AHC:Light cavalry tactics more favoured than heavy cavalry tactics in Europe by 1000.

[Petike]

This is very true. Using levy for assault would result in pile of corpses followed by mass desertion. If for whatever reason you are attempting an assault you need to have dedicated, armoured men to do the job. Even then chances are not great but way better then peasants.

Very well said. The idea that only peasants did the sieging is over the top. There is ample evidence that, at most, peasants or support staff in a feudal army just helped out with the camp, or helped with digging a siege tunnel under a fort or castle (for later lighting). They sure as hell weren't forced to charge a castle in their linen shirts, with only siege ladders and pitchforks for weaponry. Sieging was, if anything, a matter of skilled noblemen, soldiers and even trained craftsmen and engineers if need be...

Non-mercenary Roman cavalry You made a distinction there, between Roman cavalry (that was contextualised as non-Barbarian) and non-Romans cavalry qualified as mercenaries. You didn't have such distinction, and basically, only armies fighting as nations (foederati) can be really called non-Roman.

The auxilliares, just like the mercenaries, were from all over the empire. Though the mercenaries also could have been from abroad, unlike the auxilliares.

I'd say that, when it comes to the quality of the cavalry, the auxilliare cavalrymen had the upper hand in terms of military discipline and so on, while the mercenary cavalrymen had the advantage of their own custom tactics and their own hard-earned combat experiences.

And, as I've said earlier, Roman cavalry, especially heavier cavalry, did improve quite a bit by the late stages of the empire. But in the early centuries, and especially in the pre-imperial period, the overall quality of Roman cavalry was hit-and-miss.

Which is contradicted by their general use and victorious campaign against Persians or Barbarians.

But was cavalry really so essential to most if not all Roman victories ? I don't get the impression.

I'm not sure about my linguistic skills, but "not-exceptionnal" shouldn't be the same than "rubbish", IMO.

All right, rubbish might be pushing it as an expression, but I think it is true that, compared to the cavalry of other powers they were often going against, the armies of the Roman Empire had more mediocre cavalry. Possibly more generic, if cosmopolitan. Again, not a bad thing, as it gets the job done. But Romans couldn't be specialists in everything and for most of the history of the empire, cavalry was something of a weaker spot in the Roman armies (also counting mercenaries, not just legions and auxilliares).

 

[LSCatilina]

The auxilliares, just like the mercenaries, were from all over the empire.

Not really : if something they were more on local recruitment, critically with the provincialisation of armies. And you let aside the critical importance between auxiliaries and mercenaries, that is the structural and institutional integration of the former, when the latter weren't.

Roman mercenaries (as in "professional soldiers whom behaviour is first driven not by be part of a political communauty but by greed; having the triple qualities of specialist, apatrid and venality") are fairly rare, even in the latter empire.

Eventually, we're talking of communauties fighting for political/social reasons there, and that's a major point when it comes to their political/social definition.

I'd say that, when it comes to the quality of the cavalry, the auxilliare cavalrymen had the upper hand in terms of military discipline and so on, while the mercenary cavalrymen had the advantage of their own custom tactics and their own hard-earned combat experiences.

And, as I've said earlier, Roman cavalry, especially heavier cavalry, did improve quite a bit by the late stages of the empire. But in the early centuries, and especially in the pre-imperial period, the overall quality of Roman cavalry was hit-and-miss.

But was cavalry really so essential to most if not all Roman victories ? I don't get the impression.

Look at Constantine campaigns in Italy for exemple : not that it was a Constantine's proper feature, his rivals used an important cavalry as well; with the emperors (or emperor-wannabe) leading their mounted troops.
Of course, it's in the early IVth century (where heavy cavalry became standard): we're talking of a time where differenciation between "Roman" "Auxiliary" or "Laeti" is a bit moot, if not irrelevant.

compared to the cavalry of other powers they were often going against, the armies of the Roman Empire had more mediocre cavalry.

Persians or Germans would like to differ on the "more mediocre cavalry" (especially when it comes to masters of the weapon as Sassanians)

Not that Roman cavalry was some sort of imprevious super-weapon : you have as well exemple of them being a reason (less sub-tactically than tactically that said, and the commander was often in-fault for that, as in Adrianople). But overall its quality wasn't noticably worse than their opponents, the problem being more about tactical and operational concerns.

Possibly more generic, if cosmopolitan.

Cosmopolitain would imply a range of various defined identities, than a certain Romanity would dominate. But it's not even certain it was that : "being Roman as a soldier" was found on some "Barbarians" tombs, for exemple. Tactically,structurally for what matter equipment and behaviour and what may be more important for non-foederati, these soldiers weren't different : any distinction based on alleged origin is largely arbitrary, at least up to the mid and latter IVth.