Arrian’s Array Against the Alans

The republican Roman army page

The late Roman army page

The late republican Roman army page

Arrianou ektaxis kata Alanoon.

Hègeisthai men tès pasès stratias tous kataskopous hippeas epi duoin tetagmenous syn tooi oikeiooi hègemoni. Epi toutois de tous hippotoxotas tous Petraious, kai toutous epi duoin: agontoon de autous hoi dekadarchai. Epi de toutois epitetachthoon hoi apo tès eilès hèntini Aurianoi onoma. Syntetachthoon de autois hoi tès speirès tès tetartès toon Rhaitoon, hès archoon Daphnès Korinthios. Epi toutois de hoi apo tès eilès hei onoma Koloonoi. Syntetachthoon de autois Ityraioi kai Kyrènaioi kai hoi apo tès prootès Rhaitikès. Sympantoon de toutoon archetoo Dèmètrios. Epi toutois de hoi Keltoi hippeis, kai autoi epi duo, kai toutoon hègeisthoo hekatontarchos, hoosper epi stratopedou.

Pezoi de epi toutois tetachthoon, ta sèmeia anatetamena pro sphoon echontes, hoi te Italoi kai Kyrènaioon hoi parontes. Pantoon de hègeisthoo Poulcher, hoosper archei tois Italois. Bosporanoi de epi toutois pezoi iontoon hègemona parechomenoi Lamproklea, kai hoi Nomades epi toutois hypotetagmenoi Bèrooi tooi spheterooi archonti. Hè taxis de estoo eis hoplitas tessaras. Autoon de toutoon hosoi toxotai hègeisthoon. Tas de pleuras tès taxeoos phylattontoon hekateroothen hoi hippeis hoi oikeioi. Epi de toutois hoi epilektoi hippeis itoosan, kai epi toutois hoi apo tès phalangos hippeis, epeita hoi katapeltai, epeita to sèmeion tès pentekaidekatès phalangos, kai amph’ autooi de hègemoon tès phalangos Oualès, kai ho hyparchos , kai hoi chiliarchoi hois tetaktai, kai hekatontarchoi hoi tès prootès speirès epistatai. Pro de tou sèmeion au toon pezoon hoi akontistai tetachthoon. Autous de tous pezous epi tessaroon tetagmenous ienai. Epi de tèi pentekaidekatèi phalangi tetachthoo to sèmeion tès doodekatès phalangos, kai chiliarchai amph’ autooi kai hekatontarchai kai (pente) hekatontarchoi hoi tès prootès speirès, pro de tou sèmeiou au toon pezoon hoi akontistai tetachthoon. Autous de tous pezous epi tessaroon tetagmenous ienai. Epi de tèi pentekaidekatèi phalangi tetachthoo to sèmeion tès doodekatès phalangos, kai chiliarchai amph' autooi kai hekatontarchai. Epi tessaroon hoosautoos kai hède hè phalanx itoo tetagmenè.

Epi de tooi hoplitikooi tetachthoo to symmachikon, hoi te apo tès mikras Armenias kai Trapezountioon hoi hoplitai kai Kolchoi kai Rhizianoi hoi lonchophoroi. Epitetachthoon de autois hoi Aplanoi pezoi. Pantos de tou symmachikou hègemoon estoo Sekoundinos, hosper toon Aplanoon hègeitai. Epi toutois de ta skeuophora hepesthoo. Ourageitoo de hè eilè toon Rhetoon kai ho tautès eilarchès.

Tas pleuras de tou pezikou kosmountoon men kai hekatontarchai kai epi tooide tetagmenoi, phylakès de heneka hè eilè hè Galatikè parippeuetoo epi hena stoichon hekateroothen tetagmenè, kai hoi toon Italoon hippeis. Ho de eiarchès autoon epiphoitatoo tais pleurais.

To de hègemoon tès pasès stratias Xenophoon to poly men pro toon sèmeioon toon pezikoon hègeisthoo, epiphoitatoo de pasèi tèi taxei, kai episkopeotoo hopoos tetagmenoi iasin, kai tous men ataktountas eis taxin epanagetoo, tous de en kosmooi iontas epainetoo.

Hotoo men tetagmenous ienai. Elthontos de eis to apodedeigmenon choorion, tèn men hippon pasan peristènaien kyklooi eis tetragoonon taxin, tous kataskopous de pempsai eis ta hyperdexia chooria aposkopès heneka toon polemioon. Entautha de apo sèmeiou hoplizesthai sigèi, hoplisamenous de eis taxin kathistasthai. Hè taxis de esthoo hède. To men keras hekateron toon pezoon ta hyperdexia echetoo toon choorioon, hoti en toiooide hè ektaxis estai. Epitetachthoon de tooi kerati tooi men dexiooi hoi amphi Ouasakèn kai Arbèlon Armenioi, to to hyperdexiootaton tou keroos echontes, hoti toxotai eisin hoi sympantes. Protetachthoon de autoon hoi tès speirès tès Italikès pezoi. Pantoon de heisthoo Poulcher, hostis kai tès speirès tès Italikès archei. Kai toutooi epitetraphthoo autos te kai ho Ouasakès kai Arbèlos kai to ekeinoon hippikon te kai pezikon.

Epi de tou aristerou tetachthoon kai houtoi to hyperdexiootaton tou keroos echontes hoi te apo tès mikras Armenias symmachoi kai hoi Trapezountioon gymnetes kai hoi Rhizianoi lonchophoroi. Protetachthoon de kai toutoon hoi Aplanoi hoi diakosioi kai Kyrènaioon hoi hekaton, hoos probolèn men einai pro toon akontistoon tous hoplitas, tous de hyperakontizein ek tou hyperdexiou. To de en mesooi sumpan to men dexion echetoo hè pezikè phalanx hè pentekaidekatè hyper to meson tou pantos chooriou, hoti poly pleiones plèthei houtoi eisi: to de hypoloipon tou aristerou ekplèrountoon hoi tès doodekatès phalangos pezoi este epi to akron tou keroos tou aristerou. Tetachthoon de epi oktoo, kai pyknè autois estoo hè xyntaxis. Kai hai men prootai tessares taxeis estoosan kontophoroon, hois dè tois kontois makra kai epi lepton ta sidèria proèktai. Kai toutous hoi men prootostatai eis probolèn echontoon, hoos ei pelazoien autois hoi polemioi, kata ta stèthè malista toon hippoon tithesthai toon kontoon ton sidèron: hoi deuterostatai de kai hoi tès tritès kai tetartès taxeoos eis akontismon probeblèsthoon tous kontous hopou tychoien, kai hippous troosontes kai hippotèn katakanountes kai thyreooi kataphraktooi thooraki empagentos tou kontou kai dia malakotèta tou sidèrou epikamphthentos archeion ton anabatèn poièsontes. Hai de ephexès taxeis toon lonchophoroon estoosan. Enatè de epi toutois estoo taxis hoi pezoi toxotai, hoi toon Nomadoon kai Kyrènaioon kai Bosporianoon te kai Ityraioon. Mèchanai de ephestèketoosan tooi kerati hekaterooi, hoos porrootatoo prosiontoon toon polemionn exakontizein, kai katopin tès pasès phalangos.

To de hippikon xympan kata eilas kai lochous oktoo xyntetagmenon ephestatoo tois pezois, to men tois kerasin hekaterois, probolèn echon tous hoplitas pro sphoon kai tous toxotas, lochoi duo, to de tèi mesèi phalanggi, lochoi hex -- mènmion. Toutoon de hosoi men hippotoxotai plèsion tès phalangos ephestèketoosan, hoos hypertoxeuein hyper autès: hosoi de lonchophoroi è kontophoroi e machairophoroi è pelkophoroi eis ta plagia te hekateroothen hormoontoon kai to xynthèma prosmenontoon, hoi de epilektoi hippeis amph' auton Xenophoonta estoosan, kai toon apo tès phalangos pezoon hoson eis diakosious, hoi soomatophylakes, kai hekatontarchai hosoi tois epilektois xyntetagmenoi è toon soomatophylakoon hègemones, kai dekarchai hoi toon epilektoon. Estoosan de amph' auton as hekaton kouphoon lonchophoroon, hoos pasan epiphoitoon tèn phalanga hypou ti endees katamanthanoi, ekeino iooito kai therapeuoi. Hègeisthoo de ttou men dexiou keroos pantos xyn tooi hippikooi Oualès, hosper kai tès pentekaidekatès phalangos hègemoon estin: tou de aristerou hoi chiliarchai tès doodekatès.

Outoo de tachthentoon sigè estoo est' an pelasoosi entos belous hoi polemioi: pelazontoon de èdè hoos megiston kai phoberootaton alalazein xympantas tooi Evyaliooi, kai belè te apo mèchanoon kai lithous aphiesthai kai belè apo toxoon, kai lonchas hoi lonchophoroi akontizetoosan hoi te psiloi kai hoi thyreophoroi. Pheresthoosan de kai lithoi eis tous polemious apo tou Xymmachikou ek toon hyperdexioon, kai to pan akrobolismos estoo pantachothen hoos eni pyknotatos eis taraxin te toon hippoon kai olethron toon polemioon hippoon. Kai elpis men hypo tou adiègètou plèthous toon beloon mède pleion pelasein tèi pezikèi phalangi epelaunonteas tous Skythas: ei de dè pelazoien, enchrempsantas tais aspisi kai tois oomois antereisantas dechesthai tèn prosbolèn hoos karterootata kai tèi synkleisei pyknotatè tas prootas treis taxeis xynereidousas sphisin hoos biaiotaton hoion te: tèn tetartèn de hyperakontizein tas lonchas: kai tèn prootèn paiein è akontizein tois kontois apheidoos es te hippous kai autous. Apoosthentoon de ei men phygè lampra genètai, diachoorein dè tas pezikas taxeis kai epelaunein tas hippeas, mè pantas tous lochous alla tous hèmiseas: tetachthai de prootous hoitines kai prootoi epelasousin, tous de allous hèmiseas hepesthai men tois epelaunousin, en taxei de kai mè pantelei tèi diooxei chroomenous, hoos ei men phygè kartera katechoi, ekdexasthai tèn prootèn diooxin akmètois tois hippois, ei de tis epistrophè katalambanoi, epitithesthai tois epistrephousin. Homou de hoi te Armenioi toxotai epelaunontes toxeuontoon, hoos mè paraschein anastrophèn tois pheugousi, kai hoi lonchophoroi hoi gymnètes dromooi hepesthoosan: menein de mède tèn pezikèn taxin en choorai eti, alla prochoorein thatton è badèn, hoos ei ti karterooteron apantooiè apo toon polemioon, authis einai probolèn pro toon hippeoon.

Tade men ginesthai, ei apo tès prootès prosbolès phygè kataschoi tous enantious: ei de epistraphentes es kyklous hyper ta kerata parelaunein etheloien, anateinesthai men ta hyperdexiootera eti ta kerata autès tès psilès toxeias: hoos ou domikazoo mèpote asthenè tèi anatasei ta kerata gigomena idontes di' autoon oosainto kai diakopsaien to pezikon. Hyperballontoon de ta keratahekatera è hopoteron oun, pasa èdè anankè plagious men autois ginesthai tous hippous, plagious de tous kontous. Entautha dè emballontoon es autous hoi hippeis, mè akontismooi eti alla tais spathais autois sympheromenoi, hoi de tois pelekesin, hoi de Skythai gymnoi te ontes kai tous hippous gymnous echontes.

Translation

At the van of the entire army should be mounted scouts deployed in two contingents with their own commander. Behind them the Petrean horse archers, these also in two divisions: the decurions should lead them. At their rear should be deployed those from the wing of the Auriani. Those from the Fourth regiment of Rhaetians should be stationed with them, led by Daphne the Corinthian. Behind them those from the wing named Colonists. Stationed alongside them should be the Ituraeans and Cyrenaecans and those from the First Rhaetian. Demetrius should command all of these combined. Behind them the German cavalrymen, these too in two divisions, and a centurion should command them, the one in charge of the camp.

Behind these the infantrymen should be deployed, their standards carried before them, the Italians and those present of the Cyrenaecans. Pulcher, who is in charge of the Italians, should command them all. Behind them should be the Bosporan foot soldiers, commanded by their leader Lamprocles, and at their rear the Numidians under their commander Verus. The formation should be four soldiers wide. The attached archers should be at the front of their own units. The cavalrymen organic to the units should guard both flanks of the formation. Behind them should come the guard cavalrymen, and behind them the legionary horsemen, then the catapults, then the standard of the Fifteenth Legion, and with it the commander of the Fifteenth Legion, Valens, and the subordinate commander, and the tribunes and the centurions of the first cohort. In front of the standard the infantry javelineers should be deployed. These foot soldiers should be drawn up in fours. Behind the Fifteenth Legion the standard of the Twelfth should be deployed and the tribunes and centurions around it. This legion should also be drawn up in fours.

Behind the heavy troops the allied force should be formed up, the heavy armed from Little Armenia, and Trabzon and the Colchian and Rhizian javelinmen. Behind them the Apulian infantrymen should be deployed. Secundinus, who is in command of the Apulians, should lead the allied force as a whole. Behind them should be the baggage train. The wing of the Dacians and their wing commander should act as rearguard.

Centurions who are selected for this particular task should keep the flanks of the infantry in order. The Galatian wing should ride along both flanks in a single file as a guard, and the horsemen from the Italians as well. Their wing commander should ride along the flanks.

The overall army commander, Xenophoon, should lead most of the time in front of the infantry standards, should ride along the entire marching formation, and he should see to it that they march in formation, and place those in disorder back into the formation, and he should commend those that are in proper order.

This should be the marching order. On arrival at the planned spot, the entire cavalry should circle around to form a square, scouts must be sent to overwatch positions to look for the enemies. Next they should arm themselves, and after kitting out take up position in the battle formation. The battle array should be the following. Each wing of the infantry should hold high ground, as the full deployment must be in this order. On the right flank should be deployed the Armenians with Vasakes and Arbelos, holding the highest part of the flank, because they are archers all. Positioned in front of them must be the infantrymen of the Italian regiment. In command of all these should be Pulcher, who is in charge of the Italian regiment. Both Vasakes and Arbelos with their cavalry and infantry should be arranged in support of him.

On the left flank the Allies from Lesser Armenia an the light armed from Trapezus and the Rhizian javelinmen should be arrayed holding the highest part of the flank. Deployed to their front must be the two hundred Apulians and a hundred of the Cyrenaicans, in order that the heavy armed are a bulwark for the javelineers, they can hurl their javelins overhead from the high ground. The Fifteenth Legion’s infantry should hold the entire right center above the middle of the whole area, because they are by far the most numerous: the infantrymen of the Twelfth Legion should hold the remaining space on the left filling it up to the point of the left flank. They should deploy in eight ranks and their deployment should be close ordered. And the front four ranks of the formation must be of spearmen, whose spearpoints end in thin iron shanks. And the foremost of them should hold them at the ready, in order that when the enemies near them, they can thrust the ironpoints of the spears at the breast of the horses in particular. Those standing in second, third an fourth rank of the formation must hold their spears ready for thrusting if possible, wounding the horses and killing the horsemen and put the rider out of action with the spear stuck in their heavy body armour and the iron point bent because of the softness. The following ranks should be of the javelineers. The ninth rank behind them should be the foot archers, those of the Numidians, Cyrenaicans, Bosporans and Ityraeans. Artillery pieces must be deployed on each flank to fire at the advancing enemies at maximum range, and behind the whole battle formation.

The entire cavalry arrayed together in eight wings and squadrons must stand next to the infantrymen on both flanks, having the heavy armed and archers as a screen, two companies and in the middle of the formation six companies [ Gap in text ] the horse archers among them must form close to the battle line in order to shoot over it. The javelineers, spearmen, swordsmen and axe-men must guard both flanks and await the signal. The picked cavalrymen should bewith Xenophoon himself, and two hundred infantrymen from the legion, the bodyguards, and the centurions attached to the picked cavalrymen and the commanders of the bodyguards and the decurions of the elite soldiers. There should be with him [ gap in text ] a hundred light javelineers in order that while riding down the entire formation he can note and amend things if something is lacking. In charge of the entire right flank along with the cavalry must be Valens, who is commander of the Fifteenth Legion. The tribunes of the Twelfth should do so on the left.

Once thus arrayed there should be silence until the enemies come within missile range; when in range the loudest and most intimidating war cry must be raised by the whole lot, and bolts and stones must be fired from the artillery pieces and arrows from the bows, and javelins by both light armed and shield bearing javelinmen. Stones must also be thrown at the enemies by the allied force on the overwatch position, and the whole missile rain must be coming from all sides to make it concentrated enough to panick the horses and destroy the enemies. And the expectation is that the Scythians will not get close to the infantry battle formation because of the tremendous weight of missiles. If they do close in though, the first three ranks should lock their shields and press their shoulders and receive the charge as strongly as possible in the most closely ordered formation bound together in the strongest manner. The fourth rank will throw their javelins overhead and the first rank will stab at them and their horses with their spears without pause. After repulsing the enemy if there’s a clear rout, the infantry units must clear lanes and the horsemen should advance, not all squadrons, but only half of them. Those to the fore must be the first to advance. The other half should follow those that advance, in perfect formation and not in hot pursuit in order that they may continue the initial pursuit with fresh horses in case there is a complete rout, and in case they turn about to attack, they may assist those in pursuit. At the same time the Armenian archers must advance shooting their bows in order to prevent those in flight from turning about, and the light armed javelineers should advance at the run. The infantry formation should not hold its ground, but should advance at faster than the normal step in order to be a base of defence for the cavalrymen if there is stronger resistance by the enemies.

These things should happen if the are put to flight after the first charge. If they about-face and circle around the flanks, the flanking bodies of lightly armed archers should extend formation to the high ground. I do not think that seeing that the flanks become weaker through extension in this manner they will break through them and cut up the infantry. Should they overcome one or either of the flanks the horses would necessarily expose their flanks, their spears at a right angle. In that case the cavalrymen must attack not with a missile shower but with swords and and axes. The Scythians being lightly armed and having unprotected horses [text breaks off ]

Comments

The vanguard

The mounted scouts would be exploratores drawn from the ranks of both the legionary and auxiliary cavalry. The speculatores of the legion, who had once been used as spies, were by this date no longer used for reconnaissance duties, but rather as executioners and bodyguards of senior commanders. The Greek term Keltos was confusingly used to designate a German, the Celts themselves being called Galatoi. The ethnic designations of the auxiliary troops generally do not reflect the actual origins of their soldiers. Though auxiliary regiments were ethnically rather homogenous at the time of their original formation, the Roman army did not take any trouble to continue recruitment of replacements from the same background. The hekatontarchès epi tou stratopedou may be a reference to a legionary praefectus castrorum, although it may be construed as legionary centurion. This last interpretation appears less likely though, as Arrianus prefers the term phalanx to stratopedon as the Greek translation for the Latin legio, though this is parallelled in Josephus.

M. Pavkovic, 'Singulares legati legionis: guards of a legionary legate or a provincial governor?’ in: Zeitschrift fuer Papyrologie und Epigraphik 103 (1994), 223-228.
M.P. Speidel, ‘The captor of Decebalus’ in: Journal of Roman Studies 60 (1970), 142-153.

The marching order

The standards were probably carried before the unit as a whole rather than in front of their respective companies. This seems more likely as a similar arrangement is described in the marching order in Josephus and because standard bearers are also grouped together in the marching scenes on Trajan’s column. These auxiliary cohortes equitatae appear to have organic infantry archers as well as heavy infantrymen in addition to the organic cavalrymen. The incorporation of differently armed troops within the same auxilary cohort is confirmed by the description of the battel formation where there appear both hoplitai and toxotai of the Cyrenaecans. The guard cavalrymen are probably the equites singulares of the governor's guard. The marching formation of four soldiers wide differs from that described in Josephus, who mentions the infantry as marching six abreast. The breadth of the marching order was probably mainly guided by that of the available roads or paths, though other considerations such as troop strength may also have played a part. The number of four is exactly half that of the main battle deployment and may represent the army marching in half-files. There is no compelling reason to suspect different drill systems are in use because Arrianus reports multiples of four rather than three as described in the writings of Josephus, Polybius and Vegetius. In the marching order of Josephus the subunits were not at full strength with ten men from each centuria detached for pioneer duty and others serving as bodyguards. Arrianus apparently has not detached so great a number of pioneers from their units. Also the low number of legionary foot guards mentioned later on, some two hundred rather than the approximately four hundred recorded for legio II Traiana in Diocletian’s reign, may mean that the elite legionaries were marching with their own subunits rather than detached for special duty. In the case of the legio XV the primi ordines, the centurions of the first cohort, are descibed as being grouped together with the other senior officers such as the tribuni. The other centurions would have marched or ridden along their respective companies to keep their formation in order as described later on. The fact that the soldiers of the first cohort apparently did not need such close supervision by their centurions may be interpreted as further proof of their elite status, needing fewer NCO’s and officers to keep a number of soldiers in check than the other cohorts of the legion. The number of the primi ordines is unfortunately not given in the original text, preventing the identification of the legiones involved in this campaign as being organised with a five double strength centuriae first cohort or having the old type organisation of six normal strength subunits.

Flavius Josephus, Bellum Judaicum 3.115-124; 5.47-48.
A.K. Goldsworthy, The Roman army at war 100 BC-AD 200 (Oxford 1996), 180-181.
M.P. Speidel, The guards of the Roman army (Bonn 1978), 46.
E.L. Wheeler, ‘The legion as a phalanx’ in: Chiron 9 (1979), 312-313.

The allied troops

The Roman army commonly used irregular troops drawn from allied states as a supplement to its regular forces. The term of numeri for these irregular auxiliary troops current in modern literature is inaccurate. Units of all types could be termed a numerus and it carried no special meaning as force of irregulars. The place of the allied forces is similar to that described in Josephus with regular auxiliaries deployed as a rear guard.

Flavius Josephus, Bellum Judaicum 3.126; 5.49.
M.P. Speidel, 'The rise of ethnic units in the Roman imperial army' in: ANRW II-3 (1975), 202-208.

The flank guards

The use of centurions to keep the marching infantry in order is matched in the writings of Flavius Josephus. The formulation in this passage is intriguing. If centurions were specially selected for the task it apears to imply that not all centurions were involved. As mentioned above the primi ordines were evidently not selected, but there might also be others. The importance of the mounted flank guards is emphasized by the fact that the praefectus is accompanying his men rather than marching separately from his command.

Flavius Josephus, Bellum Judaicum 3.124.

The commander-in-chief

Arrianus had a great admiration for Xenophoon, the fourth century BC military commander and historian. Their primary interests coincided and both combined the functions as practising military men, historians and writers of tactical treatises. Many of Arrianus literary works share the same types of subject, the Technè taktikè for instance mirroring Xenophoon’s Hipparchikos. The use of his predecessor’s name instead of his own might seem a little bit eccentric to the modern reader, but also features in other writings of the author, including private correspondence. The army’s commander takes up a similar position as in the marching orders described by Flavius Josephus, though the rounds made by the commander-in-chief are lacking in the Bellum Judaicum. Arrianus exemplifies here an important function of the Roman commander, closely supervising his subordinates and keeping an eye on their individual performance.

Flavius Josephus, Bellum Judaicum 3.119; 5.47.
A.B. Bosworth, ‘Arrian and the Alani’ in: Harvard Studies in Classical Philology 81 (1977), 248.
A.K. Goldsworthy, The Roman army at war 100 BC-AD 200 (Oxford 1996), 156-163.

The auxiliary battle line

Archers were generally used in the Roman army for fire support overshooting ranks of heavy infantry. Arrangements like this can also be found in Tacitus and Flavius Jospehus. Their purpose was mainly to shower an area with missiles than to engage point targets. As with suppressive fire by squad machine guns in modern day combat only a fraction of the ammunition expended would hit an enemy. The archery would serve to harass opponents and generally cause wounds rather than fatalities. The lack of a significant kill ratio does not mean that this was an unimportant or senseless manner of engaging the enemy. It was generally through collapse of morale rather than butchery pure and simple that battles were decided and the effect of archery on enemy morale ensured its place in the fighting methods of the imperial army. The Roman army had a large number of archers incorporated in its structure, both as specilalist units of sagittarii as well as soldiers trained as bowmen in other formations. Even the legions of the imperial period had archers among its complement as attested by epigraphy. The depiction of auxiliary infantry as light skirmishers current in modern literature of the Roman army is a distortion of their role in the source material. The auxiliary foot soldiers were mainly heavy infantrymen that were equipped and trained to fight in manner comparable, if not identical, to that of the bulk of the legionary infantry. This is borne out by battle descriptions and by their designation as hoplitai, heavy infantrymen, in both Josephus and Arrianus. Heavy infantrymen of the auxilia are deployed here as an extension of the legionary heavy infantry line, similarly protecting the ranks of archers arrayed at their rear.

Gilliver, C.M., 'Mons Graupius and the role of auxiliaries in battle' in: Greece and Rome 43 (1996), 54-67 (contra).
A.K. Goldsworthy, The Roman army at war 100 BC –AD 200 (Oxford 1996), (auxiliary infantry battle role) 20-21 (archers) 177; 180; 184-190.

The legionary battle line

The kontos described here appears to be the pilum, the classic heavy javelin of the legionary heavy infantry with its bendable iron shank, rather than the two handed cavalry spear used by the heavy cavalry contarii. A generally applied Greek vocabulary for translation of Latin terminology does not seem to have been adopted despite centuries of contact with the Roman army. The word hyssos was used by several Greek authors, most notably Polybius, as the Greek term for pilum, while Flavius Josephus employed the word xyston. The use of the pilum as a stabbing weapon rather than a javelin is not unique: Caesar’s legionaries had done the same during the siege of Alesia according to the De bello Gallico. The description of the main battle formation lacks some details that could shed light on the exact dispositions of the subunits. It remains unclear whether the eight ranks of the legionary troops belong to the same unit, perhaps representing a contubernium deployed in single file, or that the two distinct groups of four diffently armed legionaries represented different subunits of the legion with the front four ranks perhaps belonging to the centuriae of the priores and the rear four ranks to those of the posteriores. Alternatively the front four ranks could perhaps have been those of cohortes I to V and the rear four ranks those of cohortes VI to X. The combination of both pila and lanceae as weapons used by legionaries is not unique to Arrianus. It is attested by Polyaenus for the army of Caesar, in Tacitus for a battle between legio III Gallica and Sarmatians in AD 69, in Josephus in his description of the Roman army in Judea and in Lucianus for the army of Cappadocia some thirty years after the governorship of Arrianus. These two different types of javelins were apparently a typical combination of weaponry for Roman legionaries.

Speidel’s identification of the four rear ranks of lonchophoroi as specialised legionary light infantry may be challenged by the fact that other lonchophoroi, probably elite legionaries, are designated as kouphoi lonchophoroi indicating that the others may have been heavy infantry rather than specialist skirmishers. The distinction made between psiloi and thyreophoroi among the lonchophoroi later on makes clear that not all of the javelinmen were skirmishers. The rear rank lonchophoroi are surely identical with these thyreophoroi and thus heavy infantry troops. It is not clear whether javelineers of both types would carry multiple throwing spears. Later sources indicate that lancearii could carry up to five javelins at a time. Another point of interest is whether the four rear ranks of lonchophoroi would be equipped with a different type of shield like the lonchophoroi in the general's guard.

The composite nature of the battle line portrayed here may have been inspired by the commander’s acquaintance with the details of the army organisation of Alexander the Great during the final period of his reign. For his projected Arabian campaign the army of Alexander was to have adopted a composite battle formation made up of ranks of pikemen, javelineers and archers, a type of deployment also employed by earlier Achaemenid armies as described in the Cyropaidia of Xenophoon, the great role model for Arrianus. There was little if any formal training in military theory for Roman commanders. There were no military academies to train professional officers and the paramilitary iuventus organisations for young members of the Roman elite primarily taught riding skills and individual weapons handling. Writings on military history and tactical treatises thus provided a guide for Roman commanders who did not have the modern benefits of field manuals detailing official army doctrine. The single battle formation of differently armed ranks was there to stay. Similar formations continued to be employed by the Roman and Byzantine armies for centuries to come. The defensive formation described here is very similar to a description of the Byzantine fulco or foulkon of the sixth century AD. Since there are various additional similarities between Arrian's Ektaxis and Maurice's Strategikon the tactics employed in this battle plan may have been more common than assumed and could have conformed to standard Roman tactical dispositions.

The depth of the battle line and the number of ranks was probably dependent on the strength of the subunits and the numbers recorded of three, six, four and eight may reflect deployment in half-files and files, the latter possibly being identical with the contubernia. The legionary heavy infantry centuriae of the hastati and principes in the days of Polybius counted sixty men as their normal establishment strength, those of the imperial army are supposed to have generally had an authorised strength of eighty troops. The multiples of three in Josephus may represent units that were understrength compared to their normal table of organisation. Ranks deployed for battle are described by Polybius and Vegetius as maintaining a distance of six foot. This interval was no doubt necessary for troops to hurl their javelins without wounding their comrades at their rear, but presumably the ranks would close in after spending their missiles for a much denser formation before physical contact with enemy formations. This certainly seems to be the case with the four front ranks of Arrian’s battle order as described later on. The rear ranks may have retained a greater distance to enable them to hurl more missiles. The primary function of deep battle lines was to preserve the morale of the troops and blocking ways of retreat of potential shirkers. The greater depth of Arrian’s infantry line appears to be due to the terrifying nature of the cavalry charge. Roman troops displayed more confidence in infantry against infantry engagements where shallow formations of only three ranks are attested.

The battle line described here has been seen as a Roman variation on the Hellenistic phalanx formation. However there is a very important difference between Arrian’s battle deployment and the Macedonian type phalanx. The troops of the first ranks were not armed with dedicated thrusting spears. As already remarked upon above the kontos is here not the same weapon employed by the heavy cavalry lancers, but should be interpreted as the standard heavy javelin with its bending metal shank, the pilum. Arrian even explicitly mentions the use of these kontoi as throwing weapons as well as stabbing spears, a feat that would certainly not have been possible with heavy pikes. That the term kontos was not a clearly defined technical term for a specific type of weapon can be confirmed by other sources on the Roman army. Josephus for instance also uses the word for the shorter stabbing spear used by shieldbearing Roman cavalry rather than the two handed heavy cavalry pike, though Arrian uses the word in that latter sense in the Technè Taktikè. Another major departure from the normal Helllenistic phalanx was that the battle-line was supposed to clear lanes for the cavalry to pass through after the enemy would have been repulsed. This supposes a great deal more flexibility of the Roman infantry formation than the earlier ponderous formation of pikemen. Unfortunately no specifics are given how this manoeuvre was to be executed. A recreation of a true phalanx formation was apparently adopted by the Roman army for campaigning in the East in the next century, a fact recorded in historians of the reign of Caracalla and Alexander Severus and given greater credibility by the discovery of the gravestone of a discens phalangarii of legio II Parthica at Apamea. These troops are described as using similia arma, including a dory makron or long spear, in an effort to authentically recreate the Macedonian phlanx of days long gone by. With a reported strength of six legions this must have been one of history's first and largest reenactment groups. Nevertheless it does not appear right to identify the kontos carrying kontophoroi of Arrian with the third century phalangarii armed with the dory makron.

Flavius Arrianus, Anabasis, 7.23.3-4.
Flavius Josephus, Bellum Judaicum, 3.93-95.
Iulius Caesar, Bellum Gallicum, 7.88.
Polyaenus, Strategemata, 8.23.25.
Polybius, Historiae.
Tacitus, Historiae 1.79.4; 2.29.1; 2.36.1; 3.27.3.
Vegetius, Epitoma rei miltaris, 3.15.
A.B. Bosworth, ‘Arrian and the Alani’in Harvard Studies in Classical Philology 81 (1977), 237-242.
B. Campbell, ‘Teach yourself to be a general’in: Journal of Roman Studies 77 (1987), 13-29.
G.T. Dennis, Maurice's Strategikon (Pennsylvania 1984),133-135; 145-151.
A.K. Goldsworthy, The Roman army at war 100 BC – 200 AD (Oxford 1996), .
M.P. Speidel, The framework of an imperial legion (Cardiff 1992), 14-21.
E.L. Wheeler, ‘The legion as a phalanx’in Chiron 9 (1979), 303-318.

Cavalrymen and elite troopers

Differently armed horsemen are deployed at the flanks of the infantry formation and may represent specialist troops of the individual cavalry regiments rather than each representing an entire ala with specialised weapons skills. Roman auxiliary cavalry regiments had some of their men trained in specialist fighting techniques as horse archers or shock cavalry lancers, though a number of specialist regiments were also included in its order of battle. It is unclear whether the small legionary cavalry units may have had such variation. The Notitia Dignitatum lists a unit of equites promoti clibanarii. While equites promoti are generally considered to be either legionary or praetorian horsemen in origin, achieving heavy cavalryman status was a type of promotion for auxiliary cavalrymen as well. This adds some uncertainty to the identification of the branch of service that these particular equites promoti originated from. In addition to the specialists using peculiar combat skills Roman cavalrymen were armed and trained both for missile fighting and close combat, and as is made clear at a later point in the text Arrian expected his troopers to engage their opponents first with javelins and then to close with swords and axes.

The hekatontarchai or centuriones attached to the picked cavalrymen, probably the equites singulares and possibly also the equites legionis, are likely to have been centuriones exercitatores and/or centuriones stratorum, which available evidence indicates to have been legionary officers rather than men drawn from the auxilia. The use of legionary personel in these positions is evidence of the elite nature of the imperial legionary cavalry and a reminder that though provincials of Celtic, Germanic or Thracian stock may have provided the bulk of the Roman cavalry and influenced its fighting methods, the effectiveness of these horsemen was improved by Roman training methods and discipline. The high quality of the cavalry training of the imperial army is aptly demonstrated in the Technè Taktikè or Ars Tactica, one of Arrian’s other writings.

M. Pavkovicz, 'Singulares legati legionis: guards of a legionary legate or a provincial governor?' in: ZPE 103 (1994), 223-228.
M.P. Speidel, Guards of the Roman army (Bonn 1978), .
M.P. Speidel, ‘Horsemen in the Pannonian alae’in Roman army studies II (Stuttgart 1992), 62-66.
M.P. Speidel, ‘Legionary horsemen on campaign’in: Saalburg Jahrbuch 47 (1994), 36-39.

The bodyguards

The identity of the two hundred soomatophylakes of the infantry units presents some problems. The direct Latin equivalent, protectores, is attested in other sources, mainly epigraphic, only from considerably later dates. These third century protectores appear as elite legionary cavalrymen, apparently a new expression used instead of the earlier term of singulares or secutores. The number of two hundred would round up nicely for that would be a conceivable number of legionary horsemen drawn from one full strength and one somewhat understrength legion as according to Josephus a total of hundred twenty cavalrymen were attached to a legion. Speidel has suggested that these bodyguards may have been stratores, legionary horsemen that were employed in the horse supply to the military. Here however the soomatophylakes are probably men drawn from the legionary infantry rather legionary mounted troops because of the explicit mention of pezoi, foot soldiers. The use of the term soomatophylax by Arrian may have been influenced by the similarity between these infantry guards and comparable soldiers in the army of Alexander the Great. Using Josephus as a source these troops may be identified with the lonchophoroi described as providing the legionary guards of the general. The hundred kouphoi lonchophoroi mentioned later on in this passage are therefore very likely to be part of these two hundred legionary foot guards. Whether there were a corresponding hundred kontophoroi part of the legionary bodyguards is unclear. The antesignani of Caesar’s day, precursors of the imperial lancearii, however could be armed both as heavy infantry carrying pila and as skirmishers using appropriate equipment (perhaps identical to the imperial era arma antesignana), which seems to support this theory, though Josephus seems to imply that lonchai and aspides were the standard equipment of legionary foot guards.

The commanders of the bodyguards would have been hekatontarchai lonchophoroon mentioned in Josephus, which were called praipositoi or praeposti and hekatontarchai soupernoumerarioi or centuriones supernumerarii according to a Diocletianic papyrus. The specification of these guards as light lonchophoroi seems to indicate that other lonchophoroi are heavy infantrymen and that these men should be seen as a distinct troop type. The irregular Rhizianoi lonchophoroi appear however together with other troops that are specified as light armed, strengthening the case that the corresponding heavy lonchophoroi should be the four rearranks of the main battle-line, also called the peltastai or thyreophoroi lonchophoroi. As mentioned above the kouphoi lonchophoroi are likely to be identical with the akontistai of the marching order. The identification of the kouphoi lonchophoroi as legionary foot guards has been challenged by M.P. Speidel and M. Pavkovic. They prefer to identify these soldiers as auxiliary pedites singulares rather than as elite legionaries. According to their view only a hundred of these were present with the field army with Arrianus with four hundred others stationed at one of the province’s frontier forts. The identification of the kouphoi lonchophoroi with beneficiarii is almost certainly not right. The lancea used by these troops was surely the hasta habens amentum in medio of Isidore's definition rather than the thrusting spear with decorative tip of Celtic origin used as a badge of office by soldiers as the beneficiarii and speculatores.

Flavius Josephus, Bellum Judaicum (centurion of lancearii) 6.262; (legionary guard javelineers) 3.93-95; 3.120; 5.47.
Iulius Caesar, Bellum Civile, 3.84; 1.57.
P. Beatty Panopl. 2.259-270; 2.285-290; 2,299-304.
A.B. Bosworth, ‘Arrian and the Alani’in: Harvard Studies in Classical Philology
M. Pavkovic, ‘A note on Arrian’s Ektaxis kata Alanon’in: Ancient History Bulletin 2.1 (1988), 21-23. AHB article on the net
M.P. Speidel, ‘Legionary horsemen in battle’ in: Saalburg Jahrbuch 47 (1994), 36-39.
M.P. Speidel, ‘The Caucasus frontier. Second century garrisons at Apsarus, Petra and Phasis’ in: Roman army studies II (Stuttgart 1992), 206-207.

Silence and warcries

The Roman army of the empire placed great emphasis on the silence maintained by its troops both on the march and in battle deployment. Warriors of all ages have tried to alleviate their own stress by making as much noise as possible (incidentally this is also done by chimpansee primates when engaging in warlike behaviour). The republican Roman army, which was closer to a militia than a professional army, also advanced into battle making a lot of noise, raising war cries and banging their javelins against their shields. The silence maintained by the Spartan and Roman armies was proof of their superior discipline and must have had a most unnerving effect on their opponents who had to face an enemy that apparently gave no sign of this instinctive human behaviour. The raising of the warcry only just before actual contact was also maintained by the late Roman army. Whether the Roman war cry was already resembling the Germanic barritus is unknown. Roman cavalry had apparently already adopted war cries of Germanic origin at this date, belying the usual assertion that the raising of the barritus by the late Roman army was a result of barbarisation of the fourth century army. National war cries had been raised by the various contingents of the Roman republican army and perhaps each of the constituent parts of the imperial army would use a yell of their own as well rather than a common one, adding to the cacaphony.

Flavius Arrianus, Technè Taktikè, 44.1.
A.K. Goldsworthy, The Roman army at war 100 BC – AD 200 (Oxford 1996), 50; 192-197; 229.

The advance

It is unclear in what manner the cavalry was divided in two parts. Perhaps it reflected a division between two turmae joined together in a single unit described in the Technè Taktikè and paralelling the infantry manipulus. The division of the cavalry in two parts, one in active pursuit and another defensive line as a reserve, is also described in the Byzantine military manual by Mauricius, where these troops are called cursores and defensores respectively. This handbook also describes a manner in which a similar defensive infantry formation opens up to let the cavalry pass through, though it is pure speculation whether a similar method was envisaged by Arrianus.

The faster than normal step of the infantry referred to here would probably be the gradus plenus instead of the measured gradus militaris, both described in Vegetius. That this should not be interpreted as a run seems clear because of the distinction made from the dromos of the light fighters.

Flavius Arrianus, Technè Taktikè 18.2.
Vegetius, Epitoma rei militaris, 1.9.
G.T. Dennis, Maurice's Strategikon (Pennsylvania 1984), 23-26; 48-51; 62-63; 113.
A.K. Goldsworthy, The Roman army at war 100 BC – 200 AD (Oxford 1996), 19; 182.

Precautions against flank attack

Flanking attacks are countered by withdrawing vulnerable archers to higher ground to lessen the danger of their being overrun. The use of battle-axes by Roman cavalry as close combat weapons is a peculiarity of the writing s of Arrianus. They may have been a special weapon employed by Roman cavalry regiments facing heavily armoured enemies like the cataphracti used by Eastern armies. No examples of such battle-axes have been found. It is remarkable that Arrianus here seems to imply that the Scythians or Alans are unprotected after he has given a description of heavy armour before.