Anthropology,autonomy and the art of cultural revolution

Introduction

In his recently published book A Secular Age, Charles Taylor argues that the secularity of the ‘North Atlantic world’ consists not merely in the separation of church and state, or even in the falling-off of religious belief and practice in the general population, but most pertinently in a shift in the conditions of belief. Our society is one in which belief in God ‘is understood to be one option among others, and frequently not the easiest to embrace’ (Taylor 2007: 3). This new condition signifies a ‘titanic change in our western civilization’ (2007: 12), for unlike our predecessors we are aware that ‘there are a number of different construals’ (2007: 11), standpoints and theories of reality or of the moral/spiritual life. For Taylor, this ‘index of doubt’ makes us different not only from those living in earlier ages of our culture but from ‘anything else in human history: that is, with almost all contemporary societies (e.g. Islamic countries, India, Africa) on one hand; and with the rest of human history, Atlantic or otherwise, on the other’ (2007: 1). In brief, for those of us raised in the West, naïve living out of (a construal of) reality has been eroded: we now must live reflectively, aware that our theories and experiences are both personal and provisional. ¹

A range of criticism has noted the continuing propensity of much theorizing about the Occident, a la Taylor, to imagine their negative or affirmative analyses of Western modernity via programmatic differences with the non-West. ² The work of Louis Dumont is a case in point, with his stark contrast between the individualism of Western societies and the holism of traditional societies, most clearly identified in India. Yet how widespread in social theory is this particular synonymy between reflection and the modern West, and naivety and the non-Occident? Have any other thinkers made it an integral feature of their social theory?

In this paper I examine the work of Cornelius Castoriadis, whose distinction between autonomy and heteronomy as essentially contrastive ways that different societies inhabit the world can be seen as analogous with Taylor’s formulation of reflectivity and naivety, and also with elements of Dumont’s model. Although Dumont asserts that individualism is the highest value in the modern complex of ideas, autonomy is a key (if subordinate) attribute of that ideology. The person in modern Western society is an ‘independent, autonomous, and thus (essentially) nonsocial moral being’ (Dumont 1986: 25). Castoriadis too makes a claim for the uniqueness of Western modernity and, despite radical contrasts amongst them, for the sameness of all non-Western societies, at least in comparison to this defining aspect of the Occident. Castoriadis and Taylor share other similarities as well. Taylor’s examination of the phenomenology of secularism and of modern [Western] subjectivity assumes that the conditions of belief in Muslim societies preclude their members from taking any disengaged standpoint on their own immediate realities. Castoriadis denies that Muslim societies allow their members any autonomy vis-à-vis the Islamic social institution of the world, particularly in relation to its religious law (Houston 2004).

More significantly both Taylor and Castoriadis develop the idea of the ‘social imaginary’ as a fundamental aspect of their theorizing. Despite important differences in their definition of the term – Castoriadis calls ‘social imaginary significations’ what Taylor names ‘social imaginaries’ – both agree that the significations or imaginaries that inform Western modernity constitute a radical break with pre-modern and non-Western social imaginaries. They agree as well that a central reality of modern society is ‘constructivism’: that is, to cite Taylor, that the contemporary West knows that it can ‘remake [its] political life according to agreed-upon principles’ (Taylor 2002: 110).

Taylor’s work, like Dumont’s, has been subjected to vigorous criticism for its Euro-centrism. ³ Castoriadis’ social theory, on the other hand, is probably less well-known, and accordingly less commonly engaged with. Rather than simply criticizing Castoriadis’ views for their Greco-centrism, however, in this paper I argue that his work can be used to illuminate both the emergence of autonomy in non-Western modern societies and to identify processes of autonomy in a variety of other historical contexts apart from the West. If this is so, then some of the problems associated with assumptions of Occidental exceptionality or with arguments for the unique emergence of secularity and modernity in Europe and for their consequent unfolding in the rest of the world (through colonialism) could be redirected.

To make my argument I use a case-study, the attempted formation of a new society in Turkey post-1923 by the Turkish Kemalists, to review Castoriadis’ theorizing about the social-historical world and, more narrowly, about what Arnason has called his distinctive model of modernity (Arnason 1989: 323). Castoriadis’ concept of modernity and his elucidation of what he calls its dual institution, composed of ‘two core [and antagonistic] imaginary significations of autonomy and “rational mastery”’ (1997a: 42), casts an illuminating light on one key episode in that enterprise, the 1930s making of the Kemalist cultural revolution. At the same time, that revolution’s lucid creation of a whole range of new cultural practices polemically related to the Islamic institutions of the Ottoman Empire casts into doubt Castoriadis’ reservation of the project of autonomy (or of constructivism and the experience of reflexivity) to the West. The explicit endeavour of the Turkish Kemalists to put into question established institutions and representations in the positing of new social practices challenges Castoriadis’ (and Taylor’s) assumption that it is only the West that is able to reflect upon whether its laws are just, and to change them in accordance with its answer. Indeed, the Kemalist project to lucidly institute [a new] society forces us to re-assess in fundamental ways Castoriadis’ arguments about autonomy and heteronomy and how they are related. In brief, it is the partial nature and ambiguous outcome of that Kemalist instituting that makes us question both Castoriadis’ high valuation of autonomy and his surgical distinction between autonomous and heteronomous societies.

Castoriadis and the theorizing of autonomy

Although Castoriadis as far as I know has never written on the Kemalist project – despite having been born in Istanbul and leaving there as a child in its wake – his polemical but always suggestive arguments about what he calls the ‘social-historical domain’ of human being lend themselves well to the elucidation of the cultural revolution in Turkey. For the sake of the argument let me first run briefly through the main lines of his conceptual framework. One difference between Castoriadis’ work and the more usual discussions of modernization that focus on processes of commodification, bureaucratic or scientific rationalization, colonialism, etc., is that Castoriadis is interested in the first instance in more foundational questions of the social ontology of society per se. ⁴ For this reason Castoriadis’ political philosophy has an anthropological ‘feel’ about it, in his emphasis on the self-institution of each society through its positing of what he calls its own world of ‘imaginary significations’. ‘Imaginary significations’ are described by Castoriadis as the forms, ideas and images through which society institutes itself: ‘every society defines and develops an image of the natural world, of the universe in which it lives … in which a place has to be made not only for the natural objects and beings important for the life of the collectivity, but also for the collectivity itself, establishing finally a certain world-order’ (Castoriadis 1987: 149). This stress on the specificity of different societies as well as on their totality or cognitive/informational closure shares some of the vision of Durkheimian anthropology.

Although far more precise, the term the ‘imaginary institution of society’ also bears an affinity with usage in anthropology of the concept ‘life-world’. I am thinking in particular of Mimica’s application of the term in a recent article on epistemological issues that confront the discipline. Mimica includes within the life-world a vast range of elements, including modes of experience, action and knowledge (intellection and ratiocination), existential-cognitive-affective forms and practices, intersubjective fields within which people live, a society’s ‘cosmo-ontology’, as well as people’s practical skills and practical (action-executed) deliberations (Mimica 2010). Compare this with Castoriadis’ comments on the institution of society:

[it] is always, also, nonconsciously, general and special ontology. It posits, it always has to posit, what each particular thing, every relation and every assemblage of things, is, as well as what ‘contains’ and renders possible the totality of relations and assemblages – the world. The determination, by each society, of what everything is is, ipso facto, donation of meaning to each thing and insertion of each thing into meaningful relations; it is, each time, creation of a world correlative to social imaginary institutions and dependent upon these significations. (Castoriadis 1997d: 317)

Citing Castoriadis, Mimica notes that the lives of the Yagwoia in New Guinea are increasingly encroached upon by Western capitalist civilization and its ‘ontological imaginary significations’ (although he declines to incorporate that influence in his ethnographic analysis).

The closed institution of society also means Castoriadis’ formulation shares some of the problems of the classical anthropological assumption of a globe made up of radically different and separate societies. His formulation downplays the possibility that society’s institution of a world for itself through its core imaginary significations is made in knowing relation to other societies (via trade, mimicry, enmity, amity, appropriation, transformation, development, etc.). Nor does he allow that other societies might be constituted through an entwining of antagonistic internal clusters of significations, as he posits is the case of the modern West (see above). Both of these denials disqualify the possibility that groups might knowingly fabricate themselves in and through the imaginary significations of other societies, or through their mediation of antagonistic visions and practices that constitute their own social world.

One of Castoriadis’ main concerns is the source of social creation and of the individual’s relationship to it. For Castoriadis the source of the social-historical realm is what he calls the radical imaginary, as it creates the institution of society via imaginary significations that create a world. Society institutes itself, and the instituting ‘power’ is the radical imaginary. Further, the order of any instituted society and its imaginary significations are essentially ‘arbitrary’. That is, although any instituted society ‘leans’ on the natural world, it alters nature in accordance with its own world of significations. (Not every aspect of ‘natural reality’ lends itself to transformation.) For Castoriadis the social-historical domain is characterized not by causality and determinism but by indeterminism and creation.

Because the source of social creation is the instituting imaginary or the ‘instituting society’, Castoriadis is unsympathetic to models of social or cultural creation that situate its source in the work or person of the ‘sovereign individual’. As he says, ‘the institution produces, in conformity with its norms, individuals that by construction are not only able but bound to reproduce the institution’ (Castoriadis 1997b: 7). Individuals too are social institutions, creations of an already instituted society. This is also the basis of Castoriadis’ rejection of intersubjectivity as the source of new experiences of meaning or of social creation. The ‘individuality’ of the subject is of course produced relationally or intersubjectively in her existence with others from birth, but this encounter itself occurs in the ‘torrent of meanings’ posited by the social institution. As Castoriadis says, ‘it is only through the world that one can think the world’ (1987: 106). As incarnated fragments of society, Castoriadis argues that the instituted social formation does not allow its members’ thoughtful consciousness of its self-creation. Similar to the power of the unacknowledged background framework in non-Western contexts for Taylor, the instituted society makes it impossible for the subjects it fabricates to perceive or question how they have been generated as particular social beings. ‘A true tradition is not discussed’, Castoriadis asserts (1990: 126). In particular, religion obscures awareness of society’s own self-institution. Thus in itself, or unexceptionally, the institution of society posits for its members extra-social sources for its own existence and form, ‘to exclude the idea that it might be self-institution’ (Castoriadis 1997c: 213).

Castoriadis explicitly denies that the imaginary social significations of any particular society are ‘ideological’, in the sense of a set of ideas or meanings formulated by a class, party or institution in order to veil reality. In the essay ‘Power, Politics, Autonomy’ (1991) he makes a distinction between the ‘ground power’ of the instituting imaginary and its instauration and organizing of signification and society itself, and the instituted society’s political sphere or sphere of governance, which is also the sphere of ideological discourse. For me, Castoriadis’ distinction here is too absolute: a more empirical anthropological analysis would seek to ascertain whether and in what ways even the core imaginary significations of society are partially conceivable and contested, i.e. made ideological. Nevertheless, even unexceptionally the always-present threats to ‘social heteronomy’ sketched out by Castoriadis do allow for a less severe dichotomy between religious or non-modern societies and the modern West. These threats include time or the future itself; the capacity of the psyche to thwart the schooling upon it; a certain individual reflectiveness; the crisis of instituted meaning (or the potential a-meaning of the world); wars and famine; the threat of other societies; ‘explicit power’, or the realm of ‘the political’; and most importantly, the continuing flow of the radical imaginary/instituting imaginary itself (cf. Castoriadis 1991). The radical imaginary never ceases to pull apart and put together in new ways the inherited connections and separations of previous societies. In Castoriadis’ essay ‘Institution of Society and Religion’ (1997d), the radical imagination of the ‘psyche’ is also presented as incompletely socialized by the social institution. These creative processes act to disrupt the functioning of the institution, as well as to ensure history (change and development) even in ‘primitive’ societies.

On the other hand, Castoriadis does of course consider the social exception and the history of its emergence. As with Taylor and Dumont, the exceptional for Castoriadis today is the Western world (we might also call it modernity), or what he describes as the ‘Greco-Western’ institution of society. Here in ancient Greece and then again in Western Europe, beginning in the 13th century and continuing into the present, there is a rupture in the ‘heteronomous’ institution of society, a challenge to the ‘self-occultation of society, [and to] the misrecognition by society of its own being as creation and creativity’ (Castoriadis 1997d: 327). Dumont’s conviction of the fundamental opposition between the modern West and pre-modernity is similar, even if he traces the origins of ‘individualism’ to early Christianity and religion, not to Greece and philosophy. In the modern West, Castoriadis finds uniquely ‘autonomous’ societies that explicitly recognize themselves as the origin of their own imaginary institutions, whereas for Dumont the eccentricity of modern society is discerned against the normative hierarchy of ‘the common type of holistic society’ (Dumont 1986: 25). In keeping with his understanding of the fabrication of the individual by the social institution, only autonomous societies institute autonomous individuals, subjects capable of acting ‘deliberately and explicitly in order to modify [their] law’ (Castoriadis 1997e: 340).

Summarized like this, far from an inclusive account of modernity, Castoriadis’ work might appear as the very exemplar of a self-regarding Euro-centrism or Occidentalism. Yet if it were possible to identify his key processes of lucid self-institution and cultural creation in non-Western modern societies too, then autonomy (or reflectivity and individualism) might be re-conceptualized as extra-Western as well, even while analysis recognizes the culturally and historically specific forms of those societies. As we have seen, for Castoriadis the West or modernity is not only characterized (like any society) by the imaginary significations through which it exists as a society per se, but most significantly by its ability to explicitly put those imaginary significations into question and to create new ones: that is, by autonomy.

Cultural revolution in Turkey

Is autonomy more than an ‘imaginary signification’ of Greco-Western society? Castoriadis writes that ‘we have here a being – the society of the Greek polis, certain European societies – that explicitly puts back into question and challenges the law of its own existence’ (1997e: 339–40). In other words, autonomy is both an imaginary signification of Greece that constitutes it as the Greek institution of society (and not, say, as the Persian one), and a capacity to act to knowingly alter society. Most importantly, is autonomy in its second form discoverable by other societies, in the exercise of their own explicit self-institution?

What makes any society – say, Kemalist Turkey – autonomous, then, is its history of deliberate self-creation, its capacity to posit, in its cultural revolution for example, its own forms for itself. What makes any society different however – again, say, Republican Turkey from Republican France – is its own particular instituted cluster of imaginary significations, open as they are to lucid modification. Autonomy or the capacity to posit and revoke its own laws and institutions characterizes modernity, allowing us to speak of its singularity. The varied imaginary significations that organize the human and non-human worlds in different societies make modernity plural, allowing us to speak of its multiplicity.

Castoriadis’ ideas can be appropriated to intervene in broader debates about multiple modernities ⁵ by changing their focus from philosophy to comparative sociology; that is, by opening them up to an anthropological enterprise that analyses different societies’ actual experiences and processes of conscious self-creation. This would also be one suggestive way of reconfiguring the discipline of anthropology. Alternatively, new perspectives on Europe can be achieved by exploring how selected features of Western modernity resemble versions of social relations or institutions also present in other societies. One example is found in Joanna Overing’s (1993) discussion of the freedom of the Piaroa Indians over their own labour, in comparison with Marx’s critique of the alienation of the proletariat in capitalism. Here socialism’s desired goal is given illuminating content via the Amerindian imaginary signification of work. In the first case, the explicit self-institution of any society becomes an ethnographic and empirical question, regardless of Castoriadis’ pronouncement upon its impossibility. In turn, this comparative analysis of processes of autonomy would help us to reassess some of the key categories of his work. Let us attempt to do exactly this by taking as a case study the 1930s cultural revolution in Turkey made by the Kemalists, a revolution whose social categories – although this is not the main topic of the paper – still operate to foster and curtail cultural critique today.

Many readers will be familiar with the general history of the emergence of ‘modern’ Turkey after the First World War from the ruins of the Ottoman Empire, as well as with some of the governmental strategies of the new Kemalist state in Ankara to constitute and mobilize a republican society. Similarly ambitious projects of nation-building emanated from the modernizing capital of Tehran at the same time, accompanied there too by varying degrees of hostility and intolerance towards the imaginary significations of predecessor societies. In this endeavour the Turkish republicans were the most radical in the sweep of their state activism, both in their reforms in the ‘juridico-political and institutional realms’ (Kandiyoti 1997: 113) and in their making of a cultural revolution.

The political project of the new republican nation-state after 1923 has been described in different ways, including by the unsatisfying appellations of ‘secularism’, ‘modernization’ and ‘Westernization’ (see Lewis 1961; Berkes 1964; Ahmad 1993). Occasionally it has been graced with the title ‘Turkish Enlightenment’, while often being presented as a ‘model’ for other Muslim-majority countries to emulate. None of these terms come to grips with at least three of its defining elements: the active economic and cultural relations between the Kemalist state and the Soviet Communist Party in the Republic’s first two decades [of the 1920s and 1930s]; the broader nationalist context, particularly in Europe and the Balkans, to which it was dynamically related; and the genealogy of Ottoman reflection on the adequacy of their own social institutions. The third point attests to the Ottoman’s self-conscious history of institutional reform inherited by the republicans. Influenced by all of these, Kemalism fostered the creation of new ‘social-historical forms’, investing selective significations of Ottoman society with new meanings. In Castoriadis’ words, ‘the old enters the new with the signification given to it by the new and could not enter it otherwise’ (1997b: 14).

Cultural revolution involved the Kemalists’ explicit creation and formalization of a huge range of social practices, including a new alphabet and vocabulary, new musical forms, festivals and folklore, dance genres and theatrical performance (including radio theatre), to name just a few. All of these innovations can be usefully reframed as the Republican People Party’s initiatives to sponsor new forms of embodied skills and knowledge. Since the 1990s much historical research on Kemalism’s wide-ranging educational initiatives in its foundational period has analysed the thoroughness of the attempt to explicitly fabricate a new society (e.g. Salmoni 2003, 2004). This reforming zeal was not limited to the Kemalist pedagogy dominant in schools. As in Iran, Kemalist campaigns in the areas of maternal and child health linked change in such practices to the development of a rational and ordered modernity, intimately connected to newly posited Turkish national values.

Let me proceed by describing the Kemalist revolution in Turkey through a brief case study, its initiatives in the area of music. Music reform was of prime concern for the Kemalist revolutionaries. The recording, notating and nationalizing of peasant music was one aspect of an even more fundamental project, the making of a music revolution (musiki inkılabı). According to Tekelioğlu, ‘from the moment the Republic was formed, music was given pride of place in policies relating to culture and art, a kind of “target” as leaders sought to fashion a new sort of citizen and a new nation-state’ (1996: 195). In this process certain existing musical genres and performance practices were censored, while new musical fields were developed and approved as authentic performance genres. The first music targeted for muting was tekke music, or the music of the Sufi orders, with the closing down of their lodges and rooms in 1924. Tekelioğlu notes that with the exception of the music of the Mevlevi order, the ‘mystical’ music of the brotherhoods has now essentially disappeared. The Turkish Kemalists also sought to reduce to silence anything sung in Kurdish, including the music of religious worship of Sunni and Alevi Kurds, as well as of Kurdish popular songs. The music of the Ottoman palace too, ‘classical Turkish music’, was denigrated as Byzantine and twice prohibited, briefly in 1928 and for a longer time in 1934, when the ban on its radio broadcasting lasted two years. Positioned as simultaneously both Eastern and Byzantine, Ottoman music and its teaching at the Dar-ül-Elhan (House of Tunes), the conservatorium in Istanbul founded in 1917, were abolished in 1925.

In those same years, Hamit Koşay, Turkey’s first musicologist and an expert on Anatolian music, accompanied Bela Bartok on his trip to Adana to record local songs and melodies in 1936. In the section on Halk Musikisi (Folk Music) in his Guide to Ethnography and Folklore, Koşay makes a distinction between Anatolian and Greek modal melodies, as well as establishing a connection between Anatolian and Turkish/Central Asian music. In this he follows Ziya Gökalp, who in his influential 1923 work The Principles of Turkism described Ottoman music as both Eastern and Byzantine, as well as ‘depressingly monotonous’ because it was based on modal quarter tones (cited in Tekelioğlu 1996: 201). Koşay’s disinterring of the history of musical modes in Anatolia to nationalize certain series of notes and rhythmic intervals gives some inkling of the tremendous importance ascribed to music by the Kemalists.

If certain musical sounds were censured, replacement sounds were composed of a number of new aural forms, organized by the triad West/Origin/East. The classification ‘gave the elements with which union was sought (the West and the origin), while also referring to the territory with which unification was absolutely taboo, the East’ (Tekelioğlu 1996: 195). Thus the newly identified folk music of Anatolia, when combined and scored with the harmonic devices (polyphony) of Western music, would produce the new synthetic milli (national) music. As late as the 1980s musical styles such as Arabesk (meaning Arab-like) were construed as Eastern and banned. Other forms of West/origin musical synthesis were composed and taught, including martial music.

Along with new institutions teaching Turkish folk music and dance, Western classical music too was introduced, with German composer Paul Hindemith invited to Ankara in 1935 to supervise the founding of the new Ankara School of Music. ‘His task was to oversee the establishment of a western-style conservatory, producing a symphony orchestra, soloists and composers’ (Stokes 1992: 38). Even before this, the General Director of the People’s Houses had commissioned the first opera written in the republican period (Özsoy – True Race) for the visit of Reza Shah to Ankara in 1935. Üstel (1993: 52) notes that the work, based on a Persian legend, celebrates the ‘friendship’ between the two peoples, while according to Tekelioğlu (1996: 205) Mustafa Kemal personally reviewed the libretto.

Let me conclude by referring briefly to a fascinating roundtable discussion on ‘Music and the Republic’, published in the journal Defter in 1993. The participants wonder why the Kemalists sought a revolution in music (and not in literature or painting. etc.). They propose a number of answers. The first is Atatürk’s famous declaration in 1928 that Eastern music causes apathy, sluggishness, numbness and grief, while Western music brings joyfulness and liveliness. Implicit in Atatürk’s statement, although not mentioned in the discussion, is a theory of music’s affectivity, or its power to engender listeners’ moods and behaviour. Eastern music was heard as a tranquilizer that kept the East in its backwards slumber, march music as conducive to courage and activity. We might say then that the music revolution intended to change the emotional temper of the society.

The second reason mooted is the Kemalists’ intuition that music is a key to social process. Thus discussant Ayvazoğlu cites early republican intellectual Ahmet Tanpınar saying that ‘because the musical understanding possessed by any culture is the most exalted expression of its intelligence, it is very difficult to change. Until a change in our music happens our attitude towards life won’t change’ (Ayvazoğlu et al. 1993: 9). As politics, a revolution in music then was strategically calculated, based on a conviction of the intimate relationship between music and a society’s ties to its past. Thirdly, Ayvazoğlu makes an equation between the music and language revolutions, arguing that the two most perfect channels through which any society expresses itself are music and its spoken language or mother tongue (Ayvazoğlu et al. 1993: 9). The implication is that both the new language and the new music were intended to transform communal self-expression. In brief, the music revolution facilitated the institution of a new society for Kemalists by altering their own emotional, historical and expressive dispositions.

Castoriadis’ model in the light of the Kemalist self-institution

What does this episode of the music revolution in Turkey and of the more encompassing Kemalist instituting of society reveal to us about the virtues and vices of Castoriadis’ writings on autonomy/reflectivity? Can we use this episode to modify certain aspects of that theory? As we have seen, for Castoriadis autonomy is both an imaginary signification of a particular society and a ‘means’ or capacity by which that society (and other societies?) deliberates upon and alters its social imaginary significations. Cultural revolution, to the extent that it aims at the deliberate transformation of an already existing social order or edifice and creates an education system and pedagogy to enable critical participation in a new society, is a project of autonomy or reflectivity. Plainly, the Kemalist cultural revolution sought to invent new social and aesthetic practices based not on the logic of existing social significations understood as religiously ordained but on their contribution to the making of new citizens. Although the cultural revolution was a ‘civilizing’ process spearheaded by a particular segment of society, for that segment it was also a project of self-institution, a project through which its activists created for themselves new ethics and representations of the world.

On the other hand, an authoritarian and self-mystifying aspect to the Kemalist revolution in Turkey is also plainly evident. The party-state’s attempted prohibiting of social practices analysed as obstructing the new social institution meant that much of the population were denied the possibility of alternative self-institution, of autonomy to recreate themselves in other ways. Here we see a high tension between the revolutionary re-institution of society, with its partial enabling of supporters in a revolutionary constituent assembly to engage in the law-making process, and the new state’s violent and anti-democratic suppression of opposition. At the very least we must acknowledge, as does Castoriadis for the French Revolution, that the powerful imaginary signification of the state ‘limits the work of self-institution … [and] makes the idea of revolution become identical with the idea that, if one wants to transform society, it is both necessary and sufficient to seize control of the State’ (Castoriadis 1990: 128). Equally problematically, both the interrogation of the Ottoman institution of society and the instituting of new laws and meanings by the Kemalists were justified through another cardinal modern imaginary signification, that of the nation. In the cultural revolution’s creation of a national culture and citizen, Turkishness conveyed both an inclusive status, membership in which was dependent upon the performance of particular values and practices, and a superior racial quality (against ethnic minorities). The Kemalist critique of Islam’s role as an extra-social source or origin for Ottoman institutions and laws – as Castoriadis puts it, ‘The law is made by us instead of God has given us the law’ (Castoriadis 1997f: 165, emphasis in original) – was made in the name of the Turkish nation, whose assured reality or ‘fact’ issued in a similar mystification of society’s self-institution. In the Kemalists’ attributing of certain ahistorical essences to Turkish ethnicity – republicanism, equality, a military soul, love of the army, a drive to found states – an enduring Turkishness was assigned a determining cause in the forms and practices of the new society.

This partial autonomy – or partial heteronomy – of the Kemalist self-institution of society enables us to revise Castoriadis’ formulations in significant ways. The ambiguity of the cultural revolution renders the dualistic distinction in Castoriadis’ work between heteronomous and autonomous societies unhelpful. This distinction needs to be deconstructed from both sides, through abandoning the mutually constituting characteristics that ontologically define two types of society. Most urgently – given the continuing propagation of ideologies that assign to the West a civilizing mission over others on the basis of its unique cultural genius, including autonomy, reflectivity or individualism – we need better empirical models of ‘heteronomous’ (or ‘holistic’) societies than both Castoriadis and Dumont offer, constructs that are not dependent upon a prior perspective on Western society.

The experience of the Kemalist cultural revolution enables us to assert three ostensibly banal points regarding Castoriadis’ (and Taylor’s) claim that all societies, excepting the ‘Greco-Western’, institute themselves ‘naively’ or as heteronomous. Firstly, so-called heteronomous societies, the ‘rest’ of the world both historically and in the present, are not all the same. New Guinea is not Turkey. Neither can they be reduced to sharing a common denominator, what Dumont describes as their similarity ‘only in comparison to us’ (Dumont 1986: 9), and what Castoriadis calls their ‘blind self-institution’. Secondly, the ‘rest’ do produce societies that ‘put into question the proper world of the tribe’ (Castoriadis 1997e: 339). Thirdly, non-Western societies can also be characterized by a ‘dual institution’ or even by multiple clusters of imaginary significations, and thus by members reflective about tensions or attractions between them. In the multi-religious and multi-linguistic cities of the Ottoman Muslim empire, composed minimally of Byzantine, Jewish, Christian, and Islamic significations and practices, inhabitants’ ‘life-worlds’ were partially constituted through a historical entwining of sometimes shared, sometime rival social imaginaries (cultural, experiential, intellectual, ethical, etc.). From Baghdad to Smyrna to Constantinople, practical contexts of perception and action revealed an institutional ‘porousness’, characterized by both local ‘ethnic’ languages and shared koine (common dialect), and lived out via processes of mimesis, appropriation, adaptation, enmity and amity. One example includes legal pluralism in Istanbul, in which non-Muslims had discretion over choice of court (Shari’a, Orthodox, Jewish) in which to open legal proceedings. Emerging out of that context is the famous pamphlet written by Yusuf Akçuraoğlu in 1912 (Ũç Tarzı Siyaset or Three Styles of Politics), with its lucid comparison of Ottomanism, Islamism, and nationalism as rival political possibilities for late Ottoman society.

More recent anthropological work than that cited by Castoriadis has argued that analysis of any society’s ‘classificatory scheme’ should acknowledge the many ways subjects and groups reinterpret, misinterpret, refute and reorder such a scheme (Rapport and Overing 2000; Ingold 2000: 157–171). The socially instituted and shared web of meanings that constitute society – what Castoriadis calls the magma of social imaginary significations – is surely less coherent, unified or organized than he assumes. Fieldwork-based analysis of ‘non-Western’ societies fractures Castoriadis’ assumption of their common heteronomy. Ethnographic study of the quotidian processes of everyday life shows how the moral superiority upon which a certain group’s claim for social precedence is based can be punctured by that group’s inability to live up to the demands of the ideals that set them apart. Fieldwork allows an assessment of whether this always occurring social disillusion enables subjects to identify what Castoriadis claims are the unthinkable ‘ends’ of their own society. Despite Castoriadis’ implied dismissal of non-Westerners’ accounts of the social relations in which they are involved, anthropology acknowledges participant-actors to be theorists of their own pragmatic social relations. Even as people interpret and negotiate the actions and signs of others through a socially constructed semiotics, the system of interpretation is open to extension, contradiction and new theorizing.

This deconstruction needs to occur from the other direction as well, through a reworking of the category of autonomous (or Greco-Western) society. A more modest, realistic presentation of the experience of autonomy in the West needs to be accompanied by a less strident assertion of the historic breakthrough in ancient Greece of the ‘rupture’ of lucid self-institution. In truth, Castoriadis appears to do this himself. His analysis of really existing autonomy in the West (or in ancient Greece) shows social relations in those societies to be less dominated by the imaginary signification of autonomy than the proclamation of their very definition by it can sustain. Castoriadis maintains this apparent inconsistency because he describes the incomplete extension of autonomy to every aspect of Greek or Western societies as a ‘contamination’ of the process of autonomy by a rival signification, rather than as revealing the simultaneous existence (as in Kemalist Turkey) of processes of autonomy and heteronomy in those societies as well (see, for Greece, Castoriadis 1997h and, for the West, Castoriadis 1997a). Castoriadis applies the same metaphor of contamination to ‘heteronomous’ societies, but in reverse. To escape their essential state – closure to interrogation and interpretation – they too need to be ‘contaminated’, but by the germ of autonomy. Thus, in his essay on racism, Castoriadis offers the non-West only the possibility of conversion to the uniqueness of the one culture that ‘recognizes the alterity of others’ (Castoriadis 1997f: 31). When the West doesn’t recognize this alterity – as it so rarely has in practice – this is again a result of the contamination of its constitutive difference by the signification of ‘rational mastery’ and of its ‘retreat from autonomy’. Similarly, Dumont’s text-based history of ideas approach to the development of individualism in Western society has to confront the experiences of hierarchy, inequality and lack of liberty there, seen most dramatically for him in Nazism and totalitarianism. Where do they come from? ‘They derive in the first instance from the permanence or “survival” of premodern and more or less general elements – such as the family’ (Dumont 1986: 17). Through this quick sleight of hand, the massive presence in Western society of heteronomy and hierarchy is not allowed to threaten its essential historical character.

The problem is the surgical precision with which the distinction between heteronomous and autonomous societies is maintained, and which assigns each society to one or the other category. I would argue that the distinction on that level is redundant. If democracy in Greece has been seen as an exemplary case for social theorizing about the self-institution of society, so too can the cultural revolution in Turkey. From both we learn that [some] societies are supremely heterogeneous in make-up, permeated by (and thus odd mixtures of) processes of autonomy and heteronomy, as must be the social individuals they fabricate. The example of the music revolution encourages us to consider how we might conceptualize the partial projects of explicit self-institution that exists in other societies as well.

Discerning that processes of autonomy exist in other societies is not to assume that we are witness to a more or less successful importation from the West. As already noted, Castoriadis denies the possibility that society’s institution of a world for itself through its core imaginary significations is made in knowing relation to other societies. Yet given that the Ottoman Empire never lost its political sovereignty, Kemalist selective appropriation of imaginary significations derived from European historical experience reflects not only a response to Western power but an intense and centuries-long relationship between ‘Europe’ and the Ottomans that played an essential role in their respective self-institutions. European realization of the imaginary signification of autonomy did not occur in global isolation, and neither (since the 16th century) did the growth of the European economy. Wendy Brown’s criticism of Taylor’s historical account of secularism applies to Castoriadis’ vision as well: the focus on the emergence of autonomy as an imaginary signification in Europe alone makes it ‘more provincially European, monolithic, colonial, than it needs to be’ (Brown 2007: 1).

The same point can be directed at Dumont’s genealogy of individualism, although interestingly Dumont is aware that Muslim civilization, given Islam’s close relationship to Christianity, presents a difficulty for his generalization. It is unclear whether Islamic societies constitute a holistic, an individualistic or a ‘third type’ of complex: indeed ‘on the level of ultimate values, it can be argued that this is a case of individualism’ (1977: 205). To test his hypothesis of antithetic universes of thought, Dumont recommends something akin to the exercise I have attempted here: ‘the harder way is open, of a comparison similar in its inspiration to that which I applied to India, beginning with fieldwork and issuing in a study of the modern development [of autonomy], seen this time against the background of an Islamic civilization’ (1977: 205). Taking Dumont at his word, individualism and autonomy may be elaborated ideals in the West (as is the social imaginary signification of unlimited expansion of ‘rational’ mastery), but this is not to deny by definition their presence as social practices in other societies. In short, the varying expressions of autonomy in different societies should not be indexed to an ontological distinction that constitutes the West and the rest as radically antinomic to and incompatible with each other.

Finally, I would argue for the problematic status of the distinction on different grounds as well. Castoriadis’ quarantining of autonomy from religious and non-Western societies, and the importance he gives to the Greek breach in the history of the world, leads to a relative muting of the question: what laws should be instituted? Given that lucid self-institution in itself can result in social monstrosities – as some might argue the above example of revolutionary state-nationalism in Turkey shows – the central issue is not only the necessity of explicit self-institution, but also the question of its limits. The problem can be formulated in a different way. How do we prevent self-institution in antagonistic relation to the self-institution of others, as the Kemalist social institution demonstrates in its intolerance of non-Turkish or Muslim social identities? Although Castoriadis writes that autonomy is not just an end in itself, he also argues that the first object of a politics of autonomy is to ‘help the collectivity to create the institutions that … will not limit but rather enlarge their capacity for becoming autonomous’ (1997i: 134). At least in this passage, the first laws instituted by a society that recognizes itself as the source of its own laws – that is, an autonomous society – are laws that facilitate greater recognition of itself as the origin of its own laws … and so on. Again this begs the question: what laws should we create?

Conclusion

Lest we conclude that the Kemalist cultural revolution is of historic interest only, let me finish by noting one of its chief legacies in contemporary Turkey: the continuing politicization there of cultural aesthetics as an arena of conflicted self-institution between militant Kemalist actors and religious Muslims. The contemporary conflict is not one between autonomous secularists and heteronomous Muslims, as Castoriadis’ or Taylor’s concepts would have us believe. Indeed, Muslims in Turkey write novels, make films, and record music that criticize key assumptions of Islamist politics. This cultural politics is articulated in relation to Kemalist management of and discourse on Islam. But it is also produced in relationship to different versions of Islam generated by the Islamist movement itself (for example see Muslim politician Mehmet Metiner’s 2008 autobiography in which he explicitly rejects the tenets of what he calls ‘Jacobin Islamism’). Further, the pursuit of aesthetic practices claimed to be more authentically local by religious Muslims – say the learning of the ‘traditional Turkish art’ of ebru (paper marbling), the revival in calligraphy (hat) or the popularity of the ney (the Sufi bamboo flute) – should not be understood as effects of a thoughtless, naïve or heteronomous conviction to re-institute Islamic society. Such suspicion is blind to the critical and creative work of selective appropriation necessarily involved in their production.

Castoriadis’ high valuation of the radical imaginary’s creation of new social-historical forms ex nihilo means he underplays the efficacy of self-institution through representation, interpretation, explication or evaluation (of a context, history or tradition). Muslims’ and Kemalists’ patchwork generation of artistic practices are always accompanied by a reflective discourse on the meaning of those practices. Re-discovered or familiar activity is subject to contested and creative modes of representation about it. In his worst moment, Castoriadis claims that a ‘real’ Muslim has no capacity to put socially instituted objects into question. ‘It is pyschically inconceivable for one to say, The law is unjust, when the law has been given by God and justice is merely one of God’s name attributes’ (Castoriadis, 1997f: 165, emphasis in the original). Yet, contra his prejudice, Muslims in Turkey do not seek shari’a law.

As with historic Kemalism in Turkey, contemporary Muslim counter-cultural revolution also constitutes a mode of embodied and conscious self-institution, a collective expression of autonomy. Accordingly, the important question in Turkey is clear: which political mechanisms need to be instituted to mediate the clash of rival projects of autonomy that these respective cultural revolutions herald? The remarkable but little known process of constitutional reform in Turkey over the last 15 years – with all its zigs and zags, hardly surprising given Kemalism’s compromised history of autonomy – demonstrates an ongoing determination to ‘remake [Turkey’s] political life according to agreed upon principles’ (Taylor’s constructivism). The outcome is still uncertain. That the so-called Islamist Justice and Development Party (AKP) is currently the leading actor in this project causes us to further question Castoriadis’, Taylor’s and Dumont’s attempt to objectify their own societies through a detour to others, profoundly dissimilar amongst themselves but prefigured as identical ‘in comparison to us’ (Dumont 1977: 9).