When we look at human memory and the mental objects stocked within it, we observe that beyond all possible classifications (for a general survey, see Robin 1990), a fundamental tension between two extremes systematically occurs. Memory works indeed with two different types or tools of representation. The first one could be called that of the encyclopedic model. With this term, I refer to all discourses, whether they be texts or images, which principally tend to give not only a complete view, but also an explanation of the past, even for those who do not yet have a knowledge of it. The second type, then, could be called that of the detail. With this term, I refer to all isolated units or segments which can evoke the past, but only in a subjective and metonymical way: they can do so for those who already know or still remember that particular past the details belong to, whereas the others will find those details mute and incomprehensible. Encyclopedia on the one hand and detail on the other hand, of course, cooperate in many ways. Their interaction is the rule, not the exception. Encyclopedic representations are, for instance, better understood, or are modified, thanks to the integration of details, whereas details cease to be incomprehensible when placed in the larger encyclopedic context they help to build. But what do we have to do with a detail or a set of details when for some reason the encyclopedia is missing? The classic answer is that one has to wait for further information and that until then anything goes, probably because nothing seriously works (for the role of encyclopedic knowledge in interpretation, see van de Velde 1993). This is, of course, not false, and everybody who has studied or practiced intercultural communication will know how difficult it is to manage such details. But I also believe that encyclopedia is not the only possible way to detail comprehension, and that an internal reading of details is not always a methodological error. On the one hand, I would like to argue that in some cases something of the initial shock caused by the detail resists to its encyclopedic integration, and that too great an insistence on the integrative capacities of the encyclopedia could possibly mask or erase what is really important in the details we are interested in (for further details, see the concept of effacement complementaire in Ricardou 1982). On the other hand, I would like to argue also that encyclopedia is itself an historically shifting object, and that any detail interpretation is therefore a neverending process which possibly only after a while establishes the initial strangeness of a detail, or charges it afterwards with an unforeseen opacity. For both reasons, I think it is also useful to analyze a detail in itself, even when in practice such a mental and methodological restriction is hard to realize. Around the mid-sixties, as so many other boys of my age, I passionately collected all types of bubble gum cards, i.e., cards used to merchandize small and pink chewing gum leaves (cards of movie stars, of great scenes of national history, of planes, of football players and so on). But also cards of the American Civil War: a set of 88 cards retracing mostly the battles of a conflict historians call the first modern, i.e., industrial war (Trachtenberg 1989). In those years I didn't understand a single word of English and was probably completely ignorant of what had happened in the U.S. between 1861 and 1865. Moreover, not having a television set at home, I was not daily exposed to certain mass-culture influences which could have helped me to decode my bubble gum cards. So the Civil War News, as the name of my collection was, appeared in my life a kind of cultural UFO, and I am really tempted to believe that this is still the case. In spite of all I have now learned on the subject, I go on thinking that the Civil War News remains an enigma. Even the recent discovery of the author of the series, Len Brown, has not really changed this point of view (Brown 1987). …
Read full abstract