1 All references cited only by book, chapter, and section number are from Locke, John An Essay Concerning Human Understanding,ed. Nidditch, P. H. (Oxford: Clarendon Press, 1975).Google Scholar

2 Green, Thomas Hill Hume and Locke (New York: Thomas Y. Crowell), p. 20.Google Scholar

3 Gibson, James Locke's Theory of Knowledge and its Historical Relations (Cambridge: Cambridge University Press, 1931), pp. 166-67, 176.Google Scholar

4 For example, Mackie, John L. Problems from Locke (Oxford: Clarendon Press, 1976), p. 4,CrossRefGoogle Scholar writes that Locke's “definition of knowledge…is hard to reconcile with the reality of discoveries about the physical world.” Mabbott, John D. John Locke (london: Macmillan, 1973), p. 90,CrossRefGoogle Scholar writes that there is a “logical difficulty” which “comes out when [Locke] says that this belief [in the existence of real objects] concerns the agreement of two ideas (and not the relation between an idea and a real world.)” This statement expresses concern about the apparent logical conflict between saying that knowledge of real existence involves the agreement between ideas and that it involves agreement between ideas and the world; elsewhere on this page, however, Mackie appears concerned with a different problem, that of justifying knowledge of real existence. O'Connor, Daniel J. John Locke (New York: Dover Publications, 1967), p. 163,Google Scholar states that knowledge of real existence is quite different from the other three sorts of knowledge, since “a statement affirming that something exists does not assert a relation or the lack of a relation between two ideas.” Aaron, Richard John Locke (Oxford: Clarendon Press, 1971),Google Scholar also expresses concern with the problem of conflict between saying that knowledge is perception of the agreement or disagreement of ideas, and saying that “we can know that things (which are neither ideas nor relations between ideas) exist” (p. 238). Like some of the other commentators, Aaron tends not to distinguish sharply between this logical problem about the nature of knowledge and the problem of justifying claims about real existence.

5 Yolton, John Locke and the Compass of Human Understanding: A Selective Commentary on the Essay (Cambridge: Cambridge University Press, 1970), pp. 110-11.Google Scholar

6 Woozley, Anthony “Some Remarks on Locke's Account of Knowledge,” The Locke Newsletter,no. 3 (Spring 1972), p. 13.Google Scholar

7 Locke, Of the Conduct of the Understanding,ed. Garforth, Francis W. (New York: Teachers College Press, 1966),Google Scholar section 15.

8 Essay 2.21.5; emphasis added.

9 Locke, View of the Essay, in Lord Peter King, The Life of John Locke, with Extracts from his Correspondence, Journals, and Commonplace Books (london: Henry Colburn, 1829), p. 387.Google Scholar Emphasis added.

10 Woozley, op. cit., p. 15.

11 Hofstadter, Albert. Locke and Scepticism (Columbia University doctoral dissertation, 1935) (New York: Albee Press, 1935), p. 61.Google Scholar

12 Krüger, Lorenz Der Be griff des Empirismus: Erkenntnistheoretische Studien am Beispiel John Lockes (Berlin: Walter de Gruyter, 1973), pp. 146-47.CrossRefGoogle Scholar

13 O'Connor, op. cit., p. 163.

14 Locke, A Letter to the Right Reverend Edward Lord Bishop of Worcester, Works,vol. 4 (london: W. Otridge et al., 1812), p. 357.Google Scholar

15 2.33.19. Hofstadter also cites this passage, op. cit., p. 53.

16 Locke, Elements of Natural Philosophy, The Philosophical Works of John Locke,vol. 2 (london: George Bell, 1892), p. 495.Google Scholar

17 Locke, A Letter to the Right Reverend Edward Lord Bishop of Worcester, p. 357.

18 Locke, Elements of Natural Philosophy, p. 495.

19 2.32.19.

20 Hofstadter, op. cit., p. 60.

21 Kruüger, op. cit., pp. 143-44. Though it is true that the concept of propositions relevant here is the concept of affirmations and negations composed of ideas, these propositions need not be “mental propositions” in Locke's technical sense of propositions without words altogether. He remarks “that there is so close a conn ex ion between Ideas and Words … that it is impossible to speak clearly and distinctly of our Knowledge, which all consists in Propositions, without considering, first, the Nature, Use, and Signification of Language (2.33.19). Also see 3.9.21, and 4.6.3. However, one should not construct an interpretation of Locke's characterization of knowledge which requires that known propositions have attached verbal propositions, since knowledge does not necessarily involve words at all. Locke sometimes even recommends considering ideas without words: ”the examining and judging of Ideas by themselves, their Names being quite laid aside, [is] the best and surest way to clear and distinct Knowledge … (4.6.1 ).

22 Ibid., chap. 12.

23 Arnauld, Antoine The Port-Royal Logic,trans. Baynes, Thomas Spencer (London: William Blackwood, 1851), p.111.Google Scholar

24 Claims about co-existence are a partial exception to this generalization; I believe that he vacillates between treating these as general claims about ideas and as universally quantified claims about particulars, but a case for this interpretation would take me very far beyond the confines of this paper.

25 For example, see 4.1. 2.

26 The compatibility or incompatibility of concepts is another sort of agreement and disagreement relevant to idea-theoretic truth. See 4.5.8, and note 31 below.

27 Harrison, John and Laslett, Peter The Library of John Locke (Oxford: Oxford University Press, 1965).Google Scholar

28 Locke, Essai Philosophique concernant L'Entendement Humain,trans. Pierre Coste (Amsterdam, 1735).Google Scholar

29 Arnauld, op. cit., p. 111. The French text is found in Arnauld, Antoine La Logique, ou L'Art de Penser (Paris: Presses Universitairede France, 1965;Google Scholar first published 1662). A recent translation is The Art of Thinking,trans. Dickoff, James and James, Patricia (New York: Bobbs-Merrill, 1964).Google Scholar In the Dickoff-James translation, “conviennent” is translated “belong together.”

30 Locke, View of the Essay, pp. 395-96.

31 Locke does not discuss explicitly the case of verbal propositions asserting the existence of external things here; he alludes only to the truth of two other sorts of verbal propositions: (i) trifling verbal propositions, which have only what he terms “verbal truth;” verbal truth, he writes is that “wherein Terms are joined according to the agreement or disagreement of the Ideas they stand for without regarding whether our Ideas are such, as really have, or are capable of having, an Existence in Nature” (4.5.8). The “agreement or disagreement of ideas” to which Locke refers here must be a sort of idea-theoretic truth, namely the compatibility or incompatibility of two ideas. Locke also refers here to (ii) one type of real truth, namely that which arises when “these signs are joined, as our Ideas agree; and when our Ideas are such, as we know are capable of having an Existence in Nature” (4.5.8). This comment immediately follows the other one and is plausibly assumed to employ the same sense of “agreement or disagreement of ideas,” namely that involving mere compatibility or incompatibility of ideas. Locke claims here that real truth implies this relation of ideasp/us the possible existence of real things to which these ideas apply. Clearly he is talking here about the sort of real truth that pertains to mathematics and ethics, for he thinks that such truth requires only possible and not actual application. Truths asserting real existence cannot be subsumed under either category because such truth does require the actual existence of an external thing. My claim concerns the interpretation of “agreement or disagreement of ideas” which Locke must assume in order that his account of verbal truth can apply correctly to such cases.

32 2.23.29. Emphasis added.

33 For example, one may note the fact that both of the examples of knowledge that Locke gives at 4.1.2 are cases of idea-theoretic knowledge, for he mentions here only knowledge of the identity and diversity of ideas and mathematical knowledge.

34 For example, see the passages from Mackie and Aaron cited in note4 above.

35 For example, Woozley, op. cit., p. 15.

36 I believe that Locke may not intend to call into question all sensitive knowledge of physical objects in passages like 4.2.14; but the complex question of the epistemic status of sensitive knowledge in Locke raises many issues outside the bounds of this paper.