THE ALLEGED FALLACY OF THE SENSE-DATUM INFERENCE

THE ALLEGED FALLACY OF THE SENSE-DATUM INFERENCE
Andrew Chrucky

Paper read before the New Jersey Regional Philosophy Conference (Seton Hall University, 1992) and the Eastern Pennsylvania Philosophical Association (Bloomsburg University, 1992).

Sense-data, if they exist, could conceivably provide foundations for empirical knowledge. Those who are opposed to empirical foundationalism are therefore also prone to reject sense-data and arguments for their existence, e.g., Rorty, Bonjour; while foundationalists are prone to accept the existence of sense-data, e.g., Russell, Ayer, Broad, Price, Lewis. An exception to this is the position of Roderick Chisholm who accepts empirical foundationalism but rejects the existence of sense-data.

In this paper I wish to examine Chisholm's attempt at refuting what he calls the sense-datum inference. His earliest attack on the sense-datum inference appears in "The Theory of Appearing."¹ Chisholm claims here that what he calls "the sense-datum inference" is a fallacy. Similar attacks on the sense-datum inference recur in his Perceiving and his Theory of Knowledge. I say that they are similar and not the same because Chisholm's discussion contains several ambiguities. Despite this, if any version of the sense-datum inference survives his attacks, then Chisholm has not shown that the sense-datum inference in all its forms is a fallacy. I will defend the validity of one form of the sense-datum inference. In doing so I am in large measure borrowing and expanding the arguments of Frank Jackson as found in his book Perception.²

As to Chisholm's current thoughts about sense-data, the situation is not clear. Given the present ferment of anti-foundational writings, it may be Chisholm's strategic ploy to minimize the difference between the sense-datum position and his own adverbial view. This shift is apparent from a comparison of the first edition of his Theory of Knowledge with the third edition. Where the first edition was keen on refuting the sense-datum inference, the third edition is silent about the topic. Whatever Chisholm's present thoughts on the subject, in his earlier writings he was vehemently against the validity of the sense-datum inference.

The attack is succinctly formulated in the following passage:
[The sense-datum philosopher] contends that from statements in the language of appearing we can deduce statements in the sense-datum language. That is to say, he contends that from statements such as: (A) There exists something which appears diamond-shaped, we can deduce statements such as: (S) There exists something which is diamond-shaped. . . . We may call this the sense-datum inference. And we may say that the sense-datum theory is the view which countenances this inference, whereas the theory of appearing is the view which does not.³
Two reasons are offered for not countenancing the inference. The first is that the inference is deductively invalid. The second is that it is inadvisable.

Let me comment on these two reasons for rejecting the sense-datum inference. One would think that the inference would be inadvisable if it were invalid; so it is puzzling why two criteria are being introduced for evaluating the supposedly single inference. If there is only one inference involved it should be sufficient to reject the inference for being invalid, and there is no need to bring in the other consideration of inadvisability. So the introduction of a second criticism of inadvisability suggests that the inference is also being construed not as a deduction but as something else. The implication is that the sense-data inference can also to be viewed as an abduction, specifically as an inference to the best explanation or best hypothesis.

In fact, in the paper under consideration Chisholm ignores criticizing the inference as a deduction. He rather discusses only the inadvisability of the sense-datum hypothesis. And he offers two reasons against the accepting the sense-datum hypothesis: (1) In accepting it, he writes, "we multiply entities beyond necessity;" and (2) we create a host of recalcitrant metaphysical problems. One example of such a recalcitrant metaphysical problem is Chisholm's speckled-hen problem. In any case, he argues for the preferability of what he calls the theory of appearing (his adverbial theory) as a hypothesis. His reasons are that (1') The theory of appearing does not multiply entities without necessity, and (2') it does not create recalcitrant metaphysical problems.

Whatever the merits of this line of criticism on the basis of inadvisability of sense-data as theoretical entities, I will not examine it; I want to restrict myself to an examination of the criticism of the sense-datum inference as a deduction.

We find arguments against the sense-datum inference as a deduction in some of Chisholm's other writings; notably in Perceiving and in his Theory of Knowledge.

Now I find in Chisholm's writings four different formulations of the sense-datum inference.

M appears P to S; therefore, something is P.
M appears P to S; therefore, there is an appearance which is P.
M appears P to S; therefore, S senses an appearance which is P.
M appears P to S; therefore, S sees an appearance which is P.
These may or may not be different claims, depending on how they are interpreted. They all involve an ontological inference; but (3) and (4) could additionally be construed as containing epistemic inferences. However, following Dretske, 'sensing' and 'seeing' could also be construed in non-epistemic senses, so that all four claims can be interpreted as strictly ontological inferences.

Omitting niceties about non-epistemic senses of perception words, I will restrict my examination of the sense-datum inference to scheme (1) and (2).

Chisholm's refutation of schemas (1) and (2) consists of claiming that arguments in these forms are not valid because they have the following counterexamples.

'That animal looks centaurian' entails 'There is something centaurian'.
'The pail feels empty' entails 'There is an appearance which is empty'.
'The woods sound inhabited' entails 'There is an appearance which is inhabited'.
'The curtain appears green' entails 'There is appearance which is green'.⁴
Elsewhere, he gives the following counterexamples:

'The man appears tubercular' entails 'The man presents an appearance which is tubercular'.
'The books appear worn and dusty and more than two hundred years old' entails 'The books present appearances which are worn and dusty and more than two hundred years old'.⁵
Chisholm thought that with these counterexamples he had shown the invalidity of the sense-datum inference. I disagree.

What Chisholm accomplishes with these counterexamples is to alert us to an ambiguity is the uses of appear words. And the conclusion that should be drawn is that not all uses of appear words sanction the sense-datum inference.

That Chisholm did not come to this same conclusion is difficult to understand because in Perceiving, in which he offers some of these counterexamples to the sense-datum inference, he also presents the first systematic attempt in the literature to disambiguate the different uses of 'appear' words.
APPEARANCE TALK
In Perceiving, Chisholm distinguished three uses of appear words.⁶ The first use of appear words he calls 'epistemic', which is distinguished by sanctioning the inference from 'D appears to be F to S' to 'S believes that D is F'. The role of 'appears' in this use is synonymous with 'seems'.⁷ It is used primarily as an expression of judgment, e.g., 'It seems that D is F' is equivalent to 'I judge that D is F'. I should add that Chisholm distinguishes what I take to be a species of the judgmental use of appear words. This is the hedging use of appear and seem statements: as when I make a report but do not wish to take responsibility for its accuracy or truth. Let us introduce 'appear_e' to distinguish the epistemic use.

The second use of appear words distinguished by Chisholm is the comparative use. Using the comparative sense, from sentences such as 'D appears F to S,' we can infer 'D appears like F-things usually do (in standard or optimal contexts)'. Let us introduce 'appear_c' to distinguish the comparative use.

The third use of appear words is noncomparative. This use does not allow the types of inferences sanctioned by the epistemic or comparative uses. For example, if 'D appears F to S' and S suspects, let's say, some abnormality in the lighting, then S may not believe that D is F. As to the relation of phenomenal uses of appear words to the comparative uses, the situation seems to be that comparative uses of appear words presuppose noncomparative uses.

However, I also find that Chisholm's noncomparative use of appear words has two species which Chisholm does not distinguish at the appropriate place of his discussion. One species can be identified with what Sellars generically refers to as 'ostensible perception', which has the subspecies 'ostensible seeing', 'ostensible hearing', etc., to correspond to various sensory modes. To say that there is an ostensible perception is to indicate a suspension of judgment as to whether the content of a perceptual report is true or not. Normally a perceptual report implies that the report is true. If this implication is suspended, we have a case of 'ostensible perception'. We could also characterize it as the use of a perceptual report with the achievement sense suspended. Firth's use of the term 'percept' in his "Sense-Data and the Percept Theory", I take it, is synonymous with Sellars' 'ostensible perception'. Let us introduce 'appear_o' to distinguish ostensible perceptions.

The other subspecies of noncomparative uses of appear words is what Chisholm in his Theory of Knowledge calls the "descriptive, phenomenological use."⁸ The objects of phenomenological appearances are, Chisholm tells us, the proper and common sensibles of Aristotle.⁹ Frank Jackson calls this the phenomenological or, for brevity, the phenomenal use, which is characterized "by being tied to terms for color, shape, and/or distance."¹⁰ Let us introduce 'appear_p' to distinguish the phenomenal use.
CRITICISMS OF THE DISTINCTIONS

Those who are opposed to the introduction of sense-data as phenomenologically given (and hence as sufficient to a version of a foundationalist view of empirical knowledge) have realized that the first step in the defense of sense-data is to recognize the distinct character of phenomenal appearance talk, and have tried either to eliminate phenomenal talk altogether or reduce it either to the epistemic or the comparative use, or have claimed that phenomenal talk is inscrutable. Jackson mentions the following philosophers who take these approaches: J. Bennett, J. W. Roxbee-Cox, G. Pitcher, and D. Armstrong. According to Jackson, Bennett offers a radical reduction to the comparative form. He claims that 'x appears y to S' implies 'x produces in S the kind of sensory state S usually is in when he perceives x as y'.¹¹ Pitcher and Roxbee-Cox offer an epistemic reduction, as follows: 'x appears y to S' implies 'S acquires belief z' or 'S causally receives belief z' or 'S believes z'; 'x appears y to S' implies 'S believes x is y' or 'S is inclined to believe that x is y' or 'S has a suppressed inclination to believe that x is Y'.

According to Jackson, Armstrong offers a counterfactual epistemic analysis. He says that 'x appears y to S' is equivalent to 'But for the fact that S had other, independent, beliefs S would have acquired the belief that x is y'.¹²

I would add that Richard Rorty, too, offers an epistemic reduction. He writes:
For the appearance-reality distinction is not based on a distinction between subjective representations and objective states of affairs; it is merely a matter of getting something wrong, having a false belief.¹³

Another attack on the phenomenal use of appear words is to take a skeptical stance about the possibility of demarcating them. We find Dretske, who is probably thinking of Goodman's grue paradox, expressing the following skepticism: "I know of no non-circular way to specify what is to count as a visual characteristic."¹⁴
Also Sellars denies a phenomenological or phenomenal use of appear words. In his polemics against Chisholm, Firth, and Cornman, Sellars claims that the phenomenal use of appear words is either synonymous to 'ostensible perception', or that such a use of appear words does not exist, or that it is a contrived use to describe non-conceptual states.

Chisholm, by contrast, cites Aristotle's proper and common sensibles as the appropriate occurrent sensorial properties for phenomenal terms. I am not going to argue against any of the above reductive approaches. I will assume with Chisholm that there is a distinct irreducible phenomenological use of appear words.
Now given the existence of Chisholm's four different uses of appear words, the problem which Chisholm should have posed for himself is whether the phenomenological use of appear words warrants the validity of the sense-datum inference. And the only counterexample which would have been suitable for this consideration is:

'The curtain appears green' entails 'There is an appearance which is green'
All of the other counterexamples cited by Chisholm seem to be cases of comparative uses of appear words.

'That animal looks_c centaurian' entails 'There is something centaurian'.
'The pail feels_c empty' entails 'There is an appearance which is empty'.
'The woods sound_c inhabited' entails 'There is an appearance which is inhabited'.
'The curtain appears_p green' entails 'There is appearance which is green'.
'The man appears_c tubercular' entails 'The man presents an appearance which is tubercular'
'The books appear_c worn and dusty and more than two hundred years old' entails 'The books present appearances which are worn and dusty and more than two hundred years old'.
Only example (d) exhibits a noncomparative, phenomenal use of an appearance word. And example (d), smuggled in with the other examples, is made to appear illegitimate by association. But we can resist this attempt at contamination. I find nothing wrong with example (d).

So modified, the only objection Chisholm has to the sense datum inference is that the conclusion contains the reifying words 'something' or 'appearance'.

In the third edition of his Theory of Knowledge, Chisholm is more specific about the nature of this reification. He objected to interpreting 'something' as referring to something external. With this I agree. But this objection does not preclude interpreting the word 'something' in what the scholastics called a transcendental sense, as applicable across categories.

Chisholm's own counter proposal is, in effect, to modify the formulation of the conclusion so as to exclude substantive terms altogether. He opts for an adverbial formulation. Thus, example (d) would be modified to 'The curtain appears green' entails 'I am appeared to greenly'.¹⁵

This brings the discussion to a different question -- the question of the preferability of the adverbial theory to the sense-datum theory. And this would require a new set of arguments. What I have set out to do in this paper is to argue that Chisholm has not shown that the sense-datum inference is a deductive fallacy, and this I have done.
Addendum (July 1998): In The Philosophy of Roderick Chisholm (1997), Chisholm, in the autobiographical section, writes that there were two "Copernican revolutions" in his philosophical development. One concerns his view of appearances; the other, the nature of events. He writes: "We here defend the view that appearances are individual things." (p. 35) And in the footnote #16 to this sentence, he writes, "I thus reject the "adverbial theory of appearances"." (p. 40)

Notes
¹ R. Chisholm, "The Theory of Appearing" in Philosophical Analysis, ed. Max Black (Prentice-Hall, 1963). Reprinted in R. J. Swartz, ed., Perceiving, Sensing, and Knowing (New York: Doubleday, 1965).
² Frank Jackson, Perception: A Representative Theory (London: Cambridge University Press, 1977).
³ Chisholm, "The Theory of Appearing," p. 173.
⁴ Roderick Chisholm, Perceiving: A Philosophical Study (Ithaca: Cornell University Press, 1957), pp. 115-116.
⁵ Roderick Chisholm, Theory of Knowledge (Prentice-Hall, 1966): p. 95.
⁶ Chisholm, Perceiving, ch. 4.
⁷ This was also pointed out by John Austin, Sense and Sensibilia (New York-: Oxford University Press, 1964): pp. 36-37.
⁸ Chisholm, Theory of Knowledge, p. 32.
⁹ Ibid., p. 32.
¹⁰ Jackson, Perception, p. 33.
¹¹ Ibid.
¹² Ibid., p. 40.
¹³ Richard Rorty, Philosophy and the Mirror of Nature (New Jersey-. Princeton University Press, 1979), p. 77.
¹⁴ Dretske, Seeing and Knowing (University of Chicago Press, 1969): p. 65.
D.M. Armstrong distinguishes between natural and unnatural predicates. He tells us that 'green' is a natural predicate; 'grue' an unnatural one. For a predicate 'F' to be natural "The observed Fs and the unobserved Fs must be genuinely the same, or at least objectively similar where similarity is cashed in terms of universals." What is a Law of Nature? (London: Cambridge University Press, 1983), p. 58.
¹⁵ Ibid., p. 34.