A paper presented at the Eastern Pennsylvania Philosophical Association (Bloomsburg University, Fall 1990).

REFUTING LINGUISTIC IDEALISM

Andrew Chrucky

Ted Schick has written three essays on the role of the qualitative content of experience: the earliest essay is titled "Can Fictional Literature Communicate Knowledge?"1; a more recent one is "The Semantic Role of Qualitative Content"2; and his latest essay, the one Ted presented today, is titled "The Epistemic Role of Qualitative Content.3" He sent me a copy of the latter for comment in January 1990 with some other of his published essays. I tried writing something -- but it was nothing substantial. All he got from me were some comments over the phone. I promised to do better. He then sent me a slightly revised version in May. When I got to giving him substantial comment, it was October, and his paper was already accepted for publication. I have now read all of Ted's published writings, and what follows are my comments on Ted's paper in the light of his other writings.

THE PROBLEM: LINGUISTIC IDEALISM

What is the main underlying problem that Ted is trying to deal with in the bulk of these papers? Although Ted does not formulate the problem as I will, it is a formulation that Ted accepts, especially if his papers on the role of qualitative contents are viewed in the context of some of Ted's other writings. I have in mind his essays "Rorty and Davidson on Alternate Conceptual Schemes,"4 "In Defense of the Correspondence Theory,"5 and the essay he presented before the EPPA at the Spring 1989 meeting at Wilkes University, "How is Philosophy Possible?"6 The hidden problem that is gnawing at Ted, and I should add a problem that has been bothering me also for a very long time, is how to dialectically escape from what I will call 'linguistic idealism', by which I mean the thesis that our view of reality is intimately dependent on our language -- indeed, the thesis is that all experience (knowledge) is language-laden.7

Ted comes very close to formulating the thesis of linguistic idealism, but he does not explicitly view the problems he is coping with as resulting from linguistic idealism. But in fact Ted is wrestling with linguistic idealism and its implications.

IMPLICATIONS OF LINGUISTIC IDEALISM

If linguistic idealism is correct, then we are cognitively trapped in language, and the following ramifications seem to follow:

  1. The hermeneutical circle of which Continental philosophers speak is a reality.
  2. The claim that there is no theory-observation distinction is correct: all observations are theory-laden because all observations are language-laden, and languages embody theories or conceptual frameworks.
  3. There is no neutral ground by which to compare alternative conceptual schemes. Indeed there cannot be alternative conceptual schemes for us. Either an alternative conceptual scheme is translatable into our own, and then it is not an alternative, or it is meaningless.
  4. The correspondence theory cannot be formulated without getting into a circle. Tarski's convention T: 'p1 is true iff p, does not escape the circle of language. All we have here is a sentence formulated in a metalanguage related to an object sentence. Some claim that the three expressions 'p1, 'it is a fact that p', and * it is true that p1 are logically equivalent, and from this they argue that they have the same meaning.8
  5. Talk of meanings is really talk about linguistic meanings, and meaning is for linguistic idealism, as Ted quoting Paul Churchland puts it, a matter of a linguistic network, i.e. words have meaning as a function of their linguistic roles in connection with other words. One implication of this view of meanings is linguistic holism. Ted is quite aware of linguistic holism, and that, on some version, it is forsaking the relevance of qualitative content to meaning, and he is objecting to such a version in his papers on the role of qualitative content.

There are other consequences of linguistic idealism such as (f) the dismissal of the given as a myth by Sellars, and the subsequent (g) denial of foundationalism in knowledge. Linguistic idealism has also been construed to imply (h) relativism, and it also has led to the current (i) nihilism of Rorty, which proclaims the end of philosophy.

TED'S RESPONSES TO THE RAMIFICATIONS OF LINGUISTIC IDEALISM

Several of Ted's papers are reactions to these ramifications of linguistic idealism. I will not concern myself with how Ted deals with these ramifications except for Ted's attempt at a rebuttal of (e), i.e., meaning holism, which is the topic of the paper he has presented today; and Ted's defense of the correspondence theory of truth against (d).

Let me start with Ted's argument against (e). Ted presents the following argument by Paul Churchland as found in his Matter and Consciousness9 for the irrelevancy of the qualitative content for the meaning of a word:

  1. To know the meaning of a word is to know how to use it in conversation, explanation, and prediction.
  2. To know how to use it in conversation, explanation, and prediction is to know its causal/relational properties.
  3. To know its causal/relational properties one need not know its qualitative content.
  4. Therefore to know the meaning of a term one need not know its qualitative content.
I am not sure that (i) faithfully represents Churchland's position because Churchland in the cited context seems to be talking about the meaning of a theoretical word, not just any word. But, on the other hand, Churchland's writing has enough fuzziness about it so that (i) may justifiably be attributed to him. In any case, Ted tries to refute this argument by objecting to premises (ii) and (iii). Whatever the merits of Ted's objections to these premises, the simpler way of refuting this argument is to attack premise (i). Premise (i) is false, provided, of course, we are talking about the meaning of all words; not just theoretical words. Knowing how to use a word in conversation, explanation, and prediction gives only a partial understanding of a word -- which is the conclusion Ted is driving at). A better formulation of the intention of (i) is found in the writings of Wilfrid Sellars who would substitute for (i):

(i') To know the meaning of a word is to know how to use it in making perceptual reports, inferences, and expressing volitions.
Missing from premise (i) is the perceptual context. That is why Sellars can agree with Ted that the qualitative content is relevant to a complete understanding of meanings. Without knowing how to use a word in a perceptual context, Sellars wrote, " . . . a blind man has a "partial" understanding of colour words though he is unable to use them in perceptual contexts."10 Consequently, the following statement in Ted's paper:
"Sellars, Rorty, and Churchland claim that qualitative content is irrelevant to our knowledge because it is not part of the meaning of our mental terms."
is not true of Sellars.

Turning now to another ramification of linguistic idealism -- (d) the rejection of the correspondence theory of truth, Ted tried to save the correspondence theory in his essay "In Defense of the Correspondence Theory," with what strikes me as the strategically correct tactic. He distinguished facts1, which he identified with non-linguistic states of affairs; from facts2, which he identified with true statements. But because this view of states of affairs (facts1) is not exploited in other relevant contexts, I suspect that Ted did not fully understand the implications of his position. But in what follows my refutation of linguistic idealism will rest on the recognition of states of affairs (facts1) .

So much for ramifications of linguistic idealism.

THE PROBLEM STATED IN A METAPHORICAL WAY

Let me approach the cause of the problems which bother Ted from a different perspective. Let us look at the current state of philosophy as an illness. The cause of this illness I am claiming is the doctrine of linguistic idealism, and the symptoms are the various doctrines which I have mentioned: denial of a theory-observation distinction, denial of the existence of alternative conceptual frameworks, denial of a given in knowledge, denial of foundations to knowledge, denial of the correspondence theory of truth, and so on. Using the analogy with an illness, my claim is that though Ted has not properly described the cause of the illness, he is fully cognizant of the damaging symptoms. And his efforts have been to alleviate the symptoms. Thus his cure has been palliate at best. My medical recommendation is to identify the causative agent (linguistic idealism) and kill it.

Let me gualify my remarks. I am not completely fair to Ted in saying that he has not found the cause of the illness. He has. He has understood the creature through its effects. But he has neither understood the creature as such, nor has he found a way to kill it.

THE PROBLEM STATED IN A LITERAL WAY

So, without further metaphors and analogies, what has Ted identified un-self-consciously as the cause of the current philosophical nihilism? Put simply, it is the claim that all knowledge is propositional. And I think he makes the correct inference that all those who maintain linguistic idealism have also claimed that this proposition that all knowledge is propositional entails that all knowledge is linguistically mediated.

ARGUMENT FOR LINGUISTIC IDEALISM

The argument for linguistic idealism is this:

  1. All knowledge is propositional.
  2. All propositions are composed of concepts.
  3. All concepts are linguistically mediated.

    Therefore,

  4. All knowledge is linguistically mediated. (Thesis of Linguistic Idealism)

(1), (2), and (3) are formulated by Ted in the following passages from his "The Epistemic Role of Qualitative Content":

(1') "Sellars' assumption that all knowledge is propositional . . . "

(2') "Propositions are complex entities which have concepts as constituents,"

(3') "Thus having a concept requires having a language."

(4) is an obvious derivation. For all practical purposes, Ted is aware of (4).

REFUTATION OF LINGUISTIC IDEALISM

The argument for linguistic idealism is valid. If the argument is not sound, it is because one or more of the premises are false. Ted's alleged refutation is to deny the truth of (1) and (3). My refutation consists of disambiguating (1), (2), and (3). Then, on one reading, the argument is sound; but on the other reading it isn't because (3) becomes false.

I will argue that Ted's alleged refutation fails because he denies the truth of (1) for the wrong reason. As a consequence, he also denies (3) for the wrong reason. But this creates for him a problem of reconciling a non-epistemic state of affairs with an epistemic state of affairs. For a solution, he resorts to theory of supervenience which he finds in James Van Cleve's article "Epistemic Supervenience and the Circle of Belief," to the effect that epistemic states supervene on non-epistemic states.11

I will limit myself to an examination of his rejection of (1). If my analysis is correct, then his reason for rejecting (3) is wrong, and there is no need to appeal to a theory of supervenience.

However, I want to say Ted's intuitions were right about (1) -- to the extent that he knew that he had to deal with it. We are therefore in agreement that there is something wrong with (1). Our disagreement is on how to deal with (1).

TED'S SOLUTION

In any case, Ted's strategy was to deny the truth of (1). He had claimed that this is a false statement. It is false because, for Ted, not all knowledge is propositional. In fact, he claimed that there are three types of knowing, which he called knowing that (this is his formulation of propositional knowledge), and there are the two additional types of knowing, namely: there is a knowing how to do something, which he takes from Gilbert Ryle's Concept of Mind; and there is a knowing what something feels like, which he takes from Michael Polanyi's The Tacit Dimension, and which he identifies with Bertrand Russell's 'knowledge by acquaintance'.12

His denial of (3) follows from his rejection of (1), because knowing what y feels like requires the possession of a concept, but not the possession of a language. In other words, there are, for Ted, non-linguistic concepts. Now, I agree with this conclusion, but for different reasons.

MY SOLUTION

I think that Ted is wrong about there being other types of knowledge, i.e., other than propositional. My view is that the problem with (1) is not that it is false tout court, as Ted thinks, but that it is false as usually interpreted. So my alternative to Ted's is to give (1) an interpretation; specifically I want to make, as the Scholastics used to say in debates, a distinguo. I want to distinguish two types of propositional knowledge. One type of propositional knowledge is indeed linguistically mediated, and hence (2) and (3) are correct, and (4) follows. The other type of propositional knowledge13 is not linguistically mediated, and, so (3) is false; hence, (4) does not follow.

I want to add that Ted had stumbled on the solution that I am proposing in his paper "Defense of the Correspondence Theory" when he distinguished facts1 from facts2,14 which, if properly interpreted, coincides with my solution. But because he does not utilize this distinction in his papers on the role of qualitative content, it makes me suspect that Ted was not fully aware of what he had done, or that he did not take stock of the systematic implications of his facts1 - facts2 distinction.

This is borne out by a second disagreement that I have with Ted. Ted accepts the truth of (2), while I accept it on one reading, and reject it on another.

So this is the skeletal difference between us which needs to be fleshed out.

My task will be to give arguments why Ted's alternative solution is not a solution, and to give a clarification and defense of my own solution, which is from one perspective the ramification of Ted's solution employing the facts1 - facts2 distinction.

CRITICISM OF TED'S SOLUTION

Ted's proposed solution to linguistic idealism (and I think it would be a solution if it were acceptable) is to deny (1). According to Ted not all knowledge is propositional. There is in addition to propositional knowledge, or knowing that p; a knowing how to do something, as knowing how to ride a bicycle; and there is a knowing what something feels like, as knowing what a pain feels like.

On reflection, I find Ted's discussion of this matter confused and confusing. The first problem is that I don't know how to formulate Ted's claim clearly. There is an ambiguity in his formulation which I don't think he attended to. He explicitly writes that there are three types of knowledge. He uses the word 'types.' This suggests that there is a genus, namely knowledge, and this genus has three species. But he also writes, following Ryle in this, that there are two categories of knowledge; namely, a knowing that and a knowing how. Here are Ted's words: "Ryle argues that propositional knowledge (knowing-that) and performative knowledge (knowing-how) belong to logically distinct categories because they are subject to different sorts of qualifications and accessible to different sorts of questions." But I think that this claim may be in contradiction with the sentence that precedes this one, namely: "What needs to be recognized here is that, as Ryle taught us, there are different types of knowledge." There may be a contradiction if Ted is using 'type' in the sense of 'species'. But then Ted may be using the word 'type' as Ryle sometimes does synonymously with 'category'. Its not entirely clear to me what Ted intends.

Now given these two different readings of what Ted is claiming, namely, on one reading, that there are three species under one genus, and on the second reading that there are in reality three categories, my comment is that on the first reading Ted is just wrong. As to the second reading, his claim becomes irrelevant as a criticism of (1). It is only if there are three species or types of knowledge that (1) is false.

So what reasons do I have for saying that Ted's claim that there are three types of knowledge is false? The main reason, I think, is this. If there is a species-genus relation in this situation, then it should satisfy a condition, which we can abstract from the following example. Let me formulate a clear-cut example of a species-genus relation to see what has to be satisfied. Think of the genus triangle, and the three species of triangles -- isosceles, right, and scalene. Note that we can give a definition of the genus in such a way that it applies to the species as well. A triangle can be defined as a plane figure enclosed by three straight lines. And this definition applies to a scalene, a right, and an isosceles triangle. Put abstractly the definitive properties of the genus are applicable to the species. Now if knowledge is a genus and it has the three species as Ted claimed, then there should be a definition of the genus so that it is applicable to the species.15 What could this definition be? At best, knowledge is some kind of capacity or ability. Even granting this, how is this cognitive ability to be distinguished from other abilities? I have, for example, the ability to imagine things. Is imagining a kind of knowledge? I don't think so, but I don't know how to specify the cognitive ability any more narrowly so that it will apply to knowing that, knowing how, and knowing what, but not to other abilities such as imagining. If it cannot be done, then we don't have some one category or genus with three species, but rather three different categories under the category of abilities. So I suspect that these are not three types or species, but three different categories. Ted, as I have mentioned, quotes Ryle to this effect that knowing that and knowing how belong to two different categories. But to say this is to deny that they are two types of knowledge. Let me just say that the argument I have presented is an argument from ignorance; so it is inconclusive. There may be an answer and I just haven't been clever enough to imagine it.

Be it as it may, I have other reasons for thinking that we are not dealing with species under one category but with different categories. This has to do with the technique of paraphrasing. It could also be called a reductive technique. My question is whether I can reduce Ted's three species of knowledge to three different categories. This will be done if I can eliminate the word 'know', yet preserve the sense of the analysandum in the analysans.

Try to paraphrase 'x knows that p', 'x knows how to do A', and 'x knows what y feels like'.

There are two uses of 'x knows that p' that I know. One use does not have a paraphrase, as in 'I know this is appearing red'. On this use knowledge and belief are contraries. The other use of 'x knows that p' is paraphraseable by the classical definition of knowledge as: 'p is true & x believes that p is true & x is adequately justified in believing that p is true'. Note that the word 'knowledge' in this use has been eliminated in favor of the word 'believe', which is to say that one type of propositional knowledge is a type of belief. So the category of knowledge has been reduced to the category of belief. What about a paraphrase of 'x knows how to do A'? This is paraphrased as 'x can do A' or 'x has the ability to do A'. And the attempt at paraphrasing 'x knows what y feels like' can be rendered as something like 'x has experienced y' or 'x remembers his experience of y'.

These paraphrases expose the categorial nature of knowing that, knowing how, and knowing what. I suppose that further reduction is possible, but again I don't see any common category ahead.

Let me cite one more consideration. This has to do with contrasts. Non-paraphraseable propositional knowledge contrasts with belief. While paraphraseable propositional knowledge, in the genus of belief, is some kind of correct belief as contrasted with an incorrect or erroneous belief. Now compare this with knowing how. What is there to correspond with believing correctly and erroneously? For example, riding a bike well and poorly? And in knowing what? Experiencing clearly and distinctly as contrasted with obscurely and indistinctly? Since the contrastive categories are different, the indication is that the original types belong to different categories as well.

From these three considerations, namely a search for a definition of a genus, attempt at paraphrase, and a look at contrasts, I conclude that there are not three types of knowledge, but three different categories.

EXPLANATION OF WHY THE WORD 'KNOWING' OCCURS IN 'KNOWING HOW' AND KNOWING WHAT'

I would have liked to give a satisfying explanation of why the word 'knowing' occurs in the phrases 'knowing how' and 'knowing what', but I don't have a good explanation. It may be that the uses of know in 'knowing that', 'knowing how' and 'knowing what' are equivocal in the Aristotelian sense. Aristotle distinguished three senses of 'equivocal': equivocal by chance, equivocal by reference, and equivocal by analogy. It may be that we have here a case of equivocation by chance. But I rather suspect that we are more likely involved with some kind of equivocation by reference or analogy. Aristotle's example of equivocation by reference is the word 'healthy'. In the literal sense, a living thing is healthy if all its physiological functions are functioning well. In an equivocal sense (by reference), food is healthy as a cause of a healthy animal, and urine is healthy as the product of a healthy animal. Although I have tried to work out an explanation along the lines of an Aristotelian theory of equivocation, I have not come up with a satisfactory explanation.

FORMULATION OF MY SOLUTION

My solution to the problem of refuting linguistic idealism is to interpret (1) in such a way that (4) does not follow. To illustrate by example what I have up my sleeve, I will put it this way. I claim that animals, such as rats, have propositional knowledge. And this does not entail that they have a language. If what I am saying is true, then there must be two species of propositional knowledge: a linguistic propositional knowledge, and a non-linguistic propositional knowledge. And I believe this is what Ted must have meant by his fact2 - fact1 distinction.

Immediately I must qualify my position. There are various types of propositional knowledge which are to be distinguished on the basis of the nature of the propositions involved. These could be divided into empirical and non-empirical propositions, and correspondingly empirical and non-empirical knowledge. Empirical knowledge could be subdivided into observational and theoretical knowledge. Non-linguistic propositional knowledge is to be located within the class of observational knowledge; specifically within perceptual knowledge; and more specifically, it must be identified with the class of basic perceptual knowledge.

A full defense of my proposal is too elaborate to present within the time limits at my disposal; instead I will provide a sketch of such a defense. A fuller defense of my position is worked out in my Ph. D. dissertation "A Critique of Wilfrid Sellars' Materialism" (Fordham University, 1990), which in essence is a critique of linguistic idealism.

The pertinent moves are the following:

(a) The crucial move is to distinguish a representational system from a language. This I did in the EPPA paper "Language, Thought, and Concepts in the Philosophy of Wilfrid Sellars" which I presented in the Fall of 1989 here at Bloomsburg University. It was an unmodified chapter from my dissertation. Animals, at least as developed as rats, must be credited with a representational system, but not a conventional language. If this is the case, then for the case of human beings the implication is that we are in possession of a representational system and, in addition, a conventional language. Everything else I have to say is a ramification of this distinction.

(b) The first ramification is that we have to distinguish two uses of the word 'concept'. One use is that common in psychology. Psychologists ascribe concepts to those animals which can discriminate and generalize. In this use, an animal may have a concept without possessing a language. Ted, in his paper on "Alternate Conceptual Schemes" agrees when he writes: "Yet the bat certainly organizes its experience according to some pattern and thus can be said to have a conceptual scheme, albeit somewhat rudimentary by our standards."

Let us use the neologism rs-concepts for what psychologists are describing. The prefix 'rs' is to remind us of representational systems.

By contrast, philosophers who want to use the word 'concept' to designate something peculiar to language users, deny that animals have concepts. This is true, for example, of Peter Geach and Wilfrid Sellars. They find the difference between ordinary representational systems and conventional languages in the occurrence of such logical connectors (syncategorematic terms) as 'not', 'or', 'if, 'and', 'because', etc. in conventional languages but not in representational systems. The distinctive feature of a conventional language, in the case of Sellars, is that a conventional language allows for logical inferences, whereas a simple representational system does not; instead it functions by principles of association. A concept is, then, defined as a word which is relevant to logical inferences. Animals, since they do not perform logical inferences, do not have concepts. Let us introduce 'l-concepts' for this use of the word 'concept' to remind us that it is dependent on the possession of language and logic.

(c) The second ramification of the representational system - conventional language distinction is that it requires crediting animals with propositional knowledge. It requires attributing to a representational system a structure which corresponds to a subject-predicate relation. A 'proposition', as I am using the term, is a class of sentences or sentence analogue which have a similar role in perceptions, inferences, and volitions. Thus a state of affairs by being analogous to a sentence is a proposition. Now this is a very controversial thesis. However, to deny it lands us back in linguistic idealism.

I just want to add at this point that, as I see it, the possession of a structure analogous to a subject-predicate form is neutral as to whether the creature has or does not have a language or l-concepts. The possession of a subject-predicate structure is a necessary, but not a sufficient condition for the possession of a language or l-concepts.

A representational system must have a sign vehicle. That sign vehicle is the sensation (Ted's qualitative content). And, to be consistent, Ted should have identified the qualitative content with the predicative component of his facts1, i.e. his states of affairs.

The best elaboration and defense of the thesis that sensations are the vehicles of a representational system and that they have a subject-predicate structure is worked out, in opposition to Wilfrid Sellars (ironically with Sellarsian techniques), in a series of papers by Romane Clark. He specifies his difference with Sellars by way of how they handle the following triad of claims:

(1) Perceptions are judgments [thoughts].

(2) Sense impressions are constituents of perceptions.

(3) Sense impressions are not conceptual or cognitive items.

He goes on to claim that Sellars would accept (1) and (3) but deny (2). Clark, on the other hand, accepts (1) and (2) but denies (3) .17

Clark's contribution is to take Sellars' view that thoughts are analogous to speech, and to extend this analogy to sense impressions. Speech, thoughts, and perceptions are analogous in three respects: all are intentional, all have truth values, and all are subject to reflexive, non-inferential knowledge.

The model of sense impressions ... is simply this: as predicates are to declarative sentences, as ascriptions are to assertions, as concepts are to judgments, as ascribing is to stating, so, too, sense impressions are to perceptions. . . . Sense impressions are the predicate "words" of perceptual, mental "assertions." . . . Thus, the occurrence of a sense impression is the predication of a sense quality to an object, and so it is in this way that sense impressions are cognitive or conceptual elements of sensuous beliefs.18

Clark explicitly states that his analysis applies only to 'basic perceptions':

These are basic in two senses: First, they are ascriptions of simple sensuous qualities. Second, they have no internal, logical complexity. Each is simply a (single) qualitative ascription to a (single) object of reference. 19

and, "Basic perceptions we understand to be those sensuous judgments which ascribe the proper and common sensibles to what is demonstratively before one."20 Clark notes that Everett Hall had made a similar claim, except that he wrote that "perceptions are descriptive or predicative throughout." To which Sellars responded:
how [can] a pure perception. . . be a sentence, and yet be 'predicative throughout. . . . Must not pure perceptions contain expressions referring to an object in order to be able to characterize an object?21

To this objection Clark replies that Sellars is correct on insisting on a referring component, the reference is carried demonstratively: "It is the material occurrence of the impression in the given context which provides the demonstrative reference of that experience,"22 and he also puts it this way:

Sense impressions are rather\special functions for they are defined only over a special and limited domain. They are functions taking the demonstrative references embodied in their own occurrences as arguments.23

Clark adds that Sellars1 perspicuous language, Jumblese, provides the ideal model for basic perceptions. In this language there are no predicate expressions; the names are simply written in different styles. For example, in Jumblese the sentence 'John is happy' could be written as 'John', where the predicate happy is conveyed by the way 'John' is written -- in some font, some size, location, degree of rotation, and such.

At this point, I would like to introduce the following regress argument for the existence of a non-linguistic subject-predicate form. If the recognition of a subject-predicate form is language dependent, then how is it possible to be aware of linguistic items themselves, as, for example, that 'this is the word "red"'? If the answer is that linguistic awareness presupposes meta-linguistic awareness, then obviously metalinguistic awareness presupposes meta-meta-linguistic awareness, and so on ad infinitum. The conclusion I draw is that a subject-predicate form of awareness does not presuppose a language.

(d) The next move would be to consider the role of sensations (or qualitative content, as Ted would put it) in a representational system, followed up by a consideration of how a representational system is related to a conventional language. And it would require a comment on Kant's dictum that concepts without intuitions are empty and intuitions without concepts are blind.

CONCLUSION

Unknowingly, Ted's writings have been, for the most part, attempts to deal with the problems created by linguistic idealism. He has stumbled on the thesis of linguistic idealism in the formulation of the three premises. His alleged solution claiming that (1) is false, as I have argued, is not a solution. The correct solution is buried in his paper on the correspondence theory of truth. The solution requires a distinction between non-linguistic and linguistic propositional knowledge, or, as Ted put it, a fact1 - fact2 distinction.


Notes

1 Theodore W. Schick, Jr., "Can Fictional Literature Communicate Knowledge?" The Journal of Aesthetic Education 16 (1982): 31-39.

2 Theodore W. Schick, Jr., "The Semantic Role of Qualitative Content," The Southern Journal of Philosophy 27 (1989): pp. 125-133.

3 Theodore W. Schick, Jr., "The Epistemic Role of Qualitative Content," to be published in The International Philosophical Quarterly.

4 Theodore W. Schick, Jr., "Rorty and Davidson on Alternate Conceptual Schemes," The Journal of Speculative Philosophy 1 (1987): pp. 291-303.

5 Theodore W. Schick, Jr., "In Defense of the Correspondence Theory," Philosophy Research Archives 11 (1986): pp. 319-334.

6 Theodore W. Schick, Jr., "How is Philosophy Possible?" Proceedings of the Eastern Pennsylvania Philosophical Association, April 15, 1989: pp. 22-37.

7 Linguistic idealism could also be called, by analogy to the 'ego-centric predicament', the 'logo-centric predicament'.

The source of the current preoccupation with linguistic idealism seems to be the notorious passage in Wilfrid Sellars' "Empiricism and the Philosophy of Mind," in Science, Perception, and Reality (London: Routledge and Kegan Paul, Ltd., 1963), p. 160:

all awareness of sorts, resemblances, facts, etc., in short, all awareness of abstract entities -- indeed, all awareness even of particulars -- is a linguistic affair.

Commentators have taken it literally to express the formula: all awareness is linguistic -- a formula reminiscent of esse est percipi. If we may view the latter formula as the basis for empirical idealism, then Sellars' alleged formula can be viewed as a basis for linguistic idealism. The implication of linguistic idealism is that all thoughts are linguistic.

But an interpretation attributing linguistic idealism to Sellars cannot be correct because Sellars ascribes thoughts to animals and pre-linguistic children. Therefore, the formula which should be ascribed to Sellars is: All conceptual awareness is linguistic.

8 According to Alan R. White, Truth (Anchor Books, 1970), "The fundamental supposition in all versions of this theory is that logical equivalence is the same as equivalence in meaning." p. 92.

9 Paul Churchland, Matter and Consciousness (Cambridge: MIT Press, 1984).

10 Wilfrid Sellars, "Ontology and the Philosophy of Mind in Russell," in Bertrand Russell's Philosophy, ed. George Nakhnikian (1974), p. 100.

11 The current preoccupation with supervenience is a rediscovery of what used to be called "emergent properties," as in the view of Emergent Materialism. Emergent Materialism also goes under the labels Dialectical Materialism and Non-Reductive Materialism.

12 The first and most explicit formulation and strongest defense of this tripart distinction of knowledge occurs in Ted's "Can Fictional Literature Communicate Knowledge?"

13 My use of the term 'propositional' may be misleading. What I have in mind is a knowledge of a state of affairs.

14 Ted Schick, Jr., "Correspondence Theory," p. 324.

15 This requirement would be objected to by Wittgenstein. From the fact that several things are called by one name does not mean that they have anything in common. They may form a family resemblance of properties. This is true if we are not dealing with a species-genus relation.

16 Ryle himself in the essay "Knowing How and Knowing That" wrote "I now want to prove that knowing-that presupposes knowing-how." p. 224, in Collected Papers, Vol. 2 (Barnes & Noble, 1971), which contradicts Ted's claim that the three types of knowledge are independent of each other.

17 Romane Clark, "The Sensuous Content of Perception," in Action, Knowledge and Reality, ed. Hector-Neri Castaneda (Indianapolis: The Bobbs-Merrill Co., 1975), p. 110-111.

18 Clark, "The Sensuous Content of Perception," p. 117.

I have recently discovered that L.T. Hobhouse, writing in 1896, had defended such a primitive form of cognition (of the given). And more interesting, had defended this position from the attacks of Hegel and Thomas Green. See his The Theory of Knowledge (London, 1896; New York: AMS Press, 1970).

19 Ibid., 122-123.

20 Romane Clark, "Sensuous Judgments," Nous 7 (1973), 53.

21 Wilfrid Sellars, "The Intentional Realism of Evert Hall," in Philosophical Perspectives, 110.

22 Ibid., 123.

23 Clark, "Sensuous Judgments," 54.