Number of legions at Zama

[Sean Manning]

Well, we have our views, and I think it's all worth considering, given the exiguous nature of the timeperiod.
Sean Manning:

...Thus I think it is dangerous to assume that our sources tell us of every signigicant detachment from an army. It is my impression that ancient historians just ignored noncombat casualties unless they were unusually numerous or picturesque...

I agree, Sean, and you put forth an opinion of mine with much more succinctness. Unless it was substantial, with regards to an army being vastly under paper-strength or a virulent disease afflicting it, I feel we can assume the numbers of garrisoned troops and non-combat casualties were minimal in quantity (not necessarily negligible). But the initial proposed figures, well, that's a different argument. I think revisionism has 'caught' the ridiculous exaggerations, whether the Persians and their allies at Gaugamela, Caesar's foes in Gaul, or the numbers killed, from both sides, at Chalons 451 A.D.

Indeed, after Gaugamela, Alexander received some 47,000 men from various points (not just Greece). But there were also many demobilizations.

I too cannot accept that such losses were typically negligible. Rather, I think that our sources will not tell us about them unless they were considered especially numerous. A Greek hoplite army with good cavalry support, operating in Greece in the summer, would take negligible day-to-day casualties. An army overseas, especially in forests, rough ground, or deserts, would take significant casualties outside of combat. What this means for Zama I will leave for others familiar with the campaign to consider.

Geoffrey Parker has noted that the Spanish Army of Flanders suffered about 2% attrition per month, whether in combat or not. There were more nasty diseases in 16th-17th century Flanders than the Roman world, and support for troops was somewhat more primitive (eg. tents were scarce). On the other hand, this was probably the best army in the world of its day, and it was living in fairly easy terrain (except during sieges) and the richest region in Europe.

On the other hand, early modern armies had logistical problems in part because they exceeded the capacity of their governments to support. Kingdoms with small bureaucracies and incomes took on the responsibility of supporting huge professional or mercenary armies for long periods, and things were very bad for a century or so while they learned how to do this again. Ancient armies underwent no such sudden leaps in size (excluding Philip II's creation of a standing army tens of thousands strong). So the comparison must be used with caution, but I think it can tell us many useful things we would never know from ancient sources alone.