The violent element in the narrative structure of the basic Western—shoot-out, brawl, ambush, bank raid, fistfight, wounding, duel, or massacre—like any other semantic unit in a structured discourse cannot signify anything on its own. It can only signify in terms of the structured meanings of the message as a whole. Further, its signification depends on its relation—or the sum of the relations of similarity and difference—with other elements or units. Olivier Burgelin has long ago, and definitively, reminded us that the violent or wicked acts of a villain only mean something in relation to the presence/absence of good acts:

We clearly cannot draw any valid inferences from a simple enumeration of his vicious acts (it makes no difference whether there are ten or twenty of them) for the crux of the matter obviously is: what meaning is conferred on the vicious acts by the fact of their juxtaposition with the single good action? … One could say that the meaning of what is frequent is only revealed by opposition to what is rare.… The whole problem is therefore to identify this rare or missing item. Structural analysis provides a way of approaching this problem which traditional content analysis does not.

Indeed, so tightly constructed was the rule-governed moral economy of the simple-structure Western that one good act by a “villain” not only could, but apparently had to , lead to some modification or transformation of his end. Thus, the presence of numerous bad-violent acts (marked)/absence of any good-redeeming act (unmarked) = unrepentant villain; this in turns means that he can be shot down, without excuse, in the final episode and makes a brief and “bad” or undistinguished death, provided that the hero does not shoot the villain in the back, or unawares, and does not draw first. But , the presence of bad-violent acts (marked)/presence of single good-redeeming act (marked) also can = possible salvation or regeneration of the villain, deathbed reconciliation with hero or former cronies, restitution to wronged community, at the very least, lingering and “good” death. What, we may now ask, is the meaning of “violence” when it only appears and signifies anything within the tightly organized moral economy of the Western?

We have been arguing (a) the violent act or episode in a Western cannot signify in isolation, outside the structured field of meanings which is the film or program; (b) it signifies only in relation to the other elements, and in terms of the rules and conventions which govern their combination. We must now add (c) that the meaning of such a violent act or episode cannot be fixed, single, and unalterable but must be capable of signifying different values depending on how and with what it is articulated. As the signifying element, among other elements, in a discourse, it remains polysemic. Indeed, the way it is structured in its combination with other elements serves to delimit its meanings within that specified field, and effects a “closure,” so that a preferred meaning is suggested. There can never be only one, single, univocal, and determined meaning for such a lexical item, but, depending on how its integration within the code has been accomplished, its possible meanings will be organized within a scale which runs from dominant to subordinate. And this of course has consequences for the other, the reception end of the communicative chain: there can be no law to ensure that the receiver will take the preferred or dominant meaning of an episode of violence in precisely the way in which it has been encoded by the producer.