Text entnommen aus: M. Cary, The Geographic Background of Greek and Roman History, Oxford 1959, S. V - VII, 1 - 6, 25 - 36 und 273 - 276.
1N this book I have endeavoured to make a fresh contribution to a subject whose importance is now generally recognized, the influence of geographic environment on human history, in a study of this influence on the world of ancient Greece and Rome. So far as I am aware, no attempt has yet been made to pursue this theme in a comprehensive way, except in Professor Ellen Semple's excellent book, The Geography of the Mediterranean Region, and the scope of her work does not extend beyond this region to those outlying countries which also figured in Greek and Roman history. A new survey of the entire subject in relation to the ancient Greeks and Romans may therefore be opportune.
My general indebtedness to previous writers is set forth in the bibliography at the end of the book. Of the authors therein mentioned, I am specially beholden to D. G. Hogarth, R. W. Lyde, Sir J. L. Myres, H. Nissen, A. Philippson, and Miss E. C. Semple. I also wish to put on record my obligation to T. Frank and his team of contributors to the Economic Survey of Rome, to M. Rostovtzeff, and to W. W. Tarn, from whose works I have derived much information and guidance.
My interest in anthropo-geography dates back to my undergraduate days, when I divagated from my authorized reading-list and devoured H. T. Buckle's History of Civilization. It has more recently been re-animated by A. J. Toynbee's challenging Study of History.
I gratefully acknowledge the assistance which I have received from my friend and former colleague, Miss M. S. Drower, M.B.E., who has read over my chapters on the Near East and has contributed valuable criticisms and suggestions.
I may state in advance my answer to two minor questions of geographical method which have obtruded themselves upon me. (i) A writer describing geographic features which remain discernible at the present day, but doing so in reference to past history, will almost inevitably find his pen slipping to and fro between the present and the past tenses. In my opinion such inconsistency is more than pardonable. Rigid uniformity in the use of a single tense may in such a case lead to inaccuracy or ambiguity of statement. Therefore the soundest rule is to select in each instance whichever tense is most appropriate to the particular context. (2) I have made no attempt to observe minute exactitude in the matter of measurements. Extreme accuracy may be essential in mathematical geography and in topographical studies over a narrow field, but in a general descriptive treatise the air of scientific precision thus conveyed is otiose and may be delusive. In particular, I have usually expressed the heights of mountain peaks and passes in round numbers. Approximate statements seem to me all the more preferable in this case, as there still exists much discrepancy in the figures given by modern and even by recent writers.
References to Strabo have been made according to the pages of the standard edition by Casaubon.
I. THE MEDITERRANEAN AREA
II. GREECE. GENERAL
III. GREECE. REGIONAL
IV. ITALY. GENERAL
V. ITALY. REGIONAL
VI. THE ASIATIC NEAR EAST
(i) ASIA MINOR
(3) ARMENIA AND THE CAUCASUS LANDS
VII. THE ASIATIC MIDDLE EAST
(2) THE EAST IRANIAN LANDS
VIII. NORTH AFRICA
(1) THE RED SEA BORDER
(3) CYRENAICA AND TRIPOLITANIA
(4) NORTH-WEST AFRICA
(5) WEST AFRICA
IX. WESTERN EUROPE
(1) THE SPANISH PENINSULA
X. CENTRAL EUROPE
(2) THE ALPINE LANDS
(3) THE MIDDLE DANUBE LANDS
XI. EASTERN EUROPE
(1) THE BALKAN LANDS
I. THE MEDITERRANEAN AREA.
§ 1 GEOGRAPHIC UNITY.
T HE ancient Greeks and Romans made history in all the three continents of the Old World, and penetrated far into the interior regions of Europe and of Asia. But the principal scene of their activity was on the rims of continents rather than at the centres of the land masses. Their civilization was essentially a product of the Mediterranean area, and the Mediterranean Sea was mare nostrum to them in a double sense-a political possession, but also a great formative influence.
The Mediterranean area has a homogeneous character which manifests itself in various common attributes of its constituent lands. The climate of its component countries, though subject of course to local variations, conforms on the whole to the same general type. 'Mediterranean climate', in consequence, has become a technical term which geographers apply to analogous conditions in divers other parts of the globe-southern California, central Chile, Cape Province, and South Australia. All these districts resemble the Mediterranean area in their latitude (c. 40' N. or S.), and in fronting westward upon a wide ocean; but only in the Mediterranean zone does its peculiar climate extend far inland from the ocean face. The climate of Moscow is radically different from that of London, but between Cadiz and Beirut the weather undergoes no fundamental change.
Again, in their geological structure the Mediterranean countries are mostly of one piece. Broadly speaking, the Mediterranean area as now constituted was the product of a vast upheaval in the Tertiary Age, which created a framework of heavily folded mountains round a deeply sunken trough.1 Originating in the same up-thrust, Mt. Lebanon resembles the Sierra Nevada in its tectonic forms, and the French Riviera is reproduced in the Crimean coast.
Furthermore, as a natural result of likeness in climate and structure, the Mediterranean lands are clad in a similar distinctive type of vegetation. The Mount of Olives near Jerusalem has many a counterpart in Sicily and the Balearic Isles, and the 'maquis' of Corsica is also spread widely over Greece.
1 Within this trough the Mediterranean Sea as a continuous piece of water was not formed until a comparatively recent period.
Lastly, the Mediterranean border regions, while sharing a common sea front, are marked off sharply from their hinterlands by an almost unbroken ring of mountains and deserts. The centrifugal pressure which created the Mediterranean valley built up its mountain faces into high walls, backed by more gently shelving ramps on their continental sides. By way of exit from the Mediterranean area, Nature has provided only a few commodious passages; if we do not avail ourselves of these, we must 'go over the top' or risk famine in a no-man's-land.
§ 2. Climate.
A preliminary question here arises: Was the Mediterranean climate, as we know it, already prevalent in Greek and Roman times, or is it the product of some later transformation? Climatic changes in prehistoric times, but at an age not vastly removed from Greek and Roman antiquity, have been proved on geological and archaeological evidence.' In the fifth and fourth millenniums B.C. North Africa and Mesopotamia were not isolated fringes of fertile land bordering on a wide desert tract; they formed part of a continuous belt of cultivable country enjoying a moderate but regular rainfall. Again, a variety of evidence, which will be referred to in subsequent chapters, places it beyond doubt that at the beginnings of Greek and Roman history several of the Mediterranean lands carried a denser cover of forest than at the present time, and this difference implies a heavier rainfall in preceding ages than the Mediterranean lands now receive. 2
Occasional data from Greek and Latin authors have also been pressed into service as testimony for a shift of climate since the time of their writing. The principal witness is Livy, who repeatedly records winter blizzards and prolonged summer rains incentral Italy.3 Proof of a change in weather has also been sought in the earlier incidence of harvest time in modern Greece and Italy, as compared with that of classical antiquity.4 Lastly, considerable areas in the interior of Asia which were cultivable at the beginning of the Christian era have since fallen prey to desicca-
1 V.G. Childe, The Most Ancient East, ch. 2.
2 C. E. P. Brooks, The Evolution of Climate; J. L. Myres, Who were the Greeks?, p.6.
3 H. Nissen, Italische Landeskunde, i, pp. 395 if. Cf. Appian, Bell. Civ. i. 87. I (an intensely cold winter in 83-2 n.c.).
4 Nissen, i, pp. 399-400; P. Groebe, in Drumann-Groebe, Geschichte Roms, pp. 743-4-
tion (pp. 165, 189 n.). This process, it is sometimes argued, did not stop short at the Mediterranean border.
If these reasonings hold good, the climate of the Mediterranean area in Greek and Roman times was wetter and cooler, and more akin to that of central Europe than to that of Mediterranean lands at the present day. But the evidence adduced is far from conclusive. It is not certain how far the desiccation of the Asian continent in the last two thousand years has been due to natural causes, and to what extent it is the result of human failure-the rack and ruin of raids and invasions, or the slow deterioration of an era of bad government; and in any case a change of climate in Asia, if such there has been, does not prove a shift of weather in Mediterranean longitudes. The reason why the date of harvest now falls earlier in certain Mediterranean lands may be found in the introduction of better methods of cultivation or the use of more quickly maturing seeds, rather than in a new cycle of milder winters and more torrid summers.' And the weather records culled from Livy and other ancient authors may actually be used to prove the stability of the Mediterranean climate: the reason why ancient writers reported certain hard winters or wet summers 1 was presumably because these were the reverse of normal. Finally, a comparison of the modern lines of isocarps (drawn through the most northerly points at which certain plants, e.g. figs and dates, will bear fruit) with those which can be deduced from ancient writers (especially from Theophrastus) shows a remarkable coincidence between them ; 2 and this is an important consideration, for the fig and date are highly sensitive to any change of climate. Hence according to one recent estimate the temperature of the Mediterranean area, and indeed of the whole northern temperate zone, has not varied by one degree Fahrenheit in the past thousand years.3 Local variations of climate in particular regions of this area need not of course be ruled out; indeed the extensive deforestation of certain Mediterranean lands since ancient times renders it probable that in such districts the climate is now more extreme and the summer drought more severe. But it may be assumed that the Mediterranean climate on the whole has remained unchanged since Greek and Roman times.
1 The northward extension of the Canadian wheat belt in recent years has been largely due to the establishment of new strains by Sir Rowland Biffen, which require ten days less to mature.
2 A. Jarde, Les Céréales dans l'antiquité grecque, p. 66.
3 S. F. Markham, Climate and the Energy of Nains, p. 23.
The determining factors of climate-latitude and the local distribution of land and water-combine in Mediterranean lands to produce a peculiar and distinctive cycle of weather. In the winter the Mediterranean area, which extends, roughly, from lat. 30 to 450, lies in the path of the westerly gales which are then prevalent over the Atlantic between lats. 300 and 6o°. Mt. Atlas and the Spanish plateau partly intercept these ocean winds, but the Straits of Gibraltar and the funnel between the Pyrenees. and the Cévennes provide them with regular 'cyclone tracks' into the Mediterranean basin. This basin, moreover, is a prolific source of home-made cyclones, by reason of the disparity of temperature between its heat-retaining waters and the chilled hinterlands, which causes an influx of air from the adjacent continents. The wind-eddies which are thus formed, like the swirl of emptied water round a bath-sink, originate most frequently in certain cyclone centres, such as the Gulf of Lions, the region of the Lipari Islands, the Adriatic, and the seas round Cyprus, but every part of the Mediterranean comes within their reach. Under the joint influence of these intrusive and indigenous disturbances the Mediterranean winter is a season of frequent high winds which sometimes freshen into sharp squalls and change direction rapidly.'1
The resultant churning of the air induces a heavy rainfall, rising to maxima in October and March, but copious in every winter month. The downpours are particularly heavy where steep mountain faces force the air-currents upward into a more chilly atmospheric stratum and so wring out of them an additional supply of 'orographic' rain. Of the 30 inches of. mean annual rainfall in the Mediterranean area, more than 20 are precipitated in the winter months; while the total annual supply is less than that of England, the winter quota is somewhat higher. But the rainclouds do not brood over the land, as in more northerly climes; they empty themselves in sharp cascades rather than in reluctant drizzles, and they do not leave a trail of mist or fog behind them. For all its boisterousness the Mediterranean winter is also a season öf blue skies and clear sunshine. Its occasional north winds, blowing from a frozen continent, bite shrewdly even in a sunny spell, and in wet weather they smother the uplands in snow. But in calm weather even the midwinter sun strikes with the force of an April sun in England; in the coastlands the rain
1 For some lively descriptions of sudden gusts from all quarters of heaven see Homer, Odyssey v. 291 if., and x. 46 if. (unloosening of the bag of Aeolus).
rarely turns into snow, and the snow never lies long.1 The average winter temperature of the Mediterranean lowland regions is considerably higher than that of the English plain. The mean January temperature of London (a relatively warm winter resort for its latitude) is 38°; at Rome it rises to 450, at Athens to 48°, at Palermo to 5o°, and at Malaga to 6o°. The Mediterranean lands, as a whole, are free from that continuous cold whose numbing effects have never been adequately counteracted until the advent of efficient modern methods of artificial heating. 2 But the chief advantage which the Mediterranean winter holds over those of more northerly regions lies in its longer spells of bright weather; though often wet and wild, it is never a period of settled gloom.
In summer the Mediterranean area passes under the sway of the north-easterly trade winds which then set in from the middle latitudes towards the sub-tropical zone. In Mediterranean longitudes this northerly current is intensified by a strong inward draught towards the Sahara, which from April to September sucks in the surrounding air like an immense natural furnace. Originating in the Eurasian land-mass, the predominant wind of the Mediterranean summer contains little moisture, and as it passes from cooler to hotter regions such water as it carries is evaporated more rapidly than local cloud-forming eddies can condense it. Apart from occasional trails of feathery cirrus, the Mediterranean summer sky is usually swept clear for a sun whose midday rays beat down almost vertically.3 During the dog-days the thermometer readings of the Mediterranean area rise almost to a tropical level. In London (the hot spot of England) the mean July temperature is 62°; at Rome it is 76°, and 8o° at Athens; in a few districts it goes up to 90°. In comparison with the ground temperature, it is true, the prevailing northerly wind is cool, and in the coastal districts the sea breezes that spring up in the afternoon dispel the midday's oven heat. In any case, the dryness of the atmosphere robs it of that enervating effect which saps human energy in regions of humid heat. The most adverse feature of the Mediterranean climate lies not in its high temperature so much as in its prolonged and regular drought. Apart from an
1 The rapidity of change between rain and sunshine in the Mediterranean winter explains the ascendancy of the sky-god Zeus or Jupiter in the Greek and Roman pantheon
2 Markham, op. cit., p. 41.
3 Hence to the Greeks and Romans the sun-god was the 'all-seeing' one. But the machine-like precision with which he plied his summer course did not suggest that he could regulate the weather like the 'cloud-1 acking' Zeus or Jupiter.
occasional thunderstorm, most of the Mediterranean lands are bereft of summer rain for a period that may extend over several months.
In its general climate the Mediterranean area is not a lotus land. Its winters are rougher and less balmy than seductive weather reports from its small and rare coastal strips of the Riviera type would indicate. The summer temperature rises above the level most conducive to high vitality; glare and dust are trying to weak constitutions; 1 and occasional visitations (mostly in spring time) by a torrid blast from the African desert 2 bring severe if passing discomfort. Yet it is not without reason that Mediterranean folk count themselves among Nature's favourites in the matter of climate. They enjoy an optimum annual mean of temperature 3 and of rainfall, 4 and their ratio of sunshine, which generally exceeds 2,000 hours in the year, 5 errs if at all on the side of over-measure. Some features of its winter climate, which on first impression might seem deterrent, its brisk winds and quick variations of temperature, reinvigorate body and mind after the summer's set fair weather.6 Above all, the clear, crisp, and luminous air of the Mediterranean region provides a stimulus such as few other parts of the world can offer.
It is therefore no mere accident that the Mediterranean area has been one of the world's most ontinuous centres of civilization. The Mediterranean lands have indeed passed through more than one Dark Age; but the cause of these periods of Decline and Fall should be sought in political and social failures rather than in the slings and arrows of untoward natural conditions.
1 The effect of prolonged exposure to unquenched ultra-violet rays from the clear Mediterranean sky may also be deleterious in the case of Nordic new-corners, whose pink skins do not readily absorb a short-wave radiation. (For the same reason some British settlers fail to become acclimatized in Kenya and South Africa.)
2 This wind is known as the 'Notia' in Greece, the 'Scirocco' in Italy (the plurnbeus Auster of Horace), and the 'Leveche' in Spain. It resembles the Föhn of Switzerland in being a fall-wind whose temperature is raised in the descent to a denser atmospheric stratum. In addition, it is laden with sand like a 'desert storm'. Fortunately it never lasts long.
3 590 in Italy, 630 in Greece. The English mean of 50 lies at the lower end of the optimum range.
4 30 inches: about the same as in England.
5 2,400 hours at Rome, 2,750 at Madrid (r,3o0-r,4o0 hours at London, C. 1,650 at Worthing).
6 It is partly because of its vigorous air circulation that the Mediterranean climate is classified by Ellsworth Huntington (Civilisation and Climate), together with that of other wind-swept regions (e. g. the borderölands of the central North Alantic and New Zealand), as one of the world's best.
§. 9 COMMUNICATIONS.
(I) The Mediterranean Sea. The sea which gives its name to the Mediterranean area is the largest ocean-inlet on the globe, extending over a length of 2,330 miles from Gibraltar to Port Said, i.e. as far as from Ireland to Newfoundland. Even before the opening of the Suez Canal it was the chief focus of sea communication between East and West; and ever since the invention of sea-going vessels it has been one of the world's busiest avenues of inter-coastal traffic, and one of its chief schools of navigation.4
4 In the later Middle Ages Italian seamen introduced the compass to the West and constructed the first scientific charts. In the Age of Discovery Italian navigators not only made some of the principal voyages but trained the explorers of other countries.
As a school of navigation the Mediterranean Sea can teach hard lessons. In winter it is a cyclone-track and a cyclone-factory (p. 4); at this season it is frequently swept by gales which freshen up at short notice and change direction rapidly. The capriciousness of the winds and the irregular conformation of the coasts give rise to baffling cross-seas and eddies; and if the billows of the Mediterranean never soar to the heights of the Atlantic rollers, neither do they move with the rollers' regular rhythm: height for height, they are the more dangerous because of their incalculable onset. As an inland sea in a mountain setting the Mediterranean is also exposed to sudden gusts that swoop in winter and early spring from the chilled coastal ranges on to the warmer sea-surface. A notorious example of these fall-winds is the Bora of the eastern Adriatic; but other such harpies may pounce on mariners from any high coast.
For these reasons 'the seas were closed' to the ancient Greeks and Italians in winter, in the sense that regular sailings were suspended, and short crossings only were attempted as occasion might offer. 1 The dangers that attended those who sailed out of season might be illustrated from Caesar's narrow escape on an attempted winter crossing of the Adriatic, 2 and from St. Paul's shipwreck on Malta in the late autumn. 3
But once the winter cyclones have blown themselves out, the Mediterranean waters pass over into the régime of a trade-wind with a constant direction from a northerly or north-easterly point, before which sailing-vessels can run safely and speedily. Occasionally the summer wind freshens up to gale force, and at all times the square-rigged vessels of the ancient world had difficulty in beating up against it. 4 But at night-time it renders down regularly; and at many points of the coast a brisk alternate play of land and sea breezes, arising from different rates in the rise and fail of temperature on land and on sea, will temporarily counteract the prevailing wind and allow those who are quick off the mark to steal a march upon it. 5
A feature of the Mediterranean Sea which is partly a hindrance
1 The 'close season' was reckoned from ro November to io March (Vegetius, Epitome rei militaris iv. 39). In the Middle Ages it ran from i November to i March.
2 Suetonius, Divus lulius 58, § z; Plutarch, Caesar 38, § 3. Acts of the Apostles, ch. 27.
4 The triangular 'lateen' sail, now universal in the Mediterranean, was introduced by the Saracens.
5 A night breeze helps sailers up the Dardanelles in summer-time.
but mainly a help to navigation is the feeble action of its tides. The sill at the Gibraltar Gates (p. 9) intercepts the ocean tides and breaks their force, so that they become wholly dissipated in the open water beyond C. Palos (at the south-east corner of Spain), and regain force only in a few outlying sea-funnels, such as the upper Adriatic. In general the tidal range of the Mediterranean does not exceed twenty inches.
The lack of brisk tidal movement in the Mediterranean entails this disadvantage, that its coasts do not receive a daily scouring, and that in consequence the earthy detritus of the rivers, which is concentrated by precipitation on meeting salt-water, is not washed away regularly, but accumulates in mud banks across the river mouths. A notorious example of such an obstruction in ancient times was the bar formed from the silt of the Tiber, which eventually threatened to put the city of Rome under blockade and so compelled the emperor Claudius to make a new cut for the discharge of the river; 1 and several cases of harbours being definitely ruined by silting will be mentioned in subsequent chapters. Consequently the chief emporia of the Mediterranean, instead of being located at river estuaries, like those of Atlantic Europe, are situated at a distance from them. This rule holds good of Barcelona, Marseille, Genoa, Naples, Trieste, Piraeus, and Alexandria, or of their ancient equivalents.2 In ancient times tidal inertia also diminished the value of the more deeply recessed harbours, in that sailing-vessels might have difficulty in clearing such stations in calm weather without an ebb tide to assist them. 3
On the other hand, the slightness of the tidal flow relieves Mediterranean seamen of the problems and dangers of a shifting water-line. They need not wait for flood-water before entering port, nor construct high sea-walls and docks with lock-gates. A constant water-level also facilitates beaching, and this was an important consideration in the days of wooden hulls, for in the warm Mediterranean waters these will soon become foul under Constant immersion. 4
1 Strabo, p. 231 ; Suetonius, Claudius 20, § 3.
2 By the time of Augustus lighthouses had been erected at the mouths of some rivers (e.g. the Rhône-Strabo, p. 184), but many of the river exits probably remained unlighted.
3 A. E. Zimmern, The Greek Commonwealth, pp. 30-I. It was perhaps on this account that Smyma was eclipsed by Ephesus in ancient times.
4 The Athenian ships in the Sicilian Expedition were largely rendered unfit for lack of beaching facilities (Thucydides, vii. 12. 3). Greek ship-builders preferred pine-wood, not fully seasoned, to oak for their ribs and planking, because its Sap kept the hulls comparatively clean. Oak was used for the keels only.
By its action in cutting off the waters of the Mediterranean from those of the Atlantic, the ledge of rock at the Straits of Gibraltar also prevents the formation of strong currents, such as the thermal difference between the Sea and the Ocean waters would otherwise set up. A shallow under-current from the Atlantic induces a weak surface movement along the African and Syrian coasts, and intermittent local streams may flow in narrow waters under the stress of strong winds. At two points, in the Straits of Messina and in the Euripus channel between Euboea and the Greek mainland, the play of currents is complicated by converging but unsynchronized tides. In the Euripus a rising tide and a winddriven current in one of the funnels, and a receding tide in the other, may combine to induce a veritable mill-race of seven or eight knots. In the Straits of Messina the divergent pulls of a mid-stream current, which sometimes rises to four knots, and of two coastal counter-streams causes local eddies of sufficient force to draw small vessels into their whirl (though not to engulf them). 1 But these two danger zones are marine curiosities. The general rule is that ships may drift at leisure on a Mediterranean current, but need not fear to be carried away.
A hydrographic map of the Mediterranean will show occasional straight flat stretches of sedimentary coast with a shallow approach. But its contours are for the most part the product of a landcollapse, whose lines of breakage have been subsequently warped and notched by the impact of storm-sped waves on outcrops of softer rock between the limestone buttresses. At many points accordingly the coast is as steeply scarped as the mountains of the inland, so as to present a sea face of formidable cliffs. But as a rule the water-line extends across a slope rather than along a precipice, thus providing conveniently shelving strands for running aground, and a sufficient depth of water up to the landing-point. Moreover, the innumerable bays and gulfs into which subsidence and erosion have scooped out the coast provide an endless variety of natural harbours. At some points an intrusive sea has flooded a river valley or an inland subsidence-trough, so as to form deep land-locked inlets such as the Gulfs of Actium and Corinth. At others it has not merely lapped round a mountain spur but has undercut it, thus carving out a T-shaped or 'anvil' projection and two back-to-back harbours, either of which will provide in turn a sure shelter from the prevailing wind of the
1 The Charybdis of Homer was identified with an eddy near Messina which Strabo describes with a little pardonable exaggeration (p. 268).
moment. Occasionally the mordant sea-water has ended by sawing off the protruding neck and has converted the anvil-table into an offshore island, and the result is a single harbour with a natural breakwater across its face. The Greeks and Romans often improved upon the natural harbourage by means of moles and jetties but their purpose in this was mainly to provide space for wharves or to secure the port entrances against enemy raids. 1
But the chief allurement of the Mediterranean Sea is to be found in its good conditions of visibility and the wealth of landmarks which it offers. In the winter season low cloud and driving rain may blur the field of sight, but fogs that blot it out altogether are rare. In summer the guiding lights of sun and stars are seldom dimmed, and in the dry air distant objects stand out conspicuously. Mainland chains or island peaks will show up at ranges extending to ioo miles, thus enabling ships to hold an almost straight course over long routes without losing sight of land. Before the coast of France disappears from his view a seaman will descry the tail shoulder of Mt. Cinto in north-west Corsica; with Sardinia still in sight, he will make out the islands that fringe the northern coast of Tunisia. Proceeding from Gibraltar and following the African coast to C. Bon at the apex of Tunisia, he will then catch sight of Mt. San Giuliano (ancient Eryx) in western Sicily; having cleared the heel of Italy, he will observe the bold outline of the Epirote coast; from the southern tip of Greece he may pick his way to Asia Minor through the Greek archipelago, or he may take a more southerly course with the Cretan peak of Mt. Ida in constant view. Navigation in the Mediterranean is therefore not far different from the journey of a landsman along a well-defined route, and Homer spoke aptly of the sea's 'liquid lanes'. Lacking the compass and the sextant, ancient wayfarers could nevertheless traverse the Mediterranean in all directions without being 'at sea'.
(II) The border countries. Intercourse by land within the Mediterranean area is relatively difficult. In many regions it is beset and circumscribed by close-set barriers of mountains, most of which, having been erected by a rolling-up movement of the earth, present a steep escarpment on at least one of their faces. Faultings and subsidences within the mountain systems have indented them with passes of relatively low altitude, and rivers have carved erosion valleys across them. But these valleys are
1 On ancient harbour installations see K. Lehmann-Hartleben, 'Die antiken Hafenanlagen' (Kilo, Beiheft XIV, 1923).
mostly narrow and tortuous, and any pass rising to 3,000-4,000 feet is liable to be snowbound for part of the year. The Mediterranean rivers, moreover, are more of a hindrance than a help to travel. Most of them have a torrentially rapid fail in their upper reaches, and in their lower courses they are ill suited to navigation on account of their inconstant water-level.1 Few of them attain great width, yet even the smaller streams may lack convenient crossing-points. Where the rivers pass over a limestone slab, they usually cut their bed into a deep trough, 2 and in the coastal plains they are apt to meander through marshy deltas. On soft and level ground Roman engineers were active in amending natural waterways by cutting artificial channels, but they could not force a passage through a ledge of rock, save at prohibitive cost.
In the ancient Mediterranean world, therefore, rapid communication by land was possible only where all-weather roads and bridges had been provided. These were eventually constructed in every Mediterranean country by the Romans, but even the Romans shirked the task of 'shaving' the gradients in mountain districts, so that their roads were at some points inconveniently steep for vehicular traffic. In an age when horsecollars were unknown and horseshoes had not yet passed into common use, 3 the steepness of the ramps rendered the longdistance haulage of heavy merchandise almost impossible. The transport of goods by land was mostly effected by pack-animals; in the movement of ponderous or bulky commodities the sea routes inevitably played the major part. On these grounds we may doubt whether in ancient times a 'law of isthmuses' was in general operation, according to which trading vessels would not make a continuous voyage round a peninsula or far-flung promontory, but would discharge their cargo at the promontory neck, from which point the merchandise would be conveyed overland for re-embarkation in another hold at the opposite side of the isthmus. 4 Undeniable instances of such breaking of bulk can
1 For the same reason water-mills were uncommon in the Greek and Roman world, though their principle was understood. Blümner, op. cit. i, pp. 45-9; M. Rostovtzeif, Social and Economic History of the Hellenistic World, iii, p. 1645 (a mill at Venafrum); A. W. Parsons, Hesperia, v, PP- 70 if. (another at late-Roman Athens).
2 These deep-cut river beds were a serious obstacle to the allied troops in the Battle of Italy as they advanced along the west coast south of Naples, and along the Adriatic coast north of C. Garganus.
3 G. Méautis, Revue des etudes anciennes, 1934,p. 88.
4 For this law see V. Bérard, Les Phéniciens et l'Odyssée, i, pp. 68 if. (criticized by A. E. Zimmern, The Greek Commonwealth, pp. 317-18 n.).
be adduced, but only in special cases where the sea route involved a large detour (p. 82) or was infested by pirates or privateers, or where the intervening isthmus was low-lying all the way.
On the other hand, the Mediterranean lands have, in historical times at least, been generally clear of dense forest, such as rendered large tracts of the European continent almost impassable until late in the Middle Ages. Alike in their plantations and in their natural timber standings, the trees are sufficiently spaced out to give a relatively easy passage. Even so, communications within the Mediterranean lands could never be easy without the helping hand of man.
§10. SOCIETY AND POLITICS.
The pervasive influence of Mediterranean geography on the Greeks and Romans is illustrated in numerous features of their social and political life.
The Mediterranean climate makes for an open-air existence. Where summer heat is tempered by the play of breezes, and winter chill by a clear sun, life out of doors is pleasant over the greater part of the year. This open-air habit found expression in the plans of Greek and Roman houses, for those whose owners could afford the necessary ground space were usually, laid out round one or more courts 1 - a practice which survives in the less 'Europeanized' regions of the Mediterranean. In the cities shady recesses from the summer sun and the winter downpours are almost a necessity. The streets of Mediterranean towns have therefore from time immemorial been made as narrow as is consistent with traffic requirements, the side-alleys on hill sites often being mere staircases of the 'Clovelly' type; and the Greek and Roman architects who planned 'show' streets in the major cities made a point of lining them with porticoes and colonnades. In the streets and open squares of the towns rich and poor alike met their friends, spent their leisure, and transacted much of their business ; 2 and they sought their entertainment in unroofed theatres and arenas (the concert halls alone being covered over). For political affairs it was not only the massed Popular Assemblies that met in the open. The Areopagus at Athens held session in a porch (the Stoa Bosilike), and the Dicasteries in open courts (in
1 The 'court' style of mansion was already prevalent in prehistoric Crete.
2 At Rome the 'city' men forgathered, not in a Bourse or Royal Exchange, but a lanum, i .e., by an arch in an open place. Nowadays the open-air cafés of Mediterranean towns are recognized resorts for business talks.
the literal sense); at Rome magistrates set up their tribunals in the Forum or one of its adjacent porticoes. 1
With these open-air habits went a readiness of social contacts and a general use of intercourse that made every Greek or Roman town into an informal club. Though aristocratic personages might be accompanied on their outings by an escort of retainers, to save themselves from too intimate a contact with the common folk, they did not withdraw themselves from its gaze, but courted it in their public appearances. A similar tradition of affability, or at least of accessibility, was observed by political leaders; even kings and tyrants obeyed the law of civilitas, and it was not until the third century of the Christian era that Roman emperors accepted the Oriental custom of mysterious seclusion in the recesses of a palace.
The density of settlement in Mediterranean lands is regulated to a large degree by one of their determinant features, the general scarcity of natural water-supplies in summer. The Anglo-Saxon 'tun', with its population of one hundred persons or less, could not he reproduced here, except in the pockets of softer rock where the winter rains percolated more evenly through the subsoil, so as to provide an easily tapped store of water. On the more usual limestone formations the irregular distribution of water necessitated a closer aggregation near the infrequent water-pointsat a river-side or, most commonly, in the neighbourhood of a spring. Here substantial and closely built villages, with populations of oo persons or over, would be formed, and under favourable conditions the villages would grow into towns with several thousand inhabitants. Though political and economic factors must also be invoked to explain the genesis and the siting of ancient Mediterranean cities (pp. 48 if., i i o), the primary factor of water-supply usually determined the growth of the villages which were the nuclei of the towns. 2
The habits of communal solidarity which life in compact settlements everywhere engenders were reinforced in Mediterranean lands by the need of joint action in the matter of water regulation-the provision of drainage canals for the winter floods and, more especially, the equitable rationing of supplies for
1 Eventually, however, the Roman officials withdrew more and more into the basilicae or covered halls which sprang up especially since the time of Caesar.
2 J. L. Myres, Mediterranean Culture, pp. 13-14. It has been observed by A. Philippson (Der Peloponnes, pp. 585-6) that in modern Peloponnesus the population of villages varies from 150 to 6oo, according as they stand on the softer or the harder rock-formations.
irrigation. To this extent life on the Mediterranean country-side provided a schooling for the more intensive co-operation of the nembers of a city-state.
But if natural conditons in Mediterranean lands made for communal co-operation, they did not favour agrarian communism. This form of organization is most compatible with a pastoral economy. But under Mediterranean conditions the prevailing type of land-work consisted of tillage and orchard cultivation. In these occupations, and especially in the tending of gardens, success must largely depend on the individual's devotion to his particular plot, and (in the absence of any highly organized machinery of state) this could be best assured by the institution of personal property.
Yet the natural economy of the Mediterranean lands did not play into the hands of capitalist exploitation. The generally broken and tumbled surface of the country did not favour the consolidation of plots into large units like the Bonanza farms of the American Middle West; consequently the savings in production costs that result from operations on a big scale could not be realized, save in abnormal districts. Least of all could standardization and the processes of mass-production be introduced into the numerous corners and pockets where garden cultivation was practised, for here individual skill and attention prevailed over organization. 1 The absorption of small property into the ownership of wealthy landlords, which was a recurrent feature of Greek and (more especially) of Roman history, was due in the main to political and social rather than to strictly economic causes. On the one hand, the exigencies of military service, which called the peasant away from his plot and might leave it untended för long periods, continual devastation by invaders (as in the later days of the Roman empire), and the lure of 'bread and circuses' in the cities, were so many inducements for the small owner to sell out; on the other hand, the social prestige and (in some states) the political privileges that accrued to landed property, and the general lack of other safe objects of investment, turned every moneyed person into an eager buyer. Even so, concentration of Ownership did not necessarily entail the merging of small units of cultivation into latifundia. Having regard to the conditions
1 This applied particularly to labour in the vineyards. The olive orchards required less intensive cultivation but more expensive machinery in the form of oilPresses. For this reason, and because of the olive-tree's slow rate of growth, which Withheld the hope of a speedy return on labour expended, capital played a larger part in its exploitation.
imposed by Nature, some wealthy landlords made their purchases in small and scattered parcels, and those who consolidated their holdings into compact blocks not infrequently let them out in small lots, instead of applying organized methods of capitalist exploitation. 1 The only branch of land work which gave any decisive advantage to operations on a large scale was 'transhumance' pasturing, for under proper organization hardly any more labour was required for the tending and droving of large herds than of small ones. Whatever the method of ownership, la petite culture remained the normal practice in ancient landwork.
Another natural feature of the Mediterranean lands which left its mark on Greek and Roman history was the more or less abrupt transition from the cultivated lowlands to the maquis and the summer pasture of the upper zones, and the consequent segregation of the husbandman from the herdsman. From this dissociation natural enmities sprang up in the ancient world, such as still obtain between Fellahin and Bedouin in the Near East. For the mobile and armed herdsman it was a constant temptation to supplement his scant living by raiding the plains, and to rid himself of this recurrent nuisance the lowlander had to organize his resources of greater man-power. The feud between Mountain and Plain runs like a red thread through early Roman history, culminating in the hard-fought and critical Samnite wars; and mindful of their early tradition as guardians of the Italian lowlands, the Romans subsequently pacified large tracts of provincial land by the drastic method of deductio in plana of unruly upland populations. 2 Yet the practice of 'transhumance', entailing a voluntary deductio in plana for herdsmen in the winter season, and a bargaining about the regular use of grazing-spaces in the lowlands, tended to overlay ancient feuds with emergent understandings, and the drovers' codes which Roman lawyers drew up for Italian pasturages confirmed the Roman soldiers' work of pacification. The need for highlanders to import salt for themselves and their herds also laid the foundations of friendly traffic between Mountain and Plain, and the Via Salaria, which led from the salt-pans of Ostia through Rome to the Sabine mountains, was as much a path of peace as an invasion track.
1 On this point see especially G. Salvioli, Le Capitalisme dans le monde antique, ch. 4.
2 C. Jullian, Histoire de la Gaule, iv, pp. 74-5; C. H. V. Sutherland, The Romans in Spain, p. 135; Rostovtzeff, Roman Empire, p. 222 (Dalmatia).
A more potent inducement to peace, however, was the intensive cultivation of olive and vine in Mediterranean lands. Tillers of the crop-land had periods of rest in their calendar, during which they could afford to take a holiday in the form of a campaign against some neighbouring state: even though the battle should go against them and their crops should suffer ravage, the damage was no greater than a year's hard work to follow could repair. But the labour in the vineyards was continuous, and the destruction of an olive-tree might require half a lifetime to make good. It was, therefore, no mere fancy which caused the Greeks and Romans to offer an olive branch as a symbol of peace.' 1
But the most important natural factor in ancient Mediterranean history was the Mediterranean Sea itself. This great water has its dangerous moods, and throughout the ages it has scared some of its border peoples into remaining land-lubbers. But those who have adventured it and mastered its comparatively simple rules of navigation have earned their due reward in economic prosperity or political lordship. The fundamental similarity of the Mediterranean lands has not precluded disparities of detail between them sufficient to render them economically complementary to each other, and so to draw them together into habits of economic co-operation. Some regions were natural granaries, others were ill suited to corn-production but well adapted to produce wine or oil; and the raw materials of industry were variously distributed over the whole area. Thus the inter-Mediterranean traffic in foodstuffs and textiles, in ceramics and metal-ware, grew to be of basic importance for the material civilization of Greece and Rome; but the mass-movement of staple commodities would have been impossible without the commodious highway of the Mediterranean Sea.
This sea was equally indispensable as a connecting link between the members of any comprehensive political union. The importance of naval power in the Mediterranean as an instrument of empire may be illustrated from each successive thalassocracy of ancient times, from the Minoan, the Athenian, the Ptolemaic, and the Roman lordship of the seas. 2 The part played by road
1 Fortunately the wood of the olive is exceedingly tough. The demolition Parties of an invading army, unless inured to lumber-work, might blister their hands severely before their axes had cut deep.
2 The importance of naval power as an instrument of empire was clearly understood by the Athenian writer known as the 'Old Oligarch' (Constitution of Athens, ch. ii, § ,-8). For a more general survey see J. H. Rose, The Mediterranean in the Ancient World.
communications in holding the Roman empire together should not be overlooked, but it was, above all, the Mediterranean Sea that enabled it to coalesce into an organic unity, for it alone could render possible that frequency of intercourse among its constituent parts which made them 'members of one another'. 1 Seconded by wise Roman statesmanship, the natural uniting force of the Mediterranean Sea allowed the scattered populations of its borderland to achieve a cultural bloc which has ever since been a major factor in world civilization.
1 For this reason too Polybios could write that under Roman rule Mediterranian history had become 'organic' - [griech. 'somatieidees']
X. CENTRAL EUROPE.
§ 1. GERMANY.
So far as concerns the history of Rome, ancient Germany might be roughly defined as the territory between the Rhine, the upper and middle Danube, and the Vistula. This country is, like Gaul and Britain, a land of good natural communications. Its northern half forms part of the vast plain that stretches almost unbrokenly from the North to the Yellow Sea, and the hill country of the southern portion is discontinuous and seamed with easy passage-ways. Of the German rivers, the Vistula and Oder, besides being ice-bound in midwinter, have the disadvantage of ending their course in the Baltic backwater, but the Elbe and Weser share with the Rhine (p. 251) a fairly constant flow of water and an outlet on a sea which is potentially one of the principal lines of European communications. The climate is of the Atlantic type, with a moderate range of temperature and equable rainfall. 1 Germany contains many patches of dry and open land, where crops could be grown with simple equipment; its mineral resources are large and varied, 2 and in its deposits of amber on the coasts of Samland and Jutland it formerly possessed an almost unique asset.
On the other hand, to say nothing of Roman fastidiousness in the matter of warm sun and blue skies, ancient Germany, even more than Britain, presented an unkempt appearance. 3 Its éultivable lands were surrounded with large areas of dense forest and swamp. In the German midlands the present Thuringian, FranConian, and Sudetic forests formerly coalesced into a 'Hercynian wood', whose vast expanse had captivated the imagination of Greek geographers 4 before it awed the soldiers and traders of Rome. In northern Germany extensive tracts of almost deadlevel country have too slow a run-off of their spill-water, and a hard pan of tightly rammed clay (a remnant of former Ice Ages)
1 The boundary between the Atlantic and the continental weather systems lies at the Vistula. The climate of Germany is rated among the best in the world by E. Huntington (Civilisation and Climate, ch. x).
2 The renaissance of mining in the fifteenth and sixteenth centuries began in Germany.
3 Tacitus, Germania, , § i: 'silvis horrida aut paludibus foeda'.
4 Caesar, Bellum Gallicum vi. z: a sixty days' journey across it! (Probably from Posidonius.)
274 CENTRAL EUROPE
under a thin top layer of sandstone retards subsoil drainage Until the advent of modern science and machinery Germany contained large reserves of fenland and peat-bogs.
The minerals of Germany mostly lie at a deep level; in ancient times their very presence for the most part remained unsuspected Of the seas adjoining Germany the Baltic is safe, but it was too deeply recessed from the main roads of ancient commerce. The North Sea is less boisterous than the Channel and Bay of Biscay, but it is not immune from summer gales, and at all seasons it j liable to be hemmed in with thick cloud or blurred with mist. Its Frisian and Jutish coasts are ill defined and partly submerged at high tide; and such conditions of low visibility and vanishing landmarks were peculiarly baffling to seamen trained in Mediterranean waters. 1 Consequently it remained a 'German Ocean'; and it is doubtful whether the Baltic ever carried a Greek or Roman keel.
Contrary to their early delusions about Britain (p. 263), the Romans never imagined Germany as an Eldorado, and their interest in it was almost exclusively one of frontier defence. From the economic point of view the Rhine appeared the limit of profitable expansion, and with its strong and equable flow of water it formed a good barrier against invasion; but once the Romans had drawn the remainder of their European boundary along the Danube, a re-entrant angle was formed between the two rivers which considerably added to the mileage requiring to be guarded.
In the days of Augustus accordingly the Roman armies reconnoitred in Germany for a shorter if more distant boundary, and at the outset of their quest they happened upon the best alternative line. Though the emperor could probably not have drawn a remotely correct map of Germany-not even the geographer Ptolemy could do this 150 years later-his choice of the Elbe as an alternative river barrier, and of the Czechoslovak plateau to close the gap between it and the Danube, was happily inspired, for his new German boundary, together with its prolongation along the Danube, made a clean diagonal cut across the European continent. In their advance across western Germany to establish ; their new front Augustus' generals made good use of the natural avenues of penetration. From their southern base at Moguntia-
1 Witness the disaster to Germanicus' transport fleet in A.D. 16 (Tacitus, Anus, ii. 23-4). The subsidence of the coast of north-west Germany, which was probably part cause of the Saxon migrations (Collingwood and Myres, op. cit., p. 342, may also account for the Cimbric invasions of the late second century B.C.
cum (Mainz) they followed the valleys of the Main and the Saale, and from their northern camp at Vetera (near the German-Dutch frontier) they marched up the valley of the Lippe. Their com-missariat fleets, based on Boulogne, made a circuit of the North Sea to the estuary of the Ems, Weser, or Elbe, and rowed up these rivers to the appointed meeting-place with the armies. But the last grand operation, a converging advance upon Bohemia by one force ascending the Main valley and another striking northward from the middle Danube (p. 286), had to be called off for reasons extraneous to the plan of campaign (a rebellion in the Danube lands); and a chance defeat of another Roman army which lost its way in the forests of Westphalia led to the complete abandonment of Augustus' scheme.
Augustus' plans were revived and carried into effect in an attenuated form by Vespasian and Domitian, who pinched out the salient between the middle Rhine and the upper Danube by occupying the Black Forest area and incorporated a zone of territory on the right bank of the Rhine from a point below Bonn, so as to take in the high ground of Mt. Taunus. The new frontier, which was marked out so as to follow the hill-crests beyond the valleys of the Rhine, Main, and Neckar, not only shortened the Roman line of defence, but by including the wooded areas of Mt. Taunus and the Black Forest it deprived German raiders of convenient mustering-grounds, and incidentally it annexed some of Germany's best agricultural land in Hesse and the Wetterau. One result of these frontier adjustments was that Argentorate (Strasbourg) became the starting-point of a military road across the Black Forest by way of the Kinzig valley. Thus Roman Strasbourg, though not yet a river port, acquired importance as a sally-port into Germany.
The Romans were too much preoccupied with the safeguarding of their German frontier to foster commerce with the natives; indeed they would not admit German traders into Roman territory save at selected points. 1 But they could not prevent Italian or Gaulish merchants from adventuring themselves in German lands. Finds of Roman or Gallic objects are most frequent in the Ruhr district, in Thuringia, and in north-western Germany. 2 Their distribution suggests that the traders partly used the old invasion routes by way of the Main and the Lippe, and partly worked their way up-river from the North Sea, in imitation of
1 Tacitus, Germania, 41, § 1.
2 Mrs. O. Brogan, Journ. Rom. Stud., 1936, pp. 195-222 and map.
Augustus' transport fleets.1 Numerous finds of Roman coins on German territory indicate that Roman exports (wine and metal ware) did not suffice to pay for the imports. Since the medieval trade in timber, hemp, and fur from the Baltic regions had not yet come into existence, and the amber traffic since the reign of Nero was directed to the middle Danube (p. 286),2 it remains a problem how the balance of imports over exports turned against the Roman empire and had to be settled in specie, unless the Roman slavetraders brought home large hauls of captives from the German inter-tribal wars.
1 A regular North Sea Traffic id assumed by A. Francke (Pauly-Wissowa, s. v. 'Nordsee') and A. W. Byvanck, Nederland in den romeinschen tijd, pp- 547 -8.