C H I S H O L M ' S C R I T E R I A O F I N T E N T I O N A L I T Y

JESSE L. YODER

1.1 Introduction In ten t iona l i ty has been proposed as one cri ter ion for ident ifying

men tea phenomena , and for dis t inguishing menta l f rom physical phenomena . This criterion, originally proposed by F ranz Brentano, ~ is somet imes fo rmula ted as a cr i ter ion for ident i fying in ten t iona l sentences. Much recent discussion of this cr i ter ion centers on articles by Roder ick Chisho lm3 In part icular , Chisholm has proposed three cri teria of in tent ional i ty . In Sect ion 1.2 1 state a recent f o rm u la t i on of these criteria. In Sect ion 1.31 examine recent discussion of these criteria by Alber t Flores 3 and Auson io Marras. 4 In Sect ion 1.4, I discuss a re fo rmula t ion by Chisholm of his criteria in response to criticisms by Flores and others. In Sect ion 1.5 I examine a re fo rmula t ion of Chisholm's criteria by Anson io Marras. I conclude that nei ther of Chisholm's fo rmula t ions of his criteria no r Mar ras ' revised vers ion is successful in ident ifying all and only in ten t iona l sentences.

1.2 Chisholm's Criteria for Intentional Sentences Chisho lm original ly states his three criteria for in ten t iona l i ty in

"Sentences A b o u t Believing ''5 and in Perceiving: A Philosophical Study 6. He restates these three criteria in a recent discussion: 7

I had suggested a kind of criterion of the psychological. A phenomenon is psychological, I said, if we cannot adequately describe it in our ordinary language without using sentences that are intentional. And then I formulated three criteria of intentionality, (1) A sentence is intentional if it uses a substantival expression in such a way that neither the sentence nor its negation imply that there is, or that there isn't anything, to which that expression truly applies. (2) A sentence is intentional if it has a subordinate clause such that neither the sentence nor its negation imply that the subordinate clause is true or that it is false. And (3) a sentence is intentional if it has a context that is referentially opaque. Then I added that a compound sentence is intentional provided one of its components is intentional.

Here Chisholm restates his three original cri teria for a sentence's

297

JESSE L. YODER

being intentional. Apparently Chisholm intends each criterion to be a sufficient condition for identifying psychological phenomena. The criteria cannot be individually necessary since they apply to different types of sentences. Albert Flores cites these sentences as examples of intentional sentences by each of Chisholm's three criteria: 8

Criterion 1: 1. Quine looked for the proper characterization of analyticity.

Criterion 2: 2. Othello believes that Desdemona loves Casio. Sentences 1 and 2 are each intentional according to the first two of Chisholm's criteria respectively. Sentence 2 is also intentional by Criterion 3 in virtue of the truth of the following identity statement:

3. Desdemona is the woman framed by Iago. Sentence 1 is intentional only by Criterion 1. The following compound sentence is intentional by Criterion 4:

4. Othello believes that Desdemona loves Casio or snow is white. The most correct interpretation of Chisholm's criteria, then, is that

the disjunction of the four criteria is a necessary condition for identifying psychological phenomena, while each condition is individually sufficient. This allows for the possibility that a sentence can be intentional by more than one criterion, but need not be intentional by more than one.

1.3 Criticisms of Chisholm's Criteria One criticism that has been raised against Chisholm's intentional

criteria for psychological phenomena is that sentences that contain modal terms such as 'possible' and 'necessary' satisfy one or more of Chisholm's criteria yet are not used to describe psychological phenomena. Albert Flores cites the following example: 9

5. It is necessary that Alaska is Alaska. We can assume the truth of the following identity statement:

6. Alaska is the 49th. state of the union. Substitution in sentence 5 results in the following statement:

7. It is necessary that Alaska is the 49th. state in the union. Sentence 5 is intentional by Criterion 3, yet it is not used to describe a psychological phenomenofi.

Flores cites other sentences that have been cited as counterexamples to Chisholm's first two criteria90

8. The fire needs some coal. 9. The frost may bring it about that the cliff will fall.

Sentence 8 is intentional by Criterion 1 since neither 8 nor: 10. The fire does not need some coal.

298

INTENTIONALITY

implies that any coal is available, or even exists. Sentence 9 is intentional by Criterion 2 since neither 9 nor its negation:

11. The frost may not bring it about that the cliff will fall. implies that the t ruth or falsity of the subordinate clause.

Chisholm's response to examples like these and others is to claim that the alleged counterexamples can be t ransformed into equivalent non- intentional sentences: H

These sentences are not examples counter to our thesis. Anyone who understands the language can readily transform them into conditionals which are not intentional... Instead of using intentional sentences, we could have said, 'If there should be any new epidemics, the patient would not be affected by them'... I believe that any other ostensibly non-psychological sentence which is intentional can be transformed, in an equally obvious way, into a sentence conforming to our version of Brentano's thesis. That is to say, it will become a sentence of one of two possible types: either (a) it will no longer be intentional or (b) it will be explictly psychological. Sentences about probability may be intentional, but depending upon one's conception of probability, they may be transformed either into the first or into the second type.

Fol lowing Chisholm's proposal , D . t i . Sanford proposes these transfor- mations for sentences 5 and 9:12

12. It is necessary that Alaska is self-identical. 13. It is possible that the frost will bring it about that the cliff will fall.

Sentence 12 has the name truth value as 5, while the negation of 13 implies the falsity of the subordinate clause. Hence sentences 5 and 9 are not intentional after all.

Flores correctly attributes to Chisholm the following principle:13 A. All non-transformable intentional sentences are psychological.

Sentences containing modal terms and other apparent ly non- psychological yet intent ional sentences do not count as intentional if they can be t ransformed into non-intentional sentences. Or, if they can be t ransformed into explicitly intentional sentences, they count as intentional. For example, one might transform:

14. It is probable that there is life on Venus. into

15. I t is reasonable for us to believe that there is life on Venus. Flores takes ' t ransforms' to mean "transforms into a logically equivalent sentence". He counters that if one sentence is intentional and the other is not, these two sentences have different implicatory relations and hence are not logically equivalent. 14

299

JESSE L. YODER

1.4 Chisholm's Reformulated Criterion In his reponse to Flores, Chisholm concedes that his original criteria

need to be reformulated. He gives the following definitions which are intended to provide this reformulation: 15 (DI4) T contains S autonymously in L Y Df (i) S is a proper part of T;

and (ii) for every U, if U results from results from replacing S in T by some sentence in L, then there is a P such that P is a proper part of U and the result of replacing S by P in Tis a sentence in L.

(D15) T contains S in a proposi t ional context in L Y Dr(i) S is a proper part of T; (ii) T does not contain S autonymously in L; and (iii) for every U, if U is a sentence in L, the result of replacing S by U in T is a sentence in L.

(Dl6) T contains S in an intentional context in L Y Df(i) T contains S in a proposit ional context in L; (ii) there is an S' which is inconsistent with S in L but which is such that the result of replacing S by S ' is consistent with T in L.

(D17) T contains S in an exclusively intentional context in L Y Df (i) T contains S in an intentional context in L; and (ii) there is a U such that (a) U is logically equivalent to S in L and (b) the result of replacing S by U in T is not logically implied by T in L.

In his discussion of Chisholm's reply, Ausonio Marras correctly attributes to Chisholm the following criterion:16 (P) A sentence containing a proposi t ional clause S is psychological if

and only if it contains S in an exclusively intentional context. Marras goes on to discuss the problem of modal contexts: 17

According to (P), 'John believes that Socrates is mortal' is psychological, whereas 'It is possible that Socrates is mortal'is not; the latter contains S in an intentional but not in an exclusively intentional context. And 'It is necessary that Socrates is mortal'is also non-psychological since S does not even occur in an intentional context.

As Marras says, then, Chisholm's revised criterion appears to answer objections to Chisholm's earlier criteria based on sentences containing modal terms.

Marras points out, however, that some sentences used to describe psychological phenomena are not intentional on Chisholm's revised criterion:~S

Let T be the obviously psychological sentence 'John knows that Socrates is mortal' and let S be 'Socrates is mortal'. T contains S in a propositional context but not in an intentional context, and thus not within an exclusively intentional context, since there is an S' ('Socrates is not mortal') which is

300

INTENTIONALITY

inconsistent with S but which is such that the result of replacing S by S'is not consistent with T. Hence T (and many other sentences containing verbs like 'see', 'notice' 'remember', 'realize', etc.) turn out to be non-psychological according to (P).

Other sentences that are not intentional by Chisholm's criteria are: 16. I see that my desk is brown. 17. I remember that you attended the meeting. 18. I am aware that she is wearing lipstick.

In sentences 16, 17, and 18, the result of replacing the embedded sentence with its negation yields a sentence which is inconsistent with the original sentence. Chisholm's revised criterion, then, while dealing with the problems of model contexts, appears to rule out some sentences which clearly describe psychological phenomena.

1.5 Marras' Revision of Chisholm's Criteria Marras proposes a different revision of Chisholm's original criteria:19

(i) Drop Chisholm's third criterion of intentionality in SAB [Sentences About Believing20], which fails to keep the (de dicto) modalities out of the intentional domain.

(ii) Retain the first two criteria in SAB as (singly) sufficient conditions of intentionality.

(iii) Add to those two criteria the following recursive condition: (C)Any (noncontradictory) sentence which entails a sentence which

is intentional by either the first or the second criterion is itself intentional.

(iv) Let Chisholm's first two criteria in SAB supplemented by the recursive condition (C) constitute jointly a necessary and sufficient condition of intentionality. (It will be noticed that modal sentences turn but to be non-intentional by this condition, whereas ' John Knows that P', which entails ' John believes that P', turns out to be intentional by meeting conditions (C).)

(v) Reformulate Chisholm's thesis about the psychological as follows: (D)A sentence is psychological if and only if it is intentional and is

not transformable into a non-intentional sentence. In a footnote, Marras replies to four sentences offered by John Bacon as counterexamples to his proposal. 2~ One sentence is:

19. It is a contingent matter whether A. In reply, Marras refers to an earlier article he wrote and published in Intentionality, Mind and Language. In a footnote in this earlier article, Marras makes these remarks: 22

301

JESSE L. YODER

Contingency sentences seem to be intentional by Chisholm's second criterion. For neither "It is contingent that P" (where 'contingent' means 'neither necessary nor impossible') nor its contradictory implies either that P is true or that P is false. However, since such contingency sentences are really abbreviations for compound sentences (i.e., sentences containing two or more sentences conjoined by means of such sentence connectors as 'and', 'or', 'because,' etc.), they are as such beyond the scope of Chisholm's second criterion, which applies only to noncompound sentences. Chisholm also provides a criterion of intentionality for compound sentences, according to which a compound (declarative) sentence is intentional if and only if at least one of its component sentences is intentional (Perceiving, p. 172). By this criterion, the compound sentence "It is neither necessary thatp nor impossible that p" (which is the unabbreviated form of"It is contingent that p ' ) is not intentional since neither of its component sentences is intentional.

Similar considerations apply to sentences about probability dispositions, tendencies, abilities, etc., which may be analyzed into nonintentional sentences about frequencies and / or nonintentional counterfactual sentences.

In the above quota t ion , Marras deals deftly with many modal sentences. Included in his list of s tatements he believes he can show to be non- in ten t iona l are sentences of probabil i ty. Marras apparent ly believes that probabi l i ty sentences can be analyzed into sentences abuot frequencies or into non- in ten t iona l counterfactual sentences.

The frequency interpretat ion is one of several accepted interpretat ions of probabi l i ty . 23 However, this in terpre ta t ion of probabi l i ty has difficulty account ing for the probabi l i ty of unique events�9 Consider this sentence:

20. It is probable that the sun will bu rn out in 23 bil lon years. Sentence 20 is in ten t iona l by Chisholm's cri ter ion 2. Perhaps Marras would analyze 20 as:

21. The frequency with which suns burn out in 23 bil l ion years is high. However, 21 will not do as an analysis of sentence 20 since 20 refers to a par t icular sun and since this claim is based on evidence about the sun; not on a statistical or frequency analysis of probabilit ies.

Chisholm addresses this issue in Sentences About Believing: 24 Here Chisholm says that the following sentence:

22. It is probable that there is life on Venus. can be transformed into sentences about frequencies. But unique events

�9 , . / .

occur only once so that it is a mistake to suppose that their probabthty is a mat ter of the frequency with which they occur. Chisholm also appears to be confusing giving an analysis or equivalent reading of a sentence with citing the evidence for it. The evidence for sentence 22, even if it is a

302

INTENTIONALITY

matter of the frequency with which life is found on certain places, is not part of the meaning of 22. Hence, even if the evidential sentences for sentence 22 are non-intentional, this does not make 22 non-intentional.

Probably Chisholm and Marras would prefer an intentional reading of sentences 21 and 22. Chisholm's other interpretation is as follows: 25

23. It is reasonable for us to believe that there is life on Venus. Sentence 23 is intentional by Criterion 2. However, what is reasonable to believe is determined by weighing the reasons for and against a proposition, while probability is a matter of judging the likelihood of an event's occurring. Hence it can be reasonable to believe something to which the concept of probability does not apply, as it is reasonable for me to believe that 2 + 2 = 4. Yet it is not merely probable but certain that 2 + 2 = 4. Furthermore, what is reasonable for a person to believes is relative to the background of his beliefs and purposes, so it can be reasonable to believe something improbable. For example, it is reasonable for me to believe that I can leap over a chasm if this will help me accomplish this task even if it is improbable that I will succeed.

1.6 Conclusion I have examined Chisholm's original criteria for identifying

intentional sentences, as well as Chisholm's own revision and Marras ' revision of these criteria. Marras ' revised criteria are no more successful in identifying all and only intentional sentences than either Chisholm's original or revised criteria. Since Chisholm's is perhaps the most systematic attempt to provide criteria for identifying intentional sentences, the arguments presented above cast doubt on the possibility of providing criteria that identify all and only intentional sentences. If intentionality is to be defended successfully as a criterion of mental phenomena, a better formulation of the criteria for intentional sentences than those offered by either Chisholm or Marras must be found. 26

UNIVERSITY OF LOWELL LOWELL, MASSACHUSETTS 01854

USA

303

JESSE L. YODER

NOTES

1 See Franz Brentano, "The Distinction Between Mental and Physical Phenomena", trans, by D. B. Terrell, in R. M. Chisholm, ed. Realism and the Background of Phenomenology (New York: Free Press, 1960), p. 50.

2 The most discussed Chisholm sources are R. M. Chisholm, Perceiving: A Philosophical Study (Ithaca, N.Y.: Cornell University Press, 1957), pp. 168-88 and R. M. Chisholm, "Sentences About Believing", in Minnesota Studies in the Philosophy of Science, eds. Herbert Feigl, Michael Scriven, and Grover Maxwell (Minneapolis: University of Minnesota Press, 1957), pp. 510-20. lntentionality, Mind, and Language, ed. Ausonio Marras (Chicago: University of Illinois Press, 1972) is an excellent collection of articles on intentionality, including an important article by William G. Lycan and a valuable commentary by Robert Sleigh on Chisholm's "Notes on the Logic of Believing". Also see Linda L. McAlister, "Chisholm and Brentano on lntentionality," Review of Metaphysics 28 (1974), pp. 328-38 for a discussion of how faithful Chisholm is to Brentano's thesis. More recent articles include Albert Flores, "The Thesis of Intentionality", Philosophia 7 (1978), pp.501-14; Chisholm's reply, ibid., pp. 612-15. Further discussion occurs in Ausonio Marras, "Intentionality Revisited", Philosophia 12 (1982), pp. 21-35. Also, Marion Knowles, "Some Remarks on the Intentionality of Thought", Philosophy and Phenomenological Research 43 (1983), pp, 267-79.

3 Flores, "The Thesis of Intentionality". 4 Marras, "Intentionality Revisited". 5 R.M. Chisholm, "Sentences About Believing". 6 R.M. Chisholm, Perceiving, A Philosophical Study. 7 R.M. Chisholm, "Replies", Philosophia 7 (1978), p. 612. s Flores, "The Thesis of Intentionality", p. 503. 9 Ibid., p. 505. lo Ibid. N R.M. Chisholm, "Sentences About Believing", p. 512. 12 David H. Sanford, "On Defining Intentionality", Proceedings of the XIVth.

International Congress of Philosophy, Vol. II (1968), p. 220. J3 Flores, "The Thesis of lntentionality", p. 507. 14 Ibid.,p. 511. J5 R.M. Chisholm, "Replies", p. 614. ~6 Marras, "Intentionality Revisited", p. 30. 17 Ibid. J8 Ibid. 19 Ibid., pp. 31-2. 20 R.M. Chisholm, "Sentences About Believing". 2~ Ausonio Marras, "Intentionality Revisited", pp. 34-5. John Bacon proposes his

sentences in a review of Marras' anthology entitled Intentionality, Mind, and Language. This review appears in Philosophia 6 (1976), pp. 351-58.

22 Ausonio Marras, "Intentionality and Cognitive Sentences", in Marras, lntentionality, Mind, and Language, p. 69.

304

INTENTIONALITY

23 In addition to the frequency interpretation, other interpretations of probability are the logicalinterpretation and the subjective interpretation. See Brian Skyrms, Choice and Chance, (Belmont, Cal.: Dickenson Publishing Co., 1966) for a discussion of these three interpretations of probability.

24 Chisholm, "Sentences About Believing", p. 512 25 Ibid. 26 I am grateful to Gareth Matthews for helpful comments on earlier versions of this

paper.

PRAXIS I N T E R N A T I O N A L

i

Edited by Seyla Benhabib and Svetozar Stojanovk~ In its first five years of publication, Praxls International has emerged as the voice of humanist, democratic socialism, providing a forum for the critical analysis of existing social systems, of their institutions and ideologies, of new social movements and of social theories. Distinguished contributors to past volumes have included J. Habermas, C. Taylor, A. Giddens, R. J. Bemstein, C. Offe, M. Markovic, A. Heliar, R. Rorty, M. Jay, R. Falk, and M. Lowy. Forthcoming issues will discuss Foucault and critical theory; the zero-sum utopia of the welfare state; women and work; Black neo-conservatism tn the USA; and neo-Aristotelianism. sApril, July, October and January

ubscription Rates Volume 7,1987 Individuals: s (UK) s (overseas) US$36.00 (N. America & Japan) Institutions: s (UK) s (overseas) US$95.00 (N. America & Japan) Please send orders with payment to: IdsTaylor orto: Joumals Department Journals Department Basil Blackwell Basil Blackwell PO Box 1320 108 Cowley Road Murray Hill Station Oxford OX4 1JF NY 10156 England USA

B a s i l B l a c k w e l l �9 O x f o r d a n d N e w Y o r k i i

305