Universite Pierre et Marie Curie - Paris Univertsitas


Sur ce site

Accueil >

Social physics

In the mid-1990s it became possible to publish papers on social phenomena in physical journals such as “Physica A” or the “European Physical Journal B”. This marked the beginning of a field of physics which is commonly called econophysics. However,the participation of physicists in social science research has begun much earlier.As a matter of fact, from Quételet to Pareto to Montroll many physicists contributed to the advance of social sciences. The title of Quételet’s main work, namely “Social physics (*)”, still appears as a good definition of how the approach of physics can be used in the investigation of social phenomena.

The main obstacle to the understanding of social phenomena is the fact that they are multi-faceted which makes it difficult to observe one effect at a time. In its early phase of development physics was confronted to the same difficulty. The basic methodology of experimental physics has been set up precisely as a way to disentangle multi-faceted phenomena. From steal balls to soap bubbles to feathers to oak leaves, falling bodies exhibit a wide range of behavior. We now know that there are only three mechanisms involved : gravity, buoyancy and drag due to the resistance of the fluid. It is this understanding which enables us to see various forms of “free-fall” as just different manifestations of the same phenomenon. However, to gain this understanding physicists had to perform and compare a great number of observations. As is well known Galileo’s experiments on falling bodies laid the foundations of the field of mechanics.

Although in the social sciences one cannot perform experiments in the same way as in physics, nevertheless one can carry out well focused observations (as in astronomy or astrophysics) which eventually will reveal the main factors and mechanisms which are at work. In order to observe a given factor (almost) independently from other factors onemust find situations in which the role of this factor becomes preponderant.

Thus, the effect of a lack of marital ties on suicide becomes particularly clear in situations in which (for some reason) most people cannot get married. In Roehner (2008 **) it was shown how this methodology can actually be implemented. In this observation-based approach the crux of the matter is to find statistical data for situations which enable us to observe each factor separately. In a sense this methodology can be seen as a sharply-focused version of the comparative approach which has already been used with much success by well-known social scientists from Ferdinand von Humboldt (linguistics) to Emile Durkheim (sociology) to Marc Bloch (history) to David Laitin (political science).

A last word is in order to define the role of models. Once the main factors have been clearly identified empirically it is usually a simple matter to build a mathematical model. The main virtue of the model is not, of course, that it is able to account for the very observations for which it has been designed, rather its main virtue is to provide predictions, for instance by letting one of the parameters, say a go to a “critical” value ˆa, e.g. ˆa can be zero or infinity). At this point it is essential to test these predictions. This means that one must find real situations in which a indeed takes the value ˆa. It is clear that this is a demanding challenge but it is equally clear that if this step cannot be carried out the model will be nothing but a phenomenological model without any real predictive power.

(*) Quételet (A.) 1835, 1869 : Physique sociale, ou Essai sur le d´eveloppement des
facult´es de l’homme. Muquardt, Brussels.
[This title corresponds to the edition of 1869 ; in the edition of 1835 which was
published in Paris the title is “Essai . . . ” and the subtitle is “Physique sociale”.
(**) Roehner (B.M.) 2008 : Econophysics : challenges and promises. An observationbased
approach. Evolutionary and Institutional Economics Review,March 2008
(to appear).

For more detail on this approach see :
Roehner (B.M.) 1995 : Theory of markets. Trade and space-time patterns of price
fluctuations. Springer, Berlin.
Roehner (B.M.) 2002 : Pattern and repertoire in history. Harvard University Press,
Cambridge (Massachusetts).
Roehner (B.M.) 2007 : Driving forces in physical, biological and socio-economic
phenomena. Cambridge University Press, Cambridge.