[In the following essay, Roberts examines the opposing theoretical positions of White and Arthur Marwick and defends Marwick's perspective of narrative history.]

Hayden White writes [in the essay ‘Response to Arthur Marwick,’]: ‘Historians have systematically built into their notion of their discipline hostility or at least a blindness to theory and the kind of issues that philosophers have raised about the kind of knowledge they have produced.’¹ His explanation for this blindness seems to be in terms of a set of personal failings shared by historians: their simplistic assertion that only empirical research really matters; their mistaken belief that understanding what historians do is a matter of practical knowledge available only to fellow professionals; their inability to theorize their own discourse and the suppression of the philosophical dimensions of their craft and discipline.

Such attitudes are indeed widespread among historians. But I want to suggest a much more profound reason for the anti-theoretical tradition in history: most historians feel no need for a self-conscious, separate and distinct theoretical analysis of their discipline because the dominant discourse of knowledge in history is coterminous with the common sense discourse of modern everyday life. As John Passmore has argued: ‘For the most part … there is nothing much to say about historical explanation; nothing that cannot be said about explanation in everyday life. … No wonder historians are often puzzled to know what philosophers are fussing about!’²

‘Narrative’ historians like myself, and most other historians,³ work mainly within the framework of a human action account of the past.⁴ History is viewed as a field of human action and action as the result of individual and collective reasoning in particular circumstances under the impact of a variety of social, political, economic, ideological and cultural influences (themselves contexts of action composed of and created by other human agents). The task of historians is to reconstruct the reasons for past actions. They do this by reference to surviving evidence of past human thinking, whose meaning they interpret in connection with decisions and action. This human action approach to the study of history emphasizes the freedom of individuals to act, the importance of reconstructing what happened from the actor's point of view, and the role of accident, miscalculation and unintended consequences in shaping historical outcomes.

Involved here, too, is what some writers call a philosophical anthropology of humanity: the view that human beings have certain properties, powers and predispositions. Arising out of these is the rational-purposive and self-reflective nature of human action and its intentional and intelligible character. The intelligibility of action and its expression in language—its reason-based, linguistic character—makes it possible to interpret and communicate the meaning of action.⁵

In narrative history the results of research are written up in the form of stories about connected sequences of thought and action. These narratives of action will include various descriptions relevant to the story, may include political and moral judgments about what happened, and, possibly, generalizations about past actions which ascribe to them meanings and patterns relevant to contemporary concerns. But the explanatory content of the story will be some reason-giving account of why past actors did what they did. The validity of this kind of account of the past rests on its correspondence with the direct and indirect evidence of the perceptions, motivations, goals, calculations and intentions which result in specific decisions and actions.⁶

The idea that people do things for a reason, that their individual and collective actions are the stuff of history and that it is possible to construct an evidence-based account of why past actors acted as they did is, for most of us, plain common sense. It is a set of assumptions that harmonizes with our intuition of what the world is like, corresponds with our own experience as historical actors, and forms the basis of our interaction with other human beings.

Narrative historians have been successfully practising their craft for generations, producing, within the terms of reason-giving human action accounts, an accumulation of grounded knowledge. Internal coherence and practical success, however, is no guarantee of validity and the neglect of deeper philosophical scrutiny leaves the field open to alternative (usually obfuscatory and confusing) conceptions of the nature of history. The assumptions underlying narrative history require elaboration and argument beyond the common sense propositions outlined above.

Contrary to the impression given by some philosophers and social scientists, a large and sophisticated set of theoretical resources are available to narrative historians wishing to defend their craft. These resources include:

1. The debates on individualism in the social sciences. Of particular interest to historians are efforts to validate an ontological individualism. The claim that what happens in the human world is the result of action, that the agency of action is the individual, and that while action may be constrained and influenced in various ways individuals are free to choose is of critical importance to narrative historians.⁷

2. The Collingwood school in the philosophy of history which seeks in various ways to defend and elaborate Collingwood's argument in The Idea of History that history is the history of human thought and action and that the accurate recovery of thought-action in the present is possible.⁸

3. The discussion among philosophers of the concept of action. Narrative histories rest on reason-giving explanations. For these to be valid, action must be volitional and intentional and intentions connected to reasons.⁹

4. The efforts of sociologists, particularly those working within the hermeneutic tradition, to produce a theory of action which links rational and individual action to its social dimension and context and provides an account of continuity and change in diverse human societies.¹⁰

5. The phenomenological account of the identity of form and content in narrative reconstructions of past action. Narrative historians tell stories about the past because human beings are narrative creatures and action is narrative in character. The narrative character of human existence arises out of the character of human consciousness and language.¹¹ The stories told by historians (if they are good enough) correspond to those that have been played out and lived in the past. The idea that there is a correspondence between narrative and life has been brilliantly explored by some of White's philosopher colleagues who provide a coherent alternative to his view that historians produce the narratives that they do out of linguistic necessity. Their critique of White is far more effective than Marwick's polemics and denunciations—and far more useful to historians seeking a philosophical underpinning of what they do by instinct, tradition and common sense.¹²

Where does Marwick stand on the human action account of history? Although he lays claim to a defence of the ‘historical’ approach, at times his views seem more in tune with some kind of quasi-social scientific approach. For example, in his article he comments:

Postmodernists … show no familiarity with the modes of explanation historians actually use, which certainly do not concentrate exclusively on the actions of individuals, but involve a varying balance, depending on topic and focus, between short-term human agency, contingency and convergence, and longer-term structural, ideological and institutional movements and constraints.¹³

What Marwick seems to be endorsing here is a weak version of the ‘structuration’ or ‘social scientific realist’ approach popular among some sociologists.¹⁴ The basic concept of history informing this view is that what happens in the human world is the result of a combination of agency (human action/will) and autonomous conditioning, constraining and enabling forces and structures. The truth and usefulness of this concept of history rests ultimately on a set of ontological and epistemological claims—that the human world is like this, that it consists of two different kinds of objects (individual and social), and can be known and shown to be so. In other words, it requires validation by reference to a set of metaphysical assumptions which run counter to those of narrative history. Assuming that he could overcome his disdain for anything smacking of the metaphysical, would Marwick be prepared to defend an agency-structure account of history? The answer, on balance of the evidence, is no. In Marwick's The Nature of History one can find similar statements to that quoted above, but these are counter-balanced by a clear appreciation of the past as a site of human thoughts, activities and products, by his view that ‘in a very profound sense, what happens is the consequence of the actions of individuals,’ and his final reminder in the book that structural and ideological ‘forces’ ‘are in fact created by the activities of multitudes of human beings.’¹⁵

Reclaiming Marwick for History is not, of course, the end of the discussion with the structurationists,¹⁶ or anyone else for that matter, but it is necessary in order to clarify what the argument between historians and their opponents is really all about. In the article, and in his other writings on the nature of historical knowledge, Marwick, as I understand him, seeks to defend a traditional approach to the study of the past. He does this, firstly, by explaining and defending the research practices of historians and, secondly, by attempting to puncture the theoretical pretensions of White & co. Almost all of what Marwick has to say I, as a historian, agree and identify with. But missing from Marwick's account is the crucial dimension of a presentation and exploration of the metaphysical basis of historical practise. Like anyone else, historians have reasons for the kind of research they conduct and the type of statements and truth-claims they make. The historical approach is not in opposition to the metaphysical approach: the historical approach is as metaphysical as that of any other discipline. The fact that historians prefer to practise their metaphysics rather than talk about them does not mean they cannot and should not be discussed and defended.

Notes

1. H. White, ‘Response to Arthur Marwick,’ Journal of Contemporary History 30, 2 (April 1995), 244.

2. J. Passmore, ‘Explanation in Everyday Life, in Science, and in History,’ History and Theory, 2, 2 (1962), 122, 123.

3. The purest narrative history is to be found in the works of traditional diplomatic and political historians, but narrative accounts are to be found in every field of the discipline. Indeed, even in the writings of historians explicitly hostile to the traditionalist view espoused here one often finds a crucial, defining narrative component (e.g. Eric Hobsbawm's The Age of Extremes, London 1994). This is because practically all historians share a common interest in human actors and their impact on the world and admit the importance of the connection between belief and action.

4. The most sustained defence of a human action approach to history may be found in the works of Geoffrey Elton: The Practice of History (Sydney 1967), Political History: Principles and Practice (London 1970), Which Road the Past? Two Views of History (with R. W. Fogel Yale 1983), and Return to Essentials: Some Reflections on the Present State of Historical Study (Cambridge 1991). A less well-known, but equally important, and more philosophically rigorous, advocate of the human action approach (albeit on methodological rather than substantive grounds) is the British International Relations specialist Charles Reynolds: Theory and Explanation in International Politics (London 1973), Modes of Imperialism (London 1981), The Politics of War (London 1989) and The World of States (London 1992). Noteworthy here, too, is a detectable drift in the social sciences towards the human action approach long defended by historians. See, e.g., the contributions by Vayda, McCay, Eghenter and Searle in Philosophy of the Social Sciences, 21, 3 (September 1991).

5. To sustain the formulation about language in this paragraph would require a further argument that words have determinate meaning in relation to things, actions and ideas and that this meaning can be exchanged by speakers. The postmodernist—deconstructionist alternative—that words have meaning only in relation to other words and meaning is, therefore, ambiguous and ungraspable—is criticized by, among others, the Marxist theoretician A. Callinicos, Against Modernism (London 1989).

6. The notion of causality deployed here is that of reasons as causes. There are, of course, alternative conceptions of causality in history, most of them based on the notion that causation in human affairs is, in some sense, akin to that in nature. For the reasons as causes argument see, e.g., W. H. Dray, Philosophy of History (Prentice-Hall 1964); D. M. Taylor, Explanation and Meaning (Cambridge 1970); and A. R. Louch, Explanation and Human Action (Oxford 1966). To my mind the most useful discussion of causes in history remains Elton, Political History, chap. 4.

7. For a thorough and highly enlightening review of the relevant debates see R. Bhargava, Individualism in Social Science (Oxford 1992). While the author reveals the power of ontological, as opposed to other forms of individualism, he himself ultimately rejects it in favour of what he calls contextualism, i.e. a weak variation of the claim that there are some things in the human world irreducible to individuals.

8. R. G. Collingwood, The Idea of History (Oxford 1961). On the Collingwood school see F. R. Ankersmit, ‘The Dilemma of Contemporary Anglo-Saxon Philosophy of History,’ History and Theory, 25, 1986, and other contributors in the same issue. Important contributions/commentaries in this tradition include: R. Martin, Historical Explanation (Cornell 1977); W. H. Dray, Laws and Explanation in History (London 1957); G. H. von Wright, Explanation and Understanding (Cornell 1971); L. J. Goldstein, ‘Collingwood's Theory of Historical Knowing,’ History and Theory, IX, 1, 1970, Historical Knowing (Austin 1976) and ‘Collingwood on the Constitution of the Historical Past’ in M. Krausz (ed.), Critical Essays on the Philosophy of R. G. Collingwood (Oxford 1972); P. Q. Hirst, ‘Collingwood, Relativism and the Purposes of History’ in his Marxism and Historical Writing (London 1985); L. O. Mink, ‘Collingwood's Dialectic of History,’ History and Theory, 7, 1, 1968; A. Donagan, ‘Historical Explanation: The Popper-Hempel Theory Reconsidered,’ History and Theory, 4, 1, 1964 and W. B. Gallie, ‘The Historical Understanding,’ History and Theory, 3, 2, 1964.

9. For a review of these debates see C. J. Moya, The Philosophy of Action (London 1990). Moya's conclusions are commensurate with the commonsense, practical concept of intentional action utilized by historians. The debate on free will is also relevant here. For a summary of that discussion see D. J. O'Connor, Free Will (London 1971) and, for a relevant pro-free will case, J. R. Lucas, The Freedom of the Will (Oxford 1970). I am aware that here are raised issues of unintentional action, of the unconscious, and the role of emotion as well as reason in human affairs.

10. Of particular interest is Jürgen Habermas's The Theory of Communicative Action, 2 vols (London 1989). Habermas's views are summarized by J. Bohman, New Philosophy of Social Science (London 1991), who also reviews other sociological theories of action. An alternative account of action is that proposed by various rational-choice theorists, whose starting-point are the models of economic behaviour developed by neo-classical economic theory. Their views are ably summarized and criticized by B. Hindess, Choice, Rationality and Social Theory (London 1988).

11. In relation to this point see Paul Hirst's summary of the views of Julian Jaynes in ‘The Evolution of Consciousness: Identity and Personality in Historical Perspective,’ Economy and Society, 23, 1 (February 1994).

12. See David Carr's Time, Narrative, and History (Indiana 1986) and ‘Narrative and the Real World: An Argument for Continuity,’ History and Theory, vol. 15, 1986, and Frederick Olafson The Dialectic of Action (Chicago 1979) and ‘Narrative History and the Concept of Action,’ History and Theory, 9, 3, 1970. Also: W. H. Dray, ‘Narrative and Historical Realism’ in his On History and Philosophers of History (New York 1989) and A. P. Norman, ‘Telling It Like It Was: Historical Narratives on Their Own Terms,’ History and Theory, vol. 30 1991. An important, recent contributor to this argument is M. C. Lemon, The Discipline of History and the History of Thought (London 1995). Hayden White is obviously aware of this alternative view of historical discourse—that it is driven by the nature of human action, experience, and thought—but he never seems to confront and criticize it directly. This is true, for example, of the essays published in The Content of the Form (Baltimore 1987), including the one on the work of Paul Ricoeur, who espouses a version of the narrative is life argument.

13. A. Marwick, ‘Two Approaches to Historical Study: The Metaphysical (Including “Postmodernism”) and the Historical,’ Journal of Contemporary History, 30, 1 (January 1995), 16.

14. See A. Giddens, Central Problems in Social Theory (London 1979) espec. chap. 2, ‘Agency, Structure’; R. Bhaskar, The Possibility of Naturalism (Brighton 1979); A. Sayer, Method in Social Science: A Realist Approach (London 1992); L. Doyal and R. Harris, Empiricism, Explanation and Rationality (London 1986); and A. Callinicos, Making History (London 1987). Giddens's particular approach is defended and elaborated by the historian W. H. Sewell: ‘A Theory of Structure: Duality, Agency, and Transformation,’ American Journal of Sociology, 98, 1 (July 1992) and ‘How Classes are Made: Critical Reflections on E. P. Thompson's Theory of Working-Class Formation’ in H. J. Kaye and K. McClelland, E. P. Thompson: Critical Perspectives (London 1990).

15. A. Marwick, The Nature of History (3rd edn, London 1989). The quotes are from pp. 247 and 381.

16. See my ‘Agency and Structure in the New History of the Stalinist Terror: An Individualist View’ (forthcoming)—a critique of structuration in the context of a historical case study.