There’s a useful term in the modern study of international relations, called “escalation dominance”. What escalation dominance means is that in any sort of conflict, there’s a big game theoretic advantage to being the one who decides how nasty it’s going to be. Ancient wars usually moved slowly up a ladder of escalation, from dudes yelling insults at each other across the border, to some light raiding and looting, to serious affairs where armies made an actual effort to kill and subjugate each other or conquer land.¹ Highly mobile forces tended to work best at the lower rungs of the escalation ladder, and Rome frequently allowed conflicts to stay simmering at this level. But the existence and loyalty of the legions meant that it was their choice to do so, because they could also choose to slowly and inexorably march towards your capital city killing everything in their path and doing something truly unhinged when they got there, like building multiple rings of fortifications or a giant crazy siege ramp in the desert. And, paradoxically, the fact that they could do this meant that they didn’t have to as often. Thus the threat of disproportionate escalation became the ultimate economizing measure, by preventing wars from breaking out in the first place.

If deterrence fails and you have to fight, then the next best way to economize on force is by making somebody else do the fighting for you. In the late Republic and early Empire, much of Roman territory wasn’t “officially” under Roman rule. Instead, it was the preserve of dozens of petty and not-so-petty kingdoms that, on paper at least, were fully independent and co-equal sovereign entities.² Rome actually went to some effort to keep up the charade: the client rulers were commonly referred to as “allies” [socii], and Rome took care never to directly tax or conscript their citizens. But, to be clear, it was a charade. If any of these “allies” ever wanted to leave the alliance or conduct any sort of independent foreign policy, he would not continue to be a king for long. Oftentimes the legions wouldn’t even have to show up — the terrified citizens of the client kingdom would overthrow and execute their wayward ruler themselves, in the hopes that Rome might thereby be induced not to make an example of the citizenry.

What was the point of all of this complicated kabuki theater? Once again, it’s about economy of force, this time on both the “input” and the “output” sides of the great machine of the state. On the input side: efficient government is hard to scale. Roman provincial governors were legendarily corrupt, and could get up to all kinds of mischief out there without supervision. Having a Roman ruling a whole bunch of non-Romans was also bound to cause resentment: it could lead to rebellions, or worse, tax-evasion. All of these problems were solved by pretending to have the barbarians be ruled by one of their own, a barbarian king. He could collect the taxes, and suppress revolts, and generally keep an eye on things. Moreover, as a fellow barbarian, he would know better how to keep his subjects in line, and would be less likely to commit an awkward cultural blunder. On the output side, he could also deal with border raids and other low-intensity threats. This exponentially magnified Roman military power, because it meant that instead of being stuck on garrison duty, spread out along the frontiers, the legions could be concentrated in a strategic reserve. They could then be deployed for “high intensity” operations in some remote part of the empire without worrying that they were thereby leaving the borders unguarded: operations like conquering new lands, or persuading a rebellious client kingdom that their interests lay with Rome.

If you can’t make somebody else do the fighting, then the next, next best thing is to carefully choose the time, circumstances, and location of the fight. Ideally you would muster a heavy concentration of your own forces and confront the enemy while they’re still dispersed. Ideally the ground would be thoroughly surveyed, well understood, and perhaps even prepared with static defenses. Ideally your own forces would have ample supply and good lines of communication, and your opponent would have neither of these things. It was to all these ends that in the second of the periods Luttwak surveys, the Romans built the limes³ — a massive system of defensive emplacements. These extended for thousands of miles around almost the entire frontier of the empire, but the most famous portion was Hadrian’s Wall.

John Psmith, “REVIEW: The Grand Strategy of the Roman Empire by Edward Luttwak”, Mr. and Mrs. Psmith’s Bookshelf, 2023-11-13.

---

1. This is actually also true of modern wars, and if you think you have an exception in mind, you may just not know the history that well. For instance, the current war in the Donbass wasn’t really a surprise invasion, but is best viewed as the latest and most violent stage of a conflict that’s been slowly ratcheting up for a decade.

2. Were you ever confused by who exactly this King Herod guy in the Gospel stories was? Why was there a king and also a Roman governor? He was precisely one of these client rulers!

3. Pronounced “lee-mays”, not like the fruit.