This unity can be severed, practically, in the act which is called error, where propositions expressing the truth are combined, not according to their theoretical connection, but according to what is deemed useful by him who makes the combination. It then happens that in the first place we have an empty concept, which, being without any internal rule (owing to this very vacuity), fills itself with a content which does not belong to it--for this it could have only from contact with the representation--and gives itself a _false_ subject. The opposite also occurs, that is to say, a false predicate or concept is posited, a case which will be considered further on. Limiting ourselves, meanwhile, to the first and observing that it consists in the abuse of the logical element, we shall be able to call that mode of error _logicism_ or _panlogism,_ or also _philosophism_ (since the abuse of the logical element is identical with the abuse of the philosophic element). [Sidenote: _Philosophy of history._] Logicism, panlogism or philosophism, is the usurpation that philosophy in the narrow sense wreaks upon history, by pretending to deduce history a _priori,_ as the process is called. This usurpation is logically impossible owing to the identity of philosophy and history already demonstrated, whence bad history is bad philosophy, and inversely. It may happen that the same individual who at a given moment creates excellent philosophy (and excellent history at the same time) may create bad history (and so bad philosophy) the moment after. But this amounts to saying that he who at one moment has philosophized well, may philosophize badly and err the moment after, and not by any means that the two things are possible in the same act. However, the usurpation, logically impossible, is practically effected, in which case, it is not strictly speaking usurpation, although it comes to be so considered from the logical point of view. On the other hand, the claim for the _a priori_ in history is perfectly just; for to affirm a fact means to think it, and it is not possible to think without transforming the representation by means of the concept, and so deducing it from the concept. But this deduction is an _a priori_ synthesis and therefore also induction, whereas the claim to deduce history _a priori_ would amount to a deduction without induction, not _History_ (which is, for that very reason, _Philosophy),_ but a _Philosophy of History._ [Sidenote: _The contradictions in this undertaking._]