft, like something which is intangible; and so of the rest. (Berkeley 1962, I, 8, p. 68)

This argument assumes that the only things that we can experience are representations, which is a consequence of the physical-visual model of representation, and uses this thesis to show that the physical-visual model of representation cannot work. This should have shown that the model is incoherent, but Berkeley took the problem as springing from the assumption of material objects, rather than seeing that representations cannot be static entities such as ideas.

Berkeley is led astray by his perspectivism. Since, on the perspectivist model, a representation from a particular perspective cannot reach outside that perspective, it is impossible for our ideas to refer to material or extra-representational objects. It is this version of the perspectivist fallacy that leads to Berkeley's infamous argument that material objects do not exist because we can't form a concept of them that isn't a concept. Berkeley argues:[6]

If you can conceive it possible for one extended movable substance, or, in general, for any one idea, or anything like an idea, to exist otherwise than in a mind perceiving it, I shall readily give up the cause.

.................................................

But, say you, surely there is nothing easier than for one to imagine trees, for instance, i a park, or books existing in a closet, and nobody by to perceive them. ... but what is all this, I beseech you, more than framing in your mind certain ideas which you call books and trees, and at the same time omitting to frame the idea of any one that may perceive them. ... but it does not show that you can conceive it possible the objects of your thought may exist without the mind. To make out this, it is necessary that you conceive them existing unconceived or unthought of, which is a manifest repugnancy. When we do the utmost to conceive the existence of external bodies we are all the while only contemplating our own ideas. (Berkeley 1962, I, 22-23, pp. 75-76)

But, of course, representations are not representations of themselves. They represent objects, so my conception of an unconceived object is not a conception of an unconceived concept. One could only hold this strong version of the perspectivist fallacy if one thought that representations were static entities or objects that cannot on their own power reach outside of themselves. A representation of both the object and the representation is necessary in order for us to connect them. Thus what is connected to the sign is a representation and not the object itself.[7]

The rejection of the external model of objectivity follows quickly from such arguments. Berkeley himself would have been an internalist if it were not for the role that God plays in his theory as the organizer and projector of all our ideas. Kant made this move with his Copernican revolution in the definition of objectivity. Objectivity becomes a matter of being structured in accordance with certain a priori concepts or categories, instead of a matter of being determined by an external objects. Objects themselves become internal to the system of representations; they become the result of the activity of our concepts.[8]

Apart from following Berkeley's and Hume's arguments to their logical conclusion, Kant also made a revolutionary change in the traditional model of representation. He saw that we are active in our representation of the world. This required three levels of representations: concepts, the active representations that do the structuring of our experience; intuitions, the manifold of representations that are structured; and objects or phenomena, the finished product of the structuring process and the objects of our knowledge. Kant saw that the primary cognitive act was synthesis or judgment, in which a concept was applied to the manifold of intuitions synthesizing it into an experience.

Kant retains a Representational Model of Epistemology. Knowledge is still a matter of representing, but the representations are now of other representations instead of external objects. Kant says:

As no representation ... refers immediately to an object, no concept is ever referred to an object immediately, but to some other representation of it... . A judgment is therefore a mediate knowledge of an object, or a representation of a representation of it. (A68, B93; Kant 1966, p. 54)

Truth is still correspondence for Kant (A58, B82; Kant 1966, p. 48), and representation is still a matter of similarity between the representation and its object. (A137-138, B176-177; Kant 1966, p. 121) Objectivity, however, must become a matter of the internal properties of the representational system, since representations only refer to other representations. It is characteristic of internalist views to retain the physical-visual model of representation while rejecting externalism because of arguments concerning the possibility of representation. It seems as if externalism is rejected in order to save the physical-visual model of representation. If it can be made to work, this model does allow us to know the world exactly the way it is since true representations are similar to their objects on this model. (It is no accident that Kant's views are defended by transcendental arguments that begin by assuming the possibility of knowledge and experience.)

With Kant, ideas and particular images were no longer plausible candidates for the representation that could come alive. The importance of language as the medium of representation stems from an extension of Kant's attempt to avoid the problems brought on by Descartes' substance dualism. There could be no pictorial similarity between mental ideas and physical objects. Kant had argued that knowledge cannot consist simply of ideas, but must involve the application of concepts to experience in a judgment. But in Kant the representation lost its externality. The truth of a judgment reflects our success in bringing order to experience, not a correspondence to external reality.

In analytic philosophy after Kant, language became the candidate for the representation that could come alive. In language we have an object that has the form of a judgment, yet is external to us. A linguistic representation is not just an image; its correspondence is not pictorial similarity in any straightforward sense. Language as an object has special properties that add to its allure. It has a formal structure and rules for the connection of different pieces of language in virtue of this formal structure. The correspondence between language and the world is a matter of structural isomorphism with the logical structure of the world, not any pictorial similarity. Language is also not limited in comprehensiveness of representation as a single idea is. The rules for the construction of language allow the creation of more and more comprehensive representations. So, language provides a medium of representation whose ability to correspond to reality is not hindered by mentality and whose comprehensiveness is not restricted by the concreteness of an image. Language becomes a plausible candidate for the representation that is to be made to come alive if any can. The history of the philosophy of language beginning with Frege and Wittgenstein up to the present is the story of attempts to explain how this particularly alluring form of representation can represent in virtue of its own properties.

All the attempts in the philosophy of language to make language into a representation that represents in virtue of its own properties have sacrificed the ability to explain how language functions in communication. In the ideal language project carried out by Wittgenstein and the Logical Empiricists, language represented in virtue of a structural isomorphism with the world. This relation obtained between private protocol statements and the world as experienced in the subjective sensations of a particular person. As Wittgenstein's private language argument showed, this made communication impossible. By an argument analogous to that of Hegel discussed above, the private language argument also showed that language cannot represent by itself through the mediation of private sensations, since private sensations cannot represent without themselves being interpreted.

The alternative of Frege met with equally bad results. He held that language referred to objects through the mediation of their sense. It was the sense that also allowed communication. Senses were thought to be public, mind-independent abstract entities. They were governed by formal laws of composition so that the senses of whole sentences were determined by the senses of the parts. It was language's formal structure that allowed it to map onto the formal structure inherent in the realm of senses and in the world of objects. This project made it difficult to explain communication in two ways: First, in order to be understood, these public senses had to be grasped in a private psychological act. Language may have been connected to public entities, but there was no way of explaining how our grasp of these entities was the same. Meaning and understanding had been separated. Frege gave a public theory of meaning, but a private theory of understanding. Second, due to the public nature of senses and the constraints imposed upon them by their logical structure, they could not explain the massive role that indexicals, demonstratives, and tensed statements play in communication. Attempts to solve these problems have become a sort of cottage industry in the philosophy of language, but these aspects of language cannot be explained by a theory that attempts to make language represent by itself. These types of words do not refer unless they are situated in the context of an act of linguistic representation.

What these failures should have shown is that language does not represent the world directly. It connects up to concepts or representing structures which we then apply ourselves in our particular context. We will consider how this position would solve some of the puzzles in the philosophy of language in Chapter Seven.