Hostname: page-component-78c5997874-ndw9j Total loading time: 0 Render date: 2024-11-10T07:13:45.210Z Has data issue: false hasContentIssue false
  • English
  • Français

Logicians’ Underlying Postulations

Published online by Cambridge University Press:  14 March 2022

Arthur F. Bentley
Get access Rights & Permissions [Opens in a new window]

Extract

Among the subject matters which logicians like at times to investigate are the forms of postulation that other branches of inquiry employ. Rarely, however, do they examine the postulates under which they themselves proceed. It long contented them to offer something they called a “definition” for logic, and let it go at that. They might announce that logic dealt with the “laws of thought,” or with “judgment,” or that it was “the general science of order“; More recently they are apt to connect it in one or another obscure way with linguistic ordering.

Type
Research Article
Information
Philosophy of Science , Volume 13 , Issue 1 , January 1946 , pp. 3 - 19
Copyright
Copyright © The Philosophy of Science Association 1946

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Notes

1 Sub-postulations within a wider, tacitly accepted (i.e. unanalysed) postulatory background are common enough. The present viewpoint is that of Morris R. Cohen when he writes: “The philosophic significance of the new logic, the character of its presuppositions, and the directions of its possible application are problems which have attracted relatively little reflective thought.” A Preface to Logic (1944) p. ix.

2 In his very latest publication Russell still writes: “From Frege's work it followed that arithmetic, and pure mathematics generally, is nothing but a prolongation of deductive logic,” The History of Western Philosophy, (945) p. 830.

3 Samples of logicians’ linguistic libertinism can be found anywhere, anytime, in current periodicals. Thus, for instance, in a paper just now at hand, we find “principles” of deduction referred to “intuition” for their justification, and this along with the suggestion that intuition should be “reenforced by such considerations as … ingenuity may suggest.” A few paragraphs later a set of “principles” containing wholly naive uses of the word “true” are declared to be “intuitively obvious.” Lack of humor here goes hand in hand with inattention to the simpler responsibilities of speech; (Journal of Philosophy, vol. 42, 1945, p. 450, p. 451.)

4 Compare John Dewey and Arthur F. Bentley “Postulations,” Journal of Philosophy, vol. 42 (1945), pp. 645–662. A group of postulations (A to H) is presented looking towards a natural theory of knowings and knowns. For purposes of comparison additional sketches are given of postulations (X, Y, Z) typical of current epistemologies, physiological psychologies, and prevalent forms of linguistically oriented logics.

5 In characterizing these logics as “natural,” it is to be understood that the word “natural,” as here used, is not to be taken as implying something specifically “material” as contrasted with something specifically “mental.” It stands for a single system of inquiry for all knowledge with logic as free to develop in accordance with its own needs as is physics or physiology, and to develop in system with either or both of these as freely as they develop in system with each other. See Dewey and Bentley “A Terminology for Knowings and Knowns,” Journal of Philosophy, vol. 42 (1945) pp. 225–247. Many logicians rated by us as non-natural would label themselves “naturalistic.” Thus Russell declares that he “regards knowledge as a natural fact like any other” (Sceptical Essays (1928) page 70), though our examination of his materials and procedures will give him quite the contrary rating.

6 If Otto Neurath had developed his position further than he has as yet, we could doubtless list him also on the “natural” side. He has from the beginning been much further advanced in this respect than others of his more active associates in the projected International Encyclopedia of Unified Science, of which he is editor-in-chief. His most recent publication is “Foundations of the Social Sciences,” a monograph contributed to the Encyclopedia.

7 “As far as I am aware, he (Peirce) was the first writer on logic to make inquiry and its methods the primary and ultimate source of logical subject matter,” John Dewey, Logic, the Theory of Inquiry, p. 9n. The fourth of the postulates for Dewey in the text is frequently called “the postulate of continuity,” and perhaps offers the straightest and widest route from Darwin through Peirce to Dewey's most complete development.

8 All page references are to the Logic. Professor Dewey has made further development since the Logic was published, particularly with respect to the organization of language, logical forms, and mathematics. Such advances are intimated, but not expressly set forth, in the numbered paragraphs of the text, since it is desirable, for all logics discussed, to hold the presentation to what can be directly verified by the reader in the pages of the works cited.

9 The radical nature of the advance in postulate 14 over older treatments will be plain when the postulations for Russell are considered. For the equally radical postulate 19 see postulate B-8, and its context, in the paper “Postulations,” by Dewey and Bentley, Journal of Philosophy vo. xlii (1945) p. 652.

10 All wordings are those of the section-headings of the postulates or of the immediately following text, except as follows: The sentence in 2.1 is from page 140, lines 11–12; the second sentence in 4. is from page 168, lines 3–4; the sentences in 4.1 are from page 294, lines 9–10; and page 7. lines 13–14; the sentence in 5 is from page xiii, lines 2–3; the second sentence in 7 is from page xiii, line 6.

11 For the distinction between interaction and transaction, see Dewey and Bentley, “A Terminology for Knowings and Knowns,” Journal of Philosophy, xlii (1945), p. 242, p. 246. A further development will appear shortly in the same journal.

12 This difference is well brought out by a remark of Bridgman's which Kantor quotes in order to sharpen his statement of his own position. Bridgman holds that “from the operational point of view it is meaningless to separate ‘nature’ from ‘knowledge of nature.2 “ Kantor finds Bridgman's view a departure from correct operational procedure. Dewey, on the contrary, would be in full agreement with Bridgman in this particular respect. (Bridgman, Logic of Modern Physics, p. 62; Kantor, “The Operational Principle in the Physical and Psychological Sciences,” The Psychological Record, 2 (1938), p. 6. For an appraisal of Kantor's work under a point of view sharply contrasted with that taken in the present text see the review by Ernest Nagel, Journal of Philosophy, xlii (1945) pp. 578–80.)

13 Typical confusions of logical discussion have been examined from a different point of view in my paper “On a Certain Vagueness of Logic,” Journal of Philosophy, vol. xlii (1945) pp. 6–27; 39–51. Certain characteritics of the work of Carnap, Cohen and Nagel, Ducasse, Lewis, Morris, and Tarski are there displayed. A thorough overhauling has long been needed of the procedures of Carnap and other logical positivists, both with respect to their logic and their positiveness, and this is now promised by us C. W. Churchman and T. A. Cowan (Philosophy of Science, vol. 12 (1945), p. 219). One device many logicians employ to justify them in maintaining the antiquated materials is their insistence that logic and psychology are so sahrply different that they must leave each other alone—in other words, that while psychology may be allowed to “go natural,” logic may not be so allowed. This argument of the logicians may be all very well as against an overly narrowed psychology of the non-natural type; but by the same token an overly narrowed logic results. The problem is one of full human behavior—how human beings have evolved with all their behaviors—no matter how convenient it has been found in the popular speech of the past to scatter the behaviors among separate departments of life.

14 For an extreme “mentalistic” and hopefully “solipsistic” base for logic, the procedures of C. I. Lewis may be brought into comparison by anyone sucfficiently interested. Lewis is represented by the following “postulates,” which, from any “natural” point of view, rate as disintegrating and unworkable traditions: (1) Knowledge involves three components, the activity of thought, the concepts which are produced by thought, and the sensuously given; (2) The pure concept and the content of the given are mutually independent; neither limits the other; (3) The concept gives rise to the a priori which is definitive or explicative of concepts; (4) Empirical knowledge arises through conceptual interpretation. See Mind and the World Order (1929), pp. 36ff; “The Pragmatic Element in Knowledge,” Univ. of California Publications in Philosophy, vol. 6 (1926); A Survey of Symbolic Logic (1918). A characteristic determination arising in such a background is that if “analytic facts” can “function propositionally,” then “they are called propositions“; so that “the proposition ‘men exist’ is literally one and the same with the fact that men might exist.” (Lewis and Langford, Symbolic Logic (1932) p. 472. For other illustrations of what happens to ordinary integrity of expression under such a construction see my notes on Lewis’ vocabulary, Journal of Philosophy (1941) pp. 634–5 and 1945 pp. 43–44.

15 See his discussion of Dewey's Experimental Logic (1916) reprinted as an appendix to his book A Preface to Logic.

16 All citations are from A Preface to Logic, (1944).

17 From private correspondence.

18 The book here characterized is Felix Kaufmann's Methodology of the Social Sciences (1944). Page references are not given as the presentation in the text has Professor Kaufmann's endorsement as it stands with the provisio that “he does not maintain that scientists always consciously apply the rules in their inquiries” but that “he does maintain the reference to the rules is logically implied when the correctness of scientific decisions or the appropriateness of the methods applied is judged.” “Formulations of such judgments which do not contain reference to the rules,” he regards as “elliptical.” The following citations, which Professor Kaufmann quite properly insists should be understood in the full context of the book, are assembled by the present writer who, properly also, he hopes, believes they are essential to show the manner in which expression under this procedure develops: “The contrast between deductive reasoning (in the strict sense) and empirical procedure . . . will be the guiding principle of our analysis and . . . the key to the solution of . . . problems” (op. cit. p. 3); “The most general characterization of scientific thinking” is “that it is a process of classifying and reclassifying propositions by placing them into either of two disjunctive classes in accordance with presupposed rules“ (p. 40); “The distinction between the logical order of meanings and the temporal order of inquiry” is “all-important” (p. 39); The “temporal aspect of inquiry does not enter into the timeless logical relations among propositions” (p. 30); “The fundamental properties of the system of rules are invariable” (p.232); the “genuine logical theory of empirical procedure” is ”a theory of correct scientific decisions in given scientific situations“ (p. 65); Language requires “a system of rules that gives to particular acoustical phenomena the function of symbols for particular thoughts” (p. 17); “Lack of distinction in language is, in most cases, the consequence of unclear thought.” (p. 27); “Concepts and propositions are meanings” (p. 18); in “problems of empirical science” and “logical analysis” . . . “we have to presuppose (elementary) meanings” (p. 19). Kaufmann reiterates and emphasizes his difference from Dewey in a late paper (Philosophy and Phenomenological Research, VI (1945) p. 63n.) when he states that he cannot follow Dewey when the latter dismisses “intuitive knowledge of meanings” along with “intuitive knowledge of sense-data.”

19 See the two typical marks of distinction between pre-Darwinian and post-Darwinian programs and goals suggested in the opening paragraphs of this paper. Kaufmann's tentets #5 and #7 mark steps of his advance.

20 The word “superordinate” is here employed by me as an evasive compromise. Kaufmann would say that “the meanings” are “presupposed in,” “essential to,” “logically implied by,” or “necessary for the definition of” the “inquiry.” I would say that what his development actually accomplishes is to retain them as “prior to,” “superior to,” “independent of,” or “in a realm apart from” the “inquiry.” He fully satisfies me that my wording is not what he intends, but without affecting my view that I am nevertheless describing what he in effect doss.

21 Direct compromise of particular phrases is not simple, because the whole method of expression—the “linguistic atmosphere“—varies so greatly. However, K2 may be compared with D10 and D15; K4 with D5; K7 with D19; and K10 with D18. In addition the citations about language in footnote 18, taken from Kaufmann's pages 17 and 27 are at the extreme opposite pole, so far as present issues go, from D16.

22 Or at least this seems to be the purport of such a conclusion as that “the goal of all our discussions” is “that complete metaphysical agnosticism is not compatible with the maintenance of linguistic propositions” (An Inquiry into Meaning and Truth, p. 437).

23 The sources of citations are indicated as follows:

M. “The Philosophy of Logical Atomism,” Monist, 1918, pp. 495–527; 1919, pp. 32–63, 190–222, 345–380. Page references are to the 1918 volume unless otherwise indicated.

LA. “Logical Atomism,” Contemporary British Philosophy, First Series, 1924, pp. 359–383.

RC. “Reply to Criticisms,” The Philosophy of Bertrand Russell, P. A. Schilpp, editor, 1944, pp. 681–741.

I. An Inquiry into Meaning and Truth, 1940.

24 What Russell intends by meaning is, in general, very difficult to determine. It is not that no light is thrown on the question but entirely too many kinds of light from too many points of view, without sifting. Most profitalbe is an examination of all the passages, a dozen or more, indexed in the Inquiry. See also M, 506–8; LA, 369, Mysticism and Logic, pp. 223–4.

25 For a discussion in terms of “basic propositions” see I, p. 172, p. 362, p. 414. Here the contrast between Russell and Dewey is so sharp (see Dewey, No. 14, preceding) that the extensive discussions between the two men could be reduced to a one-sentence affrmation on this point and a one-page exhibit of the context of discussion, historical and contemporary.

26 An alternative form will be found in a paper in Scientia, 1914, reprinted in Mysticism and Logic, 1917, p. 155: “Wherever possible, logical constructions are to be substituted for inferred entities.“

27 These are variously called logical atoms, ultimate constituents, simple entities, etc. “Such ultimate simples, I call ‘particulars’ “ An Analysis of Mind (1921), p. 193. They are the hardest of hard facts, and the most resistant to “the solvent influence of critical reflection.” They may be sense-data, or entities called “events” (LA, 381) or sometimes point-instants or event-particles. Mathematical-physical expressions sometimes join them among the ultra-safe. If Russell would establish definite usage for at least one or two of these words, his reader might have an easier time doing justice to him. It is particularly disconcerting to find the particulars turning out to be themselves just words, as where (I, 21) he speaks of “egocentric particulars i.e., words such as ‘this,’ ‘I,’ ‘now,’ which have a meaning relative to the speaker.” If “terms” are “words” for Russell (I would not presume to say) then the second sentence in point #7 in the text also makes particulars out to be symbols rather than entites.

28 Compare Problems of Philosophy (1912) p. 91; “On the Nature of Acquaintance,” Monist, (1914).

29 If there has been any systematic progress in Russell's work as the years pass by with respect to attitudes 8A and 8B, I have failed to detect it. The difference seems rather one of stress at different stages of argumentation. If the clash as here reported seems incredible, I suggest an examination of a particularly illuminating passage in Professor Nagel's contribution to The Philosophy of Bertrand Russell, p. 341, in which, though without directly mentioning the incoherence, he notes (a) that Russell holds that some particulars are perceived, and at least some of their qualities and relations are immediately apprehended; (b) that Russell believes his particulars are simples; and (c) that Russell admits that simples are not directly perceived, but are known only inferentially as the limit of analysis. Further light on the situation may be gained from Nagel's penetrating analysis of An Inquiry into Meaning and Truth. Journal of Philosophy, xxxviii (1941) pp. 253–270.

30 RC, 691; Principles of Mathematics, 1903; Introduction to Mathematical Philosophy, 1919; also, off and on, at any stage of his writings. Note the similar difficulty for “particulars,” point #7 and footnote 27.

31 I, p. 438, (the last sentence in the book): “Partly by means of the study of syntax, we can arrive at considerable knowledge concerning the structure of the world.“

32 For the latest illustration of Russell's general confusion of statement, pages 829–834 of his History of Philosophy, (1945) may profitably be examined. A passing glance will not suffice since the main characteristic of philosophical language is to make a good appearance. A cold eye, close dissection, and often much hard work is necessary to find out what kind of a skeleton is beneath the outer clothing.

33 Alfred Korzybski: Science and Sanity, an Introduction to Non-Aristotelian Systems and General Semantics, (1933).

34 Russell remarked to Professor Schilpp, the editor of the volume The Philosophy of Bertrand Russell, that “his greatest surprise, in the reading of the twenty-one contributed essays, had come from the discovery that ‘over half of their authors had not understood him’ (i.e. Russell).” (op. cit., p. xiii). For Moore see No. 9 and No. 10 of the skeleton construction of his logical procedure, which follows.

35 The sources of citations are indicated as follows:

CS. “A Defense of Common Sense,” Contemporary British Philosophy.“ Second Series, 1925, pp. 193–223;

RC. “A Reply to my Critics” The Philosophy of G. E. Moore, P. A. Schilpp, editor, 1942, pp. 535–677.

36 See also “Proof of an External World,” Proceedings of the British Academy, vol. XXV (1939) pp. 25–30. Professor Nagel's comment in his review of The Philosophy of G. E. Moore in Mind, 1944, p. 60–75 will be found of interest.

37 Included are physical objects, perceptive experiences taken as mental, remembered things, and other men's bodies and experiencings. “I think I have nothing better to say than that it seems to me that I do know them, with certainty. It is, indeed, obvious that in the case of most of them, I do not know them directly” …, but … “In the past I have known to be true other propositions which were evidence for them” (CS, 206).

38 “I think I have always both used, and intended to use, ‘sense-datum’ in such a sense that the mere fact that an object is directly apprehended is a sufficient condition for saying that it is a sense-datum” (RC, 639). A remarkable illustration of his careful expression may be found in the passage on page 181 of his paper, “The Nature of Sensible Appearances” Aristotelian Society, Supplementary Vol. VI (1926).

39 “Very many” is to be understood in the sense in which Moore uses the words (CS, 195), with a trend towards, but not immeditate assertion of, “all.”

40 Note the confidently reiterated “nobody can” and the “must.“

41 In the typical case, however, one concept is opposed to two or more concepts, these latter being accompanied in their consideration by explicit mention of their method of combination (RC, 666).

42 Moore has written: “I define the term (sense-datum) in such a way that it is an open question whether the sense-datum which I now see in looking at my hand and which is a sense-datum of my hand is or is not identical with that part of its surface which I am now actually seeing” (CS, 218). In simplified report his analysis in the case of “the back of my hand” discriminates “a physical object,” “a physical surface,” and a certain “directly seen” (such as one has in the case of an after-image or double-image). Moore's analysis with respect to the second and third of these has results which indicate to him that at the very time at which he not only feels sure but knows that he is seeing the second, he is in a state of doubt whether the third, which also he is seeing (and that directly in the indicated sense) is identical with the second or not; recognizing that it may be identical, in which case he is in a position of both “feeling sure and doubting the very same proposition at the same time” (paradox I); or “so far as I can see,” at any rate, “I don't know that I'm not” (Paradox II). It is to the second form of the paradox that the comment cited in the text above refers (RC, 627–653, and particularly 636–637, also CS, 217–219).

43 The analysis of Analysis which Moore offers (RC, 664–665) declares equivalence as to concepts between expressions of the form: this “concept” is “identical” with that, this “propositional function” is “identical” with that, and “to say this” is “the same thing” as “to say that.” But if we proceed to another form which also we feel we must accept, such as “to be this” is “the same thing” as “to be that,” we have, we are told, reached a paradox, which, as between expressions and concepts, remains unresolved.

44 It is significant in this connection that Moore tells us that it is always “things which other philosophers have said” that suggest philosophical problems to him. “I do not think,” he remarks, “that the world or the sciences would ever have suggested to me any philosophical problems.” The Philosophy of G. E. Moore, p. 14.

45 J. R. Kantor, op. cit. p. 223, pp. 282–3; also “An Interbehavioral Analysis of Propositions,” Psychological Record, 5 (1943) p. 328.

46 M. R. Cohen, op. cit., p. 30 Also: “Acts of judgment, however, are involved in the apprehension of those relations that are called meanings.” See also Cohen and Nagel, An Introduction to Logic and Scientific Method, 1934, pp. 27, 28, 392, where facts are made of propositions, and propositions are specifically declared to be neither physical, mental, linguistic, nor communication, and to be identifiable by the sole characteristic that whatever else they are they are “true or false.'

47 Felix Kaufmann, op. cit., pp. 18, 19.

48 B. Russell: An Inquiry into Meaning and Truth, 1940, pp. 208, 210, 217, 237 et al. Proceedings of the Aristotelian Society, Vol. XI, (1911), p. 117. Mysticism and Logic, (1918) p. 219; Monist (1918) p. 504.

49 See phrasings in Moore #1, #3, #5, et al. To Moore all such items are as familiar as the tongues of angels, and it is difficult, perhaps even impossible to find a direct cite.

50 Kaplan and Copilowish, Mind, 1939, pp. 478–484; Lewis and Langford, Symbolic Logic, p. 472; A. P. Ushenke, The Problems of Logic (1941) pp. 171, 175, 219; Roy W. Sellars in Philosophy and Phenomeological Research, vol. 5, (1944) pp. 99–100; G. Ryle, Proceedings of the Aristotelian Society, Suppl. Vol. IX (1929) pp. 80–96. An excellent start, and perhaps even a despairing finish, may be made with the Oxford Didctionary, or some other larger dictionary.