Collections: How to Roman Republic 101, Addenda: The Socii

This week, as an addendum to our series on Roman civic governance (I, II, IIIa, IIIb, IIIc, IV, V), we’re going to take a look at how Rome handles those parts of Italy it controls but which it does not inhabit. These are Rome’s ‘allies’ (socii), a euphemistic label for the Italian communities the Romans have subordinated – sometimes diplomatically, frequently militarily. Rome’s governance – or one might say, lack of governance – of these communities is one of the most unusual elements of the Roman Republic, a remarkable break with how other Mediterranean states attempted to handle their conquests.

This is no unimportant topic. I tend to think – and indeed argued in my dissertation and will argue in my first book (when it appears) – that the Roman alliance system in Italy was the single most crucial institution for the Roman Republic’s military success (though many other elements of the Roman ‘system’ matter quite a lot). At any given time during the third and second centuries, more than half of the deployed strength of Roman armies – not just the men, but their equipment (which they bought), their pay (provided by their home communities), their junior officers (also from home) and their fighting ability (never meaningfully less than that of the Romans) – was not Roman, but drawn from the many non-Roman communities of Italy.

But in the meantime, if you want to be a good socius of the blog, please share what you are reading, as I rely on word-of-mouth to gain new readers. And if you really like what you are reading and what to support it, you can become an amicus of the blog and upgrade your status to civis sine suffragio over at Patreon; supporters at the patres et matres conscripti level even get to vote on future topics (like our recent look at Greek and Phoenician colonization). If you want updates whenever a new post appears, you can click below for email updates or follow me on twitter (@BretDevereaux) for updates as to new posts as well as my occasional ancient history, foreign policy or military history musings, assuming there is still a Twitter by the time this post goes live. I am also on Bluesky ( and (less frequently) Mastodon (

Beginning of the System

The earliest indicator we have of what is going to be Rome’s socii-system is the Foedus Cassianum (‘Cassius’ Treaty’) concluded with the communities of Latium – the Latins – in 493. That is, of course, quite an early date and while we have narratives of these events from both Livy and Dionysius of Halicarnassus, we have to be quite cautious as they are operating at great chronological remove (both writing in the first century B.C.) and with limited sources (something both actually more or less admit). According to Livy (2.18) the issue had begun with thirty Latin towns conspiring in a league against Rome (which does not yet have any imperial holdings), to which Rome responded by going to war. The timing, just a few years after the expulsion of Rome’s kings and the formation of the res publica may be suggestive that the Latins had formed this league to take advantage of the political crisis in Rome, which was the largest town in Latium, in order to throw off whatever Roman influence they may have been under during the period of the kings.

In any case, the Romans win the war and impose a peace treaty the terms of which, according to Dionysius of Halicarnassus went thusly:1

Let there be peace between the Romans and all the Latin cities as long as the heavens and the earth shall remain where they are. Let them neither make war upon another themselves nor bring in foreign enemies nor grant a safe passage to those who shall make war upon either. Let them assist one another, when warred upon, with all their forces, and let each have an equal share of the spoils and booty taken in their common wars. Let suits relating to private contracts be determined within ten days, and in the nation where the contract was made. And let it not be permitted to add anything to, or take anything away from these treaties except by the consent both of the Romans and of all the Latins

This is the origin point for Rome’s use of what I’ve termed the ‘Goku Model of Imperialism’ – “I beat you, therefore we are friends.” Having soundly defeated – at least according to our sources – the Latins, Rome doesn’t annex or destroy them, nor does it impose tribute, but rather imposes a treaty of alliance on them (in practice I suspect we might want to understand that Rome’s position was not so dominant as our sources suggest, thus the relatively good terms the Latins get). The treaty sounds like an equal relationship, until one remembers that it is the entire Latin league – thirty or more communities – as one party and then just Rome as the other party.

I will still make no apologies for this visual gag.

Rome proceeds, in the century or so that follows, to use this alliance to defeat their other neighbors, both the nearest major Etruscan centers as well as the Aequi and Sabines who lived in the hills to the north-east of Rome and the Volsci who lived to the south of Latium. Roman relations with the Latins seem to fray in the early 300s, presumably because the greatest threat to their communities was increasingly not the Volsci, but Rome’s emerging regional power. That leads to a collapse of the Foedus Cassianum in 341 and another war between Rome and the Latin League. Once again our sources are much later, so we might be somewhat skeptical of the details they provide, but the upshot is that at the end the Romans won by 338.

Rome’s expansion into most areas follows a familiar pattern: Rome enters a region by concluding an alliance with some weaker power in a region and then rushing to the aid of that weaker power; in some cases this was a long-term relationship that had been around for some time (like the long Roman friendship with Etruscan Caere) and in some cases it was a very new and opportunistic friendship (as with Capua’s appeal to the Romans for aid in 343). In either case, Rome formed a treaty with the community it was ‘protecting’ and then moved against its local enemies. Once defeated, it imposed treaties on them, too. Rome might also seize land in these wars from the defeated party (before it imposed that treaty); if these were far away, Rome might settle a colony on that land rather than annexing it into Rome’s core territory (the ager Romanus). These new communities – the Latin colonies – were created with treaty obligations towards Rome.

Note the change: what was initially an alliance between one party (Rome) and another party (the Latin League) has instead become an alliance system, a series of bilateral treaties between Rome and a slew of smaller communities. And they were smaller, because Rome often took land in these wars, so that by the third century, the Roman citizen body represented roughly 40% of the total, making Rome much bigger than any other allied community. This shift was probably gradual, rather than there being some dramatic policy change at any point. Rome accrued its Italian empire the same way it would accrue its Mediterranean one: as a result of a series of localized, ad hoc decisions which collectively added up to the result without ever being intended to constitute a single, unified policy.

The Romans called all of these allied communities and their people socii, ‘allies’ – a bit of a euphemism, because these were no longer equal alliances. We’ll get into the terms in a moment, but it seems clear that by 338 that these ‘allies’ are promising to have no foreign policy save for their alliance with Rome and to contribute soldiers to Roman armies. So Rome is in the driver’s seat determining where the alliance will go; Rome does not have to consult the allies when it goes to war and indeed does not do so. The socii cannot take Rome to war (but Rome will go to war immediately if a community of socii is attacked). This is no longer an equal arrangement, but it is useful for the Romans to pretend it is.

The next major series of Roman conflicts are with the Samnites. Rome is, according to Livy, at least, drawn into fighting the Samnites because of its suddenly concluded alliance with Capua and the Campanians (though Rome had been more loosely allied to the Samnites shortly before). In practice, the first two Samnite Wars (343-341, 326-304) were fought to determine control over Campania and the Bay of Naples, with Rome fighting to expand its influence there (by making those communities allies or protecting those who were) while the Samnites pushed back.

The Third Samnite War (298-290) becomes something rather different: a containment war. Rome’s growing power – through its ‘alliance’ system – was clearly on a course to dominate the peninsula, so a large coalition of opponents, essentially every meaningful Italian power not already in Rome’s alliance system, banded together in a coalition to try to stop it (except for the Greeks). What started as another war between Rome and the Samnites soon pulled in the remaining independent Etruscan powers and then even a Gallic tribe (the Senones) in an effort to contain Rome. The Romans manage to pull out a victory (though it was a close run thing) and in the process managed to pull yet more communities into the growing alliance system. It seems – the sources here are confused – that the decade that followed, the Romans lock down much of Etruria as well.

The Greek cities in southern Italy now at last recognize their peril and call in Pyrrhus of Epirus to try to beat back Rome, leading to the Pyrrhic War (280-275). Pyrrhus wins some initial battles but – famously – at such cost that he is unable to win the war. Pyrrhus withdraws in 275 and Rome is then able over the next few years to mop up the Greek cities in Southern Italy, with the ringleader, Tarentum, falling to Rome in 272. Rome imposed treaties on them, too, pulling them into the alliance system. Thus, by 264 Rome’s alliance system covered essentially the whole of Italy South of the Po River. It had emerged as an ad hoc system and admittedly our sources don’t give us a good sense of how and when the terms of the alliance change; in many cases it seems our sources, writing much later, may not know. They have the foedus Cassianum, with its rather more equal terms, and knowledge of the system as it seems to have existed in the late third century and the dates and wars by which this or that community was voluntarily or forcibly integrated, but not the details of by what terms and so on.

The Terms

It is not even clear that all of these allies were bound to Rome by a written treaty with spelled out terms. I tend to think that they were, simply because that’s how Rome tends to do diplomacy in the periods we can more clearly observe and because our sources seem to think that the very early foedus Cassianum was a written treaty, evidently with a text that survived to them.2 In either case, the Romans had a fairly clear set of expectations for the socii. Now I am going to discuss this as a single system, but it is important to note that this is, in fact, a massive set of several dozen bilateral treaties, not a single system. Since we have the text of none of them, we can’t really talk about differences among them, but it seems almost certain there were some.

Still, we can outline the basic structure relatively well.

Socii kept their own internal systems of government and laws. We can’t always view these systems very clearly, but what we can see – often through inscriptions rather than our literary sources – suggests systems of government broadly similar to Rome’s, with powerful elected magistrates, often some kind of deliberative senate or council and voting assemblies. These internal systems continued to function and we find Italian elites among the socii quickly enmeshed in networks of patronage and hospitality with Roman elites.3 It seems likely to suppose that this would have produced some degree of convergence in the types of governing institutions, albeit that smaller Italian communities wouldn’t have needed quite so an extensive structure as Rome.

As a result, Rome maintained no officials, nor garrisons among the socii, with only rare exceptions. The socii are not quartering Roman armies, they do not have local Roman officials interfering in their business and the Romans don’t have to maintain such officials.

Rome guaranteed the safety and territory of the socii. In the Roman thought, the socii entered into the fides (‘faith, trust’) of Rome; indeed, defeated enemies that became socii will have done so through a surrender known as deditio in fidem, “giving [one’s self] over into the fides [of Rome].” Such trust was a two-way street: the socii incurred obligations to Rome but Rome also had responsibilities to them. Chief among them was that Rome had to keep them from harm. While there’s a temptation to view this obligation very cynically, the Romans clearly didn’t. Rome may not always win, but by the end of the third century, no socius could doubt that the Romans would throw away whole citizen armies before letting the socii be plundered. The Romans had, after all, engaged Pyrrhus outside of socii territory in 280, been beaten badly and then done it again in 279. They had also tried to keep Hannibal out of socii territory to disasterous losses in 218 and 217 and 216. Roman diplomats were forever “harping on about fides” (e.g. Diod. Sic. 23.1.4) as Heiro II of Syracuse complained, but this mattered to them.

Crucially, this security guarantee was as much directed against other socii as it was against external threats. Remember that Rome moves into regions by intervening in existing conflicts in those regions and when it wins, incorporates both parties into the system. That means the smaller regional states are made socii and are being protected against the larger regional states, who are also made socii. Indeed, as Michael Fronda argues4 this provides a measure of resilience to the system. If, say, Capua revolts from Rome (as it does during the Second Punic War), a whole mess of other Campanian states, concerned about Capua’s strength as the strongest Campanian community, might hew closer to Rome for safety. Thus the revolt of some communities locked in the loyalty of others who looked to Rome for protection, which in turn meant that Rome could retain the upper-hand against efforts to break up the system.

The socii in turn had to supply troops for Rome’s wars. We’ll talk about the mechanics of this in a moment. Crucially, this came with a second component, Rome imposed no tribute on the socii. That is profoundly unusual for imperial systems in antiquity, which nearly always impose some kind of tribute or tax, in either money, precious metals or bulk agricultural goods (like grain) upon the conquered. Rome’s ‘tribute,’ such as it was, was only collected in soldiers – albeit soldiers with their pay, equipment and officers (so not simple manpower).

Socii shared in the spoils of successful military activity. This is also profoundly unusual. Livy has a standard formula when reporting Roman triumphs which includes the share of the loot given out to each rank of soldier and as part of that very standard formula, he nearly always includes that the socii got the same amount per soldier (e.g. Livy 40.43.6-7, 45.43.7, note also Polyb. 10.16 which seems clear that the socii participate fully in both the gathering and division of loot). Indeed, the one time this does not happen it produces outrage; when Gaius Claudius gives the socii only half of what the citizens got, they respond by marching in silence in his triumph rather than singing the traditional bawdy victory songs in protest (Livy 41.13.8); this is the only time we hear of this and they protest not at getting no loot, but merely at getting less loot. When the loot was land, the socii seem to have had at least some access to it too; there’s a general consensus that at least some socii settled in Roman colonies placed on captured land and its also clear that the socii could rent the land that became ager publicus, public land rented out by the Roman state, generally at favorable rates.

As Tim Cornell put it,5 the Roman alliance system was, “a criminal operation which compensates its victims by enrolling them in the gang and inviting them to share to proceeds of future robberies.” In the rough neighborhood that was pre-Roman Italy and the ancient Mediterranean in general, where the consequences of losing a war could be so dire, that kind of deal isn’t such a bad one. That said, in the third and second centuries at least, most of the allies – the Latin colonies composed significantly of transplanted Romans excepted – didn’t stick with Rome out of any sense of national unity (there wasn’t any) or great affection. Rather, this seems to have been a pretty hard-nosed calculation of interests: for the elites that ran these communities, Rome protected them from outside threats, backstopped their power internally to some degree and was less bad than whatever their traditional local rival would have been.

Nuts and Bolts

That structure in turn means that there isn’t very much ‘system’ to look at. Rome maintains no bureaucrats among the socii, no magistrate whose job it is to run their affairs, no council or board to keep tabs on them. The vast tax collection apparatus of the great Hellenistic monarchies (the successors of Alexander) may have consumed fully ten percent (or more!) of the revenue they generated;6 the direct cost to Rome of this system was…almost nothing.

Instead we see two normal interactions between Rome and the socii. The first, of course, was the dilectus. We’ve discussed that before, so I’ll just offer a brief recap of the role of the socii in this process. During the first phase of the dilectus, when the military tribunes are determining which of the Roman iuniores will be called up, the consuls meet on the Capitoline with the representatives of the socii; probably local magistrates, but we’re not told this. The consuls have the formula togatorum, a document the exact nature of which is obscure – we’ve already discussed the question before – but which evidently spells out the liability each community has for recruitment, possibly as a ratio or a sliding scale of some sort or maybe an absolute count of men liable for military service in each community. In any case, the consuls inform the representatives how many men their community is expected to send and where the army will eventually form up.

And here our sources…lose interest. Polybius and Livy, upon whom we are primarily reliant here, are interested in political events in Rome, but not among the socii. Still, a few things are mentioned and others seem likely. The representatives would fan out to their communities knowing how many men they needed to send. Polybius then suggests (6.21.5), in a brief comment that each of the communities of the socii would hold their own miniature version of the Roman dilectus and that’s quite plausible. There’s a fair bit of evidence suggesting that the fourth and third century are a period of convergeance in military norms in Italy (the most visible of this is in arms and armor), so by the time the Roman system is visible to us in the third century, chances are most of the socii have military institutions that look at lot like Rome’s, just smaller, because they are much smaller. And a rapid dilectus for the allies should be possible too, because they are smaller in population and forming up much smaller units.

The socii thus take up most of the planning here. We know they appoint a junior officer and a paymaster for the unit (Polyb. 6.21.4-5). That in turn implies some system of electing officers in each of the communities of the socii; these posts would have certainly gone to local notables likely looking to rise in the politics of their own cities. And note the symmetry: the Romans elect a consul to command each army and a quaestor who handles pay; the socii also elect a commander of their detachment and a paymaster to match the role of the quaestor. That of course means the socii will have had to set aside the money to pay these soldiers, though as with recruitment, how they do that is up to them. Since they have no taxes levied on them by the Romans, they are free to raise whatever taxes they can in whatever way they wish; a tax on landed wealth, similar to Rome’s tributum is the most likely.

This also means that it is up to the socii on how to equip their troops. Italian military equipment seems to be converging, so by the second century if not earlier, the equipment of the socii seems little different than that of the Romans and units of socii are effectively interchangeable with Roman legions. Once again, the most plausible solution is that the socii do what the Romans do: restrict recruitment to me of means who are expected to purchase their own equipment. If there are ‘grades’ of equipment sets – like the velites and three ranks of the Roman heavy infantry – within the socii (entirely plausible!) we do not near about them. Likewise, elites among the socii presumably serve in the cavalry and provide their own horses; Rome tends to have more socii than citizen cavalry, though Polybius’ suggestion (6.26.7) of a fixed ratio is clearly a simplification of a general rule.

The one thing Rome administers in the logistics is the food supply: the socii receive their rations as a ‘free gift’ from the Roman people (Polyb. 6.39.14). I doubt this is generosity, but rather a desire for logistical simplification, as doing things this way lets the army keep a single stock of food supplies. If the quaestor wanted to charge the socii for the food, they’d have to coordinate with a dozen different socii paymasters to do it; far easier to just call it a free gift and move on (whereas since the quaestor handles the pay of Roman soldiers directly it is easy enough just to deduct the cost of their rations from their pay, which is what was done). Indeed, it is hard not to note that the entire system maximizes simplicity for the quaestor: pay deductions in camp for Romans keeps the movement of money entirely notional, reducing the amount of raw specie the army has to carry to make out pay, while the entire administrative burden of keeping track of the socii is offloaded on to their own leaders.

For any other kind of problem, well the socii are technically independent allies communities, so they communicate with Rome the same way any other would: they send ambassadors with messages. This is, presumably, what is technically happening every year at the dilectus. So that would follow the standard sequence: ambassadors are received by the consuls, who then bring them to speak before the Senate, which hears what they have to say and then makes recommendations – that is, a senatus consultum – to Roman magistrates. We can see this process play out in 177, reported by Livy (41.8.5-41.9.1): representatives of the socii have a problem (their people are migrating into other communities, but their recruitment burden has not changed) so they send ambassadors to Rome who have to wait until the new consuls come into office so that the consuls can convene the Senate to hear their petition, at which point they lay out their petition and, in this case, the Senate grants their request.7

Finally among the allies there were gradations of legal status. Of course you have the Romans themselves who legally full Roman citizens (cives optimo iure), but you also have Roman colonies settled with Roman citizens in new communities in Italy which often are cives sine suffragio, “citizens without the vote.” They have all of the legal protections of Roman citizens, but can’t vote or hold office in Rome, instead voting and holding office in their own communities. Rome also sometimes incorporates existing communities as cives sine suffragio, but this is sometimes a privilege and sometimes a punishment: in Campania it seems to have been a recognition of closeness in Rome, whereas in other cases Rome used the status as a mean to annex states that had been particularly hostile.8 After this came socii under the ius Latinum – the Latin Right: these were non-citizen allies but who had some part of a package of particular extra rights in the system, namely to contract lawful marriages with Roman citizens, to do business in Rome under the ius civile rather than the ius gentium (that is, to have the same commercial protections as Romans), to be able to migrate freely into Roman territory and in so doing become Roman citizens. Initially unique to the status of the actual Latins in Latium, Rome applies this status to the Latin colonies (thus their name) and so it quickly becomes disconnected with any actual Latin ethnic identity. And then finally socii with no particular special status except their alliance with Rome.

The layers of potential status gave Rome carrots and sticks to use in some of their interactions with the socii: improved status was sometimes offered, though fairly rarely, for loyal service. On the other hand, privileges might be revoked for communities which proved unreliable or disloyal. I should stress that this isn’t necessarily a regular occurrence – we have fairly few reports of status ‘upgrades’ though they do happen (Vell. Pat. 1.14.2-4; Livy 23.5; 38.36); it’s unclear if these are more frequent and just not as remarked upon by our sources.


For the third and second century, the clear significance of this system was what it meant for Roman military power. Polybius preserves the figures from an extraordinary Roman census of all of the allies in 225 (as opposed to the normal Roman census, which only counted citizens), which recorded all of the military-aged men in Italy liable for conscription (so, rich enough to serve). The numbers he gives have been tinkered at by modern historians at the edges, but only on the edges; the raw figures are as follows:9

Socii in the army60,0004,00064,000
Veneti and Cenomani20,00020,000
Note that Polybius breaks out Romans actually in the army versus liable for conscription, but I have totaled them up here.

Which is to say that the socii, collectively, represented something like 60% of Rome’s military capacity. That’s not just manpower either, because the socii pay, organize and equip these soldiers; indeed they do everything except feeding them. The number of socii that accompanied a Roman army was somewhat variable, but they generally represented a majority of the force; from 195 to 169, where we have decent information (though not for every deployment),10 for every Roman legion (normal size of 4,200 infantry and 300 cavalry) there were on average around 6,750 socii infantry and 350 socii cavalry. By the third century, these units of socii fought in the same way and with the same weapons, in so far as we can tell, as the Romans did and were effectively interchangeable with their Roman equivalents.

And these are staggeringly vast numbers. The comparison I use for my students is that Alexander the Great invaded the Persian Empire with a force of 43,000 infantry and 5,500 cavalry, with perhaps as many troops left behind to secure Greece. Rome has seven hundred thousand men liable for conscription. During the worst days of the Second Punic War, having lost more than 80,000 men in its opening years, Roman armies will approach two-hundred thousand men under arms (185,000 in 212 and 211, following Taylor, op. cit.; other scholars suggest figures as high as 225,000). This is only possible because of this system.

In battle, Roman generals typically placed the infantry components of the two Roman legions in the center, with the alae (‘wings’) of the socii on either side, such that a regular consular army (2 legions strong) presented what was effectively a four-legion wide line of battle. Then the cavalry, both Roman and socii, was normally posted to the flanks. While the junior officers – the socii equivalent of centurions – were provided by the socii, Roman generals appointed Roman officers called praefecti sociorum to help them command the alae of the socii. There were six of these fellows per legion, just like the military tribunes and we ought to understand them as functioning in much the same way.

This was a remarkably efficient and powerful system compared to the typical way empires raised military capacity from their peripheries. The normal structure was a military-tributary complex11 where the imperial center imposed taxes or tribute on the periphery and then used that money to pay soldiers. A slight variation on the system was to cut out the middle-man and simply promise soldiers land and non-free farmers (slaves or serfs) on land cut out of those subordinated communities. Such systems could grow to be very, very large, but the efficiency was low: you have to garrison these communities to keep the taxes flowing and you need a veritable army of bureaucrats to collect it.

By contrast, the socii recruit themselves, equip themselves and organize themselves. And they fight hard, because they’re fighting for their home communities, to show their bravery to their own neighbors and family (who are there with them) and for an equal share in the proceeds of victory. While the troops provided by subordinated peoples were often the weak link in Achaemenid or Seleucid armies, the Romans felt perfectly free to use socii as they would use the legions. Polybius makes almost no distinction between them, whereas Livy in his battle narratives often stresses the valor of the socii.12

At the same time, because the ‘deal’ the ROmans offered was good and avoided gross insult to the honor of the socii (unlike many military-tributary complexes, which could be quite blunt about how ‘under the boot’ their subordinated people were), the system was remarkably durable. The socii revolt en masse just twice. In 216, when Hannibal has defeated three Roman armies and is in Italy, about a third of the socii join him; Rome with the remaining two thirds (including all of the Latin colonies) is able to overcome Hannibal and put down the rebellious communities. Then in 91, about half of the socii rise up in the Social War. Their motives are complex: some of the socii clearly ‘want in’ – they want full Roman citizenship, which for reasons beyond the scope of this post, has become a lot more valuable to have by this point. Some of the socii clearly ‘want out’ – they want to break Roman domination over Italy. In any case, Rome is able to peel away the majority of the revolters by promising citizenship and subdue the rest.

For a system of imperial domination that contributed tens of thousands of soldiers to Rome’s armies annually, that is a remarkably stable system. Rome’s great victories would have been impossible without it.

The system also likely facilitated the slow but steady cultural coagulation of Italy. This process ought not be rushed in our minds; even in the second century it seems few of the socii thought of themselves as part of some ‘Italian’ polity. They were Capuans or Marsi or what have you first, socii second and ‘Italians’ never. But the system facilitated the spread of Roman and allied patronage networks across Italy, it encouraged cultural exchange, particularly the spread of Latin which begins to displace local languages by the first century. As a result, when citizenship for the socii comes at the end of the Social War (91-87), the groundwork for what is essentially a national core to the Roman Empire has been laid. By the 30s BC, Octavian’s propaganda can lean on ‘Italia‘ as a concept to rally behind (as with the oath of tota Italia recorded on the Res Gestae).

  1. trans. Earnest Clay (1940)
  2. Most scholars assume, as I do, that there were formal written treaties, but against this, note J.W. Rich, “Treaties, Allies and the Roman Conquest of Italy” in War and Peace in Ancient and Medieval History, edited by P. de Souza, and J. France (2008).
  3. Indeed, Terrenato argues this was a crucial component of Roman expansion, see N. Terrenato, The Early Roman Expansion into Italy (2019), though Terrenato’s view that elite relations were more central and conquest less so is a heterodox one.
  4. In Between Rome and Carthage (2010)
  5. In Beginnings of Rome (1995), paraphrasing Bickerman and Smith (1976)
  6. Estimates in Aperghis, The Seleukid Royal Economy (2004)
  7. In particular that the Senate block the migration of their people into Roman territory and also to close a loophole where by free socii were evidently using Roman manumission as a legal fiction to obtain Roman citizenship. The episode as a whole seems to me plausible, but this latter effort to obtain Roman citizenship by trickery is curious, coming this early; one wonders if Livy is filling out the details of the complaint with bits that would make more sense around 50 years later. Though on the other hand, Livy may well have the record of the senatus consultum that resulted from this, so perhaps he is right.
  8. Notably the Campanians retained a lot of self-government and still get referred to as socii by the sources, whereas Satricum (Livy 9.16.10) and Anagnia (Livy 9.43.24) lost most or all of their self-government when admitted as cives sine suffragio.
  9. Borrowing a chart from Taylor, Soldiers and Silver (2020), 31 who discusses these figures and the various modern scholarly takes on them in much more depth.
  10. for the raw figures, see Taylor, op. cit., 27.
  11. Not my term originally, but Carlos Noreña’s from “Urban Systems in the Han and Roman Empires” in State Power in Ancient China and Rome, ed. W. Scheidel (Oxford: Oxford University Press, 2015).
  12. There are complications with both of those sources, of course. On those complications, see Erdkamp, “Polybius and Livy on the Allies in the Roman Army,” in The Impact of the Roman Army (200 BC – AD 476), ed. L. de Blois and E. Lo Cascio. (2007).

97 thoughts on “Collections: How to Roman Republic 101, Addenda: The Socii

  1. This is really making me look forward to the book when it comes out in full. Why do these things take so long to write?

    Concerning that Tim Cornell quote: When I was a teenager, I played this MMOish browser game called Travian. You played a little community, built up troops, raided other players for resources, and generally tried to survive the anarchy. It definitely had a classical theme, and one of your possible setups was explicitly labeled “Roman”. I took that one with my profile, and went around raiding my neighbors and generally trying to force some weaker ones into my orbit. And I had some vague notions of how the Romans operated ,and tried to imitate them, or at least my understanding of them. While my setup was more like a military-tributary complex, I did very much offer protection for people under my boot and fought several bloody, unprofitable wars to protect my ‘clients’, which produced a degree of support that a lot of other guys in similar positions didn’t have. (which was generally just ‘you pay me money and I’ll stop attacking you. Anyone else does though, and you’re on your own. I tried to protect my tributaries from anyone and induce a degree of cooperation for our defense policies)

    I bring this up because my dad spotted me playing one time, asked what the game was, and when I was explaining how it worked and my overall strategy, he got this puzzled look and said “So….. you re-invented the mafia?” Weird how art imitated life that closely.

    1. The fundamentals of “I protect you and get something out of it” are very basic to social structures past the band.

    2. From my reading of where people have gotten inside mafias or triads and examined their operations without preconceptions, the deal Rome offers is tighter and more generous. Mafia protection costs but does not provide much.

      1. I suspect that for the socii the draw is less “protection” than “equal shares of loot.”

        Recall that during this time, there was no productive activity that you could possibly engage in that could approach the ROI of ganging up on somebody with some wealth and robbing them. Socii are getting a piece of that action, and there was a long time when there weren’t many years that Rome didn’t send legions someplace to chastise somebody for something.

  2. You commented in past articles in this series that Rome was much more comfortable with explicit inequality in political power than the Greeks were. I find it fascinating that this doesn’t extend to their diplomacy – in a place where they’re actually really in charge, that’s the one place where they decline to explicitly say who’s got the upper hand.

    Clearly worked for them, though.

    1. Yeah, I didn’t want to get too deep into the whys and wherefores of that because that’s a significant chunk of my book’s argument, so I want to save that for the book.

    2. As a total amateur’s guess, I’m thinking it’s a mental model sort of thing. If you’re a senior magistrate of a Greek polis that has just subjugated somewhere, your social world consists of other citizens (all ostensibly equal), some sort of resident foreigners (crush them under your sandal and make them work for you!) and slaves (Like the resident foreigners, but do it harder!) You don’t have a very robust sort of mental map for how to treat someone in a socially inferior position that doesn’t involve massive crushing subjugation and openly reminding them of that fact.

      A roman patrician, on the other hand, has to navigate the local political scene where there are numerous gradations of social superiority and inferiority, and has a more calibrated sense of differing levels of hierarchy. They can use that when dealing with “junior partners” in a degree of finesse that doesn’t culturally come easily to someone in the Greek setup.

      1. Roman aristocrats may also have felt more affinity for foreign aristocrats than for their own poorer fellow citizens. You find that attitude among upper class members of highly stratified societies at other times and places.

    3. In the article discussing the patronage system it was mentioned that they wouldn’t explicit call one the patron and the other the client, they would just call each other amicis. This seems like a similar arrangement, all our equal but one is more equal then others.

    4. I suspect that it’s the fairly simple. Within Rome, people were (reasonably) comfortable with the open heirarchy.

      Wheras the Socii would have been rapidly irritated if Rome tried to impose anything.

      There’s also the fact that if pretending that the Socii were equals or nearly equals meant they didn’t generally need to garrison the Socii, well, it’s much easier to pretend than to need to raise an army to put down a revolt.

      1. Somewhere Bret wrote or said something about how by the time of Cicero, the Romans had explicitly realized this: Cicero directly analogizes the relationship between the Romans and the socii to the relationship between a patron and his clientes. (I can’t find the article/podcast right now, so if anyone knows where Bret said this, if be much obliged.)

    5. I don’t think that this is as incongruous as it initially appears — you’re generally only their ally if you either came to them for help or got your teeth kicked in by them, so it’d be superfluous (and a little insecure) to specify that Rome was the boss.

      I’ve also gotten the impression that Rome basically didn’t have an intermediate step between being reasonably polite and murdering your entire city with EXTREME prejudice…

  3. Typos:
    “Gaius Claudius gives the socii only have of what the citizens got”
    have or half?

    “the socii thought of themselves as part of some ‘Italian’ policy.”
    policy or polity?

      1. Another:

        > At the same time, because the ‘deal’ the ROmans offered was good

        Extra capital.

  4. What would have happened if one of the socii had had their leaders line their pockets with a chunk of their military budget, and send enough men but ill-equipped, ill-paid (so they loot everywhere), ill-disciplined, and untrained? Or just not enough men?

    1. Isn’t this assuming the local city state is using tax money to pay soldiers their equipment rather than a more minimal state apparatus relying on soldiers using their own stuff? The former seems to be rarer than is assumed… This is even part of the ‘Marian’ system Bret addressed a few months ago, that does not seem to be the case in the BCs or even early ADs.

      1. You’re right that I forgot about the equipment issue that he wrote about here. “This also means that it is up to the socii on how to equip their troops. … the most plausible solution is that the socii do what the Romans do: restrict recruitment to me of means who are expected to purchase their own equipment.”

        But still: “it is up to the socii on how to equip their troops” and the socii paid their own men. “… not just the men, but their equipment (which they bought), their pay (provided by their home communities), their junior officers (also from home) and their fighting ability (never meaningfully less than that of the Romans)”. Suppose the city of Uvula Magna hadn’t had enough men of means to call up, or didn’t bother; they might have filled out the rank with guys dragged off the streets. Or they shorted the pay, or their troops ran, or … This article also says that the Romans did not govern the socii, so I can easily imagine internal corruption rotting it out when someone got greedy. A current example comes to mind. The run-up to Adrianople comes to mind.

        If so, I suspect that, after Rome had a couple of legions free, it would have led to the fetials enacting the ritual of rerum repetitio against Uvula Magna, and after a bit, a lot of loot being distributed, a big pulse of new slaves arriving in the markets, and a colony on what used to be the land of a city that used to exist.

        I’m just wondering if an example comes to mind. Greed being so common, I wouldn’t be surprised if it happened a fair amount.

        1. Your question reminded me of our host’s interview over at the hellenistic age podcast:
          Now that I skimmed through it really touched a lot of the topics of this week’s post. IIRC one of the really striking points was how (for Rome, possibly the allies) citizenship could be revoked in response to draft dodging!

        2. I suspect that having the Romans as a model they can inspect close up probably keeps their behavior more similar.
          No doubt there were Romans who ended up corrupt, but calling them “conscripts” brings to mind something different than the reality. These are still men of some substance as they can equip themselves, fighting for glory among their fellows and loot.
          They are not pressed men on a boat a thousand miles from home, with officers some of whom don’t really consider them human.
          I’m not sure the men would have accepted guys dragged off the streets – how is one of those going to protect your flank, armed with a pointed stick (which is what they have if you don’t equip and train them)?

        3. The way military service works in this kind of society, the army represents a significant fraction of the entire population. The total population of Roman Italy in this period, for instance, was only a few million people- the 700,000 liable for conscription are effectively all the free young adult men. No community is going to adopt a military system that revolves around throwing those lives away to save money, because it would be obviously self-destructive.

          1. In addition to not wanting to throw lives away now, there’s also looking forward to the future. If you go to war alongside Rome, and lose, it won’t be Rome that will be coming for you, it’ll be Rome’s enemies.

          2. “No community is going to adopt a military system that revolves around throwing those lives away to save money, because it would be obviously self-destructive.”

            Eppur si muove, as Galileo probably didn’t say. There’s an epigram somewhere (from James Nicoll?) along the lines of: Periodically, people are astonished to learn that people will do unspeakable things for money. History is full of what Barbara Tuchman called “folly” (and wrote a book about): policies that people stated at the time were self-destructive, and they pointed out better alternatives … but nevertheless the policies were done. Her first example: nothing induced Troy to bring that wooden horse thru the sacred gate to within the walls.

    2. If it’s not enough men, then that particular socus had broken the treaty, so they get an army turning up to lodge Rome’s protest with the leadership. Or enact regime change if nessecary.

      People probably bought their own equipment, which **tends** to discourage cheaping out. If a soldier **did** cheap out on their equipment, then if they survive the battle where it inevitably breaks, they get issued a replacement and their pay docked the cost.

      Ill discipline? a beating, or (if **all** of a unit were ill-disciplined) decimation of the unit (they split the unit into groups of 10, pick one from each group, and the remaining 9 from said group have to beat the unlucky sod to death) will clear that right up.

      Untrained? The veterans will presumably train the newbies, asnd if the army is completely green soldisers, the socii not being an exception won’t make much difference.

      1. Did the socii have decimation? Did Romans inflict it on socii? (I think there’s a question about whether even the Romans *ever* did decimation.) Did they have equipment available?

        1. I mean, presumably? At a minimum, the guy with Imperium in charge of the army has the authority to execute any soldier under him, so they definitely had the *right* to.

          But the basic point is that Rome could deal with indiscipline easily enough, and it wouldn’t be particularly daunting to discipline an entire unit if necessary.

    3. Part of the question is, what IS the “military budget”? We tend to think in terms of dollars and cents (ie, money), but in the past that wasn’t the case. The “military budget” would be things like food stores, military hardware, ships, and the like. While it would be possible to raid the military pantry, it would also be extremely stupid and pointless. And taking military hardware wouldn’t have worked the same way in the past as it would today. If you could command people, having people around that could enforce your commands with sharp pointy bits of metal would have been normal. It’s much later (and fictional), but Theodin’s household illustrates the point: He used the military budget to support his own household and that was considered a sign of being a good king. So in a lot of ways, lining your pockets with the military budget was not merely accepted, it was expected and sometimes seen as good leadership.

      Ill-equipped is the default state of most levies, as I understand it. Armor is expensive, and I imagine a lot of people didn’t go much beyond a minimum standard. Even sub-minimal armored troops have uses, though (scouts, spies, and the like), and can be augmented by captured/looted armor. (Looting being a common practice of war, ill-pay wouldn’t have been a huge concern to the Romans, I wouldn’t think.)

      As for discipline and training, Rome had a few ways to deal with that. Not sure how much authority the officers had over the auxiliaries, but I know within legions the belief was that the soldiers should fear their officers more than they feared the enemy. Second, Roman battle doctrine had plans for dealing with the Primus, the raw recruits who were certain to break. And there are plenty of times in battle where it’s useful to have a weaker unit, which will fall back–that lets YOU control where the enemy is going to be. Things that are part of the plan are typically not considered problematic.

      I’m also reminded of descriptions of sailors vs soldiers and marines in the 18th Century. Sailors were considered poorly disciplined by land-war standards–they wouldn’t stand in lines, they’d shoot randomly, they’d do all sorts of crazy things. So the sailors were used for all kinds of crazy things. Blowing up bridges, or taking fortresses, or the like. Critically, it wasn’t that sailors lacked discipline–a line of battle ship had to operate with smooth efficiency to merely survive, much less win a fight–it’s that they had a different KIND of discipline. And that made them useful. I think the auxiliaries of the Roman legions would be similar. They won’t fight the way Romans fight, which makes them good for those situations where Roman fighting methods aren’t very good.

      1. “but Theodin’s household illustrates the point: He used the military budget to support his own household and that was considered a sign of being a good king.”

        Yes, but it’s important to remember what part of his household he used it to support: his household guard–in other words, the guys who made sure that he could fulfill his obligations to defend his vassals. Had he spent the money on other parts of his household, like personal servants or whatnot, that would have been a very different matter.

        1. Where’s the line between “personal” and “military” for servants, though? The guy that grooms his horse and the woman who cooks his meals are serving in both military and personal capacity–in the books as well as in real life. For that matter, having servants take care of the stuff they could would benefit the military. The king’s time/effort is better spent on strategy and operations rather than petty stuff anyone can do (this is why executives have assistants now). Even stuff that looks non-military, like showing off–fancy outfits, fancy food, and the like–was a very good thing from a military perspective. A king rich enough to have expensive entertainment is a king that’s done well on the field of battle and can probably defeat you (see our host’s discussion of the horsemen in “A Song of Ice and Fire”, for example).

          So I’m not convinced that there’d be a strict differentiation in the budget between these two pools of money. There is TODAY, yes–but that’s because we operate in a very different system, with very different assumptions about our rulers and very different methods of accounting.

          To be clear: A king that didn’t spend enough on military stuff would have been viewed as a bad king, no argument from me on that count. “Chief general” is one of the three things a king is supposed to do. I’m just saying, I don’t think that societies where taxes were regularly paid in grain is going to be overly-concerned with which budget expenditures came out of. It was all property of the Crown, after all.

        2. In medieval England, everyone in the castle — except the noblewoman, her attendants, nurses for the children, and (for some reason), the laundresses — was male down to the scullions. This was because the castle was not a fortified home. It was a fortress that happened to have the lord live in it. It had a garrison, really, not a household.

          (See Life in the English Country House: A Social and Architectural History by Mark Girouard. He also did an interesting French one, but it assume you know the English one.)

  5. One thing that slightly puzzles me is that, in some ways, this system resembles classic feudalism; in both cases, the hegemon says to local elites “instead of trying to administer your lands with a central bureaucracy, I’ll let you run them autonomously provided you supply me with armed men when I need to protect you or my other vassals.” However, while this troops-not-taxes arrangement is seen as the key to unlocking huge manpower reserves for Rome, it generates the polar opposite results under feudalism, a system which (while well-suited to certain social and political circumstances) is generally regarded as incapable of raising classical-sized armies. What’s the key difference? Is it that Rome’s socii do actually have their own local systems of literate administrators, whereas medieval nobles didn’t? Is it that the Rome-socii relationship scales up better, because it’s between states and not individuals? I’d be interested to hear your thoughts.

      1. But you wrote several posts about how much power and influence elites had in the government of Rome? And in any event, it’s still a deal between a hegemon and a subordinate power.

        I suspect that one difference is that Rome had enough power on its own that (unless there was a major thing like the Social War) they had enough power to crush a few of the socii, and as you noted, the will to immediately do so (like when socii were attacked). I believe that some medieval kings didn’t have that sort of power relative to their lords.

        Were there also restrictions on socii making alliances between themselves? I wonder whether there just wasn’t any solidarity among socii. In contrast, in medieval France, I think there were the ties of marriage or fosterage, and the concept of a social class. (I seem to recall an anecdote about a later king of France haling a duke into his court, and the other dukes of France forming a disapproving array.)

        1. Few things.
          For one: Yes, the Socii, as I understand, are not allowed alliances with one another (although as we see, this does not render them incapable of organized revolt).
          The biggest thing though is that an Italian community has a deeper base of (useful) manpower and much more efficient capabilities for raising them. Feudal states a bad at raising large armies because they have forgone the apparatus of state *at all levels* and lack (indeed deliberately suppress) a social class capable of manning large armies in the absence of a well organized state. So you have a hodge-podge mostly of elites with those elites personally beholden to them (this class representing something more like the Patricians and Equites alone) and various communal militias, mercenaries, and a few well-to-do farmers with greatly atrophied or even non-existent formal systems of mobilization.
          By contrast the Italian communities are each consolidated states who posses the institutions necessary to take a census, summon eligible citizens, and then field and supply men under arms with the state apparatus missing only at the level of Italy as a unit. Similarly, they posses large numbers of free citizens with sufficient wealth to equip themselves well and cultivate this class. So you get essentially one miniature legion.

          You can see some of the difference in Brett’s discussion of the Fyrd in his Helm’s Deep series.

        2. There’s a difference between a republic where the elites have a lot of influence and quasi-sovereign aristocratic landlords

      2. Interesting. I just reread the ‘Institutions’ post from your fall of the West series, and I think it actually sketches out a good answer to my question, but possibly the opposite to the one you’ve given here. There, you do seem to suggest that republican Rome dealt principally with local elites (e.g.
        “Where there were no cities, the Romans tended to make new ones for this purpose. Roman officials could then interact with the city elites (they preferred oligarchic city governments because they were easier to control) and so avoid having to interact directly with the populace in a more granular way unless there was a crisis” and “City governments, which also administered their rural countryside, were run by a town council which consisted of the wealthiest notables of the town – the curiales – in much the same way that the Roman upper-class had dominated the running of the city during the Republic. Roman authority generally protected the curiales and their wealth from the sorts of popular uprisings that tempered many Greek oligarchies in the classical period and in return the curiales managed the population”) and that what kneecapped Rome’s successors is that late imperial centralisation sucked the life out of these provincial cities, leaving later kingdoms without many educated urbanites to whom they could outsource governance. This would appear to suggest that it was the lack of city-dwelling oligarchs, not of free farmers, that was the main problem for any feudal monarch looking to fight wars at scale. However, this article was obviously a sweeping macro-level response to a different question, so probably the two answers can be harmonised somehow.
        An interesting corollary question is why, in the more urbanised environment of the Middle-Republic-period Mediterranean, no-one else seemed willing or able to copy Rome’s alliance system when this was evidently so successful. Still, I’ve no doubt this is also covered in the book!

        1. As nothing more than an enthusiast I wonder how much the Carthaginian empire in Spain was inspired by what Rome had done in Italy. A much shorter lived example, but the end result still was Carthage extracting troops that fed its armies in the following war with Rome. And much later (and shorter) there also was Sertorius and his rebellion apparently with the same goal.

    1. Peter Heather makes the point that in late imperial society, the main division was between the the free (honestiories), from whom soldiers were drawn, and the unfree. The ‘barbarian’ federations had three classes: free, semi-free and unfree. Effectively, two-thirds of barbarians fought, as against half the Romans.

      Over the totality of ex-Roman western Europe, probably as many men were available for military service as under Rome, but no single polity had the resources or logistic capacity to support armies on the Roman scale. That said, Charlemagne and his successors could wage wars over long distances (eg the Avar campaign) and many years (it took over 30 years to defeat the Saxons).

  6. Foederati are often mentioned as a key factor in the fall of the Western Roman Empire. I know that the real picture is complicated but our esteemed host himself spoke of refusal to integrate foederati as a failure of Imperial politics. Why was the socii-system successfull when foederati-system was not? The deal seems broadly similar: communities living by their own laws under their own leaders but subordinate to Rome and requred to provide military units?

    1. I think the inconsistency between the foedus status of the Germanic tribes in Augustus’ early intrusion into Germania, and the actual behaviour of the Roman officials sent north to supervise – ie Varrus – was the major factor in bringing Arminius into rebellion. If Augustus had sent north a man with more common sense and less arrogance, the future of nothern Europe might have been drastically different.

    2. Just a guess, but I’d wager that Romes system morphed into a military-tributary system during the Imperial era, which did not mesh very well with the increasing power of the foederati.

      Also while the soci system was an treaty with an whole society, the foederati, where more treaties with local warlords who brought their own men. It’s like the difference in asking democratcly elected goverment of India to provide soldiers for a peace mission in Africa vs paying Wagner.

    3. Seems like a mix of:

      -Foederati were mixed into Roman internal politics and used/took part in internal political battles, socii were not. So Foederati are in a position to contribute to conflict in Rome, but also aren’t particularly loyal and therefore splitting up the empire isn’t a weird thing to do. Socii might also lack the loyalty, but they aren’t participating in Roman politics in the same way, so they aren’t in a position to compete for favors and create disruption in Rome which might break up alliances.

      -Sounds like romans kept up their ends with socii, but not as much with Feoderati, if I understand the history right. Responses different behaviors will also vary.

      1. Also I think a difference is how, by the time of the Foederati, the fall of the Roman republic and rise of the Emperor had shown that directly using military power for political influence works (or at least worked for some contenders), instead of how it worked during the republic, where military success was indirectly converted to political success (not via armed conflict, but by getting elected to higher offices).
        So a roman having a group of foederati under his order might try to use them directly in political battles, whereas one commanding socii wouldn’t, because military power wasn’t used in political battles (also socii might not want to rebel against Rome, because of the repercussion against their home city).

    4. Our host goes into this at the tail end of the Queen’s Latin series if you want to read his words directly, but in short; socii were integrated into the Roman system and culture and encouraged to think of themselves as Roman, with the goal of Roman citizenship held out as the ultimate carrot. The socii were lead by Romans, in a Roman fashion, and given how colonies were formed from land the Romans conquered, probably lived in and around Roman citizens.

      The foederati lived apart from Romans and were explicitly banned from becoming Roman citizens. They were “barbarians” settled on depopulated land; when there were Romans around, the foederati ruled over them as local kings rather than the Romans having cultural and economic positions above the foederati. When the foederati fought, it was as their own units under their own kings, not organized into and under the command of Romans. And there was no clear method for foederati to become Roman citizens, unlike the socii.

      So the socii had good reason to “buy into” the Roman system and see themselves as Roman (or as potentially Roman). The foederati stayed separate and were barred from becoming Roman, so they had far less reason to buy in to the Empire.

      1. This reflected the change in power between the ‘barbarians’ and the Romans. Earlier auxiliaries and drafts of people from beyond the frontiers (both frequent given Rome’s demographic needs) were divided into small groups and sent to distant places, where they would assimilate over a generation. By late Roman times this was not acceptable to the barbarians, who demanded to be kept together under their own leaders – and were numerous and strong enough to have their own way.

    5. Besides the other reasons mentioned, I suspect a problem of the late Empire was that there weren’t many wealthy neigbors to plunder anymore. Share in loot is a great motivator, but if all the best loot is already under Roman control, they are the most attractive target for an attack. (Roman elites also seem to turn towards plundering each other – through falls accusations leading to confiscation of property. Well, that is at least the impression I got from reading Ammianus Marcellinus.)

  7. Just how unique was Rome´s system of alliances in terms of being a military-only alliance?
    The Delian League/Athenian Empire was mostly tribute collecting. Most League members contributed money, only a few members provided ships. But Peloponnesian League/Spartan Empire was purely military, too – outside the directly conquered Lacedaimon of Laconia and Messenia, the allies of Sparta contributed only troops.
    So why did Roman allies expand, and Spartan allies not?

      1. And Rome alone is a LOT bigger than Sparta. Even if you throw in the rest of the Peloponnesian league (and then you should really include the socii on the Roman side).

        Also the Spartans tended to be very cautious about actually sending their own people into battle. What the Spartans often do is send a commander and a few aides to command an army made up of people not from Sparta.

        1. “And Rome alone is a LOT bigger than Sparta. ”
          Rome of 5th century BC is estimated at just 800…900 square km. Far smaller than Sparta (which was 8000 square km including Messenia) or even Athens (2200 square km of Attica mainland).
          Yes – by 225 BC the Ager Romanus was much bigger – bigger than Laconia by then. But how did “Rome alone” manage to grow so much in the 170 years previous. Not just the land – how did the actual body of Roman citizens who were not socii or not unambiguously socii manage such huge expansion?

          1. Most of the land very close to Rome was forcibly integrated. For instance the latin league after its last revolt and defeat in 340 BC was punished with roman citizenship enforced (erasing their self governing status) upon them, except for the few that didn’t join and actually retained the allied status until the social war. A lot of umbrian and etruscan communities very close by must have been like that. Rome also acquired a lot of land farther away, turned into the colonies Bret explained in his post.

    1. The Greek alliance systems originated as defensive pacts; in particular the Peloponnesian League famously saw its allies refuse to carry out offensive actions several times when they felt they exceeded the League’s legitimate goals. It wasn’t until much later (and with Persia basically buying Sparta a navy) that Spartan oligarchs such as Lysander were able to exercise hegemony over most of Greece for a whole ten years.

    2. I think our host gives us a good overview, of why Sparta failed, in his series on Sparta.

      Sparta was much more inclined to shrink it’s (not that big to begin with) citizen body more and more, until the point where it’s whole political system was simply out of whack.

    3. Sparta’s alliance system is an actual alliance system: The members are still independent, there was some force but they weren’t outright conquered like Rome’s subjects were. Which means they pursue their own policies at points, and Sparta can’t force them to join in further conquests.

      Not to mention Sparta’s politics encourage defensiveness while Rome’s encourage aggression.

  8. “Once again, the most plausible solution is that the socii do what the Romans do: restrict recruitment to me of means who are expected to purchase their own equipment.”

    I think you meant to say “to men of means”?

    1. It might be the most plausible solution, but it is not the only plausible solution. There are two alternatives that Roman Republic is actually attested as using, though these were not emphasized in Middle Republic.
      One was public equipping of soldiers.
      Because the dilectus as described during Second Punic War emphatically assumes soldiers equip themselves. Soldiers are called to Rome unarmed, then sent to homes to get their arms and meet again at assembly camp. If arms were supplied by public, the obvious thing would have been to hand out arms as soon as the soldiers were selected. There is a mention of some number of replacement arms, paid for by pay deductions, but not enough to supply all.

      We know that in later, imperial Roman army some soldiers still could provide their arms but most bore state issued ones (either to be returned or paid for by pay deduction). The change from overwhelmingly private to mostly public arms must have happened at some point which may have been late Republic.

      And then there are horses. Where Rome actually shifted from public horses to all private horses.
      In 225 BC, Rome had 26 100 knights. Of whom 24 300 were knights of private horses… most of the richest Romans. Who had to pay for their own horses and ride them to war.
      But 1800 Romans, and actually the most influential of the 26 100, and probably mostly the richer, rode horses to war… but did NOT pay for them. Because they got public horses – paid for.
      This was actually a change towards self-equipping. Because before 406 BC, Rome had had just 1800 knights total – all of them on public horses.

      And the other alternative…
      Note the emphasis on client-patron relationship. The laws that a client and patron could not bear witness against each other, and that either party who violated the relationship would be outlawed.
      And now note that the Roman Middle Republic dilectus seems to take no formal notice of client and patron. Free poor are written in the census rolls in their own name and according to their personal wealth, under a tribe – not under a patron. A rich Roman is, at most, ordered to get a horse and learn to ride as a knight of private horse. If he is richer than that, he actually has a chance to become a knight of public horse and get rid of even that. What the rich Roman did not have to do was have it enrolled under his name “equip two or five or ten clients, and report for service at the head of clients”.
      Early Republic did precisely have armies formed by individual aristocrats. Attus Clausus, a Sabine from Regillum, had a large group of clients without being a Roman, and migrated to “Rome” with his clients – who included no fewer than 500 men able to bear arms. They actually got land some distance from Rome, across Anio. Just 25 years later, the Fabii founded a fortified camp at the border of Veii, with 306 men of surname Fabius (how many of them were actual biological legitimate patrilineal descendants of an apical forefather, given the custom of monogamy, and then how many generations ago, how many freedmen or descendants of freedmen?) and friends and clients to a total of 4000.
      And in the civil wars of Late Republic we again hear of rich aristocrats mobilizing and equipping their clients. Like Pompeius Strabo who created an army in Picenum in 90 BC without being a consul yet.

      Given that Rome had tens of allies at any time (in 225 BC, 443 000 men total, and Capua with 34 000 men total was probably the biggest single contingent), how much of the allied contingents were men who, rather than equipping themselves, were equipped by their city as public? How much were clients who were equipped by their patrons and reported to their city´s dilectus as contingents of their patron?

  9. I am struck by this week’s assessment that “Rome accrued its Italian empire … as a result of a series of localized, ad hoc decisions … without ever being intended to constitute a single, unified policy” and how that is reminiscent of the debate over the British Empire coming to be “in a fit of absence of mind”?
    Indeed, how does this compare with empire formation generally? Can one distinguish between “deliberate” and “absent-minded” empires, and note any defining characteristics of those respective types? Thoughts?

    1. British empire felt much more calculated to me. It became a direct empire in India after incremental increases in direct government involved in the East India Company, which was kind of Too Big To Fail for the British elite. The colonies in North America were a deliberate attempt to set up sponsored colonies to rival the Spanish colonies.

      1. Over the history of the British Empire there were plenty of people establishing facts on the ground and then demanding British military protection (often leading to imperial incorporation, whether that was their purpose or not) when native peoples tried to disestablish those facts; see Cecil Rhodes, or the history of the Proclamation Line between 1763 and US independence; I would tend to group direct rule replacing the East India Company as its power breaks down under the same umbrella.
        Both can be true, of course: it was a very large empire, with plenty of different places and times for different policies to be enacted. I’m just pointing out sometimes growth happened in the absence (and very rarely _despite_) policy – or more accurately, in the gap created where policy conflicted; if imperial policy had been to let adventurers reap their own whirlwinds I suspect things might have evolved differently.

        1. You still get this to some extent “I travelled to some godforsaken rock the government warns me not to go to and got thrown in jail and mistreated, heeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeelp…”*

          And then if you LET them rot, the newspapers start up with the poor tragic plight of the person. Plus there is the larger example that now that government thinks it can treat your citizens however it likes.

          *Some people more or less culpable for their own misfortune, YMMV.

        2. Also British Imperialism can, imo, be divided between a more ad-hoc early empire and more conscious empire building in the 19th century.

          1. My vague impression of the history would agree, pace Rhodes – but then he’s remembered because he stands out from the crowd, right?

          2. The early one was very focused – on grabbing high cash-flow territory: sugar islands, fishing grounds, key trade control points (Gibraltar, Calcutta, Malacca …), as these brought the revenue that enabled naval expansion. This in turn brought more cash and, as well, put a brake on French and Dutch finances. British statesmen were well aware of this virtuous circle and pursued it relentlessly.

    2. I think it’s a question of distance. Once you’re far enough from the capital, local commanders are on their own a lot more, and the tendency to get acquisitive can be a lot more pronounced. Also, there’s a decent chance of the overseas government going “be cautious and don’t piss anyone off” so by definition any increase in territory in those situations will be over the objections of the central government. Peccavi springs to mind.

      There certainly are some bits of the British Empire that were at least semi-accidental, and good chunks of India and the Middle East would count under that. It also depends on your view of ‘accidental.’ There’s certainly many occasions where the choice was ‘we could invade, or we could let them do this thing we do not want them to do’ and the option was often ‘invade.’ That is a KIND of accidental, but it’s not THAT accidental. At the same time, sometimes the thing they wished to prevent was “light all the stuff on fire, kill the existing government, and impose some kind of theocratic revolutionary state where heads roll eternally.”

      A nice complicated example is Egypt, where the British and French both seem to have made at least semi-genuine efforts to leave the place independent but aligned to their interests and the Khedive just would. Not. Stop. Fucking. Up. And was busy launching imperialist invasions of his own. There’s a lot to criticize colonial governments for, but they rarely overthrew super-competent super-nice enlightened leaders.

      1. Some possible explanations why:
        1) Super-nice enlightened leaders statistically don’t exist (Ashoka and…who?).
        2) Super-nice enlighteneed leaders either deal with their subjects’ objections to whatever nonsense this week’s passel of sunburnt jackasses are up to (becoming not super-nice in the eyes of our sources) or get overthrown by someone who will, who is not super-nice by definition.
        3) Super-nice enlightened leaders are good enough to slow or even stall colonial encroachments; the one dependable thing about the British Empire unless you’re the USA, the Maori, or post-1921 is _they’ll keep trying_, so that lasts at most a generation. And things not happening is quieter, so we know less about them.

  10. Would be interesting to contrast this to Zhou Dynasty, which relied on a feudal network of lords to augment the armies. Especially since early Spring and Autumn period battles were dominated by elites fighting in chariots with a small infantry levee….

  11. The fact that Rome’s relationship with its “allies” mirrors a Roman elite’s relationship with their “friends” so closely is honestly pretty striking.

    That being said… I wonder if the Romans would have viewed referring to their subjects as “allies” and clients as “friends” as euphemisms, or if that’s just the impression that we get because our terms for those relationships imply that everyone involved is more-or-less equal?

    1. I don’t think they’d view the term friend euphemistically more than we do. There are many wildly unequal friendships. I think they probably meant it sincerely, but also were not stupid and knew it wasn’t a relationship of equals. As the good doctor pointed out above, they sure as hell meant fides. The Romans are actually extremely good allies, if you think about it. There’s a LOT of examples of them coming to the defence of their allies, even where they don’t really gain much and indeed could lose quite a bit.

      It’s also still very possible for a smaller nation to easily defeat a larger one in this era. Some luck is required, but it’s doable – and the Romans are smart enough to realize that. Just because they’re bigger than one of their allies doesn’t mean that ally couldn’t beat them if they invaded it. Even in those cases where the Romans won, that doesn’t mean it’ll go the same way next time. As time goes on and Rome gets stronger, you’ll notice the Socii get more and more restless, and I would wonder if the Romans were getting more high-handed in their dealings.

  12. If they understood what they were trying to do? Meaning, like, if a time traveler with a good working knowledge of hang gliders (but no special expertise in material science) wanted to build the earliest one?

    Wouldn’t take much. The only technological thing you really need, I suspect, is woven fabric. Wood is strong and light enough for the frame. The Paleolithic is a stretch, but certainly by the Neolithic it was probably achievable.

    Passenger-carrying kites long predate hang gliders, but if you can make a kite carry someone, the same technologies could make a hang glider. The kite is just more useful. (There aren’t really many practical uses for a not-especially-good hang glider. The main thing you can do with a kite is observe things, but that only works because a kite can ascend. Now, once you get fancy composites and aerodynamic math, there’s a lot you can do with gliders! But your early hang glider isn’t going to be that.)

  13. Ancient Romans really are such a fascinating exception in terms of their empire building and warfare. An exceptionally effective alliance system that allowed a republican city-state to gain an unprecedently large empire and eventually turn Italy into a cultural extension of itself, and a large empire that did not form along a steppe boundary (I remember Walter Scheidel pointing out how unusual that is).

    I’m surprised that alliance system didn’t inspire more imitators, especially in post-Roman Italy.

  14. If you want to vary or expand the visual gag, Takamachi Nanoha from Lyrical Nanoha is also well known to befriend people via a staggering array of laser blast that look more like what a star destroyer would do. Somehow she found her lifelong partner that.

    She’s probably at a higher % of befriending from fighting than Goku, too.

  15. I’ve always found it strange how the Foedus Cassianum begins to fray so immediately after the Gauls under Brennus apparently sack Rome.
    At first glance one might expect that Rome’s power would be diminished by this event, and that the Romans and Latins might be forced to cooperate more closely against the new Gallic threat.
    Instead, the opposite seems to happen. Rome bounces back immediately and alienates its Latin (and Hernican) allies in the process.
    Is it to do with Rome’s conquest of Veientine (Etruscan) lands just prior to the Gallic incursion? Did the humiliation of the sack hurt Rome’s reliability in the eyes of its allies? Did the trauma of the sack cause Rome to behave more aggressively and imperiously towards foreign powers?

    I realize there probably is no definitive answer as accounts don’t agree on the sack of Rome exactly (some have the Roman superhero Furious Camelman™ swoop in and save the day while quipping some edgy oneliners) but, eh, interesting to think about

    1. Wikipedia suggests (in its page on the Latin War) that it’s basically that the Romans were getting big and were a far greater threat than any invasion. Indeed, given they’re expanding into notionally Latin territory to recover, that makes sense.

      1. I suppose I am talking about an earlier context. Not the Great Latin War of 340-338 BCE but earlier Latin and Hernican defections (which our esteemed blog author describes as the “edges fraying in the early 300s”) that allegedly took place around 50 years earlier, immediately after the sack of Rome.
        The immediateness of this defection makes it seem as if the Latins had already secretly been waiting for such an opportunity even before the 300s.

        But if the Latins defected because they thought Rome was a greater threat than any foreign invader, would not the sound defeat of Rome by the Gauls (i.e. demonstrating a foreign power to be a greater potential threat than Rome) have proven that this line of thinking is in error? Would it not have created a need to band together, rather than defect?
        Perhaps the Latins blamed Rome for its pigheaded diplomacy that had invited the Gallic attack in the first place?

        Also how does Rome bounce back so quickly after having its army destroyed, its capital taken, and many older senators apparently killed, especially if the Latins were already disloyal?
        It always strikes me that the historical record seems to give Rome a ridiculous amount of “plot armour” in this instance (while simultaneously and somewhat uncharacteristically admitting that the Romans suffered an embarassing and would-be crippling defeat).
        Anyway, I am mostly just picking at imperfections in the historical record and digressing a bit from the main topic of this post admittedly.

        I suppose I can only continue to spread the gospel of Furious Camelman™, our lord and saviour who decrees that “not by gold, but by iron is our nation recovered”, presumably whilst glaring past the camera at a 45 degree angle and, in the next shot, walking away from a shattering lit oil lamp without looking back at it.

        1. Picture a playground with a bunch of little assholes who’d like to be bullies/in charge/setting the rules of the games and one kid 18 months older who’s making them all play nice or he’ll beat them up. He’s won every fight they’ve seen him in so they think he’s unbeatable and just do as they’re told and everyone mostly gets along.

          And then a complete stranger walks up to him, decks him, goes through his pockets and walks away* – leaving his invincibility shattered, but no replacement hegemon.
          _Some_ idiot’s going to fancy his chances.

          *and then gets jumped from behind by one of the large kid’s friends or something, this is not the important part of the analogy.

          1. Haha, I much enjoy that analogy!

            To be clear my thought was this:
            -If the Latins had already been itching to defect, and did so as soon as Rome was sacked (as Livy at least seems to claim), it seems nothing short of a miracle that Rome was at all able to hold on to its dominant position after losing its army, capital, much of its ruling class, and then also its allies.
            -If the Latins had not already been itching to defect, the fraying of their alliance with the Romans following the Gallic invasion seems harder to make sense of.

            Leaving aside my jokes about how Marcus Furius Camillus supposedly more or less single-handedly saves the day (which Livy offers up as the “solution” that explains Roman survival), I guess a sensible thought would be that the historical record at this stage grossly oversimplifies the actions and motives of the various Latin polities as simply “(all) the Latins defected, period”

  16. Huh– I didn’t realize the extent to which Roman expansion was basically a real-life min-maxing EU4 strategy, complete with vassal swarms and the occasional coalition war.

    1. Sounds like Rome got super lucky with increased Diplomatic Relations Cap and Diplomatic Reputation in its national ideas.

  17. I still can’t help but thinking that this system only works because Italy is small. So that it’s still viable for an army to march be it from the socii to Rome or vice versa in reasonable time. The evidence is that when Rome finally goes out of Italy, the system is immediately abandoned to standard direct annexation or tributary system.

    Other big empires like Sassanid is also much bigger than Italy so that they also can’t help but use the tributary system, because money or food is much more easily delivered to the capital than flesh humans. Also I still don’t buy that Italy is that diverse like he said in other article; this article just outright says that most of the socii have similar government system like Rome anyway, and that’s the most important culture to consider in this case.

  18. The allies cannot take Rome into war but Rome will go to war if an ally is attacked? Sounds very much like NATO to me.

    1. Big difference there is that Rome could drag it’s allies into an offensive war. NATO is strictly defensive, excepting situations where there is unanimous agreement by member states.

      1. And did just about every year it wasn’t fighting a defensive war. By the time they came to most cities, you knew you were signing up to send off men to conquer whenever and whomever the Romans wanted.

  19. That thing about the Social Wars is what I’ve been wondering about for a long time, and I only found it more puzzling after I read your explanation on the Roman assemblies. Given the fact that only a small minority of Roman citizens really had the resources and free time to attend those voting assemblies, gaining Roman citizenship actually sounds like the erstwhile citizens of those allied communities would be giving up their chance of political advancement within their own respective communities for a less certain opportunity to make their way into the Roman political order. What could have compensated for this to the extent that the socii ended up wanting to acquire Roman citizenship en masse?

    I know, I know, you’re already planning about it. It’s just been bugging me to no end.

    1. Improved legal representation/protections, and same for trade and economics. Plus, with a vote, they have formal entry into the Roman patronage system, which meant having a powerful personal ally/sponsor.

      1. This is a thing in general, legal rights are often as mcuh or more important than political rights: After all you’re far more likely to end up in a lawsuit than to be elected to office.

    2. I think the best explanation is that of H Mouritsen’s *Italian unification* (BICS, 1998), which connects it to Gracchan land redistribution and legal rights. Appian’s narrative, Mouritsen shows convincingly, is very imperial-era biased and may be anachronistic. Dual citizenship was not a “thing” into Cicero’s time.

  20. ” If there are ‘grades’ of equipment sets – like the velites and three ranks of the Roman heavy infantry – within the socii (entirely plausible!) we do not near about them.”


Leave a Reply