marjaerwin

I've been pondering the strength and overall organization of the Late Roman Army. Warren Treadgold argues that the expansion and the split between field and frontier forces occurred in the reign of Diocletian. Others such as AHM Jones argue that they came later.

Zosimus reports figures for the strengths of the armies of Constantine and Maxentius in 312, and Constantine and Licinius in 324. These are too high to account for the field armies alone. These seem to refer to the total strengths, field and frontier, of 4 different armies, and then the heavier naval forcees, and then the lighter ones. They add up to 645,000 personnel, and Agathias also reports 645,000 in the old Empire.

Treadgold suggests that Zosimus has dug up figures for Constantine, Maxentius, Licinius, and Maximinus Daia in 312, and has creatively reinterpreted them as strengths for Constantine, and Maxentius, in 312, and Constantine again, and Licinius, in 324. But this gives Maxentius by far the largest army. My suspicion had been that there were permanent field armies, of around 130,000 people total, hat these were grouped with the Praetorian Guard, and that Zosimus had assigned them all to Maxentius because of that.

In this list, Zosimus gives Maxentius 188,000 people. Zosimus elsewhere gives Maxentius 120,000 people, and the Panegyrici Latini gives him around 100,000. So it's likely that this list overstates Maxentius's strength in some fashion.

But Jones and other historians make clear that the basic distinctions in Diocletian's reign were between the legions and vexillationes, on one hand, and the old auxiliaries, on the other. There aren't enough Praetoriani, Lanciarii, Scolae, Comites, etc. to account for 130,000 people on the, uh, other other hand.

It seems most likely that Zosimus's figures refer to the forces of Crispus, Constantine I, Constantine II, and both Licinius I and Licinius II.

Yes, Constantine II was only 8 years old. I guess his father wanted his son to get an early start, and trusted his aides to offer good advice.

This solution to the "problem of Zosimus" leaves the problem of the organization of the Late Roman Army between Diocletian and Constantine completely unsolved.

Walter Scheidel, in his 2007 working paper “Roman population size: the logic of the debate” notes that the best reconstructions of Roman population distribution seem to clash with the best reconstructions of medieval and early modern population distribution. He presents the following estimates:

Peninsular Italy:

4.5 million to 5.5 million in the 1st Century CE (low count)
4.1 million to 5.0 million in the 13th Century CE

Cisalpine Gaul

1.5 million to 2.5 million in the 1st Century CE (low count)
6.0 million to 7.0 million in the 13th Century CE (?)

Higher population counts leave the distribution problem unresolved, and create other problems. Scheidel considers this one of the major problems in understanding the demography of the Roman world.

His working papers are online at http://www.princeton.edu/~pswpc/papers/authorMZ/scheidel/scheidel.html

It's fascinating, but I think Leslie White was three-quarters right with his plow theory. He argued that the heavy plow was introduced in the 6th and 7th Centuries, revolutionizing northern European agriculture, and spreading Slavic language and culture with the technology. In fact, the heavy plow incorporated new technologies, such as the coulter and the mouldboard, which developed over the entire 1st Millennium CE.

Although this probably misses much of the complexity of the issue, I think it makes sense to distinguish two agricultural traditions:

An older agriculture, which developed before the 1st Millennium CE, which was adapted to lighter soils, which used lighter plows, square fields, and two-field farming.

A newer agriculture, which developed during the 1st Millenium CE, which would adapt to heavier soils, which would develop heavier plows, strip fields, and in its later version three-field farming.

If we look at soil maps, it's clear that most arable land – excepting very rocky soils – in mediterranean Europe consists of alluvial soils and relatively light soils. Most arable land in northern Europe consists of heavy soils, including forest soils and wetland soils. About two-thirds of the arable land in Cisalpine Gaul had forest soils.

Some loess soils were accessible to the older agriculture, but most northern European soils – and Cisalpine Gallic soils – were marginal or inaccessible to the older agriculture. As the newer agriculture developed, it seems to have allowed increased settlement density in the intermediate soils, and gradual colonization of the heavier soils.

Pre-Roman and Roman settlement was generally concentrated in alluvial soils and other light soils. But, for example, Chernyakhov-culture settlement was generally concentrated near intermediate forest soils, such as those of Bukovina, circling the Carpathians and north of the Balkan mountains – as in the settlements outside Nicopolis, at Gradishte, and at Veliki Preslav.

When the newer agriculture was beginning to develop, it allowed increased settlement of intermediate soils, such as the Gothic settlement of Bukovina; as it continued to develop, it allowed increased settlement of heavy soils, such as the Slavic settlement of the Black Earth.

The newer agriculture allowed a population explosion in northern Europe – perhaps doubling or tripling between the 1st and 4th Centuries and doubling or tripling again between the 4th and 10th Centuries. But the newer agriculture disrupted existing field systems and, by opening new land to settlement, weakened established landlords and, it would seem, emperors.