THE RACES OF IRELAND AND SCOTLAND THE RACES OF IRELAND AND SCOTLAND BY W. C. MACKENZIE, F.S.A. (Scot.) AUTHOR Ol HISTORY Ol THE OiriER HEBRIDES," "A SHORT HISTORY Ol' THE SCOTTISH HIGHLAXnS," "LIFE AND TIMES OF SIMON ERASER, LORD LOVAT," ETC. PAISLEY: ALEXANDER GARDNER ^nbltchtr be ^ypointment to the l»tt Qnrm Vittons London SiMPKiN, Marshall, Hamilton, Kent & Co., Ltd, {iiy BOSTON COi.^-— ^ ESESTNUT &^» ^ASS. APR ■> 8 iS73 printed by ALEXANDER GARDNER, PAISLEY PREFACE. There are few subjects of ethnological inquiry surrounded by greater obscurity than the origin of the different races inhabiting Ireland and Scotland. A sharp conflict of theory, and a remarkable lack of definiteness, are the main char- acteristics of the discussion that centres around the subject. Any fresh views, therefore, that rest upon the foundation of scientifically ascertained facts, must be regarded as valuable aids in the solution of an admittedly difficult problem. It was with these considerations before me that I com- menced, some years ago, to study the race problem of the two countries — for they are inseparable — and embarked upon a course of independent research. There Mas room for the pursuit of original work. The confusion in which the subject is involved was rendered not less, but more, pei'plexing by a succession of treatises upon parallel lines, and all leading to no certain conclusions. In the solution of the race-problem, there is no evidence, in my opinion, equal in weight to the proofs supplied by the early forms of ancient place-names. I rest my case largely upon etymologies. They supply the most tangible evidence that it is possible to produce. Place-names cannot lie. Provided the right key can be found to unlock the treasures, they yield the pure gold of truth. But with a false key one can only fumble ; one cannot open. The application of place-names to the solution of Irish and, particularly, Scottish racial questions has been rendered largely nugatory by the method employed. Etymologists have approached the subject with their minds made up. VI. PREFACE. " Here,'" they have said, " all the names must be Gaelic ; yonder they must be Cyiniic ; in this district only Anglo- Saxon roots can be looked for ; in that, only Scandinavian." That being their attitude of mind, they have constructed Procrustean beds in which the names have been made to fit preconceived notions. I could give many instances of this method of treating both Irish and Scottish etymologies, and it may be confidently asserted of the result that it has hitherto proved a hindrance rather than a help to the study of ethnology. I do not say that my etymologies are infallibly correct ; far from it. But I do say that the names have been studied on their merits, and that my derivations are based alike upon commonsense and the facts of topography. The main theories advanced in this book are entirely new, and, if I may use the word without fear of being misunder- stood, entirely " revolutionary." I am not so sanguine as to suppose that they will meet with complete acceptance, nor so confident as to believe that they are impervious to criticism. But I have made no important assertions without supple- menting them by reasoned proofs that have satisfied me, whether or not they seem equally conclusive to others. The tests that I have applied are severe. I have dealt at some length with the legends pertaining to the race-origins, particularly in Ireland, and have endeavoured to reconstruct from them a theory of the prehistoric races, concerning whom expert opinion has not yet settled the elementary question whether they were men or myths. I have tried as far as possible to separate myth from tradition ; to penetrate the meaning of the former, and to gauge the historic values of the latter. Necessarily, this section of the subject is largely speculative, but when the speculations are in agreement with the ascertained facts of anthropology and archaeology, they are entitled to rank as working hypotheses until they are superseded by more exact knowledge. Finally, the subject of this book bristles with controversial points, and covers ground that, for its adequate exploration, PREFACE, Vll. would fill a number of volumes. I have ventured to touch a good many of these points in passing, without lingering to discuss them fully, which would have been impossible. It will be seen that the scope of the work, embracing, as it does, mythology and folk-lore, history and tradition, ety- molog}^ and anthropology, is varied and exacting. I haAC tried to make myself as clear as possible in expressing my views, even at the risk of being charged with redundancy, and I cherish the hope that whatever the faults of the work, obscurity of meaning is not one of them. If my treatment of the subject has the result of directing research into new channels for discovering the beginnings of two great and intimatel}' associated peoples, I shall feel that my labour has not been in vain. W. C. MACKENZIE. RiC'iuroxn-ox-TiiAJiES. CONTENTS. CHAPTER I., .... - The legends of Ireland and their interpretation — The Book of Invasions— Cesair — Partholan— The Nemidians — The Fomorians— Were the Fomorians Phoenicians?— Cormac's Glossary— A discussion of Beltine— The idol Crora Criiaich— The serpent-mound near Oban —The Irish " dragons "—St. Patrick as a serpent-destroyer— Pesti- lence symbolised by a serpent — The significance of cromlechs. CHAPTER n., - - - - - 17 The Firbolgs— The traditional story of their origin— The etymology of the name — A theory to explain the name — The "long-heads" of Ireland — Huxley's pregnant suggestion- The Firbolgs identified by tradition with the "Mediterraneans" — Moytura, the " heap plain " — Giants and gods — The overthrow of the Firbolgs by the Tuatha de Danann. CHAPTER III., ... - . 27 The Tuatha de Danann — The country of their origin— An account of their wanderings —The Dagda — Keating on the Dananns— The meaning of the name — The Dananns as magicians — The Fir-Sidh — Their Lapponic origin discussed — The Euhemerist theory — The Skraelings of the Norse Sagas— The Pigmies' Isle— The custom of the "knotted cord" — Selling winds in Lewis, the Orkneys, Shet- lands, and the Isle of Man — Comparetti on Shamanism. CHAPTER IV., .... - The Lapponic theory further discussed— Disproved by anthropological evidence — MacFirbis on the Dananns^ The Irish texts on the Dananns— The elves of light and the elves of darkness— St. Patrick and elf-worship — The elf-creed introduced to Ireland by Scandin- avians—Folk-lore as an aid to ethnology— A classification of the Teutonic elves— The application of elf-beliefs to the Dananns— Parallels between the Dananns and Scandinavian mythology — Thorpe on the resemblances between Teutonic and Celtic folk-lore. CHAPTER v., .... - Druidism and its significance — Druidism and the Dananns— The FffliV- Stones of Fate— An Icelandic example— The Ogam Script— Illusionism — Scottish examples of the practice of the Sian — High- land belief in the eflicacy of charms— Dwarfs and hunchbacks— The Dananns identified with the Cruithen people of Ireland— The meaning of " Cruithne "— Cruithne, " the father of the Picts." X. CONTENTS. PAGE CHAPTER VI., . .... 65 The Milesians — The two tales of Irish origins— Gadel and Scota — The stories post-Patrician— The Scottish version — Scythian and Scot — The vagueness of the name "Scythia" — Nennius on the progenitor of the Scots — The Pictish Chronicle on the Scythians and the Goths — Their common descent from Magog — How the confusion between the Goths and the Scythians arose — The Lombards and the Gael — Conclusions deduced from the evidence. CHAPTER Vn., . .... 73 The Celts — The different types of Celt— The succession of races in Western Europe — The Celtae and the Galli — A discussion of the names — The Belgae — The two branches of the Celts — Where did the Gaelic language originate ? CHAPTER Vni., ----- 82 The four stocks of the Gael — The Irish genealogies and their value — The historical aspect of the Milesian legend — Spain and the Milesians —The system of the IXmnsenchus — The different names applied to Ireland— An explanation of the Milesian names — The Basques or Vascones — A Basque element in the population of Ireland — The location of the Milesian tribes. CHAPTER IX., . - ... 90 The Iberians in Ireland — The origin of the Scots — A summary of con- clusions as to the origin of the Gael — The earliest notices of Ireland by classical authors — Ireland in the second century a.d. — Earlj' Teutonic settlements in Ireland — The earliest mention of the Scots — The Scottish hegemony in Ireland — Tacitus and Ireland — The Cherusci and the Scots. CHAPTER X., . _ - - - 99 Various hypotheses concerning the name " Scot " — Isidore's blunder — Geoffrey of Monmouth and his value as an historian —The Hibernians and the Scots — An analysis of the name "Scot" St. Patrick's distinction between the Scots and the Hibernians — Ireland indiffer- ently named Scotia and Hibernia— The Ard-rhjli of Tara — Ireland's Heroic Age — A dissertation on hair — Irish kings with Teutonic names — The Franks in the British Isles — The kilt as a Gothic dress. CHAPTER XI., - - - - - 110 The Gael— The silence of early writers on the name — Bede's evidence on the root dal—Kn analysis of the name "Gael" -The Brehon Laws and Teutonic parallels — CuchuUin : man or myth ? — The Finn Saga and its historical basis— The Flanna as professional champions — Scandinavian parallels — The dominant races described by the Senchus llur— The meanings of the provincial names. CONTENTS. XI. PAGE CHAPTER XII., - .... 121 The Gael and the Gaelic language in Ireland— How the Gaelic language was formed — St. Patrick and education in Ireland— Tradition and the ancient tongue of Ireland— Ah(/etoria — The Latin element in Gaelic — Ptolemy's map of Ireland — An analysis of the Ptolemaic names in Ireland— The general structure of the Gaelic language- Some Scandinavian legacies — The views of Dr. Joyce — Bishop MacCarthy on the Irish Picts. CHAPTER XIII., . - - . . im Antiquaries and the Picts — The different schools of theorists — The Cruithne, the Irish Picts, and the Picts of Scotland — Tighernach and the Piccardach — The meaning of '* Picars " — The Roman Picti — Picti a corrupt form of a pre-existing name — The Picts as pigmies — Sir Walter Scott on the Orcadian " Peghts " — Picts-Houses — The Picts and the elf-creed of the Teutons— The meaning of the name "Picf — Confusion between elves and Picts — Beddoe on Ugrian thralls of the Norsemen — Finn-men and Finn-women. CHAPTER XIV., ... - - iM) The various names of the Irish Picts — Rury the Great — The Golden Age of the Irish Picts — The Red Branch Knights — " Ossian " Mac- Pherson and the Irish bards — The meaning of the Irish Creei'es — The destruction of Emania — The racial affinities of the Ulster Picts — The solitary word of their language analysed — The " Danes' Cast." CHAPTER XV., . - . . - U9 The historical Picts — The Maitai and the Vecturiones (or Verturiones) — How the Picts got their name — Teutonic parallels — The " men of the elves" — Were the Picts tattooers? — Historical notices of the Picts : Herodian, Solinus, Dion Cassius— The sources of their infor- mation examined — Tacitus on the Caledonians — Shield-painting — The Pictones of Poitou. CHAPTER XVI., . . - . - i:,7 A summary of the racial ai-gument as applied to Ireland — An analysis of prefixes in Irish place-names — What the analysis proves — Anthropology and archaeology in relation to the argument — A French analogy — The Anglo-Saxon settlements in England on a different footing from the Teutonic settlements in Ireland — The composition of the English language compared with that of Gaelic — The Saxon and the Gael — The evolution of the Gaelic language — Peculiar Gaelic characteristics. CHAPTER XVII., - - - - - 182 Scotland and its legendary matter — The earliest name of Scotland — The significance of the name "Alban"— The invasion of Scotland by Agricola Who were the Caledonians ? — Galgacus or Calgacus — The Caledonian tribes self-contained units — The phj'sical features of their country — An examination of Caledonian ethnology— An analysis of the place-names mentioned by Tacitus. XU. CONTENTS. PAGE CHAPTER XVIII., ----- 192 River-names and their value — Mountain -names and their value — Ptolemy's place and tribal names in Scotland analysed. CHAPTER XIX., ... - . 209 Conclusions to be drawn from the analysis of Ptolemaic names in Scotland — The first clear view of the Pictish monarchy in Scotland — Bede on the origin of the Picts — The Anglo-Sax^oit, Chronicle and the Picts — The two divisions of the Pictish nation — The Irish traditions of the origin of the Picts— The probable sources of these traditions — The versions of the Pictish Chronicle and Nennius — Claudian on the Picts — Cymric and Scandinavian elements. CHAPTER XX., ----- 218 Gildas on the Picts — Bede on the Picts— The accounts in Geoffrey of Monmouth and Layamon — The Gaelic traditions of Pictish origins — Pictish settlements in Ireland and Scotland^The evidence of Giraldus — The Frisian settlement in Scotland — The Saxons in Scot- land — The different elements in the Scottish nation. CHAPTER XXL, ----- 228 The various theories about the Picts— The Gaelic theory as represented by Dr. Skene— The Cymric theory— The Gothic theory and John Pinkerton— Bede on the Pictish language -Sir John Rhys and the non-Aryan theory — The Pictish system of succession — Scandinavian parallels — An examination of Dr. Skene's arguments — Common elements in the Celtic and Teutonic languages— The Pictish language different from Anglo-Saxon, Cymric, or Gaelic. CHAPTER XXII., ----- 236 The Pictish words recorded by contemporaries — Scollofthes — Peanfahel —The names of the Pictish kings— The Drosten Stone and the meaning of its inscription - The incidence of languages - The dialects of modern Scots — The Pictish language the parent of modern Scots —The latter an indigenous language — How it differs from North- umbrian English — Frisian the dominant element of the later Picts — How the Pictish language became the national tongue of the Scots — The cleavage between the Pictish and the Gaelic languages. CHAFFER XXIIL, ----- 256 An analysis of Scottish river-names, mountain-names, and island- names. CHAPTER XXIV., - - - - - 274 An analysis of characteristic prefixes in Scottish place-names. CONTENTS. Xlll. PAGE CHAPTER XXV., - - - - - 293 An analysis of the oldest or most noteworthy of the provincial and town names of Scotland. CHAPTER XXVI., ----- 338 Conclusions to be drawn from the foregoing analyses — The earliest colonisation of Scotland from Ireland — A settlement of the Scots in Wales — The tradition in the Life of St. Cad7-6e— The Kingdom of Fife — The Dalriadic kingdom in Argyll — A Scottish settlement in Fife — The three sons of Ere — The extent of the Dalriadic sovereignty — The Northumbrians and the Scots — Fife an appanage of Dalriada — The relations between the Picts and the Scots — The nature of Kenneth MacAlpin's rights to the Pictish Crown. CHAPTER XXVII., ----- 350 The Romans and the Picts^The Attacots — St. Columba's mission to the Picts non-political — The Picts at Loch Ness — The Shamanism of the Picts — The Pictish monarchy on the banks of the Earn — The relations between the Picts and the Angles — The extent of the Anglic sovereignty over the Picts — The population of Lothian — The struggle for the possession of Lothian — The " Commendation of Scotland" — The English claims analysed— The cession of Lothian to Scotland— The Scottish victory at Carhani. CHAPTER XXVIIL, ----- 365 A concluding survey — The different strata of the population of Scot- land — The Britons of Strathclyde— The Northumbrian settlements in Lothian — The Anglo-Saxon, Anglo-Danish, and Norman adven- turers—The heterogeneous Scottish people — The eifect of segrega- tion — The amalgamation of the different peoples a gradual process — The "Scottish Conquest" — The decline in the Pictish power — The Kehdei and their influence — Scandinavian invasions prior to the eighth century — The Lochlans — Gael and Gall — Fingalls and Dugalls — The Gall-gaidel — The Danes and the downfall of the Pictish monarchy — The nature of the so-called Scottish Conquest — Kenneth MacAlpin as King of the Picts — Later allusions to the Picts— The Picts called "Galwegians" — The ancient divisions of Scotland — The Mormaers and Toisechs— The racial affinities of the Picts of Galloway as shown by Jocelyn's account — The incidence of Gaelic in the Lowlands— The cleavage between East and West — The Gaelic tribes in the West — The Clan Donald and their influence — The Gael of Scotland and their language called " Irish " — Racial characteristics in Scotland — The process of unification. Citations from Modf.hn Wohks, - - - 389 Index, - - - - - - -391 CORRECTIONS. Page 22 (15th line from top) : /or " distingushing " read "dis- tinguishing." 59 (4th line from bottom): for "F. H." read "J. H." Dixon. 90 (1st line of note) : for " Miss " read " Mrs." Bryant. 1 23 (1 8th line from top) : for "Flaherty" read "O'Flaherty." 131 (2nd line from bottom) : for " Ibernian" read "Iberian." 247 (1st line of note) : for " MacFirbig " read "MacFirbis." 288 (14th line from top) : for '-' Lcths" read " Leths." 332-3 : The etymology of the place-name " Stornoway " in Lewis, and Loch "Stornoway" in Argyllshire suggested in the text, is allowable only on the assumption that there has been a change of form by metathesis, an unsatisfactory solution of a topographical problem. I have now dis- covered in the Lambiama Book an ancient place-name which seems to supply the root that had previously eluded my search. It is the name Stiornu-sleinom , which is translated "Anchor Rock." The pronunciation of the word "Stornoway" by the Gaelic-speakers of Lewis strongly suggests that in this place-name an original Norse Stiornu has been preserved. By this reading " Stornoway " would mean " Anchor Bay," a name that might very well have been given to the safe natural harbour of Stornoway by the earliest Norse settlers in Lewis. (There is an "Anchor Head " between two bays in the Bristol Channel at Weston-super-Mare.) 346 (12th line from bottom) : for "to" read "of" „ (2nd line from bottom) : for "mystical" read "mythical." 360 (2nd line from bottom) : for "makes" read "make." THE RACES OF IRELAND AND SCOTLAND. CHAPTER I. The legends of Ireland and their interpretation — The Book of the Invasions — Cesair — Partholan — The Nemidians— The Fomorians — Were the Fomorians Phoenicians? — Cormac's Glossary — A discussion of Beltine— The idol Crom Cruaich — The serpent-mound near Oban — The Irish " dragons "—St. Patrick as a serpent-destroyer — Pesti- lence symbolised bj^ a serpent — The significance of cromlechs. The race problems of Ireland and Scotland are so closely intertwined as to be insej)arable. For it will be shown in the following pages that the people known as the Scots, who gave their name to Scotland, passed over to that country from ancient Scotia, the modern Ireland. The traditions and legends of these Irish settlers in ancient Alban (part of the modern Scotland) became the common inheritance of both countries, and form the connecting link in the chain that stretches forward to authentic Irish and Scottish history, and backward to traditions concerning the shadowy races Avho preceded the Gael in the occupation of Ireland. These races have provided Irish writers, more particularly, Avith plenty of scope for the exercise of ingenious guessing. The obscurity of the subject has stimulated rather than repelled persistent research. Yet it must be admitted that the result has been to envelope these prehistoric peoples in a more impenetrable mystery than ever. The prevailing I Z THE RACES OF IRELAND AND SCOTLAND. tendency of the present day is to dehumanise them; to treat them as myths; to read symbolic meanings into the records of them handed down by tradition ; or to regard them in part or in whole, not as races of real men and women who occupied Ireland before the Celts, but as a pantheon of Celtic gods and goddesses. This tendency is so contrary to the interpretations of mediaeval transcribers and commentators, that it can only be regarded as an alternative solution of the problem that has baffled investigation, or as an easy method of evading an admitted difficulty. In either case, it is not a convincing attitude. Any one endowed with a glimmer of imagination can construct a pantheon to suit his own fancy. It is not so easy to offer a sane ethnic theory that shall satisfy the requirements of modern science. It seems probable that the medisevalists have rationalised too much, and modern critics too little. What do the Irish traditions tell us about the ethnology of the country? These traditions, it is well to remember, must have been in their original form rhymed stories handed down orally by the shanachies, or professional historians, from generation to generation, from century to century, until they were clothed by monastic scribes in prose, after the art of writing in Roman characters had been acquired. Transcribed and redacted time and again by monks with views of their own, they appeared finally in the dress in w^hich we see them to-day, a dress the fabric of which was woven mainly in the eleventh and twelfth centuries. The nearer the source the purer the stream; and the same law applies to tradition. If the varnish which overlies these stories could be cleaned off, we should see the picture clearly and in its proper perspective. The deliberate emendations and the unintentional errors in which, necessarily, the traditions abound would then be obliterated, and there would be less reason perhaps for rioting in symbolism. It is the THE RACES OF IRELAND AND SCOTLAND. 3 accretion of legendary matter around the genuine traditions of the country that has concealed a good deal of the historical value, which, beyond doubt, the traditions intrinsically The Book of Leinster, a compilation of the twelfth century, contains, in common with later compilations, a record of the successive colonies that occupied ancient Ireland. The " Book of the Invasions," as this record is usually called, discriminates between the races who settled in the country and those who visited it for spoil. An account of each invasion or settlement is given with the tribal name, or the eponym, of the settlers. The etymology of these tribal names, or eponyms, has baffled philologists, and has thus added to the confusion of ideas in which the whole subject is immersed. Neither Irish nor Scottish Gaelic pro- vides an adequate key. But Cymric is of some help, and for reasons which will presently appear, Cymric is the language above all others that unlocks the door of obsolete Irish words and shows their original meaning. The first eponym that meets us is that of " Cesair," " a grand-daughter of Noah," who, with her company, arrived in Ireland — very conveniently before the Flood. It is use- less to speculate on the racial problem presented by this eponym, and even Irish antiquaries who accept the later Invasions as historical, dismiss Cesair as a myth. The word, or a similar one, is, however, used in Welsh bardic literature, where it denotes "lordship." Thus Cesair may well stand as the eponym of the earliest tribes who had dominion over Ireland . Equally nebulous are the second people who occupied Erin under the leadership of " Partholan." This eponym seems to mean land-sharers (Cymric Parthu, to divide). According to Keating, the first division of Ireland (he gives seven in all) was by Partholan, originally a Scythian, who came from Greece. He is said to have divided the country into 4 THE RACES OF IRELAND AND SCOTLAND. four parts. The tradition tells us that Partholan and the whole of his followers, numbering 9,000 people, were carried off by a plague ; yet the descendants of some survivors appear in later traditions. The third Invasion places us on slightly firmer ground. This occupation was by " the sons of Nemed," and the eponym seems to point to a race regarded, for some specific reason, perhaps for the superiority of its magic, as sacred. The Nemidians, who were the progenitors of the Firbolgs and the Tuatha de Danann (two peoples whom we shall presently meet), were brought under subjection by the Fomorians, who first appear in the time of Partholan. With the Fomorians we can commence to investigate the ethnic problem of Ireland seriously. It is difficult to imagine a race of beings with aspirations more mundane, and activities more human, than those of the Fomorians. Yet the mythologists are agreed in regarding them either as giants or as mermen. Both assumptions have a philological basis; but they cannot both be right. It is quite certain that a Fomorian cannot be at one and the same time a " giant " and a " being from under the sea." The Irish traditions describe these Fomorians without a trace of uncertainty. They were African pirates; they were Shemites who wished to separate themselves from the race of Ham; pre-eminently, they were oppressors of the men of Erin. There is nothing here that consists with the idea either of giants or mermen. Etymologically, the name " Fomorians " may be held to support the plain statements of the traditions, for it seems to mean " sea-refugees " (Cymric Fjo, flight or retreat, and Mor, the sea). That the word essentially means " pirates " would appear to be borne out by the fact that, at a later period, by a people described as " Fomorians," the Scandinavian sea-rovers are plainly indicated. If we go a step further, and ask to what nation these THE RACES OF IRELAND AND SCOTLAND. 5 African pirates belonged who exacted an annual tribute, both of children and produce, from the inhabitants of Erin, we face a question of considerable speculative interest. Is it possible to associate the Foniorians with the Phoenicians, whose explorers are believed on excellent grounds to have supplied the Greeks with the earliest description of Ireland that we possess? There are, in my opinion, good reasons for doing so. The evidence is mainly furnished by identity of religious customs, but it is reinforced by archaeological arguments that merit attention. Like the Fomorians, the Phoenicians were Africans; they were sometimes pirates; and they were the first people to visit Erin of whom a uthentic histor}' has any record. But these would be insufficient grounds of identification if there were no others. The word Beltine, applied in modern times to the fires kindled on hill-tops on May Day, was originally descriptive of a specific heathen custom of which the May Day bonfires arc (or rather were, for the practice is now extinct) com- memorative. The etymology of Beltine is disputed, modern scholars being reluctant to translate it by "Baal-fire," owing" to the supposed lack of tangible evidences of the prevalence of Baal-worship in these islands. But these evidences seem to exist notwithstanding. In cases of disputed etymology, it is well to get as far back as possible, and that rule will be followed in these pages. Peculiarly helpful, therefore, is the glossary of Irish words (obsolete or difficult to explain even in the ninth century) left by Cormac, the learned King of Munster and Bishop of Cashel, who was killed in battle in 908 A.D. We find there interpretations, a thousand years ago, of words to which a different meaning is now iattached, or the meaning of which is now altogether obscure. One of these words is Beltine. Cormac describes the custom itself in the following terras: — " Belltaine, May Day, i.e., hil-tene, lucky fire, i.e., two fires which Druids used to make with great incantations, 6 THE RACES OF IRELAND AND SCOTLAND. and they used to bring the cattle (as a isafeguard) against the diseases of each year to those fires." A 'marginal note adds: — " They used to drive the cattle between them." i A little further on, Cormac gives the meaning of Bil as Bial, i.e., "an idol god," thus showing that in the first quotation he did not, as some suppose, intend to equate bil with " lucky," or if he did, that " lucky " was a secondary meaning. There can be little doubt that Cormac's Bial stands for Baal or Bel.^ A close study of fire-customs in ancient Ireland and in modern Scotland reveals the fact that they were of two kinds, one involving the idea of sacrifice, and the other that of purification or protection. One was propitiatory and the other was preventive. The clearest account of the sacrificial class that I have seen is contained in Jamieson's Etymological Dictionary of the Scottish Language, copied from the Statistical Account of the parish of Callander in Perthshire. The quotation is as follows: — " The people of this district have two customs, which are fast wearing out, not only here, but all over the Highlands, and therefore ought to be taken notice of, while they remain. Upon the first day of May, which is called Beltan, or Baltein day, all the boys in a town- ship or hamlet meet in the moors. They cut a table in the green sod, of a round figure, by casting a trench in the ground, of such circumference as to hold the whole company. They kindle a fire, and dress a repast of eggs and milk in the consistence of a custard. They knead a cake of oatmeal, which is toasted at the embers against a stone. After the custard is eaten up, they ' Cormac's Glossary (Stokes), p. 19. * Keating says : — " It is from that fire made in honour of Bel that the 1st of May is called Biltaini or Bealtaine ; for Beltainni is the same as BeU-teine, i.e., teine Bheil or Bel's fire. THE RACES OF IRELAND AND SCOTLAND. 7 \ divide the cake into so many portions, as similar as possible to one another in size and shape, as there are persons in the company. They daub one of these portions all over with charcoal, until it be perfectly black. They put all the bits of cake into a bonnet. Every one, blindfold, draws out a portion. He who holds the bonnet is entitled to the last bit. Whoever draws the black bit is the devoted person who is to be sacrificed to Baal, Avhose favour they moan to implore, in rendering the year productive of the sustenance of man and beast. There is little doubt of these inhuman sacrifices having been once offered in this country, as well as in the east, although they now pass from the act of sacrificing, and only compel the devoted person to leap three times through the flames, with which the ceremonies of this festival are closed." There is no trace here of the element of purification, but there is a distinct suggestion of a survival of the element of sacrifice; and the worthy clergyman's surmise that the practice originated in offerings to Baal may quite con- ceivably be correct. On the other hand, the quotation from Cormac shows that Beltine in Ireland, a thousand years ago, was mainly an observance having as its object the curing of cattle-disease and the protection of the cattle from the ills of the coming year. It is not quite clear whether Cormac's fire Avas ignited in the ordinary way, or whether it was tein eight, or forced fire, commonly called need or neid-fire (A.S. gnidan, to rub; Dan. guide) . In his chapters on " Fire Customs," ^ Frazer shows the origin and widely-spread character of the need-fire, the various methods throughout the world of making these fires, * The Golden Bowjh, ii., pp. lOo-'SGo. 8 THE RACES OF IRELAND AND SCOTLAND. and the significance attached to the practice. A peculiar virtue belonged to this fire owing to its purity; it was a "living fire." In historical Rome the duty of making the sacred fire pertained to the vestal virgins and the chief Pontiff. Need-fires and perpetual-fires have a history that is full of interest. We find the " perpetual " method in Ireland as exemplified by the fire of St. Brigid (Bridget) at Kildare, which was plainly a survival of a heathen custom adapted to Christian practice.* Martin in his Western Islands gives an account of the need-fires of the Hebrides late in the seventeenth century;^ and the late Dr. Carmichael describes the custom in the same islands as practised about 1829;^ he states that in Beay (Sutherland) the need-fire was made as recently as 1830. In some cases, the people as well as the cattle, rushed between the fires to be purified. The fire-cult is usually described as an Aryan custom, but its Aryan origin is doubtful. It is intimately associated, as Dr. Peisker shows, with the Shamanism of the Ural- Altaic peoples. Describing the beliefs of the wild tribes east of the Caucasian Range, he writes: " Fire purilies everything, wards oft' evil, and makes every enchantment ineft'ective. Hence the sick man, and the strange arrival, and everything which he brings with him, must pass between two fires " ^ (the italics are mine). Here we have a root- idea substantially the same as that embodied in Cormac's description of Beltane in the ninth century, and no leSiS the same as that which induced the Hebridean crofters and the Sutherland and Perthshire farmers in the nineteenth century to drive their cattle through the forced fires, to cure them of murrain, and protect them against the power * In the Scandinavian temples there was a hallowed fire " which must never go out " (Eyrbyggia Saga). •' Description of the Western Islands, circa 1695, p. 113 (1H81). ^ Carmina Gaddica, ii., p. 340. ' The Cambridge Medieral Historj/, i., p. 346. THE EACES OF IRELAND AND SCOTLAND. 9 of enchantment during the coming year. For it is clear from a consideration of the subject of witchcraft and its various forms of expression, that cattle diseases and the spells of wizardry were intimately associated in the minds of those who practised such rites: purification and pro- tection were equally their object. It is not easy to dissociate these rites from sun-worship as being the primitive impulse from which they were derived. From the sacrifices offered to Bel there was only a further step to the rites of purification which, as we have seen, are the common possession of Aryan and non-Aryan peoples, though it appears more probable that the Aryans derived the cult from the Turanians rather than the contrary process. It would seem likely, therefore, that the two ideas, sacrificial and purifieative, gradually coalesced, thus ex- plaining the application of the name Beltine to a rite that was mainly designed for a purifying purpose. Although the survival of Baal or Bel worship in these islands is at the present day generally scouted as an exploded notion, it is difficult to evade the force of the reasoning that detects traces of that cult in such customs as that described (for example) by the minister of Callander. And it is fair to ask for an alternative and satisfying etymology of the root " Bel " in Beltine, if its identification with the Phoenician sun-god is rejected. The same root is found in the " Bell- trees " of ancient Ireland, which were apparently sacred groves.* The evidences of sun-worship, more particularly in the Hebrides,^ where ancient customs, extinct elsewhere, have persisted until modern times, are altogether too strong to be ignored. A single archa3ological argument from Ire- • It will not do to assert that Beltine is simple " Bale-fire " (A.-S. Btel, a burning), or a warning fire kindled on an eminence, because that derivation entirely fails to explain the rites associated with Beltine. The same objection applies to Cym. Btiili, an eminence. • See Martin's Western Islands. 10 THR RACES OF IRELAND AND SCOTLAND. land may be cited; and it seems conclusive. The solar discs which have been found in that island must necessarily have been associated with the solar cult. Among the ethnic Irish, a certain god (Crom Cruaich) stands out with peculiar distinctness as pre-eminently the object of special veneration. I suggest that in the several descriptions of this idol which are scattered throughout the most ancient bardic literature, the lineaments are traceable of Baal Melkarth (Moloch) the Tyrian diety that combined the beneficent and maleficent attributes of the Phoenician Sun-god. It is common ground that the May - day bonfires with their attendant customs, are survivals of pagan rites; and their symbolism, which survived to the nineteenth century, is found as symbolism as early as the ninth. The reality behind that symbolism may be seen probably in the fifth century, when St. Patrick entered upon his crusade against paganism in Ireland. The chief representative of this paganism was the idol named Crom Cruaich, situated in a plain named Magh Slecht. The idol's name has given rise to a good deal of etymological guessing. It means, literally, either " Curved Mound," if Crom is an adjective, or " Mound Serpent," if a substantive. Sir John Rhys, whose opinions are entitled to respect, suggests that the idol Crom Cruaich was in a state of decay at the time of St. Patrick, and had consequently assumed a stooping posture; an explanation which Dr. Douglas Hj^de appears to regard as satisfactory. By this reading, Crom is interpreted as the "Stooper"; but Cruaich is literally translated by Sir John Rhys as " Mound." A " Mound- stooper " is a conception that calls for an effort of the imagination. M. D'Arbois de Jubainville connects Cruaich with cruor, blood, and translates Crom Cruaich as the " Bloody Crom," an interpretation that loaves things pretty much as they were. THE RACES OF IRELAND AND SCOTLAND. 11 The literal translations given above are easily ex- plicable, if we assume that Crom Cruaich was one of those peculiarly -shaped eminences known as serpent-mounds. The best example of a serpent-mound, in Scotland at any rate, is one near Oban, which was discovered by Mr. Phene in 1871.10 The serpent-moun3 of Crom Cruaich, assuming its existence, may have escaped detection to the present day. Irish antiquaries are not agreed whether Crom Cruaich was situated in Leitrim or Cavan. It may not have been in either county, but I am convinced that when it is ultimately identified, it will be found to take the form of a serpent- mound. The artificial mound near Oban is stone-ridged; it curves like the letter " S "; and it is three hundred feet in length. It faces, looking eastwards, the triple peaks of Ben Cruachan (a name, by the Avay, that has the same derivation as Cruaich), and abuts on Loch Nell. Its situation is suggestive of sun-worship, but it is here impossible to en- large upon that suggestion. On the head of the serpent is a circle of stones, corresponding with the solar disc on the .heads of the mystic serpents of Phoenicia. In the centre of the circle, Mr. Phene found the remains of an altar which have since disappeared. Also, the circle has been proved to contain a grave, which reveals the double purpose of this dracontine structure. ii The Dinnsenchus, an Irish topographical tract of un- certain but admittedly ancient date,i2 describes Crom ** A description of this mound is given in Miss Gordon Cumming's From the Hebrides to the Himalai/aii, !., pp. 37-9. *^ The serpent-mound at Oban is not the only one in Scotland. There is one at Glenelg, and another in Lorn (Henderson's Sureivals of Beliffx among the Celts, p. 169). The author remarks (pp. 167-8) that one finds the serpent associated with a knoll in Scottish myth. The serpents figured on some of the sculptured stones may have a religious signifi- cance. " Attributed to the sixth century. 12 THE RACES OF IRELAND AND SCOTLAND. Cruaich (who is called Cenn Cruaich in the Tripartite Life of St. Patrick) ^^ in the following words: " The king idol of Ireland, namely Crom Cruaich, and around him twelve idols made of stone, but he was of gold. Until Patrick's advent, he was the god of every folk that colonised Ireland. To him they used to offer the firstlings of every house, and the chief scions of every clan." According to this description, the idol was covered with gold, and was surrounded by twelve lesser deities made of stone. If, now, we replace the idol (which, let it be assumed, was once there) on the altar of the Oban serpent-mound, we have a representation of Crom Cruaich that agrees in every particular with the description in the Dinnsenchus, not excepting even the sacrificial feature; for there is a tradition that, in remote ages, the Oban structure was the' scene of public executions. Crom Cruaich was in Magh Slecht, which may mean the " slaying plain." This interpretation appears to be more correct etymologically than the " plain of adoration," which is the usual translation. It connects the plain directly with the sacrificial rites that are mentioned in the Irish texts. • By the Phoenicians, the sacrifice of first-born children was a recognised rite in the exercise of public w^orship. The offering of first-fruits was a Semitic custom, originally derived, it is believed, from the Akkadians, a Turanian people. It was practised exclusively by Semitic peoples among the Caucasian races. In the sacrifices to Crom Cruaich, we seem to be witnessing the performance of rites appertaining to Baal ^lelkarth. A description by the late Dean Stanley of an inner temple on the Hill of Samaria, dedicated to Baal, bears some resemblance to the sanctuary " Cenn is here to be equated, perhaps, with "King," a meaning which seems to be borne out by the succeeding words quoted from the Dinnsenchus. If it means "head" or "chief," it suggests the presence of other and inferior mounds of the same character. THE RACES OF IRELAND AND SCOTLAND. 13 of Crom Cruaich. " In the centre," says Stanley, " was Baal the Sun-god; around him were the inferior deities." i* These are described by the author as Phoenician deities. In Phoenicia, the Sun-god was sometimes represented in serpentine form.i^ It has already been suggested that Crom Cruaich was dracontine, and the conjunction of Bel and the Dragon in early Irish texts can hardly be lacking in sig- nificance, particularly when we find the same connexion in the bardic literature of Wales. In the Leabhar Breac, one of the ancient Irish books, a lake on the top of a certain mountain is called Loch Bel Dracon, of which it is prophesied, in Adamnan's Vision, that it would kill, in the form of a pestilence, three-fourths of the people of the world. 1*^ It is quite conceivable that this loch may hava had a serpentine mound on its bordei's like that of Loch Nell.i''' Traces of the dracontine form are still found in gome place-names of Ireland, e.g., Cor-na-bpiast (English " beasts "), which Dr. Joyce translates as " the round hill of the worms or enchanted serpents." The familiar legend that St. Patrick drove all the snakes out of Ireland, probably originated from the well-grounded assumption that the saint destroyed the ophiolatry which he seems to have found in the island. From the Tripartite Life, we find that Cenn Cruaich's satellites were swallowed up miraculously by the earth when the saint shook his staff at them, and the chief idol himself bore the mark of the staff. This statement " Lectures on the Jewish Church, Part ii., pp. 388-9. '^ The serpent was considered to be symbolical of the solar deity. See Deane on Serpent Worship, p. 85, who calls Ophion the serpent-god of Phoenicia (p. 186). Deane (p. 94) says that the Phoenician mariners in- troduced to Western Europe the worship of a deity named Ogham. The name irresistibly suggests the mysterious Ogam script. '^ O'Curry's Lectures on the MS. Materials of Ancient Irish History, p. 427. " Water-spirits, however, usually take the form of serpents or dragons (see Frazer's Golden Bough, ii., p. 155). 14 THE RACES OF IRELAND AND SCOTLAND. seems to imply an attack by Patrick on the rites of the ethnic Irish. Curiously enough, Fionn, no less than Patrick, figures as a serpent-destroyer in Irish legend. He is said to have slain "all the savage reptiles of Erin, the two dragons of Loch Inny and the dragon of Loch Cuan, which is Strangford, the piasta of the Shannon, and the great serpent of Ben Edar, which is Howth." ^^ These reptiles and dragons may be represented in modern times by the " wurrums " feared by the Irish peasants, which infest lakes and carry off human beings. The origin of this superstition may be traceable to the impression produced in the distant past on the minds of the peasantry, by mounds shaped like serpents on the margin of lakes. The serpents covered with grass, but alive, which figured in North African myths, must assuredly mean dracontine mounds. The great sea- serpent which appears periodically to the eye of faith, may be the marine counterpart of the land dragon, or it may be the land dragon in another element, for the " beast " was apparently amphibious. The maps of early geographers are frequently decorated with fearsome monsters playfully disporting themselves in the sea. These sea-dragons illustrate the beliefs of the time: they are probably identified with such place-names as Great Orme's (Worm's) Head. The dragon-myth on sea and land gripped the imagination of our forefathers, Celts and Teutons alike, as their legends amply testify; and not of those races alone, for in one form or another, the belief is world-wide in extent. It is noticeable that, in the Irish texts, the word Crom is associated with pestilence, e.g., C?'om Chonnaill, the pestilence that appeared in the form of a beast, and was miraculously killed by Saint MacCreiche; also Crom Dubh of Connaught, by which is apparently meant the Black Death; it is translated as "the Black Maggot or Serpent."!^ '^ O'Grady, JJistorif of Ireland, i., p. 33. '3 O'Curry, Lectures, pp. 631-2. THE RACES OF IRELAND AND SCOTLAND. 15 The sacrifices to Crom Cruaich were made with the object of averting pestilence or famine. The reverence that would be paid to a god capable of causing or averting a plague is easily conceivable. It has been objected that some of the rites of the Crom Cruaich cult, may have been added to the original tradition by Christian monks who were conversant with the Scriptural accounts of the worship of Moloch. That objection seems to be met by the consistency with which the whole story, as it now appears, hangs together. It cannot well be doubted that we have here a genuinely historical picture of paganism as it existed in Ireland at the coming of St. Patrick, by whose influence the external forms of heathendom were abolished, though, in substance, some of its features were grafted on the Christian faith. On the archaeological side, there is something to be said in support of the Phoenician theory. Cromlechs in Ireland are ascribed by tradition to the Fomorians,^^ whom I am seeking to identify with the Phoenicians. It is not a little remarkable that this class of tombs, from North Africa west- wards, should preponderate along the line of the Phoenician colonies and trading centres, though, of course, they are found in other parts of the world. ^^ 2" O'Grady, Htstory of Ireland, i., p. 14.L ^1 Whether the dolmens came to Ireland with the Phoenicians, or a race akin to the Berbers, it seems to be tolerably certain that their centre of dispersion was North Africa. One of the meanings of Cat is tumulus. (See the analysis of this root in a later chapter.) It is properly applied to dolmens (cf. Keith (or Cat) Coity House at Aylesford in Kent), which adds force to the conten- tion that the latter were originally covered by mounds. Of the cup-markings on cromlechs in Scandinavia, Montelius says (Woods, p. 36): — "These were certainly used for offerings either to or for the dead." They are called "elf-mills" (compare the old custom in the North of Scotland of offering oblations of ale and milk to " Brownie " on stones with cup-receptacles for the liquid). Montelius adds : " Even at the present day, they are in many places regarded as holy, and offerings secretly made in them." Are these cup-marked cromlechs the work of 16 THE RACES OF IRELAND AND SCOTLAND. I have thus tried, by evidence which, cumulatively, may carry weight, to show that the Foniorians, a Semitic people who exacted a tribute of first-fruits from the men of Erin, were, in fact, a Phoenician colony, or a body of Phoenician sea-rovers, who imposed alike their rule and their religion upon Ireland. 22 They were followed in their domination of that country by the Firbolgs and the Tuatha de Danaun, whose identification will be attempted in the following chapters. the colonies of Semitic people who, according to Nilsson, introduced both bronze and Baal-worship to the south and west of Scandinavia ? The name "giants' graves" applied in the south-east of Ireland to cromlechs, finds its counterpart in Denmark, where they are called "giants' chambers." Probably their gigantic properties relate to the massive size of the stones, which were doubtless believed to have been raised by a race of giants. The stone circles in Scandinavia (see Worsaae) are found in conjunction with tombs of the Stone period, mounds of earth being the distinguishing characteristics of the Bronze Age. " The Irish " keeners," who were hired to howl at funerals, perpetuated a heathen custom derived apparently from a Phoenician ancestry (see Stainer and Barrett). CHAPTER II. The Firbolgs — The traditional story of their orifi^in— The etymology of the name — A theory to explain the name — The "long-heads" of Ireland — Huxley's pregnant suggestion -The Firbolgs identified by tradition with the *' Mediterraneans " — Moytura, the •' heap plain " —Giants and gods — The overthrow of the Firbolgs by the Tuatha de Danann. That the Firbolgs were a race of real men and women is common ground alike for Euhemerists and mythologists. But not for all mythologists. One of the most curious theories which have been advanced is that which makes the Firbolgs " men of the bag or womb," i.e., men " born in the ten lunar months of gestation." ^ The association of holg with " bag " lies at the root of nearly all the guesses which have been made to explain who the Firbolgs were, and to give a satisfactory derivation of their name. The prefix fir (men) is beyond dispute; the difficulty is with bolg, to which various meanings have been attached. The Irish story about these people, as preserved by Keating (a familiar name in the discussion of Irish history), is obviously a late concoction, being composed mainly of etymological elements, but based perhaps on a slender foimdation of genuine tradition. Keating states that there were three correlated peoples — the Firbolgs, the Firdhomh- noins, and the Firgailians ^ — comprehensively the Firbolgic tribes who Avere oppressed by the "Greeks." These Firbolgs, preferring exile to slavery, emigrated to Erin. ^ I'nmitirtt Traditional Jfixfon/, by J. F. Hewitt, vol. i., pp. 32 and 8'i6. - Gailion and Donihnann were names for Leinster {Silra Wadrlira, Fng. text, p. .500). 2 18 THE RACES OF IRELAND AND SCOTLAND. They had been forced by their Greek masters to dig up clay, and carry it to barren places to form soil for crops. The Firbolgs carried the clay in bags, hence their name, for holg means a " bag." The Firdhomhnoins did the digging, hence their name, for dhomhnoin means "deep." The Firgailians guarded the workers from the attacks of enemies, hence their name, for gailian means a " spear." From the time of Keating to the present day, the Firbolgs have been called " men of the bag," or " bag men," for the same reason as Keating's, namely that holg, among other things, means a " bag." That undisputed fact does not^ however, carry us very far; not further, indeed, than the threshold of enquiry. For, if the Gaelic holg (and the Cymric holgan) signifies " bag " or " sack," it means the san^ thing in Moeso-Gothic, and is found with a cognate signification in all Teutonic languages. It is the source from which are derived a number of English Avords, e.g., "bag," "big," "bulk," "bulge," "bilge," "billow," "belly," "boll," and perhaps "ball." We find it in place-names, e.g., the " Bogie " (anciently " Bolgie ") " River," " Cairnbulg," and " Dunbolg," all in Scotland, and " Moybolgue " (anciently " Maghbolg ") in Ireland, with others that could be named. The essential idea at the root of all these words is " swelling," and it will bo found that every word derived from hulg or holg possesses that characteristic. Applying this test to " Firbolg," the idea that first suggests itself is that of a nation of " paunch-bellies," an aggregation of individuals distinguished by fatness. That idea, inherently improbable as a national name, receives no countenance of any sort from tradition. Nor are we- justified on etymological or other grounds in connecting the name with the Belgae of England; and still less, perhaps, with the Volcae, the Celtic people from whom some philo- logists derive the name Walh, applied by the Teutons to THE RACES OF IRELAND AND SCOTLAND. 19 tlie Celts, and afterwards to the Romance people of France and Italy. The word Firbolg has its nearest congener among European national names in that of the Bulgars of Bolgary, a race of Ugro-Finnish origin on the Volga, whose ancestors between the fifth and seventh centuries conquered and gave their name to Bulgaria, afterwards adopting the language of the Slavonic people whom they subdued. But it is impossible to establish even a remote connexion between them and the Irish Firbolgs; the former are not found as European settlers until the fifth century. Orosius mentions a country called by him " Bulgaria," which he places near Istria on the Adriatic,^ and by the same author a Bulgarian people (" lUyrians whom we call Bul- garians ")* are placed in Thessaly. It is evident, therefore, that the ancient lUyrian people were called Bulgarians; and these lUyrians are believed to be one of the most ancient of the Mediterranean nations.^ They may be the " Bulgares " mentioned by Jordanes as a people oppressed by the Goths. Here, therefore, we may find the link we require between the Firbolgs and the Greeks of the legend who oppressed them. For wars between the Greeks and the Illyrians were frequent; and it is by no means improbable that the latter were enslaved by their formidable neighbours. They were certainly conquered by the Macedonians. Orosius relates that Philip of Macedon slew many thousands of the Bul- garians in Thessaly, and captured Larissa, their largest city. The Illyrians had a good military reputation, and " they of all people could fight the best on horses." ^ They were, therefore, a valuable asset for the Macedonian army. * King Alfred's Orosius (Thorpe), p. 257. * Ibid., p. 339. * The modern Albanians are thought to be their nearest descendants, and it is a curious fact that the modern Albanians claim a common ancestry with the modern Scots. * King Alfred's Orosius ^ p. 339. 20 THE RACES OF IRELAND AND SCOTLAND. It is not necessary, of course, to treat seriously the fiction related by Keating to account for the name Firholg, foi' that can be explained on more rational grounds. The root holg enters into combination with muir (the sea) in some early place-names to denote an inlet or a " sack " bay like the Frisian JddeJ It is found in the name " Muirbolc " (Port na Murloch, Lismore), used by Adamnan with that meaning, and in " Muirbolg " (now Murlough) in Ulster. Thus holg is in these names the equivalent of " lough " or " loch." The idea conveyed seems to be that of the sea " bulging " into the land. So, too, the Gae-bolg, wielded by Cuchullin in his famous fight at the Ford, was a spear, Avhich, on entering the body, made only one wound, but afterwards expanded into thirty barbs. And Spring was named Imhulc, perhaps because it is the time of the swelling of the buds. Applying the theory of an inlet, or bay, or loch, to explain the name of the Bulgarians (Illyrians), it is barely con- ceivable that it may relate to the Adriatic, or, in an extended sense, even to the Mediterranean Sea. But that is a venture- some hypothesis, and it seems far more probable that the Irish Firbolgs derived their name from the fact that their later location was mainly in Connaught. The numerous inlets by which the coast of Connaught is characterised, offer a plausible explanation of the name " Firbolgs," viz.: — " Bay-men," the latter being thus the equivalent of the Scandinavian name, " Vikings." ^ Tradition asserts that Erris in Mayo Avas the chief landing-place of the Firbolgs, and Mayo is peculiarly indented by bays. Corroboration of the view just stated may be found in one of the Ii-ish ^ See a discussion on "sack-inlets" in Xansen's Jn Northern Jlisff:, i., p. 93. An exact parallel is found in Mid. High German Sliich, which means both a " leather bag" (holg), and a "gulf" (boUj). Apparently, in both instmices, there has been an evolution in meaning. ** IhU/ would seem to convey the idea equally of convexity and con- cavity (cf. si mm, a bay or bosom). THE RACES OF IRELAND AND SCOTLAND. 21 texts, which informs us that the Firbolgs came to Erin " out of the East (of Ireland) beyond Slieve Alpa (which is in ^layo), and the country of the Franks and the Lochlannah."^ The allusion to the Franks and the Lochlannah seems to imply the existence of Norman and Scandinavian settle- ments in Ireland at the time the text was written, thus dating it from post-Norman times. In Eddi's Life of St. Wilfrid, there is an allusion to the tribes (apparently a servile people) who were gathered to- gether by the Picts of Scotland de utribus et folliculis Aquilonis.^^ Not improbably these tribes were located along the northern lochs on the coast. Titer and foUicidus, in a figurative sense, may well mean a " sack " bay and a " sack " inlet. The population of Ireland is now, and so far as has been ascertained, always has been, almost wholly dolichocephalic. The ancient skulls which have been observed belong either to the middle form represented by the long-barrow and river- bed elements of the population of England, or the elongated crania represented by the Scandinavian skull. The former belong to what Huxley has classified as Melanochroi, the short, dark longheads, and the latter to his Xanthochroi, the tall, fair longheads. The first is the Mediterranean or Iberian type: the other is the type associated with the Scandinavians. Retzius alludes to the likeness between the Scandinavian and what he calls the " Celtic " skull; and he states that, having on one occasion exchanged with Sir W. Wilde a typical Scandinavian for a typical Irish skull, both observers agreed that " it would be difficult to find any important difference between the two." ^^ Commenting * O'Grady, i., pp. 210-311. Lochlyn (Cyra. UychUpi) means a gulf. The Lochlannah or Scandinavians may have derived their name from the Gulf of Bothnia, or perhaps, in a wider sense, from the Baltic. "^ Cited by Skene, Celtic ^mlUmd, i., p. 2fil. Skene offered no opinion on the meaning of the words. " Prehistoric liema'ms of CalthneKa, p. 129. 22 THE RACES OF IRELAND AND SCOTLAND. upon the tall, fair, red-haired, and blue-eyed doliehocephali who are (and appear always to have been) so numerous in Ireland and Scotland, Huxley suggests that " long before the well-known Norse and Danish invasions, a stream of Scandinavians had set in to Scotland and Ireland, and formed a large part of our primitive population." ^^ I am convinced that this suggestion explains a good deal in Irish and Scottish ethnology that has presented a baffling problem to students. The descendants of the " Mediterraneans " abound in the west of Ireland at tho present day. They are a dark, long- headed, and rather short people; and their progenitors are believed to be the Firbolgs of Irish tradition. Duald MacFirbis, a celebrated Irish antiquary of the seventeenth century, distingushing between the descendants of the Firbolgs, the Tuatha de Danann, and the Milesian Scots, gives the following characteristics of the first-named: — " Everyone who is black-haired, who is a tattler, guileful, tale - telling, noisy, contemptible; every wretched, mean, strolling, unsteady, harsh, and in- hospitable person; every slave, every mean thief, every churl, every one who loves not to listen to music and entertainment ; the disturbers of every council and every assembly, and the promoters of discord among people; these are the descendants of the Firbolgs, of the Gailiuns of Liofarne, and of the Fir Domhnanns in Erinn. But, however, the descendants of the Firbolgs are the most numerous of all these." ^^ MacFirbis states that he took this "from an old book," and gives no further information about his authority. But the unflattering character which he ascribes to the descendants " Prehistoric Remahis of CaithnfKn, p. 134. " O'Curry's Lectures on MS. Materlnh, p. 223 (cf. another version, p. .580). THE RACES OF IRELAND AND SCOTLAND. 23 of the Firbolgs fits in with some allusions in the ancient texts to " a base Firbolgic clan, a tribute-paying people, scorn of the warrior-tribes of Erin." i* On the other hand, the same texts elsewhere describe the Firbolgs as being " mighty of bone and thew," but " not so comely to look upon as the warriors of the race of Milith." ^^ And Fardia, the chief of the Firbolgs, who fought with Cuchullin at the Ford, is delineated as a proud, independent warrior, of stately, mien and with flowing golden hair. There is a seeming contradiction here, but unless the text has been redacted, the explanation may be that these big, raw-boned Firbolgs were of another race, superimposed upon the smaller, darker, and less warlike Mediterraneans. They are described as " champions," and among the ancient Irish, as among the ancient Scandinavians, that word implied mercenary professional fighters, who were employed to guard the boundaries of those whose service they entered. These " fighting Firbolgs " may have thus become attached to the Mediterraneans, and in time have become their masters. Sir W. Wylde speaks of a long-headed, dark people west of the Shannon, and of a more globular-headed, light-haired stock north-east of that river,i^ by which description Huxley assumed that he meant that the latter people have broader heads than the others — " not that there was any really brachy cephalic stock in Ireland." This combination of physical characteristics in what is believed to have been a Firbolgic district, offers a curious parallel to the distinction we have been considering, and seems to support the suggestion I have made. It is conceivable that the fair, '* O'Grady's llisforii of Irelik Walkendorf (c. 1520) writes about the Skraelinger : " They were a small people who lived in underground houses and who worshipped gods." (Nansen, In Northern Mists, ii., p. 86.) Some Norse colonists of Vinland fled from the Skraelinger when they first saw them : they thought they were spirits. '* Silca (ladf/ira (Eng. text, p. .>74') says that the Dananns were " they that first introduced swine " into Ireland (or Munster). A strange dis- tinction for "gods " ! The boar was specially associated by the Swedes with the worship of Freya, "the mother of the gods." The peasantry still make images of little boars in paste in February. (See Tacitus, Germnnid (c. 45), on this custom.) THE RACES OF IRELAND AND SCOTLAND. 35 Isle, tlio peninsula is mentioned by Dean Monro in the sixteenth century, and by other writers in the seventeenth and eighteenth centuries. Now, this isle had a small so- called kirk, and the kirk had a remarkable history. For " pigmies " had been buried underneath its floor, and the bones of the " pigmies " had been dug up on various occasions — so the story ran. Its truth was proved by the bones which were there to speak for themselves, and to silence the criticisms of the sceptical. The pigmy story attracted many people to the spot to see the bones of the little men ; and the fame of the isle seems to have reached literary circles in London, for Collins alludes to it (he was among the credulous) in one of his Odes. When on a visit to Lewis early in the nineteenth century. Sir Walter Scott's corres- pondent, Dr. John MacCuUoch, tried to find the isle, but failing in his search, denied (characteristically) its very existence. Some years ago, I was more fortunate in my search, and the result of digging in the supposed kirk disclosed the existence of two underground chambers, a description of which appears elsewhere. ^^ The bones of the " pigmies " were collected and examined at the South Kensington Museum; they turned out to be the bones of various mammals (ox and sheep and lambs) and sea-birds (razor-bills and gulls). The point of all this is, that we have here a story about " pigmies " who had lived in a fairy-hill (for Luchruban has the appearance of a typical fairy-mound), which story was apparently based on a belief proved to be false. I sa,y apparently with good reason, for, although at first sight the conclusion appears irresistible that the pigmy legend derived its origin from the discovery of the small bones, further investigation showed that this conclusion was possibly a mistaken one. There is a tradition in the Luch- ^^ Proc. Soc. Ant'iq. of Scotland, vol. xxxix., pp. 248-238. 36 THE RACES OF IRELAND AND SCOTLAND. rubaa district which relates that a race of small iiion did actually reside there in pre-historic times. The tradition is that these dwarfish people were " Spaniards," who came to Lewis in 1500 B.C. (the less reliable tradition is, the more precise are its dates), the supposed Spanish connexion being unexplained. The pigmies lived on " buffaloes," probably represented by the oxen whose bones have been discovered, and they killed those buffaloes " by throwing sharp-pointed knives at them." The dwarfish people were invaded by " big yellow men from Argyll, who drove them from their ancient possessions near Luchruban." ^'^ This tradition may have pre-dated the discovery of the pigmies' bones, the finding of which would, however, accentuate and help to perpetuate the original story about the little men. From the Euhemerist standpoint, there would appear to be ground for the belief that the pigniy tradition originated from an actual prehistoric occupation of the subterranean chambers, by a small people who lived underground and as fairies became immortal. Were these Lewis pigmies Lapps? It is well to avoid attaching too much ethnological im- portance to a similarity of customs between different peoples, because it is an argument full of pitfalls. But in one in- stance at least, the coincidence of custom is so remarkable, and the custom itself is so peculiar, as to prove apparently contact, direct or indirect, Avith the British Isles by the Lapps. I allude to what may be called the custom of the knotted cord. The following quotations, set forth in parallel columns, describe the custom as practised respectively in Lapland and Scotland: — '^Communicated to the author by a resident of the district. The theories (supported by cranial evidence) of the eminent anthropologists, Sergi and Kollmann, regarding pigmy settlements in Europe in remote times, remain, I believe, unrefuted. THE RACES OF IRELAND AND SCOTLAND. 37 From Richard Eden's des- cription of the Lapps. ^^ " They tie three knots on a string hanging at a whip. When they loose one of these they raise tolerable winds ; when they loose another the wind is more vehement ; but by loosing the third, they raise plain tempests, as in old times they were accus- tomed to raise thunder and liffhtning." From J. H. Dixon's Gairloch, pp. 168-9. "On one occasion, M'Ryrie was kept several days at Stomoway by a contrary wind. He was going about the place two or three days grumbling at the delay. He met a man in the street, ^^■ho advised him to go to a certain woman, and she would make the wind favourable for him. In the moi-ning he went to her, and paid her some money. She gave him a piece of string with three knots on it. She told him to undo the first of the knots, and he would get the wind in his favour; if the wind were not strong enough for him, he was to undo the second knot, but not until he would be near the mainland ; the third knot she said he must not untie for his life. The wind changed while he was talking to her ; and he set sail that same morning. He undid the first knot on the voyage, and the breeze continued fair ; the second knot he untied when he was near the mouth of Loch Ewe, and the breeze came fresh and strong. When he got to Ploc-ard, at the head of Loch Ewe, he said to M'Lean that no great harm would happen to them if he were to untie the third knot, as they were so near the *8 Quoted by A. H. Keane in The Lapjis, p. 19. 38 THE RACES OF IRELAND AND SCOTLAND. shore. So he untied the third knot. Instantly there was such a hurricane that most of the houses in Poolewe and Liondubh were stripped of their thatch. The boat was cast high and dry on the beach at Dal Cruaidh, just below the house of Kirkton ; her crew escaped uninjured. It is said that at that time there were several women about Stornoway who had power by their arts to make the wind favourable." Selling winds cannot be claimed as a monopoly of Finnish people, but the knotted cord, so far as I have been able to ascertain, is a Finnish, and especially a Lapponic, specialty. In Regnard's Journey to Lapland, it is stated that the " knotted " way of selling the wind was " very common " in Lapland. " The very lowest sorcerers have this power, provided that the wind which is wanted has already commenced and requires only to be excited." ^^ Readers of Scott's The Pirate, will recall how Noma of Fitful-Head sold favourable winds, and in his notes on this romance, he cites Olaus Magnus, who tells of one King Eric of Sweden, called Windy Cap, in allusion to his power of making the wind blow whichever way he chose by turning his cap in the desired direction. This was just the sort of wizardry that the Scandinavians probably learned from the Lapps or Finns. The prototype of Scott's Noma was Bessie Millie, who lived at Stromness in the Orkneys over a hundred years ago, and who had a flourishing business as a seller of favourable winds to storm-stayed skippers. The goodwill of Bessie's business was acquired by one Annie Tulloch or Mammie Scott, who sold favourable winds at the rate of '" Pinkerton's Voi/ages a»d Travels, i., p. 180. THE RACES OF IRELAND AND SCOTLAND. 39 eighteen pence apiece. Her method of applying the principle of the knotted cord was to instruct the skippersl to go to sea with two reefs in the mainsail, only one of which was to be shaken out during the voyage. If both were shaken out, a contrary gale sprang up, but if the vessel were driven back to the Orkneys, a " whole-sail " breeze could be purchased from the accommodating Annie for a further consideration. 20 Professor Frazer gives the interesting information that Shetland seamen still buy winds, in the shape of knotted handkerchiefs or threads, from old women who claim to rule the storms.-^ Witches on the mainland of Scotland had other means (see Dalzell) of raising the wind, which (literally) was a far easier task some centuries ago than (figuratively) it is at the present day. The custom of the knotted cord was also known in the Isle of Man, and Thorpe says that there was a woman at Siseby on the Slei who sold winds to the herring fishers in the same manner. 22 But wherever practised, the custom was apparently borrowed from the Finnish peoples, whose wizardry was of the same character as that of the Irish Dananns.23 The exercise of magic, sorcery, and witchcraft, by what- ever name it is called. Shamanism in short, was the governing principle of these peoples. " The rule of the Shaman (or wizard) over nature," says Comparetti, " is the fundamental idea of Shamanism." He adds that, until quite recently, the Lapps were Shamanists like the Eskimos and Samoyedes, a fact confirmed by the great fame which they enjoyed in ancient times among the Scandinavians, for the truth of "' Tudor, The Orkneys and Shetlands, p. 3^i5. -' 'Ihe Golden Bouifh, pp. 32-2-6. ^ Northern Mytholof/y, iii., p. -ili-l. -' The Seid-women of the Scandinavians received money *' to make men hard, so that iron could not wound them" (Thorpe, ii., p. 214). This is analagous to the healing baths of hot milk and herbs employed by the Dananns to cure the wounded ; which was effected instantaneously. 40 THE RACES OF IRELAND AND SCOTLAND. which Comparetti quotes conclusive evidence. " There is," he says, "abundant information on this point;" and every student of the Icelandic Sagas will agree with him. Lapland witches, more properly wizards, were known in England as well. Comparetti thus describes the Shaman: — " The Shaman is more than a simple priest: he is the seer, he is the medicine-man, he is wise and powerful above all others, and is capable of miraculous actions. With his actions and his word, he dominates things and men and animals and spirits; he cures ills or prevents them; he can even produce them; he can propitiate superior beings and obtain benefits; can ensure good luck for the hunt, the fishing, the journey; can raise winds and storms, and clouds and fogs, and tempests, and can lay them, scatter them, disperse them; he can transform himself and others; he can rise in spirit into the realms of air, go down into those of the dead, and carry off their secret.'" ^4 The Dananns of Ireland had their Shamans; so, too, as we shall see, had the Cruithne of Ireland and the Picts of Scotland. 2* The Traditional Poetri/ of the Finns (Anderton), p. 172. CHAPTER IV. The Lapponic theory further discussed — Disproved by anthropological evidence — MacFirbis on the Dananns — The Irish texts on the Dananns — The elves of light and the elves of darkness — St. Patrick and elf-worship — The elf-creed introduced to Ireland by Scandin- avians — Folk-lore as an aid to ethnology — A classification of the Teutonic elves — The application of elf-beliefs to the Dananns^ Parallels between the Dananns and Scandinavian mythology — Thorpe on the resemblances between Teutonic and Celtic folk-lore. What may be called the Lapponic view of the Danann problem is not lacking in support from the evidence of anthropology and archaeology. Professor Retzius is quoted by Mr. Borlase ^ as having maintained that there was a race in Britain of Turanic origin represented by the Lapps and brachycephalic Finns, which preceded and was entirely different from what Retzius calls the " Celtic " type. Of a brachycephalic skull, found in a cist near the Knockadown group of circles at Lough Gur, Professor Harkness remarked that " it seemed to be a member of a race approximating most nearly to the modern Finn or Lapp." ^ And Borlase states that the dark races in Ireland (and Scotland) include types both of dolichocephaly and of brachycephaly. " In the wilds of Donegal," he says, " I have seen both these types." ^ He describes the burial customs of the Lapps as recorded by Scheffer, and adds: " In every particular of this account, we see precisely what archgeological research on the one hand and legend and tradition, committed to writing in the middle ages, coupled with folk-lore still in oral survival, on the other hand, lead us to believe occurred in the case of 1 The Dolmens of Ireland, p. 1,009. '//>;i Chaillu, i., 4.11). THE RACES OF IRELAND AND SCOTLAND. 45 benevolent beings; the dark elves, or dwarfs, were usually malevolent, "In Alfheim, ' says the Prose Edda, "dwell the beings called the Elves of Light; but the Elves of Darkness live under the earth, and differ from the others still more in their aetions than in their appearance. The Elves of Light are fairer than the sun; but the Elves of Darkness blacker than pitch." ^^ The Dananns of Ireland appear in both aspects. i*^ The dark elves figure in the Irish texts under the name of Luprachan, which means a dwarf. The word abhac means both a dwarf and an elf, showing an identification of the dwarfs with the elves, which is common also to Teutonic mythology. A certain Aed Enver in the Irish texts boasta that he was " of the race of Luprachan, a descendant of Dana, who in ancient days occupied Tara, and he told how the Clanna Luprachan ruled widely over Erin, teaching noble arts to the Gael, and how thoy dwelt now immortal in fairy- land." 1* This is a significant passage, for it shows the Dananns both as humans and as dwarfs or dark elves. The boast of Aed Enver is paralleled by the belief entertained by gome Scandinavians that they were descended, not from the gods but from the elves. ^^ If the Teutonic legends and traditions are studied with care, it will be found that the distinctions drawn between elves and human beings show a certain lack of definiteness. This lends support to the realist view that the originals of the elfish people were men and women, possessing in a maikod ^- Nortliern uhifiqtiifles, p. 4-14. '""The Sick-Bed of Cuchulainn " shows them as "demons." Else- where they appear as guardian spirits (the Scandinavian dlsir). See O'Grady, History of Ireland, ii., pp. 29 and 2.58. i*0'Grady, i., 150. '* Du Chaillu, Vlkliu/ Age, i., 409. "Are ye of the elves or of the gods ? " asked the daughters of King Laoghaire when the}' met St. Patrick and his companions {Trip. Life). This is an exact counterpart of the Alfar and Asar of the Scandinavians. 46 THE RACES OF IRELAND AND SCOTLAND. degree the characteristics which, in an exaggerated form — sometimes a greatly exaggerated form — have been attributed to the whole tribe of elves, dwarfs, pigmies, fairies, pixies, brownies, and other variants. The cases of the Skraellings and the Eskimos, the Lapps and the dwarfs, have abeady been cited to show how the human and non-human elements have become blended and confused. In the same way, there is every reason to believe that the Irish legends have con- founded the Teutonic Dananns with the elves and dwarfs of their mythology. The prevalence in Ireland of this elf -creed as late as the time of St. Patrick is clearly discernible. In Fiaccs Hymn, for example, we find an allusion to the fact that when Patrick went on his mission to Ireland, " the tribes worshipped elves." Complementary to this evidence, Manx legend states that Manannan MacLir (of the Dananns) and his people were " routed by St, Patrick, whereupon being of small stature, they became fairies, and lived in the ancient tumuli, using flint-arrows as the weapons with which they avenged their wrongs on human beings." ^^ This must mean that St. Patrick attacked elf -worship, and that after the introduction of Christianity, it survived only furtively and secretly in fairy beliefs. But the tradition seems to suggest, also, that a small-statured people, whose weapon was the bow, were at one time associated with the people called the Dananns. And here again we come in contact with the familiar " elf -shots " of the Irish peasantry, and the fairy arrows of the Highlanders of Scotland, both derived, perhaps, from Scandinavian legends of the miraculous archery of the primitive Finnish race of sorcerers, whom the Gothic stock displaced in Northern Europe. The inference from all this is that the elf-creed was introduced to Ireland, Great Britain, and the Isle of Man, by a people apparently belonging to the Scandinavian branch of '® Moore's History of the Isle of Man, p. 47. THE RACES OF IRELAND AND SCOTLAND. 47 the Teutonic race.^^ From their contact with the Lapps, the traditions and legends of the Scandinavians were saturated to a greater extent than those of the Germans, with storiea of underground elves; dwarfish smiths who forged magic swords and spears, and endowed them with uncanny properties; impish trolls who might be friendly or mischievous; and fear-inspiring wizards whose spells were unequalled in potency, and whose essence was regarded as divine. Place-names, as I shall show, attest the presence of a Scandinavian people in Ireland in the second century; and anthropology seems' to bring us into contact with the same people in Ireland at a period anterior to the Christian era. It is not assuming too much to suppose that these people are responsible for a good deal of the elements, common to Celtic and Teutonic mythology, that bulk so largely in the legends of Ireland and Scotland. Conquering settlers in a new country do not leave their mythology at home. If they remain segregated from the natives, they cherish their legends with conspicuous tenacity. But if they coalesce with the natives, they incorporate the indigenous legends with their own. The latter process makes folk-lore an eminently unsafe guide in determining, unaided, ethno- logical questions, though it is a useful auxiliary to anthro- pology and etymology. It corroborates, for example, the testimony of philology that the Gael of Ireland and of Scotland have a common origin, by showing us a body of legends common to both countries; and it confirms the con- clusion that when the Scots left ancient Scotia (Ireland) and settled in Dalriada (part of modern Scotland), they brought their legends and traditions with them. This applies to the Fionn Saga; and a simple explanation is thus offered of what '^ There is evidence to show that of the two creeds in Scandinavia, the Alfar and the Asar, the former was the older, the worship of Odin dis- placing it. In the earlier Edda, there are allusions to Alfl6f, the sacrifices made to the Alfar. 48 THE RACES OF IRELAND AND SCOTLAND. at one time provided a bone of contention between the sea- divided Gael. But, if we go still further back, who can say how much, for example, of the Cuchullin Saga is of imported origin, and how much is native to the Irish soil? Who, indeed, can say what share of the Fenian stories rightly belongs to our islands? Signs of a dual origin arc not difficult to discover, strongly suggestive of an admixture of races, all of them tenacious of their native lore. But this " elf " theory, in its relation to the Dananns, requires closer investigation. The uncertainty attaching to the origin of the elf-beliefs is illustrated by the tradition in Jutland concerning them. It is there related that when the fallen angels were cast out of Heaven, ^^ some of them fell on the mounds or barrows and became Barrow-folk, or, as they are also called. Mount-folk and Hill-folk; others fell into the elf-moors, who were the progenitors of the Elf -folk; while others fell into dwellings, from whom descended the domestic sprites. Now, the Barrow-folk are identical with the Irish siabh?'as,^^ which is a compound word meaning Brugh or Barrow sidh, or in other words, Mound or Blount Elves, the elves whose abodes were in the tombs. They are also the same as the " dwarfs," who, in the later popular beliefs, are generally " subterranean " ; 20 r^^^^ {^ {^ sometimes diffi- cult to distinguish them from the Norse Huldre (Hidden) folk, and Thusser or trolls. The domestic elves are the Norse '^ To this day, it is believed in Ulster that the fairies are fallen angels. '^ Tales of Berg-folk, or Barrow-folk, form the commonest type of Danish folk-lore. ^^ In the Scandinavian texts, the Si-artdlfar or dark elves, and the Dver0. Dr. R. Cruel thought that the whole of Europe was occupied by Turanian peoples of Ural-Altaic speech before tiie arrival of the Aryans. THE RACES OF IRELAND AND SCOTLAND. 77 Celtic in Roman times, have applied the adjective " Celtic " to the type which they represent. If, however, it be assumed that this type was at all times more numerously represented than the fair type; and if the well-established theory that dark stocks are more penetrative and persistent than fair stocks be accepted; then the comparative fewness in number of the fair Celts in those districts at the present day (especially in view of the inroads which a constant state of warfare must have made upon the population of the fighting class) will be well understood. It is improbable that the word " Celt " was originally anything more than a topographical designation. It may have simply meant the "forest-men:" those who, like the ancient Britons, lived on the edges, or in the cleared spaces, of the woods which covered the face of ancient Gaul. The Avord is found in modern Welsh as Celydd and Celt, a refuge or shelter afforded by a forest, which is exactly suggestive of the uses made of their forests by the Britons, as described by Roman writers. The principal woods of Britain were known in ancient times as " Caledonian " forests, the most distinctive being the great forest in the north of the modern Scotland, inhabited by the Caledonian or forest tribes. According to Caesar, Celtae was the native name of the people whom the Romans called Galli, the two names thus applying to one people. Diodorus Siculus, however, explains that the Celtae were the people who occupied the interior of Gaul above Marseilles, and the country near the Alps and on this side of the Pyrenees; while the Galli were those whose lived beyond Celtica towards the north, near the Ocean and the Hercynian mountains, and beyond the latter as far as Scythia. The Romans called the whole of these people {Celtae and Galli alike) by the common name of Galli. The distinction made by Diodor is instructive, for it seems to confirm the impression that the names Celtae and Galli wore topographical in their origin. 78 THE RACES OF IRELAND AND SCOTLAND. The name Galli is as undetermined as that of Celtae, but the meaning which probably finds most favour is that of the " mighty " or " powerful " people (Cymric Gallus, powerful or mighty).^ The Gauls are described as a vain- glorious nation, but it can hardly be supposed that they were so vain as to call themselves by so boastful a name; or, if they did, that the Romans, their conquerors, would admit their claim to it. I think that the derivation of the name must be sought elsewhere. If, like Celtae, it was topo- graphical, it may be found in the Cymric Gal, a plain, or Givalas, low land, and thus the Roman name may have been borrowed from a native source. The portion of Gaul in which Diodor places the Galli must have been of this descrip- tion, for it was the Low Countries of modern times. On this hypothesis, therefore, the Celtae took their name from the dense forests of southern and middle Gaul, while the Galli took theirs from the low, marshy district of the northern seaboard. I do not forget that there were both Celts and Gauls other than those in Western Europe, but the origin of the names remains unaffected by that consideration. It is impossible to dissociate the name Galli from the German walJi or tvealh, the origin of which is disputed. It is common ground that the word was first of all applied by the Germans to the Celtic tribes who were their neighbours, and it is sometimes derived from the tribal name Volcae,^ Avho were the Celts of Central Europe. That is a derivation which does not carry conviction. It seems more probable that ivalli is simply Gal or Gwal in a Teutonic dress, for the Cymric initial " G " is repugnant to the Germanic tongue. Thus we find George Buchanan in the sixteenth century mentioning as a curious fact that the English people of his * Pliny writes of the GalU as if the name meant " mad " or " furious ; " and it is a curious commentary on this etymology that the Irish bards allude to " the angry Britons " as a rac-ial characteristic. s The same tribe who, some writers think, gave their name to the Firbolgs. THE RACES OF IRELAND AND SCOTLAND. 79 day called Gallovidia (Galloway) " Wallowithia," and that the Gallic (French) language was called " Wallic " by the people on the borders of Germany. The Walloons of the low-lying lands {gtoalas) of Flanders must have been among the first of the Celtic tribes to come in contact with the Germans as they pressed westwards; and from their Celtic name, signifying (on my hypothesis) the " Lowlanders," the Germans may have derived the walh which they applied subsequently to all the Gallic tribes, and ultimately to all foreigners or non-Germans.'^ That the Anglo-Saxon wealh cannot have originally meant a foreigner, but a Gaul, or a person of Gallic origin is, I think, demonstrable. Anglo-Saxon arrogance could have hardly gone the length of giving the Britons, the natives of the country of which they took possession, a name signifying " foreigners." That, indeed, would be a supposition equalling in insularity the apocryphal story told of the Englishman in France, who was surprised to hear even the little children of " foreigners " speaking French. A proof of the association of ivealh with Gaul and Gwal is found in Geoffrey of Monmouth, who tells us that after the Britons were overrun by the Angles, they were called " Gualenses, Welshmen." In " Gualenses " we have a Cymro-Latin form of the English " Welshmen," but in the later forms, the " G " is dropped, and the name appears frequently in Latin documents as " Walenses." ^ It would seem, therefore, that the Wealisc or Welsh got their name from the Anglo-Saxons as denoting their Gaulish origin.^ The old Walloons — the Celtic tribes of Belgic Gaul — were '^ This root is frequently associated with the German vallcn, to wander, but the association seems to indicate a confusion of ideas. ** North-west France, or ancient Gaul, is the Valland of the Norse Sagas, and Armorica is the Wealand of the Anglo-Saxon Chronicle. " In Roquefort's Glo.isari/ the terms Walons and Gualons are used indiscriminately. Wales is le pa;/.i des WalJons; and is explained as Galfois qui est du pays de Galles. Gnlesche is explained, qui est du pays de Galles en Amjleterre. 80 THE RACES OF IRELAND AND SCOTLAND. subdued, and apparently partly expelled from their terri- tories, and partly absorbed by the Germans whom Csesar called the Belgae, a name satisfactorily explained, perhaps, by the Cymric Belgtos, rav'agers. The prevailing theory about the Belgae is that they were Celts, and many ingenious arguments have been advanced in support of that theory. It is difficult, however, to evade the force of Roman evidence to the contrary. The statement of Csesar, who had first-hand knowledge of the Belgae, and said they were Germans, is to be believed in preference to the speculations of modern critics, who try to explain away Ca3sar's words — if indeed they do not boldly assert that he was mistaken. There is, however, more to be said in favour of the evidence that the Belgae had dropped their German tongue, and adopted that of the Celts whom they had conquered. Thus some inquirers have come to the conclusion that the Belgic language resembled Gaelic rather than Cymric. That is quite a plausible conclusion, for wherever Teutonic and Cymric elements are mixed, the amalgam resembles Gaelic in its vocabulary, if not in its grammatical construction. The Belgae took possession of their territory in the south of England at a comparatively late date — probably much later than the last of the Celtic colonies from Gaul. There is no satisfactory evidence of their presence in Ireland. I must here face a problem Avhich fundamentally affects the question how the Gael found their way to Ireland. The belief is general that the Cjanric branch and the Gaelic branch of the Colts, after their supposed separation from a common stem, and before they reached these islands, co- existed in a state of independence; and that the first wave of Celtic immigration to this country Avas Gaelic rather than Cymric. No evidence of the least weight has ever been offered in support of that theory. Unable to account other- wise for the fact that the remains of the Gaulish language which survive, are plainly identical in their essence with THE RACES OF IRELAND AND SCOTLAND. 81 modern Welsh, rather than with Gaelic, philologists have been driven to the assumption that the Gaelic form of Celtic "was the language of those tribes who crossed the English Channel long before the ancestors of the Welsh people left their homes in Gaul.!*^ Where these Gaelic tribes were located on the Continent no one can say. They might have dropped from the clouds, or emerged from subterranean dwellings, for all that is known about them. Sir John Rhys, it is true, has done all that learned ingenuity is capable of accomplishing, by identifying them with the Celtae, and by attributing a Gaelic origin to the characters inscribed on the bronze calendar found at Coligny, near Lyons, in 1897. But another eminent Celtic scholar, the late Dr. MacBain of Inverness, was equally convinced that the characters on the calendar are akin to Cymric; and Sir John Rhjs himself was fain to confess that he could not explain how the Celtae reached Ireland. ii There is, in fact, no satisfactory proof that the Gaelic language, as a distinct branch of Celtic, originated on the Continent. On the contrary, the proofs are cumulative that it was formed and partly developed in Ireland; and that its traces in England and Wales, and its introduction as a spoken language into Scotland and the Isle of Man, equally derive their source from the country of its origin, namely, Ireland. We must now return to the Irish legends, and see what light they throw upon this question. "" It need hardlj' be said that the Welsh, Uke their neighbours, are a mixed race. The short, dark element in the population of Wales is notably large. This type represents, in my opinion, the predecessors of the true Celts ; it has certainly no affinity with the Gauls of the classical authors. ^^ Proceed! )i(/s of the British Academy (1905), p. 63. "There is no record," said Huxley, "of GaeUc being spoken anywhere save in Ireland, Scotland, and the Isle of Man " {Critiques and Addresses (1873), p. 176). Kuno Meyer is still more positive. " No Gael," he says, " ever set foot on British soil save from a vessel that had put out from Ireland " (cited, with approval, by MacBain in his edition of Skene's Highlanders of Scot- land, p. 383). 6 CHAPTER VIII. The four stocks of the Gael — The Irish genealogies and their value — The historical aspect of the Milesian legend — Spain and the Milesians — The system of the Dinnsenchus — The different names applied to Ireland — An explanation of the Milesian names — The Basques or Vascones — A Basque element in the population of Ireland — The location of the Milesian tribes. Irish tradition traces the descent of the Gael from four stocks, eponymised asi Hiber (or Eber), Heremon (or Eiremon), and Ir, the three sons of Miledh; in other words, the three warrior peoples. The fourth eponym is Ith, who was a nephew of Miledh. All four stocks came from Spain. According to the legends, the first to arrive in Ireland was Ith, who was slain by the Dananns, whereupon the sons of 'IMiledh avenged his death, and wrested the island from the Dananns. The three sons ^ then established the Milesian dynasty in Ireland, the south of the island falling to the share of Hiber, and the north to Heremon and Ir. A struggle for hegemony took place, resulting in the successive subjection of Hiber and Ir by Heremon, who became finally the undisputed master of the country. Now, this story is sometimes treated as strict historical fact by Irish writers, who believe that Hiber, Heremon, Ir, and Ith were actual leaders of the Gael, and came over to Ireland from Spain in the manner described by the legend. If this belief is entertained in modern times, it is not surprising to find that the mediaeval Irish bards and shan- achies gave it full credence. Upon the genealogists — and the Celts have always revelled in genealogics^was imposed ' There were really six sons, but three of them do not survive in Irish legend. THE RACES OF IRELAND AND SCOTLAND. 83 the task of drawing up tables of descent from the four progenitors of the Gael, thus linking them with the principal Irish families, who were proud of being provided with so illustrious an ancestry. Nothing was easier than to fabricate these genealogies; but nothing is of less historical value. It is astonishing to find writers who laugh at the genealogies of (say) Geoffrey of Monmouth, gravely accepting the Irish fabrications as genuine, and regarding as real men, instead of bardic myths, the Milesian monarchs who reigned in Ireland centuries before writing could have been known in the island. If oral tradition is capable of carrying us so far with safety, why not still farther? Why stop at 300 B.C., or at 1300 B.C.? Why not, in short, accompany the genealo- gists right back to Adam? It is impossible to place one's finger on the point in the Irish genealogies at which fiction ends and fact begins. If it is unscientific to reject them as wholly spurious, it is still more unscientific to base any sort of history upon them. That they are partly fictitious is obvious; that they are wholly fictitious is at least possible. Therefore, no space will be devoted in these pages to arguments founded upon their trustworthiness. The Milesian legend, however, at once assumes an historical aspect when we clearly grasp the idea that we are dealing, not with persons but with peoples. The im- migrations to Ireland were those of the Ithian people (or, as they are commonly called, the sons of Breogan), the Hiberian people, the Heremonian people, and the Irian people. Who were these people, and with whom are they identifiable? First of all, it is necessary to examine the tradition that they came from Spain. It is barely conceivable that in Celtic tradition, the country named " Spain" may be a vague and variable name, like Grecia in actual history. Grecia was applied sometimes to the south of Italy (on rare occasions 84 THE RACES OF IRELAND AND SCOTLAND. to the whole of Italy), and to Russia, as well as to Greece itself.- It is improbable, however, that any country but the modern Spain was intended by the Irish story of the Milesians, for, as I shall show, the evidence of etymology confirms that supposition. When the Milesian legend was invented (it is probably not so ancient as is generally supposed) to account for existing facts, it seems certain that it had, as its core, a tradition then existing of a Spanish descent for some of the inhabitants of Ireland. The most obvious fact for which an explanatory legend had to be found, to clothe this core with a suitable covering, was the name of the island itself. Now, for Ireland, there has been a plethora of names. -^ Those of the earliest appellations frequently appearing in the native texts are Eriu, Fodla, and Banba, which, in accordance with the usual system of Irish place-names, are represented in the legends as three Danann Queens. The Dinnsenchns, a lost topographical tract, attributed to the sixth century, and fragments of which are incorporated in the Book of Leinster, the Book of Ballymote, and elsewhere, is full of these personifications of place-names, and its in- fluence has been felt Avherever Irish etymologies have been discussed. The result is that, while Irish place-names have been the means of providing us with poetic legends, their real meaning in many cases has been obscured by fable. The system of the Dinnsenchus has made Irish etymology stand on its head. The legends are made to explain the place-names, instead of the place-names explaining the legends. This process is not confined to Ireland; it flourishes vigorously wherever the Celt is to be found, and wherever legends are loved. Place-names refusing to yield their - I have seen it argued that the Grec-ia of Irish tradition was located in Ireland. The tribes in Ireland called (irecru'u/he were probably ** Heath " men, not Greeks (see chapter ii.). •^ Quite a dozen can be enumerated. THE RACES OF IRELAND AND SCOTLAND. 85 secrets to local investigation have been dealt v^^ith after the manner of the bards; they have been turned into stories.* Kings, or queens, or heroes, or fair maidens living at a conveniently remote period, and bearing the same names as the places to be legendised, have been invented to play the part of the leading characters in many of those charming fables of the days of old, in the conception of which the Irish imagination was so fertile. " And that was how " such and such a place got its name. Ireland's place-names — many of them utterly prosaic in their origin — have yielded a rich harvest of fiction, of which we would not willingly be deprived. This system was actively at work when the Milesian legend was invented. What was the meaning of " lerne," the name given to Ireland by the Greeks; and particularly, what was the meaning of Hibernia, its most widely known name? And what was the meaning of Eriu, or Erin, the name by which the Gael called their island? No one knew. That is not astonishing when it is considered that even at the present day, there is no agreement among scholars. But the Irish shanachie never allowed himself to be beaten by a name. If he could not tell its meaning, he invented one. And so he invented the meaning of Hibernia in the usual way. Hibcr or Eber of the legend stands for the Iberni or Hibernians. Now Hibernia and Hybcria or Iberia are equations, and the ancient Iberia was the Greek name for Spain. Therefore Hiber, the son of Miledh, was brought from Spain to Ireland. Being the eldest son, he was the first of the Gaelic tribes to obtain a footing there, for that seems a reasonable implication for the story, Heremon, the second son, who secured the hegemony of Ireland, is repre- sented as a conqueror by his name, which signifies lord, or * A practice humorously recommended by Sir Walter Scott (see Lock- hart's Life). The Binnsenchns was centuries ahead of Scott. 86 THE RACES OF IRELAND AND SCOTLAND. master.^ Ir, the youngest of the three, is the eponym of Eriu or Erin, more clearly shown in the later names, Ir-land (for so it was frequently written) and Ireland. The name //• seems to be closely related to that of Ith. The latter is the name given inj the legend to Miledh's nephew, who, on account of the scarcity of corn, went from Spain to Ireland to spy the richness of the land. Now the word Ith (Cymric) means " corn," and the name of Miledh's nephew is at once explained by the nature of his errand. Ir (Cymric), on the other hand, means "luxuriant, or juicy, or green"; and the names Ith and Ir may be held to signify the division of the island into corn-land {1th) and pasture-land (Jr). Ireland of old was famed for its pastures: according to Pomponius Mela (first century), the luxuriance of grass was so great as to cause the cattle to burst! Hence the tribes of Ir are associated with pasturage and the Ithians with agriculture, for the ultimate meanings of Ir and Ith are capable of that interpretation. Related to Ir is the Cymric Irain (full of juice, or luxuriance, or greenness), which closely resembles Irin, the name given by Diodorus Siculus to Ireland. This, then, may be the real source of the name Erin; it means, in effect, the Emerald Isle. It is usually argued that Erin is an oblique case of Eriu; but that is doubtful, for both forms appear in the nominative. Either way, the root remains unaffected. To the Greek writers, Ireland was known as lerne (variants Juberna, Juverna, and Iverna) of which Hibernia is the Latin form. In the Patrician manuscripts, the usual form is Hybernia, though in some, the forms Hyberia and Yberia appear,^ thus showing clearly that Hibernia and Hiberia or '•' This is probably the meaning of the name Armin, the celebrated leader of the Cheruscans at the c-omraencement of the Christian era, and perhaps of Eormen-ric, the famous Ostrogothic Emperor of the fourth century. The word appears in Gaelic as armunn, a chief. * See Haddan and Stubbs, p. 318. THE RACES OF IRELAND AND SCOTLAND. 87 Iberia are the same names. The Spanish Iberia takes its name from the dominant people of the peninsula, who lived in the valley of the river Ebro, anciently the Iber or Hiber7 Similarly the Irish Hibernia or Iberia takes its name from the Iberni (the Hiber of the legend), the dominant people of the island, who lived in the valley of the Ivernus or lerne, now the river Maine in Kerry, or possibly the Kenmare. These people, and the river from which they took their name, both appear in Ptolemy's map of the second century. Here we have another instance of the arbitrary methods of anthropology, in labelling a particular type of cranium and pigmentation " Iberian." Like " Celtic," the word " Iberian " can be interpreted in one way by the anthro- pologist, and in quite a different way by the philologist or the historian. What, it may be asked, are the physical characteristics of the mixed people known to historians as the Ccltiberians? And is the " Iberian " type to be regarded as implying the whole " Mediterranean " stock, the short, swarthy longheads; or as being synonymous with the Basques, who, according to competent observers,^ are mainly neither short, nor swarthy, nor remarkably dolichocephalic? The Basques or Vascones — and here we have a link that almost certainly connects Ireland with Spain — must be the Irish Vascons or Bhascans, whose seat was the Sceligs or Scillies off Cape Bolus in Kerry. ^ There are frequent references in the Irish texts to the Clan Baeiscne as an element in the Fianna, and Finn him- self, the Fianna's chief, was believed by some to have derived his origin from the clan.^^ I find confirmation of the Irish texts in one of the Scottish collections of Ossianic remains, ^ Cym. Eh, issuing out ; Ehru, to pass out. ^ See Wentworth Webster on The Basque and the Kelt. • Betham, The Gaul and C_i/mhri, p. 241. '" Finn, son of Cumhall, son of Sualtach, son of Baeiscne {Silva Gadelica (Eng. text), p. J)9). 88 THE RACES OF IRELAND AND SCOTLAND. in which the sons of Fingal are described as " children of Baoisge." ^^ Indirect corroboration of the connexion is furnished by other facts. The Basques are a fine, athletic people, with a wonderfully upright carriage, due, apparently, to an unusually strong posterior base of the skull. They are exceedingly fond of athletic games, in which they display remarkable skill and strength. They have a dance like the Highland Fling, and an agglutinative language which "the devil studied for seven years without learning more than throe words." That there is a Basque element in the Irish population seems, on the whole, to be highly probable. But it would be unsafe, with our imperfect knowledge of the Basque language, to assume from fancied resemblances, the presence of Basque roots in the Gaelic language, or in the Irish or Scottish place-names. The root ur in river-names is frequently quoted as derived from the Basque ura, water, but, as we shall see, it comes more probably from an Aryan source. On the other hand, it would be still more risky to assert that Basque elements are entirely lacking in the language and the topography of the Gael. The question is at present not resolvable with certainty one way or the other. One thing, however, is certain: that the prognathism of Ireland does not come from the Basques, one of whose distinguishing characteristics is extreme orthognathism. To sum up: it is clear enough that the authors of the Milesian legend, as already suggested, had a definite traditional basis for assuming a pre-historic connexion between Spain and Ireland, which they adapted to their story of the Gael. That connexion, and the mode of using it in the legend, are clearly shown by the way in which the authors accounted for the Brigantes, a tribe located by Ptolemy's map in the south-cast of Ireland. Orosius (fourth or fifth century) mentions Brigantium as a place near Corunna, on an island " M'Callum, p. I.H. Baoisge— Biscay (Vasconia). THE RACES OF IRELAND AND SCOTLAND. 89 adjacent to which was a celebrated lighthouse. The Irish ]\Iilesian fable calls this lighthouse Breogan's Tower, and makes Breogan (otherwise Ith) the progenitor of the Clanna Breogan; in other wards, the Brigantes. The south of Ireland was occupied by the Brigantes (Breogan or Ith) and the Iberni (Hiber), and the legend is thus in agreement with the facts in stating that Hiber and Ith took possession of the south. The tribal name Brigantes has, of course, nothing to do with the Galician Brigantia. It means simply the Highlanders,!^ in allusion to the mountains of Kilkenny which formed part of the tribal lands. Similarly, the Brigantes of Britain — the most powerful of the British tribes — occupied the Highlands of the North of England; and there was a tribe of the Alps that bore the same name. Finally, a connexion with Spain was suggested by the names Galicia and the Gallseci, which names were inevitably linked with that of Gael by the authors of the legend. Some of the modern inquirers have fallen into the same error : they suppose that "Gael" and "Galicia" have a common meaning. I shall show presently what " Gael " really means; meanwhile, I am following the Milesian legend back to its sources. In the following chapters, we shall inquire into the origin of the Gael and of the people in Ireland who were called " Scots." '■- Gym. J}r!(/ari(, a Highlander. CHAPTER IX. The Iberians in Ireland — The origin of the Scots — A summary of t-on- clusions as to the origin of the Gael — The earliest notices of Ireland by classical authors — Ireland in the second century A.n. — Early Teutonic settlements in Ireland — The earliest mention of the Scots — The Scottish hegemony in Ireland — Tacitus and Ireland — The Cherusci and the Scots. Whether or not the dim figures which appear in the Irish accounts of the earliest colonies of the island, are intended to represent the neolithic tribes, including the dolmen builders, the existence of these elements in the ethnology of Ireland is amply proved by the facts of archaeology. There is no certain evidence of palaeolithic man in the island, but neolithic man is well represented at the present day by the short, swarthy longheads, who are to be found in abundance in the west and south-west. These are the so-called " Iberians," an unfortunate name to adopt, unless its meaning is well understood. The type is that of the Silures of Tacitus, the swarthy people of South Wales, who may have had a Spanish origin, as implied by the Roman historian.! The succession of metal- using men who came after the neolithic age is too ill-defined to permit of dogmatic assertion, but the Irish texts clearly suggest the concurrent 1 Miss Bryant, in her Celtic Ireland (p. 17), quotes Colmenar, a Spanish author, who states that " history informs us "" that in 200 n.c. the Biscayans took possession of Ireland, having crossed the sea in "vessels made of the trunks of trees hollowed and covered with leather." The Bay must have been abnormally smooth ! And Camden (Ed. 1695, p. 574) quotes another Spanish author, Florianus del Campo, who finds the Silures in Spain. Conceivably there may have been Silures on the River Sil, in north-west Spain ; hence, probably, the tribal name (Sii, in combination with the Basque urn, water). THE RACES OF IRELAND AND SCOTLAND. 91 use of stone and bronze; and there can be little doubt that the displacement of bronze by iron was similarly gradual. Coming to historic times, it is of more immediate interest, as bearing upon our subject, that Borlase points out that many objects in the Museum of Art of the Royal Irish Academy are comparable with those of the Merovingian period from the fifth to the eighth century. 2 And these are the objects that pertain to the Gael. This brings us to the question: who were the Gael? But before I answer that question, it is necessary to look more closely into the origin of the Scots, I have shown the probable grounds on which the Milesian legend is based, and have given reasons for supposing that the tradition of a pre-Gaelic immigration to Ireland from Spain is far from being without a solid foundation. I have shown, also, that the wanderings of the mythical progenitors of the Gael, prior to their supposed settlement in Spain, relate to the Scythians, who are equated with the Scots. A Scythic association with Spain is suggested by the fact that there was a Cantabrian promontory called Scythicum. All that can be said with certainty about these Scythians is, that they were Teutonic tribes who came from that part of Northern Europe known in mediaeval times as Scythia. That there were Teutonic invasions of Spain before the settlements of the Vandals, Alani, and Suevi, early in the fifth century, is shown by the irruptions of the Cimbri early in the second, and of the Franks in the third century. And there is some ground for believing that there were Teutons in Spain in the first century, and perhaps even earlier. According to Pliny, the Oretani of Spain were Germans, and Seneca, himself a Spaniard, alludes to the Germans having crossed the Pyrenees. If, therefore, it can be shown that the Scots were a Teutonic people, there is nothing in- herently improbable, though there is no actual proof, in ' Do/mtiitt, pp. 106,5-6. 92 THE RACES OF IRELAND AND SCOTLAND. the supposition that they may have passed over from Spain to Ireland. 3 But there is not a shred of evidence to suggest that the Celtic tribes in Ireland, included (as we shall see) in the Gael, entered Ireland as a colony from Spain. On the contrary, all the facts are opposed to any such hypothesis. My position is (1) that before the Gael as a national name came into existence, the dominant people in Ireland were immigrants from Britain, sj^eaking the same language (Cymric) as the Britons; (2) that at some unknown period, apparently between the commencement of the Christian era and the fourth century, Teutonic tribes Avho may have first come as raiders, and then as settlers, added an importajit element, probably in a gradually increasing volume, to the population; (3) that these Teutonic tribes in combination appear in history for the first time, in the fourth century, under the name of " Scots "; (4) that a struggle took place between the Celts and the Teutons for the hegemony of Ireland; and (5) finally, that the two races coalesced and formed the Gael; the Teutons, who were without women, marrying Celtic wives and adopting the Celtic language, while adding Teutonic elements which, in combination with the main Celtic fabric, brought into being the branch of the Celtic language known as Gaelic, or the language of the Gael.4 It will be seen that this theory of the origin of the Gael is in sharp conflict with the prevailing notions on the subject. Adequate proof will be required, or at any rate reasonable * The Teutonic vessels were not "hollow trunks covered with leather," nor were they wickerwood covered with ox-hides. * This theory does not, of course, exclude other elements from Gaelic : it merely states the main elements. The Irish traditions themselves suggest an abnormal mixture of elements, by the statement that Gaelic was formed from the seventy-two languages of the world {S^iichitK Mdr) ; and Giraldus Cambrensis, in the twelfth century, seems to have had a similar idea when he said that the Irish language is, " as it were, a com- pound of all other languages." THE RACES OF IRELAND AND SCOTLAND. 93 grounds must be stated, for the tenabiiity of the belief. It has ah-eady been shown that the Milesian legend suggests symbolically a welding of two peoples into the Gael, more particularly by means of the figure employed in the " marriage " of Scota. Thus, to commence with, there is a legendary basis for the belief, whatever its value may be. We turn now to the historical proofs. The earliest historical notices of the Irish people are vague and unconvincing. Strabo's account shows clearly that \ery little was known about them at the commencement of the Christian era. He describes them as being cannibals, and without the least sense of decency. But he was honest enough to add that what he related was " without competent authority." In effect, he gave these reports for what tlieyi were worth, w4iich was probably very little, if anything at all. It is curious, however, to note that even in Strabo's day, a Scythian connexion with Ireland was apparently recognized, for, in describing one of the unpleasant practices of the Irish, he proceeds to add that it was said to be a: *' Scythian " custom.^ The first sure starting-point for studying the ethnology of Ireland is provided by Ptolemy's invaluable map of (circa) 160 A.D. There we find Ireland of the second century with the tribal names of the occupiers, and with many place- names, some of which have persisted in a slightly altered form down to the present day. An examination of the tribal names clearly reveals the fact that they are of two kinds, topographical and non-topographical. That is to say, most of the names are derived from the character or location of the lands occupied by the tribes (e.g., the Iberni and the Brigantes already noticed), while others are obviously not to be interpreted in that way. The names in the first class * Mela (first century) and Solinus (second or third century) agree in calling the Irish " barbarians," but their information may not have been more exact than that of Strabo. 94 THE RACES OF IRELAND AND SCOTLAND. may bo fairly regarded as native, and those in the second class as foreign. In the latter class are two that are at once recognisable as foreign tribes, viz., the Cauci and the Manapii or Menapii.^ The Cauci (or Chauci, as the name is more frequently written) were a tribe concerning whom Tacitus tells ujs that they' were " the noblest of the Germans "; an unprovocative people; not given to rapine or plunder; yet possessing a military reputation that protected them against aggression. Pliny gives a different description of them; he says that the maritime Chauci were a proud, but poverty-stricken collection of miserable fishermen. About the end of the first century, the Chauci were settled on the coast, from the Ems to the Elbe, adjoining on the west the lands of the Frisians, by whom they were ultimately absorbed. It is a remarkable fact that about 162 a.d. (approximately the date of Ptolemy's map), the Chauci are known to have appeared in the Northern Ocean as pirates, and to have devastated the coasts of Gaul and Britain.'^ Therefore, if Tacitus was correct, their character must have changed in two generations. At any rate, the Chaucian raids show that the Teutonic tribes were not strangers to Britain about the middle of the second century ; and it seems certain that if they knew Britain well, they must have known Ireland equally well. A fertile and rich country, relatively speaking, as all reliable accounts represent it to have been, Ireland must have attracted the attention of the Teutonic rovers before the second century. It is usual to regard the first Teutonic and piratical descents on Ireland as having occurred at the end of the eighth century. Nothing could be more remote from probability. The Scandinavian forays in Ireland, which commenced late in the eighth, and were continued with in- •* Their association with the Menapian or Menavian island (Man) is doubtful. ^ Menzel, Jlistory of German i/, i., p. 105. THE RACES OF IRELAND AND SCOTLAND. 95 creasing frequency and ferocity in the ninth and later centuries, present merely a later phase of the Teutonic danger. Centuries before, Ireland had felt the weight of the Teutonic hand. The bands of wandering warriors from the Rhine to the Elbe, and the hardy Vikings from Scandinavia (the home of the best type of seaman) whoj came to Ireland as robbers, sometimes remained as settlers. Evidence is not lacking to show that the pressure of rival tribes, and perhaps even more frequently the operation of economic causes, such as famine,^ instigated the piratical raids of the Teutonic tribes more powerfully than the mere love of adventure, or even the innate greed for plunder. The fat pasture-lands of Erin offered a tempting asylum to these hungry hordes, and it may be that the dark-eyed " Iberian " women and the golden-tressed Celtic maidens w^ere not without an attractive force. These Teutons beheld the land, and found it a land flowing with milk and honey; a land where the cattle " burst " with the luxuriance of the pastures (Mela); and where fair women were perhaps not unwilling to accept as husbands the redoubtable wanderers from the waste of waters. What wonder, therefore, that we find Teutonic settlers in Ireland in the second century? The Menapii may have emigrated from the Ehine about the same time as the Chauci left their German home. It is sometimes stated that the Continental Alenapii were a Celtic tribe, but those who contend that they were Germans, like the Chauci, seem to me to have all the best of the argument. These two tribes were situated in Leinster.^ I believe that they may be regarded as forming the nucleus of the union of Teutonic tribes who were afterwards called the Scots. To those who say that the Scots were the original Gaelic * Bosworth, The OrUjln of English, Germanic, and Scandinavian Lan- iinages and Nations (note by J. H. Halbertsraa, p. 53). ' The Blani or Ehlani were in the same district, and it has been sug- ges ed by at least one writer that they were from the Elbe (Elbani). 96 THE RACES OF IRELAND AND SCOTLAND. (and Celtic) inhabitants of Ireland, and that they were there before the comniencenient of written history, there is a con- clusive answer: the name " Scots " does not appear on Ptolemy's map. If there were Scots in Ireland in the second century, it is inconceivable that the name should have been utterly unknown to Ptolemy. It is still more inconceivable that if there were Scots in Ireland centuries before the Christian era, as some writers seem to suppose, no contem- porary writers before the fourth century should have any knowledge of the name. For it is the fact that the first contemporary (Ammian Marcellin) allusion to the Scots occurs about the year 364. The inference, therefore, is unavoidable that the Scots were not the dominant people in Ireland until about the fourth century, and that their hegemony of the island cannot have been of very long duration before that date, otherwise their presence could hardly have escaped the notice of con- temporaries. When the Scots first appear in history, it is in the Teutonic character of a fierce, restless, marauding people, who crossed from Ireland to Britain for plunder, and fled thither when pursued. It is especially noticeable that before the name " Scot " meets us, Ireland gave no trouble to its bigger neighbour. The inhabitants, indeed, had acquired such a reputation for peaceful tractableness, that Agricola (so Tacitus says) believed that a single legion and a few auxiliaries would be sufficient to conquer and keep them in subjection. It is quite certain that if a people like the Scots had ruled Ireland in Roman times, no such con- ception of them could have been possible to Agricola. It cannot be objected that he had no first-hand knowledge of the island, for he had his facts from a petty Irish king, who had been expelled from his country. It is unfortunate that Tacitus has told us nothing about this king: what language he spoke, and what information, other than military, he gave to Agricola. We may perhaps infer from the silence of THE RACES OF IRELAND AND SCOTLAND. 97 the historian that the language of the Irishman was the same as that of the Britons ; and it is a reasonable assumption that the state of peace which prevailed between Ireland and Britain prior to the Scots coming upon the scene, was due to a community of race and language. Tacitus expressly states that in manners and disposition there was very little difference between the two peoples. Far otherwise was it when the Scots appeared, for the earliest notices of this people show them harassing, with merciless and persistent ferocity, the unfortunate Britons of modern Scotland. It does not necessarily follow that because we hear nothing of the Scots until the fourth century, they were a new race of immigrants who had recently arrived in Ireland. On the contrary, it may be assumed, with a far greater degree of probability, that they were a combination of tribes united by race and language, some of whom had been settled in the island for a more or less lengthy period under their tribal names, before they formed part of the gentes to whom the name of Scots was given, perhaps not earlier than the fourth century. We never hear of the Cauchi or the Menapii in Ireland after the second century. Beyond doubt, they figure in Irish history in later times under a different name or names. Analogies can be cited to show that such changes of name were not infrequent. The Continental Cauchi them- selves disappeared from history and reappeared as Frisians. The Cheruscans are mentioned for the last time by Claudian, and when they disappeared, the Saxons came upon the scene. From this circumstance, Latham drew the conclusion that they were the same people,!*^ as no doubt they were, for the identification of the Cherusci with the Old Saxons is easy.^^ And here an interesting fact is revealed as affecting the Scots. In the preface to the Acts of St. Cadroe, written in ^^ Latham, (rermani/ of Tacifnn, p. 13L " Armin, the hero of the Cherusci, was deified by the Old Saxons, as proved by the Irmin cult of Westphalia. 7 98 THE RACES OF IRELAND AND SCOTLAND. the eleventh century, it is stated that the Chorischii (who are brought from Greece) took possession of a part of Ireland, whence they passed over to Scotland, to which country they gave the name of " Scotia " after the wife of a certain son of ^neas the Lacedaemonian, called Nelus or Niulus, who married an Egyptian wife named Scota. Here we have the old legend in a new dress. I have a suspicion that this story may point to an immigration of the Cherusci to Ireland (which might explain the name Armun (lord) or Heremon in the Milesian legend) and their subsequent inclusion in the Scottic peoples. That other Teutonic tribes arrived in Ireland after the Chauci and IMenapii, there cannot, in my opinion, be the least doubt. These tribes were in a constant state of flux during the early centuries of the Christian era; and it can hardly be supposed that so tempting a country as Ireland would be overlooked. It is difficult to resist the conclusion that the Scots, who were sufficiently powerful to impose the name of Scotia upon Ireland as early as the beginning of the seventh century, and probably earlier, were a numerically strong combination, and included all the Teutons in Ireland, with the exception of those who were called in Irish tradition the Irians, in Irish history the Cruithne, and by modern writers, the Irish Picts.12 '- Welsh tradition offers striking corroboration of the statements in the foregoing pages on early settlements of Scandinavians in Ireland. According to the lola MSS., a Scandinavian king named Don came from Lochlyn and conquered Ireland. In 367 A.n. he led a mixed force of Scandinavians and Irish to Gwjmedd in North Wales. Don's son, Gwydion, we are told, was " highly celebrated for knowledge and science," and he was the first to teach the Welsh "the plays of illusion and phantasm." Here we have the Danann characteristics clearly sug- gested. In a Welsh genealogy, Serigi, a descendant of Don, is called a Gvniddellan (Gael), and his people are called Gun/ddsJ Ffrhti, or Gaelic Picts. The true meaning of Gael, Goldel, or Gvv/ddel (a Cymric form of a non-Celtic name), will be shown in a later chapter. CHAPTER X. Various hypotheses concerning the name "Scot" — Isidore's blunder — Geoffrey of Monmouth and his value as an historian — The Hibernians and the Scots^An analysis of the name "Scot"— St. Patrick's distinction between the Scots and the Hibernians— Ireland indiffer- ently named Scotia and Hibernia— The Ard-ri(jh of Tara — Ireland's Heroic Age— A dissertation on hair— Irish kings with Teutonic names — The Franks in the British Isles — The kilt as a Gothic dress. The name " Scot " has yielded an abundant crop of etymologies, some ingenious, others ingenuous, all pro- visional. I have laid stress upon the obvious equation by mediaeval writers of Scot with " Scyth," and have shown, too, how " Scythian," as a distinctive name, was associated in the Middle Ages with the Goths. Sometimes the Gothic tribes, especially the Ostrogoths, were called Scythians; sometimes the whole Teutonic race (including of course the Scandinavians) seem to be embraced in the name; but always the Scots. That fact of itself supplies an argument for the Teutonic origin of the Scots, but I do not wish to emphasise it. I believe that in its primary sense, the name "Scot" does not mean Scythian, though both words may have a common root. Gibbon thought that "Scot" meant " wanderer ";i Dr. Macpherson " boat-man," or a word with a similar signification; and Whitley Stokes gives it the meaning of "owner" or "master" (shot, property). The most curious etymology that I know of is that which equates the name with Picti, the painted or tattooed men. By associating the word " Scot " with the Welsh ysgythru, Gaelic sgath, and 0. Irish scothaim ("lop off'" or "cut"), a 1 Scots have always been wanderers, from the first time they appear in history down to the present day. 100 THE RACES OF IRELAND AND SCOTLAND. strained interpretation is obtained, in order to Kt in with the statement of Isidore of Seville that in their own language the Scots are so called from their painted bodies.- It is a fact that what Isidore says about the Scots in this statement is precisely the same as Avhat the Pictish Chronicle says about the Picts. Both statements are obviously copied from the same source, or one is copied from the other. Isidore lived in the seventh century, and the Pictish Chronicle is clearly of a much later date. Therefore, the conclusion is unavoid- able that the compiler of the Pictish Chronicle copied his statement from Isidore, correcting his palpable blunder by substituting " Picts " for " Scots "; or that both statements were copied — in Isidore's case incorrectly, and in the other case accurately — from some manuscript of a date not later, and possibly earlier than, the seventh century. There is still another alternative. The mistake in Isidore may be that of an ignorant transcriber. Whatever the explanation, it is evident, from the sense of the context, that in Isidore's statement the Avord " Scots " is erroneously written for " Picts." This, indeed, is apparent from Isidore's own description of the Picts, which repeats in effect, though not literally, what he says about the Scots. Yet it is on this palpable confusion of name by Isidore, that eminent philologists found their equation of " Scot " with the Latin meaning of " Pict." From one point of view, it is not surprising that a writer of the seventh century should confuse the two peoples, or even identify them Avith one another, for, as indirect testimony shows, their racial affinities were pronounced. Geoffrey of Monmouth asserts that the Scots were the offspring of the Picts and the Irisli. It is usual to scoff at evidence taken from so uncritical an historian, but that attitude can easily be too rigid, for there - Sir John Rhys was probably the first to suggest this etymoloey {Celtic Ih-lta'm, 1884, pp. 240-41), and he has been followed by Dr. Mac Bain (Skene's JH/jhlnnders, edition 190-2, p. 385). THE RACES OF IRELAND AND SCOTLAND. 101 is obviously a good deal of genuine history in Geoffrey. It must be remembered that he professed to be the editor, not the author, of the work that goes by his name, and was probably not primarily responsible for the eponymic and other fables which he records with indiscriminating faith- fulness. Whether, in his statements about the Scots, he writes as an author or an editor, it is difficult to say. In, either case, it is clear that at least as early as the twelfth century, and probably much earlier, there was in existence a tradition, or at any rate a belief; (first) that the Picts married Irish women, and that the Scots were descended from that union; and (second) that the Hibernians of Ireland were a different people from the Scots of Ireland. Geoffrey dis- tinguishes between the two in the clearest possible manner; and, as I shall show later on, he is supported by the state- ments of writers whose authority is unimpeachable. I find an illuminating use of the word " Scot " in a specimen of Danska Tunga, or Old Danish, believed to date from the first part of the seventh century. The word is applied to Rolf Krake, a celebrated northern warrior. He is called Hrolfr Skjotandi, which is rendered in modern Danish as Rolf den Skytte, and equated with Rolvus jacidator, i.e., Rolf the dart-man. In Old Icelandic Skot 'means missile; Skjota means to shoot with a weapon; and the Scots appear in the Icelandic Sagas as Skotar. The modern Danish Skytte is Scytte ^ in Anglo-Saxon, and Henry of Huntingdon calls the Scots gentes Scitiae. We can thus understand not only how Anglo-Saxon writers wrote Scot as Scyth, but why the word Scot was identified with the word Scythian. It may be added that in Old Frisian, Skot means a missive weapon, and the word is sometimes written Scot and Scote. '■' In The Anglo-Saxon Chronicle " Scottish " is written Sri/(f.l.se. Anglo- Saxon Sciftta = Sagittar'ms. The English words "shot" and "shoot" are, of course, from the same source as all the examples cited in the text. 102 THE RACES OF IRELAND AND SCOTLAND. It would appear, therefore, that the meaning of the name " Scot " is missile-man, and especially dart-man. The name thus falls within the same category as Angles {Angel, a sharp-pointed hook); Saxons {Seax or Sax, a knife or short sword); Franks {Franca, a javelin);* and Lombards or Longbardi {Bart or Barda, a broad axe). This classification may be called the etymology of the characteristic weapon. Of all the derivations just given, none is more convincing than the equation of " Scot " with " dart-man." There is proof of the dart being the characteristic weapon of the Scots. Personifying Britannia in one of his eulogies of the Roman General, Stilicho (himself a Goth), Claudian says that she (Britannia), thanks to her deliverer, fears not Scottish darts {tela). If this allusion means anything at all, it means that the dart was the weapon specially associated with the Scots. -^ The Scots were, therefore, a people with a Teutonic name, and with characteristics whicli can only be explained by, attributing to them a Teutonic origin. We have caught a glimpse of some of the German tribes that appear to have been included in the Scottic combination; and I shall give evidence to show that the language they spoke was probably akin to Platt-Deutsch or Low German. In the meantime, the distinction between the Teutonic Scots of Ireland and the Celtic Hibernians of Ireland must again be emphasised. There is no evidence on this question equal in value to the testimony of St. Patrick; and St. Patrick carefully dis- criminated between the Scottish reguli, the ruling caste in * The usual derivation, " free-men," is not convincing, for were not the other German tribes quite as "free" as the Franks? Whether the other etymologies are well-founded or not, the word "Saxon" is almost certainly derived from tSVw or Sa-v. ^ The German infantry in the time of Tacitus, as he plainly testifies, were skilled in the use of missile weapons. The youth about to assume arms, says the same historian, was equipped with a shield and javelin. The contemporary account of the battle of Clontarf (see Todd's Thf War of the UaedliU irlfh fhf Gdill) states that the Gael had "darts with silken strings, thick set with shining nails, to be violently cast " at the enemy. THE RACES OF IRELAND AND SCOTLAND. 103 Ireland, and the Hibernians, the mass of the people.^ And Giraldus Cambrensis (twelfth century) makes a notably clear statement on this matter when he tells us that the Scotti and Hibernenses had a general name which comprehended both, viz.: Gattheli or Gael. Thus we sec that the two different peoples, the Hibernians and the Scots had become fused into what may be fairly described as a nation, and that the national name was "Gael." This fusion must have been preceded by a long- and fierce struggle for hegemony . The fight for the mastery is typified by the accounts in the IMilesian legend of the varying fortunes of Eber, Ith, Ir, and Heremon. First the mixed tribes in the south and south-west, comprising the Iberni and the Brigantes (Eber and Ith) and others; and then the Scandinavian and other tribes in the north (Ir) and north-west, fell under the domination of the people of Low German extraction, whose centre of settlement was in Leinster (Heremon), but who must have exercised supremacy over a much wider area. These people were the Scots of his- tory. Their predominance among the Gael is shown by the fact that by the seventh century, the whole of Ireland was called indifferently by the names " Scotia " '' and "Hibernia"; and the name "Scot" was understood by foreigners as an inhabitant of Ireland.^ The Ard-righ or High King of Ireland, whose seat was at Teamrah or Tara, in Meath, was chosen now from one of the four groups of free peoples who composed the Gael, and now from another. The High Kingship was vested in the royal family of whatever group happened to be the most powerful; and it was thus a symbol of hegemony. All Irish traditions '^ Coii/fsslon, and Letter to Corotlcus (both considered genuine). See Haddan & Stubbs, ii., pp. 308 and 317. " The earhest known use of the word Scotland as a name for Ireland is by St. Laurence, Archbishop of Canterbury, at the beginning of the seventh century (Bede, B. ii., c. i). * Bede gives the general name of *' Scots " to the Irish. 104 THE RACES OF IRELAND AND SCOTLAND. agree in assigning to the descendants of Heremon (the Scots), a position of supremacy in the island, which gave them an. undisputed right to the throne of Tara. Thus they are in substantial accord with the testimony of St. Patrick, in the distinction he draws between the Scottish reguli and the Hibernians. In the Irish texts dealing with the early centuries of the Christian era — Ireland's Heroic Age — there are glimpses given, not only of the structure of society, but of the physical appearance of the Gaelic warriors. These descriptions no doubt faithfully reflect the traditional notions about these warriors, and in substance they may not be inaccurate. The simpler the description, the earlier the source. The later compilers and redactors in Ireland may be easily recognised by their profusion of adjectives. They had to cover the barrenness of their knowledge with the flowers of their rhetoric; and they were all accomplished rhetoricians. Then, as now, a multiplicity of adjectives accompanied a paucity, of ideas; but the earlier pictures of the Gael are not dis- figured by the excess of colour that characterises the decadent period. On reading these Irish accounts, one is struck by ■what would seem to be the great importance attached to the hair, the two most desirable qualities being length and fair- ness. An essay on the hair and its significance, from an historical and ethnological standpoint, would have to take account of two tribal groups in ancient Germany, the Suevi and the Franks. Tacitus, in his characteristically terse manner, marks off the Suevi from the rest of the Germans by their mode of dressing the hair. Among the Franks, long hair was the sign of a free nian.^ And Tacitus assures us that ruddy hair was a German characteristic. As a contrast to the freemen, the slaves in ancient Germany wore cropped hair, or their heads were completely shaven. '•' Latham's (lermanji of Taritus, p. 109. THE RACES OF IRELAND AND SCOTLAND. 105 We find precisely the same customs in ancient Hibernia. Long hair and fair hair— whether the " locks of golden gleam " of the Celt, or the ruddier hue of the Teuton — were regarded in the Heroic Age, or at any rate in the estimation of those who preserved the traditions of that Age, to bo characteristics of the aristocratic Gael. And, as in the case of the German slaves, so in that of the Irish bondmen, the latter had cropped or shaven heads. " They crop their hair," says Saxo Grammaticus, writing about the Irish, " close with razors, and shave all the hair off the back of the head, that they may not seized by it when they run away." ^^ Saxo's reason is not conclusive; on the contrary, it stamps him as a hitherto unsuspected humorist. The Irish serfs were shaved for the same reason as the German slaves: to distinguish them from the long-haired free-men. Probably they consisted mainly of the pre-Celtic tribes, the short, dark people physically contrasted in every conceivable way with the fair, big-bodied, long-haired, and blue-eyed tribes of Celtic or Teutonic origin. There is a Celtic word (Cym. moel, Gae. mael) which is applied sometimes to bare hill-tops, and sometimes to bald- headed men. But in a secondary sense, it implies servitude, and forms a prefixial element in Christian names denoting subjection. The familiar Scottish name " Malcolm," for example, means Mael-colum, the slave or bondman of Colura or Columba. This name, again, is cognate with Gille-colum, which has a similar signification, mael having a Celtic and gille (A. S. cild, Scots chiel) a Teutonic source, though now a typical Gaelic word.n A slave, i.e., a cropped or shaven person, and a child are both under subjection; hence the *^' Saxo Granunaficvn (Elton), p. 205. •' GiUe (cf. a sportsman's gillie) has the same force (servant) as the old meaning attached to the word "chiel" in Lowland Scotland. The Irish form is gilla, modern t/iolla. Examples of f/!ll« in modern names are Gilmore (servant of Mary) and Gilchrist (Christ's servant). There is a third class of word (Irish mot/h or niuf/) with a similar meaning. 106 THE RACES OF IRELAND AND SCOTLAND. applicability of such names to those who in adopting them wished, in Christian times, to crave the protection of their patron saints, though the custom has a pagan origin. A tentative suggestion is, that the tonsure itself may be remotely associated with this idea of servitude and humility. At any rate, the usual explanation that it originally; symbolised the Crown of Thorns placed on the Head of the Divine Master, is not satisfying, inasmuch as it suggests what is not a fact, viz., the Christian origin of tonsures. The Hibernian tonsure was not even a crown — it was a half-crown. The head was shaven from ear to ear, making the tonsure semi-circular or crescent-shaped, and with the fringe which was alloAved to remain in front, faintly suggesting, it may be, the tracing of an axe-head. Mr. Ua Clerigh thinks that this may be the explanation of St. Patrick's clerics being called Tailceanns or Axe-Heads (usually but less correctly translated " Adze-Heads ") by the ethnic Irish; and it seems plausible enough. ^^ In any. case, there is nothing inherently improbable in the suggestion that the tonsure of an Irish monk may have originally been the badge of servus Dei^'-^ A.11 this may appear discursive, but it has a bearing upon the point I wish to establish, viz., the existence of parallel customs among the ancient Germans and the ancient Irish, wdiich cannot certainly be traced, at any rate Avith the same distinctness, to other peoples. The inference is obvious. It serves to accentuate the evidence from other sources, of a 1- f/ls/on/ of Jr/'hin(l to fh' ('omitiff of Henri/ //., p. 209. See Bade on the Hibernian tonsure, or, as it was contemptuously called by the Petrine tonsurists, the tonsure of Simon Magus. There was a third (Eastern) tonsure, viz., that of St. Paul, which involved the shaving of the whole of the head. "The origin of "Culdee" is, I think, Clilfl, O'll, or Cil-De - God's servant. The Latin form of the name was Kehdei. In the Irish form, Cf'de l)e, we have the same word (reilf) as is applied in the ancient laws of Ireland to the dependants (servants) of the (j'r«d Flaith or chiefs. THE RACES OF IRELAND AND SCOTLAND. 107 Teutonic occupation of Ireland before authentic Irish history commenced to be written. The very Krst Irish King whom we meet in the earliest epoch of Irish history, of which the records can be regarded as trustworthy, had a Francic name, for Leoghaire (now Leary) who was the heathen Ard-Righ, when St. Patrick landed in Ireland, had the same name as the Merovingian Lothaire. Suibone or Suibue (now Sweeny) a Pictish name that is frequently seen in the Irish annals, denotes Swedish lineage (the Swens or Suiones as Tacitus calls them); Amalgaidh, a name that appears in the list of Irish kings, is Gothic and distinguished, for the Amalings were the royal family of the Ostrogoths, as were the Baitings of the Visigoths. ^^ Where did these Franks and Swedes and Goths come from, and when did they reach Ireland, assuming their presence there? No one can say with certainty, but that the Franks showed marked restlessness during the third century can be easily shown. Soon after the middle of that century, they ravaged Spain incessantly, especially the north and east coasts. They made more than one descent on Africa. They were so troublesome that Probus, about 277, caused several thousands of them to be transported to the borders of the Black Sea. But he could not repress them, nor induce them to settle down as peaceful colonists. They seized a fleet which lay at anchor in the Black Sea, sailed to the Archipelago, plundered the wealthy maritime cities, and landed in Sicily, where they took Syracuse. They fought the Romans below the walls of ^* Amnl is a Gothic word, meaning "mighty." Aedh and Aidan, names that are specially associated with the Scots, suggest the Anglo- Saxon E(ul (Gothic And), "prosperity" {rf. Edwin). The Fenian names Oisin and Oscar seem to contain the Teutonic Os, demigod. Oscar is purely Teutonic. 0* is generally associated with the royal (god-born) race of Northumberland (Kemble). Diarmid, also, has apparently the Teutonic prefix DIora or f>lnrn (Old German) found in compound names (carus). Aidan is found as the name of an Anglo-Saxon bishop in an Anglo-Saxon poem of the tenth century. 108 THE RACES OF IRELAND AND SCOTLAND. Carthage, and, being defeated, retreated to their ships, sailed through the Mediterranean, and, coasting Spain and Gaul, returned laden with Avealth to their own country. ^^ It was partly by means of Franks, too, that Carausius established his power in Britain; and when his assassin and successor, Allectus, was himself defeated and slain in a battle fought with the army of Constantino Chlorus in 296, the hordes of plundering foreigners who were chased out of Britain were composed mainly of Franks. ^^ Is it reasonable to suppose that a country like Ireland, endowed with a fertile soil, and enriched with mines of gold, would escape complete immunity from the visits of these rovers; or that so attractive a resting-place would fail to induce many of them to become permanent settlers in an island unprotected by the strong arm of the Roman, and already affording an asylum to their kinsmen from the Frisian coast? There was a proverb that said: " Choose the Frank for a friend but not for a neighbour." Conceivably, the native tribes in Ireland realised its truth in the third century, if the Catti, Cauci, Cherusci, and the other members of the confederacy called the Franks, became members of the Irish confederacy known as the Gael.^'^' '5 Menzel's lUstonjof Gcrmami, i., p. 113 (see Zosimus and Eumeuicus). 1" Latham, FAhnnlogti of flu Jlrltish Jxlmuh, pp. 96-7. The Franks "sacked London." Bede says that in the time of Carausius, the sea- coasts were " infested by the Franks and Saxons " (B. i., c. 6). '" The following note has a bearing upon the question of Gothic settle- ments in Ireland. Camden, quoting Sidonius, describes the apparel of the ancient Goths in these words (trans. 1695, pp. cxvii.-viii.) : — " They shine," says he (l.c.y Sidonius), " with yellow ; they cover their feet as high as the ancle with hairy untann'd leather ; their knees, legs, and calfs are all bare. Their garment is high, close, and of sundry colours, hardly reaching down to their knees. Their sleeves only cover the upper parts of their arms. Their inner coat is green and edged with red fringe. Their belts hang down from the shoulder. The lappets of their ears are cover'd with locks of hair hanging over them. . . . Their arms are hooked spears (which Gildas terms irticlnafa fi'la) and hatchets THE RACES OF IRELAND AND SCOTLAND. 109 to Hing. They wore also strait bodied coats (as Porphyrio says) without girdles." Camden adds : *' If this is not the very habit of the Irish-Scots, I appeal to their own judgments." The coincidence is certainly remarkable. In his '* Letters to Cynthia," Propertius alludes to what Professor Phillimore translates as '* wintry Goths." Camden reads this as " Irish " Getes. The confusion is between hiherna and Hibernia ; and there is a similar instance in Giles' edition of The Anglo-Sa.von Chronicle, where the statement is made that Caesar left his army to abide " among the Scots," instead of "into winter quarters." The mistake arose, as Ingram points out, from the inaccurately written MSS. of Orosius and Bede containing the words in Hifhernia and hi Hiherniam instead of iti hiherun. " Wintry Goths " is paralleled by Claudian's " icy lerne." Is there not a play upon words here ? lerne is not, and never was, in historical times, "icy." CHAPTER XL The Gael —The silence of early writers on the name — Bede's evidence on the root ddl —An analysis of the name "Gael" — The Brehon Laws and Teutonic parallels — Cuchullin : man or myth? — The Finn Saga and its historical basis— The Fianna as professional champions — Scandinavian parallels — The dominant races described by the Sfiichus Mur — The meanings of the provincial names. For a satisfying etymology of the word "Gael" (or "Goidel" as it is now usually spelt by philologists), one may search in vain. It seems to be regarded as a fact to be accepted, but not to be explained; a name to be gloried in but not to be analysed. With a Celtic probe no analysis is possible: there is no root, either in Cymric or Gaelic, to which the most imaginative etymologist can point as the source of the word. Dr. MacBain, an acknowledged authority on the Gaelic language, thought that the earliest form of the word must have been Gddilas or Gaidelas from a root gad (English " good "), but the suggestion does not carry us very far. He was compelled to look to a Teutonic source for the root; but it eluded him notwithstanding. There is no hint of the Avord in the Patrician documents, from which fact it may bo assumed that St. Patrick had no knowledge of it as a national designation. In the poetry attributed to St. Columba, the name is found (" Farewell to Erin "), and it appears also in the Amra or Elegy on Columcille (St. Columba), ascribed to the saint's contem- porary, Dalian Forgaill. The authenticity of these poems has been strongly questioned, and it is not a little curious in corroboration of this criticism, that Adamnan's Life of Columha, an admittedly genuine work, should contain not a sinffle allusion which would lead us to believe that the THE RACES OF IRELAND AND SCOTLAND. Ill name Gael was applied in his time (seventh century) to the people of Ireland, or any section of the people. Adamnan, himself an Irish monk, had an intimate knowledge of Irish affairs; and it is incredible that the word would have es- caped him had it been adopted as a national name. His Life is full of allusions to the " Scots " and the " Hibernians " of Ireland, and to the Scottic language and the Hibernian language (in his day they were synonyms); but there is not a word to show that he had ever heard either of the Gael or the Gaelic language. If native writers of the seventh century say nothing about the Gael, it would seem hopeless to look to other sources for the name. Neither in Gildas (supposed sixth century) nor in Nennius (supposed eighth or ninth century), both Cymric Avriters, nor in Bede (seventh century), an Anglo-Saxon historian, does the word appear. Yet in Bede there is a word which, I am convinced, is the root for which we have been searching: it is the word dal. When relating the settlement of the Dalreudini, an Irish tribe, in Dalriada (Argyllshire) Bede explains the name by saying that it meant the share of portion of Re'uda, " for in the language of the Scots (Irish), dal means a part." ^ This statement by Bede has a double significance: it gives us the meaning of the word Gael, and it suggests the archi- tectural method by which the Gaelic language was built up, as well as indicating one of the sources of the material employed. Taking the second point first, we find the word used in the nominative case, as well as in oblique forms, by the invaluable Cormac. Thus we know at least what it was like in the ninth (or tenth) century. We find it in the nominative plural as Gcedel, and in oblique forms as Gdidel and Gcedelu. These are substantially the forms employed ^ Ecchxil in ileal Jlrxtori/, B. i., c. L 112 THE RACES OF IRELAND AND SCOTLAND. by all the early writers in Ireland ^ who mention the name, except that there is a tendency to convert what was apparently the earlier prefixial gm into goe or goi, thus making the form Goidel, as it is frequently written at the present day. The popular form " Gael " preserves the earlier "a" and sheds the " d," owing to its quiescence in pro- nunciation, Gcedel thus becoming Gael (as it was formerly spelt) or " Gael," as at the present day, when we are in too much of a hurry to use the diaresis. But how, it may be asked, is Gcedel to be connected with Bede's dal, " which in the language of the Scots means a share " ? The answer is that it is the same word, with m Teutonic prefix (gae or ge), corresponding to the Latin con, and signifying coUectiveness. The earliest mention of the word in a Teutonic language appears in Ulphilas, i.e., in the translation into Gothic by Bishop Ulphilas in the fourth century, of a portion of the Gospels. This translation is preserved in the Silver Book, now in the University of Upsal, and numerous editions exist. I find that Ulphila's Gothic word for co-partner is Ga-daiJa. I find that in Anglo-Saxon Ge-ddl means a division, or parting, or distribution {ge-ddelan, to divide or share), and that Geddl-Iand means land belonging to several proprietors. This is simply Bede's dal, a share, with the usual prefix " ge " (Latin con). It is evident, therefore, that the mean- ing of Gadel, or Gael, is simply co-sharers or co-dividers. . The root dail, dal, or del, denoting a share, is widely distributed throughout the Teutonic languages. It appears in English in the word " deal " (as in dealing cards) and " dole " (as in doling alms), and in every instance it conveys the idea of a division. The word is characteristically Teutonic, and is not derived from a Celtic or Latin source. ' Cf. Secundius Hymn (preface); Leahhar Breac : Homily on St. Patrick; and Gilla Caomain : Chronological poem. THE KACES OF IRELAND AND SCOTLAND. 113 The conclusion is, therefore, irresistible, that it was incorporated in the Gaelic language by Teutonic contact. And that conclusion becomes a practical certainty, when we find that it is confirmed by other evidence of a similar kind. The Ancient Laws of Ireland, which embody the most important texts of the Brehon Law Tracts, afford ample illustration of the methods of land and stock-sharing that; characterised the polity of the Gael. The Senchus Mor '(i.e., the great ancient traditions), a oomjDilation of uncertain date, contains some of the best known of these Laws, which, being founded upon the unwritten traditional jurisprudence of the Brehons or Judges, possessed the invaluable sanction of custom. Their texture is thus interwoven with a social system that stretches far back in the history of Ireland, and by their means the historian and the ethnologist, as well as the jurist, are able to deduce certain conclusions that may be accepted as reliable. These Brehon Laws have been thoroughly analysed and discussed by competent authorities, and it is not my purpose to tread in their footsteps. But one clear factor that emerges from the Irish Laws may be here emphasised: and that is the grouping of society into units which were cemented by blood-relationship. The tribal system was beyond doubt in full operation. The tribe was not a family, but a group of kindred, to which was given the name of fine, and the fine was sub-divided into four " hearths," or grades of kinship. That fundamental fact affected in a marked degree the law relating to eric — the Teutonic tveregild — which lay at the root of the criminal procedure. It formed the co-partnery basis of the system of land tenure, for the division of land was tribal, each group being assigned its share, doubtless by lot. The pasture and waste land was common property, each fine, or group of kinsmen, having definite rights of pasturage which were jealously guarded. This was exactly 8 114 THE RACES OF IRELAND AND SCOTLAND. the system described by Caesar as practised by the German tribes. With them, as with the Gael, tillage land was apportioned among the family groups related by blood; and in both cases, also, the tenure was annual, each group being compelled to till fresh land every year. Caesar gives a series of reasons for the prevalence of this custom; reasons sound at the core, and, under a tribal system, altogether beneficent. The custom has come down to modern times in the run-rig system which prevailed in the Highlands of Scotland until the nineteenth century, but there was this important differ- ence: that whereas the original custom implied tribal occupation and ownership, and a regard for common interests, the modern system lacked both the stimulus of individual ownership and the ancient bond of clan senti- ment. It was, in fact, a shadow without the substance; an anachronism without either sentimental or economic war- rant; and it gave place to a system that, whatever its faults, was founded upon a recognition of facts. The resemblance between the ancient customs of Ireland, as shown by the Brehon Tracts, and those of the German tribes, as described by Roman writers, is too close to be explicable by assuming for those customs a common Aryan origin. It is true that in the ancient Cymric Laws, for example, we find many parallels with the Brehon and the Teutonic laws: as, for instance, the Cymric galanas, which is the Irish eric and the German iveregild, namely, a scale of compensation for crime. ^ We find, too, that the peculiar custom of gavelkind was observed alike in Ireland, Wales, and Kent. Still more striking is the fact that the custom of "fasting " upon a creditor (an integral part of the Lawt of Distress) was not only recognised by the jurisprudence of ^ The operation of eric was between group and group. Crime inside a group was punished by expulsion. The expelled members were called " Kin-wrecked " men. In the clan days of the Highlands, they were called "broken men." THE RACES OF IRELAND AND SCOTLAND. 115 ancient Wales, but may be found in full operation at the present day among certain tribes in India, thus pointing obviously to an Aryan source as the origin of the custom. There are distinct traces, too, in the Cymric Laws of fosterage; but it may be a question whether or not this was a native or a borrowed custom. It is certain that nowhere outside of Teutonic countries (particularly Scandinavia) and Ireland, and Scotland, do we find such remarkable examples of the persistence of this institution, and the amazing devotion which it inspired (especially between foster- brothers) as the histories of these countries afford.* The social conditions illustrated by the Brehon Laws are in harmony with those legends, (based upon genuine traditions) that are frequently treated as if they were accepted history. The Heroic Age of Ireland is regarded as representing a phase of history by some who reject the pre-Heroic Ages as fabulous. On the other hand, there are those of the mythological school who regard Cuchullin (or Cuehulain) as a sun-god. Both points of view are pro- bably untenable. It is just as easy to believe in the super- human feats of Cuchullin, as it is to believe that no such hero ever existed. The rational view to take is, that before history was written in Ireland, tradition preserved the memory of a champion, super-eminent for his feats of strength and skill, around whose person had clustered a series of legends, some well-grounded, some baseless, and others derived from actual incidents entirely dissociated, perhaps, from the romantic hero whom the bards called Cuchullin. To attempt to associate this hero with totemism, on the strength of his name (Cu, a hound) and a specific case of tabu in the legend, is venturesome. We read the story o-f ■• The whole subject of tribal custom is exhaustively dealt with in Seebohm's The Trihal Si/,sfem in ]VaIes and Tribal Cnntom in Anglo-Saxon Law. See also Maine's Ancient Irish Laws and Skene's Celtic Scotland, vol. iii. 116 THE RACES OF IRELAND AND SCOTLAND. Cuchullin and Queen Meve of Connaught, and the Red Branch Knights of Ulster, and Ferdiad the doughty Firbolg champion, as contained in the Tain bo Cuailnge (? the cow-drive), with the feeling that the bards have worked up a marvellously interesting story out of scanty material, for we may depend upon it that the more precise the details in these Irish Sagas, the more active has been the play of the bardic imagination. Yet it is impossible to escape from a sense of reality. The correct atmosphere is there; the social picture is painted by impressionists; but the colours are in harmony with what we know to be the true scheme; and we realise that we are looking upon a large truth half obliterated by a mass of small lies. These stories, therefore, — whether they relate to the first century or a later period — possess a true historic value: they are really excellent specimens of historical romances, with more romance than history, but with the history illuminated by the romance. The Finn Saga may be reasonably placed in the same category; but the evidence in favour of an historical basis is stronger. I see no reason to doubt that a certain society of champions, having as leader a man of remarkable prowess (who, during or after his life-time was called Finn or Fionn; possibly an eponymic name) attained such distinction among the other champion corps of Ireland, as to cause the name of Fianna of Leinster to become synonymous with physical strength and vigour, and to give their name to the Irish language ae a word for "giant" and " champion." On this hypothesis, there is nothing that requires explanation in the fact that the Finn legends are common both to Ireland and Scotland, for the Irish immigrants who colonised Scotland brought their traditions with them; and among these, the story of Finn MacCumhall and the Fianna whom he commanded, was perhaps the most widely known and the most tenaciously cherished. THE RACES OF IRELAND AND SCOTLAND. 117 I have already remarked upon the athletic skill of tho Basques ; and if athleticism were a Basque monopoly — which it is not and never has been — that fact would strongly corroborate the Basque origin of the Fianna. For they were magnificent athletes, carefully picked for their physical strength, and carefully trained to perform Fiannic — if not Titanic — feats. " There were seven score officers, each man of these having thrice nine warriors, every one bound to certain conditions of service." These conditions are de- tailed: ^ they are sufficiently exacting to appal an Olympic champion. And to crown all, in addition to the feats of sheer strength and physical skill, each man " must be a prime " poet versed in the twelve books of poesy " — surely a fine example of mens sana in sano corpore. The size of the champion is stated with a fine eye to contrast. In the " Colloquy with the Ancients," the few remaining members of the Fianna (including Oisin) by a feat of bardic imagina- tion, are brought down to Christian times, and hold a con- versation with some followers of St. Patrick. The largest of St. Patrick's clerics, we are told, " reached but to the waist, or else to the shoulders, of any given one of the others, and they sitting." This was the bardic way of saying that the Fianna were above the ordinary height. Societies of professional champions were recognised in Scandinavia as a useful institution. The members of these societies, and likewise individual champions, wandered over the country, offering their services to those who were ready, to give them the most liberal remuneration. Like the Fianna of Ireland, they had to offer proofs of their prowess, and they were ever ready for the test. The berserks (wearers of bearskins) were professional champions; their fits of frenzy and consequent running amok made them particularly awkward persons to have a disagreement with. The Irish Fianna seem to have a Teutonic name, for ' Silva Gadelica (English text), p. 100. 118 THE RACES OF IRELAND AND SCOTLAND. Feimiedha, an authoritative form of the word, may with some reason be interpreted as " enemies," ^ a name that must necessarily have been given to the Fianna by foes who spoke a Teutonic tongue. It is not improbable that there was a Scandinavian element in the composition of the Fianna. The mother of Finn himself was a daughter of the King of Lochlan, according to one of the Scoto-Fenian tales,'' and some of the Fenian names, as already shown, are plainly Teutonic. If we are to believe tradition, the services of the Fianna were employed mainly against Firbolgs and Scandinavians. In the first century — ^so the tradition runs — Tuathal, the High King (of the Heremonian line) imposed a horoma, or cow-tax, upon the Fir-Gaileoin, a Firbolgic people of Leinster, and the exactions gradually became more oppressive until, finally, the tribute-payers seem to have been goaded into revolt. They were then supplanted in Leinster by the Heremonians, or Scots, who employed a body of militia (the Fianna) to aid them as a fighting force in carrying out their policy. The Fianna are placed in the third century by tradition, and that is just the century during which I have assumed that some of the later Teutonic settlements in Ireland took place. Ptolemy proves that there were Teutons in Leinster in the second century, and the historical evidence, on the whole, seems to consist well with the traditional hints of a political upheaval in that province between the second and " O. Ic, Fjande ; O. H. G., Fiant ; O. Sax., Fhtml — all meaning "enemy." This is a purely Teutonic word, showing, in contrast to hostk, the objective attitude of the Teuton towards his foes, who were "the haters" (Ger. feind, the hater). The English word "fiend" is a development of this idea. According to tradition, the main duty of the Fianna consisted in defending the boundaries of the High King of Erin. 7 J. F. Campbell's West Highland Tales, xi., 349-350. Campbell (i., 62) remarks on the similarity between the Gaelic and the Norse stories. THE RACES OF IRELAND AND SCOTLAND. 119 fourth centuries. The Senchus Mor describes the three dominant races who were in Ireland at the time of the compilation of the Laws, as the Feini, the Ultonians, and the Laighin: "the three noble tribes who divided this island," as they were called. The Ultonians {JJluid or Vitas) were the people of Ulster, the so-called tribes of Ir; the Feini were associated with Tara in Meath; and the Laighin were the Leinster men. These names have been the subject of much speculation. JJluid or Vita has proved so jDuzzling that no real attempt has been made to give an etymology of the word. Probably it is a form of Cymric Gwellt, grass, signifying a grazing country ,8 which, again, fits in with the etymology of If — signifying greenness and juiciness. Feini (not to be con- fused with the Fianna) has been interpreted as meaning "farmers"^ (O 'Curry), "masters" (Atkinson), and " Phoenicians " (Shaw). But with greater probability Feini is derived from Old Icelandic Venja, to teach, for the Feini were the law-givers, and their eponym, Fenius Farsa, was a " school-master " in Scythia. Laighin or Leinster as ai place-name is interpreted in a curious fashion. In his analysis of the contents of the Book of Leinster, Dr. Atkin- son quotes the legend that " Leinster took its name from the broad lances {lag in) brought by the Black Gaill across the sea when they came with Labraid Longseck." The Black Gaill, or Dugalls, must surely be an allusion to a Scandi- navian settlement. " Broad lances " is an etymology that still holds the field, but its improbability is obvious. The name is with far greater likelihood derived from Old Icelandic laegd, a low-lying place, to which Anglo-Saxon leag (Leigh), a grassy plain, is probably related. This « According to Camden (trans. 1695), the Irish form in his day was Cui Guilli/, i.e., province of Guilly. s Old Icelandic, Vln, Old Fris., Fenne, pasture, would consist with that derivation. 120 THE RACES OF IRELAND AND SCOTLAND. derivation correctly describes the great litnestoae plain of Leinster. How the Scandinavian " ster " (sceter, mountain-pastures) was applied to three out of the four provinces of Ireland has never, I think, been satisfactorily explained. It seems to denote a wider range of! l^ordic influence than Irish historians are willing at present to admit. Possibly, too,, further study of the obscure Feni dialect, which has com- pletely puzzled Irish scholars, will show that contrary to the theory underlying all past attempts to find a key to its m.ysteries — "sages" in the dialect are mentioned in the Annals — its basis may be Teutonic rather than Celtic. The " language of the Feni " may prove to be an element in what one may call "Gaelic in the making." And now we have to consider what Gaelic really is. CHAPTER XII. The Gael and the Gaelic language in Ireland — How the Gaelic language was formed — St. Patrick and education in Ireland— Tradition and the ancient tongue of Ireland — AlH/etorki — The Latin element in Gaelic — Ptolemy's map of Ireland — An analysis of tlie Ptolemaic names in Ireland — The general structure of the Gaelic language — Some Scandinavian legacies — The views of Dr. Joyce — Bishop MacCarthy on the Irish Picts. Before there was a Gaelic language, it is obvious that there must have been a Gaelic people. The earliest un- disputed examples of what is now the Gaelic language are to be found in Adamnan's Life of St.Columha. In his Latin text, there are several words, chiefly place-names, in what he calls indifferently the " Scottish " and " Hibernian " tongue. In Bede's text, a few Gaelic words are also dis- coverable: they are called " Scottish " words: belonging to the language of the Scots of Ireland. Thus the people afterwards called the Gael, were in the seventh and eighth centuries known only by the names of Scots and Hibernians. It cannot be asserted that the names " Gael " and " Gaelic " were never applied to the people and the language during or before those centuries: all that can be said is that there' is no reliable evidence to show that the names were so used. A lack of discrimination between all these names is commonly shown in treatises dealing with affairs in mediaeval and pre-medipeval Ireland. And the confusion is accentuated by the fact, that after the Scots of Ireland became the Scots of Scotland, there were two Scots peoples in Scotland. There were those who by the Lowlanders were known as " the old Scots," and who spoke a language which the Lowlanders called "Irish": but their own name for them- 122 THE RACES OF IRELAND AND SCOTLAND. selves was " Gael," and for their language, " Gaelic." There were also the Low-country Scots who spoke a Teutonic language, but who retained the name of Scots (a name repudiated by the Highlanders) after abandoning gradually, as the result of social contact with their predecessoi's in Scot- land, the " Scots " tongue which their ancestors had brought with them from Ireland. But these fundamental, though frequently overlooked, facts in Scottish ethnology will be examined more closely in the proper place. How was this Hibernian, or Scottish, or Gaelic language formed by the Hibernians, Scots, or Gael? It must have been built up slowly and gradually, like all mixed languages resulting from inter-racial contact. Originally Cymric, like the British language, with whatever non-Aryan and pre- Celtic elements the Celts may have borrowed from their predecessors, the Irish vocabulary was enlarged and enriched by the addition of many words derived from the languages (differing only dialectically) of the various Teutonic hordes that arrived in ever-increasing numbers in the fruitful isle of the west. And when St. Patrick came to Ireland in the fifth century, the lingual development was rapid. Latin words were grafted in large numbers upon the Celto- Teutonic stem; and Ireland then learned, apparently for the first time, to read and write. It is a hard saying to many Irishmen, and in their view an incredible statement, that before the arrival of St. Patrick, Ireland had no written language. They point in refutation to the Ogam characters, and the wonderful learning of their country in mediaeval times. The Ogam argument is now rarely heard, for it is no longer tenable. Though the script is probably very ancient, its probable use as literary machinery has never been seriously suggested. As for the remarkable outburst of literary activity which followed the introduction of Christianity, the evidences of which arc furnished alike by Christian missions and by THE RACES OF IRELAND AND SCOTLAND. 123 Christian manuscripts that found their way all over Europe, it is sufficient to say that the birth of learning in Ireland was the outcome of the educational stimulus supplied by St. Patrick and his followers. The thirst for knowledge and the desire to employ a newly-found means of expression, hitherto denied to the zealous adherents of the new faith, were impulses derived from the work of the great missionary and educationist, whose name will be for ever associated with that of his adopted country. What do we learn from tradition about the ancient tongue of Ireland? There is a legend that the Irish language was formed by Gadel Glas, the eponymous of the Gael, from the seventy-two languages of the world. Another version tells us that Fenius Farsa sent from his school in Scythia his eeventy-two disciples, to learn the various languages then spoken throughout the world. ^ By a natural process of reasoning, the Feni of Meath are derived from this Fenius. According to Flaherty, Fenius composed the alphabets of the Hebrews, Greeks, and Latins, the Beth-luis-non (apparently a Runic alphabet like the Futhork) and the Ogam. This information he professed to take from For- cherne, an Irish poet, who was said to have lived before the Christian era — a statement that is hardly less sus- picious than the story about the existence of Fenius himself. It would seem probable that the whole of this legend is based upon a confused knowledge of the invention of the alphabet by the Phoenicians, who are eponymised by the name of Fenius. Or, Fenius may have been invented as the ancestor of the Feni of Meath, an undoubtedly historical people, who, by the topsy-turvy process usual in such cases, are said to have derived their name from Fenius, instead of Fenius taking his existence from them. There is a statement by Nennius which throws light upon the method employed by St. Patrick, in combining the use * Senchus Mor, i., p. 2L 124 THE RACES OF IRELAND AND SCOTLAND. of letters in Ireland with the spread of Christianity. He informs us that Patrick wrote over three hundred abgetoria, which word has been interpreted by Ware and others to mean the alphabet. ^ Tirechan, in his life of the saint, asserts that the latter baptized men daily, and taught oo.' read to them letters or abgotories; and he gives specific instances of these letters having been written for converts. Father Innes, in his celebrated essay on " The Ancient Inhabitants of Scotland,"^ quotes the glossary of Du Cange to prove that, in the Middle Ages, the words Abgatorium, Abcturium, Abecenarium, Abecenarium, and Abecedarium were used to express the A. B. C. or alphabet; and in support of his contention that the Irish received the know- ledge of letters through the Latin language, Innes argues, shrewdly enough, that the Gaelic words for a letter, a book, to read, and to write, are all derived from Latin.* Tho Romans never having entered Ireland, a knowledge of those words could only have been obtained from St. Patrick and other Latin-speaking missionaries, who find- ing no equivalents in the Irish language, expressed them in Latin, giving them only an Irish inflexion. This is not merely a plausible argument, but a fairly convincing proof that the use of letters in Ireland coincided with the intro- duction of Christianity; and it is reinforced by the fact*, that the ecclesiastical terms relating to the Christian religion which have been incorjjorated in the Gaelic language, are derived from Latin. This we should expect; but it affords presumptive evidence that the two sets of terms, literary and ecclesiastical, both obtained through a Latin medium, were introduced by the same Latin-speaking instructors, the Christian missionaries. What share, if any, Palladius had in promoting the use of letters during his short and un- 2 The word used by Nennius is Abietoria, which Innes reads as Ahgetoria. 3 Innes (1885), pp. 246-7. ♦ Ibid. THE RACES OF IRELAND AND SCOTLAND. 125 successful mission in Ireland can only be conjectured; ^ but it is clear that the reputation justly earned by mediaeval Ireland as the Isle of Learning, as well as the Isle of Saints, must be assigned to the Patrician foundation so well and truly laid in the fifth century. But it is necessary to go further back than the time of St. Patrick, and examine, as far as the facts will permit, the groundwork of the Irish language before any incorporation of a Latin element took place. The limit of our knowledge is fixed by the date (c. 160 a.d.)^ of Ptolemy's map of Ireland. Place-names, as I shall show, supply the most use- ful ethnological proofs that we possess in determining pre- mediaeval problems; and of all place-names, the names of rivers are the most valuable, for they show conclusively that the language which they disclose was the language of the people who at one time were settled in the valleys watered by those rivers. Now what do Ptolemy's river-names of Ireland tell us? Thene is not one of them that cannot be justifiably assigned either to a Cymric or a Teutonic origin; and it is a remarkable fact that some of the principal river- names are referable, as I think, to the Scandinavian branch of the Teutonic language. I would remark in the firet place on the incidence of the root Vind in these Ptolemaic river-names. We have the Vin- derius running into Belfast Lough, but the name at its mouth appears as Logia, now Lagan, the name, also, of a river in Sweden. Logia is most probably derivable from O. Ic. logr, water. Of special importance is Ptolemy's Buvinda, which Adamnan calls the Boend, now the Boyne. The latter form of the name might be referable to Buan, SAvift (Cymric), which correctly describes the flow of the current, but the analogies cited below seem to exclude that derivation. ^ Zimmer's idea that Palladius and Patrick were one and the same person has now been generally abandoned. " Variously stated as 120, 140, 150, and 160 a.d. 126 THE RACES OF IRELAND AND SCOTLAND. According to Gluck and other scholars, the Celtic Yind comes from the same root as the Gothic hveit (English "white"). The root is Vid, German hvit, with an intrusive " n," a nasalised form of Vid. The nearest Teutonic cog- nate of Vind is Danish hvid, white, and it is a question whether Vind may not have been derived direct from that source. Cymric gwyn is cognate, but the Gaelic forms, Finde and Find, later Fin and Finn, are more akin to tlie Scandinavian (see Vidua). In Buvinda we lind Vind used to denote a river with clear water, as may be seen from the use of the later Finn in river-names. Therefore, the usual etymology of Buvinda, "the river of the white cow" (in itself an unlikely name for a river) is more than doubtful. It is true that Bede (B. IV., c. 4) translates (surely Avith the assistance of an Irishman, for Bede can scarcely have had a knowledge of Gaelic) Inisbofinde (now Inishbofin) as " the island of the white heifer ' ' ; and Bofinde is the Gaelic form of Buvind. But here fiiide implies whiteness, while in Buvinda as a river-name, it has the force of clearness. The prefix Bu must be the Gaelic Bo, cow, for Adamnan, whose native tongue was Gaelic, translates the river-name Bo (the Boyle, i.e., the Bo-ail or cow-river) as Bos. Therefore,, Buvinda seems to mean the " cow-river " (distinguished by its clear water) implying, like the name Boyle, a stream where cattle were watered.'^ A point to notice in this discussion of the name Buvinda is, that here we have in one of the eariiest, if not the earliest, example of the Celtic word for " cow," as used in Ireland, a purely Cymric form; for Bu is to-day the Welsh name for cow, as it was the Irish name in the Second century. ^ There is a pretty legend that tells us how Findloch Cera in Sligo got its name. The birds sent by God to Patrick when he was in Cruaich used to strike the water with their wings, and thus made it whiter than milk, whence the loch was called "white" (Atkinson's Introduction to the Book of Ldnster, p. 38), THE RACES OF IRELAND AND SCOTLAND. 127 By the seventh eentuiy, it was in Irish, Bo. Thus we see that the Celtic elements in what was afterwards called the Gaelic language, had diverged in the course of five centuries, from the Cymric forms of the second century. And the divergence, with some show of reason, can be assigned to the fact that during those centuries, Gaelic was in the making. In Vidua, identifiable with the River Finn, we have what is apparently the Danish form (hvid) of Vind (now Finn).^ The Vinderius, also in Ulster, is shown in Ptolemy's map as a crooked stream. It is referable probably to 0. Icelandic Vindr, awry (Dan. Vind, to turn or bend) an appropriate meaning for the Lagan, in contradistinction to the Bann, which may be derived from O. Icelandic Beinn, straight, a designation that correctly describes the course of the Lower Bann. 9 Ptolemy's name for the Bann is Argita, a Cymric word {Argae, a dam, Argau, to enclose, Argaead, a shutting in) denoting a boundary, for then as now (it separates Antrim from Londonderry) it provided a natural division. The Shannon — the pride of Ireland, called by Ptolemy the Semis — has an undeniably Teutonic form. The root Sen is apparently 0. Icelandic sen, denoting a slow motion^ wdiich is applicable to the flow of the Shannon for the greater part of its course. The Barroiv, for which Ptolemy's name is the Birgns^ probably takes its name from the same ultimate source a& Brigantes, in whose country it was situated. Birgus and Barrow are related to Cymric Bri, gen. Brig (Scot» " brae "), but Ptolemy's form suggests a derivation from a Teutonic origin: 0. Icelandic Berg, a rock or hill. The river rises among the Slievebloom Mountains on the border of King's County and Queen's County. Near the Birgus, in the country of the Manapii, Ptolemy [^^ I would remark, however, that Finn in Old Icelandic means "smooth." ' There is a River Beina in Norway, and a Bane in Lincolnshire. In the North of England, Bain means straight or direct. 128 THE RACES OF IRELAND AND SCOTLAND. places the Modana, now the Slaney, itself a Danish name {Slaaen, dull). The name Modana is nearer 0. Icelandic Moda, a large river, than Cymric Mivtli, rapid, an alter- native derivation. Ptolemy's Dabrona, now the Blackwater, can hardly be attributed to Cymric du, Gaelic dub, black. It is probably derivable from Cymric Dyfru, to water, Dyjrhynt, a w^ater- >course (Cf. Ptolemy's name Sahrina for the Severn). Ptolemy's Libnius (Liffey) is referable to Cymric Lief, a flood, Llifaiv to stream, to flow. It may be added that some of the river-names of Ireland not mentioned by Ptolemy seem to betray Scandinavian origins. The Clare is the Danish Klare, with the same meaning as the English "clear." The Erne appears in O. Icelandic as Ern, rapid or vigorous, and the Suck as Sukka, noisy. Turning to the general structure of the Gaelic language, we find incorporated in it a number of words which, beyond question, are a Scandinavian legacy. The obvious reply id this fact is, that these words Avere bequeathed to Gaelic during the raids and settlements of the Northmen, which commenced (it is supposed) at the end of the eighth century. That this explanation is not conclusive I shall prove by the examination of a single word, but a very important one, in ancient Irish: the word mocu. This word occurs over and over again in Adamnan, and was clearly understood by him to mean 'clan" or "tribe"; literally the descendant from a progenitor. It is usually translated as being synonymous with mac in modern Irish and Scottish names; while mac, in turn, is equated with the Cymric mah, older map. But the strict meaning of the Cymric word is " boy " or " son," Avhoreas the Old Irish mocu stands for descendants or posterity. The difference between the two is shown in an Ogam inscription " Maqqui Erccias Maqqui Mucoi Dovinias," cited by Rhys and Jones, which THE RACES OF IRELAND AND SCOTLAND. 129 they translate as (The monument) " of MacErce, son of the Kin of Dubinn." ^^ Here the difference between " son " (maqqui or mac) and " kin " (miu:ol or moai) is clearly brought out. But both come from the same source. The first verse of the Volu-spd in the Poetic Edda, " com- posed at so remote a period in heathen times that it is impossible now to ascertain its age," contains the words mogu Heimthallar, which in modern Danish are rendered "af Heimsdal's slaegt,'" and in Latin, "posteros HeimdalU." The Old Danish mogu is thus the exact equivalent of the Old Irish mocu, and it seems certain that they are the same words. 11 The modern Irish word Sliocht (in Scottish Gaelic Sliochd), represents the ancient mocu, and is the same as the modern Danish Slaegt. There is a further curious coincidence between the word /era, occurring in the same verse of the Volu-spd (translated as hominum) and the Gaelic fear (gen: iir) a man. The Avords are practically identical, and Cymric gwr, a man, is thus more remote from the Gaelic. The Old Danish ok (and) also approaches more closely the Old Gaelic ocus than the Welsh ac. The influence upon the Gaelic language of these Scandi- navian and other Teutonic elements, affords what I believe to be the true explanation of the phenomenon which Gaelic exhibits of substituting " q," and later, " c " ("k" sound) for the Cymric "p." An initial "p" is repugnant to the Teutonic tongue, whereas it revels in the " k " sound. If this theory is tenable, the " p " and " q " puzzle is not so puzzling after all. What are we to make of these proofs of early contact with a Northern people? If the etymologies I have suggested are accepted, assent must necessarily be given to the view that, '» The Welsh People, p. 52. ^' But in Old Icelandic m6(jr means a son or boy (the Gaelic mac and the Welsh mah), and this yields the secondary interpretation of mocu .and nuic. 130 THE RACES OF IRELAND AND SCOTLAND. as early as the second century, some of the best parts of Ireland were held by settlers who spoke a Scandinavian tongue. That is a view which, to the best of my knowledge, no Irish historian has ever taken. Nor, indeed, is it possible for anyone to accept this position, if his ethnological opinions coincide with those of Dr. Joyce, whose authority on Irish place-names is believed to be unassailable. For this is what Dr. Joyce writes: — " In our island (Ireland) there was scarcely any admixture of races till the introduction of an important English element, chiefly within the last three hundred years .... and accordingly our place-names are purely Keltic, with the exception of about a thirteenth part, which are English, and mostly of recent introduction. "^^ I can conceive of no statements that can be more easily refuted than these. No one who holds the oj)inions that they embody can ever hope to be able to solve, even partially, the puzzling problem of pre-historic Ireland. I have already shown, and I shall bring further evidence to show, that the postulate of an almost unmixed race in Ireland, prior to the sixteenth century, is fundamentally unsound. Bishop M'Carthy, in his edition of Adamnan, says that " no fact in the pagan history of Ireland is more certain than that the whole country was originally held by the Irish Picts or Irians.13 With certain reservations, I subscribe entirely to that view. It is possible to go further, and say that the people whose Ptolemaic and other river-names we have been examin- ing, included the Picts. And this brings us face to face with the question: " Who were the Picts?" '•^ The Orhjin and Historji of Irish Names of Places, p. vi. Since these lines were written, the death of Dr. Joyce has left a blank in the realm of Irish research, which, however some of his views may be regarded, cannot be easily filled. " Quoted by Mr. Wentworth Huyshe in his edition of Adamnam's- Life of St. Columba, p. 1 13. CHAPTER XIII. Antiquaries and the Picts — The different schools of theorists — The Cruithne, the Irish Picts, and the Picts of Scotland — Tighernach and the Piccardach — The meaning of '* Picars " — The Roman Picti — Picti a corrupt form of a pre-existing name — The Picts as pigmies — Sir Walter Scott on the Orcadian " Peghts " — Picts-Houses — The Picts and the elf-creed of the Teutons — The meaning of the name "Pict" — Confusion between elves and Picts — Beddoe on Ugrian thralls of the Norsemen — Finn-men and Finn-women. It is not altogether surprising that the Pict has been re- garded, sometimes as a giant, sometimes as a dwarf, some- times as a fairy, and at all times as an enigma. Ever since Sir Walter Scott — himself a sound antiquary — poked fun at the hallucinations of amiable Oldbucks obsessed by pet postulates, the Pictish question has formed a centre around which many a battle-royal between rival schools of theorists has been fought. These encounters have been sometimes positively vicious in the intensity of the acrimony aroused by them. The spectacle of normally peaceful and be- nignantly gracious antiquaries seeking (metaphorically) one another's blood, because their views on the Pictish question were divergent, shows at once the perversity of human nature, and the disabilities under which even wise men may suffer when their wisdom lacks the flavouring of humour, or their sense of proportion is temporarily thrown off its balance. And after all that has happened, the Pictish question still remains unsolved, notwithstanding some con- fident assertions to the contrary. It would serve no useful purpose to tell the story of the Pictish warfare in words, or to relate the varying success of the Gothic, the Cymric, the Gaelic, and the Ibernian Schools. At the present day, it may be said that the Gael 132 THE RACES OF IRELAND AND SCOTLAND. hold the field. Di\ Skene, the great protagonist of thei Gaelic theory, a skilled lawyer in working up a brief, % sound antiquary when unencumbered with preconceived ideas, and a persuasive writer in presenting a case, has apparently prevailed upon most of our Scottish historians to accept his vieAvs, although there is a sturdy minority of Celtic philologists who keep the Cymric flag flying. Sir John Rhys, the mainstay of the non-Aryan theory, has a small and only partially convinced following, who find in, his hypothesis a way of escape from certain difficulties that have not been met by the Celts. The Goths, whose main pillar was the pompous and pugnacious Pinkerton, are for the present a wholly discredited school. What if all this confusion has resulted from a lack of co-ordination between the different points of view? What if none of them is either wholly right or wholly Avrong? For the present, my purpose is to link the Cruithnei, commonly called the Irish Picts, with the Picts of Scotland. Irish traditions gives three names, Irians, Cruithne, and Dal n'Araidhe, to the same people; and I have already shown that Cruithne was another name for the Tuatha de Danann. While Adamnan and the Annals of Ulster do not apply the name Cruithne to the Scottish Picts, the Pictish Chronicle, St. Berehan, the Albanic Duan, the Book of Deer, and John of Fordun plainly indicate that the names are interchange- able. Also, Tighernach, regarded as the most careful as well as the earliest of the Irish Annalists, frequently applies the name Cruithne to the Picts of Scotland, whom he designates likewise Picti, Pictones, and Piccardach; and he definitely connects the Cruithne with the Dal n'Araidhe. I have stated that the name Cruithne indicates a people distinguished by their elf-beliefs, or obsessed by them, as a modern word might express their attitude. I think it can be shown still more clearly that the Picts of Scotland Avere dominated by a similar belief. But let us see, first of all, THE RACES OF IRELAND AND SCOTLAND. 133 what Tighernach means by the name Piccardach, applied by him to the Picts of Scotland. Robert of Gloucester calls the Picts " picars," " pycars," and " picardes," and an explanatory note on the name " pycars " says that they were the compnye (company) of Pittes " out of Scitie (Scythia) that some clepeth Pikerdys." To the country of the Picts he gives the names of Picardye and Picardie,! and in a curious passage, he writes of " Scottes and of Picars of Denemarck (and) of Norwei." ^ "Picars" probably means plunderers. It survives in the Scots words " pikary," rapine, and " pycker," a petty thief, which in the same sense is found in archaic English as " pykar." The word " picard " is now obsolete, but it meant a small vessel for coast or river work, and is used by Leland with a similar meaning (" picardes and small ships "). The Picts were well provided with these small craft, as Gildas and Tighernach plainly indicate. It is probable that picardes were originally piratical craft, hence the association with " picars." The word " picaroon " was applied indifferently to a pirate or a pirate vessel, and the spurce of the whoie> " pickery " group of words, as applied to theft, is apparently pecus, pecoris (French picorer), shewing that the original thieves were cattle-stealers. In any case, the connexion between the Cruithne of the Irish annalists and the Picars and Picardes of English historians is, I think, fairly estab- lished. What rivets the connexion is the free use by Tigher- nach of the name Piccardach (the exact equivalent of Picardes) for the Cruithne. But what are we to make of the name "Pict?" The Latin word Picti means the painted people. Is this the original * The history of Picardy in France suggests reasons for its name being connected with the etymology discussed here. The Somme must have swarmed with "picards" sometimes. There was an early Saxon colony in Picardy. 2 Hearne, i., p. 103. 134 THE RACES OF IRELAND AND SCOTLAND. meaning of the name, or is it a Roman corruption of a name having a totally different meaning? At one time, the former conclusion was unquestioned, but the prevailing tendency to-day is to make Picti a corrupt form of some unknown Celtic word. Probably this idea originated with the difficulty of reconciling the Latin word Picti with the Anglo-Saxon names for the Picts.^ Also, it is difficult to explain why, for the first time, the Romans should give the name of Picti to a people in North Britain, three and a half centuries after they had seen a painted people in South Britain, and long after they had come in contact with tattooers in the south of Europe. Why did they give the' name of Picti to these later people, while they gave no such name to the woad-using blueskins described by Csesar? It may be argued that it was because the Picts painted pictures on their bodies, while Caeisar's Britons dyed their skins, not for ornament, but in order to intimidate their enemies. That argument is, however, too flimsy to withstand attack. And those philologists who have sought to derive the origin of the name Picti from a Celtic source, with a meaning unconnected with the custom of painting, have every justi- fication for looking behind the Latin word. But they have not discovered the Celtic word for which they have sought. The conclusion I have reached regarding the origin of the Roman Picti is, that it must be a corruption of another name, but that this name means something entirely different from "the painted people." The Anglo-Saxon forms of the name cannot easily be an Anglo-Saxon rendering of Picti. Our early chroniclers, such as Gildas, Adamnan, Bede, and Nennius employed the Latin language as their medium of expression, and consequently used the recognised Roman name Picti. In King Alfred's translation of Bede, ^ The Anglo-Saxon forms are Peohtas, Pyghtas or Pightas, and Pehtas. The Scots forms are Peychtis (old) and Pechts (modern). The Scandinavian form is Pet or Pett. THE RACES OF IRELAND AND SCOTLAND. 135 however, he gives Bede's Picti as Peohtas. Probably he knew no more than Bede knew about the origin of the Piets. But he knew, and probably the contributors to the Anglo- Saxon Chronicle knew, that Picti was a corrupt form of the original name, which they rendered as Peohtas and Pyghtas. In order to ascertain the meaning of the word Pict, we must go to the country of the Picts, the modern Scotland, or part of it. There we find the truly remarkable fact that from the Shetlands to the Border, the prevailing tradition is that the people called Picts by the peasantry were dwarfs or pigmies, with a marked predilection for underground dwellings. The word " picht," still alive in Scottish dialect, means a "very diminutive, deformed person " (Jamieson). An excellent example of the popular notion of a Pict in the nineteenth century is given by Sir Walter Scott. He tells us (in his notes on The Pirate)^ that about 1810, a missionary, " a very little man, dark-complexioned, ill- dressed, and unshaved," arrived at North Bonaldshay, one of the Orkney Islands. The inhabitants " set him down as one of the ancient Picts, or, as they call them wdth the usual strong guttural, Peghts." They produced a pair of "very little, uncouth-looking boots, with prodigiously thick soles," and appealed to Mr. Stevenson (grandfather of Robert Louis Stevenson) whether it was possible such articles of raiment could belong to anyone but a " Peght." The attitude of the people was decidedly hostile, until they understood that they had made a mistake in assuming that the unfortunate little missionary was a Pict. The following statement shows what must have been the popular conception of a Pict in the fifteenth century: — Writing in 1443, the Bishop of Orkney states that when Harald Fairhair conquered the Orkneys in the ninth century, * See also Lockhart's LIfn of Scott, where the story is related. The Orcadians considered that the " Peghts " were " no canny." 136 THE RACES OF IRELAND AND SCOTLAND. he found that the inhabitants consisted of two nations, the Papae and the Peti. The Papae are obviously Christian anchorites, the same name that was applied by the Norsemen to the Irish monks whose relics they found in Iceland. The Peti are the Picts, as we know from the name Petlands- fjordr, given to the Petland, or Pentland, Firth in the Heimskringla, and the name Petia given to Scotland by Saxo Grammaticus. The Bishop tells us that these Orcadian Peti were dwarfs who, though of little strength, were wonderful workers in the construction of their " cities." At midday, they hid in little houses under the ground.^ These " little houses " are, of course, the underground or semi-subterranean structures called in Orkney Picts-houses, or eirde-houses, or earth-houses, of which so many examples are to be found in the islands. The architectural type includes those structures above ground in the Hebrides usually called " bee-hive " houses, and known in the seventeenth century (to quote Martin) as Tey-nin-druinich,^ literally the hunch- backs' houses, though translated as " Druid's House." We find the Ronaldshay Peght and the Bishop of Orkney's Peti reproduced in the Teutonic dwarf or elf -stories. The little missionary is a counterpart of the coarsely clad " little black men " of the German tales, or the Niss of the Swedish tales, who was the size of a baby, with an old but wise face, and who wore a coarse woollen jacket and shoes like those of peasant children. Likewise, the Bishop's Peti of Orkney, the small people who hid themselves at midday, must surely be identical with the dwarfs of Northern mythology, who shun the light, the legend being that, if surprised by tha breaking forth of day, they became changed to stone. In the Alvis-mal, it is related that the dwarf Alvis had been promised Thor's daughter in marriage, but when he went ^ See The Bannatyne Miscellany p. 43, for the original quotation. * Martin's Western Islands (1884.), p. 154'. Druinneach means in Gaelic " hump-backed." THE RACES OF IRELAND AND SCOTLAND. 137 to fetch his bride, Thor cunningly detained him all night,, by asking him various questions, until the dawn, when thg dwarf, being one of those genii who shun the light of day, was obliged to depart without the bride. '^ The Peti of Orkney and the " trows," or trolls, of Shetland have all the characteristics of the Scandinavian dwarfs.^ The importance which the elf-beliefs assumed in the imagination of the Scandinavians is clearly shewn by tho fact that in Ulfiiot's Laws, it was ordered that the figure- head (a dragon) of every ship should be taken off before she came in sight of land, lest the gaping head and threatening beak should frighten the land-vcettir, the tutelary genii of the country.^ It Avould not, therefore, be surprising to find that among the names given to the Scandinavians by the other races with whom they came in contact, one of them should relate to this dominating creed. And it would seem that we find that name in the word " Pict." The word, " Pict," I believe, is derived from the same source as the English " petty " and the French petit. Accor- ing to Skeat, the origin of " petty " and petit is the Gaulish petti, from which root also comes the Wallachian pitic, a dwarf. The modern Welsh representative of petti is peth, a thing. The Teutonic cognates, or derivatives, appear to be the English "wight," German wicht, 0. Icelandic voettr, all of which imply primarily a " thing," but the words aro usually applied to a supernatural being, an elf. Similarly, the Welsh pethyn means a little thing, and pivt means any- thing that is very small; Avhile the Danish voett, an elf, " Northern Antiquities, p. 377. Du Chaillu makes a similar statement about the northern dwarfs hiding in their holes during the day, and Pennant notes the prevalent belief in Scotland about the repugnance of the fairies to the glare of daylight. * The " trows," says Scott (Lockhart's Life), do not differ from the fairies of the Lowlands or Sighean of the Highlanders. ® Landnama, c. 7, cited by Du Chaillu (i., p. 419). 138 THE RACES OF IRELAND AND SCOTLAND. is the same as the 0. Icelandic pett, a Pict, the latter aj borrowed form.^o Thus the word Pict comes, apparently, from a Celtic source, with a signification similar to that of the word Cruithne. The Scandinavian form Fett,^^ and the forms employed by English writers of the Middle Ages, associate the name with the Gaulish petti and the Welsh peth, more obviously than the picht of Scottish dialect, a word having the same radical meaning as petti, and deriving its guttural form from a Low German influence. The form " picht " may have been the source of the Latin Picti, which became stereotyped as the national name of the people to whom it was applied. The easy transmutation of the " V " (or " W' ") and " F " sounds Avith the " P " sound can be seen from examples in the various Teutonic tongues; and it is not a little curious that while the name of the Picts took, in the Teutonic languages, the initial " P " of the Cymri, the Welsh Triads took the initial " V " of the Teutons, and generally called the Picts Givyddel Vichti or Ffichti, or, as it is sometimes written, Phichtiad.^^ '" One being the (borrowed) Cymric form, and the other its Scandinavian equivalent. Pinkerton points out that the Northern nations adopted the Roman " P" to express " V " and " W." ^' In Old Icelandic, petti means a small piece of a field — a foreign word, according to Cleasby, introduced from the British Isles. ^^ In Anglo-Saxon, the Runic letter " p " was employed to represent the letter "W." It seems to have been confused sometimes with the Latin "P." For example, the name pechthelm (Weohthelm or Wecht- helm), which appears thus in Birch's Cartulariuni Saxonicum (iii. 35) must be the same name as that of Bishop Pechthelm, whom Bede mentions as a contemporary. It is difficult, indeed, to believe in the genuinely Teutonic origin of any name with an initial " P." The Welsh had another name for the Picts, viz., Peilhwr, meaning "men of the plains." A third name, Brithwr, is used in Welsh literature for the same people. Brlth means both "speckled" and "mixed." Probably it has the latter meaning when applied to the Picts, as denoting an admixture of races (cf. Brith Eingl, mongrel Angles). The name " Britons" may conceivably have the meaning of a "mixed" people, thus agreeing with the traditional composition of the Cymri. THE RACES OF IRELAND AND SCOTLAND. 139 It would appear, therefore, that the Orcadian Peti were really elves of the subterranean sort. But the Euhemeristic theory of the origin of the elf -myth in the North, appears to derive some support from certain facts of anthropology re- lating to the Shetlands. Beddoe remarks that black hair is not infrequent there, and that it is usually found in persons of a " decidedly Ugrian aspect " and melancholic tempera- ment. The same type, he adds, is found at Wick, in South Caithness, and in the north-east of Sutherland. ^^ He sug- gests that the type may represent the Ugrian thralls of the Norse invaders, or possibly some primitive Ugrian tribes. Beddoe also remarked upon a Finnish type which he had observed in the Island of Lewis,i* a type with which the present writer is acquainted. Did the elf -stories in those parts of Scotland take their rise from the presence there of Finnish thralls, who accompanied the Norse colonists,!^ either during or before the historical period? A Swedish belief is that the elves represented the souls of those who were slaves, and who tended the fields of their masters while the latter were engaged in piracy ;i^ and that belief tallies with the tradition in the Highlands that the Drinneach, or hunchbacks, were " Picts " and " labourers." The popular belief in Sweden to which I have just alluded, is really the most plausible explanation I have seen of tliQi elf-myth, for it provides a platform upon which the realist and the mythologist can both meet upon equal terms. If we postulate a slave caste of Finns, as forming part of the equipment of the Scandinavian settlers in different places of the British Isles, we can find a ready explanation of the Lapponic custom of the knotted cord, in those places where Scandinavian and Ugrian types are most prevalent, as well " The Races of Britain, p. 239. »* Ihid., p. 240. 15 It is possible that there is a radical philological connexion between " thralls " and " trolls." 1^ Northern Mi/thologi/, i., p. 93. 140 THE RACES OF IRELAND AND SCOTLAND. as the Shamanism that is such a feature of the Dananns, the Cruithne, and the Picts. We can also understand the elf- traits of the Orcadian Peti whom Harald Fairhair is said to have exterminated. There is a tradition in Shetland ^"* of certain of the natives being the descendants of " Finn-women." The Orcadian accounts of " Finn-men," who appeared occasionally on the coast in their little boats, tallj^ closely wdth the statement of Claudius Clavus (fifteenth century) about " the little pygmies a cubit high whom I have seen, after they were taken at sea in a little hide boat, which is now hanging in the cathedral at Nidaros (Trondhjem). There is likewise a long vessel of hides, which was also once taken v/ith such pygmies in it." These boats, according to Nansen, than whom there is no better authority, correspond with the Kayak and the Umiak (the women's boat) of the Eskimos. ^^ Were the Finn-men of the Orkneys Eskimos or Lapps? The " Finn " boats cai:)tured in the Orkneys and sent to Edinburgh may, with advantage, be compared with the boats at Trondhjem. '^ Tudor, The Orknei/s and Shetlands, pp. 168-9. '* Nansen, In Northern Mists, ii., p. 269. CHAPTER XIV. The various names of the Irish Picts — Rury the Great — The Golden Age of the Irish Picts — The Red Branch Knights — " Ossian " Mac- Pherson and the Irish bards — The meaning of the Irish Creeres — The destruction of Emania — The racial affinities of the Ulster Picts — The solitary word of their language analysed — The '* Danes' Cast," It was stated in the last chapter that (excluding their territorial name of Ulta or Ulster people), the Picts of Ireland were known bj three different names: the Irians, the Cruithne, and the Dal n'Araidhe, or Dalaradians. The words " Irian " and " Cruithne " have already been ex- amined. In the third name, we find the Teutonic " dal," a part or share. This root underwent a curious trans- formation in Ireland. Originally a place-name, it acquired a secondary meaning- denoting a tribe. Thus Dal n'Araidhe must have originally meant the share or portion of Araidhe, and as a fact, Dalaraidhe or Dalaradia was an Ulster place- name. But in course of time, the tribal land and the tribe alike seem to have been comprehended by the word "dal"; and latterly, Dal n'Araidhe generally meant the tribe or the descendants of Araidhe, who was killed in battle by the Heremonians or Scots in 248 a.d. He may be regarded, perhaps, as historical, though it is possible that the name is tribal rather than personal. There are parallels elsewhere (which will be noticed later) of this double meaning of "dal." But the Irish Picts had a fourth name. Believed to be descended ultimately from one Ruadhraidhe the Great, Avho is said to have commenced his reign as High King of Ireland in 288 B.C., they are frequently called the Clanna Rury in the Irish tales. It is improbable that such a person as Rury 142 THE RACES OF IRELAND AND SCOTLAND. the Great ever lived, a suspicion that is strengthened by the exactitude of his date. It was easy enough to invent a line of descent from him, with all the necessary intermediate names; but his existence is not made more convincing by a precise genealogy. The view I take of the whole of the so-called history of Ireland before the Christian era is, that it requires much stronger evidence than any that has yet been offered, to support the historical character of the numerous kings whose reigns are usually accepted as authentic. That does not imply disbelief in the general accuracy of Irish tradition; but it means that tradition is demonstrably un- able to bear the weight of precise detail with which the Irish fabricators have overloaded it. Tradition has supplied then skeleton; the bards have supplied the warm flesh, and the pulsing blood of the living story. Rury the Great may with some probability be identified with the mythic Rodric of Geoffrey of Monmouth, and Lay anion's Brut. He was the first king of the Picts to sail from " Scythia " for Scotland and Ireland; the name is the same as the Scandinavian Ruric. Ruadhraidhe (pronounced Ruari) means probably " the Red Ruler," an etymology that is more patent in what is now the English form " Roderick." The association of the Gaelic ruadh, red, with the Icelandic rjodr, red, is noticeable, all the more so, as it differs so widely from the Cymric coch, red. So, too, the Gaelic raidhe (if it means righ, king) is to be equated with the Gothic reiks, ruler, and the Scandinavian riki, kingdom. But it is believed that these words, and the whole of theic Teutonic cognates, are traceable to the Celtic rig, king, to be seen in such Gallic personal names as Cingetorix and Vercingtorix. The Latin rex is a near congener of the Gallic rix, and neither of them is far distant from the ultimate source, the Sanscrit rajan, king. But it is possible that in these Gallic names the secondary meaning of " tribes- man " may be meant by rix, for " Cyn " seems to be the THE RACES OF IRELAND AND SCOTLAND. 143 Gallic equivalent of the Cymric Cun, a leader or chief. Conceivably the latter may be the true derivation of the word " king " rather than Cyn, a tribe. The golden age of the Irish Picts was the first century a.d. That, at any rate, is the period assigned by Irish historians to the Heroic Age, and the Cuchullin saga. Whether Conall Cernach, the celebrated Ulster hero, and his king, Concobar MacNessa, and Fergus Mac Roigh (who quarrelled with Concobar) and the other champions who formed the com- munity known in Irish tradition as the Red Branch Knights — whether these were men or myths, who shall say? If they are historical, why make Cuchullin, their contemporary, a solar myth? It is true, Cuchullin performed prodigies of strength and skill that no human being has ever accom- plished, but these are merely bardic extravagances, and do. not destroy his historical character. He was the great champion of the Picts of Ulster in their struggle with th^ Connaught tribes, as described with much fertility of imagination in the Tain ho Cuailgne ; yet he himself was not, it is said, of the race of Ir. I see no reason to disbelieve in the existence of the Red Branch Knights. There is a sureness of touch about the stories of the Heroic Age that seems to show that the tradition, when first committed to writing, was well-defined. The Heroic tales have the true atmosphere of heroism, and they lack the superfluity of adjectives that inevitably betray the hand of the mediaeval inventor or redactor. It seems probable that the period should be placed later than the first, but before the fifth, century. It is impossible to re- concile the Ireland of Cuchullin with the Ireland of Tacitus; a country of warriors, with a country whose fighting qualities the Romans despised. The evidence of a reliable, historian like Tacitus must certainly be preferred to that of distorted and edited tradition. Yet both may be true, if referred to different periods. Perhaps James Macpherson 144 THE RACES OF IRELAND AND SCOTLAND. was not so far wrong, after all, in making Cuchullin and Finn (or Fingal, as he called him) contemporaries, though this is one of the points seized upon and emphasised by hostile critics in demolishing his " Ossian." There may have been too much Macpherson, and too little Ossian, in the work of genius — and a work of genius it remains in spite of everything — named the Poems of Ossian; but it has the true heroic ring, and the general picture it presents of the ethnic Gael and Pict is probably not untruthful, if allowance be made for a poet's license. Macpherson was what we call an Impressionist, and must be judged accordingly. Certainly, Macpherson is more credible in his extrava- gances than the Irish bards in theirs. The exuberance of imagination which characterised the latter, is frequently seen in their tales of the Heroic Age. But Ave have in the story of Deirdre, one of the tenderest and most moving romances that have ever come from a Gaelic pen. The three sons of Usnach, one of whom (Naoise) is the hero of this romance, were members of the first order of champions, the Red Branch Knights of Ulster, the most skilful, the most valorous, and the most chivalrous of all the Irish fighters. But why Red Branch? Branch of what or whom? That is a reasonable question to ask, but I have not observed that it has ever been answered. The Red Branch was just the Red Branch, and there's an end on't. Here it is needful to state that the Picts, or Cruithne, or Clanna Rury, once possessed, according to popular belief, the whole of Ulster. It is highly probable, as I have already suggested, that at one time they possessed, not only Ulster, but the, greater part of Ireland. In the Heroic Age, we find them centred near Armagh at a place called Emhain Macha, usually named Emania. The etymology of this name need not detain us, but it may be useful to say that it means pro- bably the River Plain (Avon Magh), the river in question being a tributary of the Blackwater. There is the usual THE RACES OF IRELAND AND SCOTLAND. 145 legendary etymology, which connects the name with a red- haired Queen Macha, who reigned at Emania. Such legends are so common in explaining Irish place-names, that it is not surprising to find nearly every Irish writer gravely citing them as adequate etymologies. Near the site of Emania is a place named Creeveroe, the ancient spelling of which is Craebh Ruadh, translated as " Red Branch." It was here that the Ulster champions met pei-iodically to exhibit their prowess; and it was from the name of this place that they derived their name of the Red Branch Knights. But is there any appropriateness in " Red Branch " as a place- name? None that I can discover. The same word Craehh appears in the ancient place-name Monaigh Craebi (the modern Moncrieff) in Scotland, and in numerous place-names of Ireland, sometimes in the form of Crew. Beyond doubt, Craebh is here to be equated with Crieff. Now the names Crieff and Moncrieff in Scotland are surely the same as tlie Welsh Cryf (Cornish Cref or Crif or Creif), meaning " strong." Where this root appears as a place-name, that place has at one time been the site of a natural or artificial ;strength, or fort. Thus in Creeveroe, if translated the Red Strength, we have at once a sensible and satisfying etymology, not only of the place-name, but of the name by which the Ulster champions were called. And if we wish to know why the fort was named the " Red Strength," we have an explanation in the tradition that the walls of the King's House were of " red yew." Creeveroe may mean the " Royal Strength," for ruadh or roe is sometimes trans- lated " royal," and thus an alternative etymology is pro- vided. But that the Creeves of Irish place-names mean strengths or forts, and not " branches," I have no doubt whatever.! 1 When dealing with Creeve as an Irish place-name. Dr. Joyce, in trans- lating it as " branch " {Place-names, 1870, p. 483) is obliged to suggest the fanciful explanation that it really means "tree," and that "tree" is associated with " games, or religious rites, or the inauguration of chiefs." lo 146 THE RACES OF IRELAND AND SCOTLAND. Emania was destroyed in 332 a.d. That is an important date to remember, for, in my belief, it marks off definitely Irish reliable history from tradition and legend. The chronology of this event is probably accurate, and the in- cident itself is, beyond question, historical. The metropolis of the Picts of Ulster was burned, and their power per- manently shattered by "the three Collas " ^ of the Hero- monian or Scottish line of kings. Thenceforward, the Pictish possessions in Ulster were narrowed down to a strip of country, now represented by County Down and the southern half of Antrim. After the Ulster Scots became predominant in that quarter, they were known by the Annalists, but not in- variably, as the Ulaid or Ulta, and their kings as kings of Uladh (Ulster); while the Picts were always called the Cruithne, and their rulers, kings of the Cruithne. The Annals of Ulster have preserved a record of the persistent antagonism between the two peoples: they contain also a record of instances in which they united to meet a common foe. Nothing proves more clearly the historical character of the destruction of Emania than the bitter memory left for many centuries by the event. Private grievances are frequently effaced by time; but national wrongs never. They may be atoned for by subsequent goodwill, but the memory of them is ineradicably engraven on the hearts of a people. They may be forgiven; but they are not forgotten. There are few more remarkable instances of this fact, than the tenacity with which the Picts of Ulster continued to preserve the memory of the burning of Emania. Dr. Hyde states that after a period of nine hundred years, the Irians (or Cruithne) refused to make common cause with the other Irish against the Normans at the battle of Downpatrick in 1260; so ^ According to Ferguson (The Teutonic Name Si/stem, p. 19), CoUa is a Saxon name. The MacDonalds are descended from one of the CoIIas. THE RACES OF IRELAND AND SCOTLAND. 147 bitterly did they resent the treatment their ancestors liad received at the hands of the Heremonians, and so deeply did the burning of Emania continue to rankle in their hearts.'^ It is to be observed that the Cruithne, in their later possessions, were confined to the part of the north of Ireland that is now by far the most prosperous corner in the whole country; for it includes the great City of Belfast. It is to be observed, further, that the people inhabiting that corner differed then, as they differ to-day, in certain respects from their neighbours. The Ulster question, it would appear, is not one of to-day; nor of yesterday. Its roots may stretch further back, even, than the times of the Tudors or the Stuarts. For the seeds of a mutual antagonism were sown in the smoking ruins of Emania nearly sixteen hundred years ago. We have now to consider, briefly, the racial affinities of the Cruithne or Ulster Picts. To what stock did they belong? What language did they speak? That their lan- guage differed from Gaelic is certain. One word, and one only of their vocabulary, has been handed down to us: the word carta. It is quoted by Cormac as meaning "a pin that is put on its shank." It is equated with the Graelic dealg, which means a pin and a thorn, a suggestive con- junction, reminiscent of the statement by Tacitus, that the ancient Germans commonly used a thorn for a pin. This Cruthinian word cartit has puzzled philologists considerably, for cartit means a shanked pin in no known language. There is little doubt that it is a compound word, as indeed Cormac's interpretation implies. The latter half of the word, viz., tit, is plainly, I think, the Icelandic tittr, pin, and car is the Cymric gar, Gaelic carr, meanig a shank. Cartit is thus a hybrid; and hybridism, as we shall see, is character- istic of the Pictish language. It is a sure sign of the mixed ^ Literary History of Ireland, p. 66. 148 THE RACES OF IRELAND AND SCOTLAND. races of which the Irish and the Scottish nations alike are composed. The Scandinavian * origin of the Cruithne would appear to have been believed in by their neighbours, for the rampart built by the Picts as a protection against the pressure of the Scots, is known traditionally as the " Danes' Cast." This rampart — the great Wall of Ulidia — extended in separate sections through the valley of the Newry River for a distance of over twenty miles. It proved an adequate barrier against the aggressiveness of the Scots, whose occu- pation, under the three Collas, of the territory comprised in Armagh, Monaghan, and Louth, (subsequently known as Oriel) was certainly effective, though their hold on the rest of northern Ulster was apparently less firm. * I use the word '* Scandinavian," throughout, in its philological sense, that is, to include the Danes. Scandinavia, strictly speaking, excludes Denmark. CHAPTER XV. The historical Picts — The Maitai and the Vecturiones (or Verturiones) — How the Picts got their name — Teutonic parallels — The "men of the elves" — Were the Picts tattooers? — Historical notices of the Picts : Herodian, Solinus, Dion Cassius— The sources of their infor- mation examined — Tacitus on the Caledonians— Shield-painting — The Pictones of Poitou. How are the elves, the " peehts " of Scottish tradition, connected with the Picts of history? Beyond any doubt, the name " Peehts " or " Pichts " was applied by the peasantry of Scotland to what I have proved to be dwarfs, or pigmies, or elves; and it is equally beyond doubt that the people so well known to historians as the Picts were not dwarfs, or pigmies, or elves. On the contrary, they were believed to be big men physically — " folk of much might," as Layamon calls them i — and it is inconceivable that had they borne the remotest resemblance to the " Peght " described by Sir Walter Scott, no contemporary writer should have alluded to the fact, and no anthropological evidence should remain to testify to its existence. So here we have folk-lore, not for the first time, apparently in conflict with history and anthro- pology; the conditions, in fact, are precisely analogous to those which we examined in the case of the Dananns of Ireland. Layamon's term, "a folk of much might," as applied to the Picts, may have its origin in the name Mcetce (Maitai)^ given by the Romans to one of the two main divisions into 1 Brut (Madden), i., p. 493. The big-bodied Caledonians of Tacitus are described by Eumenius (309 A.D.) as " Picts." That was their later name. From the description given by Gildas (a contemporary) of the physical appearance of the Picts, it is clear that there was little or nothing to distinguish them from the Scots. - Adamnan's Miath't. 150 THE RACES OF IRELAND AND SCOTLAND. which the people, afterwards collectively called the Picts, were grouped early in the third century. The McBtce are placed nearer Antonine's Wall than the Caledonians, and the description by Dio warrants the belief that they embraced the tribes south of the Grampians as far as the Wall. To- wards the end of the fourth century, a similar division is found, as described by Ammian Marcellin. He calls the Caledonians Z)^-Caledonians, perhaps in allusion to the racial admixture — Cymric and Teutonic — of that people, and gives the second group of people the name of Vecturiones, in which name may possibly appear the Latin form of the Scandin- avian Vcettir.^ Sir J. Y. Simpson, in his essay on The Cat-stane, made the suggestion (p. 40) that the Vecturiones may have been Saxon allies of the Picts,* who had then amalgamated with the latter. It can scarcely be doubted that in the fourth century, the Southern Picts were a mixture of Scandinavians and Saxons, with a Celtic element of more or less unimportance. I have shown that there is adequate ground for believing that the Picts got their name from the people upon whom ^ I am well aware that the name is now almost invariably written as ** Verturiones." Sir John Rhys introduced this form into England, and his authority has been sufficient to establish it. He founds the emenda- tion on a statement in Eyssenhardt's edition of Ammian, that the form Vecturiones comes from Gelonius, who lived in the si'cteenth century, and that it has no MS. authority (Celtic Britain (J884), p. 84). Sir John Rhys was " delighted " with this discovery, but he does not tell us how it has been proved conclusively that "r" is right and *'c" is wrong. Until this has been done, there does not seem to be sufficient ground for reject- ing Vecturiones and adopting Verturiones. George Buchanan, one of the best Latin scholars of his day, uses the form Vecturiones ; and he lived in the sixteenth century as well as Gelonius. Is it likely that he copied from Gelonius? If Verturiones is, in fact, the correct form, it may be referred to either of two Cymric words, viz., (/n^erthi/r, a fortifica- tion, or, with greater likelihood, to Leitar or Leitir : (usually in the form of Letter, both in Ireland and Scotland), the slope of a hill. Cym. Llethyr, a slope (Ger. Leiter, Eng. ladder, that which slopes or leans); Ger. Leite, slope or declivity; Gothic hleida. Linne or Lin : a pool or lake. A suffix in some Irish names, the most familiar being Dublin. Cym. Llyn. But O. Welsh Linn also denotes a marsh, a related meaning. Lough or Loch (Scotland). Cym. Llwch, an inlet, a lake. But the source is probably the Teut. root, Lek (Lak), watery, and especially 0. Ic. Logr, water. Magh, May, or Moy : a plain, or field. Cym. Mai and* Maes, Ger. matt, Eng. mead or mede (c/. Cym. Ma, a place). '^ Mam or Maum, Ireland and (rarely) Scotland: a round hill. (Lat. Mamma.) Sometimes applied in Ireland and Scot- land to a mountain pass. Mov, Mona, and Money : Money is a frequent prefix. It is usually taken from muine, a brake or shrubbery. But all the names in this class may be related to O. Norse Moinn, dwelling on a moor. Mon and Min in Scotland belong to the same category. (Gae. monadh, a moor; mbine, a bog; Cym. maum, peat, is probably related). ^ The Cym. form Mai is found in Scotland, e.c/.. May, Moy, Cambus o' May, Rothiemay, etc. THE RACES OF IRELAND AND SCOTLAND. 169 Mor : great or large (Cym. Maivr), apt to be confused with O. Ic. Mor, moor, and possibly with Gae. Muir (Gym. Mor), the sea. Muilenn {Mullin): a mill. Gym. Melin, A. S. Myln. Not of Teutonic origin, the genuinely Teutonic word for a mill being " quern " (A. S. ctveorn). Mullagh : applied in Ireland to certain hills. From Gym. Moel, piled, bare, or bald, applied in Wales to mountains with bare tops. Owen: applied in Ireland to streams. It is a corruption of Gym. Awon (Avon), a river, and even in Ireland occasionally takes the form of Aivin. Poll or Pol: pool. Gym. Pivl A. S. Pol. Doubtful whether derived from Gymric or Teutonic. Port: a haven. Lat. Portus ; Gym. Porth. Einn (in various forms): a promontory, or point. Gym. Rhyn, which, however, has various meanings, among which " Cape " may be a loan (c/. Rhinns of Galloway in Scotland). Rath: an earthen fort, and so applied to place-names. Dr. Joyce says that there are over 400 townships in Ireland with this prefix, in the forms of Ra, Rah, Raw, and Ray, and more than 700 names commencing with the word in its original form Rath (correct pronuncia- tion Ra). Now, whence is the word Rath derived? In the sense of " fort," it has no apparent affinity with Gym. Rhath, a cleared spot (cognate with 0. Ic. Rydja, 0. H. Ger. Riuti, land made fertile by uprooting; Eng. root and rid, i.e., a place ridded of trees). A form of Rath is used by Caedmon ( Burh wrathum werian) in the same sense as the Irish word for a fort, and the English equivalents ace " ward " and "guard," showing a common form of metathesis (c/. 170 THE RACES OF IRELAND AND SCOTLAND. "wraith," also written "warth"). The 0. Ic. form is varda, to watch over, to protect. The Irish raths (forts), are usually associated by tradition with the Danes, "Danish raths" being a common conjunction. Cf. A.S. wraeth, a fortification or enclosure, and 0. Ic. Reitr, a place marked out. The latter word is associated in Scot- land with "burghs" or forts, e.g., Rattar Brough (Caithness), Eattra (Borgue). Rattray (Blairgowrie and Peterhead), is probably from the same source. But it is a large assumption to suppose that all the Irish raths were forts. On the contrary. Rath signified home- stead in the Irish Laws {Celtic Scotland, III., p. 243), and is therefore related to Cym. Rhath, a cleared spot, which, in turn, seems to have been borrowed from a Teutonic source. Rath is found in a number of German place-names. In 0. Ic. Rjodr means an open space in a forest. Most of the " Raths " in Irish topography must have got their name for this primary reason. Ros or Ross : a promontory and (in the South of Ireland) a wood. Scottish topography has the word in both senses, as well as with the meaning of a moor or marsh. Cym. (Welsh and Cornish) Rhos. The source of Ross, a promontory, is probably Cym. Rhus, a start or tail (cf. Start Point in Devonshire). Ross, a wood, may be related to Rhos, a marsh. (In Cormac's Glossary, further meanings are given of Ross, viz.: " flaxseed "—a meaning still alive — and "duck- meat.") Sean {Shan): translated as "old" (Lat. Senex). But this is surely a doubtful etymology. Preferably this prefix in place-names is to be referred to a Teutonic word signi- fying herdsman {Senno, Gothic Sanja, coAvherd), and also pasture {Senne). Thus Shanbo and Shanbally may have been originally applied to a town pasture, and Shanmullagh would mean hill-pasture. THE EACES OF IRELAND AND SCOTLAND. 171 Sgor and Sgeir (Scuir and Skerry): a sea-rock, from O. Ic, Sker, a rock in the sea — a common name for rocks, especially on the west coast of Scotland. Scar and Scor in England. Sliabh (Slieve): a common name in Ireland (it is rare in Scotland), for a mountain or hill. It is usually, if not invariably, applied to a conical height, just as Mam (which see), is applied to a round hill. Why is this? Sliabh is a Gaelic cognate of " slope," which is a derivation of the word " slip," the root-idea of the latter word being found in "slippery." A hill with sloping sides necessarily has a pointed apex. Now, the English word "slip " is derived from a Teutonic base, sleip or sleup, to slip or glide (Ger. schleifen, 0. H. G. sliofan, Goth, sliupan), and a Gym. affinity is lacking. Sliabh enters as freely into mountain nomencla- ture in Ireland as Ben does in Scotland. The former has a Teutonic and the latter a Cymric origin. Indirectly these facts imply that the Gaelic language was built up in Ireland {Sliabh probablj- displaced the Gym. Pen), and transplanted in Scotland, where, however, except in isolated instances, the Teut. Sliabh failed to oust the earlier Cym. Pen or Ben. If this hypothesis is accepted, it is difficult to evade the force of the reasoning that ascribes a Teutonic element to the very texture of the Gaelic language. Mountain nomenclature is frequently both ancient and philologically suggestive. Sron : a nose, and consequently applied to a promontory. From Cym. Trwyn, a point or nose; 0. Ic. Trjona (perhaps a borrowed word). Here S is substituted for T . Sruth or Sruthair : a stream. The English word " stream " and its various Teutonic affinities are derived from the Sans, root Sru, to flow, but it is difficult to dissociate the Gaelic word from Cym. Ystrad, from which " Strath," so common in Scotland, is taken. The Gaelic 172 THE RACES OF IRELAND AND SCOTLAND. form of Ystrad is Srath, and the " t " appears to be similarly eliminated in Smth. Originally applied to a valley or strath, it may have acquired a secondary meaning by being applied to the stream flowing through the strath. Confirmation of this view seems to be afforded by Corn. Stret or Streyth, stream. Con- versely, Ystrad may come from the same root as stream: Sru, to flow, i.e., a jAace through which a stream flows. Stuaic (Stag and StooJc): an isolated rock. From O. Ic. Stakkr, a stack or cape. Suidhe {See and Sea): a seat or settlement. From 0. Ic. Setr, seat or residence, with the allied forms in all the Teutonic dialects. (In the Gaelic word, the consonants are mute.) In Ireland, says Dr. Joyce, hills, mostly crowned by earns or moats are called Suidhe — Finn, i.e., Finn's seat or resting - place. In his " Ossian," Macpherson makes use of this fact in Gaelic topography by showing us Fionn on his mountain-top. There is a mythical element here, which might be employed by mythologists to prove that Fionn was a solar deity. So, too, in proving the mythic character of King Arthur, they might point to his " seats " like those of Fionn. It is a remarkable fact that the mythology of the Finns contains a similar idea in relation to Kaleva, a word that means " rocky " from Finnic KaUio, cliff, Lapp. Galle, Kallo {Gallagh in Irish means a place full of rocks). Kaleva is a giant, evidence of whose strength is found by the people in blocks of granite that they believe him to have hurled, and in huge rocks that they call his seats. A " Son of Kalev " is called " Child of the Rock." {Comparetti, p. 209). But a much more rational explanation can be given of these "Finn," "Arthur," and "Kaleva" Seats. THE RACES OF IRELAND AND SCOTLAND. 173 Tamnach : a field. Evidently a compound word {Tamn- achadh), the prefixial Tamn, which sometimes appears as Tavn, Tmvn, and Ton, being apparently derived from Cym. Taf, a spread or a flat space, Tarbh {Tarf): a bull. (0. Ic. Tarfr, a bull). A doubtful etymology for river-names, which may be referred with greater confidence to Cym. tarfu, to expel; tarf, drive (but see below). A name like Clon-tarf, however, pro- bably means the "bulls' meadow." Tarw is a river- name in Wales (Cym. tarw, what bursts through). It means also a " bull " in Welsh. Probably we have here a derived figurative name, of which there is apparently another instance in Cymric Twrch (Gae. Tore), a hog, a Welsh river-name appear- ing in Scotland under the form of " Turk." Twrch also means " burrower," and a link is thus provided be- tween a river that burrows its way, and a hog or a boar. Similarly, a river that " bursts " its way through obstacles might fairly be compared with a bull. Tigh (various forms): a house or dwelling, from Cym. Ty, a house, Teamhair : (Tara, the capital of Ireland's High Kings in Meath; and other places in Irish topography). It some- times appears as " Tower," and that is apparently the source of the word, Avhich is usually translated as " a palace situated on an eminence." Teampull : from Lat. Templum, applied generally to ancient churches. Teine {Tin or Tinny): fire, indicating places where fires (whatever their object) were kindled. Cym. Tan, fire. But the source is the Teutonic root tand, to burn; it is found in all the Teutonic dialects (Goth, tandjan, to set on fire; O. Ic. tandri, (ire; A. S. tendan, to kindle. Eng. tinder, etc.). 174 THE RACES OF IRELAND AND SCOTLAND. Tir : land. (Lat. Terra.) Cym. Tir. Tohar {Tipper and Tihber): a well, from Cym. Divfyr, water. O. Gael. Dobur or Dohhar. Torr or Tor: a heap or tower {Turrus). Cym. Tivr. Traigh (Tray): a strand (Tractus). Cym. Traeth, a tract, or sand. Ttiaim {Toom or Tom): a tumulus. Cym. Tom, a mound. Tulach {Tully, Tulla, Tullow, TaUoiv): a small hill. Cym. Twich, a knoll. Corn, Tallic, Tallock, Tallach, what is highly placed. Vaimh {Wem and Weem): a cave. Cym. Wm, hollow. Uisce or Uisge (numerous forms): water. Cym. Wysg, current or stream. The foregoing analysis of typical place-names in Ireland shows conclusively that the people who bestowed those names upon the places where they settled, spoke, some Cymric, others Teutonic, and a few Latin, the last element being plainly post-Patrician, and mainly ecclesiastical in its incidence. The amalgamation of these elements is shown by a further examination of Irish topography, which reveals the existence of many hybrids — Teutonic prefixes, and Celtic suffixes, or vice versa — among the names. The importance of this evidence in solving the ethnological problem presented by the Gaelic race and language can hardly be over-estimated. The suggestion that these Irish place-names merely show affini- ties with Cymric and Teutonic words, without being directly derived from them, fails entirely to meet the case. Obviously, the names are not cognates, but derivatives. In the domain of anthropology, there is to be seen in Ireland the undoubted prevalence of the Nordic or North Teutonic type, mingling with the classical type of the Gallic Celts. It has been shown that the pre-Celtic and pre- THE RACES OF IRELAND AND SCOTLAND. 175 Teutonic types are also well represented; but the question of the original composition of the Gael and the Gaelic language is not directly affected by the ethnology of their predecessors, with whom there was no amalgamation. In the domain of archaeology, too, a Teutonic connexion with Ireland has been proved to exist; and Irish legend betrays distinct points of contact with Teutonic folklore: as I have shown, there is a remarkable resemblance between certain customs primitively observed alike by Teutons and Gael. But nowhere is the Celto-Teutonic blend so clearly revealed as in Irish topography. And place-names, rightly inter- preted, are unassailably conclusive. There is little difficulty in finding analogies for this mixture of peoples and languages, the closest being per- haps the coalescence of races in France on conditions re- markably similar to those postulated for Ireland. Just as the Scots, a Teutonic people, gave the name of Scotia to Ireland (and later to Scotland), and imported Teutonic elements into the Cymric language spoken by the Celtic people with whom they coalesced; so the Franks, also a Teutonic people, gave the name of France to part, and ultimately the whole of Gaul, and introduced Teutonic words into the Celto - Roman language spoken by the Gallic people whom they subdued. In one sense, the settlements of the Visigoths and Burgundians in Gaul show even a closer analogy to that of the Soots in Ireland. For while the Frankish monarchy, in alliance with the Church, each for its own ends, aimed primarily at conquest, the Visi- goths and Burgundians sought a peaceable settlement among the Gallic people. " They shared lands and goods," says Dean Kitchin,^ " with the older owners. . . . He (the German), took half of all forests and gardens, two-thirds of all cultivated lands, one-third of all slaves, and so settled down in peace." ^ Hint, of France, i., 60. 176 THE RACES OF IRELAND AND SCOTLAND. Here we have conditions entirely favourable to a mixture of languages. The Teutons found the prevailing language in Gaul to be a Low Latin dialect, necessarily interspersed with Celtic words, the latter being relics of a language which had been renounced by a conquered and decadent people in favour of the tongue of their conquerors. The submersion of the Celtic language in Gaul, by Latin, is a striking fact in the study of races. It was a sign of the decay of nationalism, which itself was the outcome of a loss of independence, and the deadening lethargy induced by the hopelessness of its recovery. In such circumstances, subdued peoples mould themselves gradually, but surely, in the shape of their masters; and in time, assimilation, more or less complete, generally takes place in language, customs, and sympathies, if equal liberties and privileges are enjoyed by the different racial units which comprehend the population. An apparent anomaly here suggests itself in the fact that the Teutonic tribes conquered the West by force of arms, but instead of absorbing and assimilating the Western nations, were themselves absorbed and assimilated, leaving only indistinct traces of their Germanic origin. So it was in France; so it was in Spain; so it Avas in Ireland. The explanation appears to be that, primarily, the Teutonic in- vaders were not settlers — they were plunderers. When they settled, they married the women of the country, and the mother-tongue of their children gradually displaced the father-tongue, as it will always do quite naturally. • The Anglo-Saxon settlements in England were on a different footing. Originating in the arrival of bands of adventurers, whose swords were for sale, the immigration developed into an organised scheme of colonisation, in which the Teutonic wives of the settlers were included. This would appear to be proved by the testimony of Bede, who tells us that Old Anglia was said to have remained depopulated (desert) from the time of the emigration to Britain " to this day." Thus THE RACES OF IRELAND AND SCOTLAND. 177 in England, owing to the comparative absence of racial in- termarriage, the Celts did not absorb the Saxons, nor did the Celtic language oust the Anglo-Saxon tongue. Nor, on the other hand, did the Saxons absorb or assimilate the Celts. Some they must have reduced to a state of bondage; many they drove westwards and probably northwards; and a minority may have been permitted to retain their lands. These lands may have remained tributary on varying con- ditions, as was the case with certain territories in Gaul conquered by the Franks; and the holders would in those instances either sink gradually into a state of serfdom, or become completely and permanently Anglicised. Numerous traces of the Celt are found in the place-names of England, but comparatively few in the English language. The Franks amalgamated the Low Latin of the LaAV Courts with their own Teutonic Law-terms. The result was "a barbarous Latin full of German words." But by the end of the eighth century, the Lingua Romana Enstica had firmly established itself as the national language of the country. At the Council of Tours in 813, homilies were read either in Romance or German, and the Army oaths of 842 show that it was not until about, or after the middle of, the ninth century, that bi-lingualism among the Franks fell into disuse. The decay of Teutonic influences in Gaul must have been accelerated by the death of Charles the Great. But although German thus gradually disappeared as a distinct and spoken language in Gaul, it left its permaneiit mark on the language of the Franks, that is, French. The dialects of Northern France contain many traces of the original language of the Franks, while in Normandy, the Scandinavian element, introduced by the Northmen, is shown in the local dialect, as well as in many place-names and naval terms. The Proven9al dialects show the influence of the Bur- gundian settlements, and in Gascony the speech of the Visi- 178 THE RACES OF IRELAND AND SCOTLAND. goths runs through the Latin texture, interspersed with some Basque remains.^ In Spain, the same Teutonic elements are found in the spoken languages. In Catalan, the Biscayan Latin is mixed with Gothic, as is the purer Latin in Castile. In Portugal, Suevic mingles with the main West Latin stream. Again, when the composition of the English language is considered, it is easily seen how the main Teutonic fabric — itself a mixture of Low German with important Scan- dinavian dialects — has been mingled with a comparatively small Celtic element, borrowed from a conquered race, and a Norman-French element of profound importance imposed by a conquering people. Romance, the language of the Court, the Church, the Law, the Schools, and the Army, never became the language of the people. There was no real blend between the Normans and the Anglo-Saxons, and the attempt to force a foreign language on an unwilling nation was foredoomed to failure. Of necessity, communica- tion between the two j^eoples had to be carried on by means of a double vocabulary, and the two languages were mutually affected by the contact. But in the end, the Anglo-Saxon of the masses triumphed, and the Romance of the classes was incorporated in, and assimilated with, the Teutonic dialects, to form, in the twelfth and thirteenth centuries, with Latin and other elements, the English language. English would be a comparatively poor language did it consist of an Anglo-Saxon element only, instead of being the richest in the world by its capacity for absorption. The purest languages are the poorest. And so it is with the Gaelic language. The power of incorporating foreign elements shown by the original Celtic, is maintained to the present day by the addition of English 8 Roquefort's Glossary explains the " Walonne " language as lanque primitive des Francois et qui s'alUra bientvt par la jonction du Tudesque et du Latin (ii., p. 737). THE RACES OF IRELAND AND SCOTLAND. 179 words in a Gaelic dress that represent new inventions and new ideas. Thus, the enrichment of the language by the importation of foreign words has proceeded apace with what a purist would, with some propriety, regard as its disfigure- ment. A loss in purity has been accompanied by a gain in flexibility of expression, in enlargement of vision, and in facility of communication. Clearly the balance of advantage lies on the side of the language that can absorb, adapt, and incorporate. 10 The Saxon and the Gael are not parted by the chasm that is generally believed to exist. Their nearness of kinship is proved more particularly by anthropology and philology. They have given to one another, taken from one another, profited by one another, by social contact in England, and by actual amalgamation in Ireland and Scotland. There is not, and there should not be, any real antagonism between them. Ideally, one is the complement of the other. Throughout the Gaelic vocabulary, the same facts pro- claimed by place-names are observed on analysis, and nowhere more prominently than in the numerals, which are plainly of Latin origin. These facts are sometimes partially obscured by the accumulation of phonetically use- less, but grammatically convenient, consonants in the modern language; and it may be remarked here that if ever the Gaelic language is to be popularised among non- Gaelic speakers, it will be necessary to simplify it by clear- ing away, as far as possible, this superfluity of mute letters. It need scarcely be said that the evolution of the language has resulted in marked divergences from original forms, and that the Gaelic of the present day is as different from the ^o In The Welsh People (p. 617) the authors quote, apparently with approval, O. Schrader {Prehistoric Antiquities, Eng. translation, p. 113), who says that "the notion of a mixed language must have more weight assigned to it than has hitherto been allowed." That is a true and pregnant statement. 180 THE RACES OF IRELAND AND SCOTLAND. oldest written Gaelic as is English from Anglo-Saxon. The numerous glosses in the oldest Irish manuscripts show that the Gaelic of the glossarists was, in turn, different from the Gaelic of an earlier date. That, of course, is only what might be expected; but among other things, it shows the absurdity of the attempts so frequently made to explain Gaelic place-names by the Gaelic of the present day. As well attempt to explain the " wicks " and " hams " of Anglo- Saxon topography by the English of the twentieth century. The Celtic element in the oldest Gaelic must have been pure Cymric. Cormac proves this by using " p " words, e.g., prem (Gae. cruim), a worm, and map (Gae. mac), a son. Thus, even by the ninth century, Gaelic had not shed entirely its Cymric characteristics. In its grammatical structure, Gaelic has points of resemblance with the Cymric, Teutonic, and classical languages, but it has certain characteristics that are peculiarly its own. It would be beyond the scope of this work to deal with the structural formation of the language, even if I were competent to do so; but two examples may be given of marked peculiarities. One is the aversion from the initial letter " p," which, under the influence apparently of the Teutonic element in the language (as already noticed), generally becomes "c" ("k" sound), and is sometimes, as in atha?' (pater) eliminated altogether, as it is in Moeso- Gothic. Another remarkable characteristic is what is known as " aspiration," a device for flexion which is absent in the classical languages. This is one of the most important elements in the phonetic and grammatical structure of the language: by means of the introduction of the letter " h," the sound is softened, and the case is altered. And here it may be said, that notwithstanding the frequency of the guttural " ch " in Gaelic (another Teutonic inheritance), the general softening of consonants, and consequently gain in euphony, is THE RACES OF IRELAND AND SCOTLAND. 181 a goal that has been successfully reached in the construction of the language; while the treatment of the vowels is such as to suggest the cooing of a dove. It is a mistake to suppose that Gaelic, as spoken by a scholar, is harsh. It is in a large measure a liquid language, full of devices for euphonious expression. Its " appearance is against it "; but its appearance is deceptive. CHAPTER XVII. Scotland and its legendary matter — The earliest name of Scotland — The significance of the name " Alban "—The invasion of Scotland by Agricola— Who were the Caledonians ? — Galgacus or Calgacus — The Caledonian tribes self-contained units — The physical features of their country — An examination of Caledonian ethnology — An analysis of the place-names mentioned by Tacitus. Scotland, rich in prehistoric monuments, is comparatively poor in legendary matter that can be separated, as a dis- tinctive inheritance, from the imported folk-lore of Ireland. The historian of Scotland can thus take, as his starting-point, the records of reliable and contemporary writers, and, unem- barrassed by confused and contradictory traditions of pre- historic peoples, construct from the scanty but sure material at his disposal a story of Scottish national life. The student of Irish affairs, before Irish history was written, is like a weary traveller wandering in a wilderness of fiction, who scans the horizon with an eager eye, looking for an oasis of fact. The student of Scottish affairs, it is true, encounters the same tangle of fiction and fact in exploring his line of country. But he recognises the legends as Irish; they have been carried across the Irish Channel ; and have changed their hue; yet their true origin is undoubted. The Scota of Scottish tradition may differ from the Scota of the Irish legend; so may Gathelus or Gadel; so may Simon or Simeon Breac. But the Scottish stories are simply edited versions of the Irish originals; they are mainly the work of that in- defatigable and highly patriotic collector of traditions relating to the Scottish people, John of Fordun. What was the earliest name of Scotland? The oldest geographers made no distinction between the northern and the southern parts of Britain. They were equally compre- THE RACES OF IRELAND AND SCOTLAND. 183 hended in the earliest name of the island, Albion, and its later name, Britain, It is generally assumed that after the the name Albion as applying to the whole island fell into disuse, it survived as the name of modern Scotland. It is true that the Irish name for ancient Scotland was Alba (a form of the word against which Dr. Skene vigorously protested) or Alban, which name, it is asserted, is the same as the Albion of Aristotle, or his personator. There is one, perhaps there are two, isolated passages in ancient Irish writers which apparently suggest the application of Alban to the whole island; but the identity of Alban with Albion will require proof of a more convincing nature. The meaning of " Albion " has never been satisfactorily determined, though philologists of the present day lean to the old conception of the "chalk cliffs" as the most tenable theory, which, in lieu of a better explanation, it possibly is. But the likelihood of Scotland retaining a name with this meaning after Eng- land had lost it, is not strong. Alban means the Highlands. It is a Cymric word, signi- fying "the upper part," and a cognate word seems to be furnished by 0. H. Ger. Alpun and Alpi (Alps) meaning " mountain pasture." Although the modern Gael applies the name Alban to the whole of Scotland, the ancient Alban comprised a much smaller area. Albania — the Latin form of Alban — as described in a tract of the twelfth century (De Situ Albanie) was co-extensive with the Caledonia of Tacitus, i.e., the part of the modern Scotland that is north of the Firths of Forth and Clyde. The Scots, whose slogan at the Battle of the Standard, in 1138, was " Albanich, Albanich! " were those who were afterwards known as the " ancient Scots," and the " wylde Scottis," living benorth the Firths. 1 ^ There is evidence in the allusions of ancient writers, as well as in the direct proof furnished by old maps, that Albania was sometimes con- sidered to be an island, the idea being that the two Firths (Forth and Clyde) actually met. 184 THE RACES OF IRELAND AND SCOTLAND. Our acquaintance with the Caledonian tribes commences with the invasion of Scotland by Agricola, of whose campaign his son-in-law, Tacitus, has left us an account, which, by reason of the conciseness of the narrative, is all too meagre. It is unfortunate that this account was not given by a writer of greater prolixity. Terseness is an admirable literary quality, but although Tacitus is the delight of the stylist, he is the despair of the ethnologist. He touched upon a number of racial questions, and settled none of them. Yet some of his statements are sufficiently precise and unambiguous. The " ruddy hair and large limbs of the Caledonians " suggested to him a " German origin." In his treatise on Germany, he states his belief that the Germans were a pure unmixed race; that a family likeness pervaded the whole; that their physical characteristics were " eyes stern and blue; ruddy hair; and large bodies. "^ When describing the inhabitants of Britain, he makes a clear distinction between the German- looking Caledonians and the rest of the inhabitants. Many attempts have been made to explain away his words, but it is not easy to evade the force of this distinction. If Tacitus is to be accepted as a reliable authority — and his father-in-law could have no object in misinforming him — we must take it as a fact that the Caledonians differed physically from the Britons, in resembling the Germans more closely. A further question here suggests itself. Did Tacitus mean that the whole of the Caledonians north of the Firths were red-haired, big - bodied men; or was his description limited to the particular tribe that gave its name to the whole body of the inhabitants? This is an important point in determining the ethnology of northern Scotland. Accord- ing to the point of view, it might be possible to argue that 2 Germania, c. 4. THE RACES OF IRELAND AND SCOTLAND. 185 the Caledonian army vanquished by the Romans in 84 a.d.^ was composed wholly of Germanic tribes; or that one tribe alone, the Caledonii, was of Teutonic origin. When Tacitus made his remark about the origin of the Caledonians, the area of his observations was partly tribal (e.g., the Silures) and partly geographical (e.g., the " tribes nearest Gaul"). Therefore the Caledonians might have belonged to either category. But it is noticeable that when he comes to describe the decisive battle in Caledonia, and the preparations that preceded it, he never calls the antagonists of the Romans by the name of " Caledonians," but invariably by the name of " Britons "; or the " various inhabitants " of Caledonia. Again, the Welsh Triads, when describing the foreign colonies that settled among the Britons, state that a "descent" was made in " Albin " by "the tribe of Celyddon"; that is, the Caledonians, or the refuge-seeking people who took shelter (Celydd) in the Caledonian forest. The inference is that this tribe settled among the native Britons. If we assume that this foreign people were a tribe of Germans (or Scandinavians) whose tribal name has been lost, the remark of Tacitus on their ethnology is freed from ambiguity, for it must be supposed that his allusion was to that tribe alone. But he called the " various inhabitants " of Caledonians by the name of Britons, because that was the national name of the majority of the inhabitants of Cale- donia, although the dominant tribe — the Celyddons — were not British by origin. One conclusion may be drawn from the name Calgacus, ^ The site of the battle is still an unsolved problem. It must have been near the sea ; Mons Grampius must be identifiable ; and for these reasons Ardoch must be abandoned. Skene is probably right in suggesting "Granpius" as the correct name of the mountain (Cymric gran, pre- cipitous, and perhaps pid, a tapering point). The usual reading is "Graupius." 186 THE RACES OF IRELAND AND SCOTLAND. or Galgacus, " the most distinguished for birth and valour among the chieftains," whom the Britons chose as their leader against Agricola. We find the name used by Adamnan in both forms (C and G being interchangeable) as " Calgach " and " Galgach " (Roboretum Calgachi, and Daire Calgaich are the old names of Derry). In modern Gaelic, the word has various meanings, but the root-idea is that of " pointed " or "stinging " (Cymric Cola, a point or sting, Colp, a dart, from which the Gaelic Colg or Calg, a spear, is apparently derived). The name Galgacus would appear to be of Cymric origin, the form being altered by Teutenic contact. It seems to mean " dart-man." It affords no certain clue to the language spoken by the person who bore the name, but it denotes the existence of a Celtic tongue in Caledonia. There is no mention of the name of the tribe to which Galgacus was attached, though the presumption is in favour of the Celyddons. A fact that stands out clearly in the narrative of Tacitus is, that the Caledonian tribes in normal circumstances were not under the effective government of a central authority. There was no organisation that gave them the coherence of nationality. They were simply separate, self-contained units, of relatively greater or less importance, mutually inde- pendent, and probably mutually antagonistic. But the moment they were threatened by a common danger, they united for their common defence. Yet a hastily formed alliance for a temporary purpose must have placed them at such a disadvantage as made their defeat by the disciplined soldiers of Agricola (auxiliaries, with a stiffening of legions), a foregone conclusion. They were, in fact, a mob opposed to an army. A curious parallel is presented by the conditions that prevailed in the Highlands during the clan period. There was the same lack of cohesion among the clans until a common object united them; but no sooner was that object served, than the old divisions were renewed, and the old THE RACES OF IRELAND AND SCOTLAND. 187 antagonisms were re-awakened. Thus, formidable although the Highlanders frequently proved themselves in their cam- paigns agair^t the Sassenach, their effectiveness was fre- quently neutralised — at Culloden conspicuously so — by the lack of that kind of discipline of which the basis is combina- tion, carefully planned, and obediently executed. The lack of inter-communication between the Highlanders in the clan days (except of a hostile nature) was due mainly to the physical features of the country in which they lived. Mountains divided them and a waste of trackless moor; and it was not until Wade's military roads were made in the first half of the eighteenth century, that a community of national feeling was established between them. If that was the case in the eighteenth century, the mutual isolation must have been much more pronounced in the first. For the country presented a dreary, unrelieved vista of marsh and forest, forest and marsh. In the great Caledonian forest, the precedent set by the Gauls and by the Britons of the south, must have been closely followed. According to Coesar (cor- roborated by Strabo) the British towns were in thick woods, fenced round with a trench and rampart, where, " to avoid incursions, they retire and take refuge." Of what race or races were the natives of Caledonia com- posed? That they were a homogeneous people is out of the question. Leaving out of account the people of the Palaeoli- thic and the Old Stone Ages, the evidences of the present day provided by archaeology, in conjunction with cranial characteristics and pigmentation, prove the existence of an important substratum of neolithic folk, the so-called Iberians of the chambered cairns, and the Bronze people of the short cist and stone circles.* The short, dark longheads are numerous in the West Highlands, in Caithness, and the Orkneys; and it is there that the chambered cairns pre- * The stone circles in Scotland seem to belong to the period of transition between Stone and Bronze. 188 THE RACES OF IRELAND AND SCOTLAND. dominate. The taller and fairer broadheads, on the other hand, are mainly in the north-east counties, where the beaker finds, associated with the brachycephalic skulls and the oldest Bronze remains, are thickly clustered in the map prepared by the Hon. John Abercromby.^ This map shows that while in western Scotland, pottery of the beaker class was found in a few sporadic sites only, there are numerous beaker sites in the south of England and all along the east coast of England and Scotland, as far as Sutherland, with a group in central England and some isolated instances in Wales. The conclusion seems to be that these beaker-men worked their way up the east coast from the south. At any rate, Mr. Abercromby's conclusion is, that although there are variations in the types of ceramic, there was probably no difference between the people who made them. The prehistoric factors in Caledonian ethnology must not therefore be overlooked, but even then, we are only on the threshold of the question. Who were the big red-headed men of whom Tacitus has given us a tantalising glimpse; and if, as I have assumed, his description was confined to a section of the people in northern Scotland, what were the racial affinities of the remainder, excluding the Stone and Bronze elements? It may be said at once that to this ques- tion no final answer can be given. We can however look for some guidance to the few place-names that Tacitus has left on record. Here, again, it must be premised that even if it be possible to reach the sources of those names with tolerable certainty, they only prove that the language from which they are derived was spoken by a people who, at one time (not necessarily in the first century) inhabited the places concerned. These place-names are only four in number, three of them (Clota, Bodotria, and Taus or Tavaus) being the names of ^ Proc. Soc. of Antlq. of Scot., vol. xxxviii. THE RACES OF IRELAND AND SCOTLAND. 189 rivers, and the fourth, Horesti, being a tribal name. Clota is the modern Clyde. As is so frequently the case, " C " here is interchangeable with " G." This is shown by the cognate river-name " Clude," which appears in Taliessin as Glut vein (Glut avon). Gjm.. glivyd, " of fair appearance " would fit, but this would appear to be a loan from A . S . gloed, shin- ing or smooth (0. Fris. glod, Ger. glatt, smooth, the primi- tive meaning of the Teut. root). The English words "glad" and " glitter " come from the same source. But the nearest approach is O. Ic. glot, to shine or glitter, and Glota is found as a Scandinavian river-name. Antonine calls the Island of Arran, Glotta, and Horsley translates the name given to the Clyde by Tacitus as " Glota." Camden, too, seems to prefer the form " Glotta." But, on the whole, it seems safer to look for the root in Cym. Clud (an early form of the Clyde), "that which carries" {Cluda to carry or convey). In certain of the Welsh Triads, Clud is translated as " progression." The idea of motion, so common in river-names, seems therefore to be present here. The Cluden and Clyth in Scotland, the various Cludachs or Clydachs, the Clywedog, and the Clwyd in Wales; and the Clody, Clodagh, and Glyde in Ireland show the same root. (Cludach and Clodagh give the river- root ach.) Bodotria (Ptolemy's Boderia) is a doubtful word, but it seems to be connected with Cym. Budraw, " to dirty or soil," and in view of the probability of the Forth being a muddier river in the first century than it is even to-day, there is no impropriety in this derivation. The later name " Forth " must surely have the same origin as the English word " ford," and as a matter of fact, it appears in 1072 under the name of " Scodwade," or " Scot Ford," and a little later, as " Scotte Wattre." The name given to the Firth of Forth in the Ork. Saga, namely MyrTiva-fjordr (murk-firth) 190 THE RACES OF IRELAND AND SCOTLAND. seems to bear out the derivation I have suggested for Bodotria.^ Tava or Tavaus (sometimes read as Tanaus): is clearly of Cymric origin. It is derived from Tafu, to spread, a root found in Welsh river-names, and applied to rivers having a wide or spreading mouth. It is found in Ptolemy's Tava, which is clearly the Tay. But the Tava of Tacitus cannot be the Tay. When, in a.d. 80 (the third year of his campaigns), Agricola encountered " new tribes," he had not yet pene- trated as far as the Forth. It was not until the following summer that he built his line of forts between the Clyde and the Forth, after an effective occupation of all the country south of the Firths. Therefore we must look for his Tava between the Humber (the country north of which he con- quered in A.D. 79), and the Forth. The Tava of Tacitus is probably the Tweed. The Teviot contains the name, its earliest forms being Teiwi and Tefe. The suffix " ot " is the Cym. ach, a fluid or river (0. Gae. oich, water), for " ch " and " th " in old documents being similar, they are frequently found to interchange in names. The oldest form of Forteviot in Perthshire was Fothuirta- baicht, and its later forms were Ferteuyoth and Forteviot; Elliot (Forfar) is in its oldest forms Elloch and Eloth; Kenneth was sometimes written Cinacha and Kenaucht; and so on. The name Teviot therefore means " the spreading water." But that description is only applicable to the mouth of the Tweed, of which river the Teviot is a tributary, though a tributary nearly equal in importance to the parent stream. I suggest as a probable solution of the difficulty, that in the first century, the river had not yet received its name of the Tweed, but was called the Tefe right down to Berwick. This would appear to be confirmed by the etymology of the word Tweed (earliest forms Tuidus, Tede, etc.), which * Myrkva-fjordr appears in the Heimskringla as a place-name in Sweden (Morkofjord). THE RACES OF IRELAND AND SCOTLAND. 191 seems to be identical with Cym. Tuedd, coast, the inference being that the portion of the combined streams nearest the sea received the name of " the coast river " to distinguish it from the Teviot beyond the junction. The Farrar and the Beauly rivers (see next chapter) supply an analogy in support of this theory. The tribal name Horesti may with same probability be assigned to the same origin as the English word " hurst " (Ger. horst), a thicket. (Cf. forst and forest.) The Horesti were north of the Firth of Forth, apparently in Fifeshire. Pursuing this examination of early place-names, I shall now analyse the Ptolemaic names of the second century in Scotland. CHAPTER XVIIL River-names and their value — Mountain -names and their value — Ptolemy's p'ace and tribal names in Scotland analysed. River-names are the most eloquent factors in topography, for they are the oldest and the least liable to change. They are more useful pointers even than the names of mountains. Tribes seeking settlements would be naturally attracted by rivers, and especially by fordable rivers; and the most desir- able lands would be the higher ground adjoining the swamps which must have resulted from the unbanked state of the streams. If the new settlers were superimposed upon older inhabitants, the existing names of the rivers would be re- tained, frequently (but not necessarily), in the original or a corrupted form. If the lands were unoccupied by other tribes, the settlers would give the rivers names in their own language, denoting their peculiarities or general characteris- tics, whether straight or crooked, smooth or rough, clear or dark, sluggish or swift, and in some cases, names denoting simple motion, or even the primitive idea of water. It may be laid down as an axiom of topographical research, that the more fanciful the names, either of rivere or moun- tains, the later is their origin. The simple minds of the barbarous tribes whose chief concern was the provision of food by primitive agriculture, by the chase, by the reiving lof neighbouring tribes, and by the tending of their flocks and herds, were unlikely to conceive poetic names for the features of the landscape. And it may be added that, in general, names which " leap to the eyes," as being eminently descrip- tive of the topography, are far more likely to be correct than those that call for an effort of the imagination. The horror of the obvious which characterises the work of some etymolo- gists is surely an unscientific attitude. THE RACES OF IRELAND AND SCOTLAND. 193 The following is an analysis of Ptolemy's river-names in Scotland: — Ahravanyius (Luce):i this probably means Aber-avon, the river-port, for Aber means a port, as well as a confluence. Aberavon is purely Cymric, and Luce is Scandinavian (see E,. Loxa). Alauna (Allan): this is obviously the Allan or Alne in Northumberland, but the name is also given to a town in Scotland on the Allan (Stirling). "Allan" is a widely distributed river-name found in England, Wales, and Scotland in various forms. The root is Al or El, and so appears in the Ale (Roxburgh), an early form of which is Alne. Conversely, Alnmouth (Northumberland), is sometimes pronounced Alemouth, the " an " of Allan (of which "ne" seems to be an Anglo-Saxon variant), being a common suffix in British river-names (it repre- sents Ajon or Avon, a river). In Cornish, the root Al or El appears in Hel, Kail, or Heyle, a tidal river. Probably it is to be traced to Cym. Elu, to move on, to go. We find the root as a suffix in such names as Cam-el (Cornwall), meaning the crooked (cam), river, and (pace the etymologists who attribute the name to their favourite gods) probably also in Camulodunum, the dunum, or hill-fort, of the crooked river, i.e., the Colne, on which Colchester is situated (c/. the Scottish Came- lot and the Camelot of Arthurian legends — both river- names). Boderia (Forth): already discussed (Tacitus group of names). '"Rivers," says Skene (Celt. Scot., i., 73), "do not change their names." And yet he makes Abravannus = Luce, Boderia = Forth, lena =Cree, etc. There is no rule without an exception, and river-names are not exempt from the application of this general truth. A change of name usually implies a change of race. The river-names of America, Australia, and Africa proclaim this fact. 13 194 THE RACES OF IRELAND AND SCOTLAND. Celnius (CuUen): Cym. Cul, narrow. CJota (Clyde): already discussed (Tacitus group of names). Deva : Ptolemy's name for the Dee. But the equation of Dee with Deva shows something lacking in the phonetics. There are evidently two elements in Deva, and the first syllable only (De), is represented in the modern name. The second half of the word gives the root Af (Cym.), conveying the idea of motion, from which the familiar Afon is derived. Wf, flow, or glide, or running, con- tains a related idea. The first syllable in Deva is Cym. Dwy, two, and Deva thus means the two streams. This view of the origin of Deva seems to be proved by the fact that the great Dee in England and Wales is called by the Welsh (and, as Camden observes, was called by them in his day), Divfyr Dwy, meaning the two waters, in allusion to the fact that the river has two head-streams. The Aberdeenshire Dee is mainly formed by two head-streams, and the Dee in Kirk- cudbrightshire is formed at its broadest part by a junc- tion with the Tarfe. (There is also a Dee in Ireland, showing the wide distribution of the name.) In Scotland, the Aberdeenshire Dee may have been called originally Dtvy - avon, the two rivers, for Ptolemy's town on the Dee shows the "Avon" termina- tion pretty clearly. So also does Devenick (in the name Banchory-Devenick), which means literally " the Dee River water." Deva and Devon have a common origin. We have a parallel case in the river-name " Dusk," or " Desk," which Davoren's Glossary translates as the two streams. (Cym. Dwy Wysg.)- 2 Loch Duich, in Kintail, Ross-shire, may supply a further parallel, for the name seems to mean the two waters (Dwy-ich), Duich and Long forming a fork. THE RACES OF IRELAND AND SCOTLAND. 195 The derivation of Deva from Diva, goddess, has no apparent warrant. The Dee, of course, was worshipped, but so were all the principal rivers. lena : a corruption of Ken, apparently. (Cym. Cain, clear.) lla (Ullie), or the Helmsdale River: (see " Ullie " in the Scottish river-names. Itis : (probably the Etive Eiver). Perhaps from Cym. Ith, what stretches out. Longus : (perhaps meant for Loch Long). Cym. Llong, a ship, and O. Ic. Lung, a warship. It should be observed that Cormac calls Long (a ship) a " Saxon " word. From this it would appear that the Celts borrowed the word. Loxa (Lossie): 0. Ic. Laxa (salmon-river) hardly fits here, but 0. Ic. L/oss (bright or shining), does. This is the jDrobable source of the name. Nabarus (Naver): the Sans, root is Niv, to flow, and cognates of Nabarus, or Naver, are found in Germany (R. Naab); Holland (R. Naba, or Nave); Spain (R. Nevia); Russia (R. Neva); and Wales (R. Never). Cym. Nof, what is flowing or moving, is apparently the Celtic root. " Navern " is an old form of the Scot- tish Naver, and the same form (" Nevern "), appears in the Pembrokeshire river. In O. Welsh it is spelt " Nyfer." Novius (Nith): Cym. root (Nof) just mentioned. Nith can- not, without violence, be equated with Novus. It is probably from Cym. Nydd, a twist, a suitable name for a sinuous river like the Nith. Tava (Tay): same root as the Tavaus of Tacitus (which see). Tina (Eden): perhaps from Cym. Eddain, to glide onward. But it may be a misplacement by Ptolemy of the Tyne 196 THE RACES OF IRELAND AND SCOTLAND. in Haddington (Old Norse, Thynja, to make a thunder- ing noise, as a rapid current does). Tuessis is placed in the position of the Spey, but there is no obvious connexion between the names. Possibly Tuessis may be conected with Cym. Ttvs, an outlet. The name "Spey" is clearly derived from O.Ic. Spyia, to spew or vomit, or (more obviously), from 0. Fris. Spey, with the same meaning. The name is due to its spates. (Probably " spate " has radically a similar meaning; Irish Gae. speid). Yarar (Beauly): the old name of the Beauly was the Farrar, still retained in the R. Farrar, which runs into the Beauly. This word can be ^^lausibly referred to O. Ic. Fara, to move or go, hence Far, a passage. Suffix dr is a nominal form from a, a river. Vor (gen. pi. Varar) means a fenced-in landing-place, and the word is used in Iceland for an inlet where boats land. But as a river-name, the idea of motion is preferable for the Scottish Varar. Cym. Ffawr, a running, a course, or Givdr, gentle, is alternatively a possible, but less likely, source. Ptolemy gives the names of a few sea-lochs {sinus, a bay or sea-loch), which will repay analysis. Lemannonius : This bay has been variously identified with Loch Linnhe, Loch Fyne, and (Skene) Loch Long. Its position suggests Loch Fyne, but its name and other circumstances lead to the belief that Loch Linnhe is meant. It must be remembered that in Ptolemy's map we cannot look for the accuracy of a modern map. The grotesque shape that he has given to Scotland — of which various explanations have been suggested — shows that his knoAvledge of the country was, to say the least, imperfect, though it is possible that Ptolemy was not responsible for this shape. THE RACES OF IRELAND AND SCOTLAND. 197 His map is simply a rough sketch, wonderfully, accurate in some respects, but inaccurate in others. The boundary of the Caledonii — from the Varar estuary to the Bay of Lemannonius — must have been a natural boundary, which is provided by the string of lochs now connected by the Caledonian Canal. In that case Lem^ annonius must be the modern Loch Linnhe, anciently Lochaber. Etymologically, this conclusion is sup- ported by the fact that Loch Leven runs into Loch Linnhe (Cym. Llyn, a lake), and Leman and Leven are variants of the same word (see R. Leven). Rerigonius (Loch Ryan): perhaps from Cym. Rhe, a run or current, and Rhigyn, a notch (cf. Bolg, a notch or bay). The modern form " Ryan " = Cym. Rhean, a, streamlet. The loch, as usual, takes its name from the river that runs into it. Vindogara : the Roman station at Vandogora (called by Richard, Vanduarium), was apparently Paisley, as proved both by its situation and by the Roman re- mains found at that town. Vanduara = Gwyndwr, or white water, by which name the White Cart, on which Paisley stands, was locally called. But Ptolemy gave a similar name (Vindogara) to what seems to be the Bay of Ayr. 3 Volsas or Volas (Loch Broom): the river-name "Broom," which gives its name to Loch Broom, is a corrupt form of Braon, Breyne, or Brune, the earliest forms of the name. It is a rapid mountain stream, and takes its name from Gae. Bran, a mountain stream, itself de- rived from O. Ic. Brana, to rush forward, or to fall violently (hence probably the Scots word Brane, mad, or furious). A clue is thus given to the Ptolemaic name, 'Horsley and Stukeley read Vind as ild, i.e., the Teutonic and not the Celtic form. (See Vind in Ptolemy's place-names of Ireland.) 198 THE RACES OF IRELAND AND SCOTLAND. for Cym. ZJll means what is abrupt, or quick, and TJllaid means a sudden driving. The small bay of Ullapool, from which the village takes its name, opens from Loch Broom. It may be a relic of the Ptolemaic name, but with greater likeli- hood it is a later Norse name, meaning Ulf's hoi, or farm. The names of three headlands called by Ptolemy the V eruvium, or Verubium, the V ervedrum or Virveclrum, and the Tarvedrum, or Tarvedum, or Tarvaidunos, may repay examination. They are the three principal capes of Caith- ness, viz., Noss Head, Duncansby Head, and Dunnet Head. Tarvedum is identifiable with Dunnet Head, as well from its position on the map as by the alternative name of Orcas, which seems to relate to the Orkney Islands. From Dunnet Head the precipices of Hoy and the outlines of the Orkney hills are visible. The form of the word now most generally accepted as authoritative is Tarvai- dunos. The usual derivation of Tarvai is from Tarhh (Gae.), a bull, and there is a theory that the promontory may be associated with some form of bull - worship. That, I think, is an absurdity. Plainly, Tarvai is de- rived from Cym. Terf, extreme, Terfyn, an extremity. This etymology appropriately describes the most northerly point in Great Britain. Dunos is apparently Cym. Dinas, a hill -fort. Dunnet Head consists of numerous hills and valleys, but the Dinas is probably represented by Brough, close to the headland. I can find no distinct evidence of the remains of any fort at Brough, but the name shows that there must have been a Burg on or near, the site ; otherwise the name is unin- telligible.'^ * It i.s conceivable that Tarvedum may be Cape Wrath {am Farph), and that Parfedum may be the correct reading. THE RACES OF IRELAND AND SCOTLAND. 199 The earliest forms of Dunnet are Donotf, Dunost, and Dunneth. I suggest that in these names we may find the Dunos of Ptolemy. In the early maps of Scot- land, the headland is called Quinic Nap, and Windy Nap. Verubium is obviously Noss Head. Noss is 0. Ic. Nos, a nose, a variant of Nes, so frequently applied to head- lands in Scandinavian districts. The old name was Cat- ness, but the " Cat " has long disappeared. Verubium may be derived with some plausibility from Cym. Wyraw, to reach out, with its related sub- stantative Wyre (which probably denoted a headland), and ub, what is high, thus denoting a high promontory. Vervedrum or Vervedum contains Wyre, already noticed, and for a similar reason. But Richard of Cirencester makes the first syllable Vin, which, if correct, must be Cym. Ffin, a limit or boundary. Vedr is, I think, Cym. Givydyr, green. This headland is notably verdant.^ The present name, Duncansby Head, is quite modern. The earliest forms are Dungalsbae, Dungsby, and (in old maps) Dunsby. " By " is, of course, the usual Danish termination, denoting originally a dwelling or farm, and now a village. Therefore, Dungal or Dung is probably a personal name, that of the dweller, or farmer. There is an 0. Ic. word, dunga, meaning a useless fellow, from which it may be derived, for the Scandinavians had an unpleasant habit of giving one another pointed nicknames. This name, in turn, would easily take the Gaelic form of Dungal. '' Ptolemy's name for the Wear ( Vedra) contains the same root, derived from the same source, (.hoydyr means both glass and green, and the name of the river would thus signify glassy or shining. (See analysis of Glas as a river-name, and Mr. Fox-Talbot's comment on verre.) 200 THE RACES OF IRELAND AND SCOTLAND. From the promontories we proceed to the islands, and here we have a wide and important field of investigation, for islands, like rivers, are tenacious of their names. An analysis of the island-names is instructive. It is needful, however, to remark that Ptolemy had only a vague idea of the relative position of the Scottish islands, and except by their names, there is no sure guide otherwise to their identification. Dumna may be intended for the Outer Hebrides. The name is doubtfully from the same source as that of the Damnonii, a powerful tribe occupying the entire basin of the River Clyde, and both sides of the Firth of Clyde. Dumn is an old form of the modern Welsh Divfn, meaning deep. We find it used by Bede in the form of " Dummoc " (Dumnoc, deep water), for Dun- wich. It seems to have been applied to places bounded or approached by a deep channel, as distinguished from shallows. A cognate form, in a Teutonic dress, is sup- plied by the name " Dieppe." Thus, " Dumna " may mean a territory approached by a deep channel like the Minch (La Manche). Ehiida: the modern name Hebrides, the Ha?budae of Pliny. The "r" is intrusive, through a transcriber's error, and the error has been perpetuated to the present day. Ptolemy gives the name Ebudae to a group of five islands, which he places between Ireland and Scotland. Two of them he calls Ebuda (close to Ireland, thus showing the vagueness of his knowledge), and the others he names respectively Epidium, Maleus, and Rhicina. By the identification of Maleus with Mull, Skene attempted to identify the others, from their position in relation to ]\Iull; but that is a futile task. The names, however, may be analysed, and the analysis may be fruitful. When, some years ago, I was writing a his- THE RACES OF IRELAND AND SCOTLAND. 201 tory of the Outer Hebrides, I had to relinquish the attempt to give any rational explanation of the word Ebuda. Since then I have discovered that in Roque- fort's Glossary of the Romance dialect, Ebicdes = terreins incultes. Therefore Ebudae, Hsebudae, or Hebrides means "the wastes." The Cym. affinity for this old Gallic word is probably to be found in an allusion by Solinus to the Hebrides, which, he tells us, were destitute of corn. (Cym. Heb, void of, and Yd, corn.) It is in the highest degree likely that in the second century, agriculture Avas practically unknown in the Hebrides, which must have been devoted entirely to pasturage and the chase. (See the speech that Tacitus puts into the mouth of Calgacus, in which it is stated that the Caledonians had no cultivated lands; but this may have been a hyperbole). It seems probable that Ptolemy's two Ebudae may be Islay and Jura, as suggested by Skene. Epidium must have been near the Mull of Kintyre, the Epidium promontory. The Epidii occupied Kintyre, and perhaps the island Epidium as well. Possibly therefore Epidium was Arran. The name Epidium is, I think, derived from Cym. Ypid, the tapering point. There is an alternative suggestion for the identifica- of Epidium. Ptolemy may have duplicated the name, first, as part of the mainland represented by Kintyre (Gae. Ceann-tir, Cym. Pen-tir, Land's End — see Pen- tire, in Cornwall), and again as an island. There would be nothing surprising in Kintyre being classed as an island, attached as it is to the mainland only by the narrow isthmus between East and West Loch Tarbert (Gae. Tairbeart, an isthmus, literally, boat-draught). A curious commentarj^ on this suggestion is provided by the doubtful story of King Magnus Barcleg having 202 THE RACES OF IRELAND AND SCOTLAND. tricked the Scottish King Edgar out of Kintyre, by- crossing the isthmus in a boat dragged from one loch to the other. By thus making Kintyre an island within the literal meaning, but not the actual intention, of Edgar's grant, he was enabled to include Kintyre with all those Western Isles between which and the mainland he could go in a boat with a r udder. ^ Robert Bruce afterwards crossed the isthmus in the same manner. Tarbert in Easter Ross and Tarbert in Harris are, in each case, a narrow isthmus which similarly provided short cuts; and Tarbert (or Tarbat) on Loch Lomond marks the place where boats may have been drawn across to Loch Long in the same way. Maleus is certainly Mull. Its earliest subsequent forms are Malea, Myl, and Mula. It may take its name from the mountainous character of the island, and the source would thus be Cym. Moel, a conical hill; also meaning " bare," and therefore applied in Wales to hills with bare tops, which is the general character of the Mull mountains. (But of. O. Ic. Muli, a projecting moun- tain.) It is to be observed, however, that the two forms which are the earliest (Ptolemy's and Adamnan's) both make the root Mai, which suggests that the meaning may be derived from Cym. Mall, a soddened state, Mallus, soddened, thus denoting a marsh or bog. Monceda : Skene reads this name as Monarina (so does Elton)^ and thus easily identifies it with Arran (Mon and Arina) . But this reading is opposed to the more authori- " Magnus Barefoot's Saga, c, U. The Saga gives Melkolm (Malcolm) as the name of the Scottish king instead of Edgar. This must be an error. Elsewhere the Sagas relate a similar incident in connexion with Beiti, a mythological Sea-king. 7 Elton gives both Monarina and Monaoida as Ptolemaic island-names, thus increasing the confusion. THE RACES OF IRELAND AND SCOTLAND. 203 tative form Monseda, and the position of the island (if that counts for anything) points clearly to Mon or Man . (? Cym. 'Mdon, habitation, or Mawn, peat (Manau) ). Rhicina : the remaining island of Ptolemy's group of five is usually identified with Rathlin, the small island on the north coast of Ireland. The old forms are Rachra, Ragharee and Reachrainn. An old name of the Isle of Thanet, Ruoichin, seems to contain the same root. In O. Welsh, Rag ynys means " an adjacent island," and Rag shows itself in these island-names. The northern division of Ptolemy's islands comprises Ocetis (amended to Sketis), Dumna, Orcades, and Thule, in the order of their latitude northwards. Orcades: the Orkneys. The origin of this name has given rise to a good deal of conjecture. It is usually attributed to Gae. Ore, a pig or a whale. The Gaelic "whale" must be a porpoise! The meaning is probably to be found in Cym. Orch, a limit, or rim, the Orkneys being the islands beyond the limit of Scotland in the North, e.g., Dunnet Head, Ptolemy's alternative name for which, as we have seen, is " Orcas." Probably the modern form " Orkneys " is from 0. Ic. Orkn, a kind of seal; perhaps a Norse interpretation of the Cymric name. In 0. Welsh the name appears as Orch, which supports the derivation I have given. Sketis (if that is the correct reading of Ocetis, which, per- haps, is doubtful), stands for Skye. The earliest forms of Skye are Scia, Scith, and Skid. The source appears to be Cym. Ysgi, cutting off, in allusion to the jagged nature of the coast-line. The Norse name for the island was Skid, a chopped piece or a splinter, which is a related idea. (Of. also Goth. Skaidan, to divide or sever.) The position of Ocetis, it may be added, does not correspond with that of Skye. 204 THE RACES OF IRELAND AND SCOTLAND. Thule : if Ptolemy had a confused idea of the situation of the western isles, how much more ignorant was he of the situation of Thule, the mysterious island so frequently mentioned by ancient geographers, and so vaguely placed by them. Ptolemy splashes it, so to speak, into the ocean, anywhere away up in the unknown North. The name is Teutonic (but the authorities are in disagree- ment as to its source), and there can be little doubt that the supposed island which caused the geographers and some Roman writers so much trouble, was Scan- dinavia itself. Dr. Nansen (In Northern Mists) gives excellent reasons for that belief. Let us now turn to Ptolemy's tribal names and see what we can make of them. Commencing with the northern ex- tremity, we find a group of " C " names, which are plainly Cymric in form. Cornavii occupied the extreme north. The name is from Cym. Corniaw, to butt. The Cornavii of Caithness and Cornwall were the people at the butt or extremity. The Cornavii of England occupied the land butting into the sea between the Dee and the Mersey. Cym. Corn, = Eng. Horn. Caerini occupied the Assynt country in Sutherland. The name seems to be connected either with Cym. Caer, a wall or fort, or Caeor (Cym.), a sheepfold. But Richard of Cirencester calls this tribe the Catini, a name obviously associated with the Cat of Caithness; and he places the tribe not on the west, but on the east coast, where, in point of fact, Caithness is situated. It is customary now to sneer at the whole of Richard as being the work of a convicted impostor. But it has been proved that his description of Britain is accurate in details that were unknown until modern research re- THE RACES OF IRELAND AND SCOTLAND. 205 vealed them; and it would appear that the compiler really had access to authoritative documents. Perhaps, therefore, Catini should be read for Caerini. Camden, who wrote more than a century and a half before Ber- tram forged Richard, asserts that Catini is the correct reading of Ptolemy; but following Ptolemy, he places the tribe on the west coast. Cat means heath. ^ Carnonacae : placed in West Ross-shire. Possibly the people of the Carr or Carron (Carr-avon). In that event, Cym. Non, stream, and Ach, river (a seeming duplication), may be represented in the name. But it may be derived with greater probability from Cym. Carnen, a heap (Cairn), thus making Carnonac mean a stony or rocky f)lace. Creones or Cerones: on the west coast of Inverness-shire. Perhaps from Cym. Cri (Crech), meaning "rough": the people of the " Rough Bounds," as the district was sometimes called. Damnonii (seeDumna): the name may be referable to the fact that they were situated on the Firth of Clyde. Similarly, the Damnonii of England were bounded on the north by the Bristol Channel. The name survives in Devon. Damnonia, the country of the Damnonii, is a name found in Gildas. Perhaps an equally likely derivation can be traced to Sansc. dJidman, signifying dignity, heroism, and similar qualities. Dom, in O. Frisian, generally found in compounds, but existing originally as an independent word, is said to be cognate with dhdman. The Damnonii in England and Scotland were clearly a most important tribe. Dominus also suggests itself as a pos- sible source of the name. * See the discussion of this root in the Scottish prefixes. In Ptolemy's map of England, a tribe called Catysuchlani, placed in the modern Hert- fordshire, has a name, the prefix in which may be the Cat, or heath, root. 206 THE RACES OF IRELAND AND SCOTLAND. Epidii : (see Epidium). Novantae occupied Galloway. The root contained in this name seems to be Gym. Bant, a high place, with the prefix Nw or Ny signifying a characteristic. This, again, agrees with the modern name, Galloway, for Gale in Cornish (Welsh Gallt, an ascent), means a high place, and Gwyddle is a woody place. Selgovae occupied the country west of the Novantae. The essential root in this name appears to be Cym . Sivl, a flat space or ground. This derivationisborneout by tliee.f . of Solway (Sulway and Sulloway), the name of the Firth receiving the rivers that traversed the country of the Selgovae (the Annan, the Nith, and the Dee). As their neighbours, the Novantae, occupied the hilly country of Galloway, so the Selgovae were the inhabitants of the plains to the east of them; hence apparently their name. The root gov is probably derived from Cym. gwyfato, to run out or flat. Otalini : coming round to the east coast, we find this tribe (in later editions Otadeni), occupying the district be- tween Hadrian's Wall and the Forth. A probable derivation for this tribal name is from Cym. 0th, what is exterior (or Wt, what is out), and Linn, a marsh, or Lleyn, a low strip of land, thus signifying the coast people. (Their territory extends from the Wear to the Forth.) Venicones were the people of Fife, Forfar, and Kincardine. The name seems to be related to Welsh Ffwynog, a meadow, and especially to Corn. Whynick, a marsh; Winnie, fenny. (Cf. Cym. Gwcen, a meadow.) Tcexali or Tmzali were the people of Aberdeenshire and Banffshire. The promontory of Tsezalorum is Buchan- ness or Kinnaird's Head. Tsexali may be identical with Texel at the mouth of the Rhine, a significant THE RACES OF IRELAND AND SCOTLAND. 207 circumstance (C/., also, the Scottish isle named Texa). The name suggests the plant teasel, or tazel (A. S. toesel), as a feature of the country. The district in- habited by the Tsezali may have abounded in dipsacus. A preferable derivation may be from Cym. TaivcJi, foggy. CJi would take the form of x (cf. TJxell for Vchel). Vacomagi lay between the Tsezali and the Caledonii. They occupied the County of Elgin, Strathspey, Strathavon, Braemar, and Strathardle. The latter part of the name is clearly the Celtic magus, a plain, and the prefix sug- gests Cym. Givag, void or empty. But this cannot mean a depopulated plain, unless it signifies that the Vacomagi seized unoccupied territory. The word may be a hybrid, the prefix being from 0. Ic. Vdkr, moist. Vacomagi would thus mean the people of the marshy plain. Caledonii : their territory and name have already been dis- cussed. Decantae may have occupied both sides of the Moray Firth, hence the significance of the prefix. Cant seems to be referable to Cym. Cant, a rim. The name Decantae would thus mean the people on both shores (of the Firth). Lugi occupied the country on the east coast of Sutherland. The name probably means the marsh people. Lug, and Leog (Gae.), being cognates. (The god Lug is some- times invoked to explain this name!) Probably Cym. Llwch, a lake, is the source of Lug. Smertae or Mertae : location near Loch Shin. Stokes makes the root Smer, to shine. If that is the fact, it seems to confirm my interpretation of the name Shin (which see) as being related to the Eng. word "shine." (" Smert " is found in personal Celtic names.) 208 THE RACES OF IRELAND AND SCOTLAND. The town names need not detain us long; several are associated with the river - names. Alauna (the river- town) on the east coast, looks like Inverkeithing; Alauna, on the west coast may be Dumbarton; Devana is the settlement on the Deva; Orrea, on the Ore^ (but the river- name is not given; it is derived from 0. Ic. Orr, swift); Tamia, on the Tay; Tuesis, on the Spey; Rerigonius, on Loch Ryan; and Vandogara, on the White Cart. Bannatia suggests Cym. Banad, broom; and Lindum shows the stem Lind, meaning marsh (0. Welsh Linn). There seems to be no reason to doubt that Victoria is a boastful name given to the site by the Romans. Castra Alata is evidently Burgh- head. This place is named Ptoroton by Bertram (Richard of Cirencester) ; and it is a curious circumstance that a local name for Burghhead is (or was, some years ago) Tor- rietown (Cym. Tivr, tower). In the ill-defined portion of Ptolemy's map near the Sol- way Firth, there is a group of three towns, the names of all of which suggest a high situation. Carbantorigon is resolvable into Cym . Caer, a fort or city ; Bant, high, or a high place (see Novantae); and Rigon (see Rerigonius). Uxellum is from Cym. Vchel, high. Trimontium probably does not mean " the three moun- tains," but " the mountain town " (Cym. Tre and Mynydd). Perhaps these towns were really in Galloway, but have been placed too far to the east. It is useless to attempt to identify them with any modern names, s Cf. Orrock in Fifeshire. CHAPTER XIX. Conclusions to be drawn from the analysis of Ptolemaic names in Scotland — The first clear view of the Pictish monarchy in Scotland — Bede on the origin of the Picts — The Am/lo-Sa.von Chronicle and the Picts— The two divisions of the Pictish nation — The Irish traditions of the origin of the Picts — The probable sources of these traditions — The versions of the Pictish Chronicle and Nennius — Claudian on the Picts — Cymric and Scandinavian elements. The conclusions to be drawn from this excursion into Ptolemaic geography Avill now be stated. Allowing for any etj^mologies that subsequent analysis may show to be unten- able, there will remain a residuum of unassailable evidence to prove the predominance of the Cymric language in Cale- donia during the first and second centuries of the Christian era. It is true that the presence of Cymric place-names in the second century does not necessarily imply a contemporary Cymric population. The Celts who originally named the places had doubtless long disappeared before Tacitus or Ptolemy recorded the names; and it is conceivable that their successors in the second century may have been of a different race, though they retained most of the place-names of the Celts. Yet the Cymric shape of the tribal names seems to prove, not necessarily indeed that the tribes themselves were Celts, but certainly that a Cymric language was spoken in some parts of Caledonia in the second century. Ptolemy's sources of information are unknown, but his informants must have got their facts about the tribes from Cynn-ic- speaking persons. The tribal names supplied by these Celts may not have been the names acknoAvledged by the tribes themselves; they may have been employed merely as names 14 210 THE RACES OF IRELAND AND SCOTLAND. descriptive of the character of the country occupied by the tribes. Thej^ are, in point of fact, mainly topographical, as I have tried to show, and so regarded, they offer only a slender clue to the ethnology of the tribes to which they were applied. But the evidence that they offer of the exis- tence of a Cymric tongue in Caledonia during the second century is unmistakable. The analysis also proves, if less decisively, that there was another philological element co-existing sporadically with the Cymric. That is shown by the place - names of Teutonic, and apparently Scandinavian, origin that I have analysed. Again allowing for error, the exis- tence of that element in the Ptolemaic names cannot well be doubted. The name Varar, applied to the Beauly Firth, would a^^pear to suggest that it was the channel by which tribes of Scandinavians entered the country. They may have been the refuge-seeking Celyd- dons mentioned by the Welsh Triads, and, if, as I have sup- posed, the boundary of the tribes whose distinctive name was the Caledons, stretched from the Moray Firth to Lo,ch Linnhe, that suggestion is not without support from the following facts. When we get the first clear view of the Pictish monarchy, we find that it was seated on the banks of the B-iver Ness. That river-name does not appear in Ptolemy. It is first mentioned by Adamnan, who tells us of St. Columba's visit to the Pictish King Brude at his capital on the Ness. The river-name " Ness " is Teutonic (c/. the Nissa in Sweden, the Neisse, Nesse, and Netze in Germany), and is ultimately derived from Sans. Nis, to flow. Ness is the Teutonic, and Netze the Slavonic form of the word. It is impossible to avoid the suggestion that this Teutonic river-name, inti- mately associated as it Avas with the Pictish monarchy when it first emerges into the clear daylight of history, may denote Teutonic hegemony ; and if the suggestion is pressed further^ THE RACES OF IRELAND AND SCOTLAND. 211 it is easy to believe that the River Nissa in South - West Sweden, situated in nearly the same parallel of latitude as the Scottish River Ness, was the centre of the district from which this ruling people may have proceeded to the North of Britain. In the time of Jordanes (sixth century), this district (West Gothland) was inhabited by a people whom he calls " Gautigoth," and whom he singles out from their neighbours as specially brave and warlike. In the time of Bede, the tradition about the place of origin of the Picts was that they had come from " Scythia." I have already examined this word to show its geographical vagueness; and have suggested a sound method of ascertain- ing what was meant by the writers who used it. Folio wing- that method, we find that the Ravenna Geographer (who must have used geographical terms in the sense in which they were understood in Bede's time) places Scythia to the west of the Vistula. But he states that " Old Scythia " was the name given by most cosmographers to Scandia, i.e., Scandinavia. Therefore, we are, I think, justified in con- cluding that by Scythia, Bede must have meant Scandi- navia. The Anglo - Saxon Chronicle gives the same account (copied, no doubt) as Bede, with the additional in- formation that the Picts came from the south of Scythia, which, we may take it, means South Sweden. It is a curious commentary on the Chronicle's statement, that Geoffrey of Monmouth and Layamon, while mentioning the Norwegians, the Dacians (Danes) and the Picts in association, says nothing about the Swedes. The inference may be that they believed the Picts to be Swedes. There is some ground, therefore, for the belief that the Picts were originally bodies of Swedes, or Goths from the South of Sweden, who settled in North Britain after ravag- ing the country and plundering the Cymric inhabitants during an undefined period. That, indeed, seems to be the inference to be drawn from the statements of Gildas. He 212 THE RACES OF IRELAND AND SCOTLAND. saygi that the Picts were a foreign nation who, in associa- tion with the Scots, harassed the Britons for a lengthy period, and who settled down in the northern part of the island only after the close of the Roman period in Britain. They remained there, says Geoffrey, , " mixed with the Britons." The facts may be that the tribe of Caledons represents the earliest settlement of Scandinavians; that a lengthy gap separates this settlement from the arrival of the later waves of Scandinavian origin ; that these new - comers for a long period led a restless life, their chief occupation, by land and sea, being that of plunderers, or Piccardach; that finally, they turned to pastoral and agricultural pursuits, and mixed with the earlier inhabitants; and that they them- selves, in turn, became the prey of hungry hordes, some from the same nest as themselves, and others from the moutlie of the Rhine, or the Weser, or the Elbe. Jordanes well calls Scandia " the hive of nations," and it is tolerably certain that during the migratory centuries, no inconsiderable propor- tion of the swarms from that hive fastened upon the east of Scotland. It is a well-authenticated feature of Scandinavian history, that owing to the redundancy of the population in relation to the means of livelihood, the pressure of famine occasionally made forced emigration a necessity; and lots were cast to decide who should go.^ It is by no means im- probable that the " refuge-seeking Celyddons " of the Welsh Triads, and the big, red-haired men of Tacitus belong to this category. The people whom the Romans called Picts may have been forced from their homes by economic causes, or in search of plunder. In any case, their numbers, at first small, but augmented by successive colonies, seems to have been considerable in the aggregate. Occupying apparently that part of the country north of the Firth of Forth which ^ Gildas, Sec. H and Sec. 21. "^ Bosworth's Ori See Lang's History of Scotland, vol. i., Appendix C. 384 THE RACES OF IRELAND AND SCOTLAND. silent but steady stream of iiiiiuigration from Ulster to the opposite coast; and here we have another factor in the mixture of races which cannot be ignored. The Dalriadic Scots left their mark mainly (and naturally) on the West Highlands. They spread over the Isles and the adjacent mainland, mixing with whatever Pictish and other elements may have preceded them, and subsequently with the Scandinavians who followed them; the combined races forming the restless and turbulent clans of West Highland history. There was probably a clear- cut division of the Scots after the succession of Kenneth MacAlpin, one division comprising the West Highlanders, and the other the subjects of the "Fife sovereignty," to whom would fall the chief spoils of Kenneth's successes. At an assembly of the Scots held at Forteviot, Kenneth's brother and successor, Donald, agreed with the Goedeli (Gael) for the adoption of the ancient laws and statutes of Aed Finn, framed in the eighth century. *' This implies, apparently, that some of his Scottish subjects Avere not Dalriads; for these statutes were already recognised b}^ the Dalriads, and had been the law of Dalriada for a century. There is thus some ground for the suggestion that there was a cleavage between the Dalriadic Scots in the west and the Scottish tribes in the east; and beyond doubt, the cleft was subsequently widened by the introduction of feudalism under the Scoto-Saxon Kings of Scotland, who succeeded the MacAlpin dynasty. If this cleavage first showed itself during the sway of that dynasty, it explains circum- stances that are otherwise obscure. Friction could hardly be avoided between the tribes who received lands in the fertile plains of the east and the centre, and those that had to be content with the barren hills of the west. A feeling of antagonism, due to a sense of unfair treatment, would be aroused in the west against the Scottish Crown. That ■" Innes, Appendix iii. THE RACES OF IRELAND AND SCOTLAND. 385 feeling may have inspired the rise of a kingdom of the Isles and the West Highlands under the hegemony of the Heads of Clan Donald (who signed treaties as independent monarchs); the coquetting with England, and the actual co- operation with that country at intervals against the Scottish Crown; the anti-national attitude of the west during the War of Independence until the firm statesmanship of Bruce *- allayed the spirit of discontent; the fight for supre- macy between east and west at Red Harlaw; the campaigns under ]\Iontrose and Dundee against the Lowlanders; and the Stewart risings, culminating in the " Forty-five " and the final rout at Culloden. It would be absurd to assert that a grievance having its inception in the ninth century, was the only cause, or even the mainspring, of this persistent spirit of revolt against the centre of authority in the south. But all nations (and the Celts in particular) have long memories for national injuries ;'^^ and it is impossible to ignore the patent existence of a feeling of rancour which found its ex- pression not only in deeds, but in actual words. How are we to explain otherwise the fact that the West Highlanders not merely repudiated the name of Scots — they called them- selves Albinnich (a territorial name) or Gael (a racial name) — but in 154-3 professed themselves to be the " auld enemies " of the realm of Scotland?^* Here we have the anomaly of the " auld Scots," as the Highlanders were called in the south, proclaiming to the world their enmity to^^"ards the realm of Scotland; and similarly (in 1543) we find their leader, Donald Dubh, expressing his willingness to take up arms against " all Scotishmen his enemies." John Elder, *'^ The descendant of a Yorkshire family, in the person of Robert Bruce, saved the independence of Scotland. The Stewarts were probably of Pictish descent. *' I have already cited the instance quoted by Douglas Hyde (chap, iv.) of the Picts of Ireland cherishing a national grievance for nine hundred years. ■•* The letter of the chiefs to Henry VHL appears in S(a(e Faptrs, iv., pp. .501-4. 25 386 THE RACES OF IRELAND AND SCOTLAND. himself a native of Caithness, in a letter to Hemy VIII., describes the Highland chiefs as the " Yrische lords of Scot- land commonly callit the Reddshanckes, and by historio- graphouris Pictis." Elder's knowledge of racial facts was no better than that of many more enlightened students in later years, but his statement suggests that the chiefs may have believed that they had inherited rights from the Picts which the Scottish Crown had ignored. ^^'^ His designation of the chiefs as " Irish " lords is a curious contradiction of their alleged Pictish descent, if Scottish Picts are meant; but it shows that by this time, not only were the Gael of Scotland and their language called " Irish " by the Low- landers (which w^as the fact), but also by natives of the north. The Scots people in the sixteenth century were the Lowlanders; and the Scots language was "quaint Inglis." As a distinction, the Gaelic-speaking Highlanders and their language were called '' Irish," in recognition of the origin of both. For, long before the sixteenth century, the Scots of the eastern and central districts had been absorbed by the more numerous Pictish people, W'hose language they had gradually adopted while shedding their own Gaelic speech. They remained, however, the dominant caste, and were thus able to perpetuate and impose upon their Pictish neighbours their distinctive name of " Scots," which gradually displaced the name of " Picts," though both peoples were called by the latter name for an undefined period after the accession of Kenneth MacAlpin.*^ *^ It is by no means improbable that some of the chiefs may have been descended from Irish Picts. There was, of course, a distinctively Scandi- navian leaven in some West Highland families, but it was introduced in post-Pictish times. *'' If we are to accept as literally accurate the accounts of the Battle of the Standard given by Richard of Hexham, there was little to choose, in point of ferocity, between the Picts, the Scots, and the Angles in the Scottish armj\ The Scots, particularly, are accused by Henry of THE RACES OF IRELAND AND SCOTLAND. 387 The name of the Picts disappeared; but the people them- selves remained. And their descendants are to be found at the present day in greatest number in the eastern counties, where the inhabitants are sharply divided in dialect and customs from the descendants of the Angles in the Lothians ; and more strongly contrasted with both, are the Gael of the north and the west. The difference is not one of language alone, but of temperament as well. Partly due to environ- ment, it derives much of its vitality from racial traits of character, the existence of which it would require some hardihood to deny. The temperamental gulf which divides the inhabitants of north-east Ulster from those of south-west Cork, is no wider than that which separates the people of Sutherland from those of Selkirk. One may go further, and assert that there are strongly marked distinctions even between dwellers in the same county. A native of Easter Ross is different in temperament from a native of Wester Ross; and between Inverness on the east and Glenelg on the west, there is a border line whence racial traits diverge. The further east one goes, the more does the Pictish blend betray its presence ; the further west one goes, the more do characteristics appear which, for convenience, may be described as " Celtic." But the process of assimilation, greatly accelerated by the dissolution of the clan system, and particularly hy Huntingdon of atrocities of a kind to which recent wars have accus- tomed us. The curious statement is made by John of Hexham that, after the battle, the King of Scotland took hostages from the Scots and Picts to stand by him in every conflict and danger ; and it is added that he fined them in a large sum of money. In the burgh seal of Stirling — a town with a Cymric name — the words Bruti Scoti (Scots brutes !) are applied in 1296 to dwellers benorth the Forth. The Stirling burghers were probably of Anglic descent, if Beddoe's surmise (Racen of Britain^ pp. 243-4) is correct, that the Angles of Lothian pushed en masse in the direction of Stirling to the ford of the Forth and the Campsie Fells. This is one more illustration of the racial admixture that had to be unified, and the racial animosities that had to be allayed, in the work of consolidation. 388 THE RACES OF IRELAND AND SCOTLAND. the steady development of educational machinery throughout the countrjs has blunted the edge of racialism since the eighteenth century, and has removed one by one the barriers that formerly divided the north from the south, the cast from the west. Temperamental distinctions remain, and will continue to remain, to lend variety to the component parts of the nation, and prevent (not unfortunately) the attainment of an ideal of dull uniformity. A blend of temperaments is a good thing for a nation, if they are not conflicting but complementary. It has been shown that the welding together of the diverse racial units, from Shet- land to the Tweed, into the Scottish nation, "was a long and arduous task. It could only have been accomplished by the hand of Time; and in the development of a national ideal, the realisation of a community of interests, the hand of Time continues to work beneficently by moulding the different elements into a state of more complete unification. CITATIONS FROM MODERN WORKS. A ctif of Parliament of Scotland, 380 Bannati/ne Miscellany, 136 Beddoe— 27je Races of Britain, 26, 43. 43, 368, 387 Betham — The Gaul and Ct/i?ihri, 87 Birch — Cartularium Sa.conicum, 138 Borlase— 7'As Dolmens of Ireland, 32, 41, 42, 72, 91, 280 Bosworth — The Origin of English, Germanic, and Scandinarian Languages and ^\Uions, 95, 212, 251, 252 Bryant (Mrs.)— Celtic Ireland, 90 Buchanan — Travels in the Hehrides, 237 Cambridge Modern History, 8 Campbell — West Highland Tales, 118, 155. Carmichael — CVmHma Gadelica, 8 Chalmers — Caledonia, 230 Comparetti— T/ie Traditional Poetry of the Finns, 40 Crichton and Wheaton —Scandinavia : Ancient and Modern, 307 Cumming (Miss Gordon) — From the Hehrides to the Himalayas, 11 Deane — Ilie Worship of the Serpent, 13 Dixon — Gairlorli, 37, 60 Du Chaillu— T/^e Viking Age, 44, 45, 137 Dunham — Denmark, Sweden, and Norway, 57, 59, 166 Elton— Origins of English History, 202, 375 Ferguson — llie Teutonic Name System, 146, 244 Frazer-7V^^ Golden Bough, 7, 13, 29, 232, 233 Haddon and Stubbs —Councils and Ecclesiastical Documents, 53, 86, 103 Henderson — Survivals of Beliefs among the Celts, 11 Hewitt — Primitive Traditional History, 17 Huxley — Prehistoric Remains of Caithness, 21, 22, 23, 42 ,, Critiques and Addresses, 81 Hyde — Literary History of Ireland, 59, 147, 385 Innes—The Ancient Inhabitants of Caledonia, 216, 341, 343, 344, 346, 348, 356, 358, 359, 377, 379, 381, 384 Jamieson — Etymological Dictionary of the Scottish Language, 6, 230, 251 Joyce — The Origin and History of Irish Names of Places, 130, 145, 252 Keane — The Lapps, 37 ,, Ethnology, 75, 76 Keating — History of Ireland, 6, 26, 31 Kitchin — Ilistury of France, 175 Lang— Hislury of Scotland, 383 390 CITATIONS FROM MODERN WORKS. Latham — The Germania of Tacitus, 97, lOi ,, Ethnology of the British Islands, 108 ,, The English Language, 217 Lockhart— i/(/e of Sir Walter Scott, 48, 85, 135 MacBain — Ptolemy s Geography of Scotland, 230 M'Clure— iJr/VwA Place-names, 280 MacRitchie — Fians, Fairies, and Picts, 62 MaMet— Nor tJm-n Antiquities, 43, 45, 57, 61, 137, 155 Martin — Description of the Western Islands, 8, 9, 136 Menzel — History of Germany, 94, 108 Moore- History of the Isle of Man, 46, 370 Nansen— /j( Northern Mists, 20, 34, 140 New Statistical Account of Scotland, 331 Nilsson — Primitive Inhabitants of Scandinavia, 34 O'Curry— Lectures on the 3IS. Materials of Ancient Irish History, 13, 14, 22, 23, 28, 43, 58 O'Grady— History of Ireland, 14, 15, 21, 23, 44, 45, 49 Sifra Gadelica, 17, 34, 61, 87, 117 Picard — . . . de jjrisca Celtop(edia, etc., 2H Pinkerton — Voyages and Travels, 38 ,, An Eiiquiry into the History of Scotland, 214 Proceedings of the British Academy, 31, 59, 81 Proceedings of the Society of Antiquaries of Scotland, 35, 188, 244 Rhyfi—Celtic Britain, 100, 139, 150 Rhys and Jones— The Welsh People, 129, 179, 232 Robertson — Index of Missing Charters, 380 Roquefort — Glossaire de la Langne Romane, 178, 284 Seebohm — The Tribal System in Wales, 245 i=ikene— Celtic Scotland, 21, 193, 222,229, 230, 236, 237, 240, 241, 275, 287, 297, 327, 341, 342, 343, 344, 345, 346, 351, 352, 370, 371, 382, 383. Skene— Chronicles of the Picts and Scots, 215, 221, 242 The Highlanders of Scotland, 81, 100, 229, 230, 376 ,, Four Ancient Books of Wales, 229 Stanley — Lectures on the Jewish Church, 13 State Papers, 385 Talbot — English Etymologies, 63, 199 Taylor — Words and Places, 319 ThovYie-Northern Mythology, 39, 49, 51, 53, 139 Todd— r/)e War of the Gaedhil with the Gait I, 102, 374 Tudor — The Orkneys and Shetlands. Ua Clerigh — History of Ireland to the Coming of Henry II., 106 Webster — The Basque and the Celt, 87 Wentz — The Fairy Faith in Celtic Countries, 50, 53 INDEX. Abravanuiis, 11)3 Adamnan, 62, 110 Adder, 256 Ailsa Craig, 264 Alauiia, 193, 208 Alban, meaniug of, 1S3 Almond, 256 Anglo-Saxon Chronicle and tlio Picts, 211 Angus, 2U3 Annan, 256 Aray, 256 Arbr.jatb, 203 Ard-righ of Tara, 103 Argyle, 293 Dalriadic Kingdom in, 342 Arran, 261 Assynt, 294 Athole, 294 Attacots, the, 350 Awe, 256 Ayr (see Aray) Badenoch, 294 Balfour, 295 B.ilor of the Evil Eye, 25 Banchory-Devenick 295 Banchory-Ternan, 295 Banff, 296 Bangor, 295 Bannatia, 208 Bannockburn, 296 Banshees, the, 31 Barra, 265 Barrow, the, 127 of the Boyne, 51 folk, 48 Basques, the, 87, 117 Bcauly, 296 Bede, 111 on the Picts, 211, 219 Beith, 296 Belgae, the, SO Beltine, 5 Bel worship. 9 Benbecula. 265 Ben Lomond, 263 Ben Nevis, 263 Berwick-on-Tweed, 296 Blair, 278 Blantyre, 296 Boderia, 193 Bodotria, 188 Book of the Invasions, 3 Brahan, 256 Brander, 256 Brechin, 297 Brehon Laws, 113 Broadheads, the, 76 Brora, 257 Buccleuch, 297 Buchan, 297 Buchanan-, 297 Bulgarians, 19 Burntisland, 297 Bute, 265 Buvinda, 125 Caerini, the, 204 Cairns, 262 Caithness, 279, 282 Calder, 257 • Caledonians, the, 184 Calgacus, 185 Callander, 297 Carbantorigon, 208 Cargill, 298 Carham, Scottish victory at, 3'U Carmichael, 298 Cai'nouacae, 205 Carrick. 160, 298 Carriden, 298 Carron, 257 Cart, 257 Cashel, 160 Catini, the, 204 Cavan, 160 Celnius, 194 Celtae, the, 77 Celts, the, 73 Cerones {see Creones) " Cesair," 3 Chauci, the, 94 Cherusci, the, 97 Clackmannan, 298 Clan Donald and their influence, 385 Claudian on the Picts, 216 Clota, 188, 194 Colonsay, 266 Comrie, 266 Conan, 257 Conn of the Hundred Battles, 58 Cormac, King, 5 Cornavii, the, 204 Craill, 299 Cramond, 299 Creeves, Irish, 145 Creones, 205 392 THE RACES OF IRELAND AND SCOTLAND. Criech, 29lt Crieff, 21*9 Cromarty, ;iOO Crom Cruaich, the idol, 10, 15 Cromdale, 301 Cruithne, the, 43, 132, 147, 211 and Dananns, 61 Cuchullin, 115 CuchuUin Saga, 48 Culdees (see Keledei) Culloden, 301 Cumbernaukl, 26(J Cumbraes, 266 Ciimlodden, 301 Cunningham, 301 Cupar, 301 Cymri, the, 267 Cymric Laws, 114 Dagda, the, 29, 51 Dalaradia, 141 Dalbeattie, 21t6 Dalriadic Kingdom in Argyle, 3 in Fife 342 Sovereignty, extent of, 344 Damnonii, 205 Dananns, the dominant power in Ire- ♦land, 27 in Scandinavia, 27 physical features of the, 43 in the Irish Texts, 44 mediri'val notions of, 50 and Cruithne, 61 Danes' Cast, the, 148 Danes and the downfall of the Scottish Monarchy, 376 Decantae, 207 Deer, 301 Derry, 163 Deskford, 257 Deva, 194 Devana, 20S Deveron, 257 Devon, 257 Dingwall, 302 Dinnsenchus, the, 84 Dollar, 302 Don and Doon, 358 Dornoch, 302 Douglas, 358 Druidism and its significance, 54 Druid stones, 58 Drosten Stone, 243 Druids of Gaul, 54 Ireland, 55 Drumalban, 303 Dugalls, 373 Dull, 303 Dumbarton, 303 Dumfries, 304 Dumna, 200 Dunbar, 304 Dundee, 304 Dunedin (we Edinburgh) Dunfermline, 304 Dunipace, 305 Dunkeld, 305 Dunnichen, battle of, 355 Dunnottar, 365 Dunoon, 306 Dunvegan, 306 Dupplin, 306 Durness, 306 Dusk Water, 257 Dwarfs, the, 50, 61 Dysart, 163 Earn, 258 Ebuda, 200 Eden, 165 Edinburgh, 307 Edington, 307 Eigg, 267 Eilean Dunibeg (see Luchruban) Elf-myth, the, 139 Elf-worship, 46 Elgin, 308 Elliott, 258 Elves, the, 44, 49, 61 of the Scandinavians, 33 Emania, destruction of, 146 Epidium, 201 Ere, the sons of, 342 Erich t, 258 Erin, meaning of, 86 Esk, 258 Ewe, 259 Fairies of Ireland and Scotland, 33 Falias, 28 Falkirk, 309 Falkland, 309 Fasque, 309 Fearn, 165 Feini, meaning of, 119 Fianna, the, 116, 117 Fife, 310 Dalriadic kingdom in, 341, 342 an appanage of Dalriada, 347 Scottish settlements in, 342 Findhorn, 259 Fingalls, 373 Finnias, 28 Finn-men, the, 140 Fionn, 166 Fionn Saga, 47 Fionn the serpent-destroyer, 14 Firbolgs. 4, 18, 26 Fire customs, 6, 7, 8, 9 Fir-Sidh, the, 31 Flan nan Isles, 267 Fleet, 259 Fodla, 21(4 Fomorians, the, 4, 25 Forfar, 310 Forgan, 310 Forres, 310 Forth, 259 INDEX. 393 Fortingall, 311 Fortrose, 311 Frey, the god, 52 Frisian settlement in Scotland, 223 Gael, the, 110 descended from four stocks, 82 origin of the, 91 and the Saxon, 179 Gaelic language, how fonned, 122 evolution of, 179 peculiar characteristics of, 180 Gaelic tribes in the west, 384 Gairloch, 311 Gala, 259 Galashiels, 311 Galcacus (see Calgacus) Gall-Gaidel, the, 375 Galli, the, 77 Galloway, 311 and the Picts, 379 Garioeh, 312 Garry, 166 Gartnait, 242 Geanies, 312 Geolfrey of Monmouth, 211" Gigha, 267 Gildas on the Picts, 218 Giraldus, Cambrensis, 223 Girvan, 259 Glasgow, 312 Glassary, 313 Glencoe, 313 Glenelg, 313 Goidel {see Gael) Golspie, 313 Gorias, 28 Goths, the, 69 Govan, 313 Gowrie, 314 Greenock, 314 Hawick, 314 Helmsdale, 315 Hibernia, meaning of, 85 Hill-folk, 48 Holyrood, 315 Horesti, 189 Huntly, 315 Iberians in Ireland, 90 lena, 195 lerne, meaning of, 85 Ha, 195 Illusionism and the Dananns, 59 Inchaffray, 315 lona, 267 Ireland, different names for, 85 earliest notices of, 93 in the second centurj', 93 Teutonic settlement in, 94 Heroic Age of, 104 Gael in, 121 Ireland, St. Patrick and education, 122 tradition and ancient tongue of, 123 place-names of, 158 Irish genealogies, 82 Irvine, 259 Isla, 259 Island nomenclature, 264 Islay, 268 Itis, 195 Jedburgh, 315 Jura, 268 Keith, 282 Keledei, the, 371 Kelso, 315 Kelvin, 259 Kerrera, 268 Kilsyth, 316 Kilt as a Gothic dress, 108 n Kilwinning, 316 Kinghorn, 316 Kinross, 316 Kirkcaldy, 317 Kirkcudbright, 317 Kirkintilloch, 317 ' * Kirkwall, 317 Kirriemuir, 317 Knapdale, 317 Knock, 288 Knotted cord, custom of the, 36 Knoydart, 317 Kyle, 318 Laighin, meaning of, 119 Lairg, 318 Lammermuir, 318 Lanark, 318 Lajjjjs, the, 32, 36 Shaminists, 39 Larg Hill, 318 Lasswade, 319 Lauder, 259, Layamon on the Picts, 220 Leeds, 271 Leinster, the Book of, 3 Leith, 259, 319 Lemannonius, 196 Lennox, 319 Leslie, 319 Leven, 259 Lewis and Harris, 269 Lia Fail, the, 55, 56, 57 or Stone of Destiny, 28 Linlithgow, 319 Lismore, 271 Lochlans, the, 372 Lochwinnoch, 319 Longheads, blonde, 75 Longus, 195 Lorn, 320 Lothian, 320 394 THE RACES OF IRELAND AND SCOTLAND. Lothian, struggle for possession of, 356 ceded to Scotland, 363 Northumbrian settlements in, 366 Loxa, 195 Luchruban, 3-1 Lug, 29, 52 Luing, 271 Lumgair, 320 Luprachan, the, 45 Luss, 320 Lyon, 260 MacAlpin, Kenneth, as Kint Picts, 377 Macpherson, James, 143 Mailcu, 242 Maitai, the, 149 Maleus, 202 Mar, 320 Marchmont, 321 Maree, 321 Markie, 260 Markinch, 322 Marnoch, 316 Maybole, 322 Mearns, 322 Melrose, 323 Menapii, the, 95 Mertae (see Smertae) Methven, 32c Miledh, 66 Milesian legend, the, S3 names, 85 Milesians, the, 65 and Spain, S3 Minto, 323 Modana, the, 128 Moffat, 323 Moidart, 317 Blona, Money, lOS Monceda, 202 Montrose, 324 Moray, 324 Mormaers, the, 380 Mora stone near Upsal, 56 /;. Mount-folk, 48 Moy, 168 Moytura, battle of, 25, 28 Muck, 271, 290 Mull, 29o Rlurias, 28 Musselburgh, 325 Nabarus, 195 Nairn, 260 Naitan (.sec Nectan) Nectan, 241 Nel, 66 Nemedians, the, 4 Nennius, 216 on the Picts, 219 Nigg, 325 of the Novantae, 206 Novius, 195 Oban, 325 artificial mound near, 11 Ochiltree, 325 Odin, 51 Ogam Script, the, 58 Orcades, 203 Oronsay, 272 Otalini, 206 Oykell, 260 Pabba, 272 Paisley, 325 Panbride, 326 Panmure (see Panbride) Papill, 272 Partholen, 3 Partick, 327 Peanfahel, 237 Peebles, 327 Peffer, 260 Pentland, 327 Perth, 327 Phcenicians and the sun-god, 12 Picars, 133 Pictish Chronicle, the, 216 Pictish Power, decline of, 370 Pictones of Poitou, 156 Picts, the, 131 houses, 136 Irish, various names of, 141 Golden Age of, 143 Ulster, 147 historical, 149 origin of name, 150 tattooers, 152 historical notices of, 152 and the Anglo-Saxon Chronicle, 211 Bede on the, 211, 219 two great divisions of, 213 Claudian on the, 216 Gildas on the, 218 Nennius on the, 219 Geoffrey of Monmouth, on the, 219 — — Layamon on the, 220 theories about the, 228 succession system of the, 231 and ScotSj relations between the, 347 and the Romans, 350 Shamanism of the, 353 — — at Loch Ness, 353 and Angles, 355 Kenneth MacAlpin as King of the, 377 called Galwegians, 379 Pigmies' Isle, 34 Place-names, Irish, 125-8, 158-174 Scottish, 188-208, 266-337 INDEX. 395 PluscarJen, 328 Pollokshaws, 328 Polmai»e,328 Portree, 328 Prestwick, 328 Provincial and town names i if Scotland, 293-337 Ptolemy's place-names, r2-">-8, 193-208 Quiraing, 328 Raasay, 272 Rannoch, 329 Red Branch Knights, 1i'3 Renfrew, 329 Rerigonius, 197, 208 Rhicina, 203 River-names, Ptolemy's, 125 and their value, 192 Scottish, 256-273 Romans and the Picts, 350 Rona, 272 Rosemarkie, 329 Roslin, 329 Rosneath, 329 Ross, 170, 291 Rothesay, 329 Rothiemurchus, 329 Roxburgh, 329 Rum, 272 Rury the Great, 141 Ruthven, 330 St. Columba and his mission to the Picts, 351 St. Kilda, 269 St. Patrick, 46 Sanquhar, 330 Saxon and the Gael, 179 Saxons in Scotland, 324 Scandinavian champions, 117 incursions, 372 Scollofth, 236 Scone, 330 Coronation Stone at, 56 Scot, the word, 68 meaning of, 99 Scota, daughter of Pharaoh, 66 Scotland in legend, 182 earliest name of, 182 Invasion of, by Agricola, 181 colonisation from Ireland of, 339 the Commendation of, 360 strata of population of, 365 Scots and Scythia, 68, 211 Scythians, the, 68 Selgova^, 206 Selkirk, 331 Senchus Mor, the, 119 Serpent- worship, 12 Shaman, the, described, 40 Shamanism , 39 of the Picts, 353 Shanachies, the, 2 Shandon, 331 Shannon, the, 127 Shetland, 272 Shiant Isles, 272 Shield-painting, l.V) Shin, 261 Shira, 261 Siabhras, the Irish, 48 Sian, 59 Sidhe, the, 62 Sketis, 203 Skraelings, the, 34 Skull, the ancient Irisli, 41, 42 Skulls, ancient, 21, 75 Slamannan, 298 Sleat, 331 Smertffi, 207 Spain and the .Milesiiins, 83 Spean, 261 Spittal, 331 Staff a, 273 Stirling, 332 Stone of Destiny. .55, 56, 57 Stornoway, 332 Strachan, 333 Stranraer, 333 Strathclyde, Britons of, 366 Sunart, 317 Sun-worship, 9 Sutherland, 334 Tacitus and Ireland, 96 Tacitus' place-names, 168-191. Ta?xali, 206 Tain, 334 Talorg, 242 Tamia, 208 Tarbat or Tarbert, 334 Tarland, 335 Tarvedum, 198 Taus or Tavaus, 188 Teamrah or Tara, 103 Teith, 261 Teutonic and Celtic folk-lore, coin- cidences between, 52 Laws, 113 Teviot, 261 Thurso, 261 Tighernach, 66 Tillimorgan, 335 Tilt, 261 Tina, 195 Tiree, 273 Tobermory, 335 Tongue, 336 Town names of Scotland, 293-337 Traditions, Irish, 1 Traquair, 336 Trimontium, 208 Trolls, the, 49, 61 Troon, 336 Trossachs, 336 Tuatha de Danaan (.sec Dananns) meaning of name. 30 396 THE RACES OF IRELAND AND SCOTLAND. Tutssis, 196, 268 Tummel, 261 Uist, 273 Ulidia, Wall of, 148 UUie, 261 Ulster Picts, 147 Uluid or Ulta, meaning of. IIU Ulva, 273 Urquhart, 336 Uxellum, 208 Vacomagi, 207 Vandogara, 197, 208 Varar, 196 Venicones, 20() Vecturiones or Verturiones, the, 150 Verubium, 199 Vidua, 127 Vinderius, the, 125 Volsas, 197 Wales, Scottish colony in, 340 Walloons, the, 79 Wemyss, 337 Whithorn, 337 Wigtown, 337 AVinds, selling, 38 Witchcraft among the Scandinaviiins, 60 Ythan, 262 DATE DUE - 1-' T HAR 3 12003 CAYLOBD PRINTED INU.S.A. BOSTON COLLEGE 3 9031 01213847 5 497815 DA 927 MACKENZIE. Bapst Library Boston College Chestnut Hill, Mass. 02167