What is Liberalism

Copyright notice: This paper is due to appear in Political Theory. This is a pre-print, and may be subject to minor changes. What is Liberalism? Duncan Bell (University of Cambridge) Like the history of anything else, history of philosophy is written by the victors. Victors get to choose their ancestors, in the sense that they decide which among their all too various ancestors to mention, write biographies of, and commend to their descendants. (Richard Rorty)1 Before we can begin to analyze any specific form of liberalism we must surely state as clearly as possible what the word means. For in the course of so many years of ideological conflict it seems to have lost its identity completely. Overuse and overextension have rendered it so amorphous that it can now serve as an allpurpose word, whether of abuse or praise. (Judith Shklar)2 Introduction Liberalism is a spectre that haunts Western political thought and practice. For some it is a site of the modern, an object of desire, even the telos of history. For others it represents an unfolding nightmare, signifying either the vicious logic of capitalism or a squalid descent into moral relativism. For others still, perhaps the majority, it is a mark of ambivalence, the ideological prerequisite for living a reasonably comfortable life in affluent democratic states – the least worst option. But what is liberalism? Across and within scholarly discourses it is construed in manifold and contradictory ways: as an embattled vanguard project and constitutive of modernity itself, a finegrained normative political philosophy and a hegemonic mode of governmentality, the justificatory ideology of unrestrained capitalism and the richest ideological resource for its limitation. Self-declared liberals have supported extensive welfare states and their abolition; the imperial civilizing mission and its passionate denunciation; the necessity of social justice and its outright rejection; the perpetuation of the sovereign state and its transcendence; massive global redistribution of wealth and the radical inequalities of the existing order. Shklar’s complaint that Acknowledgments: I’d like to thank the following for their incisive comments on earlier drafts of the paper: Robert Adcock, Chris Brooke, David Craig, Sarah Fine, Michael Freeden, Mark Goldie, John Gunnell, Joel Isaac, Ben Jackson, Ira Katznelson, Duncan Kelly, Daniel Klein, Chandran Kukuthas, Patchen Markell, Jeanne Morefield, Tim Stanton, Casper Sylvest, Colin Tyler, and Brian Young. I have also benefitted from presenting the paper (or earlier iterations of it) at seminars at Auckland, Cambridge, LSE, Oxford, Sussex, Sydney, Victoria, and York. All the usual disclaimers apply. Rorty, “The Historiography of Philosophy” in Philosophy in History, ed. Rorty, Jerome Schneewind, Quentin Skinner (Cambridge: CUP, 1984), 70. 2 Shklar, “The Liberalism of Fear” (1989) in Shklar, Political Thought and Political Thinkers (Chicago: UCP, 1998), 3. 1 1 it is an “all-purpose word” is thus unsurprising, for liberalism has become the metacategory of Western political discourse. There are several responses to “overextension”. One is simply to ignore it, deploying the term as if its meaning was self-evident. Ubiquitous across the humanities and social sciences, this unreflective impulse generates much confusion. Another is to engage in “boundary work” – to demarcate and police the discourse.3 Some influential attempts to do so have figured liberalism as a capacious tradition of traditions, with Guido De Ruggiero and Friedrich Hayek, for example, bifurcating it into British and Continental forms. The most common variation on this theme is to distinguish “classical” and “social” liberalisms.4 Another popular response is to narrate liberal history as a story of rise or decline, triumph or tragedy. A familiar rendition bemoans the lost purity of the original. Thus Leo Strauss mourned the transition from virtuous “ancient” liberalism (reaching its apogee in Athens) to debased forms of “modern” liberalism (commencing with Machiavelli), while Sheldon Wolin averred that twentieth-century liberalism had disastrously forgotten its early sceptical enunciation.5 Some neoconservatives have claimed the mantle, seeking, with Irving Kristol, “a return to the original sources of liberal vision and liberal energy so as to correct the warped version”.6 Declension has also been a recurrent libertarian complaint. When he came to pen his defence of “classical” liberalism in 1927, Ludwig von Mises grumbled that from Mill onwards the ideology had degenerated into socialism, a warning that Herbert Spencer had flagged half a century earlier.7 But the development of liberalism can also be cast as progressive. Both L.T Hobhouse and John Dewey, for example, celebrated the transfiguration of liberalism from an ideology of laissez faire to the use of systematic government intervention to reduce harmful disadvantages.8 The argument continues today with many libertarians viewing “social” liberalism as a deplorable form of socialism and many social liberals rejecting the liberal credentials of libertarianism. All sides claim to be heirs of the one true liberalism. A related policing strategy is to concede the intellectual diversity of liberalism while extracting its constitutive element(s) – its ineliminable core. This too is contested terrain. Adopting the most common line, Shklar sought to create a “modest amount of order” by characterising liberalism as a “political doctrine” with “only one overriding aim: to secure the political conditions that are necessary for the exercise of personal freedom”.9 Jeremy Waldron rightly observes that to identify a commitment to freedom as the foundation of liberalism “is to say something too vague and abstract to be helpful”, and he proposes instead that it is best defined by a “requirement that all aspects of the social should either be made acceptable or be capable of being made acceptable to every last individual”.10 Ronald Dworkin asserts that “a certain conception of equality…is the On the practice, see Thomas Gieryn, “Boundary-work and the Demarcation of Science from NonScience,” American Sociological Review, 48 (1983): 785-95. 4 De Ruggiero ([1927]) The History of European Liberalism, trans. R.G. Collingwood (Boston: Beacon, 1959); Hayek, “Liberalism” [1973] New Studies in Philosophy, Politics, Economics and the History of Ideas (London: Routledge, 1978), 113. Alan Ryan complicates matters by distinguishing between modern, classical, social and libertarian variants: The Making of Modern Liberalism (Princeton: PUP, 2013), 23-8. 5 Strauss, Liberalism Ancient and Modern (Chicago: UCP, 1968); Wolin, Politics and Vision (Princeton: PUP, 2004), 263. 6 Kristol, Reflections of a Neoconservative (NYC: Basic, 1983), 75. 7 Mises, Liberalism (Indianapolis: Liberty Fund, 2005), 153-4; Spencer, The Man Versus the State (Indianapolis: Liberty Fund, 1969 [1884]). 8 Hobhouse, Liberalism, ed. J. Meadowcroft (Cambridge: CUP, 1994[1911]); Dewey, Liberalism and Social Action (NYC: Putnam, 1935), 21. 9 Shklar, “Liberalism,” 3. On liberty as “normatively basic” see Gerald Gaus and Shane Courtland, “Liberalism,” The Stanford Encyclopaedia of Philosophy, ed. Edward Zalta. 10 Waldron, “Theoretical Foundations of Liberalism,” Philosophical Quarterly, 37 (1987), 131, 127, 140. 3 2 nerve of liberalism.”11 Others insist on a cluster of commitments. The historian Gary Gerstle, for example, suggests that liberals have always endorsed three “foundational principles”, rationality, emancipation, and progress, while John Dunn once lamented the “dismaying number of categories” thought central to liberal ideology, including political rationalism, hostility to autocracy, cultural distaste for conservatism and tradition, tolerance, and individualism. 12 Even its supposed core has proven rather elusive. In what follows I neither attempt to adjudicate between these competing interpretations nor present a new substantive liberal theory. Instead, I seek to reframe the way in which the liberal tradition is understood. I open with a critique of some existing interpretive protocols used to delimit political traditions, before introducing (in Section II) a new way of conceptualizing liberalism, suggesting that it can be seen as the sum of the arguments that have been classified as liberal, and recognised as such by other self-proclaimed liberals, across time and space. In the second half of the essay I analyse the emergence and subsequent transformation of the category of liberalism in Anglo-American political thought between 1850 and 1950. This serves as an illustrative case study of some of the methodological arguments I outline in the first two sections. While Section III traces the evolution of the language of liberalism in nineteenth century Britain, Section IV explores how the scope of the liberal tradition was massively expanded during the middle decades of the century, chiefly in the United States, such that it came to be seen by many as the constitutive ideology of the West. Above all, I contend that this capacious understanding of liberalism was produced by a conjunction of the ideological wars fought against “totalitarianism” and assorted developments in the social sciences. Today we both inherit and inhabit it. Constructing Liberalism: Scholarly Purposes and Interpretive Protocols There are at least three types of answer that can be given to the question in the title, each of which serves a different scholarly purpose. Prescriptive responses specify norms of correct or best usage. They delineate a particular conception of liberalism, branding it as more authentic – more truly liberal – than other claimants to the name. Such accounts vary in the core features recognized as constitutive, the interpretive methodologies utilised to identify them, and the normative stance assumed towards them. This is the most familiar type of answer, not least in contemporary political theory. Comprehensive responses attempt to chart the plethora of liberal languages. Rather than prescribing a favoured conception they seek to identify the actual range of usage, mapping the variegated topography of liberal ideology. These accounts vary in the interpretive methodologies employed and the temporal and spatial scope of enquiry. Explanatory responses account for the development of liberalism(s), whether understood in prescriptive or comprehensive terms. They too vary in methodology and scope. Although each kind of response is legitimate in certain circumstances, problems arise when they are misapplied or conflated. In particular, prescriptive accounts are very poor guides to understanding the internal complexity and historical development of ideologies. Furthermore, scholars adopt different methodological strategies – interpretive protocols – to answer the question. To argue about a political tradition – to compare and contrast it; to chart its decline, crisis, or ascension; to pinpoint its flaws or celebrate its strengths – it is first necessary to Dworkin, “Liberalism,” in A Matter of Principle (Oxford: Clarendon, 1985), 183. Gerstle, “The Protean Character of American Liberalism,” American Historical Review, 99/4 (1994), 1046; Dunn, Western Political Theory in the Face of the Future (Cambridge: CUP, 1991[1979]), 33. 11 12 3 construct it as an object of analysis. Political theorists typically employ one of two main protocols: stipulative and canonical. Contextualism offers an alternative.13 Stipulative accounts identify necessary (though rarely sufficient) conditions for a position to count as a legitimate exemplar of a tradition. “Liberalism” is typically constructed from interpretations of the meaning and inter-relation of core concepts, such as liberty, authority, autonomy, and equality. Such accounts employ definitional fiat to demarcate the legitimate boundaries of liberalism, insisting that only those committed to a particular set of views count as properly liberal. We have already encountered the contrasting formulations offered by Dworkin, Gerstle, Shklar, and Waldron. History is sometimes invoked in such accounts, but it is usually what Rawls aptly terms the “philosopher’s schematic version of speculative history”, and while these arguments often cite historical figures – above all Locke, Kant, Mill, and now Rawls himself – their core normative arguments can be justified independently of any past expression.14 Traditions are usually constructed around a canon of renowned thinkers, which serves simultaneously as a reservoir of arguments, an index of historical continuity, and a powerful source of intellectual authority. Canonical approaches thus distil “liberal” theoretical structures from exemplary writings. The most frequent targets for this protocol are (again) Locke, Kant, Mill, and Rawls, though a host of other figures are sometimes marshalled to fit the occasion. Leo Strauss and his epigones have divined sweeping interpretations of liberal modernity from a handful of “great books.” Pierre Manent, for instance, charts the unfolding of liberalism through a procession of figures stretching back to Machiavelli and Hobbes.15 Far from being an exclusive Straussian strategy, however, this is arguably the most common protocol for constructing liberalism. To take one prominent recent debate, canonical formulations are central to arguments about the relationship between liberalism and empire. While Uday Singh Mehta grounds his influential argument that liberalism has an “urge” to empire on readings of Locke and Mill, most rejoinders have likewise focused on canonical figures.16 Both of these methodological strategies are valuable, even essential, for achieving particular scholarly aims. Stipulative protocols can be fruitfully employed in the construction of normative political philosophies and the elaboration of ideal types for conducting social analysis. Canonical scholarship, meanwhile, can generate insightful readings of individual thinkers. But neither are capable of underwriting plausible comprehensive or explanatory accounts as they cannot shed much light on the universe of liberal languages, the plethora of competing and often contradictory claims that travel under its name. Articulated in the register of philosophical abstraction, the stipulative genre is estranged from the vicissitudes of history and political Less common in political theory, expressive protocols are widely utilised across the humanities and social sciences. They distil the meaning of liberalism through a form of reverse engineering, working backwards from observations on (aspects of) a society to the ideas purportedly underlying it. First, certain entities – for example, public policies – are classified as “liberal,” a classification usually based on the selfidentification of the relevant agents or the alleged correspondence between the entity and a putative external (“liberal”) standard. Second, the entities are taken to embody or express underlying ideas or values which are then characterized as liberal. Thus: State A is classified as liberal; “liberal state” A enacts policy B. Policy B is therefore “liberal”. B embodies or expresses liberal value or idea C. An expressive protocol is arguably employed in Dworkin’s “theory of what liberalism is” (Dworkin, “Liberalism”). This protocol has various problems, not least debilitating circularity. 14 Rawls, Lectures on the History of Political Philosophy, ed. Samuel Freeman (Cambridge: HUP, 2007), 11. 15 Manent, An Intellectual History of Liberalism, trans. R. Balinski (Princeton: PUP, 1996). 16 Mehta, Liberalism and Empire (Chicago: UCP, 1996). For an important response: Jennifer Pitts, A Turn to Empire (Princeton: PUP, 2005). Cf. Duncan Bell, “Empire and Imperialism” in The Cambridge History of Nineteenth Century Political Thought, ed. Gregory Claeys and Gareth Stedman Jones (Cambridge: CUP, 2012), 864-91. 13 4 practice. It is caught on the horns of a dilemma. Unless the stipulated commitments are conceptualised at a very high level of generality – e.g. that liberalism prioritises individual freedom, or that liberals are committed to toleration, liberty and constitutional government – they will invariably fail to encompass the deep divisions between professed variants of liberalism, yet when pitched at that level they provide little guidance for pursuing the detailed reconstruction necessary for satisfactory description or explanation. Waldron’s argument illustrates this mismatch. Maintaining that only those adopting his contracturalist view of justification count as properly liberal, he anoints Locke, Rousseau and Kant as genuine liberals, while suggesting that John Stuart Mill and numerous other nineteenth-century figures (especially utilitarians) stand in an “ambiguous relation” to the tradition. On this account, then, liberalism simultaneously pre-exists its own self-conscious formulation and was misunderstood by many of those who played a fundamental role in its propagation. At least he admits that “many liberals may not recognise” the picture he paints.17 The problem with canonical protocols is that they can rarely support the generalisations they are invoked to underpin. As Mehta’s argument shows, work in this vein often seems to assume that the ideas of canonical figures can stand in for, or be seen as sufficiently representative of, the tradition as a whole. Despite claims to the contrary, this provides a defective foundation on which to build an analysis of a tradition. Given the internal diversity of liberalism, its national and regional variation, and its polyvocal evolution, it is exceptionally difficult to ground felicitous non-trivial generalisations on the work of a handful of authors. A further problem is that this protocol often takes as given the very thing which should be investigated – the construction of the canon. The idea of a canon of great thinkers standing at the heart of a pre-constituted tradition is, in part, an artefact of the professional development of academic political theory during the twentieth century.18 It is the product of a particular moment in time, shaped by largely forgotten value-commitments and selection criteria, and arguments centred on claims distilled from the canon are thus conducted within a discursive echo-chamber. Indeed studying the processes through which the canon crystallised can reveal as much (or more) about the dynamics of political thinking as the forensic analysis of purportedly exemplary texts. Contextualist approaches to the history of political thinking need little introduction.19 The bulk of such work has focused on illuminating the patterns of early modern political thinking, and there are no general contextual histories of liberalism – indeed its methodological precepts render such a project quixotic. Contextualists have nevertheless made an important contribution to the analysis of liberalism by challenging the assumption that it can be traced to the seventeenth century. Versions of this argument have been tendered by John Dunn, Mark Goldie, J.G.A. Pocock, Quentin Skinner, and James Tully. Pocock, for example, maintains that “liberalism” was “not used in the eighteenth century, where the adjective ‘liberal’ did not bear its modern meaning, and though elements were present which would in due course be assembled by means of this formula, there was no system of doctrine corresponding to its later use”.20 Thus no significant inferences about liberalism can be drawn from the earlier period. In particular, this strand of scholarship has repeatedly questioned Locke’s elevated status as a (or the) foundational Waldron, “Theoretical Foundations,” 128, 143-44. John Gunnell, The Descent of Political Theory (Chicago: UCP, 1993). On the politics of canon formation in literature, see John Guillory, Cultural Capital: The Problem of Literary Canon Formation (Chicago: University of Chicago Press, 1993). 19 For a seminal statement, see Quentin Skinner, Visions of Politics, I (Cambridge: CUP, 2002). 20 Pocock, The Machiavellian Moment (Princeton: PUP, 2003), 579. 17 18 5 liberal.21 It is important to recognise that this is not principally a semantic argument about the absence of the word “liberalism”, but rather a claim about the range of concepts and arguments available to historical actors.22 It is about extant thought-worlds not recoverable terminology. Yet while this body of scholarship has questioned conventional accounts of liberal history, it has rarely probed how and why that very convention emerged. Michael Freeden has developed the most systematic contextualist account of Anglo-American liberalism. It is, he argues, ...that semantic field in which the political understandings of people who regard themselves as liberals, or who others regard as liberals, may be investigated. It is a plastic, changing thing, shaped and reshaped by the thought-practices of individuals and groups; and though it needs to have a roughly identifiable pattern for us to call it consistently by the same name, “liberalism,” it also presents myriad variations that reflect the questions posed, and the positions adopted, by various liberals.23 However, even Freeden tends to blur prescriptive, comprehensive and explanatory arguments. Identifying Millian liberalism as the most genuine manifestation of the ideology, he finds several alternative strands wanting, including contemporary libertarianism and “American philosophical liberalism” (social liberalism following Rawls). With its focus on state neutrality, neo-Kantian conceptions of autonomy, and the possibility of specifying fixed principles of justice, as well as its abstraction from practical political activity, the latter represents a “decisive departure” from prevailing liberal thought, while the former lacks “many of the attributes which bestow on the liberal profile its distinctive contours,” and it is thus disqualified as “a serious contender for the current mantle of liberalism”.24 On this account, while liberalism contains no ineliminable trans-historical essence a specific thread nevertheless expresses its most mature “established” form. Freeden’s explicit anti-essentialism is thus qualified by prescriptive boundary-working methodological commitments. His general approach nevertheless points to a fruitful interpretive strategy. A comprehensive contextualist analysis of liberalism should provide a framework for grasping the diverse ways in which liberal languages emerge, evolve, and come into conflict with one another, rather than trying to distil an ahistorical set of liberal commitments from conceptual or canonical investigation. A Summative Conception Thomas Nagel is surely right to proclaim that “[i]t is a significant fact about our age that most political argument in the Western world now goes on between different branches of [the liberal] tradition”.25 This ideological victory is acknowledged by both self-proclaimed liberals and their critics. At the turn of the new millennium Perry Anderson protested that “[f]or the first time Mark Goldie, “Introduction” in The Reception of Locke’s Politics, ed. Goldie (London: Pickering, 1999), I, xvii-lxxiii. For a recent powerful argument, see Tim Stanton, “John Locke and the Fable of Liberalism,” Historical Journal (forth.). 22 Ryan, Modern Liberalism, 9, reads it as a straightforward semantic claim. 23 Freeden, Liberal Languages (Princeton: PUP, 2005), 20. See also Freeden and Marc Stears, “Liberalism” in The Oxford Handbook of Political Ideologies, ed. Freeden, Stears and Sargent (Oxford: OUP, 2013), 329-48. 24 Freeden, Ideologies and Political Theory (Oxford: OUP, 1996), 227ff, 276, 278. 25 Nagel, “Rawls and Liberalism” in The Cambridge Companion to Rawls, ed. Samuel Freeman (Cambridge: CUP, 2003), 62. 21 6 since the Reformation there are no longer any significant oppositions – that is, systematic rival outlooks – within the thought-world of the West: and scarcely any on a world scale”. Writing more in sorrow than celebration, Raymond Geuss concurs: “we know of no other approach to human society that is at the same time as theoretically rich and comprehensive as liberalism and also even as remotely acceptable to wide sections of the population in Western societies”.26 Most inhabitants of the West are now conscripts of liberalism: the scope of the tradition has expanded to encompass the vast majority of political positions regarded as legitimate.27 Today there is little that stands outside the discursive embrace of liberalism in mainstream Anglo-American political debate (and perhaps especially in academic political theory), and most who identify themselves as socialists, conservatives, social democrats, republicans, greens, feminists, and anarchists, have been ideologically incorporated, whether they like it or not. Useful as they are for other tasks, stipulative and canonical protocols offer little help in interpreting this phenomenon. We thus need a comprehensive account that can accommodate the plurality of actually existing liberalisms, past and present, without smuggling in boundary-working prescriptive commitments. A plausible explanation, meanwhile, must unpack the dynamics of ideological conscription. This section introduces a comprehensive heuristic, while the remainder of the essay begins the task of explaining how the meaning of Anglo-American liberalism was transformed between 1850 and 1950. I propose the following definition (for comprehensive purposes): the liberal tradition is constituted by the sum of the arguments that have been classified as liberal, and recognised as such by other self-proclaimed liberals, across time and space. Let us call this the summative conception. Adopting it offers several benefits: it can help make sense of the discursive “overextension” and elastic usage of the term, while avoiding unhelpful claims about pure essence or authentic form. Moreover, it forces us to examine traditions as evolving and contested historical phenomena, conjured into existence by the work of many hands, shaped by scholarly knowledge-production and pedagogical regimes, and often inaugurated and remade with specific politico-intellectual purposes in mind. It allows us to grasp, that is, the intricate dialectic of intentional human action and unintended consequences that structure any rich political tradition.28 Freeden, as we have seen, characterises liberalism as “that semantic field in which the political understandings of people who regard themselves as liberals, or who others regard as liberals, may be investigated.” While I agree with most of this, it is necessary to qualify the claim about those “who others regard as liberals”.29 The problem here is that the term is commonly used to tar opponents or to create linkages between liberalism and political positions that liberals invariably reject. Witness the current fashion for American ultra-conservatives to connect liberalism to both fascism and Marxism.30 If we adopt an unqualified summative position – defining liberalism as the totality of positions termed liberal – then the tradition would now traverse the spectrum from fascism to communism, which is an implausibly expansive view. Hence the epistemic limit: only those positions affirmed at some point in time by groups of self-proclaimed liberals should be included. This allows us to map the universe of liberalism(s), though it raises another question: how widely held must a particular interpretation be for inclusion? Can any usage (by a selfproclaimed liberal) expand the boundaries of liberalism? There is no simple answer to this Anderson, “Renewals,” New Left Review, 1 (2000), 13; Geuss, “Liberalism and its Discontents,” Political Theory, 30 (2002), 320. 27 For a parallel usage to which I am indebted, see David Scott, Conscripts of Modernity (Durham: Duke UP, 2004). 28 Note that a comprehensive account is not suitable for constructing a coherent normative political theory. 29 Freeden, Liberal Languages, 20. 30 E.g. Jonah Goldberg, Liberal Fascism (London: Doubleday, 2007). 26 7 threshold question – scholars will adopt different inclusion criteria depending on their purposes and methodological inclinations. My own view is that to stake a claim for inclusion there must be sustained usage by numerous prominent ideological entrepreneurs over at least two generations. Otherwise, the bar for inclusion is set too low. That H.G. Wells declared himself a “liberal fascist” is nowhere near enough to warrant incorporating fascism into the liberal tradition, for barely anyone else followed him along this idiosyncratic path.31 But contra Freeden and others, “libertarianism” clearly meets the entry criteria. The temporal point is also important: I am not suggesting that only arguments labelled (and recognised) as liberal at Time T1 count as liberal. An argument is not expelled from the liberal tradition because it is later ascribed a different label or because liberals now happen to reject it. The tradition is constituted by the accumulation of arguments over time. Explicit justifications of imperialism, arguments seeking to limit suffrage on grounds of gender and racial difference, eugenicist attempts to “perfect” the species: all form part of the liberal tradition.32 As do rejections of these positions. Rather than attempting to sanitise or inoculate liberalism by ignoring aspects no longer considered palatable, or, more subtly, relegating those aspects to superseded historical circumstances while simultaneously extracting a pristine trans-historical core, we should recognise that liberalism has become a hyper-inflated, multifaceted, body of thought – a deep reservoir of ideological contradictions. In thinking about traditions it is productive to distinguish between the identities of agents and the arguments they invoke – between being an X (liberal, socialist, fascist) and employing forms of argument that are best characterised as X. The former is a claim about self-fashioning and the construction of personae, the latter about doctrine. Although this essay has focused on academic debates, the argument also applies to practical politics. It may well be part of the selfunderstanding of an American Tea Party devotee that they are fundamentally opposed to liberalism, but this identity-claim does not entail that they reject arguments central to the liberal tradition (as construed by the summative conception). In other words, despite espousing virulent anti-liberalism they are nevertheless committed to paradigmatic liberal positions insofar as they defend (say) neo-classical economics, libertarian social policy, and the superiority of “liberal democratic” institutions. Within political theory, the same can be said for many self-proclaimed critics of liberalism, whether post-structural, critical-theoretical, republican, communitiarian, or conservative. Another consequence of adopting the summative conception is that it dissolves a familiar but misleading picture of traditions, which are still often conceived of as self-contained bodies of thought with relatively clear and stable boundaries.33 On this view, the interstitial spaces between established traditions are populated by hybrids – liberal-socialists, liberal-conservatives, Christian-realists. However, this fails to grasp the ideological miasma of modern politics, in which most individuals simultaneously adopt positions that are claimed by assorted traditions. The most hardened Tory or Republican, contemptuous of moderate “liberal-conservatives,” is likely to propound ideas that have long been affirmed by mainstream liberals. When looking at an agent who has been classified in two or more ways – say as a liberal and a conservative – this P. Coupland, “H.G. Wells’ ‘Liberal Fascism,’” Journal of Contemporary History, 35 (2000), 541-58. Goldberg uses this example to reach the opposite conclusion. 32 One objection to this argument is that some liberal ideas/values/commitments (e.g. the normative priority of liberty) are more central to the tradition than others. I agree with this as an empirical claim. But on my view it is neither a conceptual or normative necessity that all possible legitimate liberalisms will contain those ideas/values/commitments. We can imagine future iterations with a different core. Thus the centrality of (e.g.) liberty is a historically contingent feature of liberalism. 33 E.g. Sherri Berman, The Primacy of Politics (Cambridge: CUP, 2006); Rogers Smith, Civic Ideals (London: Yale UP, 2007). 31 8 could mean several different things. It might imply that one of the classifications is mistaken, or that they adopt a hybrid position, or alternatively that decomposing the argument will yield some elements that are genuinely “liberal” and others that are genuinely “conservative”. The main problem with these options, however, is that today it is impossible to convincingly classify values (such as liberty or equality) or public policies (such as free trade or democracy promotion) as exclusively liberal or conservative (or something else). They are – they have become – both at once.34 The scholarly implications of tradition-construction can be significant, as the work of Domenico Losurdo demonstrates. His remarkable “counter-history” of liberalism places considerable emphasis on the social practices characteristic of British, Dutch and American societies.35 He contends that the British slave trade was at its peak in the eighteenth century, well after liberalism was consolidated by the settlement of 1688, and that in North America chattel slavery reached its apogee in the early nineteenth century, following the victory of liberalism in the War of Independence.36 John Locke figures heavily in both narratives. If we adopt the current conventional understanding of liberalism, as Losurdo does, this throws up a disturbing puzzle about liberal attitudes to domination, hierarchy, and exploitation, and it underpins his sweeping critique. The normative conclusions that Losurdo draws about contemporary liberalism are derived from, and are only intelligible in relation to, his interpretation of the tradition. But the puzzle dissolves if we adopt (for example) a Pocockian interpretation, because on that account neither Britain nor the United States was liberal in any meaningful sense. Interpretations of tradition often shape contemporary understanding as well as historical investigation. Liberalism before Locke At the turn of the twentieth century, the dominant prescriptive narrative about liberalism in the English-speaking world identified it as a product of the late eighteenth and early nineteenthcenturies, part of a cluster of ideological innovations that also included socialism. At the turn of the twenty-first century, the dominant narrative views it as a product of the mid-seventeenth century or earlier. In the former, the French and American revolutions and the global spread of capitalism play a starring role; in the latter, the Glorious Revolution of 1688 and the religious wars in Europe. In the former, utility, democracy, and political economy are the guiding topics; in the later, natural rights, the social contract, and constitutionalism. In the former, radicals like Jeremy Bentham take centre stage, in the latter it is almost invariably John Locke. Indeed Locke’s foundational role in liberalism is today a leitmotiv of political thought, promulgated by critics and adherents of liberalism alike. “To the extent that modern liberalism can be said to be inspired by any one writer”, Wolin wrote in Politics and Vision, “Locke is undoubtedly the leading candidate”. Stephen Holmes agrees: “The best place to begin, if we wish to cut to the core of liberalism, is with Locke”.37 The transition from one conception to the other tells us much about the It follows that those values/policies are also now part of the conservative tradition (and hypothetically others too). 35 Losurdo, Liberalism, trans. Gregory Elliott (London: Verso, 2011). Losurdo mixes canonical and expressive protocols. 36 On the problem with characterising the nineteenth-century US as liberal, see Daniel Rodgers, “The Traditions of Liberalism” in Questions of Tradition, ed. M.S. Phillips & Gordon Schochet (Toronto: UT Press, 2004), 203-33. 37 Wolin, Politics and Vision, 263; Holmes, Passions and Constraint (Chicago: UCP, 1995), 15. Rawls’s “speculative” history traces liberalism to the Reformation and the sixteenth century religious wars (Lectures, 11). 34 9 trajectory of modern politics, the sociology of knowledge, and the historicity of theoretical categories. In his compelling account of American political thought John Gunnell argues that liberalism only became a widely recognised category of general political discourse after the First World War, and only assumed an important role in academic political theory in the wake of the Second World War. Moreover, he contends that “it was not until after 1950 that there was even any extended discussion of Locke as a liberal”.38 Adding a British dimension to the story complicates this picture. Both the conception of liberalism as a tradition rooted in early modern political thought, and the identification of Locke as a foundational liberal, emerged slightly earlier in Britain than in the US, and for different reasons. Yet despite this initial variation, British and American narratives converged during the ideological battles of the middle decades of the twentieth century, creating the vision of liberalism that dominates scholarly discourse today. While the term “liberal” had long been used in English to denote assorted aristocratic dispositions, mores, and pursuits, it only assumed a specifically political meaning in the early nineteenth century. Borrowed from the Spanish Liberales of the 1812 Constitution, the term was first employed in a derogatory manner by Tories to malign their Whig opponents. During the 1820s it was reclaimed by some radical Whigs, in a classical example of rhetorical redescription, to characterise individuals and policies dedicated to non-revolutionary reform, although it also became associated with the small but vocal group of “philosophic radicals”, including the young John Stuart Mill. “Liberal” was increasingly utilised to describe the politico-economic demands of the emergent middle classes.39 Yet it remained a marginal category: during the 1820s and 30s ‘“liberals’ were not a firmly defined group and ‘liberalism’ did not securely mark out a single intellectual phenomenon.”40 It was only during the second half of the century that usage proliferated, though it was closely tied to the creed of the newly-named Liberal Party.41 Despite its increasing visibility, there was little sophisticated or thorough discussion of liberalism as an intellectual tradition until the early twentieth century, and even then it was rare. The main political theory textbook employed in Cambridge and Oxford in the late nineteenth century, Bluntschli’s The Theory of the State, didn’t use liberalism as an organising category, and nor did Sidgwick’s Development of European Polity, which replaced it in Cambridge at the turn of the century. It is barely visible in surveys of political thought written between the 1850s and the 1930s.42 The effort to construct an authoritative liberal tradition only gained ground during the Gunnell, “The Archaeology of American Liberalism,” Journal of Political Ideologies, 6 (2001), 131; Gunnell, Imagining the American Polity (Philadelphia: PSU, 2004), 183-219. 38 Jörn Leonhard, “From European Liberalism to the Languages of Liberalisms”, Redescriptions 8 (2004), 17-51. 40 D.M. Craig, “The Origins of ‘Liberalism’ in Britain”, Historical Research, 85 (2012), 482. Cf. Daisy Hay, “Liberals, Liberales and The Liberal”, European Romantic Review, 19 (2008), 307-20. For European context, see Maurizio Isabella, Risorgimento in Exile (Oxford: OUP, 2009). 41 The Liberal Party was created from a fissile coalition of Whigs, free-trading Tories, and Radicals. The name was first used officially in 1868, but it had been a common designation since the 1850s. 42 E.g. Frederick Pollock, An Introduction to the History of the Science of Politics (London: Macmillan, 1890); William Graham, English Political Philosophy from Hobbes to Maine (London: Arnold, 1899); Ernest Barker, Political Thought in England from Herbert Spencer to the Present Day (London: Williams, 1915); W.L. Davidson, Political Thought in England from Bentham to J.S. Mill (London: Williams, 1915); Ivor Brown, English Political Theory (London: Methuen, 1920); Robert Murray, The History of Political Science from Plato to the Present (Cambridge: Heffer, 1926); Lewis Rockow, Contemporary Political Thought in England (London: Parsons, 1925); C.E. Vaughan, Studies in the History of Political Philosophy, 2 vols. (Manchester: MUP, 1925). Harold Laski was an exception, identifying liberalism as an economic ideology produced by the Industrial 39 10 perceived “crisis of liberalism” in the Edwardian era. Fighting acrimonious battles over the future of the British state, and challenged by an emergent politically-conscious labour movement, some liberals elaborated edifying genealogies to underwrite the ideological legitimacy of their cause. The most common renditions of the liberal tradition identified the transition from the eighteenth to the nineteenth century as the formative moment. W. Lyon Blease’s Short History of English Liberalism, published in 1913, was typical. A polemical defence of advanced liberalism written by a legal scholar, it argues that liberalism was the product of three revolutions: the industrial (starting in the 1760s), American and French.43 In emphasising the Revolutionary-era origins of liberalism, the ideology was often prescriptively defined by a dual commitment to liberty and social equality, sometimes even democracy. This move excluded earlier Whig political thought. It was a constellation of ideas that could only have emerged after the revolutionary tumult of the late eighteenth century and the rise of a powerful middle class demanding political representation. In 1862, in one of the earliest detailed accounts of liberalism, James Fitzjames Stephen pinpointed the connection: As generally used…“liberal” and “liberalism”...denote in politics, and to some extent in literature and philosophy, the party which wishes to alter existing institutions with the view of increasing popular power. In short, they are not greatly remote in meaning from the words “democracy” and “democratic”.44 Forty years later William Dunning, a prominent American historian and political theorist, argued that “[f]undamentally, nineteenth-century Liberalism meant democracy.”45 In an essay seeking to illuminate the “Historic Bases of Liberalism”, another writer distinguished liberals from Whigs by pointing to the aristocratic character and consequences of 1688. “In none of the great documents of the time,” he contended, “do you find the suggestion that the people should share in the work of government,” for such a conception only emerged in the wake of the French Revolution, and it therefore followed that liberalism could only be a product of the late eighteenth century.46 This view only began to lose popularity in the interwar years, though it did not disappear completely. In a textbook published in 1920, for example, the author declared that the “essence of Whiggism has always been the belief in individual liberty combined with the denial of social equality” and that as such “this conception is rejected by Liberals who have a far wider experience on which to frame their social judgements”.47 Other variants of the prescriptive protocol can also be discerned, including one that reduced liberalism to a species of utilitarian radicalism. Thus A.V. Dicey wrote in 1905 of “Benthamite individualism, which, in accordance with popular phraseology, may often be called conveniently liberalism”.48 Revolution, though with philosophical roots in the seventeenth century: Political Thought in England from Locke to Bentham (London: Williams, 1920), ch. 7. 43 Blease, Short History of English Liberalism (London: Unwin, 1913). 44 Stephen, “Liberalism”, Cornhill Magazine, V (1862), 72-3. See also Leonard Hobhouse, Democracy and Reaction (London: Unwin, 1904), 166. 45 Dunning, “A Century of Politics”, North American Review, 179/577 (1904), 803. 46 P.J. Macdonell, “Historic Bases of Liberalism” in Essays in Liberalism (London: Cassell, 1897), 220. Hillaire Belloc also discussed the liberal tradition entirely in nineteenth-century terms (“The Liberal Tradition”, 1-30). 47 Brown, English Political Theory, 66. “Locke had striven hard and striven successfully for more freedom, but he had never striven hard for more equality” (66). On continuities between Whigs and Victorian liberals, see John Burrow, Whigs and Liberals (Oxford: Clarendon, 1988). 48 Dicey, Lectures on the Relation between Law and Public Opinion in England during the Nineteenth Century, ed. Richard VandeWetering (Indianapolis: Liberty, 2008), 67. 11 It is both striking and symptomatic that in Britain, so often seen as the incubator of liberalism, Locke was not widely regarded as a liberal – let alone a paradigmatic one – until nearly a century after liberalism emerged as an explicit political doctrine. Several generations of self-identified liberals somehow failed to recognise him as one of their own. While Locke’s nineteenth century biographers celebrated him as one of the greatest of philosophers, their verdicts on his political writings were far less positive. Acknowledging him as a leading Whig ideologue who exerted a major influence over eighteenth century political thinking, they almost invariably rejected his theoretical arguments as defective and obsolete.49 In so doing they painted a microcosmic picture of his general reputation during the Victorian age: “Locke meant the Essay” not the Treatises.50 Most accounts of the historical development of modern political thought contended that there had been a radical break – both intellectual and political – at the end of the eighteenth century. A new world had dawned, and there was little space in it for Lockean political theory. Liberalism was figured as the progeny of this gestalt switch. The historicist sensibility that permeated nineteenth century social and political thought was antithetical to the rationalist deductions of Locke, and accounts of natural rights, natural law, and above all the social contract, were widely denigrated as primitive. The eminent legal scholar Frederick Pollock was reiterating a popular line of argument when he claimed that Hume had shown decisively that “even as analysis the mere doctrine is useless.” He concluded that Burke had been right to ridicule the contract as “absurd.”51 Another writer, later the M.P. for the Combined Scottish Universities, used a more colourful insult, scorning it as “the veriest figment of pedantic theorizing that any mystified scholastic ever dreamed”.52 Another common response was to historically relativize Locke’s work, viewing him as a man of (and trapped in) his time. Thus the idealist philosopher W.R. Sorley loftily declared that despite the palpable weakness of Locke’s political theory, “it served its purpose as a justification of the revolution settlement in accordance with the ideas of the time.” 53 Many also questioned Locke’s originality, suggesting that his main political ideas were derived from others, above all Hooker. As G.P. Gooch wrote in his influential account of seventeenth democratic thought, “there is little in Locke that he did not find in the thinkers of the Interregnum”.54 These lines of criticism were synthesized in the first monograph on Locke’s political philosophy (originally a doctoral dissertation supervised by John Dewey): “His moral and political philosophy may well be viewed as the summation of the best thought of the seventeenth century. Though he added few ideas of his own and developed the old ideas he took from others, he is rather the ripe fulfilment of the past than the herald of the future”. The author Lord King, The Life of John Locke (London: Colburn, 1830); H.R. Fox-Bourne, The Life of John Locke (London: King, 1876), II, 524-40; Thomas Fowler, Locke (London: Macmillan, 1880); A. Campbell Fraser, Locke (Edinburgh: Blackwood, 1890); Samuel Alexander, Locke (London: Constable, 1908). 50 Hans Aarsleff, “Locke’s Influence” in The Cambridge Companion to Locke, ed. Vere Chappell (Cambridge: CUP, 1997), 278. 51 Pollock, “The Social Contract in Hobbes and Locke,” Journal of the Society of Comparative Legislation (1907), repr. Pollock, Essays in the Law (London: Macmillan, 1922), 109. See also his “Locke’s Theory of the State” [1904] 80-102. Pollock was unusual in suggesting that Locke envisaged the contract as hypothetical. For the conventional criticism, see Edwin Wallace, “John Locke”, Westminster Review, 107 (1877), 193. 52 Henry Craik, “John Locke,” Quarterly Review, 169 (1889), 490. 53 Sorley, “John Locke” in The Cambridge History of English Literature, ed. Ward and Waller (Cambridge: CUP, 1912), VIII, 390. 54 Gooch, The History of English Democratic Ideas in the Seventeenth Century (Cambridge: CUP, 1898), 358. Gooch did not link Locke and liberalism, thought he suggested that Locke’s account of self-ownership provided the theoretical basis for socialism (358). 49 12 concluded that “Locke’s theory of political society is decidedly weak” and offered little to contemporary political theory.55 Locke spoke from and about a lost world. Nineteenth century philosophers had very rarely seen Locke as a liberal or written positively about his political theory. John Stuart Mill’s assessment is indicative. In the System of Logic he praised Locke as “that truly original genius” and a hugely talented “metaphysician”, yet in the vast corpus of Mill’s work there are only a handful of references to Locke’s political writings.56 His only sustained discussion is in a book review, wherein Mill follows custom in disparaging social contract theories and inalienable rights, while conceding that their proponents rightly identified the importance of limitations on government. “This is the truth,” Mill notes, “which was dimly shadowed forth, in howsoever rude and unskilful a manner, in the theories of the social compact and the rights of man”.57 On Liberty contains one passing reference to Locke, while James Fitzjames Stephen’s powerful riposte, Liberty, Equality, Fraternity, didn’t mention him at all.58 Elsewhere, Stephen belittled Locke as confused and outmoded. The Second Treatise, he argued, …was in its day extremely popular, and its practical effects were no doubt great, as it furnished people with the best and most accessible popular justification for the Revolution of 1688. It would be difficult, however, to find a better illustration of the fact that we have travelled a very long road since Locke’s time, and have carried the metaphysical principles of which he perceived certain aspects, to consequences which have made his political speculations appear altogether superannuated and bygone. His conclusion was equally damning: it was worth studying once popular books “to consider the reasons why they now fall so flatly among us.”59 Herbert Spencer, probably the most widely read English-language philosopher of the age, wrote four major works of political theory – The Proper Sphere of Government (1842), Social Statics (1851), The Man versus the State (1884) and Part IV (“Justice”) of The Principles of Ethics (1891) – and across hundreds of pages Locke was mentioned just once, when his theory of property was casually rejected as “unsatisfactory”.60 T. H. Green, the leading philosophical light of the final quarter of the century, shared Mill’s deep scepticism about the foundations of early modern political thought, and while he expended considerable energy grappling with Locke’s epistemological writings – “at once so plausible and so hollow” – he barely mentioned his political views. Dismissive of the state of nature, pre-political rights, and contracturalism, Green ultimately regarded Locke’s arguments as incoherent and he never saw him as a fellow (or proto) liberal.61 Nor did Henry Sidgwick, who characterised Locke as a philosophically misguided Whig Sterling Lamprecht, The Moral and Political Philosophy of John Locke (NYC: Columbia UP, 1918), 6, 150-51. Locke was a “Whig and a liberal”. 56 Mill, System of Logic [1843], Collected Works, ed. John Robson (Toronto: UT Press, 1974), VII, 29; VIII, 305. In the book under discussion, George Cornewall Lewis’s Remarks on the Use and Abuse of Some Political Terms (1832), “liberalism” is absent and Locke’s views are ridiculed. 57 Mill, “Use and Abuse of Political Terms”, CW, XVIII, 11. 58 Mill, On Liberty (1859), CW, XVIII; Stephen, Liberty, Equality, Fraternity (London: Smith, 1873). 59 Stephen, “Locke on Government” (1867), Horae Sabbaticae (London: Macmillan, 1892), II, 142. 60 Spencer, The Principles of Ethics (Indianapolis: Liberty, 1978), II: 111-12. 61 Green, “Introduction to Hume’s Treatise of Human Nature” (1874) in Collected Works of T.H. Green (Bristol: Thoemmes, 1874), 1:13; Green, Lectures on the Principles of Political Obligation (1886) (Bristol: Thoemmes, 1997), 375. D.G. Ritchie was a partial exception. In the Principles of State Interference (London: Swan, 1891) he linked English empiricism with liberalism, and praised the continuing political relevance of Locke’s writings, though he derided their philosophical value (138, 128). In Ritchie’s Natural Rights (London: Swan, 1895), Locke was characterised as both an ideologue of 1688 and an early liberal (6, 239, 175, 186). 55 13 ideologue.62 In the Edwardian era, Graham Wallace added a new post-Darwinian twist to the story by arguing that Locke’s “plea for a government which should consciously realise the purposes of God” was one of many philosophical utopias rendered irrelevant by modern science.63 The same pattern of omission, disavowal and scorn emerges if we turn from political theory to historical scholarship. In Leslie Stephen’s important History of English Thought in the Eighteenth Century Locke’s ideas were relegated to a primitive past. In relativizing mode he termed Locke’s arguments a “formal apology of Whiggism” and grudgingly admitted that they “did well enough for the quiet time of the eighteenth century.” They were then comprehensively superseded: “That authority vanished when the French Revolution brought deeper questions for solution, and new methods became necessary in politics as in all other speculation”.64 Published during the same decade J.R. Green’s hugely popular history of England classified Locke as a Whig philosopher of 1688 before noting that the social contract had long been regarded as obsolete. 65 Venerated throughout Europe for his prodigious erudition, Lord Acton acknowledged that Locke had been a significant historical actor while assailing the quality of his political theory: “always reasonable and sensible, but diluted and pedestrian and poor”.66 While Acton clearly regarded Locke as a notable member of the “Party of Liberty,” he didn’t think of him as a member of the party of liberalism. In the seminal multi-volume Cambridge Modern History, planned by Acton before his untimely death, Locke was again credited as an influential Whig apologist, albeit one whose political ideas “had already been better expressed by Sidney”.67 The great F.W. Maitland also held a low opinion of Locke, cataloguing the many “grave faults” of Locke’s arguments, above all his literal belief in the historical reality of the social contract.68 Across the Atlantic, Locke’s reputation was barely higher. The standard history of political thought textbook, for example, presented a damning account of his “illogical, incoherent system of political philosophy”.69 This widespread scepticism about the quality and relevance of Lockean political thought was fortified by the historicist “comparative method”, which did so much to shape scholarship during the late nineteenth century.70 Its proponents, the most influential of whom was Henry Maine, challenged deductive models of politics and sought to root the origins and development of customs, language, social structures, and legal forms, in long-term historical-evolutionary processes. Antipathetic to early modern natural law and utilitarianism alike, it provided yet Sidgwick, The Development of European Polity (London: Macmillan, 1903), 364-7, 417-18. Locke is largely absent from Sidgwick’s Methods of Ethics (1874), Principles of Political Economy (1873), or The Elements of Politics (1891). 63 Wallas, Human Nature in Politics, 3rd ed. (London: Constable, 1909), 178. 64 Stephen, History of English Thought in the Eighteenth Century (London: Smith, 1876), II, 150, 135. However, he later acknowledged that Locke had unwittingly laid the foundations for Bentham’s radicalism: “Locke, John (1632-1704)”, The Dictionary of National Biography (London: Smith, 1893), XXXIV, 32. 65 Green, A Short History of the English People (London: Harper, 1878 [1874]), 601-2. 66 Acton, Lectures on Modern History, ed. Figgis and Laurence (London: Macmillan, 1906), 217. 67 Arthur Lionel Smith, “English Political Philosophy in the Seventeenth and Eighteenth Centuries” in The Cambridge Modern History, ed. Ward, Prothero and Leathes (Cambridge: CUP, 1909), VI, 805, 787. 68 Maitland, A Historical Sketch of Liberty and Equality (Indianapolis: Liberty, 2000), 42, 52. Written in 1875, it was only published in 1911. In his Constitutional History of England (Cambridge: CUP, 1908), Locke appeared very briefly as “that excellent Whig” (290). 69 W.A. Dunning, A History of Political Theories from Luther to Montesquieu (London: Macmillan, 1905), 368. 70 On the comparative method, see John Burrow, Stefan Collini & Donald Winch, That Noble Science of Politics (Cambridge: CUP, 1983), ch. 7; Sandra Den Otter, “The Origins of a Historical Political Science in Late Victorian and Edwardian Britain” in Modern Political Science, ed. Robert Adcock, Mark Bevir & Shannon Stimson (Princeton: Princeton UP, 2007), 37-66. 62 14 another weapon to attack the political thinking of the seventeenth and eighteenth centuries. It exercised a profound influence on historical scholarship and the emerging social sciences – perhaps especially political science – on both sides of the Atlantic.71 In the locus classicus of comparatvism, Maine’s Ancient Law, Locke made a fleeting appearance as one of the many thinkers whose ideas about the state of nature and the origins of law were fundamentally mistaken.72 For J.R. Seeley, the leading ideologue of the late-Victorian empire, Locke’s political thinking was simply too ahistorical be of value, while he didn’t even warrant a mention in E.A. Freeman’s Comparative Politics, the first book to apply the method to the development of political institutions across time and space.73 Teaching in the elite English universities reflected both Locke’s prominence as a “metaphysician” and his meagre reputation as a political thinker. At Oxford in the 1870s the Essay, though not the Treatises or Letter, was a compulsory text in moral and political philosophy.74 In Ritchie’s appraisal of the political science curriculum in 1891, the key authors are listed as Aristotle, Hobbes, Bluntschli, Maine, and Mill.75 At Cambridge, William Paley’s Principles of Moral and Political Philosophy (1785) was the standard text during the first half of the nineteenth century. While Paley briefly paid lip-service to Locke’s historical importance, he ignored his arguments and rejected the social contract on utilitarian grounds. Locke’s fortunes didn’t improve during the closing decades of the century. When Henry Sidgwick surveyed the subject in the mid-1870s, Locke failed to make the list of set authors in political philosophy, though students were expected to be familiar with Plato, Aristotle, Cicero, Hobbes, Clarke, Shaftesbury, Butler, Smith, Hume, Kant, Paley, Bentham, Whewell, Mill, and Grote.76 The History tripos paper in “Political Philosophy and General Jurisprudence” followed a familiar pattern. In 1875, for example, Aristotle, Guizot, Tocqueville, Mill, Gibbon, Blackstone, Austin and Maine, but not Locke, were listed.77 Given Locke’s tarnished reputation at the time what are we to make of his current status as the ur-liberal? One possible answer is that it is based on a mistake – that Locke simply wasn’t a liberal.78 Another response is to insist that we have now corrected the error of earlier thinkers who failed to recognise Locke’s liberalism. In other words, he had either always been a liberal or he was never one. Both positions are defensible: it is possible to extract conflicting meanings from Locke’s work. But I suggest an alternative answer: Locke became a liberal during the twentieth century. As part of a process of retrojection his body of work – or at least some stylised arguments stripped from it – was posthumously conscripted to an expansive new conception of the liberal tradition. Robert Adcock, Liberalism and the Emergence of American Political Science (Oxford: OUP, 2014), ch. 5; Duncan Bell, “Alter Orbis: E.A. Freeman on Empire and Racial Destiny” in Making History, ed. Alex Bremner and Jon Conlin (Oxford: OUP, 2015); James Farr, “The Historical Science(s) of Politics” in Modern Political Science, 66-96. 72 Maine, Ancient Law (London: Murray, 1861 [1908]), 101; Freeman, Comparative Politics (London: Macmillan, 1873). 73 Seeley, Introduction to Political Science, ed. Sidgwick (London: Macmillan 1919[1891]), 28. 74 Mark Pattison, “Philosophy at Oxford”, Mind, 1 (1876), 91. 75 Ritchie, “The Teaching of Political Science at Oxford”, Annals of the American Academy of Political and Social Science, 2 (1891), 88. 76 Sidgwick, “Philosophy at Cambridge,” Mind, 1 (1876): 235-46. 77 Jean McLachlan, “The Origin and Early Development of the Cambridge Historical Tripos”, American Historical Review, 9 (1947), 99. 78 Pocock, “The Myth of John Locke and the Obsession with Liberalism” in John Locke, ed. Richard Ashcraft and Pocock (LA: Clark Library, 1980); Goldie, “Introduction”. 71 15 Wars of Position: Consolidating Liberalism The Lockean narrative was consolidated in Britain and the US between the 1930s and the 1950s, as liberalism was reconfigured as the ideological other of “totalitarian” ideologies, left and right.79 This was achieved through two key discursive moves and across two main chronological phases. The first move deepened the retrojective extension of the liberal tradition that had already begun in both Britain and the United States. The early modern account moved from being a minority report to the dominant narrative. The second move was, if anything, even more significant: the emergence and proliferation of the idea of “liberal democracy”, when intellectuals propagated an all-encompassing narrative that simultaneously pushed the historical origins of liberalism back in time while vastly expanding its spatial reach. For the first time, it was widely presented as either the most authentic ideological tradition of the West (a pre-1945 storyline) or the constitutive ideology of the West itself (a view popular after 1945). This story began to coalesce during the 1930s, in a context of radical anxiety about the meaning and fate of liberalism. This was an era where, as Mussolini proclaimed, “[a]ll the political experiments of our day are anti-liberal.”80 Liberals and their critics fought an ideological war of position, attempting to delineate the true, prescriptive meaning of liberalism. The narrative was cemented in the more complacent post-war intellectual milieu as scholars from across the political spectrum, and from assorted academic disciplines, converged on this new all-encompassing narrative, even as they proffered radically different explanations and normative evaluations of it. Strauss, Laski, Macpherson, Hartz, and Wolin, among others, helped to fabricate the new ideological structure. Though rarely acknowledged or analysed, the transformation of liberalism did not go completely unnoticed. In a lecture delivered in 1960, Eric Voegelin observed that “in the course of the last 30 years the image of what liberalism is has changed completely”.81 Wittingly or not, we are the heirs of this ideological labour. The main conceptual shift which facilitated the emergence and popularisation of the Lockean narrative in Britain was the conscription of Whig constitutionalism into a newly expansive conception of liberalism. This move was captured by de Ruggiero in 1933: “The ambitious designs of the radicals, curbed by the tenacious forces of tradition, fused with the older Whiggism to form a composite liberalism in which the old and the new were gradually integrated and harmonized”.82 Contra Ruggiero, however, this discursive “fusion” was largely a product of the twentieth century. Consequently, liberalism came to be viewed through a wide-angle lens, as a politico-intellectual tradition centred on individual freedom in the context of constitutional government. This expansion in ideological scope was also facilitated by shifts in the philosophical current. The eclipse of idealism in the early twentieth century, as well as powerful challenges to utilitarianism, helped to open up the conceptual space in which contracturalism and natural rights arguments could emerge from the shadows. In this new intellectual environment, Locke, the arch-Whig, was recast – by default as much as design – as a seminal liberal thinker and a source of inspiration for an individualist account of political life. This retrojective process began in earnest during the Edwardian years. Hobhouse’s Liberalism, arguably the most comprehensive discussion of liberal political theory published during the first half of the century, played an important role in establishing the lineaments of the (new) Lockean On the concept, see Abbott Gleason, Totalitarianism (Oxford: OUP, 1997). Cited in Ira Katznelson, Desolation and Enlightenment (NY: Columbia UP, 2003), 23. 81 Voegelin, “Liberalism and its History” [1960], Review of Politics, 36 (1974), 504-5. 82 Ruggiero, “Liberalism”, Encyclopaedia of the Social Sciences, ed. E.R. Seligman (London: Macmillan, 1933), IX, 438. 79 80 16 tradition.83 He posited the emergence of liberalism as coeval with the development of the early modern English state. In its original Whig iteration – a theory of the “Natural Order” centred on inalienable prepolitical rights and the restraint of government – it embodied a “negative” form of constitutionalism that sought to eliminate obstacles to human progress. “It finds humanity oppressed, and would set it free”. But, Hobhouse continued, the underlying theoretical architecture was fundamentally flawed, and only during the nineteenth century was a positive dimension added, first by utilitarians and more recently by “new liberals”.84 Thus Hobhouse presented the Whigs as pioneer liberals, albeit now superseded. In addition to providing fellow liberal reformers with a powerful constitutionalist genealogy, he had another good reason for stretching the discursive boundaries of liberalism, as he was engaged in the attempt to craft a liberal socialist politics to replace the desiccated “old liberalism” of the “Manchester School” and the Benthamites.85 Yet this Lockean narrative, a precursor of things to come, remained marginal until the 1930s, and scholarly and popular discussions of liberalism were most commonly tied to the quotidian concerns of the often-embattled Liberal party.86 When R.G. Collingwood wrote the translator’s preface for Ruggiero’s History of European Liberalism in 1927, he still felt it necessary to inform his audience that the book addressed liberalism in the “continental” not the “British” sense, as a “name for principles of constitutional liberty and representative government,” rather than a party ideology.87 The First World War and its aftermath also saw early attempts to self-consciously define an American form of liberalism with its roots in the seventeenth century. Progressive scholars and publicists took the lead.88 The critic Harold Stearn was one of the first. He drew heavily on Hobhouse’s account of the true meaning of liberalism, but his historical narrative had a different emphasis, focusing in particular on religious toleration and the catalytic role of Roger Williams, the seventeenth century Protestant theologian and colonist.89 Despite dedicating a chapter to “what liberalism is” and another to the “English heritage” of American liberalism, he never mentioned Locke. Interpreting liberalism as an ideology centred on religious toleration become a popular theme in American scholarship, exemplified by Vernon Parrington’s hugely influential Main Currents in American Thought, published in the late 1920s though composed largely in the 1910s.90 Parrington argued that liberalism was developed originally in the natural law theories propounded by the early Puritan settlers, who had been transplanted from a European environment inhospitable to their radical claims into a welcoming new world in America, where liberalism could truly flourish. Though Parrington stressed the centrality of Williams – “England 83 As late as 1963, C. Wright Mills claimed that Liberalism was the best account of the subject: The Marxists (Harmondsworth: Penguin, 1963), 25n. 84 Hobhouse, Liberalism, 24-5, 8. For another clear usage, see A.W. Benn, The History of English Rationalism in the Nineteenth Century (London: Longman’s, 1906), I, 111. Compare Herbert Samuel, Liberalism (London: Grant Richards, 1902), which does not mention Locke and makes little effort to trace a genealogy. 85 Michael Freeden, The New Liberalism (Oxford: OUP, 1978). 86 Freeden, Liberalism Divided (Oxford: OUP, 1986). 87 Collingwood, “Translator’s Preface”, vii. Cf. John Morley, Recollections (London: Macmillan, 1917), I, 21. 88 On the transition from Progressivism to liberalism, see Gerstle, “Protean Character”. On the transatlantic dialogue between British and American thinkers, see James Kloppenberg, Uncertain Victory (Oxford: Oxford UP, 1986); Marc Stears, Progressive, Pluralists, and the Problems of the State (Oxford: Oxford UP, 2002). 89 Stearns, Liberalism in America (NY: Boni, 1919), 11, 16-17, 33-34. Charles Merriam dismissed it as a shallow piece of partisan propaganda: American Political Science Review, 14/3 (1920): 511-12. 90 Parrington, Main Currents in American Thought, 3 vols. (NY: Harcourt-Brace, 1927-30). Gunnell argues that Parrington’s trilogy marked the “threshold of the adoption of liberalism as an American political identity – both in politics and political theory”: Gunnell, “Archaeology”, 132. Thanks to Robert Adcock for discussion of this topic. 17 gave us her best”91 – he also assigned Locke a prominent role. The connection between liberalism and toleration helped open the door for placing Locke at the centre of the tradition. Thus whereas parliamentary constitutionalism was central to the British appropriation of Locke (via the retrojection of the Whigs), it was religious toleration (via the retrojection of key elements of Puritanism) which did much of the ideological labour in the United States. By 1918, some of the key building blocks were in place, but the ultimate hegemony of the Lockean narrative was still far from guaranteed, and it required a series of political and intellectual moves to produce it. The discursive consolidation of this extensive new account of liberalism was a product of the complex interweaving of geopolitical dynamics and disciplinary imperatives within the human sciences, especially political science and history. Indeed the academic disciplines which profess to instruct us about the nature of liberalism played a fundamental role in its transfiguration. The shift unfolded in the context of a transfer of scholarly authority from Britain to the US. Whereas British commentators had shaped the contours of interpretation in the late nineteenth and early twentieth centuries, exerting a profound influence (alongside German scholarship) on the development of American political science and history, by 1945 a decisive shift across the Atlantic was apparent. The new liberal narrative was thus largely a product of the US human sciences, though it was mirrored in Britain. The change in meaning is captured in the evolution of George Sabine’s influential conspectus of western political thought, which was one of the first major scholarly texts to discuss liberalism in detail. The first edition, published in 1937, rooted its origins squarely in nineteenth-century Britain, figuring it as a distinct ideology positioned between socialism and conservatism. (Locke was not characterised as a liberal). Moreover, like so many of his contemporaries, Sabine worried that liberalism “was a diminishing force in modern society”.92 In the revised edition of 1951, however, his account of liberalism was both more capacious and more confident, and he asserted that it now had two main senses. The first, which he associated with Fascist and Marxist critics, saw it as the “social philosophy of the industrial middle class” and thus coterminous with laissez faire capitalism. Rejecting this critique, he endorsed a far broader account of liberalism as both the “culmination” of Western history and largely synonymous with democracy.93 Here he followed political theorist Frederick Watkins, who had recently celebrated liberalism as the “secular form of Western civilization” and the “modern embodiment of all the characteristic traditions of Western politics”.94 Sabine concurred: “political liberalism has been deeply implicated in the whole development of Western culture”.95 (The ultimately unsuccessful attempt to retroject liberalism back into the ancient Greek world, thus making it coterminous with Western civilization, was one of the signature ideological moves of the era.)96 An irony appears 91 Parrington, Main Currents, I, 74. Sabine, A History of Political Theory (London: Harrap, 1937), 679. In 1941 he observed that “[to] give a practical definition of liberalism is virtually impossible”: Sabine, “The Historical Position of Liberalism,” American Scholar, 10 (1940-1), 490. 93 Sabine, A History of Political Theory (NYC: Holt, 1951), 620. In a review, Macpherson noted the shift in meaning and concluded that the “ideological atmosphere in America” made an understanding of both liberalism and Marxism “increasingly difficult”: Western Political Quarterly, 4 (1951), 145. 94 Watkins, The Political Tradition of the West (Cambridge: HUP, 1948), ix. Watkins made liberalism coextensive with freedom under the law. For an influential conservative political-theological account that adopted the same timeline but reversed the normative conclusion, see John Hallowell, The Decline of Liberalism as an Ideology (Berkeley: UC Press, 1943); Hallowell, Main Currents in Modern Political Thought (NYC: Holt, 1950). 95 Sabine’s review of Watkins, Political Science Quarterly, 64 (1949), 147-49 96 The phenomenon was noted in Francis Coker, “Some Present-day Critics of Liberalism,” American Political Science Review, XLVII (1953), 1-2. 92 18 lost on Sabine. Whereas linking democracy and liberalism had, in the nineteenth century, served to delimit its chronological scope, it was now employed to buttress the claim that liberalism was the spiritual inheritance of the West itself. Confusion reigned. As liberalism’s boundaries were conceptually stretched, so whatever fragile coherence it once had was lost. In the mid-1930s Dewey observed that “liberalism has meant in practice things so different as to be opposed to one another,” and it only got worse. 97 A decade later a noted philosopher could insouciantly observe that “we, too, have our ‘ideology,’ inherited from the past as the liberal tradition, the American creed, the Judeo-Christian heritage of Western civilization or the like.”98 For many, these ideas had become interchangeable. The tendency to construct legitimating genealogies for ideological purposes provoked the ire of a young C.B. Macpherson, who complained that too many scholars plotting the history of western philosophy substituted serious analysis with assertions of political faith, “using their history to show how long and honourable an ancestry that faith has”.99 A new piece of conceptual technology was added when the term “neo-liberalism” was coined in the late 1930s. During the last thirty years, it has served as shorthand for the valorisation of the minimal state and deregulated market, but (to add to the confusion) it was originally utilised to identify a desirable via media between unrestrained capitalism and progressive statism.100 Commentators grumbled endlessly about the theoretical muddle. One frustrated scholar marvelled in 1948 that “[o]ne finds the term employed to defend everything from classical economics to the Soviet interpretation of communism.”101 In 1955 Reinhold Niebuhr addressed the “confusion”, arguing that liberalism had come to denominate both a phase of human history, “the rise of a modern technical society availing itself of democratic political forms and of capitalistic economic institutions”, and a specific set of partisan political commitments. It also signified two “contradictory” claims, namely that liberty necessitated both the unleashing of capitalism and its radical restraint.102 A similar pattern can be discerned in Britain. The translation of de Ruggiero’s History of European Liberalism and the publication of Laski’s The Rise of European Liberalism fortified the early modern liberal narrative.103 It became the norm during the 1930s and 1940s.104 Sceptical of claims about seamless continuity, Isaiah Berlin summed up the nature and ideological appeal of what had become a popular position by 1950: “European liberalism wears the appearance of a single coherent movement, little altered during almost three centuries, founded upon relatively simple foundations, laid by Locke or Grotius or even Spinoza; stretching back to Erasmus and Dewey, Liberalism, 3. Arthur Murphy, “Ideals and Ideologies, 1917-1947,” Philosophical Review, 56 (1947), 386. 99 Macpherson, “The History of Political Ideas”, Canadian Journal of Economics and Political Science, 7 (1941), 564-5. 100 On neo-liberalism, see Ben Jackson, “At the Origins of Neo-Liberalism”, Historical Journal, 53 (2010), 129-51; Angus Burgin, The Great Persuasion (Cambridge: HUP, 2012). For a contemporary attempt to distinguish Ordo-liberalism and the work of the Mont Pelerin Society, see Carl Friedrich, “The Political Thought of Neo-Liberalism”, American Political Science Review, 49 (1955), 509-25. 101 Boyd Martin, “Liberalism,” Western Political Quarterly, 1 (1948), 295. 102 Niebuhr, “Liberalism,” New Republic (1955). He endorsed the “Lockean type of liberalism”. 103 Laski, The Rise of European Liberalism (London: Allen & Unwin, 1936), 9, 115. For a similar analysis, see Richard Crossman, Government and the Governed (London: Christopher, 1939), 69-80. Laski’s analysis was not unchallenged: “There is plenty of truth in this as a historical account, though it is a one-sided truth. But to speak of it as Liberalism shows a bad confusion of thought”. G.C. Feild, Mind, 45 (1936), 527. 104 E.g., George Catlin, The Anglo-Saxon Tradition (London: Kegan, 1939); Catlin, The Story of the Political Philosophers (NYC: McGraw-Hill, 1939); Michael Oakeshott (ed.), The Social and Political Doctrines of Contemporary Europe (Cambridge: CUP, 1939), xi-xviii; Thomas Cook, History of Political Philosophy from Plato to Burke (NYC: Prentice-Hall, 1936), 710-11; J.P. Mayer, Political Thought (London: Dent, 1939). 97 98 19 Montaigne, the Italian Renaissance, Seneca and the Greeks”.105 By the early 1960s Kenneth Minogue, a young theorist at the LSE, could confidently assert that liberalism was a “single and continuing entity...so extensive that it involves most of the guiding beliefs of modern western opinion” and that John Locke was its “founding father”.106 The new historical narrative was adopted by both critics and celebrants of liberalism. Converging on description, they diverged in both explanation and normative evaluation. From the left, for example, Laski depicted liberalism as an ideology with foundations bored deep into the bedrock of western history: “liberalism has been, in the last four centuries, the outstanding doctrine of Western civilization”. It supplied the ideological scaffolding of modern capitalism, and unsurprisingly Locke now played a starring role as the “most representative prophet” of the new age.107 This line of critique reappeared in the work of Laski’s student Macpherson and is still popular today.108 On the political right, meanwhile, Strauss, Voegelin, and others, also pressed variations on the early modern theme. Self-proclaimed liberals were only too happy to vaunt the robust durability and deep roots of their creed, bolstering its ideological armature in the face of hostile competition. Narrative convergence helped produce discursive hegemony. It was against this imposing – but quite new – ideological edifice that the contextualist scholars of the 1960s fought their rear-guard action. Arguably, the most significant conceptual move of the interwar era was the emergence of the idea of “liberal democracy”. Barely visible before 1930, in the ensuing decade it began to supplant existing appellations for Euro-Atlantic states.109 During the 1940s and 1950s it became a commonplace.110 As a global conflict over the proper meaning of democracy raged, the modifier “liberal” simultaneously encompassed diverse representative parliamentary systems while differentiating them from others claiming the democratic title, above all Italy, Germany, and the Soviet Union. As Ernest Barker observed the year after Hitler assumed power, the “issue of our time is hardly a simple issue of democracy versus dictatorship. Dictatorship itself claims the quality of democracy; indeed it claims the quality of a higher, a more immediate, spontaneous democracy”. This was, then, a clash between “two types of democracy – the parliamentary type…and the dictatorial type.”111 Liberal democracy was the name increasingly adopted to cover the former in its conflicts against the latter. Social scientists soon began to utilise the concept, usage that was refined and normalised after 1945. By 1954 Quincy Wright could assert confidently that the concept of “liberal democracy” originated in sixteenth century Europe, especially in England, and was powerfully articulated in Locke’s political philosophy.112 The Berlin, “Political Ideas in the Twentieth Century”, Foreign Affairs, 28 (1950), 357. Minogue, The Liberal Mind (London: Methuen, 1962), vii, 2. 107 Laski, Rise of European Liberalism, 9, 115. 108 Macpherson, The Political Theory of Possessive Individualism (Oxford: Clarendon, 1962); Losurdo, Liberalism. 109 E.g., M. Parmlees, “Liberal Democracy, Fascism, and Bolshevism,” Annals of the American Academy of Political and Social Science, 180 (1935), 47-54; J.A. Leighton, Social Philosophies in Conflict (NYC: Appleton, 1937); Alfred Zimmern (ed.), Modern Political Doctrines (Oxford: OUP, 1939), xiv-xix; Crossman, Government, 286-7, 294-6; J.A. Hobson, “Thoughts on our Present Discontents,” Political Quarterly, 9 (1938), 47-57; E.H. Carr, The Twenty Year’s Crisis, 1919-1939 (London: Macmillan, 1939), 37. See Oakeshott, Doctrines, xvi-xix, for sceptical acknowledgement of the linguistic shift. 110 A google Ngram graph shows this post-1930 spike in usage: http://books.google.com/ngrams/graph?content=liberal+democracy&year_start=1800&year_end=2000 &corpus=0&smoothing=3. Google Scholar also offers illustrative evidence. Scrubbed of false positives, the term “liberal democracy” is employed with the following frequency: 1900-1910 - 6; 1911-1920 - 33; 1921-1930 - 24; 1931-1940 - 143; 1941-1950 - 216; 1951-1960 - 374. 111 Barker, “Democracy since the War and its Prospects for the Future”, International Affairs, 13 (1934), 757. On the threat, see Ira Katznelson, Fear Itself (NYC: Norton, 2013). 112 Wright, “International Law and Ideologies,” American Journal of International Law, 48 (1954), 619. 105 106 20 Lockean narrative was thus frequently generalized into a broader claim about the Lockean-liberal character of Anglo-American societies, an interpretive strategy popularised by Louis Hartz and that was to have a profound effect on the emergent subfield of comparative politics.113 Conjoining “liberal” to democracy automatically (and vastly) expanded the scope of those purportedly encompassed by liberalism, as supporters of “liberal democracy” were conscripted, however reluctantly, to the liberal tradition. Liberalism was thus transfigured from a term identifying a limited and contested position within political discourse to either the most authentic expression of the western tradition or a constitutive feature of the West itself. Again, this conceptual shift was rarely acknowledged, though it didn’t pass completely unremarked. Strauss noted the peculiarity, and the “serious difficulty” for interpretation, that resulted from the “fact that here and now liberalism and conservatism have a common basis; for both are based here and now on liberal democracy, and therefore both are antagonistic to Communism”.114 The political instrumentalization of intellectual history was widespread across the Euro-Atlantic world, reaching its reductio ad absurdum in Bertrand Russell’s claim that “[a]t the present time Hitler is an outcome of Rousseau; Roosevelt and Churchill, of Locke”.115 It is thus unsurprising that history provided another disciplinary space for propagating the new vision of liberalism. The “history of ideas”, an emergent field combining history and philosophy that “rose like a new sign in the zodiac over large areas of American culture and education”, was, like political theory, transformed by émigré scholars, including Hans Baron, Ernst Cassirer, Felix Gilbert, Raymond Klibansky, Paul Kristeller, Hajo Holborn, and Erwin Panofsky.116 Its zealous proponents helped to define and defend a holistic “Western” civilization based on “liberal” values, and as such it was of “strategic” value in fighting totalitarianism.117 As the classroom became as powerful vector for the transmission of the new liberal-civilizational creed, so the Journal of the History of Ideas, founded in 1940, served as the principal venue for its scholarly elaboration. It is no coincidence that it was the only academic journal to receive a secret subsidy from the CIAsponsored Congress on Cultural Freedom.118 University curricula, then, provided institutional authority for the transvaluation of liberalism. “Western civilization” courses, which flourished from the end of the First World War until the 1960s, popularised “an interpretation of history that gives the United States a common development with England and Western Europe and identifies this ‘civilization’ with the advance of liberty and culture”. Helping to construct a mythopoeic narrative of the West as simultaneously ancient and modern, free and strong, they were the most widely taught history courses after the Second World War.119 While claims about the intellectual coherence, historical continuity, and ethico-political superiority of “the West” stretched back at least as far as the eighteenth century, it was only in the mid-twentieth century that this potent civilizational narrative came to be routinely classified as liberal. The victorious spread of liberalism and the rise of the West came to be seen as one and the same thing. Conclusion: Conscripts of Liberalism Hartz, The Liberal Tradition in America (NYC: Harcourt, 1955). Strauss, Liberalism, vii. He also wrote that “The conservatism of our age is identical with what originally was liberalism” (ix). Cf. Voegelin, “Liberalism,” 507; Hayek, “Liberalism,” 113. 115 Russell, History of Western Philosophy (London: Unwin, 1945/1948), 711. 116 Anthony Grafton, “The History of Ideas”, Journal of the History of Ideas, 67 (2006), 1. 117 Jotham Parsons, “Defining the History of Ideas,” Journal of the History of Ideas, 68 (2007), 682-89. 118 Francis Stonor Saunders, Who Paid the Piper? (London: Granta, 1999), 338. 119 Gilbert Allardyce, “The Rise and Fall of the Western Civilization Course,” American Historical Review, 87 (1982), 706; Peter Novik, That Noble Dream (Cambridge: CUP, 1988), 312. 113 114 21 The nature of liberalism has been a core concern in political theory since its advent as an academic specialism in the early twentieth century. In this essay I have criticised some prominent approaches to interpreting liberalism, introduced some methodological tools for thinking about the proliferation of liberal languages, and sketched an explanatory account of shifts in the meaning of liberalism in the Anglo-American world. The analysis has implications for both political theorists and historians. Above all, it suggests the need to be alert to the historical contingency and variability of our theoretical vocabularies and the power dynamics of traditionconstruction. It also calls into question the general utility of “liberalism” as a category of political analysis. Current debates about the nature of liberalism – in and beyond political theory – are often distorted because of the ahistorical understanding of liberal ideology that they invoke. Conducted in a discursive echo-chamber, they are often marked by a symptomatic form of collective amnesia, a problematic erasure of the political and intellectual dynamics that generated much of what is now articulated as scholarly common sense. This essay is intended as a modest contribution to the work of historical recovery. As Stephen wrote in 1862, “[t]he words ‘liberal’ and ‘liberalism’, like all other such phrases, derive a great part of their significance from the time they were invented”.120 The history of liberalism, though, is a history of constant reinvention. The most sweeping of these occurred in the middle of the twentieth century, when liberalism was increasingly figured as the dominant ideology of the West – its origins retrojected back into the early modern era, it came to denote virtually all nontotalitarian forms of politics as well as a partisan political perspective within societies. This was partly a consequence of the delegitimation of political extremes, partly a result of the vicissitudes of domestic political strife, and partly a result of political and conceptual labour performed in to the developing human sciences. Karl Popper once referred to The Open Society and its Enemies as his “war effort”, a contribution to the fight against totalitarianism. The construction of Lockean liberalism was a grander, more all-encompassing variation on the same theme. 120 Stephen, “Liberalism,” 70. 22