What does it mean for a country to reform itself? How is this actually done? How much of reform is by explicit design and deliberation, and how much by other means? What roles do norms, standards and measurements play in such reform? This book addresses these questions by looking at how Turkey has undertaken a reform of its statistics, and particularly at changes in how statistics are collected and used. Statistical tools and analyses are now pervasive (and still growing) aspects of our public and private lives, used in myriad fields of science, engineering, business, entertainment and government. Given this massive influence it is remarkable that these statistical tools for the most part are a little over one hundred years old (Porter 1986). Statistics are often seen as information or indicators, and indeed they can relatively easily be shown to function that way, for instance in periodic publications that private enterprises or public authorities find useful (e.g. as data about markets or populations for the generation of policy). However, changing the way in which statistics are compiled and used can also have effects that are less often noted, including on the phenomena the statistics are about (Scott 1998). Indeed, a main argument of this book is that one of the important ways in which rural livelihoods in Turkey—and agriculture in particular—are changing is through changes in the way statistics about agriculture are collected and used. The implications of this are significant and go beyond agriculture and rural livelihoods. For one thing, a group’s lifestyles and worldviews might be changing as a result of seemingly “technical” adjustments to administrative instruments ostensibly intended merely to observe them. This book shows how lifestyles, livelihoods and worldviews often change as much through changes in things like data collection as they do through more explicit, deliberative processes, pointing to the imbrication of the “technical,” the “social,” and the “political,” a formation Timothy Mitchell (2002) has referred to as technopolitics. Statistics and their reform may seem arcane or technical and far from the arenas in which politics is being made “loudly” (especially these days in Turkey) but in focusing on them I hope to show that it is in and through (and in response to) new statistical apparatuses and their attendant definitions and practices that “the political” is being formed and reformed in Turkey, as it is elsewhere.
The book is based on fieldwork in Turkey with statisticians, farmers and agricultural technicians, examining the role of statistics in how Turkey learns the things its application to enter the EU (however uncertain the outcome now appears) has required it to learn about itself. The impact this is having on Turkey is not commonly understood or appreciated, neither inside nor outside the country. Moreover, to the extent that the genealogies of these new knowledges and practices are often located in EU norms and directives we are able to see how the EU integration process—regardless of the country’s ultimate entry or even of the future of the EU itself—will have changed Turkey.1
Numbers and Society
That statistics have effects on—indeed that they were part of the emergence of the object known as—society, and that the relationship between statistics and society is feedback loop-like, is not a new insight; Ian Hacking, most prominently, has been pointing this out for several decades now. In a paper presented in 1980 (published in English in 1991) he wrote, “[The collection of statistics] may think of itself as providing only information, but it is itself part of the technology of power in a modern state” (Hacking 1991: 181). Another recent volume is devoted to describing “the mutual construction of statistics and society” (Saetnan et al. 2011), while in an influential volume on the performativity of the discipline of economics Mackenzie, Muniesa and Siu write regarding the relationship between economies and economics, “the issue…is not just about ‘knowing’ the world, accurately or not. It is also about producing it” (2007: 2). Verran (2010) writes of enumeration practices that they bridge the semiotic (mainly as symbols and indexes) and the material, reflecting and elaborating ideologies about the world and the forces transecting it, while also being involved in transforming those worlds. The term I will use to describe the relationship between statistical data and social and natural forms (say, farming communities and planted fields) is performative, following scholars contributing to what has been called the “performativity programme” (Çalışkan and Callon 2009: 370). In using this term I build on scholarship in several fields including sociolinguistics, feminist studies and recent substantivist approaches to economy (itself building on earlier such work), and I define performative in more detail below. For now suffice it to say that to call something performative is not to liken it to a “performance” in a theatrical sense, like someone in the role of an actor for an audience. Rather, it is to emphasize that an act of description can have effects that rearrange the relationship between the description and the phenomena the description is purportedly about. We will thus see how, to again borrow a phrase from Ian Hacking, “representing” and “intervening” are two sides of the same coin and are in fact inseparable (Hacking 1983). Ultimately, I will argue that it is in and through this kind of performativity that important work of institutional commensuration between Turkey and the EU is happening.
The starting point of this book is that the ways in which scientists and other “experts” are studying the world has become central to how the world is changing—and this includes changes in the kinds of people and collectivities we all are.2 The specific processes through which this has happened are many, but by the late nineteenth century an “avalanche of printed numbers” (Hacking 1990: 2) in the form of statistics accumulated about a growing number of phenomena was playing a central role in this dynamic. This was enabled by subtle but profound shifts around 1840 in the nature of knowledge in Western Europe and the North Atlantic, specifically what was considered a “fact” (Poovey 1998) and the increasingly important roles played by numbers and quantification (in the form of mathematics and statistics) in what Hacking (1990) and Porter (1986) describe as a new approach to and understanding of probability: a shift from a degree of certainty one might have about something, to the relative frequency with which something happens. Nineteenth century figures like the Belgian polymath Adolphe Quetelet (1796–1894), who was trained in mathematics and astronomy, began applying probability mathematics to a widening circle of phenomena, data on which was rapidly accumulating (Canguilhem 1978). He thought that the distribution of certain traits observed among a large number of people (e.g. the height of military recruits) described a curve that was similar (analogous) to the ones used in plotting errors in astronomical observations, with a few low calculations and a few high ones at the two ends and many more near the middle, which in the case of astronomical calculations was considered to probably be where the “true” value lay. This “normal curve” began to be applied in the form of analogies to more and more “social” phenomena (e.g. the ...