STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. THE CAMPAIGN OF PLATAIAI By R. T. Clabk I The campaign of September, 479 B.C., presents so many interest- ing problems that no excuse need be made for this attempt to re- examine them. Our knowledge of events depends virtually on a single document, the ninth book of Herodotos. The narrative is in the Father of History's best style, detailed, animated, and adorned with piquant incidents. Collected from a variety of sources, the information in his possession has been worked into a coherent story whose very smoothness had caused it, until recently, to be accepted as a reliable account of one of the few battles in history which may fairly be called decisive. The campaign begins with the secret departure of the Spartan forces from Lakonia for the Isthmos (c. II). 1 At that moment the Persians, who, earlier in the year, had advanced south, were in pos- session of Athens. Almost as soon as Argive messengers had brought news of the starting of the Spartans the Persian army retired into Boiotia. The seizure of the city was a political move and strategi- cally can only be considered a raid (pace Munro, JHS, 1898, p. 152). There is no record of any military activity during the occupation, and this immediate abandonment suggests that only an "expedi- tionary force " was employed. But the manner of the Persian retire- ment is somewhat mysterious. While retreating, says the historian (c. 14), Mardonios learned that another army of 1,000 Spartans had reached Megara, whereupon in the hope of taking them he wheeled his whole army round against Megara and his cavalry went on before into the Megarid. After this news came (c. 15) that the Greeks were concentrating at the Isthmos, and he therefore retired through Dekeleia. The "neighbours of the Asopians" were sent for and led him via Sphendale, Tanagra, and Skolos to Theban territory. This he "cleared" and took up a semi-fortified position. 1 The references are throughout to Herodotos ix. [Classical Philology XII, January, 1917] 30 The Campaign of Plataiai 31 It is possible to take out of the narrative more than the actual words imply, but it seems clear that we are intended to note (1) that Mardonios' final retirement was by a different route than that assigned to the first retirement, and (2) that the retirement was undertaken in some haste. The reason assigned for the retreat from Megara is that the Greeks were at the Isthmos, which is virtually no reason at all. The original retreat was changed into a sudden "dash" at another army. What army this was is uncertain. It may have been an advanced guard or it may have been the apparently perma- nent garrison at the Isthmos (c. 8) ordered to move the moment news arrived that Pausanias had left. As Megara was in no danger the move seemed to intimate that the Greek offensive had begun. It could have been no more than an intimation, yet it so excited Mar- donios that he broke off his retirement, executed maneuvers involving his whole force (all to take Megara which had been at his mercy all summer), failed in his object, whatever it was, and had to retreat by a circuitous route in some haste. That he really intended to strike at Megara is scarcely possible. The town was a walled one, had just (vide Herodotos) received a reinforcement of 1,000 Spartans, and could have been taken only by regular assault. Yet stress is laid on Mardonios' use of his cavalry. It seems clear then that what he aimed at was to intercept this force. It was not going to Megara, because Herodotos explicitly states it was at Megara when Mardonios turned. Where then was it going ? Apparently it eventually got to some point where it became an embarrassment to Mardonios' first line of retreat and necessitated the long detour. Now, when Mar- donios finally got into Boiotia he took up a somewhat peculiar position. His lines ran from Erythrai, past Hysiai, right to the Asopos in Plataian territory, i.e., south of the river. The position was not uniformly held, but troops were massed at three points. The left wing (possibly composed mainly of cavalry) was at Erythrai, the center a little north of Hysiai on the hills above the Moloeis, while the right rested on the Asopos. The camp seems to have been some- where north of the left wing on, or near, the Erythrai-Thebes road, probably south of the river (Grundy, Great Persian War, p. 449; Macan, Herodotus, chaps, vii-ix, II, 367). One is at once struck with the fact that of the three Kithairon passes one is abandoned — the 32 R. T. Clark Plataiai-Megara pass, through which runs the best road from the Isthmos to Thebes — and thereby the great strategic line of Kithairon. The explanation most in favor is that Mardonios laid an elaborate trap. The road to Thebes was left open as a bait. If the bait was swallowed, the Greeks would push up from Plataiai with Thebes as objective and deploy against the Persian right. Here the right would take them in semi-flank on the Plataiai-Thebes road, the center would take them on flank and rear, while the cavalry from Erythrai would complete the debacle. It is very doubtful, however, if such elaborate strategy can be ascribed to a fifth-century leader as is involved in this "trap." If Pausanias had an objective, it was that of all good generals, the enemy's main force. He thought it was at Erythrai, and off he went in search of it. To say his objective was Thebes and that Mardonios knew that, is to use terms which only befit a later age. But on the other hand one does expect to find that attention will be paid to those fundamental principles of strategy which date back to prehistoric times. Mardonios breaks one in calmly leaving a superior line of defense (which could make an even better line of defense for his foe) while Pausanias breaks another in gaily severing — and that without any political object — -his communications with his base. The key to the Kithairon position and to Pausanias' com- munications with the Isthmos is Plataiai, and Plataiai is therefore the key to the campaign. The point I wish to make is this: Mar- donios chose a "next best" position because he had lost Plataiai, and Pausanias could go to Erythrai because he held it. It is scarcely to be doubted that the organization of the great advance was considerably more business-like than Herodotos pre- tends — one has only to remember the fatuous anecdote of Khileus — and it is probable that troops had begun to concentrate long before Pausanias left. That general was, of course, committed to the offensive; he had to drive the barbarian out of Greece. But he had to prepare for that offensive, and it was no use collecting all Hellas at the Isthmos unless he gave them a road by which to leave it. Mardonios, who never intended to fight a battle in Attica, received the news of the advance and commenced a leisurely retreat. Sud- denly he heard that Spartans were already at Megara and were The Campaign of Plataiai 33 advancing up the Megara-Plataiai road. He realized his danger. He could indeed retire as he had planned, but he would lose the Kithairon line if they reached Plataiai before him. Hence his sudden raid in MAP I Persian Position Battle at Erythrai Greek Advance the hope of catching them on the open road. The raid failed, and the result of the seizure of Plataiai was that if the main forces moved as quickly his whole retreat was endangered unless he took a new route. Hence the circuitous retreat. Let it be said, too, that Mardonios, who was on the defensive strategically (i.e., he was meet- ing an offensive) could only have neglected the Kithairon line by compulsion. It was an ideal line for a force such as the Persian acting against the hoplite, and Mardonios knew the latter both in his weakness and his strength. 34 R. T. Clabk It will be asked why Mardonios failed to secure Plataiai. In point of fact there is little reason why he should have secured it. He had no foe to fear, and Boiotia was so entirely in his hands that isolated towns were negligible. In his invasion of Attica — an evacu- ated district — he had no need to trouble about communications if there was no foe to threaten them. It is possible Plataiai never was in Persian hands at all in 479. 1 It had been destroyed in 480 but was probably rebuilt, and there is no record of a second occupation or destruction, while the account of the Greek retreat from the river seems to indicate it was still standing. Besides, Mardonios was lulled to a false sense of security by his dreams of breaking up the confederation. When he heard of the force at Megara he had suddenly to face the possibility of attack. If this were the van- guard and Pausanias was advancing to join the Athenians at Eleusis he might be in considerable peril. The reconaissance indeed told him that the advance proper had not yet begun, but it failed to keep the Greeks from Plataiai. Only the fact that there was such a failure explains Mardonios' neglect of Plataiai now. 2 It also helps to explain his subsequent dispositions, for, if the Greeks were in Plataiai, he would naturally conclude Pausanias intended to advance into Boiotia by that route in the expectation that the Persian main force covered Thebes. Naturally he put it where it would not be expected, north of Hysiai and in the most convenient place for that decisive action which is the duty of the center force. Pausanias did not hurry his advance. He concentrated his forces, effected a junction with the Athenians at Eleusis, and there received information that the Persians were at Erythrai. On Ery- thrai he therefore directed his march, a thing he was distinctly rash in doing unless he held Plataiai. Historians almost unanimously declare that the Greek advance used all three passes. Not only does that contradict Herodotos just where his broadly accurate but uncritical knowledge begins, but it is at variance with the Greek plan 1 The little city was an Athenian ally and Mardonios was anxious to conciliate the Athenians. 2 Granted that Plataiai was now in Greek hands, it may be asked why Mardonios did not attempt to storm it. Artabazos' failure at Poleidaia is explanation enough. No general expecting an immediate attack in force could afford to begin siege opera- tions. The Campaign of Plataiai 35 of campaign. Pausanias was not advancing to a Kithairon position, but to an invasion of Boiotia via Erythrai. With the Isthmos for base, troops at Plataiai, and his lines of supply directed from his base to that village, he advanced through the pass and found Erythrai evacuated. This leads us to the first stage of the battle proper. II The advance of the Greeks into Boiotia is thus described: When they reached Erythrai and learned that the Persians were encamped on the Asopos, the Greeks formed themselves opposite, along Kithairon (c. 19). When they remained on the hills Mardonios sent his cavalry against them (c. 20) with disastrous results to the Megarians, who sent a messenger to Pausanias. He called for volun- teers: 300 Athenians responded (c. 21), beat back the enemy, killed Masistios, and were on the point of being overcome (c. 22) when the arrival of the main body defeated the final Persian attack (c. 23). Then the Greeks judged it better, both for other reasons and because that region was better supplied with water, to go down toward Plataiai, and especially to Gargaphia. They therefore marched past Hysiai, and, arriving in Plataian territory, formed their line on the plain (c. 25). The Persians, having mourned their loss, now marched to the Asopos (c. 31), and the two armies faced each other (c. 35). Here we are already in difficulties. In c. 15 we find the Persians at the Asopos and at Erythrai; in c. 19 they have left Erythrai and are gone to the Asopos; while in c. 31 they go now to the Asopos. In c. 25 we have the Greek positions carefully described as "on slight elevations and the level plain," while in c. 31 we are told the Greeks are encamped on the Asopos. After beating back the cavalry the Greeks determined to go to Plataian territory and Gargaphia (c. 24), while in c. 19 they have already formed opposite (after learning that the Persians were at the Asopos!)— what or whom t Where does the Persian cavalry come from ? In what position were the Greeks when their leader had to be informed by a herald (c. 21) of the imminent proximity of a division of his army which later (c. 23) only needed to shout to make its needs known? And why should Pausanias sacrifice 300 Athenians if he was coming up with his main army ? 36 R. T. Clakk Herodotos, I fancy, gained his fullest account of the battle from an actual combatant, probably an Athenian sous-officier of intelligence but not of sufficiently high grade to be in his leaders' confidence. No "official history" of the war was published, and the sous-officier, remembering a very complicated series of maneuvers which he did not understand, has only a confused narrative to give. Keeping this sous-officier in mind, let us examine the account. Mardonios' strategy had failed, and there was nothing for it but to make the best of it and evacuate Erythrai. He could do nothing but wait until he saw what Pausanias would do. The Greek leader on arriving at Erythrai found to his surprise that the Persian main force was on the river. He had therefore only one move possible — to go to Plataiai. Even if Plataiai was still unoccupied, he had to go there unless he desired to risk his whole line of communications. Persian troops still held the Hysiatan position, and keeping therefore on high ground he advanced along Kithairon in a westerly direction. At the head of this column would naturally be the Spartans and the Athe- nians and Megarians, i.e., what was later the left would bring up the rear. On this rear Mardonios suddenly launched a cavalry attack. The Greeks can hardly have been drawn up on the hills awaiting an attack, because they had no reason to expect one, while Mardonios would scarcely have sent cavalry uphill against hoplites in position. What happened was probably this: Mardonios, learning that his enemy was on the march, saw an excellent chance to disturb his line and prevent his reconcentration on Plataiai and his subsequent advance. If he was successful he might compel the whole army to fight on unfavorable ground. The rear was promptly thrown into confusion by the series of charges. (As no infantry was used there is no question of a regular battle.) The Megarians were on bad ground, probably caught off their guard, and had to send a herald to Pausanias for aid. This proves that the van was some distance away, at or past Hysiai. If the whole army was drawn up in battle array, presumably watching this little drama, Pausanias was betray- ing an incompetence that was not even graceful if he waited for a courier to tell him what he knew. A picked corps of Athenian archers went off, beat off the attack, and killed Masistios. Mar- donios saw he had accomplished nothing and made a last endeavor The Campaign of Plataiai 37 to spoil the advance to Plataiai. In one mass the cavalry surged down on the extreme right of the Greeks, but the main body came up and after desperate fighting the cavalry sullenly withdrew. That MAP II CAMP First Phase Greek piercing move Persian Eetreat Evolution to second position the main army means the whole Hellenic army is unlikely, considering both their position and their numbers. One corps was sufficient. Could the phrase mean the Athenian main body in contradistinction to the 300 who were really its rearguard ? The Athenian narrator would say "our main body," whence Herodotos interpreted this of the whole army. The result of the skirmish was the clearance of the Erythrai pass. Mardonios had attacked the Greek lines, had been defeated, and the Greek position was secure, with the hills held from Plataiai to beyond the Persian center at Hysiai. 38 R. T. Clark Grundy (p. 461) sees in the cavalry battle an attempt to cut the Greek line. But it is difficult to see what the Greek line was doing in such a position as to allow it to be pierced at Dryoskephalai, i.e., right and left of the pass. It could only have gone to the left. Besides, Mardonios only used his cavalry, and it was not the weapon for such a maneuver. Macan (p. 368), on the other hand, imagines that the Greek column was headed by the Megarians and was attacked as it emerged from the pass, that the Greeks came into the battle suc- cessively, and so beat off the attack. It was folly for Mardonios in that case to attack with cavalry. The ground would not permit such maneuvering of at least 20,000 men. It is only natural to suppose the Spartans led the van, and the account indicates that the attack took place after Pausanias had been at Erythrai and had given up his idea of fighting a battle there. Munro (p. 157) supposes the attack was delivered near Hysiai and that the main Greek body had gone off to attack the camp in the direction of Panakton and turn the left wing. With the Persian forces as they were, that was simply playing into the enemy's hands. Pausanias had to cover Plataiai, as I have said before. Both these views ignore the fact that the Persian center was at Hysiai (as Herodotos does, doubtless because it became the left wing before it was in action). Encouraged by his success, Pausanias resumed his offensive, and leaving the hills advanced into the plain. Duncker {History of Greece, p. 340) regarded this movement as a mistake because of its abandonment of Dryoskephalai. It was certainly bold, so bold indeed that the position to which it led the Greeks became untenable, but it necessitated a withdrawal of the Persians and made it impos- sible for Mardonios to move south for nine days. Advancing rapidly then, Pausanias sought to pierce the Persian lines via Gargaphia, thus cutting off the Hysiatan division, a maneuver which if at all successful must have brought on an attack with the Greeks on favorable ground. To execute this the wings changed. The Spartans were still heading the advance, but the original rear was becoming the left to keep the Persian right from outflanking (origin of change of wings' story?). Mardonios, however, outstripped him. The Persian center evacuated Hysiai and joined the right on the ridge. Pausanias' plan was excellent, but either it was too slowly executed or Mardonios anticipated his design. The Campaign of Plataiai 39 To understand the position it is necessary to remember the posi- tion of the Persian right wing. It never moved during the battle except to advance south, so that its original position would be its most northerly one. Herodotos says it extended irapa rbv 'kauirbv, not "across the Asopos." He is obviously viewing it from the Greek point of view, so we may conclude that the right wing was south of the river. All accounts of the battle which I have seen make the Greek position before the retreat on the hills south of the Asopos (i.e., based on c. 31 ad. fin.), and therefore place the Persian lines north of that river. This is clearly seen, e.g., in the map of Grote's History. But to make the lines extend across Al to the main Asopos is to misunderstand Mardonios' strategy, for which it was essential that the Plataiai-Thebes' road should be left open in order that the Greek lines should get as far north as possible, and thus the right of the original Persian position must have been along the hills above Al. This raises the problem of the name Asopos. Grundy (p. 470) has already applied that name to Al to explain his difficulties about the so-called "second position." If in c. 31 Herodotos means by Asopos the main stream, then there must have been a retreat of the Persians. No such retreat is mentioned nor can it be deduced from the historian's narrative. The position of the Persian right wing is carefully described as being in Plataiain territory, nor is there an indication of a retirement north into Theban territory. Again, the best authorities place the camp south of the river. If it was south, then a wing north of the river would be cut off from it and it was obviously intended as a refuge if they were driven back. I would suggest that Asopos in Herodotos means throughout the main river plus its tributary Al. To apply the same name to a tributary as to the main stream is very common in country districts everywhere, and even if the inhabitants had an individual name for Al (which is unlikely), Herodotos would almost inevitably call it Asopos. Thus this Asopos 1 in a sense is the frontier north and east of the battlefield. When Herodotos says "up to the Asopos" he means a point at the junction of Al with the main stream. His words in c. 31, rbv tomt-q peovra top 'Aaairdv seem to imply that he speaks of one and the 1 It is unnecessary to object that Al is a mere brook. Under conditions of war the merest brook may constitute a formidable obstacle. A ditch, e.g., would play- havoc with hoplite formation in action. 40 R. T. Clark same stream (raurjj = at that point, i.e., where the Greeks are), and not of a new tributary of the original stream. What had happened was that Mardonios' withdrawal from Hysiai (referred to in c. 31, as opposed to c. 19, which refers only to the withdrawal from an east-southeast position to a river position) was accomplished before Pausanias had completed his thrust via Gar- gaphia. The result was that Pausanias, instead of facing a gap not held in force, was brought up against the solid mass of the new Persian left wing (the original Hysiatan center, and was compelled to turn the rear of his advance more and more in a northwesterly direction in order to face the second enemy mass on his left. The conditions of the ground however allowed him to occupy a line parallel to and almost as strong as that of his foe. The position was stalemate. Much difficulty has been caused by the assumption, e.g., by Delbriick (Persirker Krieg, p. 96), that Pausanias' strategy was a defensive one. Here he considers that Pausanias advanced to secure a better defensive position. This is absurd. If he had wished a "better defensive position" he had the Kithairon line. Pausanias was always on the strategic offensive; the retreat was only tactical, and then, if you will, to a "better defensive position." To interpret otherwise is to misunderstand the whole situation. The advance again is regarded by Grundy (p. 473), Woodhouse (JHS, 1898, p. 41), and Bury (History of Greece, p. 290) as an attempt to turn the Persian right. This is good so far as it realizes that Pausanias was taking the offensive. But it ignores the fact that the Persians moved after the Greeks did and a real outflanking movement was impossible as long as there were any Persian troops south of Al. By Gargaphia is, in any case, not the way to outflank a right wing which stretched away to the left of it and which was never really covered by the Greek left at any period of the battle. As Awdry (Annals of the British School at Athens, 1894, p. 97) and Wright (Plataea, p. 56) point out, any advance to Thebes must have taken place over excellent cavalry ground, and the Persian cavalry was not so shattered as to be as negligible as this would imply. As we have seen, too, Thebes never was Pausanias' objective. Munro (p. 159) likewise supports this view — apparently forgetting his Panakton theory — but with a different The Campaign of Plataiai 41 explanation. He supposes Pausanias delegated to the Athenians — • the despised Athenians — the task of skirting the hills east of the road to Thebes and of turning the right by forcing the river. They MAP III Second Phase and Projected Greek Retreat "funked" and took refuge on the Pyrgos hill, with the result that Pausanias had to move his center into the plain to fill the gap. But if his piercing movement was not to result in entire defeat, Pausanias had to keep his left in position against the Persian right until the enemy moved. This is the explanation of the left-flank movement. Nor is any account taken of the Persian center. If the gap had actually occurred, Mardonios had an easy task with his superior forces to pierce the line and roll the right up with his center. Any successful flanking movement was impossible for Pausanias, and an 42 R. T. Clabk attack following an attempt at one was almost certainly doomed to fail. Macan's theory (p. 379), while this criticism does not apply to it, conceives of a double position, which depends largely on his view of the Erythrai combat but which takes account of the Persians at Hysiai. He considers the first Greek position was an immediate formation on emerging from the pass. From that they moved to a Hysiatan position facing the Persian center and separated by the valley from the Persian right. Here was spent the time till Mar- donios raided Dryoskephalai, and then Pausanias moved forward to the main Asopos. Of all this there is no word in Herodotos. It is difficult to see why, if Pausanias was in straits at Hysiai, he should have considered that a tactical advance would help him unless he intended a general attack. The text shows that he clung to his position obstinately till badgered out of it, and that all the skirmishes took place on the river. Ill The narative thus goes on: For eight days the sacrifices were unfavorable, i.e., the generals did not wish to move (cc. 36, 37). Then Mardonios sent a cavalry squadron to Dryoskephalai which cut off the baggage train (c. 39). Two more days elapsed. The Persians abandoned their lines and came down to the stream, but the Greeks did not move. Cavalry attacks were incessant, and there was continual general fighting (c. 40). By the eleventh day reinforcements had visibly increased the Greek army, and Mardonios could stand it no longer. Against Artabazos' advice in favor of a policy of bribery Mardonios decided to attack (cc. 41, 42). Alexander at night betrayed the plan to the Greeks (cc. 44, 45). The story of the change of wings follows (cc. 46, 47). Mardonios taunted the Spartans and finally challenged them (c. 48). No answer being forthcoming he sent his cavalry against the Greeks, and what was almost an attack in force ensued. On the left the Greeks were repulsed from the river and on the right from Gargaphia. As a result they became short of water and food (cc. 49, 50). It was resolved to retire to the Nesos if no general attack took place next day. Water there was plentiful and the ground was too rough for cavalry. From it half their forces would reopen the Erythrai pass (c. 51). All that day they suffered heavily, and at The Campaign of Plataiai 43 night there was a general stampede, many "never intending to go where they had been told, while others bolted to Plataiai" (c. 52) and drew up before the Heraion. Pausanias ordered the Spartans to follow, but Amompharetos refused to move. Pausanias stayed to reason with him (c. 53). The Athenians, suspecting Spartan treach- ery, kept still and sent to ask Pausanias what to do (c. 54). He answered by asking them to come to him and "act as they should" (c. 55). Morning came and found Amompharetos still obstinate. Pausanias therefore moved off along the hills, keeping to the rising ground and the base of Kithairon. The Athenians also moved, but in the valley (c. 56). Amompharetos yielded now and followed. Pausanias halted on the Moloeis to receive him. He appeared, pur- sued by the cavalry (c. 57). Mardonios now ordered a general attack on the Spartans (c. 59). This account in the main is quite clear. The Greek line extended from Gargaphia round Al. On the rising ground the Persians were on the ridge, if anything in the stronger position. The position was the direct result of the check to the piercing move, with the result that inaction, though to the taste of neither, was a necessity for both. To force either position was dangerous; to retreat in the face of the enemy equally perilous. Much has been said to account for this failure to move, but necessity is a sufficient explanation (Meyer, Griechische Geschichte, p. 409). In vain did Mardonios bring his men down from the hills to induce the Greeks to engage with him; they saw their peril and remained in their lines. After a week of this maneuvering Mardonios tried to effect a diversion by cutting the Greek communication with Dryoskephalai. A squadron of cavalry occupied the pass. The actual effect of the movement was slight — a convoy was captured — but it was a plain threat to Pausanias' whole position by outflankment on the right. Provisions suddenly became scarcer, and the Greeks were in some difficulty. Mardonios, however, could not wait and resolved to take the offensive. He was now, at least, reinforced by Artabazos and was able to use his whole division as a reserve. His counsel was better than Artabazos', for he knew he had the advantage for the moment, while Artabazos' policy meant a long process and Persia, with the issue at sea still doubtful, could not afford long processes. 44 R. T. Clark The battle began by a series of cavalry charges followed by an attack in force designed to drive the Greeks from the river. On both wings the Persians were successful. On the left they crossed Al and drove off the Athenians. On the right they beat back the Spartans from Gargaphia. As a result the water supply failed. Pausanias was in grave peril. At the least, he was outflanked on both sides by swarms of enemies, though what Awdry says (p. 91) — that the Persian cavalry was riding at will round the Greek position — can hardly be true. The Greeks had to move. Here our sous-officier comes in again. He was not of the council and did not understand what he saw. Hence the incoherence of the account and the extraor- dinary tale of the stampede. Pausanias, if he was going to retire, had to cover his original position, i.e., the Plataiai line. He could not see the result of a repetition next day of that day's happenings, and it was possible the Greeks would be so far weakened as to be forced to retire with the battle raging. He had therefore first to secure the key — Plataiai — and accordingly the center was to retire on the town the moment it was dark. The others would stay in their lines till dawn to cover the retreat and then would retire southeast on diverging lines, the right ultimately to Erythrai via the Nesos, the left to the Nesos and thence cover Hysiai. On the Kithairon line it would await Mar- donios' development of his offensive. Another serious day passed with hard fighting and distress from want of water. At night in pursuance of the plan the center retired. Greek discipline off the battlefield was never very good. There was need for haste (seen by the fact that the center formed whenever it reached Plataiai), and so the retreat must have looked not unlike a stampede. Imagine the feelings of an Athenian, not in his leaders' confidence, ordered to stay in his lines to wait for the Spartans, sud- denly seeing the whole center apparently bolt, leaving his division isolated! At dawn, or sometime before, the Athenians, crossing the stream in their rear, proceeded to the Nesos, while Pausanias retired back from Gargaphia on fairly level ground. Amompharetos was left behind at Gargaphia to conceal the retreat. 1 With daylight the Persian cavalry discovered the truth, but Pausanias had had time to 1 It is surely unnecessary to comment further on the Amompharetos incident. The Campaign of Plataiai 45 get away and be on the rising ground that slopes to Kithairon. He halted to receive Amompharetos who came slowly, holding off the Persian cavalry. MAP IV Final Phases Persian Attack Final Battle Greek Pursuit Note. — These plans do not pretend to be otherthan the merest sketch maps With Grundy (p. 487) this retreat must be considered necessary. The tradition given in Diodorus from Ephorus praises Pausanias for his brilliant choice of a second position. It is difficult to see why. The original motive was brilliant, but the position into which it led him was never tenable. Sooner or later he must retreat. Retreat being thus a tactical necessity, it will not do to lavish praise on Pausanias for the brilliant device of a "feigned retreat." Of this theory Wright (p. 63) is the most redoubtable champion. He 46 R. T. Clark considers that there was no need for the Greeks to retire for (a) it has been established by Woodhouse (p. 48) that there were other streams near (but water would then have to be carried some distance and there was severe fighting all the time) ; and (6) the Persian cavalry retired from Gargaphia at night and did not occupy it (but it had already succeeded in destroying the springs). Apart from this con- sideration altogether, Pausanias was practically surrounded, was actually outflanked and on bad ground. Wright, however, will have it that Gargaphia was deliberately sacrificed to entice the infantry to follow the cavalry to destruction. But the infantry was already engaged and had driven in the Greeks. If the retreat was "feigned," why did Pausanias delay so long, why did the center leave the battle, why did he allow the Athenians to escape taking part ? The retire- ment of the center meant the occupation in force of Plataiai, which again indicates clearly that Pausanias meant to retire on Kithairon. As it turned out, the Spartans were very nearly defeated, might even have been defeated had the Athenians not engaged the Persian right. Pausanias of course must have foreseen a pursuit and an attack and took precautions accordingly; but to induce an attack cannot have been his aim. Admittedly the "feigned retreat" was a favorite device. Wright cites Thermopylai and Marathon. But in both these cases it was executed in action, in the former to make the enemy break rank, in the latter to let the wings close in. In neither case was there a complicated maneuver of an entire force over a wide tract of country involved. The feigned retreat of the Laches (c. 191c) which did win the battle was at the wicker fence and nowhere else. Delbriick's suggestion (p. 117) that the Persians attempted a diver- sion toward Phylai and that the center was sent back to meet this, Mardonios attacking while the line was being reformed, is unthinkable. Why should the center then form north of Plataiai, outside the town ? What good, moreover, could such a move have done Mardonios ? The Greeks were, as Munro says (p. 160), demoralized, or at least within measurable distance of being so, and their victory was the result of Pausanias' quickness in taking advantage of Mardonios' fatal error. The Persian general thought the Spartans were beaten, he failed to see the Athenians in the depressions toward the Nesos, and so ordered a general charge. Pausanias saw his chance, halted his line, and sent a messenger to the Athenians to send him archers The Campaign of Plataiai 47 for use against the cavalry and to change their course so as to come to his help. Munro (p. 164) considers this message apocryphal. It is not easy to see why Pausanias did not know the Athenians would meet the right wing of the attack, and he needed them if only to make assurance doubly sure. The Persians came on confident of victory. Macan (p. 372) complains that the center took no part in the battle but it must have been the original center which had driven the Spartans from Gargaphia and now led the pursuit. They did not venture to attack hoplites hand to hand, but threw up a stockade of wicker shields from behind which they opened a galling fire of arrows. The pretext of an unfavorable sacrifice kept Pausanias' men steady and an invocation of Hera gave them heart. The left — the Tegeans — attacked, and shortly after the Spartans moved also. This indicates the Spartan lines were in a northwest direction, i.e., in that of an army making for Erythrai. Meantime the Athenians, hurrying over to help the Spartans, straggled into the pursuit and the medizing Greeks who had been the right wing faced about and engaged them. The Spartans, however, by themselves had forced the stockade, perhaps by the device of the feigned retreat for which Pausanias would then deserve all the credit Wright gives him. The rest was easy. Unable to withstand the Spartan charge, the line broke and fled to the camp, the Spartans in hot pursuit. A desperate rally took place near the Demetreion, where the Persians displayed wild heroism. In the mMee, however, Mardonios, who was leading what promised to be a successful charge, fell, and at that the rally collapsed, Mardonios' chosen troops dying where they stood. Arta- bazos, left as a reserve (doubtless on the ridge), saw that the day was lost and drew off his squadron. It is difficult not to accuse him of cowardice, since the sudden accession of his fresh troops might have broken the Greeks. But his royal master seems to have had nothing but approval for the deed, and at that we may leave it. However, there was still some fight left in the Persians. At this point the cavalry re-entered the battle. Both Munro (p. 163) and Macan (p. 370) complain of its disappearance, but it had driven in Amom- pharetos and harrassed Pausanias till it was drawn off to make room for the infantry. With great gallantry it now fought a brilliant rear- guard action the whole way to the camp, thus enabling the broken infantry to man the walls. The Spartans, driving the enemy before 48 R. T. Clark them, failed to carry it, but the Athenians coming up later — which indicates that the battle on the left was more severe than Macan (p. 384) seems disposed to admit — effected a breach through which the Tegeans dashed. The victory was won. Both Wright (p. 70) and Macan (p. 372) doubt this division of labor, but one may ask, When did sappers ever form a storming party ? There remains the problem of the center, which during the struggle of the wings had remained inactive at Plataiai. When it saw the Greeks winning, contingents were detached to join the pursuit. It cannot be too strongly insisted that the center could not move until it was certain that victory was beyond doubt. If either Pausanias or the Athenians were broken, on it might depend the whole safety of Greece. The Corinthians went along the hills, while the Megarians and Phleiasians entered the plain. Here they were suddenly attacked by a squadron of Theban horse and, being in disorder, were hopelessly scattered. Macan (p. 385) supposes there was a division of duty, the Corinthians supporting the Spartans, the Megarians the Athenians, then in extremis. Possibly, but the point is that the Greek advance on the camp had begun before they moved. Consequently they were off their guard and careless. That the battle was won is seen by the fact that if the Athenians were still contesting the ground with the right wing, the Theban horse would, granted that they could have detached themselves from the m&Me, have had to ride round, or through, the Greeks. (They may, of course have been the Dryo- skephalai squadron, but Herodotos seems to indicate [c. 39] that that was Persian.) In that case the Megarians would have seen it com- ing. It was a squadron that had got off from the rout of the right. The Athenians were already near the camp and, seeing the Megarians on level ground and unsuspecting, the squadron suddenly swept into them. There is little evidence, it may be remarked, for the view that the Athenians were guilty of remarkable remissness, to use no stronger word, and failed conspicuously to attain their objective. The whole retreat except the center was caught ere it reached its objective, and no more blame attaches to the Athenians than to the Spartans. Tactically the retreat was a failure. Mardonios' error turned it into a victory. Perth, Scotland