ttP;NRLF 
 
 E *** ' 
 
 B 4 001 37b 
 
 ksn 
 
 ■r <-■"■',/ ■ 
 
 ■■*'" > t ■ ' <•' '' ; *- " ■ 
 ■•»■*'/'-.■.-;■•"■♦■•■;•■ 
 
 
 aaEg 
 
 ■ i ■. . . 
 
 WJ 
 
 
 
 jQC 
 
 ^mij: 
 
 
 V ; --< -- ~- ■* ' » r >-' ■' 
 
 "l '"*'" 
 
 *» ' "' ■ ' / .*■-'••"'#■■ ' 
 
 ' ' 
 
 Bui L '-"' •'•■ ffl 
 
 ra 
 
 Bl .■''■•"'■ ■■.•'■'••'■ 
 
 ^yfy 
 
 iff '■■■'■■- '■■ 
 
 jQ 
 
 '' "'V 
 
 '^HwHaanVt'yni " k • 
 
 
 
 TOBttS 
 
 **vv* 
 
J> 
 
 \r 
 
Hi u ' oi Gorgon, prow mi: Pediment <>r nn- Temple of 
 Svl Minerva at Hath ($). See page V3. 
 
 frontispiece 
 
THE ROMANIZATION 
 OF ROMAN BRITAIN 
 
 ■?■ BY R HAVERFIELD 
 
 THIRD EDITION, FURTHER ENLARGED, 
 WITH TWENTY-SEVEN ILLUSTRATIONS 
 
 OXFORD 
 AT THE CLARENDON PRESS 
 
 1915 
 
OXFORD UNIVERSITY PRESS 
 
 LONDON EDINBURGH GLASGOW NEW YORK 
 TORONTO MELBOURNE BOMBAY 
 
 HUMPHREY MILFORD M.A. 
 
 PUBLISHER TO THE UNIVERSITY 
 
 tl 
 
 • | 
 
4-' 
 
 o 
 
 PREFACE 
 
 The following pages are based on a paper which I read 
 to the British Academy in 1905 and which, according to 
 the custom of the Academy, was issued both in its general 
 ; Proceedings ' (ii. 185-217) and separately. In its separate 
 form it soon ran out of print, and in 1912 the Delegates 
 of the Clarendon Press published a new edition, revised 
 and enlarged to about twice its original size. This second 
 edition is now in turn exhausted ; in issuing a third, I 
 have revised and in places recast the text, and I have again 
 increased considerably both text and illustrations. I have 
 tried to preserve the character of the work as a treatise 
 on a definite subject which seems to possess quite real 
 interest and importance ; I have also endeavoured so to 
 word my matter that the text, though not the footnotes, 
 can be read easily by any one who is interested in the subject, 
 without special knowledge of Latin. 
 
 I have to add one regret. Last June my friend Franz 
 Cumont told me that he was writing a volume somewhat 
 similar to mine, which would describe the Romanization of 
 his own country, Belgium. In the Roman age Britain and 
 northern Gaul, which includes the area of Belgium, were so 
 closely akin in many ways that such a volume from Cumont 's 
 pen could not fail to cast strong new light on Romano- 
 British problems : I hoped to learn much from it for the 
 betterment of this edition. War has come between. His 
 
 oon < iui 
 
PREFACE 
 
 volume has hccn completed ; it is brilliant, erudite, instruc- 
 tive, and in addition excellently illustrated [Comment la 
 Belgique Jut romanisSe, Vromant, Paris and Brussels, witl 
 sixty-nine illustrations). Bui it reaches me at the very latest 
 
 stage of my printing, and I can refer to it only here, in ; 
 Preface which is a postscript, and in a few footnotes. Oi 
 
 most matters common to the two books our views agree 
 
 There is. indeed, only one point of nionn nt on which the\ 
 
 do not agree. In discussing the types <>r houses used ii 
 Roman Britain and in northern Gaul, M. Cumonl is inclincc 
 to admit more <>f Mediterranean influence in reaped of tin 
 so-called ' corridor house ' than I can do (below, p, 1*2). 
 
 I have to thank Mr. I). Atkinson, Research Felloe 
 of Reading University College, for various efficient helj 
 in preparing this edition. 
 
 II. 
 Oxford. 
 
 February 9, 1915. 
 
CONTENTS 
 
 chap. page 
 
 List of Illustrations ..... 7 
 
 1. The Romanization of the Empire !) 
 
 2. Preliminary Remarks on Roman Britain . 23 
 
 3. Romanization of Britain in Language . . 29 
 
 4. Romanization in Material Civilization . . 3(5 
 
 5. Romanization in Art . . . . .48- 
 
 0. Romanization in Town-Life, Local Govern- 
 ment and Land-Tenure . . . .57 
 
 7. Romanization in Religion . . . .67 
 
 8. Chronology of the Romanization . . . 74 
 
 1). The Sequel, the Celtic Revival in the Later 
 
 Empire ....... 80 
 
 Index 
 
 81) 
 
LIST OF ILLUSTRATIONS 
 
 FIG. PAGE 
 
 Head of Gorgon from Bath. (From a photo- 
 graph) ...... Frontispiece 
 
 1. The Civil and Military Districts of Britain . . 25 
 
 %j 
 
 2, 3, and 4. Inscribed tiles from Silchcster. (From 
 
 photographs) ..... facing p. 30 
 
 5. Inscribed tile from Silchcster. (From a drcaving 
 
 by Sir E. M. Thompson) ..... 30 
 
 0. Reconstruction of the Plaxtol Inscription from 
 
 various fragments ..... 33 
 
 7. Inscribed tile from Plaxtol, Kent. (From a photo- 
 
 graph) ...... facing p. 33 
 
 8. Fragment of inscribed jar, from Ickleton. (From 
 
 a photograph) .... facing p. 33 
 
 9. Ground-plans of Romano-British Temples. (From 
 
 Archaeologia) ...... 37 
 
 10a. Ground-plan of House at Brislington . . 38 
 
 10b. Ground-plan of House at Clanville. (From Archaeo- 
 logia) ........ 38 
 
 11. Ground-plan of Courtyard House at Northleigh 41 
 
 12. Plan of a part of Silchcster, showing private 
 
 houses, the Forum, and the Christian Church. 
 (From Archaeologia) ..... 43 
 
 13. Painted pattern on wall-plaster from Silchcster. 
 
 (Restoration by G. E. Fox, Archaeologia) facing p. 44 
 
 1-1. Plan of British Village at Din Lligwy. (From 
 
 Archaeologia Cambrensis) .... 4(3 
 
 15. Plan of House at Frilford ..... 47 
 
 1(5. Late Celtic Metal Work (From a photograph) 
 
 facing p. 48 
 
8 LIST OF ILLUSTRATIONS 
 
 PIG. PAGE 
 
 17. Fragments of New Forest pottery with leal 
 
 patterns. [From Archaeologia) . . . 19 
 
 18. Castor Ware. [From photographs) . facing p. ."><) 
 
 19. Hunting Scenes from Castor Ware. (From Artis, 
 
 Durobrivae) ..... facing p. 50 
 
 ■_'<>. Fragment of Castor Ware showing Hercules and 
 
 Hesione. (After C. R. Smith) .... ~>\ 
 
 •_' l . Dragon-brooches. {From a drawing by C. ./. 
 
 Praetoriu8) ....... 0- 
 
 '-'•_\ The Corbridge Lion. (From a photograph) facing p. .">•*'» 
 
 28. Inscription from Caerwent, illustrating cantonal 
 
 government. (From a drawing) . . . 59 
 
 24. Plan of Silchester. (From the author's 'Ancient 
 
 Town-planning \ Fig. 81) ... . 68 
 
 25. Relief of Diana and Hound from Nettlcton, (From 
 
 a photograph) .... Joeing p. ?:3 
 
 20. Relief of Mercury and Rosmerta from Gloucester. 
 
 (From a photograph) . . . facing p. ?U 
 
 27. Ogam inscription from Silchester. (From a drawing 
 
 by C, J. Praetorius, Archaeologia) . . . ^- , 
 
 Not€. For the blocks of tlw* rrontispieoe and ol Pigs. •".. 5, 18, 19, 
 I am indebted to the editor and publishers of the Victoria Count) 
 History. For t In- block of Pijr. 12 I have to thank the Royal Institute 
 of British Architects. The block of Pur. 25 has been kindly lent l»\ 
 
 * • 
 
 the Bath Branch ol the Somersetshire Archaeological Society. 
 
CHAPTER I 
 
 The Romamzation of tiie Empire 
 
 Historians seldom praise the Roman Empire. They 
 regard it as a period of death and despotism, from which 
 manly vigour and political freedom and creative genius and 
 the energies of the speculative intellect were all alike ex- 
 cluded. There is. unquestionably, much truth in this judge- 
 ment. The world of the Empire was indeed, as Momnisen 
 has called it, an old world. Behind it lay the dreams ancl 
 experiments, the self -convicted follies and disillusioned 
 wisdom of many centuries. Before it lay no untravelled 
 region such as revealed itself to our forefathers at the 
 Renaissance or to our fathers fifty years ago. No new con- 
 tinent then rose up beyond the western seas. No forgotten 
 literature suddenly flashed out its long-lost splendours. No 
 vast discoveries of science transformed the universe and the 
 interpretation of it. The inventive freshness and intellectual 
 confidence that are born of such things were denied to the 
 Empire. Its temperament was neither artistic, nor literary, 
 nor scientific. It was merely practical. 
 
 Yet if practical, it was not therefore uncreative. Within 
 its own sphere of everyday life, it was an epoch of growth in 
 many directions. Even art moved forward. Sculpture was 
 enriched by a new and noble style of portraiture and by a 
 school of historical narrative in stone. Architecture found 
 new possibilities in the aqueduct of Segovia and the Basilica 
 of Maxentius. 1 But it was only practical ends — the erection 
 of buildings or the historical representation of men and 
 deeds — that woke the artistic powers of the Romans. The 
 greatest work of the imperial age must be sought in its 
 
 1 Wickhofi, Wiener Genesis, p. 10 ; Riegl, Stilfragen. p. 272. 
 
■c THE KOMANIZATION OF THE EMPIRE 
 
 provincial administration. The significance of this we have 
 conic to understand, as not even Gibbon understood it, 
 through the researches of Mommsen. By his vast labours 
 our horizon has broadened beyond the backstairs of the 
 Palace and the benches of the Senate House in Home to the 
 wide lands north and east and south of the Mediterranean. 
 
 and we have begun to realize the true achievements of the 
 Empire. The old theory of an age of despotism and decay 
 
 has been overthrown, and the believer in human nature can 
 now feel confident that, whatever their limitations, the men 
 of the Empire wrought for the betterment and the happiness 
 of the world. 
 
 Their efforts took two forms. They defended the fron- 
 tiers against the barbarians and secured internal pea* > : 
 they developed the civilization of the provinces during that 
 peace. The first of these achievements was but for a time. 
 In the end the Roman legionary went down before tin 
 Gothic horseman. The barbarians were many : * they were 
 also formidable fighters ; perhaps, without railways and 
 explosives, no generalship could have wholly kept them 
 back. But they won no rapid entrance. From the middle 
 of the second century, when their assaults became violent, 
 two hundred years passed before they won a real footing, 
 and the Roman lines were still held in some fashion even in 
 the beginning of the fifth century. Despotism did not 
 destroy, nor case steal away, the manly vigour of the Empire. 
 Through battles without and tumults within, through the 
 red carnage of uncounted wars, through the devastations 
 of great plagues, through civil discord and sedition dud 
 domestic treachery, the work went on. It was not always 
 marked by special insight or intelligence. Tin- nun who 
 carried it out were not. for tin- most pari first-rate statesmen 
 or first-rate generals. Even in the art of war they were slow 
 lo learn : they clung to an obsolete infantry, they neglected 
 
 Some recent writers, like l)ui>«>i>> iii MHangu Cagnai, pp. 84/2 67, 
 ir\ to minimise theii numbers, but tiny do not seem t<> me quite t<> 
 prove t heir cm-' , 
 
 
THE ROMANIZATION OF THE EMPIRE 11 
 
 new tactics and new engines. Their successes were those 
 (rf character, not of genius. But their phlegmatic courage 
 saved the Empire through many years and secured for the 
 lands within the frontiers an almost unbroken quiet. The 
 age of the Empire is the longest interval — indeed, it is the 
 one long interval — of peace which has yet been granted to 
 any large portion of our world. 
 
 The long peace made possible the second and more lasting 
 achievement of the Empire. The lands which the legions 
 sheltered were not merely blessed with quiet. They were 
 also given a civilization, and that civilization had time to 
 take strong root. Roman speech and manners were diffused ; 
 the political franchise was extended ; city life was estab- 
 lished ; the provincial populations were assimilated in an 
 orderly and coherent culture. A large part of the world 
 became Romanized. The fact has an importance which, 
 even to-day, we might easily miss. It is not likely that any 
 modern nation will soon stand in quite the place which Rome 
 then held. Our civilization seems firmly set in many lands ; 
 our task is rather to spread it further and develop its good 
 qualities than to defend its life. If war destroy it in one 
 continent, it has other homes. But the Roman Empire was 
 the civilized world ; the safety of Rome was the safety of all 
 civilization. Outside roared the wild chaos of barbarism. 
 Rome kept it back, from end to end of Europe and across 
 a thousand miles of western Asia. Had Rome failed to 
 civilize, had the civilized life found no period in which to 
 grow firm and tenacious, civilization would have perished 
 utterly. The culture of the old-world would not have lived 
 on, to form the groundwork of the best culture of to-day. 
 
 The Empire did not, of course, grow into a nation, in the 
 sense in which we now use that word. It resembled modern 
 Austria rather than France or Germany. But it gained — 
 what Austria has missed — a unity of sentiment and culture 
 which served some of the purposes of national feeling. Late 
 in its days, about a.d. 400, a Greek from Egypt, who was 
 also the last great Latin poet, wrote a remarkable praise 
 
12 THE ROMANIZATION OF THE EMPIRE 
 
 of Rome. She (he sang) alone of conquerors had taken bo 
 her bosom the world which she had subdued ; she had been 
 mother, not mistress, and to her men owed it that from 
 Rhone to Orontes — from the Atlantic to the sands of the 
 Arabian desert — they were all one people. 1 Claudian was 
 
 probably echoing here an earlier Greek litterateur. But that 
 neither makes him insincere nor his words untrue. He felt, 
 and felt rightly, that Ixomanizat ion was a real thing. The 
 
 Empire had passed out beyond the narrower ideal of military 
 dominion, which at its birth Vergil had set forth in famous 
 verses. 8 Rulers and ruled had assimilated ; a civilized life 
 had grown up which even its barbarian assailants learnt to 
 honour and accept and which they passed on to later ages. 
 . This Romanization was real. But it was. necessarily, not 
 altogether uniform and monotonous throughout all the wide 
 Roman lands. Its methods of development and its fruits 
 varied with loeal conditions, with racial and geographical 
 differences. It had its limits and its characteristics. First, 
 in respect of place. Not only in the further east, where (as 
 in Egypt) mankind was non-European, but even in the nearer 
 east, where an ancient Greek civilization reigned, the effect 
 of Romanization was inevitably small. Closely as Greek 
 civilization resembled Roman., easy as the transition might 
 seem from the one to the other. Rome nut here that most 
 serious of all obstacles to union, a race whose thoughts and 
 affections and traditions had crystallized into definite cohe- 
 rent form. That has in all ages cheeked Imperial assimila- 
 tion ; it was the decisive hindrance to the full Romanization 
 
 "I the Greek east. A few Italian oases were created l>\ 
 the establishment of coloniac lure and there in Asia Minor 
 and in Syria. Such, for example, were Alexandria Troas, 
 close by ancient Troy, or Antioeh in Pisidia, lately explored 
 
 1 Quod cuncti gens mm sutnus, Claudian, de cons. SHHcfionis, iii. 
 L50-9. The idea seems taken From Aelius Aristides, who in his 
 
 • Praise of Rome' called Iter vdtrw /«t)t;;/' and Bpeaks i>t the Empire as 
 
 (.tin \wpa <tii'<\j/s /.ui \v ij-vKoi' : lie unite in tin- middle of I lie second 
 
 century. 
 
 [en, vi. si? loll, in regere imperio poputoe, etc. 
 
THE ROMANIZATION OF THE EMPIRE 13 
 
 by Sir W. M. Ramsay, or Bcrytus on the Syrian coast. The 
 colonists, the speech, the constitutions of these settlements 
 were Roman, and now and again their citizens won high office 
 at Rome. From Troas, to quote one case only, came the 
 Quintilii who held four consulships in the later second cen- 
 tury and who dwelt in the largest of all the palaces in the 
 Campagna of Rome. But in one after another of these towns 
 the Roman element perished like an exotic plant. 1 The* 
 Romanization of these lands was political. Their inhabi- 
 tants learnt to call and to deem themselves Romans. They 
 did not adopt the Roman language or much of the Roman . 
 civilization. 
 
 .'The west offers a different spectacle. Here Rome found 
 races that were not yet civilized, yet were racially capable 
 of accepting her culture. Here, accordingly, her conquests 
 differed from the two forms of conquest with which modern 
 men are most familiar. We know well enough the rule of 
 civilized white men over uncivilized Africans, who seem 
 sundered for ever from their conquerors by a broad physical 
 distinction. We know, too, the rule of civilized white men 
 over civilized white men — of Prussian (for example) over 
 Pole, where the individualities of two civilized races clash 
 in undying conflict. The Roman conquest of western Europe 
 resembled neither of these. Celt, Iberian, German, Illyrian, 
 were marked off from Italian by no broad distinction of race 
 and colour, such as that which marked off the ancient • 
 Egyptian from the Italian, or that which now divides the 
 Frenchman from the Algerian Arab. They were marked 
 off. further, by no ancient culture, such as that which had ■ 
 existed for centuries round the Aegean. It was possible, it* 
 was easy, to Romanize these western peoples. 
 
 1 Mitteis, Reichsrecht and Volksrecht, p. 147 ; Kubitschek, Festheft 
 Bormnn)i (Wiener Studien, xx. 2), pp. 340 foil. ; L. Hahn, Rom and 
 Romanismns im griechisch-rom. Osten (Leipzig, 190G). One reason for 
 the loss of Roman culture is indicated by inscriptions like C. iii. 6800 
 (from the interior of Asia Minor), on which a veteran of Legio xii 
 Fulminata commemorates a wife with the purely native name of Ba. 
 This legionary must have had some knowledge of the Latin language 
 and the Roman civilization : his children probably had none. 
 
l I THE ROMANIZATION OF THE EMPIRE 
 
 Even their geographical position helped somewhat in- 
 directly to further the process. Tacitus two or three times 
 observes that the western provinces of the Empire looked 
 out on DO other land to the westward and bordered on no 
 free nations. That is one half of a larger fact which influ- 
 enced the whole history of the Empire. Hound the west 
 lay the sea and the Sahara. In the east were wide lands and 
 powerful states and military dangers and political problems 
 and commercial opportunities. The Empire arose in the 
 west and in Italy, a land that. geographically speaking, 
 looks westward. Hut it was drawn surely, if slowly, to the 
 east. Throughout the first three centuries of our era. we 
 can trace an eastward drift — of troops., of officials, of govern- 
 ment machinery — till finally the capital itself is no longer 
 Home but Byzantium. All the while, in the undisturbed 
 security of the west, Romanization proceeded steadily. 
 
 The advance of this Romanization followed manifold lines. 
 Much was due to official encouragement bv statesmen who 
 cherished the ideal of assimilating the provinces or who recog- 
 nized more cynically that civilized men are easier ruled than 
 
 I — 
 savages. 1 More, perhaps, was spontaneous. ! The definite 
 
 and coherent culture of Rome took hold on uncivilized but 
 
 intelligent provincials and planted in them the wish to learn 
 
 its language and share its benefits. And this wish was all 
 
 the keener since Roman tolerance drove no one into uni- 
 
 • formity. The compulsion to accept another speech and 
 
 another nationality which has been laid at one time or 
 
 another on Slav or Magyar or Alsatian French in modern 
 
 Europe — always with unsuccess —was no part of Roman 
 
 policy. Home made her culture more attractive by not 
 
 thrusting it upon her subjects.__ 
 
 The most potent single factor in the Romanization was 
 
 the town. Italian civilization was itself based on city life : 
 
 1 Tacitus {Agr. 21) emphasises this: ut homines ditpeni <i( rudes, 
 fO(jue in bella faciiea, quieti <t otio per vo h t pt ai e t adsuetcerent, 
 Ituriari privatim adiuvare public* ut templa jura domot tsuttuereni. 
 . . . ldque apud imperitot humaniUu vocabatur, nun para vervitutis 
 ettet. 
 
THE ROMAXIZATIOX OF THE EMPIRE 15 
 
 it was natural that the Empire should diffuse that life, 
 especially in the provinces of western and central Europe 
 which had few towns or none before they came under Roman 
 rule. [The most common step in this direction, at least in 
 the early Empire, was the foundation of coloniae, munici- 
 palities on the Italian pattern, manned by time-expired 
 legionaries (men who were citizens of Rome and spoke 
 Latin x ), laid out on Roman town-plans, 2 decorated with 
 Roman street-names and in all essentials Roman cities. 
 These coloniae were not meant altogether as missionaries 
 of culture. Primarily, they served as iii formal^f pjrtresges. 
 When Cicero 3 describes one of them, founded under the 
 Republic in southern Gaul, as ' a watchtower of the Roman 
 people and its outpost against the tribes of Gaul ', he states 
 an aspect of such a town which obtained during the first 
 century of the Empire no less than in the Republican age. 
 Nevertheless, they inevitably became centres of Roman life, 
 and though, being somewhat artificial military creations, 
 they were liable, as in the east (p. 13), to be gradually 
 merged in the peoples round them, 4 most of them escaped 
 this fate and really helped in Romanization. 
 
 Other towns were less direct official creations. Often, 
 native provincial markets or other centres of life grew so 
 far Romanized that they were held to merit the rights and 
 status of a Roman municipality, and the wisdom of the 
 Roman government in recognizing such progress was well 
 repaid by the development of fresh centres of Roman civili- 
 zation. Often, the legionary fortresses attracted traders, 
 women, veterans and others to settle outside their gates 
 but under the shelter of their ramparts, and their canabac, 
 
 1 Till about a.d. 70 most of the legionaries were Italians ; later, 
 they were recruited in the provinces but they regularly came from 
 towns which were adequately Romanized. Tiberius militem Graece 
 testimonium interrogation nisi Latine respondere vetuit (Suet. Tib. 71). 
 
 2 I may refer to my Ancient Town-planning (Oxford, 1913), eh. viii. 
 
 3 Pro Fonteio 13. So Tac. Ann. xii. 27 and 31, Agr. 14 and 32. 
 
 * Even Colonia Agrippinensis (Cologne) on the Rhine nearly fell 
 victim to this at one moment, Tac. Hist. iv. 65. 
 
16 THE ROMANIZATION OF THE EMPIRE 
 
 or ' bazaars '. to use an Anglo-Indian term, grew not seldom 
 into cities, worthy of municipal position. No doubt in all 
 these towns it was the municipal aristocracies which were 
 especially Roman. Like the German municipal elements in 
 mediaeval Cracow and elsewhere in eastern Europe, they 
 rested on a stratum which was less civilized. Yet we shall 
 see below that in many provincial Roman towns even the 
 lower stratum was Roman or Romanized. 
 
 Towns wiic not tin only factors in the process. Provin- 
 cials who seemed ripe for it often received grants of the 
 franchise individually or in large bodies. The abler pro- 
 vincials who became Romanized found careers open to them 
 :it Rome. Everywhere was practical inducement for the 
 native to enlist in the Roman culture. Weight, too. must 
 be ascribed to the drift of Italians into the provinces. This 
 was not a population-making emigration, like the present- 
 day mass emigrations of the Italian lower classes. It was 
 rather a drift of men from the well-to-do middle classes, 
 merchants and others, who formed little Roman centres 
 where neither troops nor Roman municipalities existed. 1 
 It was just such an emigration as that by which mediaeval 
 Germans helped to civilize parts of Galicia and Hungary 
 and to diffuse some sort of town-life through them. 2 If it did 
 not Romanize on the lines along which we have Anglicized 
 Australia, it was still a strong culture-making force. It 
 added its aid to the spread of town-life and to the willingness 
 of the provincial to carry Romanization through. 
 
 The process is hard to follow chronologically, since datable 
 evidence is scanty. In general, however, the instances <,| 
 really native fashions or speech which arc recorded from this 
 or that province belong to the early Empire. To that ag 
 
 ' Schulten, <le eonventibus civium romanorum : Kornemann, dedvibui 
 
 i<>, a. in provinciia imperii eonsittenHbus. For an example take an 
 inscription from Bourgea in Aquitania, /n<> salute Caesarum ef />. /?., 
 
 Minenne rl diVM Drusillue siniuin in />, r/n tuinn. ('. AgUetUS PritMU 
 ;i. :ir Aug. t e(urnliir) c(ivium) r(oinanonun). dating from \.i». 88—40 
 (C. xiii. 1. 1194). 
 : H. F. Kaindl, QeschichU <I< r Deutechen in den Karpathentandern, 
 
THE ROMANIZATION OF THE EMPIRE 17 
 
 we can assign the Celtic, Iberian, and Punic inscriptions 
 which occur occasionally in Gaul, Spain, and Africa, the 
 strange sculptures of three-headed or horned or cross-legged 
 deities in northern Gaul, the use of native titles like Ver- 
 gobret or Suffete, and the retention of native personal names 
 and of that class of Latin nomina, like Lovcssius, which are 
 formed out of native names. 
 
 In the second and third centuries of the Empire there is 
 a change. Roman elements now dominate ; in most regions 
 native survivals are few. But they are not merely few ; 
 they no longer stand out in hostile contrast with the Roman 
 civilization. The two have harmonized ; amalgamation has 
 gone forward. Indeed, one province can show from this age 
 a few native items unknown in earlier days of Roman rule. 
 In Gaul, or rather in some districts of that large area, the 
 Celtic measure of distance, the ' leuga ' of about 2,500 yards, 
 appears on official milestones in place of the Roman mille 
 passus} while the Druids, banned by the first Emperors, 
 emerge from their hiding, though in very humble fashion. 
 But these things arc plainly not due to anti-Roman or even 
 un-Roman feeling. The real position can be seen best in the 
 curious ' Gallic Empire ' of a. d. 258-73. 2 Here Roman elements 
 dominated, but they mixed in friendly fashion with native 
 things. The emperors of this state were called not only 
 Latinius Postumus, but also Piavonius and Esuvius Tetricus. 
 Its coins were inscribed not only ' Romae Aeternae ' and 
 ; Spei Publicae ', but also ' Herculi Deusoniensi ' and 
 ' Herculi Magusano '. It not only claimed independence 
 of Rome, but it modelled itself on Rome. It had its own 
 senate and consuls ; just as at Rome, tribunicia potestas was 
 conferred on its ruler, and the title princeps iuventutis on its 
 heir-apparent. We see Gaulish rulers with Gaulish names 
 appealing in some sort to native memories and at the same 
 
 1 The 'leuga' docs not occur in the first century ; in the second it 
 displaced the Roman measure in certain districts, and later it was used 
 still more freely. But it never got into Gallia Xarbonensis. 
 
 1 An admirable account of this has recently been given by M. Camille 
 Jullian in his ilisfoire rfe la Gaule, iv. ")70-9'2. 
 
 1751 B 
 
18 THE ROMANIZATION OF THE EMPIRE 
 
 moment fully accepting Roman fashions, speech and political 
 institutions. The native element in Gaul had not quite 
 died out of mind, but it had become little more than a pic- 
 turesque contrast to the Roman. If this Gallic Empire in 
 some details recalls the past, it still more looks forward. 
 Its independence, its cry of ' Gaul for the Gauls '. is a geo- 
 graphical fact, not a racial survival. It demanded indi- 
 vidual life, and doubtless also individual protection from 
 the barbarians, Tor a Roman section of the Roman world. 
 It anticipated the birth of new nations. 
 
 Progress in Romanization was perhaps slowesl in lan- 
 guage,, especially in the remoter districts of the Empire. 
 In Roman Africa. Punic was in almost official use in towns 
 like Gigthis, in the Syrtis. as late as the second century, and 
 Punic-speaking clergy were needed in outlying villages even 
 in tin- fourth century. In Gaul, Celtic is stated 1 to have 
 been spoken at the same epoch among tin Treveri, who 
 lived round what is now Trier. Presumably the native idiom 
 lingered on in the vast woodlands of the Eifel and Hunsriick 
 and Ardennes and in the hills above the upper Mosel valley, 
 from which uncouth uplanders came down to sell forest- 
 produce in towns, where they must have looked as strange 
 as the Gorals of to-day in the streets of Cracow or LembeiLr. 
 On tlie borders of Gaul and Spain, in the shadowy valleys 
 of the Pyrenees. Basque must have survived throughout the 
 Roman age, as it has done ever since. On the high plateau 
 of Asia Minor, where Greek was the dominant tongue of 
 civilized folk, six or seven other dialects. Galatian, Phrygian. 
 Lycaonian, and the rest, lived on till a very late date, espe- 
 cially (as it seems) on (he wild and remote pastures of the 
 Imperial domain-lands.- Some of these are survivals, noted 
 
 1 Jerome, comment, in epiet. ml Galakis, ii. ."». I [e i^ tin only authority 
 and his accuracy baa been doubted. Hut other survivals can Ik 
 quoted from this region ; here foi instance, in the secluded region of 
 Birkenfeld, the < elto-Roman culture is said t<» have resisted Germaniza- 
 lion long alter the lowlands had succumbed. (Cumont, Belgian* 
 
 nnnniiis,!. p. <>.->. takes the same \ iew as I have given ah<»v. , 
 
 K. I lull. //, , ,„, s. \Iiii. •_> lit ;, | : W. M. K.unsax . f> Hen. .lain, theft*. 
 
THE ROMANIZATION OF THE EMPIRE 10 
 
 at the time as exceptional, and counting in the scales of 
 history for no more than the survival of Croatian in a few 
 villages of the Italian Abruzzi or of Wendish (Sorb) fifty miles 
 from Berlin. Others are more serious facts. But they do 
 not alter the main position. In most regions of the west the 
 Latin tongue obviously prevailed. It was, indeed, powerful 
 enough to lead the Christian Church to insist on its use, and 
 not, as in Syria and Egypt, to encourage native dialects. 1 
 
 In material culture the Romanization advanced quickly. 
 One uniform fashion spread from the Mediterranean through- 
 out central and western Europe, driving out native art and 
 substituting a conventionalized copy of Graeco-Roman or 
 Italian art, which is characterized alike by technical finish 
 and neatness, and by lack of originality and dependence on 
 imitation. The result was inevitable. The whole external 
 side of life was lived amidst Italian, or (as we may perhaps 
 call it) Roman-provincial, furniture and environment. Take 
 by way of example the development of the so-called ' Samian ' 
 ware. The original manufacture of this (so far as we arc 
 here concerned) was in Italy, chiefly at Arczzo. Early in 
 the first century south Gaulish potters began to copy and 
 compete with it ; before long, the products of the Arretine 
 kilns had vanished even from the Italian market. Western 
 Europe henceforward and even Italy were supplied with 
 their ' best china ' from provincial and mainly from Gaulish 
 sources. The character of the ware supplied is significant. 
 It was provincial, but it was in no sense unclassical. It 
 drew many of its details from other sources than Arczzo, but 
 it drew them all from Greece or Rome. Nothing either in the 
 manner or in the matter of its decoration recalls native Gaul. 
 Throughout, it is imitative and conventional, and, as often 
 
 viii (1905), 79-120, quoting, amongst other things, a neophrygian text 
 of a.d. 259 ; W. M. Calder, Hellenic Journal, xxxi. 161. 
 
 1 Mommsen (Rom. Gesch. v. 92) ascrihes the final extinction of Celtic 
 in northern Gaul to the influence of the Church. But the Church was 
 not itself averse to native dialects : its insistence on Latin in the west 
 may he due rather to the previous diffusion of that language. (T am 
 glad to sec that Cumont (p. 109) agrees with me.) 
 
 b2 
 
20 THE ROMANIZATION OF THE EMPIRE 
 
 happens in a conventional art, items are freely jumbled 
 together which do not tit into any coherent story or sequence ; 
 many Gaulish potters seem to have been mainly anxious to 
 leave no undecorated spaces on their bowls. At its best, it 
 is handsome enough, though its possibilities are limited 
 l>v its brutal monochrome. But it reveals unmistakably the 
 Roman character <>f the civilization to which it belongs. 
 
 This Romanization in material things means more than 
 is always recognized. Some scholars, in particular (perhaps) 
 philologists, write as if the external environment of daily 
 life, the furniture and decorations and architecture of our 
 houses, the buckles and brooches of our dress, bear no rela- 
 tion to our personal feelings, our political hatreds, our 
 national consciousness. That may be true to-day of Asiatic 
 or African who dons European clothes once or again for 
 profit or for pleasure. It was not true of the Roman pro- 
 vincial. When he adopted, and adopted permanently, the 
 use of things Roman, we may say of him, firstly, that he had 
 become civilized enough to realize their value, and further, 
 that he had ceased to bear any national hatred against them. 
 Such hatred must have existed here and there : Tacitus 
 hints that it existed for a little while in Britain. Rut it was 
 rare ; wc can argue from the spread of Roman material 
 civilization that provincial sentiment was growing Roman. 
 
 By what process the less material aspects of provincial 
 life became Roman is less clear, because it was necessarily 
 more subtle. We seem, however, to sec. at least in Western 
 Europe, the same harmonious amalgamation of dominant 
 Roman elements with native elements that have not been 
 Wholly absorbed. In the east, of course, town-life and local 
 government and land-tenure were mainly Hellenist ie : Ro- 
 manization here made lit tie way. But in the west t lure were 
 towns enough of Roman foundation and Roman character 
 
 (p. 15), with yet an intersprinkling of native developments. 
 
 In northern and western Gaul, for instance. Roman munici- 
 palities (strictly so called) were wanting- Nevertheless, 
 
 towns sprang up here, some through Roman official encour- 
 
THE ROMANIZATION OF THE EMPIRE 21 
 
 agement and some of spontaneous growth. These towns 
 were a cross between Roman and Gallic. They were the 
 v ehefs-lieux ' of native cantonal areas and their local govern- 
 ment was native. But the titles of their magistrates were 
 borrowed from the Roman municipal terminology and their 
 government was assimilated to the Roman municipal pattern ; 
 even their town-plans were in some cases ' chessboards ' of 
 the received Italian type. 1 We shall meet some such towns 
 in Britain. In other provinces, as in southern Spain, hardly 
 a trace occurs of anything outside the strict Roman system. 
 So again in the sphere of religion. The Roman Empire 
 was generally tolerant of not-Roman worships, save in the 
 cases of Druidism and Christianity. It was rewarded. In 
 the western provinces the natives welcomed the Graeco- 
 Italian pantheon, identified their own gods with one or 
 another of its members, or, in default of identification, con- 
 tinued t^ieir old cults under new Latin names such as deae 
 matres. j Religion is seldom logical or uniform,jand the pre- 
 cise value to be put on these identifications doubtless varied 
 with every case and perhaps with every worshipper. Some- 
 times we may think we can see the old gods living on behind 
 their Roman masks and indeed keeping their power into the 
 Middle Ages. More often, Roman and native coalesced, and 
 again the exact proportions of the mingling must have in- 
 finitely varied. Some of the native cults seem to have sur- 
 vived more vigorously in the consciousness of the worshippers 
 than the others ; the one thing in which they agree is that 
 the Roman and the native are not hostile. There was 
 nothing unnatural to the provincial in honouring a Mercury 
 who was decked out in wholly Roman attributes — wand 
 and winged cap and purse and the rest — but who was placed 
 beside a provincial companion whose attributes declare her 
 the Celtic goddess Rosmerta (p. 73). The French scholar 
 Boissier once wrote that the civilized world was never nearer 
 to a common creed than under the Empire. Had it been 
 realized, it would have been a very complex creed. 
 
 1 See my Ancient Town-planning, p. 1*20 and Fig. 29. 
 
 
 V 
 
22 THE ROMANIZATION OE THE EMPIRE 
 
 It remains true, of course-, that, till a language or a custom 
 is wholly dead and gone, it can always revive under due con- 
 ditions. The rustic poor of a country seldom affect the 
 trend of its history. Hut they have a curious persistent 
 force. Superstitions, sentiments, even language and the 
 consciousness of nationality, linger dormant among them, 
 till an upheaval comes, till buried seeds are thrown out on 
 the surface and forgotten plants blossom once more. The 
 world has seen many examples of such resurrection — not 
 least in modern Europe. The Roman Empire offers us singu- 
 larly few instances, but it would be untrue to say that there 
 were none. 
 
 Komanization was. then, a complex process with complex 
 issues. It deies not mean simply that all the subjects of 
 Rome became wholly and uniformly Roman. The world is 
 not so monotonous as that. In it two tendencies were 
 blended with ever-varying results. First. Romanization 
 extinguished the difference between Roman and provincial 
 through all parts of the' Empire' but the east, alike' in speech, 
 in material culture, in political feeling and religion. When 
 the provincials called themselves Roman or when we- call 
 them Roman, the epithet is correct. Secondly, the process 
 worked with different degrees of speed and success in different 
 lands. It did not everywhere and at once destroy all traces 
 of tribal or national sentiments or fashions. These remained, 
 at least for a while- and in certain regions, Dot in ae-tive' oppo- 
 sition, but in latent persistence, capable of resurrection under 
 proper conditions. In such a case the provincial had become 
 a Roman, but he- e-oulel still undergo an atavistic reversion 
 t<» the ways of his forefathers. 
 
CHAPTER II 
 
 Preliminary Remarks on Roman Britain 
 
 One western province seems to break the general rule. 
 In Britain, as it is described by many English writers, 
 Roman and Briton were as distinct as modern Englishman 
 and Indian, and ' the departure of the Romans ' in the early 
 fifth centurv left the natives almost as Celtic as their coming 
 had found them nearly four hundred years before. The 
 adoption of this view may be set down, I think, to various 
 reasons which have, in themselves, little to do with the 
 subject. The older archaeologists, familiar with the wars 
 narrated by Caesar and Tacitus, pictured the whole history 
 of the island as consisting of such struggles. Later writers 
 have been influenced by the analogies of English rule in 
 India. Still more recently, the revival of Welsh national 
 sentiment has inspired a hope, which has become a belief, 
 that the Roman conquest was an episode, after which an 
 unaltered Celticism resumed its interrupted supremacy. 
 
 These considerations have, plainly, little value as history, 
 and the view which is based on them seems to me in large 
 part mistaken. As I have pointed out, it is not the view 
 which is suggested by a consideration of the general character 
 of the western provinces. Nor do I think that it is the view 
 which best agrees with the evidence which we possess in 
 respect of Britain. In the following paragraphs I wish to 
 examine this evidence. I shall adopt an archaeological 
 rather than a legal or a philological standpoint. The legal 
 and philological arguments have often been put forward. But 
 the legal arguments are almost wholly a priori, and they 
 have led different scholars to very different conclusions. 
 Tlie philological arguments are no less beset with difficulties. 
 
 
24 ROMAN BRITAIN : PRELIMINARY REMARKS 
 
 Both the lads and their significance arc obscure, and the 
 inquiry into them has hitherto yielded little beyond con- 
 fident and yet contradictory assertions which arc incapable 
 
 of proof. The archaeological evidence , on the other hand, 
 is definite and consistent. It illuminates, not only the 
 material civilization, but also the language and to some 
 extent even the institutions of Roman Britain, and supplies, 
 though imperfectly, the facts which our legal and philo- 
 logical arguments do not yield. 
 
 I need not here insert a sketch of Roman Britain. But 
 I may call attention to three of Us features. In the first 
 plage, it is necessary to distinguish the two halves of the 
 province, the northm^jui d^ western u plands occupied only 
 by troops, and the eastern and southern lowlands which 
 contained nothing but civilian life (Fig. I). 1 The two are 
 marked off, not in law but in practical fact, almost as if 
 one had been domi and the other militiac. [Wc shall not find 
 much t race_^f_Jjoir^a nization in the uplands. There neither 
 towns existed nor villas. Northwards, no town or count rv- 
 house has been found beyond the neighbourhood of Aid- 
 borough (Isurium), some fifteen miles north-west of York. 
 Westwards, on the Welsh frontier, the most advanced towns 
 were at Wroxeter (Viroconium), near Shrewsbury, and at 
 Caerwent (Venta Silurum), near Chepstow, and the furthest 
 country-houses two isolated dwellings at Llantwit Major, 
 in Glamorgan, and Llanfrynach. near Brecon. 2 In the south- 
 west the last country-house was near Lyme Regis, the last 
 town at Exeter. 8 These are the limits of the fully Romanized 
 
 For details Bee the Victoria County Histories <>i Northamptonshire, 
 i. 159, and Derbyshire, i. 191 . I may sa\ here thai much of the evidence 
 for the following paragraphs is to be round in my articles on Romano- 
 British remains printed in various volumes <•! this History. I am 
 indebted to its publishers for leave to reproduce several illustrations 
 from its pages. For others I refer my readers to the History itself. 
 
 See my Military Aspects of Roman Wales, notes so ami S2. There 
 was apparently some Borl of town life at Carmarthen. 
 
 West of Bxeter Roman remains are few and mostly later than 
 
 \.i>. 850. No town or country-house or Earm oi stretch of roadway 
 has ever been found here. The list of discoveries ^wsts of: one 
 
 
A 
 
 
 B 
 
 
 Birltnj 
 
 *°' 
 
 Fio. 1. (A) The Civil, (H) Tin: .Military Districts of Britain. 
 
 
 i 
 
26 ROMAN BRITAIN: PRELIMINARY REMARKS 
 
 area. Outside of them, the population cannot have acquired 
 much Roman character. Within these limits weir towns 
 and villages and country-houses and farms, a large popula- 
 tion, and a developed and orderly lit* . 
 
 This sharp division between the military and civilian areas 
 suggests that the garrison of Britain — the three legions at 
 York. Chester, and Caerleon, and the ' auxiliaries " scattered 
 in COSteUa, perhaps 80,000 or 35. 000 men in all— had little 
 influence <>n the civilization of Britain. At York, indeed, 
 a town grew up outside the fortress (p. 57). But neither 
 York nor Caerleon seem to have much affected the two 
 country-towns near them, at Aldborou^h and Cacrwent; 
 few other traces of civilization occur near either fortress, 
 and Chester lay wholly beyond the pale. Possibly, as 
 M. Cumont has observed, 1 the provisioning of the troops 
 brought landowners and farmers into contact with tin 
 Roman system. But in general Britain must have, in this 
 respect, differed much from northern Gaul and the- Rhine 
 frontier. There- six legions and their * auxiliaries r watched 
 150 miles of frontier during the- earlier Empire, and their 
 influence on the Romanization of the- border is very plain. 
 
 Second ly, the distribution of civilian life, e\ en within these- 
 limits, was singularly uneven. It is not merely that some 
 districts were the- special lu>nu s of wealthier residents. We 
 have also to conceive of some- parts as densely peopled and 
 of some as hardly inhabited. Portions of Kent. Sussex, 
 Essex, and Somerset are- set thie-k with ruins of country- 
 houses and similar vestiges of Romano-British life-. Other 
 
 early settlement on Plymouth harbour ; another near Bodmin, ol small 
 size-, dating from tlu- later first century ; a third, equally small and of 
 uncertain date, on Padstow harbour ; some- scanty vestiges e>r tin- 
 mining, principally late' ; two milestones (if milestones they be) <»f th< 
 early fourth century, at Tintagel church and at St. Hilary ; ami some 
 scattered hoards and isolated hits. Portions of the country were 
 plainly inhabited, but the inhabitants did not learn Roman ways. Iik<- 
 those who lived east of the Exe, Even tin-mining was not pursued 
 very actively till a comparatively late period, though the Bodmin 
 settlement may he- connected with tin-works dose by. 
 1 Journal of Roman Studies, h. 118. 
 
ROMAN BRITAIN : PRELIMINARY REMARKS 27 
 
 portions of the same counties, southern Kent, northern 
 Sussex, south-eastern Essex, western Somerset, show few 
 traces of any settled life. The midland plain, and in par- 
 ticular Warwickshire, 1 seems to have been the largest of 
 these ' thin spots '. Here, among great woodlands and on 
 damp and chilly clay, there dwelt not merely few civilized 
 Roman-Britons, but few occupants at all. 
 
 Lastly, Romano-British life was on a small scale. It was, 
 I think, normal in quality and indeed not very dissimilar 
 from that of many parts of Gaul. But it was, in any case, 
 defective in quantity. \ We find towns in Britain, as else- 
 where, and farms and country-houses. But the towns are 
 small and somewhat few, and the country-houses indicate 
 comfort more often than wealth. So, too, the costlier 
 objects of ordinary use, line mosaics, precious glass, gold 
 and silver ornaments, occur comparatively seldom, 2 and 
 such as do occur, seem to be almost wholly imports. The 
 great ' Lanx ', for instance, which was picked up on the bank 
 of the Tyne near Corbridge, is not only the one eminently 
 important piece of Roman silver found in the province ; it 
 is also in all likelihood a product of the eastern Empire. 3 
 jlii Roman Britain we have before us a civilization which, 
 like a man whose constitution is sound rather than strong, 
 might perish quickly from a violent shock. ! 
 
 A caution must be added. Geographically, Britain is an 
 island tied closer than is always realized to the continent of 
 Europe. The British lowlands are in the east and south ; 
 right over against them, across a narrow sea, are the low- 
 lands of the continent ; the rivers of island and continent 
 How out opposite each other ; it is easy from either shore 
 to reach the other coast and to pass up into the land behind 
 it. Li both pre-Roman and Roman times it was constantly 
 done. Therefore the same Celtic races dwelt on both sides 
 of the sea : there was frequent intercourse and the same or 
 
 1 Vict. Hist, of Warwickshire, i. 228. 
 
 - Sec my remarks in Traill's Social England (illustrated edition, 1901 ), 
 i. 141-61. 
 
 1 Journal of Ho man Studies, iv (1914), 1-12, with illustration. 
 
 ~i 
 
 \ 
 
 Y 
 
 
28 ROMAN BRITAIN: PRELIMINARY REMARKS 
 
 nearly the same civilization spread over northern Gaul and 
 Britain from the Rhine to the Atlantic. In the districts 
 of military life I his civilization was crossed by the beliefs 
 and customs and fashions of the soldiers ; elsewhere we deal 
 with a Romano-Celtic — originally Celtic — civilization which 
 requires to be studied more or less as a whole. It is useless 
 to examine Roman Britain or Roman Gaul or even much of 
 Roman Germany without constant reference to this whole, 
 and much good work attempted by modern French >r 
 German or English archaeologists has failed to \ ield its 
 proper fruit from neglect of this fact. 
 
CHAPTER III 
 
 ROMANIZATION IN LANGUAGE 
 
 We may now proceed to survey the actual remains. They 
 may seem scanty, but they deserve examination. 
 
 First, in respect of language. Even before the Claudian 
 conquest of a. d. 43, British princes had begun to inscribe 
 their coins with Latin words. These legends are not merely 
 blind and unintelligent copies, like the imitations of Roman 
 legends on the early English sceattas. The word most often 
 used, rex, is strange to the Roman coinage, and must have 
 been employed with a real sense of its meaning. After 
 a.d. 43, Latin advanced rapidly. *Ko Celtic inscription has 
 been detected, I believe, on aity monument of the Roman 
 period in Britain, neither cut on stone nor scratched on tile 
 or potsherd, and this fact is the more noteworthy because 
 Celtic inscriptions are not unknown in Gaul (see p. 31). On 
 the other hand, Roman inscriptions occur freely in Britain. 
 They are less common than in many other provinces, and 
 they abound most in the northern military region. But 
 they appear also in towns and country-houses of the low- 
 lands, and some of the instances are significant. 
 
 The town site which we can best examine for our present 
 purpose is Calleva Atrebatum (Silchcster), ten miles south 
 of Reading, which has been completely excavated within 
 the circuit of its walls. It was a small town in a stoneless 
 country ; it can never have had many lapidary inscriptions, 
 and such as there were must have been eagerly sought by 
 later builders. Nevertheless, a few fairly perfect inscrip- 
 tions on stone and many fragments have been found here 
 and prove that the public language of the town was Latin. 1 
 
 1 For these and for the following graffiti see my accounts in the 
 Viet, llisl. of Hampshire, i. 275, 282, and Eph. Epigr. ix. 984-8 and 
 1292-4 ; for the Clement inns tile see also Archaeologia, lviii. 30. 
 
no ROMANIZATION IN LANGUAGE 
 
 The speech of ordinary conversation is equally well attested 
 by smaller inscribed objects, and the evidence is remarkable, 
 since it plainly refers to the lower class of Callevans. When 
 a weary brickmakcr scrawls satis (enough) witli his finger 
 on a tile, or some prouder spirit writes CLEMENTINVS FECIT 
 TVBVL(t*wi) (Clementinus made this box-tile); when a bit 
 of Samian is marked fvk (thief), presumably as a warning 
 from the servants of one house to those of the next, or a brick 
 shows the word PVELLAM, part of an amatory sentence 
 otherwise lost, or another brick gives a Roman date, the 
 1 sixth day before the Calends of October \ we mav be sure 
 
 
 ^N^t^f/7 mfo 
 
 
 Fig. •">. Graffito on a Tile found at Silchesteb (p. 80). Pertaam 
 per/Uhu Campester LuciKanus \ Campimus, conticuerc omnes. 
 
 that the lower classes of Calleva used Latin alike at their 
 work and in their more frivolous moments (Figs. 2, 8, 4). 
 
 When we find a tile scratched over with cursive lettering 
 possibly part of a writing Jcsson — which ends with a tag 
 from the Aeiieid, we recognize that not even Vergil was out 
 of place here (Fig. 5). 1 The examples are st> numerous ami 
 remarkable that they admit of no other interpolation. 2 
 
 1 Sir I-:. M. Thompson! (ira-U and Latin Palaeography (1804), p. i»i i . 
 first suggested this explanation : Wph. be. 1298. 
 
 - I have not. of course, quoted all. To call them -as did a kindly 
 
 Belgian critic ol this paper in it-> first published form ' mi norabre de 
 Faits trop pen considerable 1 is really t<> misstate the ease. 
 
/ 
 
 . 
 
 L 
 
 /*. 
 
 ^L: 
 
 Fig. 2. ... puellam. 
 
 \ 
 
 Jk 
 
 i 
 
 Fig. 3. .FeciY tubul(um) Clcnicutinus. 
 
 I 
 
 \ 
 
 V 
 
 / 
 
 / 
 
 30 
 
 Fig. 4. vi k(alendas) Octo\l>res. . . . 
 Figs. 2-4. Graffiti on Tiles from Silciikstkr. (See p. 30.) 
 
ROMANIZATION IN LANGUAGE 31 
 
 I have heard this conclusion doubted on the ground that 
 a bricklayer or domestic servant in a province of the Roman 
 Empire would not have known how to read and write. 
 The. doubt rests on a misconception of the Empire. It is, 
 indeed, akin to the surprise which tourists often exhibit 
 when confronted with Roman remains in an excavation or 
 a museum — a surprise that ' the Romans ' had boots, or 
 beds, or waterpipes, or fireplaces, or roofs over their heads. 
 There are, in truth, abundant evidences that the labouring 
 man in Roman days knew how to read and write at need, 
 and there is reason to believe that in the lands ruled by 
 Rome education was better under the Empire than at any 
 time since its fall till the nineteenth century. 
 
 It has, indeed, been suggested by doubters, that these 
 graffiti were written by immigrant Italians, working as 
 labourers or servants in Calleva. The suggestion does not 
 seem probable. Italians certainly emigrated to the provinces 
 in considerable numbers, just as Italians emigrate to-day. 
 Bui we have seen above (p. 16) that the emigrants of the 
 Imperial age were not labourers, as they are to-day. They 
 were traders, dealers in land, money-lenders, or other ' well- 
 to-do ' persons. The labourers and the servants of Calleva 
 must be sought among the native population, and the graffiti 
 testify that this population wrote Latin. 
 
 It is a further question whether, besides writing Latin, 
 the Callcvan servants and workmen may not also have spoken 
 Celtic. Here direct evidence fails. In the nature of tilings. 
 we cannot hope for proof of the negative proposition that 
 Celtic was not spoken in Silchester. But all probabilities 
 suggest that it was, at any rate, spoken very little. In the 
 twenty years' excavation of the town, no Celtic inscription 
 lias emerged. Instead, we have proof that its lower classes 
 wrote Latin for all sorts of purposes. Had they known 
 Celtic well, it is hardly credible that they should not have 
 sometimes written in that language 1 , as the Gauls did across 
 the Channel. In Gaul, potters of Roman date could scrawl 
 their names and records, Sacrillos avot, ' Sacrillus potter ", 
 
.32 lH)MA\TZATTO\ IN LANGUAGE 
 
 Valens avoti, ' Valens potter', on a mould. 1 No such scrawl 
 
 has ever been found in Silchcstcr or indeed in Britain. In 
 Gaul, men with Roman names, Martialis and the like, could 
 set up inscriptions couched wholly or almost wholly in Celtic. 
 No such inscriptions occur in Britain. The Gauls, again, 
 could invent a special letter H to denote a special Celtic 
 sound and keep it in Roman times. No such letter was 
 used in Roman Britain, though it appears on earlier 
 British coins. This total absence of written Celtic cannot 
 be a mere accident. 
 
 No other Romano-British town lias been excavated s<, 
 fully or so scientifically as Silchester. None, therefore, has 
 yielded so much evidence. But we have no reason to con- 
 sider Silchester exceptional. Such scraps as we possess 
 from other towns point to similar Romanization elsewhere. 
 Fvr, for instance, recurs on a potsherd from the Romano- 
 British country town at Dorchester in Dorset. London 
 has yielded a tile on which, before it was baked hard, some 
 one scratched in unconventional Latin the remark, 'Austalis 
 Efoes off on his own daily for a fortnight.' Austalis— that 
 is, Augustalis — was plainly a workman: so was his critic, 
 and their fellow-workmen could presumably read and appre- 
 ciate the criticism. 2 Leicester, too, supplies a tile scratched 
 Primus- fecit x, 'Primus has made ten tiles.* 
 
 The rural country-houses and farms, mostly ill-explored 
 and ill-recorded, furnish much scantier evidence than a care- 
 fully excavated town. Yet they are not without their 
 Roman inscriptions cut on stone, for the most part dedica- 
 tions or tombstones, which prove that al hast the owners 
 or occupiers of the houses claimed to know Latin. Of the 
 more cogenl graffiti on tiles or potsherds, examples are rare. 
 
 1 Aval oi avoHs seems td be a Gaulish term for 'potter*. One 
 example, Saerittos avoi form., suggests a bilingual sentence such ;i> v, 
 And in some Cornish documents of the period when Cornish WAfl 
 definitely giving way to English. 
 
 * Austalis dibits (i.e. (lit bus) iiii vagcUur silni) cotuHm. See my 
 
 notes in /•;/>//. Epigr. vii. 1111 and Journal of Human Studies, i. 168, 
 |»late xxvi. 
 
Fig. 7. [nscribed Tile prom Plaxtol, Kent. 
 
 The to]> <>f the tile shows traces <>t' tlio line parietalem and most of the line Cabriabanu 
 
 the Lower pari shows the former in lull ami the traces <»!' the latter. Tin' third liiil 
 
 (. . . icavit) has here, as on all tlu* fragmentSj tailed to come out clearly, (p. 33.) 
 
 Fio. 55. Fragment <>i [nscribed Jar from [gkletoNj Cambs. (p. 33.) 
 
ROMAXIZATION IN LANGUAGE 
 
 33 
 
 But the man who made the tiles for a house at Plaxtol in 
 Kent thought it worth while to cover them with Roman 
 lettering ; apparently he incised the legend in three lines on 
 a wooden cylinder and rolled it over the tiles while soft, 
 thus producing a recurrent inscription, ' Cabriabanus (or 
 Cabriabantus) made this wall-tile ' which served as a sort 
 
 QfrfW 
 
 i i i ■ 1 i ' 1 3 i i ■ •*! ■ i ■ "i ■ i ■ J ' ■ i ■ 'l 
 
 Fig. G. Reconstruction of the Plaxtol Inscription from 
 
 Various Fragments. 
 
 The legend is, line 1 PARIETALEM, 2 CABRIABANVs (or 
 XTVs), 3 . . . ICAVIT ; line 1 is topsy-turvy to the rest. In the above, 
 CAB in 2 and IT in 3 are repeated twice, to show the recurrence of the 
 lettering. 
 
 of decoration (Figs. 6 and 7). 1 Again, two pieces of a 
 blackish urn found long ago in the Roman farm at Ickleton, 
 in south-east Cambridgeshire, bear a graffito which may be 
 completed ex ha]c amici bibun[t, i from this jar friends drink ' 
 (Fig. 8). 2 Yet once more, a Roman site near Easton Grey, 
 in north Wiltshire, has yielded a little bas-relief carved 
 (as it seems) in local stone with the figures of a goddess and 
 three worshippers ; the mason has roughly signed it, Civilis 
 fecit, i Civilis made me.' 
 
 1 Proc. Soc. Antiq* LoncL xxiii. 108 and Eph. ix. 1290. 
 
 ■ C. vii. 1335. 7. Now at Audley End, where I have seen it. 
 Too little remains of the jar to fix its date ; it does not suggest the 
 later Empire. 
 
 1751 C 
 
34 ROMANIZATION IN LANGUAGE 
 
 The general result is clear. Latin was employed freely in 
 the towns of Britain, not only on serious occasions or by the 
 upper classes, but by servants and workpeople for the most 
 
 accidental purposes. It was also used, at least by the upper 
 classes, in the country. Plainly there did not exist in the 
 towns that linguistic gulf between upper class and lower 
 class which can be seen to-day in many cities of eastern 
 Europe, where the employers speak one language and the 
 employed another. On the other hand, it is possible that 
 a different division existed, one which is perhaps in general 
 rarer, but which can, or could, be paralleled in some Slavonic 
 districts of Austria-Hungary. That is. the townsfolk of all 
 ranks and the upper class in the country may have spoken 
 Latin, while the peasantry may have used Celtic. No actual 
 evidence has been discovered to prove this. It is not, how- 
 ever, in itself an improbable linguistic division of Roman 
 Britain, even though the province did not. contain any 
 such racial differences as those of German, Pole, Ruthene 
 and Rouman which lend so much interest to towns like 
 Czcrnowitz. 
 
 It remains to cite the literary evidence, distinct if not 
 abundant, as to the use of Latin in Britain. Agricola. as 
 is well known, encouraged it, with the result (says Tacitus) 
 that the Britons, who had hitherto hated and refused the 
 foreign tongue, became eager to speak it fluently. About 
 the same time, as Plutarch mentions in his tract on the 
 cessation of oracles, one Demetrius of Tarsus, a 'grammarian', 
 was teaching in Britain (a. d. SO), and his t caching is recorded 
 as nothing out of the ordinary course. 1 Rather later, in 
 a.d. 90, Martial boasts that he was read in Britain, and 
 about a.d. 120 Juvenal alludes casually to British lawyers 
 taught by Gaulish schoolmasters. It is plain that by the 
 second century Latin must have been spreading widely in 
 the province. We need not feel puzzled about the way in 
 which the Callevan workman of perhaps the third or fourth 
 century learnt his Latin. 
 
 1 Sec Eph. Epigr. i\. 560 and Dessau, Herme*, xlvi. 156. 
 
ROMANIZATION IN LANGUAGE 35 
 
 At this point we might wish to introduce the arguments 
 deducible from philology. We might ask whether the 
 phonetics or the vocabulary of the later Celtic and the 
 English languages reveal any traces of the influence of Latin, 
 as a spoken tongue, or give negative testimony to its absence. 
 Unfortunately, the inquiry seems almost hopeless. The 
 facts are obscure and open to dispute, and the conclusions 
 to be drawn from them are quite uncertain. Dogmatic 
 assertions are common. (^Trustworthy results are corre- 
 spondingly scarce.) One instance may be cited in illustra- 
 tion. It has been argued that the name ' Kent ' is derived 
 from the Celtic ' Cantion ', and not from the Latin ' Cantium ', 
 because, according to the rules of Vulgar Latin, ' Cantium ' 
 would have been pronounced ' Cantsium ' in the fifth cen- 
 tury, when the Saxons may be supposed to have learnt the 
 name. That is, Celtic was spoken in Kent about 450. Yet 
 it is doubtful whether Latin ' ti ' had really come to be 
 pronounced ' tsi ' in Britain so early as a.d. 450. And it 
 is plainly possible that the Saxons may have learnt the name 
 long years before the reputed date of Hengist and Horsa. The 
 Kentish coast was armed against them and the organization 
 of the 'Saxon Shore' established as early as about a.d. 300. 
 Their knowledge of the place-name may be at least as old. 
 No other difficulty seems to hinder the derivation of 
 4 Kent ' from the form ' Cantium ', and the argument 
 based on the name thus collapses. It would be impossible 
 here to go through the list of cases which have been 
 supposed to be parallel in their origin to ' Kent ', nor should 
 I, with a scanty knowledge of the subject, be justified in 
 such an attempt. I have selected this example because it 
 lias lately been emphasized by an eminent writer. 1 
 
 1 Vino«rra<lofT, Growth of the Manor, p. 102. I am indebted to 
 .Mr. \V. H. Stevenson for help in relation to these philological points. 
 
 C2 
 
CHAPTER IV 
 Romanization ix Material Civilization 
 
 From language we pass to material civilization. Here is 
 
 a far wider field of evidence, provided by buildings, private 
 or public, their equipment and furniture, and the arts and 
 small artistic or decorative objects. On the whole this 
 evidence is clear and consistent. The material civilization 
 of the province, the external fabric of its life, was Roman, 
 in Britain as elsewhere in the west. Native elements 
 succumbed to the conquering foreign influence. 
 
 I. In regard to public buildings this is natural enough. 
 Before the Claudian conquest the Britons can hardly have 
 possessed large structures in stone, and the provision of 
 them necessarily came with the Romans. The fora, basilicas, 
 and public baths of the towns, such as have been discovered 
 at Silchcster, Caerwent and elsewhere, follow Roman models 
 and resemble similar buildings in other provinces. The 
 streets of the towns seem also to have been laid out on the 
 'chessboard' system of town-planning proper to Roman 
 municipalities of the Empire ; to this point I shall return 
 in a later chapter (p. 6-1). The temples, however, botli in 
 town and country, show as a rule something more of a local 
 pattern. They consist generally of a small square or nearly 
 square cella or shrine, with a roofed portico or colonnade 
 running round all its four sides, and an entrance usually 
 from the east (Fig. 9); the building often stands in a large 
 open irregular enclosure. This type of temple occurs at 
 Silchester and Caerwent and on many rural sites; it occurs 
 also in northern Gaul and as far east as the Rhine. It 
 differs from the ordinary classical type, and is taken by good 
 
ROMANIZATION IN MATERIAL CIVILIZATION .37 
 
 authorities to be of Celtic origin ; it may, however, be a varia- 
 tion from the classical type or even an amalgamation of 
 classical and native. 1 
 
 CA6 RWENT 
 
 SILCH G STER 
 
 ENTRANCE 
 
 ^CALE oMrflliilllT 
 
 to 
 
 xo 
 
 JL. 
 
 30 
 
 40 
 
 SO r- 
 
 _JFee.t 
 
 Fig. 9. Two Ground-plans of Romano-British Temples. 
 
 II. The private houses, by which I mean those built for 
 civilized occupants, present more complicated features. Like 
 dwelling-houses all the world over, they exhibit many 
 varieties. 2 But we can distinguish two main types, called 
 by English writers the Corridor and the Courtyard types. 
 In the corridor house the front was formed by a narrow hall 
 or corridor, which usually terminated at one or both ends 
 in a largish room projecting slightly in wing-fashion. Houses 
 of this class were common in Roman Britain. Many were 
 small and poor ; in the Frilford farmstcading, shown in 
 Fig. 15, the wing-room must have been almost the only 
 
 1 For Gaulish instances of these temples, see Leon de Vesly, Les 
 Fana de la region Normande (Rouen, 1909) ; for Germany, Bonner 
 Jahrbi/cher, 187G, p. 57, Hettner, Drei Tempelbezirke im Trevirerlande 
 (Trier, 1901) and Trierer Jahresberichle, iii. 49-06 ; they occur as Ear 
 south as the Auvergne. The English writers who have published 
 accounts of these structures have tended to ignore their special char- 
 acter. The temple unearthed at Wroxetcr in 1913 seems to have 
 belonged to the classical type, like that at Lydney. 
 
 2 In the Victoria History (Hants, Xorihanls, Shropshire, Somerset) 
 I have given some twenty-five plans, which make up a fairly repre- 
 sentative series. 
 

 \ 
 
 ENTRANCE 
 
 CerUan 
 
 Inferred 
 
 Foundation 
 
 courses 
 
 Fig. 10a. Corridor House at Brisiinoton, neab Bristol 
 
 (sec p. 39) (for scale sec Fi<j. 10b). 
 
 I 
 
 ^ 
 
 CLANVILLE 
 
 S^S^ 
 
 Fig. iou. House at Clanville, neab Andoveb, Hants. 
 
 See p. 89.) 
 
 [Hyp. hypocaust; Tess. plain tessellated floor.] 
 
II 
 
 ROMANIZATION IN MATERIAL CIVILIZATION 39 
 
 comfortable apartment, and much of the house may have 
 served farm purposes. Others were more luxurious ; cor- 
 ridor houses at Brading, and at Brislington, near Bristol 
 I (Fig. 10a), seem to have been the homes of wealthy owners. 
 In the second or Courtyard type, the rooms were ranged 
 along corridors which enclosed three — less commonly two 
 or four — sides of a spacious squarish yard (Fig. 11). Such 
 houses were naturally extensive, and many were clearly the 
 residences of rich men. In one form or another they are 
 not much less common than the corridor houses. But the 
 two types run into one another, and it is sometimes hard to 
 decide whether a house consisting of a centre and two short 
 wings should be called a corridor house with wing-rooms 
 overgrown or a courtyard house with stunted flanks. 
 
 A third and far rarer type shows a narrow oblong building, 
 generally furnished with living-rooms at each end, while 
 a double row of columns runs down its central portion. 
 Its ground-plan strangely resembles that of a great columned 
 barn, but it is possible that the* middle space between the 
 columns was really open to the sky and that the columns 
 supported the roofs of sheds or colonnades. Some houses 
 of this type possessed good mosaics and comfortable fittings ; 
 more often they were subsidiary to better houses of the 
 courtyard or corridor type, standing close by them and pro- 
 viding perhaps quarters for servants and the like. Fig. 10b 
 shows one of these houses in which parts of the original 
 sheds or colonnades have been built up into rooms. 1 
 
 Corridor and courtyard houses occur freely both in town 
 and in country ; the third type has been as yet detected 
 only in the country. It is noteworthy that no special type 
 of town-house occurs. Apart from a few shops — simple 
 structures with shop in front and living-rooms or stores 
 behind — the dwellings of Silehester (Fig. 12) and Caerwent 
 are much the same as those of the countryside, and what is 
 known of other towns, of Wroxetcr or Aldborough, tells the 
 same tale. Excavation may some day show us town-houses 
 1 Vict. Hist. Hants, i. 302, 316; Archaeol. Journ. Ixvi. X5. 
 
40 ROMANIZATION IX MATERIAL CIVILIZATION 
 
 somewhere, but we have enough evidence already to conclude 
 that the distinction between town-houses and country-houses 
 was substantially unrepresented in Roman Britain. Here, 
 however, we touch on a feature of Romano-British town- 
 life which belongs rather to Chapter VI. 
 
 Britain is not peculiar in its two main types of houses. 1 
 Like the temples described above (p. 36), courtyard and 
 corridor houses recur in very similar forms in northern Gaul. 
 From the seaeoast to the Rhine they are indeed the dominant 
 types of houses. 2 At present they are attested only as 
 country-houses, but that is perhaps because no complete 
 town-house has yet been uncovered in any Roman town 
 of this region. The general likeness of Roman Britain to 
 northern Gaul suggests that Amiens, Reims, Metz, 8 diel not 
 in this respect differ very greatly from Silchester. 
 
 The origin of these two northern types has been much dis- 
 cussed. English writers tend to think them Celtic, since thev 
 occur in Celtic lands ; they also sec in the corridor an element 
 common to both types, and suggest that the courtyard type 
 grew out of the corridor type by gradually pushing forward 
 its wing-rooms and continuing the corridor in front of them. 
 Foreign writers more often derive them from types of houses 
 used in Italy and the Greek east. Probably the material 
 does not yet exist for a full settlement of the problem ; for 
 one thing, we know too little of the rural dwellings of Italy, 
 large or small. It is clear, however, that the Italian houses 
 most familiar to us, the town-houses of Rome- and Pompeii, 
 bear no likene-ss to tin- northern hems. s. Their central 
 feature is an atrium, and there is not an atrium te> be found 
 in any house in Roman Britain. 
 
 1 The- type- of Fig. ion seems purely British. 
 
 ■ Some plana of north Gaulish and German country-houses and 
 farms are- given l>y de Caumont, Abtccduire (ed. 2, 1870), pp. •'*?{) io\\., 
 and Kropatschek, VI. Berichi tier r6m.-germ. Kommission, inio-ii, 
 pp. .">?-?.*$. For others mt tin- Annate* of the Namur Archaeological 
 Society and similar journals. 
 
 Nor perhaps even Trier : a hull-explored town-house at Trier is not 
 at all Pompeian (Bonner Jakrb. e-iii. 2'M). 
 
Fig. 11. Courtyard House at Nortiileigh, Oxfordshire, as 
 excavated in 1815-16. Room 1, chief mosaic with hypocaust ; 
 rooms 8-18, mosaic floors ; rooms 21-7 and 38-43, baths, &c. ; the 
 west wing had poorer rooms, perhaps for servants. Recent excava- 
 tions show that this plan represents the house in its third and latest 
 stage ; in the corridor (10) a part of the earlier house-front is shown 
 by dotted lines. The pottery found in the recent excavations suggests 
 that the first house on this spot was built not later than the early 
 second century. 
 
L2 ROMANIZATION IX MATERIAL CIVILIZATION 
 
 Probably the courtyard house has more connexion with 
 the south than the corridor house. The town-houses of the 
 Greek cast and the kindred houses of Timgad in Africa, of 
 Pola and Doclca in Adriatic lands — houses that arc built 
 round a small columned court or peristyle — offer faint 
 parallels to our courtyard houses. Indeed, one or two 
 houses at Silchester and Caerwent actually have such small 
 courts. 1 More definite parallels, again to the courtyard type-, 
 can be found in other houses, mostly country-houses, of 
 the same Greek type, which wen- built round large peristyles 
 comparable in size to the spacious British and Gaulish 
 yards. 9 Perhaps we may conclude that our courtyard house 
 owed much of its development to this originally Greek 
 type. And if the peristyle house excavated in 1882 at 
 Bibracte (Mont Beuvray), in mid-France, be of pre-Roman 
 date, as Dcchelctte thought, we may further guess that 
 (his type was spreading northwards as early as the age of 
 Caesar. 3 
 
 Hut the corridor house remains unfathered. To it Mediter- 
 ranean lands offer no analogies. It had neither atrium nor 
 peristyle, and the attempts of sonic scholars to dctei t pictures 
 of it on two African mosaics are not convincing. 4 The most 
 southern corridor house which I can quote was dug up years 
 
 1 Silchester, insula xiv. l (Archaeologia, Iv. 221) ; Caerwent, bouse :5 
 (Arch, lvii, plate 40). A tew Pompeian houses have no atrium and 
 belong to this type ; for instance, ins. v. ~> and vi. 15. Similarly, parallels 
 may be drawn between certain l'ompeian wall-paintings of houses 
 and certain large houses in Germany, as at Nennig, Rouhling, Wittlich 
 (see Etostowzew, Archaol. Jahrbuch, 1904, p. 103). Bui such houses 
 are rare in Germany and unknown in Britain. 
 
 2 For instance, the large house of Fannius Sinister near Pompeii : 
 a large house mar Pola (Schwalb, Rcmische Villa bei Pola, Wien, 
 lito'j), an oil-farm on the same coast (Gnirs, Jahrbuch fur AUeriurns- 
 kunde, ii. 184) : a large house at Saint-Leu in Algeria (Revue africaine, 
 
 1804, p. 280), and the luxurious house in the town of I'thina (Oudna, 
 in Tunis, see Fondation Piof, hi. 177). 
 
 Bulliot, Fouilles (lr Mont Beuvray \ Dechelette, Manuel, ii. !».*>."5. 
 
 1 KropatSChek (see p. 40, note) assumes that the corridor house was 
 common in Italy. But that is pure assumption ; certainly the BOSCO 
 
 Reale farm is quite different. His arguments suiter also from his 
 genera] neglect of all finds outside Germany. 
 
50 
 
 • • ' ■ ■ I — 
 
 Scale of Fect 
 zoo 
 
 ;<->? 
 
 Fig. 12. Part or Silchester. Showing some private houses and 
 shops, the Forum, and the Christian Church. (From the phut by 
 Sir W. Hope, issued hi/ the Society of Antiquaries.) 
 
44 ROMANIZATION IX MATERIAL CIVILIZATION 
 
 ago near Pan. 1 Perhaps, after all, we may credit it with 
 a Celtic origin. Thai is the conclusion for which we should 
 look on genera] grounds — that the larger and richer houses 
 copied foreign patterns, while the smaller oiks, like the 
 Indian bungalow, tended to follow native lines. Here, as 
 elsewhere, the Romanization of Britain combined native 
 and Roman elements. 
 
 The internal fittings of these houses show the Roman 
 supremacy more definitely. These lit tings are wholly bor- 
 rowed from Italian sources. If we cannot find in the 
 Romano-British house either atrium or impluvium, iablinum 
 or peristyle, such as we find in Italy, we have none the 1 ss 
 tlie painted wall-plaster (Fig. 13) and mosaic floors, the 
 hypocausts and bathrooms of Italy. The wall-paintings 
 and mosaics may be poorer in Britain, the hypocausts more 
 numerous ; the things themselves are those of the south. 
 No mosaic, I believe, has come to light in the whole of 
 Roman Britain which represents any local subject or contains 
 any uiielassieal feature. The usual ornamentation consists 
 either of mythological scenes, such as Orpheus charming the 
 animals, 2 or Apollo chasing Daphne, or Actaeon rent by his 
 hounds, or of geometrical devices like the so-called Asiatic 
 shields which are of classical origin. 3 Perhaps we may 
 detect in Britain a special fondness for the cable or guilloche 
 pattern, and we may conjecture that from Romano-British 
 
 1 Archaeol. Journ. xxxvi. 17. 
 
 2 There is no reason to think the numerous Orpheus mosaics Christian. 
 Christianity was not so ubiquitous as that. The scene, I imagine, was 
 popular because it included various quaint animals. 
 
 3 It has been BUggested that these mosaics were laid by itinerant 
 Italians. The idea is, of course, due to modern analogies. It does 
 not seem impossible, since the work is in a sense that of an artist, and 
 the pay might have been high enough to attract good decorators from 
 
 tin- Continent. However, no evidence exists to prove this or even 
 
 to make it probable. The mosaics of Roman Britain, with hardly an 
 exception, are such as might easily be made in a province which could 
 export skilled workmen to Gaul (p. 77). They have also the look of 
 work imitated from patterns rather than of designs sketched by artists. 
 It is most natural to suppose that, like the Gaulish Samian ware — 
 which is Imitative in much the same fashion — they are local products. 
 

 
 ' 
 
 Fig. 13. Painted Pattern on Wall-plaster from Silchester, 
 
 Showing a conventional style based on classical models (p. 44). 
 
 {Restoration by G. E. Fox, in Archaeologia.) 
 
 44 
 
ROMANIZATION IN MATERIAL CIVILIZATION 45 
 
 mosaics it passed in a modified form into later Celtic art. 
 But the ornament itself, whether in single border or in many- 
 stranded panels of plaitwork, occurs not rarely in Italy as 
 well as in thoroughly Romanized lands like southern Spain 
 and southern Gaul and Africa, and also in Greece and Asia 
 Minor. It is a classical, not a British pattern. 
 
 III. Turn now to the dwellings of the peasant poor. These 
 we know only in one corner of southern England, but within 
 this limit we know them well. On the chalk downs of Wilts 
 and north-east Dorset, Colt Hoare was busy a century ago, 
 and in 1884-90 Pitt-Rivers dug three villages wholly up — 
 at Woodcutts, Rotherlcy and Woodyates, a dozen miles 
 south-west of Salisbury — and later workers have continued 
 the search. 1 In plan these villages are not Roman ; their 
 round mud-huts and pits, their strange ditches, their shape- 
 less enclosures, date from days before or early in the Roman 
 occupation. But Roman civilization soon reached and 
 absorbed them. The ditches were filled up ; hypocausts, 
 odd but unmistakable, wall-plaster painted in Roman 
 fashion, roofing of Roman tiles, came into use ; the villagers 
 learnt to eat and drink from Samiamdishes and cups of glass, 
 and even to keep their clothes in wooden chests of drawers ; 
 some of them could read and write. 2 Meanwhile, they 
 utterly forgot their Celtic fashions ; there is no sign of the 
 Late Celtic art in any of Pitt-Rivers's multitudinous illustra- 
 tions. To these men the Roman objects which they used 
 were the ordinary environment of life ; they were no ' delicate 
 exotic varnish ', as one eminent writer has called them. 3 
 Indeed, I cannot find in our Romano-British remains the 
 
 1 R. Colt Hoare, Ancient Wiltshire (1812-21); A. Pitt -Rivers, Excava- 
 tions in Cranborne Chase, &c. (four large quartos, privately printed, 
 1887-98) ; M. E. Cunnington, Wilts Archaeol. Magazine, xxxvii. 42, 
 xxxviii. 53 ; Hey wood Sumner, Excav. on liockbourne Down (London, 
 1914). 
 
 2 Pitt-Rivers, iii. 3-G. So Colt Hoare, Ancient Wilts, Roman Aera, 
 p. 127 : ' On some of the highest of our downs I have found stuccoed 
 and painted walls, as well as hypocausts, introduced into the rude 
 settlements of the Britons.' 
 
 3 Vinogradoff, Growth of the Manor, p. 39. 
 
46 ROMANIZATION IN MATERIAL CIVILIZATION 
 
 contrast alleged by lliis writer 'between an exotic culture 
 of a higher order and a vernacular culture of a primitive 
 kind '. There were in Britain splendid houses and poor 
 oiks. 1 5 1 1 1 a continuous gradation of all sorts of buildings 
 
 and all degrees of comfort connected them ; there is no 
 discernible breach in tin scale. Throughout, the dominant 
 (lenient is the Roman provincial fashion which is borrowed 
 
 from Italy. 
 
 io.ro 'O 1033 +o so & to 4© qo too 
 t-l_l 1 1 — I 1 « I | I \ * 
 
 SCALC 
 
 Fecr 
 
 Fig. 14. Native Village at Din Lligwy, Anglesea. 
 
 We find Roman influence even in the most secluded 
 villages of tin- upland region. At Din Lligwy. on the north- 
 easl e<»ast of Anglesea, excavation (Fig. l has uncovered 
 the ruins of a village enclosure about three-quarters of an 
 acre in extent, containing round and square huts or rooms, 
 with walls of roughly coursed masonry and roofs of tile. 
 Scattered up and down in it lay hundreds of fragments <,| 
 Samian and other Roman or Romano-British pottery and 
 
ROMANIZATION IN MATERIAL CIVILIZATION 47 
 
 a far smaller quantity of ruder pieces, a few bits of Roman 
 glass, some Roman coins of the period a.d. 250-350, various 
 iron nails and hooks, querns, bones, and so forth. 1 The 
 place lies on the extreme edge of the British province and 
 on an island where no signs of proper Roman occupation can 
 be detected, while its ground-plan shows little mark of 
 Roman influence. Yet the smaller objects and perhaps also 
 the squareness of one or two rooms show that even here, 
 in the later days of the Empire, the products of Roman 
 civilization and the external fabric of Roman provincial life 
 were present and almost predominant. 
 
 1 E. Neil Baynes, Arch. Cambrcnsis, 1908, pp. 183-210. 
 
 FRILFORD 
 
 Fig.15. Plan oi Farmhouse at Frilford, Berks. (From plan bij 
 Sir A. J. Evans), See p. 39. The scale is the same as that of Figs. 10a 
 and b on p. 38. 
 
CHAPTER V 
 
 Romanizahon ix Art 
 
 Art shows a rather different picture. Here the definite 
 survivals of Celtic tradition are not perhaps more numerous 
 but are certainly more tangible. There flourished in Britain 
 before the Claudian conquest a vigorous native art, chiefly 
 working in metal and enamel, and characterized by its love 
 for spiral devices and its fantastic use of animal forms(Fig. 10). 
 This art — La Tene or Late Celtic or whatever it be styled — 
 was common to all the Celtic lands of Europe just before the 
 Christian era, and its vestiges are particularly clear in Britain. 
 When the Romans spread their dominion over the island, 
 it almost wholly vanished. For that we are not to blame 
 any evil influence of this particular Empire. All native 
 arts, however beautiful, tend to disappear before the more 
 even technique and the neater finish of town manufacture 
 The process is merely part of the honour which a coherent 
 civilization enjoys in the eyes of country folk. Disraeli 
 somewhere describes a Syrian lady preferring the polish of 
 a western boot to the jewels of an eastern slipper. W "it li 
 a similar preference the British Celt abandoned his national 
 art and adopted the Roman provincial fashion. 
 
 He did not abandon it wholly. Little local manufactures 
 of small objects witness to sporadic survivals. Such, among 
 pottery, are the New Forest stoneware with its curious leaf- 
 ornament (Fig. 17), which was used a good deal in southern 
 Britain, 1 and the better known and far more widely dis- 
 tributed Castor ware, made on the banks of the Xen some 
 
 five miles west of Peterborough. We may briefly examine 
 
 t his latter instance. 
 
 1 Victoria Hist. Hants, i. 826 ; Archacnl. .fount. \\\. :U«». 
 
 
HVO& 
 
 
 J 
 
 ; 
 
 Fig. 16. Late Celtic Metal Work (£). 
 
 Boss of a shield, of perhaps the first century B.C., found in the Thames 
 near Wandsworth, and now in the British Museum. See p. 48. 
 
ROMANIZATION IN ART 
 
 49 
 
 At Castor and Chesterton, on the north and south sides 
 of the river, were two Romano-British settlements of com- 
 fortable houses, furnished in genuine Roman style. Round 
 them stretched extensive pottery works, which seem to have 
 been active during the greater part of the Imperial period. 
 The ware, or rather the most characteristic of the wares 
 
 Fig. 17. Fragments of New Forest Pottery with Leaf Patterns. 
 
 (From Archaeologia.) See p. 48. 
 
 made in these works, is generally called Castor (or sometimes 
 Durobrivian) ware. It was not, indeed, peculiar to the 
 potters of the Ncn valley. There is evidence that, to some 
 small extent at least, it was made elsewhere in Britain, and 
 it must have been produced freely in northern Gaul, though 
 none of its kilns has yet been identified there ; possibly it 
 was produced there first and afterwards copied in Britain. 
 But Castor is the only attested centre of its manufacture 
 
 1751 d 
 
50 ROMANIZATION IX ART 
 
 on a large scale, and the cups and jars from its potteries 
 seem not only to be more abundant but also more varied in 
 decoration and sometimes more directly inspired by native 
 elements than the continental fabrics. 1 
 
 Castor ware was decorated by the method often called 
 4 barbotinc ' ; the ornament was in relief and was laid on by 
 hand in the form of a semi-liquid l slip ' with the aid of a tube 
 or other tool — just as in the later Roman Empire the orna- 
 ment was laid on glass, 8 or as in our own day it is put on 
 sugar-cakes. Every piece is, therefore, the individual product 
 of a potter, not a mechanical cast from a mould. From this 
 point of view it is noteworthy that the British Castor ware 
 directly embodies the Celtic tradition. If it was copied 
 from the Continent, the island potters either took over with 
 it an element which has all but disappeared from the 
 Gaulish work, or else they added that element. Castor ware 
 is based, indeed, on classical patterns —foliated scrolls, hunt- 
 ing scenes, gladiatorial combats, even now and then a mytho- 
 logical representation. But it recasts these patterns in 
 accordance with its own traditions and also with the vigour 
 of a true art. Those fantastic animals with strange out- 
 stretched legs and back-turned heads and eager eyes ; those 
 tiny scrolls scattered by way of background above and below 
 them ; the rude beading which serves, not ineffectively, 
 for ornament or for dividing line ; the suggestions of return- 
 ing spirals; the manifest delight of the artist in plant and 
 animal forms — all these things are Celtic (Figs. IS. 10). 
 
 When we turn to the scenes in which man is prominent — 
 a hunting picture in which (exceptionally) the huntsman 
 
 1 Good illustrations of continental Castor ware arc given in SamnUung 
 
 Xirssrn, Koln, 1011, plates 77, 78. Continental manufacture, possibly 
 near Cologne, seems to be proved by the amount of the ware found 
 in the Low Countries. North France and Germany. In Germany the 
 production is said to have begun before a.i>. KM) and to have ceased 
 soon after a.i>. 200. Its decoration is almost wholly confined to rather 
 Stereotyped animals, but the Colchester 'gladiators" urn ', mentioning 
 the Thirtieth Legion (C. R. Smith, Coil. .Int. iv. 82; C. vii. 1885. B), 
 may be Rhenish manufacture. 
 
 2 Kisa, GUu im Altrrtumc, ii. 4175. 
 
» » ' • 
 
 » I > 
 
 
 , > ' > • 
 
 1 > > 
 
 50 
 
• 
 
 
 
 c 
 
 
 o 
 
 H 
 
 o 
 
 
 v>) 
 
 i 
 
 ^N ' 
 
 
 ^Vp 
 
 
 I 
 
 ■ 
 
 g 
 
 : <^>*~, 
 
 
 1 
 
 fe 
 
 
 
 
 
 
 
 
 
 50 
 
ROMAXIZATION IN ART 
 
 51 
 
 appears, or a chariot race, or a gladiatorial show, or Hesione 
 fettered naked to a rock and Hercules saving her from the 
 sea-monster (Fig. 20) x — we do not always find the same skill 
 and vigour, i^rom of old the Celtic artist had been averse 
 to representations of the human form. When with an 
 initiative lacking in his continental rival — an initiative 
 
 Fig. 20. Hercules rescuing Hesione. (From a piece of Castor ware 
 found in Northamptonshire . C. R. Smith, Coll. Ant., vol. iv, PI. XXIV.) 
 
 which it is fair to recognize — he added this to his repertory, 
 he passed beyond his proper bounds. Now and then he suc- 
 ceeded ; more often he failed ; his Hercules and Hesione 
 are not fantastic but grotesque. In taking in new Roman 
 elements, his Celtic art lost its power and approximated to 
 the conventionalism of Samian ware. 2 
 
 1 This and the corresponding scene of Perseus and Andromeda were 
 popular in Britain and Gaul. See (e.g.) a tombstone at Chester 
 (Grosvrnor Museum Catal. No. 138), and others at Trier (Ilettncr, 
 Sieindenkmdler zu Trier, p. 200) and Arlon (Wiltheim, Luciliburgensia, 
 plate 57) and Igel. Whether the scenes generally conveyed any 
 symbolical meaning in these lands, I should greatly doubt. 
 
 2 For an account of Castor and Castor ware see Vict. Hist. Xorthants, 
 i. 100-78, 20G-13. 
 
 d2 
 
52 
 
 IIOMAMZATIOX IX ART 
 
 Brooches tell much the same tale of predominant Roman 
 fashions not unmixed with Celtic survivals. Many of those 
 found in Britain are peculiarly British. One of the com- 
 monest of Romano-British ' fibulae '. commoner in the north 
 than in the south of the island, is not only directly traceable 
 to a Celtic ancestry, but is very rare outside Britain. 1 The 
 
 examples which have 
 been found in northern 
 Gaul and Germany can 
 almost be counted on 
 t he fingers of two hands ; 
 and when a specimen 
 once turned up near 
 Frankfurt, it so startled 
 the local archaeologists 
 that they assigned it to 
 Africa. But the most 
 si riking example is sup- 
 plied by the enamelled 
 k dragon-brooches ' (Fig. 
 21). Both their designs 
 and their gorgeous colour - 
 Fig. 21. 'Dragon-brooches' found i n(r are Celtic in spirit : 
 
 AT CORBRIDGE ({). (P. 52.) 
 
 thev occur not seldom 
 
 in Britain ; from the Continent only four instances are 
 recorded. 2 Here certainly Roman Britain is more Celtic than 
 Gallia Belgica or the Rhine valley. Yet a complete survey 
 of the brooches used in Britain would show, especially in 
 the south, a dominant army of types which were equally, 
 common here and on the Continent and belong to the 
 Roman provincial civilization. The 'Aucissa 1 and 'knee' 
 and 'cross-bow' varieties may serve as examples. 
 
 1 For iIk- origin <>f the type see A. .T. Evans, Archaeol. Iv. 182 : 
 for illustrations and f<»r the distribution, my note, Arch. AeHana^ 1009, 
 
 l». 400, anil Curie, Newstead, p. :5'_>i. 
 
 * I have given a list in Arch. Aelimna, it>0!>. p. 120 : see also Curie, 
 Newstead, j». 810, and R. A. Smith. Proc. Soc. Antiq. Lond. xxii. 61. 
 In all about twtntx examples have been noted in Britain. 
 
^. 
 
 t 
 
 Fio. 22, The Corbridge Lion (p. 53.) 
 
 53 
 
ROMANIZATION IN ART 53 
 
 Perhaps it is to this survival of the Celtic spirit in a 
 Romanized Britain that we should ascribe two remarkable 
 sculptures found at Bath and at Corbridge. The Spa at 
 Bath (Aquae Sulis) contained a stately temple to Sul or 
 Sulis Minerva, goddess of the hot springs. The pediment 
 of this temple, partly preserved by a lucky accident and 
 unearthed in 1790, was carved with a trophy of arms —in the 
 centre a round wreathed shield upheld by two Victories, and 
 below and on either side a helmet, a standard (?), a cuirass, 
 besides other details now lost. It is a classical group, such as 
 occurs on other Roman reliefs. But its treatment breaks clean 
 away from the classical. The sculptor placed on the shield 
 a Gorgon's head, as suits alike Minerva and a shield (see 
 Frontispiece). But he gave to the Gorgon a beard and 
 moustache, almost in the manner of a head of Fear, and he 
 wrought its features with a fierce virile vigour that finds no 
 kin in Greek or Roman art. I need not here discuss the 
 reasons which may have led him to add male attributes to 
 a female type. For our present purpose the important fact 
 is that he could do it. Here is proof that, for once at least, 
 the supremacy of the dominant conventional art of the 
 Empire could be rudely broken down. 1 
 
 Another example is supplied by the Corbridge Lion, found 
 among the ruins of Corstopitum in Northumberland in 1907 
 (Fig. 22). It is a sculpture in the round showing a nearly 
 life-sized lion standing above his prey. The scene is common 
 in provincial Roman work, and not least in Gaul and Britain. 
 Often it is connected with graves ; sometimes (as perhaps 
 here) it served for the ornament of a fountain. But if the 
 scene is common, the execution of it is not. Technically, 
 indeed, the piece is open to criticism. The lion is not the 
 ordinary beast of nature. His face, the pose of his feet, the 
 curl of his tail round his hind leg, arc all untrue to life. The 
 
 1 For the temple and pediment see Vict. Hist. Somerset, i. 229 foil., 
 and references given there ; I have discussed the artistic problem 
 on p. 235 and Journal of Roman Studies, ii. 132. Quite recently, 
 M. Adolphe Reinach has suggested that the head embodies a definite 
 Celtic idea (Bull, du musee de Mulftouse, xxxvii). 
 
54 UOMANIZATIOX IN ART 
 
 man who carved him knew perhaps more of dogs than 
 lions. But he* fashioned a living animal. Fantastic and 
 even grotesque as it is. his work possesses a wholly unclassical 
 fierceness and vigour, and nol a few observers have remarked 
 when seeing it that it recalls not the Roman world but the 
 Middle Ages. 1 
 
 These exceptions to the ruling Roman provincial culture 
 are rare in Britain. But they are probably commoner here 
 than in the Celtic lands across the Channel. In northern 
 Gaul we meet no such vigorous semi-barbaric carving as the 
 Gorgon or the Lion. At Trier, Mctz, Arlon, Sens, there are 
 notable sculptures, but they are consistently classical in 
 style and feeling, and the value of this fact is none the 
 less if (with some writers) we find special geographical 
 reasons for the occurrence of certain of these sculptures.' 2 
 
 Exceptions are always more interesting than rules — even 
 in grammar. But the exceptions pass and the rules remain. 
 The Castor ware and the Gorgon's head are exceptions. The 
 rule stands that the material civilization of Britain was pre- 
 dominantly Roman. Except the Gorgon, every worked or 
 sculptured stone at Bath follows the classical conventions. 
 Except the Castor and New Forest pottery, all the better 
 earthenware in use in Britain obeys the same law. The 
 kind that was most generally employed for all but the meaner 
 purposes, was not Castor but Samian. 3 This ware is charac- 
 
 1 Arch. AeKana, 1908, p. 20,"> ; Journal of Roman Studies, ii. Ms. 
 
 - Michaelis, Loeschke and others assume an early intercourse between 
 
 the Hose] basin and eastern Europe, and thereby explain both a statue 
 in Pergamene style, which was found at Metz and appears t<> have been 
 
 carved there, and also the Neumagen sculptures. As all these pieces 
 
 were produced in Roman times, early intercourse seems an inadequate 
 
 cause. Moreover, Pergamene work, it rare in Italy, occurs in Aquitaniu 
 and Africa, and may have been popular in the provinces. 
 
 3 I may protest against the attempts made from time to time to 
 dispossess the term ' Samian '. Nothing better has been proposed, and 
 it has the merit of perfect lucidity, of the substitutes suggested, 
 ■ Pseudo-Arretine ' is clumsy, 'Terra Sfgillata " is a1 least as incorrect, 
 and 'Gaulish " covers only part of the held (Proc. Soc. Antiq. Loud. 
 wiii. 120). 
 
ROMANIZATION IN ART 55 
 
 teristic of Roman provincial art. As I have said, it is copied 
 wholesale from Italian originals (p. 19). It is purely imita- 
 tive and conventional ; it reveals none of that delight in 
 ornament, that spontaneousness in devising decoration and 
 in working out artistic patterns which can clearly be traced 
 in Late Celtic work. It is simply classical, in an inferior 
 degree. 
 
 The contrast between this Romano-British civilization 
 and the native art which preceded it can readily be seen if 
 we compare for a moment a Celtic village and a Romano- 
 British village. Examples of each have been carefully 
 excavated in the south-west of England, hardly thirty miles 
 apart. The Celtic village was close to Glastonbury in Somer- 
 set. 1 Of itself it was a small, poor place — just a group of 
 pile-dwellings rising out of a marsh and dating from the two 
 centuries immediately preceding the Christian era. Yet, 
 poor as it was, its art is distinct. There one recognizes all 
 that delight in decoration and that genuine artistic instinct 
 which mark Late Celtic work, while technical details in the 
 ornament (as, for example, the returning spiral) reveal their 
 affinity with the same native fashion. On the other hand, 
 no trace of classical workmanship or design intrudes. There 
 has not been found anywhere in the village even a 'fibula ' 
 with a hinge instead of a spring, or of an Italian (as opposed 
 to a Late Celtic) pattern. 
 
 Turn now to the Romano-British villages excavated by 
 General Pitt-Rivers and already mentioned in these pages 
 (p. 45). Here you may search in vain for vestiges of the 
 native art or of that delight in artistic ornament which 
 characterizes it. The ground-plans of the villages, the forms 
 of the poor cottages, are native ; the art is Roman. Every- 
 where the monotonous Roman culture meets the eye. To 
 pass from Glastonbury to Woodcutts is like passing from 
 
 1 The Glastonbury village was excavated in and alter 1892 at 
 intervals ; a full account of the finds is now being issued by Bulleid 
 and Gray (The Glastonbury Lake Village, vol. i, 1911, with a preface 
 by Dr. R. Munro). The finds themselves are mostly at Glastonbury. 
 
56 ROMAXIZATIOX IX ART 
 
 some old timbered village of Kent or Sussex to the uniform 
 streets of a modern city suburb. Life at Woodeutts had, no 
 doubt, its barbaric side. Out- writer who has discussed it 
 with a view to the present problem l comments on ' dwellings 
 connected with pits used as storage rooms, refuse sinks, and 
 burial places' and ' corpses crouching in un-Roman positions'. 
 The first feature has its parallels in modern countries and 
 was doubtless common in ancient Italy. The second would 
 be more significant if such skeletons occupied all or even 
 the majority of the graves in these villages. Neither feature 
 really mars the broad result, that the material life was 
 Roman. Perhaps the villagers knew little enough of Roman 
 civilization in its higher aspects. Perhaps they did not 
 speak Latin fluently or often. They may well have counted 
 among the less Romanized of the southern Britons. V. t 
 round them too clung the heavy inevitable atmosphere of 
 the Roman material civilization. 
 
 1 Yinogradoff, Growth of the Manor, p. 30. A parallel to the non- 
 Roman burials found by General Pitt-llivers may be found in the will 
 of a Lingonian Gaul who died in the latter part of the first century. 
 He was a Roman citizen, and his will is drawn in strict Roman fashion. 
 Hut its last clause orders the burning of all his hunting apparatus, 
 spears and nets, &c, on his funeral pyre, and thus betrays the Gaulish 
 habit (Dessau, Inscr. sel. 8379). 
 
 That earlier native forms of burial were used in Roman Britain is 
 shown by the remarkable burial mounds of the first and second 
 centuries at Rartlow Hills in NW. Essex [Archaeol. XXV, xxvi, xxviii, 
 xxix), Mersea Island {Trans. Essex Arch. Soc. xiii. 110), Rougham 
 (1/(7. Hist. Suffolk, i. 315), Gorsley Woods in East Kent (Arch. Can- 
 liana, XV. 311 , Thornborough in Rucks remains at Audlcv Mud), and 
 Youngsbury, in Herts Archaeol. Hi. 287 ■ They occur also in Belgium; 
 sec Annates dc In Soc. arch. <le Xamur, xxiv. .>(), and now Cuinonf 
 
 Belgiqut romanisie, p. 88. 
 
CHAPTER VI 
 
 romanization in town-llfe, local government am) 
 
 Land-Tenure 
 
 I have now dealt with the language and the material 
 civilization of the province of Britain. I pass to a third 
 and harder question, the administrative framework of local 
 Romano-British life, the town-system and local government, 
 and the land-tenure. Here we have to discuss especially 
 the extent to which the Roman coloniae and municipia 
 penetrated the province and the substitutes which arose 
 instead of them, and the diffusion and influence of the 
 Roman ' villa '. In respect to the towns and the local 
 government, it has to be remembered that Roman, like 
 Greek, towns were each the head of a dependent district, 
 and therefore what we might now call the town and the 
 county government more or less coincided. 
 
 I. First, the towns. Britain, we know, contained five 
 municipalities of the privileged Italian type. The colonia of 
 Camulodunum (Colchester) and the municipium of Veru- 
 lamium (St. Albans), both in the south-east of the island, 
 were established soon after the Claudian conquest of a. d. 43. 
 The colonia of Lindum (Lincoln) was probably founded in 
 the early Flavian period (a.d. 70-80), when the Ninth 
 Legion, hitherto at Lincoln, seems to have been pushed 
 forward to York. The colonia at Glevum (Gloucester) arose 
 in a.d. 96-98, as an inscription definitely attests. Lastly, 
 the colonia at Eburacum (York) must have grown up during 
 the second or the early third century, under the ramparts 
 of the legionary fortress, though separated from it by the 
 intervening river Ouse. 1 Each of these five towns had, 
 
 1 The fortress was situated on the left or cast bank of the Ohm- ; 
 the present cathedral stands wholly within its area. Tarts of the 
 
58 ROMAXIZATIOX IX TOWN-LIFE 
 
 doubtless, its dependent territory, which may have been as 
 large as an average English county, and each provided the 
 
 local government for its territory. 1 That implies a definitely 
 Roman form of local government for a considerable area — 
 
 a larger area, certainly, than received such organization in 
 northern Gaul. Yet it accounts, on a liberal estimate, for 
 barely one-eighth of the civilized part of the province. 
 
 Throughout most of the rest of the British province, 
 or rather of its civilized area, the local government was 
 probably organized on the same cantonal system as obtained 
 in northern Gaul (p. 21). According to this system, the local 
 unit was the former territory of the independent tribe or 
 canton, and the local magistrates were the chiefs or nobles of 
 the tribe. That may appear at first sight to be a native system, 
 wholly out of harmony with the Roman method of govern- 
 ment by municipalities. Yet such was not its actual effect. 
 The cantonal or tribal magistrates were classified and arranged 
 just like the magistrates of a municipality. They even used 
 tlie same titles. The cantonal civtias had its duoviri and 
 quaestors and so forth, and its ordo or senate, precisely like 
 any municipal colonia or municipium. So far from wearing 
 a native aspect, this cantonal system became one of tin 
 influences which aided the Homanization of the country. 
 It did not, indeed, involve, like the municipal system, the 
 substitution of an Italian for a native institution. Instead, 
 it permitted the complete remodelling of the native institu- 
 tion by the interpenetration of Italian influences. 
 
 We can discern the cantonal system at several points in 
 Britain. Hut the British cantons were smaller and less 
 
 Roman walls can still be traced, especially at the Multangular Tower. 
 
 The municipality lay 00 the Other hank of the Ouse, mar tin railway 
 station, where mosaics indicate dwelling-houses. IN outline and plan 
 
 arc, however, unknown. Even its situation has not been generally 
 recognized. 
 
 1 If the evidence Of milestones may be pressed, the territory of 
 Khuracum extended southwards at least twenty miles to ( astleford, 
 and that of Lincoln at least fourteen miles to LittleborOUgfa [Eph. 
 
 Epigr. vii. i 105 i.\. rj."»:{, and vii. i<>*.'?). The general size of these 
 
 municipal ' territoria ' is proved hy Continental inscriptions. 
 
LOCAL GOVERNMENT AND LAND-TENURE 59 
 
 wealthy than those of Gaul, and therefore they have not 
 left their mark, either in monuments or in nomenclature, s<> 
 
 
 m 
 
 % 
 
 ;X LEG-LEG^,,, 
 L^YD¥^BN 
 
 L 
 
 » 
 
 &• 
 
 T* 
 
 )EC 
 
 TO 
 
 , !iPVBI?CIV.I 
 
 1 %ILVfWM 
 
 r 
 
 2 
 
 %. 
 
 '/7777TT/T7TTTTT rr 7 MPM »*p*», 
 
 2 
 
 ^ * * 
 
 ^ 
 
 Fig. 23. Inscription found at Caerwent mentioning a Decree 
 of the Senate of the Canton of Silures. See p. 00. 
 
 dearly as we might desire. Many inscriptions record the 
 working of the system in Gaul. Many modern towns- 
 Paris, Reims, Amiens, and thirty or forty others -derive 
 
GO ROMAXIZATIOX IX TOWX-LIFE 
 
 their present names from those of the ancient cantons, and 
 not from those of the ancient towns. Britain has hitherto 
 yielded only one such inscription (Fig. 28), 1 on a monument 
 erected at Caerwent (Venta Silurum) by the cantonal senate 
 of the Silures to some genera] of the Second Legion at Isca 
 Silurum, twelve miles off. Only one British town was called 
 in antiquity by a tribal name and that is a doubtful 
 instance.- Xo single case occurs in which a modern town- 
 name is derived from the name of a British t ril>c :{ _>^ 
 
 We have, however, some curious evidence from another 
 source. There is a late and obscure Gvugruplnf "I the Roman 
 Empire which was probably compiled at Ravenna somewhere 
 about a.d. 700, and which, as its author's name is lost, is 
 generally ([noted as the work of i Ravennas \ It consists 
 for the most part of lists of names, copied from sources far 
 earlier than the seventh century, and very carelessly copied. 
 In general it adds very few details. But in the case of 
 Britain it notes the municipal rank of three of the four 
 coloniae, and it further appends tribal names to nine or ten 
 town-names, which are thus distinguished from all other 
 British place-names. For example, we have Venta Belgarum 
 (Winchester), not Venta simply, and Corinium Dobunorum 
 (Cirencester), not Corinium simply. The towns thus specially 
 marked out are just those towns which are also declared by 
 
 1 Found in 1908 : . . . leg. leg. [/]/, Aug. procon8ul(i) provinc. Nar~ 
 bonensie, leg. Ana. pr. }>r. proui. Lugudunen(sis) : ex decreto ordinis 
 
 respubl{i<(i) <iiit(atis) SitUTUtn. It was probably set up to Claudius 
 Paulinus, early in the third century [Athenaeum, Sept. 2(5, !!>().'{; 
 ArchaeOlogia, lix. 120; Kpfi. ix. 1012). Other inscriptions mention 
 a ciris Cuntius, a civitas CotUVeUaunorum and the like, hut their 
 evidence is less distinct. 
 
 J Icinoe in Kin. Ant. t? 4. (» may be Venta [cenorum [Victoria Hist. 
 of Norfolk, i. 286, 800). In its Gaulish section the [tin, uses these 
 
 tribal town-names ahont as often as not. 
 
 * Canterbury may seem an exception. Bui its name comes ultimately 
 
 from the Marly English Form of Cantium, not from the Cantii. In the 
 south-west and in Walts, tribal names like Dnninonii (Devonshire), 
 Demetae, Ordovices, have lingered od in one form or another ; accord- 
 ing to Prof. Hh\s, Hernieia is derivable from Brigantes, Hut these 
 cases differ widely from the Gaulish instances. 
 
LOCAL GOVERNMENT AND LAND-TENURE 61 
 
 actual remains to have been the chief country towns of 
 Roman Britain. This coincidence can hardly be chance. 
 We may infer that the towns to which the Ravennas appends 
 tribal names were the cantonal capitals of the districts of 
 Roman Britain, and that a list of them, presumably muti- 
 lated and imperfect, has been preserved by some chance in 
 this late corrupt compilation. 1 
 
 In other words, the larger part of Roman Britain was 
 divided up into districts corresponding to the territories ^ 
 of the Celtic tribes ; each has its capital, and presum- ^ 
 ably its magistrates and senate, as the above-mentioned 
 inscription shows that the Silures had at Venta Silurum. 
 We may suppose, indeed, that the district magistrates — 
 the county council, as it would now be called — were 
 also the magistrates of the country town. The same 
 cantonal system, then, existed here as in northern Gaul. 
 Only, it was weaker in Britain. It could not impose 
 tribal names on the towns, and it went down easily when 
 the Empire fell. In northern Gaul, Nemetacum Atrebatum 
 became Atrcbatis and is now Arras. In Britain, Calleva 
 Atrebatum (Silchester) remained Calleva, so far as we know, 
 till it perished altogether in the fifth century. 
 
 Municipalities and cantonal capitals furnish nearly all the— 
 known examples of Romano-British towns. Two or three 
 lesser places may have been secondary country-towns. 2 
 A spa, rather than a town proper, flourished at Bath, and 
 attracted invalids from Britain and from northern Gaul. 
 There is only one important addition to be made to our 
 list. Londinium sprang up in the earliest Roman period, 
 on a spot marked out by trade advantages rather than by 
 
 1 Ravennas (ed. Parfhey and Pinder), pp. 425 foil. ; my Appendix 
 to Mommsen's Provinces of the Empire (English trans., 1909), ii. .'J.">2. 
 The places are those now known as : Exeter, Winchester, Caerwent, 
 Cirencester, Silchester, Canterbury, Wroxeter, Leicester, Castor by 
 Norwich, and probably Chichester : to these we may add from other 
 sources Aldborough (Yorks) and Dorchester in Dorset. 
 
 2 Rochester, in Kent and Kenchestcr near Hereford are the only 
 ones which merit mention here. 
 
62 ROMANIZATION IX TOWN-LIFE 
 
 any noteworthy native settlement ; it quickly #rew to be 
 the largest and richest town in the provinee. But we never 
 hear that it won municipal rank, and its civic constitution 
 rested perhaps on a different basis. \V< know from Taeitus 
 that it began as a gathering of traders round a convenient 
 centre. We know also that the Roman provinces contained 
 many such clubs or communities of Roman traders, ruling 
 themselves on a quasi-municipal pattern (16, note 1). We 
 may think that London was, at the outset, one of these 
 communities, and that, while most of them grew into muni- 
 cipalities, it kept its original status unaltered. The Empire 
 was as full of irregularities as the Greek accidence, and 
 Roman opportunism loved to let well alone. London in 
 the fourth century gained the title — honourable, if not 
 rare — of Augusta, but remained in its quasi-municipal 
 * position. 1 
 
 On paper this represents much Romanized town-life in 
 Britain. Did the facts bear out the theory ? On the whole, 
 we may say that they did. The Romano-British towns 
 were of fair size. Silchester was by no means the biggest. 
 Roman London, perhaps even Roman Cirencester, were 
 larger than Roman Cologne or Bordeaux : Verulam and 
 others were not so far behind. 2 They possessed, too. the 
 buildings proper to a Roman town town-hall, market-place, 
 public baths, 'chess-board 5 street-plan, all of Roman 
 fashion ; they had also shops and temples, and even here 
 and there a hotel ; and it is to be noted that these were 
 present not only in the municipalities, as it seems, but in 
 
 1 Londinium is often credited with wonderful features— territory, 
 
 pomerium, citadel, jurisdiction to a mile outside its ^ates, and s,> forth. 
 No true view of it can be got, unIe8S these he put aside. 
 
 2 Within the walls, London was about 825 acres, Cirencester a little 
 
 over 240, Cologne 2 10, Verulam 2()0, Silchester, Colchester and Leicester 
 
 iio-ioo acres. Comparisons, however, arc difficult, even where the 
 walled ares is known, since sometimes (at London. Silchester, Trier. 
 
 Cologne) the walls seem to have enclosed the town at near its largest, 
 while elsewhere the walled area is hut a fragment left after Teutonic 
 invasion. Foi Bordeaux see Jullian, laser, de Bordeaux, h. ."><ss ; 
 
 at its zenith, he tells me. it perhaps covered 185 <>r 200 acres. 
 
LOCAL GOVERNMENT AND LAND-TENURE 63 
 
 the cantonal capitals as well. Whether and how far the 
 municipalities had a stronger Roman colouring than the 
 other towns, we do not know. But we can sec that their 
 Roman constitutions were realities ; witness the tiles of 
 Roman Gloucester, with their stamp RPG (respublica 
 Glevensmm) and their dating by municipal magistrates, 
 the ' duovin ' and ' quinqucnnalcs '. 
 
 NORTH GATE 
 
 SILCHESTER 
 
 kgt gate 
 
 BOO 4O0 MO 
 
 SOUTH GATE 
 
 Fig. 24. (p. 04. ) 
 
 Other details point somewhat the other way. We should 
 have expected the British municipalities, like those of other 
 provinces, to have helped in supplying the Roman army 
 with legionaries and the Roman administration with officials. 
 But, so far as present evidence goes, few Britons served in> 
 
64 ROMANIZATION IN TOWN-LIFE 
 
 tin- legions and hardly any won official rank. Again, the 
 plans of the towns known to us reveal a significant feature, 
 which I have noted already (p. 40). The dwelling-houses 
 
 in them are not town-houses, fitted to stand side by side 
 and to form regular streets ; they are country-houses, such 
 as neither did nor could combine in continuous rows; they 
 are dotted about like cottages in a village (Figs. 12. 2-1). 
 One recognizes that the town-planning of Silehcster or 
 Caerwent was introduced amid surroundings not fully urban 
 and that it represents an attempt at municipalization for 
 which the dwellers in Calleva and Venta were not ready. 
 These men learnt town-life from Rome. They did not learn 
 it in its highest form. Indeed, through all the rebuildings 
 which the spade reveals in these towns, they clung till the 
 end to their older rural fashion. 1 
 
 Those who weigh these facts against one another will eon- 
 elude, I think, that the Roman town-system of Britain was 
 a real thing. It contained native as well as Roman elements ; 
 here, as elsewhere, Romanization was a subtler and more 
 complex process than mere absorption in Rome. The towns. 
 too, were neither many nor very large ; here, as elsewhere, 
 Romano British life was on a small scale. Hut in one way 
 or another and to a real amount. Britain shared in that 
 expansion of town-life which formed a special achievement 
 of the Roman Empire. 
 
 The towns and the districts connected witli them occupied 
 most of the British lowlands. Whatever was over, fell prob- 
 ably within the Imperial domains, which covered wide tracts 
 in every province and were administered by local* procura- 
 tors ' of the Emperor. The lead-mining districts — Mcndip in 
 Somerset, the neighbourhood of Matlock in Derbyshire, the 
 Shelve Hills south-west of Wroxeter, the llalkvn region in 
 Flintshire, the moors of south-west Yorkshire — must have 
 belonged to these Domains, and for the most part an- actually 
 attested by inscriptions on lead-pigs as Imperial property. 
 Of other domain lands we meet what stems to be one early 
 1 See further my .iiirirnl Town-pkmntng, pp. 127-33. 
 
\-^ 
 
 LOCAL GOVERNMENT AND LAND-TENURE 05 
 
 instance at Silchester in the reign of Nero 1 — perhaps the 
 confiscated estates of sonic British prince or noble -and 
 though we have no further direct evidence, the history of 
 other provinces suggests that the area increased as the years 
 went by. Yet it is likely that in Britain, as indeed in Gaul, 2 
 the domain lands were comparatively small in extent. 
 Moreover, if we may trust analogies from Asia Minor, they 
 probably contributed little to Romanization (p. 18). 
 
 II. It remains to say what little can be said as to the 
 land-tenure of the province. Evidence on this point is un 
 fortunately very scanty. We know next to nothing about 
 either the size or the character of the estates which corre- 
 sponded to the country-houses and farms of which remains 
 survive. The ' villa ' system of demesne farms and serfs 
 or colonic which obtained elsewhere, was doubtless familiar 
 in Britain. Indeed, the Theodosian Code definitely refers 
 to British colonic But whether it was the only rural system 
 in Britain is beyond proof, and previous attempts to work 
 out the problem have done little more than demonstrate 
 the fact. 5 It is quite likely that here, as indeed in any 
 
 1 Tile inscribed NEHCIX.EA^GGiH, Nero Claudius Caesar Augustus 
 Germanicus (Ej>h. ix. 1267). It differs markedly from the ordinary 
 Silchester tiles, and plainly belongs to a different period in the history 
 of the site. Possibly the estate, or whatever it was. did not remain 
 Imperial after Nero's fall ; compare Plutarch, Galba, 5. The Combe 
 Down principia (C. vii. 02), which arc not military, may supply 
 another example, of about A.D. 210 (Vict. Hist. Somerset, i. .'Ill ; 
 Epfi. ix. .510). 
 
 - Ilirschfckh Klio, ii. .'J07, :508. Much of the Gaulish domain land 
 appears to date from confiscations in A.D. 197. 
 
 1 The term ' villa ' is now generally used to denote Roman count rv- 
 liouscs and farms, irrespective of their legal classification. The use is 
 so firmly established, both in England and abroad, that it would be 
 idle to attempt to alter it. Hut for clearness I have in this paper 
 employed the term 'villa' only where I refer to the definite -villi ' 
 system. 
 
 4 Cod. Tlieod. \i. ?. 2. 
 
 5 For instance, Seebolun {English Village Community, pp. 254 loll, 
 connected the sullix ' ham' with the Roman ; villa' and apparently 
 argued that the occurrence of the sullix indicated in general the former 
 existence of a ' villa \ But his map. showing the percentage of local 
 
 1751 E 
 
66 ROMANIZATION IN TOWN-LIFE 
 
 province, othei forms <>f estates and of land-tenure may have 
 existed beside the ' villa "- 1 The one thing needed is evidence. 
 Unfortunately, the sizes and relative positions of the country 
 dwellings do not, <>f themselves, reveal much in tins respect. 
 In some Rhenish districts the houses are so uniform in plan 
 and so evenly dist ributed as to suffffesl settlements of veteran 
 soldiers. In Britain the evidence at present known points 
 to a system which has grown up of itself bu1 does not show 
 
 the exact nature of that system. 
 
 In any ease, the net result appears fairly certain. The 
 bulk of British local government must have been carried on 
 through Roman municipalities, through imperial estates, 
 and still more through tribal civitates using a Romanized 
 constitution. The bulk of the landed estates must have 
 *y conformed in their legal aspects to the ' villas ' of other pro- 
 vinces. Whatever room there may be for the survival of 
 native customs or institutions, we have no evidence that 
 thev survived, within the lowlands, either in great amount 
 or in any form which conflicted with the genera] Romanized 
 character of the country, 
 
 names ending in ' ham * in various counties, disprove- his view. For 
 the distribution o! the Buffix * ham ' and the frequency of Roman 
 country-houses and farms do not coincide. In Norfolk, for instance, 
 1 ham is common, but there is hardly a Roman count rv-house or 
 farm in the county (Victoria Hist, of Norfolk, i. 294-8). Somerset, on 
 the othei hand, is crowded with Roman country-houses, and has 
 hardly any ' hams \ 
 
 1 Prof. Vinogradoff, Growth of the Manor, chap. ii. argw a for the 
 existence of Celtic land-tenures besides the Roman 'villa*. 'There 
 was room (he suggests) for all sorts of condition,, from almosl exact 
 copies ol Roman municipal corporations and Italian country-houses 
 to tribal arrangements scarcely colon red by a thin sprinkling of imperial 
 administration ' (p. «:{). This i> very probable, lint 1 find no definite 
 proof of it. It northern Gaul were better known it mighl provide 
 a decisive analogy. Bui the Gaulish evidence itsrli seems disputable. 
 
CHAPTER VII 
 
 ROMANIZATION IN RELIGION 
 
 Tile current l^li^po^, of the modern world, monotheistic 
 in character and (easTekn in origin, are exclusive ; no man 
 can be in any real seTlse Mahometan and Jew at once. The 
 polytheisms of ancient Europe contained little to hinder 
 combinations of creeds, and the Romans, being politic as 
 well as polytheistic, encouraged the process. They had easily 
 equated their own Italian gods with the gods of Greece ; the 
 provincials found it no harder to combine native provin- 
 cial cults with the Graeco-Roman religion. The western 
 half of the Empire thus became a blending-vat of worships, 
 western and eastern and Roman. The ruling element was 
 Roman. The native cults of western origin survived — at 
 least on the surface — mainly as appendages of Roman 
 deities, and even the far stronger eastern cults, Mithraism 
 and the rest, took on somewhat of Roman dress. The out- 
 come was too vague and ill-defined, and too various in 
 different lands, to be called a Roman provincial religion. 
 Rather, an equation of worships was established under 
 Roman primacy, by which a man who changed his town 
 or province, could change his gods as easily as he changed his 
 washerwoman. 
 
 This happened also in Britain. The inscriptions and 
 sculptures of our province show a mass of diverse cults 
 which were united in their use of Latin and in their common 
 Roman colouring. In detail, however, the military districts 
 differ widely, as so often, from the districts of civilian life, 
 in which the Romanized provincials dwelt. We may best 
 group our survey into (1) cults which seem strictly Roman, 
 (2) others which may be- called Komano-lVltic. and (.'$) 
 
 E 2 
 
6s R0MAN1ZAT10N IN RELIGION 
 
 others again which came to Britain from sources neither 
 
 Roman nor Celtic, but either Teutonic or Oriental. 
 
 I. Purely Roman dedications, such as an Italian mighl 
 have set up in Italy, arc common enough in the military 
 
 v area. There we meet altars to luppiter Optimus Maximus 
 and other true gods of Rome, without any intermixture of 
 non-Roman religion. But they arc altogether rare in the 
 towns and country districts. A few exceptions can be noted. 
 At Chichester in the middle of the first century a Roman- 
 izing native princelet set up a monument to Neptune and 
 Minerva. In the midlands, near Stony Stratford, a man 
 with a Celtic name. Vassinus, made some sort of offering to 
 Jove and Vulcan. A shrine in the Cotswolds contained a 
 figure of a god in lull armour, carved in stone, with tin 
 superscription deo Romtdo, 'to the- god Romulus. 1 In a 
 
 few places we meet altars set up simply to Mars or Mer- 
 cury or Aesculapius or Diana. Rut the total list of these 
 plain Roman dedications is short. Nor do we hear more 
 of the official worship of the Emperor. Dedications to his 
 Divinity (numina Augustorum, &c.) are frequent in forts 
 and fortresses. Elsewhere they are .scanty. In the colonia 
 of Camulodunum was a temple lor tin official cult of Rome 
 and the Emperor ; some years ago a boy fished (Hit of a 
 Suffolk stream a bronze head which was probably pillaged 
 from it in the rising of Boudicca. But we hear next t<> 
 nothing about the cult. It not only had no religious value ; it 
 had not even the social importance which it enjoyed in Gaul. 
 
 II. Far commoner are Romano-Celtic and native dedi- 
 cations. 1 Many of these arc dedications to Roman gods with 
 Celtic epithets, to Mars Belatueader. Mars Cocidius. Mars 
 
 Corotiacus, not to Mars simply. It does not appear that 
 the varieties of Mars which were thus created wielded different 
 
 powers, or that you prayed to Mars Belatueader lor one sort 
 
 of favour and to Mars Cocidius for another ; that doubtless 
 
 happened to some extent, hut it does not seem to have been 
 common. We may say rather that scattered, mostly local, 
 
 1 Anwyl's article in the Cambridge Medieval History, ii. 473-9, i?, 
 I fear, unsatisfactory. 
 
ROMAXIZATIOX IN RELIGION CO 
 
 cults crystallized round Roman names. It was, however, 
 only a few Roman gods — in Britain and in north Gaul 
 Mars and Mercury — who attracted Celtic epithets to them- 
 selves "at all freely. Apollo, Diana, Juno, Neptune, and the 
 rest appear comparatively seldom or even never with them. 
 On the other hand, a long series of dedications concern 
 gods whose names are purely Celtic except for their Latin 
 terminations. These are many. But they do not greatly 
 differ from those just described ; indeed, many Celtic deities 
 appear now with, now without, the Roman prefix. 
 
 If we now proceed to classify the Celtic cults of which we 
 meet remains in Britain, we must note first the absence of v 
 any hierarchy of great gods. Of Esus, Taranis and Teu- 
 tatcs, sometimes styled the Celtic Trinity, no sign emerges. 1 
 Instead, a crowd of lesser deities reveals a primitive religion 
 in much the same rudimentary state as were the religions 
 of Greece and Rome before the Olympian gods had become 
 acknowledged as supreme. Some bear names which seem 
 descriptive of character. Such was Belatucader, ' good Jit 
 war ', who was worshipped in the north and coupled with 
 Mars. Such, too, Maponus, kin somehow to the Welsh 
 ' Mabon ', a child, and habitually yoked with Apollo. 
 Others belonged to natural features. Verbeia at Ilklcy 
 was patron saint of a stream still called Wharfe ; the 
 Northumberland Cocidius (often Mars Cocidius) may have 
 begun as god of the Coquet. Others with less intelligible 
 names were clearly connected with special spots ; such 
 were Ancasta at Bittern (near Southampton), Coventina, 
 
 1 Teutates occurs oner, possibly twice, identified with Mars ; the 
 others are absent. A Chester altar (C. vii. 108) is said to read IOM 
 TANARO, but the reading is uncertain ; even if it be ri<dit, still 
 Tanarus is not (as Air. Holmes thinks. And. Britain, p. 279) the saun- 
 as Taranis. Whether these three ^ods were really so important, is 
 disputed; see Jullian, Cambridge Medieval History, ii. 404, (for) and 
 S. Reinach, Mythes et Calfes, i. 205 (against). .Mr. Holmes mistakes 
 the position when he says that 'the devotee who composed his inscrip- 
 tion to Toutates would not have wittingly ascribed to a mere local god 
 the qualities of Mars'. That is just what they did, .ill over Britain 
 and north Gaul. 
 
70 ROMANIZATION IN RELIGION 
 
 whose sacred water bubbled up within the shadow of the 
 Roman Wall, and Antenociticus, whose shrine now lies be- 
 neath a suburb of Newcastle. Sul or Sulis. thought to be by 
 origin the Celtic female Sun and identified with Minerva, 
 was goddess of the Bath waters. Nodens, kin to or bearing 
 the same name as an Irish hero. Xuada of the Silver Hand, 
 was worshipped in west Gloucestershire at Lydn< \ . 
 
 These cults and others like them :ire British. Some Celtic 
 dedications which occur in the province seem, on the- other 
 hand, to have been brought in from the Celtic mainland. 
 Mars Leucetius (the lightning god), Mars Rigisamus (most 
 royal). Mars Olludius, Apollo Grannus, belong across the 
 Channel : Grannus. god of healing waters, had a home at 
 Aachen. A Caerwent altar provides ;i signal example of 
 how such import happened. It was sel up by a quite 
 unknown man. one Nonius Romanus, to Mars Lenus or 
 Ocelus (Marti Leno sive Ocelo). Mars Lenus was a local saint 
 in the Mosel valley; Mars Ocelus has been met again in 
 Caerwent and also in the north. As the Celts of the Mosel 
 were wont to emigrate freely, it is pretty plain that Nonius 
 came thence to Caerwent ; there he wished to honour the 
 gods of his old and his new home, and equated the two in 
 one phrase. 1 Another and much better known example of 
 imported Celtic worship may be found in the Mother God- 
 desses, the deae metres. Every one who lias looked into 
 museums in the north of England or along the Rhine will 
 be familiar with the curious reliefs which show the Three 
 Mothers seated stiffly side by side, clothed in long robes 
 and strange headdress and often holding on their laps round 
 baskets of fruit. Their cult was common in north Italy and 
 south-eastern Gaul, and on the middle and lower Rhine, 
 and in Britain. But in Britain it is limited mainly to the 
 army ; its monuments occur, with comparatively few excep- 
 tions, within the military area, and the worshippers, SO 
 far as they state their professions, are nearlv all soldiers. 
 
 1 I'ljih. Kpigr. i\. 1182 and my note. We may ascribe to another 
 such immigranl tin- 'colonne au geanl ;ii Cirencester [Eph. i\. 997). 
 
ROMANIZATION IN RELIGION 71 
 
 Probably its birthplace was in the Celtic districts of northern 
 Italy and south-eastern Gaul, where the earliest dedications 
 have been found. There, during the early Empire, soldiers 
 were recruited in large numbers for service on the Rhine 
 and in Britain, and these soldiers took their native worship 
 with them. Only, from the Rhine garrisons the cult spread 
 to German and Gaulish tribes around, finding perhaps some 
 native Triad of Goddesses with which it amalgamated. 
 while in Britain it remained, for the most part, confined 
 within its military habitat. 1 
 
 III. Foreign cults were also imported into Britain from 
 non-Celtic sources. But these were confined to the haunts 
 of soldiers almost more rigidly than the Mother Goddesses. 
 One group, in its way an interesting group, consists of 
 Teutonic cults brought over by German soldiers serving in 
 t he northern British frontier garrisons. Sometimes these 
 Germans accepted the gods whom they found in their new 
 quarters ; thus, a little band of men who bear German names 
 and expressly call themselves ' Germani ', is found erect- 
 ing an altar to Maponus close by the Roman Wall. But 
 often they kept to their Teutonic deities —Mars Thingsus 
 and the Two Alaisiagae, Garmangabis, Viradecthis, the 
 Unseni Fersomari, and many more. One German cult even 
 spread a little, though not beyond military surroundings. 
 Tlie small ill-cut altars inscribed deo Ureter i or Vheteri or 
 Veteri were, as it appears, originally set up to a German god 
 Veter. Soon the worshippers forgot this and took the dedi- 
 cation to mean 'to the old god ' ; they even put it into the 
 plural and paicfs honour to the di veteres, the Old Gods 
 generally. 2 
 
 1 See Ihni, BomierJahrb'hherAxxxni. 1-200, and my paper in Archaeol. 
 Aeliana, xv. 314. Including the kindred Suleviae, &c., aboul 60 
 examples have been Pound in Britain; of these the civilian districts 
 furnish a sixth — Cirencester 4, London 2 or :{, Colchester, Bath, and 
 Lincoln, l each ; at Lincoln, as once or twice elsewhere in Celtic lands. 
 Matres have been latinized into Parcae. 
 
 - See my note on Eph. Epigr. ix. 1182. The spelling uhe~ or hve- 
 seems decisive of a Teutonic origin. The name is often written with 
 an i for one or both of the t's. 
 
72 ROMANIZATION IN RELIGION 
 
 Far more momentous to the Empire as a whole than 
 
 these little Teutonic cults were the immigrant religions 
 from the east, the worships of Mithras and Dolichenus and 
 Cybele and Isis and others. They were very powerful. 
 But in the Atlantic provinces, in Spain and western Gaul 
 and Britain, their power was limited. They were confined 
 to special arias, mid in particular to military areas. Mith- 
 r.iism. the greatesl of them all, overran Italy and central 
 [Europe and the Rhone valley which so closely copied Italy. 
 But further west and north it went only where the troops 
 went- to the Rhine frontier, to northern Britain, to the 
 legionary fortresses. From Gibraltar to Fifeshire, bareh 
 half a dozen Mithraic monuments have been recorded which 
 are not connected with the presence of soldiers. The cult 
 of the Semitic Dolichenus was equally widespread in Italy 
 and middle Europe and equally absent from Spain and from 
 all but the military districts of Gaul and Britain. Tin- 
 barbaric rites of Cybele. although (perhaps in mitigated 
 form) they invaded southern Gaul, were abhorred in the 
 west and above all in Britain. 1 If we would find eastern 
 cults in Britain, we must go to the military posts. At 
 Corstopitum on the Tyne, just south of the Wall, was a 
 military depot with some sort of settlement round it. where 
 all manner of military nun collected. There altars were 
 set up to Astarte (Ashtoreth). to Heracles of Tyre, to Doli- 
 chenus, to Sol Invietns. to Panthea (Isis ?). as will as to the 
 British Brigantia and Maponus and the German Veter. 
 Nothing of the sort occurs in the towns or country-houses of 
 southern Britain. Here, again, the influence of the Roman 
 garrisons in Britain was limited to themselves (p. 26). * 
 
 1 A. v. Domaszewski, Journal of Roman Studies, i. 54; Etoscher's 
 /./ (. Mythol. s.v. Meter. 2027. A statue from Cheaters {Lapid. Sept. 
 \ 10) is often said to represent Cybele, lait it is doubl nil. 
 
 statuettes. figurines and other small objects connected with 
 
 Oriental cults OCCUr, of course, lar beyond the limits noted in t lie 
 text. But, so lar as they were not mere enrios. they point mainly 
 
 to isolated worshippers. 
 
Fig. 25. Relief of Diana and Hound from Nettleton. (p. 73.) 
 
 (From a photograph.) 
 
 Fig. 26. Relief "i Mercury in full Roman Style, with 
 a Celtic < ««»i>i»i — . from Gloucester, (p. 73.) 
 
 (From a photograph.) 
 
ROMANIZATION IN RELIGION 73 
 
 In Britain, therefore, as in other western lands, Roman- 
 ization in religion meant, within the military area, a sentiaa 
 numinum, a kitchen-midden of all sorts of cults heaped up 
 from all quarters of the Empire. Outside that area it meant 
 a mixture of Roman and native deities. The proportions 
 of the mixture no doubt varied, as I have said above 
 (p. 21). But we find little, if anything, to suggesl thai 
 non-Roman elements were consciously preserved as being 
 non-Roman. Even in the countryside, even in the shrines 
 with ' Celtic ' plans (p. 36), dedications are uniformly couched 
 in Latin. At Nettleton, ten miles north-east of Bath, 
 chance finds seem to have revealed a ' Celtic ' temple witli 
 two reliefs of Diana. Both were fully Roman in style 
 (Fig. 25). Though no inscription survives to illuminate the 
 cult, we need not doubt that here the passers-by — whether 
 they knew it or not — worshipped Diana of the Romans. 
 At Lydney, in the sanctuary of the Celtic god Nodens (p. 70), 
 the temple-plan is Roman, the graffiti are in Latin, and 
 a representation of Nodens himself (as it seems) might 
 pass for a rude sketch of Neptune. 
 
 In all such cases Roman and native seem to be harmoni- 
 ously intertwined, but the Roman is supreme. It was. no 
 doubt, limited ; the mixture included, as a rule, only a few 
 of the Roman dominant gods. But it may be worth adding 
 that, while in northern Gaul a Roman god sometimes 
 appears along with a distinct Celtic companion. Mercury 
 (for instance) with Rosmerta, that particular manner of 
 mixing Roman and native is rarer in Britain. Here tin 
 native element asserted itself less definitely beside the Roman. 
 Now and then it occurs, as on a relief found in Gloucester ' 
 (Fig. 20). on which Mercury stands beside a goddess who 
 seems not to be Rosmerta but some other Celtic deity. 
 
 1 Catal. of Museum formed ok Gloucester . . . I860, p. «. 
 
CHAPTER VIII 
 Chronology of the Romanization 
 
 From the survey of the evidence which illustrates the 
 RomanizatioD of Britain, I pass to inquire how Ear history 
 helps us to trace the chronology of the process. A few facts 
 and probabilities emerge. 
 
 Intercourse between Britain and the Roman world began 
 when Caesar conquered Gaul. It had lasted nearly a century 
 when Claudius invaded the island in a.i>. 43. During thai 
 age south-eastern Britain learnt much from Home. Latin 
 words, as I have said above (p. 29), now appeared on British 
 coins. Arretine ware found its way. at hast in stray pieces. 
 to London (or South wark), to Colchester, to Foxton in 
 Cambridgeshire, to Alchester in Oxfordshire, to Purbeck in 
 Dorset and some similiar sites, and it was well known and 
 freely used at Silchester; the tribal capital of the Atrebates, 
 which grew into the Romano-British Calleva, must have 
 undergone some sort of Romanization Ion" before a.i>. 48. 1 
 The establishment of a Roman municipium at Verulam 
 (St. Albans) before A.D. 60, and probably before a.d. 50.- 
 points the same way. For the status of municipium w;is 
 granted in the earlier Empire especially to native provincial 
 
 towns which had, so to say. Romanized themselves, with- 
 out Roman official action or official settlement of Roman 
 soldiers or citizens, and had thus merited municipal privi- 
 
 1 For Southwark and London Bee Journal of Roman Studies, i. 146 ; 
 i he accounl of the Soul hwark piece lis Walters, Proc, Cambridgeshire 
 Antiq. Soc. \ii. 10T, is incorrect. The total amount ol Arretine (bund 
 in London is small compared with thai from Silchester and suggests 
 thai pre-Roman London a Southwark) was unimportant. For Foxton 
 & <• Babington, .inc. Cambridgeshire, i>. 64. For Alchester seem} 
 note Proc. Soc. Antiq. hand. \\i. nil. 
 
 it is very much more suitable to Claudius than to Nero, and more 
 suitable n. the earlier than t<> tin- later years <»f Claudius. 
 
CHRONOLOGY OF THE ROMANIZATION 75 
 
 leges. It is quite likely that such Romanization had com- 
 menced at Verulam before the Claudian conquest and formed 
 the justification for the early grant. 
 
 [ After the conquest, the lowlands as far west as Exeter and 
 Shrewsbury, and as far north as the Humbcr, were subdued 
 by a.d. 50. Romanization may therefore have marched on 
 at once. About a.d. 60 certainly, the insurgent Britons 
 under Boudicca (Boadicca) were able to massacre an enor- 
 mous number of Romans and ' friendlies ' -a number esti- 
 mated at the time as 70,000 — and many of the victims must 
 have been Romanized Britons; it is not impossible thai 
 this disaster arrested the civilizing process awhile. The real 
 advance seems to have come a little later, in the Flavian 
 period (a.d. 70-95). In that age many provinces stepped 
 Forward on the path to Roman culture. In Britain, towns 
 like Silchester, Caerwent, Wroxeter, 1 now take definite 
 shape, perhaps with official encouragement ; now, as we 
 may conjecture, tribal capitals were deliberately converted 
 into civilized towns, with street-plans and public buildings 
 of Roman type. Now, too, the spa at Bath developed. 2 
 Now, as Tacitus tells us, Latin began to be spoken, the toga 
 to be worn, temples, town-halls and private houses to be put 
 up in Roman fashion. Now also civil judges, legati iuridici, 
 were appointed, presumably to deal with litigation arising 
 out of the advancing civilization. 3 Tacitus states that 
 Agricola, as governor in Britain in 78-85, openly encouraged 
 this Romanization, and that his efforts met with great 
 
 1 Silchester was plainly laid out all at once, and though it certainly 
 existed in some form long before a.d. 70, the evidence of coins and 
 pottery implies that it took a big step forward soon alter 70 ; we may 
 connect that step with the laying out. At Caerwent and Wroxeter. 
 coins, pottery and brooches suggest that there was little, it any. town 
 life before the Flavian age and a good deal soon after. 
 
 2 At Bath the earliest datable stone belongs to A.D. 76, just before 
 Agricola came out (Vict. Hist. Somerset, i. 222, 269 ; Eph. Epigr. 
 ix. 99C). 
 
 3 A. v. Domaszewski, Rhein. Mus. xlvi. ."><)<) ; (II.. ix. :>:>:'>:',. inscr. of 
 Salvius Liberalis, iii. 2864 9960, inscr. of Iavolenus Prisons, both <>r 
 the Flavian period. 
 
< ll 
 
 ( IIROXOLOGY OF THE ROMANIZATION 
 
 success. We know, however, thai the movement began before 
 he reached Britain, and it would seem that he was rather 
 carrying out the policy <>f his age than his own. Anyhow, 
 the policy succeeded. In a. i>. 85 it was thought safe to 
 reduce the garrison of the province by a legion and some 
 1 auxilia '- perhaps a quarter or a fifth of its hitherto 
 st rength. 1 
 Of further progress during the second century we have 
 
 little exact information. On the one hand we find thai 
 serious risings vexed northern Britain at three points in this 
 century, about 115-120. again about 155-168, a and once 
 more about 175-180, when Caledonia was abandoned, while 
 the years which ended the second and opened the third 
 century were full of trouble. All this must have kepi even 
 the civilian area somewhat in disturbance. It was perhaps 
 at some crisis in this period that the nourishing county-town 
 of Isurium, a dozen miles north of York, had to shield 
 itself with stone wall and ditch. 3 On the other hand, the 
 development of the countryside by means of farms and 
 country-houses must have already begun. We meet early 
 traces of it in Kent and the south-eastern part of the island 
 generally, and sometimes outside these limits. Even in 
 Oxfordshire a site such as Xorthleigh (p. 11) has yielded 
 pottery which can hardly be later than the first half of the 
 second century. Even in tin- villages excavated by Pitt- 
 Rivers (p. 55), the use of Samian ware had spread before the 
 end of the first century. 
 
 Peace certainly set in after the opening of the third cen- 
 tury. It was then. I think, that country-houses and farms 
 
 1 Classical Review, 1904, p. 158 : 1905, p. 58, withdrawal of Batavian 
 
 cohorts. The withdrawal of Legio II AdintiK is well known. 
 
 J Archaeologia Aeliana, xxv. ( L904) 142 ? ; Proc. Soc. . intiq. Scotland, 
 
 wwiii. 1.11. 
 
 The town-wall of Lsurium, partly visible to-day, is buill in a fashion 
 which suggests the second century rather than the late third or the 
 fourth century, when most of the town-walls in Britain ami Gaul 
 wtr<' probabl} put up. 'rims, its masonry shows the 'diamond 
 
 brOOChing 1 which also OOCUrS On the Wall ol Pius ill Scotland and 
 
 which iiiuvt have therefore been in usejduring the second century. 
 
CHRONOLOGY OF THE ROMAN IZATIOX 77 
 
 became common in all parts of the civilized area. The 
 statistics of datable objects discovered in these buildings 
 seem conclusive on this point. Except in the south-eastern 
 region, coins and pottery of the first century are infrequent, 
 and many sites of rural dwellings have yielded nothing 
 earlier than about A. D. 250. Despite the ill name that 
 attaches to the third and fourth centuries, they were perhaps 
 for Britain, as for parts of Gaul, 1 a period of progressive 
 prosperity. Certainly, the number of British country- 
 houses and farms inhabited during the years A. d. '280-350 
 must have been very large. Prosperity culminated, it 
 seems, in the Constantinian Age. Then, as Eumenius tells 
 us, skilled artisans abounded in Britain far more than in Gaul, 
 and were fetched from the island to build public and private 
 edifices as far south as Autun.- Then, also, and, indeed, as 
 late as 3G0, British corn was largely exported to the Rhine 
 Valley, 3 and British cloth earned a notice in the eastern 
 Edict of Diocletian. 4 The province at that time was a pros- 
 perous and civilized region, where Lai; in speech and culture 
 might be expected to prevail widely. 
 
 V_No golden age lasts long. In 343 Constans had to cross 
 the Channel and repel the Piets and other assailants. 5 Alter 
 360 such aid was more often and more urgently required. 
 
 1 Mommsen, Rom. Gesch., v. 97, 100, and Ausonius, passim. 
 
 - Eumenius, Paiteg. Constantio Caesari, 21 civitas Aeduorum . . . 
 plurimo8 quibus Mae provinciae (Britain) redundabant accepit artifices, 
 el nunc exstruelionc veterum domorum el refectione operum publicorum 
 it templorum instauratione consurgit. 
 
 8 Ammianus, xviii. 2, 3 annona a Brittaniis sueta transferri ; Zosimus, 
 iii. .*>. 
 
 4 Edict. Diocl. xix. :J0. Compare Eumenius, Paneg. Constantino 
 Aug. 1) pecorum innumerabilis multitudo . . . onusta velleribus, and 
 Constantio Caesari, ll tanto taeta munere pastionum. Traces <»r dyeing 
 works have been discovered at Silchester (Archaeologia, liv. Mio. &c.) 
 and of lulling in rural dwellings at Chedworth in Gloucestershire] 
 Darenth in Kent, and Titsey in Surrey (Fox. Archaeologia, lix. 207). 
 
 6 Ammianus, xx. 1. The expedition was important enough to be 
 recorded on coins which show Constans on a galley, recrossing the 
 Channel after his victory (Cohen, 9-18, *.V:e.). On the history of the 
 whole period for Britain sec Cambridge Medieval History, i. 878. 
 
78 ( KRONOLOGY OF THE ROMAMZATIOX 
 
 Significantly enough, the lists of coins found in sonic country- 
 houses close about 850-60, while other houses remained occu- 
 pied till al>out 885 or even later. The rural districts, it is 
 plain, began then to be no longer safe: some houses were 
 burnt by marauding bauds, and some abandoned by their 
 owners. 1 In the crisis of ;3(>7-<S the ravages seem to have 
 spread over almost all the lowlands.- Therewith came 
 necessarily, as in many other provinces, a decline of Roman 
 influences and a rise of barbarism. Men took the lead who 
 were not polished and civilized Romans of Italy or of the 
 provinces, but warriors and captains of warrior bands. 
 The Menapian Carausius, whatever his birthplace, 9 was the 
 Ion runner of a numerous class. Finally, the great raid of 
 100-7 and its sequel severed Britain from Rome. A wedge 
 of barbarism was driven in between the two. and the central 
 government, itself in bitter need, ceased to send officers to 
 rule tlie province and to command its troops. Britain was 
 left to itself. Vet even now it did not seek separation from 
 Home. All that we know supports the view of Mommsen. 
 It was not Britain which broke loose from the Empire, but 
 the Empire which gave up Britain. 1 
 
 1 See, for example, the coin-finds of the country -houses ;it Thruxton 
 Abbots Ann, Clanville, Holbury, Carisbrooke, Ac., in Hampshire 
 (Vict. Hist, limits, i. 294 foil.). The Croydon hoard, deposited about 
 \.i). :{."")i (Numismatic Chronicle, l «><>.">. j». 87) may be due to the same 
 
 cause. 
 
 1 Ainniian. xxvii. 8, (>. 
 
 It is hard to believe him Irish (Rhys, Cumbrian Arckoeol. Assoc.. 
 
 Kerry Meeting, 1801). The one ancient authority, Aurelius Victor 
 (xxxix. 'jo), describes him simply as Menapiae civie. The Gaulish 
 Menapii were well known ; the Irish \h naph were very obscure, and 
 the brief reference can only denote the former. 
 
 * Mommsen, /»'"///. Gesch., v. 177. Zosimus, vi. ■"> (a.d. 108), in a 
 puzzling passage describes Britain as revolting from Home when 
 Constantine III was tyrant ( \.i>. 107—11). it is generally assumed 
 that when Constantine tailed to protect these regions, they set up for 
 themselves, and in thai troubled time Buch a step would i>e natural 
 enough. Bui Zosimus, a little later on (vi. in, a.d. U0), casually 
 
 states, in the middle Of B chapter about Italy, that HonoriuS wrote to 
 
 Britain, bidding the provincials defend themseryes, so that the act of 
 
 i os cannot have been final. l'ossii»i\ . however, as tin' contexl suggests 
 
CONCLUSIONS 79 
 
 Such is, in brief, the positive evidence, archaeological, 
 
 linguistic, and historical, which illustrates the Romaniza- 
 tion of Britain. The conclusions which it allows seem t<> be 
 two. First and mainly : the Empire did its work in our 
 island as it did generally on the western continent. It 
 Romanized the province, introducing Roman speech and 
 town-life and culture. Secondly, this Romanization was 
 not uniform throughout all sections of the population. 
 Within the lowlands the result was on the whole achieved. 
 In the towns and among the upper class in the country 
 Romanization was substantially complete — as complete as 
 in northern Gaul, and possibly even more complete. But 
 both the lack of definite evidence and the probabilities of 
 the case require us to admit that the peasantry may have 
 been less thoroughly Romanized. It was covered with a 
 superimposed layer of Roman civilization. But beneath this 
 layer the native element may have remained potentially, if 
 not actually, Celtic, and in the remoter districts the native 
 speech must have lingered on, like Erse or Manx to-day, as 
 a rival to the more fashionable Latin. How far this hap- 
 pened within the civilized lowland area we cannot tell. 
 But we may be sure that the military region, Wales and the 
 north, never became thoroughly Romanized, and Cornwall 
 and western Devon also lie beyond the pale (p. 24, note 3). 
 Here the Britons must have remained Celtic, or at least 
 capable of a reversion to the Celtic tradition. 
 
 and as Gothofredus and others have thought, the name ' Britain ' is 
 here a copyist's mistake for ' Bruttii '. In any case the * groans of tli< 
 Britons 1 recorded by Gildas, show that the island looked to Rome 
 lonji alter 410. On Constantino see Freeman, Western Europe in tin 
 Fifth Centura, pp. is, 148, and Bury, Life of St. Patrick, p. 829. 
 
CHAPTER IX 
 
 The Sequel, the Celtic Revival in the Lateb Empire 
 
 So far we have considered the province of Britain as it 
 was while it still remained in nal fad a province. Let "us 
 
 now turn to the sequel and ask how it tits in with its antc- 
 cedents. The Romanization, we find, held its own for 
 ;i while after loo. The sense of belonging to the Empire 
 had not quite died out even in sixth-century Britain. Roman 
 names continued to he used, not exclusively, but freely 
 enough, by Britons. Roman 'culture words * seem to occur 
 in tlie later British language, and some at Kast of these may 
 he traceable to the Roman occupation of the island. Roman 
 military terms appear, if scantily. Roman inscriptions are 
 occasionally set up. The Romanization of Britain was 
 plainly no mere interlude, which passed without leaving 
 a mark behind. 1 But it was crossed by two hostile forces, 
 a Celtic revival and an English invasion. 
 
 I. The Celtic revival was due to many influences. Wc 
 may find one cause for it in the Celtic environment of tin 
 province. After 107 the Romanized area was cut off from 
 Home. Its nearest neighbours were now the less-Romanized 
 Britons of districts like Cornwall and the foreign Celts of 
 Ireland and the north. These were weighty influences in 
 favour of a Celtic revival. And they were all the more 
 potent because, in or even before the period under discussion. 
 
 tin opening of the lifth century, a Celtic migration seems to 
 
 have set in from the Irish coasts. The details of this migra- 
 
 1 Much of the ornamentation used by post-Roman Celtic art comes 
 from Roman sources, in particular the interlaced or plaitwork, which 
 has been well studied by Mr. Romilly Allen, lint how far it was 
 borrowed From Romano-British originals and li<>\\ far from similar 
 Roman provincial \\<>rk on the c ontinent, is not very clear (see p. i.~>). 
 
CELTIC REVIVAL IN THE LATER EMPIRE 81 
 
 tion are unknown, and the faint traces which survive of it 
 arc not altogether intelligible. The principal movement was 
 that of the Seotti from North Ireland into Caledonia, with 
 the result that, once settled there, or perhaps rather in the 
 course of settling there, they went on to pillage Roman 
 Britain. There were also movements in the south, but 
 apparently on a smaller scale and a more peaceful plan. 1 
 At a date given commonly as a. d. 265-70 — though there 
 does not seem to be any very good reason for it — the Dessi 
 or Deisi were expelled from Mcath and a part of them settled 
 in the south-west of Wales, in the land then called Demetia. 
 This was a region which was both thinly inhabited and 
 imperfectly Romanized. In it fugitives from Ireland might 
 easily find room. The settlement may have been formed, 
 as Professor Bury suggests, with the consent of the Imperial 
 Government and under conditions of service. But we are 
 entirely ignorant whether these exiles from Ireland num- 
 bered tens or scores or hundreds, and this uncertainty 
 renders speculation dangerous. If the newcomers were few 
 and their new homes were in the remote west beyond Car- 
 marthen (Maridunum), formal consent would hardly have 
 been required. Other Irish immigrants probably followed. 
 Their settlements were apparently confined to Cornwall and 
 the south-west coast of Wales, and their influence may easily 
 be overrated. Some, indeed, came as enemies, though per- 
 haps rather as enemies to the Roman than to the Celtic 
 elements in the province. Such must have been Xiall 
 of the Nine Hostages, who was killed — according to the 
 
 1 Rhys, Cambrian Archaeol. Assoc, Kerry Meeting, 1891, and Celtic 
 Britain (ed. 3, 1904, p. 247), minimizes the invasions of southern 
 Britain (Cornwall and Wales). Bury (Life of St. Patrick, p. 288) 
 emphasizes them : sec also Zimmer, Nennius Vindicatus, pp. 84 foil., 
 
 and Kuno Meyer, Cipnnirodorion Translations. 18JK"j-(J, pp. .*>."> loll. 
 The decision of the question seems to depend upon whether we should 
 regard the Goidelic elements in western Britain as due in part to an 
 original Goidelic population or ascribe them wholly to Irish immigrants. 
 At present, philologists do not seem able t<» speak with certainty 
 on this point. Hut the evidence for some amount of invasion seems 
 adequate. 
 
 1731 1 
 
82 
 
 THE SEQUEL. THE CELTIC REVIVAL IN 
 
 traditional chronology -about a. i>. M)5 on the British coast 
 
 and perhaps in the Channel itself. 
 
 All this must have contributed to the reintroductiou of 
 
 Celtic national feeling and culture. A Celtic immigrant, it 
 
 may be, was the man 
 who set up the Ogam 
 pillar (Fig, 27), which 
 was discovered ;it Sil- 
 chester in the excava- 
 tions of 189:3. ' The 
 circumstances of the 
 discovery show that tins 
 pillar belongs to the \ cry 
 latest period in the his- 
 tory of Calleva. Its in- 
 scription is Goidelic : 
 t hat is, it does not be- 
 long to the ordinary 
 Callevan population, 
 which was presumably 
 Brythonic. It may be 
 best explained as the 
 work of some western 
 Celt who reached Sil- 
 chester before its British 
 citizens abandoned it. 
 We do not know the 
 date of that abandon- 
 ment, though we may 
 
 conjecturally put it before, and probably a great many years 
 
 before, a. i). .500. In any case, an Ogam monument had 
 
 1 ArchaeotogiOi Iiv. -j:;:;. in ; Rhys and Brynmoi Jones, Welsh 
 People, pp. 15, <;.-> ; ]'i<t>>ria llisf. of Hampshire ', i. 279 ; English Hist. 
 Review, \i\. 628 Whether the man who wrote was [risfa or British, 
 depends on the answer t<> the question set forth in the preceding note. 
 Unfortunately, we do not know when the Ogam Bcript came ftrsl int o 
 use. Professor Rh$s tells me that the Silchester example may quite 
 conceivably belong to the fifth century. 
 
 Fig. 27. Ogam Inscription 
 from Silchester. 
 
THE LATER EMPIRE 83 
 
 been set up before it occurred, and the presence of such an 
 objeet would seem to prove that Celtic things had made their 
 way even into this eastern Romanized town. 
 
 II. But a more powerful aid to the revival may be found 
 in another fact — the destruction of the Romanized part of 
 Britain by the invading Saxons. War, and especially defen- 
 sive war against invaders, must always weaken the higher 
 forms of any country's civilization. Here the assailants 
 were cruel and powerful, and the country itself was some- 
 what weak. Its wealth was easily exhausted. Its towns 
 were small. Its fortresses were not impregnable. Its 
 leaders were divided and disloyal. Moreover, the assault 
 fell on the very parts of Britain which were the seats of 
 Roman culture. Even in the early years of the fourth 
 century it had been found necessary to defend the coasts 
 of East Anglia, Kent, and Sussex, some of the most thickly 
 populated and highly civilized parts of Britain, against tin 
 pirates by a series of forts which extended from the Wash 
 to Spithcad, and were known as the forts of the Saxon Shore. 
 Sixty or seventy years later the raiders, whether English 
 seamen or Picts and Scots from Caledonia and Ireland, 
 devastated the coasts and even the midlands of the 
 province. 1 When, in the fifth century, the English came, 
 no longer to plunder but to settle, they occupied first the 
 Romanized area of the island. As the Roma no-Britons 
 retired from the south and east, as Silchestcr was evacuated 
 in despair, 2 and Bath and Wroxeter were stormed and left 
 desolate, the very centres of Romanized life were extin- 
 guished. Not a single one remained an inhabited town. 
 
 1 About a. n. 4-05 Patrick was carried oil from Bannavem Taberniae. 
 
 It tins represents the Romano-British village on Watting Street called 
 Bannaventa, near \Vhilton in Northants (Vict. Hist. Northante, i. 18G), 
 the raids must have covered all the midlands : see Engl. Hint. Review, 
 1895, p. 711 ; henee Zimmer, Realene. fur protestantisehe Theol. x. 
 ( 11)01), Art. ' Keltische Kirehe ' ; Bury, Life of St. Patrick. |>. 822. There 
 arc, however, many uncertainties surrounding this question. 
 
 2 Engl. Hist. Review, xix. C25 ; Vict, Hist. Hampshire, i. 871-2 ; 
 Vict. Hist. Shropshire, i. 217. 
 
 If 2 
 
84 THE SEQUEL, THE CELTIC REVIVAL IN 
 
 Destruction fell even on Canterbury, where the Legends tell 
 of intercourse between Briton and Saxon, and on London. 
 where ecclesiastical writers fondly place iifth- and sixth- 
 century bishops. Both sites lay empty and untenanted for 
 many years. ( )nly in the far west . at Exeter <>r at Caerwetit . 
 does our evidence not more or less forbid us to guess at 
 a continuing Romano-British life. 
 
 The same destruction came also on the population. 
 During the long series of disasters, many of the Romanized 
 inhabitants <>f the lowlands must have perished. Many 
 must have fallen into slavery^ and may have been sold into 
 foreign lands. The remnant, such as it was. doubtless 
 retired to the west. But, in doing so. it exchanged the 
 region of walled cities and civilized houses, of city life and 
 Roman culture, for a Celtic land. No doubt it attempted 
 to keep up its Roman fashions. The writers may well be 
 correct who speak of two conflicting parties, Roman and 
 Celtic, among the Britons of the sixth century. But the 
 Celtic element triumphed. Gildas, about a.d. 540, describes 
 a Britain confined to the west of our island, which is very 
 largely Celtic and not Roman. 1 Had the English invaded 
 the island from the Atlantic, we might have seen a different 
 spectacle. The Celtic element would have perished utterly : 
 the Roman would have survived. As it was. the attack fell 
 on the cast and south of the island — that is. on the lowlands 
 of Britain. Safe in its western hills, the Celtic revival had 
 full course. 
 
 It is this Celtic revival which can best explain the history 
 of Britannia minor, Brittany across the seas in the western 
 
 How much of Britain was still British when Gildas wrote, he da - 
 HOl tell us. Hut he mentions only the extreme west (Damnonii, 
 Demetae) ; his atmosphere is Celtic, and his rhetoric contains no 
 reference to a flourishing civilization. We ma\ conclude thai the 
 Romanized part of Britain had been losl i>> his time, or that, if some 
 of it was still British, long war had destroyed its civilization Unfor- 
 tunately, we cannot trust the traditional Knfflish chronology of the 
 period. For the date of Gildas, see W. II. Stevenson. Academy, 
 October *2<'>. 1805, ftc. : I see no reason to put either Gildas or any 
 part of the Epistula later than aboul 540 
 
THE LATER EMPIRE 85 
 
 extremity of Gaul. How far this region had hem Romanized 
 
 during the first four centuries seems uncertain. Towns were 
 scarce in it, and country-houses, though not altogether 
 infrequent or insignificant, wire unevenly distributed. At 
 some date not precisely known, perhaps in the middle of the 
 third century, it was in open rebellion, and the commander 
 of the Sixth Legion, which was stationed at York, one 
 Artorius Justus, was sent with a part of the British garrison 
 to reduce it to obedience. 1 It may therefore have been, as 
 Mommsen and Jullian think, one of the least Romanized 
 corners of Gaul ; in it the native idiom may have retained 
 unusual vitality. Yet that native speech was not strong 
 enough to live on permanently. The Celtic which is spoken 
 to-day in Brittany is not a Gaulish but a British Celtic ; it 
 is tlie result of British immigrants. This immigration is 
 usually described as an influx of refugees fleeing from 
 Britain before the English advance. That, no doubt, was 
 one side of it. But the principal immigrants, so far as we 
 know their names, came from Devon and Cornwall,* 2 and some 
 certainly did not come as fugitives. The King Riotamus who 
 (as Jordanes tells us) brought 12,000 Britons in a. d. 170 to aid 
 the Roman cause in Gaul, was plainly not seeking shelter 
 from the English. 3 We must connect him, and the Pifth- 
 
 1 C. iii. 1919 = Dessau 2770. The inscription must be later than 
 (about) A.i). 200, and it somewhat resembles another inscription (C. iii. 
 :J228) of the reign of Gallienus, which mentions milites vexill. legg. Ger- 
 manicianar. et Brittanicin. cum auxiliis earutn. Presumably it is either 
 earlier than the Gallic Empire of 258-7:3, or falls between that and the 
 revolt of Carausius in 287. The notion of O. Fiebiger (l)e ekurium 
 Italicarum historia, in Leipziger Sludien, w. 304) that it belongs to the 
 Aremoric revolts of the fifth century i-. I think, wrong. Such an 
 expedition from Britain at such a date i> incredible. 
 
 2 The attempt to find eastern British names in Brittany seems 
 a failure. M. de la Borderie, for instance, thinks that Corisopitum (or 
 whatever the exact form of the name is) was colonized from Cor- 
 stopitum (Corbridge on Tyne). But the latter, always to some extent 
 a military site, can hardly have sent out ordinary emigres, while the 
 former has hardly an historical existence at all, and may be an ancient 
 error for civitas Coriosolilutn (C. xiii (1), p. 191). 
 
 3 Freeman {Western Europe in the Fifth Century, p. Kit) suggested 
 
s<; THE SEQUEL, THE CELTIC REVIVAL IN 
 
 century movement of Britons into Gaul, with the Celtic 
 revival and with the same causes that produced, for instance, 
 the Scotic invasion of Caledonia. 
 
 This destruction of Romano-British life produced a result 
 which would he difficult to explain if we could not assign it 
 to this cause. There is an unmistakable gap between the 
 Romano-British and the later Celtic periods. However 
 numerous may he the Latin persona] names and 'culture 
 words " in Welsh, it is beyond question that the tradition 
 of Roman days was lost in Britain during the fifth or earlv 
 sixth century. That is seen plainly in the scanty literature 
 of the age. Gildas wrote about a.i>. 540, three or four 
 generations after the Saxon settlements had bcyun. He 
 was a priest, well educated, and well acquainted with Latin, 
 which he once calls nostra lingua, lie was also not unfriendly 
 to the Roman party among the Britons, and not unaware 
 of the- relation of Britain to the Empire. 1 Yet he knew 
 substantially nothing of the history of Britain as a Roman 
 province. He drew from some source now lost to us- — 
 possibly an ecclesiastical or semi-ecclesiastical writer — some 
 
 that a migration of Britons into (.aul had been in progress, perhaps 
 since the days of Magnus Maximus. and that l>\ 170 there was a regular 
 British state on the Loire, from which Riotamus lid his [2,000 men. 
 Hodgkin (Cornwall and Brittany, Penryn, uni) thought that the 
 soldiers of Maximus settled on the Loire about 888, and that Riotamus 
 \sa^ one of their descendants. He quotes Gildas as saying thai the 
 British troops of Maximus went abroad with him and never returned. 
 That, however, is a different thing from Baying that they settled in 
 a definite part of Gaul. For this latter statement I can find no evidence, 
 and the Celtic revival in our islands seems t<> provides better setting 
 for EUotamus. 
 
 If Professor Bury is right [Life of St. Patrick, p. 854), EUotamus had 
 a predecessor in Dathi, who is said to have gone from Ireland to (.aul 
 about \.i>. i-s to help the Romans and Aetius. Zimmei (Nennitu 
 Vind. t p. 85) rejects the tale. Hut it tits in well with the Celtic revival. 
 
 1 Mom ins. n. i'it face to Gildas (Mon, liciin. Hist.), pp. 9-10. Gildas 
 is. however, more Celtic in tone than Mommsen seems to allow. Such 
 
 i phrase as Ha at ana Britaaaia s,d Romania CetiSeretur implies a eon- 
 
 Bciousness of contrast between Briton and Roman. Freeman (Western 
 
 Europe, p. 1">) puts the ease tOO Btrongly the other way. 
 
THE LATER EMFlftE 87 
 
 details of the persecution of Diocletian and <>r the career of 
 Magnus Maximus. 1 For the rest, his ideas of Roman history 
 may be judged by his statement that the two Walls which 
 defended the north of the province — the Walls of Hadrian 
 and Pius — were built somewhere between A.D. 888 and IK). 
 Tie had some tradition of a coming of the English about 
 450, and of a reason why they came. Hut his know- 
 ledge of anything previous to that event was plainly mosl 
 imperfect. 
 
 The Historia Brittonum, compiled a century or two later, 
 preserves even less memory of things Roman. There is 
 some hint of a veins traditio seniorum. But the narrative 
 which professes to be based on it bears little relation to the 
 actual facts ; the growth of legend is perceptible, and even 
 those details that are borrowed from literary sources like 
 Gildas, Jerome, Prosper, betray great ignorance on the pari 
 of the borrower. 2 We have got here a very long way 
 beyond Gildas. He, after all, knew something of Maximus 
 and understood (however dimly) the relation of Britain to 
 Rome. The ' Historia ' goes altogether astray on both 
 points. On the other hand, the native Celtic instinct 
 is more definitely alive and comes into sharper contrast 
 with the idea of Rome. Throughout, no detail occurs 
 which enlarges our knowledge of Roman or of early post- 
 Roman Britain. 
 
 The same features recur in later writers who might be, or 
 have been, supposed to have had access to British sources. 
 Geoffrey of Monmouth — to take only the most famous- 
 
 1 Magnus Maximus, as the opponent of Theodosius, seems i«> have 
 been damned by the Church writers. Compare the phrases of Orosius, 
 vii. 85 (Theodosius) posuil in Deo spent siiain seseque adverstu Minimum 
 tyrannum sola fide maior proripuit and ineffabili iudicip Dei and Theo- 
 dosius victoriam Deo procurante suscepil. 
 
 - The story of Vortigern and Ilengist now first occurs and is obvious 
 legend. A prince with a Celtic name may have ruled Kent in 150. 
 There were, indeed, plenty of rulers with barbaric names in the fourth 
 and fifth centuries of the Empire. Bui the talc cannot be called 
 history. 
 
88 CELTIC REVIVAL IX THE LATER EMPIRE 
 
 asserts thai he used a Breton book which told him all manner 
 of facts otherwise unknown. The statement is by no means 
 improbable. But, foi all that, the pages <>f Geoffrey contain 
 
 no new fact about the first five centuries, which is also true 1 
 From first to last, the Celtic tradition preserves no real 
 remnant of recollections dating from the Romano-British 
 age. Those who might have handed down such memories 
 had either perished in wars with the English or sunk back 
 into the native environment <>f the west. 2 
 
 Bui we are moving in a dim land ofdouhts and shadows. 
 lie who wanders here, wanders at his peril, for certainties 
 are few, and that which at one moment seems a fact, is likely, 
 as the quest advances, t<> prove a phantom. It is. ton. 
 a borderland, and its explorers need to know something of 
 the regions on both sides of the frontier. I make no claim 
 to that double knowledge. I have merely tried, using such 
 evidence as I can. to sketch the character of oik n gion, 
 that of the Romano-Brit ish civilization. 
 
 1 Tims. he refers to Silehester, ami so good a judge as Stubbs once 
 
 suggested that lot- this lie had some authority now lost to us. Vet the 
 turn- fact thai Geoffrey knows only the English name Silehester dis- 
 
 proves this idea. Had lie used a genuinely ancient authority, he would 
 
 have (as in other eases) employed the Roman name. Another explana- 
 tion may he given. Geoffrey wrote in an antiquarian age, when the 
 
 ruins of Roman towns were being noted. JJoth he and Henry ol 
 Huntingdon seem to have heard of the Silehester ruins, and hot h 
 accordingly inserted the place into their pages. 
 
 The English mediaeval chronieles have sometimes been thought 
 to preserve racts otherwise forgotten about Roman times. So far as 
 I can judge, this is not the ease, even with Henry of Huntingdon. 
 Henry, in the later editions of his work, borrowed a few facts from 
 
 Geoffrey of Monmouth, which are wanting in his first edition (see the 
 Eiengwrt and All Souls MSS. ; the truth is obscured in the Rolls Series 
 text, as I have pointed out. Athenaeum, April (i, 1901). He also 
 preserves one local tradition from ( olchestei : otherwise in- contains 
 nothing which need puzzle any inquirer. Giraldus Cambrensis, when 
 
 at Home, saw some manuscripl which Contained a list of the live pro. 
 
 vinces of fourth-century Britain otherwise unknown throughout the 
 Middle \u r <s [Archaeol. Oaconirnsis 1804, p. 224). 
 
I NDEX 
 
 Aldborough (Isurium Brigantum), 24, 89, 01 note, 70. 
 Arret ine ware, 19 ; in Britain, 74. 
 Avolis on moulds of Gaulish potters, 31. 
 
 Bannavem Taberniae and St. Patrick, 83. 
 Hath, 61, 75 ; the Bath Gorgon, 53. 
 Brislington (near Bristol), house at, 39. 
 Brittany, British migration to, 84. 
 
 Caerwent (Venta Silurum), 24, 42, 70 ; a cantonal capital, 00. 
 Canterbury, derivation of name, 00 ; deserted after the Human 
 
 period, 84. 
 Cantonal system in Gaul, 20 ; in Britain, 58 foil. 
 Carausius, birthplace, 78. 
 Castor ware, 49 foil. 
 Celtic art, 48, 80 note. 
 
 Celtic languages used in Gaul and Britain, 18, 31 foil. 
 Celtic type of temples, 37 ; of houses, 38 foil. 
 Christianity, its attitude to native languages, 19. 
 Clanville, house at, 39. 
 Cloth made in Britain, 77. 
 Coloni (rural), in Britain, 05. 
 Coloniae (municipalities), 15 ; in Britain, 57. 
 Corbridge, 72, 85 ; the Corbridge Lion, 53. 
 Corn exported from Britain to the Continent, 77. 
 Cornwall. Roman remains in, 24 note. 
 
 Deae mat res, 70. 
 
 Demetrius of Tarsus at York, 34. 
 
 Dessi (Deisi) migrate from Ireland to Wales, 81. 
 
 Dens Jeter, di veteres, perhaps Teutonic deities, 7J. 
 
 Devonshire, Roman remains in. 24 note. 
 
 Din Lligwy (Anglesea), village at, 46. 
 
 Dragon brooches, 52. 
 
 Emigration from Italy into the provinces, 10. 
 
 Frilford (Berks), house at, 39, 47. 
 
 Gaulish kingdom of A.D. 258-73, 17. 
 
 Gaulish language used under the Empire, 18. 
 
 Geoffrey of Monmouth, 87. 
 
 Gildas, 84, 80. 
 
 Glastonbury, pre-Roman lake-village near. 55. 
 
 Gloucester, colonia, municipal tiles, .vr, 03 ; sculpture of Mercury 
 
 found at, 73. 
 Goidelic elements in Britain, 81; 82. 
 Gorgon at Bath. 53. 
 
90 INDEX 
 
 llcm\ oi Huntingdon, Hengwrt and All Souls MSS., 88 note. 
 Hesione and Hercules, 51. 
 Hiatoria Brittonum, 87. 
 
 Houses in Etonian Britain, their varieties, their relation to houses in 
 north Gaul, Italy, &t\, 'tf foil. 
 
 Icinos, bribe-name perhaps used of chief town, 60. 
 [ckleton (Cambridgeshire), graffito at, -i'-i. 
 Imperial domains in Britain. fit. <».->. 
 
 Jerome (St.), cited, 18. 
 
 Kent, derivation of name, :j5. 
 
 Late Celtic art, 48. 
 
 Latin used in the- provinces, 18 ; in Britain, 2!». 
 Leicester, graffito I rum. :}'2. 
 
 Leugae in Gaul, 17. 
 Lincoln, 57. 
 
 London — ■ 
 
 Pre-Roman inhabitation, 7 t, noU. 
 
 Size, 62. 
 
 ConstitutioD of town. 62. 
 
 Latin spoken in. :j'J. 
 
 Deserted alter the Roman period, 81-. 
 
 Magnus Maximns, late of his army, 86 note 
 
 Mars in Roman provincial religion, 69. 
 
 Mars Lenw rive Ocelus, 70. 
 
 Mercury in Roman provincial religion, 69. 
 
 Mercury and Rosmerta, 21, t:;. 
 
 Mithraism, distribution in western Europe and Britain, ?ii. 
 
 Mosaic floors in Roman Britain, 44. 
 
 Nettleton (on the Fosse), shrine of Diana, ?."{. 
 N( w Const ware, 48. 
 
 Xodcns, Celtic deity of Lydney, 70, ?". 
 Northleigh (Oxon), house at, 41, 76. 
 
 Ogam at Silchester, 82. 
 
 Oriental worships in Britain, ?•_». 
 
 Orpheus on mosaic floors, not Christian, 44. 
 
 Pergamene style in the Roman provinces, 5f //"/' 
 Pitt-Rivers, excavations by, 45, .*>.">. 
 Plaxtol (Kent), inscribed tiles at. :;:;. 
 
 Pompeian houses compared with British, 40, 42 /mlr. 
 Punic language, used in Roman Africa, is. 
 
 Ravenna <•< ographer, 60. 
 Religion, 'Jl ; in Britain. 68. 
 Riotamus, British chief in Gaul, B5. 
 
 Samian ware, in. 5 i . 
 
 S< el)ohm"s theory Of the BUffix * ham '. 65 note. 
 
 Silchester — 
 
 Name. (i| . 88 /"(/- ■ 
 
 Pre-Roman, 74. 
 
INDEX 91 
 
 Silchester — 
 
 Imperial domains under Nero at, 05. 
 
 Development as Romano-British town, 75 nolc. 
 
 Houses in, 42, 64. 
 
 Latin used in, inscribed tiles, 29 foil. 
 
 Temples of, 37. 
 
 Town-planning of, 43, 64. 
 
 Dyeing works in, 77 nolr. 
 
 Abandoned, 83. 
 
 Tanarus, supposed Celtic god, 69. 
 Temples in Britain, 36, 62, 73. 
 Town-planning in Roman Britain, 64. 
 Towns of Roman Britain, 57-65. 
 
 Veter [Vheter), di veteres, 71. 
 
 Vergil, tags from, known at Silchester, 30. 
 
 Verulamium, municipium, perhaps pre-Roman town, 57, ?J. 
 
 Villages in Roman Britain, 45, 55. 
 
 Yinogradoff, 35, 66. 
 
 Yortigern and Hengist, 87. 
 
 Wales, Roman, 24, 46, 81. 
 Warwickshire, few Roman remains in, 27. 
 Wroxeter (Viroconium Comoviorutn), 24, 37, 61. 
 
 York, cofonid and fortress. 26, 57. 
 
OXFORD: HORACE HART 
 l'RINTER TO THE UNIVERSITY' 
 


 HOME USE 
 CIRCULATION DEPARTMENT 
 n. . . MAIN l, BRARY 
 
 This book ls due th 
 
 'month loans may be ren.wi ! * S,am,,e,, •"»<>*• 
 6-month loans may he reoS e Alf me . 642 '3105. 
 Renewals and * C, ™&* ' »"»«'»« ><* 
 
 Zl ^S— ^spnor 
 
 AU BOOKS flRF <jmr.cpt T ; 
 
 J976 
 
 KB. CO. 
 
 *77;;t, 
 
 ... General Library 
 University of Califo 
 Berkeley 
 
 fornia 
 
 &- 
 
U. C. BERKELEY LIBRARIES 
 
m 
 
 Us? 
 
 J iilrU 
 
 til' 
 
 m 
 
 
 tb 
 
 ■.WnafyfSBto' 
 
 nil 
 
 ■ ■>■-, 
 
 .-*.™^':r^a^i«cfctt ■ aw? 
 
 CTCtJOOtM 
 
 ■■-■".■■■■