Around the turn of the twentieth-century is probably as good a place to pin the tail on the historiographical donkey as any. It's when history began to be taught not simply as a part of the suite of knowledge expected of future statesmen, but also when historians really sought to understand the societies they studied. It wasn't always correct, it wasn't always absent of bias, it wasn't often as interdisciplinary as modern history, nor as comparative, but it should be identified as the moment when a conscious move was made away from these faults.

An interesting paper on the issue is R.I. Moore's inaugural lecture at the University of Newcastle (1996).

Most would say that Thucydides was the first true historian. The way he wrote about the Peloponnesian war was fairly impartial and analytical. After that a few people tried to follow in his foot steps, like Xenophon, Polybius and Tacitus.