EXCURSUS ON THE ETHNOLOGY OF CELTIC SCOTLAND.

by the editor Alexander MacBain, M.A., L.L.D.

 

            The ethnology of the British Isles is still, despite the intelligent researches of the last fifty years, in an unsettled state. This, is greatly due to the fact that the subject draws its materials from various subordinate or kindred sciences, and no one man has yet appeared who has been able to grasp with equal power the reins of all these sciences. The archaeologist deals with the monuments and other physical remains of man's past, helped by the anatomist in deciding upon "skins and skulls," a subject also dealt with by the anthropologist, whose sphere of science is man—his race, physique, and beliefs. The historian depends on his written or printed documents; while the latest to lend his aid, as a real, not an empirical, scientist, is the philologist. Much was done in former times in using language to decide racial points; but it is since Grimm and Zeuss some sixty years ago put philology on scientific lines that any good has accrued from this subject. It is still a science known thoroughly, especially for purposes of ethnology, only by a few.

            Without going back to the cave-men, and others of paleolithic times, when Britain and its isles formed a continuous part of Europe, we come to neolithic times, when unmistakably we have man of the New Stone Age. These neolithic men were comparatively small of stature, long-headed, and dark-haired. They buried in long barrows. The Bronze Age begins with the intrusion of a race tall in stature, broad-headed, and fair-haired, with beetling brows—a splendid race physically and mentally. They buried in round barrows. Some—indeed, most—ethnologists regard these men as the first wave of the Celts; some say of Gadelic, or perhaps Gadelic and Pictish. They are allied by physique to several past and present races on the Continent—the modern Walloons, for instance, and the old Helvetii. The view maintained by the Editor is that the Gadels or ancient Gaels and the Picts both belonged to the great Aryan Race, and originally possessed the tall stature, blond hair, and long heads which are postulated for the pure Aryan. The Aryan Race, or rather the Aryan-speaking Race, is a discovery of modern or scientific philology. It was discovered some sixty years ago that the languages of the various nations—barring a very few—dwelling from Ireland to Ceylon, spoke language that ultimately came from one original tongue. In short, the chief Indian languages, Persian, Slavonic, Lettic, Teutonic, Greek, Latin, and Celtic, are descended from one mother-tongue. For a long time it has been a matter of dispute where this original language had its habitat. It is now agreed that southern Russia and ancient Poland formed the home of the Aryan tongue. The dispersion of the Aryan-speaking people began some four thousand years ago. The Celts lay on the upper reaches of the Danube until the dawn of history begins; the Latins and they were nearest of kin of any of the other leading branches. The Celts spread over Germany to the shores of the North Sea, and then, about 600 B.C., or indeed earlier, they entered Gaul and pushed on their conquests into Spain, and later into northern Italy. They were at the height of their power in the fourth century, spreading from the west of Ireland to the mouth of the Danube, and in 279 they overran Asia Minor, settling down to the limits of Galatia about 250 B.C. Such an "empire" might satisfy Rome itself. But it had no centre, and soon crumbled, after two hundred years' domination.

            The Celts all unite on one philologic peculiarity: every Aryan initial p has been lost. In the course of their dispersion over Europe they divided into two dialects over the Aryan sound qv (as in Lat. quody Eng. quantity). The one dialect made it k or q purely, the other made it p ; and we speak of P and Q Celts for brevity's sake. The Belgic Gauls, the Britons and Welsh, and the Picts, were P Celts; the Gadels or Gaels of all ages were Q Celts. Most of Gaul spoke the P variety of Celtic. The Celts, of course, pushed westward into Britain. It is usually thought that the Gadels came first. The common notion naturally is that they swarmed into England about 600 B.C., and were thence driven westward into Ireland by the advancing Belgic tribes. Undoubtedly Gadels were in Wales and Devonshire in the fifth century A.D., settled as inhabitants. These, however, are accounted for as the invaders of the Roman Province of Britain during the invasions of the Scots and Picts from 360 to 500. Indeed, in 366, and for a few years, the Province of Britain was ruled, or misruled, by Crimthann, High-King of Ireland. Theodosius arrived in 369, and drove out the invaders. As early as 200, settlements were made by expelled Gaels in South Wales. Besides this, Gaelic inscriptions of the fifth and sixth centuries in Oram are found in South Wales, and one or two in old Cornavia. Professor Rhys is the great protagonist for the view that the Gadelic tongue was continuous in Wales from the time of the first Gadels till the seventh century. On the other side, Professor Kuno Meyer asserts that "no Gael ever set his foot on British soil save from a vessel that had put out from Ireland," a dictum with which the present writer agrees.

            The tradition among the Gaels of Ireland themselves is that they came from Spain to Ireland. It is more likely that, starting from Gaul, they skimmed along the southern shore of England—perhaps the Picts were then in possession of the country—and thus arrived in Ireland. Their own traditions and there being no other trace of them in Britain before the Christian era prove this contention. As already said, the date of their arrival must be about 600 or 500 B.C.

            About the same time the Picts came across, possibly from what was afterwards the land of the Saxon invaders of England, and may have colonised Scotland first, bringing there the red-haired, large-limbed Caledonians of Tacitus. In any case, the Picts must have been the predominant race in Britain in the fourth century B.C., when the Greek voyager, Pytheas, made his rounds of the northern seas. He calls the people of Britain Pretanoi or Prettanoi; this might be a Celtic Qretani, present Gaelic Cruithne, possibly from cruth, figure, so called because they tattooed themselves, whence Lat. Picti, painted men. The fact that Pictavia was also the name of a large Gaulish province makes this last statement doubtful. It may, however, be inferred that this Greek form Prettania gave rise to the name Britain—a bad Latin pronunciation of Prettania. Prof. Rhys here objects, and points out that Pliny mentions a tribe of Britanni as situated at the mouth of the Somme, not very far from Kent; that there was such a tribe is proved by the modern town-name of Bretagne. If Prof. Rhys is right, he must postulate that part of Kent was inhabited by these Britanni, and that from this little colony came the name of the whole island. No Britanni are mentioned as in Britain, and it is likely that the tribe on the Somme were some returned emigrants from Britain. The Welsh call the Picts Prydyn (from pryd, figure), which again agrees with Gaelic derivation (Gaelic cruth, whence Cruithne, is, in Welsh, pryd). Britain is Welsh Prydain, the same word as that for Pict. Hence the Picts are the "figured" men both in the Gadelic and Brittonic languages. These are the Editor's views, and the proof must be deferred till we come to treat the Pictish question.

            We are on firm historic ground in regard to the last Belgic invasion of Celts from the Continent. The Belgic Gauls crossed over into Britain before Caesar's time, for he found them in possession of at least the eastern portion of England; the language was the same on both sides of the Channel, some tribe names, such as the Atrebates, were common to both, and King Divitiacus ruled both in Gaul and Britain., Caesar speaks of the Britons of the interior as aboriginal, no doubt referring to the west coast and to Scotland. In any case, the Belgae seem at the time of the Roman conquest to have possessed Britain as far as the Forth—at least its eastern half, being probably in much the same position as we find the Anglo-Saxons about 613. The Picts had been conquered or driven west and north; we know they inhabited all northern Scotland then, and possibly what was afterwards the Kingdom of Strathclyde. Tacitus mentions the Silures in South Wales as a dark curly-haired people, and argues their Spanish origin. These Silures are now recognised as the survivors of the Iberians of the Neolithic age. In Scotland, therefore, at the beginning of the Christian era, the racial position would be thus: Belgic Gauls in the eastern portion of the country from the Firth of Forth' to the Tweed; parallel to them in the western half, from the Firth of Clyde to the Solway, were the Picts, still retreating. The rest of the Picts filled the remaining portion of Scotland from the Firths to Cape Wrath and the Orkney Isles. The previous Iberian population, with its admixture of Bronze-age men, were absorbed by the Celts or driven westwards, where, among the Isles and on the West Coast, plenty traces of them are still in evidence. The Roman occupation of the district between the Walls, that is from the Tyne and Solway to the Clyde and Forth Wall, no doubt added a new ethnologic factor to the population there; and the Brittonic or Belgic Gauls undoubtedly came to possess Strathclyde and Dumbarton (the "dune" of the Britons). In the sixth century the Anglo-Saxons entered Scotland. The Celts called them Saxons because that tribe formed the first Teutonic raiders and invaders of Britain, the Gadelic tribes receiving the name from the Brittonic peoples. It was, however, the Angles that conquered the eastern halt of Scotland to the Firth of Forth.

            Meanwhile the Scots, who had helped the Picts to harass the Roman province for a hundred years, had acquired settlements on the Argyleshire coast and in the Isles. The Scots were simply the inhabitants of Ireland ; it was their own name for themselves. Isidore of Saville (600 A.D.) says the name in the Scottic language meant " tattooed," and, as a matter of fact, the root word is still alive in the language—Gaelic sgath, lop off; old Irish scothaim, allied to English scathe. This makes both Gadels and Picts mean "men of the tattoo." Dr. Whitley Stokes prefers the root skot, property; German schatz stock; and translates the word as "owners, masters." The first invasion of Scotland by the Scots is set down by the Irish annalists as in the latter half of the second century (circ. 160 a.d.) under Cairbre Riata, whom Bede calls Reuda (Gadelic Reiddavos "Ready-man?") Riata gave his name to the Irish and Scotch Dal-Riadas both—" the Tribal portion of Riata." Possibly additions took place during the Picts and Scots alliance of 360 to (say) 460, but in any case a great accession to the Scots on the West Coast was the arrival, in 501, of the sons of Ere from Dalriada; they founded the little kingdom of Dalriada, practically Argyleshire and its Isles, though the original Argyle extended from the Mull of Kintyre to Lochbroom, as our earliest documents show. It means "Coastland of the Gael"—Airer-Gaidheal. When the Norse came about 800, they called the Minch Scotland Fjord, which shows that the Gael practically held the West Coast entire, and the Picts held the East Coast to Pettland Fjord, or Pictland Fjord, now Pentland. The name Scot and Scotland came to be applied to the Scottish kingdom in the tenth century by English writers—the Anglo-Saxon Chronicle calls Constantine, who fought unsuccessfully at Brunanburg, in 938, King of Scotland. The Irish, who were called by this time Hibernienses, or Hiberni, by outsiders, dropped the name Scot and called themselves Goedel, or, later, Gaoidheal, "Gael." This is the name that the Highlanders still call themselves by—Gaidheal. Unfortunately, the oldest Irish form dates only from 1100—Goedel, which would give a Gadelic form, Gaidelos but Scottish Gaelic points to Gadilos or Gaidelos, and from various considerations seems the corrector form, giving a root gad, Eng. Good Gothic gadiliggs, relative; German gatie husband. The idea is " kinsman," as in the case of the native name for Welshman—Cymro, whence Cymric, Com-brox, a " co-burger," where brox or broges (plural) is from the root mrog land ; Lat. margo Eng. Inark march.

            The next invasion of Scotland, which gave her a most important accession of population in the Isles, the West Coast, and in Sutherland and Caithness, was made by the Norse about 795. Our historians seem little to understand either its extent in time and place or the great change it wrought in the ethnological character of the districts held by the Norse. Of this we shall speak at its proper place in notes on Chapter V. The Norman invasion extended even to Scotland, and Celtic earls and barons, either through failure of heirs male or otherwise, soon and in great numbers were succeeded by Normans and Angles.

            It will thus be seen that the Scottish people are ethnologicaly very mixed. The Caledonians, as Dr. Beddoe points out, still show German, or rather Walloon, characteristics. Norse features are predominant in Lewis and the northern Isles generally, though Iberian and other (such as Spanish) elements are strong. The East Coast is largely Teutonic. The old burghs were planted by the Canmore dynasty in the northern districts to keep the ordinary population in order, and towns like Inverness were from the first in the hands of Flemish and other Teutonic traders.

The Pictish Problem.

            Till criticism began with Father Innes's Essay on the Ancient Inhabitants of Scotland in 1729, the Scottish historians taught that the Picts and Scots were two separate nations living side by side, each speaking a language of its own. These historians gave their attention nearly altogether to the story and genealogy of the Scots, representing Kenneth Mac Alpin in 843 as overthrowing and even extirpating the Picts, insomuch that their language and their name were lost. Father Innes's Essay, among other things, holds that though Kenneth Mac Alpin, the Scot, had to fight for his Pictish throne, yet he was rightful heir, but he proves that there was no extirpation of the Picts. Their language, as a dialect of Celtic, like British (Welsh) and Gaelic, naturally gave way to the Court and Church language of Kenneth and his dynasty, which was Gaelic—such is his easy-going method of getting rid of a national language. Later on Pinkerton, who had an anti-Celtic craze, put the Picts in the foreground of his historic picture of Scotland before 843; he regarded them as Gothic or Teutonic—ancestors of the Lowland Scots, who wiped out the Dalriadic Kingdom about 740. The king of the straggling remnant of Dalriads, one hundred years later, became, in the person of Kenneth Mac Alpin, also King of Picts. George Chalmers (1807), sanest critic of them all, regarded the Picts as Cymric or British by race and language, and of course accepted the usual story of the Scottish Chronicles. Mr. Skene, in the first edition of the present work, in 1837, adopted Pinkerton's revolutionary ideas about the Picts and the Scottish Conquest, but with the great difference that he regarded the Picts as Gaelic-speaking, using the same language as the Scots. In fact, he held that there was no change of race or language at the so-called Scottish Conquest, which was no conquest at all, but a mere matter of succession on Kenneth's part according to Pictish law. This may be called the " Uniformitarian " theory of early Scottish history: nobody conquered anybody, and the great Pictish nation was, as before, in language and race, the main body of the Scottish Kingdom, and most certainly ancestors of the present-day Scottish Highlanders—at any rate the Northern Picts were so. The Southern Picts he allows in 1837 to be conquered by Kenneth Mac Alpin, but in Celtic Scotland he only admits that Britons were between the Tay and the Forth—the Britons of Fortrenn being mentioned in the Irish Chronicles— and gave Kings to the Picts, as the Kings' lists compelled him to admit; but these Britons were Cornish (Damnonii of Cornwall and Dumnonii of mid-Scotland, according to Ptolemy's geography, were likely the same people in Skene's view). This very plausible theory has for the last sixty years held the field in Scottish history; indeed, the popular historians know no other. The County histories of Messrs. Blackwood, of course, hold by Skene's theories; and the two latest historians of Scotland—Dr. Hume Brown and Mr. Andrew Lang—regard the Picts as purely Gaels, and kill off the Dalriads in the time of the terrible Pictish King, Angus Mac Fergus (about 740). The obscurity of Kenneth Mac Alpin's succession is insisted upon. Mr. Lang, as might be expected, is really "funny" on the subject. Writing about Prof. Zimmer's expression that the Scots "took away the independence of the Picts," he says :— " We might as easily hold that James VI. took away the independence of the English by becoming King, as that Kenneth Mac Alpin, a Pict by female descent [?], did as much for the 'Picts." Dr. Skene has retarded the progress of scientific research into early Scottish history for at least a generation. This sort of thing, as shown by Lang's case, will go on for many a day yet let Celtic scholars do what they like.

            Modern Celtic scholars have reverted to the old position of the Chronicles. Respect for the authority of contemporaries like Bede and Cormac, and, we may add, Adamnan, compels them so to do, not to mention the authority of the Chronicles; philological facts, scientifically dealt with, and considerations of customs, especially in regard to marriage, hold the next place. The present writer thinks that the topography of Pictland is one of the most cogent factors in the solution of the problem, but, unfortunately, Celtic scholars " furth of Scotland " cannot appreciate this aspect of the question except to a limited extent. If Prof. Rhys studied the topography of Pictland instead of the so-called Pictish inscriptions, it is certain that he would not distract either Celtic scholars or outsiders like Mr. Lang with his theories as to the Pictish being a non-Aryan, pre-Celtic tongue. The ingenuity wasted on this theory and on its ethnologic consequences makes the outsider yet distrust philologic ways. And here, again, the study of Scottish ethnology is retarded, though not to the same extent as it is by Dr. Skene's theories.

            We can here only summarise the arguments that go to prove that the Picts were a Celtic-speaking people, whose language differed both from Brittonic and Gadelic, but, at the same time, only differed dialectically from the Gaulish and Brittonic tongues. The language was of the P class. The arguments are these :—

 

            I.—Contemporary writers speak of the Pictish as a separate language from both Brittonic and Gadelic.

            Bede (731) twice refers to the matter:—"The nations and) provinces of Britain, which are divided into four languages, viz., J those of the Britons, the Picts, the Scots, and the English" (III. cap. 6). There may have been thus many provinces in Britain, but only four languages. In his first chapter he adds Latin as a fifth language—Britain " contains five nations, the English, Britons, Scots, Picts, and Latins, each in its own peculiar dialect cultivating the sublime study of divine truth." These statements, surely, are definite enough: Pictish is a language different from either Brittonic or Gadelic. This Skene acknowledges in the present volume, but confines it to the southern Picts; in Celtic Scotland he does like the Scottish theologian—he looks the difficulty boldly in the face and passes on!

            Adamnan (died 704), writing for people who knew that Pictish was a very different tongue from Irish, did not require to mention that interpreters were needed any more than modern travel-books do, but he does incidentally mention that Columba preached the Word twice through an interpreter, once to a peasant, and once to a chief. "On two occasions only," says Skene, does he require an interpreter, and it is at once inferred that King Brude and his court spoke to Columba without interpreters—and in Gaelic !

            Cormac, King-bishop of Cashel (circ. 900), records a word of the beria cruithnecJi or Pictish language {cartit, pin).

            The next contemporary references occur in the twelfth century, and they concern the so-called Picts of Galloway. These will best be considered under the next heading.

 

II.—The so-called Picts of Galloway and the Irish Cruithnig.

            The Picts of Galloway are mentioned as being present at the Battle of the Standard (1138) by Richard of Hexham, a contemporary writer, who informs us that King David's army was composed inter alios of "Pictis, qui vulgo Gallweienses dicuntur." The learned cleric calls them Picts; their usual name was Gallwegians. From Reginald of Durham, writing at the end of the twelfth century, we get a word belonging to these Picts, for, speaking of certain clerics of Kirkcudbright, he calls them "clerici illi qui Pictorum lingua Scollofthes cognominantur." Unfortunately, the word Scollofthes proves nothing, for like the Welsh ysgolhaig and old Irish scoloe, scholar, student—latterly, in Gaelic, servant—it is derived from Latin scholasticus; but the reference to the Pictish language implies its existence in Galloway at the time. Of course we can pit against these two references, another from the same Anglic source. Henry of Huntingdon, who writes before 1154, says: "The Picts seem now destroyed and their language altogether wiped out, so that what old writers say about them appears now fabulous." We have further an enumeration of the inhabitants of the Glasgow diocese in the charters of Malcolm and William the Lyon, which are addressed thus: "Francis et Anglis, Scotis et Galwejensibus et Walensi-bus"—Franks (Norman French), English (of the south eastern counties), Scots (Gaels possibly), Galwegians and Welsh (remains of the old Britons of Strathclyde). Here there is no mention of Picts.

            Galloway is so named from Gall-Gaidheil or "Foreign Gaels." This was the name given to the mixed Norse and Gaels who inhabited the Isles of Scotland, Man, Galloway, Kintyre, and the Western coast of Scotland. Dr. Stokes thinks that the Gaelic portion of them had relapsed into paganism. The Gall-Gaidheil afterwards formed the Kingdom of Man and the Isles, without, however, any portion of the mainland being included; and the name Gall-Gaidheil became latterly restricted to Galloway. The early history of Galloway can only be guessed at. The Brittonic people certainly had possession of it, and Dr. Beddoe regards the tall hillmen of Galloway and upper Strathclyde as the best representatives of the Brittonic race, Wales itself being very much mixed in blood. It formed part of the Kingdom of Strathclyde, no doubt; but it must have received a Gaelic population from Ireland before its conquest by the Norse. Its place-names show traces of Brittonic, Norse, and Gaelic names ; but Gaelic names are predominant. Gaelic was spoken in' Galloway and Ayr till the seventeenth century ; but the Gaels of Ayr, Lanark, and Renfrew were invaders from the north, who in the tenth and eleventh centuries imposed their language and rule on the British Kingdom of Strathclyde. It is clear, from the above considerations, that the Galwegians of the twelfth century were anything but Picts, and that their language was the same as the Manx. Richard of Hexham and Reginald of Durham, finding the Galwegians a race apart, called them Picts; and so Dr. Skene founds one of his strongest arguments that Pictish was Gaelic on the fact that the Gaelic-speaking Galwegians were Picts according to two bungling English ecclesiastics of the twelfth century.

            The Irish Picts have always the name of Cruithnig, both in Gaelic and in Latin, whereas the Picts of Scotland are variously called Cruithnig, Picts, Piccardai, Pictones, and Pictores. In Ireland there were Picts in Dalaraidhe (Down and part of Antrim), in Meath and in Roscommon. The last two were doubtless some mercenaries introduced by some King or Kinglet returning victoriously from exile. Nothing is known of them save in a wild legend about the arrival of the Picts first in Ireland and their departure to Scotland, leaving a remnant in Meath. But the Cruithnig of Dal-araidhe figure prominently in Irish history in the sixth and seventh centuries. The Irish histories relate that they were the attendants or descendants of the Princess Loucetna, daughter of Eochaidh Echbel, King of Alba: she married Conall Cernach, the great Ulster hero of the early part of the first century of the Christian era. But the Ulster Picts were evidently invaders from Scotland who settled on the corner of Ireland nearest to their own land. By the sixth century they were as Gaelic-speaking as the rest of the Irish. And hence Skene finds another proof that Pictish was Gaelic. He also misreads the history of Ulster, which he regards as having been all populated by the Picts. Ulster had in early Irish history two consecutive denotations : Ulster at first meant the province of Ulster as it is now. But the old kingly heroes of Ulster—the Clann Rudraid, descended of Ir, son of Miled— was gradually extruded from its lands by scions of the royal line of Ireland, until in the fifth century they had only Dal-araidhe or Ulidia or Uladh, which was still called Ulster and its kings still styled "Kings of Ulster." They were, of course, also King of the Picts of Dal-araidhe. Hence has arisen Skene's confusion, in which he is followed by Prof. Rhys.

 

 

III.—The Pictish Language.

            Not a line of either poetry or prose has been recorded in Pictish ; the so-called Pictish inscriptions are yet unravelled. Only two words are recorded by writers as Pictish. Bede records that the east end of the Roman wall, between Forth and Clyde, ended "in loco qui sermone Pictorum Pean-fahel, lingua autem Anglorum Penneltun, appellatur." Here pean is for penn, which is also the old Welsh for "head," old Gaelic, cenn ; and fahel is allied to Gaelic fal, Welsh gwawl, rampart. Both Skene and Rhys regard pean as British, belonging to the "Britons of Fortrenn," or if not so, borrowed from the British. Cormac records the word cartit, a pin or brooch pin, to which Stokes compares the old Welsh garthon goad.

            We have, however, ample means to judge the affinities of the Pictish language in the numerous personal and place-names recorded by classical and later writers, or still extant in old Pictavia.

 

(l) Names in the classical writers.

 

            Tacitus first mentions Caledonia, by which he means Scotland north of the Firths, and Ptolemy writes it Kaledonios. The long e between l and d is guaranteed by the old Welsh Celydon, and Nennius's Celidon; but all the same, it must be regarded as a Roman mispronunciation of Caldon—ld being not common in Latin as a combination, for early Gaelic shows Callden, now Caillinn, Scotch Keld, in Dun-Keld ; and there are three other names near at hand there with the same ending, notably, Schiehallion. The root cald in Celtic means "wood," and Caldonii would mean " woodlanders."

            Tacitus also records the Boresti in Fife; he gives the personal name Calgacus, "sworded one" (Gaelic calg, colg, Welsh caly), The much misread Mons Graupius (now Grampian), yields the root grup a non-Gadelic root in p, which argues its Picto-Brittonic character. Stokes compares it to Greek grupos, rounded (Ger. Krumm, bent). The Orcades, or Orkney Isles, give the Celtic root ore, pig, possibly here meaning "whale."

            Ptolemy (circ. 140 A.D.) in his geography, gives some 44 names connected with Pictland. Ptolemy's tribal names begin in south Pictland with the Damnonii, who stretched across the neck of Scotland from Ayr to Fife. It is usual to regard the word as a variant of the Cornish Dumnoni, now Devon (Gaelic .domhan, world, and dumno) ; both Skene and Rhys allow them to be Britons—those Britons of Fortrenn who were responsible for the Brittonic elements in the Pictish language according to the theories held by these writers. The Epidi of Kintyre are distinctly of the P Celtic branch ; the root ep or eq means horse (stem eqo, Gaulish epo}. The Carnonacai (G. earn), the Caireni ("sheep men"), the Cornavii (compare Cornwall), the Lugi (lug, win), Smertai and Vaco-magi (nagh, plain), are all good Celtic names); and to these may be added the Decantai, found also in Wales, and the Vernicones (G. fearna, alder?). The Taixali of Aberdeen, and the Cerones or Creones, are as yet unexplained as to name. The coast names come next. The Clota or Clyde is from the Celtic clu, clean; Lemannonios, now Lennox, like lake Lemann, comes from lemano, elm. The river Longos, Norse Skipafjord, or Loch Long, comes from long, -ship; Tarvedum {tarbh, bull); Cailis river (caol, narrow); Deva river means "goddess," and is a common Celtic name, more Gaulish-Brittonic than Gadelic; Tava, the Tay, has Brittonic equivalents (W. Tawe, Devon Tavy? Welsh taw, quiet). Celtic, too, must be Itys (Gaulish Itins), and Vir-vedrum and Ver-ubium (prefix ver); nor would it be difficult to explain from Celtic roots Volas or Volsas, Nabaros (nav, float ?), Ila, now Ullie (il, go), Varar, Tvesis (Spesis ? now Spey); and Loxa. Tina and Boderia or Bodotria (Forth), are doubtful. The town names are less satisfactory. Alauna, really the river Allan, a good Celtic river name (W. Alun, Cornish Alan, root pal) Lindum, G. linne, loch, water, possibly Linlithgow; Victoria, a translated name, in West Fife ; Devana, "goddess," Gaulish Divona, "tons additus divis," gets its name from the river as usual, viz., the Don, old Gaelic Deon, now Dian, being in spite of its inland bearings, really Aberdeen ; Orrea, Bannatia, and Tamia are not immediately explicable, though, as far as mere roots are concerned, they can be Celtic. Alata Castra, or Winged Camp, is supposed to be Burghead. It is a translated name. So, too, is High Bank, between the Ullie or Helmsdale, and the Varar or Moray Firth. This has recently been happily equated with the Oykel, whose "High Banks" the Norse usually made the southern boundary of their conquests, and which they called Ekkjals-bakki, or EkkjaFs Bank. The name Oykel goes along with the Oichil Hills and Ochiltr-ee, and is from Celtic uxellos, high, Welsh uchel, Gaelic uasal. The Pictish here shows decidedly Brittonic phonetics. The island names prove nothing: Ebouda, perhaps for Boud-da, now Bute ; Malaios, now Mull (inal, mel, brow, hill) ; Epidium (ech, horse) ; Ricina ; Dumna (compare Dumnoni) ; and Skitis, now possibly Skye (not ski, cut, "indented isle.")

            The historians of Severus's campaign (208-11) record but few names. The Maiatai and Caledoni are the only tribes mentioned seemingly having the north of Scotland between them, the Maiatai being next the northern wall. Adamnan calls them Miathi; the name is still unexplained. Argento-coxos was a Caledonian chief of the time; the name means "Silver-leg." A tablet found some years ago at Colchester gives us the war god's name as Medocius (G. and Irish Miadhach) and the devotee's name was Lossio Veda Nepos Vepogeni Caledo. The date of the inscription is from 232-235. Prof. Rhys has suggested that Lossio (Brittonic gen.. Lossion-os, Gadelic Lossen-as) is related to the Welsh personal name Lleision. Vepogenos, the name of the Caledonian's grandfather or uncle (possibly), is thoroughly of the P variety of Celtic, and it appears in a shorter root form (z) in the Pictish list of Kings (Vip, Vipoig), Gaelic Fiacha, a common name. Veda may be for Veida, and this in a shorter root form appears in the Pictish Kings list as Uuid, i.e.) Vid. Ammianus Marcellinus (circ. 400) gives the two tribes of Pictland as Di-calidonae and Vecturiones. The latter name has been happily corrected by Prof. Rhys into Verturiones, whence the historic name of Fortrenn, the district between the Forth and the Tay. To sum up the results of the above analysis: one-third of the names can easily be paralleled elsewhere on Celtic ground— Gaulish or Brittonic, though not on Gadelic ground; a fourth more show good Celtic roots and formative particles, and another fourth can easily be analysed into Aryan or Celtic radicals. These facts dispose of Prof. Rhys's theory of the non-Aryan and non-Celtic character of the Pictish, and it also makes so far against Skene's Gadelic view—a name like Epidi being especially decisive against a Q language. The names of northern Pictavia show no difference in linguistic character from those of the south, as witness—Deva, Devana, Vacomagi, Caelis, Smertae, Lugt, Cornavi, Caireni, Carnonacae, Tarvedum, Verubium (root ub, point, weapon); and, finally, Orcades.

 

(2) Post-classical Pictish Names.

            Contemporaries like Adamnan and Bede record but few Pictish names, and we depend on the Chronicles of the Picts and Scots for complete King lists, and on the Irish Annals as a check on these lists and as a source of further names, and especially, place-names. The lives of the Saints present some names, but this is a doubtful source. The King list begins with Cruithne, the eponymus of the race, who is contemporary with the sons of Miledh, the Gadelic invaders of Ireland, whose date is only 1700 B.C. according to the Annals. We have 66 names of Kings to cover the period from Cruithne to Brude, son of Mailcon (554-584 A.D.), the King who received Columba in 565. Imagination seems to have failed the Pictish genealogists in making this list, for they fill a long gap with 30 Kings of the same name —Brude, differentiating them by epithets that go in couples, thus : Brude Leo, Brude Ur-leo, Brude Pont, Brude Ur-pont, &c. The ur here is the Gaulish prefix ver, Welsh gur, guor, Irish fer, for, allied to English hyper and over. It is very common as a prefix in all the branches of Celtic. It is useless to take these King names seriously before Brude Mac Mailcon's time, though one figure may be historic—Nectan, son of Erp (a.d. 480), who is said to have given Abernethy to Derlugdach, abbess of Kildare. The name Nectan is common to Pictish and Gaelic; it comes from necht, pure, whose root is nig, wash. The Pictish form and pronunciation is doubtless best recorded by Bede's Naiton, which shows Brittonic phonetics in changing ct into it. Erp, the father's name, was common in Pictland, and we last hear of it among the Norse. Erp, son of Meldun, a Scottish Earl, and grandson of an Irish King, was captured by the Norse, and as a freed man went to colonise Iceland in the end of the ninth century; from him descended the Erplingi clan of Iceland. This is clearly the Pictish equivalent of Welsh Yrp (Triads) and Gadelic Ere, the latter a very common name (ere means cow, heaven). Brude appears in Bede in a more Welsh form as Bridei; Stokes equates it with Eng. proud. Mailcon, the father, may have been the famous Welsh King, whom Gildas calls Maglo-cunus, "High Chief," known later as Maelgwyn of Gwyned. The list from Brude Mac Mailcon to Kenneth Mac Alpin is in the Pictish Chronicles as follows:—

 

Gartnait filius Domelch (584-599). The name Gartnait, or Garnait, was very common in Pictland. It comes from gart, head ; Welsh, garth. It is non-Gadelic. Domelch is in the Irish Annals given as Domnach (from dumno}.

 

Nectan nepos Uerb (599-619), "nephew of Verb." Verb appears in many Gaulish and British names. In Ir. it means "cow," " blotch ;" in 0. W. gverp, stigma.

 

Ciniod f. Lutrin (619-631) ; Ir. Cinaed Mac Luchtren. The first name is our modern Kenneth {cin-aed^ "fire-kin"), common to Irish and Pictish. Lutrin is a Pictish form of Celtic Lugo-trenos, (< strong by the god Lug." Lug either means the " sun-god " or " winner."

 

Gartnait f. Uuid or Wid (631-635). The name Vid is to be compared to 0. W. guid as in Guid-lon, Guid-nerth ; fuller form Veida, already mentioned. Seemingly the root is vid^ know. It also exists in Ir. as a prefix : Fid-gus, Fid-gaile.

 

Bridei f. Uuid or Wid (635-641). Brude son of Vid, brother of above.

 

Talore frater eorum (641-653). The name Talorg and Talorgan is purely Pictish, and is the same as Gaulish Argio-talos, " Silver Brow." It is common ; there was a St. Talorgan. The phonetics of the Pictish Chronicle are here purely British (rg becoming re).

 

Talorcan /. Enfret (653-657). Talorgan, son of Eanfrid, King of Bernicia, who was an exile in Pictland. The name Eanfrid is Saxon.

 

Gartnait / Donnel (657-663). The father's name is Domnall or Donald (Dumno-valos, "World-King"), and it is Irish. He was himself likely a Scot of Dalriada.

 

Drest frater ejus (663-672). Drust is meant. It is a common name and purely Pictish. Its longer form is Drostan, old Cornish inscription Drustagni; more celebrated as Tristan or Tristram of the legends. Stokes makes the root drut, W. drud, brave, strong. Compare Eng. trust and the terminal trud in Teutonic names (Ger-trude, &c.).

 

Bridei f. Bili (672-693). Brude, son of Bill or Beli, King of Strathclyde. The name is British (Ir. Bil, good).

 

Taran f. Entfidich (693-697). Taranis was the Gaulish "thunder" deity. W. taran, thunder. Adamnan has Tarainus, a Pict. The Irish Annals give Enfidaig for the father's name, En-fidach possibly; Fidach, son of Cruithne, and Vid, already discussed, have the same root.

 

Bridei f. Derili (697-706). Brude, son of Derile. The der may be an intensive prefix, as in W. Der-guist, 0. Br. Der-monoc. There are also Dergard and Doirgarto, which came from Der-gart, gart being as in Gartnait.

 

Nectan f. Derili (706-724-729), brother of above.

 

Drust and AlRin co-reigned. The name Alpin is purely British; if native, the root is alb white, as in Alpes, the Alps. It seems allied to the name Alba, the older Albion.

 

Onnust f. Urgust (730-760). Angus, son of Fergus. Both names are common to British and Irish. They mean "Unique Choice " and "Super-choice."

 

Bridei f. Urgust (760-762), Angus's brother.

 

Ciniod f. Wirdech (762-774). Kenneth, son of Feradach. An early mythic king was called Wradech, Ir. Annals, Uuradech, that is, Feradach. The name seems both Ir. and Pictish.

 

Alpin f. Wroid (774-779). Ir. Annals, Feroth and Ferith, compare W. Gueruduc.

 

Drust f. Talorgan (779-783).

 

Talorgan f. Onnust or Angus (783-786).

 

Canaul f. Taria (783-788), mis-reading for Conall, son of Tadg, both names being purely Irish, and he seems to have been a Scot interloper.

 

Constantin f. Urgust (d. 820). Constantine is Latin; Fergus, already discussed.

 

Unnust f. Urgust (820-833}. Angus, son of Fergus, his brother.

 

Drest or Drust f. Constantin and Talorgan f. Utholl co-reigned 3 years.

 

Uwen or Eogan / Unnust or Angus (836-838). Eogan is both British and Gaelic.

 

Wrad f. Bargoit 3 years. [Possibly Dergairt]

 

Bred or Brude, son of Dergard, "Ultimus rex Pictorum " (St Andrews Priory Reg.). For Dergart, see Bridei f. Derili.

The above list, as handed down by the Pictish Chronicles, the age of which is unknown, is decidedly British in phonetics, and the names Brude, Gartnait, Talorgan, Drostan, and Alpin, are foreign to old Gaelic; but, at the same time, they are explicable from British sources. There is nothing non-Celtic in the list. It tells, therefore, both against Skene and Rhys.

 

(3) The so-called Pictish Inscriptions.

            Pictland shares with the south of Ireland, Cornwall, and South Wales the peculiarity of possessing inscriptions in Ogam character. Ogam writing is an Irish invention, coincident probably with the introduction of Christianity into southern Ireland in the fourth century. By the south Irish missionaries this style of inscription was introduced into Cornwall and South Wales; and naturally we must look to the same people as its propagators in Pictland. The south Irish conformed to Rome in Easter and other matters in 633 or thereabout. It is likely that they came to Pictland in the Roman interest some time after, and may have been mainly instrumental in converting King Nectan in 710 to adopt the Roman Calendar. The Irish Annals say that he expelled the Columban monks in 716 over his conversion to Rome.

            We should naturally expect these inscriptions to be either in Irish or Pictish, but Prof. Rhys has jumped to the conclusion that they are purely Pictish, and, as his Pictish is non-Aryan, so is the language of these inscriptions. Unfortunately they are difficult to decipher; the results as yet are a mere conglomeration of letters, mostly h, v, and n. One at Lunasting in the Orkneys is punctuated, and according to Rhys runs thus:— Ttocuhetts : ahehhttmnnn : hccvvevv : nehhtonn.

            In opposition to those who hold that Pictish was a Brittonic tongue, Prof. Rhys cites the above, and declares that if it be Welsh he will confess he has not understood a word of his mother-tongue! It is neither Welsh nor any other language under the moon. Mr. Lang quotes the inscription and says— "This appears to be not only non-Aryan, but non-human ! or not correctly deciphered. Some people seems to have dropped all its aspirates in one place at Lunasting." A word here and there is in a general way recognisable in these decipherments (as above the last word looks like Nechtan), but as yet these inscriptions are not correctly deciphered, and some, like the Golspie stone, are too weathered or worn to be deciphered. (4) Place-names of Pictland.

            Only a resume’ can be given here. The Pictish place-names are very different from names on Gadelic ground—Ireland and Dalriada. There is, of course, a veneer of Gaelic over them, as the Scots really did impose their language as well as their rule on the Picts. Place-names in the Isles and in Sutherland and Caithness must be left out of account, since they are largely Norse. From the southern borders of Ross to the P^orth east of Drumalban the names have all a marked family resemblance, partly Gaelic, partly Pictish. The prefixes aber and pet, unknown to Gadelic, are found from Sutherland to the Forth. The former means " confluence," and had two forms, aper and oper, as in Welsh {ad, od, and ber, Lat. fero); the Gaelic for aber is inver, and it has in the most common names superseded the Pictish aber. Pet means "farm," G. baile, which, in fact, has superseded it in purely Gaelic districts for a reason which the dictionary should make clear. The prefix both—farm, dwelling, common to Irish and Welsh as an ordinary noun, is widely used in Pictland to denote a bally. Pres, a bush, W. prys, a covert, is a borrowed Pictish word, and occasionally appears in place-names, as does perth, brake, in Perth, Partick (old Perthoc, Strathclyde British), and Pearcock or Perthoc (King Edward). British pen we do not find now; every one such has become kin, as in Kin-cardine, a very common name, for Pen-cardin, W. cardden, brake. Equally common is Urquhart for older Ur-charden, Adamnan's Airchartdan, "At (the) Wood." A prepositional prefix peculiar to Pictish names is for, fother, corrupted into fetter (Fetter-cairn) and foder (Foder-lettir). It is corrupted also into far (Far-letter = Foder-letter). Possibly it is an adjective terminally in Dunottar (Dun Foither of Chronicles ?), Kin-eddar (King Edward), &c. It seems to mean " lower," "under ": vo-ter, a comparative from vo, Gaelic fo, under. The extensive use of certain prefix names in Pictland is observable as compared to Ireland, where their use is rare: strath, ben, monadh (rare in Ireland), allt ("stream" in Pictland), corrie, blair, and cairn. Lan, so common in Wales, is rare, though known, in Pictland; the cill of the lona monks gave lan no chance. Ochil Hills and Oykel river have already been discussed. Space does not allow the discussion of individual place-names; nor can the influence of Pictish on Gaelic phonetics and vocabulary be touched. Such a word as preas bush, already alluded to, is easily detected as a Pictish borrow, because initial p is non-Gaelic, and its root qre, or qer, is allied to G. crann, W. prenn.

 

IV.—Pictish Manners and Customs.

            For the manners and customs of early Scotland, Skene goes to Ireland, and transfers the whole social system to Pictavia; so, as the latest example, does Mr. Andrew Lang. But surely the Book of Deer ought to have warned them all that this is utterly wrong. The public life outlined there resembles the Irish, but it is not the same. We have the king (rl). mormaer or great steward (translated earl or jarl), and toisech or clan chief: also the clan. The word mormaer means "lord"; but it must be a Gaelic translation of the Pictish word, for the Gaelic itself is hybrid {mor, great; maer, officer; from Lat. major). We have only three grades of nobility here, nor is there any trace else of more. The tenure of land is the usual Celtic one, but the only word of definite import we get is daback or davoch, four ploughlands, a term peculiar to Pictland, though extended slightly in feudal times to the West Coast and Isles. We see, therefore, that the older Pictish system underlies the Gaelic kingdom of Scotland.

            Another serious point, whose significance was lost by Skene, and found only too well by Prof. Rhys, is the Pictish rule of succession, or the marriage system. The succession to the throne (Bede) and to property (Irish writers) lay in the females; that is to say, a man succeeded to the throne because his mother was the previous king's daughter or sister. The king's brother was his heir, and failing him, his sister's son. It was the female side that was royal. A glance at the king list given above shows this: no son succeeds a father, but a brother often succeeds a brother. The fathers, too, were often outsiders:

            Talorgan, son of Enfrid, Prince of Bernicia, and called cousin of Egfrid (686); Brude, son of Bill, King of Strathclyde; Gartnait, son of Domhnall, Donald being likely a Scotic prince. This system, where maternity alone is regarded as certain, holds a low view of marriage, and is at present found only among uncivilised races. Caesar knew of the existence in Britain of promiscuous marriage; Dion tells us that the wife of Argento-coxos, a Caledonian, acknowledged promiscuity among the high-born; and Bede explains the system of his day—that the Picts got their wives from the Scots on condition of the succession to the throne being through the females.

            Here we have a custom palpably belonging to a non-Aryan race, not to speak of a non-Celtic race. It must therefore be due to the customs of the previous inhabitants still surviving among the Celts; the vanquished here took captive their victors. Whether the Pictish language was also influenced by the previous one it is hard to say; but the influence could not be much, because Celtic civilisation was much higher than the native one, and borrowing would-be unnecessary.

            To sum up the argument we cannot do better than quote Prof. Mackinnon's criticism on Dr. Skene's position: —"The question cannot, however, be settled on such narrow lines as these [Pictish if non-Gaelic would have left remains, and an interpreter was only wanted twice.] The questions of blood and language must always be kept distinct. Anthropology and archaeology may hereafter yield concrete evidence which will be decisive of this matter. As things are, the following facts must be kept in the forefront. Among the Picts, succession was' through the female. This custom is unknown among the Celts; it is, so far as we know, non-Aryan. Again, Bede regarded Pictish as a separate language. The Gael of Ireland, looked upon the Picts or Cruithnig, to use the native term, as a people different from themselves. Cormac, the first Gaelic lexicographer, gives one or two Pictish words, quoting them as foreign words, at a time when presumably Pictish was still a living language. The Norsemen called the Pentland Firth Pettland, i.e., Pictland Fjord, while the Minch was Skottland Fjord. Mr. Whitley Stokes, after examining all the words in the old records presumably Pictish, says: ‘The foregoing list of names and words contains much that is still obscure; but on the whole it shows that Pictish, so far as regards its vocabulary, is an Indo-European and especially Celtic speech. Its phonetics, so far as we can ascertain them, resemble those of Welsh rather than of Irish.’" Celtic scholars of the first rank who have pronounced on the matter are all agreed that Pictish was not Gaelic, as Skene held.