1 Collingwood, R. G., An Essay on Philosophical Method (Oxford: Clarendon Press, 1933), henceforth abbreviated as EPMGoogle Scholar.

2 Collingwood, R. G., An Essay on Metaphysics, rev. ed., with an introduction by Rex Martin (Oxford: Oxford University Press, 1998), henceforth abbreviated as EMGoogle Scholar.

3 Collingwood, R. G., The Idea of History, with an editorial introduction by T. M. Knox (Oxford: Clarendon Press, 1946)Google Scholar; rev. ed., with an introduction by Van, Jan der Dussen (Oxford: Oxford University Press, 1993), henceforth abbreviated as IHGoogle Scholar.

4 See Donagan, A., The Later Philosophy of R. G. Collingwood (Oxford: Clarendon Press, 1962)Google Scholar, esp. chap. 10, and also his “Collingwood and Philosophical Method,” in Critical Essays in the Philosophy of R. G Collingwood, edited by Krausz, M. (Oxford: Clarendon Press, 1972).Google Scholar

5 Rotenstreich, N., “Metaphysics and Historicism,” in Critical Essays in the Philosophy of R. G. Collingwood, edited by Krausz, M. (Oxford: Clarendon Press, 1972)Google Scholar.

6 Toulmin, S. ascribed to Collingwood the view that the relativity of concepts and attitudes to historical contexts entails historical relativism in “Conceptual Change and the Problem of Relativity,” in Critical Essays in the Philosophy of R. G. Collingwood, edited by Krausz, M. (Oxford: Clarendon Press, 1972), pp. 212–13Google Scholar.

7 Other such commentators include Rubinoff, Lionel (Collingwood and the Reform of Metaphysics [Toronto: University of Toronto Press, 1970])Google Scholar and Mink, L. O. (Mind, History and Dialectic [Bloomington, IN: University of Indiana Press, 1969])Google Scholar.

8 See James Connelly's “Metaphysics and Method: A Necessary Unity in the Philosophy of Collingwood, R. G.,” Storia, Antropologia e Scienze del Lin-guaggio, 5, 1–2 (1990): 33-156;Google ScholarModood's, Tariq “The Later Collingwood's Alleged Historicism and Relativism,” Journal of the History of Philosophy, 27 (1989): 101–25;CrossRefGoogle ScholarOldfield's, A. “Metaphysics and History in Collingwood's Thought” and R. Martin's “Collingwood's Claim that Metaphysics is a Historical Discipline,” both in Philosophy, History and Civilization: Interdisciplinary Perspectives on R. G. Collingwood, edited by Boucher, D., Connelly, J. and Modood, T. (Cardiff: University of Wales Press, 1995)Google Scholar.

9 Collingwood, R. G., Speculum Mentis (Oxford: Clarendon Press, 1924), henceforth abbreviated as SMGoogle Scholar.

10 At the time of writing the editorial introduction to this book, Knox had access to the manuscript of Collingwood's The Principles of History (herein abbreviated as PH). The manuscript went missing after Knox completed the editorial work on The Idea of History and only small parts of it were integrated in the text of IH. Knox based his judgement that the later philosophy of Collingwood is historicist and relativist partly on his reading of a manuscript that was not available to Collingwood scholars until recently, when the manuscript of The Principles of History was accidentally discovered in the basement of Oxford University Press and subsequently published (edited and with an introduction by Dray, W. H. and Van, Jan der Dussen [Oxford: Oxford University Press, 1997])Google Scholar. The publication of this newly found manuscript has made it possible to see more clearly the nature of Knox's editorial interventions in the text of IH, in particular, his choice of which sections of PH to include in IH. These are discussed by Dray and Van der Dussen in their introduction to The Principles of History. I would concur with Dray and Van der Dussen's view (p. xxiii) that it is difficult to see why Knox should have concluded that the later philosophy of R. G. Collingwood is historicist and relativist on the basis of a reading of The Principles of History. Rather tentatively, I would advance the thesis that the only difference worthy of note between IH and PH is one of emphasis. In PH, Collingwood's emphasis seems to be more on history as a cultural process than on the logical structure of historical explanations. Whereas the concern with history as a cultural process is present in IH, it is more fully developed in PH. It is perhaps the later Collingwood's interest in a more substantive conception of history (history as a cultural process) rather than in a merely formal conception of history (history as a kind/form of explanation) that led Knox to formulate the view that the later Collingwood fundamentally altered his philosophical outlook.

11 This term was coined by Rubinoff, L.. It was originally introduced in “Collingwood and the Radical Conversion Hypothesis,” Dialogue 5, 1 (1966): 71–83CrossRefGoogle Scholar. Rubinoff also discussed the radical conversion hypothesis in Collingwood and the Reform of Metaphysics (Toronto: University of Toronto Press, 1970)Google Scholar.

12 See Donagan, , The Later Philosophy of R. G. Collingwood, esp. chap. 10, pp. 263–65Google Scholar.

13 Metaphysics without ontology is a metaphysics of experience, a study of the conditions of the possibility of experience. Collingwood's metaphysics of experience is Kantian in its inception and, like Kant's, should not be confused with dogmatic or pre-critical metaphysics. The latter sought to uncover the nature of things rather than the structures of knowledge. I have explored the relationship between Kant's and Collingwood's metaphysics of experience in “How Kantian Is Collingwood's Metaphysics of Experience?” Collingwood Studies, 6 (1999): 29–52Google Scholar.

14 An analogous assessment of the relationship between Collingwood's earlier (Speculum Mentis and An Essay on Philosophical Method) and later work (An Essay on Metaphysics) is given by Rotenstreich. He believes that the development of Collingwood's thought is best understood as consisting in the gradual weakening of Collingwood's ontological commitments. He also seems to believe that Collingwood's epistemological reform of metaphysics is responsible for his trespassing in the area of the sociology of knowledge: “What actually emerges from Collingwood's An Essay on Metaphysics is a kind of cultural anthropology of metaphysics, a study of different world views entertained in the course of history by different individuals and by groups of individuals…. His view is formulated both from the level of his own intentionality and from the level of a historical or cultural anthropology referring to views and revealing their disguised historicity. He is caught in the dilemma similar to that of the sociologists of knowledge who look at the mind as the function of a social reality and still assume the independence of mind by way of attributing it the capacity of detachment for the sake analysis of the phenomena dealt with in the sociology of knowledge” (“Metaphysics and Historicism,” in Critical Essays in the Philosophy of R. G. Collingwood, edited by Krausz, M. [Oxford: Clarendon Press, 1972], pp. 199–200)Google Scholar.

15 By a transcendental argument I understand an argument which takes experience as a fact to be explained and which argues from the conclusion (the fact of experience) to the premises (the conditions of its possibility or logical ground). Such arguments are circular because the adoption of the premises is recommended on the basis of their ability to explain how the conclusion is possible, i.e., the premises of a transcendental argument have no independent support. Collingwood acknowledged this circularity by stating that in a philosophical argument the principles put forward in explanation of a form of experience are answerable to that form of experience. He would have probably agreed with the claim made by many other philosophers that such circularity is unavoidable and that, far from avoiding it, one should “leap into the circle.”

16 See Collingwood's discussion of the relativity of causes in EM. For a discussion of the doctrine of the relativity of causes, see Dray, W. H., “Historical Causation and Human Free Will,” University of Toronto Quarterly, 29 (1960): 357–69CrossRefGoogle Scholar, and Martin, R., “Collingwood on Reasons, Causes and the Explanation of Action,” International Studies in Philosophy, 23 (1991): 47–62CrossRefGoogle Scholar.

17 There is, for Collingwood, a legitimate sense in which “rational” explanations are “causal.” Rational explanations are causal in what he refers to as the first sense of the term cause, where the cause of an action is described as the combination of two elements: the causa ut (the agent's motivation) and the causa quod (the agent's beliefs).

18 Rotenstreich observes, for instance that although Collingwood's interest in the ontological proof is to be found both in EPM and EM, only in the former is it employed to provide a justification of the existence of the subject matter of philosophy. In Collingwood's EM, Rotenstreich says, “[t]he ontological proof does not carry religion beyond its own realm, as in Speculum Mentis; neither is it a paradigmatic argument for metaphysical thinking, qua thinking where essence involves existence, as in An Essay on Philosophical Method. Here [in EM] the ontological proof presupposes a belief. The scholastic argument is a presupposition and has no independent validity. Instead of being a paradigm of the character of philosophical thinking, the ontological proof becomes an illustration of the dependence of thinking on changing presuppositions. This change occurs in history” ("Metaphysics and Historicism,” p. 187) Along similar lines, Donagan claims that in Collingwood's later thought “Metaphysics … is not an attempt to establish categorical universal propositions about Being itself, but categorical particular propositions about what this or that people at this or that time have believed” (The Later Philosophy, p. 263). Donagan is aware that this interpretation may be vulnerable to the objection that Collingwood renewed his support for the ontological proof even in EM, when he claimed to be doing metaphysics without ontology, but Donagan explains this oddity by saying that although in An Essay on Metaphysics Collingwood “continued to recognise St Anselm's ontological proof as a major philosophical achievement,” he had, by then, re-interpreted the proof “as a specimen of reformed metaphysics” (ibid., p. 265).

19 An appropriate account of Collingwood's rehabilitation of the ontological proof would need independent treatment. I have discussed the significance of Collingwood's rehabilitation of the ontological argument in “On Collingwood's Rehabilitation of the Ontological Argument,” Idealistic Studies, 30 (2000): 173–88CrossRefGoogle Scholar. What I hope to do here is to sketch some of the arguments that may be adduced in support of the claim that Collingwood did not, even in EPM, employ the ontological proof to establish substantive ontological conclusions. For an account of Collingwood's treatment of the ontological proof, see Felser's, J. M.R. G. Collingwood's Early Philosophy of Religion and Its Development, Ph.D. dissertation (Chicago, IL: University of Illinois, 1992), Vol. 2, chap. 4Google Scholar. Collingwood's rehabilitation of the ontological proof was attacked by G. Ryle in both “Mr Collingwood and the Ontological Argument,” Mind, 44 (1935): 137-51, and “Back to the Ontological Argument,” Mind, 46 (1937): 53-57. Collingwood defended his version of the ontological proof in a private epistolary exchange with Ryle which is deposited in the Bodleian library.

20 The Collingwood-Ryle correspondence, Collingwood's letter of 9/5/1935.

21 Brentano, Franz, Psychologie von empirischen Standpunkt (Hamburg: Felix Meiner, 1925;Google Scholar reprinted in 1955 in Realism and the Background of Phenomenology, edited by Chisolm, R., translated by Terrel, T. B. (Glencoe, IL: Illinois Free Press, 1960)Google Scholar. For an account of Brentano's phenomenology, see Aquila's, R. E.Intentionality: A Study of Mental Acts (University Park, PA: The Pennsylvania State University Press, 1977), esp. chap. 1Google Scholar.

22 IH, Epilegomena 4.

23 This account of re-enactment differs from the one provided by Dray, W. H. in the following respect. Dray's major concern in his recent book, History as Re-enactment: R. G. Collingwood's Idea of History (Oxford: Oxford University Press, 1995)Google Scholar, and in his previous work on Collingwood, lies with the logical form of re-enactive or historical explanations. My account of re-enactment, on the other hand, focuses not on the logical form of explanation but on the criteria for the identity/non-identity of thought. Collingwood, I argue, assumes (very much like Locke in his discussion of personal identity) that criteria of identity are relative to the kind/sort of thing under consideration and hence that the kind-of-thing thought is/determines what are the appropriate criteria for identifying and distinguishing thoughts. I believe these two accounts of re-enactment (from the point of view of a discussion of the logical form of historical explanations and from the point of view of criteria for the identity of thought) to be complementary rather than mutually exclusive. For a more thorough account of Collingwood's account of re-enactment from the perspective of a discussion of criteria for the identity of thought, see Saari, H., “R. G. Collingwood on the Identity of Thought,” Dialogue, 28 (1989): 77–89CrossRefGoogle Scholar, and my own “Collingwood on Re-enactment and the Identity of Thought,” Journal of the History of Philosophy, 38 (2000): 87–101CrossRefGoogle Scholar.

24 Collingwood, , “Outlines of a Philosophy of History,” in IH, p. 479Google Scholar.

25 In The Idea of History, Collingwood argued this point by insisting on the necessity of distinguishing between what he interchangeably calls sensations/ feelings/mind in its immediacy/the flow of consciousness on the one hand and thought/mind in its mediation on the other. He offered the following three criteria for distinguishing between feelings/sensations and thought. First, as we have already seen, thought “stands outside time” in the sense that it is inappropriate to ascribe spatial and temporal properties to it. Feelings/sensations, on the other hand, have spatio-temporal location. Secondly, feelings/sensations, unlike thought, are private to the person who entertains them whereas thought is not (we can entertain the same thought process but we cannot have the same brain process). Finally, feelings/sensations can be counted as physical objects whereas thought cannot: thoughts are numerically distinct only to the extent that they are qualitatively discernible. Whereas Collingwood's distinction between feelings/sensations was not devised to address a claim within the sociology of knowledge, it is clear that it could be employed as a tool to undermine any attempt to conflate the object of knowledge (understood as an intentional object) with the context of its occurrence, be this the brain, as in the case of the mind-brain identity theory, or social existence, as in the case of the sociology of knowledge.

26 The critical edge of Collingwood's descriptive metaphysics clearly emerges in The Idea of History in his discussion of the three conceptions of history (common sense, “scissors and paste,” and scientific history) and his denunciations of scissors-and-paste history as pseudo history.

27 Collingwood referred to this form of history as scientific history not because its method resembled that of natural science but, on the contrary, because it is a form of enquiry which is distinct from natural science. This kind of history is said to be scientific because it has its own domain of enquiry and has therefore attained to the original Latin meaning of the word scientia, meaning a systematic body of knowledge with its own method and subject matter.

28 A similar view is put forward by Rex Martin in his introduction to a recent edition of An Essay on Metaphysics (Oxford: Clarendon Press, 1998)Google Scholar. According to Martin, the presuppositions about the nature of a science can be ordered into a scale of forms in accordance with their progressive phases of development (pp. xliv-xlv). Collingwood's progressive ordering of different conceptions of history is, for Martin, an instance of a scale of forms.

29 See EM, pp. 302–303.