Pay in the Republic

The Roman army at the start of the republic up until the 1st century B.C. was a militia force, assembled from a cross-section of the populace. There were minimum property requirements for service, which determined which branch of the army you could fight in: the richest men would serve as the cavalry, since horses are by far the most expensive possession you can bring to war, whilst the poorest would fight in the light infantry, armed with only some javelins. (And from the 2nd century B.C. onwards, even these seem to have carried swords and perhaps small shields in addition to their javelins.) The very poorest could not serve at all.

These soldiers were neither paid nor forced to fight: they were drafted, but fought because it was their duty, because it affirmed their social status and position, because they would have been scorned as cowards by their friends and neighbours had they refused, and last but not least because even in the absence of pay there was a very good chance of winning loot and booty. In the earliest days of Rome, they would be called up somewhere at the start of summer, do battle with (or raid) some neighbouring tribe or city, and return home in time for the harvest.

Later in Republican times, as wars started to broaden in scope and length and started to take place further and further from Rome, some form of pay was introduced. Livy claims this first happened during the 10-year siege of the Etruscan city of Veii.

Polybios gives us the rates of pay in the Roman legions in the middle of the 2nd century B.C. He uses Greek coins, since he's writing for a Greek audience: 4 obols a day for a centurion, 2 for a common legionary, and 1 drachma for a cavalryman. We're not entirely sure how much this is in Roman coins. Either way, this was by no means a living wage: it served more as a way for soldiers to pay for expenses whilst on campaign. It's also in this period that we have some records of men enlisting and re-enlisting in armies more than they were required, seemingly seeking to make a kind of career for themselves out of service in the legions, though it is hard to tell how common or widespread this practice was. Clearly such men could not live on the wages alone, but booty might have compensated.

It's interesting to note the comparatively low wages centurions receive: in Augustus' day, centurions were paid between 18 and 72(!) times the annual salary of a common soldier, depending on the rank and status of the centurion. Indeed, at one point Pliny the Younger sends 40,000 sesterces (10,000 denarii) to one of his clients who's gotten a centurion's commission, to "outfit himself." This is more money than a common legionary would have made in his entire 20-25 year career.

It should also be noted that the Italian allies, who made up half the army, did not get paid at all unless their home towns paid them, but Rome did provide them with free grain.

Pay in the professional army

Caesar doubles the wages of his soldiers to 225 denarii a year at the end of the 1st century B.C., so logically they had to have been about half that beforehand. Pay remained at roughly this level until the end of the 1st century A.D., when Domitian boosts it to 300 denarii. Then it stays stationary again until Septimius Severus increases it to 400 or maybe 450 at the turn of the 2nd century A.D. and also abolishes the system where cost for food and equipment is deducted from soldiers' pay. His son gives the soldiers another 50% pay boost just a few years later, which amounts to something like a 500% percent increase in wages. This may seem like an extreme increase in just a generation, until we realise that by this point the Roman empire was starting to struggle with very heavy inflation and coin-devaluation.

Indeed, in the ensuing crisis of the 3rd century, the system of monetary rewards for service starts to break down completely, as central leadership devolves in the face of endemic civil war and inflation renders the nominal payments for service meaningless. Soldiers start to maintain themselves by just seizing whatever goods and services they require, which might lead Roman provincials to see little difference between the barbarians raiding them and the soldiers supposedly protecting them.

The great reformer Diocletian makes great efforts to completely reorganise taxation again at the start of the 4th century A.D., but it remains a system of taxation in kind. In this system, a value is accorded to every kind of produce and service, and all citizens of the empire have to pay their dues in the way best suited to them: so a farmer would pay his taxes in grain, a potter in pots, a muleteer by transporting a certain amount of goods for the army for free. This system seems to have worked surprisingly well, although as Roman recovery continues throughout the 4th century, money-based rewards for the soldiers start to become more prominent again. In concurrence with this system, Diocletian also instituted great production centres for military equipment and supplies, which then made sure the soldiers were properly equipped.

Other forms of reward

The regular yearly pay was by no means the only or even the greatest reward for service. In addition to their pay, soldiers in the Empire would be paid substantial donatives whenever a new emperor took the throne, and sometimes on other special occasions. Augustus bequeathed 75 denarii to every legionary on his death, which you'll recall is a quarter of his annual salary at this point. Claudius pays a whopping 3,750 denarii to each of the Praetorians, who have elevated him to the throne. Praetorians probably got double salary anyway, but this is still about 8 years' worth of salary to them.

Then, of course, there was the retirement money soldiers got after completing their 20-25 year tour of duty. Originally, this reward came in the form of land. (One of the reasons the Pannonian legions mutinied in 14 A.D. was that they claimed they had been given useless swampland as a reward.) Later, this was more and more often replaced by a cash bonus, often in instalments, and sometimes in addition to a smaller land grand. Veterans also got legal privileges and exemption from some forms of taxation. In monetary terms, the reward after Augustus was 3000 denarii for common legionaries, which amounts to some 13 years of salary.

In late antiquity, this system too was replaced by one where veterans got land and supplies in kind again. This is because of the troubles with inflation and taxation I described earlier. So under Constantine, a soldier might get land, a pair of oxen, 100 bushels of seed, a little money, and life-long tax exemption. Should be enough to get you started, eh?

In imperial times there was less chance for looting and seizing booty than there had been during the many offensive wars of the Republic, but this too formed a possible source of rewards. If you were in an army seizing a rich city by storm, you might well do very well for yourself.

Oh, and then there were medals. Glory and recognition is hard to quantity in numerical terms, but it must be assumed that at least some soldiers were very much motivated by such things.

The cost of living

So, what was all this money worth in real terms?

It's actually very difficult to say. We have very few reliable figures for prices and wages from Roman times, outside of military records.

As I mentioned above, there were deductions made to soldiers' wages for much of the Principate. These could be substantial: the few documents we have show total deductions amounting to as much as 50-100% of the wage. This was for things like clothes, food, boots, and also camp festivals. Still, these figures also give us some insight into what stuff cost. Anyone interested in the details: I've copied out one of these papyri in this old post

The late 1st century documents we have list deductions of 240 drachmas (i.e. roughly 240 sesterces or 60 denarii) per year for food. It lists another annual 145.5 drachmas for clothes. This would then give us a rough figure of 95 out of 300 denarii (this is after Domitian's raised the wages) for basic living expenses, minus housing.

We can't be sure whether this was an inflated price (letting the government make a profit) or whether the soldier would supplement this by buying more food and clothes with his own money. Nor can we be sure whether prices were different in other sections of the empire: as is the case so often, our best records come from Egypt. (Where the climate preserved many papyri) But documents found elsewhere in the east list the exact same prices for food, suggesting this was a standardised deduction rather than anything reflecting market prices.

A careful conclusion would be that a legionary's basic wages, at least in the 1st-2nd century A.D., was more than subsistence-level but by no means amounted to real wealth. However, additions like donatives and the missio nummaria paid on retirement could make a soldier quite well off by the time he left the army. Combined with his legal and tax benefits, he could well expect to be a notable and respected individual in whatever community he chose to settle. Alternately, many veterans remained near the legionary camp after retirement, where they would maintain their old connections and bonds of friendship. When they married, their sons would often enlist in their old unit.

Legionary officers from centurio upwards, meanwhile, were very well paid indeed and would belong to the elite of the Roman Empire by any metric.

Sources:

• Goldsworthy, A., the Complete Roman Army
• Erdkamp, Paul. A Companion to the Roman Army. 2007.