Collingwood and Philosophical Methodology, Special Issue, Collingwood and British Idealism Studies. Giuseppina D\'Oro and James Connelly (eds.) 2016.

On this see Price, H. 2009. 'Metaphysics after Carnap: The Ghost Who Walks?' in Metametaphysics, eds. D. Chalmers, D. Manley and R. Wasserman. Oxford and New York: Oxford University Press, 320-47.
The manuscripts of Collingwood's lectures on moral philosophy are deposited in the Bodleian library, Oxford.

As defended, for example, by Jonathan Dancy in his Ethics without Principles (2004)
See Hume (1748), An Enquiry Concerning Human Understanding, section VII.
On this see D'Oro and Sandis 2014.
See however, D'Oro (2015 and 2016).



Introduction
Giuseppina D'Oro and James Connelly
Collingwood is the author of what are arguably the two most substantial metaphilosophical treatises of the twentieth century, An Essay on Philosophical Method and An Essay on Metaphysics. He was deeply concerned with the question of the relation between the special sciences, and how philosophy could act as a referee between what he regarded as the quintessential human science, history, and the natural sciences. He was also concerned with the very status of philosophical enquiry and whether philosophy is an historical discipline. Although often on the margins of mainstream philosophical discussion, Collingwood's account of the role and character of philosophical analysis and the relation between his two explicitly metaphilosophical treatises has been the subject of extensive discussion within Collingwood scholarship. The question as to whether or not An Essay on Metaphysics represents a substantive historicist departure away from the earlier An Essay on Philosophical Method and the exact meaning to be attached to his avowed aim of accomplishing a rapprochement between philosophy and history has been hotly debated, albeit in a rather insular manner, in a climate where philosophers mostly preferred to concern themselves with first order claims rather than ask what it is they are doing when they are doing philosophy. The recent resurgence of interest in philosophical methodology therefore provides an opportunity to reconsider Collingwood's contribution to philosophical method and locate it in the wider metaphilosophical debate.
After a period of relative neglect, questions concerning the nature of philosophical analysis have come back to the fore of philosophical discussion with the publication of explicitly metaphilosophical works, such as Williamson's (2010) The Philosophy of Philosophy and a number of high profile edited collections such as Chalmers et al. (2009) Metametaphysics and Haug's (2014) The Armchair or the Laboratory?. Questions concerning the method and subject matter of philosophy, the relation between philosophy and the special sciences, and the nature of philosophical progress, which had dominated during a period of Wittgensteinian consensus that tailed off towards the end of the twentieth century, have gained front stage once again. By and large this revival of interest in philosophical methodology has been prompted by the perceived need to articulate a defence of the autonomy of philosophy in the face of the growth of naturalistic assumptions and the widespread view that all ontological questions or questions about what exists, are ultimately answerable by the methods of natural science. As the century drew to a close the Quinean view that philosophy is continuous with the natural sciences and that any differences between the two is a distinction in degree rather than in kind became the standard view. The rise of this naturalistic orthodoxy is often traced back to the Carnap/Quine dispute and the widespread view that Quine won the argument against Carnap's defence of the analytic/synthetic distinction in his influential 1950 essay 'Empiricism, Semantics and Ontology.' Here Carnap (1950) argued that the first order existence claims made by the special sciences entail linguistic frameworks which are not truth-evaluable in the manner of first order claims. For example, the claim "there is a cat on the mat" analytically entails what Carnap called the linguistic framework of spatio-temporal objects. But while "there is a cat on the mat" has a definite truth-value, the linguistic framework of spatio-temporal objects does not, because questions concerning the ontological status of frameworks, as opposed to existence questions internal to them, are unanswerable. For Carnap while is legitimate to ask whether it is true that "there is a cat on the mat" it is a mistake to ask whether "there is a framework of spatio-temporal objects" is true because frameworks provide higher level conditions which make internal true/false possible. Quine's rejection of the analytic/synthetic distinction in 'Two Dogmas of Empiricism' (Quine 1951) was widely seen as undermining the Carnapian view that there is a principled distinction between the synthetic claims advanced within the first order sciences and the frameworks within which those claims are advanced. For Quine our statements "face the tribunal of sense experience not individually but only as a corporate body". Statements which one could not possibly conceive to be false or which appear to be irrefutable, are only more central to the web, not distinct in kind from statements whose questioning does not appear to be problematic. Quine's view that there is no distinction in kind between the claims articulated by the first order and second order levels of investigation thus strikes at the heart of the view that philosophy has both a method and subject matter that is distinctive from that of the first order sciences. While many hailed Quine as having saved the day for ontology, rescuing it from the Carnapian view that all legitimate ontological claims are internal to linguistic frameworks, a growing number of philosophers are now invoking his notion of linguistic frameworks in contemporary debates in meta-ontology where they are often mobilized as a double edged sword to a) undermine traditional metaphysical enquiry by claiming that external questions concerning the existence of linguistic frameworks are unanswerable pseudo-questions, and b) to reinstate a clear division of labour between the subject matter of the special sciences and that of philosophy.
By now the relevance of this excursus on the Quine/Carnap debate to Collingwood's metaphilosophy may have become apparent. Collingwood's later work is best known for seeking a rapprochement between philosophy and history and this rapprochement is sometimes construed as an attempt to weaken the distinction between first and second order level of analysis thereby depriving philosophy of its own distinctive subject matter. This historicization of philosophy is arguably a sort of naturalization by other means since, once the distinction between the higher level concepts or presuppositions that are normative for the practitioners of the special sciences and the first order claims internal to them is weakened, so is the distinction between philosophy and other forms of knowing. The issues at stake between Carnap and Quine are therefore at the heart of Collingwood's metaphilosophy and indeed to an ongoing debate within Collingwood's scholarship. Does Collingwood intended rapprochement between philosophy and history entail that all ontological claims are historically relative so that what exists is ultimately determined by an historical fact, i.e., what is believed at a particular time and place? And if so, did he think of the philosopher as the mere under-labourer of the historian and perhaps of philosophy as a branch of history? Or was Collinwood, on the other hand, arguing that there is a distinction in kind between the first order claims of the special sciences and the uncovering of the presuppositions that are normative for them? If so, how could this attempt to preserve a special role for philosophy be reconciled with his avowed goal of bringing about a rapprochement between philosophy and history? What is at stake in these interpretative questions is whether or not Collingwood was committed to a version of the internal/external distinction sufficiently robust to support the view that philosophy has a subject matter and method that is distinctive from that of the special sciences.
Much of what Collingwood says would appear to suggest that the first order claims advanced in the special sciences presuppose concepts or principles which are not in themselves truth-evaluable and thus different in kind from the claims advanced at first order level, to which the notion of truth and falsity does indeed apply. For example, the claim that prescribing a dose of aspirin lowers a patient's temperature is truth-evaluable, but the manipulability conception of causation (what Collingwood calls cause in sense II) that underpins such a claim is not. This suggests that the absolute presuppositions which structure the first order sciences are beyond verification, not because we are unable (because of certain limitations of the human intellect) to know whether such concepts correspond to an enquiry independent reality, but because, as they supply the criteria for truth and falsity, it is incoherent to ask whether they are true or false.
Understood in this way the task of philosophy could be seen as contiguous with the (Kantian) search for the a priori conditions of first order knowledge rather than the Quinean weakening of the internal/external distinction. To be sure, absolute presuppositions would supply a priori conditions of knowledge not in the sense that they are absolutely necessary and universal (and this is indeed a significant departure from the Kantian project) but rather because they are constitutive of the practices which they underpin. Causes, in sense I, are necessary to explain actions historically much as causes in sense II are necessary to explain occurrences in the practical sciences of nature. They are constitutive of those practices. But while this constitutive notion of the a prioricity of absolute presuppositions departs in important respects from Kant's project to spell out conditions which are absolutely necessary and universal, it is also not undermining of the internal/external distinction and with it the view that philosophy and the special sciences have a very different modus operandi.
What meaning is then to be attached to Collingwood's description of his project as accomplishing a rapprochement between philosophy and history? This is the topic of the first two contributions by Vasso Kindi and Richard Allen.
In "Collingwood, Wittgenstein, Strawson: philosophy and description" Kindi defends a sense in which presuppositions can be factual without ultimately reducing philosophy to empirical science. She distances Collingwood from the Kantian and Strawsonian projects of uncovering structures of knowledge that are absolutely universal and argues that the presuppositions uncovered by metaphysical analysis are less like analytic necessities and more like empirical propositions hardened into presuppositions as in Wittgenstein. By contrast for Allen ("Philosophy and the articulations of absolute presuppositions") Collingwood's rapprochement between philosophy and history consists in the fact that philosophy always begins its reflection from given practices. There has to be, as he says "a concrete body of knowledge or established practice for there to be any 'philosophy of' it" for the only alternative to such a philosophy would be "an entirely abstract, ahistorical and constructive one, which has no reference to the actual world". To this extent the task of uncovering absolute presuppositions is historical (since the metaphysician begins with forms of knowledge that are historically given) and also fulfils the logical project of making explicit what is already implicitly known as set out in An Essay on Philosophical Method.
Collingwood's attempt to disentangle the presuppositions which govern different forms of enquiry in An Essay on Metaphysics is rarely mentioned in relation to his account of the overlap of classes in An Essay on Philosophical Method and with few exceptions Collingwood's moral philosophy is largely under-explored, possibly because his lectures on Moral Philosophy are mostly still available only in manuscript form. Yet the discussion of the varieties of goodness in An Essay on Philosophical Method is arguably a precursor to his discussion of the need to disambiguate contested concepts such as the concept of "cause" in An Essay on Metaphysics. As a result, an understanding of Collingwood's contribution to moral philosophy, and where it sits on the spectrum of positions from consequentialism to virtue ethics is largely absent from contemporary debates in normative ethics where deontology and utilitarianism are normally seen as competing ethical positions. In An Essay on Philosophical Method, by contrast, Collingwood speaks of utilitarianism's teleological/hedonistic approach and deontology's right-based approach as articulating different conceptions of the "good". This strategy seems to dovetail An Essay on Metaphysics' decision to characterize the kind of explanations which are operative in the human and natural sciences as presupposing different senses of "cause". This approach leaves open the possibility that an action may be described as good on different grounds and yields a distinctive philosophical position which characterizes the agent's moral obligations neither in consequentialist nor in de-ontological terms. The agent's duty is to apply the notion of good that is called for in the situation in which she acts. This is not a form of particularism (for it does not deny there are moral principles) and Collingwood's metaethics, in so far as it can be gleaned from his metaphilosophy, seems to yield a very distinctive approach to normative ethics. Daniel's contribution ("Deriving Collingwood's meta-ethics: absolute presuppositions as fundamental principles of morality") focuses on Collingwood's conception of philosophy as the uncovering of absolute presuppositions but does so in the context of moral philosophy in order to make explicit his contribution to metaethics.
In "At loggerheads" Maarten Steenhagen discusses Collingwood's rejection of adversarial style in philosophical discussion and argues that Collingwood's rejection of the adversarial style has deep roots in his metaphilosophical views. This is a very distinctive aspect of Collingwood's rejection of adversarial engagement which, unlike more recent criticisms of it, is the result of his conception of the role of philosophical analysis. The goal of philosophical analysis, for Collingwood, is not to classify views into true and false but to uncover the presuppositions on which opposing claims in philosophical debate rest. Collingwood's non-confrontational approach to philosophical engagement as promoting mutual understanding, rests on his rejection of realism and his view that in order properly to be understood any answer we give must be located in the context of the questions which it seeks to answer. Collingwood's rejection of the adversarial style in philosophy is thus more than a response to a contingent political goal, such as the desideratum to redress gender imbalance in the philosophical profession.
There has been much discussion of Collingwood's relationship to Kant and Hegel in the scholarly literature. The discussion has often been fuelled by an attempt to pinpoint the exact nature of Collingwood's idealism and where he stands in relation to the idealism/realism debate. In "Dialectics and typology: narrative structure in Hegel and Collingwood" Díaz-Maldonado compares Hegel's and Collingwood's conception of history and argues that whereas like Hegel Collingwood did not see history as a mere flux of disconnected events, he did not share Hegel's view that history is the unfolding of a telos present in inchoate form from the very beginning. The notion of plan, for Collingwood, is necessary only in a Kantian sense, as a regulative idea or presupposition of historical narrative.
Collingwood was acutely interested in the presuppositions of the special sciences; accordingly, the last three essays of this special issue explore his contribution to the philosophy of science, the philosophy of history and the philosophy of art, and how Collingwood's discussion of these is influenced by his metaphilosophical approach.
In An Essay on Metaphysics Collingwood argued that the failure to disambiguate the different meanings of the term "cause" generates unnecessary conflict between forms of enquiry. Once it is acknowledged that different senses of causation answer to different explanatory imperatives, the conflict between forms of enquiry is deflated, because such conflict would be genuine only if the (allegedly competing) explanaions were answering one and the same question. But there are as many different senses of "cause" or "because" as there are different senses of "why". One must make sure not to mismatch them. As he puts it in The Idea of History
When a scientist asks "why did that piece of litmus paper turn pink?" he means "on what kind of occasions do pieces of litmus turn pink?" When an historian asks: "why did Brutus stab Caesar?" he means "what did Brutus think, which made him decide to stab Caesar?" (1946/1993, 214).
Collingwood also argues that we understand the necessary connection between events in nature by extending the connection holding between motivation and action at work in sense I. Although this point has not generally been noted, Collingwood's claim appears to be an explicit critique of Hume's belief that the origin of our idea of causal connection comes from the repeated observation of constant conjunctions and that the idea of causation could not possibility be derived from an act of introspection or reflection on the apparent necessity holding between what we will and our bodily movements. Hume went to great lengths to refute the possibility that the idea of causation could be derived from something akin to what Collingwood calls sense I of causation (where cause captures the explanatory conceptual relation holding between motivation and action) by arguing that if there were a necessary connection between the will and bodily movements we would be able (by an act of will) to change the orbits of the planets or command our internal organs. In a reversal of Hume, Collingwood claims that we would have no notion of causation in sense III (the sense in which it is used in the natural sciences) were it not for our possession of the concept of causation in sense I. Unlike Hume Collingwood did not think of cause in sense I as requiring an ontological interaction between the mind and the body, but as a necessary conceptual relation between an action and the motives it expresses. But it is nonetheless this sense of causation, which he takes to be logically primary and the origin of the idea of causation in general. In "Collingwood and manipulability-based approaches to causation" Elena Popa discusses Collingwood's claim that the sense of causation at work in history (or more generally in hermeneutic contexts) is logically prior to both the sense of cause at work in the practical and the theoretical sciences of nature (senses II and III), locates his discussion in the context of contemporary manipulability approaches to causation and shows how Collingwood's own account is embedded within his conception of metaphysics as the study of absolute presuppositions.
In recent years the philosophy of action has grown considerably. It has unfortunately lost touch with its roots in the philosophy of history understood as a geisteswissenschaft, or as the study of the mind, which is very much Collingwood's own conception of history's subject matter. As the study of mind, history is to be contrasted with the study of nature. It is concerned with actions as an expression of thought, unlike the natural sciences which are concerned with what occurs as a manifestation of laws. History is therefore concerned primarily with thoughts, be these the thoughts of our contemporaries or of those living in a distant past. However, for Collingwood, understanding the actions (qua expressions of thought) of those living in a distant past poses special challenges, for they do not always share the same thought context or epistemic premises of the interpreter. In this respect the challenges the interpreter faces in making sense of agents living in the distant past are not dissimilar from those of interpreting agents living in faraway lands. Constantine Sandis's "Period and place: Collingwood and Wittgenstein on understanding others" explores the affinities between Collingwood's and Wittgenstein's account of understanding others, be they our remote ancestors or our faraway contemporaries. While Collingwood tends to focus on period and Wittgenstein on place, they share an account of understanding others that is underpinned by the anti-causalist account of action explanation that tended to prevail prior to Davidson's (1963) essay, "Actions, Reasons and Causes".
The philosophy of art, like the philosophy of history, is one of those areas in which Collingwood's work is most often discussed even if his approach, which focuses on the production of art, has not always been met with favour in a context where discussion has tended to focus on reception rather than creation. In "What is the business of Collingwood's The Principles of Art?" James Camien McGuiggan appeals to the philosophical methodology underpinning Collingwood's aesthetics in order to show that Collingwood' approach, when duly understood, can provide a better account of borderline cases of art.
Collingwood's twin metaphilosophical works have a great deal to offer to the contemporary debate. The latter half of the twentieth century has witnessed an ontological backlash against the linguistic turn and the return of pre-critical forms of metaphysics. This backlash is evident in the titles of many recent monographs such as Devitt's (2010) "Putting metaphysics first," Heil's (2003), "From an ontological point of view" and Sider's (2011) "Writing the book of the world." The conceptions of metaphysics articulated in these works seek to circumvent Collingwood's logic of question and answer and his claim that since all questions rely on presuppositions, and all forms of knowing rely on a body of questions to which they seek answers, there can be no such thing as knowledge of pure being, which is precisely the kind of knowledge these nouveaux metaphysicians are seeking. Many of those who take exception to this renewed enthusiasm for pre-Humean metaphysics tend to seek the alliance of Wittgenstein and more recently of Carnapian linguistic frameworks to undermine the view that there are legitimate true or false answers that can be given to questions concerning the existence of frameworks. Collingwood does not tend to be mentioned in such debates. But his work would have much to offer. As a quasi-continental philosopher who writes in beautifully accessible English his work could provide the resources to articulate a much neglected post-Kantian perspective in the philosophy of mind and social science. This edited collection shows that although Collingwood's work is yet to have the kind of influence it deserves on mainstream analytic philosophy there are many scholars, both well established and new to the profession, who are taking a very close look at his work with a view to establishing how his distinctive metaphilosophical approach could be brought to bear on many of the first order philosophical issues that concern us today.

CARNAP, RUDOLF. 1950. 'Empiricism, Semantics and Ontology', in Revue Internationale de Philosophie 4 (2), pp. 20-40.
CHALMERS, D., MANLEY, D AND WASSERMAN R., eds. (2009), Metametaphysics (Oxford and New York: Oxford University Press).
COLLINGWOOD, R.G. (1933/2005), An Essay on Philosophical Method, new edition with introduction and additional material, ed. James Connelly and Giuseppina D'Oro (Oxford: OUP)
COLLINGWOOD, R.G. (1940/1998), An Essay on Metaphysics, new edition with introduction and additional material, ed. Rex Martin (Oxford: OUP).
COLLINGWOOD, R.G. (1946/1993), The Idea of History, new edition with introduction and additional material, ed. Jan van der Dussen (Oxford: OUP).
DAVIDSON, DONALD (1963), 'Actions, Reasons, and Causes', in Journal of Philosophy, Vol. 60, pp. 685–700.
D'ORO GIUSEPPINA and SANDIS, CONSTANTINE (2014), 'From Anticausalism to Causalism and Back' in D'Oro G. and Sandis C. Reasons and Causes: Causalism and Anti-causalism in the Philosophy of Action (Palgrave).
DANCY, JONATHAN (2004), Ethics Without Principles (Oxford University Press).
DEVITT, MICHAEL (2010), Putting Metaphysics First (Oxford University Press).
D'ORO G. and OVERGAARD, S. eds. (2016 forthcoming) The Cambridge Companion to Philosophical Methodology.
D'ORO GIUSEPPINA (2015), 'Unlikely Bedfellows? Collingwood, Carnap and the Internal/External Distinction', British Journal for the History of Philosophy, Vol 23 (4) pp. 802-817.
D'ORO, GIUSEPPINA (2016 forthcoming), 'Collingwood's idealist metaontology: between therapy and armchair science", in D'Oro G. and Overgaard, S (eds.) The Cambridge Companion to Philosophical Methodology.
HAUG, M.C. ed. (2014), The Armchair or the Laboratory? (New York: Routledge).
HEIL, JOHN (2013), From an Ontological Point of View (Oxford: Clarendon Press).
HUME, DAVID (1748), An Enquiry Concerning Human Understanding.
PRICE, HUW (2009), 'Metaphysics after Carnap: The Ghost Who Walks?' in Metametaphysics, eds. D. Chalmers, D. Manley and R. Wasserman. (Oxford and New York: Oxford University Press), pp. 320-47.
QUINE, W.V.O. (1951), 'Two Dogmas of Empiricism, in Philosophical Review Vol. 60, pp. 20–43.
SIDER, TED (2011), Writing the Book of the World (Oxford University Press).
WILLIAMSON, THIMOTHY (2010), The Philosophy of Philosophy (Wiley-Blackwell)