(If anyone is still visiting this site, be aware the lack of new posts is due to a compulsion to work on something scholarly. So, I provide you with an example of this crap, an appendix that I just completed. This is the life blood of classical scholarship. Why some of the Greek came through and the rest turned into gibberish and why the footnote arabic numerals become Roman numerals, I have no idea.)
APPENDIX 6: THE TOPOGRAPHY AND ARCHAEOLOGY OF MARATHON
The plain of Marathon stretches about six miles along a slightly curving SW-NE axis, averaging about two miles in width between the heights and the sea.[i] The tectonically active plain is bounded and well defined on the west, north and east by rocky, scrub-covered hills of schist and marble – Aphorismos (1555 ft), Kotroni (771 ft), Stavrokoraki (1043 ft) and Drakonera (794 ft) – that rise fairly abruptly and steeply. To the southwest Mt. Agrieliki climbs to 1827 feet on extremely steep slopes, which are presently covered with small trees at the lower levels, and its eastern foot falls about a kilometer from the coast, forming the southern entrance to the plain. A side valley containing the town of Marathona extends northwest between Kotroni and Stavrokoraki, and a smaller valley, the Avalona, runs parallel to it on the west side of Kotroni. At the western extremity of the plain Agrieliki, Aphorismos and Kotroni form a sort of recess in which the village of Vrana is located. Typical of coastal Greece, the plain itself consists of alluvial fans and has in the last twenty-five hundred years risen about ten feet due to sediments brought down from the interior.[ii] Construction and the planting of trees have dramatically changed the appearance of the central and southwestern parts of the plain in just the last forty years, but inasmuch as the Greeks were able to form up their phalanx and none of the sources mentions any impediments, it may be assumed that in antiquity the plain was primarily planted in grain and there was only a scattering of trees.[iii]
Kynosoura (164 ft), a steep-sided spur of Drakonera, juts south into the sea for about a mile and a half, forming a protective weather barrier for the northern part of the bay and especially sheltering the Schoenia, a sandy beach running southwest from the promontory for two miles. The beach rises gently to a belt of pine woods, behind which is the Great Marsh, which once covered perhaps two-thirds of the northeastern half of the plain and extended to the coast south of the Schoenia. At the eastern fringe of the marsh, hard up against the spur that becomes the Kynosoura promontory, is a small salt lake, which drains into the sea.[iv] There was until 1934, when it was drained, a smaller marsh, the Brexiza, in the southern entrance to the plain, but unlike the Great Marsh it is not described by Pausanias and classical remains all but prove that it did not exist in antiquity.[v] The coast of the plain is for the most part formed of a shelving beach with shallow waters, but the southern reaches tend to be more rocky and uneven, especially when contrasted with the Schoenia.[vi]
Prominent in the middle of the plain is the Charadra, a winter torrent or arroyo that issues from the hills above Oinoe, a village northwest of Kotroni, and flows through the valley between Kotroni and Stavrokoraki, cutting two deep channels through Plasi to the sea. Though dry most of the year, the gullies present a formidable obstacle, the banks being as high as twenty feet in places. A smaller torrent, the Rapendosa, descends from the hills between Agrieliki and Aphorismos and disappears about a third of the way across the plain. A torrent such as the Charadra will certainly not follow the same course for two and a half millennia, and in any case the central part of the plain appears to have risen about ten feet since antiquity, for the most part because of material brought down from the hills by these torrents.[vii] It is thus impossible to determine exactly where the gullies ran at the time of the battle or whether they existed at all, but since neither Herodotus nor Pausanias makes any mention of this terrain feature, it is safe to say that if it did exist, it had no impact on the battle. The plain is well watered, two fault lines producing a number of springs, the principal ones being at Oinoe, Vrana, the eastern foot of Agrieliki (“Mati”) and at Kato Souli, between the eastern base of Stavrokoraki and the Great Marsh (“Megalo Mati” or “Makaria”). Wells are found all over the plain, providing most of the water for the region today, and Pausanias says there was fresh water flowing out of the Great Marsh.[viii]
Of the villages/demes of the Marathonian tetropolis three have been more or less securely located: Oinoe at the site of the modern village of that name, Trikorynthos at Kato Souli and Probalinthos at the eastern base of Agrieliki north of the Brexiza marsh (less securely). The site of Marathonitself is still disputed, but the scant archaeological evidence now points to a spot near the coast amidst the channels of the Charadra.[ix] There were in the fifth century three routes leading from Athens into the Marathon plain. The main road, apparently suitable for carts, ran for some twenty-six miles from Athens via Pallene to the area of the Soros and then continued northeast across the plain and on to Rhamnous. A second road led northeast from Athens to Kephisia, where it split into two paths, one passing through modern Stamata and Oinoe and on into the plain through the Avalona valley, the other heading through modern Dionysos and descending to Vrana through the gorge of the Rapendosa. Each of these routes is a bit more than twenty-three miles long, and both become fairly rough and steep tracks through wooded areas once they enter the hills around Marathon.[x]
Rising above the southern plain, approximately a mile northeast of the foot of Agrieliki and a half mile northwest from the coast, is the Soros, a thirty foot high artificial mound that is generally accepted as the burial place of the Athenians who fell in the battle.[xi] About four miles west of the Soros, at the site of the Marathon museum in Vrana, is a cluster of seven middle and late Helladic tumuli, and about 300 feet northeast of these lies a seventh mound, dated to the early fifth century. Within this tomb were found the remains of one juvenile and ten adult males, leading a few scholars to conclude that this is the tomb of the Plataeans mentioned by Pausanias.[xii] This identification is almost certainly incorrect, however. Pausanias’ catalogue of sights at Marathon proceeds in a more or less direct line northeasterly from the Soros to the stone “stables” of the Persian horse, and placing the Plataean tomb, which is mentioned immediately after that of the Athenians, at Vrana represents a three mile detour from this route. Further, the battle centered on the Soros, and it is difficult to see why the Plataean dead would be carried all the way to Vrana rather than being interred in the vicinity of the Athenians, where all could be conveniently visited and honored.[xiii] The tomb contents also argue against the identification: a mix of burial styles, the presence of a boy, very poor grave gifts and a single crude inscription in Attic lettering on an unworked stone.[xiv] It is far more likely that a low mound observed near the Soros in the nineteenth century marks the spot of the Plataean burial.[xv]
Following his notice of graves of the Athenians and Plataeans Pausanius mentions a monument to Miltiades and a trophy of white marble.[xvi] About 650 yards north of the Soros are the foundations of a tower (“Pyrgos”), possibly medieval, which according to nineteenth century travelers incorporated large blocks of white marble, now all gone. About a mile and a half to the northeast of these ruins, near the present church of Panagia Mesosporitissa, are the remains of another tower, also sporting ancient marble, including column drums and an Ionic capital. The two towers are likely to mark the approximate sites of the ancient monuments from which the marble was pilfered, since in the first case Leake observed actual marble foundations and in the second the number and size of the blocks argues against being moved any great distance. That these are the remains of the Miltiades monument and the battle trophy is a tempting conclusion since the fragments indicate monuments rather than buildings or enclosures and one would expect the Miltiades memorial to be in the vicinity of the burials.[xvii]
The last battle-related item Pausanias mentions before describing the Makaria spring and the Marsh are the Persian dead, whose burial place he could not find. He was, however, informed by the Athenians that they had been thrown into a trench, and in the nineteenth century von Eschenburg found in the area off the western edge of the Marsh huge quantities of bones (“viele Hunderte von Todten”), seemingly buried in a haphazard manner.[xviii]
The Persian fleet certainly anchored along the Schoenia. This section of the coast was the most amenable to the mooring of ships, and the Kynosoura promontory protected the anchorage from the dangerous northeast winds. Inasmuch as vessels were apparently beached only for protection or maintenance, the ships would have been anchored right at the water’s edge, sterns facing inland. Assuming no more than 300 vessels and no more than thirty feet of beach space per vessel (approximately the width of a trireme with oars extended), the fleet could be moored in a single line along the Schoenia, providing for the most convenient unloading and loading and for the quickest departure.[xix] This anchorage provided immediate access to the region of the Great Marsh and the most likely site of the Persian camp, the deme of Trikorynthos. Herodotus does not mention a Persian camp, as he does in the case of Plataea and Mykale, but inasmuch as the area around the Schoenia could not comfortably accommodate 50,000 or more men for several days there must have been a separate encampment, at least for the army.[xx] A camp on the plain immediately west of the marsh is possible, but the area to the north, bounded by the hills of Stavrokoraki and Drakonera and the marsh itself, offered excellent protection against attack and controlled the road to Rhamnous. Water was more plentiful in this locale, especially from the Makaria spring, and here the marsh was apparently deeper, remaining green longer into the autumn and thus providing more fodder for the horses.[xxi] And though it is barely evidence, Pausanias in fact identifies some excavations and marks in the hills beyond the marsh as the “stables” of Artaphernes’ horse and the marks from his tent.[xxii]
More problematic has been the location of the Greek camp. Herodotus says that upon arriving at Marathonthe Athenians established their camp “in the precinct of Herakles,”[xxiii] without however providing any indication where that might have been. In his eighth Pythian ode Pindar says the games of Herakles were held “in a/the nook/corner of Marathon,” but he may simply mean that Marathon was a corner or nook of Attica.[xxiv] In the 1930s an inscription containing regulations for games at the Herakleion was found just north of the Brexiza marsh, but even the finder of the stone, Soteriades, believed it had wandered, especially since it had been refaced.[xxv] Further, there was some evidence that the area near the marsh had been sacred to Athena Hellotis.[xxvi] He pointed instead to the Vrana valley, near the chapel of St. Demetrios, where he discovered what he believed to be the early 5th century remains of a sacred enclosure, which he identified as the Herakleion, noting that St. Demetrios would be an understandable successor to the pagan hero.[xxvii] Possessing springs, providing a secure position and covering both the main road along the coast and the back roads through Kephisia, the Vrana location, which is in fact in a sort of “nook,” was subsequently accepted by many, if not most writers, as the site of the Athenian camp.[xxviii]
This all changed with the discovery in 1972 of a dedication to Herakles, this stone found incorporated in a Roman building in the area of the Brexiza marsh. Certainly, two inscriptions regarding Herakles are very compelling, and this relatively narrow area between Agrieliki and the sea might well be described as a “nook or “corner” of Marathon. A camp here makes military sense, and the location fits perfectly the epithet the dedicatory inscription assigns to Herakles – “at the gate” – and the statement in one of the Marathon epigrams that the Athenians were “before the gates.”[xxix] While certainty is a commodity in very short supply at Marathon, the Athenian camp may now be fairly securely located at the southern entrance to the plain, and Herodotus’ account must be reconciled with this location.[xxx]
[i] A brief survey of the archaeology of the Marathon area and a list of the attendant literature up to 1988 can be found in Travlos 1988, 216-21.
[ii] Higgins & Higgins 1996, 33; Pritchett 1960, 156-57; see further note 7. The sea level of the Aegean also appears to have risen about 10 feet; see Pritchett 1959, 255-56.
[iii] Nep. Milt. 5.3: arbores multis locis errant rarae; Caspari 1926, 103 (followed by How & Wells 1912, II, 112) believes the Greek center was weakened in order to accommodate trees and vines, but fear of being outflanked was a far more compelling reason; see . The plain was relatively free of trees when observed by Frazer at the end of the nineteenth century; Frazer 1898, 433.
[iv] Woods: Aesch. Eleg. 3: Maraqw&nion a!lsoj; Paus. 1.14.5: to_ Maraqw~ni a!lsoj; marsh: Paus. 1.32.7: li&mnh ta_ polla_ e(lw&dhj. Recent geophysical examination suggests that the marsh was once a lake and before that a lagoon, and in an unpublished study Richard Dunn concludes that in 490 it was in fact a lake. Pausanias describes it as a “mostly marshy lake,” but that is over 600 years after the battle. On the other hand, he describes the Persian fugitives blundering into the marsh and suffering great casualties, which seems very unlikely were it simply a lake with marshy fringes. It is also unlikely that the channel connecting the marsh/lake to the sea was used by the Persian ships, assuming it was even navigable. Marsh would make the mooring and unloading of the vessels more difficult, and the ease with which almost all the ships escaped makes more sense were they on the beach; see also note 16. There is also evidence that the northern part of the shoreline was further inland and the souther further out in 490. On the marsh and coast see Kretnz 117, 214-15 and the map 155.
[v] Soteriades 1935, 120-21; Pritchett 1960, 152-54, 1965, 83-84; Themelis 1974, 239-41; Petrakos 1995, 68-86; Hammond 1973, 186-87 believes there was a marsh in antiquity because of the powerful springs in the area and because of scholia on Pindar claiming that Athena Hellotis was so named because of the marsh at Marathon, but the rise in the sea level better explains the emergence of a marsh and the scholia are extremely vague (e0n, peri&, pro_j) on the spatial location of the marsh to Marathon, which itself could be the deme, the town or the whole tetropolis. The stone attesting to a temenos of Athena was in fact found near the chapel of St. Demetrios, a mile and half north of the Brexisa (Vanderpool 1966b, 319-20), and the scholiast may simply be wrong, Hellotis with its double lambda deriving instead from Hellotia, a daughter of Timander.
[vi] See the admiralty chart in Hammond 1973, 218.
[vii] Burn 1966, 161-62 believes there was no Charadra in antiquity because deforestation of the surrounding hills had not yet occurred, but flashfloods in the area were already proverbial: Demon FGH 327 F 8 (= Strabo 8.6.16, Zen. 5.29, Suda s.v.): Oi0nai=oi th_n xara&dran. In 1828 Leake (see his map in Hammond 1973, 183) observed the two torrents following roughly the same courses they do today, and Soteriades 1935, 132-33 concluded the Charadra followed the same course in antiquity. But in the 1960s Pritchett confirmed the earlier reports of Staes that the level of the plain at the Soros had risen some ten feet, and this together with the sherd deposits around the torrent convinced him and two separate geologists that the present course of the torrent is not that of 490; a map made in 1792 in fact shows the Charadra following a different course; Pritchett 1960, 141-42, 156-57, 1969, 6.
[viii] Paus. 1.32.6; Pritchett 1965, 84-85; Petrakos 1995, 52-55. The depth of the water table, presently as little as 23 feet near the Brexiza marsh, increases as one moves inland towards Stavrokoraki, where it is now some 65 feet; though certainty is impossible because of the tectonic activity in the area, the water table was probably higher in antiquity, before another twenty-five hundred years of alluviation.
[ix] Placing Marathon near Plasi fits the order Probalinthos – Marathon – Trykorinthos given by Strabo 9.1.22, which traces demes north along the coast. The archaeological literature on the demes is extensive; see Travlos 1988, 220-21; the most recent and/or pertinent: Pritchett 1960, 149-52, 1965, 83-88, 1969, 1-11; Vanderpool 1966b, 319-22; Marinatos 1970a, 5-9, 1972, 6-7; Themelis 1974, 229-35, 239-42; Traill 1986, 146-48. Petrakos 1995, 1-2 suggests that there was no village of Marathon but rather houses scattered about the plain, but such would constitute a big exception in Pausanias’ itinerary.
[x] In 1996 I could not find the head of the Oinoe trail, but did climb the Rapendosa track for about a half mile; it was extremely steep and rough. A description of the tracks can be found in Frazer 1898, 441-42. The excellent road observed between Stamata and Marathona by Clarke in 1801 cannot be earlier than the late fifth century; Ober 1982, 457-58.
[xi] Paus. 1.29.4, 32.3 is clearly referring to this tomb. Schliemann 1884, 85-88 believed it to be prehistoric, but the excavations of Staes confirmed the date of 490; see esp. Staes 1893, 46-63; Hammond 1973, 172-78. Pausanias describes only a grave (τάφς) with stone slabs (στh=λαι) inscribed with the names of the dead, which slabs G. Spyropoulos has claimed to have recently found in the villa of Herodes Atticus in the Peloponnesus. He consequentlty suggests that it was Herodes who erected the mound when he purloined the inscriptions of the Marathon dead, which is barely possible since Pausanias would have passed through Marathon before AD 174 and Herodes died in 177; report in Αρχαιολογία Archaeology Newsroom 8 May 2009; the book mentioned in the report, Die Architektur der Villa des Herodes Atticus zu Eva/Loukou, can not be located and is perhaps classified.
[xii] Paus. 1.32.3; Marinatos 1970a, 9-28, 1970b, 155-66, 1970c, 351-66; Hammond 1973, 197-98; Burn 91-92.
[xiii] Pritchett 1985, 129.
[xiv] See esp. Welwei 1979, 101-6, who suggests these might be the remains of scouts who were surprised and killed by the Persians; Themelis 1974, 244 believes they are normal local burials despite the absence of any females.
[xv] Clarke 1818, 27-28; Leake 1841, 101; Pritchett 1985, 128.
[xvi] Paus. 1.32.4-5.
[xvii] Miltiades monument: Leake 1841, 101; trophy: Vanderpool 1966a.
[xviii] Paus. 1.32.5: o1rugma. Von Eschenburg 1886, 10.
[xix] That the Stoa Poikile paintings (Paus. 1.15.3) show Persians fleeing into the marsh between illustrations of the battle and the fighting at the ships places the fleet at the Schoenia. On mooring the ships see Harrison 1999, 168-71; Whitehead 1993, 95-98; Herod. 6.107.2 says the ships were “moored” or “anchored”: ta_j ne&aj o#rmize. Herod. 6.114.1 shows a Persian ship stern-first, and the fact that the fleet got away so quickly suggests stern-first mooring. The Brescia sarcophagus, thought to reproduce the scenes in the Stoa Poilile, shows ships moored stern-first; see Vanderpool 1966a, pl. 35.
[xx] Following a suggestion of Macan 1895, II, 244, n. 8, van der Veer 1982, 398-99 believes there was no Persian camp.
[xxi] Most modern authors place the camp around Trykorynthos, but some have it west of the marsh: Macan 1895, II, 244-45; Munro 1926, 242; Schachermeyr 1951, 18-19; Vanderpool 1966b, 323; et. al.; Shrimpton 1980, 30-31 curiously places it near the Soros. Shrimpton 1980, 31, n. 23 argues that the pasturage available at the marsh would be dangerous to horses fed on hay and grain, but surely the Persian horse-handlers would approach this change in feeding very carefully, and in any case Trykorinthos provided easy access to water and better security after dark. See Frazer 1898, 432 for a description of the marsh before it was drained.
[xxii] Paus. 1.32.7; Frazer 1898, 432 observed “niche-like excavations” on Stavrokoraki; Leake 1841, 96 found a “small cavern” on Drakonera. Since the Persians are unlikely to have engaged in excavating rock, it is likely the story of the stables and tent later attached themselves to one or the other of these formations.
[xxiii] Herod. 6.108.1: e0n teme/nei+ 9Hrakleo/v.
[xxv] IG i.3 3 (=SEG x.2): hερακλείο[ισι]; see Vanderpool 1942, 329-37, 1966b, 322-23, 1984.
[xxvi] See note 5.
[xxvii] Soteriades 1935.
[xxviii] Including me: Berthold 1976-1977, 88-91; some of the others: How & Wells II, 109; Delbrück 1920, 54; Munro 1926, 241-42; Maurice 1932, 21; Pritchett 138-40; Burn 243; Hammond 189-90; Billows 208, who apparently did not get the memo on the second inscription.
[xxix]IG i.3 1015bis: ερακλεῖ…τὸμ Πυλίοις ἀνέθεκε ερακ[λ ˘ ˘ ¯ ˘ ˘ ¯ ˘]; see Marinatos 1972, 6; Kamanoudes 1978, 237-42; esp Matthaiou 2003, 190-94. IG i.3 503/4: αἰχμὲν / στε̑σαμ πρόσθε πυλο̑ν; see Matthaiou 2003, 194-97.
[xxx] Virtually everyone now accepts the location at the entrance: e.g., Burn 1977, 90-91; van der Veer 1982, 96-97; Evans 292; Lazenby 54-56; Krentz 118-21.