In the introduction to his doctoral dissertation on ‘the conquest of the snapshot’, the photography historian André Gunthert writes:

This ‘also’ is tied up with a shift: the attention to the technological, to its traces in the image, transforms our gaze and has us move to the ar-chaeological level. But this passage involves taking non-verbal elements into account: images, devices, diagrams, graphics, etc. Pierre Francastel, in his article ‘Valeurs sociologiques de l’espace-temps figuratif’, clearly demonstrated the importance of these non-verbal sources, and yet they create methodological problems as part of an archaeology: how (and of what) can they make an archive? It was questions of this sort that led Michel Foucault to pass from an archaeology based on discourses to an epistemology that takes dispositifs into account. I would like to give an example, using a concrete object, to illustrate these questions, one sometimes mentioned by archivists because of the singular problems it poses: the first cinematic natural colour process marketed commercially, Kinemacolor.

This process was invented in 1906 in England by George Albert Smith and was financed and marketed by Charles Urban until 1915.