The Romans' relations with Germanic people began before the annexation of Gaul, and they were not exactly to the advantage of Rome. Most notably, several legions had been defeated by a massive migration coming from the North, which aggregated several Germanic tribes (the best-known of which are the Cimbri; it also included the Teutones, whose ethnic name has had a surprising fortune). The defeat of Arausio (Orange, in southern France) was particularly humiliating, and resulted in the death of several legions; sources (such as Livy, 67.1) point to a body count higher than in Cannæ. This first encounter arguably gave the Germans a durable aura, even though later engagements were to favour Roman troops (e.g. against Ariovistus: Iulius Caesar, Commentarii de Bello…, 1.50). A recension of victories and defeats is not the good way to think about these events; for the Romans, winning was the normal thing to do. A people that managed to crush Roman troops was a force to be reckoned with.

It is also true that individual Germans were considered to be worthy warriors. It is hard to know what to do of such an assertion; fierceness is a traditional feature of barbarian people in Roman representations, and the more barbarous you are, the best fighter you are (actually, there might be some truth in this idea). Germanic societies, before they came in contact with the Romans, were fragmented in several small tribes, which must have known a good deal of infighting, therefore guaranteeing a constant military training. It must also be underlined that a good part of the male population of a given tribe must have taken part in these fights; this is the reason why Germanic coalitions succeeded in fielding troops that matched the Roman Empire's, even though their lands were far less populated than, say, Italy or (the province of) Africa.

Another element of importance is that tribal societies were much harder to control and conquer than “states.” The Roman troops had no problem in conquering the Hellenistic west: they only had to topple the local king (sometimes the ruler went as far as bequeathing his kingdom to Rome) and to replace his bureaucrats by their own (often by recycling members of the former élite). On the other hand, if you vanquish a tribe (or even wipe it out), you still have to deal with numerous neighbours. Indirect control is also harder, because you have to deal with multiple actors. And even if you manage to exercise some kind of overlordship over the area, revolts are probable — much more than in societies where people are used to be submitted to distant powers (a good example for the Germans would be, in Tacitus, the account of the revolt of the Batavi, in 69 AD (Annals, 4.15 and sq.). Finally, it is very probable that taxes on Germania would never have repaid the mere cost of military occupation/frontier troops; Romans preferred to have a stable border supplemented by a series of client kingdoms than to annex profitless territory (except for the occasional venture of a general, but the system of the Early Empire limited this kind of military adventurism which had been so characteristic of the Late Republic).

The picture is quite different with Gaul. 1st century BC Gaul was already quite “state-like” in many regards. By the end of the 1st century, many local kings had been replaced by aristocracies, in a way that could be compared to the transition between the Roman monarchy and the Republic. The conspiration described in the first book of the Commentarii mentions several of these oligarchs, many of whom had ties with the former monarchs and wanted to get their throne back. Social stratification was therefore quite developed. Moreover, some tribes, such as the Arverni (whose territory centered on the Massif Central, a mountainous region in southern France) controlled wide areas of land, and were able to mobilise important armies. The denarius had already been implemented as a kind of common money, which suggests a high level of political coordination. If we think, once again, of the state model in terms of “contagion,” such a level of development may be imputed to the role of Marseille, a beacon of Greek presence in the West — whose influence is clearly shown by the adoption of Greek in Gallic epigraphy, among other elements.

Voilà for the Republic and the early empire. The picture is different in later periods: Roman influence (in the guise of subsidies, army recruitment, trade) had contributed to a concentration of power in the Germanic world, and therefore to the apparition of larger confederations. There was still an outstanding military culture amongst the Germans, who were arguably more threatening than before, but the gradual evolution towards vast entities and the increasing levels of social stratification had changed the situation.

(I must confess that this post is mostly a reworking of previous contributions)