Çağlar Keyder is Professor of Sociology at Binghamton University and at Bogazici University in Istanbul. He has written extensively on economic history, historical sociology and political economy, mostly with a focus on the Ottoman Empire and Turkey, and always with an interdisciplinary perspective. One thread in his early work was the analysis of divergent capitalist trajectories in Western Europe and the Ottoman Empire through the nature of agrarian transformations in the two regions. His work on the Ottoman Empire was foundational in pointing toward a re-interpretation of the Empire's economic and social history from a world-systems perspective. This impressive body of work on late Ottoman economic history greatly contributed to what can only be called a ‘paradigmatic shift’ in the scholarship on this region. This was followed by his path-breaking book on modern Turkey, State and Class in Turkey: A Study in Capitalist Development (1987), which has become a landmark in the Marxian analysis of Turkish political economy. By analysing the state elite as an autonomous agent in capitalist development, with its own interests and agenda, this book opened the door for research on the contested development of a bourgeois class, and on the political accommodation between the state and businessmen — questions that have since become central in the study of capitalist trajectories in ‘late-comer’ contexts. In his later works, Keyder's focus has shifted to the analysis of complex urban transformations in Istanbul in the period of neoliberalism and globalization as well as to the changing welfare regime of Turkey in the same period. His work in these fields has greatly contributed to our understanding of the impact of globalization on social, economic and spatial dynamics in Turkey. In authoring and (co-)editing an impressive number of articles, books, collections and journals, Keyder has communicated to younger scholars his penchant for the fruitful synthesis of political economy and historical sociology. Professor Keyder's most recent work has responded to the challenges of globalization, neoliberalism and the current crisis. A limited selection of his publications is listed below.
TK: I would like to start with a partly biographic question. Unlike many scholars in the field, your academic focus has changed rather substantially over the four decades. Your first works in the 1970s deal exclusively with the relationship between agrarian structures and the development of capitalism in different geographies. But starting with the late 1980s your work seems to have moved somewhat away from such questions into analysing the functioning of neoliberal globalization in the context of cities. Could you explain how and why this shift in your academic orientation took place?
ÇK: Well, my interests change as the issues change. But, of course, one is also in dialogue with prevailing academic concerns. Initially, my interest in agrarian structures was political. The orthodox expectation of concentration in land and expropriation of the peasantry were integral to the class analysis of the time. This orthodoxy did not seem to apply in the case of Turkey. It was important to point this out in order to be able to understand the political triumph of pro-market populism of centre-right parties who received the majority of their support from the peasantry. I had written my dissertation on Turkey in the 1920s, when the vast majority of the population were peasants, and trade and credit networks operated on the basis of links with small producers in agriculture. So my interest in the peasantry ran in the direction of showing that the agrarian structure was dominated by smallholders who gradually became market-oriented petty commodity producers. In other words, for the larger part of the Turkish countryside, there wasn't much potential for a class struggle between landlords and landless peasants.
I was also influenced at the time by the new and growing social science literature on the peasantry, which demonstrated their rapid transition into small commodity producers. These debates around the peasantry and commodity production readily linked with the enormous interest in Marxist historiography. The transition debate was just being discovered in the 1970s and subsequent interventions to this debate, especially by Robert Brenner, helped me understand the relative roles of the market and of class relations in agrarian transformations. Obviously, all of this was crucial to analysing developing countries and their future in the world economy.
There was one more dimension to studying the agrarian structure, which had important implications for the history of the Ottoman Empire and the dominant modality of its incorporation into the world economy. This was the debate relating to the nature of the Ottoman social formation. There was, at the time, a lively discussion on the Asiatic Mode of Production, which centred on the conceptual problem implicit in it. In an orthodox reading, Marxism had to deny the possibility of such a mode of production where the state constituted its own bureaucracy as a tax-collecting class, and was itself not reflective of relations that obtained in the civil society. In other words, the state was also the basis of the formation of a bureaucratic cadre that continued to serve and dominate the state apparatus. My understanding of the Turkish society and my reading of Ottoman history led me to the conclusion that the state in fact held such a position, that the patrimonial structure was such that political control over society meant that the Palace and the bureaucratic establishment were the effective rulers.
These rulers, however, were not able to perpetuate their rule through inter-generational transfer of office and wealth. Those families that converted their political position into landed or commercial capital were, by and large, destroyed after a few generations. In other words, the patrimonial order was effectively perpetuated and was not allowed to transition into a feudal one of hereditary privilege. Only in the nineteenth century, under European pressure and protection, had the possibility emerged of the independent evolution of a bourgeois class, which was comprised mostly of non-Muslim merchants, bankers and later manufacturers — all in the urban context. The state had not yielded on the agrarian issue and the peasantry had remained independent and not feudalized. The merchants and the bankers served as intermediaries between the world economy and the ‘independent’ peasantry.
This was a very different trajectory to, for example, that of many Latin American countries, or even Egypt, where world-economic integration had implied the emergence of large-scale, export-oriented, estate agriculture. Without the counterweight of a class of large landlords, committed to exporting to the world market, there had been no tradition of liberalism in Turkey; no confident social class with any chance of hegemony in the society or ability to challenge the state class. So, talking about the agrarian structure and trying to show its continuing domination by smallholders implied taking sides as far as the character of the pre-capitalist social formation was concerned. More importantly, it also implied a different analysis of social and political transformation than what at that time was suggested by the prevalent political strategies (again dominant in Latin America) based on a struggle against an oligarchy defined by landholding and comprador networks. As you can see, I am a firm believer in the importance of agrarian origins.
TK: In your seminal work State and Class in Turkey, you successfully apply this framework to the analysis of the development of capitalism in post-Ottoman Turkey arguing that the State played quite crucial and central roles, which could be seen as hampering the development of an independent bourgeoisie as well as a strong civil society. Could you elaborate on this argument?
ÇK: For social scientists working on development there is no possibility of escaping the state. During most of the twentieth century we were debating and analysing national development and its actors. Development was not something that happened on its own; it had to be initiated, promoted, planned. Since Friedrich List or even since Colbert in the seventeenth century it was clear that the state had a role to play in development. Within the Marxist approach this focus on the state posed a problem concerning the class character of the state and the motivation of the state managers. In my analysis of the Turkish case, based on the premise of a state class dominating the society, the story of capital accumulation or the formation of a bourgeoisie could be told as the struggle to establish a workable balance with state control. The problem with Turkey — and this prefigures the experience of a host of post-colonial societies — was that when the Republic came to existence after World War I, ‘de-colonization’ implied the ousting and destruction of the Greek and Armenian bourgeoisie that, through links with the global economy of the pre-war era, had been in the process of establishing a robust class position in the society. During and after the formation of an independent Turkey out of the wreckage of the Empire, they were killed, deported or were subject to population exchange. Basically, the new Republic ended up with many fewer capitalists and a decimated civil society, hence a much diminished potential for capitalist development. The bureaucrats thus gained a new and strengthened position bolstered by the ideological claim of having saved the country and the nation from foreign domination. From this point of view, Turkey's experience prefigured the social and ideological formation of new states in the post-World War II era.
As we know, very soon after the formation of the new Turkish Republic the Great Depression followed and this allowed the bureaucrats in Ankara to come up with relatively protectionist measures — even though they did not really have a prior project to do so. What followed was a statism with autarkic aspirations, which was prolonged by war-time isolation. It is important to note that relative economic isolation continued in the post-War era, into the 1970s. Under American hegemony, as long as Turkey's politics conformed to the larger project of Cold War containment, Turkish bureaucrats were given a lot of leeway. Until national development fell into crisis during the 1970s, they could implement their import-substitution policies and development plans without much objection from the outside.
What you are alluding to in the question is the ensuing struggle when statist industrialization succeeds. When it does, there emerges a new class of businessmen who owe their nurturing to bureaucratic subsidy and protection, but who feel sufficiently grown out of the infant stage that they want to negotiate their share in political power. This is the struggle between the state and the bourgeoisie. Of course, until the neoliberal era, this new bourgeoisie's bid largely remained tentative; and, in the absence of ideological hegemony, they resorted to supporting military interventions and contributed to maintaining Turkey's stunted political life.
For their part, the state managers never enjoyed the prestige and the coherence that apparently characterized, for instance, the South Korean case. The Turkish state was not predatory but it was not a successful developmentalist one either. Perhaps it was not sufficiently autonomous; it easily gave in to populist temptation. It certainly was not successfully embedded, preferring the coercive mode. The bourgeoisie was always wary and never trusted the bureaucrats fully, while the bureaucrats were jealous of their prerogatives and were not above undermining particular businessmen whose fealty they suspected. This uneasy accommodation between the two parties is an instance of the state/development nexus that has been the subject of some of the most inspiring contributions to development studies since the 1980s, such as Evans (1995), Kohli (2004), Amsden (1989) and Chibber (2003). Eventually, it was the global triumph of neoliberalism which tipped the scale toward the owners of capital.
Now, under the present crisis, the state–bourgeoisie relationship may be undergoing a new transformation, as a sort of neo-mercantilism is becoming the rule. The Turkish state, which since 2002 has been governed by a pro-business and pro-market single-party regime, seems to want to become a trading state, where business strategy is dictated by political choices. Once again, we find the bourgeoisie cowering — not so much before the bureaucrats but before the party officials. This history suggests to me that the Turkish state, as an administrative framework, has never been fully controlled by the bourgeoisie even if it has become resolutely pro-capitalist.
TK: So, most of your empirical work — whether on agrarian structures, state–capitalism relationships or, more recently, global cities — focuses geographically on the former territories of the Ottoman Empire. In fact, during the 1970s and 1980s several scholars affiliated with the Braudel Center in Binghamton have made influential and timely interventions to social sciences by forcefully demonstrating the theoretical and empirical incompleteness of social science theories due to their blindness to such geography (see, for example, Wallerstein et al., 1987). Could you explain why you have decided to look for answers to your general theoretical questions in this geography and what we have learned, or we can still learn, from analysing social structures, state–society relations and capitalist trajectories in this particular geographical context?
ÇK: Of course, there is a trivial answer to this: I am from this geography. My primary identity is being born in the former imperial capital city of Istanbul and having lived here most of my life. But the non-trivial answer is that in the late 1970s and 1980s it seemed that this entire geography was neglected in social sciences. There was of course a very rich historical literature but when compared to all the conceptual innovation deriving from Latin American experience and then the East Asian experience, the Middle East looked as if it contributed very little to the literature on development and even less to historical sociology. This is why it was gratifying to bring the Ottoman Empire into focus within World Systems studies. The perspective of peripheral incorporation provided a rich framework within which to consider the problems of economic and political transformation during the nineteenth century. One could integrate into the account the impact of the world economy, the networks of trade and credit, processes of marketization in agriculture, the growth of port cities, the emergence of new groups such as market-oriented farmers/merchants and an intermediary bourgeoisie, and the attempt by the state to resist and to accommodate outside pressure.
All these themes in historical sociology could be found in abundance in the experience and history of Middle Eastern societies. It is not the case of course that the Ottoman Empire — stretching over the Balkans, what is now Turkey and most of the Middle East — experienced a uniform transformation. We are not talking about a smooth geography. Social structures and agronomic conditions varied in different regions, but since a single legal and administrative framework applied to the entire realm, the process of incorporation and subsequent transformation could also be studied in the comparison of various regions. Along with colleagues in Turkey and elsewhere, we compared agrarian change, port cities and post-imperial state formation in various studies ranging over Greece, Syria, Turkey and Egypt. This is an ongoing project that extends the initial attempt started in Binghamton in the late 1970s. Of course, the focus of most of these studies has been historical sociology in the sense that different regions in the Ottoman Empire ended up with quite divergent processes of transformation and modernization. It is not very easy to draw a direct line from these diverging paths to the current situation. But, nonetheless, this is an area that is worth exploring.
TK: Now, I would like to move on to your more recent works which deal almost entirely with neoliberal globalization. In this endeavour you seem to privilege two topics to understand how neoliberalism plays out. The first of these is the social, economic and spatial transformations in what we can call ‘globalizing cities’ and particularly Istanbul. The second one is the changing regimes of social policy around the world and particularly in Turkey. To start with the first: what can we learn from a close study of such cities that we cannot learn in other contexts in our attempts to understand neoliberalism?
ÇK: I live in Istanbul and it is fascinating to observe the layers of its history, how the built environment of the various periods — the Byzantine, Ottoman and the Republican — survive, find new ways to interact and serve new functions. Istanbul became a typical third world city during the period of national development. Its population exploded with migration from the interior of a previously rural population. The city of my childhood, which was relatively small — only one million in population — and still held a post-imperial aura, became shabby, dirty and polluted, surrounded by shanty towns constantly under construction, and callous towards its history and geography. I got interested in the global-city literature because it pointed to a new kind of fortune for a city like Istanbul which seemed to hold all the right cards. Globalization could herald a change in the course of its development, where Istanbul would play a new role at the centre of regional and global networks. In fact with the fall of the Soviet Union there were already the beginnings of such networks, mostly at the informal level but promising new paths of transformation. My initial instinct was advocacy. I wrote as a patriot of the city identifying the opportunity that came with globalization; in later writings, after global networks started dominating Istanbul's evolution, I turned to their social impact — the polarizing tendency, the social exclusion of the population who played only a small part in the dynamic, the risks to cultural heritage and historical memory that ill-advised catering to business and tourism posed. Of course, by focusing on the global city in Turkey, one could also see what was going on in the rest of the country; one could observe how the social contract and the bargains of the national development era were being dismantled. The newly rich, the most globally connected, those capitalists, bankers and professionals who played in the world league, were all in Istanbul, as were the newly unemployed, dislocated and the excluded. So, a focus on the global city seemed to be the best vantage point to observe the course of globalization as the economy, the society and the culture of national development were being transformed.
The other dimension of global city development is that in a sense it completed the transformation that had begun much earlier, towards the full domination of the market over all commodities. What I am referring to here is again something that relates to what I said earlier about the Ottoman social formation. Under the Ottoman Empire the state was never fully comfortable with the notion of private property in land. The fiction of the land being owned by the Sultan is well known; the implications of this on capitalist development are also fairly clear. But there is another consequence of this ambivalence in the property regime: in the case of Turkey, as in many rapidly urbanizing countries during the same era, the accommodation of rural migrants depended very much on the existence of public ownership. Most of the land around cities where the shanty towns were built was in fact not privately owned — in other words it was land that belonged to the treasury.
Under the impact of globalization competition over urban land became more intense; in order to attract investment ambivalence had to be eliminated as populist policies were abandoned in favour of a strict property regime. What this means is that over the last two decades or so, urban land has become more commodified; and if you remember the Polanyian perspective, the commodification of land is one of the principal factors in the denial of moral economy (Polanyi, 1944). In this sense the global city's development also amounts to the full domination of the market in the urban arena.
TK: The commodification process in land is something that I also worked on for my dissertation: I specifically looked at urban renewal projects in shanty towns that demolish massive informal developments and relocate people into formal subsidized housing built by the central government. This is a very significant process implemented by a quite conservative and neoliberally oriented, yet at the same time very popular party that has been gaining enormous support during the last three elections from the urban poor. Strongly inspired by your own work, I treated this transformation as a shift in the social policy of Turkey from one mostly based on informal relations in land markets to a more formalized regime of social provision. How would you characterize this argument? And why, in your opinion, would the poor vote for a party with a blunt capitalist agenda? Can this party's new social policy regime provide an answer to this puzzling reality?
ÇK: I have to say that political attitudes and voting behaviour are not entirely transparent to me. You are quite correct in pointing to the ambivalent behaviour of the ruling party toward the poor. On the one hand they are unflinching advocates of a neoliberalized market; on the other hand they have made various attempts at providing a safety net and some measure of social protection for the poor. Obviously, their popularity cannot be explained only through their appeal to material interest, because they also represent a certain liberalizing ideology (at the level of a religious nativism) against the strict enlightenment secularism of the previous period. However, it is certainly correct to link the transformation in the land market with social policy.
The land situation, for a very long time, provided an ersatz social policy toward the poor and new immigrants into the city. For one, the new immigrants were coming from a background of small farming. In other words, they were poor but not exactly expropriated and they were not driven off their land. So, it was a question of a not-too-desperate movement from the countryside to the city. Second, for a long time it was the case that, when they did arrive in the city they could, with modest payments either to some right-holder or to a protector, occupy a piece of land on which they could start to build a dwelling. Over time, most of these illegal developments were acknowledged — not fully legalized, perhaps, but certainly strong possession rights were obtained. Now, with stronger commodification of the land, this has become impossible for newer immigrants. In fact, since the 1990s, not only has occupying land become difficult in terms of any accommodation with the police or with municipal government, but newly elected politicians have also been reluctant to act as patrons to the newcomers because they find the promise of revenues and tax incomes they might obtain through new kinds of development much more lucrative.
Now, the main reason for me to become interested in social policy at this point was because during the 1990s a new group of immigrants started coming to Istanbul and to some other cities in Turkey. These were the Kurdish population, originating mostly in Eastern and Southeastern Anatolia. We need a parenthesis here because although I have talked about the legacy of the Ottoman Empire as one of small ownership in land, the Kurdish area had a rather different social and agrarian structure. In this region, tribes were sedentarized in the second half of the nineteenth century. When the Empire pressured and forced tribal leaders to settle, their followers were transformed into dependent peasants who lived in a sharecropping-like arrangement with their leaders. Thus, the agrarian structure that characterized Kurdish areas was much more feudal than in the rest of Turkey. Also, this area was quite late in its incorporation into the market. The underdevelopment of the area triggered the rise of political movements which gained strength as former sharecroppers were expropriated. In fact, the Kurdish movement initially started as an anti-feudal movement by landless peasants, but transformed into an ethnic-national movement, arguably due to the intransigence of the Turkish state.
By the 1990s, the political situation had deteriorated to what can only be described as a civil war between Kurdish guerillas and the Turkish army, leading to larger waves of migration of displaced populations. In the 1990s poor Kurdish peasants arrived in large cities (both in western and eastern Turkey) and were confronted not only by cultural exclusion and discrimination, but also a situation where land to squat on was no longer available. So, in that sense the ersatz social policy of permitting new immigrants in the city to occupy land and informally build their own houses could no longer function. Let us not forget that this was also the time when neoliberal policies and structural adjustment were transforming public enterprises and making employment leaner and more flexible, and therefore the employment market did not have the same kind of carrying capacity that it used to have either. This meant that a new and more formal social policy would be needed if there was going to be any kind of accommodation for this population, who would otherwise be destitute. It is in this context that we have to evaluate the attempts to do something about social policy — as a response to the damage caused by the new global economy, but also as a political attempt to ‘deal with’ the Kurdish question.
TK: The kind of social policy reform or transformation that we see in Turkey under a very neoliberal party, which seems to be somewhat of a puzzle, is actually being repeated over and over in a lot of the developing world. What makes this even more puzzling is that in almost all advanced capitalist countries welfare spending and social policy in general is in decline. When you look at budgetary decisions you see that in the developing world the budget allocated to social policy is increasing while it is decreasing in the advanced capitalist countries. How would you explain this somewhat anomalous situation?
ÇK: This is very true and it is a little-studied development: that in some countries such as Turkey, Brazil, Mexico, South Africa, Korea and Taiwan, there have been significant increases in social expenditures over the past ten to fifteen years. First, we must remember that a lot of these third world countries had what we might call a Bismarckian version of social policy, which meant that states were willing to provide pensions, healthcare and unemployment benefits to a relatively small segment of the population, who were formally employed, mostly in the public sector. At the same time they were willing to neglect those who they considered to be the informally employed, or the poor self-employed. Now this might have been sustainable in largely rural societies when the public sector dominated the economy and provided the major dynamic for development, and when there was a possibility that formal employment would expand. But as neoliberalism and structural adjustment contributed to the dismantling of the public sector and as more and more people ended up in informal sectors in the cities, there were new demands. There was also pressure from international organizations such as the World Bank and the IMF to dismantle the Bismarckian model and to initiate a more universal social policy that would also benefit the non-working or the informally working poor. So, in that sense, programmes that were targeted at public employees or formal, unionized industrial workers began to include larger groups of people in their coverage. But, at the same time, as a result of the interrelated processes of neoliberalism and large scale de-ruralization, there emerged in the cities a population that was always precariously employed and whose livelihood could not be seen as stable or guaranteed. The realization of this growing problem led international organizations, the World Bank in particular, to advocate poverty reduction policies. This was a process of what the economists might call bribing because while the market-guided transformation of the economy was going on, there was a huge population that was falling by the wayside. The new social policy initiatives were designed to alleviate the burdens falling on this population. The obvious example is conditional cash transfers in countries such as Brazil, Mexico and Turkey which work to provide poor families with basic incomes in exchange for sending children to school.
In Turkey, we have a government which has been reluctant to reduce the pace of neoliberal globalization. On the other hand, it has been willing to undertake new types of social policy. The most important of its initiatives has been the newly introduced healthcare system, with almost universal coverage, which has been one of the most significant elements in the ruling party's popularity. Leaving the Bismarckian model behind and creating more of a citizen-based and rights-based social policy has become an option for governments that were reluctant to slow down the globalized transformation of their countries. In countries like Turkey and Brazil, we have seen rapid increases in social expenditures, both in terms of combating poverty and also in terms of establishing universalist policies such as in health and social assistance, with the consequence that the orbit of social policy has expanded tremendously.
TK: You seem to suggest that Turkey has performed rather well in the last decade under the rule of the single-party government formed by the Justice and Development Party. Given the strong criticism that this party's economic policies and political authoritarianism has generated, your comments seem to be rather controversial!
ÇK: In some areas it hasn't done too badly, but we are still in the midst of the crisis. As I just said, social policy has been relatively successful, and as a consequence income distribution has not worsened. The government has behaved in a very Keynesian fashion, accelerating investment in infrastructure and expenditures in housing. This latter category, which is in part driven by social policy concerns, is interesting, because it does not require expenditures out of the current budget. A government agency was formed and empowered to privatize public land in peri-urban areas. This agency ‘sells’ land to private sector developers who in turn finance the construction of social housing. Since the existing housing stock dating from the shanty town construction era is of terrible quality, there is a rush to upgrade. Banks have been happy to oblige in the case of middle-class developments, while social housing is made available at low and mostly manageable monthly installments. We must not forget that land development and housing construction have been the pilot sector in recent crises, starting in the late 1990s in Thailand and continuing with the mortgage bubble in the US. The Turkish economy may also face similar problems due, on the one hand, to an overbuilding of middle-class apartments, and on the other to debt-servicing difficulty when the employment situation worsens.
There is, however, a longer-term development in the political field that we should be carefully observing. One manifestation of the ‘double movement’ during the current crisis — the reaction against the untrammelled working of the market — will certainly be greater state involvement in the making of economic decisions. As in the Great Depression, the unfolding of the current crisis will likely bring about new forms of state engagement with the economy. Neoliberalism is being abandoned, but we don't yet know what will be the equivalent, in the current conjuncture, of the various forms of statism, ranging from war economies in Europe to conservative-developmentalisms in peripheral countries such as Turkey, which defined the spectrum of responses to the Great Depression. I believe that the single-party government in Turkey is rehearsing a regime of tight control in the social and political spheres (in part through their deployment of Islamic nationalism as ideology), and of political domination over the economy. They are dismantling the liberal framework of autonomous regulatory bodies and strengthening the economic bureaucracy. They employ various instruments of control over investment decisions and play favourites among business groups. In terms of foreign trade and investment, the government has taken on a managing role, in a sort of neo-mercantilist stance. It is likely that they will begin to engage in industrial policy in the near future. It may well be that this attempt at dirigisme is a harbinger of what we will see in countries able to divest themselves of the strict discipline of international organizations while also avoiding the clutches of international finance. This is not to say that the crisis will thus be evaded, but this stance may serve to deflect some of the discontent with the economy by highlighting national goals and achievements.
Let us not forget, however, that such management comes at a steep political price. The crisis will deliver new sorts of political balances everywhere. In the case of Turkey it is likely that tolerance for criticism and dissent will diminish, all powers of the state will be collected in the hands of the governing party, the dose of nationalism will increase, and we will get a civilian version of the authoritarian governments that we are accustomed to. The only difference is that they used to function under the sponsorship of the military; now there will be the dicta of an elected party, reflecting the tyranny of the majority.