Hybrid Proper Names WOLFGANG KUNNE

Transcription

1 Hybrid Proper Names WOLFGANG KUNNE Can Frege possibly mean what he says? In his essay "The Thought" we read: 1 [A]... [O]ften... the mere wording, which can be made permanent by writing or the gramophone, does not suffice for the expression of the thought... If a time-indication is conveyed by the present tense one must know when the sentence was uttered in order to grasp the thought correctly. Therefore the time of utterance is part of the expression of the thought. [B] If someone wants to say today what he expressed yesterday by using the word "today", he will replace this word with "yesterday". Although the thought is the same its verbal expression must be different in order that the change of sense which would otherwise be effected by the different times of utterance may be cancelled out. The case is the same with words like "here" and "there". [C] In all such cases the mere wording, as it can be preserved in writing, is not the complete expression of the thought, the knowledge of certain circumstances accompanying the utterance, which are used as means of expressing the thought, is needed for us to grasp the thought correctly. Pointing the finger, hand gestures, glances may belong here too. [D] The same verbal expressiorvcontaining the word "I" in the mouths of different men will express different thoughts of which some may be true, others false. If we take literally what Frege says in parts [A] and [C] of this text then the following claims are made: If someone expresses a thought (Gedanke) through the use of a sentence S, the means for expressing the thought often is not sentence S by itself, but 5 together with the circumstances of its utterance. In such cases the complete thought-expression is an admixture consisting of a linguistic entity and a slice of non-linguistic reality. 2 That situation occurs whenever the sentence used is index - ical, i.e. contains an indicator like "I", "today", "here", or like the tense-modifiers. Already two decades earlier this rather unusual claim about the composition of certain thought-expressions is foreshadowed in one of Frege's manuscripts: 3 1 ( , p. 64). In [C] the Geach translationreads "conditions" for "circumstances" I prefer the more literalrenderingof Frege's "Umstdnde". In [D] the translations has "utterance" instead of "verbal expression". (Frege wrote "Wonlaut".) How could there be one utterance "in the mouths of different men"? 2 Means of expressing a thought are not thoughts, I assume. Thus Frege can hardly use the term "circumstances" to refer to facts; for he takes facts to be true thoughts (cf. 1918, p. 74). 3 Frege, (1969, p. 146). Unfortunately Frege here goes on to suggest that for "I" one could substitute salvo sensu the name of the speaker. His criterion of difference in sense reveals this to be mistaken: If NN suffers from amnesia she might very wellrefuseto accept the thought which she could express by "I am NN". Cf. Dummett, (1981, p.121). Mind, Vol October 1992 Oxford University Press 1992

2 722 Wolfgang Kunne [E] The mere words [sc. "I am cold"] do not contain the whole sense, but it is also a matter of who utters them... Words like "here" and "now" only obtain their full sense from the circumstances in which they are used. In this paper I would like to show that Frege's remarks on the composition of indexical thought-expressions, if taken far more literally than is usually done, suggest a rather non-standard conception of the make-up of certain proper names. I shall outline some of the consequences of this view and try to make it look less bizarre than it looks at first sight by comparing what I call "hybrid" proper names with quotation-names. As an aside, there are also a few observations concerning similar (and similarly neglected) ideas in Wittgenstein and Schlick. As a preliminary, let us first remind ourselves of certain features of the Fregean view of sense (Sinn) which stand out in parts [B] and [D] of the passage quoted above. Suppose we hear a voice on a tape saying things like (51) I have bloodtype A (52) Today is a religious holiday in many countries (53) The weather here is sometimes dreadful and we neither know whose voice has been recorded nor when and where the tape recording was made. Under these circumstances we understand what we hear and then again we don't understand it. 4 We understand it insofar as we have a certain lexical and grammatical competence: we know which meaning those sentences as type-sentences have in the English language. Let us call this type of understanding: grasping the linguistic meaning of 5. This conception of meaning enters into many (folk) semantical categories. If we classify (SI) and "Ich habe Blutgruppe A" as synonymous, we ascribe to these sentences the same linguistic meaning. If we call a sentence like "This is a bank" equivocal, we ascribe more than one linguistic meaning to it. An expression is linguistically meaningless in a language L if it is judged ungrammatical by competent L-speakers or if it is or contains at least one pseudo-word for which no L -speaker could provide what would be accepted as a correct (ostensive or verbal) explanation. In another respect we do not understand those utterances of (S1), (S2) and (S3): we do not grasp what Frege calls the thoughts expressed. Knowledge of the linguistic meaning of a sentence can only coincide with knowledge of the thought expressed if the sentence does not contain any component which can change its 4 What follows in this section is presented on a larger canvas in my (1983, pp , ).

3 Hybrid Proper Names 723 denotation in our world while its linguistic meaning remains constant. Indexical singular terms 3 are such sentence-components. Plainly the linguistic meaning of sentence (S2) does not change every day at midnight. Equally clearly, the sentences (S2) and (S2*) Yesterday was a religious holiday in many countries do not have the same linguistic meaning: they are not synonymous. Frege, however, claims in [B]: The sense changes if (S2) is uttered on successive days. And Frege would also claim: The sense can be the same if we first utter (S2) and on the following day (S2*). Or consider spatial indicators. The linguistic meaning of sentence (S3) does not change just because the speaker takes a journey, and the sentences (S3) and (S3*) The weather there is sometimes dreadful do not have the same linguistic meaning. And yet Frege suggests in [B]: The sense changes if (S3) is uttered at distant places, and it can be the same if the speaker utters (S3) here and (S3*) there. Obviously the notion of sense in Frege's theory cannot be equated with our intuitive conception of linguistic meaning. The Fregean sense expressed by means of a declarative sentence is a thought, and when Frege talks about the sense of a part of a thought-expression, he always means the contribution of this part to the thought expressed. Frege's theory provides no technical term for what we have called linguistic meaning. 6 According to Frege's compositional view of thoughts the sense of the proper name "2" and the sense of the concept-expression (Begriffswort) "( ) is a prime number" combine to form the thought that 2 is a prime number. The sentence "2 is a prime number" is a complete expression of that thought. Now, as Perry pointed out several years ago, we run into trouble if we try to apply this compositional view to indexical thought-expressions (1977, pp. 479ff). Consider example (SI). Knowing the linguistic meaning of the word "I" we know that it (standardly) denotes its respective speaker or writer. When both Paul and Mary utter sentence (SI), the first word of their respective utter- 5 As to my terminology, the following hints might suffice: A linguistic expression (be it sentential or subsentential) is indexical iff it either is an indicator or contains at least one indicator. I shall concentrate on those indexical expressions which give you a sentence when you insert them into the n open places of an /i-place predicate. Since I take singular terms to be such sentence-forming operators on predicates, I do not hesitate, pace Anscombe, to classify "I" as a singular term A singular term is an indicator only if it is semantically unbreakable and its denotation varies systematically with certain features of the circumstances under which it is uttered while its linguistic meaning remains constant. We shall soon see that denotation even in the case of singular terms is not to be equated with Fregean reference (Bedeutung). 6 Cp. Burge (1979, pp ).

4 724 Wolfgang Kunne ances has the same linguistic meaning. Nevertheless, according to [D], the thoughts expressed by Paul and Mary are different. Thus in understanding the word "I" as an element of the English language one does not thereby grasp the Fregean sense which combines with the sense of the concept-expression "() has bloodtype A" to yield the thought which only Paul expresses when he utters (SI). Since Paul utters a quantifier-free sentence the completing sense can only be the sense of a proper name (Eigenname), in Frege's admittedly rather broad use of this term. The proper name in Paul's utterance of (SI) is therefore not the indicator "I", but... But what? Perry somewhat hesitantly ascribes to Frege the answer that the completing sense is provided by an indicator-free definite description which denotes Paul and which, at the moment of his utterance of (S1), is taken by him to do so. 8 Perry then goes on to subject this answer to devastating criticism. But there is no hint whatsoever in Frege's essay that this really is his view; on the contrary, his rather startling remarks on the incommunicability of the thoughts expressed by solitary users of "I" cannot consistently be combined with a "disguised description" view. 9 Passages [A], [C] and [E] suggest a rather different answer. Let us take the "circumstances" of an utterance of sentences like (SI), (S2) and (S3) to consist of all and only those non-linguistic entities which have to be identified if an evaluation of the utterance in terms of truth and falsity is to be possible. Then Frege's answer to the above question would be this: In an utterance of (SI) the proper name consists not only of a token of the indicator "I" but also of the speaker. It is the sense of this proper name which combines with the sense of the predicate in (SI) to yield the thought expressed by Paul. Let us call a proper name hybrid if and only if it consists of more than just a verbal expression. What refers to (bedeutet) the speaker is not the indicator "I" but a hybrid proper name which contains an occurrence of this indicator. The indicator by itself does not refer to any object (Gegenstand) whatsoever, and thus a fortiori it does not refer sometimes to this and sometimes to that object. So even in the face of indexicality the semantical predicate "x refers to y" can remain a two-place predicate: we do not have to take it as a fragment of a longer predicate with further places for constituents of the occasion of the utterance ("in the mouth of speaker 5 at time /"). The hybrid proper name which is used in Mary's utterance of (SI) has a different reference (Bedeutung), and as for every sense there is at most one reference 10 it has not the same sense as the one used in Paul's utterance. Notice that a hybrid proper name which contains a token of "I" 8 Perry (1977, p. 485): "This is, I think, as near as we are likely to come to what Frege had in mind." Kaplan once wrote: "Frege... seems... to believe that all names, even such demonstrauves as "I", are to be analyzed as disguised definite descriptions" (1975, p. 725, cf. p. 717). 9 So much the worse for these remarks, Perry would answer. But his objections against the very idea of incommunicable Fregean thoughts have been defused by Gareth Evans in 5 of his (1981). 10 (1892, pp ).

5 Hybrid Proper Names 725 always refers to a part of itself, to its non-linguistic component. We shall come back to this remarkable feature of some hybrid proper names in the last section of this paper. The proper name used in an utterance of (S2) is hybrid as well. It consists of a token of the indicator "today" and the time of its utterance. It refers to the day to which the time of the utterance of this word belongs. And the hybrid proper name used in an utterance of (S3) consists of a token of the indexical description "the weather here" and the place of its utterance." In all these cases it is the thought-expression, not the thought expressed, which contains a person, a time, or a place: Frege never dreamt of identifying thoughts with Russellian singular propositions. The fact that persons, times and places themselves have neither Fregean sense nor linguistic meaning is no obstacle to the view that they are (sometimes) parts of significant expressions. After all, the proper name "Phosphorus", too, has several parts which have no kind of meaning. Suppose a sentence containing just one indicator is uttered twice. Then the two utterances make use of one and the same hybrid proper name if and only if they contain the same non-linguistic object and tokens of the same indexical expression. 12 Hence only utterances of "I" by the same person, and only simultaneous utterances of "the present moment" are occurrences of the same hybrid proper name, and this name-type obeys the ancient principle unum nomen, unum nominatum which is notoriously offended against even by "Aristotle". Was Frege himself prepared to accept the idea of hybrid proper names as a consequence of the incompleteness of indexical thought-expressions? He was indeed. In one of his posthumous writings he remarks on the demonstrative description "this person": 13 [F] A concept-word combined with the demonstrative pronoun or definite article often has... the logical status of a proper name in that it serves to designate a single determinate object. But then it is not the conceptword alone, but the whole consisting of the concept word together with the demonstrative pronoun and accompanying circumstances which has to be understood as a proper name. The proper name in question is hybrid: it consists of more than just a verbal expression. And what is the remaining part? Again Frege, somewhat vaguely, speaks of "accompanying circumstances", and [C] suggests that he might be thinking of pointing gestures and glances. 14 But they matter only to the interpreter of the utterance, and even the interpreter does not always need such clues: just imagine a motionless speaker, saying with closed eyes "This is a horrible noise (desperate smell)". I propose that we assimilate the demonstrative case to 1 ' I take the place of an (oral) utterance to be the region of space which is occupied by the utterer while she is making the utterance. 12 Ben Hoefer made me reflect on this. 13 (1914; cf p. 230) 14 Cp. Peirce: "The pointing arm is an essential part of the symbol without which the latter would convey no information" (1932, 2: 293).

6 726 Wolfgang Kunne that of 'T'-utterances and take the object demonstrated to be the essential nonlinguistic part of a hybrid proper name containing a demonstrative. Thus understood a hybrid proper name containing a "this" also conforms to the principle unum nomen, unum nominatum. Before reflecting on further consequences of this option let us look at two philosophers who clearly took it. Moore reports of a lecture by Wittgenstein given in the early thirties: 15 He made a distinction... between what he called "the sign" and what he called "the symbol", saying that whatever was necessary to give a "sign" significance was part of the "symbol", so where, for instance, the "sign" is a sentence, the "symbol" is something which contains both the sign and also everything which is necessary to give that sentence sense... He illustrated this by saying that if a man says "I am tired" his mouth is part of the symbol. Desmond Lee, after a discussion with Wittgenstein in , made the following note: If we say of an object, "This is 3 ft. high", then the object is part of the symbol.(1980, p. 115); (cf. pp. 7, 26 f) Using this terminology we can say that incomplete thought-expressions like (SI) to (S3) are signs. These signs are components of symbols (complete thoughtexpressions), the other components of which are the (relevant constituents of the) circumstances of the sign's utterance. Evidence for the affinity of Frege's and Wittgenstein's views in this regard is not limited to indirect sources like those quoted above. In the Philosophical Remarks which Wittgenstein wrote in , he says: 17 What characterizes propositions of the form "This is..." is only the fact that the reality outside the so called system of signs somehow enters into the symbol. And in the Brown Book which Wittgenstein dictated in , one finds the following reflection: 18 Nothing is more unlike than the use of the word "this" and the use of a proper name I mean the games played with these words, not the phrases in which they are used. For we do say "This is short" and "Jack is short"; but remember that "This is short" without the pointing gesture and without the thing we are pointing to would be meaningless. What can be compared with a name is not the word "this" but, if you like, the symbol consisting of this word, the gesture and the sample. 13 Moore (1959 p. 262). In the Tractatus Wittgenstein had used the distinction sign/ symbol quite differently (cp ft). 17 (1975, p. 120). 18 (1964, p. 109).

7 Hybrid Proper Names 72 7 What Wittgenstein in all these passages calls a symbol is a hybrid proper name in the sense of this paper which, for expository reasons, conforms to Frege's idiosyncratic use of the term "Eigenname"." And what Wittgenstein in the first and the last of those passages variously calls "significance", "sense" or "meaning" shares with Fregean sense the property of not being identical with linguistic meaning, for obviously "I am tired" and "This is short" are immaculate as English type-sentences. The idea of a symbol containing a non-verbal element is also present in the Philosophical Investigations, namely in Wittgenstein's remarks on ostensive definitions. I will touch upon that topic in the last section of this paper, when I talk about samples. (Incidentally, Wittgenstein's use of the term "sample" at the very end of the last quotation is misleading. The context shows that here he is not concerned with the use of "This is short" for the purpose of an ostensive definition of "short". So long as one uses such sentences to formulate a statement rather than a rule it hardly makes sense to call the object pointed to a sample.) Montz Schlick also made use of the idea of symbols (hybrid proper names) in his 1932 London lectures on "Form and Content"; perhaps in this as in so many other respects he was inspired by his many discussions with Wittgenstein in Vienna: If one morning the mail should bring you a letter containing nothing but a green leaf, you would not be able to make anything of it; you could record it as a simple fact, but it would not "mean" anything to you. On the other hand, the curious occurrence would have the character of a communication, it would be an actual message, if the leaf were accompanied by some explanation or if you had received some instruction concerning it. It might be a leaf someone promised to send to you from his garden, or there might be a note saying "I found this on my desk" or "Please observe the colour of the leaf or "This is the colour I spoke of yesterday", etc. In all these cases the object itself enters into the language as part of it, it has exactly the same function that a picture or description or other sign would have: it is itself a symbol in the symbolism called "language". The only peculiarity of this case is that the symbol has the greatest possible similarity to the signified object. (1979, vol. II, p. 303) IV If, in the case of demonstratives, Frege's view about incomplete thought-expressions is spelled out along Viennese lines then Frege cannot treat Macbeth's midnight utterance of (S4) This is a blood-stained dagger 19 Perhaps Wittgenstein was led to his conception by Frege's "The Thought". We know that he was familiar with the essay, as he wrote a letter (unfortunately not preserved) dated September 16, 1919, to Frege in which he thanked him for sending a copy of the essay and made "critical comments" about it. Cf. Gabriel (1979).

8 728 Wolfgang Kunne as he treats utterances with vacuous proper names like "the prime number between 20 and 22" or "the inventor of the Perpetuum mobile". Whereas in the latter cases it is only a lack of reference (and not one of sense) which has to be complained of, in the case of an utterance of (S4) by a speaker who is hallucinating like Macbeth a more radical failure would have to be lamented: no thought is expressed at all. If we take this option the claim which Gareth Evans (unjustifiably, I think) made on behalf of Frege for all vacuous proper names can be justified for the case of vacuous demonstratives: The speaker does not succeed in expressing a thought (Evans 1982, esp Ch. 1). This verdict is not concerned with the speaker's psychological state at the moment of making his utterance; it does not exclude that our speaker is thinking a thought at the moment of his utterance he might be convinced, for example, that there is a dagger before him. 22 The verdict is based on a damage of the symbols the speaker uses. This damage resembles the one we find in the following inscription presumably produced by a desperate printer: (S5) " " is a letter with which many Greek words begin. In both cases there is only a fragment of a proper name. Let me insert a note of caution here before making a final point. The conception of hybrid proper names as sketched above should not be expected to help us with the often discussed identity problems surrounding the notion of Fregean thoughts. Consider the following hypothesis: For two hybrid proper names to have the same Fregean sense it is a necessary as well as sufficient condition that they contain the same non-linguistic entity and indicator-tokens which have the same linguistic meaning (two "now"s, for example, or a "jetzt" and a "now"). Frege could not accept this hypothesis. That he does not hold the condition to be necessary can be seen from passage [B] above. And if we apply the test for difference in sense which is so memorably used on the first pages of liber Sinn und Bedeutung we soon recognize that it is not sufficient either. Consider this example. Mary is talking to a man on the phone. Simultaneously she looks out of her window and sees a man in a phone box. In this situation it might come as a great surprise to her to be told: That man [visually presented] - that man [auditorily presented]. Hence the hybrid proper name-tokens occurring in this informative utterance have different senses, and yet they contain the same man and tokens of the same indicator. V A hybrid proper name containing a token of"/", of the present moment", of "the place which is occupied by me now" 21 or of "this", refers to a part of itself. This 22 Cp. my (1987, pp ); and my (1990, pp ). 23 Please recall footnote 11.

9 Hybrid Proper Names 729 feature is perhaps less bizarre than it appears at first sight. After all, there are many expression-tokens which undeniably refer to parts of themselves. The definite description in the next line, for example, (56) The third word in line (S6) is a noun refers to a part of itself. More interesting for our purposes is a certain kind of quotation-name. Of course, the quotation-name (the expression consisting of quotation-marks and the sign enclosed) in the next line: (57) "e" is a letter with which many Greek words begin should not be taken to refer to that which is surrounded by quotation-marks, because that token, that inscription certainly does not have the property of occurring in many words. 24 What is said in (S7) is true only if the quotation-name does not refer to a concrete inscription but to what concrete inscriptions are tokens of. Under this reading the quotation-marks there are type-quotes. But we sometimes use quotation-names which do refer to one of those parts of which they consist, and we could introduce a special sort of quotation-marks for this sort of quotation, e.g. Reichenbach's token-quotes: 25 (58) v e' is an occurrence of a Greek letter. The quotation-name in line (S8) does not refer to (numerically) the same object as the next one: *e'. And both names refer to what is literally one of their parts. In this respect they resemble those hybrid proper names which were mentioned at the beginning of this section. And there is another remarkable property shared by quotation-names and these hybrid proper-names: They cannot lack a reference. In the light of Frege's view as spelled out in this paper it is as incorrect to classify an indicator (or an indexical definite description) by itself as a proper name, as it would be incorrect to classify quotation-marks by themselves as proper names. Both indicators and quotation-marks are expressions which can be used to generate proper names. How do quotation-marks perform this role? 26 In 24 Some letter-tokens do have the property of occurring in several word-tokens: Think of inscriptions in a crossword puzzle. 25 Reichenbach (1947, p. 284 ff). 26 It used to be said that quotation-names are semantically as unstructured as are many personal names. To be sure, the expression "Epsilon" really is a letter-name of the same semantic character as the personal name 'Tarski". Neither name is semantically structured: the occurrence of "silo" in "Epsilon" is irrelevant to an understanding of "Epsilon", just as the occurrence of "ski" in 'Tarski" has no bearing on an understanding of the latter. But is the quotation-name of the fifth letter of the Greek alphabet semantically unstructured? If it were, one should be able to ask sensibly with regard to it: Which letter has that name? That very question can be posed as regards the name "Epsilon" and can be answered by poinung to an object: "This letter is called Epsilon". That question, however, does not make sense if asked of the quotation-name of the same letter. If we know the role of quotation-marks, we eo ipso know for any quotation-name what it refers to. This would be a miracle if quotation-names did not have any semantic structure. For further discussion and references to the literature cp. Kunne, (1983, pp ); and Davidson (1984, pp ).

10 730 Wolfgang Kunne exactly the same way as the indicator "this". 27 After all, instead of (S7) we could write something like: (S7*) e <r- This is an occurrence of a letter with which many Greek words begin. Here the role of quotation-marks in (S7) is fulfilled by the indicator "this" and the arrow (used as an inscriptional gesture-substitute). And (S8) is nothing but a notational variant of (S8*) e < This is an occurrence of a Greek letter. Now if indicators are parts of quotation-names it is not really surprising that quotation-names can usefully be compared with hybrid proper names. Since the entities referred to in our last two numbered lines are themselves linguistic in nature, we do not hesitate to think of them as part of the thought-expression. Sometimes we find it almost as easy to characterize a non-linguistic object in the same way: when the object is used as a sample. "If I want to tell someone what colour some material is to be, I send him a sample, and obviously [sic] this sample belongs to language", albeit, as Wittgenstein whom I am quoting adds, not to the "Wortsprache" (1975, 16). Of course, one cannot very well claim that in an utterance of (SI), (S2) or (S3) the speaker, the time or the place of her utterance are used as samples. But in all these cases we can think of those entities, which have to be identified if an evaluation of the utterance in terms of truth and falsity is to be possible, as part of the thought-expression, and if I am right this is exactly what Frege took them to be. Philosophisches Seminar der Universitdt Hamburg Von-Melle-Park 6 D-2000 Hamburg 13 Germany REFERENCES WOLFGANG KUNNE Burge, Tyler 1979: "Sinning against Frege", Philosophical Review 88, pp \ Davidson, Donald 1984: Truth and Interpretation. Oxford: Clarendon Press. Dummett, Michael 1981: The Interpretation of Frege's Philosophy. London: Duckworth. Evans, Gareth 1981: "Understanding Demonstratives", in Parret, Bouveresse (eds.), Meaning and Understanding, Berlin: de Gruyter The Varieties of Reference. Oxford: Clarendon Press. Frege, Gottlob 1892: "Uber Sinn und Bedeutung", in Zeitschrift fur Philosophic undphilosophische Kritik, NF 100 (1892), transl. in Geach, Black (eds.), Translations from the Philosophical Writings of Gottlob Frege. Oxford: Basil Blackwell, Cp. Davidson (1984); KUnne (1983).

11 Hybrid Proper Names : "Logik", in Hermes et al. (eds.), Frege, Nachgelassene Schriften, Meiner: Hamburg, 1969, transl. in Gottlob Frege, Posthumous Writings Oxford: Basil Blackwell : "Logik in der Mathematik", in Nachgelassene Schriften, Posthumous Writings. Oxford: Basil Blackwell, : "Der Gedanke. Eine logische Untersuchung", in Beitrdge zur Philosophie des deutschen Idealismus 1( ), transl. in P.T. Geach (ed.) 1977: Logical Investigations, Gottlob Frege, Oxford: Basil Blackwell. Gabriel, Gottfried et al. (eds.) 1979: Frege, Wissenschaftlicher Briefwechsel Hamburg, Meiner. Kaplan, David 1975: "How to Russell a Frege-Church", Journal of Philosophy 72, pp Kiinne, Wolfgang 1983: Abstrakte Gegenstdnde, Semantik und Ontologie. Frankfurt/M, Suhrkamp. 1987: "The Intentionahty of Thinking", in Mulligan (ed.), Speech-Act and Sachverhalt, Reinach and the Foundations of Realist Phenomenology, Dordrecht: Nijhoff : "What One Thinks", in Rapp, Wiehl (eds.), Whitehead's Metaphysics of Creativity, New York: State University of New York Press. Lee, Desmond (ed.) 1980: Wittgenstein'sLectures Oxford: Basil Blackwell. Moore, George Edward 1959: Philosophical Papers. London: Allen & Unwin. Peirce, Charles Sanders 1932: Collected Papers. Harvard: Harvard University Press. Perry, John 1977: "Frege on Demonstratives", Philosophical Review, 86, pp Reichenbach, Hans 1947: Elements of Symbolic Logic. New York: The Free Press. Schlick, Moritz 1979: Collected Papers. Dordrecht: Reidel. Wittgenstein, Ludwig 1970: Philosophische Bemerkungen. Frankfurt/M: Suhrkamp. 1964: The Blue and Brown Books. Oxford: Basil Blackwell. 1970: Philosophische Untersuchungen. Frankfurt/M: Suhrkamp. 1975: Philosophical Remarks. Oxford: Basil Blackwell.