The social sciences of quantification

Regarded as a leading figure in critical thought and what has come to be known as French pragmatic sociology, his influence in the fields of sociology, political science, economic histor

Logic, Argumentation & Reasoning 13

Logic, Argumentation & Reasoning

Interdisciplinary Perspectives from the Humanities and Social Sciences

Volume 13

Series editor

Shahid Rahman

Logic, Argumentation & Reasoning explores the links between Humanities andthe Social Sciences, with theories including, decision and action theory as well

as cognitive sciences, economy, sociology, law, logic, and philosophy of sciences

It’s two main ambitions are to develop a theoretical framework that will encourageand enable interaction between disciplines as well as to federate the Humanitiesand Social Sciences around their main contributions to public life: using informeddebate, lucid decision-making and action based on reflection

The series welcomes research from the analytic and continental traditions,putting emphasis on four main focus areas:

• Argumentation models and studies

• Communication, language and techniques of argumentation

• Reception of arguments, persuasion and the impact of power

• Diachronic transformations of argumentative practices

The Series is developed in partnership with the Maison Européenne des Sciences

de l’Homme et de la Société (MESHS) at Nord - Pas de Calais and the UMR-STL:

8163 (CNRS)

Proposals should include:

• A short synopsis of the work or the introduction chapter

• The proposed Table of Contents

• The CV of the lead author(s)

• If available: one sample chapter

We aim to make a first decision within 1 month of submission In case of apositivefirst decision the work will be provisionally contracted: the final decisionabout publication will depend upon the result of the anonymous peer review of thecomplete manuscript We aim to have the complete work peer-reviewed within

3 months of submission

The series discourages the submission of manuscripts that contain reprints ofprevious published material and/or manuscripts that are below 150 pages/85,000words

For inquiries and submission of proposals authors can contact the editor-in-chiefShahid Rahman via: shahid.rahman@univ-lille3.fr or managing editor, LaurentKeiff at laurent.keiff@gmail.com

More information about this series at http://www.springer.com/series/11547

Isabelle Bruno • Florence Jany-Catrice

ISSN 2214-9120 ISSN 2214-9139 (electronic)

Logic, Argumentation & Reasoning

Interdisciplinary Perspectives from the Humanities and Social Sciences

ISBN 978-3-319-43999-0 ISBN 978-3-319-44000-2 (eBook)

DOI 10.1007/978-3-319-44000-2

Library of Congress Control Number: 2016948621

© Springer International Publishing Switzerland 2016

This work is subject to copyright All rights are reserved by the Publisher, whether the whole or part

of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on micro films or in any other physical way, and transmission

or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed.

The use of general descriptive names, registered names, trademarks, service marks, etc in this publication does not imply, even in the absence of a speci fic statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG

The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

To the memory of Alain Desrosières (1940–2013)

1 Introduction The Social Sciences of Quantification in France:

An Overview 1Isabelle Bruno, Florence Jany-Catrice and Béatrice Touchelay

Part I Sociology of Quantification: Alain Desrosières’s

contributions

2 Introduction to the work of Alain Desrosières: the history

and sociology of quantification 17Michel Armatte

3 Alain Desrosières’s reflexive numbers 33Luc Boltanski

4 Alain Desrosières’s spectacles: one lens realist,

the other constructivist 43Emmanuel Didier

5 Fromstatistics to international quantification: a dialogue

with Alain Desrosières 55Roser Cussó

6 Learning from the history of the probabilistic revolution:

the French school of Alain Desrosières 67Fabrice Bardet

Part II The Statistical Argument in the Neoliberal Era

7 Quantifying the effects of public action on the unemployed:

disputes between experts and the rethinking of labour market

policies in France (1980-2000) 83Etienne Penissat

vii

8 Counting the homeless in Europe:‘compare before

harmonising’ 97

Cécile Brousse

9 The statistical backbone of the new European economic

governance: the Macroeconomic Imbalance Procedure

Scoreboard 109Gilles Raveaud

10 Evaluating public policies or measuring the performance

of public services? 129Florence Jany-Catrice

Part III Uses of Quantification : Power and Resistance

11 Private accounting, statistics and national accounting

in France: a unique relationship (1920-1960s) 139

Béatrice Touchelay

12 Figures for what purposes? The issues at stake

in the struggles to define and control the uses of statistics 149Marion Gilles

13 The uses of quantification: power and resistance The example

of unemployment statistics 161Pierre Concialdi

14 Statistical argument: construction, uses and controversies

Prices and purchasing power 171Alain Gély, Bernard Sujobert and Béatrice Touchelay

15 The quantification of the social sciences: an historical

comparison 183Alain Desrosières

List of contributors

Michel Armatte is a statistician and economist, teacher in Paris-Dauphine University

and research worker on history of statistics, probability and econometrics at Centre A Koyré

He was a long date compagnon of Alain Desrosières His last book is on Economics as an Engineering (2010)

Fabrice Bardet is a senior researcher at University of Lyon (ENTPE) He is specialist on sociology of quantification and urban development He holds a PhD in Political science from

the University of Paris 1 La Sorbonne (2000) and the habilitation to be PhD supervisor (HDR)

from Lyon 2 University (2013) He has recently published La contre-révolution comptable Ces chiffres qui (nous) gouvernent (Paris, Les Belles Lettres, 2014)

Luc Boltanski is director of research in sociology at the French School for Advanced Studies

in Social Sciences (EHESS) Regarded as a leading figure in critical thought and what has come to be known as French pragmatic sociology, his influence in the fields of sociology, political science, economic history and social economy has been wide-ranging Among his

latest published works in English are, to name but a few, Mysteries and Conspiracies Detective Stories, Spy Novels and the Making of Modern Societies (Polity Press, 2014), On Critique: A Sociology of Emancipation (Polity Press, 2011) or The New Spirit of Capitalism

(with Eve Chiapello, Verso, 2007)

Cécile Brousse was a senior official at the French National Institute for Statistics and

Economic Studies (INSEE) and a researcher at the Center for Research in Economics and Statistics (CREST) She is currently working at the Directorate for Research, Studies and

Statistics (DARES) in the French Ministry of Labour, Labour Relations and Solidarity

Isabelle Bruno is associate professor in political science at University of Lille (France) and

member of the Lille Centre for European Research on Administration, Politics and Society (CERAPS, CNRS/Lille 2) Among other research interests, she works on benchmarking as a technology of fact-based government and on EU processes of quantification She recently co-edited with E Didier and T Vitale a special issue on « Statactivism: State Restructuring,

Financial Capitalism and Statistical Mobilizations », published in Partecipazione e Conflitto The Open Journal of Sociopolitical Studies (vol 7, n° 2, 2014)

Pierre Concialdi is currently working as an economist at France’s Institute for Social and

Economic Research (IRES) Earlier he worked as a deputy director at the Center for Research

on Incomes and Costs (CERC) His studies have focused on income inequalities and poverty, social protection and the financing of social security, tax progressivity, low-wage employment and unemployment

Roser Cussó is Professor at University of Paris-I She works on the history and sociology of

international quantification (statistics produced by intergovernmental and supranational organizations), and more generally on international and transnational technical activity Her interest is specially focused on the impact of expertise on the public debate and on the

ix

functioning of democracy Her main research areas are education, population, and minorities,

as well as related economic questions

Alain Desrosières (1940-2013) is the founder of the social history of statistics After

graduating from France’s Ecole Polytechnique, he was concerned about social issues and, therefore, chose to study at the National School of Statistics and Economic Administration (ENSAE) He became an administrator at the INSEE in 1965 His commitment to the dissemination of statistical information and his interest in history led him to play a key role in the development of the critical approach to statistics Analyzing in detail their conditions of

production, their uses and their impacts on societies, Desrosières published in 1993 The politics of large numbers : a history of statistical reasoning (Harvard University Press, 1998)

as a first synthesis of his thought and research Concerned with the effects of neoliberalism on democracies, he dedicated work to the role and place of statistics in new forms of governance, notably marked by the withdrawal of the Welfare State His studies were published in 2008 in

the two volumes of The Statistical Argument Desrosières has played a leading role in the

evolution of the social sciences of quantification both in France and in the Anglo-Saxon countries

Emmanuel Didier is a former student of Alain Desrosières and became a friend even more

than a colleague He is a CNRS researcher and at the same times a visiting professor at the UCLA Institute for Society and Genetics where he acts as the deputy director of a joint laboratory called EpiDaPo In addition to writing books on the socio-history of quantification which is his main topic, Emmanuel Didier edited and wrote the introduction to the last book

of Alain Desrosières 2014, Prouver et gouverner, un analyse politique de la statistique publique La Découverte, Paris

Alain Gély is a public statistician, former INSEE official and CGT activist involved in the

National Council on Statistical Information (Conseil national de l'information statistique or

Cnis) He takes part in the “comité du label”, committee which issues the certificates of general interest and statistical quality in the public statistical surveys

Marion Gilles is a sociologist, research officer in French National Agency for the

Improvement of Working Conditions and a Research Associate at the Max Weber Center (University of Lyon) She has defended a thesis at the EHESS on the process of quantification

of health at work and its effects within companies, entitled "Counting to weigh in? The quantification of ‘health at work’ : in between institutionalization and conceptual ambiguities"

Florence Jany-Catrice is full Professor in economics at the University Lille (France) She

conducts her researches at the Clersé (CNRS-UMR8019) She is an honorary member of the

Institut universitaire de France (IUF, 2006-2012) Her researches concern mainly the

economy of quality in the field of the service economy: quality of labor and employment;

quality of output, wealth and their statistical measures She published The New Indicators of Well-Being and Development (Palgrave McMillan) 2006, with J Gadrey; and La performance totale: nouvel esprit du capitalisme? (2012, Presse Universitaire du Septentrion)

Étienne Pénissat is a researcher (CERAPS, CNRS) He conducts research on the sociology

of quantification He worked on the construction and the uses of labor statistics in France since 1945 He has also investigated the role of social statistics in the construction of the European Union He published with Jay Rowell: « The Creation of a European Socio-

economic Classification: Limits of Expert-driven Statistical Integration », Journal of European Integration, 37 (2), 2015, 281-297

Gilles Raveaud is associate professor of economics at the Institute for European Studies of

the University of Paris 8 – Saint-Denis A co-founder of the French Post-Autistic Economics Movement, he has specialized on the defence of pluralism in economic teaching (see

“Teaching Introductory Economics in a Pluralist Way: Why and How”, in Robert GARNETT,ErikOLSEN,MarthaSTARR,(eds.), Economic Pluralism, London, Routledge, 2010, pp 250-

261) He has also done critical work on the European Employment Strategy (see “The

European Employment Strategy: Towards More and Better Jobs?”, Journal of Common Market Studies, 2007, Vol 45, N° 2, pp 411-434)

Bernard Sujobert is a public statistician, former INSEE official and CGT activist involved in

the National Council on Statistical Information (Conseil national de l'information statistique

or Cnis) He is also a member of the board of governors of the Institut pour le développement

de l’information économique et sociale (IDIES)

Béatrice Touchelay is professor of history at the University of Lille (France), a researcher at

the Institut d’histoire du Septentrion (IRHiS/CNRS) Specialist of history of public statistic,

of history of private data and accounting, professionals of accountants, accounting practices

of enterprises, she studied also history of taxations and publics policies in France and Europe, lobbying of businessmen and associations and development of expertises

xiList of contributors

I Bruno – F Jany-Catrice– B Touchelay

Université de Lille, CERAPS – CLERSÉ - IRHIS

Lille, France

e-mails : isabelle.bruno@univ-lille2.fr ; florence.jany-catrice@univ-lille1.fr ; beatrice.touchelay@univ-lille3.fr

Introduction The Social Sciences of Quantification in France: An Overview

Isabelle Bruno, Florence Jany-Catrice and Béatrice Touchelay

Abstract This edited volume contains various contributions from economists, historians, political scientists, sociologists and statisticians, all of whom share the same approach to

quantities Whether they be dealing with national, European or international statistics on the

homeless, occupational health or economic governance, they all emphasise the extent to which the numbers are based on conventions and they all call into question their assumed obviousness by examining the exercises in quantification that produced them To put it another way, they all explore the 'black boxes' constituted by the indicators, categories, scoreboards and other accounting or statistical tools that serve both as evidence and as instruments of government This is the fundamental lesson to be drawn from the work of Alain Desrosières, statistician, historian and sociologist of statistics, whose work has inspired and guided a large number of studies since the end of the 1990s The aim of this book is to give a general idea of the research that has its origins in the approach pioneered by this eminent thinker on large numbers and the politics underlying them

This edited volume1 contains various contributions from economists, historians, political scientists, sociologists and statisticians, all of whom share the same approach to

quantities Whether they be dealing with national, European or international statistics on the

homeless, occupational health or economic governance, they all emphasise the extent to which the numbers are based on conventions and they all call into question their assumed obviousness by examining the exercises in quantification that produced them To put it another way, they all explore the 'black boxes' constituted by the indicators, categories, scoreboards and other accounting or statistical tools that serve both as evidence and as instruments of government (Desrosières 2014) This is the fundamental lesson to be drawn from the work of Alain Desrosières, statistician, historian and sociologist of statistics, whose work has inspired and guided a large number of studies since the end of the 1990s The aim of this book is to give a general idea of the research that has its origins in the approach pioneered

by this eminent thinker on large numbers and the politics underlying them (Desrosières 2002)

1 We are deeply grateful to Richard Sobel for having suggested the idea for this book and to Marie-Aude Depuiset for managing the whole enterprise We warmly thank Andrew Wilson for the high quality translation of

most chapters This book received the support of the French program “ADA” (Argumenter, décider, agir) carried

by the “Maison Européenne des Sciences de l’Homme et de la Société” (USR 3185, Lille, North of France) The ADA program received some financial support from the “Métropole Européenne de Lille” (2011-2014)

1

© Springer International Publishing Switzerland 2016

I Bruno et al (eds.), The Social Sciences of Quantification, Logic, Argumentation

& Reasoning 13, DOI 10.1007/978-3-319-44000-2_1

1 The development of the social sciences of quantification

The social sciences of quantification encompass three convergent approaches historical, socio-political and socio-economic) that are mutually enriching

(socio-1.1 A socio-history of quantification

When it is open to the other human and social sciences, history occupies an important place in the 'French school' of the social sciences of quantification that was founded by Alain Desrosières Any attempt to understand historical phenomena, which offer particularly fertile terrain for approaches that foreground tensions and the combining of perspectives and approaches that Desrosières valued very much, requires a mix of tools from practitioners in different disciplines As Desrosières put it, the analysts concerned need to wear a number of different hats and to don several pairs of spectacles in succession His interest in history, both that of the French National Institute for Statistics and Economic Studies (INSEE) and that of the French public statistical institutions that preceded it, was due in no small measure to his curiosity about people, about the actors in an institution in which he spent the whole of his working life This curiosity impelled him to go and interview Henri Bunle (1884-1986), for

example, who had joined the Statistique générale de la France just before the First World War and who was appointed acting director of the Service National des Statistiques following

the arrest of René Carmille (1886-1945)2 in 1943 Another document that Desrosières was fond of was the photograph of Henri Bunle blowing out the 100 candles on his birthday cake

at the ceremony organised to mark the occasion in 1984, which was held in the reception rooms of the Ministry of Finance in the Rue de Rivoli and attended by both Francis Louis Closon, INSEE's first Director-General, and Alfred Sauvy (Desrosières 2003b) Out of affection for his institution, but without allowing his gaze to lose any of its critical acuity, he played a large part in illuminating this history A number of articles3 outlining the evolution

of the French public statistics institutions, for example the one on the higher vocational

schools, l’école d'application (Desrosières 1995), comparing the French system to that in

other European countries or analysing the evolution of studies on the history of statistics (Desrosières 2003) helped to shine a light into some hitherto obscure corners He was the instigator, with others, of the workshops on the history of statistics organised by the INSEE research unit at Vaucresson in June 1976, and of the publications that followed (Affichard 1987) These workshops enabled links to be established between historians and practitioners and highlighted the value of the historical approach in giving practitioners the detachment required to use the tools of quantification His involvement in gathering together illustrations and writing the book commemorating the 50th anniversary of INSEE's founding, which was distributed internally in 1996 (INSEE 1996), gave substance to this heritage The addition of

2 René Carmille, comptroller general of the French Army, was the founder of the Demographic Department that aimed to mitigate the abolition of the military recruitment offices after the armistice of July 1940 Carmille then managed the SNS, which was created out of the merger of the Demographic Department and the SGF in October

1941

3 Articles published in Économie et statistiques or in Courrier des statistiques, a journal, of which Alain

Desrosières was a pillar, that sought to disseminate the results of statistical studies widely and had a very strong historical slant On the organisation of European statistical systems, see for example (Desrosières 1999, 3-4)

I Bruno et al.2

an historical perspective to the annual national statistics workshops (Journées de la statistique) which he played a major part in organising, helped to ease the INSEE's past His

commitment to preserving the evidence of the institution's past4 and his interest in historical studies helped to show the tensions and uncertainties linked to INSEE's past Alain Desrosières shared his taste for history with Raymond Lévy-Bruhl, INSEE's Secretary-General during the chairmanship of Edmond Malinvaud, whereas their jobs tended rather to bring them into conflict (Desrosières and Touchelay 2008) They were among those who made it possible for the archives lodged at the Ministry of Finance's Economic and Financial Archives Centre at Savigny-le-Temple to be consolidated, which included both the INSEE archives and those of the descendants of the National Statistical Service’s founders5

Desrosières' taste for history was combined boldly with a sociological approach that was alert to the actors, their positions and their dominance relationships In this way, he opened the way to research in social history and political sociology focusing on the power struggles around the production of 'large numbers'

1.2 Quantification: between science and politics

As paradoxical as it might seem, statistics is not one of the classic objects of investigation in political science This appears paradoxical in the light, firstly, of the history

of statistics, which is coextensive with the construction of the state, as the word’s etymology suggests6 and, secondly, of the questions of power and power relations that are an inherent part of the production of ‘large numbers’ However, the paradox dissipates when it comes up against two common presuppositions In the first, a certain scientific neutrality is attributed to

statistics, which are assumed to be objective and impartial, in other words apolitical In the other, they are regarded above all as persuasive data that may serve as support for an

argument, as evidence and as tools for scientific proofs themselves or for the actors that are the objects of such proofs Such presuppositions are obstacles to an understanding of the political dimensions of statistics, obstacles that the sociology of quantification makes it possible to overcome The discipline developed in France around Alain Desrosières, who himself drew inspiration from, among other things, science and technology studies, on the one hand, and governmentality studies, on the other

This approach is based on two basic postulates The first is that data are not, as their Latin etymology might suggest, ‘something given’ Rather, they are constructed in accordance with certain procedures, using certain measurement tools and with numerous choices being made throughout the entire process of quantification Quantifying is a task, a social activity,

4 For example, he lent Béatrice Touchelay a large hand-written exercise book containing all the minutes of the SGF's technical committee (from 1917 to 1937), of which he took great care

5 For example, the collection lodged at the Centre des archives économiques et financières de la France

(Savigny-le-Temple, 77) by Xavier Jacquey, which he had inherited from his father, Pierre Jacquey, who was a close collaborator of René Carmille at the time when the Demographic Department and then the national statistical service were being set up under the Vichy regime The documents are available at shelf marks 5A-

an inextricably technical and social practice It is technical in that it involves measurement and social in that it involves agreement and convention, hence the now classic equation whereby to quantify = to agree + to measure (Desrosières and Kott 2005) The second is that science and politics are not mutually contradictory but have to be conceptualised together

(Latour 2008) More specifically, just as political scientists examine processes of politicisation and depoliticisation, so it is necessary to shed light on the processes leading to the politicisation, technicisation and quantification of objects and problems These are eminently political processes, if only because quantification tends to ‘depoliticise’ its objects,

to render them undebatable and to remove them from the workings of democracy

Over the past ten years, an increasing volume of research, theses, publications and seminars has been devoted to the issues at stake in quantification, its uses and its effects on government practices and public action Over the last 15 years, the political sociology of quantification has become a thriving area of research in France It belongs to an older, international tradition of studying ‘statistical reasoning’ and its categories As noted above, this is a tradition more rooted in history Besides Desrosières’ work, mention should also be made of that of Martine Mespoulet (2008) and Alain Blum (2003) on Soviet statistics, Giorgio Israel on ‘the mathematisation of the real’ (Israel 1996) or of Theodore Porter, the

author of standard texts on the subject such as The Rise of Statistical Thinking, 1820-1900 and Trust in Numbers: The Pursuit of Objectivity in Science and Public Life (both published by

Princeton University Press in 1986 and 1995 respectively) We do not claim that this list constitutes an exhaustive genealogy of the sources that inspired a new generation of researchers in the political sociology of quantification Rather, it is intended simply to underline their dynamism by providing a small handful of examples

It is on this breeding ground that a series of studies on the ‘politics of quantification’ (Bardet and Jany-Catrice 2010), including those contained in the present volume, has flourished since the beginning of the 2000s Over and above the production of statistics, sociologists of politics have concerned themselves with the reception given to the figures produced and the uses made of them, both by governments, with their fondness for indicators, barometers, dashboards, rankings and benchmarks as means of shaping public action7, and by the governed, within the context of a form of activism that might be described as

‘statactivism’(Bruno et al 2014a; Bruno et al 2014b) They have investigated a number of different areas whose diversity reflects the proliferation of quantification activities in most spheres of state intervention (police, employment, education, research, health, etc.), at local and national, European and international level The chapters in this volume reflect this dual diversity

The production and use of statistics originate in quantification processes that are political, since they are a locus for conflicting interests and mutually antagonistic visions of the world They also generate political effects resulting from the exercise of power as well as

from acts of resistance to that same power This insight may appear obvious to some, but is by

no means self-evident to many others Training in political science, as in most of the other

7 This approach is illustrated by the chapters by Roser Cusso at international level, Cécile Brousse and Gilles Raveaud at European level and Étienne Pénissat in the case of France

I Bruno et al.4

social science disciplines, usually ignores these questions around the production of statistics

It would be advantageous to raise awareness of this insight, since it helps to create a critical mind and a valuable reflexivity These qualities are useful, both for citizens inundated with figures by the media and those who govern them as well as for professionals who have to use statistics as decision-making tools or to evaluate their own performance, whether they work in private companies, in public administration or even in an association operating in the social and solidarity economy

There is another insight that deserves to be more widely disseminated in addition to the lessons to be drawn from social history and the political sociology of quantification, namely that provided by socio-economics

1.3 A rich and creative contribution taken up by too few economists

With the exception of those claiming affiliation to the French convention school8, very few economists have been keen to adopt the socio-economic approach to quantification that Desrosières’ work encourages After all, adoption of such an approach requires an unpresuming and reflexive re-examination of tools, methods and research practices more generally In short, it demands a new, more penetrating epistemology that mainstream economics, which aspires to the status of ‘science of measurement’, is unfortunately increasingly keeping at arm’s length9 The discipline of economics contains within it the epistemological duality underlined by Desrosières On the one hand, there is a realist epistemology, favoured by orthodox action theories As Desrosières (2000) notes, the starting hypothesis here is that reality and truth pre-exist measurement and are revealed by it It is a position that permeates facts and research questions It is all the more widespread since some

of those who argue that facts are constructed end up reaching an accommodation with the realist position, in order to maintain the effectiveness of their thinking and practice On the other hand, acknowledgement of the constructed nature of the statistical apparatus and its measurements reveals the choices that are made when the apparatus is converted into

‘cognitive montages’; it also reveals the constraints and practices that shape the implementation and uses of that apparatus

Metrological realism has become the nerve centre of a section of the social sciences The 2012 book by Etienne Wasmer (former winner of the prize for the best young economist

in France and professor at Sciences po) and Marc Ferracci (from the Centre de recherche en

économie et statistique) entitled État moderne, État efficace is a paradigmatic example

(Ferracci, Wasmer 2012) The authors’ argument, which combines ideas from political science with mainstream economics, is based on the notion that only a state evaluated by

‘scientific methods’ is at all likely to behave rationally and to be ‘a good state’ From their perspective, scientific methods are those that come as close as possible to counterfactual reasoning and make use of innovative statistical and econometric techniques Their argument,

8 The French school of economic conventions has been progressively created by Olivier Favereau, François Eymard Duvernay, Robert Salais, André Orléan, but also Jean-Pierre Dupuy (philosoph) and Laurent Thévenot (sociologist), some of them being trained by the French ENSAE (National School of Statistics)

9 For a critique of the epistemology of the science of measurement, cf for example (Mouchot 2003)

1 Introduction The Social Sciences of Quantification in France: An Overview 5

which is more subtle than it appears and undoubtedly more effective as well, is a plea for the state to agree to allow itself to be evaluated by these methods However, this is complicated, since public and collective action is often multidimensional The authors respond, in effect, that this is not a problem: in order to become modern and efficient, the state has to agree to

engage only in actions susceptible of accurate measurement Citizens will then have quantified proof of a ‘causal link’ between state action and its effects: ‘Thus the objectives [of

public policies] have to be simplified; they should be limited to a small number of quantifiable indicators whenever possible and the promoters of the legislation in question should be made accountable for the results’ (Ferracci, Wasmer 2012, 170)

Constructivism, for its part, is a common denominator for a large section of the other social sciences, and serves as a starting point for a productive dispute with mainstream voices For institutional reasons10, only a small number of academic economists adopt a constructivist position, namely those who consider economics to be a social science11 For several decades now, some researchers, while favouring interdisciplinary approaches, have been opening up the ‘black boxes’ of a number of public statistics, thereby triggering disputes about what counts and what should be counted The targets of their criticisms are usually indicators that have become both the sole permissible representations of progress and, at the same time, revealing of the major crises that characterise contemporary capitalism The debates that have had particular resonance internationally include those around the measurement of wealth (sustained in particular by the work of Jean Gadrey and Florence Jany-Catrice in Lille, Dominique Méda in Paris, Anne Le Roy and Claudine Offredi in Grenoble and Isabelle Cassiers and Géraldine Thiry in Belgium); representations of unemployment and employment (to which Robert Salais, Nicolas Baverez, Bénédicte Reynaud (2013) and Jérôme Gautié (2002) have made notable contributions); or even the calculations and representations of the debt Opening up the ‘black box’ that is the historical process leading to the construction of the indicators and the statistics on which they are based gives substance to the choices made, making it possible to escape from the determinism of the assumed laws of economics and potentially giving rise to a plurality of measurements, whether of inequalities (of income or property)12, poverty (absolute and relative inequalities, living conditions, etc.) or wealth (Stiglitz-Sen-Fitoussi Commission) They may also give rise to competing measures produced

by organised civil society This was the case in France with the CGT trade union federation,

which constructed a consumer price index in the 1970s, the organisation ACDC (‘Les autres chiffres du chômage’/Alternative unemployment figures) in the 2000s, with its alternatives to the ILO convention on measuring unemployment, and the Réseau d’alerte sur les inégalités

(RAI or Warning on Inequalities), which produced alternative measures of inequalities

10 See the studies by the French Association for Political Economy (Association française d’économie politique, AFEP) which, in its diagnosis, reveals the institutional reasons that have led to the gradual stifling of pluralism

of thought, methods and theories in economics This eradication of heterodox thinking is obviously not confined

to France http://assoeconomiepolitique.org Accessed 23 February 2016

11 They now have a famous advocate with Thomas Piketty’s most recent postures

12 See, among others, (Piketty 2014)

I Bruno et al.6

2 Meetings that gave birth to a common project

United under the banner of the ‘social sciences of quantification’, a whole community

of economists, historians, political scientists, sociologists, management specialists and statisticians is now involved in re-examining the problems and concepts discussed in Desrosière’s work, which has served as a link between these various disciplines and inspired a profusion of exchanges and collaborative ventures in various places and on various subjects One such event was a workshop entitled ‘From the politics of large numbers to target-driven politics Statistical argument in the era of the New Public Management’ was held on 15

March 2013 at the Maison européenne des sciences de l’homme et de la société (MESHS) in

Lille, North of France It brought together researchers from various disciplines, some established, others at the start of their careers, with a view to entering into dialogue on the question of quantification It was this workshop that gave rise to the book you are holding in your hands

The starting point for this book was a contemporary issue we regard as fundamental, namely the ‘target-driven politics’ associated with what is also known as the ‘modernisation’

of the state and the ‘stabilisation’ of its finances, notably by means of French and European

laws and measures such as the organic or fundamental law on the finance acts (Loi organique relative aux lois de finances or LOLF), the general review of public policies (Révision générale des politiques publiques or RGPP), the modernisation of public action (Modernisation de l’action publique or MAP) and the European budgetary pact Our

hypothesis is that this ‘target-driven politics’ has replaced the ‘politics of large numbers’, i.e the use of macroeconomic data for economic forecasting and planning, that characterised the social state in the decades following the Second World War Desrosières highlighted this shift

in the objectives of statistical argument Regarded for a long time as a convention resulting from a compromise and endowed with scientific reality, it is now increasingly seen as a depersonalised and controversial technique used to support the arguments of those in power More broadly, a number of studies have explored the relationships between neoliberal governmentality and quantification technology On the one hand, authors have analysed the symptoms of neoliberalism, which is characterised in particular by a quest for instrumental organisational rationality that is to be applied in exactly the same way to market and non-market organisations alike They have revealed the disastrous role played by policies based on disembodied figures that merely provide support for this instrumental rationality On the other hand, new thinking has emerged, some of it drawing on the analytical framework developed

by the French ‘convention’ school, particularly around the ‘economics of worth’, with multidimensional measures of performance being proposed Taking as their starting point the plurality of registers in which the actors in question operate, these studies show the extent to which these instruments of power can be circumvented and quantification used to support programmes of resistance or even alternative policies

Without seeking to be exhaustive, the aim of this book is to show how fruitful and productive this multitude of interdisciplinary perspectives can be To this end, it includes contributions from specialists in various disciplines, including sociologists, historians, economists and political scientists, as well as statisticians, management specialists and trade

1 Introduction The Social Sciences of Quantification in France: An Overview 7

unionists These authors analyse the ways in which quantification is being deployed at all levels of public and collective action They explain the effect quantification has on the social world and show how the actors promote it, adapt to it or resist it

The articles gathered together in this book are based on several observations The first is that societies have been overtaken by ‘quantophrenia’ (De Gaulejac 2005), which some

commentators have no hesitation in describing as pathogenic (Supiot 2012) As a

consequence of this ‘quantophrenia’, that which is not counted ends up being gradually discredited This is the substance of the arguments advanced by authors such as Dominique Méda (1999), Marilyn Waring13 or Jennifer Bué, Isabelle Puech and Thomas Coutrot (2004)

This observation provides the stimulus for a reinterpretation of target-driven policies and an analysis of the networks of actors who use statistics as their sole mode of argumentation These contributions are also all based on the insight, which is due largely to Desrosières, that indicators, when consolidated through processes of institutionalisation and legitimation that are often specific to them, become constraints and resources comparable to legal rules and norms These processes both legitimise what is constructed and bury the meta-data and conventions that served as a basis for the construction of the indicators The ways in which the large international organisations (OECD, Eurostat, etc.) present their statistics are most certainly extreme examples of this concealment These processes of institutionalisation and legitimation turn certain key indicators (purchasing power, the consumer price index, GDP, unemployment rate, etc.) into coordinating mechanisms and reference points for (individual and collective) action

Inspired directly or indirectly by Desrosières’ work, all the contributions are also an invitation to maintain the tension between an internal and an external position One way of embodying this tension is the idea that it would be intellectually and politically ineffective to reject measurement solely on the grounds that it will always be manipulated, manipulable or the result of a social construction; this after all places researchers in a radically critical position, the future of which is extremely uncertain On the contrary, the implicit message of the various contributions to this book is that, under certain conditions, it is necessary to take the risk of reinforcing the analysis and proof with alternative figures, which may be intellectually useful and politically productive This position is closely akin to that opened up

by ‘statactivism’ (Bruno et al 2014a; Bruno et al 2014b) Deconstructing or revealing everything that is alienating in, for example, target-driven modes of performance is an exercise that the social sciences, or at least those that have lost their reflexive capacities over the years, should resume in order to develop a plan for freeing themselves from such thinking While statistical input is not sufficient, it is certainly a necessary condition for revisiting theories and research practices suffused with individualist, utilitarian and market ideology

13 See Who's Counting? Marilyn Waring on Sex, Lies and Global Economics, National Film Board of Canada,

1995

I Bruno et al.8

3 Outline of the book

The book opens with a summary of Alain Desrosières' contributions to the sociology of quantification The chapters in the first part contain testimonies from several generations of researchers who were profoundly influenced by Desrosières' work The five authors all enjoyed varying periods of real intellectual collaboration with Desrosières Michel Armatte is

a statistician and was Desrosières' 'accomplice' in many seminars at ENS and elsewhere ('Introduction to the work of Alain Desrosières: the history and sociology of quantification') In a very personal chapter, the sociologist Luc Boltanski pays homage to the work of his friend and colleague, Alain Desrosières by situating his career, preoccupations and work in the intellectual context of his time He shows how his research on quantification fuelled the debates in the social sciences which, from the 1970s to the first decade of the 21stcentury, raised and revisited the questions of social reality, veridiction and critique, as well as those of power and institutions, and in particular the power and institutions of the state Thus the unique position that Desrosières contrived for himself within the social sciences is

‘indissociably epistemological and political’, since his sociology opens up the ‘black box’ of numbers in order to initiate public debate and not to close it off, thereby facilitating a shift from numbers that claim absolute validity to what Boltanski calls ‘reflexive’ numbers’ Emmanuel Didier is a sociologist who was very close to Desrosières; he compiled and wrote the introduction to the collection of articles published in the posthumous collection cited above (Desrosières 2014) He also assisted Desrosières in coordinating seminars at the EHESS on the 'politics of statistics' ('Alain Desrosières' spectacles: one lens realist, the other constructivist') Fabrice Bardet is a political scientist who was supervised by Desrosières when researching his doctorate on the influence of the central statistical service on defining the boundaries of the French regions in the second half of the 20th century ('Learning from the history of the probabilistic revolution: the French school of Alain Desrosières') Roser Cusso is a sociologist who benefited from Desrosières' presence on the examining panel for

her accreditation to supervise research (habilitation à diriger des recherches)14 ('From

"international statistics " to "quantification in international organisations": a dialogue with Alain Desrosières') These authors explain, each in his or her own way, how Desrosières influenced their intellectual trajectories and describe the part of the journey they travelled together This tribute, based on personal testimonies, does not claim to be exhaustive, but it does bear witness to the vitality of the school of thought and analytical framework Derosières initiated

The second part takes as its subject statistical argument in the neoliberal era Far from being set in stone, the forms of statistical argument vary in space and over time Alain Desrosières identified several configurations of the state, which he characterised in terms of different, historically situated combinations of statistical forms, modes of action and ways of conceptualising the social sphere His purpose in identifying these different configurations was to 'endogenise the construction of statistical tools in the context of historical analysis of

14 The habilitation à diriger des recherches (HDR) is the highest degree awarded in the French university system

and authorises holders to supervise PhD theses and to apply for full professorships

1 Introduction The Social Sciences of Quantification in France: An Overview 9

the forms of the state' (Desrosières 2003a, 219) In the 1990s, a new 'neoliberal' state emerged

to supplement the four earlier configurations: the engineer state (from the 17th century onwards), the liberal state (from the 18th century onwards), the welfare state (from the end of the 19th century onwards) and the Keynesian state (from the 1940s onwards) This new configuration was characterised notably by its own specific statistical apparatus, which is illustrated in the chapters by Cécile Brousse, Étienne Pénissat and Gilles Raveaud

In the neoliberal era, the production of statistics requires the construction of national spaces of equivalence in which national results obtained as part of intergovernmental coordination programmes can be quantified and compared The construction of such spaces cannot be taken for granted Rather, they are the result of an arduous and deeply political process of categorisation and quantification The attempt to produce a census of the 'homeless'

in Europe, which is examined by Cécile Brousse, is a good example The efforts of Eurostat

to construct an indicator of homelessness on behalf of the European Commission provoked a dispute between the statisticians and the representatives of European charitable associations serving in the working group organised by Eurostat Although the group succeeded in producing a harmonised definition of the homeless category, there were many debates on which measurement tools were to be used In fact, each member state has its own particular way of assisting the homeless and therefore its own way of counting them, as is shown by the diversity of data gathering methods, units of account (individuals or households) and modes

At the other end of the statistical chain, where such harmonised figures are put to use, indicators and dashboards are developed in order to evaluate and monitor national public policies Gilles Raveaud investigates the methodology deployed in the construction of a

battery of indicators incorporated into the Macroeconomic Imbalance Procedure Scoreboard,

which is used to monitor member states' economic governance in order to force them to maintain a certain level of budgetary discipline He demonstrates very clearly how this statistical apparatus shapes the political recommendations made by the European Council This scoreboard is both an evidentiary tool and a tool of government; it provides decision makers with information but also helps to shape their decisions In this case, this form of statistics gives European policies a neoliberal slant

Another characteristic feature of statistics in the neoliberal era, according to Alain Desrosières, is the use of randomised evaluations in public policies (evidence-based policies) This method, which involves comparing a sample in receipt of government assistance with a control sample, was adopted belatedly in France in order to evaluate, for example, the

implementation of the revenu de solidarité active (RSA), which provides a minimum income

I Bruno et al.10

for unemployed and underemployed workers15 Étienne Pénissat highlights the controversies around the scientific nature of such experimental data and investigates the political and bureaucratic issues that they raise by examining the statistical surveys carried out in an attempt to quantify the effects of employment policies on the employed from the 1980s to the 2000s as well as the uses made of them by officials in the French Ministry of Labour By combining the sociology of quantification with a sociology of the actors, he shows how these new forms of statistics helped to transform state intervention in employment policy

The third part, finally, examines the uses of quantification It considers the controversies and the political and bureaucratic power struggles that developed around large numbers and the various forms of resistance to them Statistics are now applied to a varied and expanding range of socio-economic problems In the economic sphere, statistics are obviously a key component of the arguments advanced, both for experts who manipulate the data produced by the public statistical service with a certain degree of dexterity, and for those who use them In the sphere of work and employment, statistics are regularly taken to task because of what they cover or do not cover, because of what they reveal, how they do so and for the ways of looking at social realities that they help to shape

Long deployed in the economic and social spheres (purchasing power, unemployment, work, employment), statistics are also being applied to contemporary issues, which they have, incidentally, played a part in developing, particularly in areas such as the environment or occupational health

The articles in this third part were selected with a particular aim in mind, namely to critique the simplistic idea that statistics are always a tool at the service of the strong and powerful16 Each of the authors demonstrates the degree of conflict that is possible in the production of statistics, as is shown by the history of the measurement of 'purchasing power',

as well as the conflicts around their uses

Alain Desrosières’s career as a statistical engineer and his intellectual curiosity, his attraction to the various human and social sciences and his interest in the link between the forms of the state and the development of public statistics and between the macro and micro levels of the economy and the wider society led him to investigate the connections between public statistics and private accounting Considered as conventions, statistics and private accounting are both the result of a process of standardisation and of a collective desire to have available data that will serve to inform the dialogue between public and private actors Béatrice Touchelay’s chapter extends this approach

In his chapter on 'The uses of quantification: power and resistance', Pierre Concialdi, a researcher at IRES, outlines the main controversies that punctuated the publication of the unemployment figures in France at the end of the 2000s He shows in his shrewd analysis that

a capacity for resistance is essential in a democracy as a means of rallying the necessary

15 Introduced in 2007, the RSA is a social welfare benefit and an element of France’s active labour market policy It is intended to provide individuals aged at least 25 and over (and from 18 to 24 in the case of single parents) and with few resources with a ‘minimum’ level of income The actual level varies with the composition

of their household The specific feature of the RSA is that it is conditional: beneficiaries are obliged to look for work or to map out and implement a career plan

16 A prejudice already demolished by Ted Porter (1995)

1 Introduction The Social Sciences of Quantification in France: An Overview 11

opposition forces and notes that the trade unions continue to be key actors in that resistance However, he also shows that while resistance can be effective in opposing the official unemployment figures, challenging the conventions on which the measurement of unemployment is based is considerably more problematic, since such challenges remain the exclusive preserve of the public authorities In a dialogue between researchers and trade unionists, Alain Gely, Bernard Sujobert and Béatrice Touchelay outline a socio-history of the notion of purchasing power and its concomitant, the consumer price index They describe the main controversies that have punctuated its contemporary history, particularly those that raged at INSEE, and the shifts in the current uses of the notion of 'purchasing power' One of the distinctive features of this chapter is that it sets the deliberations of an academic alongside those of trade unionists in the national statistical institute Finally, in a chapter entitled 'Figures for what purposes? The issues at stake in the struggles to define and control the uses

of statistics', Marion Gilles explores the conflict around the uses of statistics, with particular reference to statistics gathered by occupational health physicians The significance of this chapter is that it shows that questions around processes of institutionalisation and legitimisation, far from being specific to public statistics, are very real issues for data produced and disseminated by a profession 'without the support of any institution' The author emphasises the ambivalent nature of statistical resources, which can sometimes be used in support of arguments about the links between work and health and on other occasions be deployed as benchmarks used by employers Thus she shows that statistics can be hijacked and used against those who produce them

While these three articles obviously do not provide an exhaustive survey of the uses to which statistics can be put, they do all show that, although statistics are frequently used to the advantage of those in power, they can also under certain conditions play a vital role in challenging and resisting both the conventions underlying the measurements as well as the measurements themselves

Conclusion

Alain Desrosières was not only a major thinker about large numbers but also a creator

of bonds, as this book testifies Always willing to go and discuss an idea, to respond to a research proposal, to pose a new question or to present his work to the most varied audiences,

he was able to create synergies, encourage questions and broaden the horizons of all those whom he took under his wing Rather than just leaving us a box of conceptual and methodological tools, he bequeathed us a collective intellectual enterprise that is still flourishing and now extends well beyond the confines of this book The overview it offers is representative, if partial, and will, it is to be hoped, encourage readers to pursue their exploration of the social sciences of quantification through the rich and diverse bibliography that has already been produced

I Bruno et al.12

Affichard, J (Ed.) (1987) Pour une histoire de la statistique (2 vol.) Paris: INSEE

Bruno, I., Didier, E., Vitale, T (eds) (2014b) Statactivism: Statistics and Activism

Participazione e conflitto The Open Journal of Sociopolitical Studies, 7(2), 198-356 Bué, J., Coutrot, T., Puech, I (2004) Conditions de travail : les enseignements de vingt ans d’enquêtes Toulouse : Octarès

De Gaulejac, V (2005) La société malade de la gestion Idéologie gestionnaire, pouvoir managérial et harcèlement social Paris : Seuil

Desrosières, A (2014) Prouver et gouverner Une analyse politique des statistiques publiques Paris: La Découverte

Desrosières, A (2003a) Historiciser l’action publique: l’État, le marché et les statistiques In

Laborier, P., & Trom, D (eds.), Historicités de l’action publique, Paris, PUF, 2003, p

207-221

Desrosières, A (2003b) Naissance d’un nouveau langage statistique entre 1940 et 1960,

Courrier des statistiques, 108, 41-52

Desrosières, A (2002) The Politics of Large Numbers A History of Statistical Reasoning

Boston: Harvard University Press

Desrosières, A (2000) L’État, le marché et les statistiques Courrier des statistiques, 95-96,

3-10

Desrosières, A (1999) La statistique publique dans les pays européens, Courrier des Statistiques, 91-92, 3-4

Desrosières, A (1995) D’une école de statistique et d’économie à l’ENSAE et l’ENSAI:

1942-1996 Courrier des statistiques, 75-76, 47-53

Desrosières, A., & Kott, S (2005) Quantifier Genèses, 58, 2-3

Desrosières, A., & Touchelay, B (2008) Raymond Lévy-Bruhl (1922-2008): un statisticien

innovateur et un serviteur de l’Etat Courrier des statistiques, 124, 49-52

Ferracci, M., & Wasmer, E (2012) État moderne, État efficace: évaluer les dépenses publiques pour sauvegarder le modèle français Paris: Odile Jacob

INSEE (1996) Cinquante ans d’INSEE ou la conquête du chiffre Paris: INSEE

Israel, G (1996) La mathématisation du réel Essai sur la modélisation mathématique Paris:

Seuil

Reference List

1 Introduction The Social Sciences of Quantification in France: An Overview 13

Latour, B (2008) Pour un dialogue entre science politique et science studies Revue Française de Science Politique, 58 (4), 657-678

Méda, D (1999) Qu’est-ce que la richesse? Paris: Flammarion

Mespoulet, M (2008) Construire le socialisme par les chiffres Enquêtes et recensements en URSS de 1917 à 1991 Paris: INED

Mouchot, C (2003) Méthodologie économique Paris: Points

Piketty, T (2014) Capital in the Twenty-First Century Cambridge, Mass.: Harvard

University Press

Porter, T M (1995) Trust in Numbers The Pursuit of Objectivity in Science and Public Life

Princeton: Princeton University Press

Supiot, A (2012) The Spirit of Philadelphia: Social Justice vs the Total Market London,

New York: Verso

I Bruno et al.14

Part I Sociology of Quantification: Alain

itinerary : training at Ecole Polytechnique and National School of Statistics, meeting with

Pierre Bourdieu, first studies on History in Statistics in 1976 and on ocupational and social

categrization with Thevenot in the 1980’s, and publication of his major book of 1993 – la politique des grands nombres, with the new resource he found besaide sciences studies school

and Bielefeld works on the probabilistic Revolution We are then debating of some problems

he assessed like the gap between statistics as tool of proof and as mean of government, between realist or constructivist approaches of statistics, or again the differences between measurement in natural and social sciences, with the prominent part of conventions After

2000 he launched an ambitious program of Sociology of quantification which concerned not only statistics but also comptability

Alain Desrosières passed away on 15 February 2013 at the age of 72 He was my friend, my colleague and my remarkable partner in a research project that I would like to describe here because, above and beyond his human qualities and the history of statistics, to which he was the first person in France to dedicate a specific book (Desrosières 1993), he developed an original approach to analysing the relationship between a society and its instruments of quantification (including statistics and national accounts) that liberated us from the painful oscillations of the 1970s between outright sycophancy and admiration for their formal power to illuminate and guide social progress and the most radical critiques that denounced, not unjustifiably, the power they held over our lives

My companionship with Alain, which dates from the beginning of the 1980s, was born out of our shared experience of training as engineers This training was a channel through which were conveyed the benefits of modernity, as well as its snares and delusions, that is the mystique of progress through science, which was strengthened significantly by the double success of structuralism in mathematics and the social sciences Our re-evaluation of this mystique was aided considerably by a certain reflexive distance inherited from Marxism, the anti-colonial struggles, the events of May 1968 and critical sociology The result was a certain uneasiness with the day-to-day experience of our training and then of our respective institutional situations, in which part of our social identity contradicted our personal identities Alain was a graduate of the Ecole Polytechnique and I had studied at the Ecole Nationale Supérieure des Arts et Métiers and this imposed on both of us a certain obligation to acquiesce in our institutions’ practices and values, which included ‘ragging’ of new students,

traditions, militarism, machismo, elitism, esprit de corps, a technocratic ideology and

17

© Springer International Publishing Switzerland 2016

I Bruno et al (eds.), The Social Sciences of Quantification, Logic, Argumentation

& Reasoning 13, DOI 10.1007/978-3-319-44000-2_2

paternalism, none of which we found congenial The cooperation with Algeria in Alain’s case and with Tunisia in mine was also difficult to stomach The injunctions to cooperate after the wars or processes of independence were shot through with ambiguities How was it possible

to assist without dominating? Who should be assisted, the people or the elites in power? In order to escape this uneasiness and reconcile our professional lives with our critical position towards society, we both took the same route via the ‘humanities’, a combination of history and sociology that enabled us to merge the best aspects of our training (the ability to reason and formalise) with research agendas that enabled us to tackle the societal issues that concerned us so much From that moment on, our venture progressed, partly in the institutions

to which we were affiliated and partly on their margins Thus for more than 30 years I was engaged in a long dialogue with Alain Desrosières on the social history of statistics, which took the form of a point by point discussion of a bibliography, of his writings or of mine, and

of shared teaching at the National School of Statistics (ENSAE)

This collaboration could not be taken for granted because we were not on the same side of the ATM ATMs, the automatic teller machines that are found on the walls of banks, have two points of access One is inside the bank and enables bank staff to fill them up with banknotes; the other is outside the bank, in the street, and is used by customer who, by keying

in their personal identification numbers (PINs), can obtain cash in the form of banknotes, which are conveyed from inside the bank to the outside through a small slot And of course customers know nothing about how ATMs are filled with banknotes and bank staff know only

a little about the customers’ needs that prompt the withdrawals The same applies to the statistical ‘data’ that are produced by some statisticians who, for some users, are analogous to the bank staff who are unaware of all their customers’ needs, while those same users are wholly ignorant of the way in which these ‘data’ are produced In his capacity as a civil servant at INSEE and in his first post overseeing the accounts for ‘lace and guipure17’, Desrosières should have identified with the government statisticians who produce data and post them through the small slot in an ATM to those who seize upon them, such as academics like me, who turn them into what they call ‘evidencing tools’ in other disciplines, such as economics and social sciences, or managers and administrators, who turn them into tools used

to govern populations and things In fact, neither of us conformed to this norm of a defined division of labour between producers and consumers of ‘data’, which of course can scarcely be regarded as ‘givens’, in either economic or epistemological terms However, this made us very much aware of this arbitrary division between producers and consumers and of the need to go beyond it

well-For Alain Desrosières, the project took the form of a commitment to research and to the identification of research questions It is this project whose various stages, and the gains made, I would now like to retrace My aim is not simply to pay tribute to him but also to provide an introduction to his work on the social history of statistics that might persuade readers of the benefits of ridding ourselves of the two opposing caricatures usually associated with it: progress through quantification versus lies, manipulation and alienation Approaching

17 A coarse, large-patterned lace without a net ground

M Armatte18

statistics simultaneously as an ‘evidencing tool and a tool of government’ offers a much more comprehensive and effective perspective for understanding its role, both past and present, in the management of government and of individuals This will offer us an opportunity to initiate

or continue a debate on and an assessment of Alain Desrosières’s arguments and to contrast them with other, more familiar interpreters in this field Let everyone take possession of them, discuss them and enrich them

A graduate of the Ecole Polytechnique (year of admission 1960) who chose to serve the state in a post as close to social demand as possible, Alain Desrosières joined the public administration as a senior civil servant at INSEE, the French national statistical institute Thus

it was with one foot in the vast state apparatus dedicated, in collaboration with the statistical departments of the various ministries, to producing statistics and economic studies and another in the network of sociologists and economists affiliated to the CNRS or to university departments that he constructed an original approach to the use of statistics in society The initial focus of this approach was on the social and economic data produced by the institution that employed him, using an interpretative framework borrowed from Bourdieu, who had been his professor at the ENSAE Alain Desrosières later (2003) explained what he had taken from this ‘unlikely encounter between statisticians and Bourdieu’, a teacher whose intimate knowledge of Algerian society he was very soon to share (Alain did his military service there and returned on official business in the course of the ‘Work and Workers in Algeria’ project (Bourdieu et al 1963), together with a favourable attitude towards statistics as an evidencing tool (an ‘initial legacy’ he called it), which was unusual among critical sociologists, and – finally and above all – a ‘requirement for reflexivity that led me to examine the historically and socially situated dimensions of the technical tools themselves’ (the ‘second legacy’) Like Bourdieu et al (1968), he very quickly convinced himself of the fact that social data are not

‘givens’ in either the epistemological or economic sense, since they are constructed and costly

to produce Rather, they result from a process of knowledge production for which the social and cognitive prerequisites require clarification Alain also adopted the concepts of field and habitus in order to explain the reproduction of inequalities and dominance relationships This doxa derived from Bourdieu constituted the framework for his early studies of the French productive system (Desrosières 1972) and the marriage market (Desrosières 1978), two subjects associated with his early posts at the Department of Enterprise (1965-72) and then at the Department of Population and Households (78-87) and for which he drew on Benzecri’s data analysis techniques and the early deliberations on this methodology with Michel Volle and Ludovic Lebart Alain recently talked about the vogue for such analyses among sociologists (Desrosières 2008) Having outlined the method’s origins in the concept

of correlation, developed in biometrics and psychometrics, he describes it as opposed to the econometric modelling method that was cast in the same mould The former is characterised

by description and exploration, the topological principle (distance and proximity) and the language of individuals, properties and groups, the latter by inductive inference, causation theory and the explanation and language of variables However, he went beyond these epistemological considerations by engaging in an historical survey of the social uses of this

2 Introduction to the work of Alain Desrosi res è 19

method in the representation of social space in the 1970s, providing cartographic evidence of

an equivalence between the space occupied by groups and that in which practices are located This empirically obtained equivalence corroborated Bourdieu’s notion of field, as well as his theory of social reproduction and mobility, but it was also applied to analyses of family budgets (M Glaude) and ways of life (S Juan), or even of ‘socio-styles’ (lifestyles) (B

Cathelat) These analyses brought this social mapping to the attention of the press (the Nouvel Observateur), albeit not without a certain degree of slippage from academic standards,

particularly with regard to the transparency of the processing procedures

Bourdieu’s sociology was still present in the background of his work on occupational categories, which begun when he was at the Research Unit between 1974 and

socio-1978 and was systematised between 1979 and 1982 The father’s socio-occupational classification was ‘the favourite variable for adherents of Bourdieu’s initial legacy’ And so

he set about the task of recasting the socio-occupational classifications as he had been asked

to do, drawing on an early study by Guibert, Laganier and Volle and referring explicitly to

Distinction (Bourdieu 1979) The main idea was to ‘denaturalise’ the categories and to

‘historicise and sociologise the statistical tools, particularly the classifications’ Desrosières (2003) highlights the contradiction that had to be overcome at the time between using numbers in order to obtain a secure scientific foothold and destabilising that foothold by arguing that the evidence was socially constructed by the dominant groups

For Desrosières, the historicisation of statistics began in 1976 Together with Jacques Mairesse, he organised the Vaucresson seminars on the history of statistics that were to lead

to the publication by INSEE of the two volumes (1977, 1987) of Pour une histoire de la statistique After the small number of studies by B Gilles and J Dupâquier, this publication

paved the way for a French history of statistics Alain’s contribution to the 1977 publication was twofold He wrote a short article with Mairesse and Volle on the high points of French statistics, of which March, Carmille and Gruson are the heroes, which owes a great deal to the Michel Volle’s research on ‘the profession of statistician’ (1980) and industrial classifications (1982), which was published some time afterwards He also features in this book as the author

of a comprehensive, well documented report on the history of socio-occupational classifications, in which he discusses the paradox between the apparent mathematical rigour

of statistical records and the no less apparent social complexity of the conventions governing their classification Statistics had found its historian, but one who focused solely on its administrative side, where the ‘data’ were produced The ATM was not transcended Alain distanced himself through formalisation in order better to identify the procedures used to gather and classify data and he was to reveal to me that he did not suspect at the time of the Vaucresson seminars that it would be possible to link it to a history of mathematical statistics

as a method of proof

The sociologisation of statistics, which vied with its historicisation, was an equally important question As early as the beginning of the 1980s, Alain Desrosières sought out sociologists both within and outside his institution Seconded on a half-time basis to the

M Armatte20

Centre de Sociologie Européenne, where he kept company with Michael Pollack and

organised with him a seminar on ‘the social history of social science policies and techniques’,

he put considerable effort into the first Statistics and Sociology seminars, held in October

1982, which attracted 200 participants, half of them from the statistical departments of the various ministries and half from CNRS and university research centres The report on the seminars emphasises the main difficulties encountered in bringing the two disciplines together due to the heterogeneity of their objects of enquiry, values and methods and their client-supplier relationship (the ATM again) However, it also makes the case for overcoming these difficulties through a joint study of the construction of the instruments and procedures of sociological research Thus the dialogue had to be continued Alain was always to describe

himself as a statistician and sociologist

His work on classifications led to the construction, with Laurent Thevenot, of a new

classification (known as the PCS) Les catégories socioprofessionnelles (1988) occupational categories), a short book that was to make his name known to the general

(Socio-public, sets out the arguments underlying the PCS Alain Desrosières drew close to the Groupe de sociologie politique et morale (Political and Moral Sociology Group, or GSPM)

and to Thévenot’s work on the economics of social coding (l’économie du codage social 1983), investment in forms (investissements de formes 1986) and the economics of worth (les économies de la grandeur 1991) in which he examines statistical operations At the same

time, Alain discovered the writings of the convention school (Eymard-Duvernay, Orléan, Salais, Dupuy, Favereau) from which he was to borrow the idea of a multiplicity of conventions of equivalence and a plurality of corporate logics (Salais et Thevenot 1986) And

by way of contrast to the sociology of conventions, historical considerations led him, during this transitional period when the Fordist regime was clearly being abandoned in favour of the

as yet uncertain neo-liberal regime, to investigate the regulationist school as well

Desrosières did not choose between history and sociology At the beginning of the

1980s, a third way presented itself, that forged by the British science studies school, which

was engaged at the time in developing an approach to the sociology of scientific knowledge The strong programme formulated by David Bloor constituted a method of investigation that was causal (i.e explanatory), impartial (agnostic as to what was true or false), symmetrical (in its study of both winners and losers) and reflexive (it applied these principles to itself) Its

adaptation to the French case by Bruno Latour, in Les microbes guerre et paix (published in English as The Pasteurization of France) and Irréduction (1964) (Irreductions in English) and

La science telle qu’elle se fait (1989), and by Michel Callon, in his translation and

actor-network theories, is equally abrasive This approach provides tools with which scholars can distance themselves from the standard epistemology and macro-history and adopt a micro-sociological approach to the actors by studying laboratory controversies and procedures It is also a method for conceptualising both science as a cognitive and social construction and hence for conceptualising the history of statistics, including mathematical statistics, in terms

21

2 Introduction to the work of Alain Desrosi res è

of social issues, translation, conflicts and controversies and the construction of the notions of proof and truth18

The third influence on, or rather impetus for, Alain’s research was the internalist history of mathematical statistics, expounded by B Bru and P Crepel with regard to Laplace and Condorcet in the new seminar on the history of probability and statistics organised by Bru, Barbut and Coumet (with which he was associated from its beginnings in 1983), and Sigler’s 1986 book, which was taken up in my initial presentations at the EHESS19 Alain’s renewed interest in mathematical statistics and probability, despite his lack of interest in modelling, dated from this period The tools made available by the science studies approach and convention theory were to enable him to link two bodies of research findings The first included the results of Quetelet, March and Cheysson’s analyses of administrative statistics, comparisons of different European statistical systems (his admiration for Szreter’s work in the

UK dates from this period) and institutional statistics (relating to the SGF, SNS, CNS, ENSAE), which he began to compile and publish in ‘Histoire de formes’ (Desrosières 1985),

‘Masses individus moyennes” (Desrosières 1988) and ‘Les spécificités de la statistique publique française’ (Desrosières 1989) The second was the body of data on the mathematical history of statistics and probability that he had read in the work of Kendall, Stigler and

Benzecri and in the two volumes of The Probabilistic Revolution, edited in 1987 and 1989 by

Kruger, Daston, Heidelberg, Gigerenzer and Morgan and based on the proceedings of an international seminar held in Bielefeld In this book, Alain found a broad, international and interdisciplinary approach to statistics that for the first time linked the mathematical and statist traditions

It was to this linkage that his major book, La politique des grands nombres, was

devoted It was published in 1993 and has been translated into several languages20, the English version appearing in 1998 This book is an integrated narrative, woven from these two histories of statistics, the social history of institutions and their methods of production, gathering (surveys or registers), coding and recording, on the one hand, and the cognitive history of the formal transformations to which the products of these operations were subjected and their enlistment as evidencing tools (regression or factor analysis), on the other The duality of these two points of view had struck him for a long time This history, which combined the two strands but could not be properly unified as long as the historiographies resisted, is spread out over more than two centuries but reads perfectly well because it does not indulge in any useless erudition; Desrosières forced himself to investigate all the innovations and controversies encountered from the perspective of the sociology of scientific knowledge Yule, for example, transposes the methods of regression analysis used in

18 Even recently, (in ‘Est-il bon est-il méchant’), Desrosières was citing Latour’s Irreductions as a source of

stimulation for conceptualising the notions of convention of equivalence and incommensurability in contrast to the use of long statistical series in quantitative history, the abuses of econometric modelling and the widespread use of benchmarking in the 2000s

19 Ecole des hautes études en sciences sociales, Paris

20 It is said to have been translated into 17 languages, including Uzbek! This information remains to be confirmed, however

M Armatte22

biometry to economics Within what framework does he do this, why and from whom? Alain Desrosières launches into a description of the issues at stake in the treatment of poverty in England at the beginning of the century, of which a regression-based comparative study of indoor and outdoor assistance constitutes one episode He analyses a large number of controversies: German statistics vs political arithmetic; Peuchet’s literary statistics vs Duvillard’s calculations; aleatory contracts vs probabilism and degree of faith, realism and nominalism in the theory of averages, homogeneity vs heterogeneity of populations, Quetelet’s average man vs Galton’s genius, causalism vs contingency in Pearson’s work, French and English administrative statistics, case studies vs surveys, census vs polling, eugenicists vs hygienists, various principles for classifying deaths, various interpretations of chance and the notion of the model in econometrics (Koopmans vs Vining, Keynes vs Tinbergen)

These controversies are strung like pearls on one or two threads that constitute his favoured topics and which often take the form of tensions between two poles The first tension arises out of the duality between the history of mathematical sources and the history

of their uses, between the history of the state and the history of science It is further intensified by the fragmentation of historiographies: the history of ideas and of theories, the history of facts, the history of quantification and formalisation, the history of the production

of information and the history of science and of expertise were at the time still largely confined within their own separate fields (Armatte and Desrosières, 2000) A second conflict, this time methodological in nature, between internalist and externalist histories does not entirely match up with the first It is commonplace to observe that the history of mathematics draws almost exclusively on mathematical concepts and ideas and that the history of the state draws on a wide range of external knowledge in the political sciences However, it is no longer unusual today to view the history of quantification and modelling from a sociological perspective Desrosières is fairly successful in doing this by adopting an anthropological approach to science based on the analysis of controversies, combining in almost every case a cognitive section with one concerned with the social issues at stake

The third source of tension that runs right through the book from introduction to conclusion opposes a realist vision of statistics (it describes a reality) to a constructivist vision (reality is the result of a statistical construction based on conventions) This tension recurs in several historical contexts: what is the legitimacy of a realist interpretation of an average, an indicator, a correlation or a model? How are we to characterise the various ways in which authors have challenged this realism: Moreau de Jonnès’s nominalism, Cournot’s rationalism, Pearson’s idealism, the constructivism of the post-modernists, etc.? This question of realism overlaps with that of the dual cognitive and social nature of statistics, although the one cannot always be reduced to the other: although realism is more common in representations of the

hard sciences, this is not a general rule ‘The gap between technical object and social objects – dating back to the seventeenth century – is now a deep one (…) (and) it is both difficult and indispensable to conceive these objects as simultaneously constructed and real, conventional, and solid For want of such an endeavor, statistical information runs the risk of oscillating

23

2 Introduction to the work of Alain Desrosi res è

endlessly between opposing and complementary states: an undebatable reference situated above these debates, and a target of polemical denunciations destroying the complex pyramid

of equivalences’ (Desrosières 1993, p.398) Is statistics used to describe or as a basis for

action? Is it a ‘science of the state’ or a means of processing large numbers? Does one have to

be a mathematician, a sociologists or even an economist in order to recount its history? Desrosières does not always come down in favour of one or other of these alternatives It is a question of ‘attitude’, he says In each individual case, he suggests certain facts that point in one direction and others that point in the opposite direction

The conclusion of his book returns to these tensions In the first, the opposition is

between ‘two linguistic registers: that of description and science (there is) and that of prescription and action (we must) In the second, it is between ‘two attitudes to the question of

reality, one realist (or objectivist) and the other relativist (or historicist) From these two antitheses, Desrosières derives a four-box grid of the possible combinations Over and above the commonest case, in which realism and the scientific approach are combined, he invites readers to consider the case of a constructivist approach to formal knowledge (which makes it possible ‘to reconstruct a genesis and the social practices that have led to a solid statistical object’) and that of a realist and objective approach to public action, which ultimately has

been the doxa of modernity since Bernouli’s Art of conjecture The final case, that of a

constructivist (or deconstructivist) approach to decision-making, to action and to the management of things and people, is based on the hypothesis that ‘the conventions defining the objects certainly give rise to realities, even though these objects are resistant to proof and

to any attempt to unravel them’ To conclude, statisticians seek to ‘do things that hold’, for both formal and social reasons In other words, they construct objects (indices, classifications, models, etc.) that have both formal qualities that enable them to provide a coherent and realistic representation of reality and social legitimacy as a convention, as an element of society These two properties make them socially useful as links, as bases for negotiation, in short as tools of government In order for these objects (categories, concepts, measures) to hold, they have to be based on conventions that ‘must simultaneously remain undebated, so that life may follow its course, and debatable, so that life can change its course’ It seems to

us that this last sentence in the book presages an attempt to resolve this contradiction through change, though a series of expert regimes combining knowledge and power in a certain way and through the resulting cycle of life and death of social conventions These were to be

themes of his writings in the next decade

In the postface to the second edition of the Politics of Large Numbers (2000),

Desrosières assesses the reception given to his book and takes stock of the studies published since 1993 Thus one can read the long list of publications and collaborations that influenced him He skimmed through the ‘internalist’ histories by Benzecri, Hald and Stigler, decried the

‘Whiggish’ perspective that dominates the history of economic thought and deals with statistics ‘at best as an instrument’ but appreciated the seminal studies by Mary Morgan, Judy Klein and Michel Armatte on econometrics He visibly preferred the more ‘externalist’ studies by Ted Porter on the construction of objectivity, by Margo Anderson and Stephen

M Armatte24

Fienberg, Jean-Pierre Beaud and Jean-Guy Prévost, Silvana Patriarca and Simon Szreter and,

in France, by Luc Blondiaux on polls, Eric Brian on the intertwining of the government and geometry in production of figures in the classical age, de Martine Mespoulet, Emmanuel Didier on national statistical systems and their methods of investigation (registers, censuses, surveys) and by Berlivet, Dodier, Gollac and Spenlauer on the public policy evaluation Philosophers also exerted a notable influence on him, including Hacking, to whose lecture on

constructivism (Construction of What ?, 1999) at the Collège de France he was respondent

However, Foucault, whose lectures (Sécurité Territoire Populations, 1978 and Naissance de la Biopolitique, 1979) were not published until 2004, was not yet part of the landscape Desrosières was fascinated by the Sokal affair, the great controversy of the time Alan Sokal,

a physics professor at New York University, perpetrated a publishing hoax that sought to smash to pieces the deconstructivist approaches of the science studies school Finally, taking stock of the various critiques of his book, he acknowledged that the history of states, the producers of sources useful to their own governance, and the history of the methodology, of the enrolled forms and of the philosophy of reality for which they are the vectors, were not clearly linked in 1993 He noted, however, that the reactions to his book were also distributed around the two poles without ever focusing on both at the same time Alain Desrosières did not yet have any miracle solution for this

The solution was to emerge with an article, which, it seems to me, plays a key role as a crossing point at which the epistemological and political approaches start to be linked This is

the celebrated article entitled The state, the market and statistics, the first version of which, written in 1997, was published in 2000 in the Courrier des statistiques, in 2001 in the Cambridge History of Science (Porter and Ross eds., vol.7), in 2003 in Historicités de l’action publique (Laborier and Trom) and is reproduced as Chapter 3 in the Argument statistique

volume 1, 2008 In this article, Desrosières identifies five ideal-typical forms of the state from the point of view of its role in the economy:

- the engineer state which directly replaces private initiative with state intervention

(Colbert’s royal factories, Ecole polytechnique and the Saint-Simonianism of

post-revolutionary years, New Deal and Manhattan Project in the USA in the 1940,

French economic planning in the 1950s);

- the liberal state trusts in the market and its mechanisms to produce the optimal

state of equilibrium The market economy requires price and production data in order to justify, support and provide a theoretical underpinning for these mechaisms (England of the Corn Laws, France in 1860, agricultural statistics in the USA at the end of the 19th century);

- the welfare state is concerned with the protection of workers against the risks of

unemployment, illness and accident (1880-1930 with the Office du Travail (Department of Labour) and insurance schemes; eugenics; price indices; budget surveys;

- the Keynesian state seeks to put in place the apparatus required to regulate, at a

global level, the balance between supply and demand, and to this end makes use of national accounting, economic modelling and planning;

25

2 Introduction to the work of Alain Desrosi res è

- the neo-liberal state that emerged in Europe after the 1980s reduces state

intervention in economic exchanges to a minimum and relies on decentralised power, the supposed efficiency of market and the supervision of individual behaviour It makes extensive use of incentives and competition

Each of these five configurations, which can be likened to expert regimes that combine knowledge and social control in a particular way, is associated with a favoured set of statistical tools and methods that are used to support the exercise of power and the implementation of public policies Thus data collection technologies and, more generally, population and production statistics, are adapted to the needs of the engineer state; price statistics, together with standard microeconomic theory, are used to support the liberal state; budget surveys and probabilistic insurance techniques are used by the welfare state, while national accounting, consumption and employment surveys and econometric techniques meet the needs of the Keynesian state The crisis of these last two models, which are relatively intertwined, was followed by the shift towards neo-liberalism of the Thatcher-Reagan years that put an end to the methods of forecasting and planning based on macroeconomic data; they were replaced by the widespread use of benchmarking, which is based on the rational expectations hypothesis, competition between economic agents and ex-post evaluations of their performance using sets of indicators that are only weakly linked to each other by logical

or statistical relations The French organic law relating to financial legislation, the ‘open method of coordination’ and the opening up of new markets (‘tradable pollution rights’) at the European level are the most striking examples Here too, Alain drew on the work of numerous researchers who cannot be listed here for want of space

This article constitutes an undoubted advance in Desrosière’s efforts to better articulate the dual nature of statistics as both an evidencing tool and an instrument of government within an historical framework in which sudden breaks and continuities succeed one another However, the principal difficulty is still the matching of these abstract, ideal-typical forms to clearly situated historical regimes, which is both necessary, since the ideal types are merely generalised from historical cases, and impossible, since historical regimes of the same type (Colbertism and Gaullism, Keynesian and Soviet planning) each have their own specific characteristics and cannot be likened to each other

We have now reached the late 1990s, when Alain Desrosière’s devoted himself, quite logically, to dealing with the fallout from his book and, in the first decade of the new century,

to an attempt to synthesis his research themes into a sociology of quantification He inaugurated the first course at ENSAE on the history of statistics (he was soon joined by Michel Armatte), which provided an exceptional opportunity to engage the future managers

of the public statistical service (particularly those destined for a career in the state engineers corps) in a reflexive exercise informed by both history and sociology that would enable them

to see behind the formal objects that they manipulate and understand their role in measuring

and constructing the state In 1992, he joined the editorial committee of Genèse, which gave

him an opportunity to put his approach and arguments across to a wider audience, not only

M Armatte26

through his own publications but also through those of the young scholars who were following in his wake and whose work he followed voraciously In May 2002, he launched the first seminars on the sociology of quantification, which were attended by some 100 participants The presentations were, curiously, divided between two themes: the history of statistics (Porter, Didier, Amossé, Mespoulet, Berlivet, Beck and Armatte) and the history of accounting (Miller, Chiapello, Lemarchand, Godechot and Capron) This programme is testimony to Alain Desrosières’s desire to widen the scope of his research The parallel between statistics and accounting is, after all, very tempting, by virtue in particular of their dual ability to describe and constitute an economic reality by using an equivalent (counting of individuals in the former case, counting of monetary units in the latter) Increasing familiarity with the writings of Miller and, in 2006, the collaboration with Eve Chiapello on a joint article on positive accounting theory were to typify this attempt, which remains incomplete, to establish a wider field of enquiry within which a sociology of quantification could be developed

At the beginning of the new century, he joined the A Koyré Centre as an associate researcher and regularly attended the seminars on the governance of science and climate modelling This was an opportunity for him to incorporate his work on statistics more effectively into a more general approach to science in society in the 20th century and to include modelling in his approaches to quantification (Pestre 2001, Dahan et Pestre 2004, Dahan 2007, Armatte 2010) In February 2006, he contributed to the launch of the first

Seminars on the history of statistics at INSEE; a second series was held in March 2008

Retirement, which he took in 2005, scarcely changed his work habits (he continued to go to his office at INSEE), did not lead to the break-up of his networks and afforded him little additional time to assuage his insatiable curiosity

The articles he published in 2007 for the Centre Cournot and in 2008 with the Presses des Mines are evidence of a final significant advance in his thinking They are focused on his

new slogan: to quantify is to agree + to measure This leitmotiv is the means whereby the

realist and constructivist approaches to statistics could finally be truly linked together by reducing the gap between them: the real could not be denied by an absolute form of

constructivism in which the act of measurement totally created the object (What is not socially constructed? Hacking had asked) and, conversely, the role of quantification in the government

of men and things could not be denied by an absolute form of realism in which things supposedly had an independent existence prior to being measured Bachelard and Bourdieu had already declared that the real is constructed The idea that quantification is based on negotiated conventions, since it requires a certain degree of agreement on the thing that is to

be measured, such as ‘the qualities of the quantities’ for example, banishes the spectre of realism and, symmetrically, that of relativism It surreptitiously introduces the social into the very heart of the logic underlying orders, categories and metrology and gives the lie to the notion that these conventions are arbitrary This approach to quantification can, incidentally,

be virtually duplicated in a very similar approach to modelling, as I suggested to Alain

27

2 Introduction to the work of Alain Desrosi res è

Desrosières’s work in recent years was centred around the publication in 2008 by the Presses des Mines of a collection of articles in two volumes, whose titles ‘Pour une sociologie historique de la quantification’ and ‘Gouverner par les nombres’ clearly reflect the two approaches that he had sought throughout his life to combine: statistics as an evidencing tool and as a tool of government The introduction to the first volume develops this theme, taking

as its starting point the equation ‘to quantify is to agree and then to measure’ The introduction to the second volume, which is entitled ‘Les mots et les nombres Pour une sociologie de l’argument statistique”, takes as its starting point the triptych of state, market and statistics These two introductions, one dealing with a sociologically informed metrology based on the notion of convention, the other with policymaking informed by statistical instruments, specify the scope of this sociology of quantification It is a twofold approach, social and cognitive simultaneously, to the construction of the figures that underlie the imposition both of evidence and of the power to govern In a final article, which might be regarded as a summary of his thinking, Alain Desrosières (2011, 2014) resumed and extended his analyses of the neo-liberal regime, drawing together the threads of his arguments in a highly pertinent way He introduces a new idea, namely that indicators retroactively influence the behaviour of agents, as actors undergoing quantification This idea supplements the notion

of performativity that Michel Callon deploys in order to account for the changes to reality brought about by scientific theory

The moving tribute paid to him by his colleagues at INSEE on the occasion of his retirement, the encomium he received from the historians gathered in Berlin in May 2006 for the study day organised in his honour by the Centre Marc Bloch and the Max Planck Institute for the History of Sciences, and the meeting of his closest friends at Vertrieux on the occasion

of his 70th birthday in 2010 reflect, more objectively than my own personal memories, the aura of respect and friendship that surrounded Alain Desrosières, well beyond the close-knit circles of family and colleagues What stands out from these tributes is not only a body of work that has influenced minds, served as a point of reference and encouraged others to dedicate themselves to scholarship but also a consistent series of portraits of the man, of his qualities and his aptitudes Alain was an excellent reader of others’ work, with very incisive and forceful views on their strengths and weaknesses He was what several people have called

a ‘mailleur’, that is a connector able to bring together diverse intellectual worlds When he

was awarded an honorary doctorate by his colleagues at UQAM, in Montreal, Alain set out very clearly what he owed to others and what he thought he himself had contributed through his work As far as I am concerned, the verdict is clear: I owe him an enormous debt of gratitude, not only for his friendship but also for our intellectual exchanges, the one always going hand in hand with the other And in writing this introduction I hope I have given as many people as possible access to his work

M Armatte28

We're here take a small part of the bibliography of books and articles Alain Desrosières established itself until 2001 and completed by us The * et ** identify articles from tomes 1

and 2 of the book l’argument statistique

Armatte, M (2010) La science économique comme ingénierie Quantification et Modélisation, Paris : Presses des Mines

Bourdieu, P., Chamboredon, J.-C., & Passeron, J.C (1968) Le métier de sociologue, Paris :

Mouton

Bourdieu, P., Darbel A., Rivet J.-P., Seibel C (1963) Travail et travailleurs en Algérie, Paris : Mouton

Dahan, A (Ed.)(2007) Les modèles du futur, Paris : La Découverte

Dahan A & Pestre D (Eds) (2004 Les Sciences dans et pour la Guerre, Paris : EHESS Desrosières, A (1972) Un découpage de l'industrie en trois secteurs Economie et statistique,

40, 25-39

Desrosières, A., Mairesse, J & Volle, M (1976) (with) Les temps forts de l'histoire de la

statistique française Economie et Statistique, 83, 19-28

Desrosières, A (1977) Eléments pour l'histoire des nomenclatures socio-professionnelles In

Pour une histoire de la statistique - Tome I : Constributions, Paris : INSEE, 155-231

Desrosières, A., & Thévenot, L (1979) Les mots et les chiffres : les nomenclatures

socioprofessionnelles Economie et Statistique, 110, 49-67

Desrosières, A (1982) Un essai de mise en relation des histoires récentes de la statistique et

de la sociologie In Actes de la journée d'étude "Sociologie et statistique", Paris : INSEE -

Société Française de Sociologie, Tome 1, 161-182

Desrosières, A (1985) Histoires de formes : statistiques et sciences sociales avant 1940

Revue Française de Sociologie, 26 (2), 277-310

Desrosières, A., & Thévenot, L (1988) Les catégories socioprofessionnelles Paris : La

Découverte

Desrosières, A (1988) Masses, individus, moyennes : la statistique sociale au XIXe siècle

Hermes, 2, 41-66

Desrosières, A (1989) Les spécificités de la statistique publique en France : une mise en

perspective historique Courrier des statistiques, 49, 37-54

Desrosières, A (1990) How to make things which hold together : social science, statistics

and the state In P Wagner B Wittrock & R Whitley (Eds), Discourses on society,

Dordrecht: Kluwer Academic Publishers, 195-218

Desrosières, A (1993) La politique des grands nombres Histoire de la raison statistique,

Paris : La Découverte

29

2 Introduction to the work of Alain Desrosi res è

References

Desrosières, A (1994) La commission et l'équation : une comparaison historique entre les

Plans français et néerlandais Note INSEE-CREST, n° 36/J310, du 31 mai 1994

Desrosières, A (1997) Refléter ou instituer : l'invention des indicateurs statistiques In

Dupoirier, E & Parodi, J.-L (Eds), Les indicateurs socio-politiques aujourd’hui, Paris :

L’Harmattan, 15-33 *

Desrosières, A (1998) The Politics of Large Numbers A History of Statistical Reasoning,

Cambridge: Harvard University Press

Desrosières, A (1999), La commission et l’équation : une comparaison des plans français et

néerlandais entre 1945 et 1980 Genèses, 34, 28-52 **

Desrosières, A., & Armatte, M (2000) Méthodes mathématiques et statistiques en économie :

nouvelles questions sur d'anciennes querelles In Beaud, J.-P., & Prevost, J.-G (Eds.), L'ère

du chiffre, systèmes statistiques et traditions nationales Montréal : Presses Universitaires du

Québec, 431-481

Desrosières, A (2000) L’État, le marché et les statistiques Cinq façons d’agir sur

l’économie Courrier des statistiques, 95-96, 3-10

Desrosières, A (2000) L’histoire de la statistique comme genre : styles d’écriture et usages

sociaux Genèses, 39, 121-137.*

Desrosières, A (2001) Managing the Economy: the State, the Market and Statistics In

Porter, T & Ross, D (Eds) The Cambridge History of Science, vol.7 : Modern Social and Behavioral Sciences Cambridge: Cambridge University Press

Desrosières, A (2001) Entre réalisme métrologique et conventions d’équivalence : les

ambigüités de la sociologie quantitative Genèses, 43, 112-127 *

Desrosières, A., & Chiapello, E (2003) Les paradoxes de la métrologie économique : le cas

étrange de la Positive Accounting Theory In Breviglieri, M., Lafaye, C & Trom, D (Eds) Sens de la justice, sens critique Paris : Economica

Desrosières, A (2003) Bourdieu et les statisticiens : une rencontre improbable et ses deux

héritages In Encrevé, P., & Lagrave, R.-M (Eds) Travailler avec Bourdieu, Paris :

Flammarion, 209-218

Desrosières, A (2003) Du travail à la consommation : les usages des enquêtes des budgets de

famille Journal de la SFdS, 1-2, 75-111 **

Desrosières, A (2003) Historiciser l’action publique L’Etat, le marché et les statistiques In

Laborier, P., & Trom, D (Eds) Historicités de l’action publique PUF : Paris, 207-221.* Desrosières, A (2003) Les qualités des quantités Courrier des statistiques, 105-106, 51-

63.**

Desrosières, A (2005) Enquêtes versus registres administratifs : les deux sources de la

statistique publique Genèses, 58, 4-27 **

Desrosières, A (2006) Les recherches de Ian Hacking sur l’histoire des usages des probabilités et des statistiques dans le raisonnement inductif, http://www.jehps.net *

M Armatte30