Abstract
An English speaker in uttering the words, (0) ‘The earth is round’, says, or asserts, the same thing as a French speaker uttering the words, (0′) ‘La terre est ronde’. 1 The thing asserted is a proposition, the proposition that the earth is round. That there are propositions, as distinct from the sentences that express them, is a commitment of psychology and other human sciences, which ascribe beliefs and other propositional attitudes. The existence of propositions is an integral part of our ordinary conceptions of consciousness and cognition, and therewith of our ordinary conception of what it is to be a person.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Footnotes
Analysis, 10, 5 (1950), pp. 97-99; reprinted in L. Linsky, ed., Reference and Modality (Oxford University Press, 1971), pp. 168–170. See also Church’s “A Formulation of the Logic of Sense and Denotation,” in Henle, Kallen, and Langer, eds., Structure, Method, and Meaning: Essays in Honor of Henry M. Sheffer (New York: Liberal Arts Press, 1988), pp. 3-24 at 5-6n; “Intensional Isomorphism and Identity of Belief,” Philosophical Studies, vol. 5 (1954), pp. 65-73, reprinted in N. Salmon and S. Soames, eds., Propositions and Attitudes (Oxford University Press, 1988), pp. 159-168; and Introduction to Mathematical Logic I (Princeton University Press, 1956), p. 62n.
The following is a partial list: Tyler Burge, “Self-Reference and Translation,” in F. Guenthner and M. Guenthner-Reutter, eds., Translation and Meaning (London: Duckworth, 1977), pp. 137–153, and “Belief and Synonymy,” Journal of Philosophy, vol. 75, no. 3 (March 1978), pp. 119-138; Rudolf Carnap, “On Belief-Sentences: Reply to Alonzo Church,” in Carnap’s Meaning and Necessity (University of Chicago Press, 1947, 1956), pp. 230-232; Donald Davidson, “The Method of Extension and Intension,” in P. A. Schilpp, ed., The Philosophy of Rudolf Carnap (La Salle, 111., 1963), pp. 331-349, at 344-346; Michael Dummett, Frege: Philosophy of Language (Cambridge, Mass.: Harvard University Press, 1973, 1981), at pp. 372-373, and The Interpretation of Frege’s Philosophy (Cambridge, Mass.: Harvard University Press, 1981), at pp. 90-94; Peter Geach, Mental Acts (London: Routledge, 1957), at pp. 89-92, and “The Identity of Propositions,” in Geach’s Logic Matters (Oxford: Blackwell, 1972), pp. 166-174; Steven Leeds, “Church’s Translation Argument,” Canadian Journal of Philosophy, vol. 9, no. 1 (March 1979), pp. 43-51 (I thank Mark Richard for providing this reference); Brian Loar, Mind and Meaning (Cambridge University Press, 1981), pp. 29-30, 152; Hilary Putnam, “Synonymy, and the Analysis of Belief Sentences,” Analysis, vol. 14, no. 5 (April 1954), pp. 114-122, reprinted in N. Salmon and S. Soames, eds., Propositions and Attitudes (Oxford University Press, 1988), pp. 149-158; W. V. Quine, “Quantifiers and Propositional Attitudes,” reprinted in L. Linsky, ed., Reference and Modality (Oxford University Press, 1971), pp. 101-111, at 110; Mark Richard, Propositional Attitudes (Cambridge University Press, 1990), at p. 156ff; Israel Scheffler, “An Inscriptional Approach to Indirect Quotation,” Analysis, vol. 14, no. 4 (March 1954), pp. 83-90, and “On Synonymy and Indirect Discourse,” Philosophy of Science, vol. 22, no. 1 (January 1955), pp. 39-44, at 43-44n. While no attempt is made here to respond adequately to each of these critiques, much of what will be said here is applicable to a number of them.
Frege, Philosophy of Language, p. 372.
“The Identity of Propositions,” p. 167.
More generally, it is my considered view that the attempt to avoid an ontology of extra-linguistic abstract entities by an appeal to intra-linguistic substitutes is philosophically misguided. The reasons for this judgment are complex. As an excessively brief summary, I mention that the ontology of everyday discourse is replete with abstract entities other than propositions. The philosophical security that is supposed to be afforded by replacing propositions with sentences is largely an illusion, since sentences, no less than propositions, are abstract entities. Many sentences-infinitely many, in fact-are too long to be written or uttered by any conceivable creature. Moreover, principles concerning the identification of sentences are not nearly as “extensional” as is sometimes supposed. This last difficulty often manifests itself in the need to resort not merely to sentences as such, but to sentences as sentences of a particular language, or to sentences as meaning that such-and-such, etc.
For many applications, the requirement of full preservation of meaning (whereby, as a consequence, one believes what is expressed by the analysandum if and only if one also believes what is expressed by the analysans) may be weakened to mere logical equivalence, and in some applications, to such wider equivalence relations as modal equivalence or a priori equivalence, with little effect on the overall force of the argument.
Church’s conception of linguistic expressions and their semantics is such as to require that such semantic attributes as sense, denotation, and truth value always be relativized to a particular language. Others favor a conception according to which relativization to a language is unnecessary or even inappropriate. Cf. Peter Geach’s protests in Mental Acts, pp. 88-89; and David Kaplan, “Words,” Proceedings of the Aristotelian Society, vol. 64 (1990), pp. 93–116. Although Church’s presentation of the Translation Argument assumes his view on this issue, the argument can also be presented with the opposite view presupposed, or with neutrality (as it is here).
One addendum that would obviously be inadmissible for the purpose of eliminating commitment to propositions is the phrase ‘as expressing the proposition that the earth is round’, or more simply ‘as meaning that the earth is round’. (Cf. footnote 7 above.) Church explicitly considers a possible expansion of (2) that logically entails this inadmissible one, admitting that such an expanded version may yield (1) as a logical consequence. He does not note that the expansion in question would not be suitable for the elimination of propositions. (It may even lead to circularity, if the proposed analysis is extended to such constructions as’ sentence S means in language L that the earth is round’ in addition to (1).) Instead he notes, correctly, that the particular expansion of (2) he considers does not preserve the meaning of (1). Again he utilizes translation to crystalize the point. In addition, he considers embedded constructions like ‘Jones believes that Chris believes that the earth is round’, arguing that alleged analyses of this and of its translation may even differ in truth value in their respective languages. (Cf. footnote 8 above.) Carnap says in response (op. cit., p. 230, in a highly compressed paragraph) that he intended precisely such an expansion of (2), while conceding that Church’s objection is correct. Carnap’ s overall response to Church is unclear and puzzling. In Meaning and Necessity, pp. 63-64, scarcely a page after presenting his analysis of statements of belief, Carnap says of analysis in general that although analysandum and analysans must be logically equivalent, they need not be intensionally isomorphic-or as Carnap also puts it, the analysis must preserve intension but need not preserve intensional structure. Davidson (op. cit.) and Putnam (op. cit.) argue, in effect, that Carnap’s analysis of statements of assertion and belief in particular, therefore, is not intended to capture their meaning but only something logically equivalent-thus making Church’s objections inapplicable. In sharp contrast, Carnap concedes not only Church’s objection, but futhermore that (1) is strictly not even a logical consequence of the intended version of (2) (pp. 230-231). Curiously, Carnap also endorses Putnam’s response (p. 230), while proffering an alternative version of (2) as a scientific replacement (presumably now logically equivalent) for (1) (pp. 231-232). The textual evidence suggests, however, that Carnap confused Putnam’s response on this point with Putnam’s response to a separate objection by Benson Mates concerning embedded constructions. In “Intensional Isomorphism and Identity of Belief,” Church extends the Translation Argument against Putnam’s response to Mates. See also the introduction to N. Salmon and S. Soames, eds., Propositions and Attitudes, (Oxford University Press, 1988), pp. 13–14 n l0. The issues here are numerous and quite complex. Cf. footnote 28 below. Insofar as embedded constructions are involved, the dispute is intimately related to issues concerning Frege’s notion of indirect sense. See footnote 25 below.
Quine, op. cit., at pp. 109-110. The expression ‘believes-true’ is, however, significantly misleading. For discussion, see my “Relational Belief,” in P. Leonardi and M. Santambrogio, eds., On Quine (Cambridge University Press, 1995).
Cf. G. E. Moore, “Russell’s ‘Theory of Descriptions’,” in his Philosophical Papers (New York: Collier Books, 1959), pp. 149–192, at 156-157, where he anticipates Langford’s translation test.
”Über Sinn and Bedeutung”, translated as “On Sense and Reference,” in R. M. Harnish, ed., Basic Topics in the Philosophy of Language (Englewood Cliffs, N.J.: Prentice Hall, 1994), pp. 142–160, at 144. I follow Church in translating Frege’s use of ‘Bedeutung’ as ‘denotation’ rather than ‘meaning’ or ‘reference’, and Frege’s use of ‘Gedanke’ as ‘proposition’ rather than ‘thought’. I also use the word ‘concept’ in Church’s sense, which is very different from Frege’s use of ‘Begriff’.
Meaning and Necessity, §30, pp. 129-133.
The Interpretation of Frege’s Philosophy, at pp. 89-100. This represents a turnabout for Dummett. In his earlier work, Frege: Philosophy of Language, he considered the thesis that the indirect sense of (0) in English is the customary sense of (4), only to dismiss it as “rather implausible” (p. 267). Both this thesis concerning indirect sense, and the exegetical thesis that Frege’s theory implies the former thesis, are defended in Gary Kemp, “Salmon on Fregean Approaches to the Paradox of Analysis,” Philosophical Studies, vol. 78, no. 2 (May 1995), pp. 153–162, wherein specific references to Frege’s writings are provided (at p. 160) in support of Dummett’s interpretation.
Dummett points out that the inference in question is actually invalid, but he allows (pp. 98-99) that it is validated by the addition of a third thesis (which Frege held and which Dummett rejects) as premiss, to wit, that “the sense of an expression is the way in which its [denotation] is given to us.”
Bishop Berkeley argued, against Locke’s doctrine of “abstraction,” that one cannot conceive of, for example, a color without also conceiving of some shape or other of that color. According to Dummett, Frege did not hold, in the same spirit as Berkeley, merely that one cannot conceive of a sense in any way without also conceiving of it as the sense of some expression or other. It is not that any conception of a sense we have, must be accompanied in cognition by a conception of it as the sense of some expression. The point, rather, is that in conceiving of a sense, we conceive of it only as the sense of e, for some expression e or other, since we have no conception of the sense other than as the sense of e, for some expression e, for such a conception to accompany. (This will be made more precise below. I am grateful to Charles Chihara for bringing this distinction to my attention.)
Carnap, op. cit., at p. 129; Dummett, Frege: Philosophy and Language, pp. 266-269.
Dummett does not explicitly speak in a relativized manner of the sense of an expression in a language. The parenthetical references to the language l are included here for the benefit of those whose view of linguistic expressions and their semantics requires for propriety the relativization to a language. See footnote 9 above.
”Uber Sinn und Bedeutung” at p. 149 of Harnish.
In The Interpretation of Frege’s Philosophy, at p. 89, there occurs an otherwise inexplicable switch, wherein the discussion, suddenly and without notice, changes its focus from an identification of the indirect sense of a sentence like (0) with the ordinary sense of the corresponding analogue of (5), to an identification of the indirect sense of ‘Aristotle’ instead with the customary sense of ‘the sense of ‘Aristotle’ ‘.
Cf. the argument given by Dummett op. cit., p. 95, top paragraph. See also Kemp, op. cit.
It must be admitted that the construction ‘Chris believes the earth is round’, without the word ‘that’, is perfectly grammatical English, and (0) occurs nonextensionally therein. This phenomenon, however, does not immediately yield the result that ‘believes’ is a nonextensional (ungerade) operator. It is arguable that (0) is induced to shift to the indirect mode here by other (perhaps pragmatic) factors. (Indeed, linguists commonly refer to the phenomenon as “‘that’ deletion.”)
Cf. my Frege’s Puzzle (Atascadero, Ca.: Ridgeview, 1986, 1991), at pp. 5–6; and “Reference and Information Content: Names and Descriptions,” in D. Gabbay and F. Guen-thner, eds., Handbook of Philosophical Logic IV: Topics in the Philosophy of Language (Boston: D. Reidel, 1989), pp. 409-462, at 440-441.
The Translation Argument is not supportive of all aspects of Fregean theory. As is shown in my article “A Problem in the Frege-Church Theory of Sense and Denotation,” Noûs, vol. 27, no. 2 (1993), pp. 158–166, while the argument supports Frege’s thesis concerning indirect sense, it thereby leads to an inconsistency when combined with a Fregean solution to the Paradox of Analysis, of a sort advocated by Church in his review of the Black/White exchange concerning the Paradox of Analysis, The Journal of Symbolic Logic, vol. 11 (1946), pp. 132-133. I argue in the article that relinquishing the Fregean solution to the Paradox of Analysis threatens the Frege-Church theory, by collapsing Frege’s and Church’s original argument for the pivotal distinction between the sense and denotation of an expression. (Whereas this difficulty exposes a potentially serious weakeness in the Frege-Church theory of sense and denotation, it does nothing to weaken the force of the Translation Argument.)
Mental Acts, pp. 91-92. (I thank Saul Kripke for providing this reference.)
Burge, “Self-reference and Translation,” especially pp. 141-144; Dummett, loc. cit. Burge’s response to the Translation Argument is also endorsed by James Higgenbotham, in “Linguistic Theory and Davidson’s Program in Semantics,” in E. Lepore, ed., Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson (Oxford: Basil Black-well, 1986), pp. 29–48, at 39n. (I thank Higginbotham for providing the last reference.)
This point is explicitly acknowledged by Dummett (Frege: Philosophy of Language, p. 372; The Interpretation of Frege’s Philosophy, p. 90). It is more or less acknowledged by Bürge (op. cit., p. 141), although his overall response to the Translation Argument is explicitly part of a program to avoid commitment to propositions (pp. 152-153). Geach espouses an extreme skepticism regarding the notion of synonymy. He argues that (3’) is “of the same force as” (his version of) (2), and that (3’) and (2) are “reasonably equivalent,” but specifically stops short of declaring that they are either the same, or different, in meaning. By the same token, Geach does not propose (2) as a meaning-preserving analysis of (1). It is unclear therefore why he does not concede that any author of (1) is committed to the existence of the proposition that the earth is round (this being what the Translation Argument is aimed at demonstrating), and let this be his reason for recommending that the author substitute (2), which lacks any such commitment. (Cf. Quine, op. cit., and my “Relational Belief” for discussion.) In any event, Dummett’s reply has the advantage over Geach’s (and Quine’s) that it does not depend (at least not to the same extent) on any implausible or otherwise controversial skeptical theses concerning synonymy. Burge claims that the sentence he proposes for (2) involves self-denotation, in such a way that the best translation will preserve this feature at the expense of denotation. The details of Burge’s argument will not be pursued here. One minor correction should be noted, however. Contrary to Burge’s initial claim (which he credits to W. D. Hart) that translation of a self-denoting sentence either preserves self-denotation or denotation but not both, translations may be given that preserve neither one. (For whatever it is worth, it is even probable that such translations have been given in actual practice.) Burge’s argument requires only the weaker claim that no translation of a self-denoting sentence preserves both denotation and self-denotation.
Frege: Philosophy of Language, p. 372. But see also The Interpretation of Frege’s Philosophy, p. 90.
Heidelberger, Review of Dummett’s Frege: Philosophy of Language, in Metaphilosophy, vol. 6, no. 1 (January 1975), pp. 35–43, at 42-43; Lewy, Meaning and Modality (Cambridge University Press, 1976), at pp. 64-66; Linsky, Oblique Contexts (University of Chicago Press, 1983), at p. 8. (I thank C. Anthony Anderson for providing the last two references.) See also Saul Kripke, “A Puzzle about Belief,” reprinted in N. Salmon and S. Soames, eds., Propositions and Attitudes (Oxford University Press, 1988), pp. 102-148, at 142n25. In his discussion in The Interpretation of Frege’s Philosophy, Dummett attributes both the thesis that the indirect sense of (0) in English is the ordinary sense of (4), and the exegetical thesis that Frege held the former thesis, to Heidelberger. Dummett also there accuses Heidelberger of inconsistently conjoining these theses with an endorsement of the Translation Argument (pp. 91, 94). These attributions are dubious. Although Heidelberger defends the Translation Argument against Dummett’s criticism, he explicitly declines to endorse the argument and instead expresses sympathy for the alternative criticisms of it by Davidson and Putnam (p. 43n). Moreover, he does not straightforwardly propose either of the theses in question. Instead he correctly attributes to Frege the thesis that the indirect sense of (0) in English is the ordinary sense of (5), while mis-identifying this thesis with the alternative thesis which Dummett had branded “rather implausible” (p. 37; see footnote 15 above). As noted earlier, Dummett also fails to distinguish between these two theses concerning indirect sense, and therefore fails to distinguish properly between the corresponding exegetical theses concerning Frege on indrect sense. (It should also be noted that whereas Heidelberger explicitly attributes to Frege the thesis that the indirect sense of (0) in English is the ordinary sense of (5), he also attributes to Frege a fallacious argument for that thesis which Frege does not give, and which is in fact inconsistent with his views.)
The Interpretation of Frege’s Philosophy, pp. 90-91. The premiss is explicitly rejected by Kemp, op. cit., and Stephen Leeds, op. cit. Curiously, Dummett ulitmately endorses the premiss (p. 94), on the ground that (1) has the same “conventional significance” in English that (1) has in French (p. 99). It is unclear why Dummett insists nevertheless that it is illegitimate for Church to assert this premiss, on the same or very similar ground, in the course of his argument against the proposed analysis.
For more on this sort of issue, especially as it relates to Burge’s response to the Translation Argument, see my “A Problem in the Frege-Church Theory of Sense and Denotation,” at p. 166nl5.
Frege’s Puzzle, loc. cit. (footnote 24 above), pp. 58-60 and elsewhere (especially pp. 78-79, 84-85, 100, 114-115).
“Intensional Isomorphism and Identify of Belief,” in Salmon and Soames, p. 168n22. In a similar vein, Church had written in “A Formulation of the Logic of Sense and Denotation,” at p. 5n: “This device is not essential to the explanation, but is helpful in order to dispel any remnants of an illusion that there is something in some way necessary or transparent about the connection between a word or a sentence and its meaning, whereas, of course, this connection is entirely artificial and arbitrary.”
Russell, “Knowledge by Acquantance and Knowledge by Description,” reprinted in N. Salmon and S. Soames, eds., Propositions and Attitudes (Oxford University Press, 1988), pp. 16–32.
In calling a condition “purely descriptive,” I mean to preclude its being such a condition as might be expressed in the form ‘the condition of being [identical with] that very F’ It is assumed here that Church’s notion of a concept likewise excludes such conditions. Cf. my Reference and Essence (Princeton University Press and Basil Blackwell, 1981), pp. 14–23 (where the term ‘descriptional’ is employed instead of ‘descriptive’). Russell construed acquaintance in a very strict sense which excluded the possiblity of acquaintance with particulars other than oneself or mental items directly contained in one’s consciousness. The distinction itself can be drawn independently of this severe restriction, however, and is clearly legitimate with regard to more familiar notions of acquaintance. One such notion is that of having perceptual, or other natural or “real,” cognitive contact with a particular person or object-the sort of connection that is sufficient to enable one to form beliefs or other attitudes about the object (in an ordinary sense). A somewhat stricter notion imposes the further condition of knowing who or what the object is, in an ordinary sense. (Some philosophers, having evidently overlooked the possibility of perceiving an object without knowing who or what the object is, have confused these two broader notions of acquantance, See my “How to Measure the Standard Metre,” Proceedings of the Aristotelian Society, New Series, vol. 38 (1987/88), pp. 193-217, especially at 200-201n, 213ff.) Throughout I use the term ‘identification’ for a notion of acquaintance implying knowledge of who or what the object is. Some contemporary neo-Russellian theories of meaning deny that one who knows the content of (1) automatically thereby knows what proposition it is that Chris believes (i.e., automatically thereby knows at least one proposition that Chris believes). Such theories may hold instead that one who knows the content of (1) is thereby acquainted with the believed proposition in some less familiar way, and may not know exactly what proposition is in question. Knowing what F so-and-so is (as special cases, knowing what proposition Chris is hereby held to believe, or knowing what person-or who-so-and-so is) may be a matter knowing of so-and-so, de re, that so-and-so is it (him/her), believing this fact about so-and-so while conceiving of it in a special, identifying way. (See my Frege’s Puzzle, pp. 103-118, on the notion of believing a proposition while taking it a particular way; see also footnote 43 below.) Even such theories, however, will generally recognize an important epistemological difference between the contents of (1) and (6), such that knowing the fact described by (6) falls well short of knowing that fact described by (1). It may be acknowledged, for example, that one who knows the content of (1)-by contrast with one who merely knows the content of (6)-ipso facto kows something de re about the proposition that the earth is round, namely, that it is something Chris believes (even, perhaps, without knowing exactly what proposition is in question). This is sufficient for my primary purpose in the discussion to follow. For related discussion, see Mark Richard, “Articulated Terms,” in J. Tomberlin, ed., Philosophical Perspectives, 7: Language and Logic (Atascadero, Ca.: Ridgeview, 1993), pp. 207-230. It is possible that there is a gradation of notions of acquaintance. This would not make Russell’s distinction between knowledge by acquaintance and knowledge by description untenable; on the contrary, it would make for a multiplicity of legitimate Russellian dis-tinctions. The distinction on which I rely in the discussion to follow need not be made completely precise, and will be sufficiently obvious to support the conclusions reached by its means.
As explained in the previous note, Russell himself held a respresentational epistemology with regard to all particulars other than oneself and the mental contents of one’s consciousness. Contemporary Russellians have typically favored a less restrictive episte-mology on which one knows various concrete particulars “directly,” i.e., without appealing to individuating qualitative concepts of those particulars.
Cf. the infinite-regress argument in Russell, op. cit., at pp. 28-29 of Salmon and Soames.
One may not straightforwardly conclude from the infinite-regress argument that according to the strict representational sort of epistemology in question, in order to conceive of a thing x, one must in that very act of conceiving apprehend each of infinitely many concepts. For although the epistemology requires that the act of conceiving of x necessarily involves an act of apprehanding a concept c 1 of x, it does not require that the act of apprehending c 1 necessarily involves conceiving of c 1. One might even label the act of apprehending a concept without conceiving of it a kind of “direct acquaintance” with the concept. One may not even conclude from the infinite-regress argument that according to strict representational epistemology, in order to conceive of anything one must already apprehend each of infinitely many concepts. What is required on Prege’s doctrine of the hierarchy is that one be able to acquire an apprehension of any one of the infinitely many concepts on demand, so to speak, in order to comprehend a sentence that embeds an expression within a sufficiently large number of ungerade operators. Strict representational epistemology nevertheless yields a possible answer to Russell’s famous “no backward road” observation and the difficulty it is supposed to raise for Frege’s doctrine. It is true that on Prege’s theory, for anything x there are countless concepts of x, no one of which can be singled out as privileged or as the “designated” or “standard” concept of x. But according to strict representational epistemology, in conceiving of a concept c 1 through one’s acquaintance with it, one is thereby apprehending a special identifying concept c2 of c 1. By attending to what one is apprehending, one can conceive of c2 through one’s acquaintance with it, and hence through one’s apprehension of a special identifying concept c3 of c2, and so on. Thus it seems that one need only attend to what one is apprehending to generate a proper Fregean hierarchy starting from a single concept. Even if there is no “backward road” (privileged branch) from a thing that is not a concept to a concept of the thing, it seems there may be a backward road from a concept to a concept of the concept. (I am indebted here to remarks made in a seminar by Saul Kripke and to later discussion with C. Anthony Anderson.) Cf. my “Reference and Information Content: Names and Descriptions,” loc. cit., and “A Problem in the Prege-Church Theory of Sense and Denotation,” at p. 163.
It must be noted, however, that Church rejects Frege’s notion of indirect sense. See “A Problem in the Prege-Church Theory of Sense and Denotation,” pp. 164-165nl0.
A number of philosophers have noted differences of meaning beween such terms. For a sample of the relevant literature, see D. M. Armstrong, “Materialism, Properties and Predicates,” Monist, vol. 56, no. 2 (April 1972), pp. 163–176, at 174; Jaegwon Kim, “On the Psycho-Physical Identity Theory,” American Philosophical Quarterly, vol. 3, no. 3 (July 1966), pp. 227-235, and “Events and Their Descriptions,” in N. Rescher, ed., Essays in Honor of Carl G. Hempel (Dordrecht: D. Reidel, 1969), pp. 198-215, at 205-206; Bernard Linsky, “General Terms as Designators,” Pacific Philosophical Quarterly, vol. 65 (1984), pp. 259-276; N. L. Wilson, “The Trouble with Meanings,” Dialogue, vol. 3, no. 1 (June 1964), pp. 52-64. (The last is evidently the ancestor of the other discussions.)
“Intensional Isomorphism and Identity of Belief,” footnote 25 (p. 168 of Salmon and Soames).
Presumably understanding (0), as a sentence of English, entails knowing what (0) means in English, but it is arguable that understanding, in a strict sense, requires more than this (perhaps acquiring the knowledge of the meaning of (0) in a special computational manner). There are delicate issues, on which the present discussion is neutral, concerning the extent to which the English meaning of (5) captures the way of conceiving the proposition in question that is possessed by one who correctly understands (0) as a sentence of English. See footnote 36 above.
See footnotes 19 and 36 above. The first conjunct expresses strict representationalism with regard to concepts. On a Fregean theory the word ‘as’ occurring in Dummett’s constructions ‘… is given to us as-’ and ‘conceives of … as-’ must be itself regarded as an ungerade operator. Dummett’s Thesis, as formulated here and as given by Dummett himself, thus involves quantification into a nonextensional context. (I am not claiming that Dummett would accept my specific formulation of his thesis, only that he explicitly endorses the thesis itself, which I formulate thus, and attributes the same thesis to Frege. The textual evidence for these claims is clear.)
For example, in “On the Scientific Justification of a Conceptual Notation” (reproduced in his Conceptual Notation and Related Articles, T. W. Bynum, ed., (Oxford University Press, 1972), pp. 83–89), Frege says that “we think in words …, and if not in words, then in mathematical or other symbols” (p. 84). See footnote 15 above.
In light of this, Frege undoubtedly rejected even the weaker, Berkeleyan thesis mentioned in note 17 above.
It is arguable that a person completely color-blind from birth may nevertheless learn of the color green (de re) that grass and emeralds are that color, by learning that they reflect light of such-and-such wavelengths under visually normal conditions. Even so, there still seems to be some knowledge that those of us who see green (as green) have, and that the completely color-blind person lacks, concerning the color of grass. In some sense, the completely color-blind person still does not know how grass looks, phenomenologically, with regard to color. By contrast, if Dummett’s theory were correct, none of us could know of the proposition that the earth is round (de re), that (0) expresses it in English (let alone could we understand what (0) means in English-see footnotes 36 and 43 above).
This notation is normally used for definitions within a single language. For present purposes we may think of the union of all of the separate languages spoken by a particular speaker as constituting a single comprehensive language. (Any resulting lexical ambiguities may be resolved by means of disambiguating subscripts.) Alternatively, the notation is easily extendible to accommodate inter-language definitions, for example by writing: [e 1,l 1] =def [e 2,l 2].
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer Science+Business Media Dordrecht
About this chapter
Cite this chapter
Salmon, N. (2001). The Very Possibility of Language. In: Anderson, C.A., Zelëny, M. (eds) Logic, Meaning and Computation. Synthese Library, vol 305. Springer, Dordrecht. https://doi.org/10.1007/978-94-010-0526-5_27
Download citation
DOI: https://doi.org/10.1007/978-94-010-0526-5_27
Publisher Name: Springer, Dordrecht
Print ISBN: 978-94-010-3891-1
Online ISBN: 978-94-010-0526-5
eBook Packages: Springer Book Archive