Collections: Total Generalship: Commanding Pre-Modern Armies, Part IIIb: Officers

This is the continuation of the third part of our three(ish) part (I, II, IIIa, IIIb, IIIc) series looking at the role of the general in commanding pre-gunpowder armies in battle. Last time we looked at how an army’s discipline could limit or expand the options available to its general: drill creating synchronized discipline could expand the ‘McDonalds Menu’ worth of things individual components of an army could do, allowing for the execution of more complex plans. At the same time, creating that synchronized discipline was so expensive that most armies didn’t do much of it, especially for infantry forces. That in turn left a commander with little ability to get those forces to execute complex plans or react on the fly, but for forces that were just expected to either hold a position or advance forward in a line, complex plans were less essential.

This week we’re going to look at a closely related factor: the presence of officers and their relative command independence. Officers are both the conduit through which the general relays decisions but also the means by which those decisions are carried out. At the same time, they can also be decision makers in their own right. In both cases a suitably developed command system was essential for actually employing the wider menu of options that synchronized discipline could provide.

And if you want to be an ACOUP officer, you can support me on Patreon (please note that supporting at the patres et matres conscripti level does not entitle you to run for the consulship nor command the legions of ACOUP). And 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.

An Officer By Any Other Name

But before we go any further, we need to clarify what we mean by this term ‘officer.’ Militaries, as I have stressed before, recreate their civilian social institutions; since different societies have different institutions, that produces differences in the division of class structure.

One of the key divides that shows up repeatedly is the division between commoners elevated to positions of minor authority and aristocrats in positions of command. Indeed, we have this in our modern armies: this is the distinction between non-commissioned officers (NCOs) which are generally soldiers promoted from the lower ranks and commissioned officers who enter the military typically at the rank of lieutenant (or its equivalent). That system is a hold-over from the origins of modern military structure in Europe’s early-modern gunpowder armies: the commissioned officers (whose commission, to be clear, came from the king) were drawn from the aristocracy and often the nobility, while their non-comissioned subordinates were drawn from the common soldiery who were in turn recruited from the peasantry. Because no peasant could become a knight or noble (under normal circumstances) so too no sergeant could (under normal circumstances) become a lieutenant; the social orders were thus kept separate, civilian mirroring military.

But exactly where that divide occurs and its significance varies wildly. In the modern military structure, a second lieutenant, the lowest commissioned officer, typically commands a platoon of c. 40 soldiers while a sergeant1 typically leads a squad or section (around 10 soldiers). By contrast in the Roman army, the centurion – a senior NCO like the sergeant promoted from the common soldiers – led a century of 60-80 soldiers; more senior centurions (the primi ordines) commanded an entire cohort (480 soldiers).2 So the line between NCO and commissioned officer can be drawn in very different places in different societies.3

Likewise an important distinction is often made around the emergence of a professional officer class (following on the emphasis made on this point by Samuel P. Huntington in The Soldier and the State (1957)) and the officers who came before which Huntington describes as ‘amateurs.’ This is, I’d argue, a mistaken understanding of the role of military aristocrats in their societies; they were no more amateurs at war than a person born into serfdom is am amateur farmer. Neither the serf nor his military aristocrat overlord is a professional, but neither of them is an amateur either; the categories are not exclusive. Nevertheless there is a difference between truly ‘amateur’ officers (which certainly existed), military aristocrats for whom war is a born-calling, and professional officers.

For this discussion, however, much of this complexity is beside the point. That isn’t to say it is unimportant generally (it is very important generally) but that for this topic here, we can define ‘officer’ very broadly and so avoid having to untangle the particular social context of each military in question. Consequently we’re going to just define officer here as anyone who has command over other combatants, effectively lumping NCOs and commissioned officers (and their ancient equivalents) together. That isn’t because those distinctions are unimportant (they are important!) but because they aren’t quite relevant here, since we are really focused on how command structures enable armies to respond with ability and execute complex plans. It is command, not social status per se that we’re focused on.

How Much Officer Do You Need?

Pinning down rules of thumbs on how many officers an army needs (again, including both NCOs and commissioned officers), by contrast, is relevant but deceptive. Students (and occasionally scholars) can easily enough notice that the ratio of officers to soldiers (‘rankers’ we might say) has gone up steadily overtime since the early modern period, at the same time that armies to became capable of progressively more complex operations. The inference is easy to draw: a higher ratio of officers (that is, more officers) corresponds to a greater capability to execute complex plans.

And in the very broad strokes this seems to be true. The Spartans, our sources note (Thuc. 5.68; Xen. Lac. 11.4-5) had more organizational divisions commanded by more officers than other contemporary hoplite formations and seem also to have been able to maneuver a bit better than other hoplite formations, albeit still nowhere near what would be possible for later Roman or Macedonian armies.4 It’s difficult to precisely parse out the organization (in part because Thucydides and Xenophon differ in some details), but Xenophon gives the mora as the large scale unit (estimated around 576 men; six of which made up the army), each of which had 29 officers (see chart below), so one officer for each 19 or so regular Spartiates. As Xenophon notes, this greater degree of organization with many smaller units nested in larger units enabled the Spartans to perform maneuvers thought by the Greeks to be quite difficult to do. This is succeeding over a very low bar, since a hoplite phalanx was one of those formations which struggled to do much more than march forward in a line, but it still speaks to the importance of officers.

Fast Forward to the organization of the Macedonian army under Philip II and Alexander and by all indications the system has become more sophisticated, with an even greater number of officers. We don’t have a description of the detailed structure of Alexander’s army, but what we do have is a later description by a philosopher, Asclepiodotus, of an ‘ideal’ sarisa-phalanx for the later Hellenistic period. The basic unit, the syntagma, consisting of 256 men as he describes it has eighty various officers (many of which we would classify as NCOs, but again we’re not making those distinctions here). That’s now one officer for every 2.2 regular soldiers. Those syntagma were then grouped into larger formations; under Alexander these were the taxeis, of which there were six. In later Hellenistic armies, particular parts of the phalanx are specified by their shields or commanders, but we often have limited visibility into the subdivisions of the phalanx.

Again, it is worth warning that we cannot be sure the degree to which Asclepiodotus may have embellished this system, but it does seem to be accurate in the broad outlines at least.

And we know from Alexander’s campaigns and the subsequent wars of his successors that this was a much more flexible, reactive formation than any hoplite phalanx. Alexander is able to do things like refuse his left flank (to give himself more time to win on the right) and later Hellenistic phalanxes are able to do things like form square under combat conditions (App. Syr. 35), while often operating with attached light infantry and cavalry support. This was a clockwork mechanism substantially more complicated than anything the poleis of Greece had put together on land and it demanded a lot more officer accordingly.

Via Wikipedia, a fourth century fresco of Macedonian soldiers from the Agios Athenasios tomb. It is possible but by no means certain that the two men to the left who wear purple cuirasses are meant to represent officers (hegemones); there is some evidence to suggest that only these more senior Macedonian officers (who were, as you see in the chart above, in the first rank) were fully armored as a matter of regulation, though this issue is an unsettled one and personally I have my doubts that regular Macedonian soldiers would be so poorly armored.

So it seems like we have a pretty clear pattern from most hoplites to Spartiates to Macedonians that ‘more officer’ means ‘more reactive and flexible formation.’ And indeed, some scholars have noted such.5 And then we get to the Romans.

There is little question that the Roman legion of the Middle Republic was a more flexible, reactive adaptable fighting formation than the Hellenistic phalanxes it faced. Our sources tell us as much (Plb. 18.31-32; Liv. 44.41.6) and that same conclusion emerges pretty clearly in both the battle narratives we have an our understanding of how the legion was expected to function in battle (discussed last time). Even the basic mechanics of the legion’s fighting method – the three ranks engaging in sequence, forming and closing gaps as they do – appears well beyond the capabilities of the Hellenistic phalanx (and indeed this leads the latter to struggle and eventually fail to cope with the former). And while as a rule the Hellenistic phalanx maneuvers by taxeis, in the legion notionally the maniples could be independent maneuvering units as the fight shifted from hastati to principes to triarii; a much smaller unit of maneuver.

So we ought, by the theory, to have yet more officers still, right? And yet we don’t! The legion (or 4,200 or so excluding cavalry) is divided into thirty maniples, each split into two centuries. Each century was commanded by a centurion so that gives us sixty centurions; centurions were organized in seniority order, so they are not all peers but organized in a hierarchy. Above this was the commander himself (holding imperium; a consul, praetor, proconsul or propraetor) along with a number of military tribunes (frequently junior aristocrats, typically six per legion these fellows could be delegated to command a legion in battle), the praefecti sociorum, Romans put in command of the allied units and the assigned quaestor who handled pay and finance but might also be asked to command in a pinch. Below the centurions, we have attested in the imperial period the decani, each of whom commanded a contubernium or ‘tent group’ of six men. The contubernium but not the decanus is attested for the period of the Republic, but I tend to think that there must have been decani in the Republic too; as the commanders of six men they’d probably double as file-leaders since the standard Roman file was of six. The legionary cavalry (just 300 per legion; the Romans liked to rely more on allied cavalry) were divided into ten turmae (of 30) each commanded by three decuriones (so that’s 30 decuriones total).

That means for a standard double-legionary army (9,000 Romans, we’ll put the socii and their praefecti to the side) had 1 commander, 1 quaestor, 12 military tribunes, 60 decuriones, 120 centurions and perhaps 1,200 decani; 1,394 officers for 7,606 regular soldiers, a ratio of around 1 officer to every five and a half soldiers. That is decidedly less than the Hellenistic phalanx and yet the Roman legion is, as noted, more flexible and response than the Hellenistic phalanx. There must be some key element we are missing here.

And indeed, there is.

Independent Command

To understand the difference, we need to understand what officers are expected to do. In practice, we can divide their roles into three categories. First, they have organizational responsibility; the officer is supposed to make sure the men under their command show up, get fed, have all of their equipment and so forth. Second, officers execute decisions made by their superior officers; if the army commander sounds the charge, all of the officers along the line execute that charge by leading their units forward. Finally – and most important for our question – sometimes even relatively junior officers are expected to make their own tactical decisions.

In modern armies, this last concept goes by the term ‘mission tactics’ or the German term Auftragstaktik although there is a lively debate about the exact origins of the term and the idea, with the traditional view (challenged but not overthrown) being that the modern tradition of Auftragstaktik comes from the Prussian or German Imperial military tradition. Obviously that is all much after the period we’re interested in, but as I think we’ll see some degree of what could be termed ‘mission tactics’ are a frequent occurrence in sophisticated armies. This was not a brand new idea invented in the early 20th century in Germany, but a common idea that surfaces and resurfaces when conditions for it are right (much like synchronized discipline).

At its essence, the idea behind mission tactics is that rather than the senior officer (like a commanding general) giving his subordinates a rigid plan to merely execute (that second function of officers), they instead are given a clear goal to accomplish within certain parameters (forces, time frame, etc.) and left to work out the details themselves. The idea is then those junior officers, rather than being straight-jacketed into a single big plan, can instead craft plans that utilize their local knowledge of terrain and conditions and also react quickly to change plans as they become aware of changing conditions. In essence then the junior officer has a degree of independent command, within the confines of the goals given to them by their superior.

Doing that in practice however doesn’t merely require having the officers, it requires a certain culture for those officers where they feel empowered to improvise within the broader goals and plans set by the general. Naturally that also requires generals who are willing to allow their officers to improvise, which is something that does not necessarily come naturally to military culture which is in every other respect predicated on strict adherence to orders and deference to rank. And that adherence isn’t insane: if as the general you order your cavalry to screen the flanks of your army and instead they ride off on a raid the whole army is made vulnerable. They very need for coordination we’ve been discussing makes adherence to orders by all of the component parts essential; this demands that adherence be flexible and that’s a hard balance to strike.

Technology also plays a role here. Mission tactics are, in some respects, a necessary adaptation to modern battlefields: the overwhelming amount of modern firepower forces dispersion, meaning soldiers are more spread out (to take advantage of cover and concealment). As an army becomes more dispersed, subordinate officers and their units are naturally going to be out of sight and contact with their larger elements, which demands that those junior officers be prepared to make decisions on their own.

By contrast, prior to the late-1800s, armies tended to be much more concentrated with soldiers often lined up shoulder-to-shoulder to take advantage of either massed firepower or the mutual protection we discussed last time. For an army fighting in formation, the general may not want every officer to have the ability to act independently – after all, for the army to function everyone needs to be moving together in that formation. You wouldn’t want one file leader running off on their own and there’s basically no condition where you’d need that file leader to exercise their own judgement anyway – they just need to keep their file in its right position.

So these armies tend to have some officers who operate entirely within the first two officer functions (organizational/administrative and executing the decisions of others). These officers are important of course: getting everyone fed matters and they can allow for more complex formations. Xenophon, for instance, comments on how the Spartan enomotarchs allow the Spartans to form up in variable depths and the same is absolutely true of the sub-divisions of the Hellenistic phalanx – it’s easy to see how you could either double or halve the width of a syntagma, for instance, by having the hemilochites form new files directly to the side of their lochagos in the first case or by stacking the lochagos’ file behind his dilochites in the latter case.

But some armies also delegate degrees of command independence downward; it is typically ‘downward’ in the sense that allowance for independent decision-making begins at the top and decreases as one goes down the ranks. Thus the crucial question becomes: how far down the chain of command is independent action encouraged and how prepared are those officers to exercise that independence? Precisely because of the limitations in both gathering information and communicating orders we’ve already discussed, delegating that authority – giving those lower officers the authority to deviate from the plan when appropriate – is crucial to producing an army that can react to unexpected developments.

The answer can vary pretty widely. In Alexander’s army, it appears that the taxiarchs – the commanders of the six taxeis (we might say ‘regiments’) of the Macedonian phalanx – could maneuver independently. Simmias, for instance, ordered his taxis to hold position because he could see (as Alexander could not) that the Macedonian left was in a lot of trouble and if he broke contact with it, it risked being surrounded (Arr. Anab. 3.14.4). That was a deviation in the plan for sure and it left Alexander’s successful right wing’s left flank hanging in the air, but in the event it was probably the right choice – had Alexander won on the right but Parmenion been surrounded and destroyed on the left it would have made for a bitter victory indeed. It’s also clear from Arrian’s narrative both here and at Issus that the taxeis each could maneuver independently, though given the vulnerability of the phalanx on its flanks and rear great effort was made to maintain a mutually supporting line; later commanders (Pyrrhus of Epirus and Antigonus III Doson) seem to have experimented with ‘articulated’ phalanxes where the taxeis were joined by lighter or more flexible troops (often covered by elephants) to allow them to maneuver a bit more freely. However at the same time, the sense one gets for later Seleucid and Ptolemaic armies – the largest and most powerful of Alexander’s successors – is that the phalanx became less flexible and more rigid.

That said, a taxis was generally a fairly large unit; its exact size would have varied as the strength of Alexander’s central phalanx did, but under Alexander there were always six (except at the Hydaspes where only five were engaged); for the roughly 9,000 phalangites at Granicus (334) or Issus (333) that would mean we’re looking at a unit of c. 1,500 (one is tempted to suppose the ‘paper strength’ might have been exactly 1,536, which would make a pleasing symmetry of six taxeis each composed of six syntagma of 256 men each).6

Via Wikipedia, a map of the Battle of Gaugamela. It is worth noting that while this map is certainly serviceable, there is a better and more detailed pair of maps of the battle in the Landmark Arrian (2010), edited by James Romm.

Independent action in the Roman army went much lower, down to the individual centurions where necessary. This makes sense: after all, in the standard Roman triplex acies each maniple needs to maneuver on its own and both centuries of the maniple need to be able to move relative to each other to create and close gaps. Caesar’s accounts of his campaigns are full of centurions acting on their own initiative (typically attacking in brave and often reckless fashion), like Pullo and Vorenus7 (Caes. BG 5.44) or Gaius Cratinus (Caes. BCiv. 3.91). The same can be seen of the more-or-less equivalent commanders among the socii, the praefecti cohortis; one such praefectus of the Pelignians rallied his unit to the attack by taking their battle standard and throwing it into the middle of the enemy battle line, essentially forcing his men to charge to salvage their honor (Plut. Aem. 20.1-2). Livy records a similar trick in a different battle where a different Pelignian praefecti does the same thing to rally his men to attack and it goads the Roman centurions nearby to follow suit, precipitating an engagement by the whole army (Liv. 25.14). Beware of Pelignians throwing standards, is what I’m saying here. If a Pelignian throws a flag at you, run.

Via Wikipedia, a funerary monument for Marcus Caelius, centurion primus pilus of legio XVIII (written here as XIIX). The small decorated disks on his breastplate are phalerae, essentially medals. In his hand he carries one of the badges of a centurion’s position, the vitis or vine staff, used both to dress the ranks and discipline soldiers.

But note that means that independent action extended all the way do, potentially, to units of 120 (a maniple) or even 60 (a century) in the legion. I want to be clear that we only see this sort of independent action generally in extremis, but of course that is when you want to see it; if everything was going to plan one follows the plan. Nevertheless the difference at this scale is obvious: the smallest unit with independence in a Macedonian army was a taxeis of perhaps 1,500, while the Romans allowed potentially similar independence to units as small as a maniple of just 120.

And this relates back directly to the flexibility and success of the Roman legion. Consider again Cynocepehelae (197); Flamininus, the general upon realizing his left wing was falling back but his right was advancing rode to his right to try to drive his legion through and win the battle and seems to have carried through the pursuit, but an unnamed military tribune had enough command authority to gather some of the unengaged maniples of the victorious Roman right to flank and destroy the victorious Macedonian right (Plb. 18.25-26). One can see a lot of the elements coming together here: the engaged maniples need to be able to wheel about and attack in a new direction – that demands discipline but it also demands centurions who are able on seeing what the tribune is trying to do to be able ‘drive’ their units that way. And of course it requires everyone – the soldiers, the military tribune, and the centurions to believe that such a bold deviation from the battle plan using a quite large portion of the Roman force on the right was an acceptable thing for a less senior officer to decide to do at the spur of the moment. They had to think, ‘the general will reward me for my initiative’ and not ‘the general will punish me for my disobedience’ which comes down to command culture.

Likewise, we’ve mentioned Bibracte (58BC) and the feat of having the rear line of the triplex acies about-face and attack in the opposite direction of the front two lines. What I find notable here is that Caesar does not say that he did that, despite the fact that Caesar is very quick to take credit for his tactical brilliance by name – Caesar is forever telling his reader all of the great things Caesar did (in the third person). Instead Caesar says “the Romans wheeled about,8 advancing in two groups, the first and second lines such that they resisted [those enemies] driven from the summit; the third such that they blocked those [enemies] arriving” (Caes. BG 1.25; emphasis mine of course). Caesar isn’t ordering his – he’s not even on horseback at this point; one is left to assume this was a quick reaction probably by the senior-most centurions in the rear-most cohorts.

It’s this sort of expected level of independent action in extremis which in turn also means that Aemilius Paullus can trust his centurions to make good decisions and simply give the order for the component elements of his legion at Pydna (168) to maneuver separately and thus turn the one large battle into many little ones, breaking up the phalanx (Plut. Aem. 20.8-10). In essence he was releasing his centurions to advance or retreat on their own initiative, in their own time, which created a battlefield too dynamic for the Macedonian phalanx to cope with.

Even though the Roman army likely had fewer officers, it had more of them who could react quickly and independently, though again we must be quick with caveats. Both army’s approaches to their officers and maneuver units fit fairly well with their style of fighting. Pikes, like the sarisa (and later European pikes, for that matter) generally need to move in large groups to be effective. By contrast, as Polybius himself notes, the Roman sword-shield-and-javelin fighting style (with heavier armor too) could work well in smaller more flexible groups.

Different Needs for Different Missions

Now it is easy to look at this and conclude that having more officers and officers with a greater degree of freedom to act independently is just better and on the balance it probably is better, but there are complications well worth addressing.

First, of course, these officers have to come from somewhere and this reintroduces the distinction between officers drawn from the aristocracy (leading to your modern commissioned officers) and those drawn from veteran soldiers (leading to modern non-commissioned officers). Even for modern armies, attracting the requisite talent to staff both a competent officer corps and NCO corps is difficult; the problems mount for pre-modern societies where educated, literate men are a scarce commodity in general. And especially if you are going to expect these junior officers (commissioned or otherwise) to exercise independent command authority they need to be either trained, experienced or both, which demands keeping them in the army for extended periods – these cannot just be farmers you call up for a few months out of the year (like, for instance, your average hoplite). A society’s aristocracy may provide some of these men, but probably not enough to thoroughly officer large armies. Consequently the resource demands of producing these kinds of fellows in quantity are significant.

And that leads us right back to the same problem we had with discipline: in many cases the expense of producing a flexible, adaptable army is going to either exceed the resources of the state or fail to be cost efficient when compared to a less flexible but much cheaper ‘amateur’ army that is still cohesive and can get the job done.

Finally, of course, civilian social structures are going to exert pressure on this system too. The Macedonian officer corps was built out of a preexisting social institution, the ‘companions’ (hetairoi) of the king. The Roman command system also follows Roman social hierarchy. But not all societies lend themselves to that kind of organization; a decentralized state that has many local nobles and notables is going to struggle because those men will expect by dint of their position to lead the troops from their territory in battle. And while a ruler might try to enforce centralization (a perilous venture, for the aristocrats will oppose it!) it may not even be a good idea because those decentralized institutions often produce the essential cohesion that holds the army together. Democratic states may well fear that a large, professional officer corps is a threat to the democracy (a fear not entirely unfounded), while autocrats often fear well-trained junior officers with command independence because they are the logical breeding ground for military coups.

And while there is a clear advantage to the sort of army that practices synchronized discipline and has lots of officers that can act independently to allow it to execute complex plans, these armies do not always win. Sometimes they lose to highly motivated amateurs fighting on their home turf (think either the failure of the Persian invasions of Greece or Roman defeats on the frontiers in the fourth and fifth centuries) because morale and cohesion (next week!) matter a lot too. Sometimes they lose because they find themselves facing alternative non-state military systems which are also flexible and reactive but through organic social institutions. Sometimes they lose due to poor generalship (these complex armies require a high degree of competency to run; a Roman army does not require a genius at the helm but it does require the general to get a lot of finicky but basic block-and-tackle sort of work done). And sometimes they lose because of plain bad luck. This is a particular problem because in most cases (Rome seems the clear exception) highly trained, disciplined and professionalized armies of this sort tend to be fragile – expensive and difficult to replace quickly.

But the structure of most strategy games is such that we rarely see this. Factions games like Mount & Blade or Total War or Age of Empires may have different units, but it is extremely rare they they offer different levels of command and control itself. Non-state Gallic and German armies are every bit as centrally directed in Total War: Rome II and Paradox’s Imperator as Roman or Macedonian armies. The emergence of infantry-based gunpowder armies in Medieval II: Total War and Empire: Total War doesn’t bring about any change in the way those armies can be controlled or directed. If this is represented at all, it is usually the way it is shown in Europa Universalis IV: a ‘discipline’ stat that serves as a flat bonus to damage or defense. Ironically, one of the rare exceptions to this rule was Crusader Kings II, where standing ‘retinue’ forces, precisely because they were standing forces, could be used more flexibly at least at the operational level; this feature was then sadly removed for Crusader Kings III, where professional retinues now behave exactly like levies except that they do more damage and have higher defense.

The great missed opportunity I see here is player choice in force composition, forcing the player to make a choice (or perhaps in choice of action, accept a reality) in terms of how controllable their army would be. The absence here is particularly notable in games with fantasy settings that feature ‘endless horde’ style armies like Warhammer‘s Skaven or Orcs; forcing a player to account for an army that is much less controllable might make these factions feel different in a way that would also make historical sense.9 But the same sort of decision ought to be a factor in many historical games as well.

All of this also combines to give players of these games a sometimes unrealistic vision of exactly how agile an army can be and the impression that their historical counter-parts must have been ‘unbeatable’ (particularly in the case of horse archers, which can be controlled in these games with such unreal precision that they can only ever be out of position by player mistake; real horse archers were powerful, make no mistake, but not undefeatable).

Next week we’re going to turn to an even less realistic element of these portrayals of pre-modern warfare: morale and cohesion.

  1. There are higher grades of sergeant in, for instance, the US Army (First Sergeant, Staff Sergeant and so on), but these ranks of sergeant typically advise a more senior officer rather than directly command a larger unit
  2. And the most senior of those centurions, the primus pilus had authority over the entire legion, though typically an aristocrat (either an office holder, a military tribune or a legate) was usually in command of a legion or several legions with the senior centurions advising him, much the same way senior sergeants work today. That system is formalized under Augustus with the creation of the legatus legionis, a specific legionary commander drawn from the senatorial aristocracy.
  3. And while we’re here different social structures also impact the degree to which that line is permeable. In the United States military, for instance, there is a whole institution – Officer Candidate School – providing a route for non-commissioned officers to earn commissions as officers. That sort of jump was substantially rarer in early modern Europe and functionally non-existent in the Roman army – the lack of rank mobility mirroring a lack of social mobility in their civilian society.
  4. Its important to note, however, that we really only have a complete view of the command structure for Spartan armies and not their contemporaries. it’s clear that Thucydides and Xenophon want to cast this structure as deeply unusual, but the vision of a completely junior-officer-free hoplite phalanx may be to a degree mirage of our sources, on this note E. Wheeler, “The General as Hoplite” in Hoplites: The Classical Greek Battle Experience (1993), 134-5. Though this point too cannot be pushed too far: Xenophon and Thucydides write for an audience which would have known the organization of the Athenian hoplite phalanx extremely well and so would presumably have agreed with the judgement that the Spartan organization was exceptional.
  5. E.g. F. Naiden, “The Invention of the Officer Corps” Journal of Historical Studies 7.1 (2007): 35-60
  6. And if you are thinking, “Wait, Alexander has more than 9,000 infantry!” Yes, he does. The main phalanx is the largest single component of his infantry, but there were also the hypaspists and the infantry agema, two elite units of Macedonian infantry, along with large numbers of Greek mercenaries who likely fought as hoplites, plus yet larger numbers of lighter troops.
  7. This is where HBO’s Rome got the names, but both men were centurions historically.
  8. conversa signa, literally ‘turning the standards,’ which is of course how you would wheel the unit, the soldiers being drilled to maintain their position relative to the standards.
  9. There is a mechanic like this, the ‘rampage’ mechanic that some units, mostly wild animals or lizardmen units, have, though the fact that the consensus of elite players seems to generally be that rampage is such a huge liability on these units that they’re not worth having unless there are no other options is striking. I think the problem here is that most of the major ‘rampage’ units are high cost units that are both very powerful but also very vulnerable in a bad match-up. By contrast, having a similar mechanic on, say, Bretonnian, Skaven or Orc common infantry where once in combat they are essentially uncontrollable would still work for those units since they’re mostly ‘tar-pit’ units anyway.

127 thoughts on “Collections: Total Generalship: Commanding Pre-Modern Armies, Part IIIb: Officers

  1. The part about roman independent action reminds me of a story of the early roman republic. But I cant remember the names, maybe someone in the comments knows it. The general orders his cavalry officer (who is also family) to scout but not engage. The officer scouts, but seeing an opportunity, leads a successful cavalry raid. The general gets mad, and orders his execution, citing Titus Manlius his example. the army likes the officer so they try to defend him, and the incident ends up with a shout match in rome where the general demands the execution while the senate tries to defuse the situation.

  2. I love how “throwing of the banner” is so universally understood that it’s used by the Romans and nowadays is a standard visualization for attack moves in strategy games. Good design.

    1. I am trying to think of a game that indicates attacking by throwing the banner into the unit being attacked–do you have any examples? It strikes me that throwing your banner is always going to be an exceptional high-risk, high-reward thing: you lose some of your the ability to command the troops, and if you can’t get it back you are in disgrace.

      1. They’re referring to game UI, not actual animated banner-throwing. Like how orders in Mount & Blade Bannerlord are visualized as giant banners.

        1. I understood that they meant the UI — I have just never seen that before. Like the other reply here, I have seen banners in many games for setting way-points or marching orders and I have always understood it like an indication that the unit will take its banner to that location. It strikes me as quite exceptional to throw your banner to the enemy so that your soldiers have to fight desperately to get it back. I was just sincerely curious if there was a game whose UI represented an attack with that image.

      2. A banner is often used to indicate a unit’s destination in strategy games, so the player can see at a glance where this or that unit is headed in addition to where they are right now. It probably evolved separately from the actual practice of throwing the banner, though, because of that utility of visually depicting a unit’s current destination, and a banner is just a more immersive way of depicting that than a giant glowing arrow. Games that let you set multiple waypoints often represent each waypoint with a separate banner, and if you change your destination partway through, the banner magically throws itself from one enemy unit to another even if the unit it corresponds to is three hundred yards away.

  3. It is implied here that the Achaemenid Persians had an “army that practices synchronized discipline and has lots of officers that can act independently”. That is a bit surprising, Roel Konijnendijk often writes that we know very little of how the Achaemenid Army was organised. At least this would be a counterpoint to the trope that the Persians were an “Asiatic horde” whipped together by the King

    1. If I’m not mistaken, many Persian troops were levied from the Satrapies, and therefore carried their own traditions and organization, in some cases dating back to the beginning of written human history. Assyria had been Empire-building when the Greeks were still steppe tribes, after all. So I think a large variety is to be expected, from troops that were extremely well organized to ones that had barely any organization (by Greek standards).
      The Greeks are by right the Barbarians in this relation.

      1. If I’m not mistaken, many Persian troops were levied from the Satrapies, and therefore carried their own traditions and organization, in some cases dating back to the beginning of written human history.

        In many cases, those traditions and organisation consisted of levies raisef for a specific campaign and then disbanded afterwards, in a manner not dissimilar to that used by the Greeks. Even where there had been professional or semi-professional armies, like Assyria, I don’t think there’s much evidence of the Persians maintaining these armies after their conquest. There’s not much evidence of the Persians *not* maintaining these armies, of course, but still we can’t just assume that the satrapal forces maintained their traditions from before the Persian conquest the best part of a century ago. (And in the case of Assyria specifically, they’d been under Babylonian rule for almost a century before being made a Persian province, giving further time for their military tradition to degrade.)

        1. Come to think of it, Herodotus says somewhere that the Persians deliberately demilitarised the Lydians to stop them rebelling. Granted he might not be accurate in this.

        2. I doubt any ancient culture could have kept large bodies of men under arms on a regular basis. Something about work needing to be done or everyone starves.

    2. See Ancient Warfare III.6 (Achaemenids used battle standards to communicate to their troops, no evidence for this anywhere in the Greek world before the death of Hephaistion) https://www.karwansaraypublishers.com/products/issue-iii-6-pdf and parts of chapters 4 and 6 of this PhD thesis (Achaemenids may have had a decimal organization and definitely a sophisticated bureaucracy to pass commands from senior officials to soldiers on the ground) https://www.steiner-verlag.de/Armed-Force-in-the-Teispid-Achaemenid-Empire/9783515127752

        1. Thank you, Shaun. Now that I see the table of contents your book seems really interesting! I had forgotten about the battle-standards

      1. Having battle standards isn’t the same as using complex tactics. Anglo-Saxon and Viking armies would carry standards into battle, but their main tactic was generally to just form a shield wall and advance against the enemy — much like an ancient Greek phalanx, in fact.

        1. And imagine how hard it was to line up and advance *without* battle standards to communicate commands, spot units and commanders when everything had become confused, and spot particular units in a camp? And those are the three functions which Greek authors tell us Achaemenid battle standards had (there were probably more since Greeks did not use or understand them).

          1. I don’t see how that’s supposed to rebut my point. Sure, an ancient army probably needs standards before it can carry out complex tactical manoeuvres, but it doesn’t follow that an army with standards was capable of such manoeuvres.

            Also, the claim that the Greeks didn’t even understand the use of standards is one that needs backing up.

  4. “Neither the serf nor his military aristocrat overlord is a professional, but neither of them is an amateur either; the categories are not exclusive. ”

    Not exhaustive.

    For not exclusive, you would have to say rather that they were both.

  5. I agree that taking even some of this stuff into account might be interesting in video games. I personally find the effective micromanagement in games like Total War overly gamey and rather boring. A fantasy setting, such as the ones in the recent Total Wars, seems ripe for all sorts of differences in unit controllability to spice up the formula.

  6. I think there’s a point to be made here about the concept of ‘amateur’ as used in the time of Huntingdon, when it was much less likely to mean unskilled or untrained– or ‘amateurish’.
    The Olympics are probably the example that tracks the best today, but the distinction between amateur and professional cricketers, who played alongside each other, was maintained into the 1960s (the amateurs also known as ‘gentlemen’ if you want a sense of the class element).

    1. ‘Amateur’ originally comes from the Latin root meaning ‘love’, as in someone who does something for love of it rather than for money (then they would be a professional).

      In astronomy it remains a useful distinction without a pejorative connotation, with the two groups mostly (but not entirely) inhabiting different fields of expertise: amateur astronomers tend to know more about what can be seen in the night sky by eye, or directly through a telescope, and have historically contributed tons of valuable data and discoveries (e.g., as recently as 2019 the first known interstellar comet was detected by an amateur). Professional astronomers, in contrast, would probably have a hard time picking out more than a few constellations if outside at night*, but delve into the math (so…much…math…) to plan and carry out much more detailed observations than could be made by the human eye (including outside the visible spectrum, or from spacecraft). Thus to say someone is an amateur astronomer carries no insult, merely a description.

      *A stereotype, as someone *can* be both an amateur and a professional astronomer, but generally pretty accurate in my experience.

  7. The discussion of unit independence reminds me of Nelson before the Battle of Trafalgar – I don’t know a lot about naval history but I’ve been involved with a few Trafalgar Night Dinners which are still a big deal for the Royal Navy and make a focus of extolling Nelson as an example of leadership.

    He famously gathered his captains before the battle and did not merely give orders but ‘sold’ them on his (pretty radical) battle plan, also giving them autonomy with the wonderful line “No captain can do very wrong if he places his ship alongside that of the enemy”.

    1. Age of Sail command is fascinating because you have incredibly high bandwidth discussion before the battle, but virtually none during it. Nelson and his “band of brothers” had been working together for years, and would have had many, many lengthy in-person discussions about “what we do when the French come out to fight”. What if we have a strong westerly wind _and_ we’re close to a lee shore _but_ the French outnumber us? All right, now what if the French are upwind of us and outnumber us but there’s only a light wind? What will Collingwood probably do if his squadron is separated and upwind of the body of the fleet and engaged by a superior French force? What will he expect me to do? And so on.

      But once battle was joined, you’re basically hoping that your fleet can see the few bytes of information that your flags convey – until you lose all your masts and stop being able to hoist flags at all, of course. Nelson didn’t send a single signal after the first shot was fired at Trafalgar. The last signal he sent was “18”, as the fleets were approaching: “Engage the enemy more closely”. And he left that flying as long as he could.

  8. Great writing as always. Couple of misspellings I picked up.

    “Our sources tell us as much (Plb. 18.31-32; Liv. 44.41.6) and that same conclusion emerges pretty clearly in both the battle narratives we have an our understanding of how the legion was expected to function in battle (discussed last time).”

    we have an our –> we have **and** our

    “Caesar isn’t ordering his – he’s not even on horseback at this point; one is left to assume this was a quick reaction probably by the senior-most centurions in the rear-most cohorts.”

    Seems to be missing a word between “his” and the hyphen — probably troops or army?

  9. I wonder about the level of “backflow”, the degree to which military culture influenced the larger social structure. The Romans, for instance, especially in the Republic, were a hugely militarized society, where everyone who was anyone important spent a significant amount of time in the army. And I seem to remember reading, although I know decent sources are very thin on the ground for the Roman monarchial period, that they are thought to originally have fought in a phalanx very much like a Greek polis, virtually no command structure and all.

    Was the change away from that and into the precursors of the much more flexible, much better documented system they have centuries later driven by a differing social structure, or by the tactical and operational demands that they were facing? And if it’s the latter (And I get the sense it is, although I’m rather ignorant here), then the social structure must have allowed for a more significant delegation of authority, it just wasn’t exercised; but at the same time I can’t imagine delegating that authority, especially in a society where meritorious conduct in battle was so lionized, wouldn’t have effects once those people mustered out of the legions and got involved in civilian political life back home.

    1. Some of it probably goes down to the statement Brett previously made about the behavior of the Republic. Something along the lines of “one can understand the policy of the Republic as focused on the attainment of bigger and better armies so that the elites could win bigger battles to gain more political power at home.” The sense I get from that and from a casual knowledge of the military reforms is that generally, once they started winning with these new systems, very few people cared, since the fundamental social structure didn’t change, the elites still commanded overall.

    2. I found myself sort questioning, well not the idea but the exact way it was suggested by Dr. D.

      I am not sure that militaries tend to replicate the institutions of the society they come from. However, they would almost always replicate the culture, but in a filtered form. This would continue until the culture of the military was itself forced into a new mold by circumstances, driving changes back towards the parent culture.

      But man that’s a really complicated issue. I have in mind as a model the American military, where I think you can very easily the liberal-but-a-little-aristocratic culture of the day replicated in the Continental Army, but also the cultural shift necessary to win the Revolutionary War returning to drive corresponding change in American Society.

      I bring this up because there was a distinct lack of a formal model to follow. The colonial institutions were basically taken down and re-founded, and the military structure only partially followed from them. I’d probably suggest that civil and military structure followed similar, simultaneous development rather than one drawing from the other.

      1. I think the primary difference- and here the American colonials were just anticipating what Britain and most of Europe would come around to as the Industrial Revolution (and mass armies) transformed society so much – is that almost all Continental Army officers came from the middle class out of necessity because there was no real aristocracy to draw from. Henry Knox was a bookseller; Nathanael Greene ran a foundry; Benedict Arnold was an apothecary (and part-time smuggler).

          1. It’s a terminology different but not a social difference: What’s called the lower nobility on the continent is called “gentry” in Britain. In practice they are (and were recognized as) the same kind of people.

            Though I do note that eg. 17th century Sweden sees a certain amount of churn: You need officers to helm all those armies, and there’s not enough nobles around, so you make more nobles. It’s all theoretically still at the discretion of the king, but at least for a time there seems to have been a pretty solid expectation that you’d get ennobled after a certain period of service or distinguishing yourself. (this caused the 18th century reaction that closed off all ennoblement in the 18th century which had it’s own knock on effects, but that’s a different story)

            (to take an example, by the late 1600’s there were roughly five times as many nobles as in the 1500’s)

          2. In Sweden of 16th and 17th cemturiea, the major commoner landholders could get “free” of taxes by equipping and recruiting a cavalryman. (In fact, if you were willing to do this, you could get even four other farms’ taxes allocated to help you in the enterprise.) Of course, should the cavalryman die, you would need to send a new one to maintain your tax-free status, so this was a risky proposition but brought a lot of social capital.

            You could either recruit an outsider to ride for you, go yourself or send a son. It was possible to advance from the ranks to a battlefield officer promotion, and finally, to nobility (nobility would come around the time of getting your own company), but this was not easy. Most rusthållare never became nobles.

          3. Coming back to this a while later, this reminds me of some of the “fyrd”-like systems among some of the Germanic cultures (such as in Anglo-Saxon England). Dr. Deveraux has mentioned these at times.

            In such a system, one of the major ways that the typical household’s resources are mobilized to support the state may be that every X households are required to pool their resources and equip a single capable, adequately armed and able-bodied fighting man, who can then be called up for military service as needed.

      2. The American military and the revolutionary war is a very exceptional case where the society was trying to create new institutions, new social structure at the highest level, at the same time as it was creating an army.

        Even so, the new American Continental army looks to my inexpert eye a hell of a lot more like the British army than, say, the armed forces of the Iroquois and the Five Nations. European army officers could train and command American troops with only minor differences.

        1. Yes, the number of european officers who signed up to fight for the rebel cause was significant (either for ideological reasons, or just to get experience, the entire “There’s a war going and we need some experience so let’s go join it” is one of those fascinating bits of the time period)

          1. Coming back to this later, I think it’s connected to the fact that the European officer class still had its roots in the medieval landed aristocracy. To such officers, “military officer” isn’t just a job you do for the state you live in; it’s part of your basic identity, because you were raised from birth within a culture that promotes such activities as a core part of your worth as an individual.

            Which means that there is a strong incentive to go looking for opportunities to win glory and experience and prestige fighting in someone else’s war, as long as they can pay and as long as your own country doesn’t need you in a way that successfully plays on your own loyalties.

  10. “The absence here is particularly notable in games with fantasy settings that feature ‘endless horde’ style armies like Warhammer‘s Skaven or Orcs; forcing a player to account for an army that is much less controllable might make these factions feel different in a way that would also make historical sense.”

    For Warhammer, at least, this is a relic of the fact that the game descends from a tabletop miniatures game where there is no automation of decisions. Only the player can choose where to move his units. (Though Warhammer Fantasy Battle did include pursuit rules for when a unit breaks an opposing unit, which can cause your cavalry to run off the board in pursuit!)

    And of course so many of our computer wargames descend from tabletop wargames, where that unrealistic amount of command and control is necessary for the game to function at all. It’s only with the advent of computers that you could even start to think of letting the game system take control of your units in any but the most rudimentary ways. (Like the aforementioned pursuit rule.)

    1. While very elaborate systems of command and control loss are hard to do in a board game, relatively simple yet effective ones can and do exist. Something like Hornet Leader, for instance, manages quite well for all bandit actions in a thoroughly automated system to provide opposition in a solitaire game.

      I don’t play Warhammer, but I would think that a horde of aggressive, low-tier ‘conscript’ hordes wouldn’t be too hard to automate with a relatively simple script. The unit MUST advance to attack in such and such situations and it MUST hold its position in all others unless maybe you have a specific officer unit stacked with it at which point your options expand or something. You could work something like that into a hex and chit game, or a miniatures one.

      Another possible alternative is something you see fairly commonly is to have a card-side system with ‘command cards’. Offhand, I can think of the historical (Napoleonic wars) Manoeuvre, and a Star Wars Battle of Hoth game both have the system: You can only have your units move or attack or whatever you want to do when you draw the appropriate card from the deck, and those cards are generally keyed to certain types of units and are NOT evenly distributed. The lower-tier units simply don’t get to move all that often, and spend a lot of time standing around, unable to do much more than retaliate if attacked.

      A third option, which I admit I’ve only seen once, (It was an ACW game I’m blanking on the name of. Maybe Terrible Swift Sword?) revolves around a ‘devolved’ command system. Every so often, your opponent gets to move some of your units, and you can be damn sure they’ll make ones that help them and not you. Making your conscripty units partially controlled by the enemy who will make them charge unsupported or stand still when everyone else is attacking or whatnot could do a good job of reflecting the lack of command and control.

      1. This would also be easy (and arguably fun) in the video game format. It would also be pretty easy to adjust by faction. Orcs, for instance, could very easily have relatively low morale base but as they build up a more powerful Waaaaagh! their morale skyrockets… but they may also just hurl themselves forward in a way the commander does *not* want, and making counter-play much easier.

        And those Khornate Bloodletters might just ignore you completely and immediately charge the foe just because they do that, even though the rest of your army might be standing around confused.

        I also think a flipside might be interesting, like Skirmish units that literally can’t stand up in melee and will always retreat if threatened, so the player can’t choose how to use them at all, and the soldiers won’t just stand there and die simply because it’s convenient for the general.

      2. You could also simply play with the moves allowed. Low control units can move, medium ones can move and turn, high control can also move back… Or maybe, you have to expand a “command point” to do anything but the most basic thing (representing officer’s action), of which the Romans have a virtually unlimited supply but the orcs nearly none…
        Simpler solutions, less realistic, but easy to implement…

        1. The flash game Hex Empire has world war two like unit counters playing a relatively simple game, but each player only has 5 moves per turn. I always interpreted it as a logistical challenge, but on a lower scale, it could be considered C&C as well.

      3. You’re right that some different command structures have evolved for wargames, but those are mostly recent innovations. I would say that complex solitaire systems and card-driven tactical systems have only really come into their own in the last fifteen years or so. It’s a continually-evolving system of mechanics, too.

        Obviously, wargames were being ported to PC long before that, and so PC wargames inherited a lot of features of earlier tabletop wargames. I don’t expect that to last forever, however. The newer generations of tabletop wargames are putting more constraints on players, not fewer, and I expect those sorts of constraints to make their way over to computer wargames at some point.

        In fact, I could see a designer tap into some of the ideas Dr. Deveraux is positing in this series and decide that it could be the basis for a really excellent PC game. Maybe they could even decide to hire our gracious host for a consultancy role?

        1. It’s not really that new. Outside of Manuevere, which came out in 2008, all of the games I listed are from 1993 or earlier. Some of that is biased from the games I pick up, but I don’t think it’s as recent of a phenomenon as you’re claiming. And if you’re expanding the scope to “all command restraints” it gets even bigger. My single favorite wargame is World in Flames by ADG, and that has as one of its biggest mechanics is a command limit system where even if you have the troops and supplies to attack into X, you probably don’t have the ability to organize, plan, and execute everything you want to do, and that organizational power must be rationed. That was originally published in 1985, the “Final” (but not really) edition I did most of my playing in was officially published in 2003 and that was after a years long playtesting setup where the rules were largely available.

          1. Advantage of being old is that I played a lot of tabletop wargames and boardgames in the late 1970s and 1980s. You’re right that there are older examples of limited command and control, but robbbbbb is right that it is a mostly recent innovation.

            The dominant – not exclusive, but very common – style of tabletop gaming before personal computers had a lot of complexity and record keeping because that was necessary to make the games more realistic and challenging. For those games it made sense to let the players have a lot of command and control because they were doing everything else related to movement and combat anyway. If you are scribbling down movement orders and stacking resource counters it would seem silly for the game rules to pronounce “no you can’t move that unit/counter where you want to”.

            Once computer gaming became widespread, a lot of tabletop gamers – again, not all – decided that having a computer doing all the record keeping and calculations was more enjoyable. Tabletop games have largely – again not wholly – IMHO been redesigned to focus on different aspects that computer games are not so good at. One outcome has been thinking about how best to represent command and control problems in a reasonable time frame, which has led to ideas that were rare in earlier periods becoming widespread today.

        2. There is/was a PC game called Kohan that has/had a regiments system. Once battle was joined the regiment largely did its own thing and player control was limited. The game had a sequel but has disappeared since.

          I dream of a modern day remake centred around this mechanic.

          In the meantime, there is Dominions…

      4. Obviously you cannot (short of telepathic mind control) issue orders to enemy units. But there could be a sort of morale effect (scary noises, display of slain foes, appearance of badass warrior with awe-inspiring reputation) that could be projected as a counter to enemy cohesion. A very successful morale attack sows panic and the enemy runs or at least breaks. A mediumly successful morale attack creates reluctance by the enemy to advance. Or a successfully feigned appearance of weakness or a taunting provocation causes the enemy to attack out of order exposing themselves to counter-attack.

      5. You wouldn’t even need particularly complex pre-set scripts for the horde units. The defining feature of units that need more officers isn’t that they behave randomly nor according to preset scripts, it’s that you can only give them simple commands and will have a lot of difficulty issuing new ones. I think Bret has used a comparison to dumb fire rockets before, which is an easy and probably sufficient way to model a horde: You launch it in a specific direction, it engages the first enemy unit it hits, and you don’t regain control of it until it is no longer engaged, at which point you can launch it in another direction (possibly with some delay for messenger travel time, if we want to incorporate that part of it).

        1. Maybe you could simulate this with a rule that every unit needs an officer leading it, and then let different armies have different numbers of officers to simulate their historical organisation. So, e.g., a Roman army could bring lots of officers, allowing them to divide their troops into lots of units and so execute comparatively complex battle plans; a fifth-century BC Greek army, on the other hand, might only be able to bring a few along, meaning that their army has to be organised into a few big blocks and limiting their tactical options.

          Another, not necessarily mutually exclusive idea, would be to make the player roll for “disorganisation” whenever a unit moves. Poorly-drilled troops would be likely to fail the roll if they try to do anything more complex than advancing straight ahead over open ground (and they might fail even then), whereas well-drilled troops would be able to execute more complex manoeuvres (wheels, about face, etc.) with only a minimal chance of failing the roll. A failed roll could give a debuff to a unit’s combat and moral stats, unless it stays still for a turn (to represent the unit pausing for everyone to find his place in the line again).

      6. There are excellent tabletop wargame systems that replicate this rather well, actually, and have for some time. Formerly The Gamers’ American Civil War Brigade series had this mechanism: you write down orders as General in chief and then roll and hope they get 1. actually delivered quickly by an estafette (who might get shot on the way) and 2. be implemented by your subordinates (corps level) who may or may not want to obey based on his stats. Lots of ways this could go wrong, not least the “charge and take this hill” that is only implemented after too many turns have passed since the order was issued, when the enemy units are now occupying the formerly unoccupied hilltop… Combined with excellent morale and stragglers rules (not rare to see half your units strength lost to stragglers as fighting goes one), it made for great ACW wargaming, if quite frustrating. See https://boardgamegeek.com/boardgamefamily/119/series-civil-war-brigade-gamers and https://boardgamegeek.com/wiki/page/Civil_War_Brigade_(CWB)_Series

      7. the biggest problem with a command card style system is it often limits command ability in a weird way when dealing with certain types of formations- they generally end up set up to deliver a system where a unit does nothing unless commanded, making bigger armies less maneuverable. But this ignores the fact that in reality, many very non-maneuverable armies could be told to move forward and fight quite easily (in fact, it was getting them to do anything but walk in a straight line at the enemy that was the problem).
        Frenzy sort of solved this on tabletop warhammer. by making ill disciplined but bloodthirsty units, like khorne bersekers, move directly towards the nearest enemy, which allowed them to be baited out of formation, but was countered by the rule also granting an additional 50-100% damage, ensuring that frenzy cat-and-mouse was vitaly important to certain matchups. I think it doesn’t really fit total war as turn based manouver made fully kiting a frenzied unit essentially impossible unless they were commanded by a complete idiot, and the combat rules meant that you often had to completely sacrifice the bait (you couldn’t micromanage an eagle so it flew off right as the charge hit home without a risk of it fleeing the battlefield entirely, and the distances between armies on garage-fittable tables meant that you’d often have to let the frenzied unit hit something, which would, faced with an enemy that did almost twice the regular amount of damage, instantly evaporate).

    2. The discussion of Warhammer tabletop rules seems to be missing the “Animosity” rule for the Orc army, which is precisely such a rule simulating horde command structures. Occasionally an Orc unit will attack regardless of whether that is the command you would prefer to issue, and rarely would even attack your own units. This strongly affects the play style for the Orc army, because you need to plan for some of your troops attacking even in situations where it would be better to hold or withdraw. You could include special Black Orc officers who would allow you to ignore animosity at the cost of killing a few of the troops to maintain order.

      1. But that is an exceptional case in Warhammer applied only to Orcs, the default assumption elsewhere in Warhammer is near-perfect control. In other tabletop wargames from the same time frame *all* troops have some potential to behave in ways that the player didn’t intend or want.

        As is so often the case this is more about “what is fun for our players?” and less being “realistic” because the players in a Warhammer game are not looking to recreate historically accurate battles and the challenges of being a general in such. Warhammer also seeks to emphasise differences between factions and troop types rather than look for similarities, so having special rules just for Orcs fits the game.

        (I’m not saying that Warhammer gamers *always* ignore realism – the player happily moving ridiculously impractical armoured vehicles around on the Warhammer game board may be equally good at realistic phalanx tactics or dreadnought battleship command in other games.)

        1. It’s true that the default, but depending on edition other psychology rules definitely impair your command and control: Frenzy, Fear and Terror etc. All impacted what your units could do (eg. you had to pass a leadership test to charge a unit that caused fear and/or terror) (you also had to pass one if charged by one, but the consequences were worse in that case, rather than just not charging your unit would break) Pursuit would also require a leadership check to prevent, for instance, and there were similar other things (IIRC; reforming, IE: deepening or shallowing your formation required an LD test too back in the day)

    3. I do not ethat Warhammer has, at times, had precisely these kinds of things: Units with the Frenzy trait will always charge if possible. Orc units have to talk leadership tests and roll on the infighting chart regularly, etc. Goblins fear elves so have to take leadership tests whenever they want to charge them, etc.

    4. Tabletop is where the units come from, but the engine and game systems come from Total War, a computer-only franchise that started out modeling Japanese Warring States (ie early modern) warfare.

  11. I seem to remember Caesar’s Civil War stating that at one battle Pompey’s forces held their ground rather than charging, expecting Caesar’s line (which had started its charge at the usual distance) to exhaust itself covering double the ground; and then stating that all the centurions spontaneously stopped the charge, reformed, marched forward, and then charged from the appropriate distance for a standing foe all without orders.

    Caesar holds that the morale effect of charging/being charged while standing was significant, and that the centurion’s quick action denied the enemy any fatigue advantage.

    If I am remembering correctly, this would seem an extraordinarily good example of such low level initiative, in that they not only disobeyed a just given order to charge, but they all did so and all expected everyone else along a long line to do so without time to coordinate the action when they saw that the situation was not as expected.

      1. Another good example of a large force all acting against orders in coordination because the junior officers decided it was the right thing to do.

        The battle I was remembering was Pharsalus, and did in fact have the account I remembered.

  12. The Youtuber Officially Devin once suggested that chaff like Ork warriors or Brettonian peasants should only be controllable when your general is right next to them. Outside of this range they will follow a fixed script, which is to charge the enemy and pursue them as long as possible. Or more likely, be slaughtered in droves and run the hell away.

    1. Or more realistically for the peasants, to stand in a line for a short period. Maybe deploy them more like defensive structures, like a line of stakes — basically no control of them, just put them down and they stand there and act as an obstacle until the battle is over, or until an enemy breaks in combat with them (at which point they charge the enemy down — whether you want them to charge or not!).

  13. You introduction about what constitutes an officer had me wondering if staff officer roles and other non-command roles billeted for officer ranks are officer roles in the functional sense or merely officer roles in terms of rank and hierarchy. For example, my father was an Army chaplain who retired as a colonel. There were times in his career when he held authority over other soldiers — whether other chaplains or chaplain assistants (or even just one chaplain assistant at the battalion level). But there were also times in his career when he did not. Since his functional area was finance, he also had times where he primarily exercised oversight over civilians, not service members. I’d imagine the same is true for JAG, physicians and similar specialties. And even in more traditional military branches, officer career progression moves them through billets that aren’t command roles. I suppose some of these are like the praefecti sociorum in that they could “command in a pinch,” but others expressly prohibit exercising combat command, as with chaplains and physicians. Has the modern military conflated officer ranks with officer roles in a way pre-modern societies didn’t?

    1. A specific thing for the US military, and government in general, is that it has conflated ranks with pay scales, so that likely had a lot to do with it.

      Incidentally, this may also be an example of the military replicating the institutions of the society it’s drawn from, although it’s also possible that it goes the other direction. Most companies (at least in my line of work, which is the tech industry) have a leveling system which is integrally connected to pay bands. One of the big organizational complications is how to differentiate managers from extremely experienced employees who are not managing people. This is very much not a solved problem, and different corporations do this differently. However, some of the resulting solutions look remarkably similar to the separation between enlisted and officer promotion tracks in the US military. In a lot of ways, it’s the same problem to solve, after all.

    2. Staff officer roles in particular descend from the noble entourage of a general, and typically those positions are waystations in the career of a command officer.

      Very different from the specialist officer ranks like that chaplain one you describe.

      1. Wasn’t the Prussian general staff that kicked off the modern iteration of the general staff model organized as sort of permanent class of advisors and planners — often with the same advisor and commander attached to one another throughout their career? I thought they had a separate career track as a way of ensuring that Prussia wouldn’t need to cross their fingers the next time a war came around and hope for a someone like a Napoleon who had a particular genius at the organizational responsibilities.

        1. This was a source of their advantage – in every contemporary army, staff officers were still selected by the “friend of a general” method, with no standardized or mandatory training.

          It took until the Prussian victories of 1860-70 for European armies to realize they needed an equivalent. eg the École de Guerre was founded in 1874, and Sandhurst’s importance greatly increased.

          The US was ahead of the game on this, as part of its professional-feds/state-militia combination.

  14. “By contrast in the Roman army, the centurion – a senior NCO like the sergeant promoted from the common soldiers”

    Well, kinda sorta. This was certainly true in the Republic and early Principate, but at least by Trajan’s time if not earlier it was possible for a young man of equestrian rank to be “commissioned” directly as a centurion. This was an alternative cursus to the traditional one of auxiliary prefect-tribune-cavalry prefect. Why this came about, or what the attraction of it might have been, is unknown; it is possible that there were more applicants than available prefectures, or it may be that some equestrian families only had the dosh to pay for one son’s career on the more expensive path. At any rate, starting off as a centurion at age 20 or so would in the long run create a much better shot at becoming a pilus prior or primus pilus before retirement than for those who only started after a dozen years as rankers.

    Note also that a senior centurion had a status much more akin to a senior field-grade commissioned officer, say a colonel. A primipilus after completing his enlistment was eligible for the post of praefectus castrorum or camp prefect, which besides being the legion’s ‘quartermaster’ was also considered to be third in command, able to take over in the event the legate and tribunus laticlavius were absent or incapacitated; or a full-term primipilus could leave the army and receive not only a senior civil service post (again, like a modern colonel), but equestrian rank if he didn’t already have it.

  15. Question for anyone whose served in the Armed Forces: Do you think Officer Candidate School is an important prerequisite to becoming a good CO? Or do the higher-ups just assume it is because of the traditional NCO-CO divide? My current layman theory is that in pre-modern armies, the NCO-CO divide was more important socially than logistically. With pre-modern war being (relatively) more simple, a competent soldier could have risen smoothly from the bottom rank to the top if the social system allowed for it. With modern war, even though our social system (in the U.S.A.) allows for more upward mobility, the need for extra training is greater dur to the complexities of modern warfare.

    What say ye?

    1. My understanding is that the system was originally in place because of a class divide that isn’t so stark anymore, but it’s still a useful system.

      The issue is time, or age. The military is a physical occupation, even at the top to some degree. You generally want your generals to be under 70, and the current retirement age for US generals is 64. But in order to become a general, you’d need decades of experience in various roles and functionalities and it’s honestly a waste for a promising young officer to have to spend a decade going for private to private first class to corporal to sergeant, before he becomes a lieutenant. It’s better throw him into the deep end as a second lieutenant, have him get a taste for what it’s like at the bottom for a couple years, while also gaining leadership experience.

      1. So does it really take a decade to go up all the enlisted ranks?

        I guess my real question is, what is the difference between the highest enlisted rank (Sergeant Major?) and the lowest CO rank (Second Lieutenant?). that the latter needs extra schooling while the former does not?

        1. In the current system, the first is a veteran who knows how things are actually done, who knows who to talk to if you need new equipment fast or how to practically organize things. The second is a person learned in how things are supposed to go in theory, and in keeping the larger plan in mind, but with little on the ground experience.

          Ideally they work together to make a plan that fits the doctrine and theory but which will actually be doable, and communicate it well to the lower commanders.

        2. Part of it is the assumption is that the Sergeant Major has, for lack of a better term, gone through the school of hard knocks, and the second lieutenant has not. Also, usually, there is a theoretical difference in their roles, namely that officers give orders and non-commissioned officers carry them out.

        3. It is important to note that the rank of “lieutenant” originated as, essentially, the rank of “trainee captain.” The very word ‘lieutenant’ means ‘one who holds instead of another,’ and the thing being held is, implicitly, “the status of being the captain.” A lieutenant is someone who does the captain’s job only because the captain is not available.

          A captain is a necessarily experienced and proficient commander of soldiers on a relatively large scale (100 soldiers or so). Because of the nature of modern warfare, these soldiers carry diverse equipment, have complex supply needs, are often subdivided by terrain and out of direct line of sight of each other, and encounter an extremely diverse variety of threats.

          Being a captain is a hard job, and it requires considerable education and training (including practical experience) to make sure that the captain can issue good orders and maintain the morale and physical state of the soldiers under their command.

          The best way to understand the difference between a lieutenant and a sergeant is that a lieutenant is an inexperienced trainee captain, and a sergeant is not. Sergeants are what rank-and-file soldiers are routinely expected to grow into if they gain experience and don’t demonstrate any particularly serious character defects that would prevent them from exercising low-level command.

          Captains are not simply a developed form of rank-and-file soldier. To do their jobs, they require extensive specialist knowledge that rank-and-file soldiers are under no obligation to keep track of. Obtaining that knowledge would not help the soldiers who make up 95% or so of the army to do their jobs, and it takes years of study in most cases, so there is no point in creating a routine process by which the rank and file eventually become captains.

          It would be a waste of time for all involved, because it would require soldiers to study “captaining” skills they don’t need, and the trainee captains to study “soldiering” skills they don’t need. A captain does not need to be an excellent soldier for the same reason an orchestra conductor does not need to be an excellent violinist.

          With that said, in a modern army free of class-warfare nonsense, here is no fundamental reason a sergeant cannot become a captain (by way of becoming a lieutenant first). It’s just that this isn’t something that needs to happen so routinely that it would be the standard promotion track.

    2. What’s the alternative, offices just go into the same basic training program as enlisted soldiers? Seems pretty logical they would have a different program.

      1. My understanding of the system is that almost nobody enlists directly as an officer, *unless* they already have at least a college-level education. Most officers enlist during college via the ROTC. Others (I think) start out as enlisted troops, go to college, and then apply for OCS. A few start out with military academies such as Annapolis or West Point; there may be private military academies which serve as something of a pipeline, though I’m not familiar with that side of things.

        Either way, they always go through at least some level of basic training. Not alongside the enlisted, mind you, but they still have to learn how to wear a uniform, march, and shoot, all the usual military stuff.

        1. Broad strokes, you are correct. The overwhelming bulk of officers in the US DOD come from ROTC (AFROTC, NROTC etc) pipelines. Specifics vary, but at bare bones, they take about a 3 credit class each semester, have to do physical or practical training a few times a week, have weekend long training sessions 1-4 times a year, and at least one important (can easily be failed, removing you from the program (Marines), or your grade there is competitive, feeding into what your employment options are on graduation (Army, AF)) training block for about a month or so in the summer between junior and senior year.

          If you’re already a college graduate (BA/BS… Army has an option to take associates and finish later, but keeping thing broad) the Army and Marines in particular have a decent share of OCS contracts. Marines send you to OCS, and if you fail you’re a civilian. Army sends you through the enlisted pipeline, then OCS, and if you fail you finish your tour as a soldier. Not sure of the specifics for Navy and AF–the OCS slots are surge capacity, and when I went through there really wasn’t much demand for them to fill (late 2000s).

          Then there are various enlisted commissioning programs where an already enlisted soldier, typically junior NCO (4-8 years service, E-3 to E-5, maybe junior E-6) is enabled to finish a bachelors degree and do OCS.

          Senior Military Colleges (the actual DOD designation for The Citadel, VMI, etc) are part of the ROTC pipeline, with some specific differences that are finer grain than things I have already broad brushed past in this summary.

          Source: Commissioned USMC from The Citadel

      2. That’s how Israel does it – fresh officers are recruited from the ranks coming out of basic training, and sent off to extended officer training.

      3. Why not? Coming from a country with decades of conscription system, sounds perfectly natural Everyone has the same X months of basic training program as the everyone else. After basic training you could get sorted to various underofficer / officer training tracks according to demonstrated and evaluated ability (and after that, there is opportunities for active service career).

        I had first difficulties understanding the English-language system when I first started reading English-language media, especially when I tried to mentally translate it as a kid. The only “commission” I had heard about was the leadership of the EU and the commissars in the Soviet Russia.

    3. The Peter Principle suggested that the class-based NCO/Officer break was organizationally useful because it prevented competent people from being promoted until they reached a job they could no handle. You got senior sergeants who were good at their jobs and were basically stuck as sergeants. I’m not sure how that propagates into more modern systems

    4. I would like to note that the Continental countries had a much less clear-cut view on the officer/NCO career path separation. For example, in the Imperial Russian army, a young man with some university education could become an officer by enlisting, then serving half a year in the ranks, some time as a junior NCO and then being promoted to lieutenant. (Admittedly, they also had a career path for direct commission.)

      Similarly, in the current German system, even if you are contracting with the Bundeswehr to become an officer, the first half a year of your service follows the rank-and-file career path then includes a training and a short practice as a junior NCO, after which you go to the university to get your master’s degree, then get some more military education, only then getting promoted to lieutenant.

      The American way of training officers so that they definitely don’t have any experience of serving in the ranks is actually somewhat odd, even in historical comparison. Even decidedly much more aristocratic countries have seen it useful for officers to have some experience of how the people downstairs live.

  16. The lack of independent officers always bugged me in the Total War series. If I want to go play Alexander with the cavalry, it would be nice for the rest of the army to react intelligently to things.

    1. You actually can, with some difficulty. You’d have to split up your army into two stacks on the campaign map, one with the cavalry you want to directly command and the rest with the rest of the troops. You then set the reinforcement army to AI control and only directly manage the cavalry.

      I’ve never done it myself, since I can’t stand the AI sending my troops to die on its idiotic play, but you can do it if you want. And while I’m not sure it’s “intelligent”, it will try to react to things.

      1. At least in more recent Total Wars (Rome II, Attila, 3K, Troy, I think Warhammers but haven’t played as recently) you can simply group units and set them to AI Defend or AI Attack and get a pretty reasonable experience (this also does give you the ability to change that order or perform a degree of micromanagement to handle some things the AI isn’t so good at).

  17. You compare the Roman centurion to modern NCO, and that seems to be a common comparison nowadays (seen it in both historical fiction, fantasy, and the way Romeaboos in the modern military operator), but I’ve seen some analyses that say that’s not really the apt comparison.

    That the Centurion wasn’t really comparable to a sergeant, deriving command authority from an officer, but more like a captain-colonel (depending on seniority of the centurion) in his own right. And far from rising up from the ranks of common legionaries, many centurions were from wealthier, though not Senatorial, background and were given fast track promotions, or even direct appointment to command ranks.

    1. Gettysburg may be the most famous example in American history, but there are many, many, many others.

      For instance, the Battle of Spotsylvania Court House, a year later, was defined by much the opposite. Shortly before, Grant let Sheridan ride off with pretty much the entire Union cavalry force on a long range raid in the general direction of Richmond. And the lack of good cavalry scouting of the Confederate position resulted in the Union army taking a slow and hamhanded approach to deployment, losing the opportunity to cut the Confederates off from the critical road junction the battle was being fought over. Sure, Sheridan captured a bunch of prisoners and supplies, and his cavalry killed Jeb Stuart, but the price was an ugly stalemate battle that set the tone for the rest of the Overland Campaign that was to follow.

      And there are no doubt hundreds of examples just from well-recorded history of “our army would have done better if the cavalry had been in the right place at the right time.”

  18. This bit was one of the first bits of military theory I learned, thanks to Ender’s Game. Say what you will about Orson Scott Card, or how well Ender’s Game fits with Speaker for the Dead and its sequels, or how good any of the EG follow-ups were (especially the ones centered around Bean, who is actually a genetically engineered supergenius), but the original novel is a grim coming-of-age story about child soldiers which also coherently presents Card’s thesis about what it takes to win a war—skilled, trustworthy subordinates executing mission tactics with small units, and inflicting such severe violence on the enemy that they can’t get back up (which is also severe enough to scar the perpetrators of said violence).

    1. Underscored by a deeply held conviction “to the core of his soul, that he can only do what he and the other [officers] work out for themselves… that no matter what happens, no [one] will ever, ever step in to help him in any way.” So, not just empowered to make independent decisions, but utterly dependent on it. Though that is perhaps largely informed by the nature of the campaign leaving navies separated by uncrossable light years.

      1. Though that is perhaps largely informed by the nature of the campaign leaving navies separated by uncrossable light years.

        Three points.
        1. As I recall, the same principles apply to Ender’s command in Battle School, aside from the odd plan that required the entire team to work together for some complicated maneuver (like the Battle of Three Armies).
        2. The ansible makes those light-years very crossable for command purposes…which is emphasized by the fact that the fleet commanders weren’t even born until after the fleets launched.
        3. Orson Scott Card chose to write a story about an interstellar campaign, and chose to limit his FTL technobabble devices to communication and not transportation. For all the criticism I have for Ender’s Shadow and its sequels, and the apathy I felt for the when I read Speaker for the Dead in high school, Orson Scott Card is clearly a skilled enough writer to pick a setting which suits the points he wants to make as a writer, rather than just tacking his opinions onto whatever sounded cool to write.

        1. I think Card’s depiction of command as involving a commander who is convinced that no one will step in to help him, and Card’s creation of a setting where interstellar distances can be crossed instantly by messages but not by ships, are things that grew up together.

          Remember that in science fiction, it is common for the story to start with known material facts, then extrapolate from them an artistically compelling realization about the likely consequences of those facts. It is likely that Ender’s Game originated as a sort of “marriage” of two separate ideas:

          1) The realization that interstellar warfare at sublight speeds would have no possibility of reinforcements; your fleet lives or dies on the strength of how well it can use whatever it brought with it.

          2) The compelling but inherently separate vision of child soldiers being raised as weapons.

          The ansible then serves as a deliberately engineered link between these two concepts, and creates extra elements that graft the two together.

          On the one hand, it fleshes out the nature of Battle School and what the children are being trained to do. The school is training commanders, not soldiers, because the soldiers must already have been launched to the target decades ago.

          On the other hand, it expands upon the nature of the style of command Battle School ‘must’ (abusively) train into its students. Because they will be commanding fleets that cannot be reinforced, and the command is by remote control under circumstances that frankly will make it easier for a single command team to run all the battles successively, like a simulator exercise, rather than having a dedicated commander to get to know each of the individual attack forces.

          1. Thanks for this, I now have a better head canon reason for the way Battle School operates.

            It does occur to me that the way the armies train in the battle room (just the zero-G shoot-em-ups, not elsewhere) is not very efficient? A standard army before Ender has 40 soldiers, in four platoons of 10, and a commander. So the platoon leaders, 10% of the total, have junior command roles that in the book are described as very restricted. Only the army commanders, 3% of the total rounded up, are developing high level command skills.

            Ender introduces a more flexible organisation with smaller toons, now approaching 20%, but still not great.

    2. One aspect of the Battle School in Ender’s Game is that absolutely everyone is fighting with the same equipment. There’s no combined arms training, no officers who say are specialists in interception vs specialists in defence suppression. I would guess this is to encourage potential officers to think about tactics and behaviour rather than burying themselves in weapon statistics.

      Makes me wonder though, could the Romans be so flexible on the battlefield because they had simplified their army so much? By the time the Romans fought against the various Macedonian successor states the latter had moved on from Philip/Alexander “pike phalanx plus cavalry wedge” to trying to incorporate elephants, scythed chariots, cataphracts, Persians, … Meanwhile a Roman army, to a first approximation, is 45% legionaries, 45% very similar auxiliaries, and the last 10% cavalry that nobody expects to do very much.

      1. I feel like that description of the Roman army hides a lot of complexity. Most obviously, the auxiliaries aren’t all going to be the same, and they’re certainly not going to be like the legionaries.

        Having a reliably standardized legionary core to every Roman army probably helped, though I think not for the reasons you’re suggesting. If every army is built around soldiers you can trust to know tactics X, Y, and Z, you can teach the officers how best to mix and match those tactics to handle various tactical situations. If you don’t know whether your army will be capable of all/any of those maneuvers, there’s not much point in educating officers how to best use them.

        1. Heck, during the middle Republic the legionaries weren’t the same — the hastati, principes, and triarii were all equipped differently, even though they performed roughly the same function on the battlefield.

  19. Is there any relationship between how far down the hierarchy independent command is expected and how far down the hierarchy you find commissioned officers rather than NCOs?

  20. For our host or other experts: I’ve been interested in the subject of junior officers in the Western European medieval period opposed to the classical examples presented here. Can you recommend a good place to start on the subject? Obviously I expect that it would vary greatly in time and place, but anything on basic practice would be helpful.

    1. Verbruggen’s The Art of War in the Middle Ages is a good start. Contamines’ War in the Middle Ages is more recent. Mid-later medieval cavalry fought in ‘conrois’ – small groups which trained together,and grouped under higher-level leaders (so a ‘knight’ and his off-sider might lead a group of 10-15 riders). Medieval infantry was usually drawn from town or district militias, and often mustered in groups of ten under their decennier, with a vingtenier (20-er) and a ‘fiftier’. These had a counterpart in civilian life in the very common fraternities and oath-groups.

  21. I’m quit curious as to how a levy would be officered, a local magnate would in the medieval period by with his knights in the cavalry. So a local funtionary such as a sheriff or maybe even wealthy peasent (i.e. miller) with a cadre of permanent militia memebers could have acted as the NCO’s to the levy infantry of a baron or earl.

    1. See Part IV of the Helm’s Deep series, about the organisation of Rohan and the similarities with early medieval armies.

      https://acoup.blog/2020/05/22/collections-the-battle-of-helms-deep-part-iv-men-of-rohan/

      My understanding is that an ancient / medieval army only levies actual peasants when the threat is immediate and local. No-one wants to march peasants off to invade someone else. (They’re not much use, you’d have to feed them, you want food to eat when you come back, …)

      The levy in a medieval army is more often from towns and cities, which have a higher population density and are rich(ish) enough to both be able to spare them for a not too far away battle and provide food and equipment. The towns usually have their own social structure even when nominally vassals to some lord, so the major, guild leaders, and other rich folk provide the leadership. At the lower levels the levy organises itself on neighbourhood / guild lines as described for the Army of Rohan.

      1. In scandinavia, that for various reasons had to rely more on levies, the commander was usually some local functionary (sometimes even the parish priest) but there are some indications that soldiers simply chose their own officers.

      2. If you look at eg Charlemagne’s regulations, it is clear that peasants were levied – in numbers, for service as far away as Hungary or the Spanish March. A rural population usually can spare quite a few young men with little or no impact on productivity – for war or emigration or as part of the continual drift into the demographic sink urban areas. Later urban militias were certainly better-equipped and more cohesive, but deeply rural areas (Switzerland, Albania, Croatia, Gascony, Scotland and so on) were major sources of effective military manpower.

        1. Part of it is that agricultural work is heavily seasonal: You can make do with quite a lot less manpower in the summer than you need in spring (for plowing/sowing) or especially autumn (for harvesting)

          1. There’s a reason why the old-fashioned view of an idyllic country life is herding, not farming. Sitting around watching the beasts in the field is actual work. A farmer waiting for something to grow is able to do something else and probably will.

        2. Notice that all of those areas are also marginal in terms of agricultural productivity. Peasants as well as nobles can run into the second/third son problem, and have fewer options for how to deal with it.

  22. Ok, so let me just get this down:
    *scribble scribble*
    “Beware…of Greeks…bearing gifts…and beware…of Pelignians…flinging flags.” Got it!

  23. As someone who self identifies as a logistics nerd, I like the decani, leading 6,. Why is a man leading 6 soldiers called a decanus? Because every contubernium had a group of servants attached, to do all the stuff that isn’t hitting strangers with sharp pieces of iron. Also that raises the size of the standard double-legionary army our hosts talks about by another 66%. Thats a lot of mouths to feed if you think about it. Espacially if we tend to not talk about more than half of them.

  24. The old Total War games (Rome I, and both Medival I and II), had noble cavellery that could charge with out orders, or even worse not disengange when ordered. I remember it to be a major pain in the back.

    1. Apart from the cavalry you mentioned the only other example for units acting on their own I can think of are the “Beserkers” in Heroes of Might and Magic IV, who automatically go and attack the nearest enemy on their turn. They’re pretty powerful at least, so they go as hard as they die hard. Sometimes it’s actually beneficial to block their path with other troops, or to slow them down with a movement-reducing spell that you’d normally only apply to enemies.

  25. Bret, I’m interested in whether you’ve played much of Imperator as of the 2.0 patch? It made some very interesting changes to the military system- implementing a levy/standing army split very similar to what was present in CKII, where levies’ force compositions are culturally determined and immutable and have to be commanded by the governor of the province they’re raised from, while standing armies (“legions”) are flexible in both and can be drilled in peacetime.

    It’s not revolutionary by any stretch of the imagination, but it is a much more nuanced and interesting model than Paradox have implemented in any of their other games (excepting HOI, though that’s approaching a very different period with very different goals), and a fair advance on CKII’s I think.

  26. There’s an incredibly janky game called Oriental Empires, set in Warring States China, that had some interesting wrinkles when it came to raising troops: You could either get peasant troops (cheap, shitty, cost population) noble troops (good but small numbers, and could only be recruited from your noble-class pops, who you couldn’t tax and didn’t do much other than serve as recruitment) or professionals (incredibly expensive and really only a late game option)

    The interesting was how they interacted, raising troops increased unrest from thier social status, and while you could use troops to stamp down on unrest *peasant troops only reduced noble unrest and vice-versa*, while professional troops reduced unrest from both. So if you relied heavily on peasant leveis you also had to pay much more attention to their unrest factors.

    The game itself was janky, but that I thought was fairly interesting.

  27. The webcomic Erfworld was set *inside* a putative turn-based wargame, where most units would attack enemy units by default. (Scout units would hide, instead.) Some units (mostly Commanders, namely Warlords and Casters) had Leadership, which in this context is like free will, the ability to choose whether to engage, for themselves and any units they were leading.

    (Having an in-character view of a wargame is weird.)

  28. One thing that “clicked” for me on reading this is why military aristocrats so often fought as cavalry. Normally this is ascribed to them being the only ones with enough wealth to raise horses, which is true, but I think another reason is that cavalry, by dint of their superior mobility, are often used for flanking and other relatively complex manoeuvres, which of course requires a relatively high level of training to pull off without becoming disordered. If your social structure allows you to raise an army with, say, 10% military aristocrats who can spend all their time training for war, and 90% levies who, however good their equipment and individual skill-at-arms might be, don’t really have much opportunity to practise formation drills, it makes sense for you to give the 10% the roll that requires the more complex manoeuvring, and to give the 90% the much simpler task of engaging the enemy head-on to pin them.

  29. The NCO/officer distinction has a new social basis in modern draft armies: conscripts serving their normal term can be NCOs, while officer rank requires signing on as a professional for at least some extra time (as little as 1-2 years for junior officers) to account for extra training.

    (There are also professional NCOs in such armies, it’s just that their rank and role aren’t conditioned on their professional status like for officers.)

    1. Actually, many conscripted armies have also conscripted officers. You don’t, at least formally, sign on as a professional but simply get ordered for a longer service. For example, the Swedish army had this system during the Cold War, and Finnish Army and Austrian Bundesheer have it still. The reserve officer is called to duty like any other reservist, but when in service, they are equal to the career officers. Of course, to become a reserve officer or to serve successfully in officer duty requires you to have a very good motivation, so your status as a conscript is a formality to some extent.

      In the Finnish system, the reserve officer gets the training to lead a platoon in the field, and in a war-time mobilised army, most platoon leaders and company deputy commanders would be reserve officers. Advancement to higher positions is relatively slow on the reserve officer career path, so a reservist reaches a position as a company commander usually in their mid 30’s, early 40’s. By that time, they have had usually a civilian university education (not a formal prerequisite, though very common), work experience as a manager in civilian life (also not a formal prerequisite) and quite a lot of extra military training. A few reserve officers may get a command-level assignment and a major’s rank, but this is rather rare (40 or so majors promoted a year, compared to roughly 250 captains and some 1,500 second lieutenants promoted yearly in the reserves). Reserve officer promotions to lieutenant colonel make national news, as they happen only a couple of times a decade.

      1. You do have to voluntarily sign on to extra time, that’s the important distinction.

    2. Conscripts served as officers in the 20th century American conscript army (until we moved to a volunteer force). My grandfather (in WWI) and (I think) John Kerry (in Vietnam) are examples. As I understand, accepting a commission was voluntary, but the alternative was serving in the ranks, not going home.

      As chance would have it, my grandfather’s original unit was eventually sent overseas, whereas the unit he was assigned to as a lieutenant after OCS was not. John Kerry’s course was different.

      1. The draft term was two years; IIUC officers had to sign on to 3-4 years for their commission.

  30. What’s really remarkable is how long the Macedonian form of army organisation persisted after Alexander. More than two millennia after Alexander’s death, the French army in 1914 advanced to the Marne in taxis.

    Sorry.

    1. Actually, even the French national military history museum – in Paris – which has one of those taxis in its collection, admits that those carried to the Marne front for the counterattack were a relatively minor part of the force. About 3,000 men got to the battlefield this way. The taxi bit’s principal influence was as a press morale builder for the French civilians.

  31. I got bored with the series around Dominions 3, but Dominions 5 has units with ‘undisciplined’ tag. Those units can’t be given battle orders and will just attack closest enemies. In contrast, other units can be set to wait for a few turns then attack, or fire at rearmost enemies, attack big monsters, or alternate shooting and advancing. The best example of such units are independent barbarians.
    Some commanders might also be insane, and it causes them to sometimes ignore orders or do something else. This is generally caused by magic or lovecraftian beings. I believe also some Tartarians you free come with this.

  32. Age of Empires 2 has had since its release in 1999 unit stances and attack/follow/patrol moves as well as formations, allowing for very intricate commands. Especially patrolling in defensive stance comes close to what one might call Auftragstaktik.

Leave a Reply to PaulCancel reply