loosely divisible into chronological periods, though their dating is by no means uniform (e.g. the Neolithic, Bronze and Iron Ages begin at different times in different areas).

We can make some external chronological checks from the comparatively precise records kept by Hittite, Babylonian and Egyptian priests. These, being mathematically based, provide us with dates of an accuracy unobtainable in Greece: king-lists, for instance, assignable in terms of our own calendar. The Egyptian Pharaoh Amenhotep III reigned 1412-1376 BC; seals and scarabs of his consort Queen Tiy have been found in Crete at the Late Minoan (LM) II level, and also at Mycenae among material of Late Helladic (LH) III. As the former ends, and the latter begins, about 1400, we can peg this date with other corroborative evidence, which rules out the possibility of these artefacts being antiques as the median transitional period.

Linguistic studies, too, if properly interpreted, can tell us a good deal about tribal movements and indigenous population layers. Greek belongs to the widespread Indo-European family of languages, and was first brought to the Balkans (though probably not in any form which we would readily recognize) between 2100 and 1900 BC (see pp. 37-8) by successive waves of invaders from beyond the Danube and the Black Sea area. The loan-words these newcomers borrowed from the local population - vine, olive, fig, wheat and the sea, among others - suggest very clearly that they were pastoral high landers from the inland steppes. They had no term for a bath tub: that too is significant. Such clues, like the later distribution of local Greek dialects, can provide us with a useful pattern; but just what such a pattern means in specifically historical or political terms is less certain. Conquest? Racial fusion? Commercial penetration? Without a written record how can we be confident of distinguishing between them? The effective boundaries between one dialect and another must always have been shifting: to treat a dialect area as a clearly defined political or geographical unit can lead to serious misconceptions.

The best fixed criteria for significant change in the population of any given area are still new burial customs, changes in house style, tools, pottery and weapons, or the appearance (detectable by craniometry) of a new physical type. Scientific aids such as the Carbon-14 test, which measures the dissipation-rate of radio-activity, can help in dating early artefacts, but the possible margin of error in conducting the test itself still remains uncomfortably wide. Climate and geography discussed more fully below - can also provide a check on movements, since primitive peoples disliked moving from one climatic zone to another, on the grounds that this disrupted their traditional economy. Thus migrations (see pp. 46-7) tended to be horizontal (i.e. E-W or

W-E) rather than vertical (N-S or S-N), and to keep roughly the same height above sea-level, except in very unusual circumstances. Mycenaeans, Phoenicians and Greeks alike tended to colonize only the coastal fringes, and to leave the hinterland alone, especially when as in Anatolia - it was mountainous plateau.

Lastly we have the retrospective tradition, both oral and written, of myth and legend. This is enormously valuable, but at the same time has to be handled with the sharpest and most cautious critical acumen. The epic poets and mythographers embroidered and slanted much of their material, misunderstood more, and got a great deal plain wrong; but as often as not there turns out to be a solid substratum of fact underlying their fancies. The Trojan and Theban cycles, for example, both seem to contain a solid core of genealogical and narrative truth. Archaeologists are busy digging up Thebes' seven gates, while numerous Near Eastern finds in that neighbourhood (including cylinder seals) suggest that the story of Cadmus and the Phoenician alphabet may likewise rest on relatively firm historical tradition. We should not underestimate the tenacity of oral memory, which tends to

6 Opposite: marble Cycladic doll, c. 2000 BC. These delight

ful

figurines - either standing, as here, or squatting (the latter known as 'fiddle idols' be cause of their resemblance to a violin) - continued to be produced through most of the third millennium BC.

7 Terracotta model represent

ing a scene of sacrifice to the dead, from Kamilari, near Phaestos, Crete. Such a find, coupled with the attention be stowed on Cretan burials, sug gests a belief in the after-life. The couples with tables of

offerings in front of them

may be the deceased them selves, or perhaps chthonian (underworld) deities.

be lost with the advent of writing. The Trojan Expedition in Homer belongs very largely to a half-misunderstood Bronze Age, even though the similes are drawn from Homer's own eighth-century Ionia. How much of Agamemnon's world, though, would we guess at from the technically much harder' evidence of the Linear B tablets? In the beginning was the Word: without written records, our account of any civilization, however intuitive and sympathetic, must inevitably remain what T. S. Eliot called 'hints and guesses, hints followed by guesses'. In the last resort perhaps that is true, to some extent, of all historiography.

Climate and geography

The author of that oddly named treatise On Airs, Waters and Places (perhaps Hippocrates himself, certainly a Hippocratic physician writing in the late fifth century BC) reveals, long before Montesquieu, a very clear awareness of environmental influence. "The people of Europe', he writes (meaning, of course, Greece), 'differ from one another both in stature and in shape, because of... severe heat waves, severe winters, copious rains and then long droughts, and winds, causing many changes of various kinds' [S23]. These factors, he asserts, lead to 'wildness, unsociability, and spirit'. To call the Mediterranean climate of Greece 'temperate', as people sometimes do, is highly misleading. It ranges between extremes of heat and cold, with violent winds (most noticeable on the islands) in winter and summer alike. During July and August the whole Aegean is scoured by a dominant north-east gale known in antiquity as the 'Etesian Winds', and today as the meltemi. Autumn and spring rains tend to be brief but torrential. The effect of these seasonal deluges on rivers, combined with the mountainous, deforested nature of the landscape, is to produce torrents which are either dry-bouldered or in spate, and either way completely non-navigable. Snow is common in the moun tains, and can lie on peaks like Parnassus or Olympus as late as June.

The combination of summer drought, winter rains and long spells of winter sunshine is ideal for long-germinating plants such as the fig, vine and olive, which between them sum up the basic core of Mediterranean farming. For a country so dependent on wheat and barley as a staple cereal, Greece is surprisingly ill-adapted to largescale arable farming. Its soil, except in the plains, is thin, and less than one-fifth of the whole Greek land-mass is, or was ever, cultivable. This had a profound effect on the country's political and economic history, since from a comparatively early date much of its wheat (and timber) had to be imported from unreliable sources abroad. On the other hand, because of its dry, clear air, Greece suffered remarkably few epidemics in comparison with other European countries, and those

which did occur (e.g. the great plague during the Peloponnesian War: see pp. 136-7) tended to be brought about by extraneous causes. Malaria was largely confined to low, marshy areas like central Boeotia, where the anopheles mosquito could breed. Since Greece is traversed by high mountain ranges, movement in winter tended to be minimal: storms at sea and snow-bound passes kept all but the most persistent travellers at home. War, therefore, tended to be a seasonal affair. There were exceptions to this rule; but between November and February, for the most part, winter quarters prevailed.

For so small an area Greece offers considerable climatic variations. The northern part of the country, including Macedonia and Thrace, is not Mediterranean at all, but lies on the main Continental shelf. The west coast is wetter, and warmer in winter; optimum Mediter ranean conditions are found in Attica, and on the Aegean islands. One sharp climatic distinction, not dependent on latitude, is that between mountain regions and littoral. The result, during antiquity, was a social gap: coastal settlements tended to be more developed, while the hinterland remained backward. Differences were - and still are - accentuated by the tree-belt level. Above three thousand feet the Greek uplands provide summer pasturage (in those areas where deforestation has not left a wilderness of barren limestone). One or two states, such as Sparta, possessed both valley and upland territory

8 Aerial view of Mount Olympus, the highest moun tain in Greece (9,573 ft) and traditionally the home of the Olympian deities. This gigan tic massif, lying on the borders of Thessaly and Macedonia, acted as a bulwark against invaders from the north, though

it could, with effort, be cir cumvented.

9, 10, 11 Different in so many respects, the cultures of Crete and Helladic Greece reveal a

common dependence on sheep and (to a lesser degree in Crete) cattle. Below: a unique offer ing from Middle Minoan East

Crete (Palaikastro): a votive

bowl containing over two hundred model sheep, and their shepherd. Opposite: the general setting looking east (above) and the Acropolis. (below) of Gla in Boeotia - a cattle-baron's stronghold with ample pasturage round Lake Copais.

suitable for grazing; but most were restricted to either one or the other Attica and Corinth, for instance, are more or less all lowland, while Arcadia forms a great mountain plateau - and this made a great difference to their economy and social outlook. The only extensive low-level pasturage was in Boeotia and Thessaly: no wonder that from prehistoric times (as the great cattle-pound of Gla, by Lake Copais, eloquently testifies) these areas formed the home of rich and aristocratic cattle-barons, who could breed horses and afford a cavalry arm in battle. Chariots, too, were restricted to the plains, and not merely on grounds of expense: Greek mountain trails, in antiquity as today, could be negotiated by nothing more complex than pedestrians and pack-animals. Wherever possible, the Greek preferred to ship his goods from harbour to harbour along the coast.

Another marked effect of Greek climate is on social life. Weather can be bad: talking of February Hesiod says, with good reason: 'Beware of the month Lenaion, bad days, that would take the skin off an ox.' But trouble is, for the most part, limited to a three-day blow,