Primitive Athens as described by Thucydides

by Jane Ellen Harrison

INTRODUCTORY.

The traveller who visits Athens for the first time will naturally, if he be a classical scholar, devote himself at the outset to the realization of the city of Perikles. His task will here be beset by no serious difficulties. The Acropolis, as Perikles left it, is, both from literary and monumental evidence, adequately known to us. Archaeological investigation has now but little to add to the familiar picture, and that little in matters of quite subordinate detail. The Parthenon, the Propylaea, the temple of Nike Apteros, the Erechtheion (this last probably planned, though certainly not executed by Perikles) still remain to us; their ground-plans and their restorations are for the most part architectural certainties. Moreover, even outside the Acropolis, the situation and limits of the city of Perikles are fairly well ascertained. The Acropolis itself was, we know, a fortified sanctuary within a larger walled city. This city lay, as the oracle in Herodotus[1] said, ‘wheel-shaped’ about the axle of the sacred hill. Portions of this outside wall have come to light here and there, and the foundations of the great Dipylon Gate are clearly made out, and are marked in every guide-book. Inside the circuit of these walls, in the inner Kerameikos, whose boundary-stone still remains, lay the agora. Outside is still to be seen, with its street of tombs, the ancient cemetery.

Should the sympathies of the scholar extend to Roman times, he has still, for the making of his mental picture, all the help imagination needs. Through the twisted streets of modern Athens the beautiful Tower of the Winds is his constant land-mark; Hadrian, with his Olympieion, with his triumphal Arch, with his Library, confronts him at every turn; when he goes to the great Stadion to see ‘Olympian’ games or a revived ‘Antigone,’ when he looks down from the Acropolis into the vast Odeion, Herodes Atticus cannot well be forgotten. Moreover, if he really cares to know what Athens was in Roman days, the scholar can leave behind him his Murray and his Baedeker and take for his only guide the contemporary of Hadrian, Pausanias.

But returning, as he inevitably will, again and again to the Acropolis, the scholar will gradually become conscious, if dimly, of another and an earlier Athens. On his plan of the Acropolis he will find marked certain fragments of very early masonry, which, he is told, are ‘Pelasgian.’ As he passes to the south of the Parthenon he comes upon deep-sunk pits railed in, and within them he can see traces of these ‘Pelasgian’ walls and other masonry about which his guide-book is not over-explicit. To the south of the Propylaea, to his considerable satisfaction, he comes on a solid piece of this ‘Pelasgian’ wall, still above ground. East of the Erechtheion he will see a rock-hewn stairway which once, he learns, led down from the palace of the ancient prehistoric kings, the ‘strong house of Erechtheus.’ South of the Erechtheion he can make out with some effort the ground plan of an early temple; he is told that there exist bases of columns belonging to a yet earlier structure, and these he probably fails to find.

With all his efforts he can frame but a hazy picture of this earlier Acropolis, this citadel before the Persian wars. Probably he might drop the whole question as of merely antiquarian interest—a matter to be noted rather than realized—but that his next experience brings sudden revelation. Skilfully sunk out of sight—to avoid interfering with his realization of Periklean Athens—is the small Acropolis Museum. Entering it, he finds himself in a moment actually within that other and earlier Athens dimly discerned, and instantly he knows it, not as a world of ground-plans and fragmentary Pelasgic fortifications, but as a kingdom of art and of humanity vivid with colour and beauty.

As he passes in eager excitement through the ante-rooms he will glance, as he goes, at the great blue lion and the bull, at the tangle of rampant many-coloured snakes, at the long-winged birds with their prey still in beak and talon; he will pause to smile back at the three kindly ‘Blue-beards,’ he will be glad when he sees that the familiar Calf-Carrier has found his feet and his name, he will note the long rows of solemn votive terra-cottas, and, at last, he will stand in the presence of those Maiden-images, who, amid all that coloured architectural splendour, were consecrate to the worship of the Maiden. The Persian harried them, Perikles left them to lie beneath his feet, yet their antique loveliness is untouched and still sovran. They are alive, waiting still, in hushed, intent expectancy—but not for us. We go out from their presence as from a sanctuary, and henceforth every stone of the Pelasgian fortress where they dwelt is, for us, sacred.

But if he leave that museum aglow with a new enthusiasm, determined to know what is to be known of that antique world, the scholar will assuredly be met on the threshold of his enquiry by difficulties and disillusionment. By difficulties, because the information he seeks is scattered through a mass of foreign periodical literature, German and Greek; by disillusionment, because to the simple questions he wants to ask he can get no clear, straightforward answer. He wants to know what was the nature and extent of the ancient city, did it spread beyond the Acropolis, if so in what direction and how far? what were the primitive sanctuaries inside the Pelasgic walls, what, if any, lay outside and where? Where was the ancient city well (Kallirrhoë), where the agora, where that primitive orchestra on which, before the great theatre was built, dramatic contests took place? Straightway he finds himself plunged into a very cauldron of controversy. The ancient agora is placed by some to the north, by others to the south, by others again to the west. The question of its position is inextricably bound up, he finds to his surprise, with the question as to where lay the Enneakrounos, a fountain with which hitherto he has had no excessive familiarity; the mere mention of the Enneakrounos brings either a heated discussion or, worse, a chilling silence.

This atmosphere of controversy, electric with personal prejudice, exhilarating as it is to the professed archaeologist, plunges the scholar in a profound dejection. His concern is not jurare in verba magistri—he wants to know not who but what is right. Two questions only he asks. First, and perhaps to him unduly foremost, What, as to the primitive city, is the literary testimony of the ancients themselves, and preferably the testimony not of scholiasts and second-hand lexicographers, but of classical writers who knew and lived in Athens, of Thucydides, of Pausanias? Second, To that literary testimony, what of monumental evidence has been added by excavation?

It is to answer these two questions that the following pages are written. It is the present writer’s conviction that controversy as to the main outlines of the picture, though perhaps at the outset inevitable, is, with the material now accessible, an anachronism; that the facts stand out plain and clear and that between the literary and monumental evidence there is no discrepancy. The plan adopted will therefore be to state as simply as may be what seems the ascertained truth about the ancient city, and to state that truth unencumbered by controversy. Then, and not till then, it may be profitable to mention other current opinions, and to examine briefly what seem to be the errors in method which have led to their acceptance.

CHAPTER I.
THE ANCIENT CITY, ITS CHARACTER AND LIMITS.

By a rare good fortune we have from Thucydides himself an account of the nature and extent of the city of Athens in the time of the kingship. This account is not indeed as explicit in detail as we could wish, but in general outline it is clear and vivid. To the scholar the remembrance of this account comes as a ray of light in his darkness. If he cannot find his way in the mazes of archaeological controversy, it is at least his business to read Thucydides and his hope to understand him.

The account of primitive Athens is incidental. Thucydides is telling how, during the Peloponnesian War, when the enemy was mustering on the Isthmus and attack on Attica seemed imminent, Perikles advised the Athenians to desert their country homes and take refuge in the city. The Athenians were convinced by his arguments. They sent their sheep and cattle to Euboea and the islands; they pulled down even the wood-work of their houses, and themselves, with their wives, their children, and all their moveable property, migrated to Athens. But, says Thucydides[2], this ‘flitting’ went hard with them; and why? Because ‘they had always, most of them, been used to a country life.’

This habit of ‘living in the fields,’ this country life was, Thucydides goes on to explain, no affair of yesterday; it had been so from the earliest times. All through the days of the kingship from Kekrops to Theseus the people had lived scattered about in small communities—‘village communities’ we expect to hear him say, for he is insisting on the habit of country life; but, though he knows the word ‘village’ (κώμη) and employs it in discussing Laconia elsewhere[3], he does not use it here. He says the inhabitants of Athens lived ‘in towns’ (κατὰ πόλεις), or, as it would be safer to translate it, ‘in burghs.’

It is necessary at the outset to understand clearly what the word polis here means. We use the word ‘town’ in contradistinction to country, but from the account of Thucydides it is clear that people could live in a polis and yet lead a country life. Our word city is still less appropriate; ‘city’ to us means a very large town, a place where people live crowded together. A polis, as Thucydides here uses the word, was a community of people living on and immediately about a fortified hill or citadel—a citadel-community. The life lived in such a community was essentially a country life. A polis was a citadel, only that our word ‘citadel’ is over-weighted with military association.

Athens then, in the days of Kekrops and the other kings down to Theseus, was one among many other citadel-communities or burghs. Like the other scattered burghs, like Aphidna, like Thoricus, like Eleusis, it had its own local government, its own council-house, its own magistrates. So independent were these citadel-communities that, Thucydides tells us, on one occasion Eleusis under Eumolpos actually made war on Athens under Erechtheus.

So things went on till the reign of Theseus and his famous Synoikismos, the Dwelling-together or Unification. Theseus, Thucydides says, was a man of ideas and of the force of character necessary to carry them out. He substituted the one for the many; he put an end to the little local councils and council-houses and centralized the government of Attica in Athens. Where the government is, thither naturally population will flock. People began to gather into Athens, and for a certain percentage of the population town-life became fashionable. Then, and not till then, did the city become ‘great,’ and that ‘great’ city Theseus handed down to posterity. ‘And from that time down to the present day the Athenians celebrate to the Goddess at the public expense a festival called the Dwelling-together[4].’

One unified city and one goddess, the goddess who needs no name. Their unity and their greatness the Athenians are not likely to forget, but will they remember the time before the union, when Athens was but Kekropia, but one among the many scattered citadel-communities? Will they remember how small was their own beginning, how limited their burgh, how impossible—for that is the immediate point—that it should have contained in its narrow circuit a large town population? Thucydides clearly is afraid they will not. There was much to prevent accurate realization. The walls of Themistocles, when Thucydides wrote, enclosed a polis that was not very much smaller than the modern town; the walls of the earlier community, the old small burgh, were in part ruined. It was necessary therefore, if the historian would make clear his point, namely, the smallness of the ancient burgh and its inadequacy for town-life, that he should define its limits. This straightway he proceeds to do. Our whole discussion will centre round his definition and description, and at the outset the passage must be given in full. Immediately after his notice of the festival of the ‘Dwelling-together,’ celebrated to ‘the Goddess,’ Thucydides[5] writes as follows:

‘Before this, what is now the citadel was the city, together with what is below it towards about south. The evidence is this. The sanctuaries are in the citadel itself, those of other deities as well[6] (as the Goddess). And those that are outside are placed towards this part of the city more (than elsewhere). Such are the sanctuary of Zeus Olympios, and the Pythion, and the sanctuary of Ge, and that of Dionysos-in-the-Marshes (to whom is celebrated the more ancient Dionysiac Festival on the 12th day in the month Anthesterion, as is also the custom down to the present day with the Ionian descendants of the Athenians); and other ancient sanctuaries also are placed here. And the spring which is now called Nine-Spouts, from the form given it by the despots, but which formerly, when the sources were open, was named Fair-Fount—this spring (I say), being near, they used for the most important purposes, and even now it is still the custom derived from the ancient (habit) to use the water before weddings and for other sacred purposes. Because of the ancient settlement here, the citadel (as well as the present city) is still to this day called by the Athenians the City.’

In spite of certain obscurities, which are mainly due to a characteristically Thucydidean over-condensation of style, the main purport of the argument is clear. Thucydides, it will be remembered, wants to prove that the city before Theseus was, because of its small size, incapable of holding a large town population. This small size not being evident to the contemporaries of Thucydides, he proceeds to define the limits of the ancient city. He makes a statement and supports it by fourfold evidence.

The statement that he makes is that the ancient city comprised the present citadel together with what is below it towards about south. The fourfold evidence is as follows:

1. The sanctuaries are in the citadel itself, those of other deities as well as the Goddess.

2. Those ancient sanctuaries that are outside are placed towards this part of the present city more than elsewhere. Four instances of such outside shrines are adduced.

3. There is a spring near at hand used from of old for the most important purposes, and still so used on sacred occasions.

4. The citadel, as well as the present city, was still in the time of Thucydides called the ‘city.’

We begin with the statement as to the limits of the city. Not till we clearly understand exactly what Thucydides states, how much and how little, can we properly weigh the fourfold evidence he offers in support of his statement.

‘Before this what is now the citadel was the city, together with what is below it towards about south.’ The city before Theseus was the citadel or acropolis of the days of Thucydides, plus something else. The citadel or acropolis needed then, and needs now, no further definition. By it is clearly meant not the whole hill to the base, but the plateau on the summit enclosed by the walls of Themistocles and Kimon together with the fortification out-works on the west slope still extant in the days of Thucydides. But the second and secondary part of the statement is less clearly defined. The words neither give nor suggest, to us at least, any circumscribing line; only a direction, and that vague enough, ‘towards about south.’ It is a point at which the scholar naturally asks, whether archaeology has anything to say?

But before that question is asked and answered, it should be noted that from the shape of the sentence alone something may be inferred. That the present citadel is coextensive with the old city is the main contention. We feel that Thucydides might have stopped there and yet made his point, namely, the smallness of that ancient city. But Thucydides is a careful man, he remembers that the two were not quite coextensive. To the old city must be reckoned an additional portion below the citadel (τὸ ὑπ’ αὐτήν), a portion that, as will later be seen, his readers might be peculiarly apt to forget; so he adds it to his statement. But, by the way it is hung on, we should naturally figure that portion as ‘not only subordinate to the acropolis, but in some way closely incorporated with it. In relation to the acropolis, this additional area, to justify the arrangement of the words of Thucydides, should be a part neither large nor independent[7].’

Thus much can be gathered from the text; it is time to see what additional evidence is brought by archaeology.

Thucydides was, according to his lights, scrupulously exact. It happens, however, that in the nature of things he could not, as regards the limits of the ancient city, be strictly precise. The necessary monuments were by his time hidden deep below the ground. His first and main statement, that one portion of the old city was coextensive with the citadel of his day, is not quite true. This upper portion of the old burgh was a good deal smaller; all the better for his argument, had he known it! Thanks to systematic excavation we know more about the limits of the old city than Thucydides himself, and it happens curiously enough that this more exact and very recent knowledge, while it leads us to convict Thucydides of a real and unavoidable inexactness, gives us also the reason for his caution. It explains to us why, appended to his statement about the city and the citadel, he is careful to put in the somewhat vague addendum, ‘together with what is below it towards about south.’

To us to-day the top of the Acropolis appears as a smooth plateau sloping gently westwards towards the Propylaea, and this plateau is surrounded by fortification walls, whose clean, straight lines show them to be artificial. Very similar in all essentials was the appearance presented by the hill to the contemporaries of Thucydides, but such was not the ancient Acropolis. What manner of thing the primitive hill was has been shown by the excavations carried on by the Greek Government from 1885-1889. The excavators, save when they were prevented by the foundations of buildings, have everywhere dug down to the living rock, every handful of the débris exposed has been carefully examined, and nothing more now remains for discovery.

When the traveller first reaches Athens he is so impressed by the unexpected height and dominant situation of Lycabettus, that he wonders why it plays so small a part in classical record. Plato[8] seems to have felt that it was hard for Lycabettus to be left out. In his description of primitive Athens he says, ‘in old days the hill of the Acropolis extended to the Eridanus and Ilissus, and included the Pnyx on one side and Lycabettus as a boundary on the opposite side of the hill,’ and there is a certain rough geological justice about Plato’s description. All these hills are spurs of that last offshoot of Pentelicus, known in modern times as Turkovouni. Yet to the wise Athena, Lycabettus was but building material; she was carrying the hill through the air to fortify her Acropolis, when she met the crow[9] who told her that the disobedient sisters had opened the chest, and then and there she dropped Lycabettus and left it ... to the crows.

A moment’s reflection will show why the Acropolis was chosen and Lycabettus left. Lycabettus is a good hill to climb and see a sunset from. It has not level space enough for a settlement. The Acropolis has the two desiderata of an ancient burgh, space on which to settle, and easy defensibility.

The Acropolis, as in neolithic days the first settlers found it, was, it will be seen in [Fig. 1], a long, rocky ridge, broken at intervals[10]. It could only be climbed with ease on the west and south-west sides, the remaining sides being everywhere precipitous, though in places not absolutely inaccessible. For a primitive settlement it was an ideal situation. Two things remained for the settlers to do: first, they had to level the surface by hewing away jagged rocks and filling up cracks with earth and stones to make sites for their houses and their sanctuaries; and second, they had to supplement what nature had already done in the way of fortification; here and there to make the steep rocks steeper, build a wall round their settlement, and, above all, fortify that accessible west and south-west end and build an impregnable gateway. Kleidemos[11], writing in the fifth century B.C., says, ‘they levelled the Acropolis and made the Pelasgicon, which they built round it nine-gated.’ They levelled the surface, they built a wall round it, they furnished the fortification wall with gates. We begin for convenience sake with the wall. In tracing its course the process of levelling is most plainly seen. The question of the gates will be taken last.

Fig. 1.

In the plan in [Fig. 2] is shown what excavations have laid bare of the ancient Pelasgic fortress. We see instantly the inexactness of the main statement of Thucydides. It is not ‘what is now the Citadel’ that was the main part of the old burgh, but something substantially smaller, smaller by about one-fifth of the total area. We see also that this Thucydides could not know. The Pelasgic wall following the broken outline of the natural rock was in his days covered over by the artificial platform reaching everywhere to the wall of Kimon. At one place, and one only, in the days of Thucydides, did the Pelasgic wall come into sight, and there it still remains above ground, as it has always been, save when temporarily covered by Turkish out-works. This visible piece is the large fragment (A), 6 metres broad, to the south of the present Propylaea and close to the earlier gateway (G). In the days of Thucydides it stood several metres high. Of this we have definite monumental evidence. The south-east corner of the wall of the south-west wing of the present Propylaea is bevelled away[12] so as to fit against this Pelasgic wall, and the bevelling can be seen to-day. This portion of the Pelasgic wall is of exceptional strength and thickness, doubtless because it was part of the gateway fortifications, the natural point of attack.

Fig. 2.

Save for this one exception, the Pelasgic walls lie now, as they did in the day of Thucydides, below the level of the present hill, and their existence was, until the excavations began, only dimly suspected. Literary tradition said there was a circuit wall, but where this circuit wall ran was matter of conjecture; bygone scholars even placed it below the Acropolis. Now the outline, though far from complete, is clear enough. To the south and south-west of the Parthenon there are, as seen on the plan, substantial remains and what is gone can be easily supplied. On the north side the remains are scanty. The reason is obvious; the line of the Pelasgic fortification on the south lies well within the line of Kimon’s wall; the Pelasgic wall was covered in, but not intentionally broken down. To the north it coincided with Themistocles’ wall, and was therefore, for the most part, pulled down or used as foundation.

But none the less is it clear that the centre of gravity of the ancient settlement lay to the north of the plateau. Although the north wall was broken away, it is on this north side that the remains which may belong to a royal palace have come to light. The plan of these remains cannot in detail be made out, but the general analogy of the masonry to that of Tiryns and Mycenae leave no doubt that here we have remains of ‘Mycenaean’ date. North-east of the Erechtheion is a rock-cut stairway (B) leading down through a natural cleft in the rock to the plain below. As at Tiryns and Mycenae, the settlement on the Acropolis had not only its great entrance-gates, but a second smaller approach, accessible only to passengers on foot, and possibly reserved for the rulers only.

Incomplete though the remains of this settlement are, the certain fact of its existence, and its close analogy to the palaces of Tiryns and Mycenae are of priceless value. Ancient Athens is now no longer a thing by itself; it falls into line with all the other ancient ‘Mycenaean’ fortified hills, with Thoricus, Acharnae, Aphidna, Eleusis. The citadel of Kekrops is henceforth as the citadel of Agamemnon and as the citadel of Priam. The ‘strong house’ of Erechtheus is not a temple, but what the words plainly mean, the dwelling of a king. Moreover we are dealing not with a city, in the modern sense, of vague dimensions, but with a compact fortified burgh.

Thucydides, though certainly convicted of some inexactness as to detail, is in his main contention seen to be strictly true—‘what is now the citadel was the city.’ Grasping this firmly in our minds we may return to note his inexactness as to detail. By examining certain portions of the Pelasgic wall more closely, we shall realize how much smaller was the space it enclosed than the Acropolis as known to Thucydides.

Fig. 3.

The general shape of the hill, and its subsequent alteration, are best realized by Dr Dörpfeld’s simple illustration[13]. A vertical section of the natural rock, it is roughly of the shape of a house ([Fig. 3]) with an ordinary gable roof. The sides of the house represent the steep inaccessible cliffs to north and south and east; the lines of the roof slope like the lines of the upper part of the hill converging at the middle. Suppose the sides of the house produced upwards to the height of the roof-ridge, and the triangular space so formed filled in, we have the state of the Acropolis when Kimon’s walls were completed. The filling in of those spaces is the history of the gradual ‘levelling of the surface of the hill, the work of many successive generations.’ The section in [Fig. 4] will show that this levelling up had to be done chiefly on the north and south sides; to the east and west the living rock is near the surface.

Fig. 4.

It has already been noted that on the north side of the Acropolis the actual remains of the Pelasgian wall are few and slight; but as the wall of Themistocles which superseded it follows the contours of the rock, we may be sure that here the two were nearly coincident. The wall of Themistocles remains to this day a perpetual monument of the disaster wrought by the Persians. Built into it opposite the Erechtheum, not by accident, but for express memorial, are fragments of the architrave, triglyphs and cornice of poros stone, and the marble metopes, from the old temple of Athena which the Persians had burnt. Other memorials lay buried out of sight, and were brought to light by the excavations of 1886. The excavators[14] were clearing the ground to the north-east of the Propylaea. On the 6th of February, at a depth of from 3-4½ metres below the surface, they came upon fourteen of the ‘Maidens[15].’ The section[16] in [Fig. 5] shows the place where they had slept their long sleep. We should like to think they were laid there in all reverence for their beauty, but hard facts compel us to own that, though their burial may have been prompted in part by awe of their sanctity, yet the practical Athenian did not shrink from utilizing them as material to level up with.

The deposit, it is here clearly seen, was in three strata. Each stratum consisted of statues and fragments of statues, inscribed bases, potsherds, charred wood, stones, and earth. Each stratum, and this is the significant fact, is separated from the one above it by a thin layer of rubble, the refuse of material used in the wall of Themistocles. The conclusion to the architect is manifest. In building the wall, perhaps to save expense, no scaffolding was used; but, after a few courses were laid, the ground inside was levelled up, and for this purpose what could be better than the statues knocked down by the Persians? Headless, armless, their sanctity was gone, their beauty uncared for. In the topmost of the three strata—the stratum which yielded the first find of ‘Maidens’—a hoard of coins was found: thirty-five Attic tetradrachms, two drachmas, and twenty-three obols. All are of Solon’s time except eight of the obols, which date somewhat earlier. Besides the ‘Maidens,’ on this north side of the Acropolis other monuments came to light, many bronzes, and among them the lovely flat Athena[17], the beautiful terra-cotta plaque[18] painted with the figure of a hoplite, and countless votive terra-cottas.

Fig. 5.

The excavations on the south side of the Acropolis have yielded much that is of great value for art and for science, for our knowledge of the extent of the Pelasgian fortification, results of the first importance. The section in [Fig. 7], taken at the south-east corner of the Parthenon, shows the state of things revealed. The section should be compared with the view in [Fig. 6].

Fig. 6.

Fig. 7.

The masonry marked 2 is the foundation, deep and massive beyond all expectation, laid, not for the Parthenon as we know it, but for that earlier Parthenon begun before the Persian War, and fated never to be completed. At 4 we see the great Kimonian wall as it exists to-day, though obscured by its mediaeval casing. All this, if we want to realize primitive Athens, we must think away. The date of Kimon’s wall is of course roughly fixed as shortly after 469 B.C., the foundations of the early Parthenon are certainly before the Persian War, probably after the date of Peisistratos. We may probably, though not quite certainly, attribute them to the time of the first democracy, the activity of Kleisthenes[19], a period that saw the building of the theatre-shaped Pnyx, the establishment of the new agora in the Kerameikos, and the Stoa of the Athenians at Delphi. Laurium had just begun to yield silver from her mines. Themistocles, before and after the war, was all for fortification; the Alkmaeonid Kleisthenes may well have indulged an hereditary tendency to temple building.

Save for the clearing of our minds, the date of the early temple-foundations does not immediately concern us. Their importance is that, but for the building of the Parthenon, early and late, we should never apparently have had the great alteration and addition to the south side of the hill and the ancient Pelasgian wall would never have been covered in. Let us see how this happened[20].

We start with nothing but the natural rock, and on it the Pelasgian wall (1). Over the natural rock is a layer of earth, marked I. Whatever objects have been found in that layer date before the laying of the great foundations; these objects are chiefly fragments of pottery, many of them of ‘Mycenean’ character, and some ordinary black-figured vases.

It is decided to build a great temple, and the foundations are to be laid. The ground slopes away somewhat rapidly, so the southern side of the temple is to be founded on an artificial platform. The trench (b) is dug in the layer of earth; then, just as on the north side of the hill, no scaffolding is used, but as the foundations are laid course by course, the débris is used as a platform for the workmen. A supporting wall (2) is required and built of polygonal masonry; it rises course by course, corresponding with the platform of débris. And then, what might have been expected but was apparently not foreseen, happens. The slender wall can be raised no higher and at about the second course the débris unsupported pours over it, as seen at III.

The débris, unchecked, fell over as far as the old Pelasgian wall. How high this originally stood it is not possible now to say; but, from the fact that outside the supporting wall the layers of débris again lie horizontally, and from the analogy of another section taken further west, which need not be discussed here, it is probable that the old wall was raised by several new courses, and that the higher ones were of quadrangular blocks, as restored in [Fig. 7].

So far all that has been accomplished is the raising of the old Pelasgian wall and a levelling up of the terrace to its new height. That these terraces were raised step by step with the foundations of the Parthenon is clear. Between each layer of earth and poros fragments—just as we have seen in the similar circumstances of the north wall ([p. 15])—is interposed a layer of splinters and fragments of the stones used in the building of the foundations. This can clearly be seen at II. in the section in [Fig. 7].

It may seem strange that Kleisthenes, or whoever built the earlier Parthenon, did not at once utilize the Pelasgian wall and boldly pile up his terrace against its support. But it must be remembered that the space between the Parthenon and the Pelasgian wall was very great; an immense amount of débris would be required for the filling up of such a space, and it was probably more economical to build the polygonal supporting-wall nearer to the Parthenon. Anyhow it is quite clear that the polygonal wall was no provisional structure. Its façade shows it was meant to be seen, and that the terrace was meant for permanent use is clear from the fact that it is connected by a flight of steps with the lower terrace under the Pelasgian wall ([Fig. 8]). It is clear that whoever planned these steps never thought that the lower terrace would be levelled up.

Doubtless whoever filled in the terrace to the height of the raised Pelasgian wall believed in like manner that his work was complete. But Kimon thought otherwise. We know for certain that it was he who built the great final wall, the structure that remains to-day, though partly concealed by mediaeval casing [Fig. 7] (4). Plutarch[21] tells us that after the battle of Eurymedon (469 B.C.) so much money was raised by the sale of the spoils of the Persians that the people were able to afford to build the south wall. We know also that this wall of Kimon was at least as much a retaining wall to the great terrace as a fortification. For the filling up of the space between the Pelasgian fortification and his own wall Kimon had material sadly ample. He had the débris left by the Persians after the sacking of the Acropolis. The fragments of sculpture and architecture that bear traces of fire are found in the strata marked IV, and there only, for it is these strata only that were laid down after the Persian War[22]. The last courses of ‘Kimon’s wall’ (5) were laid by Perikles, and he it was who finally filled in the terrace to its present level (V).

Fig. 8.

The relation of the successive walls and terraces is shown by the ground-plan in [Fig. 9][23]. The double shaded lines from A to E and D show the irregular course of the old Pelasgic wall. The dotted lines from B to F show the polygonal supporting wall of the first terrace. It ran, as is seen, nearly parallel to the Parthenon. Its course is lost to sight after it passes under the new museum, but originally it certainly joined the Pelasgic wall at C. At B was the stairway joining the two terraces. Next came the time when, as the rubble fell over the wall, larger space was needed, and a portion of the Pelasgic wall was utilized and raised. This is shown by the thick black line from B to E coincident with the Pelasgic wall; the masonry here was of quadrangular poros blocks. The coincidence with the Pelasgic wall was only partial. At GH there jutted out an independent angular outpost, and again at EF the new wall is separate from the old; at FD it coincided with the earlier polygonal terrace wall. Kimon’s wall is indicated by the outside double lines, and in the space between these lines and the wall HEK lay the débris of the Persian War. Above that débris lay a still later stratum, deposited during the building operations of Perikles.

Fig. 9.

The various terraces and walls have been examined somewhat in detail, because their examination helps us to realize as nothing else could how artificial a structure is the south side of the Acropolis, and also—a point, to us, of paramount importance—how different was the early condition of the hill from its later appearance.

Before we pass to the consideration of the second clause in the historian’s statement, ‘together with what is below it towards about south,’ it is necessary to say a word as to when the old fortress walls were built and by whom. Kimon and Themistocles we know, but who were these earlier master-builders?

A red-figured vase painter of the fifth century B.C. gives us what would have seemed to a contemporary Athenian a safe and satisfactory answer—‘There were giants in those days.’ The design in [Fig. 10] is from a skyphos[24] in the Louvre Museum. Athena is about to fortify her chosen hill. She wears no aegis, for her work is peaceful; she has planted her spear in the ground perhaps as a measuring rod, and she has chosen her workman. A great giant, his name Gigas, inscribed over him, toils after her, bearing a huge ‘Cyclopean’ rock. She points with her hand where he is to lay it.

Fig. 10.

On the obverse of the same vase ([Fig. 11]) we have a scene of similar significance. To either side of a small tree, which marks the background as woodland, stands a man of rather wild and uncouth appearance. The man to the left is bearded and his name is inscribed, Phlegyas. The right-hand man is younger, and obviously resembles the giant of the obverse. He is showing to Phlegyas an object, which they both inspect with an intent, puzzled air. And well they may. It is a builder’s staphyle[25], or measuring line, weighted with knobs of lead like a cluster of grapes; hence its name. Phlegyas[26] and his giant Thessalian folk were the typical lawless bandits of antiquity; they plundered Delphi, they attacked Thebes after it had been fortified by Amphion and Zethus. But Athena has them at her hest for master-builders. All glory to Athena!

Fig. 11.

It is not only at Athens that legends of giant, fabulous workmen cluster about ‘Mycenean’ remains. Phlegyas and his giants toil for Athena, and at Tiryns too, according to tradition, the Kyklopes work for King Proetus[27], and they too built the walls and Lion-Gate of Mycenae[28]. At Thebes the Kadmeia[29] is the work of Amphion and Zethus, sons of the gods, and the fashion in which art represents Zethus as toiling is just that of our giant on the vase. The mantle that Jason wore was embroidered, Apollonius of Rhodes[30] tells us, with the building of Thebes,

Of river-born Antiope therein

The sons were woven, Zethus and his twin

Amphion, and all Thebes unlifted yet

Around them lay. They sought but now to set

The stones of her first building. Like one sore

In labour, Zethus on great shoulders bore

A stone-clad mountain’s crest; and there hard by

Amphion went his way with minstrelsy

Clanging a golden lyre, and twice as vast

The dumb rock rose and sought him as he passed.

Sisyphos, ancient king of Corinth, built on the acropolis of Corinth his great palace, the Sisypheion. He is the Corinthian double of Erechtheus with his Erechtheion. Strabo[31] was in doubt whether to call the Sisypheion palace or temple. Like the old Erechtheion, it was both fortress and sanctuary. In Hades for eternal remembrance, not, as men later thought, of his sin, but of his craft as master-builder, Sisyphos[32], like Zethus, like our giant, still rolls a huge stone up the slope. Everywhere it is the same tale. All definite record or remembrance of the building of ‘Cyclopean’ walls is lost; some hero-king built them, some god, some demi-god, some giant. Just so did the devil in ancient days build his Bridges all over England.

Tradition loves to embroider a story with names and definite details. The prudent Attic vase-painter gives us only a nameless ‘Giant.’ Others knew more. Pausanias[33] had heard the builders’ actual names and tried to fix their race. He tells us—just as he leaves the Acropolis—‘Save for the portion built by Kimon, son of Miltiades, the whole circuit of the Acropolis fortification was, they say, built by the Pelasgians, who once dwelt below the Acropolis. It is said that Agrolas and Hyperbios ... and on asking who they were, I could only learn that in origin they were Sikelians and that they migrated to Acarnania.’

Spite of the lacuna, it is clear that Agrolas and Hyperbios are the reputed builders. The reference to Sicily dates probably from a time when the Kyklopes had taken up their abode in the island. The two builder-brothers remind us of Amphion and Zethus, and of their prototypes the Dioscuri[34]. Pliny[35] tells of a similar pair, though he gives to one of them another name. ‘The brothers Euryalos and Hyperbios were the first to make brick-kilns and houses at Athens; before this they used caves in the ground for houses.’

The names of the two ‘Pelasgian’ brothers are, as we know from the evidence of vase-paintings[36], ‘giant’ names, and Hyperbios is obviously appropriate. The names leave us in the region of myth, but the tradition that the brothers were ‘Pelasgian’ deserves closer attention.

In describing the old wall we have spoken of it as ‘Pelasgian,’ and in this we follow classical tradition. Quoting from Hecataeus (circ. 500 B.C.), Herodotus[37] speaks of land under Hymettus as given to the Pelasgians ‘in payment for the fortification wall which they had formerly built round the Acropolis.’ Again, Herodotus[38] tells how when Kleomenes King of Sparta reached Athens, he, together with those of the citizens who desired to be free, besieged the despots who were shut up in the Pelasgian fortification.

A Pelasgian fortification, a constant tradition that Athens was inhabited by Pelasgians—we seem to be on solid ground. Yet on a closer examination the evidence for connecting the name of the fortification with the name ‘Pelasgian’ crumbles. In the one official[39] inscription that we possess the word is written, not Pelasgikon, but Pelargikon. In like manner, in Thucydides[40], where the word occurs twice, it is written with an r. Pelargikon is ‘stork-fort,’ not Pelasgian fort. The confusion probably began with Herodotus, who was specially interested in the Pelasgians.

Why the old citadel was called ‘stork-fort’ we cannot say—there are no storks there now—but we have one delightful piece of evidence that, to the Athenian of the sixth century B.C., ‘stork-fort’ was a reality.

Immediately to the south of the present Erechtheion lie the foundations of the ancient Doric temple[41], currently known by a pardonable Germanism as the ‘old Athena-temple.’ For its date we have a certain terminus ante quem. The colonnade was of the time of Peisistratos; it was a later addition; the cella of the temple existed before—how much before we do not know. The zeal and skill of Prof. Dörpfeld for architecture, of Dʳˢ Wiegand and Schrader for sculpture, have restored to us a picture of that ancient Doric temple all aglow with life and colour and in essentials complete[42].

Fig. 12.

Of all the marvellous fragments of early sculpture recently discovered, none is more widely known nor more justly popular than the smiling, three-headed monster known throughout Europe as the ‘Blue-beard.’ He belongs to the sculptures of the west pediment of the inner pre-Peisistratean cella of the ‘old Athena-temple,’ a portion of which is shown in [Fig. 12]. It is tempting to turn aside and discuss in detail the whole pediment composition to which he belongs. It will, however, shortly be seen ([p. 37]) that our argument forbids all detailed discussion of the sanctuaries of Athena, and the pediments of her earliest temple have therefore, for us at the moment, an interest merely incidental.

Thus much, however, for clearness sake may and must be said. The design of the western pediment fell into two parts. In one angle, that to the left of the spectator, Herakles is wrestling with Triton; the right-hand portion, not figured here, is occupied by the triple figure of ‘Blue-beard,’ whose correct mythological name is probably Typhon[43]. He is no protagonist, only a splendid smiling spectator. The centre of the pediment, where, in the art of Pheidias, we should expect the interest to culminate, was occupied by accessories, the stem of a tree on which hung, as in vase-paintings, the bow and arrows and superfluous raiment of Herakles.

It is a point of no small mythological interest that in this and two other primitive pediments the protagonist is not, as we should expect, the indigenous hero Theseus, but the semi-Oriental Herakles; but this question also we must set aside; our immediate interest is not in the sculptured figures of the pediment, but in the richly painted decoration on the pediment roof above their heads.

The recent excavations on the Acropolis yielded a large number of painted architectural fragments, the place and significance of which was at first far from clear. Of these fragments forty were adorned with two forms of lotus-flower; twenty had upon them figures of birds of two sorts. Fragmentary though the birds mostly are, the two kinds (storks and sea-eagles) are, by realism as to feathers, beak, legs, and claws, carefully distinguished. The stork (πελαργός) in the Pelargikon is a surprise and a delight. Was Aristophanes[44] thinking of this Pelargikon when to the building of his Nephelokokkygia he brought

For brickmakers a myriad flight of storks.

One of the storks is given in [Fig. 13]. The birds in the original fragments are brilliantly and delicately coloured. Their vivid red legs take us to Delphi. We remember Ion[45] with his laurel crown, his bow and arrows, his warning song to swan and eagle.

Fig. 13.

There see! the birds are up: they fly

Their nests upon Parnassus high

And hither tend. I warn you all

To golden house and marble wall

Approach not. Once again my bow

Zeus’ herald-bird, will lay thee low;

Of all that fly the mightiest thou

In talon! Lo another now

Sails hitherward—a swan! Away

Away, thou red-foot!

In days when on open-air altars sacrifice smoked, and there was abundance of sacred cakes, birds were real and very frequent presences. To the heads of numbers of statues found on the Acropolis is fixed a sharp spike to prevent the birds perching[46]. They were sacred yet profane.

The lotus-flowers carry us back to Egypt. The rich blending of motives from the animal and vegetable kingdom is altogether ‘Mycenaean.’ Man in art, as in life, is still at home with his brothers the fish, the bird, and the flower. After this ancient fulness and warmth of life a pediment by Pheidias strikes a chill. Its sheer humanity is cold and lonely. Man has forgotten that

Earth is a covering to hide thee, the garment of thee.

There are two sorts of birds, two sorts of lotus-flowers, and there are two pediments. It is natural to suppose, with Dr Wiegand, that the eagles belonged to the east, the principal pediment. There, it will later be seen ([p. 47]), were seated the divinities of the place. Our pediment decorated the west end, the humbler seat of heroes rather than gods. There Herakles wrestled with the Triton; there old Blue-beard—surely a monster of the earlier slime—kept his watch; and over that ancient struggle of hero and monster brooded the stork.

The storks themselves are there to remind us that the old name of the citadel was Pelargikon, and that Pelargikon meant ‘stork fort’; by an easy shift it became Pelasgikon[47], and had henceforth an etymologically false association with the Pelasgoi. Etymologically false, but perhaps in fact true, for happily the analogy between the Pelargic walls and those of Mycenae is beyond dispute, and if the ‘Mycenaeans’ were Pelasgian, the walls are, after all, Pelasgic.

We have seen that both Thucydides and the official inscription write Pelargikon; their statements will repay examination.

Thucydides, after his account of the narrow limits of the city before Theseus, returns to the main burden of his narrative, the crowding of the inhabitants of Attica within the city walls. ‘Some few,’ he says[48], ‘indeed had dwelling places, and took refuge with some of their friends or relations, but the most part of them took up their abode on the waste places of the city and in the sanctuaries and hero-shrines, with the exception of the Acropolis and the Eleusinion, and any other that might be definitely closed. And what is called the Pelargikon beneath the Acropolis, to dwell in which was accursed, and was forbidden in the fag end of an actual Pythian oracle on this wise,

The Pelargikon better unused,

was, notwithstanding, in consequence of the immediate pressure thickly populated.’

The passage comes for a moment as something of a shock. We have been thinking of the Pelargikon as the Acropolis, we have traced its circuit of walls on the Acropolis, and now suddenly we find the two sharply distinguished. The Acropolis, though closed, is surely not cursed. The Acropolis is one of the definitely closed places, to which the refugees cannot get access; the Pelargikon, though accursed, is open to them, and they take possession of it; the two manifestly cannot be coincident. But happily the words ‘below the Acropolis’ bring recollection, and with it illumination. What is called the Pelargikon below the Acropolis is surely that appanage of the citadel which Thucydides in his second clause mentions so vaguely. The ancient polis comprised not only ‘what is now the citadel,’ but also together with it, ‘what is below it towards about south[49].’ Thucydides would have saved a world of trouble if he had stated that ‘what is below towards the south’ was the Pelargikon; but he does not, probably because he is concerned with dimensions, not with nomenclature.

The Pelasgikon meant originally the whole citadel, the ancient city as defined by Thucydides. This was its meaning in the days of Herodotus. In the Pelasgikon the tyrants were besieged ([p. 25]). But by the time of Thucydides the Acropolis proper, i.e. much the larger and more important part of the old city, had ceased to be ‘Pelasgic’; the old fortifications were concealed by the new retaining walls of Themistocles and Kimon. It was only at the west and south-west that the Pelasgic fortifications were still visible, hence this portion below the Acropolis took to itself the name that had belonged to the whole; but this limited use of the word was at first tentative. Thucydides says, ‘which is called the Pelargikon.’ This is quite different from the definite ‘the Pelasgian citadel’ used by Herodotus. The neuter adjectival form is, so far as I know, never used of the whole complex of the Acropolis plus what is below.

From Thucydides we learn only that what was called the Pelargikon was below the Acropolis. ‘Below’ means immediately, vertically below, for when, in Lucian’s Fisherman[50], Parrhesiades, after baiting his hook with figs and gold, casts down his line to fish for the false philosophers, Philosophy, seeing him hanging over, asks, ‘What are you fishing for, Parrhesiades? Stones from the Pelasgikon?’ An inscription[51] of the latter end of the fifth century confirms the curse mentioned by Thucydides, and shows us that the Pelargikon was a well-defined area, as it was the subject of special legislation. ‘The king (i.e. the magistrate of that name) is to fix the boundaries of the sanctuaries in the Pelargikon, and henceforth altars are not to be set up in the Pelargikon without the consent of the Council and the people, nor may stones be quarried from the Pelargikon, nor earth or stones had out of it. And if any man break these enactments he shall pay 500 drachmas and the king shall report him to the Council.’ Pollux[52] further tells us that there was a penalty of 3 drachmas and costs for even mowing grass within the Pelargikon, and three officers called paredroi guarded against the offence. Evidently the fortifications of the Pelargikon, partially dismantled by the Persians, had become a popular stone quarry; as evidently the state had no intention that these fortifications should fall into complete disuse. The question naturally arises, what was the purport of this surviving Pelargikon, why did it not perish with the rest of the Pelasgic fortifications?

The answer is simple: the Pelargikon remained because it was the great fortification of the citadel gates. According to Kleidemos, it will be remembered ([p. 11]), the work of the early settlers was threefold; they levelled the surface of the citadel, they built a wall round it, and they furnished the fortifications with gates. Where will those gates be? A glance at the section in [Fig. 1] shows that they must be where they are, i.e. at the only point where the rock has an approachable slope, the west or south-west. We say advisedly south-west. The great gate of Mnesicles, the Propylaea which remain to-day, face due west; but within that great gate still remain the foundations[53] of a smaller, older gate ([Fig. 2], G), built in direct connection with the great Pelasgic fortification wall, and that older gate, there before the Persian War[54], faces south-west.

Fig. 14.

This gate facing south-west stands on the summit of the hill, and is but one. Kleidemos ([p. 11]) tells us that the Pelargikon had nine gates. That there should be nine gates round the Acropolis is unthinkable, such an arrangement would weaken the fortification, not strengthen it. The successive gates must somehow have been arranged one inside the other, and the fortifications would probably be in terrace form. The west slope of the Acropolis lends itself to such an arrangement, and in Turkish days this slope was occupied by a succession of redoubts ([Fig. 14]). Fortified Turkish Athens is in some ways nearer to the old Pelasgian fortress than the Acropolis as we see it to-day. We shall probably not be far wrong if we think of the approach to the ancient citadel as a winding way ([Fig. 15]), leading gradually up by successive terraces, passing through successive fortified gates[55], and reaching at last the topmost propylon which faced south-west. These terraces, gates, fortifications, covering a large space, the limits of which will presently be defined, formed a whole known from the time of Thucydides to that of Lucian as the Pelargikon or Pelasgikon.

Fig. 15.

Lucian indeed not only affords our best evidence that, down to Roman days, a place called the Pelasgikon existed below the Acropolis, but is also our chief literary source for defining its limits. We expect those limits to be wide, otherwise the refugees would not have crowded in.

The passages about the Pelasgikon in Lucian are two. First in the ‘Double Indictment[56],’ Dike, standing on the Acropolis, sees Pan approaching, and asks who the god is with the horns and the pipe and the hairy legs. Hermes answers that Pan, who used to dwell on Mt Parthenion, had for his services been honoured with a cave below the Acropolis ‘a little beyond the Pelasgikon.’ There he lives and pays his taxes as a resident alien. The site of Pan’s cave is certainly known; close below it was the Pelasgikon. This marks the extreme limit of the Pelasgikon to the north, for the sanctuary of Aglauros ([p. 81]) by which the Persians climbed up was unquestionably outside the fortifications. Herodotus[57] distinctly says, ‘In front then of the Acropolis, but behind the gates and the ascent, where neither did anyone keep guard, nor could it be expected that anyone could climb up there, some of them ascended near the sanctuary of Aglauros, daughter of Kekrops, though the place was precipitous.’

A second passage[58] in Lucian gives us a further clue. Parrhesiades and Philosophy, from their station on the Acropolis, are watching the philosophers as they crowd up. Parrhesiades says, ‘Goodness, why, at the mere sound of the words, “a ten-pound note,” the whole way up is a mass of them shouldering each other; some are coming along the Pelasgikon, others and more of them by the Areopagos, some are at the tomb of Talos, and others have got ladders and put them against the Anakeion; and, by Jove, there’s a whole hive of them swarming up like bees.’ A description like this cannot be regarded as definite proof; but, taking the shrines in their natural order, it certainly looks as though in Lucian’s days the Pelasgikon extended from the Areopagos to the Asklepieion. The philosophers crowd up by the regular approach (ἄνοδος) to the Propylaea; there is not room for them all, so they spread to right and left, on the right to the Asklepieion, on the left to the Areopagos; some are crowded out still further on the right to the tomb of Talos[59], near the theatre of Dionysos; on the left to the Anakeion[60] on the north side of the Acropolis.

Yet one more topographical hint is left us. In a fragment of Polemon[61] (circ. 180 B.C.), preserved to us by the scholiast on the Oedipus Coloneus of Sophocles, we hear that Hesychos, the eponymous hero of the Hesychidae, hereditary priests of the Semnae, had a sanctuary. Its position is thus described: ‘it is alongside of the Kyloneion outside the Nine-Gates.’ It is clear that in the days of Polemon either the Nine-Gates were still standing, or their position was exactly known. It is also clear that, whatever was called the Nine-Gates was near the precinct of the Semnae. The eponymous hero of their priests must have had his shrine in or close to the sanctuary of the goddesses. Moreover the Kyloneion or hero shrine ties us to the same spot. When the fellow-conspirators of Kylon were driven from the Acropolis, where Megacles dared not kill them, they fastened themselves by a thread to the image of the goddess to keep themselves in touch; when they reached the altars of the Semnae the thread broke and they were all murdered[62]. The Kyloneion must have been erected as an expiatory shrine on the spot.

When we turn to examine actual remains of the Pelasgikon on the south slope of the Acropolis ([Fig. 2]), we are met by disappointment. Of all the various terraces and supporting walls, only one fragment (P) can definitely be pronounced Pelasgian. The remaining walls seen in [Fig. 16] date between the seventh and the fifth centuries. The walls marked G in the plan in [Fig. 16], but purposely omitted in [Fig. 2], are of good polygonal masonry, and must have been supporting walls to the successive terraces of the Pelasgikon; they are probably of the time of Peisistratos[63], but may even be earlier. It is important to note that though not ‘Pelasgic’ themselves they doubtless supplanted previous ‘Pelasgic’ structures. The line followed by the ancient road must have skirted the outermost wall of the Pelargikon; later it was diverted in order to allow of the building of the Odeion of Herodes Atticus. The Pelasgikon of Lucian’s day only extended as far as the Asklepieion; the earlier fortification must have included what was later the Asklepieion[64], as it would need to protect the important well within that precinct.

Thucydides has stated the limits of the ancient city, ‘what is now the citadel was the city together with what is below it towards about south.’ We now-a-days should not question his statement. The remains of the Pelasgian fortifications disclosed by excavation amply support his main contention, namely, that what is now the citadel was the city, the conformation of the hill and literary evidence justify his careful ‘addendum’ together with what is below it towards about south.

But, as noted before, the readers of Thucydides were not in our position, they knew less about the boundaries of the ancient city, and though they probably knew fairly well the limits of the Pelasgikon, even that was becoming rather a matter of antiquarian interest. Above all, they were citizens of the larger city of Themistocles, the Dipylon was more to them than the Enneapylon. Thucydides therefore feels that the truth about the ancient city needs driving home. He proceeds to give evidence for what was, he felt, scarcely self-evident. If we feel that the evidence is somewhat superfluous, we yet welcome it because incidentally he thereby gives us much and interesting information as to the sanctuaries of ancient Athens.

The evidence is, as above stated ([p. 8]), fourfold.

CHAPTER II.
THE SANCTUARIES IN THE CITADEL.

τὰ γὰρ ἱερὰ ἐν αὐτῇ τῇ ἀκροπόλει καὶ ἄλλων θεῶν ἐστί.

There are sanctuaries in the citadel itself, those of other deities as well (as The Goddess).

Needless difficulties have been raised about this sentence, and, quite unnecessarily, a lacuna in the text has been supposed[65]. Though the form of the sentence is compressed, the plain literal meaning is clear. The first piece of evidence that Thucydides states is that in the ‘citadel itself other divinities “as well” have sanctuaries.’ To what does this ‘as well’ refer? Obviously to ‘The Goddess’ mentioned in the clause next but one before as presiding over the Synoikia, ‘The Goddess’ who was so well known that to name her was needless.

It has been proposed to read the sentence thus: ‘There are (ancient) sanctuaries in the citadel itself both “of the goddess Athena” and of other deities as well.’ This is true, but it is not what Thucydides says and not what he means. He does not desire to make any statement whatever about the sanctuaries of Athene or their antiquity; both propositions are for the moment irrelevant; he wishes to say what he does say, that ‘there are sanctuaries in the Acropolis itself, those of other deities as well (as The Goddess).’ It is the ‘other deities’ not ‘The Goddess’ who are the point.

Fig. 16.

But Thucydides always leaves perhaps rather much to the intelligence of his readers. It may fairly be asked, why is the existence of these sanctuaries of ‘other deities’ an argument in support of the statement that the Acropolis was the ancient city? Once fairly asked, the question answers itself. The Acropolis in the time of Thucydides was a hill sacred to Athena, it was almost her temenos; the other gods, Apollo, Zeus, Aphrodite, had their most important sanctuaries down below, all over the great ‘wheel-shaped’ city. Athena had from time immemorial, it was believed, dwelt on the hill; any statement about her shrines would prove nothing one way or the other. But in the old days, before there was any ‘down below,’ any ‘wheel-shaped’ city, if the ‘other gods’ were to be city gods at all they must have their shrines up above. Such shrines there were on the Acropolis itself; this made it additionally probable that the Acropolis was the ancient city. The reasoning is quite clear and relevant, and the argument is just the sort that an Athenian of the time of Thucydides, with his head full of the dominant Athena, and apt to forget the ‘other gods,’ would need to have recalled to his mind.

The citadel of classical days, with its ‘old Athena temple,’ Parthenon and its Erechtheion lies before us in [Fig. 16]. The ‘old Athena temple’ and the Parthenon belong to ‘The Goddess,’ where then are the ‘sanctuaries in the citadel itself which belong to other deities’ of which Thucydides is thinking?

For such we naturally look to the north side of the Acropolis, where lay the ancient king’s palace ([Fig. 2], C). About that old palace westward there lay clustered a number of early altars, ‘tokens’ (σημεῖα), sacred places and things (ἱερά). Later these were enclosed in the complex building known to us as the Erechtheion. It is by studying the plan of this later temple that we can best understand the grouping and significance of the earlier sanctuaries.

The Erechtheion as we have it now is shown in [Fig. 17]. Its plan is obviously anomalous, and has puzzled generations of architects. It was reserved for Professor Dörpfeld, with his imaginative insight, to divine that the temple, as we have it, is incomplete; and, further, to reconstruct conjecturally the complete design. In the light of this reconstruction the Erechtheion, as we now possess it, became for the first time intelligible.

Fig. 17.

Fig. 18.

This reconstruction is shown in [Fig. 18]. The temple in the original plan was intended to consist of two cellas, each furnished with a pronaos; the east cella is marked on the plan ‘Athena-Polias Tempel,’ the west cella is marked ‘opisthodom,’ i.e. opisthodomos or back chamber. Between these two cellas is a building divided into three chambers, marked in the plan ‘Poseidon-Erechth(eus)-Tempel.’ The middle chamber of the three is entered by two porches, a large one to the north, a smaller one—the famous Karyatid porch—to the south. This middle chamber alone of the three was probably provided with a low roof as shown in the sketch in [Fig. 19]. A building so complex cries aloud for explanation. It has become symmetrical, but what is its significance? What for us its connection with the sanctuaries of ‘other deities as well’?

Fig. 19.

To understand the new temple we must go back to the times before it was built[66]. It was intended—though ultimately this intention was not fully accomplished—to replace other existing sanctuaries, and these were first the old temple of Athena, and second the old temple of Erechtheus. The ‘old temple of Athena’ appears on the plan ([Fig. 18]) to the south of the Erechtheion; the very scanty remains of the old temple of Poseidon-Erechtheus are seen running diagonally under the western part of the new Erechtheion.

The ‘old temple of Athena’ consisted, it is clear, of two parts: to the east the actual cella of the goddess; to the west, divided into three chambers, the opisthodomos or treasure-house. We are concerned wholly, it must be noted, with the ‘other deities,’ not with Athena; for from the consideration of Athena and her sanctuaries Thucydides has dispensed us; but the arrangement of the new Erechtheion cannot be understood without some reference to the disposition of the old temple of Athena.

Perikles intended to demolish not only the old Erechtheion but also the old temple of Athena, and to supplant them by a common sanctuary. The east cella in the old Athena temple was to be replaced by an east cella for the goddess in the new; the opisthodomos to the west of the old temple by an opisthodomos to the west of the new. Between these parts of the old Athena temple three chambers were to be devoted to replacing the old Erechtheion. It is difficult by help of ground-plans to realize the different levels of the temple, but those who have been on the spot will remember that the new cella of Athena is on the same level as the old. The Erechtheion with its different levels is a striking contrast to the Parthenon, where, as we have already seen, the slope of the ground was levelled up and that at enormous expense. This preservation of different levels in the Erechtheion is in itself sufficient evidence of the sanctity of the different cults to be enshrined. The longer complex structure, with its different levels and its five chambers, was intended, as Perikles planned it, to be entered by the two porches, north and south. Structurally these would reduce the effect of undue length, but they had also another purpose—the north porch contained the trident mark of Poseidon, the south the grave of Kekrops.

The plan of Perikles was never completed. By some one’s machinations, whether of architect, priest, or politician we do not know, he was—as before in the building of the Propylaea—frustrated, and obliged to be content with a truncated scheme. The new Erechtheion almost certainly had been begun before the outbreak of the Peloponnesian War. When Perikles found that his plan was not accepted in full, he did not design a new temple but made a compromise obviously intended to be provisional. He was again frustrated in the execution even of this modified scheme, which was not completed till much later. The Erechtheion that we know has the east cella for Athena complete and the two porches, but two only of the three intended midway chambers were built, and the westernmost one, as appears on the plan, is slightly reduced in size. The west cella was never even begun. It is probable that Perikles never succeeded in transferring the image of Athena from her old temple to the new cella, but this question[67] it is not necessary we should here decide.

Setting aside those portions of the Erechtheion which were intended to supply the place of the old temple of Athena, namely the east cella and the proposed opisthodomos to the west, we have now to consider what were the ancient sanctities (ἱερά) of ‘other deities’ which the three central chambers and the two porches were planned to enshrine. They are as follows:—

1. The hero-tomb of Kekrops.

2. The Pandroseion.

3. Three ‘tokens’ (σημεῖα).

a. A sacred olive tree.

b. A ‘sea’ called after Erechtheus.

c. A trident mark sacred to Poseidon.

1. The hero-tomb of Kekrops.

We begin with Kekrops because, by almost uniform tradition, with Kekrops Athens began. The Parian Chronicle[68] sets him at the head of the kings of Athens, and the date assigned to him is 1582 B.C., before Kranaus, before Amphictyon, before Erechtheus. Thucydides[69] names him as the typical early Athenian king. ‘Under Kekrops and the first kings,’ he writes; Apollodorus[70] says definitely, ‘the indigenous Kekrops, whose body was compounded of man and snake, first reigned over Attica, and the country which before was called Attica was from him named Kekropia.’ Herodotus[71] looked back to a day before Athens was Athens and when there were no Athenians at all: ‘The Athenians,’ he says, ‘at the time when the Pelasgians held that which is now called Hellas, were Pelasgians and they were called Kranai; under the rule of Kekrops they were called Kekropidae; but when Erechtheus succeeded they changed their name for that of Athenians, and when Ion, son of Xuthus, became general, they took from him the name of Ionians.’

Herodotus touches the truth. Kekrops was not the first king of Athens, he was king before there was any Athens, long before. He was the ancestor of the clan of the Kekropidae. At some very early date—the Parian marble may very likely be roughly right—the Kekropidae got possession of the Acropolis and called it Kekropia. Kekropis was the name not only of one of the four original Attic tribes but also of one of the later ten[72]. But though the clan kept its old name it lost the headship of Kekropia. Kekrops had only one son, Erysichthon[73], and he died childless; that is the mythological way of saying that the kingship changed families. Then came the time when the leading clan were Erechtheidae, descendants not of Kekrops, but of Erechtheus. These are Homer’s days. He knows nothing of Kekrops and Kekropia, only of ‘the people of Erechtheus[74].’ Then still later came another change; those who once were the people of Erechtheus became the people of Athena, Athenians. But Kekrops and Kekropia were first, probably long first. Kekrops is the hero-founder, the typical old-world king. It is Kekrops whom Bdelycleon[75], tormented by modernity, invokes:

‘Kekrops, oh my king and hero, thou that hast the dragon’s feet.’

Kekrops was half man, half snake. His ‘double nature’ gave logographers and even philosophers much trouble. Was it because he had the understanding of a man and the strength of a dragon, was it because, at first a good king, he later became a tyrant, or because he knew two languages (Egyptian and Greek), or because he instituted marriage? The curious will find it all in Tzetzes[76]. Eager anthropologists have seized on Kekrops as a totem-snake, but the average orthodox mythologist is content to see in his snake-tail the symbol of the ‘earth-born’ Athenians. This interpretation grazes the truth, but just misses the point. The hybrid form is of course transitional. Kekrops is sloughing off his snake form[77] in deference to the inveterate anthropomorphism of the Greek. He was once a complete snake, not because he was a totem-snake, not because he was an ‘autochthonous hero,’ but because he was a dead man and all dead persons of importance, all heroes, become snakes.

No one has done so much to obscure the early history of Athenian religion as Athena herself, by her constant habit of taking over the attributes of other divinities[78]. The eponymous hero of each victorious tribe, Kekrops and Erechtheus in turn, is a home-keeping, home-guarding snake (οἰκουρὸς ὄφις). But by the time of Herodotus[79] the sacred snake supposed to live on and guard the Acropolis lives in the sanctuary of Athena, and is almost the embodiment of the goddess herself; when the snake refused the honey-cake it was taken as an omen that ‘the goddess had deserted the Acropolis.’ By the time of Pheidias the snake is just an attribute of the Parthenos, and was set to crouch beneath her shield. But Pausanias[80] has an inkling of the truth; he says, ‘close beside the spear is a snake: this snake is probably Erichthonios.’ The real relation of goddess and snake was simply this: the original pair of divinities worshipped in many local cults were a matriarchal goddess, a local form of earth-goddess, and the local hero of the place in snake form as her male correlative; such a pair were Demeter and the snake-king Kenchreus at Eleusis[81], such were Chryse and her home-keeping nameless guardian snake on Lemnos[82], such were Eileithyia and Sosipolis at Olympia[83], such were ‘the goddess’ and her successive heroes Kekrops and Erichthonios or Erechtheus; only, as will later be seen, in this last pair another goddess preceded Athena.

Kekrops then was a dead, divinized hero embodied as a snake; the natural place for his worship was his tomb, probably the earliest sanctuary of the Acropolis. Clement[84] of Alexandria says, ‘the tomb of Kekrops is at Athens on the Acropolis,’ and Theodoretus[85], quoting Antiochos, adds that it is ‘by the Poliouchos herself,’ the goddess of the city. We might safely assume that a hero-tomb was a sanctuary, but we have express evidence: in an honorary decree[86] respecting the ‘ephebi’ of the deme of Kekrops it is ordered that the decree shall be set up ‘in the sanctuary of Kekrops,’ and from another decree[87] we learn the name of a ‘priest of Kekrops.’

But our most definite evidence as to where the tomb of Kekrops lay comes from the famous Chandler inscription[88] now in the British Museum. This inscription is exactly dated by the archonship of Diokles (409-408 B.C.). It is a statement of the exact condition in which the overseers of the unfinished temple took over the work, what part was half finished, what unwrought and unchannelled (i.e. columns), and what were completely finished but not set up in their place. The various parts of the temple are described as near or opposite to such and such an ancient shrine, and fortunately among these descriptions occur more than one mention of the Kekropion. The following[89] is decisive: ‘Concerning the porch beside the Kekropion the roof stones above the Korae must be....’ The porch of the Karyatids, or to call it by its ancient[90] name, the porch of the Korae, the Maidens, was beside, close to, the Kekropion.

So far all is certain. The tomb of Kekrops was close to the porch of the Maidens; but in which direction? We should expect it to be north-west, because in that direction, as will be immediately ([p. 48]) shown, lay the precinct of Pandrosos, daughter of Kekrops. Professor Dörpfeld[91] places it conjecturally at D ([Fig. 16]), and the site is almost certain. It has been already noted that the west wall of the present Erechtheion was set back a short distance within its original plan. It may have been to avoid trenching on the tomb of Kekrops. Moreover, at the south end of this wall there is a great gap in the ancient masonry of about 10 ft. long by 10 high. The gap is evident, though it was filled up by modern masonry. It is spanned by an enormous ancient block of stone, 15 ft. by 5. Here probably was buried the serpent king.

Fig. 19.*

With the serpent king and his prophylactic tomb clearly in our minds, we turn with new eyes to examine certain fragments of sculpture discovered in the recent excavations. Nothing perhaps caused more surprise when these fragments came to light than the size and splendour of the snake-figures. We have already seen ([p. 27]) that the western pediment of the Hecatompedon held two sea-monsters, a Triton and Typhon; the eastern pediment held two land-snakes of even greater magnificence. The design of this pediment as restored by Dr. Wiegand[92] is as follows ([Fig. 19*]). In the apex is seated Athena; to her right hand a figure seated and crowned, and therefore a king or a god; this figure survives, but the figure which must have balanced him to the left of the goddess is lost for ever. Athena is supreme; the surviving figure is usually called Zeus, but from his subordinate place it seems to me that it is more likely he is either a subordinate god, Poseidon, or a local king, Erechtheus. Possibly Athena is seated between Poseidon and Erechtheus.

It is, after all, not the seated protagonists of the pediment, be they Olympians or local kings, who most interest us, but the two great snakes who in the angles keep watch and ward. These snakes are often described as ‘decorative’ or ‘space-filling.’ But surely they are too alive, too large, too dominant to be mere accessories. One of them is shown in [Fig. 19*] in detail, so far as he can be represented by an uncoloured reproduction. In the original he is blue and orange, and his companion in the other angle is a vivid emerald green.

Herodotus[93], it is true, speaks of one snake only as guardian of the Acropolis, the snake who when the land was beset by the Persians, would not eat its honey-cake; but then Herodotus writes as if he had no personal knowledge: ‘the Athenians say there is a great snake.’ In the story of Erichthonios tradition, and good Attic tradition, knew of two. Hermes in the Ion of Euripides[94] says, referring to Erichthonios,

‘To him

What time she gave him to the Agraulid maids

Athena bound for watch two guardian snakes;

In memory whereof Erechtheus’ sons

In Athens still upon their nursing babes

Put serpents wrought of gold’;

and on the well-known vase in the British museum[95] depicting the scene, two snakes appear. We need not say that the two snakes of the pediment are a duplicated Kekrops, but we may and do say that they are two hero-snakes, guardians of the city, and we may further conjecture that they were an old pair, male and female. This conjecture brings us to the woman counterpart of Kekrops, the snake king, his ‘daughter’ Pandrosos.

2. The Pandroseion.

Kekrops and his faithful daughter Pandrosos were not far sundered. The situation of the Pandroseion is, within narrow limits, certain. It was an enclosure to the west of the present Erechtheion. The invaluable Chandler inscription[96] speaks of ‘the pillars on the wall towards the Pandroseion.’ This must refer to the west wall, on which were four engaged pillars at a height of about 12 feet from the ground. In another inscription[97], found during the pulling down of the ‘Odysseus’ Bastion, mention is made of two pediments, one towards the east and the other ‘towards the Pandroseion.’

We know, then, certainly that the Pandroseion was west of the present Erechtheion. We know also that it was close to the ‘old temple of Athena.’ Pausanias[98], in passing from the one to the other, distinctly says: ‘The temple of Pandrosos adjoins the temple of Athena.’ As Pausanias distinctly says there was a temple (ναός), not merely a temenos or sanctuary (ἱερόν), it is disappointing that excavations have yielded no trace.

In actual cultus and topography we have found Kekrops side by side with one woman figure, Pandrosos. In current mythology he has three daughters, of whom is told the thrice familiar story of the child and the chest[99]. It will repay examination.

The child Erichthonios is born from the Earth in the presence of Kekrops. His real mother, Earth, gives him up to the tendance of Athena; such is the scene familiar on terra-cottas and vase-paintings. Athena places him in a chest or wicker-basket, and gives him to the three daughters of Kekrops, Pandrosos, Herse, Aglauros, with strict orders not to open the chest. The two sisters, Herse and Aglauros (or according to some versions all three), overcome by curiosity open the chest, and see the child with a snake or snakes coiled about him. In terror at the snake, who pursues them, and fearing the anger of Athena, they cast themselves down from the Acropolis.

The story is manifestly absurd, and in some of the elements plainly aetiological.

The suicide of the disobedient sisters is easily explicable. Half way down the Acropolis, below the steepest portion of the rock, were a number of shrines and tombs. Why were they there? Clearly because the persons after whom they were named had thrown themselves down, or been thrown down, from the top. Such a shrine was the tomb of Talos[100], near the Asklepieion. Daedalos was jealous of Talos, and threw him down from the rock. Such was also the shrine of Aegeus[101], below the temple of Nike Apteros, where Aegeus in despair at the sight of the black sail cast himself down. Such was the sanctuary of Aglauros[102] on the north side of the Acropolis. Somebody must have cast herself down to account for the situation. When one sister only is mentioned she is naturally Aglauros, but all three are often allowed to commit suicide for completeness sake.

Of the three sisters, Herse was not a real person[103]; she has no shrine, she is only a heroine invented to account for the ceremony of the Hersephoria. The cult of Aglauros is below the Acropolis and manifestly separate from that of Pandrosos, and Pandrosos alone for the present need be considered.

Pausanias, after stating that the temple of Pandrosos adjoins that of Athena, says that she was ‘the only one of the sisters who was blameless in the affair of the chest intrusted to them.’ As Pandrosos had a shrine so revered it would have been awkward to make her out guilty. He then, without telling us whether or no he perceives any connection, proceeds to describe ‘a thing which caused me the greatest astonishment and is not generally known.’ The thing that so astonished Pausanias was the ceremony of the Arrephoria[104]. Maidens called Arrephoroi bore upon their heads certain sacred things covered up; these they carried by night by a natural underground passage to a precinct near to that of Aphrodite in the Gardens. There they left what they had been carrying, and brought back other things also wrapped up and unknown. From the analogy of other mystery cults we may be sure that the objects carried were some sort of fertility-charms, and they would be carried in a chest or wicker basket, a cista or a liknon, veiled that the sacred thing might not be seen. The girl-Arrephoroi might not look into the sacred chests. Why? The answer was ready, the goddess they served, Pandrosos, had also her sacred chest into which she and she only had not looked.

The personality of Pandrosos is hard to seize and fix. One thing is clear; ‘Pandrosos’ is not a mere ‘title of Athena.’ She manifestly, as daughter of Kekrops, belongs to that earlier stratum before the dominance of The Goddess. Later Athena absorbed her as she absorbed everything else. In official inscriptions she usually comes after Athena, and is clearly a separate personality. Thus the epheboi[105] offered their ‘sacrifices at departure (ἐξιτήρια) on the Acropolis to Athena Polias and to Kourotrophos and to Pandrosos,’ and women swore by her, though not so often as by Aglauros. We have one ritual particular that looked as though between her and Athene there was at some time friction. Harpocration[106] in explaining the rare word ‘ἐπίβοιον,’ ‘that which is after the ox,’ says, quoting from Philochoros, that it was the name given to a sacrifice to Pandrosos. If any one sacrificed an ox to Athena it was necessary to sacrifice a sheep to Pandrosos. Pandrosos was in danger of being effaced by Athena, and some one was determined this should not be; all that ‘The Goddess’ could secure was precedence.

We have found, then, a maiden goddess who was there before ‘The Goddess,’ nay, who may have herself been ‘The Goddess’ before Athena claimed the title. Pandrosos belongs to the early order of the Kekropidae, before the dwellers on the hill became Athenians. It is possible that her presence throws some light on the beautiful, but as yet enigmatic figures of the ‘Maidens’ who have been restored to us by the recent excavations. Who and what are they?

The ‘Maiden’ whose figure is chosen for the frontispiece of this book was found alone, somewhat later than the rest, in October, 1888, not like the others ([p. 16]) North of the Erechtheion, but near the wall of Kimon to the South, between the precinct of Artemis-Brauronia and the West front of the Parthenon. There is a certain fitness in this, because though in dress, adornments, colouring, general type, she is like the rest, her great beauty will always make her a thing apart. The torso and head were found separate, and about the torso there is nothing specially noteworthy. The unique loveliness is all of the face, and it escapes analysis. There are, however, peculiarities worth noting. The right eye is set much more obliquely than the left. This gives an irregular charm and individuality; the unusually high forehead emphasizes the austere virginal air, and the same may be said of the straight chest and long thin throat. But the secret of her beauty is still kept; standing as she does now among the other ‘Maidens,’ she is a creature from another world, and for all their beauty the rest look but a kindly mob of robust mothers and genial housewives.

The statues in question, which now number upwards of fifty, have been called by the name ‘Maidens,’ a name current among archaeologists. It is open to objection, because ‘maidens’ (κόραι) meant in the official language of the inscription already quoted[107] the ‘Caryatid’ figures of the Erechtheion. The word has, however, one great advantage, it is vague and commits the user of it to no theory as to the significance of the statues. The word korè meant to the Greek not only maiden, but doll or puppet or statue of a maiden. We need only recall the familiar epigram with the dedication to Artemis[108]:

Maid of the Mere, Timaretè here brings

Before she weds, her cymbals, her dear ball

To thee a Maid, her maiden offerings,

Her snood, her maiden dolls their clothes and all.

Here the korai are actual dolls, but in Attic inscriptions we find the word korè used of a statue[109], thus, ‘a korè of gold on a pillar’; or again in a dedication to Poseidon, ‘he dedicated as firstfruits this korè.’ A korè is one form of an agalma, a thing of delight.

The statues, then, may be called ‘Maidens,’ but the word is too vague to help us much as to their significance, and it is their significance, who and what they are, not their value in the history of art that here concerns us.

The question is generally put thus, Are they statues of Athena, or are they statues of mortal women dedicated to her? priestesses or merely worshippers? Statues of Athena they are, I think, certainly not; they have neither helmet, spear, shield, nor even aegis. Athena may appear sporadically without characteristic attributes, but that a series of fifty statues of Athena should be dedicated without a single hint of anything that made Athena to be Athena is scarcely possible.

Are they, then, mortal maidens? For priestesses their number, restricted as they are by style to a short period of years, is too many. If they are mere mortal worshippers, it is at least strange that in the only two cases where we have inscribed bases they are dedicated by men. In one case we have the simple statement: ‘Euthydikos son of Thalearchus dedicated[110]’; on the other, Antenor, it is stated, makes the statue, Nearchos dedicates it as ‘firstfruits of his works[111].’ Would Nearchos dedicate a statue of mortal woman as ‘firstfruits of his works’? We seem to be at an impasse.

But there is surely a third solution open to us. The maidens need not be mortal because they are not Athena. There was a time before the armed maiden with spear and shield and aegis came from Libya or the East, a time when another maiden ruled upon the hill and was ‘The Goddess.’ Is it not at least possible that the maidens are made in her image, and that when the armed goddess took possession of the hill, when the ancient Kekropidae and Erechtheidae became Athenaioi, the maidens of the old order passed into the service of the maiden of the new? that we must think of their type as shaped at least for the worship of Pandrosos rather than Athena? The type of the warlike goddess was not fashionable in Greece. The Greeks, if any people, held firmly the doctrine that

A woman armed makes war upon herself.

The woman armed and disarmed, the Amazon in defeat, they made beautiful and poignantly human, but the woman armed and triumphant, Athena Nikephoros, remained a cold unreality. The korè of Eleusis is not armed, but at Corinth and at Sparta there was that strangest of all sights—the image of Aphrodite armed[112]. Whence she came is, as will later be seen ([p. 109]), not doubtful. In Cythera[113], Pausanias tells us, ‘the sanctuary of the Heavenly Goddess is most holy, and of all Greek sanctuaries of Aphrodite this is most ancient. The goddess is represented by a wooden image armed.’ The Cythereans called their armed Oriental goddess Cytherea. Did the Athenians call the same armed goddess ‘Athenaia’? Be that as it may, before her coming they worshipped the unarmed maiden.

Before we pass from Kekrops and Pandrosos to the later order under Erechtheus, the traditional events reputed of the reign of Kekrops must be noted. There are three:—

1. The contest between Athena and Poseidon, of which Kekrops acted as judge.

2. The introduction of the worship of Zeus.

3. The institution of marriage.

The discussion of the contest between Athena and Poseidon really belongs to the Erechtheid period, and must stand over till then. The introduction of the worship of Zeus and the institution of marriage are probably but the religious and social forces of the same advance, and may be taken together.

In front of the Erechtheion, Pausanias[114] tells us, was an altar dedicated to Zeus Hypatos, on which no living thing was sacrificed, but only cakes (πέλανοι). Pausanias does not here say that the altar was dedicated by Kekrops, but, in his discussion of Arcadia[115] and the human sacrifice of Lycaon, he says, ‘Kekrops was the first who gave to Zeus the title of Supreme, and he would not sacrifice anything that had life, but he burned on the altar the local cakes which the Athenians to this day call pelanoi.’ What probably happened was just the reverse of what Pausanias describes: there was an old altar to ‘the Supreme,’ the Hypatos; at some time or other this was taken over by the immigrant Zeus; the shift was attributed to Kekrops.

Zeus was essentially of the patriarchal order, i.e. of a condition of things in which the father rather than the mother is the head of the family, gives his name to the children, and holds the family property and conducts the family worship. Nothing could be more patriarchal than the constitution of the Homeric Olympus. Such a condition of things is necessarily connected with some form of the social institution known to us as marriage. Accordingly we learn from Athenaeus[116], quoting from Clearchus the pupil of Aristotle, that ‘At Athens Kekrops was the first to join one woman to one man: before connections had taken place at random and marriages were in common—hence as some think Kekrops was called “Twyformed” (διφυής) since before his day people did not know who their fathers were on account of the number’ (of possible parents). The story of the contest between Athene and Poseidon was later mixed up with the same tradition of the shift from patriarchy to matriarchy. St Augustine[117] says that the women voted for Athena, and their punishment was to be, among other things, that ‘no one was hereafter to be called by his mother’s name.’

We pass to the three tokens (σημεῖα), the first of which is

a. The sacred olive-tree.

The holy bloom of the olive, whose hoar leaf

High in the shadowy shrine of Pandrosos

Hath honour of us all.

Apollodorus[118] says, ‘After him (Poseidon) came Athena, and having made Kekrops witness of her seizure, she planted the olive which is now shown in the Pandroseion.’ A ‘seizure’ indeed, and not from Poseidon but from the elder goddess Pandrosos. Athena is manifestly an interloper; why should Pandrosos have other people’s olive trees planted in her precinct? The olive is but one of the many ‘tokens’ or attributes that Athena wrested to herself. It was there before her, Kekrops quite rightly holds it in his hand.

The olive-tree grew in the Pandroseion, it also grew in the older Erechtheion. Herodotus[119] says, ‘There is on this Acropolis a temple of Erechtheus, who is called earth-born, and in it are an olive-tree and a sea which, according to the current tradition among the Athenians, Poseidon and Athenaia planted as tokens when they contended for the country.’ There is no discrepancy, the Pandroseion must have been included in the older Erechtheion.

By a most happy chance, among the fragments of decorative sculpture left us is one on which is carved ‘the holy bloom of the olive,’ in three delicate sprays. The real sacred olive was old and stunted and crooked[120], but the artist went his own way. The fragments are grouped together in a conjectural restoration[121] in [Fig. 20]. All that is certain is that we have a Doric building and adjacent to it the wall of a precinct over which the olive is growing. Against the wall of the building is the figure of a woman in purple, wearing peplos and himation. Against the wall of the precinct once stood a man. Only one leg of him is left. The two figures might be part of a procession. The woman, standing full face, may belong to the same composition, but this is not certain. She wears a red chiton and bluish-green himation. On her head is a pad (τύλη), for she is carrying some burden. One of her arms is lifted to support it. We think instinctively of the Arrephoroi. The figure, though very rudely hewn, has something of the lovely seriousness of the other ‘maidens.’ The whole composition may have belonged to a pediment of the earlier Erechtheion, but its pictorial character makes it more probably a votive relief for dedication there, and representing some scene of worship at the ancient shrine.

Within the older Erechtheion we have further

(b) A cistern or ‘sea,’ called after Erechtheus. With it may be taken

(c) A trident-mark, sacred to Poseidon.

Fortunately about the position of these two sacred things there is no doubt. Underneath the pavement of the westernmost chamber (c) of the present Erechtheion is a large cistern[122] hewn in the rock, and at A in the North porch are the marks of the trident.

Fig. 20.

The two things together, the sea-water in the cistern and the trident-mark, were both associated with Poseidon. Pausanias[123] says they were said to be ‘the evidence produced by Poseidon in support of his claim to the country.’ Apollodorus[124] says, ‘Poseidon came first to Attica and smote with his trident in the middle of the Acropolis and produced the sea which they now call Erechtheïs.’

Athena produced the olive-tree, Poseidon the salt well and the trident-mark as ‘tokens’ or evidence of their claim. This is manifest aetiology. There had been on the Acropolis from time immemorial certain things reputed sacred, a gnarled olive-tree, a brackish well, three holes in a rock. It was the obvious policy of any divinity who wished to be worshipped at Athens to annex these tokens. Pandrosos had the olive-tree before Athena. The name of the well Erechtheïs shows that it was a ‘token’ of Erechtheus rather than of Poseidon.

Such sacred trees, such ‘seas,’ such curious marks existed elsewhere; Pausanias[125] himself notes in another inland place, Aphrodisias in Caria, there was a sea-well. What impressed him as noteworthy about the well at Athens was that when the South wind was blowing it gave forth the sound of waves, but then as he does not say if he waited for a South wind, the ‘sound of waves’ may have been a detail supplied by the guides.

The trident-mark belongs to a class of sacred things that will repay somewhat closer attention. Fresh light has been thrown upon it by a recent discovery. In examining the roof of the North Porch, with a view to repairs, it was observed that immediately above the trident-mark an opening in the roof had been purposely left. The object is clear; the sacred token had to be left open to the sky; it had to be sub divo. This is manifestly more appropriate to a sky-god than to a sea-god.

Our best analogies are drawn from Roman sources. Ovid[126] tells us that when the new Capitol was being built a whole multitude of divinities were consulted by augury as to whether they would withdraw to make place for Jupiter. They tactfully consented, all but old Terminus. He stood fast, remaining in his shrine, and still possesses a temple in common with mighty Jupiter:

And still, that he may see only heaven’s signs

In the roof above him is a little hole.

When place was wanted for an Olympian, be he Zeus or Poseidon or Athena, the elder divinities were not always so courteously consulted. We do not even know whose open air token Poseidon seized.

Servius[127], commenting on ‘the steadfast stone of the Capitol,’ tells the same story. There was a time when there was no temple of Jupiter, that is there was no Jupiter. Augury said that the Tarpeian mount was the place to build one, but on it were already a number of shrines of other divinities. Ceremonies were performed to ‘call out’ by means of sacrifice the other divinities to other temples. They all willingly migrated, only Terminus declined to move: this was taken as a sign that the Roman empire would be for all eternity, and hence in the Capitoline temple the part of the roof immediately above, which looks down on the very stone of Terminus, was open, for to Terminus it is not allowable to sacrifice save in the open air. Terminus was just a sacred stone or herm, incidentally to the practical Romans a boundary god. Another Roman god, Fidius[128], had in his temple a roof with a hole in it (perforatum tectum), and Fulgur, Caelum, Sol and Luna had all to dwell in hypaethral temples[129]. Wherever the lightning struck was in Greece holy ground, to be fenced in but open always above to the god who had sanctified it, to the ‘descender,’ Kataibates[130]. Kataibates became Zeus Kataibates, Fulgur Jupiter Fulgur, but the lightning and the ‘descender’ were there before the coming of the Olympian, and the threefold mark preceded Poseidon.

In picturing to ourselves therefore the ancient sanctities of the Acropolis, we have to begin with certain natural holy things that were there from time immemorial, that were holy in themselves, not because they were consecrated to this or that divinity. Such were the olive-tree, the salt sea-well, the trident-mark—we are back in a time rather of holy things than divine persons. Successive heroic families, in possessing themselves of the kingship, take possession of these sanctities; they are as it were the regalia. In the time of the Kekropidae, Pandrosos, daughter and paredros of Kekrops, owns the olive-tree; in the time of the Erechtheidae the well is called Erechtheïs, and all the sacred things are included in an Erechtheion. It is worth noting that though Poseidon claimed the well and the trident-mark he never gave his name to either, and though Athena boasted of the olive-tree and snake, neither was ever called after her.

The name of Erechtheus or Erechthonios marks a stage definitely later than that of Kekrops. In the reign of Kekrops we hear nothing of foreign policy. He is engaged in civilizing his people, in marrying them, in teaching them to offer bloodless sacrifice. But the reign of Erechtheus is marked by a great war. He fought with and conquered Eumolpos, king of the neighbouring burgh Eleusis. Kekropia has taken the first step towards that hegemony she was to obtain under Theseus.

Erechtheus, not Kekrops, is the king-hero known to Homer; the two passages in which he and his city are mentioned are significant. In the Odyssey[131], Athena, having counselled Odysseus, leaves him to make his entrance alone into the house of Alkinoös, while she betakes herself home. ‘Therewith grey-eyed Athene departed over the unharvested seas and left pleasant Scheria and came to Marathon and wide-wayed Athens, and entered the good house of Erechtheus.’ Here manifestly Athena has no temple, she has to shelter herself in the good house of Erechtheus (Ἐρεχθῆος πυκινὸν δόμον). That is how it used to be in the old kingly days, the king was divine, his palace a sanctuary.

But in the Catalogue of the Ships[132]—allowed on all hands to be a later document—things are quite otherwise. Among the captains of the ships were ‘they that possessed the goodly citadel of Athens, the domain of Erechtheus the high-hearted, whom erst Athene, daughter of Zeus, fostered, when Earth, the grain-giver, brought him to birth;—and she gave him a resting-place in Athens, in her own rich sanctuary; and there the sons of the Athenians worship him with bulls and rams as the years turn in their courses.’

The passage is a notable one. The singer is manifestly in some difficulty. Athena by his time is supreme; she has a goodly temple: it is she who offers hospitality to Erechtheus, not Erechtheus to her. Yet the singer knows the early tradition that the goodly citadel belongs to the king Erechtheus, he also knows the ritual fact that annual sacrifice was offered to him. This ritual fact of the sacrifice to Erechtheus is attested by Herodotus[133]. He tells us that the Epidaurians were allowed to cut down sacred olive-trees to make statues from, on the express condition that they annually sacrificed victims to Athena Polias and Erechtheus. Here the goddess joins in the honours, a fact not expressly stated in Homer, though probably understood.

So far we have Erechtheus, hero-king, snake-king, like the earlier Kekrops and Athena. Athena, it is evident, is the later intruder, but we have had no evidence of Poseidon. Poseidon’s position at Athens is a very peculiar one. Unlike Erechtheus, he has no temple called after him, he cannot give his name even to a salt sea-well, his trident-mark is probably to begin with a thunder-smitten rock; unlike Athena he never gets the people called after him, and yet, spite of all this, his worship is ancient and deep-rooted, and from him rather than from Zeus or Athena the old nobility of Athens claimed to be descended.

We are so accustomed to regard Athena as the Alpha and well-nigh the Omega of Athenian religion that the priority of Poseidon, one of the ‘other gods,’ needs emphasis. The Athenians themselves, however, at least the more conservative[134] among them, recognized it. Poseidon they knew was son of Kronos, and Athena daughter only of the younger Zeus.

‘O Sea-Poseidon and ye elderly gods’

exclaims the youth in the Plutus when he holds the torch to the wrinkles in the old woman’s withered face. When, in the Frogs, Euripides is made to utter what is taken to be a fine old conservative sentiment, Dionysos answers ‘Good by Poseidon, that!’ When in the Knights Nicias the household slave—conservative after the manner of his class—hears that the new demagogue is a black-pudding chandler, he exclaims in horror,

‘A black-pudding chandler, Poseidon what a trade!’

The choice of Poseidon by the conservative party was no mere chance; they believed in him, they swore by him, because they thought they were descended from him. In the case of one noble family, the Butadae, this descent was no mere chance tradition; their family tree was written up in the Erechtheion itself, and they claimed to be descended from a certain Butes, son of Poseidon and brother of Erechtheus. When Pausanias[135] entered the later Erechtheion he saw in the first chamber three altars, ‘one sacred to Poseidon on which sacrifices are offered to Erechtheus in accordance with the command of an oracle, one to the hero Butes, and one to Hephaestos; the paintings on the wall represent the family of the Butadae.’ It is often said that Erechtheus is merely a ‘title’ of Poseidon; this was the view of the lexicographers. Hesychius[136] explains Erechtheus as ‘Poseidon at Athens.’ But the statement about the altar shows that they were not originally the same, the command of an oracle was needed to affiliate them. It is a noticeable point moreover that Poseidon has no temple of his own, only an altar in the ‘dwelling’ (οἴκημα) called the Erechtheion. This sanctuary bearing the kingly name, remains his ‘steadfast house’ and is an eternal remembrance of the days when the king was priest and the god’s vicegerent on earth.

But there came a time when kings ceased to be in the old full sense incarnate gods, and then the kingly function was split into two offices, secular and spiritual. Of this at Athens we have traces in the narrative of Apollodorus[137]. He says ‘on the death of Pandion his sons divided the paternal estate and Erechtheus took the kingship, but Butes took the priesthood of Athena and of Poseidon the son of Erechthonios.’ It was the family tree of the royal priest Butes that was religiously preserved in the Erechtheion. The ‘paintings’ on the wall could of course only go back to the rebuilding of those walls in 409 B.C., but the genealogical tree would go back to time immemorial. In the Lives of the Ten Orators[138] we hear of Lycurgus, the Eteobutad, as follows. His ancestors derived from Erechtheus, son of Ge and Hephaestos, but his immediate ancestors were Lycomedes and Lycurgus, whom the people had honoured with a public funeral. And the descent of his family from those who held office as priests of Poseidon is on a complete tablet in the Erechtheion written up by Ismenios son of Chalcideus and there are wooden images of Lycurgos and his sons, of Habron, Lycurgos and Lycophron made by Timarchos and Cephisodotos the son of Praxiteles. And Habron dedicated the tablet to his son, and coming in succession to the priesthood he resigned in favour of his brother Lycophron. Hence Habron is represented handing over the trident to him.

By such family trees, by the genealogies and successive priesthoods of royal priestly families, was ancient chronology kept. Argive chronology it will be remembered was reckoned by the years of the consecration of the successive priestesses of Hera[139]. The record was kept in the ancient sanctuary of the Heraion and the statues of the priestesses were set up in front of the temple[140].

With the question of the cult of Athena we have not to deal, but as Poseidon is emphatically one of the ‘other gods’ a word must be said about the subordinate position he comes to occupy. This position is remarkable. To the conservative party as we have seen he was a god of the first importance; it is very noticeable that the chorus of Knights[141] sing first to ‘Poseidon lord of horses’ and only second to ‘Pallas, She of the Citadel.’ Their normal orthodox relation, Athena first, Poseidon second, is reflected in the hymn at Colonos. Yet when we come to examine the ritual of the two divinities we find that their priesthood was conjoint; the Butadae held the priesthood not only of Poseidon but of Athena[142].

These difficulties, these incongruities in tradition, would no doubt be easily solved did we fully know the origin of the cults of Poseidon and Athena. This at present is hidden from our eyes. Kekrops, Pandrosos, Erechtheus, are obviously local. Their worship never spread beyond the hill of Athens, but Poseidon and Athena were worshipped over the whole of Hellas, and whether in Athens they were indigenous or imported cannot at present be certainly said. Herodotus[143] emphatically states that Poseidon originated in Libya, ‘for none except the Libyans originally possessed the name of Poseidon and they have always worshipped him.’ It is in Libya also that this same Herodotus[144] notes that the dwellers round lake Tritonis sacrifice principally to Athena and next to Triton and Poseidon, and from the Libyan women the Greeks obtained the dress and the aegis of the statues of Athena.

If we may hazard a glimpse into things remote or dark, it may be conjectured that the worship of Poseidon and Athena came from Libya to Attica from a people geographically remote, but with racial affinities[145]. That in Libya Athena was, as Herodotus notes, the more important of the two. An old matriarchal goddess, transplanted to Athens in the days of king Erechtheus, she fell when social conditions were patriarchal rather than matriarchal to a subordinate place. Poseidon rather than Athena stood at the head of the Athenian family trees. He headed the conservative aristocratic party. But at some time of political upheaval, possibly even as late as the time of Peisistratos[146], the tide turned, and the ancient matriarchal goddess, as patron of the tyrants and the democracy, reasserted herself. It is Athena not Poseidon who brings Peisistratos back in her chariot to Athens. All this, the prior supremacy of Poseidon, the resurgence of Athena, is reflected in the myth of the Eris, the rivalry, the contest of the two divinities for the land, in the aetiological myth of the planting of the olive-tree and the smiting of the rock with the trident.

To resume, among the ‘other deities’ are first and foremost Kekrops and Erechtheus, ancient eponymous kings, Pandrosos the daughter and paredros of Kekrops and later affiliated to these the immigrant Poseidon. Their ‘sacred things’ are the tomb of Kekrops, the olive, the ‘sea,’ the trident-mark. The list does not exhaust the ‘other deities’ worshipped on the Acropolis; Zeus had altars, Artemis perhaps from early days a precinct. Herakles, though probably an oriental immigrant, was worshipped on the Acropolis at a very early date. It has been one of the sudden corrections sometimes so sharply administered by archaeology to our prejudice that, among the ancient poros sculptures of which so many remains have come to light, Herakles is prominent, Theseus conspicuously absent. But the group of deities and sanctities that cluster round the Erechtheion are sufficient for our purpose, and for that of Thucydides. They show that the Acropolis was the polis for the simple reason that ‘there are sanctuaries in the citadel itself, those of other deities as well’ (as the Goddess).

CHAPTER III.
THE SANCTUARIES THAT ARE OUTSIDE THE CITADEL.

καὶ τὰ ἔξω πρὸς τοῦτο τὸ μέρος τῆς πόλεως μᾶλλον ἵδρυται, τό τε τοῦ Διὸς τοῦ Ὀλυμπίου καὶ τὸ Πύθιον καὶ τὸ τῆς Γῆς καὶ τὸ ἐν Λίμναις Διονύσου (ᾧ τὰ ἀρχαιότερα Διονύσια τῇ δωδεκάτῃ ποιεῖται ἐν μηνὶ Ἀνθεστηριῶνι) ὥσπερ καὶ οἱ ἀπ’ Ἀθηναίων Ἴωνες ἔτι καὶ νῦν νομίζουσιν, ἵδρυται δὲ καὶ ἄλλα ἱερὰ ταύτῃ ἀρχαῖα.

Thucyd. II. 15.

Let us recapitulate. Thucydides has made a statement as to the city before the days of Theseus.—Before this, what is now the citadel was the city, together with what is below it towards about South. In support of this statement he has adduced one argument. The sanctuaries are in the Citadel itself, those of other deities as well (as the Goddess). He now adduces a second, ‘And those that are outside are placed towards this part of the city more (than elsewhere). Such are the sanctuary of Zeus Olympios, and the Pythion, and the sanctuary of Ge, and that of Dionysos-in-the-Marshes (to whom is celebrated the more ancient Dionysiac Festival on the 12th day in the month Anthesterion, as is also the custom down to the present day with the Ionian descendants of the Athenians); and other ancient sanctuaries also are placed there.’

This second argument we have now to examine:—

By ‘this part of the city’ it is quite clear that Thucydides means that portion of the city of his own day which he has carefully marked out; i.e. the citadel plus something, plus ‘what is below it towards about South’; by this we have seen is meant the upper citadel plus the Pelargikon. This second piece of evidence is, like the first, adduced simply to prove the small limits of the ancient city. But Thucydides has expressed himself somewhat carelessly. Readers who did not know where the sanctuaries adduced as instances were, might and have taken ‘towards this part of the city’ to mean ‘towards about South.’ The proximity of the two phrases and the appearance of a relation between them, if in fact there be no relation is, as Dr Verrall[147] observes, ‘a flaw in composition which would not have been passed by a pupil of Isocrates.’ The carelessness of Thucydides is, however, excusable enough. He assumes that the position of the shrines he instances is known as it was by every Athenian of his day. He also assumes that the main gist of his argument is intelligently remembered, that his readers realize that he is concerned with the character and dimensions not the direction of his ancient city.

All that Thucydides tells us is that the sanctuaries outside the ancient city are ‘towards’ it[148]: strictly speaking he gives us absolutely no information as to whether they are North, South, East or West. But ‘towards’ implies approach, and, if we are told that sanctuaries are ‘towards’ a place, we naturally think of ourselves as going there and as finding these sanctuaries on and about the approach to that place.

As to the direction of the approach to the Acropolis there is happily no manner of doubt. In Thucydides’ own days it was where it now is, due West; in the days before the Persian War, the days when the old sanctuaries grew up towards the approach, it was South-West. We know then roughly where to look for our ‘outside’ sanctuaries; they will be about the entrance West and South-West. We must however remember that the whole ancient entrance with its fortifications, the Enneapylon, covered a far wider area than is occupied by the Propylaea now; it took in the whole West end of the hill and part of the North side, as well as part of the South. The area included to the South was, as we have already seen ([p. 34]), much larger than that to the North.

The Sanctuary of Zeus Olympios and the Pythion. The two sanctuaries first mentioned, those of Zeus Olympios and of Apollo Pythios, are linked together more closely than by mere topographical juxtaposition. In the Kerameikos Apollo Patroös[149] had a temple close to the Stoa of Zeus Eleutherios; down near the Ilissos, Zeus Olympios had his great sanctuary ([Fig. 49]), and near it Apollo Pythios had a temenos, and here, where Thucydides is speaking of the most ancient foundation of the two gods, father and son, they are manifestly in close conjunction. This is fortunate for our argument. For it happens that, whereas we know the exact site of the earliest Pythion, of this earliest Olympieion there are no certain remains. From the known site of the Pythion and from the close conjunction of the two we can deduce within narrow limits the unknown site of the Olympion.

Possibly at this point, if the reader knows modern Athens, the words ‘the unknown site’ of the Olympion will rouse an instinctive protest. Surely the site of the Olympieion, with its familiar cluster of Corinthian columns, is of all things most certain and familiar. It lies South-East of the Acropolis not far from the Ilissos (see [Fig. 49]). A moment’s consideration will however show that this Olympieion, though familiar, is irrelevant, nay impossible. It is too remote to be described as towards the ancient city, it is too recent to be accounted an ancient sanctuary. It was, as Thucydides quite well knew, begun by Peisistratos[150].

We begin by fixing the site of the Pythion, happily certain.

Literature alone enables us within narrow limits to do this. In the Ion of Euripides[151] Ion, learning that Creousa comes from Athens, presses her for particulars about that ‘glorious’ city. As a priestling he is naturally interested in all canonical legends, but what he is really eager about is the ancient sacred spot which linked Athens to Delphi. The nursling of Delphi eagerly asks

And is there there a place called the Long Rocks?

Cre. Why ask this? Oh the memory thou hast touched.

Ion. The Pythian honours it and the Pythian fires.

Cre. Honours it! he honours it! Curse the day I saw it.

Ion. What is it? You hate the haunts the god loves best.

Cre. Nothing. Those caves could tell a tale of shame.

But this is not what the pious Ion wants and he turns the subject.

The place at Athens dearest to the Pythian, the place his lightnings honour is on the Long Rocks, and there, we may safely assume, was the god’s earliest sanctuary.

The prologue of the same play tells us where the Long Rocks were, namely on the North of the Acropolis. Hermes, who brought Ion to Delphi, speaks[152]:

‘A citadel there is in Hellas famed,

Called after Pallas of the golden spear,

And, where the northern rocks ’neath Pallas’ hill

Are called the Long Rocks, Phoebus there by force

Did wed Creousa.’

Nor is it Ion only who knows that this place was honoured by the Pythian fires, it is no mere ‘poetical’ figure. Strabo[153], in speaking of a place called Harma in Boeotia, says we must not confuse this Harma with another Harma near Pyle, a deme in Attica bordering on Tanagra. In connection with this Attic Harma, he adds, the proverb originated ‘When it has lightened through Harma.’ Strabo further goes on to say that this Harma, which is on Mt Parnes, to the North-West of Athens, was watched by certain officials called Pythiasts for three days and nights in each of three successive months; when a flash of lightning was observed a sacrifice was despatched to Delphi. The place whence the observation was taken was the altar of Zeus Astrapaios, Zeus of the Lightning, and this altar was in (or on) the (Acropolis) wall between the Pythion and the Olympion.

Euripides, it is clear, is alluding to this definite ritual which of course would be familiar to Ion. That ritual he clearly conceived of as taking place near the Long Rocks. Near the Long Rocks must therefore have stood the altar of Zeus of the Lightning, on the wall between the Olympieion and the Pythion. Not only the Pythion but the Olympieion must therefore have been close to the Long Rocks. The word used by Strabo for wall (τεῖχος) is strictly a fortification wall, and we should naturally understand it of that portion of the Pelargikon which defends the North-West corner of the citadel and abuts on the Long Rocks ([Fig. 2]). It is just here, close to the Pelargikon that we should, from the account of Pausanias[154], expect to find Apollo’s ‘best loved’ sanctuary. Pausanias on leaving the Acropolis notes the Pelargikon, or as he calls it Pelasgikon, and immediately after says ‘on the descent not to the lower parts of the city but just below the Propylaea, is a spring of water, and close by a sanctuary of Apollo in a cave; they think that it was here he met Creousa, the daughter of Erechtheus.’

Pausanias says ‘a sanctuary of Apollo in a cave.’ It is the fact that the sanctuary is in a cave that strikes and interests him. He does not call it a Pythion. But by another writer the actual word Pythion is used. Philostratos[155] describes the route taken by the Panathenaic ship thus: starting from the outer Kerameikos it sailed to the Eleusinion, and, having rounded it, it was carried along past the Pelasgikon and came alongside of the Pythion, where it is now moored. The Panathenaic way has been, as will later be seen ([p. 131]), laid bare; for the moment all that concerns us is that the Pythion is mentioned immediately after the Pelasgikon and was therefore presumably next to it. Philostratos puts what he calls the Pythion in just the place where Pausanias[156] saw his ‘sanctuary in a cave’; the two are identical. Further, any doubts as to where the ship was moored are set at rest by Pausanias himself. He saw the ship and noted its splendour. It stood ‘near the Areopagus.’ The Pythion must have stood at the North-West corner of the Acropolis ([Fig. 46]).

Even if we relied on literary evidence only we should be quite sure that the Pythion of which Thucydides speaks was somewhere on the Long Rocks, at the North-West end of the Acropolis. Happily however the situation is not left thus vague; the actual cave of Apollo has been found, and thoroughly cleared out, and in it there came to light numerous inscribed votive offerings to the god, which make the ascription certain.

From the lower tower at the North-West corner there have always been clearly visible to any one looking up from below three caves ([Fig. 21]), a very shallow one immediately over the Klepsydra, and two others nearer together and somewhat deeper separated from the first by a shoulder of rock. On the plan in [Fig. 22] these are marked Α, Β and Γ. The question has long been raised which of the three belonged to Apollo and which to Pan. As Pausanias[157] first mentions the sanctuary of Apollo in a cave and then passes on to tell the story of Pheidippides, manifestly à propos of Pan’s cave, it has been usual to connect Α with Apollo and Β and Γ, one or both, with Pan.

Fig. 21.

But the identification has never been felt to be quite satisfactory. The cave Α is really no cave at all; it is a very shallow niche. It is impossible to imagine it the scene of the story of Creousa. Moreover it bears no traces of any votive offerings having been attached to its wall, nor have any remains of such been found there.

Between cave Α and cave Β there is a connecting stairway α, α′, α″, but it should be carefully noted that Α has no direct communication with the upper part of the Acropolis nor with the Propylaea. The steep staircase that leads down now-a-days from near the monument of Agrippa to the little Church now built over the Klepsydra looks very rocky and primitive, but really only dates from mediaeval or at earliest late Roman times. It was made at the time that the so-called ‘Valerian’ wall was built, which starts from the Klepsydra and reaches to the Stoa of Attalos ([Fig. 46], dotted lines).

Fig. 22.

We pass to cave Β, which formerly was believed to belong to Pan. Recent excavations[158] leave no doubt that it was sacred to Apollo. The back wall and sides of this cave are thickly studded with niches for the most part of oblong shape, but a few are round. About in the middle of the cave is an extra large niche, which looks as if it had contained the image of a god. Many of the niches still show the holes which once held nails for the fixing of votive tablets. As the cave became unduly crowded with offerings they overflowed on to the rock at the left hand.

So far we are sure that cave Β was a sanctuary, but of whom? If Α did not belong to Apollo we should expect that Β, as next in order, was Apollo’s cave. The ground in front of Β has been cleared down to the living rock and the results of this clearance[159] were conclusive. Exactly in front of Β there came to light eleven tablets or pinakes all of similar type, and all bearing inscribed dedications to Apollo, either with the title ‘below the Heights,’ or ‘below the Long Rocks.’ Cave Β is clearly a sanctuary of Apollo.

The votive tablets are all of late Roman date; it is probable however that owing to the small space available, they superseded earlier offerings of the same kind. The type scarcely varies. Specimens are given in [Fig. 23]. The inscription is surrounded sometimes by an olive wreath and sometimes by a myrtle wreath with characteristic berries. Occasionally the wreath is tied by two snakes. Two inscriptions may serve as a sample of the rest. On No. 1[160] ([Fig. 23]) is inscribed ‘Good Fortune G(aios) Ioulios Metrodorus a Marathonian having borne the office of Thesmothetes dedicated (this) to Apollo Below-the-Long (Rocks).’ In the second[161] instance ([Fig. 23]) the dedicator states that he is ‘King’ (Archon), and the dedication is to Apollo ‘below the Heights.’ Clearly the two titles of the god were interchangeable.

These dedications are of capital importance. It is little likely that unless the custom had been of immemorial antiquity the archons would have sought out an obscure cave-sanctuary in which to place their commemorative tablets. Was there not the temple of Apollo Patroös in the Market Place and the splendid Pythion down near the Ilissos?

Fig. 23.

They chose the cave-sanctuary of Apollo in which to place, at the close of their term of office, their votive tablet because it was in this ancient sanctuary that they had taken their oath of fidelity on their election. At the official scrutiny[162] of candidates for the archonship enquiry was made as to the ancestry of the candidate on both father’s and mother’s side. But it was not enough that he should be a full citizen, he was also solemnly asked whether he had an Apollo Patroös and a Zeus Herkeios and where their sanctuaries were. The Athenians, in so far as they were Ionians, claimed descent through Ion from Apollo and of course through Apollo from Zeus. The sanctuary in the cave was therefore to them of supreme importance. This scrutiny over, the candidates went to a sacred stone near the Stoa Basileios, and there, standing over the cut pieces of the sacrificed victim, they took the oath to rule justly and to take no bribes, and they swore that if any took a bribe he would dedicate at Delphi[163] a gold statue commensurate in value.

The archons had to prove their relation to Apollo Patroös and to dedicate a gold statue if they offended the Pythian god under whose immediate control they stood. Moreover it was not enough that they should swear at the Stoa Basileios. The oath was doubtless older than any Stoa Basileios in the later Market Place. After they had sworn there they had to ‘go up to the Acropolis and there swear the same oath again[164].’ Then and not till then could they enter office. And whither on the Acropolis should they go? Whither but to the cave where a little later they will dedicate their votive tablets, and where still the foundations of an altar stand, the cave of their ancestor Apollo Patroös and Pythios?

Whether the second oath, on the Acropolis, was taken actually in the cave-sanctuary cannot be certainly decided; the votive tablets make it probable and they make quite certain that the cave-sanctuary was officially used by the archons. This fact it is necessary to emphasize. Until these inscriptions were brought to light Apollo’s cave was thought to be of but little importance, curious and primitive but practically negligible. Now that it is clear that the archons selected it as their memorial chapel, such a view is no longer possible. It was a sanctuary not merely of Apollo Below-the-Heights but of the ancestral god, the Apollo Patroös of the archons. Moreover—a fact all important—this Apollo ‘Below-the-Heights’ being Apollo Patroös was also Apollo Pythios. Demosthenes in the de Corona[165], calling to witness his country’s gods, says ‘I call on all the gods and goddesses who hold the land of Attica and on Apollo the Pythian, who is ancestral (πατρῷος) to the state.’ The sanctuary in the cave was a Pythion. Apollo coming as he did to Athens from Pytho was always Pythian whatever additional title he might take, and every sanctuary of his was a Pythion; his most venerable sanctuary was not a temple but a hollowed rock.

The Pythion lies before us securely fixed, primitive, convincing. With the ‘sanctuary of Zeus Olympios’ it is alas! far otherwise. Given that the Pythion is fixed at the North-West corner of the Acropolis, and given that, according to Strabo (see [p. 69]), it was so near the Olympieion that the place of an altar could be described as ‘between’ them, then it follows that somewhere near to that North-West corner the sanctuary of Zeus Olympios must have lain. We may further say that as Thucydides, it will be seen, notes the various sanctuaries and the city-well in the order from East to West, and begins with the sanctuary of Zeus Olympios, it lay presumably somewhat to the East of the Pythion. To the East of the Pythion, near to the supposed site of the temenos of Aglauros, was found an inscription[166] with a dedication to Zeus, but, as inscriptions are easily moveable, no great importance can be attached to this isolated fact. Of definite monumental evidence for the existence of a sanctuary of Zeus where we seek it, we must frankly own at the outset there is nothing certain[167]. It must stand or fall with the Pythion.

Before examining such literary evidence as exists it is necessary to note clearly that Thucydides mentions not a temple but a sanctuary. The great temple near the Ilissos, begun by Peisistratos[168], and not completed till centuries later by Antiochus Epiphanes and Hadrian, is usually spoken of as a temple (ναός), but we have no grounds whatever for supposing that on or near the Long Rocks there was a temple, but only a sanctuary[169], which may very likely have been merely a precinct with an altar. Such a precinct and altar might easily disappear and leave no trace. This is of importance for the understanding of what follows.

When we come to literary evidence one point is clear. Before Peisistratos began the building of his great temple there existed another and earlier place for the worship of Zeus, and this is spoken of as not a temple but a sanctuary. Pausanias[170], when he visited the great temple, wrote, ‘They say that Deucalion built the old sanctuary of Zeus Olympios, and, as a proof of the sojourn of Deucalion at Athens, point to his tomb, which is not far distant from the present temple.’

It has usually been assumed that this earlier sanctuary was on or near the site of the later temple, but, as Prof. Dörpfeld[171] has pointed out, this is no-wise stated by Pausanias. He only says that there was a tomb of Deucalion, not far from the present temple, and that the existence of this tomb made people attribute to Deucalion the building of the early sanctuary. Where the early sanctuary was he does not say. It should be noted that he is careful to use the word sanctuary, not temple, in speaking of the foundation of Deucalion.

From this it follows, I think, that when we hear of a sanctuary of Zeus Olympios, not a temple, there is a slight presumption in favour of its being the earlier foundation. In the opening scene of the Phaedrus[172] an ‘Olympion,’ i.e. a sanctuary of Zeus, is mentioned. Socrates and Phaedrus meet somewhere, presumably within the city walls, for Socrates is later taxed with never going for a country walk. Socrates says, ‘So it seems Lysias was up in town.’ Phaedrus answers, ‘Yes, he is staying with Epikrates in yonder house, near the Olympion, the one that used to belong to Morychus.’ The favourite haunt of Socrates was the agora; a stroll by the Ilissos was to him a serious and unusual country walk. Our Olympion at the North-West corner of the Acropolis would fit the scene somewhat better than the great temple near the Ilissos; but that is all, the passage proves nothing.

A question more important perhaps than any topographical issue remains. Do we know anything of the nature of the god worshipped in the ancient sanctuary, or of the character of his ritual? The question may seem to some superfluous. Zeus is surely Zeus everywhere and for all time, his cloud-compelling nature and his splendid sacrificial feasts familiar from Homer downwards. But then what of Deucalion? Deucalion is a figure manifestly Oriental, a feeble copy of the archetypal Noah. Why does he institute the worship of our immemorial Indo-European Zeus? Are there two Zeuses?

There were, at least at Athens, two festivals of Zeus. Thucydides[173] himself is witness. He tells us of the trap laid for Kylon in characteristic fashion by the Delphic oracle. Kylon was to seize the Acropolis ‘on the greatest festival of Zeus.’ But this ‘greatest festival’ was alas for him! not of the Zeus he, as an Olympian victor, remembered, but of ‘Zeus Meilichios,’ and—significant fact for us—it, the familiar Diasia, was celebrated ‘outside the city.’ This ‘outside the city’ cannot fail, used as the words are by Thucydides himself, to remind us of our sanctuary, also ‘outside.’

What may be dimly discerned, though certainly no-wise demonstrated, is this. The name Zeus is one of the few divine titles as to which philologists agree that it is Indo-European. But the name Zeus was attached to persons and conceptions many and diverse, and here in Athens it was attached to a divinity of Oriental nature and origin. Meilichios[174] is but the Graecized form of Melek, the ‘King’ best known to us as Moloch, a deity who like the Greek Meilichios loved holocausts, a deity harsh and stern, who could only by a helpless and hopelessly mistaken etymology be called Meilichios the Gentle One. His worship prevailed in the Peiraeus, brought thither probably by Phenician sailors, from his sanctuary there came the familiar reliefs with the great snake as the impersonation of the god. It was this Semitic Melek whom Deucalion brought in his ark. When this Semitic immigration took place it is hard to say. Tradition, as evidenced by the Parian Chronicle[175], placed it in the reign of the shadowy Attic king Kranaos, about 1528 B.C.

The sanctuaries of both Zeus and Apollo are alike outside the ancient city. Zeus had altars on the Acropolis itself; Apollo, great though he was, never forced an entrance there. The fact is surely significant. Herodotus[176], it will be remembered, marks the successive stages of the development of Athens: under Kekrops they were Kekropidai, under Erechtheus they were Athenians, and last, ‘when Ion, son of Xuthos, became their leader, from him they were called Ionians.’ Ion was the first Athenian polemarch[177].

One thing is clear, Ion marks the incoming of a new race, a race with Zeus and Apollo for their gods. From the blend of this new stock with the old autochthonous inhabitants arose the Ionians. Zeus and Apollo were called ‘ancestral’ at Athens because they were ancestral; the new element traced its descent from them, and presumably the affiliation was arranged by Delphi; but Apollo, though his sanctuary was on the hill, never got inside.

Ion had for divine father Apollo, but his real human father was Xuthos. This Xuthos, as immigrant conqueror, marries the king’s daughter Creousa. Xuthos was really a local hero of the deme Potamoi[178], near Prasiae. He came of Achaean stock, and therefore had Zeus for ancestor. Hermes, in the prologue to the Ion[179], is quite clear. There was war between Athens and Euboea:

And Xuthos strove and helped them with the sword

And had Creousa, guerdon of his aid,

No home-born hero he, but son of Zeus

And Aiolos, Achaean.

And again[180], when Ion questions his unknown mother as to her husband:

Ion. And what Athenian took thee for his wife?

Cre. No citizen: an alien from another land.

Ion. Who? For a well-born man he needs had been.

Cre. Xuthos, of Zeus and Aiolos the offspring he.

The tomb of Ion, significant fact, was not at Athens but at Potamoi, and Pausanias[181] saw it there. Well may the sanctuaries of Zeus and Apollo stand together.

To return to the question of topography. That the cave marked Β on the plan is sacred to Apollo admits, in the face of the inscribed votive tablets, of no doubt. But a difficulty yet remains. It was noted in speaking of the cave above the Klepsydra that it was too shallow and too exposed to be a natural scene of the story of Creousa. The same objections, though in a somewhat less degree, apply to the cave marked Β. The difficulty, however, admits of an easy solution.

The excavators proceeded to clear out cave Γ, and here they found nothing, no votive tablets, no altar, no inscriptions. But in carrying on their work further East they came on a fourth cave, of a character quite different from that of Α, Β, or Γ. The fourth cave, Δ, has a very narrow entrance; it communicates by a narrow passage with Δ′ and also with Δ″, but Δ″ has been turned into a small Christian church, of which the pavement and a portion of a brick wall yet remain. Here at Δ we have a cave in the full sense of the word, and here we have in all probability the cave or caves, the ‘seats[182]’ (θακήματα) of Pan.

But, be it remembered, Pan was a late-comer; his worship was introduced after his services at Marathon. In heroic days, the time of the story of Creousa, the Long Rocks were shared by the Pythian god and the daughters of Aglauros. The hollow triple cave marked Δ′, Δ″, Δ‴ was once the property of Apollo, and it saw the birth of Ion; later it was handed over to Pan, and is again, as in the Lysistrata[183], the natural sequestered haunt of lovers. Kinesias, on the Acropolis, points out to Myrrhine that near at hand is the sanctuary of Pan for seclusion, and close by the Klepsydra for purification.

In the countless votive tablets[184] to Pan and the nymphs, the type varies little. We have a cave, an altar: round the altar three nymphs are dancing, usually led by Hermes, and, perched on the side of the cave or looking through a hole, Pan is piping to them. The three nymphs, three daughters of Kekrops, were then dancing on the Long Rocks long before Pan came to pipe to them. Concerned as we are for the present with Apollo and his Pythion, it is only necessary to note that their shrine, the sanctuary of Aglauros, must have been near the cave of Pan, somewhere to the East. Euripides[185] speaks of them as practically one:

O seats of Pan and rock hard by

To where the hollow Long Rocks lie

Where, before Pallas’ temple-bound

Aglauros’ daughters three go round

Upon their grassy dancing-ground

To nimble reedy staves.

Where thou O Pan art piping found

Within thy shepherd caves.

Exactly where that sanctuary of Aglauros was excavations have not established. At the point where the cavern is closed by the little modern church, begins a stairway, consisting of seventeen steps (θ-κ-λ-μ-), cut in the rock. These steps manifestly lead up to the steps already known, which lead down, twenty-two in number, from the Erechtheion. This is probably the ‘opening’ (ὄπη) down which the deserting women in the Lysistrata[186] were caught escaping. Still further East is a long narrow subterranean passage, a natural cleft in the rock π-π′, and at the end of this, just above the modern Church of the Seraphim, is supposed to be the sanctuary of Aglauros. Here were found a niche in the rock, the basis of a statue, and some fragments of black-figured vases. Here again there is communication with the Acropolis, but only by a ladder ascending the cliff for about twenty feet at a precipitous point. Moreover the upper part of the stone stairway is of mediaeval date so that it is not likely that the ascent was an ancient one.

The Sanctuary of Ge.—The site of this sanctuary can, within very narrow limits be determined.

Pausanias, in describing the South side of the Acropolis, after passing the Asklepieion, notes the temple of Themis and the monument of Hippolytus. Apropos of this he mentions and probably saw a sanctuary of Aphrodite Pandemos ([p. 105]); he then says ‘there is also a sanctuary of Ge Kourotrophos and Demeter Chloe’; immediately afterwards he passes through the Propylaea. The sanctuary of Ge must therefore have been at the South-West corner or due West of the Acropolis, and presumably somewhere along the winding road followed by Pausanias (see Plan, p. 38). From the account of Pausanias[187] we should gather that Ge Kourotrophos, Earth the Nursing-Mother, and Demeter Chloe, Green Demeter had a sanctuary together; perhaps they had by the time of Pausanias, but the considerable number of separate dedications[188] to Demeter Chloe makes it probable that at least in earlier days these precincts, though near, were distinct.

The union of Ge Kourotrophos and Demeter Chloe is not the union of Mother and Maid, it is the union of two Mother-goddesses. Of the two Demeter belongs locally not to Athens but to Eleusis. Ge Kourotrophos is obviously the earlier and strictly local figure. But Demeter of Eleusis, from various causes, political and agricultural, developed to dimensions almost Olympian, and her figure tended everywhere to efface that of the local Earth-Mother, hence we need not be surprised that the number of dedications to Demeter is larger than that of those to Kourotrophos. Kourotrophos appears among the early divinities enumerated by the woman herald in the Thesmophoriazusae[189], and the scholiast, in his comment on the passage, recognizes her antiquity: ‘either Earth or Hestia; it comes to the same thing; they sacrifice to her before Zeus.’ Suidas[190] states that Erichthonios was the first to sacrifice to her on the Acropolis, and instituted the custom that ‘those who were sacrificing to any god should first sacrifice to her.’

The Sanctuary of Dionysos-in-the-Marshes.

The name Dionysos at once carries us in imagination to the famous theatre on the South side of the Acropolis ([Fig. 16]), and we remember perhaps with some relief that this theatre is, quite as much as the Pythion, ‘towards’ the ancient city; it lies right up against the Acropolis rock. We remember also that Pausanias[191], in his account of the South slope, says ‘the oldest sanctuary of Dionysos is beside the theatre.’ He sees within the precinct there two temples, the foundations of which remain to-day; one of them was named Eleutherian, the other we think may surely have belonged to Dionysos-in-the-Marshes. It is true that the ground about the theatre is anything but marshy now, nor could it ever have been very damp, as it slopes sharply down to the South-East. Still, from an ancient name it is never safe to argue[192]; in-the-marshes may have been a mere popular etymology from a word the meaning of which was wholly lost.

But a moment’s reflection shows that the identification, though tempting, will not do. Thucydides himself ([p. 66]) seems to warn us; he seems to say, ‘not that precinct which you all know so well and think so much of, not that theatre where year by year you all go, but an earlier and more venerable place, and, that there be no mistake, the place where you go on the 12th day of Anthesterion, and where your ancestors went before they migrated to colonize Asia Minor.’

It is most fortunate that Thucydides has been thus precise, because about this festival on the 12th day of Anthesterion we know from other sources[193] certain important details which may help to the identification of the sanctuary.

The festival celebrated on the 12th of Anthesterion was the Festival of the Choes or Pitchers[194]. On this day, we learn from Athenaeus[195] and others, the people drank new wine, each one by himself, offered some to the god, and brought to the priestess in the sanctuary in the Marshes the wreaths they had worn. On this day took place also a ceremony of great sanctity, the marriage of the god to the wife of the chief archon—the ‘king’ as he was called. The actual marriage took place in a building called the Boukoleion, the exact site of which is not known; but certain preliminary ceremonies were gone through by the Bride in the sanctuary in-the-Marshes. The author of the Oration ‘against Neaera[196]’ tells us that there was a law by which the Bride had to be a full citizen and a virgin when she married the king, she was bound over to perform the ceremonies required of her ‘according to ancestral custom,’ to leave nothing undone, and to introduce no innovations. This law, the orator tells us, was engraved on a stele and set up alongside of the altar in the sanctuary of Dionysos in-the-Marshes, and remained to his day, though the letters were somewhat dim.

But this, though much, is not all. The orator goes on to tell us why the law was written up in this particular sanctuary. ‘And the reason why they set it up in the most ancient sanctuary of Dionysos and the most holy, in the Marshes, is that not many people may read what is written. For it is opened once only in each year, on the 12th of the month Anthesterion[197].’ Finally, having sufficiently raised our curiosity, he bids the clerk read the actual oath administered by this pure Bride to her attendants, administered before they touch the sacred things, and taken on the baskets at the altar. The clerk is to read it that all present may realize how venerable and holy and ancient the accustomed rite was. The oath of the attendants was as follows: ‘I fast and am clean and abstinent from all things that make unclean and from intercourse with man, and I will celebrate the Theoinia and the Iobakcheia to Dionysos in accordance with ancestral usage and at the appointed times.’

We shall meet again the precinct, the altar, the stele, the oath; for the present it is all-important to note that the precinct In-the-Marshes was open but once a year, and that on the 12th of Anthesterion. It is impossible, therefore, that this precinct could be identical with the precinct near the theatre on the South slope[198], as this must have been open for the Greater Dionysia, celebrated in the month Elaphebolion (March-April).

The precinct In-the-Marshes has been sought and found; but before we tell the story of its finding, in order that we may realize what clue was in the hands of the excavators, it is necessary to say a word as to the time and place of the festivals of Dionysos at Athens.

Thucydides himself tells us that the Dionysiac festivals were two, an earlier and a later. His use of the comparative—‘Dionysos-in-the-Marshes,’ he says, ‘to whom is celebrated the more ancient Dionysiac Festival,’—makes it clear that, to his mind, there were two and only two. The later festival, the Greater Dionysia, was celebrated in the precinct of Dionysos Eleuthereus; the time, we noted before, was the month Elaphebolion.

The ‘more ancient Dionysiac Festival’ is of course a purely informal descriptive title. But it happens that we know the official title of the two Athenian festivals, the earlier and the later[199].

1. The later festival, that in the present theatre, was called in laws and official inscriptions ‘the (Dionysia) in the town’ (τὰ ἐν ἄστει), or ‘the town Dionysia’ (ἀστικὰ Διονύσια).

2. The more ancient festival was called either ‘the Dionysia at the Lenaion’ (τὰ ἐπὶ Ληναίῳ Διονύσια), or ‘the (dramatic) contest at the Lenaion’ (ὁ ἐπὶ Ληναίῳ ἀγών), or, more simply, ‘the Lenaia’ (τὰ Λήναια).

We have got two festivals, an earlier and a later, the earlier called officially ‘Lenaia,’ or ‘the dramatic contest at the Lenaion’; but were there two theatres also, an earlier and a later? Yes. Pollux[200] tells us there was a Dionysiac theatre and a ‘Lenaic’ one—just the very word we wanted. And to clinch the whole argument we find that the ‘Lenaic’ one was the earlier. Hesychius[201], explaining the phrase, ‘the dramatic contest at the Lenaion,’ says, ‘there is in the city the Lenaion with a large enclosure, and in it a sanctuary of Dionysos Lenaios. In this (i.e. presumably the enclosure) the dramatic contests of the Athenians took place, before the theatre was built.’

This ‘theatre,’ where the plays were performed before the theatre of Eleuthereus was built, was no very grand affair; its seats, it would seem, were called ‘scaffoldings’ (ἴκρια). Photius[202] in explaining the word ikria says, ‘the (structure) in the agora from which they watched the Dionysiac contests before the theatre in the precinct of Dionysos was built.’

Photius, while explaining the ‘scaffolding,’ gives us incidentally a priceless piece of information. This early theatre was in the agora. But then, to raise a time-honoured question, to which we shall later ([p. 132]) return, where is the agora? This question for the present we must not pursue. But the ancient theatre consisted of more than ‘scaffolding’ for seats. It had what was the central, initial, cardinal feature of every Greek theatre, its dancing place, its orchestra; and we know approximately where this orchestra was. A lexicographer[203], explaining the word orchestra, says, ‘a conspicuous place for a public festival, where are the statues of Harmodios and Aristogeiton.’

The agora, conducted by successive theorists, has made the complete tour of the Acropolis, but the statues of the Tyrant-Slayers cannot break loose from the Areopagus,—beneath which ‘not far’ from the temple of Ares, Pausanias[204] saw them. The statues, according to Timaeus, were at the site of the ancient orchestra[205], from the scaffolding of which ‘in the agora’ the more ancient festival (the Lenaia) was witnessed. Here then, somewhere near the Areopagus, we must seek the sanctuary of Dionysos-in-the-Marshes.

The Lenaia, though more ancient than the ‘city Dionysia’ was no obscure festival. Plato[206], in the Protagoras, mentions a comedy which Pherecrates had brought out at the Lenaia, and it can never be forgotten that for the Lenaia, in 405 B.C., Aristophanes wrote the Frogs[207]. The chorus of Frogs[208] assuredly remember that their home is in the Limnae. There they were wont to croak and chant at the Anthesteria, on the third day of which festival, the Chytroi or Pots, came the ‘Pot Contests,’ probably the earliest dramatic performances that Athens saw.

‘O brood of the mere, the spring,

Gather together and sing

From the depths of your throat

By the side of the boat

Co-äx, as we move in a ring;

As in Limnae we sang the divine

Nyseïan Giver of Wine,

When the people in lots

With their sanctified Pots

Came reeling around my shrine.’

The excavations which have brought to light the ancient sanctuary of the Limnae were not undertaken solely, or even chiefly, with that object. Rather the intention was to settle, if possible, other and wider topographical questions: where lay the ancient road to the Acropolis, where the ancient agora, and where the city well, Kallirrhoë. Yet, to some, who awaited with an almost breathless impatience the result of these excavations, their great hope was that the precinct of the Limnae might be found; that they might know where in imagination to picture the ancient rites of the Anthesteria and the marriage of the Queen and those earliest dramatic contests from which sprang tragedy and comedy. The wider results of the excavations will be noted in connection with the Enneakrounos; for the moment it is the narrower, intenser issue of the Limnae that alone concerns us.

So far our only topographical clues have been two. (1) Thucydides has told us that the sanctuary in the Marshes with the other sanctuaries he mentions was ‘towards’ the ancient city; we have fixed the Pythion at the North-West corner of the Acropolis, and as his account seems to be moving westwards, we expect the Dionysiac sanctuary to be West of that point. (2) We know also ([p. 87]) that the ancient orchestra was near the Areopagus. We look for a site for the Dionysia which shall combine these two directions. If that site is also a possible Marsh, so much the better; and here indeed, in the hollow between the Pnyx, Areopagus, and Acropolis, water is caught and confined; but for artificial drainage, here marsh-land must be. This, by practical experience, the excavators soon had reason to know.

Fig. 24.

A portion of the results of the excavations begun by the German Archaeological Institute in 1887[209] and lasting for upwards of ten years is to be seen on the plans in Figs. [24] and [35]. The enlarged plan of a portion of the excavations ([Fig. 24]) for the moment alone concerns us. The first substantial discovery that rewarded the excavators was the finding of the ancient road. It followed, as Professor Dörpfeld had always predicted it would, the lie of the modern road. Roads being strictly conditioned by the law of least resistance do not lightly alter their course. The present carriage road to the Acropolis is a little less devious in its windings than the ancient one, that is all ([Fig. 35]).

Fig. 25.

Just below where the ancient road passes down from the West shoulder of the Acropolis, and at a level much higher than that of the road itself, the excavators came on a building of Roman date and indifferent masonry, which proved to be a large hall, with two rows of columns dividing it into a central nave and two aisles. To the East the hall was furnished with a quadrangular apse. Within this apse was found an altar[210] decorated with scenes from the worship of Dionysos, a goat being dragged to the altar, a Satyr, a Maenad, and the like. This altar would in itself rouse the suspicion that we are in a sanctuary dedicated to Dionysos, but fortunately we are not left to evidence so precarious.

Of far greater interest than the altar, and indeed for our purpose of supreme importance, was another discovery. In the apse, with the altar mentioned and other altars, was found the drum of a column ([Fig. 25]), which had once stood in the great hall; columns just like it are still standing, so that it belongs without doubt to the building. On it is an inscription[211], divided into two columns and 167 lines in length, which from its style may be dated about the third century A.D. Above the inscription, in a relief in pediment form containing Dionysiac symbols, two panthers stand heraldically, one to either side of a cantharus; above is the head of a bull. Inscriptions arranged in this fashion on columns are not unusual in the third century A.D.[212]

The inscription contains the statutes of a thiasos, or club of persons calling themselves Iobakchoi, who met in a place—the hall where the inscription was set up—called the Bakcheion. This is our quadrangular building marked Bakcheion on the plan ([Fig. 24]). The rules, which are given in great detail, are very interesting, but for the present one thing only concerns us—the name of the thiasos, the Iobakchoi. Iobakchos was a title of Dionysos, a title probably derived from a cry uttered in his worship, and, we remember ([p. 85]) with sudden delight, the Gerarae, the attendants of the Queen, promised in their oath to celebrate, in accordance with ancestral usage, the Iobakcheia.

But the building, and even the traces of an earlier structure that preceded it[213], are of late date; we are on the spot, and yet so far the sanctuary in the Marshes eludes us. But not for long. Digging deeper down, to the level of the ancient road, the excavators came on another and an earlier structure, the triangular precinct marked on the plan, and here at last evidence was found that settled for ever the site of the sanctuary of Dionysos-in-the-Marshes.

The sanctuary, for such we shall immediately see it was, is of triangular shape, and lies substantially lower than the roads by which it is bounded. The sides of the triangle face approximately, North, East and South-West. The precinct is surrounded by an ancient polygonal wall, a portion of which from the South end of the South-West side is shown in [Fig. 26]. The material is throughout blue calcareous stone, but the masonry is by no means of uniform excellence or of the same date. At various periods the wall must have undergone repairs. The space enclosed is about 560 square metres. Owing to the fact that the precinct lay deeper than the surrounding roads, sometimes to the extent of two metres, the wall is supported in places by buttresses, only one of which is of good Greek masonry; the rest seem to have been added shortly before the ancient precinct fell into disuse.

Fig. 26.

A notable point about this precinct wall is that there is no trace of any large entrance-gate. We expect a gate at the South-West side, where the precinct is skirted by the main road. Here the wall is well preserved, but there is no trace of any possible gate. The only feasible place is at the South end of the East wall, where there seems to have been a break, and towards this point, as we shall see, the small temple is orientated. Here, then, and in all probability here only, was there access to the precinct.

At the North-West corner the excavators came on a structure so far unique in the history of discoveries. They found a walled-in floor 4·70 m. by 2·80. This floor is carefully paved with a mixture of pebbles, stone, and cement, and is inclined to one corner at an angle of 0·25 m. At this lowest point there is a hole through the wall enclosing the floor, and outside, let into the pavement, is a large vessel, 0·50 m. in diameter, quadrangular above, round below. They had found, beyond all possible doubt, what they had never dared to hope they might find, an ancient Greek wine-press or lenos, and at the finding of that wine-press fled the last lingering misgiving. In [Fig. 27] is a view[214] of the wine-press, which shows clearly how it lies just in the corner of the triangular precinct, with its South-West wall (in the front of the picture) abutting on the Panathenaic way. The stucco floor of the wine-press comes out in dead white. In the background can be seen, to the right, the North aisle of the rectangular Bakcheion, and, to the left, the foot of the Areopagus rock.

Fig. 27.

The wine-press, which is shown in section in [Fig. 28], had, like the precinct, had a long history. It had been rebuilt more than once. The paved floors of two successive structures are clearly visible. The upper one is smaller than the lower, and, of course, of later date. It is, however, below the level of the Bakcheion, and must have been underground when the Bakcheion was built. The lower wine-press is at the same level as the Lesche, on the opposite side of the road, which is known to be of the 4th century B.C. Under this 4th century wine-press is a pavement which must have belonged to a third, yet earlier structure. It may be noted that these wine-presses are in every respect exactly similar to those in use among the Greeks to-day. The wine-press within the precinct is not the only one that came to light; scattered about near at hand were several others. Two can be seen on the plan in [Fig. 35]. It was indeed a place of wine-presses, a Lenaion.

Fig. 28.

Fig. 29.

The wine-press in itself would mark the precinct as belonging to Dionysos, but there was more evidence forthcoming. In the centre of the precinct is the foundation in poros stone of a large altar, 3·10 metres square ([Fig. 29]). In this foundation there once were four holes; three of them remain, and the fourth may be safely supplied. These holes are evidently intended for the supports on which the actual altar-table rested. Such altar-tables are familiar in vase-paintings, and seem to have been in use specially in the cult of Dionysos; they held the wine-jars offered to the god, and baskets of fruit such as those on which the attendants of the Queen took their oath ([p. 85]). Moreover, the actual altar-slab of just such a table has been found in Attica, and it bears an inscription to Dionysos Auloneus[215]. Yet another important point remains. On the West step of the altar foundation a long groove is sunk in the stone. Its purpose is obvious. Both on the Acropolis and elsewhere in sacred precincts such grooves are found, and they served to contain the bases of stelae, on which decrees, dedications, and the like were inscribed. Is it not at least possible that we have here not only the altar on which the Queen took her oath, but the groove in which was set up the very stele on which it was inscribed, the stele which stood ‘alongside of the altar’ (παρὰ τὸν βωμόν)?

We have, then, a precinct secluded from the main road; within it, open to the air, a great altar. But inside this precinct not a single inscription nor any sort of votive offering has come to light. In a precinct so important this at first sight seems strange. The explanation lies to hand. Votive offerings are meant to be seen, meant to show forth the piety of the worshipper as well as the glory of the god. Was it worth while to dedicate an offering in a precinct that was open but for one day in the whole year? Apparently not. This was essentially a ‘mystery’ sanctuary, with no touch of the museum.

In the sanctuary of Dionysos-in-the-Marshes we expect not only precinct and altar but an actual temple, the existence of which we know, not from Thucydides, but from the scholiast[216] on the Frogs of Aristophanes. Commenting on the word ‘marsh’ he says, ‘a sacred place of Dionysos, in which there is a dwelling and a temple of the god.’ Callimachus in the Hekale says,

‘To him, Limnaios, do they keep the feast

With choral dances.’

The ‘dwelling’ may be some building that contained the wine-press; the temple happily has been found, and its position in relation to the precinct is strange and significant.

The foundations of the temple came to light in the South corner of the precinct. It is of small size (3·96 by 3·40 m.), and consists of a quadrangular cella and a narrow pronaos. From its small size it seems unlikely that the pronaos had any columns. The masonry is very ancient. The walls are polygonal, and the blocks of calcareous stone of which they are made are on the South-West side unusually large. In the foundations of the side-walls a few poros blocks occur. There are no steps serving as foundation to either cella or pronaos. From this Professor Dörpfeld concludes that in all probability this temple is earlier than the temple of Dionysos Eleuthereus, close to the skenè of the theatre. The temple of Eleuthereus belonged to the time of Peisistratos; it is more carefully built than the one newly discovered, and it has one step. Early though the newly discovered building undoubtedly is, it was preceded by a yet earlier structure, the walls of which, marked on the plan, lie beneath its foundations.

Quite exceptional is the relation of the temple to the precinct. It does not lie in the middle, and is, moreover, separated from the inner part of the precinct by a wall and a door that could be closed. This separating wall is however apparently later than the temple, which possibly at one time stood free within the precinct. The separating wall is only explicable on ritual grounds. It made it possible for the temple to be accessible all the year round, whereas the precinct, save for one day in the year, was closed.

Are we to give to the ancient sanctuary the name Lenaion? To the sanctuary itself probably not. The meaning of Lenaion, it would seem, is not ‘sanctuary of the god Lenaios,’ but rather ‘place of the wine-press.’ It is noticeable that writers who could themselves have seen the sanctuary never call it Lenaion. Thucydides[217], the writer of the oration against Neaera[218], be he Demosthenes or Apollodorus, and again Phanodemus[219], as quoted by Athenaeus, all speak of it as the sanctuary of Dionysos-in-the-Marshes. Isaeus[220] calls it the Dionysion-in-the-Marshes. On the other hand, when contemporary authors speak of the dramatic contest which was held not in honour of Dionysos Eleuthereus but at the older Dionysia, they speak of the contest as at or on the Lenaion, never as in-the-Marshes. The natural conclusion is that the name Lenaion is applicable to the place where the contests actually took place, namely to the ancient Orchestra and perhaps its immediate neighbourhood. The district of the wine-presses naturally had its dancing place, and that dancing place was called the Lenaion. To this day the peasants of Greece use for their festival-dances the village threshing-floor.

In the theatre of Eleuthereus Dr Dörpfeld[221] has given back to us the old orchestra. He has shown us deep down below the successive Graeco-Roman and Roman stages the old circular orchestra built of polygonal masonry ([Fig. 16]). On this old orchestra, with only wooden seats for the spectators, were acted, we now know, the dramas of Aeschylus, Sophocles, Euripides, nay tradition[222] even says, and we have no cause to doubt its veracity, that Thespis was the first (in 586 B.C.) to exhibit a play in the ‘city’ contest (ἐν ἄστει).

But ancient though it was, before it, as we have seen, came the orchestra in the Limnae. Dr Dörpfeld had hoped that his excavations would give back this orchestra too; this hope has not been fulfilled. Traces have been found of a circular structure on the South slope of the Areopagus and are marked on the plan ([Fig. 46]), but they are of uncertain date, and, if they mark the site of any ancient building, it is probably that of the Odeion of Agrippa. The old orchestra lay at the North-West corner of the Areopagos.

Tradition records the beginning of the contests ‘in the city,’ i.e. in the theatre of Eleuthereus, but the beginnings of the other festivals, the Lenaia and the Chytroi, held in the Limnae, are lost in the mists before. The two are in all probability but different names for the same festival, or rather the Chytroi is the whole ceremony of the third day of the Anthesteria and Lenaia the name given to the dramatic part of the ceremonies. But though we do not know the beginning, and though, as will presently be seen, the ‘Pot-Contests’ went back in all probability to a time before the coming of Dionysos, we have hints as to how the end came, how the splendour and convenience of the great theatre of Eleuthereus gradually obscured and absorbed the primitive contests of the orchestra in the Limnae.

It was, we know, the great statesman Lycurgus who, in the 4th century B.C., built the first permanent stone stage in the theatre and made the seats for the spectators as we see them now. So pleased was he, it would seem, with his theatre that he thought it useless and senseless to have plays acted elsewhere. Accordingly in the Lives of the Ten Orators[223] we learn that Lycurgus introduced laws, and among them one about comic writers ‘to hold a performance at the Chytroi, a competitive one, in the theatre,’ and ‘to record the victor as a victor in the city,’ which had formerly not been allowed. He thus revived the performance which had fallen into disuse.

Lycurgus meant well we may be sure, but he was a Butad[224], he ought to have known better than to pluck up an old festival by the roots like that and think to foster it by transplantation. The end was certain; the old precinct, deserted by its festivals, was bit by bit forgotten, overgrown, and at last in part built over by the new Iobakchoi.

The precinct had lost prestige by the time of Pausanias[225]. Had the temple of Dionysos-in-the-Marshes been above ground he would assuredly not have passed it by. Near to where the precinct once was he saw a building, a circular or semi-circular one, which may have been a last Roman reminiscence of the orchestra, and still of note though it did not occupy the same site; he notes ‘a theatre which they call the Odeion.’ It is probable that this was the theatre built by Agrippa and mentioned by Philostratos[226] as ‘the theatre in the Kerameikos, which goes by the name of the Agrippeion.’

Before leaving the sanctuary in-the-Marshes, a word must be said as to the Anthesteria or, as Thucydides calls it, ‘the more ancient Dionysiac Festival.’ I have tried elsewhere[227] to show in detail that the Dionysiac element in the Anthesteria was only a thin upper layer beneath which lay a ritual of immemorial antiquity, which had for its object the promotion of fertility by means of the placation of ghosts or heroes. On the first day, if I am right, the Pithoigia was an Opening not only of wine-jars but of grave-jars; the second, the Choes, was a feast not only of Cups but of Libations (χοαί); the third, the Chytroi, not only a Pot-feast, but a feast of Holes in the ground and of the solemn dismissal of Keres back to the lower world. That the collective name of the whole feast Anthesteria did not primarily mean the festival of those who ‘did the flowers,’ but rather of those who ‘revoked the ghosts[228].’

But in trying to distinguish the two strata, the under stratum of ghosts, the upper of Dionysos, I never doubted that the Pot Contest on the day of the Chytroi belonged to Dionysos. Dionysos and the ‘origin of the drama’ are canonically connected. It has remained, therefore, something of a mystery how Dionysos, late-comer as he was, contrived to possess himself of the ancient ghost-festival and impose his dramatic contests on a ritual substratum apparently so uncongenial. Religions are accommodating enough, but some sort of analogy or possible bridge from one to the other is necessary for affiliation.

The difficulty disappears at once if we accept Professor Ridgeway’s[229] recent theory as to the origin of tragedy. The drama according to him is not ‘Dorian,’ and, save for the one element of the Satyric play, not Dionysiac. It took its rise in mimetic dances at the tombs of local heroes. When Dionysos came to Athens with his Satyr attendants he would find the Pot-Contests as part of the funeral ritual of the Anthesteria. He added to the festival wine and the Satyrs. Small wonder that comedy, as in the Frogs, was at home in the Underworld, and could in all piety parody a funeral[230] on the stage.

Thucydides has given us four examples of sanctuaries outside the polis which are ‘towards that part’ of it, but again, as in the first clause, he seems to feel that if he has spoken the truth it is not the whole truth, so he saves himself from misunderstanding by an additional clause, ‘and other ancient sanctuaries are placed here.’

It would be idle to try and give a complete list of all the sanctuaries that were situated in this particular region, still more idle to decide of what particular sanctuaries Thucydides was thinking. The precinct of Aglauros and the Anakeion on the North side, the sanctuary of the Semnae and the Amyneion on the West, the sanctuary of Aphrodite Pandemos and that of Themis on the West and South-West are all ‘towards’ the approach. Three out of these, the Amyneion, the sanctuary of the Semnae, and the sanctuary of Aphrodite Pandemos, are of such interest in themselves and so essential to the forming of a picture of the sanctities of ancient Athens that a word must be said of each.

The Amyneion. The Amyneion, or sanctuary of Amynos[231], is known to us only through monumental evidence, brought to light in the recent excavations. Its discovery is one of the things that make us feel suddenly how much of popular faith we, relying as we must almost wholly on literature, may have utterly lost.

Fig. 30.

If after leaving the precinct of Dionysos in-the-Marshes we follow the main road for about 35 metres, we come on a precinct ([Fig. 30]) of much smaller size and of quadrangular shape, which abuts on the road and along the North side of which a narrow foot-path leads up to the Acropolis. The precinct-walls are of hard blue calcareous stone from the Acropolis and neighbouring hills, and the masonry is good polygonal. The entrance-gate (A), which has been rebuilt in Roman times, is at the North-West corner. A little to the East of the middle of the precinct, and manifestly of great importance, is a well (B). The natural supply of this well was reinforced by a conduit-pipe, which leads direct into it from the great water-course of Peisistratos, which will later ([p. 119]) be described. Near the well are remains of a small hero-chapel, and within this was found the lower part of a marble sacrificial table (C), decorated with two snakes. The masonry of the precinct wall, the well, and the shrine all point to a date at the time of Peisistratos. Even before the limits of this precinct were fairly made out the excavators came upon a number of fragments of votive offerings of a familiar type. Such are reliefs representing parts of the human body, breasts and the like, votive snakes, and reliefs representing worshippers approaching a god of the usual Asklepios type. Conspicuous among these was a fine well-preserved relief ([Fig. 31]), depicting a man holding a huge leg, very clearly marked with a varicose vein, exactly where, doctors say, a varicose vein should be. The inscription[232] above the figure is unfortunately so effaced that no facts emerge save that the dedicator, the man who holds the leg, was the son of a certain Lysimachos, and was of the deme Acharnae. The style of the letters and of the sculpture dates the monument as of about the first half of the 4th century B.C. It was clear enough that the excavators had come on the precinct of a god of healing, and a few decades ago the precinct would have been labelled without more ado as ‘sacred to Asklepios.’ We should then have been left with the curious problem, Why had Asklepios two precincts, one on the South, one on the West? We know that Asklepios made his triumphant entry into the great precinct on the South slope in 421 B.C.; if he had had a precinct on the West slope since the days of Peisistratos, why did he leave it?

Fig. 31.

Fig. 32.

But now-a-days in the matter of ascription we proceed more cautiously. We know that votive-reliefs of the ‘Asklepios’ type are offered to almost any local hero, that local heroes anywhere and everywhere are hero-healers[233]. Hence local hero-healers were gradually absorbed and effaced by the most successful of their number, Asklepios. In literature we hear little of the hero-cult of an Amphiaraos, but his local shrine went on down to late days at Oropus. Fortunately in our precinct we have inscriptions that leave us no doubt. On a stele[234] ([Fig. 32]) found there we have an inscription as follows: ‘Mnesiptolemè on behalf of Dikaiophanes dedicated (this) to Asklepios Amynos.’

At first we seem no further; we have the familiar Asklepios worshipped under the title of Amynos, Protector, Defender. A second inscription[235], however, makes it certain that Amynos is not merely an adjective attached to Asklepios, but the cultus title of a person separate from Asklepios. This inscription, of the latter half of the 4th century B.C., is in honour of certain persons who had been benefactors of the thiasos (ὀργεῶνες) of Amynos and of Asklepios and of Dexion. We know who Dexion was; he was Sophocles, heroized, and he, the mortal, came last on the list. Sophocles had a shrine apart, or it may be a separate shrine within the larger one. The same inscription[236] goes on to order that the honorary decree was to be ‘engraved on two stone stelae, and these to be set up, the one in the sanctuary of Dexion, the other in that of Amynos and Asklepios.’

Sophocles[237] though, to us, he is first in remembrance, comes last in ritual precedence; Amynos is first. The history of the little shrine is instructive. Not later than Peisistratos, and how much earlier we do not know, the worship was set up of a local hero with the title Protector, Amynos. At some time or other, perhaps shortly after the pestilence at Athens, which the local Protector had been powerless to avert, it was thought well to call in a greater Healer-Hero, Asklepios, who meanwhile had attained in the Peloponnesos to enormous prestige. The experiment was tried carefully and quietly in the little precinct. Amynos kept his own precedence. No one’s feelings are hurt; the snake of the Peloponnesos is merely affiliated to the local Athenian hero-snake, the same offerings are due to both, the pelanoi, the votive limbs. But the new-comer is too strong; Asklepios waxes, Amynos wanes—into an adjective. Asklepios outgrows the little precinct and betakes himself to a new and grander sanctuary on the South slope.

The precinct and worship of Amynos, though it has no mention in literature, is preserved to us perhaps through its association with the dominant worship of Asklepios; but Amynos was probably only one among many heroes who had their chapels and their family worships scattered along the main road of the city where countless little buildings remain unidentified ([Fig. 35]). If the supposition suggested above ([p. 99]) be correct these local heroes must have had choral dances about their tombs, those choral dances affiliated by the late-comer Dionysos, and ultimately leading to the development of the drama. At the festival of the Anthesteria these local ghosts would be summoned from their tombs on the day of the Pithoigia; on the day of the Chytroi they would be fed and their descendants would hold a wake with revels and dancings.

The Sanctuary of the Semnae Theai or Venerable Goddesses. The site of this sanctuary is practically certain. Euripides[238] in the Electra makes the Erinyes, when they are about to become Semnae, descend into a chasm of the earth near to the Areopagos. Near to the Areopagos there is one chasm and one only, that is the deep fissure on the North-East side, the spot where tradition has long placed the cave of the Semnae[239]. A cave they needed, for they were under-world goddesses. Their ritual I have discussed in detail elsewhere[240]; here it need only be noted that it was of great antiquity and had all the characteristic marks of a chthonic cult. As under-world goddesses the Venerable Ones bore the title also of Arai, Imprecations; they were for cursing as well as blessing; the hill it is now generally acknowledged took its name from them rather than from the war-god Ares. Orestes it will be remembered[241] came to the Areopagos to be purified from his mother’s blood, and he found the people celebrating the Choes; he found them, if our topography be correct, close by, in the precinct of Dionysos-in-the-Marshes.

The Sanctuary of Aphrodite Pandemos. Harpocration[242] in explaining the title Pandemos tells us that Apollodorus in the sixth book of his treatise About the Gods said that this was ‘the name given at Athens to the goddess whose worship had been established somewhere near the ancient agora.’ His conjecture that the goddess was called Pandemos because all the people collected in the agora need not detain us, but the topographical statement coming from an author who knew his subject like Apollodorus, is important. We have to seek the sanctuary of Pandemos somewhere on or close to the West slope of the Acropolis, somewhere near the great square which as we shall see ([p. 131]) stood in front of the ancient well-house and formed the ancient agora.

Pausanias[243] mentions the worship of Aphrodite Pandemos in a sentence of the most tantalizing vagueness. After leaving the Asklepieion he notes a temple of Themis and in front of it a monument to Hippolytus. He then tells at length the story of Phaedra and next goes on ‘When Theseus united the various Athenian demes into one people he introduced the worship of Aphrodite Pandemos and Peitho. The old images were not there in my time, but those I saw were the work of no obscure artists.’ Immediately after he passes to the sanctuary of Ge Kourotrophos and Demeter Chloe and then straight to the citadel.

Of the actual sanctuary of Aphrodite Pandemos not a trace has been found. From the account of Pausanias coupled with that of Harpocration we should expect it to be somewhere below the sanctuary of Ge and above the fountain Enneakrounos, near which was the ancient agora, and of course outside the Pelargikon. When the West slope of the Acropolis was excavated[244] in the upper layers of earth about 40 statuettes of Aphrodite were found, and these must have belonged to the sanctuary. Inscriptions[245] relating to her worship were found built into a mediaeval fortification wall near Beule’s Gate. These, as not being in situ, cannot be used as topographical evidence, but they give us important information as to the character of the worship of Pandemos.

The first[246] of these inscriptions ([Fig. 33]) dates about the beginning of the fifth century B.C. ‘[...]dorus dedicated me to Aphrodite a gift of first fruits, Lady do thou grant him abundance of good things. But they who unrighteously say false things and....’ Unfortunately here the inscription breaks off so the scandal will remain for ever a secret. Aphrodite, it is to be noted, is prayed to as a giver of increase. She does not seem yet to have got her title of Pandemos, but as this occurs in the two other inscriptions found with this one, and they probably all three came from the same sanctuary, this Aphrodite is almost certainly she who became Pandemos.

Fig. 33.

Fig. 34.

The second inscription ([Fig. 34]), dating about the middle of the 4th century B.C., is carved on an architrave adorned with a frieze of doves carrying a fillet. The architrave is broken midway. Only the left-hand half is represented in the figure. This inscription[247] again is partly metrical, forming an elegiac couplet.

‘This for thee, O great and holy Pandemos Aphr[odite,

We adorn with gifts, our statues.’

Beneath in prose and in smaller letters come the names of the dedicators. Pandemos is here quite plainly the official title of the goddess.

The third and latest inscription[248] is carved on a stele of Hymettus marble. It is exactly dated (283 B.C.) by the archon’s name, the elder Euthios. It records a decree made while a woman called Hegesipyle was priestess. The decree, which is too long to be here quoted in full, ordains that the astynomoi should at the time of the procession in honour of Aphrodite Pandemos ‘provide a dove for the purification of the temple, should have the altars anointed, should give a coat of pitch to the roof and wash the statues and prepare a purple robe.’

Aphrodite Pandemos was a ‘great and holy goddess,’ giver of increase. She was no private divinity of the courtesan; the second inscription tells us that she was worshipped by a married woman, who is her priestess. It is literature and not ritual that has cast a slur on the title Pandemos; the state honoured both her and Ourania alike ‘according to ancestral custom.’ Plato[249] in his beautiful reckless way will have it that because there are two Loves there are two Goddesses, ‘the elder one having no mother, who is the Heavenly Aphrodite, the daughter of Ouranos; to her we give the title Ourania, the younger, who is the daughter of Zeus and Dione, and her we call “Of-all-the-People,” Pandemos.’

The real truth was that Aphrodite came to the Greeks from the East and like most Semitic divinities she was not only a duality but a trinity.

When Pausanias[250] was at Thebes he saw the images of this ancient Oriental trinity and he knew whence they had come. ‘There are wooden images of Aphrodite at Thebes so ancient that it is said they were dedicated by Harmonia and that they were made out of the wooden figure-heads of the ships of Cadmus. One of them is called Heavenly, another Of-all-the-People, and the third the Turner-Away.’ The threefold Aphrodite came from the Semitic East bearing three Semitic titles: she was the Queen of Heaven[251], she was the Lady of all the People, Ourania and Pandemos, what the third title was which the Greeks translated into Apostrophia we do not know; as already noted it took slight hold. At Megalopolis[252] we see how the third title of the trinity faded. There close to the house where was an image of Ammon made like a Herm and with the horns of a ram, there—significant conjunction—was a sanctuary of Aphrodite in ruins, with the front part only left and it had three images, ‘one named Ourania the other Pandemos, the third had no particular name.’ So it was that the Greeks lost the trinity and kept, all they needed, the duality.

The Greeks themselves always knew quite well whence came their Heavenly Aphrodite, she of Paphos, and she of Kythera. Herodotus[253] is explicit. He is telling how some of the Scythians in their passage through Palestine from Egypt pillaged the sanctuary of Aphrodite Ourania at Ascalon. ‘This sanctuary,’ he says, ‘I found on enquiry is the most ancient of all those that are dedicated to this goddess, for the sanctuary in Cyprus had its origin from thence, as the Cyprians themselves say, and that in Kythera was founded by Phenicians who came from this part of Syria.’ Pausanias[254] says ‘the first to worship Ourania were the Assyrians, next to them were the dwellers in Paphos of Cyprus, and the Phenicians of Ascalon in Palestine. And the inhabitants of Kythera learnt the worship from the Phenicians.’

The Oriental origin[255] of Ourania, Queen of Heaven, the armed goddess, the Virgo Caelestis, was patent to all; but Aphrodite in her more human earthly aspect, as Pandemos, goddess of the people and of all increase, was so like Kourotrophos, like Demeter, that she might easily be thought of as indigenous. Yet her ritual betrays her. For the purification of her sanctuary we have seen there was ordered a dove. Instinctively we remember that when Mary Virgin[256] went up to the temple of Jerusalem for her purification she must take with her ‘a pair of turtle-doves or two young pigeons.’ In the statuettes of Paphos, Aphrodite holds a dove in her hand; the coins of Salamis in Cyprus are stamped with the dove[257]. At the Phenician Eryx when the festival of the Anagogia[258] came round, and Aphrodite Astarte went back to her home in Libya, the doves went with her, and when they came back at the Katagogia, a white multitude, among them was one with feathers of red gold, and she was Aphrodite.

CHAPTER IV.
THE SPRING KALLIRRHOË-ENNEAKROUNOS ‘NEAR’ THE CITADEL.

καὶ τῇ κρήνῃ τῇ νῦν μὲν τῶν τυράννων οὕτω σκευασάντων Ἐννεακρούνῳ καλουμένῃ, τὸ δὲ πάλαι φανερῶν τῶν πηγῶν οὐσῶν Καλλιῤῥόῃ ὠνομασμένῃ—ἐκείνῃ τε ἐγγὺς οὔσῃ τὰ πλείστου ἄξια ἐχρῶντο, καὶ νῦν ἔτι ἀπὸ τοῦ ἀρχαίου πρό τε γαμικῶν καὶ ἐς ἄλλα τῶν ἱερῶν νομίζεται τῷ ὕδατι χρῆσθαι.

The argument now stands as follows. As evidence that the old city was the present citadel with the addition of what is below it towards about South Thucydides has adduced two facts: 1st, that the sanctuaries are in the citadel, those of other deities as well (as the Goddess); 2nd, that those that are outside are placed towards this part of the city more (than elsewhere). Instances of such outside shrines are the sanctuary of Zeus Olympios and the Pythion, and the sanctuary of Ge, and that of Dionysos-in-the-Marshes. This last is defined, to prevent confusion with the later sanctuary of Dionysos Eleuthereus, as the scene of the earlier Dionysia. Finally, other ancient sanctuaries also (not named) are placed here.

We next come to the third fact adduced as evidence, namely, a statement as to the position of the ancient city spring, as follows: ‘And the spring which is now called “Nine-Spouts,” from the form given it by the despots, but which formerly, when the sources were open, was named Fair-Fount—this spring (I say) being near, they used for the most important purposes, and even now it is still the custom in consequence of the ancient (habit) to use the water before weddings and for other sacred purposes.’ Was ever argument stated in fashion more odd, involved, and utterly Thucydidean?

A spring which was once called Kallirrhoë and now Enneakrounos is ‘near,’ i.e. is near the ancient city as above defined, and is now used for weddings and the like. Why does Thucydides, who is ‘least of all mortals a gossip,’ tell us about the water and the weddings? Why refer to the history of the fountain at all? Because, as in the case of the Anthesteria, the reference to things ancient is part of his argument. The train of thought is this. The water of Nine-Spouts is now used for weddings. Why? On the face of it there seems no particular reason. The fountain ‘Nine-Spouts’ has water enough and to spare. But the fountain ‘Nine-Spouts’ was not always there, it replaced ‘Fair-Fount,’ and this spring the ancient Athenians used only for ‘most important’ purposes. Again, why? Well, clearly because there was not enough of it for general use. It was ‘near,’ and yet they reserved it for special purposes. We may gather, then, from the account of Thucydides, though he does not expressly state it, the despots not only changed the name but increased the ‘water supply[259].’

As to where the spring was, save that it is ‘near,’ Thucydides says absolutely nothing. It might be North, East, South, or West. We who have followed him step by step down the western slope, from the Olympieion and Pythion to the sanctuary of Ge and to the sanctuary of Dionysos-in-the-Marshes, expect to find ‘Nine-Spouts’ somewhere near these sites, somewhere in the depression enclosed by Acropolis, Pnyx, and Areopagos. But we must bear in mind that this expectation is based on our identification of the previous sanctuaries, not on any words of Thucydides about the spring.

But when we ask, as we inevitably must, where did Pausanias see the famous fountain, we are in better case. Pausanias[260] saw ‘Nine-Spouts’ near to the Odeion, and the Odeion he saw immediately after the statues of Harmodios and Aristogeiton, on the slope of the Areopagos. Immediately after the Enneakrounos, ‘beyond the fountain,’ as he says, Pausanias[261] saw the temples of Demeter and Kore, which can scarcely be separated from the Thesmophorion on the Pnyx. Somewhere adjacent to both Pnyx and Areopagos we should, from Pausanias, expect to find ‘Nine-Spouts,’ and there find it we shall.

It is fortunate for us that Thucydides was so explicit about the fountain. He gives us not merely a fountain called Fair-Fount but a fountain called Fair-Fount that was turned into Nine-Spouts. This is fortunate, because the word translated ‘Fair-Fount,’ Kallirrhoë, is a term so general that it might be applied to almost any spring. If in travelling through Greece to-day you stop to drink from a spring and ask your guide its name, he will, three times out of four, tell you it is Mavromati, Black-Eye, because that is a term so general as to be safely applicable. So at Athens there was, certainly in later days and possibly even in the time of Thucydides, another Kallirrhoë far away on the Ilissus. As Socrates, in the Axiochos[262], was going out towards Kynosarges and had reached the Ilissos he heard some one shouting to him, and turning round he saw Kleinias running towards Kallirrhoë. Clearly this was another Kallirrhoë, not the one near the Pnyx. How this duplication of Kallirrhoës at Athens arose will later ([p. 143]) be considered. The Kallirrhoë we are in search of is the Fair-Fount which became the Nine-Spouts, that and no other.

It is worth noticing how quickly the spring lost its old name. People were, no doubt, very proud of the new Nine-Spouts. Herodotus[263] naively assumes that in the days of the Pelasgians Fair-Fount was called Nine-Spouts. The Athenians said that their expulsion of the Pelasgians from Attica was justified, for ‘the Pelasgians who were settled under Hymettus used to make excursions thence and do lawless deeds. Their daughters used constantly to go to the Enneakrounos for water, for at that time the Greeks had no household servants, and whenever they came the Pelasgians used to offer them violence out of insolence and contempt.’ There must have been people alive in the days of Thucydides whose fathers remembered the change made by the despots, yet the name Fair-Fount was, when Thucydides wrote, evidently a matter of antiquarian knowledge.

The question now before us is, Have we evidence that a spring, naturally small but reinforced and rearranged at the time of the despots, existed in the district enclosed by the Pnyx, Areopagos, and Acropolis? A glance at the plan in [Fig. 35] will show that such evidence does indeed exist. In the Pnyx rock at the point marked Y is the spring Kallirrhoë, Fair-Fount. It has been reinforced by water from the district of the Ilissus, brought in the conduit of Peisistratos. In front of the ancient Kallirrhoë once stood a Fountain-House, also of the date of the despots, the Fountain-House called Nine-Spouts, Enneakrounos.

The evidence for this threefold statement must be examined in detail. But first a word must be said as to the geological conditions of the site so far as they bear on the water-supply of Athens.

For her water-supply, and especially for her drinking water, Athens depends, has always depended, not on her rivers but her wells. In describing the Enneakrounos Pausanias[264] says, ‘There are wells throughout all the city, but this is the only spring.’ His statement as regards the spring is not strictly correct. Besides Kallirrhoë the ancient city possessed two natural springs, and these both on the Acropolis itself, the Klepsydra at the North-West corner and the spring in the precinct of Asklepios on the South slope. About the wells he is right. The plain on which Athens stood was, owing to its geological structure, amply supplied with wells. Its uppermost stratum is of calcareous stone, the material of which the hills of Lykabettos, of the Mouseion, and the Acropolis are all formed. Through this stratum rain can freely filter. But beneath this calcareous layer is a second stratum of slate and marl; this is practically impermeable, and here water collects into wells.

Fig. 35.

Wells, then, occur sporadically all over Athens and the Athenian plain, but nowhere in such abundance as in the district under discussion[265]. The Pnyx and the Mouseion on the one side, the Areopagos and Acropolis on the other form, as will readily be seen by reference to [Fig. 46], a sort of trough, in which both rain and subterranean water are caught and must necessarily accumulate. As the ground slopes towards the North and the West the water accumulated cannot make its way towards the Ilissos. Its only outlet is the narrow and inadequate passage between the Pnyx and the Areopagos to the Eridanos. It is not surprising that, though the district lies high above the bed of the Eridanos, it was somewhat marshy. That its watery character was early turned to account and led to a dense population is shown by the fact that no less than 100 wells have been sunk within its narrow limits. These wells will be seen dotted about all over the plan in [Fig. 35]. These wells for subterranean water are frequently reinforced by cisterns for collecting rain-water. The cisterns are easily distinguished from the wells by the fact that they are lined with cement. Sometimes an old well which has presumably run dry has been turned by a coat of cement into a cistern. It is very remarkable that, long before the days of Peisistratos, elaborate systems existed for collecting water, in wells, cisterns, and conduits; one canal extended as far as the Odeion of Herodes Atticus, and followed a course almost coincident with that of Peisistratos, which it long preceded. Its complex of wells is clearly seen at T in [Fig. 35], a little to the North of the ‘Branch Conduit to Koile.’

It is beside our purpose to examine in detail the artificial water-supply[266] of the district before the time of Peisistratos. That such a system existed is worth noting, because it shows that the district is a good site for the Limnae, and also that it was from early days thickly populated.

Our immediate concern, however, is to fix, if possible, the site of Kallirrhoë. Nor is this difficult. As the traveller goes by the modern carriage road from the ‘Theseion’ to the Acropolis, and as he nears the Pnyx he will see on his right a number of rock-chambers and channels cut in the rock, originally buried out of sight but laid bare by the making of the modern road. These are shown in [Fig. 35] to the right and left of the spot marked Kallirrhoë, and appear more plainly on the enlarged plan in [Fig. 38], where they are marked r¹-r¹⁰. They are a succession of rock-hewn wells and cisterns and channels, dating from early Greek to Roman times. Their number is additional evidence that the rock of the Pnyx had a regular system for collecting water, but of the series two only concern us, those marked r⁶ and r⁷.

An enlarged plan of the wells r⁶ and r⁷, with their connecting passages and chambers, is given in [Fig. 36]. A detailed description of it is important, because these chambers, recognized as forming the ancient Kallirrhoë, are now closed to the public by a locked gate, behind which few visitors to Athens penetrate.

Fig. 36.

A narrow stairway, a-b, leads into a chamber (Y) hewn in the heart of the rock. This chamber is about 4 metres square, and has an arched roof. Immediately opposite the entrance to Y, in the Western wall, a niche 1·80 m. deep has been cut (C). In this niche the shaft of a well (r⁷) has been sunk 2 metres deep. This is clearly shown in section in [Fig. 43]. In front of the well was a barrier, so that water could be drawn without fear of falling in. Over the well, about 0·80 metre above the pavement, was a small niche, which may have held an image. From the entrance of the chamber Y, about 1·30 metres high from the ground, there is a channel, n-p, worked in the rock. It has a slight inclination towards the niche C, and was obviously meant to collect the water that oozed from the vaulted roof and the walls. Later it was used as a conduit for the new water-supply brought by Peisistratos. Remains of a lead pipe and a terra-cotta conduit were found at m.

For,—doubt is impossible,—we have here in the niche at C the ancient Kallirrhoë. The large rock chamber Y marks it out from the other wells. Its importance down to Roman times is shown by the fact that the chamber Y is paved with a rich mosaic, the patterns of which are like those made elsewhere in Athens in the time of Hadrian. The ancient well must have kept its sanctity, otherwise it would not have been so adorned. After the well had run dry, and when the water-supply was purely artificial, the walls and ceiling were carefully cemented and the cement was later renewed. Such a coating would of course have been impossible when the roof and walls were dripping with natural water.

At the right hand of the entrance to Y was a passage, e-f, leading down by steps into a large elliptical chamber, r⁶. This chamber, presumably a cistern, was paved in Roman days with marble slabs, but below the marble pavement is a stucco pavement of Greek date. From this cistern leads a channel, i, which may have led to the well-house of Peisistratos, or, as suggested in the restoration ([Fig. 43]), to a smaller subordinate fountain.

The supply of water at Kallirrhoë was slender. We have seen that efforts were made to reinforce it by well-sinking, by conduits, by cisterns. But, though the Athenians found the water of Kallirrhoë adequate for their ritual baths, they had other needs, and, as the city grew and grew, the effort to cope locally with the increasing demand proved futile. There was a crying need for water from a distance, a great popular need such as the despots loved to supply. Water was needed, and water was brought in a supply practically inexhaustible, from the district of the upper Ilissos.

By a happy chance in the history of excavations, long before the search for the aqueduct of the despots began, another aqueduct, the work of another despot, had been brought to light—the aqueduct that Polycrates made for the Samians. At the close of his account of Polycrates, Herodotus[267] tells us he had lingered long over the affairs of the Samians ‘because they possessed three of the most wonderful works ever accomplished by the Greeks.’ The first and the only one of these wonders that concerns us was a great aqueduct bored through a mountain 150 fathoms high. The length of the tunnel, he goes on to say, was seven stadia, the height and the breadth eight feet each way. Through this tunnel there went a second passage, 20 cubits deep by three feet wide, through which the water is carried along in tiled pipes from a great spring to the city of Samos. The architect of this tunnel was a Megarean, Eupalinos, son of Naustrophos.

Possibly, pace Herodotus, even if the Samians had had no aqueduct he would anyhow have told us the story of the ring; be that as it may, his account of the first wonder, the aqueduct, is invaluable, and has been fully substantiated. Never was a town by nature worse off for its water-supply than Samos, and rarely has one been supplied by a more astonishing piece of engineering. The ‘great spring’ Hagniades has been found[268], the tunnel with its double channel, even the very earthenware pipes laid down by Eupalinos. We know perfectly well what to expect in an aqueduct made by the despots.

The excavators naturally sought for the conduit of Peisistratos in the immediate neighbourhood of Kallirrhoë, and there, close up to the Pnyx rock, they found it, at a distance of about 40 metres from the rock chamber Y. From that point up to the South of the Odeion of Herodes Atticus its course has been completely excavated. It is best seen in Professor Dörpfeld’s official plan ([Fig. 46]). Just South of the Odeion the conduit could not be cleared out, because of its damaged condition and the mass of débris that had fallen over it. Between the Odeion and the Dionysiac theatre it runs beneath an ancient road, and passes within the precinct of Dionysos, between the earlier and later temples. Beyond that point its course has not been excavated in detail, but beneath the modern Russian church a conduit passes which must be its continuation, and this leads on to the water-course[269] discovered long ago, now utilized for watering the Royal Gardens. This water is known to come from the upper valley of the Ilissus ([Fig. 49]).

The main conduit ran, then, from the upper valley of the Ilissus to the great reservoir basin marked on the plan in [Fig. 35], but from this main conduit several branches can be traced; the most important are the branch tunnel that leads to the district of Koile and a smaller branch that goes off to water the Amyneion. Other ramifications can be traced, the object of which is not always clear; they probably occur at points where in piercing the tunnel veins of water were reached, and some served to bring to the main conduit subsidiary supplies from the Hill of the Muses and from the Acropolis.

Only those, as Professor Dörpfeld[270] himself remarks, who have taken the trouble to get right down into the tunnellings and cross tunnellings and explore them thoroughly so far as they can be explored, can form any idea of the magnitude of the work. Sometimes it is possible to stand upright in the conduit, some portions can only be reached on the hands and knees. The fact is borne in upon any one and every one who has made even a brief exploration, he feels himself unquestionably exploring what must have been the main artificial water-supply of ancient Athens, and here, if such a supply were needed, must have been the centre of the ancient city life.

The aqueduct is dated securely by comparison with the work of Eupalinos at Samos as of the time of the despots. Two striking analogies are observable between the aqueduct of Peisistratos at Athens and that of Polycrates at Samos. These are the character of the pipes, and the system of shafts. The separate pieces of the pipes at Athens are from 0·60 m. to 0·61 in length, not counting the junction points. They are made of fine yellowish clay; inside they are protected by a red glaze, outside they are left rough, except that at each end they are glazed and have a double stripe of glaze round the middle and round each end. In length and diameter they correspond with the Samos pipes, which Professor Dörpfeld carefully inspected for comparison[271]. The Samos pipes also are actually decorated with stripes, only the stripes at Samos are incised, those at Athens painted.

The same correspondence is notable in the way the pipes are joined together: both at Athens and Samos the pipes are soldered together with lead, and provision is made at both places for cleaning them. An elliptical shaped hole large enough to admit the hand is left, and is provided with a cover. A specimen of the Athenian pipes is shown in [Fig. 37], and side by side with it a section of the conduit with the pipe in position.

Fig. 37.

The pipes bear abundant traces of long use and frequent repair. In quite early days they seem to have got crusted with lime deposit from the water, and in some cases quite choked up, the water then flowed over the pipes and flooded the main channel to two-thirds of its height. In some places, where the rock was soft, it seems to have got worn away and fallen in, and portions of the tunnel became useless. New borings were made for about 30 metres and new pipes put in; these were quadrangular instead of round, but in the disused portion of the tunnel the old round pipes still lie about.

Secondly, as at Samos, at intervals of from 30 to 40 metres, both tunnels alike are provided with shafts, which served when the tunnels were first made for the clearing away of the rock fragments, and which were made use of for the like purpose when the conduit was excavated. These shafts are sunk perpendicularly; one of them reached down to a depth of 12 metres, so low does the conduit in places lie.

Of cardinal importance to us is the point at which the conduit debouches, because near to that point we may hope to find the fountain-house ‘Nine-Spouts.’ The conduit ends in an arrangement which is somewhat surprising, and which will be best understood by reference to [Fig. 38]. To the extreme left, at a point near letter B, the conduit emerges. It here consists of a massive channel built of blocks of poros stone, indicated by the thick black lines on the plan. At point a⁴ it ends in the Pnyx rock. But, and this is the odd thing, at a³, about eight metres before the channel ends, a pipe issues from the stone channel and running parallel to the Pnyx rock conducts the water to the main reservoir (Haupt-Bassin). A similar arrangement has been observed in the aqueduct at Samos. There, too, the conduit pipe leaves the rock channel before it ends. It is conjectured[272] that this was a plan intended to mislead an enemy who might desire to cut off the water-supply.

The conduit actually debouches at a⁵ into the great reservoir from which the new fountain-house Nine-Spouts must have been fed. Here, at the reservoir, we find indications of three successive structures. First a structure of very early date, possibly of the time of Solon. Second that of Peisistratos. Third a late Roman structure. Of the two earlier structures no masonry remains, but the position and dimensions can roughly be made out by markings on the Pnyx rock, out of which the West side of the basin was hewn. The exact size of the original basin, which was smaller than the later one, cannot now be determined. In the time of Peisistratos it was enlarged and deepened; the floor of the basin was sunk nearly 1·50 metres deeper. The great basin of Peisistratos was lined with masonry, the blocks of which have now disappeared. In Roman days the place of the great basin of Peisistratos was taken by a quite small structure. This change must have taken place before the building of the late Roman villa which occupied the place where once the ‘Nine-Founts’ stood. When the villa was built the great reservoir had for some time been disused, and the water from the aqueduct, not being needed on the spot in any large quantity, was carried by pipes to the lower city to the North for the supply of the new Roman market-place. These alterations as to water-supply, it should be noted, are of the first importance in questions of topography, and change in the direction or the extension of an aqueduct is naturally the index of a shifting of population.

Fig. 38.

The restoration by Professor Dörpfeld ([Fig. 38]) is, it must clearly be understood, to a large extent conjectural. It must be consulted strictly in conjunction with the plan in [Fig. 35], where the actual remains of Greek date are clearly marked in solid black lines. So used it can be of great service in helping us mentally to reconstruct scattered fragments of masonry that would otherwise be unintelligible.

Some of the details of the restoration have been suggested by the water-works discovered at Megara, which are in some respects better preserved than those at Athens. At Megara are extant not only a great conduit to bring water from a distance but an elaborate arrangement for utilizing it consisting of a reservoir and a pillared draw-well besides a fountain house. It is very probable that the works of Theagenes served as a model to Peisistratos, and therefore before the draw-well and fountain house of Peisistratos are discussed a word must be said of the excavations at Megara.

Fig. 39.

Pausanias[273] begins his account of the city of Megara somewhat abruptly thus. ‘In the city there is a fountain. And Theagenes built it for them. About him I have already mentioned that he gave his daughter in marriage to Kylon the Athenian. This Theagenes, having possessed himself of the tyranny, built the fountain, and from its size, its decorations, and the number of its columns, it is worth looking at. Water flows into it called the water of the Sithnidian nymphs.’ After the excavations at Athens, the fountain or, as perhaps it is best called, the well-house of Theagenes at Megara was sought and found[274] at the bottom of the Eastern Acropolis of Megara, called Karia. The aqueduct leading to the reservoir was excavated for a considerable distance, and proved to be a structure closely resembling those found at Athens and Samos. Eupalinos it will be remembered was a native of Megara. The draw-well, the supporting walls of which are well preserved, was about 15 by 20 metres in size and built of Kara limestone, a material much used in the 6th century B.C. for the foundations and stylobates of buildings. All round the side whence water was drawn was a low parapet wall. This wall shows signs in many places of being worn away by the friction of ropes and dripping of water. The block shown in [Fig. 39] is closely paralleled by the block found in Athens and placed beneath it for comparison.

Not only, then, at Athens did a despot build a well-house and artificially increase a supply of holy water. The original spring at Megara was sacred to the Sithnidian nymphs; we do not know what nymphs guarded Kallirrhoë at Athens; there were plenty about, for to this day close at hand is the Hill of the Nymphs. Dionysos who dwelt so near was called Limnaios, He-of-the-Marshes, Phanodemos[275] says, because he invented the blending of must with water; hence, he adds, ‘the springs are called Nymphs and nurses of Dionysos, because water mixed with wine increases it.’

We return to the water-worn stone, the details of which are shown in [Fig. 40]. This stone is of great architectural importance. From it can be deduced not only the date of the building to which it belonged, but also something of its dimensions and general appearance. The date is fixed by the clamp mark at C. The clamp itself has disappeared, but its shape is proved by the mark of its insertion. Clamps of the

shape only appear at Athens in buildings of about the date of Peisistratos, e.g. on the earlier temple of Dionysos Eleuthereus. Our stone belonged to a building of the date of Peisistratos. As regards the character of the building, it is clear from the curve at e which is a segment of a circle, that the stone was at this point cut away to receive a pillar. The unworn condition of the stucco at b leads Professor Dörpfeld to conclude that the stone was a corner stone, the angle protecting the stucco from friction. The distance between these two points, e and b, gives the measurement of the intercolumniations. From this one stone it is certain that a draw-well of the date of Peisistratos existed and that it was surmounted by a colonnade. Its appearance must have been somewhat that of the draw-well (Schoepf-brunnen) restored in [Fig. 38]. We pass to the consideration of the fountain house Nine-Spouts.

Fig. 40.

The great open square marked ‘place of the Enneakrounos’ ([Fig. 38]) is really the site of Nine-Spouts. This is clear from many considerations. 1. Nine-Spouts must have stood over or in front of Fair-Fount which it superseded. Over it would be an impossible situation, because of the Pnyx rock, so we may securely place it in front. 2. Nine-Founts must have stood about two metres below the level of the basin, from which it was fed, in order that the water might flow easily in. 3. At K 2 and K 3 are the beginnings of two ancient subterranean canals which must have been intended to carry off the superfluous water from Nine-Spouts. 4. Straight down to this open place comes the footway from the Acropolis and thither also all the rest of the roads ultimately converge. 5. The place must have been in Greek times an open place, as no foundations of Greek buildings have been found, only the remains of a great Roman house, and under it countless wells.

This Roman house consisted of a large atrium with a peristyle of twelve columns and several small chambers surrounding it. The walls are a patchwork of materials of all kinds, and even the bases of the columns are made up of fragments from other buildings. One of these fragments belonging to the draw-well we have already discussed, another, we shall immediately see, belongs to Nine-Spouts itself.

Fig. 41.

Can we form any mental picture of Nine-Spouts? Fortunately vase-paintings come to our aid. It is not a little remarkable that in the decoration of black-figured water-vases (hydriae) of the 6th century B.C., there appears a sudden fashion in fountain-houses. Of hydriae so decorated the British Museum contains no less than ten. One of these[276] is reproduced in [Fig. 41]. The Fountain-House depicted is of the usual shape, a tetrastyle Doric portico. The architectural details are very clear, the triglyphs and guttae standing out in white. In actual architecture they would both be painted blue. Four maidens are water-drawing. Two of them are hanging up wreaths. Over three of them their names are inscribed Iope, Rhodopis, Kleo. But what at once arrests our attention is the arrangement of the water-spouts. Facing us are three, a lion’s head and two horsemen, to either side of these is a lion’s head spout; that makes not a Nine-Spouts but a Five-Spouts. But, drawn in perspective as they must be, do not the side spouts each represent three? It is at least probable that we have an arrangement like that restored in [Fig. 38], three spouts facing, and three at each side. Lion-spouts are of course frequent in Fountain-Houses. The horsemen of our vase are unique; they give the Fountain-House a dashing despotic air.

Fig. 42.

We know then just what sort of architectural fragments, we might expect to find; we can imagine a fragment that would be conclusive. A ‘Doric’ portico might belong to more than one kind of building, a lion’s head spout could belong only to a Fountain-House. No lion’s head has been found, but instead, what is as good for our purpose, a stone hollowed out for the reception of a lions head. This stone is shown in [Fig. 42]. Not only is the space for the lion’s head evident, but behind is clearly visible the hole for the pipe. The block is of blue calcareous stone such as is found both on the Acropolis and the Pnyx. Of exactly the same limestone is a small remnant of a polygonal wall from the South boundary of the precinct of the Fountain-House.

The plan in [Fig. 38] makes the general disposition of the place of the Enneakrounos clear, the large reservoir behind (Haupt-Bassin), immediately in front of it the draw-well (Schoepf-brunnen), and to the right of the reservoir, and of course equally fed by it, Nine-Spouts (Lauf-brunnen). In front a great open space. What is matter for conjecture is the exact site and size of Nine-Spouts. A clear view of the relation of Nine-Spouts to Fair-Fount is given in the sectional restoration[277] in [Fig. 43]. There we see the vaulted rock chamber Y, the actual well, Kallirrhoë, to which it led, and in front of it, the modern road intervening, Nine-Spouts or Enneakrounos itself. In front of that again the open space, possibly once enclosed, was the heart and centre of the agora.

Fig. 43.

Before we pass to the question of the agora it may be worth while to notice that the well-house, Enneakrounos, Nine-Spouts, was known as late as the seventeenth century to have been on the West slope of the Acropolis. In the curious old plan, then drawn by Guillet and Coronelli[278], a portion of which is reproduced in [Fig. 44], we have on the West slope not only a well against which in the key to the plan is marked ‘Enneakrounos,’ but also close to it the ruins of a small theatre, which may well stand for the Odeion as seen by Pausanias. In another plan of the seventeenth century, usually known as the plan of the Capucins, both theatre and Enneakrounos are missing, and in their place stands the so-called ‘Theseion.’ On close examination it may be seen that on the Capucin plan, the theatre, the Enneakrounos, and some other buildings have been obliterated and other monuments drawn in over them. It may be taken therefore as certain[279] that, in the seventeenth century, remains of an ‘Enneakrounos,’ and of a theatre-like building near it, existed.

Fig. 44.

We have had to reconstruct the Nine-Spouts as best we might from the analogy of well-houses on vase-paintings, from the remains of the well-house at Megara, and from a few scattered, though significant stones. We have also inferred its importance from the vast system of water-works of which it was the manifest goal. But there is another witness to its past greatness. It is the place where all ways meet. The irregular square in front of the well-house Nine-Spouts and in part occupied by it was manifestly a great centre of the city life. The complex of ancient roads is best seen in [Fig. 46]. The great Panathenaic way passes along its Eastern side, but that is not all. The branch roads from the Areopagos converge thither. Most important of all for us, straight down from the Acropolis gate, skirting the Amyneion, there descends a narrow footway. By this we may be sure the King’s daughters descended to fetch water from Kallirrhoë.

A word must be said as to the nature and surroundings of the main ancient road, which topographically is of capital importance. Somewhere along its course must have lain the ancient Agora. Our first impression is, unexpectedly, of narrowness, just as it is when we stand on the other Sacred Way, at Delphi. On the Panathenaic way five persons can only just stand abreast; the chariots must have gone in single file. It is in fact a narrow Oriental street. It is bounded on either side by walls of good polygonal masonry and is hemmed in, as is seen on the map, by houses and precincts. Beneath the road is an elaborate system of drainage pipes with shafts by which they could be entered for cleaning purposes. There are of course many cross-roads, two to the left leading to the Areopagos, one to the Pnyx, another to Koile. The footway leading straight to the Acropolis has already been noted.

One of the best preserved portions of the road is that which runs along by the Western side of the precinct of Dionysos-in-the-Marshes. Here the polygonal walls on both sides are well preserved. Almost opposite the wine-press we come on buildings which, from inscriptions, can be dated as of the sixth and fifth centuries B.C. These consist of an open exedra, quadrangular in shape and of polygonal masonry. Inside this precinct is a small shrine with no columns, in front of it an altar of poros stone. Both material and technique point to the sixth century B.C. To whom the shrine is dedicated is not known. Thucydides could perhaps have told us. In the course of the century next following the shrine must have fallen into disuse. As the level of the road rose it would, once disused, speedily get covered up. That this was actually the case is clearly shown by the fact that a building of the fourth century B.C. was superimposed. It extended right back to the Pnyx rock. Two boundary stones of this later building are still[280] in situ in the wall bordering on the main road; on each is inscribed ‘Boundary of the Lesche’ (ὅρος λέσχης). Immediately next to the South comes a building of polygonal limestone masonry. Two inscriptions show that this building was mortgaged, so it must have been a private house. Beyond this there is nothing of special interest till we come to the great open place in which stood the fountain Nine-Spouts.

The careful engineering of the road, its elaborate drainage, the way it is close packed on either side with houses and sanctuaries leave us no doubt but that in it we have the one and, it appears, the only chariot-way from the agora to the Acropolis. The shrines that line this regular approach lie essentially and emphatically towards that part of the city.

So far we have considered the road as an approach, but it must always be remembered that historically we have to reverse our procedure. The city grows from the central hill, not towards it, and that outward growth is clear. It may be traced on the map in [Fig. 46]. The ancient agora lay in the hollow between the hills directly overlooked by the assembly place on the Pnyx; then as it outgrew these narrow limits it was forced bit by bit round the West shoulder of the Areopagus, and there turned Eastward by the hill Kolonos Agoraios, on which stands the ‘Theseion’; below that hill was the Stoa Basileios, which in the fifth century B.C. was assuredly part of the agora. The agora could not spread Westward; the hill prevented that; it was forced always Eastward, first in Hellenistic days as far as the Stoa of Attalos, then in Roman days to the Gate of the Roman Agora and the Tower of the Winds. Such is its long but simple story. If we follow the water-course of Peisistratos and its later Roman extension we shall not go wrong.

The houses that covered the square in front of Nine-Spouts, and into which fragments of the well-house were built, are all of Roman date. Clear them away, and we have, as has been seen, a great quadrangular space in front of the city well, a place to which all ways converge ([Fig. 46]). Surely here, if anywhere, is the ancient agora, close to the city gates.

It is remarkable that, visiting Athens half a century before the excavations began, an English scholar, Christopher Wordsworth[281], by sheer light of common sense, saw that here, and here only, could the ancient agora be, and here he marked it on his quaint, rudimentary map ([Fig. 45]). His words are, as contrasted with later confusions, memorable. ‘In order,’ he says, ‘to obtain a distinct notion of the natural characteristics of the spot to which we refer, let us consider it in the first place as abstracted from all artificial modifications; let us imagine ourselves as existing in the days of Kekrops, and looking upon the site of Athens. In a wide plain, which is enclosed by mountains except on the South, where it is bounded by the sea, rises a flat, oblong rock lying from East to West about fifty yards high, rather more than one hundred and sixty broad, and about three hundred in length. It is inaccessible on all sides but the West, on which it is approached by a steep slope. This is the future Acropolis or Citadel of Athens. We place ourselves upon this eminence and cast our eyes about us. Immediately on the West is a second hill, of irregular form, lower than that on which we stand and opposite to us. This is the Areopagus. Beneath it on the South-West is a valley neither deep nor narrow, open both at the North-West and South-East. Here was the Agora or public place of Athens. Above it to the South-West rises another hill, formed like the two others already mentioned of hard and rugged limestone, clothed here and there with a scanty covering of herbage. On this hill the popular assemblies of the future citizens of Athens will be held. It will be called the Pnyx. To the South of it is a fourth hill, of similar kind, known in after-ages as the Museum. Thus a group of four hills is presented to our view, which nearly enclose the space wherein the Athenian Agora existed, as the Forum of Rome lay between the hills of the Capitol and the Palatine.’

Fig. 45.

The secret of Dr Wordsworth’s insight lies in the words, ‘we place ourselves upon the eminence and cast our eyes about us.’ He stood on the actual hill, realized, as Thucydides did, that that was the beginning of things, noted the shape of the hill and its only possible approach, and saw that the developments of the city must lie that way, towards that part, as Thucydides would say. Half a century later Prof. Dörpfeld, coming with the trained eye of the engineer and architect, made, quite independently of Dr Wordsworth, the same observation. The valley enclosed by the Acropolis, Areopagus, Pnyx, and Mouseion, was then utterly barren of visible remains; other archaeologists had placed their agora where ancient remains were visible, North or South of the Acropolis; Prof. Dörpfeld, in defiance of orthodox tradition, placed it West, and there his excavations, as we have seen, brought to light the sanctuary of Dionysos-in-the-Marshes, the ‘Nine-Spouts,’ the Panathenaic Way, and the host of sanctuaries, houses, wine-presses, wells, and water-courses that encompassed the ancient agora.

Later we shall have to examine what it was that led other scholars and archaeologists astray; for the present we must return to Thucydides. He never mentions the agora, his thoughts never for a moment stray from his city before Theseus. He has shown its meagre extent and the immediate proximity of its most ancient sanctuaries, and to clinch his argument he returns to the citadel itself and its ancient name; he resumes the whole argument (see [p. 8]) in its last and most emphatic clause.

Because of the ancient settlement here, the citadel as well (as the present city) is still to this day called the city.

Thucydides is strictly correct both as regards official and literary usage. An examination of official inscriptions shows that down to the Peace of Antalcidas (387-6 B.C.) the Acropolis was officially known as polis[282]. The new form ‘in the Acropolis’ first appears in the year of the peace[283], and from then on is in regular use. In literature, both in prose and verse, polis is still uniformly used after a local preposition, e.g. towards the polis, in the polis; but when there is no local preposition the word acropolis is employed. Thus, in the Knights of Aristophanes[284], when the Sausage-Seller sees the Goddess herself coming from the polis with her owl perched on her, and there is no shadow of doubt that Athena is coming from the Acropolis; but Lysistrata[285] says, ‘to-day we shall seize the Acropolis,’ where there is no local preposition, though the sense would have been clear with polis. As Dr Wyse[286] has pointed out, it was easy for the word polis to go on being used for the Acropolis, because the Athenians had another word (ἄστυ), which they used in such phrases as ‘in town,’ ‘to town.’

We have learnt from Thucydides all he has to tell us, and in the light of recent excavations he seems to have spoken clearly enough. The limits of his ancient city have been confirmed by the discovery of the old Pelasgic fortifications. We have seen with our own eyes two of the ancient sanctuaries which lay towards his city, the Pythion and the sanctuary of Dionysos-in-the-Marshes; and from literary evidence inferred the two others, the Olympieion and the sanctuary of Ge. We have noted that, in the order in which Thucydides names them, they occur in succession from East to West; and, most convincing of all, near to the last-named sanctuary we have found Nine-Spouts, and not only Nine-Spouts, but the old Fair-Fount that was before it. Thus all seems clear and simple; Thucydides, Pausanias, and modern excavations tell the same harmonious tale.

Fig. 46.

From Antike Denkmäler II. 37.