The Statistical Society of Paris


Jean-Jacques Droesbeke wrote the article Les Racines de la Societe Française de Statistique (The roots of the French Statistical Society).

The first part of the article is the French Statistical Society - Introduction. We give an English version at THIS LINK.

The third part of the article is the From the Association of University Statisticians to the Association for Statistics and its Uses. We give an English version at THIS LINK.

The fourth part of the article is The French Statistical Society. We give an English version at THIS LINK.

The References to the article are given at THIS LINK.

2. The Statistical Society of Paris

Created in 1860, the main goal of the SSP was to "popularise statistical research through its work and publications". From the beginning, this society has developed in a context dear to economists, but it will also turn to demographers, actuaries, doctors, not to mention politicians. In order to give a foundation to its projects, the SSP publishes, from the year of its creation, the Journal de la Société de Statistique de Paris (Journal of the Statistical Society of Paris) which will be published regularly for nearly 140 years and which will become the Journal de la Société Française de Statistique (Journal of the French Statistical Society) in the context of the merger mentioned above.

Although the SSP is of particular interest to us here as the "root" of the SFdS, it was not the first association of statisticians incorporated in France.

2.1 Precursors

During the first half of the 19th century, several statistics associations were created in France, and more particularly in Paris. This situation is not surprising in that the governments of the day saw the collection of statistical information as an instrument for the formulation of various policies. This utilitarian design will lead to a real craze, resulting in quantification applied to many areas of economic and social life (see Desrosières, 1993).

Thus the beginning of the 19th century will see the birth of a Bureau of Statistics, so great is the ambition to create "large statistical memoirs". Launched by the chemist Jean Antoine Chaptal - while he was Minister of the Interior, called to this position by Bonaparte from 1801 to 1804 - this office will alternately experience periods of success and depression to end its existence in 1812. Chaptal will grant much attention to the construction of various statistics, noted especially through the prefects who will receive from him a small book of 70 pages entitled Essay of Statistics and written in 1801 by Jacques Augustin Mourgues.

Jean Baptiste Nompère de Champagny, who took over the Ministry of the Interior in 1801, then Charles Etienne Coquebert de Montbret, appointed head of the Bureau in 1806, will try to be good successors of Chaptal, also using the assistance of prefects to obtain various information. But the achievements of the surveys they will conduct are very uneven. Despite the marked fervour for this type of approach, specific to the last years of the Empire, they lack an important quality: reliability.

In this context, we note the foundation, in the first months of 1803, of a 'Statistical Society' in Paris, under the auspices of the Institut de France and the government of the time, an association of which we have, in fact, few traces. It is only known that it meets for the first time the 16th pluviose of the year XI - that is to say, 5 February 1803, in the Gregorian calendar - under the presidency of Edmé Mentelle, a member of the Institute, assisted by the Vice President René Desgenettes, a doctor by profession, and Secretary Louis Ballois, initiator of the project, all under the protection of Jean-Jacques Cambacérès. Unfortunately for him, the life of this society will be short, especially because of the death of Ballois in late 1803. It seems that this society was extinct by 1804.

Another statistical society was created in Marseille on 7 February 1827. Recognised as a public utility in 1832, it grew rapidly, thanks in particular to local financial support and the efforts of its perpetual Secretary, Pierre-Martin Roux, who will assume his responsibilities, from 1829, for more than 30 years.

Composed of 30 to 40 active members during this period - but supported also by more than 200 corresponding members - the 'Société Royale de Statistique de Marseille' (Royal Society of Statistics of Marseilles) will know a remarkable growth (See Depoid, 1961). The country-wide fame it enjoys even allows it to organise a banquet, in 1862, to celebrate the birth of the SSP.

From 1870, however, its activity will be reduced and this association will no longer have a large membership, at the beginning of the 20th century, only a few historians and archaeologists. After the 1914-1918 war, it will become the 'Société de Statistique, d'Histoire et d'Archéologie de Marseille et de Provence' (Society of Statistics, History and Archeology of Marseille and Provence). In 1950, it will merge with the 'Institut Historique de Provence' to become the Historical Federation of Provence, for which statistics is no longer the centre of interest.

Two years after the creation of the Statistical Society of Marseille - in 1829 - César Moreau founded the 'Société Française de Statistique' Universelle. This one is inspired by English learned societies - that Moreau discovers while he is vice-consul of France in London - and more particularly by the 'English Statistical Society', founded in 1826 and which precedes the 'Manchester Statistical Society' and the 'London Statistical Society'. This new association has an ambitious program, as evidenced by its first objective: "To form in Paris, in France, in Europe, in the world, a centre of unity for the studies of Statistics". Its growth is rapid since it will count on more than a thousand members in 1836, of which a hundred reside abroad. Placed under the protection of the King, it is located on the street, its offices occupying Number 2 of the Place Vendome. Its activities and interests are disseminated in the Journal of the Activities of the Society of Universal Statistics. Unfortunately for it, this society declines after 1843, following the resignation of Moreau and internal dissensions. The Revolution of 1848 will be its death knell. It will be absorbed by the 'National Agricultural, Manufacture and Commercial Academy' which, in turn, will disappear around 1900.

A last association must be mentioned here. Set up in 1830 by Charles Etienne Coquebert de Montbret and Sébastien Bottin - whose collecting, gathering and classifying habits will allow him to leave his name to posterity - the 'Société libre de Statistique' (Free Statistical Society) run from the Office of Jean-Antoine Chaptal, then at the end of his life, and Charles Dupin, particularly appreciated for his work in differential geometry. Its existence is of very short duration. It seems that the only action we have traced is its attempt to merge with the Moreau Society, which will end in failure.

2.2 Some highlights of the SSP journey

Our intention is not - as we have already said - to detail the different stages of the life of the SSP, but rather to highlight some important facts.

The creation of the SSP is strongly linked to a political event that takes place in 1860: the signing of a commercial treaty between France and England that promotes free trade advocated by liberal economists. The main architect of this treaty is Michel Chevalier, counsellor to Napoleon III. Faced with the various disputes that are expressed against this treaty, a small group of men plans to create an association of Statistics that could defend it "scientifically". Besides Michel Chevalier, it is composed of Maurice Block, Joseph Garnier, Clément Juglar, Leonce de Lavergne, Auguste de Malarcé, Hippolyte Passy, Louis Villermé and Louis Wolowski. They will soon be joined by the Director of the General Statistics of France, Alfred Legoyt - successor to Alexandre Moreau de Jonnes in this post - who will be in charge of drafting a statutes and encouraging membership. The first members gathered at the Hôtel de Ville in Paris, where the first session of the association was held on 5 June 1860, under the chairmanship of Michel Chevalier.

Many founders of the SSP come from the 'Société d'Economie Politique' (Society of Political Economy), founded in 1842, with which the SSP will maintain good contacts in the decades that follow. It is also the Journal of this association, the Directory of Political Economy, created in 1844 by Joseph Garnier and Maurice Block, which will serve as a model for the future Journal of the Statistical Society of Paris.

Recognised as a public utility in 1869, the SSP experienced a difficult period during the War of 1870. New statutes were proposed in 1872, at the same time as a transformation of the name of the Association into 'Société de Statistique de France'. Only this last proposal will be rejected by the Council of State.

In 1882, the SSP considered the constitution of a 'Central Commission of Statistics' in France, taking up a suggestion made by Quetelet during the international congresses organised between 1853 and 1876 (see Droesbeke, 2003). Thus, a 'Higher Council of Statistics' will be created in 1885, the 25th anniversary of the SSP. At that time, the latter has nearly 400 members, some of whom are famous - we think in particular of Sadi Carnot and Felix Faure, two future Presidents of the Republic. Some associate members complete the panorama, among who we can note the presence of Baron Haussmann and Ferdinand de Lesseps.

The statutes of 1872 having foreseen an annual rotation of the presidents, it would be too long to state here all their names. If, however, we had to cite a few of them, limiting ourselves for the moment to the first 25 years of existence of this association, we would choose, apart from the founding members, some characters whose work or actions would deserve to be further explored. First there is the physician and demographer Louis-Adolphe Bertillon, whose article on the theory of averages, published in 1876 in the Journal de la SSP, is an important reference for historians of statistics. We can also note the presence of engineer Emile Cheysson or the economist Léon Say, grandson of Jean-Baptiste Say, leader of the liberal economy.

During this period, which goes from 1860 to 1885, the Journal de la SSP deals mainly with economics, demography, health, and social problems (illiteracy, morality of the population, etc.). Some articles are devoted to the role of statistics. The methodological aspects mentioned in this review concern mainly the study of means and methods of graphical representation. While the contributions to the Journal are essentially descriptive, as one might expect, they are often intended to illustrate the authors' interests or to support their theses. The titles of these articles are often explicit and their content can sometimes be described as "surprising" in our time. To convince oneself of this, one has only to go through the table of contents of the various notebooks of the journal. Let us mention, as examples, some titles among others: Statistical studies on ways to reduce the mortality of Europeans in hot countries (1860), On the alleged physical degeneration of the French population compared to other European populations (1863 ), Natural Children before Statistics (1868), The Role of Alcoholic Beverages in the Increase in Cases of Madness (1873), Stuttering Statistics in France (1874), Abandoned Children (1881) or an Essay on the comparative morality of the various classes of the population and mainly of the working classes (1872). There is also a Note on the unpopularity of statistics and its causes (1864). An article stands out somewhat from others. Published in 1876, it is due to Irénée Bienaymé and is entitled Large numbers in statistics. On a principle which M Poisson had thought to discover, and which he had called the "law of large numbers".

The year 1885 is also an important date for the rise of statistics: it is during this year, in fact, that on the occasion of the 50th anniversary of the 'Statistical Society of London', the 'International Statistical Institute' is created, intended to relaunch the international meetings initiated by Quetelet in 1853. The International Statistical Institute will organise its second session in Paris in 1889, thus contributing to the events set up for the centenary of the Revolution French. This institution will return to this city in 1909 to hold its 12th session during which the SSP will celebrate its (almost) fiftieth anniversary.

During this second period, which goes from 1885 to 1910, the SSP has a fairly stable membership, of the order of 400 people. Among its most famous members, we must certainly mention the names of Raymond Poincaré and Paul Doumer, both future Presidents of the Republic. Note also the presence of Emile Borel who will become President of the Association in 1922 and to whose actions we will return later. The Bertillon family - whose role in nineteenth-century French statistics is generally emphasised - is represented at the time by Jacques, a doctor by occupation, who will preside over the society in 1897. We think that we must also mention some dignified presidencies of interest: they concern the economist Paul Leroy-Beaulieu in 1889, Pierre-Emile Levasseur in 1900 - who was one of the first in France to write a doctoral thesis in economic history - and especially, Lucien March in 1907, whose work on statistical methodology and its application in economics is internationally known.

The themes addressed by the Society's Journal during the first 25 years are now supplemented by articles on statistical education and its methodological developments as well as by some new contributions among which we can mention a report on the electric machine used for the census Hollerith used by the Austrian statistical services at the time and studied by the French services for possible use.

Several facts relating to the SSP deserve to be noted for the period from 1911 to 1935. First, there is an increase in the number of its members which, in 25 years, goes from just under 400 to 750. In addition to the Presidents of the Republic already mentioned, the society will include a winner of the Nobel Prize in Medicine (Charles Richet, winner in 1913) and the Nobel Peace Prize (Léon Bourgeois, awarded in 1920).

It will also contribute to the creation in 1922 of an 'Institute of Statistics attached to the University of Paris' (ISUP), an essential step in the process of dissemination of this discipline. Unfortunately, the steering committee of ISUP will soon encounter significant financial problems. To confront them, five of its members - Alfred Barriol, Clement Colson, Michel Huber, Lucien March and Henri Truchy - propose, in 1927, the creation of a foundation, "The Statistical Science", in order to "contribute to the development and the prosperity of the Institute of Statistics of the University of Paris". One month will suffice to endow this foundation with the character of a public utility.

For two years Emile Borel will teach at the ISUP the calculation of probabilities and its applications to statistics. He will be replaced at the end of 1924 in this position by Georges Darmois. The latter, like the economist François Divisia - whose work on indices deserves to be known by statisticians - and André Liesse - who will publish in 1902 a book on Statistics, its difficulties, processes and results - will also participate in the activities of the SSP.

A characteristic of this association, during this period, is the growing membership of members working in the insurance sector and in that of banks and credit institutions. It is also necessary to underline the important relations of the SSP with public statistics.

The centres of interest of the Journal of the Society were at the time mainly focused on demography and economics, but there are also some articles closer to the statistical approach, mainly due to Lucien March.

The period from 1935 to 1960 does not see any major changes from the previous situation. By consulting, in particular, the list of its presidents, one realises that the fields which concerned this association remain the economy, demography, finance, and insurance. There are also some mathematicians among whom we can note the presence of Maurice Fréchet.

The Journal of the Society is also very diverse in its content. Among the authors dealing with statistical subjects, we can mention the presence of Fernand Chartier, Jacques Desabie, Daniel Dugué, Maurice Fréchet, Edmond Malinvaud, Eugène Morice, George Morlat, Daniel Schwartz or even Pierre Thionet, the latter having twelve publications between 1945 and 1960. Other authors complete the list in more applied fields: Maurice Allais, Jean Fourastié and Alfred Sauvy are some examples.

The last period of the SSP goes from 1960 to 1997. It sees the birth in 1974 of the 'French Statistical Society' whose statutes will be deposited in 1976. Created with the intention of extending to the national level what was, on paper, hitherto reserved for Paris and in order to protect this denomination, it will not really have its own activities. But its existence will facilitate the birth of the SFdS, as will be seen later.

The SSP will also create, in 1975, the French Statistician Award awarded "one year to a foreigner of great renown, a second year to a young French or foreign statistician, a third year to a confirmed French statistician". The first prize will be awarded to Robert Horvath, professor at the University of Szeged (Hungary).

Despite these initiatives, the SSP will see its recruitment stagnate and even decrease gradually. At the same time, another scholarly association will see the light of day at the end of the 60s and experience a growing development. This is the ASU. It is therefore not surprising that from the 1980s, many presidents of the SSP will work towards a rapprochement with this other "root" of the SFdS.

The third part of this article is the From the Association of University Statisticians to the Association for Statistics and its Uses. We give an English version at THIS LINK.

The fourth part of the article is The French Statistical Society. We give an English version at THIS LINK.

The References to this article are given at THIS LINK.

Last Updated February 2018