Die Idee ist eine alte: Die Sprache, die ein Schüler spricht, sie entscheidet über seinen Schulerfolg. Bernstein hat dies in seinem Konzept des restringierten Codes verarbeitet. Bourdieu und Passeron haben sie in ihrem Konzept des linguistischen Kapitals implementiert und am Beispiel von Studenten geprüft:
“If, as often happens, it is forgotten that the categories defined within a student population by criteria such as social origin, sex , or some characteristic of their school record have been unequally selected in the course of their previous schooling, it is impossible to account fully for all the variations which these criteria bring out. For example, the scores obtained in a language test are not simply the performance of students characterized by their previous training, their social origin and their sex, or even by all these criteria considered together, but are the performance of the category which, precisely because it is endowed with the whole set of these characteristics, has not undergone elimination to the same extent as a category defined by other characteristics. In other words, it is a version of the pars pro toto fallacy to suppose that one can directly and exclusively grasp even the intersecting influence of factors such as social origin or sex within synchronic relations which, in the case of a population defined by a certain past that is itself defined by the continuous action of these factors over time, take on their full significance only when reinstated in the process of the educational career. We have her chosen to adopt a deductive method of exposition because only a theoretical model such as one which interrelates the two systems of relations subsumed under the two concepts of linguistic capital and degree of selection is capable of bringing to light the system of facts which it constructs as such by setting up a systematic relationship between them. In complete contrast to the ‘pointilliste’ verification which applies fragmentary experimentation to a discontinuous series of piecemeal hypotheses, the systematic verification set out below seeks to give experimentation its full power of disproof by confronting the results of theoretical calculation with the findings of empirical measurement.
The same theoretical model enables us to understand why the most constant and also the most powerful relations are, at the level of higher education, those which link the degrees of linguistic competence to the characteristics of the previous academic record. It is principally through the medium of initial streamings (type of secondary school and first year -sixteme – selection) that social origin predetermines educational destiny, i.e., both the chain of subsequent school career choices and the resulting differential chances of success or failure. It follows from this., first, that the structure of the population of selection-survivors is constantly changing, even with respect to the criterion governing elimination, which has the effect of progressively weakening the direct relation between social origin and linguistic competence (or any other index of scholastic success). Secondly, at every stage in their school career, individuals of the same social class who survive in the system exhibit less and less the career characteristics which have eliminated the other members of their category, depending on the severity of the selection to which their class is subject and the level of education at which the synchronic cross-section is taken”
(Bourdieu & Passeron. Reproduction in Education, Society and Culture, pp.72-82).