Grand Tactics at Gaugamela Author(s): A. M. Devine Reviewed work(s): Source: Phoenix, Vol. 29, No. 4 (Winter, 1975), pp.

374-385 Published by: Classical Association of Canada Stable URL: http://www.jstor.org/stable/1087282 . Accessed: 13/02/2013 02:49
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Classical Association of Canada is collaborating with JSTOR to digitize, preserve and extend access to Phoenix.

http://www.jstor.org

This content downloaded on Wed, 13 Feb 2013 02:49:02 AM All use subject to JSTOR Terms and Conditions

GRAND TACTICS AT GAUGAMELA

A. M.
LIKE
DEVINE

accounts not long afterit was fought. subject of a numberof conflicting and As a result,it has been a perennialsource of difficulty, indeed of formodernscholars. In particular,extensivedisagreement controversy, over Alexander's tactics on this field has been productive of several in what actuallyhappenedduringthe attempts recentyearsto reconstruct course of the battle.' Most of these attemptshave been marredby the the intoa preconceived evidentdesireoftheirauthorsto force engagement tactical mould, one derived essentiallyfromthe militarysystemsof the Too oftenhave we beheldAlexander and nineteenth centuries. eighteenth the cocked hat of Napoleon or performing Prussian drill of the wearing Frederickthe Great. Hence it is my aim in thispaper to put forward yet anotherattemptat reconstructing events of the battle of Gaugamela the -this time, however,avoiding the intrusionof tactical anachronisms. This will involve, among other things,the presentationof a somewhat view of Alexander'suse hereof the grand tacticsof attack in unorthodox order. oblique
THE MACEDONIAN ORDER OF BATTLE

SO MANYofthedecisive of battles history, becamethe Gaugamela

SinceGaugamela fought an openplain,2 normally on was the important of factor terrain an indirect inAlexander's role and played only plans, had on littleeffect the courseof the battleotherthan that of facilitating of thePersian tactics doubleenvelopment cavalry. influence with Its was, in visible therather of unusualnature theformation takenup however, forces to of bytheMacedonian prior theopening theengagement. Having no naturalobstacleon whichto resteitherof his flanks, and hence to expecting be easilyoutflanked the numerically by superior Persians,
'The most interestingrecent reconstructionsof the battle are to be found in Sir William Tarn, AlexandertheGreat(Cambridge 1948), Part I Narrative, 46-51, and Part II Sources and Studies 182-190; G. T. Griffith, "Alexander's Generalship at Gaugamela," 7HS 67 (1947) 77-89; A. R. Burn, "Notes on Alexander's Campaigns, 332-330," 7HS 72 (1952) 81-91; J. F. C. Fuller, The Generalshipof Alexander the Great (London 1958) 163-180; E. W. Marsden, The Campaign of Gaugamela (Liverpool 1964) 40-64; R. D. Milns, Alexander the Great (London 1968) 110-126; F. Schachermeyr, Alexander der Grosse(Vienna 1973) 267-276; R. Lane Fox, AlexandertheGreat(London 1973) 233-243; and P. Green, Alexanderof Macedon (Harmondsworth 1974) 286-296. These works will be cited by author's name. 2For the site of the battlefield, Schachermeyr, see 268, sketch-map4, and 270, note 311.

374
PHOENIX,Vol. 29 (1975) 4.

This content downloaded on Wed, 13 Feb 2013 02:49:02 AM All use subject to JSTOR Terms and Conditions

GRAND TACTICS AT GAUGAMELA

375

as Alexander drewup his armyin whatmightbe looselydescribed a tactical"square." At bothends of the Macedonianfront wereplaced Becauseof thenecessity it of drawnback from en ichelon. flank-guards, the tacticalflexibility, angle at whichtheseflank-guards maintaining someweredrawnback couldnothave beentoo great.It was probably of like army's deployment thing 450.3As a consequence theMacedonian its in thislong-based was formation, secondline of infantry trapezoid of wouldnot failto unableto coverthefulllength its rear,a factwhich of have an effect, albeita minor one,on thecourse thebattle. of of The flank-guard theright wedgewingtookthe form a roughly of units,each withits front shapedarrangement cavalryand infantry Persianbattle-line. These unitswere parallelto that of the opposing in lines,whichcould,whenthe need arose,be deployed three potential into actuallines, setat an angleto themainfront still easilyformed three In ofthearmy, nowfacing from centre. therearmost of but the outward in these from to right, from left front rearward, to and lines, posted order wereone halfof the Agrianians commanded Attalus, Brison's by [12],4 Macedonianarchers[13], and, rightflankrear of the entirearmy, in of Cleander's [14]. mercenary infantry Positioned front theAgrianians in and thearchers, thesecondofthethree werethe7rpbpopoL irres lines, and Ariston. [15],under Aretes, thePaeoniancavalry [16],under Finally, in front them, hence front thewhole of in of and were the stationed body, of as the mercenary cavalry Menidas[17],forming, it were, pointofthe "wedge."5 The leftflank-guard similarly was withitsconstituent units arranged, drawn back en echelon approximately same angle. Roughly at the
3There can be no question but that the flank-guardswere placed at an angle to the enemy battle-line and not simply, as Marsden, 48-51 and diagram 2, has them, to Alexander's allegedly oblique front.The phrase is is used by Arrian to des~brartaLtv cribe the positioningof both the right(3.12.2) and left (3.12.4) flank-guards. the case In of the former,the reason given for this disposition is el 7rovavdAyKt KaraXa2&dvotL 1 dwvarlr'7ra vyXElc arV dXayya. Now, if this flank-guardwere already parallel to the Persian front,there could be no question of using it to extend (dvaarrdat) the while using it to close up (vuyKXE^iau) the phalanx would involve a phalanx still further, retreatin the face of the enemy, which could well be dangerous, if not disastrous. 4The numbers in square brackets correspond to those used to indicate the relevant units in Figs. 1 and 2. 5Thisinterpretation Arr.3.12.2-3 is based on Burn, 85-86. Griffith, of 78-79, however, interpretsthis passage as meaning that Menidas' cavalry was, in fact, posted on the extremerightflankrear of the army, at an angle to the main front,with the rest of the in flank-guard line withit on its left. Like that of Marsden, Griffith's exegesis is open to doubt on tactical grounds: for,if it is accepted, then, as Marsden observes (48), "it is to extremelydifficult explain why the Persians did not attempt to stop Alexander's movement to the right by attacking the Companions themselves" (rather than by clashing with Menidas' cavalry).

This content downloaded on Wed, 13 Feb 2013 02:49:02 AM All use subject to JSTOR Terms and Conditions

Persian Positions

23
2 21810 a8
20 24

24

25
FIG.

1.

ALEXANDER'S

ORIGINAL

ORDER

OF BATTLE

(BASED

ON

This content downloaded on Wed, 13 Feb 2013 02:49:02 AM All use subject to JSTOR Terms and Conditions

95

24

24

24
FIG.

2.

ALEXANDER'S

ATTACK

IN OBLIQUE

ORDER

(THE

SITUATION

AT A

This content downloaded on Wed, 13 Feb 2013 02:49:02 AM All use subject to JSTOR Terms and Conditions

378

PHOENIX

of like the flank-guard the otherwing,it, too, was deployed wedge-shaped in threepotentiallines: two of cavalry, supportedin rear by a thirdof At infantry. the point of this "wedge" stood the mercenary cavalry of Andromachus [23],whilebehindthemwerestationedthealliedcavalry[21], under Coeranus, and Agathon's Odrysianhorse [22]. The thirdand rearmost line was made up of Sitalces' Thracian javelin-men [18], together with two units omittedby Arrian,but mentionedby Diodorus, namely, the Cretan archers[19] and the Achaean mercenary infantry [20].6 Formidable as these flank-guardswere, the real strength of the Macedonian forcelay in its centre. Here was arrayed the iron core of Alexander'sarmy, front-line the infantry phalanx. Drawn up in orderfrom in rightto leftwere: first, theiraccustomedplace of honour,the Hypaspists [2], under Nicanor, then the taxeis of Coenus [3], Perdiccas [4], Meleager [5], Polyperchon[6], Simmias [7], and Craterus [8]. On either were stationed the best cavalry units of the entire flankof the infantry army: on the left,the superb Thessalian cavalry [10], under Philip, supportedby a body of allied Greek horse[9], commandedby Erigyius,and, on the right, the Companion cavalry [1], officiallycommanded by Needless to say, Darius Philotas, but led in battle by Alexanderhimself. could not,and would not,riskthe destruction his own first-rate of cavalry in a direct assault on so invulnerablean enemy front.Rather, common sense dictated that the Persian attack would be made against the Macedonian flanks;hence Alexander'sadded concernwith providingparticularly strongmobile defencesforthese flanks.
THE BATTLE

The battle itselfwas precipitatedby Alexander's movementcbs rb r' beb6v,7which alarmed Darius, who was afraid that, if the Macedonian armyreached the groundwhichhad not been levelledby the Persians,his in of would (supposed) advantage in weaponry, the form scythe-chariots,
here 6Diod. 17.57.4; cf. Curt.4.13.31,Arr.3.12.4-5. The reconstruction of the left is flank-guard basedon Burn,86-87. 3.13.1.It is generally thattheMacedonian advancetooktheform an of 7Arr. accepted towardsthe Persianleft.Although is probablethat,due to the it oblique movement factor simplephysical of the did convenience, Macedonians advanceat an angleto the front not an Darius wouldnothave enemy (though at too pronounced angle;otherwise, fearedthat theywouldmoveoffthe prepared before contactwithhis ground making their has withtheavoidance enemy of "mineforces), Aufmarsch cometo be connected fields" an attackinoblique (Curt.4.15.1;cf.Polyaenus 4.3.17),and with opening general order(Diod. 17.57.6).However,the propinquity any kindof minefield rendered of is dubiousby thePersianattempt envelopAlexander's to whiletheattackin flank, right oblique orderdoes not appear to have takenplace untilthe crisisof the battle (see below,382).

This content downloaded on Wed, 13 Feb 2013 02:49:02 AM All use subject to JSTOR Terms and Conditions

GRAND

TACTICS

AT GAUGAMELA

379

he be nullified.8 Therefore orderedthe advance-guardof his own leftwing (probably the thousand Bactrian cavalry and the regimentof Dahae, also a thousandstrong,mentionedby Curtius4.12.6) to ride around the Macedonian right wing, which Alexander was leading in person, and halt its march.This placed Alexanderon the defensive, and, in orderto regain the initiative,he detailed Menidas [17] to charge the enveloping Persians. The counter-attack, however,failed,and Menidas' small force was drivenback by the superiorweight (a merefourhundredhorsemen9) of the enemy cavalry. Alexander now ordered a charge to be made by Aretes'prodromoi with [15]10and the Paeonians of Ariston[16], stiffened Cleander's mercenary carried [14]. As thissecondcounter-attack, infantry out by farmore formidable Macedonian units,was meetingwith greater success than the first, Bessus, Darius' leftwing commander,threwinto the fraythe rest of the Bactrians,possiblyas many as eightthousandin These were themselvesrepulsed,and theirbattle-formations number.11 broken,aftera long, hard fight12 culminatingin a vigorous charge by the troops of the Macedonian flank-guard.13 Meanwhile, the Persian were demonstrating their ineffectiveness with a charge scythe-chariots against the Macedonian right centre. They were promptlyrouted by Balacrus' javelin-menand the remaining halfof the Agrianiancontingent [11], who had been stationedin a forward positionin orderto screenthe Companion cavalry [1]. Now came the crisisof the battle as Alexanderhalted his lateral march and began a directadvance on the Persian positions.The movementof some cavalry,whichDarius had told off supportthe troopsattempting to the envelopment the Macedonian right, of had lefta gap in the front line of the Persian left centre.14 was towards this gap that Alexander, It seizing the tactical initiative once more, now wheeled, and, making a of wedge (FqBoXov) the Companion cavalry [1] and part of the infantry led phalanx [2, 3, 4, 5, 6],16 the attack in the directionof Darius himself. At this development,the Great King fled.
8Arr. 3.13.2; cf.3.8.7. 3.5.1. See also J. R. Hamilton, 5 (1955) 217-218. CQ 9Arr. 1oAretes' "second"charge Arr.3.14.1 and 3.14.3 appearsto be a doublet theone at of at Arr.3.13.3. "Cf. Curt.4.12.6. 3.13.4.The repeated assaultsofBessus'cavalryindicatethatthisstruggle was 12Arr. oflengthy Alexander's duration, perhaps, just after ending, majorattackon theenemy whilethe factthat the heavierlossesherewereincurred the Macedonians centre, by thattheretreat thePersianleftwas due moreto demoralization, a result of as suggests ofDarius' flight from field, the thanto actualtactical at defeat thehandsofAlexander's troops. in the 80-81,and Burn,87, note6, are agreed excluding Companion "aGriffith, cavalry from partof theaction;cf.Tarn II 185-186. this 14Arr. 3.14.2; cf. Curt.4.15.20. 3.14.2 and 3.14.4. 16Arr.

This content downloaded on Wed, 13 Feb 2013 02:49:02 AM All use subject to JSTOR Terms and Conditions

380

PHOENIX

I suggestthat thisevolutioninto a wedge-shapedformation, evidently carried out only during the actual crisis of the battle, has led to the Macedonian army being creditedby both Curtius and Diodorus with a battle-linein oblique orderfromthe verybeginning the engagement.16 of The identityof the source, or sources, utilized by these two historians fortheirbattle-narratives itselfa major crux,and need not concernus is here.It is possiblethat Curtiusand Diodorus bothmuddledtheiraccounts of Gaugamela themselves.However, it is morelikelythat theyinherited an already defectivetradition.Nor, indeed,is it hard to see how such an erroneousversiongot its start. It may, at first, have been a case of mere For, whatever its original tactical disposition,the over-simplification. frontline of the Macedonian army was, at this stage of the conflict, actually advancing in oblique order.The wedge of Arrian3.14.2 was, it should be stressed,no mere metaphor,but rathera complex formation of units arranged en echelonin two oblique lines, slanting in opposite in directions but meeting a broad point.The left-hand side of thismassive units of the phalanx from as wedge was made up of the first-line infantry far left as (and including) Polyperchon'staxis [6], while the ilae of the both the muchshorter side Companioncavalry [1] constituted right-hand and the broad apex of the formation."A sharper point was doubtless provided by the person and body-guardof Alexanderhimself, swinging leftwardto assail Darius. At the same time, the flank-guard the of Macedonian rightwing, also in oblique order,was swingingoutward to the right in close combat with the Bactrian cavalry of Bessus.18 A militaryignoramus,especially if he had not been presentat the battle, mightwell have become obscure,or even genuinelyconfused,when endeavouringto describeso complicateda tactical situation.Not so, however, the experiencedsoldierPtolemy,who (we may reasonablyassume) rode with the 0auttXK throughout engagement.Ptolemymay not tic the have had a very clear idea of what was going on in other parts of the field-his coverageof the proceedings the Macedonian leftis farfrom on it is only thanks to him (and to Arrian,who precomprehensive-but serves his account, at least in outline) that an accurate reconstruction of Alexander's oblique order tactics, and hence of the decisive moment of the battle, can be made. Meanwhile,in the centre,due to both the pace of Alexander'sadvance towardsthe gap in the Persian front and the evolutionof his rightwing into wedge-formation, gap developed in the Macedonian army's own a
16Curt.4.15.1; Diod. 17.57.6. of 17For alternative reconstructions Alexander's wedge, see Fuller, 171, diagram 11, and 174; Marsden, 57; Milns, 118 (diagram) and 123; Schachermeyr,273; and Green, 293. 84-85, does, that Alexander had to wheel "sThereis no reason to suppose, as Griffith, to his rightand drive Bessus' wing offthe fieldhimself.

This content downloaded on Wed, 13 Feb 2013 02:49:02 AM All use subject to JSTOR Terms and Conditions

GRAND TACTICS AT GAUGAMELA

381

frontline. While all the units fromPolyperchon'staxis [6] rightward in surgedforward the greatattack on the Persian centre,Simmias,unable to push forward rapidly,and hearingthat the leftwas in difficulties, as halted his own taxis [7] to succourit. Throughthe resultant gap therenow a forceof Indian and Persian cavalry fromthe disintegrating penetrated enemycentre.19 Reaching the Macedonian baggage-park[25],20 they fell its defenders, overwhelmed them,and liberatedthose of theirown upon whom they found there as prisoners.At this, the Macedonian troops second-line [24],in accordancewithpreviousorders,facedabout infantry and put the raidersto flight. While all this was taking place, the left,under the command of Parmenion,was sustainingan attemptedenvelopment,as well as a frontal commandedby Mazaeus. assault, by the as yet undefeatedPersian right, Being thus hard-pressed,Parmenion despatched riders to Alexander, asking for assistance.2' It is almost certain, despite statementsto the that this message was never delivered. For contraryin our sources,22 whenParmenion'smessengers reached the Macedonian right, theyfound that Alexander was already far advanced in his pursuit of Darius and returned, apparently without having accomplished their mission.23 Fortunately,however,forthe Macedonian cause, Parmenionsucceeded, withouthelp,in checkingthe Persian attack withhis Thessalian cavalry fromthe fieldfiltered to [10]. Then, as news of Darius' flight through the faltering enemyright,the repulserapidlybecame a rout.
"The account of an attack on the Macedonian camp itself,by cavalry sent fromthe Persian rightby Mazaeus (Curt. 4.15.5; Diod. 17.59.5-8; Plut. Alex. 32.3), is probably as apocryphal as the fantasticstoryof the refusalof Darius' mother,Sisygambis, to be rescued (Diod. 17.59.7; cf. Curt. 4.15.10-11), with which it is connected. Plut. Alex. 32.3-4 and Curt. 4.15.6-8 renderthe alleged attack on the camp even more dubious by linkingit with the tale of a weak-kneedParmenion demanding aid froman unperturbed, if contemptuous,Alexander (see below, note 21). 20Arr.3.14.5-6. As this raid was dealt with by the second-line infantry(Arr. 3.14.6), it is evident that its object (whetherintended or not) was merelythe fieldbaggage-park, and not Alexander's fortified four-day camp (Arr. 3.9.1; Curt. 4.10.15), some 16-17 miles away, or either of his two alleged night-camps (Curt. 4.12.17 and 4.12.24; cf. Arr. 3.9.3-4, which has only one such camp), also some 3-4 miles distant. 21Anappeal forhelp fromParmenion at this stage is quite credible, and need not be associated with any deliberately discreditingstory of cowardice or irresolutionon his not part, as it is in Plutarch and Curtius (above, note 19), though, very significantly, in Diodorus (note especially, 17.60.8). For the role of Callisthenes as the creator of the anti-Parmeniontradition,see Tarn II 352-353 and 357, and J. R. Hamilton, Plutarch: Alexander, A Commentary (Oxford 1969) 89. 22Arr. 3.15.1; Curt. 4.15.6-8 and 4.16.2-3; Plut. Alex. 32.3-4 and 33.6. 21Diod. 17.60.7. The dust and chaos of battle (Curt. 4.15.32-33; Diod. 17.60.4), togetherwith the likelihood that Alexander was by then a couple of miles or so fromhis original position at the start of the battle, renderit most unlikely that anyone coming from the left wing could have reached him while he was pursuing Darius. Cf. Fuller, 178, and Lane Fox, 241.

This content downloaded on Wed, 13 Feb 2013 02:49:02 AM All use subject to JSTOR Terms and Conditions

382

PHOENIX

Whetheror not he receivedParmenion'sappeal forhelp,Alexanderdid turn back, most likely because of the loss of Darius' trail at the River darknessof evening.24 Lycus (probably the Khazir Su) and the growing It was duringthis rearwardride26 that Alexanderbecame involved in a shortbut bloody clash witha large body of retreating Persian cavalry,in company with some Parthian and Indian horse. Here, according to Arrian,26took place the hardestfought cavalry action of the entirebattle. After having slain sixty of Alexander's Companion cavalry [1] and wounded his favourite, the of Hephaestion,27 remnants the Persian horse brokethrough Macedonian squadrons and made good theirescape. the When he had assured himselfof the fact that his army was completelyvictorious,and that hence therewas no more gloryto be won on the field, Alexander renewed his pursuit of Darius. Meanwhile, Parmenion advanced and took the Persian camp, along with the enemy's baggage-train and elephants. Having crossed the Lycus, Alexander bivouacked on the farside and restedhis cavalry until around midnight, when he rode on to Arbela, some fortymiles away, where he hoped to When he reached Arbela the capture both Darius and his field-treasury. as following day, he did in factfindtherethe Persian field-treasury, well as his defeatedopponent's chariot,but no Darius. The Great King had riddenon to meet an ignominious end-at the hands of his own subjects.
24Curt. 4.16.16-19; cf. Plut. Alex. 33.6-7. That "Alexander was no Prince Rupert, and Gaugamela no Edgehill"-as Marsden, 61, puts it-has become a dogma in recentstudies of the battle. It is highly probable that, after the battle of Ipsus in 301 B.C. (where Demetrius, pursuing the beaten cavalry of Seleucus and Lysimachus too far from the field,lefthis father,Antigonus Monophthalmus, to be defeated and killed), the historiographical emphasis changed from one of execrating Parmenion for being responsible, throughhis "unnecessary" appeals forassistance, forthe escape of Darius (Curt. 4.16.3 is most significant here), to that of glossingover the tactically irresponsiblecharacter of Alexander's prolongedpursuitof the Great King (note Curt. 4.16.29-30). The linkingof the historianAristobuluswith the battle of Ipsus at Arr. 7.18.5 is certainlyevidence for this view. 3.15.1-2; Curt. 4.16.20-25. Hardly the vast envelopingmovement,aimed at the 26Arr. of complete encirclement the Persian centreand left,that Marsden, 61, seems to make it. Alexander lacked both the required effectivesuperiorityin numbers and the tactical position fromwhich to carryout such an encirclement.In any case, the grand tactics of central penetrationwould have held a greaterappeal, because of theirdirectness,forthe king's "heroic" spirit (here note, for example, Arr. 1.11.7-8; 1.12.1; 3.3.1-2; 5.26.5; 6.9.3-6; and 6.24.2-3; Plut. Alex. 8.2), as well as being better suited to the task of of dealing with the rather loose tactical formations the Persians. 2"Arr.3.15.2; cf. Curt. 4.16.20. The importance and ferocity ascribed to this clash by our sources probably have theiroriginin the personal reminiscences Ptolemy himself. of "'Arr. 3.15.2. No tactical conclusions can be drawn fromArrian's inclusion here of Coenus and Menidas; cf. Griffith, 86-87, and Milns, 124. It is evident from Diod. 17.61.3 and Curt. 4.16.31-32 that Arrianhas simplyconflateda mentionof the wounding of Hephaestion and the loss of sixty Companions in this encounter with a casualty list of senior officers; Burn, 88, and also R. M. Errington,CQ 19 (1969) 237. cf.

This content downloaded on Wed, 13 Feb 2013 02:49:02 AM All use subject to JSTOR Terms and Conditions

GRAND

TACTICS

AT GAUGAMELA

383

CONCLUSION

the Having thus reconstructed battle of Gaugamela, we may approon conclude with an appraisal of Alexander'sgeneralship this priately field.That he showed great competencehere goes withoutsaying. What is open to question is the consistencywith which Alexanderdisplayed in thiscompetence his conductof thebattle. Certainly foresight the shown in providing for mobileprotection the flanks the Macedonian of adequate army duringthe defensivephase of the action meritsimmediatepraise. Yet the wisdom of Alexander's order for Menidas and his mercenary on obvious. It cavalry [17] to open the engagement theirown is farfrom to certainlyseems irresponsible launch a small body of horseon a charge of formation enemycavalry,even if,as was the against a much stronger case here, support were close at hand. Viewed in this light, Menidas' charge does appear to have been an error-and an errorof carelessness at that. But simplecarelessness, is especiallyat crucialmoments, hardlytypical of Alexander.Everythinghe did was carefully planned in advance. This is not to say that he nevermiscalculated.Rather,his mistakes,like those of othermen of great military political genius,were by and large due or to passion or megalomania.Such a mistake,as we shall see, was his prolonged pursuit of Darius fromthis very battlefield.And that was the resultnot of any impetuousdecision,but ratherof yet anotherattemptif despite a similarfailureat Issus-to carryout his long-standing, overly hazardous, plan to end the war at one blow by seizingthe personof the Great King. Likewise,Menidas' chargewas no ill-considered manoeuvre, but part of a carefully laid tactical plan that happened to misfire. was, It in fact,a device to "draw" the enemy.28 the GranicusAlexanderhad At sacrificed advance-guardofcavalryin orderto throwthe Persianhorse his guardingthe fordinto confusion.29 Only then had he launched his main attack with the Companion cavalry. Here, at Gaugamela, Alexander faced a similarproblem-how to open the enemy'sranks forthe decisive assault. Since Bessus' Bactrians and the Dahae were movingaround his flankin a disciplinedmanner,30Alexanderhad to "draw" them.The only to way that he could do thiswas by sendinga small unit forward attack, and be driven back in disorder-a disorder which would include the Alexandercould pursuers.Having thus dissolved theirbattle-formation, exploit theirdisorganization sendingin stronger by cavalry units. In the event this is exactly what he did. And, but forthe vastly superiornumbers of the Persian cavalry,he would probablyhave won his battle there
28I am particularly gratefulto ProfessorE. Badian forsuggestingthisinterpretation to me. 29Arr.1.15.1-3. IOCf.Arr. 4.5.4-5.

This content downloaded on Wed, 13 Feb 2013 02:49:02 AM All use subject to JSTOR Terms and Conditions

384

PHOENIX

The factthatthistactical and then. shouldnotblind to its us ployfailed It haveworked and had here, notthe before, should viability. had worked as beenexpecting At anyrate, it. Alexander Persians, seems very possible, for didmakea mistake which must heldaccountable, ofunderhe that be the of withwhichDarius could counter estimating strength the forces move.But theinitialfailure by no meansirretrievable. hisopening was On thecontrary, openedthewayfor stab at theheart thePersian of it a sincein orderto meetthe threatDarius had uncovered left his army,
centre. Brilliant and subtle tactician that he was, Alexander,like Napoleon, workedon the principle "a plan withbranches."Once his originalplan of had bogged down in a hard-fought cavalry engagementon the right, Alexanderput into effect main branch-the central attack in wedgeits formation. This major attack was the high-light the battle in more of ways than one, and, as such, warrantssome discussion.The manoeuvre itselfwas well-conceived,and-apart fromthe consequent disjunction in the Macedonian front line-skilfully executed. That Alexanderchose this tactic, ratherthan, say, an ordinaryfrontalassault, was evidently due to two separate considerations.First, he meant to seize the opportunityto envelop the internalleftflankof the Persian centreby way of the gap that had just opened. Secondly, he sought to enable his Companion cavalry [1], the tactical spearhead of the wedge,to advance ahead of the infantry, while neitherdeprivingitselfof close support fromthe nor disruptingthe army's main battle-line. In devising the phalanx manoeuvreof attack in wedge-formation, more technically, double in or, Alexanderevinced a striking in the sphere of oblique order, originality grandtactics.For althoughtheoblique orderattack had been inventedby Epaminondas some forty years before,31Alexanderappears to have been the firstgeneral to develop the manoeuvrein any tacticallysignificant But fromthis point on Alexander's tactics are not so commendable. of Putting his long-term political interestsahead of the requirements the immediatetactical situation,he at once set offin pursuitof Darius himself.Admittedly, the successfulcapture of the Great King's person would have outweighedany incidentaldisaster,such as the destruction of the Macedonian leftby an undefeatedPersian right.But it must be conceded, in view of the ease of retreat affordedDarius by the open terrainof the battlefield,that there was never very much chance of Alexanderactually overtakinghis quarry. Moreover,across the enemy's
well-authenticatedoccurrenceof the oblique order attack, at the battle a"For the first of Leuctra in 371 B.C., see Diod. 15.55.1-15.56.4. 32The "Alexandrine wedge" seems to have made its debut duringAlexander's Illyrian campaign in 335 B.c. (see Arr. 1.6.3). way.32

This content downloaded on Wed, 13 Feb 2013 02:49:02 AM All use subject to JSTOR Terms and Conditions

GRAND

TACTICS

AT GAUGAMELA

385

natural line of retreattherelay the River Lycus, completewith a bridge that could be brokendown by the Great King or-as indeed happenedafterhis crossing.33 Although hopelesslyblocked by his fleeingfollowers it is possible that Alexander was unaware of the problem of the riveruntil he got to it, the natureof the terrain rear of the Persian in crossing positioncould scarcely have escaped his notice. In these circumstances, the tacticallycorrectcourse of action would have been forhim to return to the battlefield properas soon as the routof the Persian centrewas seen to be complete. That he did not in fact do so is an instance of tactical delinquencywhichcan be counted against him, at least fromthe standpoint of a critiqueof pure generalship.His Homeric attitude to war had again led him offin fruitless quest of the "spolia opima." in Needless to say, Alexanderwas fortunate having as the commander of his leftso competenta tactician as Parmenion,to whom belongs the here-as at Issus-what was virtuallya separate battle. gloryof winning Parmenion'swing of the armyhad certainlybeen placed in jeopardy by the king's preoccupationwith Darius. The gravity of the situation is indicated by Parmenion's (unanswered) appeal for help. Nevertheless, two considerationscan be advanced in partial defence of Alexander's conduct. In the first place, he had providedhis generalwiththe excellent Thessalian cavalry [10] to act as a nucleus foroffensive action. In the his apparent confidencein the ability of Parmenion and his second, troops to withstandthe onslaughtsof the Persian rightwas justifiedby the successfuloutcome of a remarkably similarsituationat Issus.34 It is generallyagreed by militaryhistoriansthat Alexander's victory at Gaugamela was a tactical masterpiece.This judgment, though substantially correct, does require some qualification.As we have seen, Alexander's tactics in this battle, while on the whole brilliantin their and subtlety,were not altogetherflawlessin theirexecution. originality The masterpiecewas marredby the victor's heroicpretensions and misconceived orderof priorities. we must remember, was, after Gaugamela, the victoryof a verygreat but as yet immaturegeneral.35 all,
UNIVERSITY OF ALBERTA, EDMONTON

33Curt.4.16.8-9 and 4.16.16-17. 34Arr. 2.11.1-3; cf. 2.11.6-7 (suspicious); Curt. 3.11.13-15; cf.3.11.16 (also suspicious). See above, note 24. 35I am greatly indebted to ProfessorsE. Badian, A. B. Bosworth, D. Fishwick and J. King-Farlow, as well as to an anonymous referee,for their helpful comments on earlier draftsof this paper.

This content downloaded on Wed, 13 Feb 2013 02:49:02 AM All use subject to JSTOR Terms and Conditions