ORIENTAL RESEARCH INSTITUTE LIBRARY ACCESSION Ho. $ s- ^-.cA CAli No. 2»>\ » X1^^^^^3)^ SRI VENKATISWARA UNIVERSITY TIRUPATI CAMBRIDGE UNIVERSITY PRESS LONDON : Fetter Lane NEW YORK The Macraillan Co. BOMBAY, CALCUTTA and MADRAS u,Lt:d. The Macmillan, Co. of Canada, Ltd, TOKYO Maruzen-Kabushiki-Kaisha ; All rights te&etved THE CAMBRIDGE ANCIENT HISTORY EDITED BY J. B. BURY, M.A., F.B.A. S. A. COOK, LITT.D. F. E. ADCOCK, M.A. 4448 VOLUME I EGYPT AND BABYLONIA TO 1580 B.C. SECOND EDITION CAMBRIDGE AT THE UNIVERSITY PRESS 1928 FIRST EDITION 1923 SECOND EDITION 1924 .J REPRINTED 1928 |fi!W^l>r* V&M8Y 4 v ~ '••- «'•*"" 3PJRINTEI) IN GREAT BRITAIN PREFACE TO THE FIRST EDITION Cambridge Ancient History is designed as the first part JL of a continuous history of European peoples. The last part, the Cambridge Modern History^ has long since been complete, and the middle section, the Cambridge Medieval History^ is in course of publication, Starting with the remote and dim beginnings, upoh which some new rays of light fall every year, the Ancient History will go down to the victory of Constantine the Great in A»D. 324, the point at which the Medieval lakes up the story, -The history of Europe begins outside Europe* Its civilization is so deeply indebted to the older civilizations of Egypt and south-western Asia that for the study of its growth the early history of those lands is more important than the barbarous life which Celts, Germans, and others lived within the limits of Europe* Europeans, who wish to follow the history of their own development from its origins, must first of all become acquainted with the civilizations of Egyptian, Sumerian, Hittite, Semitic and other peoples of north-eastern Africa and south-western Asia, and therefore our first volume is concerned mainly with these peoples, Behind the civilizations of Babylon and Egypt lies a vast and still little known tract of time during which man was gradually toiling up towards that relatively high stage of civilization he had reached when he first appears to us in his written records* The discoveries which have rewarded the geologists, geographers, *md anthropologists of the last few decades have made it feasible to attempt a reconstruction of the story of man in Europe and its environs throughout those prehistoric millenniums. The story of the land-masses prior to the formation of the present con- tinental system can in some measure be written down and its significance apprehended. It is not out of place to recall -that the written history of one of the peoples of Palestine, which represents only the unscientific ideas of an Dearly age, was up to very recent 1|mes thought by learned tfaejx to furnish an authentic account of the beginnings of the earth and the 'human race, £ To-day a large though scattered mass of geological and archado^ logical facts supplies us with a little genuine knowledge p^Sit our ancestors were doing and making at a time wheii; ift£a and water and climate differed appreciably from what th^^te how, a time long anterior to that once commonly thougfit to be the date VI PREFACE of the creation of the universe itself. To ignore what is now known, little as it is and precarious as it may be, about palae^- lithic and early neolithic man, would be indefensible in a work which aims at explaining how Europe came to be what it is to-day. The activities of the palaeolithic age have helped to build modern Europe, and its effects persist; individuals of 'Ami- gnacian' descent, physically true to type, are among us stilL The first two chapters of this volume, by Professor Myres, show how the story of primitive man may be read by his latest descendants, and how the darkness before the 'dawn of history* may be illuminated by a brilliant interpreter. Chapter m, on the history of Exploration and Excavation, is designed to give the reader some notion of the arduous, qjid some- times romantic, work of a century which has revolutionized cmr knowledge of the Near East, In an account, necessarily brief, of archaeological discoveries in Egypt, Mesopotamia, Syria, the Hittite and Aegean areas, and Cyprus, the writer, Professor Macalister, shows how archaeological data have been classified and interrogated, and how unknown scripts have been deciphered and forgotten languages recovered. It seemed desirable to state the fundamental chronological problems which face the historian in regard to the early history of Mesopotamia, Egypt, Palestine, and Greece; to show how archaeological and historical evidence have been co-ordinated; and in the case of conflicting systems of chronology to explain which has been adopted and why. Chapter iv will help the reader who is not familiar with prehistoric research to understand how it has been possible to frame a definite chronological scheme, especially when the data, as in the case of Crete, are purely archaeological, Thus the first four chapters are preliminary. In chapter v Dr S. A* Cook gives a general account of the Semitic area, famous as a stepping-stone between three continents and as the home of three great religions. This chapter is a prelude to the later history of the Semites* It describes generally the mind of the Semite as revealed in his beliefs and practices, in his history and his treat- ment of history, while it tells what is known about the early history of Syria and Palestine down to the close of the Hyks<y period, circa 1580 B.C., the lower limit of this volume, In the four chapters (vi to ix) devoted to Egypt, Professor Peet treats the early predynastic age on the basis or the archaeo- logical evidence, and describes Egyptian life and thought under the Old and Middle Kingdoms (chapters vi> ix), while the his- torical events, and the historical sources, the administration and PREFACE vn the social conditions., of these two kingdoms, are dealt with by Dr H, R. Hall (chapters vn and vm). Three chapters (x to xn) on the earlier period of Babylonian history, by Professor Langdon, include an account of the interest- ing culture of ancient Susa and a discussion of the problem of the Sumerian invaders, and portray the history of the notable con- querors Sargon and Naram-Sin, in what may be called the Golden Age of the Sumerians. Mr Campbell Thompson (chapters xin to xv) continues the story, and also contributes a full description of the Golden Age of the Semitic Babylonians — the age of Hammurabi and his Code of Laws, the discovery of which (in the winter of 1901—2) threw a brilliant light on the character of society in that part of the Near East, four thousand years ago. * Ifi the chapter (xvi) on early Egyptian and Babylonian Art Dr HalPs wide knowledge of ancient art and his familiarity with the collections in the British Museum have enabled him to illustrate the aesthetic temperaments of the peoples concerned, to discriminate the periods of artistic freshness and decline, and to throw light on the difficult problems of borrowing and foreign influence. The Editors regret that it was impossible to provide illustrative plates without unduly increasing the price of the volume; but in the Bibliography to this chapter the reader will find references to illustrated books. Finally, Mr Wace has contributed the chapter on the early civilization of Aegean lands. Thirty years ago the chapter would have been a blank, because there was absolutely nothing to say. One of the finest triumphs of archaeological research has Jbeen the discovery in Crete of a wonderful and unsxispected civilization in contact with Egypt and Asia. This ancient meeting of ea$ft and west offers problems which unite the classical and the Sepiitic scholar, the Egyptologist and the student of 'Bible-lands/ \ Our first volume, then, while it contains a survey of the &arly history of a large network of inter-related lands, down to the occupation of Egypt by the Hyksos and of Babylonia by the Kassites (events which may perhaps be associated with sweeping movements in Indo-European lands to the north), may also be Regarded as a general introduction to those that will follow it. In the next volume a new age opens up, an age characterized by what we may perhaps call internationalism : Greeks whose names were well remembered in Greek records will come upon the $tage and the curtain will rise upon Old Testament history. Any exposition of the history of early ages down to 3000 years ago and even beyond, must be in a very hig|i degree provisional* viii PREFACE This is due to the fortunate circumstance that new evidence is continually and rapidly accumulating- Conclusions historians draw to-day from the records at their disposal about Babylonia", Egypt, Asia Minor, and the Aegean may be upset, corrected, amplified, or transformed by a new discovery to-morrow. Since the writing of this volume was begun, writers who had completed their contributions have seen cause to change some of their state- ments in the light of new evidence which happened to be revealed in the meantime. Obviously there is a limit to this and experts must not expect to find a reference in every case to the npwyettes de Ijt^derniere heure* Even as we are writing, Sir Arthur El^ans publishes the news that his latest excavations at Cnossus (the spring of 1922) have disclosed the fact that the end of the second phase of the c Middle Minoan ' civilization was due to lin efrtih- quake. We may note that this disaster was not contempor- aneous with the volcanic eruption which wrought ruin in Them and Therasia (see below, p. 603) 1. The appearance of some new evidence, to enable us to decide finally between conflicting views of the chronologies of Egypt and Babylonia, is much to be desired. In accordance with the opinion of the great majority of scholars we have adopted the * shorter' dates (see chapter iv, i, iii). It is desirable to impress upon the reader that the precision with which the dates are assigned is based partly upon ancient lists and computations assumed to be tmstworthy, but partly also upon modern calcula- tions of a few crucial dates as to which there is no definite unanimity. The date adopted here for Hammurabi is not accepted by some high authorities. And as to Egypt, Dr Hall is unable to accept the view of Professor E. Meyer and other historians who follow him, that the Xllth Dynasty ended in 1788 B.C*; and he puts back the date by more than two centuries. This view affects both the earlier Egyptian dates and the chronology of the early Aegean periods which depend on Egyptian synchronisms* 'Early Minoan HI/ which the latest investigations of Sir Arthur Evans have shown to extend from the Vlth to the Xlth Dynasty,, is on our chronological scheme 200 years earlier than it is on the scheme which he has adopted. See pp. 173, 656 syy, In a co-operative work of this kind, no editorial pains coma avoid a certain measure of overlapping; and in fields, where there * Weidner's recent discussion of Sargon's expedition to the west, and of the oldest historical relations between Babylonia and the Hittite area, may- be mentioned as another example of the progressive character of studies in this field (see p. 647, 6). PREFACE IX Is so much uncertainty and such wide room for divergencies of Xiews? as in the first two volumes, overlapping must mean that occasionally different writers will express or imply different opinions. It has not been thought desirable to attempt to eliminate these differences, though they are often indicated or discussed. Such inconsistencies may sometimes be a little inconvenient for the reader's peace of mind, but it is better that he should learn to take them as characteristic of the ground over which he is being guided than that he should be misled by a dogmatic con- sistency into accepting one view as authoritative and final. It will easily be understood that it is not possible to give chapter and verse for every statement or detailed arguments for every opinion, but it is hoped that the work will be found service- able to professional students as well as to the general reader. The general reader is constantly kept in view throughout, and our aim is to steer a middle course between the opposite dangers, a work which only the expert could read or understand and one so * popular' that serious students would rightly regard it with indifference. In this connexion, the problem of transliterating occurs, and a quite satisfactory solution has not been found. Conventional and accepted spellings have been retained, but where usage varies the more correct are used (for Instance Mohammed, Nebuchad- rezzar), For classical Greek names the Latin forms are adopted (as in the yournal of Hellenic Studies). In regard to oriental names, we have thought it reasonable to assume that general readers are indifferent to what experts know; and experts do not always agree as to the precise spelling. We have followed generally Breasted, Hall, and King, and the Encyclopaedia Biblica, but attention has been paid to the lists drawn up by the Royal Geographical Society, and to the transliteration of Arabic recom- mended by the British Academy (vol. vm). The difficulty of transliterating unvocalized Egyptian names and of Interpreting names in cuneiform is commented on below (pp. 119, 126), Some modern technical transliterations are as formidable-looking as the hieroglyphs themselves. In Egyptian and in the other languages «}h is adopted instead of s or the like; s for £, ts, etc.; k for q, etc.; and kh for the harder guttural fc, &. But Hatti and Habiru haye been written because *Hittite* and * Hebrew* are so familiar; a$d Hammurabi is now well enough known to dispense even ^ith a diacritical point. Names when they first occur are sometimes written with their proper vowel-lengths, etc.; but as a rule dia- critical marks have been avoided (although :, Kaahshi may be x PREFACE thought clumsier than Kassi), and more or less conventional spellings (e.g. Ashur) have been freely employed. On the? other hand, an attempt is made in the Index to register some of the more correct spellings which for one reason or another deserve attention, but could not be introduced into the text without making it unduly technical1, We wish to express our indebtedness to contributors for their readiness in carrying out editorial suggestions, in avoiding archaeological and other technicalities and in restricting the use of footnotes; for advice on questions of transliteration and on other difficult questions which arose from time to time; and foFthe preparation of the bibliographies and the lists of kings, Mr Wace is indebted to Sir Arthur Evans for his kindness in reading the chapter on the Aegean and Early Greece,, imcr the Aegean section of the chapter on Chronology. Professor Myres wishes to express obligations to Professor H. J. Fleure, to Mr Harold Peake, F.S.A., and to Mr L. H, D. Burton. Dr Cook wishes to thank Dr H. R. Hall, Professor Kennett and Dr Nicholson for help in revising chapter v. He is particularly indebted to Professor A. A. Bevan, who read two proofs, and made many valuable criticisms and suggestions. But for the views put forward in that chapter the writer has sole responsibility. Special thanks are due to Professor Myres for the Table facing p. 660, and for the preparation of Maps i— 6* For permission to use Maps 7, 8 and 1 1 we are indebted to the publishers of the Encyclopaedia Biblica^ Messrs A. & C, Black; to Messrs Chatto & Windus for Maps 9 and 10 (from the first and second volumes of the late Dr Leonard W. King's A History of Babylonia and Assyria from Prehistoric Times to the Persian Conquesi)\ and to Messrs Methuen & Co. for the plan of Babylon on p. 504 (from Dr H. R, Hall's The Ancient History of the Near East from the Earliest Times to the Battle of Salamis)* The index has been made by Mr W. E. C, Browne, M.A., former scholar of Ernmamiel College. The design on the outside cover represents Hammurabi, king of Babylonia, and is from the head of the stone monument on which is inscribed the famous code now known after his namdt on the original he is depicted standing in the conventional attitude of adoration before the sun-god, Shamashu, the god of righteousness and justice* J. B- B, 8. A. C. F, E. A. * See the letters a, c, d, g, h> j, k, q, s, t and z in the Index. PREFACE TO THE SECOND EDITION ^ ||" ^HE demand for a new edition of the first volume of the JL Cambridge Ancient History has come much sooner than the Editors ventured to anticipate, and they have not been able to do more than make some corrections and modifications which could be effected without disturbing the paging* TSe remarks which they made at the top of page viii of the Preface have been amply justified since the volume was first sent to press. j[n Egypt, the Aegean, Babylonia, Palestine and Syria, excavations have continued and interesting discoveries have been made. At Byblus, for instance, new information has been gained touching the extensive relations between Egypt and Phoenicia during the Middle Kingdom (see below, p. a 2 6), The successful diggings at el-'Obeid and Kish have supplied archaeological and historical data, of which the bearing on the period covered in this volume cannot yet be justly estimated. We may point to Mr C. L. Woolley's report (The Times^ Jan. 19, 1924) of a monu- ment of A-an-ni-pad-da, son of Mes-an-ni-pad-da (on whom see below, p. 367), and Professor Langdon's addition to the kings of Kish (tb. Jan. 22, 1924). But the information which is thus being accumulated must be submitted to a careful criticism, and that takes time, as experience shows that the full significance of fresh material cannot be evaluated at once. This is especially true of the problems of chronology, which for the early Sumerian period have assumed a new aspect through Professor Langdon's publi- cation of a very important list of the early kings. Although, with the ever-present prospect of other historical inscriptions coming to light, we cannot treat this document as decisive, yet, as its im- portance is unquestionable, it seemed desirable that some account of it should be given in this edition, and on page xiii sq* will be found a statement drawn up on the basis of Professor Langdon's publication and of some notes which he has kindly sxijpplied. A fly-sheet containing all the more important corrections and additions to this volume will also be issued separately with volume ii. Some reviewers made the justifiable criticism on volume i that it suffered from the absence of illustrations. The Editors are glad to be able to state that the Syndics of the University Press have Xll agreed to publish a volume of plates which, it is hoped, will appear in the course of I fit,, It remains for the Editors to express their cordial thanks to the contributors for help in the preparation of the new edition, particularly to Mr A, J, B.Iace in the account of excavations in the Aegean (Chap, in Section vi), and to Mr Campbell Thompson for the translation of the Kassite names thich is given on p. xv. JIB, SIC, Fl'i NOTE ON THE CHRONOLOGY OF THE EARLY SUMERO-BABYLONIAN PERIOD PROFESSOR LANGDON has recently published an important inscription, part of the Weld- Blundell collection1 in the Ashmolean Museum, Oxford. It is a large prism with eight columns of closely-written chronological material which gives the entire Sumerian lists of dynasties before and after the Flood to the end of the Isin dynasty in 2076 B.C. A small tablet in the same collection contains the names of the ten kings who reigned before the Flood, for which period it gives 456,000 years. The dynastic prism aas only eight kings before the Flood and assigns to them a duration of 241,200 years. Other important dynastic lists in fragmentary condition have been found in the Nippur Collection. These agree with the Oxford prism in giving twenty dynasties from t&e Fld%d to the Isin dynasty inclusive, and 125 kings. The first dynasty reigned at Kish (p. 365, 1. 18 from end). It included 23 kings, who are said to have reigned 24,510 years, 3 months and 3^ days. The figure recalls the 'World-year' of 25,920 years, the approximate period of the sun's apparent revolution through the twelve signs of the zodiac; but it is unlikely that the precession of the equinoxes was known even in the age of the most advanced Babylonian astronomical knowledge (Langdon, op, tit* p. 3, n. 6, cf. Kugler, Stsrnkunde und Sterndienst in Babel, u, 24—32). The longest and shortest reigns of this dynasty are 1500 and 140 years respectively; the names differ somewhat from the list on p. 665, and the name of Zukakipu (the 'scorpion') is replaced by Daggagib. The first dynasty of Erech (p. 366) counted twelve kings, reigning 2310 years. The name of the second king of the dynasty of Ur (p. 367, 1. 19) may preferably be read Meskem-Nannar. The dynasty of Awan (the identification with Awa& should be omitted on pp. 366, L 21 sf.9 438, L 14, from end) had three kings ruling 3 56 years. The details on p. 367 (lower half of the page) are considerably affected by the new prism. A list of seven kingdoms now intervenes between the semi-historic period and the northern Semitic kingdom of Akshak. The second dynasty at Kish, which succeeded that at Awan, may be placed about 3700 B.C.; to its eight kings the prism assigns 3195 years. The next dynasty ruled at Kharnazi and its king Khadanish is said to have ruled 360 or 420 years, the figures are presumably errors for six: or seven years* The sovereignty then returns to Erech in the south (c* 3400 B*C.), where the name of only one king, Enugduanna, is known. It is probable that the names of Lugalkigubnilakh and Lugalkisalsi are to be inserted here. After this second kingdom of Erech we reach the second kingdom of Ur, where four kings ruled 1 08 years. The capital now shifts to Adab for a period of 90 years, and then far to the north at Maer, where a dynasty of six kings (Ansir, [Lugaltar]zi, the rest are mutilated) reigned 136 years. It seems evident from the texts that the two succeeding kingdoms of Kish (the third) and Akshak were contemporary. If, therefore, we may follow the new source, it may be computed that these djpiasties were founded about 2967—6 B.C., in which case the first approximately fixed date in Sumero-Babylonian history will have to be placed more than 200 year* lower than that given on p. 367, L 4 from end. Moreover, it would now seem that the old third dynasty of Kish disappears (sec p. 667 [8], and n. 4); the two kings Urzaged and Lugal-tarsi belong to the second dynasty of Erech, and Mesilim possibly to the Awan dynasty (Langdpn, opf cit. 1 Oxford Editions of Cuneiform Texts, n. The Weld~$lundett collections, vol. ki. Historical inscriptions, containing principally the chronological prism, W-B 444. Oxford, 1923. C.A.H.I. xiv CHRONOLOGY OF SUMERO-BABYLONIAN PERIOD p. 6 /f.). The first kings of ELish of whom we have contemporary records apparently belonged to other kingdoms, and claim the title because of its dignity. Oa p. 3^3,^ L 9 jf. read: who followed the second kingdom at Kish and the brief dynasty of Khamazi (r. 3400). The third (not/0«r/j£) dynasty of Kish was founded by Kug-Bau, as the name should now be read instead of Azag-Bati (p. 370 last par., and 1. 7 from end). Ur-Nina was contemporary with the rulers of Maer, not Akshak (p. 379, L 14 from end). On p. 380, 11. 9—o? omit the words: convincing evidence. * * dynasty, and ib. 1. 6 from end, for Uruazagga the better reading now is Uru-kugga, Rimush (p. 408, L 19 from end), according to the Oxford prism, reigned nine years. Manishtusu was his elder brother (p. 409, L 21 Jf.). Naram-Sitx was his son (contrast /^,), although Babylonian tradition calls him son of Sargon (p. 4r^foot)1. For 22 read 24 (/<£.last line); and note that the prism gives a much lower figure for his reign — probably 38 years (p. 413, L 6). The fifth dynasty of Erech contains only one king, Utukhegal, to whom is ascribed a reign of 7 years, 2 months and 7 days (p. 434, last par.). T o Dungi^p. 43^, L 6) is ascribed a reign of 47 (not 58) years, and Langdon reduces all the figures in his reign (11. 4—18, and also p. 456, 1. 21 from end) by eleven. The length of the reign of Bur-Sin (p. 4575 L 20) is given as nine (not «<gvijf) years. Finally, on the basis of the Oxford prism and other evidence Langdon arrives at dates generally lower than those adopted in this volume. Starting from Kegler's brilliant interpretation of the tablet of observations of the planet Venus for the twenty- one years of the reign of Ammi-zaduga, the tenth king of the First Babylonian Dynasty, and in consultation with the Oxford astronomer, Dr Fotheringham, he now holds that the beginning of this dynasty may be placed at 2 1 69 E.G. The astronomical calculations in themselves are not entirely final, and the argument also turns upon the precise beginning of the year in certain contracts relating to the division of the date-harvest and the renting of fields in the seventh-eighth months. The date which Langdon now adopts is fifty-six years lower than that adopted in this volume (pp. 404, 1. 3, 479, 67.3)> but he definitely rejects the much lower dates for the dynasty which are held by Weidner (viz. 2057, see p. 672, xu i) and Kugler (viz. 2049), Langdon maintains the date 2357 for the beginning of the dynasty of Isin (pp* 471, 672); but, besides the modification of the earliest approximately fixed date (viz. 2967—6, see above), other important changes are suggested arising out of the Oxford prism. Thus, the Maer-Akshak-Kish domination (p. 373^ L 14) may be dated 3 103-2777, For the Kug-Bau dynasty (/£„ L 7) he suggests 2967-2873, and a similar reduction of about 120 years becomes necessary on p, 378, L 12 (viz. 2967-2873)* So the date of Sargon becomes 2752 (pp. 368, L 16 from end, 403, 1. 8). Lugal-zaggisi begins to reign in 2777 (pp. 39 5, L 21, 402, L %), The fourth dynasty of Erech Is dated 2571-2542 (p, 423, L 9), and that of Gutium becomes 2541-2416 (pp. 423 f$.9 670). Ur-Bau's date is 2620 (p. 373, L 26). The end of the last dynasty of Ur is fixed at 2328 (p. 377, L 13), and Dungi and Bursin are dated respectively 2391 and 2345 B.C. ,(pp, 437, 1 5, 457, L 19), These dates indicate the complexity of the chronological problems, and the difficulty of obtaining conclusive results, owing to the serious differences among tfce ancient souf ces themselves and the frequently very intricate character of the astro- nomical and other questions. They are not to be regarded as final, but it seemed desirable that a general statement of the evidence published by Prof* Langdon should be made accessible in this edition. 1 Prof. Lang-don adds that Sargfon claims to have collected ships from Melukhkha, Magan and Diknun at the quay of Agade (addition to p, 404, L 14), ON i i ^ i ring is a translation of the Babylonian renderings of the names of tie twenty-one kings, diiefly Kassites, mentioned on p, ^ of volume i; 'J) "Traeiin" 'Offspring of k Lord of Us' 'Servant o or 'Help of Bel " s Jil <( U Protect(ion) of [Stall]11 ion)oftkLordofknJs]' eiiDotieW TABLE OF CONTENTS CHAPTER I PRIMITIVE MAN, IN GEOLOGICAL TIME BY JOHN L. MYRES, O.B.E., M.A., D.Sc., F.S.A. Wykeham Professor of Ancient History, Oxford PAGE I. JUKE SETTING OF THE STAGE i Definition of History I Nature and Man ..»,.,.„. 2 IL PRE-GLACIAL GEOGRAPHY 3 Theseaof 'Tethys* 4 Tertiary mountain-building 6 Crust movements ..,,,.... 8 Beginning of the Mediterranean * 9 Tertiary flora and fauna . . , . . . . . r I Africa separated from Asia 13 The Highland Zone 14 Relation to the Southern Flatland 16 African fauna , . . . . . . * * 17 III. THE GLACIAL CRISIS 1 8 Effect upon flora and fauna . . . , . . . 19 IV. THE PRINCIPAL HUMAN RACES 21 Mongoloid man »,..»,».. 22 His extension 24 African fauna and African man * . * , * • • 2 5 Sequence of human types 27 The white races , ,..,,.. 28 V. PALAEOLITHIC MAN IN THE SOUTH AND EAST , , * . 31 The Nile Valley ......... 33 Domesticated plants and animals . * , . . * 35 Links between Egypt and Europe 36 Man in Syria and Arabia 37 The Semites 38 Palestine 39 The Euphrates and Mesopotamia ...... 40 V L THE ICE AGE IN THE NEAR EAST ,,..., 42 Conditions in Armenia and Iran .43 VII. THE ICE AGE IN EUROPE 45 Mousterian man ......... 4^ Later types * ... .*..«» 48 Later palaeolithic cultures * S° VIII. THE CLOSE OF THE OLD STONE AGE „ . . • • • 52 The kitchen-middens « . - * , , * , • S3 Swamp and forest in north-west Europe 54 xviii CONTENTS CHAPTER II NEOLITHIC AND BRONZE AGE CULTURES BY J. L. MYRES PAGE L THE HIGHLAND ZONE AND ALPINE MAN « . » . . 57 The forests . 58 Varieties of man .*.»..... 59 Forest culture and polished implements . ... - 63 II. CHARACTERISTICS OF NEOLITHIC CULTURE . 5 5 Inventions ......... $7 Eurasian and Eurafrican cradle-lands . . . ^9 Pottery and pottery styles ....... 70 III. REGIONAL TYPES OF NEOLITHIC CULTURE: ALPINE EUROPE . „ 71 The lake-dwelling 73 IV. REGIONAL TYPES: THE DANUBE BASIN . , . . . 75 Daimbian pottery ......... 77 South-eastern Europe and Asia Minor 79 V. REGIONAL TYPES: THE TRIPOLJE CULTURE . .... 80 VI* TKE CULTURE OF THE NORTH-EASTERN STEPPE . . , . . 82 Waggon-dwelling culture; languages ...... 84 VII* THE CULTURE OF ANAU AND SUSA . , . * . . . 85 Contact with the west 4 88 VIII. THE RE0-WARE CULTURE OF THE NEARER EAST .... 89 The influence of Cyprus and Syria ««*... 90 IX. THE CULTURE OF THE EASTERN MEDITERRANEAN . * . . 92 Early Aegean culture ..... 93 X. THE CULTURE OF THE WESTERN MEDITERRANEAN AND ITS OFFSHOOTS * 94 'Megalithic* origins . <, * . * » „ * 95 XI. THE CULTURE OF THE BEAKER-FOLK: . , * . . xoo XII. THE COMING OF BRONZE . . . . . „ . .103 Aegean influence . „ . f , . . . .105 XIIL THE HALLSTATT CULTDRE . . . . « . . 106 The horse . . . . * , . * . 107 First appearance of iron . . . . . . . 109 Cremation . . . . . . , . . j to CHAPTER III EXPLORATION AND EXCAVATION BY R. A. STEWART MACALISTER, L,iTT*D.> F.S.A. Professor of Celtic Archaeology, University College, Dublin I. THE RELATION op ARCHAEOLOGY TO HISTORY . . » . . r 1 a Petrie's pottery test . . . . . . . , * XI4 II. EGYPT : (a) Surface exploration , . . . , . .116 (&) Decipherment * . . . . . „ ,117 (c) Excavation , ,120 III. MESOPOTAMIA: (a) Surface exploration . . . „ , , » 122 (ff) Decipherment , . » . , . » • . 123 (r) Excavation . , m , , , ^ ,127 CONTENTS xlx IV* STRIA AND PALESTINE: PAGE (a) Surface exploration. . . . . „ . .130 (<£) Decipherment . . . . . . ...132 (<r) Excavation . . . . . . . .132 V. THE HITTITE EMPIRE . . . . . . . . • *35 VI. THE AEGEAN CIVILIZATION . . * . . » * .136 The work of Schliemann , , . . . . 137 Periods of Cretan culture. . , . . . . .139 VII. CYPRUS ........... 142 Decipherment of Cypriote , . . . . . 1 44 CHAPTER IV CHRONOLOGY I. MESOPOTAMIA BY STANLEY A. COOK, Lrrr.D. Fellow of Gonville and Caius College, Cambridge Introduction ........... 145 Mesopotamian usage . . . . . . . . . .147 The limmu lists ........... 148 The testimony of Berosus . , . . . . . . .150 Assyrian data » * . . . . . . . . .153 Dates of Hammurabi and Saigon . . . . . . . .155 Table of dates 156 II. THE OLD TESTAMENT BY S. A. COOK Character of data . . . . . » . , . .156 Period of the monarchy . . . . . . . . .158 Exilic and post-exilic period . . . . « . . * ,162 Pre-monarchical period > . . . . . . . .163 General character of the chronology . . „ . . . .165 Table of dates 1 66 III. EGYPTIAN CHRONOLOGY BY H. R. HALL, D.LITT., F.S.A. Keeper of Egyptian and Assyrian Antiquities, British Museum Direct sources .... ....... 166 Sothic cycle ........... 168 Date of Xllth Dynasty . , 169 Date of Menes . . „ . . . . . . . .171 Institution of the calendar, 4241 B.C. ....... 172 Table of dates 173 IV. PREHISTORIC GREECE BY A. J. B. WACE> M.A. Formerly Fellow of Pembroke College, Cambridge j Late Director of the British School of Archaeology, Atb*™ Archaeological periods , ... tyrj Early and Middle Minoan 175 Late Minoan . . . . . . . . , ^ , [76 Greek legend and tradition . . '. . . , *"- , 178 Helladic and Minoan co-ordinations . . , . . ,179 Thessalian periods ..*.., ...180 xx CONTENTS CHAPTER V THE SEMITES BY S. A, COOK PAGE I. PEOPLE, LANGUAGE AND MOVEMENTS . , - . . 1 8 1 Geographical limits . . . . . - - ,182 The * sons7 of Noah: Shem 184 The Semitic languages . . . . . - , .186 The alphabet 189 Migrations and trading movements - „ , . . .190 Semitization of immigrants . . . . . . * ' 192 Influence of Arabia , . .193 II. TEMPERAMENT AND THOUGHT ....... 194 Psychology of the languages . „ . . - .195 Religious characteristics »....- . 197 Polytheism and Monotheism . . . . . . .199 Semitic and non-Semitic thought .*.,.. 203 The extremes of the Semites . . , . , . .205 IIL SOCIAL AND POLITICAL DEVELOPMENT ...... 206 The^W 207 Attitude to the divine powers „ . . , „ . .209 Fundamental ideas , . . , . . . . ,210 Men and the gods , . . . . . . „ .213 The sanctity of kings . . . . . . . .214 Historical vicissitudes , . . , . . . ,216 IV. TREATMENT OF HISTORY- . . . . . . . .2x7 Treatment of tradition . . . . , * . ,219 Attitude to development . . . . . . . .231 The writing of history , . . , . » . ,222 Historical ideas . , . * , * „ . ,223 V. SYRIA AND PALESTINE * .... , . ... . 225 The story of Sin tihe . * , „ . . * ,226 Amor and Mesopotamia . * . . . . . ,230 Amorite gods P . * . . . » . .231 The Hyksos 233 Native Palestinian traditions . , , . » , .234 Genesis, chap, xiv , . . . . . . . .236 Paucity of historical material , . , » ^ . .237 CHAPTER VI EGYPT: THE PREDYNASTIC PERIOD By T. ERIC PEKT, M.A* Professor of Egyptology, Liverpool University L THE EVIDENCE OP THE CEMETERIES „ » . „ , ,238 Predynastic burial .....»»», 239 Predynastic settlements . . , , , . , ,241 Pottery and stone vases .»,.*.*, 243 Physical type, language and religion . - „ „ . 244 CONTENTS XXI PAGE II. JC^iTA FOR HISTORY ......„., 247 Introduction of tlie Calendar ....... 248 Sources for the predynastic period . , * . , .250 Historical slate palettes . . . . . . . .251 Ivory- knife-han die from Gebel el- Arak . „ . . .252 Original home of the predynastic Egyptian . * . . .254 Indications of eastern origin , , . „ . . .255 CHAPTER VII THE UNION OF EGYPT AND THE OLD KINGDOM By H. R, HALL L THE LISTS OF KINGS: DYNASTY I . . . „ . ,257 Sources . . . . . . . , . .258 Infiltration of aliens . . . . . . . . 261 Hamites and Armenoids ........ 262 Kingdoms of the north and south . . . . . .265 Pre~Menic kings ......... 266 The originals of Menes . . . . . . .267 Narmerza .......... 268 The court of Semti ......... 270 The dead and mummification ..,..,. 272 II. DYNASTIES II— IV . . . . . . , , .274 Zos&r and the first pyramid ...»**. 276 The age of Snefru ......... 278 Pyramids of Gizeh. . . . . . . . 28 r Zenith of Egyptian art . . . . . . . .282 Mycerirxus . . , . , * " . , . .282 III. THE CLOSE OF THE OLD KINGDOM . . . . . . .284 The 'son of the Sun-god* 285 Art and religion . . . . . . . . .286 The 'admonitions of Ptahliotep' . , . . . .288 Unis and the pyramid at Sakkarah ... ... 290 Pepi ........... 291 Uni in Palestine ......... 293 Entrance of negroes . . . . . . . .295 The Heracleopolites ........ 297 CHAPTER VIII THE MIDDLE KINGDOM AND THE HYKSOS CONQUEST BY H. K. HALL DYNASTIES XI AND XII ........ 299 Amenemhet I and the god Amon . * . , . 301 The Instructions of Amenemhet . , . , . .303 The story of Sinuhe . . . . . . . 3 04 The works of Senusret (Sesostris) I . . . * , . 305 Relations with Crete . , . . « * . .307 Senusret III, the historical Sesostris . * . ,,, . - 308 Amenemhct III , * , , . , ,r . • . . 309 xxii CONTENTS PAGE II. THE HYKSQS , 310 North Syrian movements . . . . . - . . 3 F2 Yekeb-hal, Khian and other kings . . . . . - 3 1 3 Expulsion of the Hyksos . . . . - « « 3r4 III. THE INTERNAL CONDITIONS OF THE AGE . . . - • .315 Life of the people 3*7 Officials and soldiers . . . . . » - 3 * & Tombs and religion . . . . . . • .321 The priesthood 323 Religious literature . . . - . . . . .324 A Messianic prophecy . . . . . . . ,325 CHAPTER IX LIFE AND THOUGHT IN EGYPT UNDER THE OLD AND MIDDLE KINGDOMS BY T. E. PEET General Egyptian character - , * • . . . * ,326 I. THE ARCHAIC 3PERIOD AND THE OLD KlNGDOM , . . . .328 Local and solar cults „ . . . . . . 329 Osiris - . . . . . , . . . -S3^ The^ 334 The tomb, death, and the hereafter . . . * * -33^ II. THE EARLIER INTERMEDIATE PERIOD, VllTH TO Xril DYNASTIES . . 340 Language and writing . . . » . . . .341 Early literature . . . . . . . , » 343 Pessimism . . . . . , , . * * 345 III. THE MIDDLE KINGDOM . . . . . , . ,346 Moral standards . . . . . . . . .347 'Story of the Eloquent Peasant' 349 Coffin Texts . „ * 351 Belief in a judgment . . .*. , . » »353 Hfke, rnagic and morality. . . . . „ . 354 CHAPTER X EARLY BABYLONIA AND ITS CITIES BY STEPHEN H. LANGDONT, M*A.7 B,D., Pn.D* Professor of and Shiliito Reader in Assyriology, Oxford L PHYSICAL CHARACTERISTICS * * , , * . . 356 The Euphrates and Lower Mesopotamia , . * . 358 Sumer and the date-palm , . , » . . 3 6C II. THE ORIGIN OF THE SvMERIANS . . . . . * ,361 The cultures of Anau and Sxisa * * , . . „ ,362 IIL EARLIEST TRADITIONAL DYNASTIES . . . . , . .364 The first city-states . * . „ . . . m .365 The third dynasty of Kish . . . , , . ,368 The fourth dynasty of Kish . . * , , . 370 Sumerian writing and religion • , . . » . 371 CONTENTS xxili PAGE TV TgE RECORDS OF THE CITY-STATES . . . . . . * 373 Lagash . . . . . . . . . . -373 Enkhegal and Ur-Nina . . . . . . . .374 Shuruppak and its legends . . . . . . • 377 The dynasty of Ur-Nina, 3100 B.C. . . , . . 378 Eannatum and Enannatum . . . . . . .380 Entemena and his son . . . . . , . .382 Rise of priests of Lagash . . . . . „ . .385 Social reforms of Urukagina . . . . . „ ,387 Inroad of Lugal-zaggisi . . . « . , . .388 V. OTHER crriEs , . . . . . . . . .389 ~ Umma . ..,.*,.... 389 Adab .*.-**..,.. 39° Nippur 391 Isip and Larak ......... 393 Kish ........... 393 Cuthah 394 Sippar 395 Erech. ........... 396 Larsa (Ellasar) ......... 397 Ur 398 Abu Shahrein (Eridu) . . . . . . . .399 Myth of Adapa ......... 400 Ashur ........... 401 CHAPTER XI THE DYNASTIES OF AKKAD AND LAGASH BY S. H. LANG DON I. THE RISE OF THE DYNASTY OF SARGON ...... 4O2 Stories of his origin , . , . . . . .403 Conquests in the west ........ 404 The foundation of Agade ....... 407 Accession of Rirnush ........ 408 Manishtusu .......... 409 Contemporary monuments . . . . , „ .410 Purchase of estates . . . . . , . . .411 II. NARAM-SIN AND THE DECLINE OP THE DYNASTY OF SARGON . . .412 Deification of Naram»Sin . . . * . . * .413 His conquests ......... 414 Expedition to Magan . * . . . . . .415 The * Stele of Victory* ........ 41,7 Submission of Elam, Lagash and Nippur ..... ;^8 ]<• Reign of Shargalisharri ........ f;T^v The rise of Gutium , ; 1421 Period of anarchy . . - . * . » .,; ;;V; 422 III. GUTIXJM AND LAGASH ....... 423 The kings of Gutium * . . . . . ./ . 424 Ur«Bau of Lagash . . . . * , » . 425 xxiv CONTENTS PAGE IV. THE KINGDOM OF GUDEA OF LAGASH „ . . . - ¥ 426 The statues of Gudea • 4"^s Contemporary art and literature . . « - • 43 2 Overthrow of dynasty of Gutium . . . . . .434 CHAPTER XII THE SUMERIAN REVIVAL: THE EMPIRE OF UR By S. IL LANG DON I. UR-ENGUR AND DUNGI .435 Might of Ur-Engur 43 6 Conquests in the east . . . . . . „ .438 Submission of Susa ........ 440 II. LAGASH AND OTHER CITIES OF THE EMPIRE . - . . .441 Sumerian liturgies . „ . * . . , . .443 The principal cults . . . . . . , , 4 44 Conditions in Akkad ..,.*..« 44^ III. THE EASTERN PROVINCES ..<....,. 447 Early deities of the east . . . . . . . ,448 Semitic infusion ......... 450 IV. THE NORTHERN ANZ> WESTERN EXTENSION . . . . . ,451 Ashur ...**.... 451 Subartu . . . * 452 Cappadocia and its Semitic colony „ . . . . ,453 V. THE DECLINE OF SUMERIAN POWER . . . . - . .456 Bur-Sin .......... 457 Gimll-Sm ..„„....„* 458 Ibi-Sin and his overthrow » « » . * * .459 Sumerian law and calendar *»..*.* 461 The influence of the Sumerians . * . . . .462 CHAPTER XIII ISIN, LARSA AND BABYLON By R. CAMPBELL THOMPSON, M.A., F.S.A. Fellow of Mcrton College, Oxfc rd L THE POWER OP THE SEMITES .,.,.»*. 464 West Semitic elements .»«.«.*. 466 Amor (Amurru) ..,.,.... 467 Early Assyria ...*,,„»*. 468 Kara-Euynk and Kerkuk . , . . . . . .470 11^ THE DYNASTY OF ISIN ....»..,. 47o Overthrow of Ibi-Sin ..»...«. 47 r Contemporary laments . „ . * . . . .472 ^Chedorlaomer* and G-encs;s xiv . „ f » . .473 New Sumerian activity ..«*.... 474 An Amorite raid .*..,..*. 476 The wars of Gnngunurn - . , , * . , » 477 Larsa ........... 478 The First Dynasty of Babylon . » , » „ , .479 CONTENTS xxv PAGE Relations with Larsa and Isln . . . * , . .480 Defeat of Larsa by Elam . , - . . . . 483 Elamlte kings. ....... 484 Rim-Sin's successes against Isin . . . . , .485 Fight for Isin and Larsa . . . . . . . .486 III, HAMMURABI .......... 487 Conquest of Elam ......... 488 Temple and other works . . , . . . , .489 Campaigns in the north ........ 490 His law-code ......... 492 Extent of his empire ........ 493 CHAPTER XIV THE GOLDEN AGE OF HAMMURABI BY R. CAMPBELL THOMPSON I. THE COUNTRY .......... 494 Communications by water . . . . . . .495 Ships and houses ......... 497 The date-palm ......... 499 Animals and birds . . . . , . . . .500 The Tigris .......... 501 II. BABYLON . 503 Plan . 504 Nebuchadrezzar's buildings . . . * . . .505 Tower of Babel 508 III. GOVERNMENT AND SOCIETY ........ 508 The patesis 509 Judicial procedure . . . . . . . . . 511 The levy 514 Capital offences, penalties . . . - . . .516 Social castes " . . . -5*8 Slavery 520 IV. PRIVATE LIFE 522 Matrimony . . . . . . . . .523 Divorce and adultery . * , * . . . .524 Children and inheritance . . . . - . « .526 Loans ........... 528 V. RELIGIOUS INSTITUTIONS . - . . . - * * 529 The gods 529 Religious beliefs . . . . . . . . -53* The temple and its staff 532 Priestesses and temple- women , . ... . . .536 VI. ORDINARY LIFE, DEATH, LITERATURE ...... 546 The crops -54* Food 543 Coinage, metals and pottery . . » . . ,.545 A love-letter ........ * 547 Burial . * . . 548 Myths and legends * . - . . , % » 55° xxvi CONTENTS CHAPTER XV THE KASSITE CONQUEST BY R. CAMPBELL THOMPSON PAGE I. THE END OF THE FIRST BABYLONIAH DYNASTY * . . . . 552 The Kassites and their language . . - * . 553 The kings of the Sea-country 555 Their advance . . . . - - * - 5 $7 Abeshu' and his artificial Hoods . . . . . 558 A Hittite raid • - .561 Decline of Babylonia . . . . • * * ,562 II. THE KASSITE DYNASTY . . . . . • . . -5^3 Internal conditions . . . . . « - . .564 Religion and art . . . . » * . . .567 Prelude to the 'Amania Age' 568 CHAPTER XVI THE ART OF EARLY EGYPT AND BABYLONIA BY H. R, HALL I. EGYPTIAN ART ..«,..»* . 570 Use of naetals . * * - , . . fc » 571 Archaic art . . . . - . . * , * 573 Portraiture * , . ,,574 Small art .....*,*. 576 II, INTERRELATIONS WITH BABYLONIA * . * . . , - 577 Prehistoric pottery . * . , . . . , .578 The Gebel el~Arak knife-handle . , . . . .580 Use of stone t . * . . . . . . 582 II L BABYLONrrAN ART . . . . . . , » * 5 84 The copper lions of el-*Obeid * * , . , « ,585 Relations with the west * . . « * » m ^ CHAPTER XVII EARLY AEGEAN CIVILIZATION BY A. J* B. WAGE I. CRETE . . . . . 589 Transition to Earty Minoan Age . , coo Conditions «... „ * . . SOI Middle Minoan Age . . * « * 593 The script . . , JQ j Palaces of Cnossus and Phaestus „ . ^ * 595 Middle Minoan culture . . * * . 596 Transition to Late Minoan , . * CONTENTS xxvii PAGE II. T,PE_CYCLADES * . . 599 Earl/ Cycladic culture Relations with Crete III. THE HELIADIC CIVILIZATION . Early Helladic Period Middle Helladic Period . Minyan ware Cretan influence IV. THE THESSALIAN CIVILIZATION First Thessalian Period . Second Period Third Period First and second cities of Troy , Supremacy of Crete Appearance of Mycenae . 600 602 603 604 606 607 608 609 609 610 6ix 612 614 615 LIST OF ABBREVIATIONS 617 BIBLIOGRAPHIES Chapters I and II ............ 619 Chapter III 625 Chapter IV 628 Chapter V 630 Chapter VI 636 Chapter VII 637 Chapter VIII 640 Chapter IX 643 Chapters X-XI1 645 Chapter XIII 649 Chapter XIV 651 Chapter XV 652 Chapter XVI 653 Chapter XVII 655 SYNCHRONISTIC TABLE 656 COMPARATIVE TABLE OF PRINCIPAL SEQUENCES AND CORRELATIONS IN SELECTED REGIONS BETWEEN NORTH-WESTERN EUROPE AND MESOPOTAMIA FACING 660 LIST OF EGYPTIAN KINGS OF THE OLD AND MIDDLE KINGDOMS, c. 3 500-1 580 B.C 661 LIST OF KINGS AND PATESIS OF SUMER AND AKKAD %$ KINGS OF ISIN, LARSA, BABYLON, ETC. , - . 672 GENERAL INDEX . . , . . . - .676 CONTENTS LIST OF MAPS, ETC. PACK 1, Stages in the growth of Land-masses , , , , FACING 16 2, The Ice Age », 48 3, Zones of Vegetation, to illustrate sequence of climatic regions „ 64 4, Olive, Vine and Orange Areas of the Mediterranean , „ 64 5, Principal Neolithic cultures „ no 6, Europe showing the principal lines of Early Bronze Aje Intercourse „ no 7, Trade-routes of Hither Asia „ ^224 8, Egypt „ 324 9, Babylonia, stowing the sites of Ancient Cities, , , ^ 400 10, Babylonia, Assyria and Mesopotamia , , , . „ 4/14 n. Syria, Assyria and Babylonia „ 566 12, Map to illustrate Early Aegean culture , , , , ,. 614 Plan of Babylon , 504 CHAPTER I PRIMITIVE MAN, IN GEOLOGICAL TIME L THE SETTING OF THE STAGE ¥ TflSTORY, in its common and more popular sense, is the JLJl $tudy of Man's dealings with other men, and the adjust- ment of working relations between human groups. But there is a larger sense, in which Human History merges in Natural History, and aludieS the dealings of Man with Nature; and it may be ob- served that it has been only by slow degrees that any human group has attained to such vision of the unity of mankind, or of civiliza- tion, as might constrain it to regard other human groups as more than a peculiarly intractable element in its own natural surround- ings. An austere conception of War — that under certain circum- stances Right has no court of appeal but Might — survives to remind us that Man has not yet wholly rid himself of this con- fusion between things and alien persons; and the most modern conception of international right so far accepts this fact of an alienation between the higher functions of human groups, however reasonable, as to take differences of language — of the medium, that is, for interchange and reconciliation of ideas, — as the best guide when and where, for the present, it is safer to keep human groups apart, and let them manage their affairs as far as possible each in their own way. History, in the narrowest sense of all, as the interpretation of written evidence for arrangements made for right living within a human group, or between such groups, accepts implicitly the same criterion, and stops short where such evidence is not available. Linguistic Pa^a525|^^y: goes a little further back, in the study of the distnf^^ groups, and of such relations between them as loan-words, or structural likenesses in the speech, mgy suggest. But the spoken word does not fall to the ground, like the spent missile or the broken vessel, to be its own memorial of human achievement: it vanishes in air, so that the philologist deals not with originals, but at best with the reminiscence of an echo. To recover, therefore, what men were doing, or maKliig, still more what they were thinking or desiring, befofe the dawn of history, the sole available method is that of the archaeologist, C,A,H«I I 2 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP, merging as it does In that of the geologist : since these alone^ handle and interpret original creations of men's thought and will, and contemporary elements of the physical surroundings of those men. Where the tree falls, there shall it lie, and where the lost implement or shattered potsherd, or worn-out man fell, there have they lain, for all that any one cared then, or knows now. It is the careless- ness (in the literal sense) of the river as to the gravel which it carried, and an equal carelessness of those men as to what happened to their leavings., that justify such a hypothesis of the credibility of these data, and make prehistoric times at least a penumbra of history, ^ ^ "TSfor"are we compelled any longer by prejudice or authority to regard those times as catastrophically short, any more than we must believe that Rome was built in a day. Man's prehistory merges in the pageant of the animal world, and of the planet-wide arena on which it has been in progress. Mountain and sea-basin too have their history. Their geographical distribution has varied in immemorial years; the faith that can remove mountains is the same in kind as that in which the historian brings together armies and frontiers, 'bone to his bone/ showing *all the kingdoms of the world in a moment of time/ Such 'historical* geography and 'historical' ethnology are a proper prelude to the history of the ancient world; and much, even within that history, cannot fully be understood without them. Ancient peoples come upon the stage of history, not all together, but in a certain order, and by their proper entrances; each with a character and make-up con- gruous with the part they will play* The pageant— or is it the drama ? — of history presupposes the formation of that character, and its equipment, in the green-room of the remoter past; and the sketch of the growth of initial 'cultures,* which follows now, is intended, like the hypothesis of a Greek play, to describe how men came by those qualities of build and temperament, those aims in life, atid the means wherewith they were attempting to achieve them. For, to the student of prehistory, a * culture is nothing more or less than this — the total equipment with which each gene- ration of men starts on its career, in whatever external conditions; to the archaeologist, no less, it is literally that equipment whirh the men of each generation were discarding, when they and it respectively ceased t6 be of Significant use. To see how the stage itself was set for this pageant, we must look back beyond the moment "when the first characters enter it. For it has been Nature, rather than Man, hitherto, in almost every scene* that has determined where 'the action shall He* Only at a I, n] 'NATURE5 AND 'MAN3 3 comparatively late phase of that action3 does Man in some measure shift t*he scenery for himself, * And by Nature and Man are here meant neither supernatural force nor superhuman design, altering the arrangement of us and our surroundings like chessmen on a board. Nature, adopted in our speech from Latin natura^ an unlucky mistranslation of Greek $>vcn^ stands as a common and inclusive term for all 'physical' events that happen; its Greek original being a verbal substantive signifying the fact of growth, the 'way things grow/ the mere processes of a world as apprehended by a mind. It has nothing to do, SB its Latin antecedents might suggest, either with birth or any sort of coming-into-being; nor with any question 'what shall it be in the end thereof ?* These are matters outside * natural9 history and human history alike. All history is the mere study of processes, of the 'way things grow* in the old Greek sense; for to this, modern thought has laboriously but unequivocally reverted> after long preoccupation with beginnings and endings, with cos- mogony and eschatology of all kinds, in the centuries between Greek science and our own. Within this Nature, so presented as a process or coherent sequence of occurrences, and so far as we know (by inference of me and you, each from experience of the rest of us corporeally participant in what goes on) a part of this Nature, stands Man, perceiving what goes on, learning what that is, conceiving it as alterable by inventive effort, and striving accordingly, with experience of what we call results, great or small, of that strife. By Man, then, in what follows, is meant the collective total of such perceiving, learning, inventing, striving and experiencing * selves/ myself and yours and theirs. By races of men, are meant groups and sequences of such selves linked by corporeal similari- ties propagated by natural process within each group: by peoples or nations^ groups of selves exhibiting peculiarities of interpreta- tion, invention, and effort sufficiently similar for their results to be cumulative and coherent; and by cultures or civilizations the accu- mulated and coherent results of such similarities in the activity of selves like you and me. IL PRE-GLACIAL GEOGRAPHY The stage of human history is a wide one from the firs|.;;®vieii disregarding those varieties of man in inner Asiaiorrjc«jg3ferial Africa which come latest and most incidentally iB~to,^i::S|ory? the stage even of ancient history is the whole home o£ffieP ''white races/ 4 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. from the Atlantic coast of Europe to the Persian plateaux, from the Sahara to the Baltic; the north-western quadrant of the? land- mass of the Old World, To understand even the actual configuration of this area, some of which Is very complicated — still more, to understand the changes which have occurred in the form and extent of the land-masses since they have been inhabited by man — we must review the whole series of events which have resulted in the formation of the present European peninsula, of the sea-basins which lie north and south of it, and also of its eastward continuation into Hither Asia, a similarly constituted highland with comparatively low-lying* flat- lands to north and to south. For, if we trace this series of events far enough back, we reach, at all events, the more immediate reasons for those strongly marked contrasts in the composition and structure of Its rocks, which have so profoundly affected the habitability and human prosperity of each component region, through the peculiar distribution of its plants and animals, and eventually of its breeds of Man. Herodotus, attempting to summarize the contrast between the northern flatland and the Aegean cradle of the Greeks, describes Scythia as a land where there are no earthquakes and they grow corn for sale. That immensity of arable is itself the corollary of the flatland's long immunity from geological stress, and its accu- mulation of successive sediments, as sea-floor or dusty desert. The recurring earthquakes in Greece and Italy, through ancient and modern times, are sufficient evidence that the process of mountain building is not yet complete, and the rarity and discontinuity of cultivable soils illustrate the dislocation and wear-and-tear inci- dental to such a process. The catastrophic geology of Genesis and the Psalms voices the same experience of Nature's workings among a people of the Nearer East. Let us summarize, then, the main course of that period of planetary history, within which the history of Man is one of the more recent episodes. The chalk which composes the 'white walls' of England, the massive limestones of the 'hills which stand about Jerusalem/ and the similar grey limestone which gives its wilder grace to the land- scape of Greece, were formed by deposition on the floor of a gre»t sea which, covered all, and more than all, of the stage on which history has played its greatest drama hitherto. This sea, to which geologists give the picturesque name of *Tethjy/ belongs to the second of the three great schemes of oceans and continents, whose distribution can be distinguished in the long course of the earth's history. It had taken shape as the result of that period of violent I, ii] SECONDARY DISTRIBUTION OF LAND AND SEA 5 planetary convulsion which closed the * primary* phase, and its q^liteSration, with the exception of the Mediterranean Pontic- Caspian, and Caribbean basins, marks the change from the 'secondary' to the 'tertiary' in which human history is the most recent episode. Unliie the modern Atlantic and Pacific Oceans of the * tertiary' phase, which (whatever their breadth) extend from the Arctic to the Antarctic circle, Tethys had its greatest diameter from east to west, and was comparatively narrow from north to south. Eastward it abutted on an ancient s Angara' continent, of whi^h the solid core lay in north-eastern Asia, with more recent extensions further south: westward it opened into a Pacific Ocean. Southward it was bounded by another ancient continent, 'Gpnd- wana-lan^L,' which had once extended in one vast oblong from weat of South America to east of Australia, but was already foun- dering in places, so that growing gulfs in its southern margin were separating South America from South Africa, and South Africa from Australia; first symptoms of the South Atlantic and Indian Oceans that were to be. Similar collapse of its northern margin allowed the waters of Tethys to form a deep bay between Brazil and Morocco ; and a long gulf between East Africa, on the one hand, and, on the other, a 'I^muxian^ peninsula connecting South Africa through Madagascar with peninsular India. Both of these eventually broke clean through to meet the southern gulfs, and insulated South America and ' Lemuria* for ever* Round the north end of 'Lemuria/ there was in due course open sea between Tethys and the new Indian Ocean; and meanwhile the rise of the first mountain structure of south-eastern Asia connected the Australian fragment of old 'Gondwana' with the southward ap- pendages of * Angara-land,* so that a single continent extended from Arctic Siberia to New Zealand. Northward, * Tethys' had probably sea-passage, of uncertain and perhaps varying width, to an Arctic Ocean, between 'Angara- land' and Scandinavia, one of the oldest and most massive corner-stones of the whole fabric. West of this again, between Scandinavia and Britain, a narrower strait extended far north, and perhaps reached the same Arctic Ocean. Beyond this, the rugged Caledonian highlands of Britain stood outpost on the eastern margin of a 'Laurentian* continent. The south coast of ' probably crossed the north Atlantic along the modern bulging then southward round the nascent Appalachian ch retreating northward near the Pacific coast of North ^A^S^c^ till it approached (or even joined) eastern * Aftgara^fflS'l^eyond the north Pacific, All north of this coastline seems €01 have been solid 6 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. land, with Greenland and Labrador at its core; but from time to time a wide lakeland covered the * middle west' of North AiAerica. Round these ancient shores, under the influence of solar heat, the general planetary circulation of winds and sea-currents played then as now. The resulting climates however were different, by reason of the shape of the sea-basins, and the altitude of the land- masses. In particular, the long trough of 'Tethys/ lying wholly in north temperate atid subtropical latitudes, and landlocked towards the north from Mexico to Scandinavia, served — like the Medi- terranean of to-day, but on a vaster scale — to mitigate and aspimi- late in an exceptional degree the climates of its foreshores, and still more those of its islands. For though most of 'Tethys* was open water, a lar/*e region between north Africa and Scandinavia was broken by large is- lands, ruinous fragments of continents older still, like Scandinavia itself, and the Caledonian highlands, Snowdonia, atid the Malvern and Mendip Hills, imbedded in the margin of *Laurentia.* One such forms now the plateau core of Spain and Portugal; Sardinia, Corsica, Elba, and the rugged *toe? of Italy are peaks of another, which we may call * Tyrrhenia*; the Caucasus, the Bohemian high* land, the Ardennes, are others, round whose skirts old shingle- banks and other shore deposits replace the clean limestones characteristic of the greater depths. So early in the history of the planet was the site of our European and Mediterranean region con- spicuous for its abnormalities, and its juxtaposition of old and new. The * tertiary* period of crust-history, which is still in progress — for the term * Quaternary,1 signifying those recent phases when Man's presence can be demonstrated, is a needless concession to self-esteem — is characterized, like its * primary * and * secondary * predecessors, by vast readjustments of the crust, breaking up the Laurentian and Indo-African continents, and crumpling the cre- taceous sea-bed of 'Tethys* into a series of elevated ridges* These folds result; from two series of lateral stresses. The one, thrusting outwards from Angara-land to east, south and west, has caused a series of southward-bulging * arcs' (like the rucks in a tablecloth when a heavy book is pushed across it) which define the present continent of Asia* Such arcs form the half-submerged island*- chains, Aleutian, Kurile, Japanese, Lu-chu; the grand sweep through Burma, and the Malay peninsula with its insular pro- longation to the Moluccas; the Himalayan range and the Hindu- Kush; the Iranian arc which traverses Baluchistan, south and west Pefsia, and Kurdistan; and further west, the Tauric and Dinaric systems which bound respectively Asia Minor on the south* and I, n] TERTIARY MOUNTAIN-BUILDING 7 the Balkan peninsula on the west,, as far as the head of the Adri- atic. Then follows the southward and westward-bulging Atlas range, and its prolongation into south-eastern Spain. Within these outer arcs rise other folds obviously concentric with them, most easily recognizable in north-eastern Asia, and behind the Hima- laya, but perceptible also in Iran and northern Asia Minor. Be- tween the folds, lie less crumpled areas, at higher or lower levels. The plateau of Tibet stands now at over 1 5,000 ft., the Tarim basin at over 3000 ft., and the core of Asia Minor at about xooo ft. above the sea; the Behring, Japan, and China Seas, on the other hand^have bottom at 12,000—9000 ft. down; the Gulf of Oman at 6000 ft., and the southern lobe of the Caspian at about 2000 ft* Similarly, outside each greater arc, the margin of old Gondwana- land*has been forced down and under, in the Bay of Bengal, in the Persian Gulf, — where the whole of Arabia has been tilted like an ill-laid paving slab — -and in the eastern basin of the Mediterranean, where the north African foreland has been fractured stepwise, so that, while the Libyan shore is beset with quicksands, the greatest depths are off the Peloponnese and Rhodes. The folds of the other series result not from southward but from northward thrusts, and overhang similarly sunken * forelands,' this time on their northern side. Examples are the Altai range between Mongolia and western Siberia, the Caucasus, and the whole Alpine series, — Balkans, Carpathians, Alps, and Pyrenees. The course of these European folds is complicated by several factors, chief among which is the presence of those older lands already mentioned, both north of the Alpine folds, in Bohemia, the Black Forest and Vosges, and the Auvergne, and within the folded area, as in Spain, *Tyrrhenia,' and Hungary; the stubbornness of which has not merely accentuated the transverse amplitude and overfolding of the ridges themselves, but has compressed them lengthways into the 2, -shape presented now by the Carpathians and Balkans and caused the spiral distortion of the Pyrenees, Alps, Apennines, Atlas, and the Spanish-Balearic arc. Finally, local relaxations of these strains brought about the collapse of whole regions of the crust, either parallel to the trend of the folded arcs, or transversely. Examples of longitudinal subsidence are the Black Sea and southern Caspian, carrying away both ends of the Caucasus and another great segment of mountain range between the Crimea and the Balkans : another is the Adriatic, nipped between the Dinaric arc and the Apennines* ^fraft&Verse fracture and collapse are illustrated, within the mass of 013 Angara- land, by the long * trough-fault' or 'rift yall^' df the Red Sea, 8 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. which Is prolonged between Crete and Rhodes right across the junction of the Dinaric and Tauric arcs, submerging the Aegean archipelago, and breaking down a shattered trough through Macedonia and Serbia to the Hungarian plain. A branch of this same rift forking west across the Dinaric folds depressed the Gulf of Corinth; another diverging eastward further south forms the Gulf of Akaba, the Dead Sea, and the trough of Code-Syria, and may be traced far athwart Armenia. All these are only classical examples of the main types of crust movement to which the tertiary transformations of old * Tethys' are due. Crust-movements of such amplitude occupied a vast period of time. And all the while, rainfall and frost were^denuding and dis- secting the land surfaces; rivers were transporting the debris, and depositing it in lake basins and coastal seas; limestones and fharls were accumulated in deeper waters; and at times along the lines of severest distortion and fracture, volcanic matter was discharged molten from beneath* Principal stages in this tertiary derangement of what had been the cretaceous sea-bottom of Tethys may be summarized as fol- lows. Their importance for us, over and above their contribution to the actual distribution of land and water, of mountains and plains, is that in conjunction with the changes of climate resulting from such rearrangement of lands and seas,, they have restricted or extended the regions which this or that type of vegetation could occupy, and the range of the animal forms which such vegetation fed, and so contributed in due course to localize and differentiate the main varieties of Man* Foldings and upheavals of the old sea-floor began earliest, as they have since reached their greatest amplitude, eastward in the heart of Asia, where the Himalaya, Kuen-lun andTienshan ranges, with the plateaux of Tibet and Mongolia uplifted between them, intervene between Angara-land and the * Leirmrian* sub-continent, of which only fragments soon remained, represented by Madagas- car and peninsular India. Elsewhere too during this stage there was widespread exposure of the sea-floor; especially along that east-and-west axis of upheaval which eventually becomes the 'Highland^Zone* of western Asia and southern Europe. Anl without being elevated, many of the remaining sea-basins dried up altogether, leaving vast deposits of salt and gypsum, like those which are forming now in the waste heart of Persia, Renewed submergence followed, from the westward ocean* as far south as Kordofan, and as far east as Khorasan* But the Hindu- Kush and Iranian arc barred off for ever from Tethys its old south- J, n] FIRST PHASES OF A MEDITERRANEAN SEA 9 ward gulf; and a mere bulging of the African continent cut off the eljj uf "depression in western Sahara from what we may now begin to call the Midland Sea; for it is the first phase of the Mediterra- nean of to-day. But the fauna and flora of the lands which were appearing now along the line of the Alpine folds were still essenti- ally of such Indo-African type as had spread thither during the period of exposure. And such they long remained; for these lands were mainly Insular, and as the Laurentian continent still limited the Atlantic northwards not far from the line joining Newfoundland to Cornwall, the oceanic currents which bathed their ^shores maintained a subtropical climate, warm, moist, and equable. Furthe^ folding and upheaval of the western arcs extended and consolidated the mountain zone of the Nearer East as a long pro- montory connecting the high plateaux of Asia with these mid- European islands, and these again with the British promontory of Laurentia, along the very ancient line of folding represented by the Ardennes and the Mendips. The result was to bisect the Mid- land Sea into a southern or 'Mediterranean' and a northern or *Sarmatian* basin, which henceforth have separate histories until almost modern times. A further result was that the sinuous Apen- nine-Atlas ridge encircled a -'West Mediterranean' basin, which though it communicated usually with both the Atlantic and the East Mediterranean, was occasionally cut off from both, and in late Miocene times was so much reduced by evaporation that none of its deposits of that age are now above water level. There was therefore ample communication between the new mid-Europe and the Moroccan lobe of the old Africa* The East Mediterranean long retained much of the character of its predecessor the Midland Sea, The highland arcs along its north border included Crete and Cyprus; the Adriatic had not yet SLink outside these arcs, nor the Aegean within them. The moun- tains of Media and Elam were still very imperfectly developed, and the Arabian slab of Gondwana-land had not yet been frac- tured or even tilted under their stresses. The southern border of this sea lay therefore far to the southward across Africa, from a Moroccan Gulf, south of Atlas, to Abyssinia, Hadramaut, and the mountain ridge of Oman; with an easterly gulf extending far Into Iran* It was separated however from all seas to the south-east, as its marine fauna show, by the ridge already mentioned connecting the Asiatic with the African continent. Occasionally disconnected from the Atlantic by elevation of the lands round tb^ttfestern basin, it underwent repeated phases of evaporation; stud Indo-African io PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. plants and animals still occupied its northern margins^ leaving their remains for example in Sanies and Attica, "* The northern or Sarmatian Sea had a similar though separate history. It extended repeatedly far east> to lake Balkash and the foothills of Altai and Tienshan, and far north round the base of the Urals, an old ridge accentuated by the same tertiary stresses as the mountain-zone which bounded this basin on the south. Caucasus was sometimes insulated, but usually formed part of its southern margin, with only gulfs or lakes outflanking it south- ward. Westward communication with the Atlantic was interrupted earlier, oftener, and more completely than in the Mediterranean area, thanks to the growing intimacy between Mid-Europe and those ridges and stacks of old land which we have seen ^embedded in the Laurentian foreshore. Between the rising Alps and- the Bohemian and mid-German highlands a long gulf remained^ or in high-and-dry periods a drainage basin which we may already call 'Danubian*, but the strong northward and outward bulge of the Carpathians eventually cut off these lowlands and the sunken Hungarian basin, to form inland lakes. An outlet through the Iron Gates to the Pontic basin cannot be demonstrated till later. As the land-masses of Mid-Europe and also of North Africa and Western Asia increased in extent, the climate of the whole region became drier: the c Sarmatian' sea shrank into a * Pontic* scries of lakes, connected only by flood channels, if at all, but including then a region so far to the south-west as the present north Aegean* Eventually one of the deep fiver valleys, which dissected the ex- posed Sarmatian sea-floor, cut back into the high ground in the re-entrant angle between the Carpathians and Balkans, opened a new outlet for the waters of the Hungarian and Bavarian basins already mentioned, and created the Danubian drainage system. In this period also a long trough, faulted across Mid-Europe^ deter- mined the upper basin of the eventual Rhine, though it was long before this lakeland was tapped, like the Danubian, by a river cutting back from the north through the old Taunus highland from Coblenz to Bingetl, The same period of uprise and continental climate affected the Mediterranean also. The rising escarpment of Media and Elam cr.t off its Iranian gulf, which became silted, first with river deposits* then, as its waters evaporated, with a crust of salt and gypsum. And as the folded escarpment rose, very steep and lofty, the foreland in front of it to the south-west was forced down and under, till the great quadrangular slab which we call Arabia was snapped off" from Africa> and tilted bodily, downwards at the foot I, n] TERTIARY FLORA AND FAUNA n of the new Zagros range, but with a free broken edge upreared to wqptwatJrd, and long troughs of dislocation and subsidence between itself and the African continent. The Red Sea trough opened for long into the Mediterranean, like the Nile trough to the west of it; but was closed at its south end by the main ridge from Asia to East Africa. Further north, the same fractures crossed the Mediterranean floor, so that the free edge of the Arabian slab, or rather the de- tached strip of it which forms the Lebanon range, was thencefor- ward the eastward limit of that sea. The movement, violent as it appears in retrospect, was however gradual, and progressed from south to north so that the drainage basin formed on the tilted slab remained Connected with the Mediterranean through North Syria, and fee Jordan valley, lying in a smaller and earlier rift than that of the Red Sea and for long a tributary of a great river system of north-eastern Africa, still contains species in common with the Nile and the Euphrates* But, in time, Mesopotamia too became a separate basin like Iran, accumulating its own river sediments, and in dry periods its beds of salt and gypsum. The gradual coherence of new land-masses where the Tethys basin had been, and the restricted communication between the remaining seas and the Atlantic, affected the climate of the whole region profoundly and adversely, and the fauna and flora were modified accordingly. Surviving representatives of the first occu- pants of tertiary Europe are now only recognizable in the Malay zoological region, and to some extent in tropical west Africa. For our present purpose we need only note that it Is in these two regions alone that the great anthropoids, gorilla and orang-utan, survive; that it is certain that various monkeys, and probable that creatures ancestral to Man, were among these * Malayan * occupants of mid-Europe; that the most * simian* varieties of Man himself, the dwarfish, heavy-jawed, and long-armed Negritos, have a similarly discontinuous distribution surviving only in central Africa, in Malaya and beyond, and to sorae extent ijp^sputhern India; and further, that the only creatures really intermediate be- tween these and the anthropoids, are Pithecanthropus from a deposit considerably later in Java and the Broken Hill skull from Rhodesia, It is not without reason, therefore, that search has been made for human handiwork, even in eocepe and miocgae beds. But^tjbfii; * eoliths* collected in Belgium from miocene deposits h^^etiibt yet been generally accepted as such: those from graves ililing the sides of the present Nile valley are rather betft^jrtS^te'd, but must still be viewed with reserve. ia PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. As the climate became less favourable, the Malayan fauna gave place in the north-west to characteristically * African' types, which persisted in the new European (or rather *Eurafrican') region until the close of the Pontic stage. Then, rather abruptly,, and very widely, this * African' fauna was itself replaced by new forms, dis- tinctly 'Arctic/ advancing apparently from that Laurentian con- tinent which had existed all the while west of Britain, and prob- ably had extended also far eastward beyond Scandinavia as the Sarmatian sea evaporated; since similar 'Arctic' forms can be traced penetrating Asia too, as far as its Himalayan crest* The * African' withdrawal was of course gradual and unequal ; typical forms sur- vived in Bessarabia, for instance, later than elsewhere, and Spits- bergen and Greenland still had magnolias and plane-tpees during the Pontic phase. Within the folded zone, especially, there were secluded regions favourable to the survival of the old warmth- loving forms. The progressive folding of maturer mountain-ranges, and the development of more recent folds, such as the Apennines and the Jura, accentuated this subdivision of the north-western or * Eurafrican ' land-mass. A fresh period of submergence follows, probably due to relaxa- tion of the folding stresses, and collapse of ill-supported blocks, The British promontory of Laurentia, and probably the whole southern seaboard of that continent, began to give way, so that the Atlantic ocean, which had long ago been extended southwards from Tethys to the Antarctic, spread northwards now into cooler latitudes. Since some of these sinking areas — for example, the Aquitanian region of France — adjoined the west Mediterranean, Atlantic marine fauna had access once more to the Mediterranean basins. As the water surface of these seas increased, the climate became moister, and the weathering of the highlands and main river valleys more destructive. It is in this period that the drainage systems of the lower Rhine, the Seine and other rivers of the English Channel, the Loire, Garonne and Guadalquivir, and the Wady Draa, south of the Moroccan Atlas, were established, in deep-lying gulfs; the Rhpne is another example of drainage con- sequent on such subsidence. Further east, the Aegean depression, already noted, began Qo admit Mediterranean waters to basins hitherto belonging to the Pontic lake-system. The Pontic region, too, receiving ample rain- fall once more, regained its old continuity from the Carpathians as far as lake Baikal. The trough-valley of the Nile was being opened, as we have already seen, as far south as Assuan, and re- ceived copious drainage from the high west edge of the Arabian I,n] SEVERANCE OF AFRICA FROM ASIA 13 slab; the first fractures had begun along the line of the Red Sea, and the remains of considerable lakes in the Dead Sea and Orontes region suggest similar subsidences further north. The Mesopo- tamian basin was by this time quite cut off from the Mediterranean by the tilting of Arabia, though the barrier from Lebanon north- ward was of no great width or height. When the tilting movement came to a crisis, and Arabia broke away from the African con- tinent, the Red Sea trough opened first as a gulf of the Mediter- ranean. But the foundering of the next block south of Arabia admitted the waters of the 'Indian1 Ocean into this gulf from the south; 'and a similar inbreak through the Hormuz strait converted the Mesopotamian lake, which had formed along the sunk eastern edge of Arabia, into a ' Persian gulf of the same southern ocean, extenfiing all along the foothills of Zagros, and also towards Anti- Taurus, and Anti-Lebanon. In this fashion, while the Mediter- ranean remained limited eastward almost at its present shoreline, the whole region between it and the Iranian plateau became almost wholly separated from what remained of old Gondwana-land, both in peninsular India and in east Africa, with a new and narrow isthmus, twice constricted, at Suez and north of the Lebanon, instead of the old broad land-avenue from Iran to Abyssinia. The consequence of this separation will be seen to be of the utmost importance, when we consider the distribution of Man, and of the modern fauna and flora generally; for it is with the severance of Africa from southern Asia, on the one hand, and the replacement on the other of ' African * plants and animals north of the Mediterranean by northern forms from Scandinavia and the Laurentian foreshores about Britain, that the modern period of tertiary time may fairly be said to begin. It will be evident from what precedes, that by this time not only Europe but the whole north-west quadrant of the Old World land-mass had been shaped approximately to its modern propor- tions : only the precise distribution of sea and lake over the shal- lower hollows in its surface being liable to shift, according as either the land rose or sank locally, or the supply of moisture varied over its landlocked basins. The broad features of this large group of regions, the eventual home of the 'white races* of man, may there- fore be summarized in modern geographical terms. It consists, essentially, of the Alpine 'folded highland/ whose structure ai^ conformation we have been tracing, bounded both northward^Scl southward by abrupt outward slopes overlooking depr^€^i btit undisturbed and level * forelands/ Included wi^^f'^iife ' folded region are numerous plateaux more or less ^le(vat©dy and more or 14 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. less buried under later sediments- And westward;* where the Alpine folds fade away towards the foreshores of the new Atlantic Ocean, and around the British remnants of Laurentia, now de- tached,, there is a * continental shelf of varying width, and liable to moderate oscillations of level. Three main regions are therefore to be distinguished here: (i) the Highland Zone itself; (2) its Northern Foreland, from the North Sea to the foothills of Tienshan and Altai, with its south- eastern half liable to be submerged in *Sarmatian* or *Ponto~ Caspian * lakes; (3) its Southern Foreland, from Morocco to Mesopotamia, continuous and undisturbed at a fairly high average level in latitudes remote from the Highland Zone; more broken and depressed further north, till its fractured slabs sink beneath the waters of the Persian Gulf, the east Mediterranean, and the lake region of southern Tunis. As already described, it is by no geo- logical accident that the west Mediterranean basin lies north, not south, of the Atlas folds; within the Highland Zone, that is, not adjacent to it like the eastern basin, The later history of these three principal regions must be traced separately, if only because the altitude of the Highland Zone has long been sufficient to give it a markedly cooler and moister climate than either of the Flatlands; so that its greater rainfall has sculp- tured it very deeply, and wrought upon its surface abrupt and complicated scenery of mountain and valley; the varied rocks thus exposed contributing directly, and still more (by their detritus) indirectly, to accentuate local differences in the soils and eventual flora of each drainage area. As its limits lie obliquely from north- west to south-east between latitude 50° in central Europe and 2 5° in south Persia, the larger changes of climate have affected its main regions serially from one extremity to the other. Its uplands have been sufficiently continuous at most periods to permit the spread and withdrawal of consecutive types of vegetation; yet the deep engraving of its passes has permitted the transmission of comparatively lowland flora from one basin to another* And what is evident for vegetation applies equally to all animals which are susceptible to changes of climate and food supply* This Highland Zone3 then, may conveniently be regarded, Aa its main characters, as a single geographical region. Frequently and for long periods, it has been a promontory based on central Asia, or a long isthmus, connecting a south-eastern continent with a wide and old land in the north-west. At all times its upland conformation, moister climate, and denser forest vegetation have secluded it from the Flatlands on either flank. In so far as there I, n] EURASIAN AND EURAFRICAN FLATLANDS 15 has been interaction. It has been the Highland which has had the initiative; because in periods of excessive moisture it has been from the" foothills of the Highland that forest has spread over adjacent plains; whereas in periods of drought, the extension of steppe con- ditions into the foothills has been retarded by the residual rainfall around the heights. Only by glaciation, It would seem., could the Highland vegetation be devastated from within, and even so under the most favourable conditions for reoccupatlon from the less frost- bitten highlands continuous with it to the south-east. And as we shall see in due course, such glacial devastation did actually occur, between the close of the pleistocene period and the beginning of our own. North of the Highland Zone lies the Northern Flatland. It is alm*t featureless from Altai and Tienshan to the Baltic and North Sea; except for the narrow transverse fold of the Ural range, which however fades away southward before reaching latitude 5*0°. But beyond those almost accidental depressions of its western margin, which form our 'narrow seas,' this Flatland Is limited by two considerable mountain-masses, Scandinavian and British, of great age and stability; and beyond these to the north-west ex- tended formerly a long arm of that old Lauren tlan continent which still encircled the north Atlantic, long after it had ceased to occupy it; and it was probably the subsidence of this Laurentian land (represented now by the * Wyville-Thomson Ridge, * on the ocean floor from Britain to Iceland and Greenland) and the circumstance that the breach of continuity lay west and not east of the Scandi- navian and British mountain ranges and involved general redistri- bution of currents between the Atlantic and Arctic Oceans, that determined the profound changes of climate to which allusion has already been made (p. 12). Of secondary importance are the minor oscillations which deter- mined whether the northern parts of the Flatland, east and west of the Ural divide, should be above or below water; and thereby assigned to the remainder, and to the whole north face of the Highland Zone, a climate either moist enough to fill the Sarmatian depression with lakes or a sea, or dry enough to exhaust this re- servoir and reduce the whole northern Flatland to a cold desert as Inhospitable as the hot desert on the south side, to which we tutu now. The southern or Eurafrlcan Flatland Is almost as simple ,ift its main features as the northern, and far more uniform in defSalL As this region lies within the planetary trade-wind b$l&$!i£tiS devoid of abruptly folded ridges which might precipitate ra,lia^— except the 16 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. Atlas to the north, which belongs to the folded Mountain Zone, and the highlands of Nigeria and Abyssinia, — it has always been less well watered than the regions north and south of it, which are moistened respectively by the westerlies and the equatorial rain- belt. As no cause seems to be known which could displace the equatorial belt of perennial rainfall and dense forest., to any con- siderable degree, the normal result of a pluvial or glacial crisis in the northern hemisphere has been to contract this trade-wind belt and its desert regime; and conversely. The only other important variable affecting the Southern Flatland has been the greater or less extent of submergence, in the Mediterranean and Mesopo- tamia, mitigating or accentuating the dryness of northerly winds, The mitigating influence of the Atlantic has of course been per- sistent, but has been neither great nor far-reaching, after the dis- appearance of that old gulf or lake-basin south of the Ahaggar plateau after the miocene period. It should be noted however that in periods of greater rainfall both this plateau and the Tassili and TIbesti uplands further north and east,, have attracted sufficient moisture to feed large rivers, running some southwards to the Niger, others northwards into the Mediterranean, by the Wadi Irharhar and the Tunisian Schotts* Direct land-contact between the southern Flatland and the Highland Zone is interrupted for the middle third of its length by the persistent water-surface of the cast Mediterranean basin, last remnant of old *Tethys'; and again far eastward, by the Gulf of Oman and the Mesopotamian Gulf, formerly much larger and wider than now. Between these two sea-barriers, outer ridges of the Tauric arc radiate south-westward and southward into north Syria and Cyprus, and this highland prominence is continuous southward with the upstanding edge of the great Arabian slab and detached fragments of it, as far as the peninsula of Sinai, forming a causeway along which migrations of momentous im- portance have occurred repeatedly. In the west Mediterranean the Atlas range, which must always be regarded as being geo- fraphically continuous, as well as structurally, with the ranges of icily and Italy, and also of south-eastern Spains has the west Saharan Flatland along its steep southern face; but the continuity of the Eurafrican land-mass here Is qualified by the depth, and usual submergence, of the west Mediterranean depressions. Only at either end of this western basin have there been intermittent land-bridges from Atlas; north-eastward through Sicily to the . Apennine arc, concentric with the Alps and repeating on a small .scale some features of the Syrian causeway; and through Spain, an I, n] AFRICAN9 AND 'ARCTIC' FAUNA 17 old highland comparable in size and structure with that of Asia Minor? to the broad coast-plains of the Atlantic seaboard north of the Pyrenees, It results from these northward avenues of the southern Flat- land, that there has been long intercourse between its inhabitants and those of the Highland Zone, at both ends of their long frontier; simple, marginal, and almost uniformly from north to south over the Syrian causeway; intermittent, complicated, oscillatory, and far-reaching, in the 'Eurafrican* west. Such oscillations and, no less, the general replacement of 'Afri- can' by * Arctic* forms of life throughout the whole north-west of the Old World, were caused, or at all events greatly accelerated, in the pleistocene period by the onset of a profound change of climate, very severely felt all over the new European sub-continent of Eurafrica, but by no means confined to this region; for the *Ice Age' or * Glacial Period7 of the Old World has its counterpart in the New, and even very similar sub-periods. There have also been 'Ice Ages* in the southern hemisphere, but there is no proof that they either coincided or alternated with those in the northern, and they had no known influence on mankind. With the northern, and especially with the European Ice Age it was otherwise. That the replacement of African occupants, on the other hand, was as gradual as it was, was due to the fact that the Ice Age was not continuous, but had its * interglacial* phases, which permitted African forms to return northward, and also allowed Asiatic species to move westward into Europe along the Highland Zone; and we shall see that this oscillation had profound significance for Man. For if we compare the earliest known distributions of the other primates with the actual distribution either of their modern repre- sentatives, or of the principal races of man, it becomes clear that whereas the four-handed, and also many of the four-footed mem- bers of this 'order* of animals, retained mainly arboreal habits, and consequently were withdrawn southward and eastward into Africa and Malaya, as the subtropical forests were restricted by the general change of climate, one intermediate variety, two-handed an4 two-footed, and thereby more able to accommodate itself to the accidents of life in the open, became so far master of its fate as to outlast the forest, and enter on a career of pedestrian adventure and manual exploitation. We do not yet know at what stages in this acclimatization to the parkland and grassland sequel of the retreating forest this biped primate achieved its three primary controls over its surroundings — control over dead matter, in the C. A,BM i8 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. shape of boughs and stones, prolonging the reach, and enhancing the force, of its natural hand-stroke; control over the wayward energy of fire, the scourge and the terror of all other animals; and therefore not only comparative security against carnivorous animals, but control over the fund of sustenance and energy supplied by animal flesh. But we do know already, from an implement-strewn surface of old land underlying some of the earliest glacial debris of East Anglia, that some sort of tool-using, and animal hunting * precursor' of ourselves ranged so far as this to the north-west before the climate was as yet quite glacial; and from similar indi- cations in the Nile gravels, and on the surrounding desert, that subtropical drought restricted him as little as subarctic cold. How far these early traces, or remoter relics such as the Trinil brain-case from Java, or the Broken Hall skull and other bones from Rhcdesia, may be connected with ancestors of any actual variety of Man we must consider in fuller view of the effects of the glacial crisis* III. THE GLACIAL CRISIS The causes of this Ice Age have been much discussed, and are still obscure: recent investigations lay greater stress on geo- graphical factors, such as the distribution of land and water, the elevation or depression of the region, and other circumstances favourable to intense snow-fall at certain places and seasons, than to those astronomical explanations by nutation of the earth's axis, or precession of the seasons, which were formerly popular. It is at all events certain that the severest glaciations occurred in periods of submergence, and that the repeated relaxations of glacial auster- ity coincide with greater exposure of land-surfaces, and with a continental climate drier rather than warmer, since dry air, how- ever cold, precipitates little snow; without copious snow there is nothing to feed a glacier, much less a continental ice-sheet; and under dry cold winds on the lowland the snout of the best-fed glacier shrinks rapidly by sheer evaporation* The same circum- stance goes far to explain why the main ice cap of the Old World lay so far towards its western edge, exposed to wet westerly winds off the north Atlantic which as we ixave seen had only recently attained its modern extent. In the same ways the evidence for ex- tensive glaciation on the mountain ranges of Caucasus, Armenia, and especially of Central Asia coheres with that for a wide water surface in the Ponto-Caspian lakeland, and for submergence of western Siberia, Of such glacial maxima there have been recognized three in I, m] THE FOURFOLD CRISES OF THE ICE AGE 19 most parts of France, and four on the north side of the Alps and Py^ene&s, and in north-western Germany, followed by two oscilla- tions during the final retreat over those districts which lay nearer to the principal snow-caps. The second, or * Mindel, * spell in the Alpine series (corresponding with the later part of the first, further north) was the severest; submergence was deepest, temperature lowest, and the Scandinavian ice sheet widest, covering all but the south coast of Britain, and meeting the glaciers of the Alps (while the Rhone glacier, for instance, extended to Lyon) and those of the Carpathians and Urals so that their margins, like the glaciers of the Caucasus, bordered and replenished the Sarmatian sea. Outlying ice-caps, mainly of this phase, have been traced on the Pyrenees, Apennines, and Dinaric and Tauric chains; in Armenia, Zagr8s, and the north Persian ranges; and over the whole moun- tain knot of the Pamirs and Hindu- Kush, from its Sarmatian shore to an ocean-gulf which flooded the Punjab. Over these vast areas, therefore, all life was obliterated temporarily, and round their margins and interspaces was reduced to sub-arctic desolation. There is strong reason for believing that the climatic oscilla- tions of the whole north-west Quadrant synchronized and formed part of a single great planetary episode. Not only is the fourfold glaciation of north-western Europe repeated around the Alps and represented in a fourfold * pluvial' sequence in the Nile Valley; but the glacial maxima represented by deposits in Nebraska, Kan- sas, Illinois and Wisconsin respectively, though not necessarily contemporary, seem to repeat the relative intensity of the Gunz, Mindel, Riss, and Wtirm maxima of the European Ice Age. It is therefore permissible to treat as standard the southern Flatland, where there would seem to have been least bi*each of continuity in plant and animal life, and interpret the more broken and complicated sequence in the western and the northern Flatlands, and also in the Highland Zone, by reference of their main episodes to the principal stages in the south. Before dealing with the human occupants of these regions, and their redistribution during and after the Ice Age, it is convenient to note briefly the effects of any such crisis on the distribution of anTmals and plants, partly because these effects can be more fully illustrated, partly because it was in response to changes in his animal and vegetable surroundings that man's first human efforts seem to have been made. It follows directly that in any displacement of climatic zones the corresponding flora and fauna were displaced accordingly, with due allowance for peculiarities of soil or configuration which either 20 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. permitted the maintenance of any elements of such plant and animal associations, or accelerated their retreat. In this connexion, it is important to observe the normal sequence of the types of vegetation; round the margin of perennial snowfield or ice sheet* frozen treeless 'tundra* with transitory herbage after the spring thaw; then dwarf birch and stunted pine, passing to coniferous forest, and through this into mixed deciduous forest; oak, beech, and nut-bearing trees such as chestnut and walnut predominating in succession. Forest however may be interrupted, on soils un- favourable to trees, by other types of vegetation; on loess, repre- senting ancient deposits of wind-blown dust from adjacent desert, by precarious steppe or grass-land; on limestone, by the treeless turf of chalk-downs or wolds, owing to the withdrawal of surface water by underground channels; on ancient and imperviouslrocks, especially where these adjoin a wind-swept seaboard, by the dry bitter heather and gorse of moorland. Marshland too, and the travels of river valleys, have their special * plant associations/ >rming open glades between the forests which clothe the higher ground. This normal sequence is of course retarded also locally by altitude, which increases -rainfall, and reduces mean temperature. Parnassus for example has pines above its olives and buy-trees, and alpine flowers above its pines. Further south, in the 'Mediterranean' type of climate, with wet winter and rainless summer, deciduous trees give place to evergreens, and tall forest to thickets or shrubs; and as drought and warmth increase, even shrubs stand further apart,, in an under- growth of tough resinous bushes, and spring-flowering bulbs, annuals, and grasses* Eventually grasses, halfa-rush, and spiny leathery camel-fodder predominate, until they too fade out before drifting sand and sun-tanned rock. As climate becomes milder, the zones of vegetation move north- wards, and uphill; but as trees take centuries to mature, the shift of vegetation may lag behind that of climate* On the other hand, adverse shift of climate rapidly destroys the less hardy plants, for they cannot retreat and only acclimatize slowly; more .mobile forms of life, such as the larger animals and man, will cither follow their habitual food-plants or maintain themselves in aust<frer climate by change of diet, by growing winter-fur, by taking shelter in caves, or, in man's case, by appropriating the hides and fur of other animals, In an oscillating climate, therefore, such as that of this Ice Age* recurrent necessity offered exceptional stimulus to invention. It is man's inertia, rather than any initiative, his obstinate reluctance I, iv] THE PRINCIPAL HUMAN RACES 21 to abandon a mode of life once adopted, his recourse to any compromise — * rather to endure the ills we have., than fly to others tha? we know not of* — and, in the result, his unique ability to conquer Nature by reasoned conformity with Nature's ways, that differentiates him from all animals but those, such as horse and dog, in which he has apprehended and elicited faculties remotely analogous to his own, IV. THE PRINCIPAL HUMAN RACES We are next concerned with the human stock, or stocks, which occupied these regions before and during the Ice Age. It has been noted already that the geological evidence points to prolonged geogttiphical severance between the plateaux of Central Asia, with their vast folded mountains and their eastward and northward forelands (including the whole of ancient Angara-land), and all that westward prolongation of the folded zone, with its forelands, which we have been discussing. This geographical severance at the narrow and almost impassable neck of high land where the Hindu Kush intervenes between Afghanistan and the Pamirs has its human counterpart in the segregation of the ancestors of the yellow-skinned, straight-haired Mong-oloid stock from all westerly varieties; for although anatomical evidence of its ancestry is not yet collected, enough is known, as we shall see (pp. 48, 59), about the slowness of the development of human types (for example, in pen- insular Europe) to justify the belief that this ancient seclusion of central and eastern Asia, lasted none too long for the differentia- tion of a kind of man so well-marked physically and even men- tally, In the same way, the correlation of the black-skinned, woolly- haired stocks with the Malayan fauna, which is suggested by their actual distribution, would seem to postulate a period of time comparable with that suggested above for the Mongoloids, within which the no less highly-specialized negroid physique could be developed from a precursor more widely distributed, especially north-westward, and presenting those features in which both the negroid and the white stocks differ from the yellow. On the northern slopes of the Asiatic core the supply of moisture during the Ice Age brought the Altai glaciers down to 6000 ft* from sea-level, far lower, that is, than sufficed to close all avenues from central Asia to the lowlands of Siberia and Tur- kestan. On the Himalayan side, monsoon winds from an Indian Ocean which covered the Punjab and Bengal, furnished snow 22 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. more copious still, and moraines are found as low as 3000 ft. But though the comparatively narrow neck of high land between rthe valleys of Indus and Oxus was wholly beset by its ice-cap, it is not necessary to suppose that within the great plateaux of central Asia there was perennial snow, or a wholly uninhabitable region. Rather the vast accumulations of loess, the deposit of countless dust storms, suggest a 'continental' climate with wide variations, and the possibility of at least seasonal occupation by fleet grazing- animals, such as the horse. It is indeed to an intimate parasitic connexion with such an animal "host/ in some siich circumstances, that we have probably to ascribe the highly specialized type of man characteristic of this region now. The yellow skin-colour of Mongoloid man gives him protective camouflage in sandy desert and dry-grass steppe; the structure of his straight wiry half, and its rarity except on the scalp, suggest adaptation to a continental climate; while its extreme length in both sexes serves to disguise the characteristic profile of the human head and neck, and approxi- mate it to that of a quadruped seen from behind. From the rather prominent jaw combined with globular brain-case may be inferred long habituation to some food which minimized the pull of the jaw muscles on the side-walls of the skull; and the only rood which fulfils this condition is milk and its products, on which nomad Tartars still live almost exclusively: the absence of face-hair, the short concave nose with spread nostrils, the peculiar infantile lips, the wide flat face and obliquely set eyes, are adaptations we should expect if for ages this milk was absorbed direct from the udder; and the short legs of some Mongoloids, and poor development of the calf-muscles in all> suggest that, like Tartar infants nowadays, the parasitic proto-IVtongol sat tight upon his host between meals, and shared its wanderings. On the steppes of glacial Europe, man hunted and ate the horse; if we suppose that in central Asia, during the same and perhaps in long earlier periods, he made friends with him and lived upon his friendship, we seem to have a clue to the paradox of the emergence of a highly specialized breed of man from a region which had been for a very long time so little suited, except on these terms, to sustain him at all. The absence of rfhy widespread relics of such occupancy explains itself on the same hypothesis* Men who did not hunt or fight, had no more need of coups-de-poing than of supra-orbital ridges or a fighting-jaw^ such as characterize the negroids or the 'Neanderthal' type in Glacial Europe, As they must travel with their animal hosts or perish, they had no choice but to desert their ailing relatives when they I, xv] SPREAD OF MONGOL MAN 23 fell behind; Interments therefore are not to be expected, nor a group-psychology which sets much value on human life, or gives out-let to futile emotion. Almost inhuman in his normal apathy, the Mongol can display almost equine savagery when provoked by panic or ill-usage. The development of so peculiar a type presupposes not only a large continuous region? of appropriate physique, but also com- plete seclusion. The high plateaux had supplied the former for a very long time, since loess-land is so inhospitable to trees or shrubs that wide oscillations of climate only affect the density of its vege- tation without changing the quality. Seclusion has been assured by the great altitude of these plateaux, the ruggedness of the sur- rounding ranges, and the dense rain-forest of their monsoon-swept outvtfftjrd slopes. While therefore it has been exceptionally difficult for alien folk to intrude, it has been relatively easy for Mongol man to emerge, on one of two conditions — either that he parts company with his milk-giving host, and takes to hunting, as has happened in the north-east, or to agriculture, as in the south-east of Asia; or else, if he is to retain his nomad pastoral habit, he must wait till the climate has become so dry that Jhere are clearings of grassland through the forest belt. Even then he can only proceed so far as he finds grassland still in front of him; and this has only happened at two points: to the west, through the great avenue between Altai and Tienshan, and to the north-east, down the valley of the Hoang-ho; and even here it only happened far on in post-glacial time. It would be beyond the plan of this chapter, to discuss in detail the subsequent spread of Mongoloid Man through the Asiatic foreshores of his plateaii-home; but his western and north-western expansion has so profoundly influenced the course of history in the modern world, that it is necessary to trace at least the outlines of them, so far as they can be recognized; and also to make quite clear their upward limits in time, which appear to be very narrow. The older drainage of the southern and more elevated plateau, south of the Kuen-lun ranges, issued to the south-east, towards what is now the Malay Archipelago, but the Brahmaputra, cutting b^lck through the eastern Himalayas, where they have been inter- sected by the great Malayan folds, has captured the southernmost of these drainage areas; the Hoang-ho similarly has captured the northernmost, and the Yangtze the majority of those which lay between, leaving only a small remainder to feed the Sal wen and the Mekong. Consequently the main avenues of human movement have long been towards the eastern lowlands,, and the vast alluvial 24 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. area deposited by Chinese rivers, thus reinforced., has received and acclimatized most of the human overflow from the interior* m From the northern and less elevated plateaux of Mongolia, however, the older drainage was mainly north-eastward; and here owing to the conformation of the eastern arcs, the eventual recipi- ents have been on the one hand the Amur, on the other the north- ward-flowing Lena and Yenisei. Here the continental core of old 'Angara-land/ which is embraced between these two rivers and has been an immemorial reservoir of ancient forms of life, has also formed the * asylum" into which have descended successive types of flora and fauna discarded from the plateau-margins in successive periods of austerity. The human population here, so far back as it can be traced, belongs to such discarded fauna, and is consequently Mongoloid, but of far less specialized types than those whichchavc never left the plateaux. It is from this Angara reservoir, and the mountain arcs which prolong the north margin of the plateaux and encircle them eastwards, that the whole north-eastern promontory of Asia has received its human population; and similar types, essentially yellow-skinned and straight-haired,, have passed on through it to Alaska and the New World, Westward, the long-continued submergence of the Siberian lowland from the Yenisei to the Urals prevented all expansion until very recent times: and the present belts of tundra and forest vegetation are post-glacial. As far west as the longitude of Moscow they are of east Siberian origin, and it is only here that this Siberian forest meets mid-European forests advancing in the opposite direction; so that there is overlap of competing species, with a slight balance of advantage on the side of the eastern types. The importance of this is that with the forest, and its animals,, man has spread also, from east, as from west; coalescing in the same longitude as the species of trees. And over and above the disputable evidence of hybrid physique around the line of coalescence, the Mongolian antecedents of all groups cast of that line are betrayed by the fact that they have the reindeer domesticated, and do not hunt it, as Redskins do? and as did the men of the glacial west so long as wild reindeer survived there* Quite distinct from all this, and representing a very much lat<£r phase of redistribution, is the exodus from the western gate of Mongolia. Here, about latitude 45°, the roughly parallel ranges of Altai and Tienshan stand (on an average) two hundred miles apart; the descent by this avenue onto the Kirghiz steppe is easy and manifold; the head waters of the Irtish have already cut back into the plateau, and an earlier affluent of the Sarmatian sea once I^iy] AFRICAN FAUNA AND AFRICAN MAN 25 did the same, through the gap east of Lake Balkash, But this avenue only becomes passable under a special conjunction of cir- cumstances; the Kirghiz steppe, which all lies below 1500 ft., and much of it below 600 ft., must be neither submerged nor sand- swept; yet the avenue itself must be free of snow-cap and conver- gent glaciers; the forests on the outer slope and in the passes amist be discontinuous enough to permit pastoral nomads to pass with their flocks; and thirdly, there must be sufficient inducement to leave the plateaux at all. Obviously there is not here any large margin between one set of obstacles and the other. Moreover, ex- cept when the Sarmatian sea-floor is exposed, the Kirghiz steppe itself leads only to the Urals, where progress is barred again by forest. It is intelligible therefore that over long periods this western avenfae was not open for man; or if traversed at all, it served rather to admit western hunters from the steppes, or foresters along the foothills, than to let out the pastorals of the high plateaux; and the actual mixture of races all along this edge of the plateaux sug- gests that for a long while, and very widely, it was the west that was the aggressor, as indeed its cultures would lead us to suspect. This summary outline is enough to show what seems to have been going on in the Asiatic continent which bounds the North- west Quadrant on the east. Its significance is that so far as can be seen, High Asia and its characteristic type of man remained utterly secluded from the North-west Quadrant until post-glacial time, and may be quite left out of its history. We have next to deal with the African region which adjoins it on the south. This African region, like the core of highland Asia, consists of ancient and stable land, on the northern half of which cretaceous and subsequent limestones have been laid down without serious disturbance over an area which has gradually diminished during tertiary times. In the north-west the multiple Atlas ranges belong to the Alpine folds, not to flatland Africa. Eastwards the con- tinuity of this vast flatland has been broken, as we have seen, by the sunken troughs of the Nile and Red Sea, as the Arabian slab was tilted and detached. Similar depression and tilting in front of th.e Tauric and Dinaric arcs submerged successive long strips of the north margin to the Mediterranean sea-floor, but the greater part of the Libyan flatland stood fast, and the Cyrenaic plateau was even forced slightly upwards, Here there has been oscillation, even within historic times, for the harbour of ancient Leptis is high and dry now, whereas at Cyrene the sea has invaded the Greek theatre, With no barriers due to configuration, the distribution of plants 26 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. and animals over this large area closely follows the climate. Equa- torial rainfall, resulting as it does from the general atmospheric circulation,, may confidently be assumed as a permanent factor' of strictly limited range, and has probably never extended much farther north than latitude 20°. As the only really high ground is in Abyssinia., far to the south-east, the effects even of the present monsoon winds are minimized, and moreover, before the nearer sections of the Indian Ocean subsided, there was no reason for these winds to blow so far west at all. There has therefore been nothing since cretaceous times to interfere with the normal se- quence of trade-winds and westerlies over all northern Africa; and the only calculable effect even of the Scandinavian and Alpine glaciation would be to shift each of these zones southward towards the equatorial rainbelt, and narrow them both. The distribution of plant and animal life lay regularly therefore, as now, in zones of latitude: tropical forest in the south, passing through parkland into steppe and desert, and thence through steppe into evergreens followed by deciduous and coniferous forest, and sub-arctic moor- land and tundra. In the days of the early tertiary archipelago, the trade-wind zone was submerged, and there was therefore no desert; tropical plants and animals of old * Malayan' type flour- ished northwards almost to the Arctic circle, and those of the modern * temperate* zone were represented only in the interior of Laurentia, With the emergence of the western Sahara, and of the mid-European peninsula, 'Malayan' types were restricted to the Tropics, and replaced by * African * like those of the modern savannah region. On the establishment of a Mediterranean sea and European sub-continent north of it, * African* types were restricted in their turn, and replaced by * Arctic* forms from Laurentia; which have their counterpart in the modern flora of temperate North America, and are still fringed on the Atlantic and Mediterranean seaboards by the cLusitanian* remnants of genera widespread in America* Of the human associates of this pre-glacial vegetation we have no direct evidence from Europe; but the modern human type which characterizes the zone now occupied by the restricted * African* fauna, is the negroid, both in Africa itself, and (as tfate aboriginal type) in the present Malayan region, and among the * African * fauna (with its lion, tiger, and elephant) which has fol- lowed the 'Malayan' into southern Asia. The Broken Hill skull, from a deep bone-deposit in a Rhodesian cave, was found associ- ated with a distinctive * African* fauna, and is reported to display no general character subversive of this statement* I, iv] SEQUENCE OF HUMAN TYPES IN AFRICA 27 But here a distinction must be made. Among the vast majority of * African * (that is to say * negro ') men, the prominent carnivor- ous-looking jaw is accompanied by a markedly long-shaped skull, giving purchase to the powerful jaw-muscles and itself compressed by them. This, like the deeper blackness of the skin, has been commonly regarded as a special adaptation to 'African* zoological conditions; for other types survive isolated, not only in the heart of equatorial Africa and of the Malayan region, but far to the south where the edges of negro-land reach the Limpopo swamps and the Kalahari desert; types which though generally negroid, are of abnormally small stature, inclined to steatopygy (an abnormal development of superficial fat, especially among the women) and general hairiness, and with a yellowish or leathery tinge in their blackness, and a far less long-shaped head than either the standard negroes of Africa, or their * Malayan' counterpart in Melanesia. The trans-Malayan counterpart of the Bushmen, Vaalpens, and Strandloupers of South Africa is now easily recognizable in the Tasmanians1. That these types are ancient, and that they were already associated with the * African" fauna before it disappeared from Europe, is rendered probable, first, by the occurrence of negroid individuals along with north-western or Eurafrican races in palaeolithic deposits at Mentone, and in carvings palaeolithic and later; by the survival of a pygmy type into early neolithic times at S chaff hausen; by the frequent steatopygy of late palaeolithic and also of neolithic statuettes; by the representations of similar types in neolithic Egypt; and by other traces of a far wider distri- bution than now, in Africa itself. Besides the very long head of the standard negro type other characteristics, such as high stature and great physical strength, the more purely black pigment, the woolly scalp, the lack of body-hair, the prominent heel and slender calf, and the everted lips, may be regarded (like the more striking pecu- liarities of the Mongol type) as secondary adaptations to a highly special regime — in this case the tropical rain-forest, during the restriction of the * African* fauna to its eventual range south of the desert belt. Analogous local adaptations of a genetically * African' Type, associated in its geographical range with survivals of an * African' fauna, may be regarded as sufficiently accounting for the 'oceanic' negroes; for the negroid 'Dravidian* survivals in 1 There are no doubt other factors to be taken into account in tljese cor- relations, such as the build of the skull-base and the spinal calumny iail that is attempted here is to illustrate analogies, which might be multiplied, between the remoter races of the two regions in question. 28 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. southern India and beyond; for the ancient descriptions of * Asi- atic Ethiopians' in Mekran and in the extreme south of Arabia, around the margin (that is) of the sunken regions of Indo-Africa; and for the curious survival in the Mediterranean, and even in France and Britain, of types which combine certain characteristics of negro and of white man without any of the common marks of the half-breed, If an anthropologist were required to indicate an extant type of man to illustrate such common characters, he would choose the widespread and loosely interconnected group which includes the aboriginal elements of the population of Ceylon and peninsular India, and a long series of remnants further west; through southern Persia, and parts of southern Arabia, merging in thje darker- coloured and slighter built elements of the mixed 'Hamitic' p<3f>u- lation of north-eastern Africa, and in a superficially similar strain which is perceptible among outcasts and derelicts of the Mediter- ranean region and recurs as far afield as the British Isles, though here there are few precise observations yet. Summing up the relations which have existed between the negro and the white races on the African continent we reach the following result. The climatic zone represented by the Saharan desert, though it has varied in width, has been maintained long" enough to serve as an impermeable screen between the negro and the white stocks, except along a narrow coast belt fringing the Atlantic, and perhaps in the Nile Valley. Only the rare 'negroid* individuals in the palaeolithic caves of the Riviera suggest that during exceptional northward shift of the climatic belts, African man may have reached south-western Europe, temporarily and in small numbers: though others would explain these facts not by northward incursion of ready-made African * negroids/ but by the former presence, along the whole length of the region immediately south of the Highland Zone, of 'dark white* types such as those already mentioned, We have thus reconstituted, so far as it is known, the earlier distribution of the yellow-$kinned, straight-haired, round-headed Mongoloids, in the secluded upland heart of Asia; of the black- skinned, woolly-haired, and long-headed negroes of Indo-Africatf antecedents; and of the very indeterminate group of varieties which range from the Dravidian and other "dark-white* stocks to the 'poor-whites' of^the Near East and the Mediterranean, Having associated the peculiarities of their physical build, with the preva- lence of geographical conditions likely to give rise to them, we turn to the more complicated problems presented by the so-called I, IY] THE WHITE RACES OF MAN 29 * white race" of the north-west Quadrant, Here the criteria of statute, hair-texture, skin colour, and headform seem at first sight to fail us, in the medley of tall and short peoples; slim or thickset; blondes, auburn s, and brunettes; with all varieties of wavy or curly hair, and of florid or pasty complexions; with eyes brown, hazel, grey or various shades of blue; and with heads rivalling the average proportions alike of Mongol and Negro, and presenting besides very marked variation, in the height and contour of the brain-case, and in the modelling of face and jaw, In the long controversy which has been provoked by these anomalies, the following have been the principal turning points. Blumenbach selected a Georgian type from the Caucasus to illus- trate whaj he regarded as the embodiment of the qualities of the whftc race as a whole, and gave to the group a name the full appropriateness of which is appreciated only when it was realized what a medley of men is harboured in the Caucasus itself, Huxley insisted on the importance of the varieties of skin and hair, and distinguished within the whole group a blonde and a brunette section. Sergi recognized a closer structural relationship between the long-headed brunettes of the Mediterranean, and the long- headed blondes of the Baltic shores than between either of these and the broad-headed men of the Alpine zone; Bogdanof proved that the long-headed people of neolithic Russia and western Siberia belonged to the Baltic or * Nordic" type, not to the Medi- terranean type as Sergi had supposed, and were to be classed as blondes; Lapouge realized that the broad-headed strains, distri- buted through the mountain zone of central Europe, over an area tapering somewhat from east to west, and extending beyond this zone far into western Russia, into the Netherlands and Denmark, and into the south and east of Britain, originated not by local adaptation of various longer headed peoples to highland altitudes or other geographical conditions, but by the intrusion of a fresh 4 Alpine* race, anatomically distinct in its general build as well as in its characteristic head form. Ripley associated this European c Alpine' type with the great mass of even broader-headed varieties which occupy Asia Minor and the mountain zone eastward as far tLs the Pamirs. Deniker discriminated within this broad-headed complex, at least three brunette sub-types, the short thickset 'Cevenole' of central France and Savoy, the tall, well-propor- tioned 'Dinaric' variety of Dalmatia and Albania, and the very peculiar 'Armenoids* of Asia Minor, with their heads abruptly flattened behind; to which it was an easy corollary, that the blonde Alpines of north-eastern Europe had arisen by interbreeding with 30 PRIMITIVE MAN, IN GEOLOGICAL TIME" [CHAP. 'Nordic' blondes,, and his 'Littoral7 and "Atlantic ' types by similar interbreeding with * Mediterranean * brunettes. Keith distinguished between those broad-headed folk who entered Britain across 6ie North Sea, coining from north-eastern Germany, and those who entered across the Channel and originated west of the Rhine. More recently Peake has restated the evidence for separating alto- gether from any * Alpine/ that is to say south-easterly immigration, those broad-headed peoples, of northern Mongoloid descent who came westwards with the spread of the Siberian forest, round the northern edge of the old Sarmatian lake-land. It only remains before summarizing present knowledge, as heretofore, in brief narrative form, to note tentative identification by de Ouatrefagcs of the * Cro-Magnon type' of late-palaeolithic man with recent Berber and Guanche strains; the separation established by Scfiliz of* the old long-headed population of the Danube Valley both from the Nordiclong-heads of the Baltic area, and from the Mediterranean folk of the south-west, and his affiliation of it to the late palaeo- lithic hunting-folk; and Fleure's recent confirmation of the long- suspected survival, in the moorlands of central Wales, of a breed anatomically indistinguishable from the widespread * Aurignacian* type, of the same remote period. For the steps by which these main positions have been won, and consolidated into a realm of knowledge, reference must be made to current hand-books and the literature on which they are based, The problem of the * white races* is simplified in some degree by the severe glaciation of northern and central Kurope, which is the central event of 'Pleistocene * and * Quaternary * times; since the origin of the modern population of the glaciated regions is to be sought not in any general survival of earlier kinds of man within them, but in their reo ecu pation by plants, animals and men alike, from unglaciated areas. It is therefore only in these adjacent areas that questions of continuous descent can arise; and the actual distribution of the Mongoloid and Negroid varieties, and still more the reported occurrence of non-Mongoloid and pro-Mon- goloid remains on a number of sites around the fringe of South America, offer a strong presumption that the human species had already spread very widely before the glacial crisis deranged itft distribution. The close association of Negroid survivals with the discontinuous African fauna makes it certain, as we have seen,, that man accompanied this fauna before its disruption^ and probable that he was associated with it when it was still in full occupation of the North- Western Quadrant. Human remains do in fact occur with those of * African * animals, in numerous European deposits I, v] PALAEOLITHIC MAN IN THE NILE VALLEY 31 belonging to fairly early phases of the Ice Age; and the later and better 'attested varieties of * eoliths/ belonging to phases not long an*tecedent5 would be accepted by many people as evidence of a tool-using mode of life, if there were found contemporary traces of men who might have used them. In any case, the * Chellean' types of implements (p. 46), which are contemporary with the earliest human remains, are clearly not by any means primitive, but pre- suppose much experience in the improvement of handy stones, V. PALAEOLITHIC MAN IN THE SOUTH AND EAST The sequence of early forms of man and of his handiwork was first established laboriously and by comparison of many sites in western Isurope; and it is only recently that it has been realized that in the Nile valley we have a single continuous series of de- posits, outside the glaciated area, and free from its destructive austerity, but near enough to it to be affected by marked alterna- tions of moist and dry climate which can now be securely linked to the main periods of the Ice Age in Europe. With this clue to guide us, we can more easily seize the outstanding features of the European series, among their bewildering complexity of detail. The deep narrow Nile-gulf which was formed, as we have seen, in the pliocene period, across an otherwise featureless plateau, became at the close of that period a series of long lakes fed partly from the upper Nile, but partly also by considerable lateral streams whose gravel-screes, washed from the plateau surface during a period of considerably greater rainfall than now, contain chipped flints of * eolith* types; and such * eoliths7 are found on the plateau also. In a period of increasing subsidence and more abundant rain these lakes were gradually silted up by the deeply stratified Melan- <?/w>-beds, which overlie the lateral screes as high as 1 80 ft* above the present flood level. A rain maximum, which may be taken to represent the first glacial crisis in Europe, accelerated this silting, and made good hunting on the plateau for some kind of man, whose implements, of the Chellean* type familiar from interglacial gravels in west- ern Europe, are found both there and in the Melanop$is~bzd$. Breasted ascribes to this phase, on account of their deeply weather- stained appearance, certain earlier rock-engravings of animals and even of boats, on the precipitous edges of the plateau, A first inter- pluvial drought (representing an interglacial mitigation of the cli- mate of the north-west) terminated this silting; but the Nile stream, fed as now by tropical rainfall further south, continued to flow 32 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. over the dry lake-beds, and cut into them a deep canon. It should be observed here that any northward shift of the desert regime should be accompanied by some extension of the tropical rain belt, and probably also of the area affected by the monsoon rains, since the Indian Ocean was by this time as extensive as now; and that these rains, then as now, would continue to feed the Nile stream, however arid the climate of its lower valley. In this period of drought and erosion, man seems to have maintained himself on the dry lake-bed, for Chellean implements are found among those remnants of its surface which forms the 'upper-terraces' on either side of the gorge. Next, a second rain-age, corresponding with the second or 4 Minder glaciation of Europe, flooded the gorge and set the lateral torrents to work again; and as a rise in the sea-levelf like that which submerged much of the Atlantic seaboard, checked the main stream, a fresh series of gravels, analogous to those of the Somme and Thames, were deposited to a height of 90-100 ft. above present flood level. In the new screes, as well as on the old plateau surface outside the valley, implements are found, of the more advanced *Acheulian* fabric, showing that man became again ubiquitous as the region became refertilized; and in the early part of the second interpluvial pause, he spread once more, as in France and Britain, on to the gravel beds, and scattered imple- ments there to the margins of the new gorge which was being cut through them by the main river* Once again in a third rain-maximum, corresponding with the third or *Riss* glaciation, fresh gravels were laid down in this inner gorge, not so copious as the earlier series, but partly covering the recent lateral screes, and standing in some places as much as 30 ft* above modern flood level; since the actual valley has been eroded In them during a third interpluvial drought* And once again, man ranged over the surface of these gravels, and left his implements there, as well as on the plateau, where they He on the desert surface mixed with all their predecessors* Then follows the fourth rain-maximum, a comparatively mild one, corresponding with the fourth or 'Wtlrm* glaciation. The gorge, of which the bed lies not less than 60 ft. below modern flood level, began to accumulate the first deposits of the present alluvium; which are shown by borings to contain human imple- ments at nearly all depths. A first pause in the deposition of this alluvium, corresponding probably with the * lower forest' period around the North Sea and the Channel, allowed man to descend through the fens to the river margin, and accounts for the presence, I,v] CONTINUOUS HUMAN OCCUPATION 33 at a depth of 50—60 ft. not merely of implements but of rough fragmehts of pottery, and animal bones of domesticable if not domesticated species. As the rate of deposition since the thirteenth century B.C. aver- ages 4-08 inches in a century, this depth would represent a period of 15,000—18,000 years, assuming that the rate remained uni- form. But as there were certainly oscillations, and probably more rapid deposition at first, this estimate can only be approximate, and should perhaps be reduced. A second and a third access of alluvium, corresponding with the lower and upper peat-moss periods in Europe, and separated by ill-defined pauses, has raised the flood plain to its present level, at which it covers not only the edges of th§ last-eroded gorge, but part of the valley floor of the fourtS rain-maximum between the 'lower terraces* already men- tioned, which in some places now rise only about 20 ft. above the flood plain. It will be seen from this sequence of events that there is every reason to believe that the Nile valley, and the margins of the desert plateau on either side of it, have been occupied by man continu- ously, though with varying density of population, at least from the beginning of the pleistocene period. Wherever any of its succes- sive land surfaces remain in the valley itself, his implements have been found representing successive stages of skill analogous to those of western Europe; and on the surface of the plateau* which has been exposed continuously, implements of all periods are found indiscriminately, and constitute a more nearly uninterrupted and graduated series than anywhere else. The only serious gap, inevitably, is in the period immediately preceding the settlements on the present alluvial surface, because this alluvium is in process of deposition, and its encroachment, since the last interpluvial pause, on the surface of the lower terrace, has been burying the sites and tombs of immediately preceding phases, As human occupation has been thus continuous and (in the more fertile intervals) widespread, and as there is nothing in the physique of the earliest known inhabitants of the alluvial surface to suggest that they have been of anything but the local variety of EiJrafrican man, it seems probable that the arts of life represented in all these deposits are of indigenous, or at least quite local de- velopment. The last period of alluvial aggression has however been a very long one, and while the pauses in it may be presumed to represent phases of greater drought than now, the periods of more rapid deposition should be interpreted conversely as periods of moister climate, and consequently of less complete isolation from C. A.H.I 3 34 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. the comparatively well-watered and fertile region of Palestine and Syria, where a similar though not yet so perfect sequence* of im- plements is being found. At these phases therefore allowance mxist be made for the possibility of intrusions of Palestinian and Syrian man, anticipating that which is known to have occurred when the so-called 'Gizeh' type entered and dominated Egypt in early dynastic time, And the gradual dilution of this alien type on that occasion imposes caution in assuming,, from the approximate purity of the predynastic inhabitants of the valley, that no such intrusion had ever occurred earlier. The importance of such caution will be understood when we take stock of the predynastic culture, more fully to be described in Chap, vi, and compare it with the distribution of |ome of its chief elements elsewhere. First, the types of implements preserve almost without qualification the ancient technique of mere chip- ping and flaking. The grinding and polishing, characteristic of neolithic implements in Europe and along the Highland Zone, are employed only late and in a supplementary way. The flaking on the other hand exhibits a climax of unparalleled delicacy just before the first apparition of copper implements, in immediately predynastic time. This obstinate adherence to the flake-technique cannot be merely due to the abundant supply of suitable flint, in the rocks of the valley sides; for the upper valley exhibits a large variety of crystalline and volcanic rocks, and pebbles from these rocks are included in old gravels downstream. Considering the proximity of the large West Asiatic region of ancient and highly developed skill in grinding and polishing such pebbles, the persistence of the flake-technique in Egypt is therefore a strong presumption of technical isolation; and the rare occurrence of polished celts, of the fully formed neolithic types common In Western Asia, points rather to occasional trade than to local manufacture* This isolation is confirmed by the original and unparalleled sequence of the pottery-forms, at all events down to the point at which appears the * red-polished* ware which is common to Egypt, Syria, and Cyprus in immediately predynastic time. And the fact that so many of the Egyptian pot-forms seem to depend on those of vessels cut out, or rather ground out^ from hard stone, makes all the more remarkable that abstention from the grinding-tcchniqiie for implements, which has been noticed above* With the ovoid forms of many of these stone vases should be compared those of the perforated stone mace-heads, which likewise betray great technical skill in shaping and perforating refractory rocks, I, v] DOMESTICATED PLANTS AND ANIMALS 35 It must however be remembered that it is just at this period that thfi Nile-valley series is least continuous and complete. Be- tween the two main groups of the oldest remains (the refuse heaps and the burials beyond the advancing edge of the alluvium) there is a notable discrepancy. For from the moment when the population left their earlier settlements, descended into the fens, and began to domesticate cattle and practise agriculture, until the time when the fens were completely reclaimed, the dead were probably buried in the alluvium, or on its margin, and the earlier tombs are there- fore covered more or less deeply by the later alluvium. That this was so, is shown by one or two of the oldest known burial-grounds, which not only lie on the very edge of the present alluvium but have been proved to extend beneath it. Yet even these show phases of culture which are highly developed and in some respects already decadent; and throughout the long 'predynastic' period for which burials are available, there is further decadence, especially in the finer stonework. Moreover, even in the earliest known graves, objects of lapis lazuli, which must be of foreign origin, are found occasionally, and also objects of copper. Probably therefore the greater part of the purely neolithic stage of Egyptian civilization still remains to be disinterred from tombs on the valley floor beneath the recent alluvium, and from sites on old flood plains within the alluvium itself. Other arts of life, represented in the earliest burials, are the use of wattled huts, basketry, matting and vegetable thread; of leather and wood-work, and of bone, ivory, and shell for ornaments. Rouge and green malachite were used for paint. Agriculture is represented by flax, millet, barley, and wheat; of the latter grain, both the variety called emmer (Triticum dicoccum^ which is found wild in limestone uplands in Syria, and Moab, and in western Persia) was grown, and also the cultivated wheat (TrMcum vu/gare). Goats, sheep, and short-horn cattle were kept, all apparently of African varieties, and attempts were still being made even in early dynastic times to domesticate ibex, gazelle, antelope, deer, and other desert ruminants. There were domestic geese and ducks from the fens, and from early paintings it would seem that the os?xich was familiar, if not kept in captivity like the gazelle. The dog was known, and in early dynastic times there were special breeds for sport and other purposes. The only beast of burden was an African variety of ass. There was organized irrigation, and probably an ox-drawn hoe, the prototype of the plough, as it is depicted in early-dynastic hieroglyphs. The river was navigated in large house-boats; there was fishing with hooks of delicate 3 — * 36 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. flint work, and immemorial hunting on the desert and in the fens, for the prowess of chiefs was symbolized by a fobe^ of leopard-skin. That in fertile periods a similar neolithic culture spread widely westward over north Africa is clear from early Egyptian records, depicting the Libyans as pastoral folk, with herds of cattle, and asses; and from the survival, even now, of pot fabrics and basketry of predynastic technique and decoration. Connecting links between the palaeolithic series of the Nile valley and of western Europe are not yet numerous; but enough has been found, especially in Algeria, and around Gafsa in south Tunis, to support the conclusions drawn from the climatic regime, and geological confirmations of its effects, and from^he general character of the modern population, which includes Aurigifkcijin remnants like those of Plynlimmon and Dordogne, and is other- wise strikingly uniform with the older elements in western Europe. Eoliths have been recorded around Gafsa and in Algerian quaternary beds; pre-Chellcan and Chellean types from Gafsa ; and the later deposits around Gafsa reveal a typical North African culture equivalent to the later palaeolithic of Europe. Rock-draw- ings from Algeria resemble those in the later French and Spanish caves. The area of this 'Capsian' culture (so called from the ancient name of Gafsa itself) seems to cover all northern Africa as far south as the oasis of Ghadames; it passes over westward into the later palaeolithic of Spain and southern France, and extends east- ward into Syria, And the continuity, now abundantly evident for this later * Capsian ' culture, is indicated for earlier periods also by more scattered finds of implements of all fabrics common to the Nile and to west Europe, in the wide ill-explored area between Gibraltar and North Syria. We may safely assume, therefore, es- sential continuity of human occupancy of this region from before the first pluvial period, qualified only in its extent by the climatic oscillations, which cohere, as in Egypt, with those of the European Ice Age. We may conclude also that all interglacial ebb and flow of human types from the south towards the Atlantic seaboard was essentially the marginal expansion or contraction of this large Eurafrican region. Of the physical characters of Burafrican nfcn we learn more at present from the remains on European sites, than from the ill-explored areas further south; in Egypt, unfortunately, no such remains have been found before the time of the predynastic graves, which belong, as we have seen, to the latest alluvial phases* Before turning however to the palaeolithic series in Europe, the question confronts us ; what was happening east of the Nile valley, I, v] EARLY MAN IN SYRIA AND ARABIA 37 and south of that section of the Folded Highland which affronts so abnfptly the Levant and the Persian Gulf; namely on that large Arabian flatland which lies dislocated and tilted askew between great Africa and greater Asia, and along the Palestinian isthmus which connects its north angle with the Highland Zone? In its earlier stages, as we have seen, this flatland was itself a part of Africa: and the great fractures which determined the geography of the Red Sea and the Nile valley did not wholly break this con- nexion. Both at the northern and the southern end of the Red Sea, there had been frequently continuous land, and in dry periods this sea shrank through evaporation, as the Dead Sea has shrunk now. But the great slab of Arabia itself, tilting steadily under Iranian fold-stresse^s, became structurally secluded behind its abrupt western escarpment, and offered to its occupants an independent, if rather restricted career. Until the comparatively late disruption of the Hormuz Strait, the waters of the long Mesopotamian lake — an Adriatic of the Nearer East — restricted the land area of this peninsula, and mitigated its climate. The dimensions of its east- ward-flowing drainage systems testify to former fertility; and we must probably conceive it as having long enjoyed a regime not unlike that of peninsular India, with Lebanon and Bashan playing the part of the Ghats. That it was inhabited by man, with a palaeolithic culture re- sembling that of north Africa, is proved by implements from Sinai, Palestine and Phoenicia, the only districts which have been sufficiently explored; they range from CM ouster ian* types onwards; that is to say, from at least the third pluvial maximum. Of the sequence of physical types, we know nothing: provisionally it may be assumed, from the sequence of artefacts, that if * Neanderthal' Man ranged over this region, as is suggested by the * Mousterian ' implements, he was extirpated, as in the west, by men of generic- ally Eurafrican stock; for the actual inhabitants, though far from uniform, are in essentials akin to their western neighbours. It would be natural to expect some traces of the 'Grimaldi7 negroids of the Riviera caves, and of the *poor white' strains (already men- tioned) which are common to the Atlantic seaboard, the Mediter- nfihean, and peninsular India; and superficial observation supports this; but there has been no accurate survey as yet. All that can be stated at present is that a modern population, of generically Eurafrican stock, shows larger local modifications than the present uniform regime would lead us to expect; and it may be inferred from this, that with more copious vegetation the main drainage areas were formerly better secluded, and permitted such differ- 38 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. entiation. The most Important of these local varieties is a com- paratively broad-headed type in the extreme south of Arabia; but there is no evidence as to its antiquity here, and it may only result from intercourse in historic times with trading centres in north Syria, which as we have seen is an ancient dependency of the Highland Zone. Such are the physical circumstances in which was fashioned one of the most notable of human stocks, the Semites of Arabia. Physically they are akin to their 'Hamitic* neighbours beyond the Red Sea and throughout Eurafrica, and strongly contrast with the men of the Highland Zone who have spread southward along its Syrian projection, or overflowed from time to time along the margins of the tilted slab. Culturally they have been habituated for long ages, like the Nile-plateau folk, to alternations'of moisture and drought, which however never seem to have permitted any extensive growth of forest, except on the monsoon frontage to the south-east, nor, on the other hand, ever to have extinguished the grassland vegetation entirely. They are therefore typically grass- land folk. They have domestic animals of their own, goat, camel,, and ass, all native to Arabia; the sheep, and eventually the horse, have been acquired by them from outside; in both cases from the north. They have a remarkable type of linguistic structure, re- motely shared only by the Hamitic group, which lies nearest to them, otherwise; and a temperament and outlook more coherent and persistent than that of any other of the greater races. In the moister spells, such people multiply over the widening grassland more rapidly than any alien can habituate himself to pastoral life, or to precarious agriculture c between the desert and the sown/ On the other hand, in spells of drought, Arabia erupts like a volcano, pouring floods of highly organized and mobile tribes across its land frontiers north-eastward, northward, and across the Jordan rift into coastland Syria and Palestine, perchance even into Africa. There has been percolation also, more insidious, but of wide effect, across the Red Sea, and especially its southern strait, where the transit into Africa is shorter, and timber for boats is more avail- able. Such periodic exodus of Arabian tribes can be traced back inferentially to the third millennium at least; and it need not tee supposed that the earliest recorded movement was by any means the first. The predynastic regime of Upper Egypt, for instance, seems to be partly due to such a movement crossing the Red Sea to Koseir, and reaching the Nile at Coptos by a trail which can be followed now. And it has already been hinted that the old popu- lation of the Nile valley may have been so supplemented even I, vj THE PALESTINIAN ROAD FROM THE NORTH 39 earlier. The physical resemblance between Arabian and Eurafrican man is* however close enough to make detection difficult, even if eafly evidence were found. See also pp. 182 sqq^ 193, 254. Allusion has already been made to the prolongation southwards of spurs from the Highland Zone through North Syria, to form with the high western edge of Arabia a continuous highland causeway along the abrupt eastern margin of the Mediterranean, and then along the Gulf of Akaba to loftier and steeper escarp- ments fronting to the Red Sea. In structure most of this causeway is Arabian, but its exposure to wet winds from the west has given it a Mediterranean climate and a considerable rainfall; it is the 'good land beyond Jordan, flowing with milk and honey/ which tempts the nomads of Arabia in all ages, yet has never acclimatized them to itsfclf. For the vegetation is partly old African, with tropi- cal survivals still in the hot moist jungle of the Jordan gorge; partly Mediterfanean, spreading along the coast plains and sea- ward foothills; but always mainly Asiatic, reinforced, ever since the junction of highland causeways above mentioned, from the Highland Zone at its north end. Of its earlier human occupants we have little but a few Mousterian implements; but in the first neolithic culture in Palestine the people are of the highland breed; they burn their dead; and their implements, pottery, and other equipment are in strong contrast both with everything Egyptian, and with the grassland influences which predominate later. Arabian man has occupied the 'good land' again and again; but the moist air seems to be fatal to him, and many of the peasantry of south Palestine are hardly to be distinguished from their neolithic pre- decessors, But the Palestinian complication is not the only one which qualifies the homogeneity of Arabia. Very ancient interaction of the streams which furrow the south face of the Highland Zone has reduced its drainage systems in this region to three. The Cilician rivers, trending south-westward, have created a secluded alluvial foreshore on the Gulf of Alexandretta, peopled, in all ages, by tribes who have come down from the mountain region inland, or, more rarely, have landed from oversea. The Tigris, flowing smith-eastward, is joined below Mosul by the two Zab rivers from the Median highlands in what was once another such foreshore, at the head of a Mesopotamian gulf; but it could only attain its eventual importance when that foreshore spread along the foothills of Zagros and merged with the similar deltas of the Diyala, and eventually of the Kerkhah and Karun further south again. The latter even now has a separate mouth west of the Shatt el-Arab. Here again, as in 40 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. Palestine and Cilicia, vegetation and other occupants spread outwards from the valleys, and coastwise, as these delta foreshores encroached on the gulf. But between the head-waters of the Tigris and those of the Ctlician rivers., one southward stream, Euphrates, has cut back deeper and further than its neighbours and intercepted not only the original headwaters of the Tigris, between Malatiu and Diarbekr, but far larger areas of old westward drainage as far as Erzerum and the slopes of Mount Ararat. Thus reinforced, Euphrates has excavated, in successive periods of elevation and copious rainfall, a wide and deep valley, well-watered and fertile throughout, athwart the sunk north-eastern slope of the Arabian slab, reaching the gulf formerly at el-Der, later at Ana, and (at the beginning of the modern phase) at Hit, where it cicscemis a last terrace of solid coast-line almost to the present sea-level. To plants, animals, and people of the foothills and the Syrian parkland, this long fertile valley has always offered sustenance far out into the steppe and desert which it traverses, By this emphatic frontier of the Euphrates channel, a roughly triangular area of southward sloping plateau — -a miniature Arabia. about as large as Ireland — is marked off from the Syrian and north Arabian plateau, and has never wholly been rejoined in history. This is Mesopotamia, the 'land between rivers/ for the Tigris delimits it no less clearly eastward from the foothill country below the Zagros ranges. Its structure is continuous with that of Syria; its climate is essentially the same, giving it (at present) desert and steppe regime in the south, and parkland nearer the hills. Its only river, the Khabur, rises in the highland, where it threatens to behead what is left of the Upper Tigris at Diarbckr; but till this happens its drainage-area is not sufficient to give It geographical importance; except that where it falls into the Euphrates close below el-Der, its delta forms a cultivable plain opening on the main valley. It will be seen at once that like Syria to the westward, Mesopotamia forms a region of transition, occupiable from the highland north of it, as far as its parkland extends at any given period; but offering wide steppe-pasture to any nomads of Arabia who may succeed in putting their flocfeS across the Euphrates, These then were the geographical and economic factors down to the time when the present sea-level was established* and the Euphrates delta, propagated south-eastward from Hit, began to coalesce with those of the Tigris and the Diyila round the Meso- potamian gulf-head, which then lay between Baghdad and Samarra, I, v] SUMER AND AKKAD 4I We might compare an immature Lombardy with, the Ticino pre- paring to join deltas with the Po. 'What has followed, while the joint delta pushed its alluvial steppe and dense fen-margin seaward over the 550 miles which separate Hit from the modern coastline., is disputed, and must inevitably be obscure. As in Egypt, the population, human and other, of the alluvial flood-plain may be presumed to have been derived from the shores of the gulf as it silted up. But these shores, as we have seen, were themselves peopled from different sources; the deltas of the eastern torrents, from the Zagros foothills; the Tigris banks, with sparse but continuous offshoots of the occu- pants of its upper valley; the fertile bed of the Euphrates, with similar elements, longer segregated however from their highland and parkland ancestry. Arabia, on the other hand, established a longer and longer land frontier with the growing flood-plain, as happened to Libya while the Nile trough was being silted up; it overflowed this frontier with its own aborigines, wherever steppe conditions were established; and this Arabian element became more important as two conditions were fulfilled; first, as the north- ern part of the delta between the main rivers, and behind its ad- vancing fen-frontage on the gulf-head, was assimilated in climate and vegetation to the steppe of southern Mesopotamia; second, as the main Euphrates stream took a more easterly course (as it eventually did), leaving a larger expanse of alluvium from Kerbela southward undefended by any considerable water-channel against Arabian immigrants; and this, too, nearly opposite the point where intercourse is easiest with the comparatively hospitable Nejd oases in the heart of the peninsula. On the other hand the establishment of an important bifurcation of the Euphrates threw a new channel across to the Tigris near Baghdad, and interposed a fresh obstacle to nomad intruders from Mesopotamia into what we may hence- forward call by its historic name of Babylonia, or by the older names of Sumer and Akkad, its principal sub-regions, which differ slightly in accordance with their respective situations. That under these circumstances the southern or Sumerian half of the growing delta should be more exclusively populated from the foothills of Za|>Tos, and that the northern or Akkadian half should show greater affinities with Arabian and Mesopotamia!! people, would seem to be inevitable, and is generally admitted. It is however unnecessary, in view of the geographical antecedents, to attribute all such northerly or westerly affinities to the earliest Semitic migration of which there is historic record. The same factors had been co-operating already for a long time. 42 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. VI. THE ICE AGE IN THE NEAR EAST We have next to see how this region of the 'Two Rivers/ and the sections of the Highland Zone adjacent to it, northwards and eastwards, were affected by the glacial crisis. As evidence is at present scanty, conclusions must be more general,, and a wider survey will best bring out the most essential points. As far as the head of the Adriatic,, a single series of events has been reconstructed in greater detail for all western Europe, and to this we shall re- turn later. East of the Adriatic, information is less copious, but the main course of events is fairly clear. The Carpathians, Dinaric ranges, and the Thracian mass of Rhodope were heavily snow- capped, and glaciated locally, and similar conditions prevailed on the coast-ranges of Asia Minor, both north and soutfi. Caucasus and Armenia, rising to greater altitudes, were glaciated more severely; and Lebanon, flanked by the Mediterranean on the one side, and with the shores of the Mesopotamian Gulf not so far off as now, on the other, had an ice-cap exceptionally heavy for its latitude. But a large part of Asia Minor probably remained fertile and habitable throughout. What is far more characteristic of this region, as of the Aegean depression and the Hellenic promontory, is the severely pluvial denudation, accentuating the rugged high- lands, and smothering the foothills in vast sheets of gravel and sand. In these the rivers cut fresh gorges during the drier intervals, which were also periods of emergence and consequently of longer and steeper gradients. The older drainage of Asia Minor had been longitudinal, towards the Aegean subsidence; sections of It are recognizable In the headwaters of the Euphrates, Halys and Iris; and the great westward avenue past Afium-karahissar probably represents its main outlet seaward* But later upthrusts of the west end of the Tauric arc closed this outlet, and converted the central plain into a lake-land, where some salt and gypsum were deposited as in Iran, But this had all happened in prcglacial timesj and the subsequent development of Asia Minor was differ- ent. For it was an immediate result of the subsidences already mentioned in the Black Sea region^ to accelerate erosion in the torrents on the new north coast, and two of these, Sangarius afiid Halys, cutting back clean through the Paphlagonian range*, drained the greater part of the central lake-land, and kept all Its floor fresh and habitable except the small central basin or Lake Tatta, The present drought is recent, and in part remediable; even in the fifth century B.C. the district west of the Halys was 'richest in sheep and corn of all known 1ri tads' for Herodotus. I, vi] GLACIAL CONDITIONS IN ARMENIA AND IRAN 43 Further east, the Armenian ice-cap extended at times almost to the plateaux, east and west; but in milder intervals there was con- tinuous highland country, full of small plateaux, glacier-fed gorges and lake-basins, from eastern Asia Minor to western Iran; prob- ably even at the worst some sort of corridor by way of Sivas, Kharput, Diarbekr, and the "Upper Tigris; while the triangular uplands of North Syria, between Adana, Damascus and Mosul, do not seem to have been glaciated at all. Even now, though the water-surface of the Persian Gulf has been greatly restricted, the deflection of the jo-inch rain-line north-eastward beyond the Lebanon reveals an exceptionally moist and equable climate, and associates this margin of the Arabian slab with the highlands to the north, and with the Mediterranean sea- boatfl, rather than with the rest of Arabia. It is no wonder that we have record of elephants in one of these Syrian valleys as late as the twelfth century B.C., or that the Macedonian veterans of Alexander the Great made here their most enduring settlements. But the well-marked ridge which is followed by the caravan route from Damascus to Palmyra makes the transition from parkland to steppe rather abrupt; and as long as the Lebanon retained any con- siderable ice-cap- — and it was certainly glaciated severely — there was little or no communication between north Syria and the south. The spread of Highland Man into Palestine (p. 39) was probably quite post-glacial, The course of events further east has been less easy to discover. Through the extension of an * Indian Ocean * along its southern margin, and through the re-establishment of the Sarmatian sea on the north, the climate of Iran necessarily became moister and more equable than it had been while its salt and gypsum beds were accumulating. Allowing always for its more southerly latitude, and ampler size, we may compare its geographical position with that of Asia Minor now, between the Levant and the Black Sea; and as it retains many elements of its old Indo-African fauna, there cannot have been any such climatic break as occurred further west. At most the salt and gypsum-covered waste in its centre has been larger or smaller, and more or less occupied by lakes; and i£S present drought is consequent on the quite recent shrinkage of the Sarmatian sea. During the glacial crisis, its high marginal ranges were snow-laden, but not severely glaciated except in the north-west and north-east; the diluvial thaw was consequently not very destructive; and as the main basin was never tapped by in- ward-cutting torrents, like Asia Minor (though some intermont basins in Zagros have begun to be drained by streams flowing 44 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. south into the Persian Gulf) its local reserve of moisture was only slowly dissipated, at the cost however of greater ultimate Salinity. East and west of it, however, the ice-caps of the Hindu Kush and Armenia, where the marginal ranges converge, isolated this region no less completely than did the seas to north and to south; for until the diluvial debris became continuous along the western frontage, the Persian Gulf extended, as we have seen, at least to the point where the Tigris emerges from the foothills of Kurdistan; while raised beaches of it have been traced far west of the Euphrates, All these regions therefore were habitable during the greater part of the Ice Age, and there are Chellean implements on the sur- face in Iran and Arabia, and in gravels containing mampoth bones on the Caspian shore, to show that they were inhabited widely by Man. The diluvial thaw, however, brought disaster here* As was natural so far south, it was very rapid, once the cold crisis was over; violent torrents seamed deeply the superficial sediments of the Arabian slab, and spread masses of debris, among which the older rivers followed uncertain courses, like the Oxus later on the Sarmatian sea-floor. Then, to diluvial rains succeeded drought and drifting sand, before any grassland, still less any forest regime could be established, sufficient to disintegrate this debris and ac- cumulate soil. In the foothills of Zagros similar torrents, descend- ing more abruptly, spent their diluvial energies within a narrower radius; so that the eventual course of the Tigris skirts and even erodes their fan-shaped screes of gravel Rapidly at first, and after- wards more gradually, the northern part of the old Persian Gulf was filled up by these converging deltas; while further south finer sediment accumulated with proportionate speed. There was up- ward earth movement, too, after long subsidence, for at Hit the Euphrates has cut down to an older shore line, and its rapids are now wearing through the sill of this* The disastrous effect of this diluvial phase was to eliminate Mesopotamia as a focus of post- glacial culture, and to postpone effective occupation till an alluvial area had been created beyond it. The contrast, in every respect, with Egypt on the one hand, and with the Po valley on the othfir, is complete,, and of historical Importance, I, vnj EARLY MAN IN EUROPE 45 VII. THE ICE AGE IN EUROPE We return now to peninsular Europe, west of the Adriatic and the Black Sea. Here the fourfold Alpine maxima of the glacial period corresponds as we have seen, with the pluvial maxima of the Nile, and with repeated glaciation, fringed by diluvial rainfall, of the Balkan lands and western Asia. Earlier study of the river gravels and caves of the Atlantic seaboard, from Britain to Spain, and most of all in northern and central France, has recently been supplemented by research along the north side of the Pyrenees, and in many parts of Spain; among the caves of the central German and Bohemian highlands; in the widespread deposits of interglacial loess along the Rhine, in the Danube valley, along the Margins of the north German lowland, and beyond the Car- pathians, in Poland, and Ukraine, There is controversy still as to the perspective of the earlier human finds; the principal question being whether these are later than the third or * Riss ' glaciation, or go back Into the milder interval between this and the second or * Minder crisis, which was the severest and most extensive of all. In what follows, the longer intervals are adopted, in the belief that these accord more closely with the pluvial series on the Nile, In Europe, as in Egypt, 'eolithic' objects from preglacial deposits, have been claimed by some observers as human handiwork. In the light of the Egyptian material, which offers very similar forms, the probability that they are so is somewhat increased; but it is too early yet for an accepted verdict on most of them. In East Angiia however the presence of preglacial Man seems already secure (p. 18), On the longer reckoning above adopted, it is in the * first iiiter- glacial * deposits which preceded the boulder clays and moraines of the *Mindel* glaciation, that the earliest human fragment has been found, a lower jaw chinless but recognizably human, from a gravel-bed near Heidelberg. No implements have been discovered in this deposit. A more perfect skull of a quite different and more modern-looking type comes, together with 'eoliths' and rolled bones of subtropical animals, from a very early river-side deposit arPiltdown, near Lewes, above, though not actually with, which lay gravel with implements ruder than those next to be mentioned, but generally similar in type. The date of the Piltdown deposit is still disputed, but it cannot be later than the earlier glacial gravels of the Thames, and may be considerably earlier. Next in the * second interglacial' debris of the thaw-swollen Somme, and other west European rivers, occur numerous imple- 46., PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. ments, of the ruder fashion typical of the gravels of Chelles, and then in more skilful workmanship, at St Acheul. Both styles are chipped from natural nodules of flint3 so as to leave one end pointed, and the butt naturally or designedly rounded for grasping in the hand. These are but flood-spoil from camping grounds on the river banks, and tell little about their makers and users except that they haunted the drinking-places of the large African fauna whose bones are in the same gravels. Here, though we have their implements, we have at present no trace of the men themselves, It is only when the third glaciation draws on, and the African fauna were being replaced, except the woolly mammoth, by rein- deer, musk ox, arctic fox, marmot, and other Arctic animals, that man and his prey alike took shelter from the weathcj: in natural caves, and 'Mousterian* scrapers and borers, rudely fashioned by retouching the fresh edges of the flakes formed in shaping the 'Acheulian* coup-de-faing^ give a glimpse of the scraping and piercing of bones and hides during such sojourn, and the first hint of woman's knack of finding secondary uses for the waste from man's chase and chipping. The colder climate was enforcing the invention of clothes. In a Jersey cave of this period the hunting weapons predominated near the opening; domestic scrapers and hacked bones of animals further in; the remains of a child lay a little outside the entrance. In one French cave, an old man had been buried intentionally in the floor, crouched as so many savages sleep. This phase also has a wide distribution, from the south of England to Spain and Portugal, Algeria and Tunis, the plateau edge of the Nile valley and the Syrian margin of Arabia; eastward too through mid- Europe as far as Hungary, south Russia* and the Caucasus; and the style of the new implements Is still very uni- form. But outside these limits, lMousterian* settlements have not been found as yet, whereas Acheulian and Chellcan implements are common in South Africa^ and even further afield; and it is possible that it is to this period that we should assign a great divergence between human experiences within and beyond this ' Mousterian ' area, for reasons to be stated later (pp. 47—50). Like their implements, the men of this Older Palaeolithic Age are of very uniform type; and as this 'Neanderthal' type differs markedly from all subsequent varieties of men and also from the older but more modern-looking type represented by the Piltdown skull and some other finds of various early periods, it must be noted briefly at this point. Into the difficult question of its place in the human genealogy, this is not the place to go. Its distribution 48 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. definitely human chin. Well represented by examples from Britain, France, and as far east as Predmost in Moravia,, this is a^type in which not only "there are no salient features which cannot be matched among the living races of the present day/ but it remains the predominant element among the modern inhabitants of se- cluded districts such as the Plynlimrnon moorland in central Wales; it is common still in the west of Ireland, in the Dordognc, in Sardinia, about Guipuzcoa in Spain, and in parts of Tra-os- Monte5? and has been noted in the oases south of Algeria and Tunis, and among Egyptians, Somalis, and elsewhere in north-eastern Africa, The numerous earlier allusions to 'Neanderthaloid' indi- viduals in modern European populations probably refer to these Aurignacians, who look 'primitive* enough when contrasted with the majority of modern men, but are separated by almost, as -great an interval from the real Neanderthal type. From characteristics of these survivors it is possible to supplement the evidence of the early skeletons. The forehead is narrow, with marked hollows in the temples, above the heavy eyebrows; the orbits are long and narrow, the cheeks are high and broad. The nose is broad, the jaw prominent, and the chin rather weak; the stature is low, the carriage loose and ungainly, and the arms very long: the hair, eyes, and complexion are dark, and the whole body is very hairy. In some individuals, the resemblance to a common type of aboriginal Australian is well-marked; and the similarity between Aurignacian skulls in Europe and the prehistoric skulls from Lagoa Santa, in Brazil and other remote localities round the margins of South America, suggests that this type had once almost as wide a dis- tribution as that of the older types of implements. It does not however seem to have been recorded, as those have been, from tropical or southern Africa; and its extreme hairiness and the wavy texture of individual hairs distinguishes it altogether both from the Negroid and from the Mongoloid breed- With the occupation of western Europe, therefore, by Auri- gnacian Man begins a continuous series of events and material remains running on to modern times* There are moments in this series where continuity of civilization cannot be directly traced, but continuous descent is sure, and therewith continuity of tradi- tion, which above all other human characters engages the atten- tion of historians. Other breeds of man have intruded later, as we shall see, from the south-east along the Mountain Zone, and from the north-east as the Siberian forest extended towards the Volga basin; but in western Europe, Aurignacian man has never been wholly superseded, and still forms coherent groups such as the I, viz] CRO-MAGNON AND GRIMALDI TYPES 49 Plynlimmon moorlanders, and the secluded settlements already mentioned in Spain, Portugal, and Algeria. We are confronted however at this stage with a new turning point of advancement, the participation of more than one distinct breed of man in a single tradition of culture, and in exploitation of the same region. For, side by side with this Aurignacian type, at least two other varieties of man made their appearance in west- ern Europe during the warmer and drier period now in question. One of these, represented by the * Cro-Magnon ' skeletons, is both less widely distributed, and of larger and more modern-looking build; and has left, like the Aurignacian, its descendants among the modern population of France, Spain and North Africa, and also (according to some observers) round the western Baltic, Its genefal similarity with the Aurignacian has led to the presumption that it spread likewise from the south-west. The other type, distinctly negroid, is best represented by skele- tons from the Grimaldi cave near Mentone. There can be little doubt of its African affinities, and there are two other indications of such African types in Europe : a pygmy breed of somewhat negroid appearance, from an ill-dated deposit at Schaffhausen, near Constance; and the well-marked steatopygy which character- izes negroid Bushmen and Hottentots of South Africa,, and of some other negro breeds further north, and is represented in European drawings and sculptures of the female figure from Auri- gnacian times to the neolithic art of Malta. Steatopygy however is not an exclusively negroid character, and as it has been observed among the living in the same secluded districts along the western Pyrenees as the Aurignacian individuals already mentioned, all that results from the Grimaldi and Schaffhausen finds is that African varieties either had not yet entirely disappeared from southern Europe, or occasionally returned thither during periods of exceptional warmth* In general, however, it may be assumed that the present climatic zones were henceforward fairly well established; and that negro man, in the mass, did not range north of the Sahara desert, nor, except sporadically as now, along the Atlantic seaboard of North Africa. *The mode of life of * Aurignacian * man differed, no less than his build, from that of the Mousterians whom he superseded; and there is no reason for an archaeologist to dispute the consensus of anthropologists that there is no trace of intercourse between the two types* If they met, it was as independent competitors for the means of subsistence; and the almost total disappearance of Mous- teriati man after the arrival of Aurignacian, suggests that it was c.A.H-r 50 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. 'war to the knife' between them. The Tasmanians had no better fortune, after the arrival of Europeans in their country; and while the Tasmanians were rather less brutish than Mousterians, Euro- pean culture claims some advance over Aurignacian. Aurignacian industry shows great superiority over Mousterian. Flint implements are fashioned no longer from cores, but from selected Hakes, and are trimmed by careful flaking on both sur- faces., not on one only as heretofore. There are special types for boring and graving, knives adapted for a drawing-cut, and chisels for push-scraping. And the uses of all these are illustrated by an increasing quantity of bone tools, some of them shaped to be hafted with wood. There are bone dress-pins, beads and pendants; and bone whistles suggest concerted signals for action at a dis- tance. We may infer that men hunted now in a horde, and obeyed a leader; and that women took some pains with their attire. More noteworthy still is the beginning of pictorial art; en- gravings on bone and eventually carvings in relief and in the round; larger drawings and paintings on the walls of caves* The subjects are the animals most hunted by man, as their remains in his settlements show, reproduced with a sympathy and accuracy of observation, and a vigour of draughtsmanship and modelling, which have rarely been equalled. Human figures are rarer; usxntlly representing women, with rather prominent jaw, long-braided hair, and frequent steatopygy. Both men and women arc usually shown very hairy, and are seldom clothed, except when they mas- querade in complete hides of animals* A late and perplexing class of linear designs may represent huts or other constructions of timber. This civilization extends in time from the beginning of the interglacial period already mentioned, until far on into the last recurrence of glacial or rather pluvial conditions, for the fourth or 'Warm' glaciation was a comparatively small affair. It passes through several phases, Aurignacian in the special sense; Mag- dalenian, a well-marked regional culture of the Atlantic coast plain, best illustrated in the Dordogne cave of La Madeleine, which belongs to the beginning of a late spell of austerity; and Azilian, first identified in a remarkable cave in the foothills of the Pyrenees, which shows Magdalenian art and industry much de- generated, and only retrieved in interest by its use of painted symbols to distinguish hoarded pebbles of uncertain use. It is the first advance from delineation of objects to the visual representa- tion of ideas. On some sites in France, between Aurignacian and Magdalenian I, vn] THE LATER PALAEOLITHIC CULTURES 51 deposits, which in general form a continuous series of develop- ment, otcur fresh and very characteristic types of implements best represented at Solutre, near Macon, where hearths and burials of their makers lie immediately over a vast deposit of horse-bones marking the climax of the dry steppe regime, and probably some kind of late-Aurignacian slaughter-ground. Their graceful 'laurel- leaf and Svillow-leaf ' blades, single edged knives, and one-shoul- dered points suitable for missiles,, economize labour by skill in the choice and manipulation of material, and might be mistaken for a high quality of neolithic work. This use of missile weapons is itself a new invention, appropriate to the hunters of so swift an animal as the horse. Ivory beads, in country now devoid of ele- phants, suggest either wide range of movement, or some form of exch^fhge* Sketches of animals, on pebble and bone, and hoards of yellow, red, and brown colouring matter, indicate artistic tastes; insignia and whistles imply organized action against swift and in- telligent game. It has even been suggested that Solutrean horses were tame; but no horse-bits have been found, nor proof of any special breeding. This Solutrean episode is noteworthy because here for the first time we have intrusion of one culture, abruptly and temporarily, into the region where another was in process of development. And as this intrusion occurs at the climax of a cycle of dry continental climate when conditions were most uniform and the obstacles of forest, river, and morass were minimized, it has been commonly interpreted as due to the intrusion of fresh people of the tall heavy- built long-headed stock represented by individuals from Brunn and Predmost in Moravia. Whence these people and their culture originated is not yet clear, as no human remains can be identified as theirs. Their flint technique, as has been already noted, suggests that they inherited Acheulian tradition, and it is possible that this tradition survived somewhere further east, while central Europe and parts of the west were passing through the Mousterian de- cadence and receiving Aurignacian culture from the south-west. So severely continental a climate in western Europe suggests austere drought further inland, and accords with the deep accu- mulations of loess over the greater part of the northern flatland, during late palaeolithic times. It was certainly a good country to leave. The lack of Solutrean remains south of the Pyrenees (except one isolated find at Altamira), and their comparatively frequent occurrence in lower Austria, Bohemia, Hungary a&d eastern Poland, almost preclude the alternative of a southern origin, which was formerly thought possible; and the striking resemblance of 4—2 52 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. certain Nile valley implements, which has been adduced in favour of this, must be taken in connexion with the similar workihanship of the earliest implements of neolithic Susa (which will be "dis- cussed on p. 85); and would be explicable by just such an exodus southward from the north-eastern steppe, as has brought nomad raiders more than once from Turkestan to Mesopotamia and the borders of Egypt within historic times* We shall see reason to suppose, later on, that the earliest neolithic people of the south Russian steppe, and probably also of the Danubian region, may be descendants of these Solutrcan hunters, withdrawn as rapidly as they had come, when the Magdalen ian climate reafforested the western plains, and restricted their hunting grounds. Another small group of Solutrean remains is notable as the first indication of man's presence in Scandinavia, the soxfchern promontory of which must therefore have been released at least temporarily from the ice-grip; and there seems reason to believe that there may be Solutrean blood in the earliest men of that region whose remains have been preserved; though their date is very much later. Indeed, at any period when the forest zone was restricted to the foothills of the Carpathians and the central German highlands, there was no physical obstacle to the move- ments of hunting hordes between the shores of the Black Sea and Caspian, and those of the Baltic, or the margin of the Scandinavian ice-sheet, VIII. THE CLOSE OF THE OLD STONE AGE On the other hand, the steadily increasing moisture of the Mag~ dalenian climate restricted the habitable areas; for the forest en- croached on the hunting ground, and horde-hunting tribes do not easily adapt themselves to forest life. Arts and industries degener- ated, especially when the antlers of the reindeer gave place to the less workable tines of the forest-ranging deer, as the material for harpoons and spearheads* The barbed harpoons themselves betray the growing importance of fishing, as the rivers increased in size, The abundance of miniature flints, at Tardenoise and many similar sites, suggests that wooden clubs or spears were armed with them, as was customary later in the Alpine lake-dwellings; and indicates that timber was more plentiful* Then, for causes which are still obscure, the distinctive * Cap- stan' type of culture, which we have already seen to have been best and earliest represented near Gafsa in Tunis and widely distri- buted from Tripoli to Morocco, spread northwards through Spain I, vm] CLOSE OF THE OLD STONE AGE 53 and France (where its local varieties have been commonly known as 'Caiftpignian') into Belgium: it shows a revived interest in flint work, and some of its forms recall a far older technique, which had certainly lasted long in north-west Sahara, and apparently also all across north Africa. This early African style has some resemblance to the Mousterian; and we may compare the relation already suggested between the Acheulian and Solutrean tech- niques. In any case the Campignian style was of southerly origin, and marks a last palaeolithic attempt to reoccupy the west of Europe, perhaps during some spell of drier weather. But this adventure failed, like the Solutrean irruption, and Campignian survivors merged in the disorganized remnants who harboured in cave shelters in Spain and the south of France, in open settle- ment on the downs along the Marne and Somrae and in Belgium (where there is some reason to believe that at Flenu and Spicnncs there was also immigration of rude tribes from the north-east), and in fishing and hunting stations along the Atlantic coast from Portugal to Scotland and Denmark. Here immense refuse-heaps of shells, bones, and implements mark a last stage of collapse of the old hunting folk, like the modern Yahgans of Terra-del-fuego and the * Strandloopers ' of Cape Colony. These 'kitchen-middens' represent a long period, during which the interior of the continent was for the most part forest or swamp, and men hunted or gathered shellfish along the strand without wandering far, except occasionally seaward for fishing. Only three almost accidental acquisitions betray some overlap between the desperate state of these survivors of the Old Stone Age and the new world which was coming into being within the dreaded oak-forest. The dog, in this extremity, became man's messmate and fellow-hunter; occasional implements of neolithic fabric were acquired somehow, and refurbished by flaking as if in mere ignorance of their proper handling; and the clay linings of old leathern cups and bowls, accidentally burned at first and there- by hardened in the fire, gave a first notion of pot-making, to be imitated by degrees, but without improvement of form. All three discoveries suggest contact, at least occasional, with some other kind of man, to whom forest and swamp were familiar, and habit- able. And both forest and swamp contained such men, as we have now to see* Further north, the swamp, engulfing by degrees much that had been tundra and cold steppe, north of the central German high- lands, had long since been assisted in its dreary advarice by con- siderable subsidence of the whole of north-west Europe, so that 54 PRIMITIVE MAN, IN GEOLOGICAL TIME [CHAP. the period during which the Scandinavian ice-sheet shrank finally back, and exposed the south promontory of Sweden, waS one in which the Baltic was an open gulf of an enlarged North Sea "that washed the 'hundred-foot terrace* of its Scottish coast. The silt set free by the annual thaws varied slightly in quality, as the season changed^ and the banded clays which it formed in this Baltic gulf form an uniquely continuous record, so minutely graduated that it has been possible to reckon within a few centuries the interval between then and now. From this vast natural chronometer it would appear that the coast of Scania was released about 12,000 years ago, and northern Sweden about 5000 years later. The re- lease of the north German lowland was of course rather earlier, perhaps about 15,000 B.C. Those inshore * Yoldia* clays, so called from the chief marine shell which they contain, were later raised above water so far that the Danish archipelago became dry land, and the Baltic a lake wherein Ancyfas and other freshwater shells superseded the marine Yoldia. This rise greatly increased the swamp-covered area, and seems to have permitted the westward spread of a peculiar culture, best illustrated by the Maglemose settlement in eastern Denmark* Afloat or stranded^ according to the season, a raft was constructed of pine trunks from the coniferous forest fringe which encroached on the swamp margin as it rose and dried; and from this precarious home men fished and hunted, of a distinct breed which seems to have moved westward from the cold steppe of northern Eurasia, and may have been of ultimately Mongoloid origin. At its greatest extension, this type may have made touch with the Magdalenian hunters of France, if it be admitted that one of its men has been found with them in the Chancelade cave* But it borrowed little from them, and only in its retreat, when the forest restricted Its swampy hunting grounds, did it absorb something of the Mag- dalenian artistic spirit, perhaps from hunting parties out of the west who had wandered onto the north flank or the forest and re- mained there. "With the Maglemose culture may be connected other swamp-land settlements round Lake Ladoga and in the coastlands east of the Baltic; and it seems likely that a Mongoloid clement among the modern Finns> and probably the main strain of the Lapps, are descendants of these people* What forced the retirement of the Maglemose culture was no less the aggression of the sea than that of the continental forest. The Baltic became open gulf again, rather more so than at present^ for it is to this phase that the ^o-foot terrace * of Scotland belongs* Marine shellfish entered, such as the periwinkle^ which gives its I, vm] SWAMP AND FOREST IN N.W. EUROPE 55 name to this L//#mr^-stage; and following them the shellfish-eat- ing folk of the kitchen-middens wandered along the north German coa^t as far as Lettland, where the pine forest closes upon the shore. Here they persisted long; and the miserable Fenni described by Tacitus in the first century A.D. may well be a last remnant of them. The swamp-culture of the north-east, as will be seen from this sequence of events, coexisted with a considerable part of the palaeolithic decadence. At least two minor advances of the snow-cap of the Alps can be traced during the long withdrawal of the Scandinavian ice-sheet, and the general mitigation of the climate of western Europe was to this extent delayed and interrupted. It would probably be safe to place the Maglemose culture at about the same period as the spread of Campignian influences northwards over France, and it is certainly older than the kitchen-middens, since these crowd closely on the modern coastline, which was submerged in the Maglemose period. The part played in north-western Europe by the swamp-culture, and by those alien men from the north-east who are its repre- sentatives, was but slight and of short duration. The continental forest on the other hand, which had been spreading intermittently across Europe, northward and westward in the wake of the re- treating ice-sheet, fringed by birch, hazel and pine, but itself com- posed mainly of oak and other deciduous trees, with the zone of beech, walnut, and chestnut following on an average some five hundred miles behind the pines, had reached and smothered all country where trees could grow, as far as the Atlantic seaboard, and southern Britain at least, by a date which may roughly be estimated not far short of 7000 B.C. The palaeolithic remnant had retreated before it till only the kitchen-midden folk survived on the very strand-line, and discontinuously even there. In the in- terior, a few exceptional moorlands, bleak downs, and the larger expanses of thirsty loess in the Rhine and Danube basins and in the north German plain, remained comparatively treeless oases where hunting folk might live. And if this had been all, the Old Stone Age might have passed out of human experience, a withered b?anch of the 'Tree of Life.1 That this was not so is due essentially to two factors. One is the sequence of climatic belts already noted, which provides that a northward shift of the westerly winds is accompanied by commen- surate though not necessarily equal shift of the * Mediterranean ' and * desert-zone* climates, and consequently by ampler* accom- modation for human activities of the Eurafrican type. The other DI n III nil V CHAPTER II NEOLITHIC AND BRONZE AGE CULTURES I. THE HIGHLAND ZONE AND ALPINE MAN WHAT then was going 021, meanwhile, within the Highland Zone? For several reasons, evidence from this region is very scanty. Much of it is ill-explored from every point of view; still more — and especially in its best explored west-end where lateifrperiofls are exceptionally well exhibited and have been care- fully studied — is out of reach for the same reason as is so much the evidence for Interglacial man elsewhere; namely that the nearer we approach the centres of glaciation, the more completely do later glacial deposits cover the surfaces of the earlier; so that in Switzer- land and south Germany, for example, human record hardly begins before the neolithic age. Further south-east the scale of acci- dent is loaded the other way: for, in proportion as glacial action passes Into pluvial, it is not excess of deposits but the wholesale removal of them by rain-fed torrents that limits observation. A very large proportion of the land surface of Asia Minor, for in- stance, has no 'surface deposits/ in the ordinary sense, at all; even in the greater valleys, which are themselves rare, the upper terrace gravels have been severely dissected; and the lower have been covered by alluvium, deposited often within historic times. Consequently, it is almost exclusively by inference from other data, such as the distribution of racial types to-day, and certain indications of the course of events in immediately prehistoric times, that the prehistory of this great region must be reconstructed pro- visionally. Limiting conditions are supplied by the climate, vege- tation, and consequent mode of existence imposed here upon man in general. Like all other highlands this literally * Alpine* Zone has always Imd a cooler and moister climate than the lowlands north and south of it; and in periods when the submerged areas on its Mediterra- nean and Sarmatian flanks were extensive, this humidity was greatly accentuated, It must be inferred from this that the whole region has been predominantly and persistently a forest area. General changes of temperature would replace subtropical by temperate or subarctic species, but would not necessarily alter the forest area. 58 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. A period of general drought would draw the forest margin inwards and upwards among the foothills; a pluvial period would expand it into the plains; and a heavy snow-cap would devastate it amdng the peaks and ridges5 and down the glaciated valleys. But none of these agencies would avail to destroy the forest regime altogether; that catastrophe was reserved for the hand of man; and even man has not devastated it wholly as yet. It follows that the grassland and parkland, fauna, whether African or Arctic, which is so widely associated elsewhere with the first signs of man's presence,, did not pervade the Highland Zone at all generally. In alluvial valleys, and in the large interment plains and forelands, such as the Danube valley, which are characteristic of the region and were reserved to grassland by their mantle of interglacial loess, it was possible for small herds of elephants to wander, as they did still in north Syria in the twelfth century B.C.; and for the lion to maintain himself as he did in Palestine until, at least, the tenth century,, in Macedon until the fifth, and in the Mesopotamian foothills until the present time. But these animals were never characteristic of the great mass of the highland: their place was taken by bear, wolf, and ruminants large and small. Man, hunting in the open, as he hunted in the lowlands of western Europe, or on the great steppes and parklands, had there- fore no inducement to occupy the forest area: at most his mode of subsistence brought him along the larger rivers such as the Danube, and its tributaries. It is significant that all the earlier individuals whose remains have been found hitherto within the Highland Zone are of the Neanderthal type; that the only large froup of Neanderthal men hitherto recorded is that from tlie rapina cavern in the headwaters of the Save; and that almost all the Neanderthal men have been found, along the western outliers of the highland core of Europe, To draw conclusions from the distribution of so few examples is risky, and the fragments from Kent's Cavern and one of the Gibraltar caves impose caution already; but there is another reason for expecting that the Nean- derthal type may be found to represent an early forest man, differ- entiated by his surroundings-, as well as by long descent,, from his Aurignacian contemporaries on the grasslands and parklands ouf- side. Whatever the relations of Neanderthal man to 'the Highland Zone, the Aurignacian stock at all events seems to have originated elsewhere, and to have only penetrated it locally and 'marginally* Here again, however, it is possible that Aurignacian relics scattered nearer its core may have been obliterated by the last outspread of glacial debris. And these last glacial deposits are sufficiently widely distributed to show that in the period which in westerti Europe is that of transition from palaeolithic to neolithic culture, practically the whole of the main highland was divested, not only of any human population it may have harboured inter- glacially, but of all save the most alpine vegetation. In any case we know enough about the changes of climate within the glacial period, to presume wide oscillation of contrasted types of man, as the forest spread or shrank again. As the highland was surrounded from north-east to south-west by tundra and cold steppe, while southward and eastward its slopes were washed by Mediterranean and Pontic Seas, there was only one avenue by which, when the climate was mitigated finally, it could be rgoccupied by that sequence of plants and animals which it exhibits now. This avenue is from the south-east, and consists of the long Asiatic continuation of the Highland Zone itself; for the Hellespont * river, * as Greek geography rightly named it, offered no real obstacle, and the occurrence of alpine flora in Crete and even in the larger Cyclades illustrates the regional continuity between the highland shores of the Archipelago itself. We have therefore to conceive the Highland Zone as a single great region, peninsular and self-contained; thrust westward into the heart of Europe from its Armenian summit, where it joins, base to base, its twin eastward promontories, the north Persian ridges and the Zagros escarpment south-eastward, and the diminu- tive but vitally important southward causeway through Syria into south Palestine. North of the Armenian mountain knot, and inti- mately associated with it, in climate, flora and fauna, lay the trans- verse ridge of Caucasus, steep-fronted towards Sarmatian seas or their flatland bed. Though we have no direct evidence yet as to the older human population, the modern inhabitants of this Highland Zone give an important clue, and the known course of events in the long neo- lithic and chalcolithic periods confirms that clue impressively. From end to end, the dominant type in historic times is distinct and characteristic; interrelated by well-marked broad-headedness and high-headedness; by wide and high orbits, set level or droop- ihg outwards, with almost no trace of brow-ridges; by broad cheek- bones and palate; by a characteristic wide square jaw with its * hinge-ends long, massive, and rising nearly at right angles with the plane of the teeth. It has broad shoulders and hips, broad hands and feet, with thick wrists and ankles, and a generally thickset build; dense parchment-like skin, sallow in the shade, and leathery under the weather; eyes hazel or brown, dark brown wavy hair, 60 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. long In both sexes and very copious on the body, with profuse beard in the men. Its nearest affinities are with the other white-skinned and wavy-haired types, and with these it has formed numero'us intermediate varieties3 within which its own bodily features appear to be in the Mendelian sense * dominant,1 so that, once introduced into a region, it tends to persist and become accentuated with time, Its purer varieties are all found within the Highland Zone: its occurrences in north-west Africa, Spain, and the Canaries are not sufficient to establish the south-western origin formerly proposed; and the broad-headed strains which connect it north-eastward with the Mongoloid population of the Eurasiatic woodland, — whose other physical features are very different, — may be attributed rather to admixture between independent types spreading in op- posite directions, than to any propagation of such strains inter the Highland. The * Alpine' type in fact may be regarded as essentially of Alpine origin. On account of its great width, this type of skull was long classed with the Mongolian; but the general build and lofty proportions of the brain-case, and still more the peculiarities of the face and jaw? should have precluded this; and the absence of skin pigment, the wavy hair, and the copious beard and body-hair, force the con- clusion that we are dealing with a stock of quite other origin, more likely to be akin to the other * white* races, b>ut nevertheless strongly contrasted with these, in its head-form and bodily build. Moreover, between the highland home of * Alpine* man, and the still loftier plateaux, which we have seen reason to regard as the Mongoloid * cradle/ the narrow but gigantic ridges of the Hindu Kush and Pamirs have been long and almost continuously glaciated, as we have already noticed; their flanks are dissected by ancient transverse gorges; and below ice-level there Is vast extent of dense Inhospitable forest, fed, like the snows above them, by wet monsoon winds. Such human elements as have worked their way round this vast ice-cap since its last contraction have moved wholly from west to east, not from the Mongolian habitat into the Alpine; and Mongol admixture in highlands west of the Hindu Kush can always be traced to another and quite recent origin^ namely to nomad pastorals Intruded transversely from the low* lying grasslands of Turkestan, which m all but the latest phases of the Sarmatian sea lay submerged and therefore as Impassable as the snow-cap. Enough seems to be known of the correlation between diet and the form of the jaw, and of the pull of the jaw-muscles on the tem- poral and parietal region of the skull, to warrant the suggestion II, i] PRINCIPAL VARIETIES OF ALPINE MAN 61 that the peculiar combination of a short and massive jaw, suited rather Tor crushing than for cutting or tearing, with a musculature soTeeble as to be accompanied by almost no lateral compression of the brain-case, points to a long-continued mode of subsistence quite different from that of the carnivorous hunters of the steppes and parklands of Eurafrica. And we have already seen that the Highland Zone has necessarily been at all periods more or less completely a forest area, ill-adapted to maintain the large land- animals of the parkland except quite locally and sporadically, but abounding in many kinds of trees and shrubs bearing fruits or nuts, from conifers to chestnut and walnut, and from cranberry, crab-apple and sloe, to the characteristic fleshy-fruited apricot and peach of Persia and Armenia, and the vine, mulberry, fig and olive whi&h are common to the foothills of the forest zone and the ever- green flora of the Mediterranean region south of it. We shall see reason also to suspect that the first domesticated grasses, wheat, barley, and millet belong to genera which inhabit this same mar- ginal belt between the forest and the southern grasslands; and that they were cultivated by men of Alpine stock as far west as the Swiss lake-basins, and as early as we have any evidence of modern man in that section of the highland (p. 72). In this connexion it is perhaps worth noting, that Greek eth- nology, which so often formulates conclusions which it has been reserved to modern observers to substantiate, clearly distinguished between an earlier phase of subsistence, that of the * nut-eating' men (ySaXaz/^c^ctyot az/8/>es), and a later 'meal-eating* culture (az/Spes aX^Tya-rat); and that, in a very ancient stratum of Greek myth and ritual, the Power to whom the gift of grain-food was ascribed was worshipped with sacrifice of the pig, a typical forest-ranger. But within the Highland itself, the Alpine type varies, and the actual distribution of its principal varieties gives a clue to its prob- able cradle-land. Most accentuated is the 'Armenoid' variety, of the Ararat mountain region, with a head characterized by very lofty vault and outward-drooping orbits, and so abruptly flattened behind, that it has been ascribed by some observers, both ancient and modern, to artificial deformation. This variety predominates throughout the central section of the Highland, and is also not un- common throughout south-eastern and east-central Kurope. Least peculiar in the two respects already noted and distinguished rather by its smoothly globular cranium, and by a jaw broad but not so angular, are the West European varieties, especially in AuVergne and Savoy, and the most easterly groups from nolth Persia to the Pamirs and beyond; and the general likeness between these 62 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. remotest groups is in fact such as to suggest that they represent a quite early phase both of differentiation and of outward srpread. Intermediate types, characteristic of east-central and south-eastern Europe, and commonly described as Dinaric5 show this globular head becoming more angular and cubical; and have their counter- part in Caucasus, western Persia, and along the southern margin of the highland thence towards the 'Dinaric' area westward, with notable offshoots southwards through Syria, Their distribution suggests a later stage both of specialization and of dissemination, around the central area already described, where alone the develop- ment has attained to that extreme 'Artnenoid* phase whose dis- tribution is least wide and also apparently least early. The relative antiquity of these successive phases of growth and spread can be stated approximately; for the outermost western or 'Cevenole7 type made its appearance in the Alps and in France during the transition from late palaeolithic to neolithic culture, At Ofnet, in Bavaria, it made its appearance in the Azilian phase, mixed with Aurignacian people, and already interbreeding with them; eastward, on the other hand, the human remains from Anau show that no such * Alpine' type had reached the north margin of Persia until after the second desertion of this early settlement (p. 85). Similarly the broad-headed intruders into Egypt at the beginning of the dynastic series, and into Crete and the Cy chides at the beginning of the Minoan Bronze Age, belong to the second phase, which may therefore be dated about 4000—3500 B.C.: and the first known occupants of Cyprus, and of Troy, in the earliest Bronze Age are of the same type, Fully developed 'Armenoid* remains, on the other hand, do not seem to be found anywhere until the second millennium at earliest* At present, therefore, it seems safe to regard this * Alpine* group of broad-headed types as representing phases of a special development within the Armenian mountain mass, or rather (since this region was certainly subjected to severe glaciation during the Ice Age) within the mountain-girt plateau of Asia Minor imme- diately west of it; large enough, isolated enough, and at all relevant periods habitable enough to become the cradle of such a sequence of varieties; sufficiently well connected with large similarly quatf fied regions eastward and westward and sufficiently liable from its geographical position to periodic changes of climate, to serve as a reservoir of population, like the highlands of Atlas and the Iberian peninsula in earlier times, and like the Arabian and Eur- asian reservoirs later on. Surprise has sometimes been expressed, that even considering II, i] FOREST CULTURE AND POLISHED IMPLEMENTS 63 how little scientific research there has been in this region, traces of palaeolithic culture are still so rare here, especially in view of the quite common occurrence of neolithic implements of polished stone in all parts of it. There is however good reason why flaked implements should be in any case rare in such a region. Though fairly well adapted for attacking wild animals, cutting up game, and dressing hides, and even for shaping and decorating imple- ments of bone and antler, the flaked implement is comparatively ineffective for felling trees, splitting logs, dressing planks, or pounding roots, bark or nuts* Moreover, though a large part of the great flatlands consist of, or rest on, flint-bearing strata — cre- taceous or derivative — and are as open country as they actually are, mainly because these limestone surfaces are inhospitable to tree^ in the highland zone, on the other hand, these beds are either absent — which accentuates its forest aspect, seeing how precarious is tree growth over limestone — or so distorted, or even deficient in flint and chert, that the supply of this material was scanty, and (what was worse) discontinuous. Collateral evidence is that in Egypt, where timber was rare and exotic, the flake- technique persisted and underwent cumulative refinement from Solutrean to chalcolithic times; and that in the kitchen-middens of north-west Europe acquaintance with polished implements in- creases pan passu with the northward advance of oak-forest, dis- placing conifers, just as these and the dwarf-birch had previously invaded the cold steppe. A further point is, that even before ac- quaintance with the polished technique began, there is a complete revolution in the mode of employment of stone implements gener- ally. The tapering pyramidal point for stabbing, and the longi- tudinal edge for cutting or ripping, are supplemented, and even- tually replaced in the more massive implements, by the transverse edge for hacking and clearing, under shock (rather than pressure) applied to the butt-end. This is conspicuous in the Campignian technique, which will be remembered as marking a last palaeo- lithic aggression in the moist forest-ridden west. The form of the butt-end, too, frequently suggests the use of some form of haft; ajid hafting itself presumes familiarity with wooden staves and clubs, and therefore with parkland at all events. Again, in any region where roots and tubers formed any considerable part of the food supply, the mere act of breaking the ground with pick or hoe, whether of wood, of bone or antler, or of hafted stone, automatic- ally smooths the surfaces of the implement about the point or edge, and leads to a natural finish, familiar among the digging-hoes of shell which are used by some Pacific peoples, and among stone 64 NEOLITHIC AND BRONZE AGE CULTURES [CHAP, hoes from pre-Columbian sites in North America. From this it is but a single step to the artificial improvement of blunted or splin- tered edges by grinding, not by flaking; and certain implements from the Nile valley show every stage of this advance in technique, though they are always exceptional there, and late. On these grounds, the inference seems justified., that whatever other causes may have been in operation, forest life, and especially the hacking of timber and the grubbing-up of edible roots? favoured the de- velopment of such types of stone implements, and methods of manufacture, asactually occur earliestandmost persistently through- out the Highland Zone, and are least and latest represented in comparatively treeless regions such as the Nile valley and the wide flatland of north Africa. ^ f It would be premature to correlate in more than the most t&nta- tive way the polished-stone technique exhibited in this region, both by catting, cleaving and grubbing implements, and by those for crushing and rubbing which so commonly accompany them, with the probability already noted, that the type of skull and jaw char- acteristic of Alpine man may result from long habituation to a diet of nuts, roots, and other vegetable foodstuffs needing steady mastication rather than the biting and tearing which meat re- quires, and so thoroughly received from the long highly-muscu- lated jaw? and prominent incisors and canines both of negro man and of the long-skulled Aurignaclan hunters of Eurafrica and the north-west* But the coincidence is noteworthy, and the roughly concomitant spread of * Neolithic * culture and of * Alpine * types of man is more striking still. That the earlier stages of such spread should be ill-represented is only what might be expected in view of the prolonged glaciation and widespread diluvial deposits of those western sections of the Highland Zone which alone are adequately explored* But it seems clear that the advanced stage of neolithic industry which is repre- sented even in the earliest settlements around the Swiss and Italian lakes, which had an Alpine population, presupposes a long and homogeneous development; and the occasional introduction of implements in fairly advanced phases of this polished technique into the Danish kitchen-midden s? which are the leavings of Aun- gnacian people, among a multitude of flaked implements of Campigniati and other late palaeolithic makes, suggests that in north-western Europe at all events there was just such an overlap of the older and the newer industries, as is proved for the long- " leaded and broad-headed stocks themselves by the mixed Azilian deposit, at Ofnet in Bavaria, which is of sufficiently late date to MAP 3 offline treeless1 areas" Coniferous forest Deciduous (broad leaved) forest Evergreens Temperate gr/js-f lands merging into steppe MAP 4 OUVE,VINE & ORANGE AREAS OF THE MEDITBWAJSTEAN" n by W, 6-AX. Jotoston !.¥ Ealiv II, ii] CHARACTERISTICS OF NEOLITHIC CULTURE 65 invite comparison with the earliest broad-headed remains from Grenellfe near Paris, and Furfooz in eastern Belgium, and with the ^earliest lake-dwellers among the Alps themselves. II. CHARACTERISTICS OF NEOLITHIC CULTURE Formerly, when attention was still mainly directed to the various types of stone implements found accidentally in surface soil., the contrast between flaked and polished technique seemed to be of greater value as an indication of date, than jiow, when the long overlap in time of these two techniques has been established on evidence from tombs and stratified sites, and when the significance of each fabric is better understood, Though the terms * Palaeo- lithic** and '^Neolithic/ have remained in common use for the older and later phases of the Stone Age, they are now applied in a secondary sense, to denote strongly contrasted phases of general advancement; and it is important to realize wherein this contrast consists. The men of the Older Stone Ages took the world as they found it, and made little attempt to alter it. They chipped natural stones into weapons for cutting and stabbing; they wrapped them- selves in skins and furs stripped from their prey. But the animals which they hunted and the fruits they gathered were wild, their shelters were natural caves, they buried their dead (at best) in a hole in the cave floor. With the late exception of hafted spears, they had no notion of construction1, and no use for timber, or for any movable object not easily held in the hand. The sole hints of cooperation or of social order are occasional whistles, and carved staves which may have symbolized rank. We may probably fill in the picture from the habits of merely hunting peoples on the open lands of Siberia, North and South America, South Africa and Australia; except in so far as all these have acquired the dog, which cannot be traced back earlier than the kitchen-middens. The New Stone Age, from its first beginnings, reveals a quite different outlook on nature. Even the implements illustrate this; their materials are varied, and presume search and selection, me- thodical and gradual improvement, constructive skill in hafting, a£Td appreciation of the elastic quality of wood, for long axe-helves, and above all for the bow, which appears first in the transitional rock paintings of Spain (p. 94). With axe and adze, man's do- minion over the forest was assured; and with chisel and saw, his mastery over the timber he had to fell. Loose stones he had already 1 Some French observers, however, have interpreted certain linear designs in Magdalenlan caves as representing wigwams and pitfalls for game. C,A«H. i 5 66 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. piled together occasionally, at all events to conceal jEiis dead. Car- pentry and roof-construction were only a matter of time. To utilize other waste products of nature., fibrous twigs, grass, bark, fruit, rind and the like, for binding, wrapping, and eventually for plat- ters, bags,, and baskets, was another elementary step in the same direction, supplementing the use, which was already ancient, of hunters' debris, bone, sinew, hide, and fur; for man seems to have used up his own leavings before appropriating those of nature. The caulking of such vessels with another waste product, ubiqui- tous mud, led on to substitution of mere clay, hardened by fire to earthenware, for perishable skins and basketry. Other waste pro- ducts, nuts, kernels, pips, and grass seeds squandered after a meal, and found germinating in spoiled earth round old encampments, propounded problems of their own. It has been suggested^ too, that early beliefs connecting such germination with human life and death may have been suggested by unforeseen .growth of fresh plants over old graves. But this obviously did not occur till the graves were in the open, not in cave floors; and the whole notion that by deranging natural soil, natural vegetation may be sup- planted by a 'crop/ more edible and fertile, and of man's own selection, is quite outside the hunter's range of ideas. The other new notion, of captivating, rather than capturing, wild creatures, and making them domestic,- — that is, *at home' around the camp, — is less alien to the hunter*s thought, and in its simpler forms is not easily reconcilable with the plans of the plant-grower: between Cain and Abel, in the story,, there was early feud, for the grazing herd draws no distinction between natural and cultivated green-stuff, except to prefer the latter. But funda- mentally the pastoral creed is the same as that of the cultivators; * man's place in nature* is at the source of life, to multiply and replenish the earth, These various forms of exploitation increase subsistence, but they demand effort: Mn the sweat of thy face shalt thoti eat bread/ When the world is 'so full of a number of things/ as it becomes for either cultivator or pastoral, a human group is so far released from ^ the rigorous restrictions self-imposed on hunting-hordps, that it may increase its population not merely safely "but with advantage: many hands make light work. But many hands, many herds, and much store of foods which must be gathered at one harvest for the twelvemonth, mean much fear of attack from without; and other waste products, loose stones^ dead trees, mere soil, were piled into a ring-fence round the settlement. It was only gradually that unpenetrable defences challenged invention II, n] EXPRESSIONS OF INTELLECT AND EMOTION 67 in the aggressor, and differentiated Implements of war from mere hunting-tools. These principal aspects of invention, — which is reason's ad- justment of the materials and the forces of nature to fulfil desires, — all come into view as the New Stone Age dawns, and separate it from the Old. Of the other great inventions we have to wait long even for the next; when fire, already in use to harden clay, should be applied also to soften stone, and extract metals therefrom; and when animals already tame should yield man not only nutriment but a new source of power. Four others, on a higher intellectual plane, come only slowly into sights observation of the sun and moon in their seasons, first hinted by sundry circles and crescents in neolithic &rt, superseding the palaeolithic masterpieces of animal portraiture; the curiously abstract quality of nmch else in neo- lithic ornament, as if number, mass, and proportion were felt to have an interest of their own; a conception of value, which may fairly be presumed among people who, though sedentary, are found to have acquired, for whatever reason, commodities from afar like turquoise or amber; and a new self-consciousness and introspection, displayed in emphasis on details of technique in decoration, and in the choice of men and their acts and works, rather than natural forms, for pictorial record. Still higher aspects of advancement than these are even less easy to detect without risk of private interpretation. Whatever the first purpose of those emphatically feminine figures, whose neo- lithic types eventually come to be associated with the profoundest conceptions of early religion, their origin is not here, but long be- fore in Aurignacian time. The same applies to the first impulse to representative art of other kinds (in any magical implications which it may have had), in which Magdalenian draughtsmen are unexcelled; to the motives for careful disposal of the dead, which as a custom is at least Mousterian; and to the beliefs and emotions aroused by the primeval mystery of fire. All these we may take to have been already traditional at the close of the Old Stone Age : the New added only those fresh glimpses of the significance of life, wkich were suggested by experimental acquaintance with the be- haviour of animals and plants. Religious conceptions such as those of a * Good Shepherd ' or of the * Bread of Life ' can hardly have anticipated economic discoveries from which they draw their sym- bolism; while they may be but little subsequent. In estimating the significance of forest and dense parkland as a factor in the transition to the neolithic stage, the distribution of such conditions should be considered as a whole. The composition 68 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. of actual forests in Europe shows them to have resulted from spread and coalescence around at least three centres. The clearest example of this is in the north-east, where the characteristic birches, pines and oaks of the Alpine Highland, spreading beyond the foot- hills of the Carpathians over central Russia, have met about longi- tude 45°? a westward-spreading woodland composed of Siberian species. That their narrow zone interpenetration., once established by contact of their advancing margins, has remained in long equi- librium is shown by the distribution of early cultures, and varieties of man; all eastward of this zone being related to widespread Asiatic types, and nearly all west of it to Alpine, Danubian and Baltic. Only comparatively recently have Asiatic people succeeded in establishing themselves west of this zone, and acquiring un- familiar woodcraft. To the south-west the course of events has not been so clearly traced* The greater range of latitude here, and the wider variations of climate which the Atlantic seaboard has undergone, have per- mitted far greater oscillation. The woodlands of Spain and Africa Minor have been repeatedly continuous with those of west-central Europe, and have developed but few distinctive forms, so that their coalescence is harder to detect. The former existence of large forest regions between the Pyrenees and the Sahara is however established; the Spanish forest flora stands more closely related to the north African than to the Alpine; but deep interpenetration is shown on the one hand by the * Spanish' chestnut and walnut, which intrude from the Alpine highland, and on the other by the occurrence of evanescent *Lusitanian* types as far north as the British Isles. That the latest interpenetration was but recent is demonstrated by the comparatively narrow range of typically * Alpine' forms beyond the Garonne, and especially by the fact that some of the most important of them, like the chestnut and walnut, are notably serviceable to man. All considerations, therefore, drawn from the peculiarities of forest life, such as the neolithic skill to grind implements instead of chipping, and the exploitation of nuts and other tree fruits for food, apply in a measure to the forested highlands beyond fjhe Garonne, as well as to those of central and south-eastern Europe. In the same way, arguments based on the early development ' of agriculture along the south-eastern margins of the highland to- wards Mesopotamia and Syria, or in the Nile valley, apply also to the foothills of Atlas and Pyrenees, to the margins of the Iberian table-land, and to other well-watered coastlands of the west Medi- terranean; and account must be taken of ancient and persistent II, nj EURASIAN AND EURAFRICAN CRADLE-LANDS 69 tradition that cereals were introduced into Greece by a Sicilian goddesS, and that wheat and barley grew wild in lands of the western sea. Similarly, the domestication of animals, though certainly very early in Egypt, Arabia, and Mesopotamia, cannot be assumed to be necessarily derived from these regions. On the one side, the breeds of domestic animals at Anau in the foothills between Trans- caucasia and Iran, and in early settlements along its western park- land fringe, are distinct from those of the * Ancient East/ On the other, those of western Europe, differing from both these groups, and no less from those of the Alpine lakeland, may be independent and Eurafrican. Most significant of all, whereas the earliest pottery of the 'Ancient East* mainly copies the forms of gourd vessels, appro- priate to a region of large irrigable alluvia, traversing ill-watered and inhospitable flatland, those of the western Mediterranean, and of the whole Atlantic seaboard, exhibit intimate dependence on fine basketry, such as still supplements pottery for storage purposes throughout the Atlas region and Iberia, in fact wherever the char- acteristic esparto-grass dominates all open country, and furnishes unsurpassed material for this kind of gear. The Nile valley, lying towards the western edge of this esparto-region, participates in both techniques. In the earliest graves, pottery of indigenous stoneware models is associated both with swollen unornamented gourd-forms in polished redware and with dull brown or black plates and saucers, quite different in profile, and copiously incised with angular geometrical schemes as closely reminiscent of bas- ketry as the shapes of the vessels themselves* That this kind of evidence should come into the reckoning at all, is a measure of the gulf which separates the study of the Old Stone Age from that of the New; and attention must be drawn, at this point, to the fresh source of information as to human habits and activities which is derived from objects of baked clay. During palaeolithic times, almost the only evidence for man's mode of life is supplied by his implements of stone, and latterly of bone and antler; and for his artistic capacity, by the carved decoration of* these, and by the engraved and painted walls of his cavern- homes. Henceforward, though the successive types of implements remain of very great importance, their evidence is supplemented almost everywhere by that of pottery, more varied and far more expressive. There are special reasons for this eloquence of * pot- sherds.* First, clay is eminently plastic^ unlike stone, wood or fibre, it has no 'grain* or texture of its own; it is therefore^/?/?, and can 70 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. be modelled Into any form characteristic of the natural ' grain' or texture of any other material; all objects of pottery are therefore literally figments of the potter's -will, fictions (to vary the phrasft) of his memory or imagination. 'Hath not the potter power over the clay?' But the potter, and still more those who will use his pots, are creatures of habit. A hunting or a pastoral people, if it makes pottery at all, will make its clay vessels like hunter's game-bags, or the leathern bowls and flasks of the nomad dairy; or forest-folk will imitate wooden bowls3 or basketry; agriculturists,, strawplait or gourds. Moreover, the practice of primitive peoples suggests that sometimes pottery has originated accidentally, through" leaky vessels of these other materials,, temporarily caulked with clay, being dropped into a fire. For? plastic as it is to begin with, clay once * fired' is unalterable^ whereas many materials which it irused to replace are perishable; it may even3 in the case supposed, not merely retain the form of the basket of which it was the lining, but even the impress of the basketwork; examples of such im- pressions on early and primitive pottery are worldwide, and serve to record whole industries whose actual products have disappeared* But however indestructible in detail, pottery is so/rajf/fc, as to be practically irreparable, once broken ; consequently there is enor- mous waste, as every housekeeper knows, and accumulation of discarded fragments. It provides therefore exceptionally copious material, and as every fragment is an original work of art, the evi- dence of pottery justifies broader and surer generalizations than almost any other human document; every potsherd in any waste heap being the response of somebody's hand and brain to some- body's need, at the same time individual and communal, industrial and aesthetic. A further consequence of this fragility is that pottery is seldom carried far from the place of manufacture: its presence characterizes a settled mode of life, and signals the neighbourhood of a settlement; though on the other hand, the absence of pottery from any district is no proof of the nonexistence of a nomad popu- lation* The utter uselessness of pottery, once broken, except as extemporized scrapers, or as builder's ballast to level a new floor, is the main cause of its archaeological value; for where broken pottery is cast out of a settlement, there it is allowed to lie and .accumulate, layer over layer, later over earlier; so that the 'se- quence-dating' derived from such a rubbish-heap is as secure as the sequence of the fossils in the sedimentary rocks, and of the highest value as evidence for changes of style3 that is to say* of the notions, industrial and aesthetic, of successive generations oir makers and breakers of pottery. As breakage and replacement arc con- II, in] MIGRATION AND CONFLATION OF STYLES 71 stant, clay almost ubiquitous, and pot-transport risky, the pottery- series ift. any settlement is exceptionally continuous and coherent; the'smallest changes of style are recorded infallibly, directly, and immediately; and every other object cast upon the same waste- heap is conserved automatically in stratified order, and can be dated by the potsherds around it, between older ones below3 and later ones above. See p. 1 1 3 sq. Further, being ubiquitous and plastic, clay is also cheap. It is the poor man's substitute for materials which he cannot afford. It may therefore record not only the equipment of daily life, but the fashionable shapes of articles of luxury, such as vessels of gold and silver* It is also the mean man's, subterfuge on occasions of cus- tomary sacrifice; to equip the dead, for example, with a cheap and durable imitation of valuable originals retained for the use of sur- vivors* This is the special interest of all funerary pottery, for it correlates each isolated * tomb-group* with the waste-heaps of the settlement to which it belongs. In general, therefore, the potsherds of any people record continuously and accurately the general cul- ture and style of successive periods; the local and daily variations both of needs and of the satisfaction of them; the more abrupt innovations resulting from intercourse with neighbours similarly recorded; and the revolutions due to immigration. Conquest, in particular, may leave its memorial in wholesale destruction, and clearance of the debris of war, and in collateral production, after- wards, of objects in distinct fashions — 'peasant style' and 'palace style' — for the respective use of old compatriots and new masters. And as it usually happens in simple societies that pot-making is women's work, supersession of an indigenous by an immigrant style is strong presumption that the newcomers brought their own women with them; whereas, if they intermarried with the natives, there may be perplexing combinations, for example of indigenous shapes and technique with imported ornaments, to please the master's eye. In either case, eventual coalescence of racial or social elements may be signalled by the rise of a new mixed style, which is sometimes of striking originality (see pp. 81, 87, 90, 97, 101). III. REGIONAL TYPES OF NEOLITHIC CULTURE: ALPINE EUROPE We have seen that the first indication of the westward spread of a new variety of man is the appearance of broad-headed indivi- duals, side by side with Aurignacian long-heads, in a remarkable burial-place in an Azilian cave at Ofnet in Bavaria, and in caves at 72 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. Laugerie Basse and elsewhere in central France. Isolated ex- amples in several districts of Spain, of uncertain and pfobably rather later age, may indicate that the spread of forest conditibns had carried the same human movement far to the south-west; but allowance has to be made here for the effects of a later sea-borne movement which will be discussed in its place below (p. 104), Next, and far more significant,, because associated with fresh elements of culture, is the broad-headed population of the pile- dwellings which occupy all the principal Alpine lakes* Here or- ganized communities were occupying settlements on artificially constructed platforms supported by wooden piles driven into the lake bottom, and communicating with the shore by a gangway. These communities maintained themselves both by haunting and fishing, and by collecting wild fruits and nuts from the forests: But they also practised agriculture from the first, and must therefore have brought this art with them when they first ventured into lake-land* Their wheat? barley, millet and flax are of the same species and varieties as were cultivated in the earliest known settle- ments on the Nile alluvium., and in the earliest stratum at Anau. Oats and rye, on the other hand, they did not grow, though the wild plants have a wide range in Europe* Perhaps they gathered them wild,, as people still gather them for food In outlying villages of Germany; but If so, it is odd that no grains of them fell overboard. Domesticated animals only became known here later; and this again corresponds with the sequence of events at Anau, Their implements include harpoons, perforators, and scrapers of botie and deer-antler perpetuating Magdalenian and Azilian forms, flaked flints like those of Azil and Taraenoisc, and especially many miniature flakes^ one use of which is here demonstrated by their occurrence mounted lengthways like saw teeth in wooden hafts* Early Egyptian reapers used sickles of the same construction* But along with these are numerous implements formed from natural pebbles of compact stone* selected for oval or cylindrical form, and improved either by splitting them longitudinally or by grinding a naturally wedge-shaped end on one or both of its faces to form a cutting edge. In^these tough or granular materials flaking is almost impracticable. Similar pebble-shaped implements with ground and polished edge are found in many regions, in surface soil and other post-glacial deposits, and mark the beginning of the New Stone Age wherever they occur. Gradually the grinding and polishing were applied to the whole surface of such implements to improve their symmetry, but it was long before oval or tapering pebble- shape at the butt-end was replaced by flat sides and more or less ll^ni] THE LAKE-DWELLING CULTURE 73 rectilinear profile. Flat edges In particular are a mark of advanced technicfue, and comparatively late date. Another important innovation is the perforating of hammer-stones and eventually of hammer-axes, effected first with sand and a blunt stick; later with a tubular drill of reed. A similar drill was used along the reed-fringed Nile at an early predynastic stage. The frequent fires to which the pile-dwellings were liable must have familiarized their occupants with the effects of fire on clay, even if they had not this knowledge already; and their pottery, which is found even on the earliest sites, Is so primitive that It may well be original. Much of It is clumsy and formless, but the more shapely pots take their forms almost exclusively from leather vessels. Atjfirst there is no attempt at ornament; later, modelled rims* and ridges, and roughly scratched patterns betray the influence of basketry and textiles, first of local, and later also of non-Alpine styles from the Rhine and Danube basins. The earliest lake-dwellers buried their dead ashore, In earth graves or slab-lined cists. But at a quite early stage it became cus- tomary to burn the bodies, and bury the ashes, with such personal ornaments as endured the fire, in a rough clay pot, closed with a saucer. As this custom of cremation destroyed direct evidence from skeletons, it prevents positive conclusions as to later changes of race; but, by drowning and other accidents, enough individuals escaped a regular funeral, to justify not merely the view that the population of the lake-regions remained broad-headed through- out, as it still essentially is, but the hypothesis that in most early periods peoples who burned their dead were probably of broad- headed ancestry. Later exceptions will be noted and discussed as they occur (pp. 81, 101). The sudden and widespread establishment of the lake-dwelling culture and of the broad-headed type almost explains Itself. From Its very construction, a lake-village could not be expanded inde- finitely, and consequently if its home population outgrew, super- fluous members had to go elsewhere and construct a fresh one. And as the Alpine lake-basins are interconnected both downstream and by passes between valley-heads, what may be described as longitudinal propagation was easy, both within the highland and along the rivers which issue from it. And in fact such movements have been traced, into the Danube valley, and beyond it into the rivers of the north German plain and the peat-mosses of the Danish peninsula; along the whole course of the Rhine and widely over northern France, Belgium, and Britain, where lake-villages were numerous, especially in mosses and bogs in Scotland and 74 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. Ireland. Some of them remained in use until Roman times, as at Glastonbury, and even later still, changing their industries and arts, but not their structure or organization. c On the steep southern face of the Alps, the abrupt transition from highland to alluvial plain checked such expansion for long; but towards the end of the Stone Age a sudden movement spread lake-dwellings from Lakes Como and Maggiore as far as the main channel of the Po; a little later, when bronze was already in occasional use, a similar but more vigorous emigration from the eastern Alps occupied all the lower valley, crossed the main river, and advanced, in the specialized stream-bed settlements known as 'Terremare* (from the use made of their debris as a fertilizer by the modern peasants),, as far as the passes of the Apennines. A few adventurous parties passed on into lower Italy, and one such settle- ment exists close to Taranto. Reserving the details of this pro- foundly important movement, for the chapter on prehistoric Italy, in VoL ii, it must be noted here that by bursting the triple barrier of Alps, fenland, and Apennine forest, which had hitherto se- cluded Italy, this migration of lake-dwellers established a con- tinuity of race and of culture between that peninsula and the tributaries of the middle and upper Danube, which has had profound influence throughout all later ages. It is a commonplace that the history of this peninsula is that of * Italy and its Invaders'; and the first of these invaders arc the Alpiae lake-folk, and their descendants in the *terremare* villages. Eastward, subsequent changes have been so numerous and far-reaching, that equivalents of these Alpine lake-dwellings are not easily found. Quite early examples occur as far cast as Laibach; later settlements widely on suitable sites throughout the Hungarian lowland; and the influence of their culture extends as far north as Bohemia, In Bosnia, the remarkable settlement at Butmir, and less famous sites in the same region, illustrate special adaptation of the pile-structure to dry valley-bottoms, in many ways analogous to Italian *terremare/ Herodotus, iti the fifth century B.C., graphically describes the pile-dwellings of Lake Prasias in the Strymon basin ; and the occurrence of pottery of typical lacustrine and *terremare' forms, in this part of Macedonia and elsewhere in the Balkans, confirms and amplifies his testimony- Further afield again, ancient descriptions of pile-dwellings in waterlogged valleys of North Syria and Georgia, unverified as yet by excavation, suggest that our Alpine lake-settlements are to be regarded as a westward section of a very large region of early and essentially homogeneous culture, adapted to the conditions of a II, IT] REGIONAL TYPES: THE DANUBE BASIN 75 moist forest-clad lake-land, such as Asia Minor and much of the highland region eastward of it must have constituted during the long * pluvial' period which was the counterpart of the Ice Age in Europe. The same climatic changes which have restricted the forests and displaced north-westward the forest-fauna and forest- type of Man have not only disrupted this area of culture, but have also destroyed much of the evidence of its former extent; for the torrential discharge of the modern seasonal rainfall has scoured out most of the alluvium from the valleys, leaving only the numer- ous early types of polished implements, — in which this whole region abounds, though it is apparently devoid of chipped flints1, — to testify to the former existence of such a culture. The actual area of continuous lake-dwelling culture has thus beei» very much reduced by adverse physical changes, aggressive from the south-east. It has also been superseded, both on this side, and around the margins of its Alpine citadel, by other types of culture better accommodated to these changes, which (as we have seen in this extreme instance) have been on the whole by way of less moisture and greater warmth, and consequent curtailment of lake-land and forest. IV. REGIONAL TYPES: THE DANUBE BASIN The first great change indeed, affecting the lake culture itself, is typical of what was going on. This is the introduction of do- mesticated animals: and as all these, in the Alpine lake-villages, are of breeds not derived from the wild species of the region, but identical with domesticated breeds of the Near East, ancient and modern, and with some of those known in neolithic Egypt, it may be inferred that their arrival in central Europe results either through exchange from tribe to tribe from the south-east, or through direct immigration of pastoral people possessing such flocks and herds. Both would be impracticable as long as a dense and continuous forest covered south-eastern Europe to the Car- pathians and the Hellespont. Either would be comparatively easy %$ soon as a drier climate, with more seasonal rainfall of the Hel- ladic and Mediterranean type2, began to break up the forests into 1 Occasional reports of such implements, of Mousterian type, belong to a period so much earlier, that if verified they would not affect the impression created by the dearth of anything later. 2 Mention must be made here of the cardinal discoveries of Roumanian and Russian pedologists as to the sequence of climate and types of vegetation along the outer face of the Carpathians and on the adjacent steppe. 76 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. parkland and meadow; a process which is always accelerated by the presence either of porous limestones like those of the Dinaric and Balkan chains, or of the loess deposits which (as we have s£en) cover so much of the lowlands of Hungary and Bulgaria, as well as of Rouznania, Galicia and Ukraine, Similar loess deposits occupy large areas of Moravia and Bo- hemia, of the Upper Danube and its northern tributaries, of the Neckar, Main5 and middle Rhine, West of the last-named river lay the north-and-south barrier of the Jura, Vosges and Ardennes, accentuated by the denser woodland which is fed by the wet winds of the Atlantic seaboard, and still clothes their western slopes and masks the passes between them. These loess-lands had been the last prairie hunting-grounds of Magdalcnian and A/ilian man, as the wet forests of the transition period closed up, alon^ the highlands to the north, from the Carpathians to the Taunus and the Black Forest. When the neolithic period opened they were still occupied by long-headed folk only slightly modified from the late-palaeolithic Aurignacians, but distinguished from their rela- tives north and west of those forested highlands, by a fuller oval headform, less angular, and associated with other characters which persisted long, and only gave way gradually before the later ex- pansion of broad-headed Alpine foresters. Here were all the con- ditions for the spread either of pastoral or of agricultural folk, so soon as the loftier, and therefore more forest-bound regions to the south-east became passable. The evidence for such passage, and for the period at which it was achieved, is as usual twofold: from the copious relics of a new and distinctive culture, and from the physical remains of the people themselves during and after its introduction. The earliest neolithic culture of these wide and interconnected lowlands, from the Balkan lands to the headwaters of the Danube, and. the basins of the Neckar, Main, Upper Elbe and Oder, is curiously uniform in type. There are regular settlements in the open valleys, usually grouped in clusters within reach of a con- siderable stream* They combine pastoral with agricultural life, and possess the same primitive crops as the Alpine lake-folk, and the same herds as the lake-folk acquired eventually. Their habitual implements are of the split-and-ground pebble type, but show a characteristic improvement on those of the lake-folk, and of the great south-eastern region of the Balkan lands and Asia Minor, in that they are ground nearly flat on one face,, and only left rounded on the other. For timber«working3 and still more for hoeing, this adze-like celt, set transversely in its haft, had obvious 11,1V] DANUBIAN 'BAND-POTTERY' 77 advantages, as with careful usage it maintained its own edge, like the shell-adzes used for sago-getting in Melanesia, and the convex maftock of the Levantine peasant. The pottery of these settlements consists mainly of small globular vessels, rather more than hemi- spherical, rounded below, and usually without rim or handle. There is no hint of imitation of any kind of structure such as leather or basketry, and the outer surface, smooth and uniform, is treated as a single open field for a continuous scheme of decoration which returns into itself, and has earned for this technique the nicknames of ' band-pottery' (Eand-keramikj Ceramique a rubans)^ and of the * free-field' style. The designs are rendered by continuous lines incised in the clay before firing; either rectilinear zigzags, or curved into lobes,^waves, or coils, sometimes rather complicated, and always quite irrespective of any limits but those imposed by the general shape of the pot. There is no attempt to emphasize or dis- tinguish its parts, for indeed it usually has none: at most there may be a collar-band following the edge of the opening. This is so different from the commoner 'skeuomorphic ' decoration of an object by enhancing its natural texture or structural elements (for example in the * western ' and * north-western y styles to be described later), and so closely resembles the * free-field7 ornament employed by those modern peoples who make their vessels of gourds, — whose natural surface is uniformly smooth and of imperceptible texture, — that it has been suggested with much probability that the Danubian * band-pottery* likewise originated so, and that consequently its origin must be sought further south-east in regions, such as Asia Minor and Syria, where gourd-plants occur naturally, and have been in immemorial use, as the earliest pot-fabrics of these and adjacent regions attest. Here again allowance must be made for the known shift of climate, and account taken of the remarkable gourd-types of the first pottery of Cyprus, and less distinctively of certain early fabrics in the Cyclades and Crete, which like Cyprus lie under the lee of this continental area; and also of a distinct gourd-element in the neolithic pottery of Egypt, which is not aboriginal there, but intrudes itself at an early phase among indigenous forms mainly derived from vases of stone. Once again it looks as though we were witnessing such an exodus from Asia Minor, both to south-east and to north-west, as we have already had occasion to infer as a probable consequence of the desiccation of this section of the Highland Zone, and as indicated by the distribution of the varieties of broad-headed man (p. 61 sgl). We have only to add, to complete the evidence at present available, that it is during the 78 NEOLITHIC AND BRONZE AGE CULTURES [CHAP, period represented by the neolithic * band-pottery ' that we have the first Indications of the spread of broad-headed man 'among the population of the Danubian region. Northwards, as we have seen, this Danubian culture occupied the loess-lands of Moravia and Bohemia., and reached the middle Rhine- It also influenced temporarily a large area beyond it in the direction of Belgium* But as the heavily forested ridges of the Carpathians and the central German highlands limited its north- ward range, so the Vosges and Jura barred extension westward, and it was not long before all its Rhine-ward provinces fell under alien influence from the north-west, of which account will be taken later (p. 9 8 jy.). Southward, its influence is clearly perceptible in the later technique of the lake-dwellers; but as it never affected Italy, the migration of the 'terremare '-builders must have oceitrred before this phase. Further to the south., the large western tributaries of the Danube, and especially the Save, received this culture early and developed it in a rather special fashion which makes the results difficult to correlate with the main Danubian types* Not enough is yet known of this district as a whole, to determine whether the remarkably rich settlement at Butmir in Bosnia is typical or not, nor to assign it to its proper phase; but it seems certain that the spiral ornaments, extraordinarily varied and beautiful, which were in vogue there, are on the one hand a local and perhaps spontaneous elaboration of the curvilinear elements common to nearly all schools of the * free-field* style; on the other, that the Butmir style of pottery, once established, was in wide demand (as actual exports show) and had a range of influence even wider^ from Thessaly and Macedonia to the Carpathians, and eventually far beyond towards the Dnieper, It has even been thought, chiefly by Teutonic ob- servers., that the spiral decoration which became so popular in the Minoan Bronze Age of the Aegean may have resulted from con- tact with this Bosnian school; but the contrary view is widely held, and until the relative dates of Minoan and of Bosnian culture are better established, this question remains open, It may even be that the Bosnian culture, lying so near as it does to the Adriatic coast, may have stood in more direct relation than is usually supposed to the neolithic art of Malta and the west Mediterranean., which also makes striking and very early use of spiral decoration; but here too intercourse cannot be asserted yet; priority even less. Though the general culture, and especially the technique of implements and pottery, of the whole of this Danubian region shows generic similarities^ each principal district developed pecu- II, iv] SOUTH-EASTERN EUROPE AND ASIA MINOR 79 liarities of Its own, of which those of the Bosnian area are only an exceptionally striking example; and these idiosyncrasies became more marked as time went on. It may be inferred, first, that the various groups of people were on the whole sedentary, as their agricultural habit suggests. Then, from the very gradual spread of broad-headed folk, among a mainly long-headed population, it would seem that this type of civilization spread rather by inter- course than by conquest; from the open situation of the settlements and the rarity of weapons of offence, that they were in no great fear of disturbance; and from the frequency of their villages and tombs, and the repeated reconstructions of their huts, that this peaceful development lasted a long time. The sam£ gradual and pacific advance characterizes also the next ftoteworthy change. As long as culture remained purely neo- lithic, and in most parts for some while after, the pottery, if it shows any designed interference with the natural colour of the clay, is baked black with the aid of a smoky fire, or of charred vegetable matter in the clay itself, or of a dressing of graphite. The surface is burnished by friction, and the incised ornaments are eventually enhanced by a filling of white earth. But about the time of the first introduction of copper, an improved method of firing came into use which took advantage of the presence of iron oxides in the clay to produce a brick-red surface, or imitated this by a wash of more ferruginous clay. Burnishing and white-filling went on as before. Now the earliest copper objects, — flat axe- blades, leaf-shaped daggers, awls, and dress-pins, — repeat with only slight variation the forms characteristic of the earliest metal- age in predynastic Egypt, in Syria, and in Cyprus; and the infer- ence that the Danubian region was acquiring its higher industries from the south-east, by way of the Hellespont, is confirmed by the fact (p. 8 9) that all over Asia Minor, similar but more emphatic replacement of polished black-ware by red-ware accompanies the spread of metal-working. This is well illustrated in the stratified site at Hissarlik on the Hellespont, where the first city has the black-ware and the second the red-ware technique. And the fact th&t these two settlements are separated by a layer of natural soil, showing that this site was for a while uninhabited, confirms the general impression that whatever intercourse there may have been, om Asia Minor into Europe, at an earlier stage, it had ceased for a while, and was renewed (and with it the importance of the Hissarlik site) when the new metal-traffic was becoming frequent, and was eliciting a return traffic in amber from the Baltic shores. As the second city seems from its contents to have been destroyed So NEOLITHIC AND BRONZE AGE CULTURES [CHAP. not later than 2000 B.C. and to have existed for a long while before that catastrophe, we have here a rough lower-limit of the period within which this traffic was established; and the foreign objects found in this second city give further cross-references to the cul- tures of other regions, as far afield as Sicily and Malta (p* 97)^ and the third civilization of Anau (p, 87), V. REGIONAL TYPES: THE TRIPOIJE CULTURE To present intelligibly the next two phases of the neolithic cul- ture of Europe and the crises which introduce them, it is necessary to range further afield3 into regions hitherto unaflrected, so far as is known, by the emergence of broad-headed man either in Asia Minor or in Alpine Europe. His relations with the Syrian highland, and with Egypt, have been discussed already, and the circumstances which hindered his general extension along the North African coast. We shall see later by what stages his culture, though not necessarily his race, passed south-eastward and eastward into the region of the ancient * painted pottery5" culture of Susa and Anau (pp. 85 jyf.). And we shall see that there is reason to believe that the site of Anau reveals that * painted7 culture in oscil- lation between the highland and the northward steppe, and in- debted for the technique of its forms, as well as of its ornament, partly to wood-using foresters, partly to leather-using pastorals from the steppe or its oases. It does not need much imagination to suggest that a steppe- or oasis-culture of this kind is unlikely to be confined to one section only of the steppe-margin; and that it is most likely to be recovered at any section of that margin where the steppe is bounded by mountain and forest as abrupt as the Kopet Dagh above Anau, and as liable to oscillations of climate, and alternate advance or retreat of the forest and its parkland fringe* Such conditions actually occur on the eastward face of the Car- pathians, and the Roumanian and Russian students of what is for those countries a problem of high practical importance to the national economy have demonstrated such oscillations throughout post-glacial time; though they have not yet established correlation in detail with those exhibited at Anau* It was therefore no surprise to geographers when the discovery was announced of a * painted pottery' culture on a number of iso- lated sites distributed oasis-like over the trans-Carpathian steppe, between the Dnieper and the Danube,, and supplemented by two other groups, one along the north side of the Carpathians, through- out Galicia> the other occupying sites in SiebenbUrgen on the II, v] « PAINTED-WARE' CULTURE OF TRIPOLJE 81 reverse flank of the Carpathian arc, as Susa and Moussian stand on the 'reverse flank or the north Persian highlands, looking over Mesopotamia and exploiting its lowlands, just as the cis-Carpathian sites spread down from Siebenbtirgen into the Hungarian plain. Like the culture of Anau, the Tripolje culture (so called from the best known of the trans-Carpathian sites) has two main phases. In both, the dead were burned, and it has been inferred from this that the people were of 4 Alpine' origin; but this does not neces- sarily follow, and the racial question may be left open for the present. The first phase seems to be purely neolithic; its decorative painting, like the first style at Anau, is simply geometrical; and it seems to beiimited to the flat land, except one brief incursion into the effrlier neolithic culture of Thessaly, with which it is at present linked only by a few casual finds in Macedonia and Bulgaria. The second, which is separated from the first by a considerable pause, during which sites were evacuated and reoccupied, as at Anau, shows marked development of its vase-forms, and still more de- cided change in its decoration; for in the interval it had acquired an empirical, though not very intelligent, acquaintance with the curvilinear ornaments of the Danubian ^band-pottery/ at a period when the latter was already strongly influenced by the Bosnian spiral designs. The painted trans-Carpathian spirals, however, never reproduce their prototypes with the close understanding of their geometry which characterizes the Bosnian school, but are Introduced haphazard in bizarre confusion among the triangles and other linear schemes which were already traditional. After a fairly long existence, to judge from the depth of de- posits on the Galician and Roumanian sites, — though there Is nowhere the vast depth of debris which Is common to Anau and Susa, — the Tripolje culture ceases abruptly and uniformly. Its sites were deserted and not reoccupied; and the cause of their evacuation is indicated by the occurrence, over the whole region of their distribution, of burial tumuli in a late phase of the neolithic culture ascribed by Russian observers to the * kurgan-folk1 ' or *rgd-ochre-people? (see below p. 83), who had long been in occupation of the central steppe, but seem to have been held aloof from the Tripolje folk along the course of the Dnieper. The occa- sion, and the cause, of their irruption can only be guessed, for it 1 Kurgan is a local word for a burial mound. These people will be herein- after described as the 'Tumulus-folk.' The red ochre with which they be- smeared their dead has been thought to be a survival of palaeolithic, perhaps Solutrean, observance (see below, p. 83), C.A.H.I 82 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. cannot at present be correlated exactly with other events, though, curiously enough, pottery resembling the later Tripolje style ap- pears suddenly in Thessaly? at a longish Interval after the "first incursion (p, 81), whether we measure it in phases of the local styles, or by the depth of superimposed settlements. It may re- present an arrival of dispossessed folk from beyond the Danube. It is certain, however, that once let loose on Roumania the * tumulus-folk' were checked westward only by the Carpathians, and that southward they crossed the Danube, spread their tumuli widely over Bulgaria and Thrace, and penetrated into north-western Asia Minor, where their tumuli overlook the 1 lellespont and follow the Sangarius valley as for as the Phrygian plateau. It has been suggested, with some probability, that it was they wJ?o destroyed the second city at Hissarlik; at all events one skull from this city, wholly different from its contemporaries, closely resembles the * tu- mulus-folk* type; and if so, their irruption would be approximately dated not later than 2000 E.G., and would range with other great movements (pp* 91, 107) which were in progress about that time* While the left wing of this irruption from the steppe swung southwards in this fashion, the right or northern wing pressed on outside the Carpathians, scattering the Tripolje folk of Galicia into Silesia, Moravia, and Bohemia, The effects of this movement must be followed at a later stage (p, 101). Other survivors of the Tripolje culture seem to have taken refuge with their cis-Carpathian kinsmen and to have introduced disorganized elements of their culture* and especially their painted decoration, rather widely within the Dnnubian region., from north- ern Serbia to Bohemia* Perhaps this dissemination had already begun, from the cis-Carpathian sites, for the relative dates are uncertain; but the general similarity of these derivative painted techniques rather points to a single impulse, of not very early date. VI. THE CULTURE OF THE NORTH-EASTERN STEPPE While in western and north-western Europe the passing of the Old Stone Age can be traced in- fairly full detail,, the record^ is as yet less copious in the east, Roumanian and Russian studies of post-glacial deposits make it certain that the deposits of loess which indicate dry steppe and desert conditions, though generally con- tinuous,, were interrupted several times by moister periods which allowed soil to form, far out from the Carpathians towards the Dnieper. Nearer the Carpathians, and further south towards the Balkans, not only are these layers more numerous, but they can be II, vi] CLIMATE AND CULTURE OF THE STEPPE 83 correlated with the flood-wash of the lower Danube and other Roumanian rivers, and with other soils so richly impregnated with vegetable matter that they are regarded as evidence for forest, like that of the Carpathian foothills but extended for some distance into the plain. The forest regime attained therefore here too, and more than once in post-glacial time, a wider extension than now; and the changes of climate which this presupposes are indicated also by wider distribution of swamps and other shore-deposits of an enlarged Black Sea. As similar oscillations are established on the low ground between Black Sea and Caspian, at Anau, and around the southern foothills of the Ural range, it may be inferred that the old Sarmatian sea-basin still exercised its moderating in- fluence over^the whole Eurasian lowland, whenever the westerlies shiftefl far enough north to supply it with rain. But these oscillations only affected the margin; and meanwhile it was only gradually that the Asiatic forest, already mentioned in other connections (p. 24), was enabled by the shrinkage of the Scandinavian and Ural ice-caps to spread round the northern edge of what seems to have been continuously steppe or desert, at all events in its central area. This region has already been suggested (p. 51 sg?) as the probable reservoir of the Solutrean hunters who intruded into western Europe at the end of the Aurignacian period, and as their probable refuge when they withdrew, with the steppe-fauna, when the Magdalenian moisture set in. That they did not permanently lose access to central Europe is clear from the occurrence of similar long-headed individuals in the Ofnet burial-place, mixed with early representatives of the new broad-headed folk of the Alpine forest region; and that their culture penetrated atone time right across the Iranian section of the Highland Zone is suggested by the discovery of implements of the peculiar Solutrean technique on sites over- looking Mesopotamia, and even in the Nile valley; though the occasions of these deposits cannot yet be dated. Many of the later palaeolithic folk on west- and mid-European sites had the habit of supplying their dead with a quantity of pojvdered red ochre ; in what belief as to its efficacy we can only guess from occasional instances of the same custom among modern savages; there is obvious symbolism in so durable a representation of blood. It is therefore of the first importance, that the same practice is habitual among the earliest inhabitants of the Eurasian steppe, a tall, heavy-built and long-headed race not very different from those western types, burying their dead in surface graves, and marking these with earth-mounds, the only possible monument in the tree- 6—* 84 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. less and stoneless loess-land. These mounds (for which the local word is kurgan) do not seem to begin until the fine Solutrean tech- nique had been lost, and their earliest contents are more roughly worked implements, and hemispherical pots of clay — durable sub- stitutes for the simple bowls of gourd or leather, available to a prairie folk. As horse-bits, and later on, fragments of wooden cars on wheels, are found in these mounds, we must infer that the horse had been domesticated, and that we have here an early phase of the waggon-dwelling culture which still occupied this grassland when it was visited by Greek explorers later on. Dates for these two inventions, locomotive animals and wheeled transport, cannot as yet be fixed; but they presuppose a combination of level un- obstructed country, with the presence of the wild hon?e, and access to parkland timber supply, which is nowhere so fully realised as in this region; and at no period so favourably even there, as in the late palaeolithic phase of moist climate, and consequent encroach- ment of such parkland far out into the steppe wherever its dusty soil was tolerant of trees. At the climax of the moist phase we have probably to picture this region wholly grassland at its centre, wholly encircled by forest, and with its southern half invaded by the swampy shores of a continuous Ponto-Caspian lakeland. It is not the purpose of this chapter to deal at length with the history of language, for though the periods of this may be as- signed an order of sequence, they can seldom be dated, because words, unlike implements, do not fall to the ground after use, But it may be noted here that the population of a region so long secluded, so vast in itself, and so absolutely devoid of internal ob- stacles can hardly have failed to acquire a fairly uniform vocabulary for such elements of their common experience and culture as the open sky, sun, moon and stars; open water, with some sort of boat, and swampland with geese and ducks; open grassland, with cattle and horses; but also parkland trees, with axes to fell them, and gourds for vessels; and the structural details of a waggon- home for its journey over paths and fords. Whether such people had also knowledge of the simpler agriculture would obviously depend on eventual intercourse with kindred men of the parkkyid, or with some other culture on the forest margin or beyond it; for on the grassland itself, as every nomad knows, even to scratch the surface may be to wound irremediably the delicate film of vegeta- tion on which depends alL Such vocabulary seems to have been among the oldest common possessions of Aryan-speaking folk; and there is now general agreement that whatever their subsequent adventures, the original speakers of this type of language probably II, vn] EARLY MAN ON THE MARGINS OF IRAN 85 Inhabited this region; while some observers go so far as to identify tbftm with these c tumulus-folk/ VII. THE CULTURE OF ANAU AND SUSA We reach next, in our survey of early neolithic cultures, the eastern section of the Highland Zone, separating the northern steppe from the lowland of Mesopotamia, where the earlier phases of civilization have been already noticed (pp. 42 syg."). In this eastern section the record is still fragmentary, in spite of the brilliant work of the French Mission in Persia. Palaeolithic culture, of the normal types, has not been detected. This is what would be expected, if the southward shift of climate clothed its abrupt escarpments on eithei^hand with forests impenetrable by hunting folk. At most, during a climax of drought, there might be occasional incursions, such as the rare occurrence of Solutrean types of implements to the south-westward has suggested already. The survival of dark- skinned folk akin to the older races of India and beyond, in the more extensive and better watered ranges of the south-western margin, suggests that this region long retained the character rather of a westward appendage of that great south-easterly region, than of an eastward extension, either of 'Africa- Arabia,' or even of the Highland Zone; and the intense glaciation of the Armenian and north-west Persian mountain-knot gives a reason for this long isolation from the west. Scanty human remains from Susa and other sites on its south-western margin, and from the ancient site at Anau, on its northern, agree in supporting this notion of its early human population, and at Anau in particular there is evi- dence that this type persisted, even where it would least have been expected, far on into early historic times. Further south, the same type seems to be figured in the Sumerian art of Babylonia early in the third millennium. The remarkable analogies between the earliest culture at Anau and on the Susan group of sites, need not therefore surprise us, nor the remote antiquity to which this common culture appears to^o back with essential continuity of development; for the lacuna between the second and third phases at Anau has been shown to be due to an encroachment of northern steppe-desert, which evacuated the site temporarily without destroying the civilization of which Anau was seldom more than an outpost. The unusual depth of continuously deposited debris on all these sites — at Susa 27 ft. for the first culture and nearly 50 for the second; at Anau 45 ft. for the first and 40 for the second — is 86 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. presumptive evidence for very long duration, unless the contrary can be proved. We may compare the 25 ft. of pre-Minoan neo- lithic debris at Cnossus, and the occurrence of pottery at 50—60 ft. below the present flood-plain of the Nile* The second culture is succeeded, after an interval of desertion, by a third, 59 ft. thick, the later part of which contains objects not much later than 2000 B.C. Without accepting therefore estimates based on rate of accu- mulation on later sites in other regions, it is permissible to regard the beginning of the Susa-Anau cycle of civilization as falling within the same scale of time as is indicated by the Baltic sedi- ments for the close of the European Ice Age* Comparison of the most recent reckonings reveals indeed very striking similarities. Breasted, relying on the actual r&tc of alluvial deposit in Egypt, dates the beginning of the present Nile alltSrvium, and the first human occupation of it, 60-80 ft. below the modern surface, to about 18,000—15,000 B.C.; a second * floor* of occupa- tion (at 35 ft.) to about 10,000 B.C,, and the earliest tombs still exposed along its edge to about 4000 B.C. Baron de Geer's study of the annual increment of laminated clays in the Baltic area sug- gests 20,000 B.C. for the retreat of the Scandinavian ice from the north German plain; 15,000 B.C. for the release of the south end of Sweden (which very soon received Solutrean immigrants from the south) and 8000 B.C. for its northern districts. Pumpclly and Huntington begin the first settlement at Anau, in south Turkestan about 9000 B.C.; the second,, which succeeded it, about 6000 B.C.; and the third, after an interval of desert-drought, about 5200 B.C.; ending with another drought about 2200 B.C. De Morgan and Montelius allow 20,000 years for the whole series at Susa; Evans and Montelius 14,000 for that at Cnossus, The drought which evacuated Anau between 6000 and 5000 B*c* would thus corre- spond with the period of elevation and more continental climate in the Ancylus period of the Baltic area, by which time the Eurasiatic tundra and forest belt had completed their fusion with the west European, and were allowing Mongoloid folk to penetrate into the Baltic area. See also p. 579* The material cultures of Susa and of Anau present close sia^i- laxities. That of Susa is described below (pp. 361 $$$•'). At Anau the first culture in the lower part of the * North Kurgan" site begins likewise with hand-made pottery, of simple but shapely forms based partly on leather work, partly^ as usual, on a still older pot- fabric; the decoration, carefully applied in dark paint, is borrowed from other techniques, and is already so conventional that its ancestry remains doubtful. The material culture of these folk, that II, vn] FIRST AND SECOND CULTURES AT ANAU 87 is, must be considerably older than anything deposited on this site. The principal implements are small flint flakes, probably for insertion in a wooden haft, like those which appear in western Europe late in the palaeolithic decline, and at the beginning of the Alpine lake culture; and perforated mace-heads fashioned from pebbles of hard rock, such as occur in the earliest Nile-valley settlements, and also in lake-dwellings in the Alps. The huts were of mud-brick; their rectangular plan suggests the use of timber for roofing. Spin die- whorls attest the arts of spinning and weaving. Wheat and barley were cultivated from the first; but the earliest bones of ox, horse*, sheep and pig are those of wild species, like the gazelle and red deer with which they are associated. There are foxes and wolves, but no dogs. Gradually, however, ox, pig, horse, and ^wo kinds of sheep were domesticated into special breeds. The occurrence of small objects of turquoise, and of copper and lead, in the later phases of this first culture, shows that in some region with which Anau had intercourse these mineral resources were already exploited; but proves little or nothing as yet as to the rela- tive date of objects at Anau itself. The human remains, which occur at all depths, are long-headed: without accepting as more than provisional the first descriptions of them as *negrito' or 'Dravidian' they may be taken as proof of the extension of a south Asiatic type over the west Iranian plateau and its mountain rim. A notable observance of these people was the burial of young children beneath the house floors. The later part of this first culture lies in a phase of gradually increasing drought; and the second culture, which succeeds it or (more accurately) invades it rather suddenly, brings little change in essentials. Sling-stones became common, stone pivots for the doors, and baking-ovens made from a large pot, Lapis lazuli and cornelian supplement turquoise, and daggers of copper are found. Agriculture proceeds as before, but the camel, goat, a new horn- less sheep, and the dog are added to the domestic animals. This, and the new fabrics of pottery, of smooth red or grey ware, un- decorated except for dark smoke-mottling on the red ware, sug- gest wider intercourse with another, and in the main more south- westerly region. This is just what would be expected if drought had disorganized the forests of the highland at its narrowest point, namely between the Caspian and Mesopotamia; for we may re- member that one of the earliest fabrics of pottery in Syria and Asia Minor, is a red-ware with various blackened by-products (p. 79; cf. p. 89 below), and that a similar fabric appears in predynastic Egypt (p. 34). 88 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. At the end of the second period, Anau had become so dry that the site was abandoned. When it was reoccupied the settlement was not on the old pile of debris, but on a lower mound a litfeie to the south. The people of this * South Kurgan' and their habits were the same as of old, including the practice of child-burial; but their pottery was now wheel-made and kiln-baked, and its decora- tion, painted as at the first, was more elaborate; the painted and the red-ware styles, moreover, have been combined in subsidiary fabrics; the red-ware and grey-ware have incised ornaments like the earliest pottery of early Asia Minor and Cyprus., and some of the forms recall those of early North Syrian fabrics. Elaborately incised clay figures of women, cattle, and wheeled carts indicate fresh contact with the grassland to the north, and with the North Syrian culture far to the west. The shapes and ornaments qf the spindlewhorls have a general resemblance to those of Cyprus and Hissarlik. Copper is supplemented by occasional bronze, and the daggers of the second culture by sickles, lances and arrowheads, There are also arrowheads of flint and obsidian, and ornaments of marble, alabaster, and blue-glazed paste like that of Egypt* A single gable-shaped seal-stone with its surfaces engraved respec- tively with a man and two winged griffins is another link with the Syrian culture, and has even been claimed as of Cretan type* This third culture also was expelled from Anau by a dry spell, more severe than the former one,, and the pause was long enough for the deserted mound to be devastated by rain-wash, till the climate improved once more and a fourth culture brought iron objects to Anau, probably not much earlier than Persian times. So detailed a survey of the scries at Anau may be justified by several considerations. First, to emphasize its close similarity with the Susan culture,, in quality, in duration, and. in the sterile interval between an earlier and a later period, on adjacent sites at Anau, but at Susa actually superposed* Secondly, because in the second culture at Susa, which corre- sponds with the earliest sites on the Sumerian ulluvium, a fresh set of influences, exemplified in the undccorated red-ware and grey- ware, appears in competition with the old painted-ware, in much the same way as in the second and third cultures at Anau, Bofn series point towards a distinct centre of culture further west, and the only culture which has such a red-ware tradition is that of early Syria, which has ancient relations with Egypt on the one hand, and with the highland-girt plateau of Asia Minor on the other; the latter a smaller replica, in respect of physical geography, of that of Iran. II, vmj RELATIONS BETWEEN IRAN AND SYRIA 89 Thirdly, because the more copious use of copper, even in the lowest layers at Susa, and still more in the tombs belonging to it, suggests that in this region, as at Anau, this copper is not originally local, but comes from another source, to which Susa had the easier access. This again points westward, to the Syrian culture or be- yond it. Fourthly, the occurrence of painted ware, resembling more or less closely the later stages of that of Anau and Susa, throughout North Syria, in south Palestine, in Cyprus (where it can be seen intruding into a purely red-ware culture), and locally also in Asia Minor, suggests a phase of reaction, later (as the sequence in Cyprus shows) than the widest expansion of the red-ware culture, in which the painted-ware tradition profoundly affected the Syrian region. Thiarphase cannot be precisely dated yet, but the presence in Egypt, under the early dynasties, of painted fabrics alien to the Nilotic styles, probably gives a downward limit for its arrival in Syria, and consequently for the previous spread of the red-ware culture eastward* The latter, on this reckoning, should be not far from contemporary with the beginning of the dynastic regime in Egypt, and the first culture of Anau would be altogether predynastic. VIIL THE RED-WARE CULTURE OF THE NEARER EAST The red-ware culture has already been noted in two connex- ions: (i) as the source of the new elements which are intruded into the second culture at Anau, and confront the Susan culture at the phase when it began to spread onto the Sumerian alluvium; and (2) on an earlier page (p» 79) as the probable source of the red-ware technique which has been traced spreading widely over the Danubian region. We have now to define its range and ex- amine its origin. The region over which it seems to be at home extends from Palestine on the south, to the Hellespont westward, and to the Upper Euphrates, or possibly rather further east; covering that is, the whole of the Anatolian or peninsular section of the Highland Zone, together with its Syrian appendage between the north end of Arabia and the eastern gulf of the Mediterranean. The earliest pottery of this region is illustrated best in the first city at His- sarlik, which has only very slight acquaintance with copper; in tombs at Yortan Keui and a number of casual finds all over Asia Minor, and in the lowest layer at the stratified site at Sakje- geuzi in North Syria. Its forms are partly close imitations of gourds, partly of skin vessels; the clay is densely blackened, and hand-burnished; the ornaments are simple bands, triangles and go NEOLITHIC AND BRONZE AGE CULTURES [CHAP, lozenges, with sparing use of punctured dots within the outlines, all incised3 and emphasized with white paste. Locally this fine * black-ware' degenerates into ashy grey? and loses its burnished surface. This early culture seems to be purely neolithic, with plump pebble-like celts rubbed to a blunt edge, and very little use of flaked flint. With the spread of copper implements a marked change takes place in the technique. Black polish gives place to a clear^brick-red, degenerating to chestnut-brown, as the black de- generated to grey. The forms become more gourd -like; open bowls3 long-necked jugs with one handle or none, wide-mouthed jars with cylindrical neck and two handles or more. Incised decora- tion becomes rarer, and is supplemented with ornaments modelled in relief. At Hissarlik, in the second city, many jars fcave human faces on the neck, or on a deep cover which fits over it, * It is in this period that the first exploitation of Cyprus takes place, and it is here, in a culture transplanted fully formed into a fresh locality, that its other characteristics have been most closely observed. Cereal agriculture was practised, as well as the growing of domesticated gourds; oxen and sheep were kept; the copper, which is abundant here, was worked extensively, and exported. The earliest forms of implement are the flat celt, the leaf-shaped dagger, and a longer dagger with a hooked tang to secure it in a wooden haft, The latter is peculiar to this culture; the former two are common to it and to predynastic Egypt., where the majority of the forms are quite different. The technique also of the red-ware is identical with the predynastic Egyptian, though its forms are wholly different; even the few gourd-forms among the Nile pottery being quite otherwise treated. The question now arises,, did Egypt or the Syrian culture ori- ginate copper-working, and transmit it to the other? In Egypt copper appears as a luxurious adjunct to a highly developed in- dustry or flaked flint, with very little grinding of implements, though hard stones were skilfully worked into vases; and it Is only very gradually that flint work declines and copper becomes com- moner; the transition is incomplete at the opening of the dynastic series about 4000 B,C* In the Asiatic red-ware region a small selejp- tion from the Egyptian copper-types appears suddenly amid the polished-stone culture, together with the red-ware pottery: Syria adds one new type of its own, and then remains long stagnant. There is copper ore in Syria itself, and in many parts of Asia Minor? but it would seem that it was the richer copper of Cyprus, exploited by men of the red- ware culture, which excelled competitors, and stereotyped these few forms over so large a I I,vxn] ORIGINALITY OF NEAR-EASTERN CULTURE 91 region. At first sight the Egyptian copper industry would seem to have priority. But the same question of priority arises as to the .origin of cultivated grains, wheat, barley and millet. Their wild forms are found along the Highland Zone, from Syria eastward; the same cultivated varieties are already in use from the first at Anau, and in predynastic Egypt. But Anau had had a very long career before the first irruption of the red-ware culture, and had copper from the first. Its domesticated animals, which it acquired some while before the red-ware came, are on the one hand derived from local species, on the other identical with the breeds of predynastic Egypt. Had Anau, or Egypt, priority? Or were both indebted to that intermediate region where the red-ware cul- ture arose £ In the present state of our knowledge of this * Middle Kitsgdom ' of the Near East, the answer remains in suspense, In another line of advancement the originality of the Syrian culture is less disputable. It is with the reoccupation of Anau by its third culture that the first clay figures of nude women appear. At Hissarlik they begin in the first city, and are copious in de- generate clay and stone types, from the second onwards. In south- western Asia Minor, similarly, they are found in the black-ware technique, and beyond the margin of this region they are part of the repertoire of neolithic Crete, and of the early bronze age of the Cyclades; in the latter case contemporary with a local school of red-ware. In Cyprus they are frequent in the local red-ware and even in a fairly early phase of it. In other parts of Asia Minor, and throughout Syria, they occur in various early techniques, in more and more traditional and grossly accentuated forms. Though a few female figures in local red-ware have been found in predynastic Egypt, they are unconventionalized and even this type had no regular vogue. In Palestine, where it became popular in the Bronze Age, there are only late and secondary types. In Babylonia it was unknown till the time of Hammurabi, and then became popular; and Hammurabi's people are thought by some authorities to have come down the Euphrates out of Syria, about 2300 B.C. (see p. 467). In Syria itself alone, on cylinders of rather earlier date, the conventional type can be traced in course of development. Everything therefore points to the creation of this artistic type, and of the religious con- ceptions which it symbolizes, within the region dedicated in his- toric times to the 'Great Mother of Asia/ With the exception of the figures of palaeolithic women, — no relationship with which can be established at present for this Asiatic type, — it is the earliest * ideal type' in history; and the earliest cult of which we know thp meaning as well as the symbol. 92 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. IX. THE CULTURE OF THE EASTERN MEDITERRANEAN We come now to the last great region, and tradition of culture, which remains to complete the survey of our neolithic cosmos — the Mediterranean itself and the districts interconnected by it. Like the painted-ware culture of western Iran, and the red-ware culture of Syria and Asia Minor, the neolithic Mediterranean culture passes over so gradually into that of the full Bronze Age, that its development and relationships to neighbouring civiliza- tions can only be traced within a broad period of time, as well as over a wide extent of country. Local advancement was uncon- formable within its limits, and precocious varieties overlapped the more belated. And from two of its most prolific areas, currents of influence were projected beyond and athwart the regions and cul- tures which have been outlined already, to an extent which has profoundly influenced all subsequent history, Contemporary with the earliest known phase of prcdynastic civilization on the margins of the Nile alluvium, occur rare ex- amples of an alien fabric of pottery, which has provisionally been described as Libyan, that is to say, they are thought to be intrusive from the west. The clay is dark-brown or blackish, hand-made and burnished; the forms are open bowls and cups, sometimes on three or four short feet. The ornament is incised in simple geometrical forms, suggestive of basketry, sometimes rather elaborate, and always emphasized by careful filling with lines or dots. White paste is used, as in the old black-ware of Asia Minor, which shares the liking for tripod supports,, but has little love for basketry. Very scattered finds further west in northern Africa link these stray vessels with an amazing wealth of distinct but similar fabrics on neolithic sites in Malta, representing a long series of develop- ment, which culminates later in the great stone-built monuments at Hajiar-Kim, Mnaidra, Hal-Tarshien, and at Gtgantea in Gozo; Sardinia has another local school, and characteristic tripod vases? at Anjelu-Ruju; Sicily has similar but less fantastic fabrics, self- coloured and richly incised, at Stentinello and Villafrati; south- Italy has others, at Matera and Pulo di Molfetta, very early modi- fied, however,, by contact with other cultures to which reference Is made later (pp* 104 sgy^). Further north,, the Rhone valley has settlements of similar culture, as far inland as the great Camp de Chassy, near Macon. By far the most important regions, however* in which this widespread Mediterranean culture occurs? are Crete and Spain, II, ix] AFFINITIES OF EARLY AEGEAN CULTURE 93 In Crete, below the first Bronze Age layers at Cnossus (see Chap. rxvn), which are as old or older than the first Egyptian dyn- astfes and therefore not later than 3500—3000 B.C.., lie neolithic deposits about 25 ft, in thickness. From trial pits in these de- posits comes self-coloured pottery incised with simple linear and dotted ornament, showing general resemblance both to the other * Mediterranean ' fabrics above mentioned, especially in respect of the vase forms, and also rarer points of correspondence with the neolithic c black-ware' of Asia Minor. Almost identical pottery occurs locally in cave-deposits on the Syrian coast, but nothing similar is known in Cyprus. Further north in the Aegean, Melos, Amorgos, and some other islands show late and specialized phases of a similar culture, already affected both by the black-ware tech- nique, and by the red-ware of Asia Minor which superseded it. These Cycladic schools belong to the first period of the Aegean Bronze Age; they had intercourse with the earliest Bronze Age culture of Minoan Crete, and so indirectly with Egypt, and may be regarded as contemporary with Dynasties IV— VI, or not later than 2500 B.C. Aegean neolithic culture thus lies in a sort of sea- girt enclave between the black-ware culture of neolithic Asia Minor, the southernmost margin of the great Danubian region in Thrace, Macedon and northern Greece, and those scattered off- shoots of the trans-Carpathian painted-ware culture which pene- trated the Balkan highland and established themselves in the Thessalian plain. Its affinities are almost wholly with the other Mediterranean coastlands, but in default of information from the long stretch of north African coast opposite — which has under- gone progressive submergence since the beginning of the Nile- alluviation — it is difficult to define its exact relations with its west Mediterranean counterpart. As the Cretan neolithic was super- seded about the beginning of dynastic Egypt by the bronze-age 'Minoan' culture, with fresh vase-forms, painted decoration, azid engraved seal-stones, its principal interest is a proof of the very long period occupied by the * Mediterranean ' neolithic period before the dawn of the Minoan. This is in full accord with the Recurrence of those * Mediterranean * types of incised pottery in early predynastic tombs, with which this section of our enquiry started, For more detailed discussion of the Minoan series itself in Crete and the Cyclades, see pp. 139 sqq^ 1 74 sqq^ and Chap* xvn. Its share in the propagation of a bronze-using culture outside its Aegean cradle-land is outlined briefly below, pp. 103 $qq« 94 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. X. THE CULTURE OF THE WESTERN MED1, TERRANEAN AND ITS OFFSHOOTS The great Spanish peninsula stands in a totally different rela- tion to the neolithic culture of the west Mediterranean basin, from that of Crete in the eastern. It is in itself a little continent, of about the same size as Asia Minor, more diverse in its configu- ration, and of at least equal variety and abundance of resources. Its two great central plateaux drain westwards to the Atlantic; as Phrygia and Cappadocia drain away northwards into the Euxine. The northern is more completely isolated, and has but a narrow foreshore astride the Douro mouth. The southern, by the two- fold access of the Tagus and Guadiana, communicates with the maritime lowlands of southern Portugal, and is reached with Kttle difficulty also by the headwaters of the Guadalquivir, from Anda- lusia,) the Lydia of Spain, Back to back with these central plateaux and facing onto the Mediterranean like JLyeia and Pisidiu in Asia Minor, are the narrow but very habitable coastlands of Almeriaj Alicante, and Valencia, with the Balearic chain, like Rhodes and Cyprus, inviting exploration seawards. Then comes Catalonia, a counterpart of Cilicia, with the long Kbro trough cut back far into the continent and opening a back door to the two plateaux of the interior. Finally, round the abrupt end of the Pyrenees and beyond lie more such lowlands, with access by the gap of Carcassonne to the vast coast plain of western France, and by the Rhone to central Europe* It would perhaps not strain analogy unduly to compare with these the Syrian coast, in some at least of its early relations with Mesopotamia. Throughout palaeolithic time this vast region had been the vehicle and the recipient of alternate phases of culture; Chellean, Acheulian, Mousterian, and at least one raid of Solutrean, from the north; Aurignacian and afterwards Capsian from north Africa, a twin continent which has no counterpart in the surroundings of Asia Minor., though its Saharan background has played repeatedly the same part in western history as Arabia has in the Near East, The long Magdalenian decadence affected the lands south of thp Pyrenees but little, and only late. Cave draughtsmanship at Al- tamira and other sites in the north-west achieves finer and maturer triumphs, and hands on eventually its own traditions to eastern and south-eastern districts-, where the rock-shelters show stag, oxen, and perhaps bison, hunted by men armed with bow and arrow, who sometimes fight among themselves, as at Morella, and whose women are shown at Cogul wearing long skirts, and II, x] 'MEGALJTHIC* ORIGINS 95 engaged In ritual dance. Even here, however, the period of cold moistuf e with consequent wide extension of forest restricted the descendants of the old hunters to these and a few other sheltered districts. Kitchen-middens accumulated along the Portuguese coast, and in the interior the subsequent deposits are mostly in caves. Rare early examples of broad-headed men show that the new people from the Alpine forest region began to spread beyond the Pyrenees, and a considerable population of this type estab- lished itself in the district around Mugem in southern Portugal. This crisis past, the whole peninsula was the prize of the next comer; and we have probably to make large allowance for our defective knowledge of Morocco and all northern Africa, in esti- mating Ibeyian originality. The small south-eastern coastlands, and ^especially that of Almeria, acquired early elements of the Mediterranean neolithic culture, and developed it rapidly; with regular settlements round caves and on hill tops, subsisting on the chase, with bow and arrow, and on simple terrace agriculture, like all branches of this Mediterranean culture. But the steep high- lands, still heavily forested, prevented expansion into the interior. From similar origins on the coast between Cadiz and Huelva, the Andalusian lowland was exploited with more success. But the main centre of advancement was the larger lowland of south Portugal. Here the kitchen-midden folk, reinforced as we have seen by 'lost tribes' of Alpine ancestry, and probably now by settlers from the Andalusian coast plain of the Guadiana, multi- plied rapidly, and created a culture of their own. Its industries are those of the other coast-districts, grafted on to those of the kitchen- middens, but matured early, rapidly, and distinctively, in this large and exceptionally favourable region. Most important of all, it is here that we first meet the custom of burying the dead, or at all events those of the more important families, in artificial chambers formed of upright blocks of untrimmed stone, and roofed with others, all as large as there was man-power to handle. Originally they were probably covered with a mound of earth, at least to the level of the cap-stone. From rude beginnings these 'megalithic* bjarial-chambers developed through a well-defined series of forms; the mere chamber, round or polygonal, according to the size of the wall-blocks; the chamber with corridor entrance, necessarily of some length as the diameter and height of the mound increased; the corridor with a terminal alcove replacing the chamber; the mere corridor with lateral alcoves or apses; and only after this, by gradual reduction of scale, the mere trench or cist below the natural surface, still lined and roofed with slabs in the ancient way. 96 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. There seems no need to infer alien influence at any stage; even the corbelled cupolas which replace the megalithic cap-stone "are but another case of ^necessity mother of invention/ Such works pre- sume co-operation, and no ordinary degree of social coherence; and people so constituted and so situated had a whole world at their feet. As the climate became drier, and the forest more pene- trable, they pressed up the great valleys, onto the southern plateau and eventually beyond it into the Ebro basin, where they found and mastered the backwood settlements from the Catalan sea- board* They reached the Mediterranean coast around Valencia; they occupied Andalusia, and were only prevented by the rugged highlands of Granada and Murcia from transforming likewise the secluded Almeria culture. The latter was to have its turn later on. The great abundance, variety and excellence of their arrowheads betray their chief means of aggression; the growing perfection of their pottery,, grey or black-polished, incised with white-filled linear ornaments, of skill and beauty, attests their sense of style; everywhere their great burial chambers demonstrate their effici- ency and energy, Nor were they checked by the sea. The * talayots* of the Balearic Islands are a local adaptation of 'megalithic' architecture to a dis- trict where soil was too precious for mound-building, and must be replaced by rubble from the fields. The * giants* graves* of Sar- dinia show development from simpler types to the phase when the corridor had outlived its terminal chamber, but not yet developed alcoves in its sides; the great monuments of Malta and Gozo show the supreme achievement of successive paired apses, dwarfing the corridor, roofed with cupolas of ashlar masonry, and supplied with side-doorways cut through a single slab* A distant but apparently early outpost is the group of burial chambers in the heel of Italy; Corsica has another such. To what extent the north African coast was occupied, the small, late, and little-studied 'megaliths' of Roknia and Enfida do not clearly inform us; the impressive 4se- nams* of Algeria and Tripoli are now known not to belong to this culture at all, but to oil-presses of Roman date; and the * mega- lithic' structures of Nubia and Moab have been too little explored to permit more than conjecture as to any affinity with the west Mediterranean culture: they seem to be rather cists than dolmens, and if so, are comparatively late. The same applies to a reported group of large-stone monuments in eastern Thrace, to those of the Crimea, and to another limited and coherent * megalithic* area on the Pontic coast of Georgia, connected, apparently, by some isolated examples south of the Caspian, with a vast region to the II, x] ITS EASTWARD LIMITATIONS 97 south-east, including most of India and extending into the Pacific, where chambers of similar construction are found sporadically. Though the area exploited by the * megalith' builders included the whole of the western Mediterranean, and perhaps extended beyond it eastward, and though the total period of this exploita- tion is shown by the successive types of the monuments to have been a long one, its influence was not permanent. In Malta, after a brilliant climax, in which many concurrent styles of decoration were attempted, including an experiment in spiral ornament which seems rather to descend from still earlier western attempts in Azilian times, than to be the result of intercourse either with Bosnia or with the Aegean, the neolithic culture seems to have been cut off suddenly and in its prime. In Sicily, which was apparently little effected by it, perhaps because its climate and soil made its forests too difficult, except close around Palermo and in the south- eastern corner, the primitive neolithic culture of Stentinello gave place to a strange and alien 'First Sicel' style, as at Castelluccio, which arrived fully developed, with geometrically painted pottery which has its only near counterpart in immemorial leatherwork design among the peoples of western and central Sahara, and in the primitive-looking pottery of the Aures and the Kabyle country of Algeria. An African origin for it, as for the Stentinello culture, is supported by the distribution of the characteristic rock-hewn chamber-tombs in which it is found. These recur in Malta, where painted pottery of rather different style is found with that of the 'megalith' culture; and also in Tunisia. But it must be remem- bered, on the other hand, that painted ware resembling that of Thessaly occurs at Matera in the heel of Italy — within the mar- gin, that is, of the 'megalith* culture, though not actually on a *megalithic* site — and that there was certainly intercourse between the painted-ware culture of Sicily and the second city of Hissarlik far away in north-western Asia Minor, a peculiar type of carved plaque in bone and ivory being common to both, and occurring also in neolithic Malta. It might have been expected, and was indeed formerly sup- pc^sed, that the neolithic culture of Malta and Sicily owed some of its characters to Aegean initiative. But this has not yet been proved, and at present such correlation as is possible tends to show that the west Mediterranean culture long developed independently, and was for the most part earlier than the great Minoan Age, The earliest links are supplied on the one hand by the bone plaques already mentioned, which are dated at Hissarlik not later than 2000 B.C.; on the other, by the painted pottery of Matera, which C.A.H.I 7 98 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. if it be of Balkan origin, belongs to an even earlier phase. As both these links are subsequent to the spread into Sicily and* Italy of the 'bell beaker' culture, to which reference must next be made, they serve to emphasize the relative earliness of the western cul- ture, and its independence of anything Aegean. And it has been noted already that even the neolithic settlement at Cnossus, seems rather to be an early northward offshoot of an essentially Medi- terranean culture with its cradle in maritime Africa, than itself originally Aegean* While the 'megalith '-building culture was permeating the west Mediterranean in this way, it was achieving even wider and more arduous expansion northward along the Atlantic seaboard. That this expansion took place mainly coastwise, and not over- land, is suggested by the distribution of the monuments in«» west- ern Europe, and especially of the different types* Principal early centres are, first, the promontory of Brittany, whence *megalithic' enterprise diverges, northward to Britain and Ireland, and north- eastward past the Low Countries to Denmark and southern Swe- den; secondly, this Scandinavian area, whence the whole of the western half of the North German plain was occupied, as far as the foothills of the central highlands. Meanwhile, the whole of lowland France was exploited, mainly up the Atlantic rivers, but also directly by land past the Pyrenees, and probably also from Catalonia along the Mediterranean shore, into Provence and up the Rhone, along an earlier line of exploration already noted (p, 92). That the whole of this vast area remained in fairly full inter- course with the motherland of the megalith-culture is clear from the occurrence of the same varieties of tomb-plan in nearly every region, usually in the same order of development, as is shown by the sequence of associated implements. The pottery varies locally, within a general uniformity of technique. But the individuals buried in these tombs vary in type, so that it is not possible to speak of a Megalith-people/ but only of a megalithic culture and a social structure imposed by its originators on the natives among whom they came. In the British Isles, these are more or less pure descen- dants of Aurignacian and other old long-headed stocks. In Scsyi- dinavia and the whole north-western area of the Continent, they are the tall massive long-headed folk who had apparently been developing there since the dispersal of the Cro-Magnon and Solu- trean hunters; they seem to be an early offshoot of the c Tumulus- people ' of southern Russia, and are the ancestors of the present 'Nordic' blondes. On the Atlantic seaboard, and all across France, there is the mixed population of Magdalenian survivors and Alpine II, x] THE NORTH-WESTERN PROVINCE 99 intruders, by this time much interbred except in the central high- land of Auvergne, where the forest remained intact longest, and the Sevenole type of broad-heads purest, and also least affected by 'megalithic' innovations. To follow the tfmegalithic* culture in detail as it made its way up the valleys leading into the Central German highland is impossible here. It is essential only to note that the strong forest barrier of the Vosges and Ardennes, and around the headwaters of the Marne, Seine and Loire, checked progress from the west, while the Rhine and Weser invited intrusion from the north. Consequently the Rhine, and its eastern tributaries Main and Neckar, early received elements of 'megalithic' culture from the seaboard, and greatly modified the^)ld Danubian culture which had exploited these areas beforehand. A temporary advance of Alpine lake-dwellers down- Rhine was met and repelled, so that northern elements penetrate even into the lakeland, and with them some Nordic men. Further east, the forest-frontier of the Danube basin seems to have held firm for a while, though northern traits were already becoming common locally, in pottery and implements, before the next crisis came. Further east still, local cultures more or less clearly based on the * north-western * megalithlc tradition, established them- selves along the upper courses of all the North German rivers as far as the Vistula, but failed like the Rhenish and Thuringian intruders to penetrate into Bohemia or Silesia, which remained essentially Danubian. Bohemia however was being affected about this period by an Alpine outflow similar to those down the Rhine and into northern Italy (p. 74), all probably due to some passing austerity of Alpine climate uncorrelated yet with events elsewhere. And before the crisis with which we have next to deal, Bohemia was also being influenced by the * painted-ware * culture from be- yond the Carpathians; so that our survey of events in neolithic Europe has now returned upon its starting point. Not merely was Europe itself by this time plotted out among well-defined regional cultures occupying its principal lowland and loessland areas, but the barriers of highland and forest which ha4 separated those areas hitherto were beginning to break down before human aggression from outside* We distinguish, that is, not merely eventual Hispanic and Gallic provinces on the Atlantic seaboard, an eventual Rhine-land and Danube-land, and Bohemian and North German regions, distinct from these and from each other; but also historic avenues like those of Carcassonne, Moravia, and the Lower Rhine. 7— * ico NEOLITHIC AND BRONZE AGE CULTURES [CHAP, XL THE CULTURE OF THE BEAKER-FOLK Meanwhile, a second impulse originating within the Spanish peninsula was to produce even more far-reaching effects than those due to the 'megalith '-builders. A good deal of the decoration, and some of the forms, of all early pottery in the neolithic Mediterra- nean, from Portugal to Crete and the * Libyan' vessels in Egypt, shows the widespread use of various kinds of basketry. This is natural enough when we consider that this culture is bounded southward by the grassland margin of Sahara, and that the most characteristic plant of all this grassland and of the plateaux of Spain itself is the half a or esparto rush, one of the finest materials for basketry in the world. But at a late period in the * megalith* culture something more specific occurs: the * bell-beaker' type of pottery, more closely imitated, both in form and incised decoration, from flexible rushwork vessels than any earlier or later type, is so suddenly intruded among existing Spanish forms, and followed by so remarkable a fresh outburst of exploitation, that there is much inducement to ascribe it to the intrusion of some fresh stimulus, perhaps from the African side, like the mediaeval coming of the Moors. Whatever the cause, the effects are certain. Over- running all parts of the peninsula, and reaching Mediterranean localities so remote as Sardinia, western Sicily, and Remedello near Brescia, in the far north-east of Italy, the 'bell-beaker* cul- ture crossed the Pyrenees, and penetrated almost all districts of France. Following the old coastal route to Brittany, it passed over to Britain and Ireland, and affected also profoundly the large region beyond the Netherlands which the * megalith '-builders had already made their own. That it was not a mere distribution of trade-objects is clear from the fact that the bell-beakers themselves are of local materials and various techniques; that it was not only the half a baskets them- selves that were traded a,nd imitated locally — though this, too, is probable — is shown by the simultaneous appearance of other kinds of objects, and by the shift not merely of whole provinces of cul- ture but of the frontiers of physical types, in the same direction as the spread of the * bell-beakers/ which, wherever they appear, are a storm-signal of profound disturbances, from Denmark to Buda Pesth. It has even been doubted whether those at Remedello are transmarine or transalpine intruders. Neglecting, as before, the bewildering details where they are known, and supplementing provisionally the no less baffling scarcity of data at some important points, we may yet present a II, xi] NEW WESTERN ACCESS TO CENTRAL EUROPE 101 general outline of the course of the 'bell-beaker' movement, and its principal effects. Ii? general, the 'bell-beaker' movement followed the main lines of the 'megalithic' culture, overtaking it however on its frontiers and passing beyond them. In one respect, however, it created a new situation altogether; for whereas in eastern France the 'me- galithic' advance had been held up by the forested highlands west of the Rhine, the * bell-beaker * folk, better organized and better armed, especially with highly-developed archery, forced this bar- rier (perhaps already weakened by previous clearings towards its main gaps north and south of the Vosges) and broke through into the Danube valley. We may speak confidently here of invasion, because the change of culture is not only sudden, but is accom- panied by replacement of the old Rhenish and Danubian popula- tion by the moderately broad-headed stock which had long been characteristic of the region of Atlantic drainage. The open villages and peaceable habits of the Danubian valley-folk made them an easy prey : remnants of them survived here and there in the foot- hills of the central highland, but this barrier also was obsolete, and the northern and western groups of * bell-beaker ' folk coalesced as they advanced, and occupied even the secluded Bohemian area. Further east still, parts of Silesia remained in occupation of a Danubian remnant; but a * bell-beaker* has been found as far down-stream as Buda Pesth. The main flood of invaders, however, was stayed in the more hilly country between Bohemia and the Austrian Alps, where the valley narrows, and the old Alpine cul- ture with its secure lake-settlements offered better resistance, and diverted the invaders northward into Bohemia and Moravia* This long-secluded region now became the centre of a fresh movement, the origin of which is obscure, though its results were revolutionary. Its population was by this time chaotically mixed, partly old Danubian, partly Alpine, partly new western invaders, and perhaps partly of more easterly and south-easterly origin; for the tumulus-building steppe-folk who, as we have seen, displaced the Tripolje culture from Galicia (p. 8 i sq.), seems to have pressed forward thus far about this time, while their southern kinsfolk made chaos in the Balkan lands. And out of this crucible of diverse stocks a new and remarkable type of man emerged, broad-headed like the Alpines, heavy browed like the steppe people, with mas- sive square face and jaw like the men of the old north-west, and with something of the high-vaulted brain-case of the Dinaric and Balkan roundheads. Their industries were in the main those of the 'bell-beaker' culture, and their east Alpine connexions kept them. 102 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. in remote touch with the nascent copper culture of Italy; ^but they buried their dead in cist-graves resembling the latest *megalithic* tombs, covered however by conspicuous earthen tumuli, nof oval like the * long barrows' of neolithic Britain, but circular like those of the steppe people. It is one of the few instances where a new kind of man has come into existence under conditions where the ante- cedents are in any degree knowable, and whose racial history ex- presses so clearly the qualities of the brain within the new type of skull. It was apparently not long before the 'round-barrow folk/ as we may conveniently call them, outgrew their Bohemian cradle, and dominated the Danube valley, and much of the eastern Alps, coalescing with the already mixed folk (Alpines, western in- vaders, and Danubian remnants), whom they found *there. West- ward they spread into Thuringia; eastward into the Hungarian and Galician lowlands. But their main achievement was to the north-west, where they overran the lowland as far south as the Seine, penetrated into Denmark and Scandinavia^ and built their 'round barrows ' in south Sweden and south-western Norway. At the estuaries of the Elbe, Weser and Rhine, they took to the sea, and occupied the eastern districts of Britain, from the Thames to the Forth, driving the long-headed folk of the 'long barrows' into the forests, but not disturbing the more civilized 'megalithic* folk of Kent and the south and south-west. Here too their 'round barrows' indicate their distribution; and the 'beaker' types of the pottery in them clearly betray their affinities. And wherever they went, they settled and have remained, the ancestors of the 'John Bull* type of Englishman and the kindred continental stocks, The old long-headed Nordic people, whom they disturbed, partly coalesced with them, partly enlarged their own borders northward at the expense of the representatives of the old * Arctic" culture, till they were checked, partly by the climate, partly by the Mongoloid ancestors of the Lapps who had been working their way round the head of the Baltic as soon as the shrinkage of the last Swedish glaciers made this possible. In the Mediterranean, the 'bell-beaker' culture produced com- paratively small effects, so far as our present information goes^ It reached Sardinia and Sicily, but apparently not Malta; and there are no known traces of it on the north African coast. And its vogue appears to have been short. There seems to be good reason for this, as the west Mediterranean, and even the Mediterranean coast of Spain itself, began now to come under a fresh influence, which was to change the whole outlook of this region. It is only in this direction that we may hope to gain even relative dates. II, xii] THE BRONZE AGE 203 XII. THE COMING OF BRONZE The movement which initiated the Minoan bronze age culture in Crete and the Cyclades does not seem to have been confined to the Aegean, Its sources were multiple, and are not to be sought only in Egypt, though intercourse between the Nile and Crete was early, active and persistent. The implements and the pottery, both red-ware and painted, have much in common, as the very names of these styles imply, with Asia Minor and Syria and with that far-easterly culture which penetrated these regions early. Further west, the connecting links are scanty, but the fact that copper-working began early in south-eastern Spain, that the first copper implements there are the leaf-shaped dagger and the flat celt, ^and that with the copper appear fresh vase-forms and an imperfect red-ware technique, which spread rapidly and widely, suggests that this western copper-industry was not an independent discovery, but resulted from intercourse with the Levant. It was not, however, the * bell-beaker' regime of the plateau, but the smaller, more secluded, and hitherto more backward culture of the Almeria coastland, which acquired and exploited the new know- ledge; and the reason for this is certainly the wealth of copper ores in the coast ranges of Murcia and Granada, near enough to the sea to be accessible to prospectors, well supplied with timber for fuel, and perhaps already provided from the same source with seafaring vessels and oversea connexions of its own. Once intro- duced, the new industry developed rapidly; improved types of implements were designed; and the discovery, perhaps accidental, that certain ores yielded a yellower metal, resembling the gold which already circulated as a rarity in neolithic Spain, led to the employment of this for ornaments, which were traded into the interior for some while before the new alloy, as this yellow * bronze' was later discovered to be, was used for implements also, when its greater toughness was appreciated, and produced designedly with the aid of * tin-stone/ This mineral is widely distributed in certain districts of the far interior, and was soon traded to the copper- working districts, and eventually also abroad. At Hissarlik bronze is found in the second city, not later than 2000 B.C., and probably a good deal earlier, in weapons of Asia Minor type; in Egypt it appears first under the Vth Dynasty, not later than 2800 B.C.; aiid in Crete it goes back earlier still, almost to the beginning of the Minoan series. Here also therefore Spanish priority in discovery cannot be proved: the transmission of knowledge Is far more difficult to io4 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. B detect than the transport of commodities; but it is significant that "in the tombs at Anjelu-Ruju, in Sardinia, which belong to a pre- metallic stage, and have a purely western culture, 10 out oT 63 bodies are not of Mediterranean type, and are indistinguishable from the broad-headed stock of Asia Minor, which certainly was entering eastern Crete early in the Minoan age, and must there- fore be presumed to have had already some seafaring skill. These Sardinian Immigrants had not been there long, for there had not been time for them to mix their blood much with the natives. Other patches of broad-headed folk have been recognized in Gerba island, off western Tripoli, and in the hill-country of north- eastern Tunis, but the earliness of their arrival here has not been demonstrated. In Spain direct evidence of such * prospecting* aliens has not been recorded yet. It can hardly be accidental, however, that the nascent copper- Industry in the west is accompanied, like that of Asia Minor and Cyprus, by active production of silver. This metal however was for long of local importance mainly, the ease with which It tar- nishes in a moist climate making it far less popular in the north- west. Another invention, this time definitely Spanish, did much to popularize the western metal industry. The leaf-shaped dagger, already broadened at the base, was fixed transversely (like a flat celt) in a long handle, and the * halberd' so constructed was'in wide demand. Together with other western types (elongated or ex- panded celts, the triangular dagger itself, and a longer swordlike blade), it was introduced into Italy, where the discovery of copper ores in Elba and Etruria set that peninsula fairly soon on an inde- pendent career; while Its nearness to the great Danubian province, now mainly dominated (as we have seen) by people of the * bell- beaker' culture, gave it an insatiable market for its metal work, traded against Baltic amber, and perhaps tin from the Central German highland. Later on, the Danube basin, and particularly the Hungarian region of it, began to exploit its own wealth of ore and fuel, and created a culture of its own; but central and north- western Europe long depended almost exclusively on Italian models, and in great part on Italian traffic. As in the west Mediterranean, so along the Atlantic seaboard, the Spanish metal traffic with its special series of forms followed in the wake of the bell-beaker culture. Halberds of early Spanish type have been found on the Upper Danube, and were widely copied in the north-west, as far as Ireland. It appears to have been about the time of the Bohemian exodus II, xnj COPPER AND BRONZE IN CENTRAL EUROPE 105 (p. 101) that the knowledge of copper began to penetrate Into westerif and central Europe; in the west mainly from Spain, and so, Jh the wake of the 'bell-beaker' folk, into the Upper Danube valley; in the centre mainly from Italy, greatly aided apparently by the arrival near Brescia and elsewhere in north-eastern Italy, and eventually as far south as Latium, of parties of people ex- hibiting mixed Alpine and Danubian physique, and burying their dead contracted in earthen graves, in old Danubian fashion. As a similar settlement has been found near Landshut in the Inn valley, it looks as if the famous Inn-Adige route across the Alps was already in use and in the hands of people from the north side. It must be reserved for a later chapter to describe the improvements in copper a#d bronze objects which were made there, and how they«cvere imitated in local factories north of the Alps, as northern ores were discovered and copper-working spread. Here it is suffi- cient to note that in all the earliest and some of the most important of the later types, such as the socketed celt, Italy supplied the models for all central Europe from the Carpathians to the Rhine, and competed, by way of Savoy and the Rhone valley, with the Spanish types which were already current further west. Far to the south-east, it is true, the copper and bronze of western Asia Minor seem to have been traded into Balkan lands through the second city at Hissarlik, and indirectly this traffic may have extended far, for amBer occurs in the second city, and celts, daggers, and pins, — spiral-headed or with an eyelet in the stem, and all common to Hissarlik, Cyprus and Syria, — are found in early lake-dwellings in Austria, But the disturbances due to the dispersal of the painted- ware people from Ukraine, and to the inroads of the 'tumulus- folk/ seem to have dislocated this traffic for a while; and it is not until a much later period, when the Late Minoan culture had at last reached the Hellespont and the north shores of the Aegean, not much before 1300 B.C., that its highly-developed swords, perforated axe-heads, and characteristic spiral decoration began to influence* the bronze work of Hungary and eventually of Denmark and Scandinavia. • Considerably earlier than this, however, and probably not much later than the days of the second city at Hissarlik, Aegean ex- plorers began to sail westwards, and penetrate into the western Mediterranean. Occasional finds betray their intercourse with south Italy as early as the Middle Minoan period, not later than 2000 B.C., trading with the Lipari islands for a rare decorative mineral, influencing the local pottery of Cassibile in Sardinia, and making at least one voyage as far as Marseilles. Their bronze io6 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. swords, of rather later date, reached Sicily when the c painted Sicel* style was in its decline; and in the Late Minoan peridd, after 1400 B.C. they had regular settlements on the east coast of Sfcily, and another at Tarentum. Their wares now reached the head of the Adriatic, and influenced the native metal-work of Este on the old Adige-route to the north, and of Etruria, probably by way of Bologna, which was then the great centre of intercourse between northern and central Italy. XIII. THE HALLSTATT CULTURE It was also perhaps by way of the Adriatic, rather than through the Macedonian passes, that Minoan manufactures, f.nd particu- larly the later types of bronze swords, reached the Middle Danube, and more especially the centres of a new culture which was de- veloping, under combined Italian, Hungarian, and Danubian in- fluences, in valley bottoms among the Austrian and Dalmatian Alps. Of the material culture of central Europe one great trading centre, Hallstatt, among the great salt-beds to which it owes its name and its exceptional wealth, gives, a little later, an unusually full glimpse; for this * Hallstatt culture* not only dominates all the Upper Danube, but exercises widespread influence over middle Germany, over central and northern France, and over Britain and Ireland. Its characteristic swords, modelled at two removes on the Late Minoan type already mentioned, travelled even further into Bosnia, Macedonia, Hungary, East Prussia, Posen, Hanover, Schleswig and Scandinavia and in later varieties into Spain and the British Isles. It was in fact the first culture so general as to deserve the name of European, and with its spread about 900— 800 B.C. this survey of the prehistoric world may close. Outlines of its distribution are given by the finds of a character- istic leaf-shaped sword with broad-flanged handleplate, a 'superior weapon' which cut its way rapidly in the hands or men of superior organization, across a large part of central Europe,, and betrays their occasional incursions into the coast lands or the Mediterra- nean, as far as the Greek islands and Egypt. ^ Several fresh factors contributed to this rapid expansion, and give the Hallstatt culture its distinctive quality. In the first place, this is the first great regional culture which made systematic use of the horse for riding as well as for driving. The horse had been hunted for food since palaeolithic times, but there is no clear evi- dence even of its domestication as a milch-animal, outside the high plateaux of central Asia, until a comparatively late date. II, KIII] THE COMING OF THE HORSE 107 The first positive record is in a Babylonian tablet of about 2 100 B*.C,, where it is described as the 'ass from the east,' or 'from the**nountains/ and was therefore still a recent acquisition among the ass-using folk west of the Zagros range (p, 501). Its arrival here is commonly referred to that Irruption of fresh peoples from Iran or beyond, who founded the barbarian Kassite dynasty of Babylon about 1750 B.C.; and as there is no reason to believe that the great plateau of Iran itself was even then In much better condition than now to support an indigenous pastoral civilization, it is probable that this irruption originated further to the north-east, on the Sarmatian flatland, and that it is to be connected, in Its significance, if not precisely in date, with the irruption of Aryan-speaking folk into India .from the same northern reservoir, and with that west- ward outflow of the "tumulus-folk* across the Dnieper, which broke up the painted-ware culture of Tripolje and penetrated through Galicia into Bohemia, and through the Balkan lands into north-west Asia Minor (pp. 82 sqq.}. The rapidity and violence of these eruptions from the northern grassland, far exceeding in extent and effects all earlier move- ments of which we have any clear Indication, were themselves probably due less to the sudden urgency of unsettlement, than to the acquaintance of the unsettled peoples with unprecedented means of rapid and concerted movement, namely the domestic horse, as steed rather than milk-giver; though the practice of mediaeval and modern horse-riding nomads shows that the two functions are compatible, and that commissariat troubles almost disappear in such a mode of life, provided only that there is ample grazing. This however is precisely the most difficult condition to be at- tained within the Highland Zone and to the south of it; and it is only on the grassland itself, and in Hungary, Thrace, Thessaly and other Intermont plains in the Balkan lands; on the Phrygian and Cappadocian plateaux In Asia Minor; in the larger basins of north Syria; and in a few secluded troughs within the Median and Persian mountains, that local centres of horse-breeding and horse- manship were established permanently. The position of Egypt is ambiguous, as usual. It has been suggested that the peculiarities of the thoroughbred 'barb* variety point to an independent domes- , tication of a north African breed of horse now otherwise extinct; but it is noteworthy that Egypt does not seem to have made any use of horses at all, eastern, indigenous or western, until after the period of oppression by Asiatic invaders which separates the Xllth Dynasty from the XVIIIth; that is, until about 1600 io8 NEOLITHIC AND BRONZE AGE CULTURES [CHAP. B.C.; and by this time the horse was apparently being already im- ported into Crete, only a little before the period at which the northern aggressors were beginning to break through into ^the coastlands of the Aegean. Once introduced, however, the horse found congenial quarters around the fen-margins of the Delta, and Egyptian chariotry met Hittite chariots and cavalry on equal terms in the Syrian wars of the thirteenth century. As in Cappadocia and Syria, so in Thrace and above all in Hun- gary, and eventually throughoxit the Danube valley, horse-driving, and eventually horse-riding conquerors organized and led their very mixed native levies, in every direction where there was pros- pect of loot and lands. The Phrygians, for example, passed over into Asia Minor in the thirteenth century, on the same track as earlier ^tumulus-folk,' and wrecked the decadent empire of* the horse-driving Hittites. Some think that the Homeric *Achaeans' represent another such incursion through Macedonia and Thessaly as far as * horse-grazing Argos.' The terremare-culture of the Po valley came to an abrupt and violent end through a similar invasion out of Styria and Krain, where most graphic representations of these sporting and fighting people are found on bronze vessels of rather later date. North of the Carpathians again, other bodies of essentially simi- lar horse-owning folk traversed the North German plain as far as Denmark, with similar social and political consequences. The sub- sequent adventures of these, and of the eventual Danubian and mid-German invaders of the maritime west, belong, however, to a later volume* It is not to be expected that the whole story of the coming of the horse should be based upon direct evidence of equine remains or of horse-bits and other horseman's gear. Enough, however, seems to be known of the general culture of the horse-owning peoples, to supplement such direct evidence as there is, by that of their weapons, ornaments, and other property. Of these, the swords already mentioned are the most significant; for among a multitude of earlier types developed by local craftsmen, especially in Hun- gary, from the old straight-edged daggers imported as we hav^ seen from Italy, and perhaps earlier still from Asia Minor by way of the Hellespont, there appears at last one, derived from an Aegean pattern, which gave these restless northern peoples what was in the literal sense the "superior weapon' against all adversaries. This * leaf-shaped* sword combined for the first time the advantages of thrust and of cut; and its long flat tang running the full length of the handle and furnished with lateral flanges gave the structural II, XHI] FIRST APPEARANCE OF IRON 109 security of a girder where this was most absent from all earlier blades.*Its occurrence as far to the south-east as Egypt, along with other mid-European types, all belonging to the period of the great sea-raids of the years about 1200 B.C.; in Cyprus where it was eventually manufactured locally; and as far west as Spain and Ire- land, is the best proof of its efficiency as a weapon. From it were developed not only the specifically 'Hallstatt* swords of the tenth, ninth and eighth centuries, but the swords of the Greeks of classi- cal times, and less directly that shorter Spanish sword which was eventually adopted by the Romans. Another notable invention must be brought into retrospect here, and may fitly close our story; for it was during the domination of the leaf-sh%ped sword that bronze began to give place to iron as the material for cutting weapons; though rather in the south than in the home of those swords themselves. Until some first-class site has been properly explored in Asia Minor or North Syria, cer- tainty is unattainable at the most crucial points in the history of the new metal : but from the fragmentary material at present avail- able, the following points seem to be made out. Egypt had occa- sional, perhaps accidental, acquaintance with iron as a rarity, from late predynastic times, and received Syrian iron as a precious metal in tribute under the XlXth Dynasty, but made no general use of the metal till Greek times. Babylonia had no early iron, and though Assyria had it occasionally from the thirteenth century onwards, there was no iron industry there till later, and iron was mainly obtained from the highland district of Commagene between North Syria and Asia Minor* In Palestine, literary references presume that iron was in use as early as the eleventh century; and iron weapons occur at Lachish and other Philistine sites after the arrival of the sea-raiders at the beginning of the twelfth. In North Syria, an iron-using culture intrudes from the north-west in the twelfth century, and it is about the same time, after the collapse of the Minoan sea-power, and of the old coast-land civilization of Cilicia before similar intruders from inland, that iron weapons become common rather suddenly in Cyprus. m As a precious metal for jewellery, Cyprus, like Rhodes, Crete, and the Minoan area generally, had known iron since about 1400, and it was perhaps through Minoan intercourse that iron finger- rings became customary in parts of peninsular Italy. At Hissarlik, iron does not appear till after the destruction of the sixth city, which occurred not earlier than the twelfth century; and there is no reason at present to believe either that Asia Minor obtained its knowledge of iron from Europe (as has been suggested) or that it no NEOLITHIC AND BRONZE AGE CULTURES [CHAP. was brought to Europe directly by the Hellespontine route. In the north the * leaf-shaped ' swords are regularly of bronze, irbn only coming into use gradually during the 'Hallstatt* period, and super- seding bronze only at its close; later, that is, than in Greece, and later still than in Cyprus, where the weapons of * leaf-shaped" type are in iron throughout, from about the eleventh century. Traces both of the older use of iron as a treasured rarity, and of its later use for tools and weapons, occur in the Homeric poems, but with- out precise clue to the relative dates of the passages. That eventu- ally a great iron-working centre arose in Noricum, and repaid to Rome the north's ancient debt to Italian bronze, is iindisputed; and it may be that those who introduced the * leaf-shaped ' sword into Cyprus during the twelfth-century sea-raids parsed on the Levant's knowledge of iron-working to the north, by way of the Aegean or the Adriatic; but at present, priority seems to lie with the North Syrian source, with the possibility that this in turn may be found to be derivative from some other centre beyond Taurus, such as the Chalybes in north-eastern Asia Minor from whom early Greece obtained afterwards its finer quality of steel* A third revolution in custom, of a less material kind, finds its first illustration on any sufficient scale, in the great burial-ground at Hallstatt. The custom of cremation, as an alternative to burial, was of old standing in Europe; for it appears almost (though not quite) at the beginning of the lake-dwelling occupation of the Alpine region, and is also characteristic of the painted-ware cul- ture in Ukraine. But it is not confined to Europe, The Medi- terranean region knows it not, but practises interment uniformly, until after the first northern aggressions; and the painted-ware culture of Anau and Susa has simple earth graves or cists* In Palestine, however, cremation was practised in a very early phase of culture at Gezer, and, though it was superseded there later by burials which seem to represent the first Semitic immigrants, yet "they made a very great burning' for King Asa of Judah in the tenth ceti tury,- and only omitted it for King Jehoram, for special reasons, in the ninth. So it may be that at this far southern outlier of what we have already seen to be the larger habitat of * Alpine/ man an old forest-usage was retained as long as there was fuel to spare. In late neolithic times, Alpine cremation spread into Italy with the 'terremare'-folk, and similarly on the north side of the Alps it came gradually into general use, passing over, for example, into Britain and Asia Minor alike. Something must however be allowed here for the dispersal of the Tripolje people westward, over the MAP 5 EUROPE SHOWING THE PRINCIPAL NEOLITHIC CULTURES MAP 6 <£?: EUROPE SHOWING THE PRINCIPAL LINES OF EARLY BRONZE AGE INTERCOURSE Land above 30Q0 feet II, xra] CREMATION IN CLASSICAL TIMES m middle basin of the Danube, and also for the prevalence of cre- mation among the Aryan-speaking invaders of India, and there- fore ^probably among other folk also on the northern grassland, At Hallstatt itself, the actual process of replacement is illustrated by many examples, interment being first supplemented by partial cremation, for example of the head, feet or abdomen, and only gradually superseded by total incineration. This, however, is a late instance; in Thessaly, cremation appears with the earliest 'leaf- shaped' swords, probably about the eleventh century, and the splendid pyre-funerals of the Homeric poems may anticipate this by a few generations, In Greece and in Italy, as in Herodotus' description of Thrace in his own day, both rituals persisted side-by-side^till Christian doctrine restored the aboriginal usage of thi south. Into the significance of this conflict of beliefs as to the latter end of Man, this is not the place to go : let Herodotus, with whose wisdom we began, close the story in his own way. 'Him who has left them, they bury in the earth, with gladness and sport, recount- ing all the evils from which he is now free and in perfect bliss' , . , 'for three days they lay out the body, slay all sorts of sacrifice, and hold a feast, ending their mourning first; and then they bury, hming to ashes, or merely interring and cast a mound, and hold sports of every kind, in which the chiefest prizes are for single combat, Such is the Thradan's funeral'— on the margin between two worlds, CHAPTER III EXPLORATION AND EXCAVATION L THE RELATION OF ARCHAEOLOGY TO HISTORY ¥N every department of human life the past century has wit- JL nessedthe gradual growth of free enquiry. Documents formerly regarded as infallible have in recent years been made the subject of the severest criticism. Neither the sanction of long habitual acceptance, as in the case of the classical historians, nor the endorsement of the divine verbal inspiration attributed to the Hebrew Scriptures, has exempted the writings named from this treatment. The statements which they contain have been put to every conceivable test. Along with the textual and literary criticism of the documents themselves, there have advanced pan passu the exploration of the obscurer literatures of North European and of Oriental nations, the observation and tabulation of the rites, customs, and beliefs of peoples in primitive stages of civilization, and the excavation of ancient cities and settlements. A wealth of illustrative material has thus been collected, which has undoubt- edly illuminated many formerly dark passages in the historical records, It is not to be supposed, however, that archaeological or ethno- logical research can supersede the labour of the historical critic, or that the results of such work can be called in, definitely to cor- roborate or to refute his conclusions. Doubtless the archaeologist may discover an inscription which, referring to some historical event, may supplement, or correct, the account of the same event in the pages of some ancient historian. But even such an inscrip- tion must itself be submitted to criticism. Oriental monarchs were not above exaggerating their mighty deeds beyond all reason, and allowance must be made for this weakness. Archaeological research consists principally in the discovery and the classification of the common things of daily life — houses, personal ornaments, domestic utensils, tools, weapons, and the like (see pp. i $q^ 66-70)* These are occasionally of value even to the historical critic; for example, they may help to expose anachronisms. If, to suggest a possible concrete case, a narrative should describe a community as using tools or weapons of iron, at a time when, as contemporary deposits CHAP. Ill, ij ARCHAEOLOGY AND HISTORY 113 indicate, it had not yet emerged from the earlier Bronze stage of cultnre, then the critic must re-examine Ms texts. Either the docmrnent is wrong in this particular, or, perchance,, the word which he has rendered 'iron9 may be found, to have some other signification. The reader will understand that this illustration is merely put forward as an example of the kind of assistance which the archaeologist may render to the historian. Archaeological evi- dence of this nature must, however, be cross-examined, like every other evidence. In a case such as we have supposed, the archae- ologist must satisfy the historian that the deposits upon which he bases his deductions are fairly representative of the state of culture of the whole community, and not merely relics of some insigni- ficant and t^ickward group of people living within its borders, but haviag no direct connexion with the course of history. In archaeological study we cannot always deduce causes from the observed effects with mathematical certainty: the evidence is often ambiguous, and frequently there are no indications to en- able us to choose among several possible solutions of a problem. We may, for example, find a layer of ashes in a stratified city-site. The historian may tell us of a conquest or of a raid about the time of this deposit; but it is at least an even chance that the fire which produced the ashes was a mere accidental conflagration, of which no documentary record has been preserved. Indeed, it is in most cases desirable for the archaeologist to form his conclusions as to chronology and allied problems independently of the historian3 and for the two fellow-labourers to settle any differences by gradual approximation. To excavate merely with the purpose of 'confirming' written history is to court inevitable disappointment, and, what is worse, to do most serious injury to the sites examined. Out often thou- sand recorded events, not more than one or two can possibly leave any permanent record upon the aspect of the sites which witnessed them. Even the scars of war quickly heal on the face of the earth. Abraham, Joshua, Samuel, David, Isaiah, Paul march in a ma- jestic procession through Palestine, but we ransack the land in Xgin for their faintest footprint : they live in the written word alone. An explorer who should be so foolish as to go in pursuit of their relics would neglect, and very probably destroy, the countless valuable remains which he would actually meet. The true function of archaeological research is to discover the conditions amid which lived such heroes of old as we have mentioned; to show them, no longer as solitary, more or less idealised or superhuman, figures, but as men of like passions to ourselves moving with other men, C.A.H.I 8 u4 EXPLORATION AND EXCAVATION [CHAP. in a busy world engrossed in its secular interests, and making daily use of the common things of life. To excavate with the sole purpose of adding to the stock of written history, by the discovery or tablets, papyri, or inscriptions, is an equally fatal error. It would not be too severe to describe many excavations that have been made as mere 'tablet-piracies/ So engrossed has the excavator been in finding libraries of tablets — the importance of which no one would dream of minimizing— that he has neglected the pots and the pans, which are essential if he is to fill in the picture of the ancient life of the region. To Professor W, M. Flinders Petrie belongs the credit for calling attention to the importance of 'unconsidered trifles/ and he has shown it at many times during his long career as an ex- cavator. To mention one striking instance, by his preliminary re- connaissance at Tell el-Hesy, the site of Lachish, in Southern Palestine, he determined for all time the principles of the dating of Palestinian pottery. He proved, by comparing stratum with stratum in the mound that covered the remains of this often-re- built city, that every age had its own style of pot shapes or orna- ment, and of clay baking. At different times different foreign in- fluences were brought to bear upon the craftsman. So completely can the evolution be systematized, that, thanks to Prof. Petrie, whose scheme has not been modified by his successors except in occasional details, it is possible to date a Palestinian mound as unambiguously as if it had been full of inscriptions. Even from horseback an observant traveller can often assign approximate limits of date to an ancient site in the country. Among other advantages, the pottery-test affords a valuable check by which the modern identification of ancient sites can be tested and controlled. Many such identifications, made in the early days of research, chiefly on the unstable foundation of similarity of name, must now be abandoned, as the potsherds show that the date of the site, and the date assigned in the literary documents, do not correspond. Seeing that the comprehension of certain his- torical events (as, for example, military movements) often depends upon an exact understanding of topography, the unimportant sherds which the archaeologist collects may thus not infrequently become of at least an indirect value to the historian. The antiquities of the Near East have attracted the attention and interest of travellers from the days of Herodotus. But for the purposes of this chapter it is hardly necessary to go back further than the beginning of the nineteenth century. Before that time these monuments were a matter rather for intelligent curiosity 111,1] POTTERY AND EARLY RECORDS 115 than for serious scientific study. We recall how the Spectator^ in his first* number (i March 1711), describes himself as making *a Voyage to Grand Cairo, on purpose to take the Measure of a Pyramid/ and adding, 'as soon as I had set rny self right in that Particular, [I] returned to my Native Country with great Satis- faction/ In short, having acquired a disconnected scrap of in- formation, in itself of only moderate interest, he made no further use of it. He was typical of his time. It was, indeed, impossible for any progress to be made in re- search so long as the inscriptions remained undecipherable. The outward appearance of Egyptian hieroglyphs was probably familiar to some Europeans at all times. The more remote cuneiform of Mesopotamia and Asia Minor was naturally for long quite un- knovrti; but from the time of the publication of Pietro della Valle's delightful letters describing his extensive Oriental travels, there was at least a scrap of knowledge available with regard to the as- pect of that mysterious script, for the writer named has repro- duced five characters which he saw at Persepolis; and has stated reasons for his supposition, which proved correct, that they should be read from left to right. The letter in which he gives these Old Persian characters is dated 21 October 1621. But the only sources of knowledge on which would-be decipherers could draw, down to the beginning of the nineteenth century, were the writers of Greece; and, as has been subsequently proved, even in the meagre information which they vouchsafe on these obscure points, they were blind leaders of the blind. In this chapter it is proposed to give a brief survey of the history of the archaeological researches of the nineteenth and twentieth centuries, which have so greatly enlarged our knowledge of Egypt, Babylonia and Assyria, and neighbouring lands; have revealed the empire of the Hittites; and have discovered the unsuspected civilization that flourished in the lands of the Aegean in the third and second millennia B.C. In setting forth the material, we shall follow the order in which a pioneer expedition to any country would naturally conduct its researches. First would come a survey o&the country, with an enumeration of the remains above ground; secondly, the collection and decipherment of its inscriptions; and thirdly, the excavation of its cities and burial places. 8—2 n6 EXPLORATION AND EXCAVATION [CHAP. II. EGYPT (a) SURFACE EXPLORATION Many accounts of the wonders of Egypt have come down to us from the hands of early writers. Herodotus has provided rich material for controversy with his descriptions of Lake Moeris and of the Labyrinth; and, in later times, the pyramid-fields accessible from Cairo attracted the attention of many mediaeval travellers and pilgrims. It is, however, hardly worth our while to expend space upon these fragmentary allusions. It is a strange and prob- ably a unique fact that the foundations of scientific Egyptology were laid in a military expedition. With the army that Napoleon conveyed to Egypt in 1798., in pursuance of his enter prises/ there were a number of draughtsmen and of keen scientific enquirers, and these made so good a use of their time and opportunities that they collected an unprecedented mass of topographical informa- tion* On the materials which they brought together is based the great Description de V&gyfte published in many volumes by the French Academy between 1809 and 1813, in which we find the first systematic account of the monuments of the Valley of the Nile. The inscriptions reproduced in this publication, although the copies were not without faults — as was to be expected, considering that they were in a script as yet completely unintelligible — fur- nished a quantity of useful material for those who first seriously attempted to unlock the secret of the hieroglyphs. When, as we shall see in the next section, Champollion had made some progress with the decipherment of the inscriptions, he was enlisted in the second great survey, that of Rosellini in 1828, Xhis enterprise considerably enlarged the body of knowledge accessible on the subject of the topography of the country; and the reading of the inscriptions made it possible for the first time to arrange the monuments in some historical order. The result of the expedition was not published, however, till after Champollion "s death in 1831* These explorations had confined themselves to the lower part of the Nile valley — that below the Aswan cataract. The study^of Egyptian remains In Nubia, and as far south as Khartum., was the work of the next survey, that of Lepsius in 1 840, This scholar had become the most prominent authority of the day on the Egyptian language. Not only did he examine the surface, but at Memphis and other places he made excavations. He thus greatly enlarged the geographical limits within which Egyptian remains were to be studied, and, further, in returning, he discovered and published Ill, n] EXPLORATION IN EGYPT 117 the important inscriptions left behind by the ancient Egyptian miners *at the copper-bearing parts of the Sinai Peninsula. By this time the main facts as to the surface antiquities of Egypt had been ascertained and put on record. All the monu- ments that a traveller would see in journeying through the country had been noted and delineated. Of course hardly a year has passed since then without adding some detail — a new inscription or graf- fito, for example — but in the main the statement may stand, that Lepsius exhausted the general topographical study of the country. The peculiar conformation of Egypt, a long narrow strip on each side of a river, and bordered by uninhabitable desert, makes it possible for a single expedition to cover the whole ground in a way hardly possible in any other country. More scientific and artistic cartclgraphy may have been undertaken since the time of Lepsius: the Egypt Exploration Fund is carrying out a detailed archaeo- logical survey; and the erection of the Aswan dam, which involved the submersion of important ancient remains over a wide extent of territory, necessitated a close examination of the country af- fected. But most of the significant work subsequent to Lepsius has been excavation rather than exploration. (£) DECIPHERMENT The few particulars of value regarding the Egyptian hiero- glyphs, preserved by a number of more or less obscure Greek writers, would never have been sufficient to enable a scholar to decipher a single inscription. Till the discovery at Rashid (Ro- setta), near Alexandria, by an officer of the Napoleonic expedition, of a bilingual inscription, in Egyptian and Greek, and till its sub- sequent analysis by European scholars, no solution of the riddle of Egyptian writing could be found. The Rosetta stone is a large slab of basalt bearing an inscrip- tion three times repeated. At the bottom the text appears in Greek, at the top in hieroglyphics; in the middle it is given again in Egyptian, but in a cursive simplification of the hieroglyphic called 'Demotic/ The hieroglyphic part of the inscription is much broken, and every line has lost its beginning and end; the Greek portion is also imperfect at the bottom. With regard to the con- tents of the inscription, all that we need say is that it is a proclama- tion by the priests of Memphis, setting forth the good deeds of Ptolemy Epiphanes, and decreeing that his statue shall be set up in every temple in Egypt. The Rosetta stone is often supposed to have been the sole key used for the solution of the problem of decipherment. But this is ii 8 EXPLORATION AND EXCAVATION [CHAP. not the case. The decipherers proceeded on the method adopted by readers of the common substitutionary form of cryptograms. The first step in such a process would necessarily be the dfeter- mination of the phonetic meaning of the characters; and the lan- guage of the Egyptian part of the inscription being unknown, this could not be done unless with the aid of -proper names^ which would be common to the Greek and the Egyptian. The Rosetta stone, in its present condition, happens to contain but one proper name — that of Ptolemy, In Zoega's De origine et usu obeliscorum (1797), the happy speculation was adventured that groups of characters surrounded with an oval ring or cartouche are proper names, or else especially sacred formulae. They are, as a matter of fact, royal names; and such a group of characters occurred in the Rosetta inscription at places corresponding to the appearance of the name riTOAEMAIOS (Ptolemaios) in the Greek version. An obelisk from the island of Philae supplied in 1822 the neces- sary further material. This bore inscriptions in Greek and Egyp- tian; it had the name IITQAEMAIOZ, and corresponding to it a cartouche identical with that on the Rosetta stone; it had also the name K A EOF"! AT PA (Cleopatra) with a different cartouche corre- sponding. Now the hieroglyphic letters in these cartouches (setting them out in a row) were respectively as follows; (A) w i x a 3 4 5 67 (B) Here it is obvious that A I is the same as B 5, and thus must be equated to the Greek fl (/>), which comes in the same positions in the Greek form of the names. Likewise A 4 is the same as B 2, and by the same argument must be equivalent to A (/). Again A 3 and B 4 are alike, and must therefore somehow represent the Greek O. In B, letters 6 and 9 are the same, and must have been regarded as equivalents to A, We have now got a framework of known letters, with gaps between them that can be filled imme- diately by reference to the Greek; we can thus identify A 2 as T, A 5 as M, and presumably A 7 as £ (s\ the preceding letter being in some way representative of the group of Greek vowels A!O. Similarly, in B, we learn to treat letter I as K, 3 as the equivalent of E, and 8 as P (r), 7 is T, but here we are introduced to the differentiation of cognate sounds, for it is a different character which is used for the same Greek equivalent in the first name. As Ill, n] DECIPHERMENT OF HIEROGLYPHS 119 for the two remaining characters, their explanation came in due time wB.en it was recognized that they always follow and distin- guish divine female proper names ; the one is /, the feminine suffix, the other an egg. An alphabet of eleven phonetic characters was thus obtained. The list was extended by applying it to other cartouches found in the publication of the French Academy, some of which contained the names of Roman Caesars or of Ptolemaic monarchs or queens, known from accessible historical sources. A few letters of each of these being determined, the rest followed automatically, as in the solution of a cryptogram. When a sufficient body of phonetic characters had been determined, the application of the key thus obtained to*the body of the Rosetta inscription proved that the Egyptian language was the ancestor of the modern Coptic. This tongue therefore provided a clue, making it possible to identify the common words with their Greek equivalents, and to systematize the grammatical structure of the language. Thus were the founda- tions laid on which three generations of Egyptologists have built ever since. The study, however, grows in complexity as it ad- vances. The language is now known to have changed greatly during its long life: the ancient Pyramid texts are in a very differ- ent form of the language from the inscriptions of the later empire. The principle that the script represents the consonantal framework of the language only (adopted by the Berlin school of Erman, but first enunciated by Brugsch in 1 857), and that the characters once supposed to be vowels are not so primarily, has added serious difficulty to the grammatical study (see p. 341 $$.*)* Readers of books on Egyptology are often perplexed by the variety of spellings adopted by different scholars in rendering Egyptian words, and especially proper names. To make them pro- nounceable at all, the vowels must be supplied; but in most cases there is little or no guidance to the correct vocalization. And even the nature of the nuances which distinguished the sounds of the so-called homophonous consonants is not always certain: thus, there are several kinds of d^ k and h sounds, as there are in the Semitic languages, but the nature of the differences between them cannot always be determined. There is thus an unavoidable differ- ence of opinion among scholars as to the true principles of trans- literation of words written in hieroglyphics : and whenever Hero- dotus or Manetho happens to give us a Grecized form, that form is often adopted for simplicity's sake. It is regrettable that personal and international jealousies have done much to obscure the question of the man or men first respon- 120 EXPLORATION AND EXCAVATION [CHAP sible for this great addition to knowledge. Four names stand out prominently in the history of the pioneer researches : the renowned Oriental scholar, De Sacy; Akerblad, Swedish Minister at R5me; Thomas Young; and Jean Frangois Champollion. It was the de- motic text which first attracted attention in the Rosetta Stone, and a facsimile was prepared by the Society of Antiquaries of London for distribution among scholars. De Sacy and Akerblad first pub- lished dissertations upon it in 1802, and the latter succeeded in identifying correctly fourteen of its characters. Dr Thomas Young (1773—1829)3 one of those singular * Admirable Crichtons* who were more numerous in former generations than they are now,, in these days of specialization, published in 1814 a study of the de- motic characters with an alphabet embodying AkerbfexFs results* but without sufficient acknowledgment of his predecessor's ^rork; it is true, he seems to have arrived at his own results independ- ently, but he had certainly read Akerblad's essay before he pub- lished, his own. Later, in 1 8 1 8, he contributed an extensive article on Egypt to the Encyclopaedia Britanmcay embodying all his re- searches, with an explanation (on the whole very incorrect, it must be confessed) of about 200 hieroglyphs. The true founder of scientific Egyptology, so far as the lan- guage is concerned, was undoubtedly Champollion (i79o~""I832)> who, again, worked on similar lines to the other investigators, and (as it would appear) with full knowledge of the progress of their researches. It is regrettable that, as Young absorbed the work of Akerblad, so he seems to have quietly appropriated the work of Young, without doing justice to his useful pioneer labours. He possessed, however, a thorough knowledge of Coptic, which his rivals lacked, and without which even in modern times a scientific study of the language is impossible; and thus equipped, he was able far to outdistance his competitors, and to win for himself a permanent fame which a little more generosity to his fellow- labourers would not have diminished. (0 EXCAVATION It is impossible to give any complete survey of the history cf Egyptian excavation. A few details only can be mentioned. We have already seen that Lepsius conducted some excavations at Memphis and elsewhere during his survey of the country. But the real founder of excavation work in Egypt was Augusta Marietta, the first director of the Cairo Museum. For the thirty years fol- lowing 1850 Mariette had the whole work of excavation in his own hands under an exclusive permit; and as all Egypt was virgin Ill, u] CHAMPOLLION, MARIETTE AND FOLLOWERS 121 soil before him, it is not surprising that his name is associated with some of the most epoch-making discoveries that have been made in trie country. The Serapeum of Memphis (the cemetery of the sacred Apis bulls), the temple of the Sphinx at Gizeh, and the cemeteries of Sakkarah, are among the chief fruits of his labours, as well as the clearance of the great temples of Abydos, Medinet Habu, Der el-Bahri, and Edfu from the rubbish that centuries of Arab neglect had allowed to accumulate. At the same time in- numerable inscriptions, works of art, statues, and other minor ob- jects enriched the museums of Paris or of Cairo. After his death in 1880, while the Cairo Museum continued., very properly, to reserve the right to prevent unique or otherwise valuable objects from leaving the country, permission was extended to representa- tives*of other countries to increase knowledge by conducting ex- cavations on their own account. The reign of Maspero (afterwards Sir Gaston Maspero) in the Cairo Museum was brilliantly inau- gurated by the opening of the pyramid of Unas at Sakkarah, on the walls of whose chamber were found the priceless * Pyramid Texts,* documents of absolutely inestimable value for the philo- logical history of the Egyptian language, and for the study of the early development of Egyptian religious ideas. In 1883 the English Egypt Exploration Fund (now the Egypt Exploration Society) was founded, and has since done steady work in several departments: excavation, surface exploration and sur- vey, including the publication of inscriptions; and Greco-Roman studies, more especially the decipherment and publication of the immense stores of papyri which certain sites have yielded. France, Germany, Switzerland and the United States of America followed suit in establishing and maintaining excavating expeditions. Every year produces a bewildering mass of new material. Among the more striking discoveries are the cache of Royal Mummies (Mas- pero and Loret, 1 88 1), the sites of Tanis and of Naucratis (Petrie, 1884—5% the Tell el-Amarna correspondence (1887), the site of Bubastis (Naville, 1887—9), Tell el-Amarna, and the numerous relics of the Heresy of Ikhnaton (Petrie, 1891), the Fayyum ex- cavations, which resulted in the identification of the Lake Moeris and of the Labyrinth of Herodotus, the Greco-Roman mummy portraits, and the first great collection of Greco-Roman papyri (Petrie, 1888), the treasures of the pyramids of Dahshur (De Morgan, 1 8 94), the tombs of the early dynastic kings at Abydos (Amelineau, 1895, continued in later years by Petrie and by Naville), the * Israel7 stele of Merneptah (Petrie, 1896), the *pre- dynastic ' race, first found at Nakadah (Petrie, 1896), the Tombs of 122, EXPLORATION AND EXCAVATION [CHAP. the Kings near Thebes (various explorers and years — most re- markable, probably, being the rich tomb of the parents of Queen Tiy, and of the Queen herself, by T. Davis, 1 904-6), the tonlb of Osiris at Abydos (Naville, 1902, and following years), the Ara- maic papyri at Elephantine (i 905 and following years), the de- tailed study of the temple of Der el-Bahri, resulting in the dis- covery of the wonderful Hathor cow-figure and many other works of art (Naville, a work of many years). This may suffice to give some idea of the work that has been no less energetically pursued since the date at which this enumeration stops1. IIL MESOPOTAMIA (*) SURFACE EXPLORATION The testimonies of early travellers regarding the 'remains of Mesopotamia are for the greater part confined to the two biblical sites of Nineveh and Babylon — the former close to the town of Mosul, the latter not far from Baghdad. The tradition of the identification of these two sites was never quite lost, although some travellers express themselves as sceptical about them. From the point of view of archaeological discovery, however, the criticisms and observations of the seventeenth and eighteenth century pioneers are of historical interest only. As in the case of Egypt, it was the nineteenth century which witnessed the real beginning of scientific work in Mesopotamia. In the important department of surface surveying the first name that calls for mention is that of C. J. Rich (1787— 1 820), Resident in Baghdad of the East India Company, who employed his leisure in visiting and surveying the gigantic mounds and fields of ruins which represent the ancient cities of Babylonia and Assyria. By his labours, accurate knowledge became available for the first time as to the real magnitude and outward appearance of the remains at such sites as Babylon, Arbela, Nineveh and others* During, and soon after, Rich's official residence at Baghdad, the best-known remains were visited by Buckingham (i 8 r 6), R» 1C Porter (i 8 1 8), Mignan (1837), 9* ^ Eraser (1834), and other travellers, all of whom added details of importance to the facts already recorded^ But a detailed survey of the whole region was still lacking : there was as yet nothing but the observations of single travellers upon individual sites. This want was in part supplied by the survey, in 1835—7, of the courses ^of the Euphrates and Tigris, by a British expedition tinder the direction of General Chesney. The maps which embody 1 On the work of the Egyptian Research Account, the Egyptian Exploration Society, etc., see p. 625 (r). Ill, in] EXPLORATION IN MESOPOTAMIA 123 the results of this work were the foundation for all later topo- graphicM investigation. Although the purpose of the Chesney ex- pedition was commercial and political rather than scientific, science profited in no small degree by the results of the work, and a great stimulus was given to further exploration. Of more immediate scientific value were the surveys of Assyria by JL F. Jones (1852), and the unfinished reconnaissance of Babylon under Selby, in the early sixties. (<£) DECIPHERMENT The decipherment of the cuneiform characters was a task of considerably greater difficulty than that of the Egyptian hiero- glyphs. In the latter study, the investigators had the assistance of inscriptions in a form of writing and language so thoroughly well- known as Greek; and when the Egyptian words themselves began to emerge from the hieroglyphic mystery which enshrouded them, their similarity to their Coptic progeny made the further task of interpretation comparatively smooth. In the case of the cuneiform characters, however, the decipherers had first to contend with the complexity and apparent want of individuality of the symbols themselves, and with their enormous number. Nor was there any available translation, in a known language, of any of the inscrip- tions to be analysed. True, the key was ultimately obtained, as in all such cases, by the use of bilingual inscriptions; but the 'trans- lation' had itself to be first made intelligible. Though the deci- pherers were not obliged (in current phrase) to interpret ignotum per ignofius, it is certainly true that they had to interpret ignotum far ignotum. Already in 1765 Niebuhr had surveyed the imposing ruins of Persepolis, a site which had shared with Nineveh and Babylon the interest of the earlier travellers, but by reason of their magnitude rather than of their historical associations; and he had pointed out that the cuneiform characters, while possessing a superficial simi- larity in all inscriptions, were not always identical in detail. There were certain inscriptions which appeared to be in a simplified form <?f the script (Old Persian) : in these comparatively few characters — some forty — were employed, and the words were divided by a single oblique wedge. In others a more complex script, em- ploying many characters, was used (Susian or Elamite). A third form (Babylonian) was distinguished by a still greater elaboration of the groups of wedges which formed the individual characters, and by an even greater number of signs. In many cases these three forms were found side by side, and it was natural to suppose 124 EXPLORATION AND EXCAVATION [CHAP. that such groups were a threefold presentation of the same state- ment, like the three parallel legends on the Rosetta stoneC The first successful attempt at deciphering any of the cilnei- form characters was made by G. F. Grotefend, who in 1 802 identi- fied three royal names in the simplified script. Being ignorant of Oriental languages, he was unable to make further progress; but he had succeeded in ascertaining the correct values of about one- third of the Old Persian letters. His dissertation, presented to the Gottingen Academy, was refused publication; not till 1893 was it unearthed and printed from Grotefend's MS, as a landmark in the history of cuneiform studies. It is noteworthy that the first steps towards the unriddling of the hieroglyphs of Egypt and the cuneiform characters of Mesopotamia were made*:*! the same year. Meanwhile at Naksh-i-Rustam. inscriptions in Greek aftd in the Pehlevi script of Persian afforded the clue, in the hands of De Sacy, to the latter character; and Anquetil-Duperron (stimulated by Thomas Hyde's History of Persian Religion, 1700)., succeeded by his studies of the Avesta in further advancing the knowledge of the Old Persian,, the language which ultimately proved to be of great value for the decipherment of the cuneiform. Even so early as 1823, Saint-Martin had shown that a vase, first published in 1762, with inscriptions in Egyptian and cuneiform, was a bi- lingual : the cuneiform being the Persian equivalent of the Egyp- tian, shown by ChampolHon to read 'Xerxes the Great King/ Grotefend is the Akerblad or the Young of cuneiform studies; their ChampolHon is Sir Henry Creswicke Rawlinson* This mili- tary officer, stationed in Persia on diplomatic duty, without any previous knowledge of Grotefend's work, succeeded independ- ently in finding his key; and as he possessed the knowledge which Grotefend lacked of ancient tongues, such as Zend (cognate with the Old Persian language of the cuneiform inscriptions), he was able to carry the solution of the problem to the end, The discovery of the key rests on what may not unfairly be called a happy guess. Near Hamadan Rawlinson discovered two short inscriptions, set forth in the trilingual form already observed at Persepolis. When the texts of the ' simplified * script, on which it was natural to begin the study, were set side by side, it was found that they were prac- tically identical, save in two places. In the twelfth line of inscrip- tion A there was a word, which we may denote by #, and corre- sponding to it in inscription B was a different word, y* In the nine- teenth line of inscription A a third word, z, was found, and in the corresponding place of B the word x reappeared. Rawlinson assumed that these three words were the names of kings; that the Ill, in] DECIPHERMENT OF CUNEIFORM 125 Inscriptions were the proclamations of successive kings, who In the cotfrse of their inscriptions referred to themselves and to their fathers; and that therefore -the name of the king In the earlier of the two inscriptions would appear in the place assigned to the name of the king's father In the later. By trial this theory could be tested; all that was needed was to find, if possible,, three successive monarchs whose names would fit the alphabetic characters. This proved to be actually the case with Hystaspes, Darius, and Xerxes, in their Old Persian forms; and the identification of their names with the three words gave Rawlinson fourteen characters of the alphabet. With the knowledge of Zend which he possessed, the extension of the decipherment to cover the whole ground of the simplest fb#m of cuneiform, now proved to be Old Persian, was only a matter of time. The two more complex forms still remained unknown, but the conquest of the Persian inscriptions provided the necessary key. On the borders of Media stands the great isolated rock of Behistun, rising sheer from the surrounding plain to a height of about 1700 feet. On a portion of the surface of this rock, chosen at a height of 300 feet — no doubt to guard against the wilful de- facement of later generations — and specially prepared for the pur- pose,, Darius Hystaspes had caused to be engraved, in the year 516 B.C., a long account of the glories of his reign, and of his triumphs over his many enemies, foreign and domestic. This in- scription, again, is trilingual, and being of considerable length it offers much valuable material to the decipherer; moreover, owing to the nature of its contents, it contains a large number of proper names, the first landmarks to be noted by those who would un- ravel an unknown script. With wonderful perseverance Rawlinsoti overcame the dangers and difficulties of climbing the rocks and of copying and making squeezes of the inscriptions, a work which he began in 1835—7, and continued at intervals, as his official duties gave him opportunity, till, in 1847, he was able to publish a com- plete translation, with full grammatical analysis, of the Persian text. Rawlinson then proceeded to the third (the Babylonian) cunei- form script and with Immediate success. Other labourers were soon attracted into the field, chief among them being Edward Hincks, an Irish clergyman, to whom much of the credit of ex tending the conquest into the more difficult field of the associated scripts is due. Oppert, de Saulcy, and Talbot are also to be teamed among these pioneers. The discovery that Babylonian and Assyrian — for both began to emerge when the force of the characters was determined — were Semitic tongues, closely cognate with Hebrew, 126 EXPLORATION AND EXCAVATION [CHAP. to some extent lightened the labour. The decipherment was sub- mitted to a final test in 1857, when the Royal Asiatic "Society challenged students to prepare translations, without collaboration, for official comparison, of the long inscription on the newly-found cylinder of Tiglath-pileser I. Rawlinson, Hincks, Talbot, and Oppert all submitted renderings; and when these were unsealed and compared, the President of the Society was able to announce that the renderings were so close as to leave no doubt that the true key had been found, and that the new science, loosely named "Assyriology/ was thus established on a solid foundation. The difficulties of the decipherment of the Semitic cuneiform inscriptions, as compared with the Persian, lie in the following facts : the characters are not alphabetic, but syllabic or ideographic; in consequence, as was first pointed out by Botta, words o£ft be written in different ways, using different syllabic combinations, or even ideograms alone; while even a small inflexion of a word may change its external aspect. Moreover, each character has not always one meaning alone, and each syllable is not necessarily represented by one sign alone. Very many of the characters have more than one sound attached to them, to be determined from the context; and there are many groups of different characters having some one phonetic significance in common. This difficulty was felt by the ancient scribes themselves; who, for their own guidance and for the help of their pupils, drew up elaborate sign-lists, which have proved of great service to modern interpreters. This ambiguity will explain the bewildering variety of form which historical and legendary names have assumed in the hands of different scholars — Ut(a)-Napishtim and Shamas-Napishtim; Ur-Bau and Ur~ Engur, and so forth. It is not only such coincidences of reading as the test in 1857* that have given modern scholars their security in reading cunei- form script. Their assurance comes also from the conformity of the Babylonian and Assyrian languages — they are almost one — thus revealed, with the grammatical formulae of Semitic speech; from the Aramaic endorsements sometimes found upon contract- tablets, which give in a well-known script transliterations of namer recorded in the cuneiform documents; from certain late cuneiform renderings of Greek words; from fragments of the Aramaic trans- lation of the Behistun inscription found on Elephantine Island; and from the discovery at Boghaz Keui of the cuneiform (Baby- lonian) version of the treaty between Ramses II and the Hittites, of which the Egyptian version had long been known. The second group of cuneiform characters is now known to Ill, m] SEMITIC AND SUMERIAN I2? represent a later form of the old language of Susa or Elam : It is of an agglutinative construction. The decipherment of this group is di??e largely to Westergaard and Norris (1844—52). Meanwhile*, Hincks in 1850 recognized that the cuneiform used for the writing of the Semitic languages of Mesopotamia was not originally devised for the purpose, but had been borrowed from some other source which was variously called 'Scythian' (Hincks), * Akkadian' (Rawlinson) or 'Sumerian' (Oppert). This was the first seed of the great * Sumerian * controversy. For as in- vestigation advanced, another language made its appearance, of agglutinative form, a language which had been assiduously studied by the Babylonians, for tablets with vocabularies and interlinear translations -from this unknown tongue into Babylonian were dis- covered. It seemed impossible that the very ancient Babylonian civilization should have been preceded by another, altogether for- gotten, which had in fact taught its arts to the Babylonians : and some scholars accordingly sought refuge from such a conclusion by supposing that the parent culture was a purely artificial con- trivance of the Babylonian priests. This view is no longer held, but there is still some difference of opinion as to the priority of Sumerian or Semitic; and the uncertainty whether particular proper names are to be regarded as the one or the other accounts for some of the above-mentioned differences in transliteration* See further, pp. 357, 364 sy.y 371, CO EXCAVATION The story of formal excavation in Mesopotamia begins in 1842, with the exploration of the mounds now called Kuyunjik and Khorsabad respectively. Paul Botta was appointed French consu- lar agent at Mosul in that year, and his ambitions were aroused by the sight of the great mounds almost at his door. Dependent on his own personal resources, he at first attacked Kuyunjik, but with discouraging results; afterwards, acting on a hint given him by a friendly native, he turned his attention to Khorsabad, and almost immediately uncovered the great sculptures of the palace o£ Sargon* Three years later (Sir) Henry Layard opened the mound of Nimrud (the biblical Calah), and in it discovered the palaces of Ashur-nasir-pal, Sargon, Shalmaneser, and other great Assyrian monarchs. He found the huge winged bulls, the graphic sculptures of Ashur-nasir-pal, the black obelisk of Shalmaneser, and other priceless treasures which now adorn the British Museum. In his second expedition (1849-1850) he found at Kuyunjik the palace- 128 EXPLORATION AND EXCAVATION [CHAP. sculptures of Sennacherib, and the first of those great and varied store-houses of vanished literature, the palace libraries? But to make a full list of Layard's numerous and varied discoveries,reven to enumerate the mounds which he examined, would here be im- possible. It is little wonder that popular enthusiasm was aroused, for the public mind was at the time being rendered uneasy by the advances of biblical criticism, and the enunciation of the doctrines of evolution, with all that they implied. The obvious bearing of these discoveries on the credibility of the Scriptural history did not fail to kindle an interest that mere art or archaeology could not hope to arouse. While we must in fairness not forget that in the fifties of the last century the principles of scientific excavation— a new disci- pline— had not been established, we cannot withhold criticism from the methods of these early pioneers. It was a fatal mistake to wander from mound to mound, pitting here and there in search of sculptures and of piles of tablets. Not only were the architectural monuments disturbed, but the eyes of the local natives were opened to the value of loot so easily gathered. The expense of completely working out one single site would indeed be great, but the scien- tific results would exceed to an extent incalculable those of the same amount of scrappy work at a large number of sites. Layard's work was continued, after his departure from the scene of his labours, by his former companion, a cultivated native of the country, Hormuzd Rassam. He, too, had the good fortune that proverbially attends the beginner: in 1853 he found at Kuyunjik the palace of Ashur-bani-pal, with its sculptures and its library; many years later (1878) lie discovered the famous bronze gates of Balawat, and in 1882 he finished his career with the identifica- tion and partial excavation of the site of Sippan So far Assyria had been the centre of interest. In 185*4 Baby- lonia began to attract the attention of the digger, the work of Loftus on the site of Erech being the first extensive work of the kind in that country. He was followed by Layard, who excavated, but without conspicuous success, in Babel, Nippur, and one or two other sites. The great mounds of Babylonia were found to ofj^r no such immediate and startling results as those of the Assyrian palaces. The buildings were for the greater part in brick, and their interest was religious rather than artistic or domestic* This does not, of course, detract from their scientific importance; but it makes it less easy to interest those on whom the explorers have to count for financial aid. Nor did the early Babylonian excavations produce such great literary harvests as the libraries of Assyria* Ill, in] THE DELUGE TABLET 129 Unable to linger over these early excavations we have to pass over witii a bare mention Taylor's digging at Abu Shahrein and his examination of the temple of the moon-god at Mukayyar (Ur) in 1 855,, and Rawlinson's investigation, about the same time, which finally settled the character of the tower-temple of Nebu- chadrezzar, called in modern times el-Birs, and popularly sup- posed to be the ruin of the Tower of Babel (p. 503). The next great discovery was made in the British Museum. On the 3rd of December 1872, George Smith, a man of unusual natural gifts, who had at first been employed as an assistant in a quite subordinate capacity, read a paper before the newly-founded Society of Biblical Archaeology which began with these words — surely the boldest announcement ever made of an epoch-making find : 4 A short time back I discovered among the Assyrian tablets in the British Museum an account of the Flood/ This was the first of that wonderful series of Cosmogonic legends that Assyria has yielded to us, new examples of which still come to light from time to time, and which, joined with the progress of geology and of historical criticism, have revolutionized the current conceptions of the early chapters of Genesis. Enthusiasm was aroused to an extraordinary height, and the Daily Telegraph subsidized an ex- pedition to be conducted by Smith in search of more tablets. In this and in a later expedition, at the charges of the British Museum, Smith collected a great literary spoil from the already much-plundered mound of Kuyunjik. Unfortunately on his third expedition, in 1876, to make further excavations at Nineveh, he collapsed in health and died at Aleppo. The centre of gravity of interest now once more shifted to Babylonia. The French consul at Basrah, De Sarzec, turned Ms attention to the mound of Telloh, afterwards identified with the Babylonian city of Shirpurla, or Lagash, as the name is now gener- ally read. His success was astonishing. The famous sculptures and other relics of Gudea and other early Sumerian kings revealed a style of art hitherto unknown, told us of the unimagined greatness of these ancient and forgotten monarchs, opened an unsuspected chapter in the world's history, and, among other results of scarcely less importance, showed us primitive forms of the writing which developed into the cuneiform script. The Germans now entered the arena, hitherto occupied exclusively by France and Britain; and Moritz and Koldewey by their work at the sites of Surghul and el-Hibbah still further enlarged our knowledge of the Sumer- ian empire: this expedition began it£ work in 1887. In the follow- ing year America joined in, and the expedition to Nippur, begin- 130 EXPLORATION AND EXCAVATION [CHAP. ning in 18885 and continued at intervals jfor many years, had rich results, the publication of which is still in progress. De Sarzec died in 1901, and two years later the French ronce more attacked Telloh, greatly adding to the material already col- lected from that magnificent site. But during the present century the interest of these Babylonian sites has been to some extent eclipsed by the results of De Morgan's Persian mission, beginning with the first excavation at Susa in 1897-8, Such 'finds* as the great obelisk of Manishtusu, king of Kish, the 'Victory Stele* of Naram-Sin, with its most graphic and life-like reliefs, and above all the world-famous code of Hammurabi, one of the most striking archaeological discoveries ever made, have secured for the Susa expedition a place in the front rank of undertaking* of the kind from which it can never be dislodged. * Of later work in Babylonia and Assyria we may mention the further British Museum excavations at Kuyunjik by L. W. King and R. Campbell Thompson (1903—5), and those of Koldewey at Babylon, carried on continuously until the outbreak of war in 1914. However in 1918 Thompson and Hall were able to resume excavation and among other discoveries at Abu Shahrein (Eridu) found indications of a presumably prehistoric occupation by some Elamite or related people. See further. Chap* x, pp* 373 IV. SYRIA AND PALESTINE 0) SURFACE EXPLORATION The unique interest of Syria and Palestine has made the surface of this region familiar to the pilgrims of many generations. But here also, as in the other countries considered in this chapter, scientific exploration begins in the nineteenth century. Seetzen in 1801, and Burckhardt, who discovered the rock-city of Petra in 1809, as well as Buckingham in 1821, may be said to have been the pioneers; these men penetrated, in the face of great difficulties, into recesses of the land till then unvisited. Costigan in 1835 first attempted to navigate the waters of the Dead Sea and to explore its mysterious shores. But he fell a victim to the deadly fevers of Jericho^ as did Molyneux in 1847, his successor in the same enter- prise, and also, many years afterwards, Tyrwhitt Drake, an officer of the Palestine Exploration Fund. The first man who endeavoured to make a reconnaissance of the whole country was Edward Robin- son, an American Congregationalist minister, whose studies ex- tended over the years from 1838 to 1852, and whose book, Bib- lical Researches in Palestine^ notwithstanding the primary exegeti- cal interest which the title expresses, marks an era in Palestinian Ill, iv] EXPLORATION IN PALESTINE 131 geography. Robinson was followed by Tobler (1845—1866) and Guerin (1852 and following years), whose rich topographical col- lectlolis paved the way for the Ordnance Survey conducted under the auspices of the Palestine Exploration Fund. The admirable cartographic work of Van der Velde (1851—2) must not be for- gotten; and while his maps were superseded only by the Palestine Exploration Fund's survey, the wonderful photographic survey carried out under military auspices during the recent war marks a very great advance in this department. In Syria, which had been neglected in comparison with Pales- tine, and the thorough exploration of which Is still a geographical desideratum, good work was done in the middle of the nineteenth century by I>e Vogue and Waddington, by the former in studying the atfcient churches, and by both in collecting the Semitic and Greek inscriptions, which are so numeroiis in various parts of Syria, Valuable surveys were made in Phoenicia (1860) by Renan; and, in recent years, by an American expedition under H. C. Butler. The Palestine Exploration Fund, the first public body to devote itself exclusively to work in this region, was founded in London In 1865. After some preliminary work of excavation in Jerusalem, it embarked in 1 8 70 on the Ordnance Survey above mentioned, under the direction of Conder and Tyrwhitt Drake. The latter was a young zoologist, but endowed with a marvellous capacity for acquiring languages. He had already accompanied the noted linguist, E. H. Palmer, in an adventurous march through the desert of the Tih, and seemed well fitted by his physical powers to take part in so laborious an enterprise as the survey. But he died shortly after the work began. His place was taken by a young lieutenant of engineers, afterwards to be famous as Lord Kitchener of Khartum. The maps, and the series of volumes accompanying them, containing name-lists, descriptions of fauna and flora, and brief — sometimes too brief — notices of the surface appearances of the various sites and mounds and fields of ruins scattered so richly over the country, are the basis of all later topographical work. It cannot be claimed that the work was as full as it might be, foT there are still plenty of gleanings left, even for an explorer who does not dig; and the knowledge since acquired on the subject of the chronology of pottery, as explained above (p. 1 14), reopens the question of many of the identifications of ancient sites proposed by the explorers. Still, these defects are mere sun-spots, and the work will always stand as a monument of industry and enthusiasm. Some later surveys, such as those of Schumacher and of Musil, in the land of Moab, may be noticed in passing* 1 32 EXPLORATION AND EXCAVATION [CHAP. (6) DECIPHERMENT There is nothing to say in this section under the head $f de- cipherment, as all the inscriptions of Palestine and Syria are in known scripts and languages — though the interpretation of many of them offers no little difficulty. The Hittitc hieroglyphs are treated in a later section. Of the inscriptions in which the mys- terious Philistines may have recorded their sentiments, and which, when they appear, may be expected to offer problems analogous to those of the hieroglyphic and cuneiform mysteries, none has as yet rewarded any excavator,, and the * speech of Ashdod ' still re- mains unknown* (0 EXCAVATION The excavations which have been carried on in the Holy Land, like those in Mesopotamia, have all been partial. Lack of funds, and the limitations imposed by the Turkish Imperial Permits, which required the work to be completed in two or at most three years, have hitherto prevented the attainment of the ideal of carry- ing out an excavation to the very end* The sites dug have been as follows; isolated spots in and around Jerusalem (Warren, 1867— 70); Tell el-Hesy, the ancient Lachish (Petrie and afterwards Bliss, 1891-2); the Soxith Wall of Jerusalem (Bliss and Dickie, 1 894-7); Gath (?), Azekah (?), and Marissa (Bliss and R. A, S. Macalister, 1899—1900); Gezer (Macalister, 1901—8); Beth-Shemesh (D. Mackenzie and F. G, Newton, 1910—12), Megiddo (Schumacher, 1903— 4);Taanach(Sellin, 1902—3); Jericho (Sellin and Watzinger, 1909—10); Samaria (Lyon and Reissner, 1908—10). The above list down to Beth-Shemesh contains the sites dug for the Palestine Exploration Fund; Megiddo and Jericho were excavated under German auspices, Taanach under Austrian, and Samaria under American. Since the War excavation has been undertaken at other sites, such as Ashkclon (Garstang, 1920—2). The results which have rewarded this activity have been very different from those obtained by Egyptian and Mesopotamian ex- plorers. It is not in human nature for a Palestine explorer to r#ad without a feeling of envy of the rich epigraphic and artistic harvest gathered by his brethren in Egypt and Mesopotamia. The Moabite Stone, found almost at the beginning of scientific excavation (in 1868), is always before his eyes as a stimulus and encouragement; but so far that extraordinary monument stands alone, and many hundreds of tons of earth have to be removed before a find of really outstanding importance can be expected* Ill, iv] PALESTINE, SYRIA AND ARABIA I33 The chief discoveries that have been made on Palestinian soil (including in the denomination Syria and the land across the Jordan) have not been numerous. In epigraphy, the stele of Mesha, commonly called the Moabite Stone, stands easily first as a