Ideology

Defining ideology is no simple feat. Its meaning has changed over time and competing conceptions are in use. Ideologies, however, – and thus also their conceptualization – have a definite history. I will therefore proceed by giving a – necessarily brief – historical account of the emergence and development of this concept, which will provide the basis for a definition of ideology and its implications.

The term ideologue – just as, and closely linked to, the term social science – makes its first appearance during the French Revolution. Replacing the old Académie Française and Académie Nationale, in 1795 the Directorate established the Institut nationale des sciences et des arts in order to bring all the sciences together. This Institut was organized into different Classes, one of which was the Classe des science morales et politiques which, in turn, was divided into sections. The first of these sections was devoted to the study of sensations and ideas, led by Destutt de Tracy and Cabanis. The members of this section were commonly called les ideologues. Their work was based on the assumption that thought was ultimately rooted in physical nature so that mental and moral phenomena could, in principle, be traced back to their physical roots. They were influenced particularly by Condillac, Locke, Maupertuis, LaMettrie, d’Holbach, Helvetius, and Montesquieu. In this context, ideology simply meant the systematic, social scientific, study of ideas, and the term had no pejorative connotation.

Politically, this study of ideas was clearly directed against traditional prejudices – especially religious thought – whose unfounded nature the materialist approach was designed to uncover. Yet, having experienced the terror of the revolution, les ideologues distanced themselves from the revolutionaries’ concern with political rule and manipulation and focused on explaining the deep structural forces shaping society. Although they were broadly liberals, their work could be appropriated by conservatives and socialists alike (Wokler, Reference Wokler2006).

Although a link between the physical or social environment and types of thought had been contemplated before – Bacon’s ‘idols’ (Reference Bacon1996: 227, 228), preconceptions developed within different social spaces, and Montesquieu’s spirit of the laws (Reference Montesquieu1949) conditioned by the environment, are examples – the elevation of such epistemologies and theories into the Institut nationale and their rapid diffusion throughout social and political thought attests to a society in which two radically different ontological and epistemological positions had developed; to a situation in which different sections of society rested their truth claims on utterly different evidence – such as the Bible on the one hand and empirical evidence found in (human) nature on the other.

Yet, this non-pejorative conception of ideology was not to survive for very long. Les ideologues were opposed to Napoleon’s imperialist policies; the latter, in response, denigrated their ‘pompous attempt to build castles in the air’ (Freeden, Reference Freeden1996: 14) as out of touch with the practice of politics and power by calling them ideologistes. Indeed, Napoleon proved his point by shutting down the entire Classe in 1803. In Napoleon’s usage, the term ideology stood for abstract and powerless thinking, detached from reality, and especially the reality of practical politics. This conception of ideology does not play a major role in the social sciences; and yet, since Napoleon’s intervention ‘the problem implicit in the term ideology – what is really real? – never disappeared from the horizon’ (Mannheim, Reference Mannheim1960: 64).

Marx’s work provides the next milestone in the development of the conception of ideology. He famously argued in the German Ideology that ‘life is not determined by consciousness, but consciousness by life’ (Marx and Engels, Reference Marx and Engels1970: 47), and thus linked specific ideologies, or world views, to social classes. In capitalist society, argued Marx, the universalized exchange of products between equals – in liberal ideology presented as the general principle of society – hides the real and unequal social relations at the level of production on which this society rests. Yet, presenting this partial truth of relations of equality as the whole truth, Marx argued, was ultimately only in the interests of the ruling classes who profited from this arrangement. Ideology here is the theoretical expression of capitalist relations of production – and more specifically, the world view which presents the particular interests of the bourgeoisie as universal interests, thus justifying the existing social and political order.

Marx’s conception of ideology was a major innovation. First, it introduced the negative connotation – ideology as an instrument of domination – into the social scientific concept. It secondly implied that ideologies are a necessary feature of class societies. And, thirdly, it introduced the problem of ‘false consciousness’. Not all workers necessarily embraced the socialist – and thus, from a Marxist perspective correct – world view. The phenomenon of sections of the working class ‘buying into’ the liberal ideology was called ‘false consciousness’ – that is, subjective belief in, and support of, an ideology that stands in objective contradiction to one’s interests. For subsequent theorizations of ideology, this phenomenon clearly suggested that while there was a link between the particular social, political, or cultural circumstances of groups and their respective world views, this link did not take the form of a strict causal relationship. It was this problem of ‘false consciousness’ that later came to inspire the work of Antonio Gramsci (Reference Gramsci1971), Louis Althusser (Reference Althusser1984), Cornelius Castoriadis (Reference Castoriadis1998) and others. In addition to complicating the relationship between social position and ideology, the phenomenon of ‘false consciousness’ also clearly implied that ideologies worked to a great extent at the level of the unconscious – they were not lies but rather misperceptions.

This new conception of ideology, however, did not just signify an intellectual development of the term but also new historical developments. While the liberal ideologues had used it to show up the unfounded nature of traditional beliefs or ‘prejudices’, all the while claiming access to the ‘real truth’ through their own empirical methods, now the ‘socialists’ demonstrated the ideological nature of the liberal world view and claimed a nonideological position for themselves. As a weapon in political struggle, the charge of ideology had thus been handed down from the liberals to the socialists – reflecting the now dominant position of liberalism in society.

Karl Mannheim’s classical study Ideology and Utopia (1960), originally published in 1929, represents the next major development in the conception of ideology. Tracing the history of ideology as a political weapon – first wielded by the liberals against the Church and the ancien régime, then by conservatives and socialists against the liberals, and finally at the time of his writing by all ideologies, including the fascist ones, against all others (1960: 33) – Mannheim pointed out that ‘it is precisely this expansion and diffusion of the ideological approach which leads finally to a juncture at which it is no longer possible for one point of view and interpretation to assail all others as ideological without itself being placed in the position of having to meet that challenge’ (1960: 66).Footnote 2 Mannheim thus identified the Marxist, socialist, or communist position as just as ideological as its liberal, conservative, or fascist counterparts.

Mannheim distinguishes between a particular conception of ideology and a total conception of ideology. ‘The particular conception of ideology is implied when the term denotes that we are sceptical of the ideas and representations advanced by our opponent. They are regarded as more or less conscious disguises of the real nature of a situation, the true recognition of which would not be in accord with his interests’ (1960: 49). In contrast, the total conception of ideology refers to the Weltanschauung or world view of a particular age or ‘concrete historico-social group’, that is, to the ‘total structure of the mind of this epoch or of this group’ (1960: 49, 50). The difference between these two conceptions of ideology is important because the former only questions a part of the adversary’s claims while continuing to assume common ground on which truth claims can be made. In contrast, the total conception of ideology questions not just the one or other concrete claim of the adversary but his entire conceptual apparatus ‘as an outgrowth of the collective life of which he partakes’ (1960: 50). And only in this total conception can the term ideology be applied to historical epochs or social groups. In short, the term ideology implies ‘that there is a correspondence between a given social situation and a given perspective, point of view, or apperception mass’ (1960: 51).

What Mannheim identified, historically and theoretically, was the shift from a juxtaposition of ‘true’ vs. ‘ideological’ world views to a situation in which all world views were regarded as ideological and pitched against each other. This shift had serious consequences for the social sciences. If all knowledge of the world was linked to, enabled and constrained by, the social, political, and cultural circumstances of its emergence, then knowledge was by definition ideological. Whereas ideology as the study of ideas was originally simply one way (among others) of fulfilling the promise of the social sciences to generate ‘true’ and ‘objective’ knowledge – it was the very success of this way of thinking which ultimately led the concept of ideology and its spread throughout society to undermine the basis, the very possibility, of attaining ‘true’ and ‘objective’ knowledge.

In the social sciences, this problem was addressed by two main responses (Mannheim, Reference Mannheim1960; Freeden, Reference Freeden1996). The first lies in positivist, the second in normative theory. Normative theory focused on the development of logical and consistent normative principles which were then juxtaposed to the existing imperfect order. This normative solution to the problem has led in political theory, as Freeden notes, to a prioritization of the study of political philosophy and a marginalization and denigration of the study of ideology – as a mode of thought entailing elements of irrationality, of emotion, of political praxis (1996: 134).

Consequently, the study of ideologies, especially in American political science, was left to a ‘positivist empiricism’ that identified and investigated ideologies as a wide spread social phenomenon (Freeden, Reference Freeden1996: 15). In this positivist conception, ideologies are not understood as aberrations from a ‘right’ way of thinking but as systems of political ideas fulfilling important, often integrative, functions in modern society. Positivist conceptions thus rightly point towards the manifold and crucial social functions of ideologies, leaving their role in obscuring and perpetuating structures of domination as the focus of critical conceptions largely unaddressed (Freeden, Reference Freeden1996: 19).

On the basis of this brief historical sketch, it is now possible to formulate a cumulative conception – based on the relatively undisputed characteristics – of ideology which can serve as a basis for the ensuing discussion. Ideologies are systems of political thought arising out of, and reflecting, the economic, political, and cultural experience of particular social groups. They function to integrate these groups and to mobilize them for political action – against competing groups and their ideologies. In class or stratified societies, ideologies are ubiquitous forms of political thinking and play a constitutive and often integrative role for the entire society (most obviously in the institutionalized form of party politics) (Freeden, Reference Freeden1996: 14, 22, 23, 552). Since ideologies reflect the socio-economic practices of particular groups, they do contain elements of truth – but of a historically and socially limited nature. In the course of political struggle, for the purposes of mobilizing broader sections of society, however, these particular truths are presented as general truth. Moreover, in the course of this struggle, ideologies constantly adjust to changing conditions, thus giving rise to internal diversity and historical development.

For the purpose of the following argument it is important to draw attention to the political and epistemological implications of this concept. The concept of ideology, as its history shows, attests to the essentially fragmented nature of politics – to the political realm as a space populated by different social groups and their respective world views engaged in a constant struggle for power (Freeden, Reference Freeden1996: 22, 23). And it is this contested nature of the political realm which has significant implications for political knowledge. For if the fragmented nature of this realm constitutes and circumscribes human thinking – which ‘is now virtually a truism’ (Freeden, Reference Freeden1996: 14) – then political knowledge can only ever be particular knowledge. For the social sciences in general, and International Relations in particular, the concept of ideology thus raises the question whether, and if so how, it is possible to transcend the ideological limits of political knowledge.

Liberal theory as methodological ideology

Although Moravcsik does not define his concept of ideology, it can, to some extent, be reconstructed from the context. The ‘ideological’ liberal approaches, according to Moravcsik (Reference Moravcsik1997: 514), ‘either collect disparate views held by “classical” liberal publicists or define liberal theory teleologically, that is, according to its purported optimism concerning the potential for peace, cooperation, and international institutions in world history’. He notes that these approaches ‘offer an indispensable source of theoretical and normative inspiration’ (1997: 514). But they do not satisfy the ‘more narrowly social scientific criteria’ deserving the title ‘theory’ (1997: 514). Such a theory, in contrast to the ‘ideological’ approaches, is defined by ‘a set of positive assumptions from which arguments, explanations, and predictions can be derived’ (1997: 514).

Thus, what is wrong with the ‘ideological’ approaches is, first, that they bring together ‘disparate’ views – views that are somehow logically unconnected. Secondly, these views are often taken from ‘classical’ writers – raising questions about their continuing validity in the contemporary world. Thirdly, they may be ‘teleologically’ defined – that is, governed by political desires and aspirations rather than facts. In singling out these aspects, Moravcsik actually touches upon a number of the characteristics of ideologies. For ideologies, as set out above, do contain contradictions; they are by definition historically bound and thus also changing; and they are geared towards political action.

Moravcsik’s conception of a proper theory is thus designed to avoid precisely these weaknesses. In contrast to the ‘ideological’ approaches, Moravcsik claims that his own theory is ‘logically coherent, theoretically distinct, empirically generalizable’: it ‘follows from explicit assumptions and generates a rich range of related propositions about world politics that reach far beyond cases of cooperation among a minority of liberal states’ (Moravcsik, Reference Moravcsik1997: 547). First, logical inconsistency is dealt with by demanding that the theory be derived ‘from a limited number of microfoundational assumptions’ (1997: 515). Secondly, the theory has to be clearly distinguished from other theories. Thirdly, the problem of historical and social, political, cultural limitations and variations is addressed by demonstrating through empirical support that the theory applies generally – in this case, as Moravcsik claims, to liberal and nonliberal states, economic and security areas, conflictual and nonconflictual situations as well as to individual states and international relations as a whole (1997: 515). Finally, emotions, passions, values, political desires, and aspirations are kept out of this theory through the requirement of empirical substantiation. Claims about the future come in the form of predictions based on past empirical evidence. Moravcsik thus solves the problem of ideology methodologically. He applies a positivist–empiricist method which is designed to keep the social, political, and cultural biases and limitations characteristic of ideologies out of theory.

On the basis of this reconstruction of Moravcsik’s understanding of ideology, and of the means by which his theory aims to address the problems of the former, it is now possible to follow his own suggestion and to judge the project ‘on its own terms’ (Moravcsik, Reference Moravcsik1997: 548).

The starting point and core element of Moravcsik’s attempt to move beyond an ideological formulation of liberal theory lies in the link between microfoundational claims and the derivation of liberal theory from these claims. Explanations and predictions are then drawn from the latter and need to be empirically substantiated in order to support the theory. It is this connection between the core assumptions, the theory itself, and the resultant explanations and predictions which guarantee logic and coherence rather than the disjointed or even contradictory claims associated with ideologies. Moravcsik formulates three fundamental empirical claims: first, that the ‘fundamental actors in international politics are individuals and private groups who are on average rational and risk-averse’ with their differentiated interests; secondly, that state preferences represent the interests of some subset of society; and thirdly, that ‘the configuration of interdependent state preferences determines state behavior’ (1997: 516, 518, 520).

The idea now, is to derive the theory of international relations from these general assumptions. However, Moravcsik acknowledges that his empirical assumptions are ‘thin’ and ‘content-free’; ‘they do not define a single unambiguous model or set of hypotheses’ because they ‘do not specify precise sources of state preferences’ (1997: 524). He thus identifies three variants of liberal theory – ideational liberalism, commercial liberalism, and republican liberalism – that contain these empirical assumptions in a richer context (1997: 524). Yet, by starting with these richer theories in which he then identifies his general assumptions, Moravcsik has reversed the order by which a proper ‘theory’ in his own terms has to be established. This reversal, I will now show, ruptures the logical coherence of his theory.

Ideational liberalism ‘views the configuration of domestic social identities and values as a basic determinant of state preferences’ (1997: 525). The core claim of ideational liberalism, Moravcsik holds, is that ‘foreign policies will … be motivated in part by an effort to realize social views about legitimate borders, political institutions, and modes of socioeconomic regulation’ (Moravcsik, Reference Moravcsik1997: 525). Entailed in this claim that ideas matter, he argues, is his first microfoundational assumption: that rational and risk-averse individuals and private groups will pursue the realization of their respective preferences. Yet, while it may be true that ideational liberalism entails the general claim that human beings rationally pursue their particular vision of a legitimate form of political and economic organization, there is nothing distinctively liberal about this claim. Indeed, a host of other approaches would readily agree. Rather, what makes a liberal theory liberal is the provision of a rational argument to the effect that legitimacy resides only, or to a greater part, in the particularly liberal forms of political and economic organization. And this particularly liberal claim cannot be derived from the general assumption.

The same problem arises in Moravcsik’s discussion of commercial liberalism. Commercial liberalism, he argues, cannot simply be equated with free trade policies because the market creates incentives for openness and closure; costs and benefits are unequally distributed leading different sections of society to favour different – protectionist and free trade – policies at different times. Moreover, while ‘trade is generally a less costly means of accumulating wealth than war, sanctions, or other coercive means … governments sometimes have an incentive to employ coercive means to create and control international markets’ (Moravcsik, Reference Moravcsik1997: 530). Again, Moravcsik identifies an assumption at the core of this commercial liberal theory which can also be found in his microfoundational claims: namely that commercial actors engage in cost-benefit-analysis of transborder economic interaction, which in turn feeds into the formulation of state preferences. Just as before, however, the commercial liberal theory cannot be derived from this claim. On the contrary, the idea that economic actors engage in cost-benefit-analysis is hardly distinctive for liberalism. In order to make a theory of international commercial relations liberal, it would have to make explicit how a liberal calculation of costs and benefits differs from other such calculations.

Similarly, it is hard to think of an approach that would, in principle, disagree with the claim of republican liberalism that ‘the mode of domestic political representation … determines whose social preferences are institutionally privileged’ and that government policy is ‘biased in favor of the governing coalition or powerful domestic groups’ (Moravcsik, Reference Moravcsik1997: 530). As Long has rightly pointed out: ‘Simply suggesting that governments represent a section of society is hardly descriptive of liberal ideas on the role of government at all’ (1995: 498). It is only when Moravcsik cites the democratic peace thesis as a case in point that the specific liberal variant of this general claim becomes apparent. For while the latter is based on the assumption that the cost of war determines whether or not people – and governments – will support it, this calculation pays no attention to regime type. On the basis of this calculation, liberal and nonliberal states will undertake ‘cheap’ wars with other liberal and nonliberal states while abstaining from ‘costly’ wars across the board. Thus, the specifically liberal character of the democratic peace thesis lies not in a cost-benefit-analysis of war but rather in the claim that liberal regimes are less likely to go to war (at least with each other) (Moravcsik, Reference Moravcsik1997: 531, 532). And this claim cannot be derived from the general assumption that political actors undertake cost-benefit analyses.

These examples reveal Moravcsik’s general assumptions as theoretically sterile. In his attempt to derive a liberal theory from these assumptions, Moravcsik basically moves from the statement ‘people engage in cost-benefit analyses’ to ‘liberal people engage in liberal cost-benefit analyses’ – and here the fruitfulness of his assumptions comes to an abrupt halt without having specified a single characteristic of liberalism. Ideational, commercial, and republican liberal theory are thus not derived from Moravcsik’s microfoundational claims. Instead, the reversal of the proper order – from specific to general claims – suggests that the particular liberal claims of these theories (all of them ideological, according to his own judgement) are generally valid simply because they can be stripped off their specificity, leaving behind the empty shell of a general assumption. Yet, since this empty shell – of cost-benefit-analysis, for instance – does not logically imply the particular legitimacy of market democracies, their generally peaceful nature, or the general rationality of free trade, the chain of logic is broken. In short, inasmuch as Moravcsik’s microfoundational assumptions are generally valid, they are not liberal, while his specifically liberal claims are not generally valid.

Anticipating such criticism, however, Moravcsik concludes that his liberal theory may simply be called ‘ “societal”, “state-society”, “social purpose”, or “preference-based” theory’, distinguished from others by the core claim that it stresses ‘the pattern of state preferences’ (as opposed to ‘the distribution of resources, and the institutional provision of information’ in realism and institutionalism, respectively) as ‘the most fundamental type of IR theory’ (Moravcsik, Reference Moravcsik1997: 548, 549). In this ‘minimalist’ formulation, liberal theory’s distinctiveness lies in the core claim that domestic preferences ultimately determine international politics. Yet, while this claim may distinguish his theory from structural realism and institutionalism, it is, for example, shared by most strands of Marxism as well as classical political theory and traditional realism, as Waltz (Reference Waltz1979: 18–37, 60–65) famously pointed out.

Moreover, this claim itself ultimately derives from the assumption of rational and risk-averse individuals. These individuals thus constitute the fundamental and irreducible source of social and international relations leading to a strictly linear, or inside-out, theory moving from the level of the individual to that of society and thence to international relations. In this, Moravcsik’s theory is indeed distinct, at least from Marxism, because the latter conceives of the individual itself as the product of social (and hence also potentially international) relations. Consequently, social and international relations are here not conceived as the product of a linear development but stand in a dialectical relationship with each other and the individual. I will return below to the question of how far this conception of the individual as the irreducible source of social and international relations – and thus also the linear directionality of Moravcsik’s theory – finds empirical support. Here it suffices to conclude that the ‘minimalist’ conception of Moravcsik’s theory as simply stressing societal preferences for the constitution of international politics lacks distinction from a range of other theories.

Taking the particular liberal approaches as a starting point, however, undermines the general validity of the theory, as I will now show. Moravcsik attempts to demonstrate the general validity of his theory by providing evidence from different historical periods, involving different actors and different issue areas as well as different levels of analysis (individual states and international relations) (Moravcsik, Reference Moravcsik1997: 515).

The determination of general validity depends, of course, crucially on the criteria by which historical data are included and excluded. In order to avoid citing outdated sources, Moravcsik applies the acid test of the continuing relevance of historical data in contemporary international affairs. Thus, he includes Adam Smith because this author recognized limitations to free trade – just as contemporary liberals do – and excludes laissez-faire policies on the grounds that they are clearly outdated (1997: 527, 529). This method raises two problems, however. First, it relies on a quite subjective interpretation of the relevant data as well as contemporary international affairs. After all, Adam Smith is also frequently cited in support of laissez-faire policies and, indeed, both interpretations can be substantiated (Wyatt-Walter, Reference Wyatt-Walter1996). In addition, it may equally be argued that recent neoliberal policies of liberalization, privatization, and deregulation prove that laissez-faire policies are and have been an integral part of liberalism.

The second problem with ‘contemporary relevance’ as a criterion for selection is that, ideally, it requires a definition of ‘contemporary’. Such a definition is, however, sorely missing. Moreover, what counts as ‘contemporary’ can only be defined with reference to the ‘historical’ (or outdated). This being the case, a focus on ‘contemporary’ relevance does not save the scholar the labour of engaging with historical data – rather, it presupposes such engagement. In sum, the failure to define ‘contemporary’ leads to a situation in which historical evidence is treated as a conglomerate of more or less useful data, examples, arguments and events relegated to a ‘take it or leave it’ status in accordance with the author’s preconceived notions of liberalism.

Next, Moravcsik claims that his theory is generally valid across issue areas: it fruitfully links a broad range of previously unconnected theories – in this case the ideational, commercial, and republican liberal theories or the realm of ideas, economics, and politics: ‘economic development has a strong influence on the viability of democratic governance, with its pacific implications; liberal democratic governments tend in turn to support commerce’ which can ‘lead to transnational communication and … promote secularizing cognitive and ideological change’ (Moravcsik, Reference Moravcsik1997: 534). And yet, what unites these approaches under the heading of liberalism are not Moravcsik’s microfoundational claims but rather, as he himself notes, the fact that their rich particular claims are mutually constitutive (1997: 533).

Finally, Moravcsik attempts to show that his theory is valid for liberal and nonliberal actors alike. Thus, cooperation amongst the states of the Holy Alliance and their conflict with Republican France indicates that both, liberal and nonliberal actors, tend to cooperate with other like-minded states while ending up in conflictual situations with those whose institutions they find less legitimate (1997: 527). This example supports the claim that states pursue the realization of their respective socio-political norms and values; that is, it supports the general microfoundational assumption. What it does not support, however, is the general validity of the specifically liberal claims of this theory. The latter clearly holds that some forms of political institutions (democracy) or social organization (market economies) are more legitimate than others – a claim which the Holy Alliance obviously did not embrace. Here, again, it is the missing logical link between the general and the specifically liberal claims that undermines the evidence: general evidence appears to support the general claims, and evidence taken from liberal spaces the liberal claims.

This reduction of evidence to the intraliberal sphere and, indeed, to its contemporary achievements only, finally, leaves Moravcsik’s conclusion completely unsubstantiated. He notes that ‘among advanced industrial democracies, a stable form of interstate politics has emerged, grounded in reliable expectations of peaceful change, domestic rule of law, stable international institutions, and intensive societal interaction’ (1997: 535). This evidence suggests that ‘the most powerful influences in world politics today are … the transformation of domestic and transnational social values, interests, and institutions’ (1997: 547). The suggestion is that the transformation of traditional (nonliberal) social values, interests, and institutions into liberal ones has led to a new form of international politics characterized by peace and cooperation. Yet, the evidence provided refers neither to the process of transformation which the now liberal states have undergone, nor to international relations at large. Instead of referring to the historical process of transformation, the evidence simply takes stock of the outcome; and instead of providing any evidence on the effects of this transformation on international relations in general – including those between liberal and nonliberal states – it focuses entirely on intraliberal relations.

In sum, Moravcsik’s empirical liberal theory of international relations does not live up to his own requirements for a proper social scientific theory. At its core, such a theory has to be derived from a set of core assumptions which establishes its logical coherence. Yet, while Moravcsik formulates core assumptions, he does not derive his theory from them. Instead, he reads the particular, rich, and varied claims of traditional liberal approaches back into his general assumptions. This disjuncture between the general and the specifically liberal claims of his theory subsequently undermines his attempts to establish its distinctness as well as its general validity. The general assumptions of this theory can be substantiated, but they are not distinctively liberal, while the specifically liberal claims find no general support. Moravcsik’s conclusion embodies these problems most clearly. While he claims to provide a general theory of the transformative forces of international relations, the evidence underpinning this claim is restricted to intraliberal relations – that is, particular and not general practices – and the achievements or outcomes of these relations – that is, static and not transformative or processual evidence. In short, where Moravcsik’s empirical liberal theory of international relations is general, it is not liberal; and where it is liberal, it is not general.

Liberal theory as political ideology

Moravcsik’s liberal theory of international relations thus does not satisfy the criteria of his own methodological conception of a nonideological theory. Yet, ideology, as set out in the beginning of this article, does not just signify a methodological problem – it is, above all, a politico-historical phenomenon. That is, ideology is not just bad science, it represents a particular world view justifying and propagating certain policies. It is in this sense, I will now show, that Moravcsik’s liberal theory of international relations can indeed be described as ‘deeply ideological’.

History plays a crucial, and threefold, role in the constitution of ideologies. First, ideologies are themselves historical phenomena – that is, they have historical origins and undergo historical transformations. Secondly, they present and propagate certain interpretations of history. That is, their promises regarding the future are based on a particular reading of the past. Hence, thirdly, historical data provide the crucial evidence on which ideologies rest. Moravcsik’s liberal theory of international relations is no exception. It takes up historical data and uses them to construct a picture of the historical development and achievements of liberalism. In a first step, I will reconstruct this substantive history embedded in Moravcsik’s liberal theory of international relations. In a second step, I will systematically add significant historical data excluded by Moravcsik resulting in a more encompassing historical narrative. The latter, I will show, reveals Moravcsik’s theory as a prime example of a political ideology – and not just a methodologically flawed piece of scholarly writing.

The most basic assumption of Moravcsik’s theory is that rational and risk-averse individuals and private groups are the fundamental actors in world affairs. This rationality, as we have seen, is identified with market democracies as the most legitimate form of socio-political organization, and with the foreign policies of free trade and pacific intraliberal relations. The ideational, economic, and political elements of this theory are linked and mutually constitutive. Thus, market economies produce wealth which in turn underpins liberal or democratic governance leading to peaceful foreign policies. These provide further support for commerce and economic development which, through transnational communication, promotes secularism and ideological change (Moravcsik, Reference Moravcsik1997: 534).

A historical narrative then supports these assumptions: ‘Global economic development over the past five hundred years has been closely related to greater per capita wealth, democratization, education systems that reinforce new collective identities, and greater incentives for transborder economic transactions. … Hence over the modern period the principles of international order have been decreasingly linked to dynastic legitimacy and increasingly tied to factors directly drawn from the three variants of liberal theory: national self-determination and social citizenship, the increasing complexity of economic integration, and liberal democratic governance’ (Moravcsik, Reference Moravcsik1997: 535). This historical development has led to a stable form of interstate policies amongst advanced industrial democracies, ‘grounded in reliable expectations of peaceful change, domestic rule of law, stable international institutions, and intensive societal interaction’. In short, ‘liberal theory argues that the emergence of a large and expanding bloc of pacific, interdependent, normatively satisfied states has been a precondition for such politics’ (Moravcsik, Reference Moravcsik1997: 535).

The substantive picture which emerges is thus one of linear historical development from the initial recognition of the rationality of market economy and government by consent through their progressive realization in domestic settings, to their gradual change of the nature and principles of international politics. And in those areas in which the liberal principles have been most fully realized, they have led to peace, prosperity, and cooperation in international affairs. This historical narrative, then, has political implications. If the endorsement of liberal values and the realization of liberal institutions in the past has led to such desirable outcomes as peace, prosperity, and cooperation – and if broader sections of the international system are to benefit from these achievements – then ‘the most powerful influences in world politics today are … the transformation of domestic and transnational social values, interests, and institutions’ (Moravcsik, Reference Moravcsik1997: 547). In other words, the achievement of these aims for broader sections of the international system requires the transformation of traditional nonliberal values, interests, and institutions.

This narrative rests, however, on quite selective evidence. I will, therefore, now systematically add crucial and previously excluded data to the story. My aim here is not to replace one historical narrative with another but rather, in keeping with the meaning of ideology not as ‘untrue’ but rather as ‘true’ from a particular point of view, to add previously excluded dimensions and perspectives to the story. And these, I will argue, suggest a different conclusion.

In Moravcsik’s narrative, the rational individual provides the starting point for the historical development, first, of economic wealth, which in turn provides the basis for the establishment of liberal societies, and subsequently of a ‘liberal’ form of international relations between such societies. It thus requires substantiation from the historical period prior to the emergence of liberalism, that is, from the period of transition from nonliberal to liberal society. In order to illuminate precisely this period, I will use the work of John Locke in which ‘the central elements of the liberal outlook crystallized for the first time into a coherent intellectual tradition expressed in a powerful, if often divided and conflictual, political movement’ (Gray, Reference Gray1986: 11). Locke therefore provides a perfect bridge between the initial development of liberal ideas (MacPherson, Reference MacPherson1962: 262; Rapaczynski, Reference Rapaczynski1987: 14; Ward, Reference Ward2006: 691) and the eventual constitution of liberalism as a self-conscious theoretical and political position in the nineteenth century; in short, Locke provides insights into precisely that period of transition addressed in Moravcsik’s first assumption.

Like Moravcsik, Locke argues that human beings are rational and that this rationality is ultimately embodied in government by consent and the protection of private property. This is so because the state of nature of all men is ‘a State of perfect Freedom to order their Actions, and dispose of their Possessions, and Persons as they think fit’ (Locke, Reference Locke1994: 269). Yet, in order to uphold this freedom, the individual has to preserve the self. And thus, Locke argues that ‘every Man has a Property in his own Person … the Labour of his Body and the Work of his Hands … are properly his. Whatsoever … he hath mixed his Labour with … thereby makes it his Property’ (Locke, Reference Locke1994: 287f). The perfect freedom of the individual, then, is based on self-possession, property, and property in the fruit of one’s labour. Political government in the state of nature reflects this freedom. ‘Men are naturally free, and the Examples of History shewing, that the Governments of the World … had their beginning laid on that foundation, and were made by the Consent of the people’ (Locke, Reference Locke1994: 336). And since this freedom is based on property, the ‘great and chief end therefore (of government) is the Preservation of their Property’ (Locke, Reference Locke1994: 351).

While the interpretation of these core assumptions is contested, largely because Locke uses the term ‘property’ sometimes in the sense of material property alone and sometimes including life and liberty, it is the material sense of property which is pertinent to Moravcsik’s argument that economic wealth provided the basis for the constitution of liberal societies. Locke establishes the original right to private property already in the state of nature in chapter V of the Second Treatise, and in that context, he uses the concept consistently with reference to material property alone (Arneil, Reference Arneil1996: 133; Armitage, Reference Armitage2004: 604). And it was this material conception of property which was subsequently widely taken up to justify appropriation of common land in the domestic and international sphere for the explicit purpose of wealth creation that underpins the development of liberalism in Moravcsik’s argument. Locke, thus, does not just concur with Moravcsik’s claim that market democracies – based on private property – embody rationality, but also with the claim that government by consent and the protection of private property are mutually constitutive. This latter point, however, contradicts Moravcsik’s claim that it is his own theory that fruitfully links hitherto unconnected liberal approaches. Instead, these connections lay at the heart of early liberal, or protoliberal, thought.

Yet, while Locke’s foundational claim broadly supports Moravcsik’s, the fact is that Locke had considerable difficulties to substantiate this claim, which throws a new light on this theory. After all, it was hardly the case that all, or even most, human beings possessed themselves or the fruits of their labour, and history was replete with examples of authoritarian government. Thus, while Locke held in theory that all human beings were rational, in practice he thought it ‘evident, that there is a difference in degrees in men’s understandings, apprehensions, and reasoning’s, to so great a latitude, that one may, without doing injury to mankind, affirm, that there is a greater difference between some men and others in this respect, than between some men and some beasts’ (Locke, Reference Locke1959, II: 446). This gap between the claim that all human beings were rational and the historical fact that most human beings lacked sufficient (or the right kind of) rationality to endorse the liberal form of socio-political organization, meant that an extension of the franchise would not lead to the establishment of societies based on private property and government by consent.

Locke’s account thus attests to the fact that rational individuals of the kind presupposed in Moravcsik’s theory were historically in extremely short supply and contra Moravcsik not generally available as a trigger for the development of liberalism. Irrespective of the precise date one might choose for the emergence of liberalism, in the eighteenth century ‘political ideas and attitudes still looked largely to the past, and were still in the main rooted in religion’ (Anderson, Reference Anderson1961: 4). How exactly these ‘traditional’ attitudes came to be turned into ‘liberal’ ones is thus crucial for a historical narrative of the emergence and development of liberalism. And while Moravcsik, mistakenly presupposing the general existence of such rational individuals, has no need to address this problem, recognizing it, Locke did. His work can thus once again serve to add an important element to the historical narrative.

In order to uphold his general claims, Locke had to show how society could be based on the principles of private property and government by consent in the absence of a majority of individuals supporting such developments or, conversely, how the majority of the population could be made sufficiently rational to establish and maintain such polities. If private property was the basis of individual freedom, Locke argued, property owners would demand that government protect private property and hence their freedom. He thus advocated the extension of full political rights to property owners and the establishment of paternal or despotical government over others. ‘Paternal Power is … where Minority makes the Child incapable to manage his property; Political where Men have Property in their own disposal; and Despotical over such as have no property at all’ (1994: 384).Footnote 3

In accordance with this argument, the extension of full political rights to wider sections of society simply required that more, and ideally all, people be turned into property owners. But where was this additional property to come from? Private property, argued Locke, was more productive than common property and thus of greater benefit to all of humankind (1994: 296–298). It was therefore justified to turn common into private property: God gave the land ‘to the use of the Industrious and Rational’ (Locke, Reference Locke1994: 291). People could simply attain property by mixing their individual labour with the original common property. The privatization of common property was thus the solution to the problem.

But there was simply not enough common land left in England to provide the vast and rising number of poor with property. Locke thus looked abroad: ‘Yet there are still great Tracts of Ground to be found, which (the Inhabitants thereof not having joyned with the rest of Mankind, in the consent of the Use of their common Money), lie waste, and are more than the People who dwell on it, do, or can make use of, and so still lie in common’ (Locke, Reference Locke1994: 299). It was this common land in America which could be used, at least in principle, to furnish all individuals with property and thus make them eligible to full political rights. In short, ‘Locke … was offering the New World, specifically the colonial settlements of America, as validation of his sociopolitical philosophy’ (Lebovics, Reference Lebovics1986: 577).Footnote 4 The lack of rational individuals who would and could establish liberal policies thus required, first, the denial of full political rights to individuals and communities who lived on the basis of common property in the domestic and international realm. Secondly, the privatization of such communal property would then widen the number of property owners in society and thus eventually enable an extension of the franchise.

Historical evidence shows that this mechanism did indeed come to play a crucial role for the establishment and development of liberalism. Locke’s theory was used to justify enclosures in England. On the one hand, land owners increasingly derived their right to political participation directly from property which led to a huge increase in the members in the House of Commons. On the other, Locke’s work was frequently cited in Parliament in support of private enclosure acts which, between 1710 and 1815, transferred 6.5 million acres or 20% of the total land from common into private property (McNally, Reference McNally1988: 8, 9, 62; Perelman, Reference Perelman2000: 175).Footnote 5

The very same practices, based on the very same arguments, were employed in the international sphere. ‘Preachers, legal theorists, and politicians’ used Locke’s theory of property to base first the land claims of the British colonists and then those of the American citizens on the enclosure and cultivation of land (Arneil, Reference Arneil1996: 169). The same argument was also influential in Australia, New Zealand, and Canada throughout the eighteenth and well into the nineteenth century (Ivison, Reference Ivison2003: 93). And, indeed, it was in settler societies like New Zealand, the USA, and Australia that the promise based on such wider distribution of property – the introduction of universal franchise – was first realized.

Moreover, these policies are of continuing relevance in the contemporary world – even if today’s commons do not any longer mainly consist in (other people’s) land. The latest round of ‘privatization’ and ‘liberalization’ targeted communal ownership of water and electricity supplies, education, health care, and the establishment of intellectual property rights over natural products and their uses, to name but a few – both in the domestic and the international sphere.

In Moravcsik’s historical narrative, it is global economic development which has led to greater per capita wealth, democratization, and education; in short, to the establishment of liberal polities. What this narrative obscures, however, is the fact that the initial economic development, precisely because the rational individuals he presupposes were generally not available, was based on the systematic political oppression and economic expropriation of the vast majority of the population, in the domestic and the international sphere. The implementation of these policies generated conflict and required the use of force (Jahn, Reference Jahn2007a, Reference Jahnb). The transformation of nonliberal into liberal societies was thus not the evolutionary process characterized by an extension of peace, prosperity, and cooperation as which it appears in Moravcsik’s narrative. Rather, it took the form of a political struggle.

And it was this struggle, the need to deal with resistance and opportunities, setbacks and advances, in a variety of settings and constellations which led to the development of different versions of liberalism: interventionist, noninterventionist, free market, Keynesian, democracy, human rights, order, and stability liberalism are the hallmarks of the history of liberal internationalism and continue to bedevil its performance and definition in the contemporary world, as more historically oriented authors point out (Smith, Reference Smith1992: 217f; Hoffmann, Reference Hoffmann1995: 174; Zacher and Matthew, Reference Zacher and Matthew1995: 111f; Doyle, Reference Doyle1997: 207f; Richardson, Reference Richardson1997; Richardson, Reference Richardson2001). Attention to this variety of liberalisms, and a contextual historical analysis of their conditions of emergence, rise, fall, and transformation would have alerted Moravcsik to the fact that liberalism did not just enter this world as a benevolent rational force which gradually conquered ground by the authority of example, but rather as a sectarian political position which had to fight its way to the top and adjusted its goals and means to the given circumstances. In short, ignoring the historical diversity of liberal thought and practice obscures the fact that ‘liberalism is above all a political project’ (Long, Reference Long1995: 502).

More systematic attention to the origins and development of liberalism, moreover, undermines a second core element of Moravcsik’s theory and historical narrative. Theoretically, Moravcsik moves in ascending order from the individual to the state to international relations; and this directionality is repeated in his historical narrative: rational individuals establish liberal polities, and these in turn practice a new form of international relations – characterized by peace and cooperation – amongst each other. And yet, it was communal property, above all in the international sphere, that played a necessary and constitutive role for economic development and the establishment of domestic liberalism. Without reference to American land, Locke could not have maintained his theory. And, in practice, economic development in Europe in general, as well as the industrial revolution in particular, were dependent on the colonies (Washbrook, Reference Washbrook1997; Marks, Reference Marks2007). Hence, the emergence of liberalism was by no means an endogenous process – thus undermining Moravcsik’s ‘minimalist’ conception of a strictly inside-out explanation of international relations. The nonliberal world is, consequently, also not the virgin territory awaiting valorization through the extension of liberal values, as which it appears in Moravcsik’s inside-out narrative. Instead, the contemporary nonliberal world is in part the product of systematic interaction with the liberal world over the past three centuries. This history of interaction, however, is systematically excluded from Moravcsik’s historical account.

Moravcsik’s criteria for the inclusion and exclusion of data – their empirical relevance in contemporary intraliberal relations – thus results not just in a partial (in the sense of incomplete), but in a systematically biased historical narrative with serious political implications. Presented as an evolutionary endogenous process, the establishment of liberalism in a number of countries – and the resultant generally peaceful and cooperative relations between these countries – indeed implies that the wider spread of these values and institutions will logically lead to an extension of these achievements. Yet, the fuller historical narrative presented above suggests otherwise. It shows, first of all, that the constitution of liberalism historically involved political struggle – thus questioning the possibility that a transformation of nonliberal values, principles, and institutions today will take the form of peaceful processes. Secondly, the development of liberalism historically required the political subordination and economic expropriation of communities based on alternative forms of sociopolitical organization – suggesting that the liberal gains made in the course of such transformation in some parts of the world or for some sections of society have to be paid for by the production of political and economic dependency in others. Thirdly, the historical evidence shows that the nonliberal world of today is already a product of systematic interaction with liberalism – clearly suggesting limits to liberalism’s transformative potential.

And yet, Moravcsik’s theory explicitly accounts for this dark and even violent side of the development of liberalism – in contrast to more ‘utopian’ positions (1997: 528, 530). Liberalism can nevertheless be presented as a worthwhile goal because its shortcomings are seen as historically contingent and its benefits, at least in principle, generalizable. In this vein, Moravcsik argues that protectionism, war, slavery, and colonialism can become rational (foreign) policies even for liberal states in situations characterized by a lack or imbalance of liberal developments – such as monopolistic features of the domestic or international economy, lack of economic interdependence, or incomplete democratization (1997: 529, 530, 532). Hence, the geographical expansion and deeper social penetration of liberal values and institutions will gradually pave the way for a generalization of liberal achievements. Yet, this hypothesis does not solve the core problem which Locke had encountered centuries ago. The very fact that liberal values and institutions are unevenly realized indicates that they are not universally endorsed. Hence, the further expansion of liberalism is bound to encounter barriers. The analysis of Locke’s solution to this problem suggests, however, that attempts to transform nonliberal into liberal groups require the provision, or at least promise, of property that will turn their members quite literally into stakeholders of a liberal society. This property, in turn, has to be appropriated from other groups who thereby lose their prior rights. The logic of expanding liberalism thus entails the production of nonliberal constituencies and undermines the possibility of universalizing the liberal achievements.

Liberal theory thus posits logical limits to the generalization of liberal achievements. Liberal history, however, may provide a practical solution: after all, the partial use of force has been successful in establishing and expanding liberalism in the past and may thus continue to provide the means for its universalization. Yet, while this course of action may well provide a certain level of prosperity, the forceful imposition of liberalism can, at least in the short run, not be equated with the beneficial outcomes of cooperation and peace. Moreover, such policies constitute the exact opposite of Moravcsik’s ‘minimalist’ assumptions: instead of providing the ultimate source of social and international relations, individuals and their preferences in this case become objects of international power. While it is impossible to predict the outcome of such a scenario in practice, it is clear that it fundamentally contradicts Moravcsik’s liberal theory.

The possibility of continuing such policies finally raises the question whether the benefits of such a generalization of liberalism outweigh its price. This question is indeed political rather than theoretical and goes to the heart of the problem of ideology. The expansion of liberalism has so far benefitted some sections of (international) society who may thus judge its benefits highly; and it has disenfranchised others who may well not be willing to pay the price. Both these judgements are rational and legitimate within the context of their own position and thus exemplify the political, that is contested, nature of liberalism. Yet, in presenting the benefits that liberalism has historically bestowed only on some sections of society as generally available, Moravcsik, no doubt unconsciously, obscures the political nature of this struggle and propagates the particular interests of liberalism’s beneficiaries. In doing so, this theory is indeed deeply ideological in the political sense.

The roots of ideology in liberal theory

Moravcsik’s liberal theory of international relations, I have argued, fulfils all the criteria for an ideology. It is ideological on his own terms since it does not manage to establish and maintain logical coherence, theoretical distinction, or general validity. It is also ideological judged in terms of the broader understanding of ideology set out at the beginning of this paper. Here, an ideology is understood as the world view of a particular social group or age which functions to integrate and mobilize this group in its struggle for power (or its maintenance). To this end, ideologies present – for the most part unconsciously – their particular world views and political aims as serving a general interest.

The fate of Moravcsik’s theory, however, has potential implications for International Relations in general. For this theory is the product of a project explicitly designed to move beyond ideology – and hence attests to the fact that the author is unconscious of the mechanism that turns his theory into an ideology. In order to avoid such unintended consequences in International Relations more generally, it is thus necessary to clarify the reason for this ironic outcome. It lies, in the case of Moravcsik’s liberal theory of international relations, in a foreshortened understanding of the implications of ideology. Moravcsik’s methodological solution to the problem of ideology, I will now show, entails a substantive conception of politics which denies the essentially contested nature of this realm of theory and practice.

The history of ideology set out at the beginning of this paper suggests that national and international political life is defined by competing ideologies struggling for dominance. In this struggle, chance plays an important part, the constellation of forces changes continuously, new actors emerge and old ones change, diversify, contract, merge – in short, the political realm is characterized by constant flux rather than regular interaction between static entities (Mannheim, Reference Mannheim1960: 103). The observer is a participant in this struggle not merely because of his values or interests, but also because political problems present themselves to him in a particular manner, affecting his most basic categories, leading to ‘actual differences in styles of thought – differences that extend even into the realm of logic itself’ (Mannheim, Reference Mannheim1960: 104). Under these circumstances, political knowledge is by definition knowledge inseparable from interest and motivation; it can only ever be particular knowledge, limited in time and space.

It is this implication of ideology, the fact that it operates at the most general level of political thought, which Moravcsik overlooks. Moravcsik’s treatment of historical data provides a good example. Since his theory is supposed to be relevant for today, Moravcsik logically determines that only data which are still significant in, and for, a contemporary conception of liberalism will be taken into account. This logic, however, depends itself on a linear conception of history which implies that policies that appear irrelevant in the current period have been historically overtaken and thus do not have to be included in the study. A cyclical conception of history, by contrast, would assume that elements of liberalism that have been relevant in the past are likely to regain significance in the future. The logic of this conception of history would thus demand particular attention to data from the past. Moravcsik’s conception of ideology, as expressed in his methodological antidote, is clearly oblivious to the implications of ideology at this level of the conceptual framework, and thus also to his own ideological conception of history, which is almost single-handedly responsible for the evidentiary shortcomings of his theory. After all, this linear conception of history determines his methodological criteria which in turn lead to the systematic exclusion of all data that might suggest an alternative – whether cyclical, dialectical, or otherwise – with the result of a linear history of liberalism shorn of its political struggles, its ups and downs, its internal variations, and all relations with internal or external competitors. This history of liberalism thus obscures precisely what the concept of ideology reveals – that the political realm is defined by struggle. Once posited, the linear conception of history simply reproduces itself.

But it is not just Moravcsik’s understanding of history that implicitly denies the ideological nature of the political realm – so does the positivist method he employs more generally. Historically, positivism is one of the two main responses to the problem of ideology developed in the social sciences (the other being normative theory). It recognizes the particularity of ideological thought as problematic and counters it with the demand for general knowledge. In positivism, ‘nothing is regarded as “true” or “knowable” except what could be presented as universally valid and necessary’ (Mannheim, Reference Mannheim1960: 149). Yet, this solution entails two problems. First, it assumes that underneath the messy superficial layer of ideological struggles, there are to be found general laws of politics – thus denying the essentially, and not just superficially, contested nature of the political realm. Secondly, as Mannheim points out, the equation of ‘truth’ with ‘generality’ is false. There are undoubtedly truths which, by virtue of their particularity, ‘are accessible only to a certain personal disposition or to a definite orientation of interest of a certain group’ (Mannheim, Reference Mannheim1960: 149). Yet, having made that equation, positivism is left with only two choices: either it simply does not study the necessarily particular phenomena of the political sphere, or it presents such particular knowledge as general. The fate of the study of ideology in political science demonstrates the implications of this choice for an entire discipline: normative theorists do not study ideologies because of their particular and illogical nature, while positivists only take stock of their general social functions – with the result, as Freeden points out, that political science fails to study the arguably most ubiquitous and influential form of political thought.

Yet, since the time of Mannheim’s writing, positivism has developed; it recognizes in principle the temporal and spatial limitations of knowledge and, in response, tends to restrict the empirical scope of investigation. In this vein it may be argued that if only Moravcsik had restricted his claims to liberal individuals, societies, and their foreign policies, the empirical basis of his theory would be stronger. But such a conscious exclusion of liberalism’s relations with its competitors over time necessarily restricts the explanation of liberal behaviour to endogenous forces alone. It thus fails to recognize the essentially political nature of the phenomenon which lies precisely in the fact that the relations between these parties are crucial determinants of their respective development and behaviour.

The methodological solution to the recognition of temporal and spatial limits of knowledge thus does not lie in a restriction of the empirical scope of the study. Rather, it requires a methodology consciously focusing on the spatial and temporal boundaries of the object in question. Ironically, therefore, while positivist approaches tend to hang on to the general assumptions and limit the historical and spatial scope of their application, the recognition of the limitations of political knowledge requires exactly the opposite: namely the assumption of the particular and contested nature of political phenomena and thence a widening of the historical and spatial scope of investigation, so as to include the origins, development, and external relations of its subject matter.

The widespread use of positivism in International Relations thus does suggest the need for a conscious and critical reflection on the implications of ideology for the discipline as a whole. This does not mean that it is impossible to generate general knowledge or that positivist approaches are generally useless. Rather, it means either that knowledge generated by positivist approaches may not address the specifically political dimension of international affairs, or that their particular insights about international politics are presented as general – and thus turn into ideologies.

The challenge that ideology presents to International Relations as a political science lies thus in formulating knowledge appropriate to the essentially contested nature of the political realm. And neither positivism, nor normative theoryFootnote 6 meets this challenge since both aspire to ‘a standard that simply does not exist in any form of political thinking’ (Freeden, Reference Freeden1996: 551). A clear statement of this challenge, however, provides the basis for a solution. For taking the essentially contested nature of politics and its continuous changes seriously implies that particular knowledge is neither necessarily untrue, nor unimportant – for the observer develops the principles of organization that enable him to understand the world from his particular point of view (Mannheim, Reference Mannheim1960: 58). In other words, Moravcsik’s liberal theory of international relations contains some important truths – if not about international relations in general. By expunging politics – the particular and contested nature of liberalism – from his theory, Moravcsik quite literally fulfils Mannheim’s observation that where politics disappears ‘ “administration” takes its place’ (1960: 170). That is, Moravcsik’s theory systematically expresses and explains the world view fitting a dominant liberal power whose interest and need lies in the ‘administration’ of the world rather than in political struggle. This, surely, is important political knowledge for participants and observers of current world affairs.

What makes ideological knowledge ideological is, therefore, not its particularity – which it shares with all forms of political thought – but rather the fact that it hides this particularity. Nonideological knowledge thus requires an explicit engagement with, and exploration of, its limitations.

In practice, a nonideological conception of liberalism or liberal internationalism must therefore engage the two main characteristics of the political realm – struggle and change – through an historical investigation of liberalism’s competition with alternative political projects. It was, after all, a particular historical context that generated liberal or protoliberal ideas and political goals amongst the members of a small section of society. And it was precisely this historical, social, and political particularity that required competition with, and struggle against, alternative views and projects in the course of its realization. This struggle, in turn, generated ‘many disputes within liberalism concerning its critical faculties and co-optation by powerful interests’ (Long, Reference Long1995: 503). Tracing this struggle historically accounts for the internal variations of liberalism, for its external limitations, for the variety of means employed in this competition, for its achievements and for the fact that its ‘promises have not been kept’ (Long, Reference Long1995: 505). While such a nonideological alternative is not a liberal theory of international relations in general, it surely is a general theory of liberal international relations. And as such it is extremely useful in illuminating the dynamics at play in a liberal world order.