The International Labour Organization as a producer of statistical knowledge Workshop 25-26. 2. 2016 International Research Centre "Work and Human Lifecycle in Global History" Humboldt University Berlin (re:work) Theresa Wobbe (Potsdam) und Sandrine Kott (Geneva) Abstracts Roser Cuso The ILO and the international statistical program The ILO s labour statistics program of 1919-1945 was less developed than that of the LoN- EFO (Economic and Financial Organization) (Cussó, 2012a; 2012b). The former produced some conventions and recommendations on labour statistics over this period but their harmonization decisions do not seem to have been translated into internationally comparable figures, especially in the case of such central issues as unemployment. The International Conference of Labour statisticians met four times during this period. A report on Methods of statistics of unemployment was issued in 1925. The possibility of having at least a common source for unemployment statistics was evoked but could not be counted on. Notable efforts were made to produce information on wages (1931). Bertillon s classification of professions was also largely reproduced in the reports, while admitting its limited application (1923). We suggest two hypotheses to explain this gap. First, labour statistics were more difficult to collect and harmonize as compared to economic and financial data. Labour information was, and still is, a delicate domain. It directly reflects the government policies related to social rights. Measures of unemployment, for instance, are in general attached to worker s access to unemployment insurance or to some form of social service or unionism. Definitions of strike concern, in one way or another, the right to organize them. This difficulty is documented by Liebeskind Sauthier and Lespinet-Moret. They note that there was no stable and comparable international definition of unemployment until 1947 : «il faudra attendre 1947 [ ] pour qu une législation internationale en matière de méthodes et nomenclatures soit établie» (Liebeskind Sauthier & Lespinet, 2008 : p9). This is confirmed by Liebeskind Sauthier (2008). Martin Bemmann, Freiburg The Establishment of 'World Economic Statistics' and the ILO during the 1920s Infrastructures, Networks and Classifications In my project on the establishment of international economic statistics after the Great War I focus on the statistical work of the League of Nations, on its attempts to make the world countable and on effects of this work for economic policy debates and decisions. By drawing on the examples of China, Japan and the Soviet Union I am looking at how the League s experts endeavoured to establish a well working statistical infrastructure and how they tried to incorporate data into their emerging concept of world economic statistics which was quite different in content, scope and reliability as well as it was the result of different surveying methods. Furthermore, I am looking at debates in Germany and Great Britain which took place in politics, economics and the broader public around the international economic
conferences of Genoa (1922), Geneva (1927) and London (1933) as well as on discussions about the post-war economic order which occurred during World War II. Doing this, I am interested to what extent international economic statistics were actually used in these debates and whether they influenced arguments and decisions of national actors. Since the statistical work of the League in the economic sphere was quite extensive I am concentrating on attempts which were made to establish global statistics on (industrial) production and trade. Moreover, I am still in a phase of empirical research. As a consequence, I cannot say that much on labour statistics. However, ILO statisticians were involved in almost all statistical endeavours of the League and I will highlight three issues which raise questions which in turn seem to be of interest for further investigations on the emergence of international (labour) statistics. First, this concerns the involvement of ILO members (Harold Butler, Royal Meeker, Karl Pribram) in committees on the establishment of statistical institutions and infrastructure in 1919 and 1920 as well as in gatherings concerned with shaping foundations for international statistics on production, trade, prices, consumption etc. Due to these activities, ILO, next to the League, the International Statistical Institute and the International Chamber of Commerce, became one of the leading organisations which shaped the international statistical landscape and its infrastructures during the post-war years. This raises questions about institutional rivalries and their impact on actual statistical work, classifications, concepts and methods, about the importance of ILO for securing data from employer s and worker s organisations (due to their tripartite structure), and about the influence ILO could exert on national institutions in order to standardise surveying and displaying methods. Secondly, and closely connected to institutional aspects, there is the issue of (informal) networks of involved actors which seem to have been of significant importance for the establishment of international statistics. Besides statisticians working in the mentioned organisations, economists and statisticians working in universities and national institutions (state and private) were also involved in shaping classifications, concepts and methods. Therefore, it is of great interest to look at the emergence and the development of such networks which could influence even strategic decisions of what should be made an issue of gatherings and how it was approached by officials from the League (and certainly also from ILO). Furthermore, when the focus is set on global statistics, it is most important to learn who was involved in shaping statistical infrastructure, concepts and methods for surveying developments in regions outside of Europe and North America. Thirdly, the central issue of classification is highlighted. I will present two aspects from my work for which ILO was important. On the one hand this is ILO s participation in drawing up a system of classification of industries and occupations, on the other its constitutional need for a definition of and reliable statistics on industrial importance of single countries. Looking at these aspects the central question comes up and should be answered, to what end (international) organisations established international statistics and who did actually used them for what purposes. Furthermore, it has to be asked to what extent statistics or statistical concepts, classifications and methods which were supposed to understand certain problems still influence our way of looking at certain aspects, even if the original problem has vanished. In other words: The importance of path dependencies should be investigated. Dhermy-Marail "The scientific agenda of ILO : methods and objects in circulation".
I can describe different statistical methods (tables, co-variations, index-numbers, calculus of probability), regarding to their origin, and regarding to their respective objects. Léa Renard The statistical construction of family workers, 1920s-1980s The contribution aims at discussing the role of statistics in establishing and circulating legitimate representations of labour, by shortly picturing the shifting classification of family workers (Mithelfende Familienangehörige) in the German census during the interwar period. We investigate both the inclusion/exclusion of the category into the taxonomy of status in employment (Stellung im Beruf) and the criteria of definition. We argue that the search for unified, homogenized statistical representation of the working world led to the institutionalization of gainful employment as an emerging standard. If we focus on this national moment in the first part, we will then shortly draft the benefits, in our view, of intertwined analyses of both national and international production of statistical classifications for the understanding of how categories emerged and circulated in transnational fields before and after 1945, based on the example of the redefinition of the category by the ILO. Yann Stricker Classification and the Regulation of Labour Migration In my PhD project I analyse how, from the interwar period onwards, international migration became the object of vast statistical knowledge production. In the presentation I will focus on the International Labour Organization (ILO) where efforts to quantify human movements led to the making of first worldwide statistics on the topic during the 1920ies. In the 1920s and 1930s the organization not only published Monthly Records on Migration, but also extensive historical reports on International Migrations. Grounded on the theory of the history of knowledge, I ask how and in what way it became historically meaningful to create quantitative knowledge about people on the move. The ILO s venture to categorise and quantify human movement throughout the world was linked to ideas of internationally balancing the supply and demand of labour by organising the flow of workers between countries. Although these ideas and plans were inspired by practices in colonial empires, the perspective of international statistical knowledge on the movements of people was a new one. By looking at the making of migration statistics in the ILO in the interwar period, I want to ask in my presentation, what it meant to create an international gaze on a colonial world. Daniel Speich Political communication changed when world statisticians met in Washington in 1947 My contribution focuses on a world convention of statisticians that was held in Washington in 1947. This was a multiple event under the auspices of the International Statistical Institute (ISI) and the United Nations (UN) that included conferences by many specialized bodies such as the Inter-American Statistical Institute, the Econometric Society and the International Union for the Scientific Investigation of Population Problems. I will first sketch the organizational details of the event and highlight the (modest) role of the ILO in it.
The second step is to explore the epistemic tensions between the universal claims of statistics and local knowledge production. From Washington, in 1947, a very local view gained global ground. One topic at the conference was statistical sampling. A second topic (on which I focus) was universalizing categories and classifications in order to produce a solid comparative world-view. Why was this such a pressing issue specifically in 1947? One reason might be that under the emerging new world order several institutional bodies (including the ILO) competed for the epistemic function of a centre of calculation (Bruno Latour), which localizes the power of defining statistical realities. Third, my paper argues for three phases in global political communication, i.e. an imperial style that quickly lost ground post-1945; an emerging global epoch that imagined the world as a family of nations ; and a subsequent era of post-national global governance. With respect to global statistics, I assume that the Washington meeting marked the end of the age of empires during which the world had been imagined as being fundamentally bifurcated between civilizing overlords and passive recipients. Now, a new vision gained ground that called for a universal framework in which all nations could be compared. Subsequently, global political communication changed because it increasingly could refer to standards that were issued by technical bodies (like the ILO, the UN or OEEC) and not by imperial metropoles. Statistics emerged as a new universal language that could not be contested on political grounds, but worked as a medium in order to articulate political claims. Further research will show in how far this move paved the way for the conception of a global governance that needs no governments. Isabelle Lespinet-Moret ILO s Industrial Hygiene and Safety at Work Service and statistics Industrial Hygiene and Safety at Work Service, in ILO, deals with occupationnal diseases and risqs, within the ILO. One of the missions of this technical Service is to collect the information and to build medical and technical knowledges, by means of the experts worldwide and the civil services. This collection and this construction is made in a transnational way. Several ways of research for the information open to them: the survey, the bibliography and the statistics. Some of national statistics are reused, others are established by the researchers of the Department. Questions that arise on this matter are the origin, the constitution of these statistics and their use. The homogenization between national ways of the statistics is a recurring question. If first and foremost the used or established statistics are the ones which concern occupational accidents, sector in which exists a knowledge of the actuaries, their use for the professional diseases and the poisonings is more hesitating Emmanuel Reynaud Wages and Statistics: the ILO s First Decade Wages are a major area of work for the ILO in its first decade. The Peace Treaty which establishes the Organization in 1919 sets, in its preamble, the priority objective of providing an adequate living wage and, in its article 427, the principle of the payment to the employed of a wage adequate to maintain a reasonable standard of life as this is understood in their time and countries. The presentation will examine the implications of these statements on ILO s statistical work in its first decade from two angles. First, it will deal with the Office work on statistics, in conjunction with the International Conferences of Labour Statisticians and other international conferences. In addition, it will look at the relationship between the statistical work of the Office on wages and the standard setting activities of the Organization.
Susan Zimmermann 'Native Labour' International: Terminology, Conceptuality and the Geography of the ILO Intervention This paper discusses two major dimensions of how the International Labour Organization approached, in terms of terminology, conceptualization and geographical reach of its related major instruments the Forced Labor Convention of 1930 (C29), the Recruiting of Indigenous Workers Convention of 1936 (C50), and the Contracts of Employment (Indigenous Workers) and the Penal Sanctions (Indigenous Workers) Conventions of 1939 (C64 and C65) the question of native labor in the interwar period. First, I look into the relationship, and sometimes tension, between the language the ILO developed when addressing native labor in international law, i.e. the legal language of the ILO s International Labor Standards on the one hand, and the broader discourse within the ILO on native labor on the other. This relationship and the ILO s inbuilt efforts to develop a legal language perceived as appropriate when addressing native labor in international law, shed light on important dimensions of how the ILO aimed to respond to both the racialization of the global labor force and labor relations worldwide, and the question of colonialism writ large, i.e. layered sovereignty as characterizing the international system at the time. Second, I look into the relationship between pre-existing territorial legislation on key elements of native labor, namely forced labor and certain aspects of contract or indentured labor on the one hand, and the ILO s international instruments aimed at regulating these same labor relations on the other. A critical interrogation of this relationship sheds light on both the impact of preexisting imperial/colonial labor law on ILO labor legislation, and the ILO s efforts to remold this heritage as it worked towards the internationalization of the regulation of native labor. Pauli Kettunen The ILO debates on a Universal Validity of "Government"- "Workers"-"Employers" Representation and its Transformative Meanings.