Short URL for this page:

[image ALT: Much of my site will be useless to you if you've got the images turned off!]
Bill Thayer

[image ALT: Cliccare qui per una pagina di aiuto in Italiano.]

[Link to a series of help pages]
[Link to the next level up]
[Link to my homepage]

This webpage reproduces an article in
American Journal of Philology
Vol. 53, No. 4 (Oct.‑Dec. 1932), pp360‑363.

The text is in the public domain.

This page has been carefully proofread
and I believe it to be free of errors.
If you find a mistake though,
please let me know!

 p360  The Sacred Treasure and the Rate of Manumission

[This note suggests emending nongenta to nonaginta in Orosius 6.15.5 and et to i.e. in Pliny 33.56.º The conclusion is that the aerarium sanctius of Rome contained gold and silver in 49 B.C. valued at about 12,000,000 denarii, that Sulla had emptied the treasury in 82 B.C., and that the average number of manumissions between 81 and 49 B.C. was nearly 16,000 per year.]

Numerals were so recklessly copied by the medieval scribes that historians in encountering inconsistencies in them frequently give up in despair and omit what might prove to be important information. In attempting an account of Caesar's finances one comes at once upon the confused statements of the amount which he found in the "sacred treasury" in 49 B.C.1 Pliny (33.56) has the amount as follows: laterum aureorum XV, argenteorum XXX, et in numerato |CCC|. This is the text given by Mayhoff, as well as by Sillig, based upon a fair agreement of the better manuscripts. But Pliny does not tell how much the lateres weighed. Orosius (6.15.5) is somewhat more definite, giving — probably from Livy — definite weights: auri pondo quattuor milia centum triginta et quinque, argenti prope nongenta milia. If the gold mentioned by Pliny (15,000 lateres) was the same as the 4135 pounds of Orosius, each later would weigh about 3⅓ ounces and would coin into 10 aurei of the type last coined, that is, the aureus of Pompey weighing one thirty-sixth of a pound (Bahrfeldt, Röm. Goldmünzen, p29). Hence the lateres would be worth about 1,000 sesterces each. This fact, since the Romans regularly reckoned large amounts in 1,000 sesterces, is striking enough to justify the belief that Orosius and Pliny are correct about the gold.

The statements about the silver are not as easy to control, but there is a coincidence here also that seems significant: the "30,000,000 numerato" in Pliny happens to correspond with Orosius, if we read prope nonaginta for prope nongenta in the latter; for 90,000 pounds of silver will make just over 30,000,000 sesterces. This coincidence will probably justify the emendation,  p361 especially since nongenta (about 450 tons!) is absurd, as Mommsen says (Münzwesen 401).

Then what are we to do with Pliny's 30,000 lateres of silver for which Orosius has nothing? If silver bullion, like gold, was stored in pieces worth 1,000 sesterces, then the striking conclusion emerges that the 30,000 lateres make precisely the same amount as the numerato; in other words the numerato is an explicative calculation of the value of the lateres; and instead of et we should probably read i.e.2 Hence it would seem that, with two slight emendations, the amount given by Orosius (4135 pounds of gold and 90,000 pounds of silver) agrees precisely with that given by Pliny (15,000 lateres gold — each worth 1,000 sesterces — that is, 30,000,000 sesterces). The coincidence in the values of the gold and silver lateres, the added coincidence in the equivalence of the silver lateres with the numerato, and finally the fact that all the lateres prove thus to contain metal to the value of 1,000 sesterces seem to me to provide a confirmation of these easy emendations. What Caesar then found in the sacred treasury was a store of gold and silver worth nearly 12,000,000 denarii, enough to pay his legions for several months at the new rate that he introduced.

Now we may take the next step. We know that the aerarium sanctius was established in the fourth century B.C. to provide funds from a five per cent manumission tax in case of Gallic invasions (Livy 7.16.6), and that it was emptied in 209 during the darkest period of the Hannibalic war (Livy 27.10.11). After that it is seldom mentioned during the Republic, though Cicero refers to it in 59 (Att. 2.16). The inscriptions of the Empire frequently record the collectors of the tax (Hirschfeld, Verwalt. 106). There was therefore no intermission in the collection.

But what interests us here is that we may discover that Sulla also emptied this treasury before Caesar. There is no explicit record of the fact, though we ought to assume it because Sulla  p362 was so far bankrupt in 82 that he not only proscribed some two thousand men for their wealth but also melted up the treasures of temples (aurea atque argentea templorum ornamenta . . . conflata sunt: Val. Max. 7.6.4; cf. App. B. C. 1.96). If he used other temple treasures, he could hardly have avoided taking the special store of the five per cent tax. But we can, I think, reach certainty in the matter. We have noticed that the ingots of gold, weighing 3⅓ ounces, would be worth 1,000 sesterces if the gold-silver ratio was that of Pompey's issue of aurei, that is, 36 to the pound. Pompey's aureus was probably struck in 80 B.C.,​3 and it was sixteen per cent lighter than the gold coins struck by Sulla's coiners in 86‑82 and about eleven per cent heavier than those of Caesar in 49‑44. In other words the ingots that Caesar found were cast at a ratio that the treasury had established after 82 and before 80. Hence we can be sure that Sulla had emptied this treasury and, since there was no serious occasion to use it between 82 and 49, we may be sure that what Caesar found was the accumulation of that period.

We may hazard a third step. Since the treasury had accumulated about 12,000,000 denarii in 32 years from a five per cent manumission tax we may discover something about the number of slaves set free at Rome, if we know the average price of slaves. Our data for Rome are very scarce and usually concern high prices paid for very valuable slaves. Our best collection of prices comes from the famous manumission records of Delphi (Foucart, Main d'oeuvre, 107), and these may fairly be used because Greek prices of slaves were largely determined by the Roman purchases at the famous market at Delos. The average price at Delphi is 400 drachmas. To be sure, a price set with a view to the tax that has to be paid will naturally be somewhat lower than normal, but that was as true at Rome as in Greece. We also know that, when the cornº dole became very lavish at Rome after 63, many owners freed their decrepit  p363 slaves for the state to feed them (Dio 39.24). Such slaves would be very cheap. On the other hand the slaves that bought their freedom or that won manumission because of good service were doubtless above the average. These two classes of cheap and expensive slaves will perhaps offset each other. If we allow for some difference between the Greek and the Roman market, we may, I think, assume about 500 denarii as the average manumission price at Rome during the last century of the Republic. The sum of 12,000,000 denarii would then represent nearly 500,000 manumissions during 32 years, or about 16,000 per year. This will seem a reasonable number to those who recall how freely slaves were freed at Rome. Cicero, for instance, though far from being wealthy, had some ten freedmen in his familia. My conclusion, then, is that the sacred treasury contained metal in 49 B.C. valued at about 12,000,000 denarii, that Sulla had emptied the treasury in 82 B.C., and that the average number of manumissions between 81 and 49 B.C. was near 16,000 per year.

Tenney Frank

Johns Hopkins University

The Author's Notes:

1 For the seizure of the treasury see Dio 41.17; Cic. Att. 7.21; App. B. C. 2.41; Plut. Caes. 35; Caes. B. C. 1.14.

2 It would be easy to assume a confusion in the manuscript between the standard medieval abbreviations of et and id est, but I fear that the error is just one of the all too many that we have to attribute to Pliny's excerptors and research helpers, for Pliny's own comment indicates surprise at the size of the amount.

3 Mommsen and Bahrfeldt date these coins of Pompey in 81; others have proposed 61. They are triumphal coins, and the head of Africa upon them and their relative scarcity speak strongly for an early date. Pompey probably did not triumph till 79 so that they were probably not designed before 80 B.C. But it is likely that they have the ratio adopted by the treasury for gold when new accumulations came in after 82.

[image ALT: Valid HTML 4.01.]

Page updated: 4 Feb 19