| La Lettre du GERPISA | no 122 (avril 1998) |
A new look at parts interchangeability is allowing for a reinterpretation of the history of industry, as Yves Cohen's article invited us to do during the February 13th discussion workshop 1. The idea of substitution is not a new one, whether it involves replacing a part by another one, or replacing human labor by mechanical labor. However, during the 19th century, a specific evolution in industry consisted in moving from repair interchangeability to assembly fine interchangeability without adjustment. Another evolution, initiated by engineers in the 1920s, set out to render workers interchangeable.
Within workshops, interchangeability took away from the worker tris capacity to evaluate this work, which in the past he had been able to do based on this experience and intuition. For example, evaluating the part produced by the worker towards the tool was transferred to the tolerance calibre. This transition participates in a long movement which, in the production realm, tends to diminish the importance of the human factor, systematically considered by Method engineers as a source of hazard and imperfection. The expansion of calibre usage in industriel production is thus accompanied by social recomposition: production workers loose part of their qualification, whereas the tolerance calibre user gains in importance. However, the need for human judgment remains: the apparatus can not
decide what is good or bad, and in the final analysis, it becomes necessary to train workers how to use the calibre without pushing the part to "the limit". Likewise, in tool/equipment workshops, those in charge of production, when distributing tasks, take into accourt the capacity of different workers to "get the maximum" out of a machine, and to master its particularities and weak points. In the same manner, for the automobile, the spare part is in fact not absolutely interchangeable, and some sort of adaptation know-how is required in repair networks. As such, the analysis of techniques can not progress if it remains prisoner of a strictly technical conception of evolution and practices.
Consequently, issues linked to interchangeability raise the question of socio-productive systems within the firms, particularly their formation. In the history of mechanical firms, more especially automobile ones, the advent of interchangeability generally brought on assembly line production. Nonetheless, one may wonder if this filiation was inevitable. This subject was already brought up in Gerpisa from other research. Likewise, linking interchangeability and mechanical industry, which was effective, is not an automatic process: immense progress in the armement industry interchangeability was also due to government financial subsidies. Diversity in combinations is an acquired quality from this first program. Finally, in socio-productive or socio-technique formation modalities, the history of intentions in transforming practices or techniques has to be carefully studied. Even though one can say that Ford is not simply an inheritor and that it put an original system into practice, one does not know what its intentions were. In the same manner, it appears that the Uddevalla concept, which partially constitutes a rupture, does not correspond to ideas its initiators had before, and it still remains cifficult to determine whether or not there was an initial principle.
In the plant itself, interchangeability becomes a part of formation - a formalization which Yves Cohen calls the organizational triangle: Analyses, Methods, Controls. Through the organization of these services, firm management allows for the co-existence of collective subjects to which it then attributes distinct functions. This disassociation represents a renewed source of conflict between these functional poles whose force structures the firm's existence. In these conflicts, the positions of Analyses, generally perceived of as being hegemonic with regards to Methods, should be questioned and even reexamined: for this, it is necessary to examine the origin of the conflict. The Methods service was introduced in France in the 1920s, accompanied by the trend towards large-scale series. This service guarantees for the establishment of interchangeability, and a standardized parts system. In essence, it serves as the distributor of the "mechanical spirit" in plants. According to certain accounts, the Methods service had hegemonic designs, and in 1969 even obtained the dissolution of Quality 1 Direction at Renault. Nevertheless, and despite its strength, the conflictuality between Analyses, Methods, and Controls should be nuanced: disassociating functions can also allow for the establishment of distinct viewpoints all aiming to achieve the firm's goals. In addition, this disassociation is also linked to the choices firms make with regards to their suppliers. At Peugeot, its establishment coexisted with the maintenance of a local subcontracting network. However, for Parisian constructors such as Citroen and Renault, it was accompanied by a trend towards forceful integration that, in the future, should be studied more carefully, and which is perhaps related to a need to master the apprenticeship of such an organizational change in a much more efficient manner.
1 - Yves Cohen, Calibres, "Tolérances, hiérarchies et doigtés. Interchangeabilité et action hiérarchique dans l'automobile à l'exemple de Peugeot, 1910-1940", Les cahiers d'histoire et de philosophie des sciences, forthcoming 1998.
2 - Patrick Fridensson, "Fordism and quality: the French Case, 1919-93", in Haruhito Shiomi and Kazuo Wada, Fordism transformed, the development of productionmethods in the automobile industry, Oxford, Oxford University Press, 1995, p. 160-183.