the students to talk about a single set of events and a single set of agents and allowing other events and characters into the picture in a subordinate role at best.
If this is the form of the mediational means that seemed to be used by the students, what can be said about the processes and reasons for using it? The first and most obvious point is that the main reason the students used it was that they had no other mediational means available. Some students who recognized this made explicit statements about it. For example, one student stated at the outset that she considered the history of the United States to be the history of the land and its early inhabitants. After making this point, however, she went on to say, "Since . . . I know relatively little about the geological and anthropological aspects of the ancient United States, I will deal with aspects of history with which I am more familiar. Suffice it to say, however, that I do not consider the history of the 'US' to begin at this point." Following this disclaimer, she went on to write about European settlers, their relations with Native Americans, and so forth.
In general, then, even though roughly half the students expressed some form of discomfort with the story line they employed, they proceeded to use it. I argue that this tendency reflects the fact that the students invoked a single cultural tool. Many of them tried in one way or another to engage in "tactics of resistance" ( de Certeau, 1984) against the power of this mediational means to shape their text, but these tactics were generally quite ineffective, both because the students had access to no well-developed alternative cultural tool and because the tool they were using had an integrated and exclusionary structure that made it very difficult for other voices to participate in the production of the texts in any elaborate way.
What this suggests is that in addition to examining ways in which individuals master accounts of the past, we need to learn much more about how they enter into specific kinds of "use relations" with these accounts. Do they accept them wholeheartedly, or do they question or even reject them? By viewing the production of historical narratives as a form of mediated action, it is possible to keep these questions in focus. In addition to inserting an essential dimension into our understanding of how the teaching and mastery of historical knowledge takes place, this focus will perhaps allow us to address more fully issues of what citizens' historical beliefs are and how these beliefs are used for beneficial and detrimental purposes.
An earlier version of this chapter was presented at the Conference on "Cognitive and Instructional Processes in Social Sciences and History," Autonoma University, Madrid, October 23-25, 1992. The research for this