The Battle Tactics of Alexander the Great Chapter 1 Part 2 By Kathleen Toohey (The above photo of a bust of Alexander was taken in the İstanbul Archaeology Museum in 2002.)  Heavy Infantry In contrast to light infantry, the situation with heavy infantry is a little simpler. Again, the first point that needs to be considered is what the characteristics of heavy infantry are. As has already been noted in the above examination of light infantry, the primary distinguishing feature is to be found not in its arms and armament, but in the importance placed on the maintenance of order in the ranks by units of this type in combat. “Without good order the hoplite are is worthless”. So said Aristotle. Anderson, Military Theory, p. 96. As a basis for the examination of the Macedonian phalanx to follow, it will be helpful to first consider the Greek hoplite’s style of fighting. Speaking of the hoplite phalanx, Garlan says; “In order to hold off the enemy at the end of their lances while protecting their right flank from enemy projectiles the hoplites had to fight in close formation. Moreover, to prevent an easy break in the formation, followed by a general rout, the men were drawn up several ranks deep (never less than four, normally eight). Hence, too, the necessity of engaging battle on level terrain, of bringing up troops and having them charge in waves in unison over the entire front, of keeping the rhythm of the advance steady, to overcome the tendency of each soldier to veer imperceptibly to the right, seeking the cover of his neighbours shield.” Garlan, War in the Ancient World, p. 124. The shield, in a very real sense, epitomises the spirit of the hoplite phalanx. The significant feature of the hoplite shield, as opposed to that used by the light infantry peltasts, is that it offers protection to more than just the man who wields it. Anderson, Military Theory, p. 15 & n.8, p.112, where he makes the point that the close connection between the shield and the phalanx is taken for granted by the ancients in general. The reported comment of Demaratus of Sparta that he cites on p.15, to the effect that the shield is carried “for the sake of the whole line” is important in this regard. Specifically, in formation, the individual hoplite was protected on his left side and much of his front, by his own shield. His right side was protected by the shield of the man next to him. This protection, however, was only available as long as the formation was rigidly maintained, with each shield facing directly forwards. If one man in the front rank were to flinch and pull his shield in to his side for a moment, the man next to him would instantly become vulnerable. If a man was killed, the right side of the man next to him would be exposed until the man in the next rank, standing behind the man who fell, moved forward to take his place. In that interval, not only was the next man in the front rank to the left, in danger of being killed, it is quite likely that he would expose the man next to him while trying to protect himself on the right. This is the reason for the phalanx formation having to be several ranks deep, so that the forward pressure of the rear ranks could ensure that gaps in the ranks were filled as quickly as possible. And this is why considerable emphasis was placed on the structure of the file, with the best men being placed in both the front and rear ranks. Anderson, Military Theory, p. 95. The need to maintain a continuous shield line was paramount. If it could not be maintained, as Garlan notes above, the formation would easily be broken, and once that happened, a general rout would quickly follow. This need also explains why the spear was the primary weapon of the hoplite, not the sword, See Anderson, Military Theory, pp. 37 – 39, for the relative importance of the spear and the sword. because it could be stabbed out well in advance of the shield line. (It was probably used to prod and probe the opposing shield line, when facing a similar unit, in the hope of slipping through and doing damage. For an analysis of how the hoplites used their spear in fighting, see Anderson, Military Theory, pp. 86 – 88.) The shield and the shield line, then, would have been important factors in building a strong sense of unity and interdependence in any hoplite unit. And it is clear that, in formation, they fought, essentially, as a unit, and that they would fight on a man-to-man basis only if the ranks were broken. Anderson, Military Theory, p. 86. Anderson here cites Plato’s Laches, where the value of a new style of training hoplites is discussed, (Plato Laches 181E – 183D.) Such a system would require a high degree of discipline in the ranks and a considerable emphasis, in training, on precision of movement, to maintain order in the unit. This is precisely what we do find. It is apparent that hoplites were trained in only a few standard weapon movements, each clearly intended to be used by the men in the front rank(s) simultaneously on hearing the specific command, (Anderson, Military Theory, pp. 86 – 87. Anderson presents a detailed discussion, drawn from ancient accounts, of hoplites being drilled in specific movements, both to facilitate quick deployment into battle order from a marching column, and to maintain order while moving in phalanx formation. The command structure of the phalanx, with each taxis divided into lochoi (and these being further subdivided so that even the leader of each file was an officer in command of the men behind him - which might be as few as four), must be viewed as being essentially for purposes of controlled movement and order, only (Anderson, Military Theory, p.100). For a detailed analysis of hoplite training and drill, see Anderson, Military Theory, pp. 84 – 110, and note also his diagrams of formation deployment, figures I – V, pp. 388 – 397 See also Pritchett, W.K., The Greek State at War: Part II. (Berkeley, University of California Press, 1974) pp. 208 – 245. Thus it can be said that the true distinguishing feature for heavy infantry, just as it was for light, is not its armament, but its role in the army of which it is a part. In the period with which this paper is concerned, that role was to provide a solid centre to which the more flexible flanking units (light infantry and cavalry) could be hinged. The above discussion of hoplites throws some light on the role of the Macedonian pezhetairoi and hypaspists. In the past, there has been some debate about the nature of the Macedonian infantry. Griffith, History of Macedonia II., p. 424, Milns, R. D., “The Army of Alexander the Great”, Fondation Hardt Entriens Tome XXII (1976) p. 120. Afterwards cited as Milns 2. It seems fairly certain that the pezhetairoi wore no breast plate. Griffith, History of Macedonia II, pp. 422 – 423. This, and some misleading phraseology in Arrian, has led some scholars to view the Macedonian Phalanx as light infantry rather than heavy. Thus Best refers to a “phalanx of Macedonian peltasts”, Best, J.G.P., Thracian Peltasts and their Influence on Greek Warfare. (Nordhoff, I Gronigen Wolters, 1969) p.142. and Hogarth presents a fanciful picture of a strange mixed battalion in which javelin men are sandwiched between men wielding pikes. Milns 2, p. 119. Hogarth’s proposition has been more than adequately demolished by Milns. Milns 2, pp. 120 – 124. When the above discussion on light infantry and hoplites is considered, Bests phalanx of peltasts must be seen as a contradiction in terms. See Griffith’s comments on it, History of Macedonia II. In a discussion about the hypaspists, Ellis comments that the major weight difference between peltasts (light) and hoplites (heavy) “is that the former wore no armour”. Ellis, J.R., “Alexander’s Hypaspists Again” Historia XXIV (1975) p. 618. Griffith notes that the breastplate was not a uniform hoplite feature. Griffith, History of Macedonia II. pp. 422 – 423. This is a classic example of the modern scholar’s failure to understand where the real distinction lay between heavy and light infantry. The statement simply confuses his arguments, as at one point he clearly relates the Macedonian phalangites to heavy infantry, only, after highlighting the difference, to have to acknowledge that the phalangites too, wore no breastplate. Ibid. Such confusion need not arise if the nature of heavy infantry is properly understood, and there is good evidence to argue that the troops of the Macedonian phalanx should be regarded as heavy infantry. This largely involves comparisons with the features of the hoplite phalanx related above. Firstly, the soldiers of the Macedonian phalanx were at least as well trained and proficient in drill as the best Greek hoplite units. Consider the elaborate drill exhibition that Alexander’s phalanx performed before the Taulantians (Arrian I. 6. 1 – 4), and the description of the deployment of Alexander’s column into phalanx formation on the march to Issus. Arrian II. 8. 2. Also, the command structure of the Macedonian phalanx bears a strong resemblance to that of the hoplite phalanx, Set out by Griffith, History of Macedonia II. pp. 419 – 420. and probably had the same function as argued above. Moreover, in set battles, the phalanx clearly took the hoplite role. Lacking a breastplate, however, the men’s defence depended both on their shields and, more importantly, the greater length of their sarissae. Griffith, History of Macedonia II, p. 421. Griffith compares the effect of a phalanx armed with sarissae, to a warship that outguns its opponents. As we have seen, the effectiveness of the shield, the role of which would have been the same here as for hoplites, depended on the maintenance of strict order. The use of the sarissa is more problematic. Brunt estimates its length at a maximum of 18 feet. He bases this conjecture on Polybius description of how the later, longer sarissae of the first five ranks “protruded before the front line”. The same, he argues, must have been true for those of the first three or four ranks, in Alexander’s time. Arrian (tr. P. A. Brunt) (London, Heinemann, 1976) (Loeb Classical Library Vol. I), Introduction, p. lxxviii. The Introduction and Appendixes of this volume, will subsequently be cited as Brunt 1. See also Griffith, History of Macedonia II, p. 421. Since it needed both hands to wield it, the sarissa was probably not as easily managed as the hoplites’ spears. The hoplite spear, as I have argued earlier, should be regarded as a probing weapon, but it is hard to see how the sarissa, which might have been up to twice as long, For the length of the hoplite spear, see Anderson, Military Theory, p. 37. could have been used in this way. Rationally, it would seem to be just too easy to brush them aside and step past them. Once a couple of ranks had slipped past, the limited amount of controlled movement of his weapon available to the sarissa wielder, would have been reduced even further. Its length would have prohibited it being pulled out of the body of the advancing unit for another thrust, as a hoplite’s spear could, and the press of men moving down its length on either side, would restrict its potential lateral movement. This would make it easier for enemy men in the following ranks to avoid it. This problem, however, can be at least partially overcome, if we consider the possibility that the sarissa was not used in a thrusting and stabbing manner, but was instead simply held firmly Or at least as steadily as possible. See discussion below, figures 1 – 3, and the further discussion I have now inserted. in front as the unit advanced, or engaged the enemy. Now here it is also important to remember that the shield used by the Macedonians was somewhat smaller than that used by hoplites. The estimates of the shield size varies, Brunt estimates a diameter of two feet. (Brunt 1, p. lxxviii) The little archaeological evidence that we have suggests that it was actually .8 of a metre, only .1 of a metre less than the approximate standard size of a hoplite shield. Pritchett, W. K., Ancient Greek Military Practices: Part I. (Berkeley, University of California Press, 1971) pp. 146 – 147, 150 – 151. The smaller shield would mean that the separate files of the Macedonian phalanx could be spaced closer together than the files of a hoplite phalanx could. Pritchett, Ancient Greek Military Practices: Part I, p.154. This would mean that, when a hoplite file penetrated the spear line of a Macedonian phalanx, the space occupied by hoplites with shield to the front, would be larger than the space between the adjacent sarissae. If the hoplites tried to pass between them with shields facing front, this would cause the leading sarissae to slip past the front rank into the body of the hoplite phalanx at a slight angle to the line of the hoplite file - an angle that would increase in size as the two units came closer together. So, instead of passing down the line between the files, the sarissae, would be driven directly into the men in the ranks following, down the line of their files. If they were only being thrust into the hoplite body, they would probably cause a considerable degree of disruption to the second and following ranks, but little real damage. However, if the sarissae were held firmly and if the Macedonians in their attack made just a steady, but unrelenting, advance, Such as Adcock describes, Adcock, Greek and Macedonian Art of War, p. 26. the damage done to the following ranks could be considerable. The densely packed hoplites, already thrown into disorder by the spears of the leading rank, would become even further confused as the spear tips of the second, third, and perhaps even fourth ranks of the Macedonian phalanx penetrated deeper, wounding and killing as they advanced. The problem is that there would probably be too little time available for this principle to operate effectively, because, having penetrated the line of sarissae, the hoplites in the front rank would have the advantage. They would be better protected than the Macedonians and they would still be able to use their spears, whereas the Macedonians in the front rank would have to drop their sarissae and defend themselves with their swords. Once the front ranks actually began to fight, the advance of both parties would quickly slow to a halt, and the effectiveness of the sarissa would be lost. Such a scenario does not accord well with the standard picture of the virtual supremacy of the Macedonian phalanx, and I would expect such a style of operation to have produced casualties far greater than those that we do find recorded. For Issus, Diodorus and Curtius agree on a figure of 300 infantry; Curtius III. xi. 27 Diodorus XVII. 36. 6. Arrian says 120 Macedonians of note fell; Arrian II. 10. 7. For Gaugamela, the figures range from 300 in all; Curtius IV. xvi. 26; to 1000 foot, in P. Oxyrhynchus 1798 (cited in fn.4 to Diodorus p.293). Diodorus gives a figure of 500; Diodorus XVII.61.3. The above interpretation is based on the assumption that all of the sarissae that projected from any single file were pointed in the same direction, straight ahead (Figure 1). Suppose, though, that they were not all pointing in the same direction. Consider the following arrangement (Figure 1). Instead of pointing straight ahead, the sarissae of the front rank are inclined to the left, at an angle of perhaps 20 0-30° to the line of the file. Those held by the second rank point 10° to the left, and those of the third rank are pointed slightly to the right (Figure 2), or straight ahead if four ranks are lowered, with those of the fourth rank pointing slightly to the right (Figure 3). See. In both figures 2 and 3, the ranks are assumed to be spaced, for convenience, at an interval of three feet, front to front.) In this case, the gap between files would be slightly wider than the smaller shield would require. However, as shall become apparent, for combat purposes, the sarissae themselves would effectively act as both armour and shield. (Author’s note: Sometime after writing this, the thought came to me that there is actually a simpler, and more probable alternative to Figures 2 and 3 that would produce much the same effect. Because of both its length and weight, the sarissa is an intrinsically cumbersome weapon to use. It would consequently be quite difficult for individual soldiers marching together to hold this long spear rigidly pointing straight ahead, as in Figure 1. Some degree of wobble would be unavoidable, and this wobble would give an effect in combat very similar to that produced by the kinds of formation illustrated in Figures 2 and 3. I have subsequently found a quite vivid representation of this on line in a short video clip titled, “La falange macedone” (The Macedonian Phalanx), on YouTube: https://www.youtube.com/watch?v=-MX2Njyhouw&t=8s. The video is an excerpt from a longer Italian television program apparently from the Atlantide series on the Italian channel LA7. There is no information as to when the original program was made, or the title of the episode. The clip was posted to YouTube 15 Jan, 2013. The brief description accompanying the clip states that the excerpt “aims to show the operation of the Macedonian Phalanx”. The late Dr Minor Markle III, formerly of the University of New England, features prominently in the clip as an Academic commentator. He appears both in a brief scene demonstrating the forging of a sarissa spear head, and a later demonstration of how the weapon would have been used by a unit of pezhetairoi both on the march and in formation for battle. Unfortunately, the commentary is in Italian and much of what Markle himself says is talked over by an Italian translation. While what is being said remains unclear to me, the visual evidence speaks for itself. The young men taking part in the demonstration, about 30 – 40 of them, are all clearly very fit and strong, and may possibly be young soldiers or similar hired for the experiment given the precision of the marching and parade skills. Yet, despite this, even when they are standing in formation with their weapons at the ready, you can clearly see the reconstructed sarissa they carry visibly swaying back and forth, up and down, not to any great extent, but obvious nonetheless. The effect becomes more pronounced when they begin to advance in formation. Interestingly, when Markle has the column deploy from column into battle formation, the front second, third and fourth ranks shift slightly to the right of the man in front in a tight echelon formation. This would seem to be an innovation by Markle to manage the practicalities of having the first four ranks project their sarissae straight ahead. The result is a formation where the sarissae of the following ranks do not align straight ahead in parallel with those of the front rank, but are each in turn angled slightly to the right of that of the man in front. The effect is very much like that displayed in Figure 3 above. I only discovered Markle’s published work on Philip, Alexander and the Macedonian sarissa after I completed my thesis. And I visited with him for a couple of days soon after that. I know that Minor was very interested in conducting experiments to experience the practicalities of the Macedonian sarissae, particularly as a cavalry weapon. These are discussed in detail in his 1977 article “The Macedonian Sarissa, Spear and Related Armor”, Markle III, Minor M., “The Macedonian Sarissa, Spear and Related Armor”, Journal of Archaeology, Vol. 81, No. 3, 1977, pp. 334 – 338. The photos, together with a very brief discussion, also appear in a later publication; Markle III, Minor M., “Macedonian Arms and Tactics under Alexander the Great”, Studies in the History of Art, Vol. 10, 1982, pp. 106 – 107. where he talks of his work with a skilled horseman named Glenn Moody to explore how the sarissa might have been used in combat. The article includes four photos of Moody demonstrating how Markle saw the weapon being carried and used by Macedonian cavalry in combat. On a personal note, I did meet and stay with Minor for a brief time shortly after completing my degree. I have fond memories of our meeting. He was a nice man with a great enthusiasm for the subject. I’m sure I will have talked about my own theories, though I can no longer be sure of just what issues we discussed. (At the time, before the academic results were published, I had hopes of pursuing my work through higher studies. As it happened, I failed to secure a scholarship, and had to abandon the idea for financial reasons.)) The effect of such an arrangement, used in association with a steady advance, might have been roughly as follows. Because of their angle, the spears of the front rank could not be as simply pushed aside as before. Instead, they would push the shields aside, probably at the same time turning the facing of their bearers, and certainly exposing them to the sarissae of the second and later ranks. Admittedly, this is pure speculation, but it does, I feel, provide at least a good working model of how the sarissa may have been used in a reasonable and effective way, that is more in keeping with its formidable reputation. Unlike the previously examined hypothesis, this way of using the sarissa would attack the front rank of the enemy, as well as those behind. The effect of this on the enemy would probably have been dramatic. The men in the front rank, faced with a line of sarissae that they could not easily penetrate, would probably flinch back. In doing this, they would press against the ranks behind, while these, at the same time, would be trying to push the men in front of them, forward. As the Macedonian sarissae then penetrated deeper into the enemy formation, the following ranks would also begin to flinch, and the enemy’s advance would slow to a halt and then change to a back-marching retreat. In this system, then, there would be no need for the advance of the Macedonian phalanx to be actually halted, though it might be slowed for a time, when initial contact was made. The rate of advance need not, and probably could not, have been very great, so that the breakdown of the enemy’s units would be a slow process. Such a picture also accords well, then, with the roles that Adcock sees the phalanx playing, of thrusting back and pinning down the enemy line. Adcock, Greek and Macedonian Art of War, p. 26. Once the sarissae had penetrated the front ranks, lateral movement by a unit would become impossible. As to the basic tactical unit of the pezhetairoi, the obvious choice is the taxis. Brunt 1, p. lxxvi. Milns,2, p. 101. But it is perhaps more appropriate to regard the real tactical unit of the phalanx as a group of taxeis, up to and including all the taxeis of the pezhetairoi and the hypaspists. As will be shown subsequently, on the couple of occasions when the phalanx was split, See the following chapters on Issus and Gaugamela. either by enemy action or by circumstances causing a gap to develop, the result could very easily have been disastrous. The reason for this rests in the vulnerability of the flanks of the phalanx. The Macedonian phalanx was essentially designed for frontal assault. If attacked in the side, its only defence would be to have the men on that side drop their sarissae (if not straight away, then probably as soon as the enemy front rank passed the tips of their spears and came within striking range), and defend themselves with their swords as best they could. (The possibilities of countering such an attack by wheeling the unit or by a change of frontage, are not considered here, as I am merely considering the vulnerability of the phalanx to opportunistic flank attacks produced by the changing circumstances of actual combat that would develop at a rate greater than such manoeuvres could counter. Such manoeuvres would probably also be excluded in battle; i.e. after combat has been joined; both because there would not be sufficient space for this, and because the unit, when threatened with a flank attack, would probably already be engaged in combat on its front.) Against a unit of formed heavy infantry attacking its flank, the Macedonians, disordered and poorly armoured, would have been clearly outmatched. Even against light infantry, the advantage would probably lie with the attackers, as they would be used to fighting in a disordered state. Any comment on their chances of coping with a cavalry attack would be frivolous. The only real defence against a flank attack was to prevent the opportunity for it arising. There were three ways to do this. One would be to protect those troops whose flanks are most vulnerable, by placing units on their flanks, which are less vulnerable to flank attack, and which are able to act as a buffer between the units that they are protecting and the enemy. The use of light infantry for this purpose has been discussed above, p. 4. The second way is to cover a unit’s flanks with troops posted beside and behind it, so that any enemy units trying to attack its flanks can only do so by exposing their own flanks to attack by these covering units. (Such a principle can be observed operating in Alexander’s troop disposition at Gaugamela - See chapter 4.) Finally, a flank attack may be countered by having a continuous front of sufficient length to preclude flank attacks by forming a continuous line across a limited geographical area, as at the battle of Issus (See chapter 3.) In Alexander’s army, the flanks of the army generally consisted of both cavalry and light infantry. As the numbers of these troops performing this role in Alexander’s army was considerable, a real threat of a flank attack to the phalanx, could arise only if a gap formed either between different taxeis of the phalanx, or hypaspists, or between the ends of the phalanx, en masse, and the units on the flank next to it. Therefore, to protect itself, the phalanx would have to be drawn up with no appreciable interval between taxeis, A point that Marsden stresses, Marsden, Campaign of Gaugamela, Appendix III, Frontage of the Macedonian Army, p. 74, N. B. 1 –“ There is no interval between infantry units”. and move as a single unit. Having to move as a single unit, it would also have to fight as a single unit. For this reason, the phalanx of the pezhetairoi and hypaspists, combined, should be regarded as Alexander’s basic tactical unit for heavy infantry, for most purposes. On the one occasion when Alexander seems to interposed a different type of unit between taxeis of the phalanx, it can be shown to have been a tactical error, from the consequences that developed. (The matter and the probable reason for the division will be discussed in chapter 3.) Even here, there would basically be only two units to the phalanx. In battle then, the tactical value of the taxis should be regarded as only becoming manifest when circumstances necessitated making a convenient break in the phalanx. There is, of course, the problem of whether or not the hypaspists were armed in the same fashion as the pezhetairoi, to be examined. The main arguments for the case that the hypaspists were differently armed to the rest of the phalanx, are based on three points; some representations on coins that have been inferred to show a hypaspist; their use, separate from the pezhetairoi, with light infantry in forced marches; Milns 1, p. 187. and from the concept that they formed a flexible link between the phalanx and the companion cavalry. Ellis, “Alexander’s Hypaspists Again”, p. 618. Adcock, Greek and Macedonian Art of War, p.26. Delbruck, H., History of the Art of War: Vol. I, (tr.W.J. Renfroe Jr.), London, Greenwood Press, 1975, p. 179. Milns and Tarn have fairly demolished the effectiveness of the first two of these arguments. Milns 1. p.188. Tarn, W.W., Alexander the Great: Vol. II, London, Cambridge University Press, 1948, pp. 153 – 154. Afterwards cited as Tarn 2. As to the link hypothesis, as will be demonstrated in the following chapters, this concept of the need for a special link becomes unnecessary when the possibility of Alexander’s use of an echelon form of attack is considered. An echelon formation is defined as “A formation in which the subdivisions are placed one behind another extending beyond and unmasking one another wholly or in part”. Hayward, Brig. P.H.C., Jane’s Dictionary of Military Terms, London, MacDonald & Jane’s, 1975, p. 61. Garber, M.D., A Modern Military Dictionary, Washington, P.S. Bond, 1942, 2nd edition, p. 87. An echelon attack could be made from a line formation by the units forming that line advancing at regularly timed intervals starting from one end and progressing unit by unit down the line to its other end. For these reasons, the hypaspists, for the purposes of this paper, shall be considered to be no different in equipment and function, to the pezhetairoi. I would, however, offer one suggestion, that it might just be possible that the hypaspists, and perhaps at least some of the pezhetairoi, were trained in the use of an alternative weapon, for use on occasions, such as in rough terrain, in towns and cities, or in pursuit, where use of the sarissa was impractical. This other weapon may have been the javelin, or it may have been the standard hoplite spear, but it must be recognised that if it did exist, we have no reliable record of it. I am prompted to make this suggestion not by the fanciful arguments of Hogarth or Von Domaszewski; these have been adequately refuted by Milns; Milns 2. pp. 118 – 124. but by consideration of a passage in Arrian’s narration of the pursuit of Darius. In this passage, Alexander has 500 of his cavalry dismount and give their horses over to the use of the fittest of the infantry. In Brunt’s translation, the infantry are described as being “ordered to mount the horses, carrying their usual infantry arms”. Arrian III. 21. 7. I would query the concept of ‘usual’ infantry arms. It seems to me absurd to think that these men, who were being mounted on horses so that the pursuit can procede at the fastest rate possible, should be expected to take with them their cumbersome sarissae, which would at least slow down their rate of progress, if they could be carried on horse at all. The use of an alternative weapon such as a spear, perhaps borrowed on this occasion from other troops such as the Agrianians, perhaps the usual weapon carried in pursuits, seems to me to be the best way of overcoming this difficulty. It is highly unlikely that this mounted infantry consisted only of Agrianians and Archers (the only other infantry involved in the pursuit apart from the units of the phalanx Arrian III. 20. 1.), as were that the case, the probability is that Arrian would have said so, rather than give a vague account of their selection according to how well they had kept up their strength. Nor do I think it likely that they went armed only with a sword, as the sword was not, at this time, regarded as a particularly effective weapon. See n. 42 above. One final point, in this regard, needs to be made. The use of this infantry in the mounted pursuit, should not be taken as weakening the previous arguments concerning unit integrity and the need to maintain order in the ranks. There is no indication of how they would have fought if called upon to do so, but the fact that they took infantry weapons strongly argues that they would have fought on foot. Probably, the heavy infantry involved would have assumed a hoplite phalanx formation, though perhaps on an extended frontage, with a consequently reduced depth. Indeed the fact that the number of infantry taken was 500 may mean that what Alexander took with him was a regrouped pentacosiarchy of the phalanx, For the place of this unit in the structure of the phalanx, see Milns 1, p. 195. in which case, no Agrianians or archers would have been included. The taking of an actual pentacosiarchy also seems more likely, because it would permit the men involved to function as a unit. And the basic command structure of the phalanx would be maintained, as the men involved would simply have been transferred from one unit to another. This makes much more sense than to think that Alexander simply lumped 500 men of different types together, and then expected them to be able to function properly in battle. Of course this is, and indeed has to be, largely speculation. However, the possibility that the hypaspists and pezhetairoi were trained in the use of an alternative weapon, The two passages that form the base of Hogarth’s case that members of the phalanx used missile weapons – Arrian I. 21. 2; III. 26. 3. – while clearly inadequate for the purpose of supporting Hogarth’s arguments, do suggest that these men knew how to use weapons (restriced probably to javelins). Hogarth’s argument is cited in Milns 2, p.119. does provide a solution to some difficult points raised by Arrian’s narrative. This includes a possible compromise position for those still trying to reconcile the inconsistencies that Arrian seems to leave in his presentation of these types. For the set battles, however, both types should be regarded as being armed with sarissae. 10