Roman Legionnaires in ACKS: An Equipment and Pay Review

Roman Legionnaires in ACKS: An Equipment and Pay Review

While working on a forthcoming paper critiquing the economic assumptions of ACKS (fairly good, but subtly incorrect in interesting ways), I noticed a discrepancy in the pay assigned to ACKS soldiers and the price of wheat.

Per this useful source, and the excellent worldbuilding blog maintained by a professor of classical history here, during the Principate, a modius of wheat, purchased in the Roman provinces, might cost around 8 asses, or half of a denarius (prices in Italy could easily be double that). Meanwhile (time of Augustus), Roman legionaries were being paid 225 denarii per year, allowing them to purchase some 37.5 modii of wheat per month (assuming no skimming by their centurion, and no deductions for gear; we know the first was common and the second was standard). Our contemporary non-Roman sources indicate substantial corruption supplementing legionnaires’ pay, but this is impossible to quantify.

An ACKS bushel of wheat costs 0.5gp. Light Infantry A (3 javelins, shield, shortsword, and leather armor) earn 6gp per month, Heavy Infantry C (spear, shield, sword, and chain armor) earn 9gp per month, and Heavy Infantry A (spear, shield, sword, and banded armor) earn 12gp per month. That is, 12, 18, and 24 bushels of wheat, not counting maintenance cost of their gear (counted separately in the cost of a unit).

Four modii almost exactly make one bushel. The Roman legionnaire was earning no more than 9.4 bushels per month, less than ACKS Light Infantry! Yet Roman legionaries were definitely “heavy” infantry, with auxiliary infantry earning 1/3rd as much pay! In my forthcoming work on the economy of small farms, I cannot find a way to reconcile these very different assumptions. While the Roman solider does receive a 3,000 denarii lump-sum pension, this is after 25 years of service, making this only a fractional pay increase even before compound interest reduces the net present value of this sum. The Praetorian Guard, however, received pay roughly triple that of the common legionnaire, bringing them roughly into line with ACKS Heavy Infantry A. Perhaps the common Roman is paid in a patriotic glow that the men of the Auran Empire are sadly undesirous of?

Nature of Roman Armor

This misadventure also highlights that none of the ACKS troop types is equivalent to Principate legionnaires. The Principate armed its legionnaires with 2 javelins, a shortsword, a shield, and a heavy helm (the Imperial Gallic helm is on the borderline between heavy and light helms, I’ll round up). So far, this sounds like Light Infantry A! The real rub is the armor. While some might consider the iconic lorica segmentata (never called that by the Romans; their name for it has been lost) to be “banded plate”, I argue that it cannot be seriously considered to give AC 5.

For reference, a Dark Ages Viking wearing a chain-mail byrnie, covering the neck with an aventail, the entire chest, including the armpits, the whole of the arms in mail sleeves, and lengthy mail skirting around the hips and down to the knees, with a light helmet, has AC 4. That is, this is the heaviest armor an ACKS Jutlandic barbarian, based clearly on Vikings, can wear. AC 5 calls to mind the splinted mail of the 13th century, with head-to-toe protection and splinted plates over some areas, likely combined with a “crusader” great helm. AC 6 plate is late 14th century plate armor with mail gussets, and the Gothic armor of the late 15th and 16th century is AC 7 (tier 2 masterwork plate armor).

By contrast, the lorica segmentata covers the chest and shoulders. Oh, and a few strips of studded leather dangling over the gonads. Likewise, we know from archeological findings that the lorica hamata (mail) was much more common during the Principate, and did not cover the arms, neck, or hips, while the lorica squamata (scale) was less common than mail but still more common than the rare segmentata, and again had similar coverage.

Thus, I believe that Roman legionnaires should be modeled in ACKS as wearing armor that grants AC 3, plus a shield.

Roman Gear

The kit of the legionnaire is as follows: two javelins for 2/6 stone, one shortsword for 1/6, one shield for 1 stone, AC 3 armor for 3 stone, a dagger for 1/6 stone, a heavy helm for 1 stone, a week’s rations for 1 stone, a large sack for 1/6, a mess kit for 1/6, their share of their file’s tent for 3/6 stone, an entrenching tool for 1 stone, and a full waterskin for 1 stone, and a tinderbox for 1/6 stone, and a stake for constructing the camp for 1 stone, total weight 10 4/6 stone. This is enough to reduce their expedition speed to 6 miles, but each tent-group (8 men) was issued a mule, with a normal load of 15 stone. The mule carries some of the legionnaires’ load, enough that they can march 12 miles per day (5 1/6 stone), plus water for itself for one day (5 stone), and feed for itself for 6 days (15 stone). The mule is overburdened, reducing its speed to half, equal to the soldiers. The mule will likely also be burned with various bric-a-brac that isn’t regulation, much to the commander’s dismay. In combat, the heavy bundle the legionnaires carry is ditched, slimming them down to 5 4/6 stone, enough to add back on their waterskin and their money-pouch while still moving at 30’ per march action.

Historically, the Romans deployed at intervals of 5’ (ibid), yet they were definitely able to act as Formed Foot, but also as Loose Foot (the maniple swap). During the deployment phase, the general chooses which formation to deploy them in, and can switch formations as an action in lieu of attacking. If desired, treat this as a bonus proficiency, per AXIOMS 4. Personally, I believe this should be a standard ability of units that qualify to be either LF or FF (as the legionnaire does).

Stats For Roman Legionnaires In D@W

A double-century (company) as a whole has HD 1-1, 6uhp, and make 2 attacks at 11+ per round (regardless of weapon), and have a range of 2 when throwing their javelins. The cohort replicates these statistics but has double uhp and damage. Their combat speed is 2/3/4. Speed and range double when deployed at platoon scale. As heavy infantry (by doctrine and training), their morale is 0. Because they carry so few javelins (and no container to store them in), they can only discharge missiles once per battle without resupply (Judge’s discretion; note that later legions would carry lighter plumbata to resolve this issue, they would deal 1d4 damage). As mentioned above, they can swap between FF and LF, and have an expedition speed of 12 miles per day (note: claims by Roman military writers that their armies could march 20 miles per day were propaganda, see here for actual archeological evidence on their speed, 10-12 miles per day).

BR and Wages

The BR for a company of legionnaires, deployed as Loose Foot, is 1.749253455, or 1.5. Deployed as Formed Foot, it is 1.357449124, or 1.5. Commanders will usually deploy their legions in loose order (5’ spacing), as they did historically. Plumbata-armed Dominate legions, as Loose Foot, have BR 1.261656673, rounding to 1.5, and as Formed Foot 1.168200623, rounding to 1. Dominate legions are definitely Loose Foot. This data was calculated via D@W: Battles.

Via D@W, we can calculate that the monthly wage of the Roman legionnaire should be around 9gp. This is equal to 18 bushels of wheat per month. To recapitulate our earlier calculations, in the time of Augustus, they got (source) around 10 asses per day, with a provincial cost of wheat of around 8 asses per modius. By Nero’s time (ibid), the wage had jumped to 16 asses but the price of wheat to 16, reducing their net income (Domitian increased wages; gee, I wonder why). Given that 4 modii make one bushel, our ACKS legionnaires are getting double pay. Clearly, the Principate counting on making up for penalties to Unit Loyalty resulting from being underpaid (no official guidelines for this exist; talk to your local Judge) by simply never encountering calamities (that’s loser talk!). For the more important Praetorians, triple pay was given, giving a loyalty bonus over the standard (at least, in theory…).

Typical Officers

A Roman cohort is a battalion-scale unit of 480 men. It is composed of 6 80-man centuries. For D@W play, bundle three into two company-scale units, led by their senior centurions (the junior gets to watch how it is done). Centurions therefore will likely range from 3rd level (high enough to lieutenant a platoon) to 5th (high enough to command in a platoon-scale battle or lieutenant a company in company-scale battles).

In the real world, centurions received 16 times the wage of the common soldier (ibid). That comes to the equivalent of 72gp, enough to raise them to 2nd level purely from campaign income, and enough to pay their monthly henchman wage and then some. The son of a prosperous commoner who bought a commission hits 1st level in 12 months, and 2nd level in another 4 years. A tribunus militum angusticlavius, likely the son of an eques, gets 50 times standard legionnaire pay, or 225gp, enough to rise to 4th level. The Legate gets 70 times base pay, for 315gp, not quite enough to reach 5th level on its own. Of course, the tribune will inherit his father’s estate sooner or later, while the Legate certainly has his own senatorial estates, both adding to their monthly income. Still, it seems that Roman armies were severely under-officered by ACKS standards. Perhaps they made up the difference with legitimized corruption…

Tactics

The real-world Roman army tended to win battles by advancing at a solid clip to the enemy, throwing their javelins at them from close range, then charging in, counting on their armor (superior to most of their foes) and the shock of their discipline to break their enemies’ morale.

The sample ACKS analogue should operate similarly. When fighting enemy infantry with 90’ movement speed, they deploy as Loose Foot. When the enemy is more than 7 hexes away (for platoon scale double these distances), they delay until after the enemies’ initiative (if they can) and then hustle. The delay prevents the enemy from pulling a double-move on them. At 5-7 hexes away they advance on their initiative at a march and ready an attack (diegetically, they throw their pila as the enemy closes). They do not charge, as this would leave their AC lowered to the enemies’ counter-attack. At 4 hexes they advance at a march and throw their javelins. If the enemy attempts to retrograde after javelin discharge, the Romans will be with range of a charge, even if the enemy hustled. At 1-2 hexes distance they advance and engage in melee, if the enemy is not already wounded. They charge if they are out of javelins (have made at least one set of attack rolls) and are within range and not engaged, or if they are within range, not engaged, and the enemy has uhp remaining equal to half their maximum +1 (that is, if dealing any damage would trigger shock rolls).

If they encounter infantry with 60’ movement, they proceed as before, except they hustle unless that would put them within 4 hexes (that is, unless they are at 6 hexes or closer). At 5-6 hexes, they delay, march, and ready, and at 4 hexes they delay, march, and throw javelins. At closer distances they march and attack. They do not charge, even if they lack javelins. Instead, they make sure that they can withdraw if hit, withdrawing the maximum 2 hexes, then closing the distance on their initiative.

Encountering 120’ or faster units, the legions maneuver to threaten a fixed point, then storm it.

When encountering archers, the legions deploy as Formed Foot and defend with their shields instead of readying to receive an attack or hustling, and otherwise use much the same tactics. These are all general guidelines.

Conclusion

In summary, ACKS does not have an exact parallel to Principate Roman legionnaires. This paper develops such a mechanical parallel, and notes differences between ACKS’s economic assumptions and the Roman army’s pay scales. A forthcoming paper will provide further analysis of these economic assumptions, revealing both some major errors while also showing how ACKS manages to provide high-verisimilitude simulation despite these (in effect, the errors tend cancel out; this speaks to careful research that does not fall into the trap of reporting “surprising facts” uncritically).

2 Likes

First, on the comparison of wages, it is correct to conclude that Roman heavy infantry is underpaid. ACOUP says nearly as much, and we can confirm that by comparison to other forces:

Roman citizen-soldiers (by definition, both soldiers and tax-payers, assidui ) opt to pay themselves quite a lot less (2 obols = 3 asses a day) – being compensated more in honor and their political role in the Republic.

Contrast to the Athenian rowers from the same ACOUP post - they make one denarius per day, so roughly 120HS per month, enough to purchase 60 modii of wheat, i.e. 15 bushels. A typical rower in ACKS makes 6gp per month if continuously fielded, equating to 12 bushels. That’s a little less than the notedly generous Athenian wage rate, and on the whole seems very reasonable.

Likewise, Alexander’s soldiers made 1-2 drachma per day, i.e. 15-30 bushels, pay as either light or heavy infantry in ACKS. Cavalry was paid twice as much as infantry, which matches up neatly with ACKS’ wages for light cavalry (30gp per month => 60 bushels, right at the upper end of that range).

The wage you cite for the Prateorian guard then we might take as an actually-appropriate wage, were they not compensated with honor and station, as you suggest.

Roman mail armor should be AC 4 when combined with their other leathers. Axioms 20 discusses and justifies this in detail.

In Battles, I think you’re zooming out too far by looking at the maniple swap as a unit changing formation. A maniple is 120 men - that’s one company! The velites are light infantry A, being the least armored, while the ranks further in are heavy infantry A or C (and the triarii are veterans, neatly meeting the standard 3:1 ratio). A full cohort of 480 men is one division of a company-scale battle, four units (velites, hastati, principes, triarii). This lets the velites perform their standard role as a skirmisher screen that withdraws on contact, while the rest retains the iron discipline and firm morale they’re known for.

1 Like

Don’t have it, sadly. Might buy it, would you recommend? I have 1-8, 9, and 11. I’m considering buying 10, 12, 15, 19, and 20, plus Heros and Hoplites. I just don’t want to buy something that the II already has, or BtA does, and I know a lot of that material got bundled into the revised printing.

Good catch on Alexander’s soldiers, and the rowers. Bu the velites/hastati/princepes/triarii division seems to no longer have been a relevant distinction by the Princepate, which is the time period I’m looking at. Agree that Light A/Heavy C/Heavy C/Vet. Heavy C fits the Middle Republic very nicely (as Archon no doubt intended). As for my proposed legionnaires, they definitely formed up at 5’ intervals, per Tactius, so they are Loose Foot by ACKS’s definitions.

I realized that I had a section written in my head, but didn’t write it down. Behold, what was supposed to be the third-from last section:

Exp Total of Roman Ranks

As codified by Augustus, entry into the equites was gated behind a property requirement of 100,000 denarii, while the senatorial order required 250,000 denarii. What level does this rank entail, if converted to xp? A denarius buys roughly the same as 0.25gp, so an eques has 25,000 xp, about the same as an adventurer-tier advanced start, and comfortably 5th level. A senatorial rank Roman has 62,500 xp, just barely below the threshold for 7th level. Given that a legion has, at most, 7 equites (counting the primus pilus as one, since he’s due to be promoted shortly), and two men of senatorial rank (the legate and the tribunus militum laticlavius), they can only directly command around 8 companies, with lieutenants for 7, but have 45 companies! (see RR p. 437) The Romans are under-lieutenanted even by the standards of JJ p. 247, and very short on commanders.

1 Like

Axioms 20 is interesting research background and justification for why all the ACs are the way they are, why armor is modelled as avoidance rather than DR, and some formulae for determining AC of piecewise or custom armors. It’s not highly gameable, but it is quite interesting, and it is extremely unlikely to ever released in other contexts, if that’s useful.

Ah, that makes sense as the distinction then. The Middle Republic is what I’m most familiar with; Aura is in ~371, but never underwent a dramatic shift from Republic to Empire and so I imagine their military was reformed differently.

Checking the JJ, the expected NPC wealth by level table matches your level estimates nicely. I think it’s a very safe bet that an imperial legate is at least 8th level, and able to command at battalion scale; 6th-7th is the minimum, but for desirable and contested offices it doesn’t seem unreasonable to conclude that they’d be filled by higher level characters. With two characters of 8th+ level, a legion could be effectively deployed at battalion scale.

That said, the outcome that they’re short on commanders is a very typical one. Most realms I’ve built out run into that problem and have to disperse some of their forces into lower scales or bring on mercenary officers. It makes sense that Augustus was highly reliant on the Senate to provide him capable commanders of sufficient level. The Republic was more effective at creating 8th-10th level characters (it just lacked many of the 13th-14th level characters the Empire would create) because the (in ACKS mechanics) oligarchic system of rule could effectively cycle people into positions just long enough for them to soak up campaign XP and level. It’s an interesting set of tradeoffs.

2 Likes

Well, I bought AXIOMS 20. I don’t regret the purchase; there’s lot of neat stuff, but I think it is still incomplete. The formula for partial armor coverage shares the same problem as BRP and GURPS: it assumes that all blows are randomly distributed across skin area, maybe with a static fudge factor for certain areas. But, realistically, if I see someone wearing chest armor and no limb armor, I’ll aim all of my blows at the limbs. I don’t know how you’d mathematically model that, but its not a static value.

Example: in BRP, it is reasonably workable to wear greaves, gauntlets, helmet, pauldrons, cuisses, and absolutely nothing over the entire chest, abdomen, and hips. While this is evocative of the Red Sonja fantasy, realistically, you’re just going to get stabbed in the chest. There is the Choose Location special effect, which then obviates the entire point of piecemeal armor.

Ultimately, I think the best estimate of armor’s protection is “how long can someone in this panoply, taking the Defend action, survive against a Normal Man with a rondel dagger (1d6 piercing).” Unarmored lasts 2.5 rounds per hit. AC6+shield lasts 20.

Nice check on the JJ table, I agree that the laticlavius probably barely qualifies for senatorial ranks, while the Legate is much more comfortable.

You will be aiming for the unprotected parts, yes. Which the armored fighter will be protecting more actively than the protected by armor parts. Making it harder to land a blow where it will have an effect, → penalty to hit, AC is correct representation. In ranged combat, you need to aim at a smaller target or hope to randomly hit it with massed fire.

1 Like

Oh, penalty to hit is definitely the correct method. I just don’t think it is as simple as skin area=chance of hit. I.E., torso covers ~36%, but I think there are diminishing returns to piling more armor on the torso. That is, 3mm steel over the torso does not provide equivalent protection to 1mm across the whole body, it provides lesser. The AXIOMS 20 rules assume constant returns.