Command and Control

[Patrick Waterson]

A thread to look at the command capabilities and restrictions of Biblical, classical and mediaeval armies.  What could commanders do with their troops, and how did they do it?

Pick a favourite era or army and add your thoughts.  We can also look at some examples.

[Paul Innes]

One of my favourites is the series of manoeuvres pulled off by Claudius Nero at the Metaurus in 207 BC during the 2nd Punic War.  First of all he manages a forced march to take some of his legionaries with him, to reinforce the army facing Hasdrubal.  Then he sneaks them into camp so that the Carthaginians don't realise he has arrived.  During the battle itself he moves uncommitted infantry from the Roman right facing rough ground past the rear of the army to reinforce the left, which helps turn the flank of the entire Carthaginian army.  A combination of strategic and tactical acumen, plus trickery, as well as the command control to get away with it.  Moreover, he does all of this with infantry!  One of the most underrated pieces of generalship, probably because he is eclipsed by his contemporaries Scipio and Hannibal.

Events like these seem very rare, and even then one usually thinks of cavalry commanders...

[dwkay57]

An interesting opening item by Paul as the original background to this thread was that army commanders - certainly in the classical Greek period - didn't have too much control over their troops once pointed in the right direction and battle had commenced.

To give this thread some initial context, in a follow up to a battle report I submitted, there was a good discussion as to the degree to which ancient generals could issue fresh orders, whether this had to go down a formal hierarchical command structure (that we might recognise in more modern times) or potentially had to give instructions direct to individual units.

"We" have moved the thread to here to allow the original thread to get back to its own purpose and look forward to an interesting discussion.

[Patrick Waterson]

The important factor in making Claudius Nero's reallocation of troops possible was that Hasdrubal had posted the opposing Gauls in terrain the Romans found too daunting to attempt.  Hence Nero's wing of the army was not in contact, and it seemed to have occurred to him that the Gauls would not be able to cross the terrain either if he did take his troops elsewhere (had he but known it, they were much the worse for wear from drink and practically immobile).  Hence he was in the happy position of being able to issue orders as if in a pre-battle situation, with the sides not in contact.  Better still, he thought of a very good place for his men to go.

Hence he in effect redeployed them for a march behind the active battlefield (where Hasdrubal's veteran troops were holding their own) and round to face the Carthaginian flank, then put them in order of battle, advanced, and Robertus was his uncle.  He does not seem to have done much with them after that, but then he did not need to.

Another classic (and extremely rare) Roman improvisation on the battlefield was at Cynoscephalae: Philip V's left had collapsed under the assault of the Roman elephants (curiously, all Roman victories against Macedonian phalanxes seem to have involved elephants messing up a key part of the phalanx) and was being pursued by the hastati.  An unnamed tribune (well, we are sure he had a name, but it has not been preserved) redirected the principes and triarii of one legion against the rear of Philip's successful right and brought off a classic victory.

In each case, an astute leader was able to get disengaged troops moving in a different direction with spectacular effect.  This we can perhaps ascribe to a talented leader seeing an opportunity and making use of a capability that existed but was very rarely employed.  This gives us two fundamental options when preparing rules for players:

1) Assume the player has a basically historical (and not particularly talented) set of subordinate leaders.  He does his battle planning before the battle and can change little during it but can, if his army used reserves, keep a reserve under his own hand (direct command) and use it as he sees fit subject to staying broadly within that army's tactical procedures (e.g. no pike reserve units - everything we have indicates pikemen all went into the fighting line).  If his army used charismatic leadership (e.g. Macedonian) he can take direct command of a wing and be creative with that.

2) Allow the purchase (extra points cost) of initiative-capable subordinates, who are able to issue fresh orders and/or generate extra movement capabilities for troops under their command.

These options need not be mutually exclusive.  An initiative-capable subordinate could also reliably carry out an ambush, flank march or the other manoeuvres beloved of wargamers but not often used by historical generals.  Ordinary subordinates might be assigned such battlefield tasks but with very little (15-20%?) chance of acting at the right time (failure = 1 turn delay, with the enemy apprised of the ambush) or of turning up at all.

The basic idea is that leadership rather than command and control was the big limiting factor if troops were well-trained and experienced; if troops were untrained and/or undisciplined then command and control became the limiting factor.

[Paul Innes]

Thanks, Patrick, those are useful observations.  I'm not familiar enough with the sources on the Metaurus to be able to go into any detail, but the mention of Cynoscephalae makes me wonder if it was the Triarii "reserve" that Nero shifted across the field at the Metaurus.  I think you're right about the rough terrain being a factor - presumably Nero left enough force behind on the right to screen the opposition, allowing him to make his successful redeployment.

I suppose that a follow-up to all of this would be to ask how best to model such possibilities on the tabletop, bearing in mind limitations on leadership and/or command control.  I know that many rulesets have a sort of engagement distance, i.e. units within a certain specified range of the enemy are considered to be involved, if not yet directly in melee as such.  This in turn would mean that availability of some sort of reserve would then need to be specified, which does not necessarily mean that the capacity to take advantage of the asset also exists.  Similar comments could be made about the difficulty of refocusing successful cavalry commands after they have defeated their initial opposition, exactly the sort of thing that failed to happen at Raphia and, perhaps, the Seleucid right at Magnesia.

Interesting thread, I think!

[Patrick Waterson]

Good thoughts, Paul.

Nero took to the Metaurus a force of 6,000 infantry and 1,000 cavalry - the pick of his army's troops.  His whole army consisted of 40,000 infantry and 2,500 cavalry (Livy XXVII.40) so looks like four legions at emergency establishment plus four alae of socii.  Cavalry is about half the expected strength, not unusual for this stage of the war.

So what did he take?  Evidently not a whole legion, but rather selected crack troops.  Four legions fielded 2,400 triarii and their corresponding alae 4,000 extraordinarii.  Together this makes 6,400 - which is a usefully close figure to Livy's 6,000 even if not conclusive proof.  The Roman cavalry contingent at full strength would be 1,200 but if we take it as reduced strength (200 per legion) then it amounts to 800, while the allied cavalry would be 1,700.  One third of 1,700 (the extraordinarii cavalry) would be 5-600 which, added to the Roman cavalry, gives 1,300-1,400.  It looks as if he may have taken his triarii and extraordinarii north, with his Roman and extraordinarii cavalry.

In the battle, Polybius (XI.1) writes that Nero 'withdrew his men from the right wing' whereas Livy (XXVII.48) says that Nero 'detached several cohorts' from his force, so on balance it seems he did not send everyone and my guess is that he sent the extraordinarii,  leaving the triarii to front the passive Gauls.  Or I may be placing too much emphasis on Livy's use of 'cohortes' and Nero may have done it the other way round.  Either way, a good leader with crack troops outside an enemy 'zone of influence' was able to implement a fresh set of orders.

On the tabletop, the key to getting the right mix in this instance may be to assign an 'initiative rating' to subordinate commanders (perhaps based on points value) and allow non-combat moves ('march moves' as they were called in WRG 7th) when outside an enemy 'zone of influence' (basically a buffer distance of about bowshot).  If units are separated by difficult terrain, the 'zone of influence' may be decreased, perhaps in line with the terrain-imposed movement penalty (so a 240-yard distance becomes a 120-yard distance if the intervening terrain halves movement, for example), allowing initiative-capable sub-commanders to manoeuvre troops despite being within what would normally be 'influence' distance.

Reserves (for armies which kept them) would usually be lined up behind the front line, and perforce out of any opposing zone of influence (if they come within such a zone - even if friendly troops intervene - they cease to be 'reserves' and lose their notional movement options).  Reserves under a C-in-C could act more or less as the C-in-C wished, subject to the limitations of the troop type, but what of reserves when the C-in-C is not commanding them but is instead leading one of the wings (or the centre) of the army?  In these circumstances the reserves would either be under a sub-commander with initiative (in which case the player could use them imaginatively) or they would not, in which instance they could be used only to plug gaps or relieve forward units (unit type permitting).

I think an important part of this, particularly for infantry formations, is to have a 'manoeuvre mode' and a 'combat mode'.  Reserves, or troops qualifying as reserves, would be considered as being able to use 'manoeuvre mode' and hence march in directions other than straight ahead.  Troops with enemy to front within zone of influence distance would be limited to straight ahead or, if the right kind of mobile troops, straight ahead or directly astern.  Exceptions might permit oblique movement for troop types of good quality who did so historically.

Cavalry commands, especially Hellenistic, might require a little fine-tuning for C-in-C or subordinate inclinations.  Antiochus III seemed to have a trademark feature of leading his right-wing cavalry to sucess as the expense of losing control over the rest of the battlefield.  Alexander never had this problem: even in mid-pursuit of Darius he could bring his cavalry back to a chosen sector of the battle in response to a call for help.  This seems to be a personality-related thing where the army concerned is regular and well-trained: if the army is irregular and/or poorly-trained then the unit may tend to act impetuously or truculently and hence inflexibly regardless of what the commander may wish it to do.

An interesting instance is the cavalry action at Zama: Laelius and Masinissa were both experienced and capable cavalry leaders with good cavalry contingents, but after their initial success were unable or unwilling to show up again until dangerously late in  the battle.  Rather than having an uncontrollable troops problem or a diff commanders problem I think what happened there was that the Carthaginian cavalry wings held together and conducted a 'fighting retreat' which occupied their more numerous opponents for some considerable time until they were able to break the Carthaginians.  There may also have been the matter of the Carthaginian camp and some perhaps enticing booty, though one would have thought that Laelius at least would have been able to exercise enough authority to prevent delay through looting.

Again, the determining factor here seems to be the enemy 'zone of influence': redeployment was very limited or simply not done within it, but was possible outside it if an officer with sufficient wit and ability was in charge.  There also arises the question of battlefield orientation: if it is not known what is transpiring on the rest of the battlefield, there is no impulse to change plans and detach units.  Nero at the Metaurus was probably aware that the action on the left had stalemated - he could probably see the heights - and was very aware that there was nothing doing on his front, so the idea of helping out elsewhere crystallised, although a dumber general might have sat and watched for the whole day.  Antiochus at Raphia and Magnesia was surprised to discover how the battle had come apart in his absence, and had he been aware that things were not going well he might have been able and willing to break off pursuit, return to the field and see what he could do.  Even Alex at Gaugamela was happily pursuing until apprised that Parmenio was in a sticky situation - had Parmenio kept quiet Alex would not have turned back.

This suggests that a good view of events on the relevant part of the battlefield and/or a send-for-help provision from troubled subordinates may be the key elements in deciding if and when a successful cavalry wing reorients to sort out matters elsewhere.  The view from the saddle may be a matter of radius/elevation/line-of-sight and I would suggest the send-for-help provision be linked to the ability of the sub-commander in question: a good general will realise sooner that he is in a sticky situation and will send for help in good time, whereas a poor one will tend not to realise the terminal nature of his problems until too late.  Given one player handling the entire army, each general could have a 'help!' rating based on his initiative rating (or quality rating, or whatever) and if/when the player desires to redeploy his successful cavalry he makes a 'help' roll which, if failed, limits the cavalry wing in question to beating up visible opponents.  If successful, cavalry not in an enemy zone of influence can get into 'movement mode' and travel elsewhere - with their leader.  I would suggest the player can attempt a 'help' roll at any time, but be limited (perhaps by subordinate intiative ratings) in the number he can make over the whole battle.

As the zone of influence concept is likely to be central, indeed critical, to determining what a unit can and cannot do, training and commander permitting, we may wish to specify that a retiring or fleeing unit has a conditional zone of influence: it acts to keep pursuers heading directly forwards (or rather directly for said unit) unless a general with initiative is present, in which case the retiring/fleeing unit's zone of influence may be disregarded by friendly troops in his radius/under his command.  Combine this with terrain modifiers to zones of influence (which can also allow careless troops to march past an enemy-held wood while in movement mode, which is not good news if the enemy pops out to attack them) and we may have the start of the basis of a system.  Naturally, a unit in movement mode contacted by an enemy in combat mode will not fight well.

These are a few ideas to get us started; we shall see how concepts progress.

[rodge]

I alway thought that Sulla demonstrated some fairly handy Command and Control at Chaeronea 86 BC:

Plutarch 'Life of Sulla' 19.1
'Archelaüs now extended his right wing to envelop Sulla's line, whereupon Hortensius sent his cohorts against him on a quick run, intending to attack his flank. But Archelaüs wheeled swiftly against him his two thousand horsemen, and Hortensius, forced aside by superior numbers, was keeping close to the hills, separating himself little by little from the main line, and getting surrounded by the enemy.
When Sulla learned of this, he came swiftly to his aid from the right wing, which was not yet engaged. But Archelaüs, guessing the truth from the dust raised by Sulla's troops, gave Hortensius the go-by, and wheeling, set off for the right wing whence Sulla had come, thinking to surprise it without a commander. At the same time Murena also was attacked by Taxiles with his Bronze-shields, so that when shouts were borne to his ears from both places, and re-echoed by the surrounding hills, Sulla halted, and was at a loss to know in which of the two directions he ought to betake himself.
But having decided to resume his own post, he sent Hortensius with four cohorts to help Murena, while he himself, bidding the fifth cohort to follow, hastened to the right wing. This of itself had already engaged Archelaüs on equal terms, but when Sulla appeared, they drove the enemy back at all points, obtained the mastery, and pursued them to the river and Mount Acontium in a headlong flight. '

[dwkay57]

The "send for help" concept from a subordinate is an interesting idea and one I hadn't pondered on till now. My own rules allow for an army general to issue fresh instructions - based on his own knowledge (i.e. visible sight) - to a corps commander but hadn't thought about messages going back up the chain.

Most of the comments so far mention named individuals (apart from the unnamed Tribune) who seem to be commanding sections of an army.  Does this imply that there was generally some structure of command? Collective nouns such as "battles", "wings" and "commands" spring to mind.

In my own rules armies are usually split into corps each of which has a commander who has one of 27 different personality types. The type of personality affects how he reacts depending upon his current situation. Whilst this doesn't permit to a great extent the "intelligent initiative" an army commander might hope for, it does encourage either wild excesses or cautious retreats (usually at the wrong moment).

[Patrick Waterson]

The usual arrangement for a Biblical or classical army seems to have been centre and wings, with the centre in Biblical and wings in classical armies containing the bulk of the mobile troops.

Hence the standard classical arrangement would be: right wing, left wing, centre.  The C-in-C would take one of these and his subordinates the other two.  Below this there would not usually be anyone acting with independent initiative, though exceptions did occur.

Rodger's mention of Sulla at Chaeronea also reminds me of Caesar at Alesia and Dyrrhachium: he and Antony both had a habit of collecting a handful (sometimes quite a large handful) of cohorts and going with it to a looming crisis point, sometimes more than once in the same engagement.  The Romans seem to have developed the cohort as a battlefield manoeuvre unit (individual maniples were a bit too small to pack much punch and a bit too numerous to start collecting and leading all over the field) but only a few commanders (essentially the very talented ones) seem to have made a habit of using them as 'fire brigades' in this fashion.  Then again, it tends to be only the talented commanders whose battles arte described in any detail, so I do not know how far we can go with such an assumption.  I would suggest: if in doubt, do not let run-of-the-mill commanders use initiative this way.

The 'commander personalities' in David's system are nice.  In real life, a good C-in-C would tolerate such excursions a maximum of once before replacing the individual concerned.  A bad C-in-C might not even realise his subordinate was doing anything counterproductive, so it might be an idea to require a good commander to have reliable subordinates (charging extra points accordingly) but allow a bad commander to have any sort of subordinate.  I would suggest one centre commander, who in a classical army usually has most if not all of the infantry, and two wing commanders, who each take a wing of cavalry with perhaps some supporting light and/or fast-moving infantry.

Greek armies seem to have had one effective leader, the C-in-C, but the action sometimes (e.g. Delium in 424 BC) segregated into two or more almost independent struggles, usually victorious right making headway against defeated left.  I cannot offhand think of any such segment exhibiting initiative by itself: clever alterations part-way through the battle seem to have been entirely due to the C-in-C.

All civilised armies exhibited organised subdivision and officers at several levels, but for the most part these seem to have confined their battlefield activities to making sure their units deployed in a straight line, advanced in parallel with their neighbours, charged together at the right time and/or halted when everyone else did.  If one does not have such officers at each level of command one's army is going to be an uncontrollable mob.  The impression I get is that the larger the army, the more effort officers at each level of command put into making the whole thing move homogenously, and the less they put into independent action.

Another point that recurs is that infantry, or at least non-skirmishing infantry, seems to have advanced and manoeuvred as a single entity when possible - a battleline.  Furthermore, default speed of advance appears to have been about half that which would be possible - Late Roman armies in particular are noted as using a deliberate advance in slow time, which has the advantage that anyone encountering a patch of difficult terrain can filter through or round it and reassemble without losing cohesion or getting left behind.  Not everyone did this: Caesar was surprised by the rapid advance of the Nervii and their allies at the Sambre in 57 BC - and again by Pharnaces' Pontic army at Zela ten years later.  We may note, incidentally, that at the 9th and 10th legions formed the left, the 8th and 11th the centre and the 7th and 12th the right, but Caesar does not name commanders for any of these groupings.  We might guess that he gave Antony the left and Labienus the centre, but he does not say so.  That nobody was commanding the right is evident from Caesar's actions once he arrived there.

Biblical armies are probably outside the immediate scope of the discussion, but from what I can determine (and most of this is inference) they seem to have had the C-in-C in the centre, commanding the best troops (which includes the chariotry and its supporting infantry or cavalry, plus missile troops and royal guards) while the wings seem to have been mainly infantry and operated in support of the centre.  Thutmose III, our most famous and extensively written about Biblical era campaigner, seems to have encouraged a collection of generals who in their memoirs mention 'accompanying His Majesty' on his campaigns, pointing to their subordinate role and perhaps pointedly omitting mention of independent action on their part while with His Majesty (although mention of saving His Majesty from the attentions of an elephant was deemed worth recording).  Israelite kings had a general of the army (singular) and 'mighty men' who commanded thousands, but again it seems to have been the king who led the army and fought in the part that mattered, perforce the centre.

This tradition seems to be reflected in Persian (Achaemenid) armies, where the King-of-Kings was invariably in the middle of his army with his best troops, and apparently thought or felt that it would be in this sector that the battle would be decided.  Cavalry were assigned to the flanks, often with a supporting infantry contingent, but the best troops were reserved for the centre and the King, including a mounted royal guard and His Majesty in a chariot.  This I see as a vestige of the days when battles would be decided by a (supported) chariot clash in the centre of the lines - assuming both sides were drawn up for battle rather than one side sneakily hiding behind a city and attacking the other on the march!

Out of interest, David, what are the 27 commander/sub-commander personality types?  It sounds as if you have put some work into this.

[dwkay57]

Patrick,

The 27 "personalities" are based around three characteristics each of which has three values, hence the 27 combinations. In thinking about the characteristics that may affect battlefield competence or activity, I concluded that these were: a degree of decisiveness; strength of leadership; and sense of loyalty. Fairly similar to other rule sets.

Decisiveness, I call "Boldness" and commanders are classified as: Rash, Reliable or Cautious
Leadership is rated as either: Strong; Adequate or Weak
Loyalty is classified as either: Loyal, Respectful or Dubious

Each commander's character can be diced for either at the start of each battle or (as I do) as the army is formed so personalities become long standing banes in their general's side.
Irregular and Barbarian commanders is diced for as a character and then the general can decide which command they are assigned to. This tries to reflect the appoint of friends in suitable places.
Regular commanders have the dice rolled for a command and then the general decides which dice applies to which characteristic. This tries to simulate some form of organised promotion avoiding the extremes.

You can  of course vary this. For instance my Persian Xerxes rates loyalty very highly, so he is allowed to up the loyalty score by decreasing one of the others.

On the battlefield, their "personality" affects how the interpret the situation. The situation is defined by the same set of factors which give seven statii ranging from "excellent" to "dire". Then by using a matrix of 27 rows (one for each commander type) and 7 columns (one for each situation) I can determine how a commander will react. So for instance:
- If the situation is "excellent" then a rash/strong/loyal commander will order a general advance, whilst a reliable/adequate/respectful commander will follow orders, and a cautious/weak/dubious (think BlackAdder II) will initiate charges.
- If the situation is "dire" then those three commanders would hold position, withdraw and retreat respectively.
Of course the troops under their command might (and do) have other opinions.........

As corps commanders tend to get killed, I've had to create a mechanism for generating sub-ordinates on the fly, but that is probably too much detail for now.

[Patrick Waterson]

That seems to be a worthwhile system.  I would suggest that Republican Roman commanders usually be diced for just before a battle (simulating the unpredictability of the consular election process) and others treated pretty much as you mention, perhaps allowing players of armies that historically started out with several successful campaigns under their belt (e.g. Alexandrian Macedonian, Carthaginian led by Hannibal or Roman led by Caesar) to reroll egregious cases on the basis that they would mostly have been weeded out in earlier campaigning.

One historical quirk perhaps to consider is that if a commander had clear orders for a given situation he tended to follow those orders irrespective of his personality or inclinations.  His particular personal proclivities tended to come to the fore in unexpected situations.  Hence a tribune might carry on in reliable but undistinguished fashion through several campaigns until one day when he sees an inviting enemy flank and rear off to his legion's eight-o-clock position.  Then he looks at his chart ... will he carry on stumping along following his hastati or will he seize the opportunity and the initiative with both hands and bring his principes and triarii down upon the unsuspecting enemy rear?

These are just thoughts.  Carry on with the good work as you see fit.

[Erpingham]

The 27 "personalities" are based around three characteristics each of which has three values, hence the 27 combinations. In thinking about the characteristics that may affect battlefield competence or activity, I concluded that these were: a degree of decisiveness; strength of leadership; and sense of loyalty. Fairly similar to other rule sets.

Reading this, it reminds me of Tony Bath's Hyboria-inspired campaign rules (this is a good thing - fond memories).  Sadly, I think it is a little too complex for my current tastes.  But it does seem to lend itself to the creation of "artificial intellegence" in the other side of a solo game.  Do you use it this way, David?

[dwkay57]

Anthony,

Yes I do use it in solo games and I do have a copy of Tony Bath's "Setting Up a War games campaign" on my bookshelf. I did re-read it for inspiration but couldn't figure out how the characteristics he suggests could be fitted in to tactical reaction. Any suggestions or ideas for those or other characteristics that might be suitable?

I've just checked the data sheet in the rules and about 47% of the 27 x 7 reactions result in "Follow Orders". So it sounds like I'm on the right lines.

[Patrick Waterson]

On the general subject of command and control, we might also examine it from the ground up.  The classic building block of classical armies was the file, an 8-16 man follow-my-leader subunit.  If we look at Hellenistic armies (because they are easy and we have Asclepidotus' tactical manual http://penelope.uchicago.edu/Thayer/E/Roman/Texts/Asclepiodotus/2*.html), a file (stoichein) consists of 16 men under a lochagos; paired with another such it becomes a double file (dilochia) of 32 men under a dilochites; two of these form a 64-man tetrarchy (tetrarchia) led by a tetrarch (tetrarches).

The 64-man tetrarchy is the basic subunit that builds up to a 128-man taxis; two taxeis form a 256-man syntagma or syntaxiarchia.  The syntagma/syntaxiarchia is the basic field unit which adds a number of 'ektakoi', men outside the ranks, namely:

"... an army-herald, a signalman, a bugler, an aide, and a file-closer. The first was to pass on the command by a spoken order, the second by a signal, in case the order could not be heard because of the uproar, the third by the bugle, whenever the signal could not be seen for the dust; the aide was there to fetch whatever was needed, while the supernumerary file-closer was to bring up the straggler to his position in the company" - Asclepiodotus, Tactics, II.9

Exactly where the signalman would have to stand in order to be seen by the unit leaders, or someone who could inform the unit leaders, is an interesting question.

This file-closer is not to be confused with the ouragos who was the last man in each file and who was meant to keep the rear end of the file moving up behind the front end.  The supernumerary file-closer was to catch and return to position any man who dropped out despite the best efforts of the ouragos at the end of each file.

From this we can conclude that a signal (advance, halt, retire) would be repeated along the line by the 'herald', signalman and/or bugler of each syntagma.  The command organisation involved the files being led from the front, so the officer in charge of the syntagma is not in proximity to the signallers - from this we can deduce that he was not expected to issue commands except to his own unit, along a 16-man frontage.  The signallers are there to repeat signals from the C-in-C or the sub-commander in charge of the phalanx.

From anecdotal evidence it would appear the Romans used a similar system, based on the buccina (military trumpet).  One observes this to be not dissimilar in size to a Celtic war-horn, but cunningly bent for easy portability.  http://en.wikipedia.org/wiki/File:Buccina_1.png  Lacking a 'herald' would mean that detailed or unusual orders could not be conveyed, but the Romans relied on standardisation and perhaps if unusual orders were required the six tribunes assigned to a legion could undertake the necessary verbal exercise.

[Erpingham]

From anecdotal evidence it would appear the Romans used a similar system, based on the buccina (military trumpet). 

Taking our lead from Byzantine manuals, we could suggest that shouted words of command were also commonplace as in more modern armies.