When modern thinkers abandoned the Christian 'answers,' they still felt an obligation to answer the questions that went with them--to show that modern thought was equal to any challenge, as it were. It was this compulsion to "reoccupy" the "position" of the medieval Christian schema of creation and eschatology--rather than leave it empty, as a rationality that was aware of its own limits might have done--that led to the grandiose constructions of the 'philosophy of history.' And naturally these constructions drew more attention to themselves than did the modest idea of possible progress that was overextended (and discredited) in their service.
Although the context in which reoccupation theory emerges is the explanation of a certain overextending attitude of arrogance typical of early, un-self-critical modernity, Blumenberg finds other explanatory uses for it. One might say of it what Blumenberg himself says, incredulously, of the secularization thesis: "If one took the frequency of its application as evidence, there could be no doubt about the historical applicability of the category of secularization. Its productivity seems to be unlimited" (13). In the chapter entitled "The Epochs of the Concept of an Epoch," Blumenberg uses it to dispel the logic by which epochs seem to be necessarily nominal, not real.
The nominal concept of epoch which held sway in the prior "epoch of the concept of an epoch," as exemplified by Bishop Bossuet's Discours sur l'histoire universelle, had as its function the clarification of times. Bossuet, Blumenberg writes,
had still related the concept of an epoch to the privileged standpoint of the contemplator of history for whom the comparison of ages was to be possible. It is not history but this contemplator of history who halts at a resting place so as to survey what happens before and after and thus to avoid anachronisms, the errors that consist in confusing one age with another. (460)
According to Bossuet, these historical vantage points are like the "principal countries" which one marks on a general map of the world as reference points for the others; their function is "to help the memory in the knowledge" of its object. The student of history is supplied with a framework whereby he can frame a material conditional to determine when any given event might have occurred. In principle then, no necessity attaches to the use of any particular collection of epochs.
Blumenberg rather confusingly subverts Bossuet's model of historical error by showing what happens if one takes the material conditional in a more substantial sense of dependence:
The quality of the 'epoch' presents itself, to begin with, as the summation of those features that protect the historian from leveling off the course of history into the monotony of what is always the same, and thus from the error of thinking that anything can happen at any time. Independently of the clerical universal historian's actual intentions, this would perhaps in fact be the most comprehensive definition of the possibilities of error in historical cognition. (460-1; emphasis added)
The sense "that anything can happen at any time" would mean for Bossuet only that a student of history has so blank a sense of the content of history that they are reduced to brute memorization when trying to fix the time of an event's occurrence. Blumenberg brings out the ambiguity of the material formulation arising out of the study of history that "[not] anything can happen at any time": it can be turned to the use of historicism, which would take it to mean that a substantial rather than merely logical dependence prevents any given event from occurring in any but its proper epoch.
The possibility of this subversion of the formulation fits out the mnemonic function of epochs to put pressure on a new historiography which otherwise would have nothing to do with the chronological historiography of a Bossuet. Why should historicism bind itself to strict chronological articulation? In the shift from nominal articulation for the convenience of the student to real articulation of history prior to any studied apprehension, does not the justification for a requirement of chronological exactitude disappear? But insofar as something makes the real principled division of history present itself as an image of the division of history in the mind, or rather as an immanentization of what originally was in the mind, it may be expected to serve the same function as it did in mind--preventing "confusion of times."