In the Roman Empire, what was the ratio of free people to slaves?

Factual Questions
1

We often read that the Roman Empire was “based on slavery.” Which raises the question of how many slaves there were – and I’ve never read even an approximate figure, in any book on Roman history. A ratio of, say, ten slaves to every free person would have been impossible, wouldn’t it? How would the masters prevent rebellion?

Also, was there any significant number of free people who owned no slaves? Or was practically everybody either a slave or a master?

And what about the Byzantine Empire? For some reason, the histories of that period that I’ve read don’t even mention slaves. Did the institution die out after the Empire converted to Christianity?

2

As to your last question, slavery existed throughout the Byzantine empire till it fell, and after that, of course, it was still practiced by the Ottomans.

3

Seems the best you can hope for are some educated guesses per below.

4

From what I’ve read, a Roman was considered desperately poor if he could not own even one slave.

So I gather most Romans owned at least one slave.

5

Coincidentally enough, despite the fact that my own field of study is 20th century America, i’ve been reading about Rome recently while doing research for a professor.

Based on the works i’ve been looking at, many of which are economic histories and labor studies, the summary provided by Whack-a-Mole, above, seems to fall well within the accepted limits. One of the more recent studies, Metropolis and Hinterland: The City of Rome and the Italian Economy, 200 BC - AD 200, by Morley, summarizes earlier findings and attempts to arrive at a conclusion.

He points out that the figures of Beloch and Brunt have tended to dominate the field, giving a total population for Italy of about 6 million, with about 2 million of these people (or about one third) being slaves in the late first century BC and early first century AD. In contrast, Tenney Frank offers considerably greater figures, putting the free population at up to 10 million, with another 4 million slaves.

Morley notes that deciding between Beloch and Brunt’s figures, on one hand, and Frank’s on the other, cannot be done “on philological grounds alone.” Instead, he attempts to look at the figures “with respect to economic and demographic considerations.” From an economic viewpoint (agricultural production, population sustainability, etc.) he concludes that Beloch and Brunt’s Italy was “slightly underpopulated,” but that Frank’s was “far too crowded.” From a demographic viewpoint, taking into account rates of natural increase, the burden of wars, migration patterns, etc., he comes to a similar conclusion, but seems to lean more towards the lower end of the scale.

Population figures for this period are notoriously difficult to determine, even with the relatively comprehensive nature of the Roman census. And it’s worth noting that these figures i’ve been giving are for the late Republic and early Empire only, and do not necessarily carry over to the later Empire and the decline of Rome.

6

ivylass: From what I’ve read, a Roman was considered desperately poor if he could not own even one slave.
So I gather most Romans owned at least one slave.

I don’t quite see how that can be so. If, as the estimates seem to suggest, the number of slaves was less than half that of free Romans, and if many well-to-do people owned dozens of slaves (and I’d guess that the very wealthiest had hundreds, at least), then there would be something like two-thirds or three-quarters of the free Roman population that didn’t own any slaves.

It’s hard to see how they could all be “desperately poor”.

7

Still, I guess it could easily work out that the majority of free Roman families owned at least one slave, even though the ratio of slaves to individual family members would be much less than 1. Maybe that’s what your sources were intending to convey?

8

Don’t know where Ivylass was reading, but I suspect it might be coming from writings of a Roman upper-class enough to learn to write and have the time to do so. And I’m no expert on Roman writing, but my guess is that ‘anybody’ in this case really means ‘any patrician’ (i.e. upper-class person).
And that it’s sort of the equivalent of two society ladies talking about how Heather’s new boyfriend is so poor he can’t even afford to vacation on Martha’s Vineyard. Imagine!

9

Recall that the Roman Empire was mainly a two-class system. It is easy to think that nine tenths of the free population were desperately poor, that is, peasants. In fact, that is exactly what you would expect.

10

Up to the end of the Republican era, the censors kept track of the citizen rolls. Each citizen was categorized into one of five classes, according to his financial status. The very poorest citizens, however, were excluded from any of these classes, and, up until 107 B.C., were not even allowed “the privilege” of military service. These were the people who lived hand-to-mouth, those with little more in the way of worldly possessions than the clothes on their backs. When they made their obligatory appearance at the censors’ booth in the Forum to be counted, they presented no more evidence of livelihood or estates than their own selves, and were thus given the sneering nickname capite censi, or “head count.”

While I can’t find any explicit reference to back me up, I believe that owning even a single slave would have been sufficient to place a citizen at least in the fifth class. Thus the use of slave ownership as a symbol of social status.

11

A typical Roman citizen would have been a farmer, probably a tenant, but if he was lucky also owning a little land. Farming was a family affair and a man, that is the head of a houshold or pater famlias, would be well off if he could afford a slave or two to help out. It must have been a hard life, but to the Roman way of thinking this was an ideal to be looked up to. Yes, most such Romans were very poor.

Of course, some farmers - that is some families - grew to be very rich and owned huge estates employing vast numbers of slaves.

I don’t know what proportion of the population was in slavery, but it is clear that the proportion varied significantly with time. When slaves were cheap, estate owners would treat slaves as little better than farm animals and essentially work them to death. On the other hand, when there were fewer slaves, owners would set them free. Freedom did not break the link between a slave and his master, the freedman would remain a patron of his former master. The motivation to set slaves free was that they would become tenant farmers working the land rented from the former master - this was simply more efficient, freedmen would work harder than slaves.

It is surprising that a freed slave would automatically become a Roman citizen, doubtless this is the origin of many of the landless head count.

12

Not only that, but in the period after the institution of the corn dole, setting a slave free often meant that he was fed at public expense, rather than at the expense of the master. And, as you note, former slaves often continued to work for their masters after being freed, and i believe that Roman law also forbade a freed slave from going into trade in direct competition with his master.

So, if you were a slave who had worked in a pottery shop for a master who had a pottery business, you could not, upon being freed, start your own shop. Consequently, many former slaves had little option but to continue with their former masters.