1 Part of a draft of this paper was read at the June, 1973, meetings of the Canadian Philosophical Association in Kingston, Ontario; another part was read at the December, 1973, meetings of the Eastern Division of the American Philosophical Association in Atlanta, Georgia. This work was supported in part by a SUNY Faculty Research Fellowship.

2 Ontological Relativity and Other Essays (henceforth abbreviated in the text as “OR“) (New York, 1969), p. 87. I shall be citing the following additional work by Quine, CrossRefGoogle Scholar:

(RR) The Roots of Reference (La Salle, III., 1973).

(MR) “Methodological Reflections on Current Linguistic Theory,” in Semantics of Natural Language, ed. by Davidson, and Harman, (2nd ed.; Boston, 1972CrossRefGoogle Scholar).

(GT) “Grades of Theoreticity,” in Experience and Theory, ed. by Foster, and Swanson, (Amherst, 1970Google Scholar).

(WB) The Web of Belief, with Ullian, J. S. (New York, 1970)Google Scholar

(PL) Philosophy of Logic (Englewood Cliffs, 1970).

(PPLT) “Philosophical Progress in Language Theory,” Metaphilosophy (1970).

(RIT) “On the Reasons for Indeterminacy of Translation,” Journal of Philosophy (1970).

(R) “Replies,” in Words and Objections, ed. by Davidson, and Hintikka, (New York, 1969)Google Scholar.

(SM) “Stimulus and Meaning,” in The Isenberg Memorial Lecture Series: 1965-1966 (East Lansing, 1969).

(WP) The Ways of Paradox and Other Essays (New York, 1966).

(WO) Word and Object (Cambridge, Mass., 1960).

3 P. 8 of Hanson, Norwood Russell, Patterns of Discovery (Cambridge, 1958)Google Scholar─as quoted in Scheffler, Israel, Science and Subjectivity (Indianapolis, 1967), pp. 14–15Google Scholar. Chapter One of the latter book provides an excellent perspective upon views commonly attributed to Hanson, Kuhn, and Polanyi.

4 While Quine is prone to say (e.g., WB 16) that for any given speech community this criterion picks out all and only the ostensively learnable sentences, it actually picks out somewhat more than the ostensively learnable sentences. This was brought to my attention by Edwin Martin, Jr. in his article, “The Intentionality of Observation,” Canadian journal of Philosophy (1973), p. 124. He says there that for the community of people who speak English, the one word sentence ‘Red'

is both intersubjectively agreed upon and could be learned ostensively. 'Unred', or ‘Not red', thus will also be intersubjectively agreed upon, for disagreement here would doubtless be reflected in disagreement over ‘Red’ and there is no disagreement over ‘Red'. But ‘Unred', Quine is inclined to say [see GT 11], could not be learned ostensively.

5 Peter Geach gives the following brief characterization of mental acts:

In historical or fictional narrative there occur reports, not only of what human beings overtly said and did, but also of what they thought, how they felt, what they saw and heard, and so on; I shall call the latter kind of reports “reports of mental acts.” The psychological character … of mental acts is expressed by the use of various psychological verbs, such as “see”, “hear”, “hope”, “think”. (Mental Acts, p. 1)

6 “I use stimulations in meaning-and in meanings primarily of observation sentences. Reference, even on the part of observation terms, is in my view theory-enveloped and thus subject to the indeterminacy of translation” (R 299). (In this paper, I shall leave uncontested Quine's questionable assumption that it makes sense to attribute truth or falsity to observation sentences taken holophrastically. On this issue see S. G. Harding, “Making Sense of Observation Sentences,” forthcoming in Ratio.)

Observation sentences, which he has attempted to define with careful avoidance of making reference to reference, are, Quine avers, unequivocally determinate.

The predicament of the indeterminacy of translation does not extend down to observation sentences …. The equating of an observation sentence of our language to an observation sentence of another language is the merest matter of empirical generalization. (SM 57; see also, e.g., PPLT 12)

Incidentally, for excellent discussion of the role of Quine an observation sentences in translation, see J. Margolis, “Quine on Observationality and Translation,” Foundations of Language (1968).

7 Quine himself elsewhere acknowledges the essential point: “Since I propose to dodge the problem of defining observation by talking instead of observation sentences, I had better not define observation sentences as sentences that report observations” (R R 39 ).

8 Strawson, P. F., “Persons,” repr. in Introduction to the Philosophy of Mind, ed. by Morick, (Glenview, 1970), p. 302Google Scholar.

9 I should here perhaps note that Morick's tokening of (3’), “Harold Morick thinks Harold Morick sees water,” no less than his tokening of (3), “I think I see water,” is what Quine calls an introspective report, a report by a speaker on his own current state of mind (see WB 20).

10 It is important to keep in mind that in the preceding discussion (in the text) we were actually considering not the possibility of restricting the notion of an observation sentence to sentences entirely free of indexicals, but rather the possibility of restricting that notion to something cooked up to improve Quine's criterion. The sentences considered were not indexical-free; they were only personal pronoun-free sentences. There can be completely indexical-free observation sentences only on pain of self-contradiction, as I shall try to make plain.

Quine has noted that for every indexical sentence ─ be it observational like, e.g., “It is raining,” or not, e.g., “You owe me ten dollars“─ there is a corresponding “eternal” (i.e., indexical-free) sentence:

Corresponding to ‘It is raining’ and ‘You owe me ten dollars’ we have the eternal sentence ‘It rains in Boston, Mass., on July 15, 1968’ and ‘Bernard J. Ortcutt owes W. V. Quine ten dollars on July 15, 1968', where ‘rains’ and 'owes’ are to be thought of now as tenseless. (PL 13)

I believe that while such correspondences between indexical sentences and eternal sentences may be used to some philosophical purpose or another, they cannot be used to expunge subjectivity from observation sentences in Quine's sense, for the simple reason that by Quine's definitions eternal sentences are tenseless, whereas observation sentences are tensed. Unlike eternal sentences, observation sentences, along with those who drink Pepsi, are members of the Now Generation: they are tied to current stimulations.

Commonly an observation sentence will cease to be an observation sentence when we change only the tense of its verb. For the remembrance of a past observation is not itself an observation. ( WB 19)

For those familiar with Quine's terminology we can make essentially the same point by noting that: observation sentences are “occasion” sentences, eternal sentences are “standing” sentences, and nothing is both an occasion sentence and a standing sentence:

The observation sentence is a species of occasion sentence. (SM 7)

Eternal sentences are standing sentences. (WO 193)

Occasion sentences, as against standing sentences, are sentences … which command assent or dissent only if queried after an appropriate prompting stimulation. (WO 35-36)

It is also worth adding that the whole idea of personal pronoun-free but otherwise indexical observation sentences seems to be completely arbitrary. It seems arbitrary to grant observationality to

(a) There's a six-footer,

but to deny it to

(b) He's a six-footer.

Finally, let me briefly draw attention to yet another qualification Quine would have to make were he to hope to deny observationality to introspective reports by means of disallowing the occurrence of certain pronouns within observation sentences. “It” would have to be a forbidden word along with the personal pronouns, or else the introspective report

(c) It hurts

would pass as observational. But if Quine were to forbid “It” from being a part of observation sentences, he would have to give up saying, as in fact he does say, that

(d) It is raining

counts as an observation sentence. It is difficult to imagine how Quine could escape this unhappy conclusion without invidiously distinguishing between the “It” in (c) and the “It” in (d).

It would be idle to adduce any more reasons why Quine couldn't plausibly deny observationality to introspective reports by way of banning all or some kinds of indexicals or pronouns. (Of course, it should be noted, Quine has shown little tendency towards making such a move. Where, e.g., Neurath sought context-free canonical protokollsätze, Quine has on the whole been content to count as observation sentences ordinary sentences like “It is raining,” “This is water,” “There is water here,” and [ WO 35, 42] “His face is dirty.“)

11 “The stimulus meaning of a sentence, for a given speaker, is the class of all stimulatory situations in the presence of which he will assent to the sentence if queried” (MR 450 [emphasis mine]).

12 I here oversimplify a bit, for actually Quine does not have a coherent position on whether or not reports like (9)─i.e., reports of relations between percipients and external things─are observation sentences. Sometimes he allows that they are:

Typical observation sentences are ‘Red’ (or ‘This is red', ‘I see red’), ‘Rabbit', 'It is raining'. (RR 40 [Emphasis mine.])

And sometimes he doesn't:

Physical theory is underdetermined even by all possible observations. Not to make a mystery of this mode of possibility, what I mean is the following. Consider all the observation sentences of the language …. Apply dates and positions to them in all combinations, without regard to whether observers were at the place and time. (RIT 179 [All italics except the first are mine.])

13 If the reader prefers, let him substitute sentence (4) or (5) for (10).

14 Quine's position would degenerate into the following sort of circularity:

Question: Why do you count “Rabbit” as observational in your sense?

Answer: Because all members of the adult community at large assent to “Rabbit” under rabbitoid surface stimulations.

Question.: Who qualifies for that community?

Answer: Only those who assent to “Rabbit” under rabbitoid surface stimulations.