But, it may be objected, is it not possible that the Celts of the Bronze Age had settled far away from the Alps, on the coasts of north-western Europe. Possible, perhaps, but what is the value of mere possibilities? We have seen it stated, and the Cambridge handbook is only a specimen of many publications that accept the view, stated most definitely that the Gaelic branch of the Celts introduced the Bronze Age to Britain and Ireland. Surely something more than a mere possibility, some shade or degree of probability should appear in support of teaching so positive.

Now let us suppose that the dominant Bronze Age population of Britain and Ireland were Celts, as we are instructed to believe. Let us see what would follow from this position. It would follow, beyond question, that the peculiar art and works of the Bronze Age in Britain and Ireland would be mainly connected with the art and works of the Bronze Age in those parts of Europe which were likewise inhabited by Celts, rather than with other parts of the Continent. I cannot find that any such connection has been established or is believed in by archæologists.

The Brythons, we are told, were Belgic invaders who introduced the Iron Age. Not the faintest probability has been brought forward to establish this very precise and positive doctrine. Coffey places the close of the Bronze Age in Ireland and the coming of iron into general use at about 350 B.C. It is admitted that the Celts of central Europe were in possession of iron about four centuries earlier. This affords a most cogent argument that, during the intervening four centuries, there was no such social and industrial continuity between central Europe and these islands as must undoubtedly have been if both regions and the intervening parts of the Continent had been occupied by Celtic populations.

Again, if the Brythons or Belgic Celts, armed with iron, were able to cross the channel and displace the western Celts in Britain, it would surely have been much easier for them to cross the Marne and the Seine and displace the western Celts in Gaul. The theory seems to presuppose that an invasion was necessary to bring the Iron Age into Britain, but the same theory would have it that the Iron Age found its way into Ireland without any invasion, for it leaves the Bronze Age Goidels of Ireland to learn the use of iron in some more pleasant way than by meeting iron-headed spears in the hands of Belgic conquerors.

It is certain that after the withdrawal of the Roman legions from Britain and after the Anglo-Saxon invasions, there were Gaelic populations in various parts of western Britain, in Argyllshire, North and South Wales, and the Cornish peninsula. Rhys supposed these to be the remnants of the Gaelic population which, in his view, had occupied all England during the Bronze Age. There is sufficient evidence to show that they were fresh settlements made by the Irish of Ireland during and after the collapse of the Roman power in Britain.

The "P and Q" element in the theory is equally unsound. It is certain that, where the Irish Celts retained the consonant Q in their language, the British Celts replaced it by P. But no such distinction has been shown to have existed between the language of the western Celts and the language of the Belgic Celts on the Continent. Such phonetic changes as the substitution of P for Q spread in an almost mysterious way through languages. Their spread may be arrested by a geographical barrier so considerable as the Irish Sea, but it was not at all likely to have been brought to a stand by the waters of the Seine and Marne. Nor can a phonetic change of the kind be taken as necessarily corresponding to any racial or political boundaries. In all the western dialects of Latin which grew into the Romance languages, the initial W of Germanic words was changed into GW, and this identical change also took place in the Welsh language, but not in Irish. It took place in Spanish, yet that does not appear to prove that the Welsh are more near akin to the Spaniards than they are to the Irish, nor, if history happened to be silent, would it prove that Britain after the Roman occupation was peopled by a Spanish invasion which did not extend to Ireland.

There is one serious argument which has been adduced in support of the view that Britain was in Celtic occupation during the Bronze Age. The existence of the word kassiteros, meaning "tin," is traced in the Greek language as far back as about 900 B.C. There seems very good reason for thinking that kassiteros was a Celtic word adopted into Greek. From this it is argued that the metal itself came from the Celts to the Greeks, which seems reasonable enough. It is further argued that the Celts must accordingly have been in possession of the country which produced the metal, and that this country was Britain. The conclusion is that the Celts were in occupation of Britain earlier than 900 B.C. It seems to me, however, that the fact, granting it to be a fact, that the metal tin reached the Greeks bearing a Celtic name is by no means proof that it came from a country inhabited at the time by Celts. If you visit the Zoological Gardens in the Phœnix Park, you will be invited, before you reach the entrance, to purchase for the delectation of the monkeys a certain vegetable product, the name of which, upon inquiry, you will learn to be "pea-nuts." No one will be rash enough to deny that "pea-nuts" is an English word. I have not the least idea where pea-nuts grow, but I am quite certain that the fact of their being named "pea-nuts" is no proof that they grow in England or in any English-speaking country. It is very good proof, however, if proof were needed, that the trade in pea-nuts has passed through the hands of English-speaking people. If kassiteros is a Celtic word, as I think it very probably is, it proves no more than that, when the Greeks learned this Celtic name for tin, the trade in tin passed towards them through the hands of a Celtic-speaking people. If it was British tin, which again is not improbable, I suggest that it came to Greece by an overland route through the Celtic region in Mid-Europe, probably along the Rhine and the Danube or to the head of the Adriatic. As a matter of fact, the Greek writer Poseidonios states that in his time British tin reached the Mediterranean by an overland route. "It is brought," he says, "on horses through the interior of the Celtic country to the people of Massilia and to the city called Narbon."

There is, then, no evidence from archæology, history, or language, sufficient to establish even a moderate degree of probability for the theory of a Celtic occupation of Ireland or Britain during the Bronze Age.

On the other hand, taking Coffey's approximate date of 350 B.C. as the beginning of the period of the general use of iron in Ireland, we shall, I think, find sufficient evidence to warrant the belief that the Celts reached Britain and Ireland about that time, and not earlier, at all events not considerably earlier than that time.

Why not earlier? I think we have conclusive grounds for believing that the Celtic migrations to Ireland cannot have begun very much, if at all, sooner than the fourth century B.C. Before stating these grounds, let us ask is there any discoverable reason for supposing that the Gaels inhabited Ireland from a time many centuries farther back. I think it possible that those who in modern times have entertained this view have been influenced by the dates assigned to the Gaelic immigration by Irish writers like the Four Masters and Keating. These dates may be taken to correspond closely enough with the estimates of archæological authorities for the commencement of the insular Bronze Age, and, in the absence of evidence to the contrary, it might be imagined that they were founded on some basis of tradition.

It is not the habit of popular tradition to encumber itself with chronology. There is no known instance of ancient reckoning in years and periods of years that is not based on some era, on the accepted date of some real or supposed event or events. Nowhere in Irish tradition has any trace been found of the existence of a native system of chronology before the introduction of Christian learning. In a paper published in the Proceedings of the Royal Irish Academy (July, 1910), I have shown how the extant written chronology of the Irish Invasions was first originated. The method was not unlike Sir John Rhys's series of equations.

The Irish historian found in Latin histories a set of definite epochs by which antiquity was divided: the beginning of the Assyrian empire, the beginning of the Median empire, the beginning of the Persian empire, the usurpation of the Magi in Persia, and the beginning of Alexander's empire. The chronology of the Irish Invasions was settled by the easy process of making each invasion coincide exactly in time with each of these epochs. It is evident that no traditional value can be attached to a chronological system of this kind.

But, it may be objected, the very remoteness of the time assigned to the Gaelic invasion by Irish historians may reflect the popular belief in its remoteness. If that be so, then the earlier the historian is the more near he is to the popular tradition. In the paper just cited, I have shown that, in the earliest known version of this chronology of the Invasions, the Gaelic migration to Ireland coincides with the date of Alexander's empire, 331 B.C. That is not very far from the date assigned by Coffey for the end of the Bronze Age in Ireland, about 350 B.C. For my own part, I attach no traditional value to this coincidence, but if it pleases anyone to insist that Irish prehistoric chronology has a traditional value, then it must be conceded that tradition, as far as it is valid, is altogether favourable to the view that the Gaelic occupation of Ireland belongs to the end, and not to the beginning, of the Bronze Age.

The migratory movements of the Celts on the Continent have a bearing which cannot be ignored on the time of the Celtic migrations to Britain and Ireland. So far as I am aware, no modern investigator has suggested that the Celts were not already in the Iron Age at the time of their expansion into Italy and Spain. Why then should it be imagined, in the absence of any positive indication to the purpose, that they occupied these islands more than a thousand years earlier?

If I am not mistaken, the archæological evidence is fairly decisive on the point. Archæologists are agreed in dividing the Celtic Iron Age into two main periods, the Early Celtic or Hallstatt period, and the Late Celtic or La Tène period, also called the Marnian period. Each of these periods is taken to consist roundly of about four centuries, and the two periods on the Continent together correspond roughly to the last eight or nine centuries before the Christian Era. The Late Celtic period is abundantly represented in the antiquities of Great Britain and Ireland, but the objects that have been found in either country belonging to the Early Celtic Period are extremely rare. On this head Coffey writes as follows ("Bronze Age in Ireland," page 5):

"It must be remembered that the Continental Hallstatt period is not at present well represented in Great Britain and Ireland, and though, under Hallstatt influence, certain Continental Iron-Age types such as bronze caldrons, trumpets, round shields, etc., found their way into Ireland, we cannot as yet definitely separate this period from the end of the Bronze Age."

In fact, "sporadic finds" are all that represent the Early Celtic period in Ireland, in Britain, and even in the neighbouring regions of the Continent. It will not be questioned that during the Hallstatt period there was quite sufficient intercourse of trade between the islands and the Continent to explain these sporadic finds as importations.

The main fact is that, so far as archæological research has ascertained, the Early Celtic period of the Iron Age is substantially absent from Ireland and Britain, whereas the Late Celtic period is abundantly represented. The Bronze Age in Ireland comes down to about 350 B.C., and its Continental affinities are not specially or notably Celtic. The Bronze Age is succeeded in both Britain and Ireland by the Late Celtic period of the Iron Age. The inference, to my mind, is obvious, that the Celts did not reach either Britain or Ireland until the Late Celtic period, i.e., until the fourth or fifth century B.C. This conclusion agrees well with all that is known of the migratory movements of the Celts on the Continent.

Let us now revert to the Belgic migrations and consider their bearing on the matter of the Celtic colonisation of Ireland. The Belgae, we have seen, were a Celto-Germanic group which, according to Cæsar and Tacitus, occupied the lands stretching from the Rhine to the Seine and Marne, and expelled from that region the Celtae proper. There is no indication in what Cæsar says that in his time this movement was one of remote antiquity. In fact, it is perfectly clear that it was a movement by no means exhausted but still in active progress when he took command of the Roman armies in Gaul. The attempted migration of the Helvetii in the first year of his command, B.C. 58, was a part of this movement. A little later, Cæsar had to repel similar attempts of the Usipetes and the Tencteri to cross the middle Rhine and settle in Gaul; and these, according to Dio, were two Celtic peoples. Still later, in the time of Augustus, the Ubii migrated from the eastern to the western side of the Rhine. From all this it is clear that the Belgic migration was a continuous movement and that its force was far from being spent at the time of the Roman conquest of the country west of the Rhine. Cæsar indicates that there were powerful Belgic settlements west of the Rhine during the great wandering movement of the Cimbri and the Teutones, i.e., about half a century before he began his Gallic campaigns. There is nothing, however, to show that these settlements were of earlier date than the second century B.C., and I have seen no reason for thinking that they could have been much earlier.

We now come to the question of the Belgic invasion of Britain and its probable date. In Rhys's theory, which is still accepted in England, the Belgic invaders were the first to establish the Iron Age in Britain. I claim to have shown good grounds for believing that there was no Celtic occupation of Britain before the Iron Age. I have already suggested that, if this Celto-Germanic movement was brought to a standstill on the banks of the Marne, it was not likely to have succeeded in over-running all England at the commencement of the Iron Age in England. It will be seen that the Celto-Germanic migrations extended not merely to Britain but also to Ireland, and I suggest that if these Celto-German Belgae had been the first people to come over armed with iron, they would have made an easy conquest of Ireland as well as of England.

Let us look at the actual evidence of the Belgic conquest of England. The sole historical witness on the point is Julius Cæsar, and this is his testimony:

"The interior of Britain is inhabited by those who say that, according to tradition they are natives of the island; the maritime part by those who had crossed over from Belgium [meaning Belgic Gaul] for the sake of plunder, nearly all of whom are called by the same names of states as the states from which they originated and came thither, and having made war they settled permanently there and began to till the land."

From this it is clear that Cæsar was informed of two populations in Britain, one which was more ancient and claimed to be native, another which resulted from comparatively recent invasion. The older population he assigned to the interior, the more recent to the seaboard. What did Cæsar mean by the seaboard, the maritime part? Sir John Rhys has no difficulty in supposing that Cæsar did not mean the whole seaboard of Britain or if he did mean it that he was not fully informed, for according to Rhys's theory, the older population, which he supposed to be Gaelic, continued to inhabit the western seaboard of England and Wales. I also agree that, whatever Cæsar may have understood, his statement about the maritime part must be taken in a restricted sense, for no one believes that the Celtic occupation in Cæsar's time extended to the seaboard of the northern parts of the island. I agree also with the view that the traditional natives of whom Cæsar speaks probably included the earlier Celtic colonists, whose settlements dated, according to my argument, from the fourth century B.C., about three centuries before Cæsar's time. The more recent maritime settlements, in that case, would have been very recent in his time, and I think that his statement leads us to that conclusion. These later settlers on the seaboard, he tells us, are known collectively by the same names as the states on the Continent from which they originated. Now this is a statement about a fact likely to be within Cæsar's personal knowledge. He was certainly well acquainted with the names of the states of Belgic Gaul, and there is no reason why he should have said that populations retaining the same names existed in his time on the British coast if he did not know it to be a fact. His testimony on this point, touching a matter within the scope of his personal observation, is of higher evidential value than any other part of the statement quoted. Cæsar does not himself name these states, but in the two following centuries the names of the various states of Britain are given by Ptolemy and other writers, and when we compare these names with those of the states of Belgic Gaul, we find that they coincide only in three instances. These are the Parisii on the foreland north of the Humber, the Atrebatii in the district of Berkshire, and the Belgae, eastward from these to the Bristol Channel. There are some eighteen other states enumerated in Britain, so that the coincidence of names amounts to only one in seven, a proportion which by no means corresponds to Cæsar's words, fere omnes, "nearly all." Except for the Parisii, who occupied the promontory north of the Humber, the states bearing names also found in Belgic Gaul are located in southern England, south of the Thames and the Bristol Channel. One of these, and the most extensive, bears the general name Belgae, which certainly does not suggest that the remainder of the population was also Belgic. Now the fere omnes, "nearly all," in Cæsar's statement cannot refer to such a small minority of the states of Britain. Therefore, either Cæsar was grossly in error, in which case there is not much to be built on his whole statement, or, if he stated the truth, which is much more likely, then there were Belgic settlements on the British seaboard in his time which had lost their identity and passed into insignificance a century later. This I take to be true, for it will be seen that there were also Belgic settlements on the Irish coast after Cæsar's time and that as states they had disappeared a few centuries later. It is indeed quite possible that the Belgae so named, in southern England, consisted of a collection of colonies from various states of Belgic Gaul, whose names were preserved in Cæsar's time, but not one of which was sufficiently populous or otherwise considerable to be worth naming by later writers. There may have been similar Belgic colonies on other parts of the southern and eastern seaboard of Britain, none of them considerable enough to be reckoned as a state. At all events, I submit that Cæsar's statement, far from justifying the assumption of a Belgic conquest on a grand scale, comprising the greater part of Celtic Britain, is rather contrary to that assumption; also, that it cannot reasonably be taken to refer to settlements made in Britain at the close of the Bronze Age three or four centuries before Cæsar's time.

I have referred to the existence at one time of Celto-Germanic settlements on the coast of Ireland. The authority on the point is Ptolemy the geographer, who flourished about A.D. 150. In the south-eastern angle of Ireland, the region of Wexford, he places a population named Brigantes. There was a very extensive state of this name in the north of Roman Britain. Its territory extended across the country from the North Sea to the Irish Sea. Whether the Brigantes were or were not Belgic colonists in Britain and Ireland, I find no means to determine. North of the Brigantes, on the Leinster coast, Ptolemy locates the Manapii. It can hardly be doubted that these were a Belgic people, a branch of the Menapii,2 whose territory on the Continent lay in parts of the countries now called Belgium and Holland. North of the Manapii on the Leinster coast, Ptolemy places the Cauci. The topography of Ireland from the time of Saint Patrick onward is very copious and minute, but no trace has been discovered in it of these three peoples in the location ascribed to them by Ptolemy. It seems to me possible that the Manapii may be represented in later times by a scattered people called the Monaigh or Manaigh. Some of these dwelt in eastern Ulster, near Belfast. Another branch of them dwelt in the west of Ulster, and their name is preserved in that of the county Fermanagh. It is interesting to note that the Irish genealogists derive the origin of both from Leinster. The only trace known to me in Irish tradition of a people similarly named on the south-eastern seaboard is found in the name of Forgall Monach, the father of Emer who was wife of Cú Chulainn. Who were the Cauci? Their name, in the Germanic form Chauci, was that of a people of the German seaboard bordering on the North Sea, who are described in Smith's Ancient Geography as "skilful navigators and much addicted to piracy." Tacitus praises them for their love of justice and says that, though ready for war, they do not provoke war. It must be remembered, however, that Tacitus was an extreme "pro-German." Elsewhere, he tells of incursions made by them against neighbouring peoples. We find, then, two peoples, the Menapii and the Chauci, on the Belgic and German shores of the North Sea, and also on the Leinster shores of the Irish Sea; and this shows that in Ireland as well as in Britain there were Celto-Germanic settlements about the beginning of the Christian era.

Cæsar is the earliest known writer to give the name Brittania to the island of Britain and the name Brittani to its people. In earlier writings the name of the island is Albion. In Cæsar's term Brittani, there seems to be a confusion of two existing names, one Brittani, the name of a small local population, the other Pretani which is recognised to be a British and probably Gaulish equivalent of the Irish name for the Picts, Cruithin, more anciently Qreteni. Cæsar fixed the name Brittani in Latin usage, but the form Pretanoi continued after his time to be used by Greek writers. Polybius and Ptolemy apply the adjective Pretanic to the two islands, and a still later geographical tract in Greek says, "the Pretanic islands are two in number one called Albion and the other Ierne." The Pretanic islands means the Pictish islands, and this name for them must have been taken from the Gauls. It points to a time before the Celtic occupation, when the Pretani or Picts were still regarded as the principal people of both islands. Here we have another indication of the relatively late period of the Celtic occupation. Cæsar learned that the natives of Britain had some curious marital customs which he did not observe among the Gauls, including the Belgae, on the Continent. A later writer, Solinus, in whose time the customs of the Britons were more intimately known to the Romans, ascribes a similar custom, not to the Britons but to the inhabitants of the Hebrides. Both accounts are based on a well-established fact, recorded also in Irish writings, the custom of matriarchy which was peculiar to the Picts. Cæsar's statement is readily explained, if we understand that the Gauls, from whom his information was likely to have been derived, still spoke of Britain and Ireland as the Pictish islands, and regarded this social custom, which was foreign to them, as a Pictish custom. In the time of Solinus, the Romans knew that the Picts were limited to the northern parts of Britain, and the story is accordingly told of the people of the Hebrides. If a custom peculiar to the Picts was spoken of in Cæsar's time as common to the inhabitants of Britain, and if Britain and Ireland were then still regarded in Gaul as Pictish islands, I suggest that this was because the Celts of Gaul did not look upon the two islands as having been mainly occupied from any remote period by a people akin to themselves.

The conclusions which I wish to draw in this lecture are: that neither Britain nor Ireland was colonised by the Celts until the Late Celtic period, corresponding to the period which followed the Bronze Age in these countries; that the Belgic or Celto-Germanic settlements were of still later date, and extended to Ireland as well as Britain; that the Belgic settlements in England were not so widespread as they are represented in modern British writers; and that the distinction between the ancient Gaels and Britons does not correspond to the distinction between the Celtae and Belgae of Gaul in Cæsar's time.

III. THE PRE-CELTIC INHABITANTS OF IRELAND

In the second lecture, I remarked how the name Iberians has been adopted to fill a vacuum as regards the naming of the population which occupied Great Britain and Ireland before the Celtic immigration. This kind of naming is unscientific and misleading. It implies that the ancient population thus artificially named can be identified as a branch of the population which actually bore that name in Greek and Latin literature. From this implied identification other equally unwarranted assumptions are likely to follow. Rhys expended a vast amount of study, ingenuity, and argument in the effort to show that very definite traces of a language akin to modern Basque survived in ancient Ireland and Scotland. On this point it may be remarked that we do not even know that the Basque population was originally Iberian. Ethnologists are agreed that, apart altogether from the Celtic migrations, there must have been a mixture of very distinct races in south-western Europe in prehistoric times. If there was a mixture of races, there was also no doubt more than one language, and if the Basque language has been able to survive the conquests of Celt and Roman and Goth, and last until our own time it may also well have survived the extinction of other languages in south western Europe.

So far as the Iberian theory is not mere vacuum-filling, it appears to rest on a single passage of Tacitus. He is describing the Silures, a British people whose territory was in the south of Wales, and who offered a very fierce resistance to the Romans. "The swarthy complexion of the Silures," he says, "the prevalence of curly hair among them, and their position over against Spain, argue that the ancient Iberians must have crossed over [from Spain] and occupied their territory." We have often heard the occurrence of similar physical traits in the west of Ireland ascribed to a more recent Spanish mixture. It all amounts to this, which Irish tradition bears out, and which nobody questions, that these western isles contain descendants of an ancient dark-complexioned population, probably already of mixed race, which existed in western Europe before the arrival of the fair-complexioned people, whose distinctive features appear by all indications to have originated in the lands forming the basin of the Baltic Sea.

If I am right in suggesting that the Greeks adopted from the Gauls the name Pretanic Islands, as a joint name for Britain and Ireland, it follows that the Gauls themselves supposed the chief population of both islands, before the Celtic occupation, to have been the Pretani, i.e., the Picts. During the early historical period, the Picts are chiefly known as the people of the northern mainland of Scotland, north of the Grampian mountains. The Venerable Bede speaks of their language as still existing in his time, the early part of the eighth century, and as being distinct from the Irish and British languages.

We have abundant and clear evidence that the Picts were at one time widely spread throughout Ireland. Early Irish writings recognise the existence, in their own time, of sections of the population known to be Pictish. The Picts were especially numerous in Ulster. They are described as a subject population, spread over the whole of ancient Oriel, which at that time comprised the counties of Armagh, Monaghan, Tyrone and the greater part of Derry and Fermanagh. There was also a large Pictish element in Connacht, and there were smaller groups, traditionally known to be Pictish, in Munster, Meath, and various parts of Leinster. In Ulster, the ruling or dominant population of a large belt of territory, extending from Carlingford Loch to the mouth of the Bann, is named in the Annals both by the Latin name Picti, and its Irish equivalent Cruithni or Cruithin, which is the Irish form corresponding to Pretani. They continue to be so named until the eighth century, when apparently their Pictish identity ceased to find favour among themselves. It may be observed, however, that, while some proper names which contain non-Gaelic elements survived in ancient Ireland, no trace has been discovered of any language other than Gaelic continuing to be spoken in any part of Ireland within the traditional memory of the people. From this it will appear that the Gaelic language had become universal throughout Ireland some centuries before Irish history and traditions began to be written. The earliest writing of Irish history still extant belongs to the closing years of the sixth century.

In the case of the Picts, we find an interesting example of the method that recommended itself to the learned folk of ancient Ireland when they desired to fill the vacuum. In the Irish "Nennius," the Picts are said to have come of the stock of the Geloni, a people of Scythia mentioned by Herodotus. The explanation of this curious piece of history is found in a passage of Virgil, in which he speaks of the picti Geloni, i.e., the painted Geloni. They were supposed to dye their skin with some colouring stuff. In one of the versions of the wanderings of the Gaels before they reached Ireland, instead of sailing the Mediterranean they marched from Scythia across Europe. On their way they fraternised with a people called the Agathyrsi, who dwelt in Thrace. They made a compact with these people, with the result that later on a body of the Agathyrsi, having taken the name of Picts, followed in the track of the Gaels and came to Ireland. On their way they passed through a part of Gaul, where some of them remained, and were afterwards known as Pictavi. From these is named Poitou in France. Virgil is at the back of this story also. In a verse of the Æneid, he speaks of the picti Agathyrsi, "the painted Agathyrsi."

From these instances, we can see how closely Virgil was read in the ancient Irish schools. We can also see from what materials our ancient scholars could weave their legends of antiquity. And later on we shall see how similar materials and a similar process enabled the Latin scholars of ancient Ireland to construct their accounts—for they have more than one account—of the origin and early wanderings of the Gaelic people.

Another considerable element of the ancient population was the Iverni, as they were called by Ptolemy in the second century. Ptolemy locates them in the middle of southern Ireland. The Irish form of their name in the time of our most ancient writings was Érainn, more familiar in later usage in the accusative form Érna. They have been sometimes called Erneans in English. In the older heroic literature, the Iverni or Érainn are the chief people of Munster. In an important early tract, which gives the names and distribution of the principal subject communities throughout Ireland, the Sen-Érainn are placed in the district of Luachair, i.e., in the north of Kerry and the adjoining parts of the counties of Limerick and Cork. The peoples enumerated in this tract are regarded as being not of Gaelic origin. Sen-Érainn means the old or original Iverni, and the term is used to distinguish them from others also called Érainn, who were of free status and are attached by the genealogists to the Gaelic stock. My opinion is that the dominant element in every part of Ireland during the historical period, including the dynastic families and higher nobility, was Celtic. Otherwise, if we suppose that large communities of pre-Celtic inhabitants continued to exist under rulers and nobles of their own stock down to medieval times, the universality of the Gaelic language as far back as tradition reaches would be hard to account for. I suppose that, when a Celtic dynasty and nobility became established over a non-Celtic commonalty, the old name of the community became attached to them all. So we find that Giraldus calls the nobles who invaded Ireland in his time Angli, giving them the name of the subject people over whom they had ruled in England, though they had been barely a century in England and some of them not nearly so long. I think the same is probably true of the free and dominant Picts in the north-east, i.e., that they consisted of a common population of Pictish stock ruled by kings and nobles of Celtic origin.

Not only in Munster but also in Connacht, Meath and Ulster, our ancient genealogists recognise the existence of Ivernian communities. Rhys put forward the view that the Iverni were only a southern division of the Picts, but this view cannot well be reconciled with Irish tradition, which seems always to distinguish between Picts and Iverni, and recognises Picts in southern Ireland and Iverni in northern Ireland. For example, in county Antrim, Dál Riada, the north-eastern portion, was Ivernian, and the rest of the county for the most part was Pictish. We are on safer ground in regarding the Picts and the Iverni as two fairly distinct peoples.

From the Iverni the whole island took the names by which it was known to the ancient Irish, the Britons, the Greeks, the Romans, and therefore no doubt to the Celts in the neighbouring parts of the Continent. But we have seen that the original Iverni, in Irish tradition, were a remnant of the pre-Celtic population. Ireland therefore was named by the Celts, as Britain and Ireland were jointly named, from an older population which the invading Celts found in possession. The Romans changed Iverni into Hiberni, through a process known as popular etymology. Hiberni suggested to them the Latin word meaning "wintry." Though Ireland was known to some Latin writers to be by no means a wintry country, but quite the contrary, this verbal resemblance naturally caught the imagination, and one Latin poet actually speaks of "glacialis Ierne," ice-cold Ireland.

The Irish and Welsh names of Ireland are not directly taken from the name of the Iverni, but evidently from an older form which must have been Ivéri. Both the Irish name Éire (formerly Ériu) and the Welsh Iwerddon go back to an older name Iverio, and this older name is actually found in the writings of Saint Patrick in the slightly disguised Latin form Hiberio. The Irish genealogies corroborate this view that the name Iverni is itself a derivative from an older name Iveri. A common feature in genealogical lore is the tracing of a people's descent from an ancestor of the same name. It is found in the Bible, in the genealogies of the Arabs, in the legends of the Greeks, and in our own legends, for example, when the Gaels are said to have taken their name from an ancestor named Gaedheal Glas. In like manner all the pedigrees of the Érainn or Iverni in the Irish genealogies are traced to an ancestor named Iar. Iar is a word of two syllables, and represents an older form Iveros. From this and from the Irish and Welsh names of Ireland, I infer that the people called Iverni were at a still earlier period called Iveri. The change in the name of a people from a simple to a derivative form is of very common occurrence. Thus, instead of Angles, people now say the English, instead of Scots, the Scotch; in Irish, the names for the English and the Welsh have undergone a similar change; and so with numerous other names in many countries and languages.

Rhys derives the old Celtic name of Ireland, Iverio, from a word cognate with the Greek piaira, meaning "fat," and understands Iverio to mean the fat, i.e., the fertile country. This explanation, however, will not hold good if, as I think, the name Iverio means the country of the Iveri, unless we suppose the name Iveri to be Celtic and to mean "the fat people!" But we have seen that, in Irish tradition, the original Iverni were a pre-Celtic people, and we are under no necessity to discover a Celtic origin for their name.

For my part, granted that this people bore the name Iveri, changed afterwards into the adjectival form Iverni, I see no serious difficulty in supposing that this name was a local variant of Iberi, the name by which the people of Spain were known to the ancient Greeks and Romans.

Authorities on Irish archæology are agreed that the Early Stone Age is not exemplified in the most ancient remains of human occupation that have been discovered in Ireland. The explanation for this is supplied by the geologists. Some thousands of years ago, the conditions of perpetual snow and ice that at present prevail in the Arctic regions extended much farther into the temperate zones. The northern parts of Europe were covered with perpetual ice. Ireland lay entirely within this glacial zone. The southern limit of the ice ran through the south of England and eastward across the Continent. The time during which this southward extension of ice lasted is called the Glacial Period. Already before that time, Europe was inhabited by man, and the Early Stone Age or Palæolithic Age is held to have preceded the Glacial Period.

The condition of Ireland during that period was like the present condition of Greenland, under a heavy covering of ice formed by the accumulation of snow. By its own weight the ice kept moving from the mountains into the valleys and plains, and from the higher land level into the surrounding seas. Under its moving action, the solid rock-formation of the mountains was ground down and rounded off and scooped into hollows, and great sheets and ridges of stones, gravel, sand and boulder-clay were accumulated on the slopes and low grounds. It is evident that any traces of human life and habitation that may have existed before this process were not likely to be found after it.

The consequence is that the earliest traceable population of Ireland was Neolithic, i.e., belonged to the Late Stone Age. By the Stone Age is meant that time in which the use of metals was still unknown, and in which the most durable material of implements used by men was stone. Needless to say, they also used wood, bone, and any other material that came to hand. The Late Stone Age is distinguished from the Early Stone Age by the use of polished and finely shaped stone implements.

In England, according to eminent authorities already quoted, the descendants of Palæolithic Man survived and are still the prevalent type. In Ireland, they did not survive, and whatever Palæolithic blood is in our veins to-day is due to immigration. Regarding the Neolithic population of Ireland, whatever is to be said belongs rather to archæology than to history. In Britain, we are told, the Neolithic population consisted of at least three distinct races, one which had remained there from Palæolithic times, and two new races, or rather a mixture of two races, which came in from the Continent. One sees how futile it is to attempt to fix upon such a population a name like Iberian. It is assuming a knowledge which does not belong to us.

The Late Stone Age was followed by the Bronze Age, but between the two came a transitional period now generally recognised, in which copper replaced stone as the most durable material of manufacture. This Copper Period is well exemplified in Ireland. Bronze, the distinctive material of the Bronze Age, was made by adding a small proportion of tin to copper, producing a metal very much superior to pure copper for the manufacture of tools and weapons. So far as I have been able to learn, the presence of tin in quantities that could be worked is unknown in Ireland. There seems to have been no scarcity of bronze, and from this I conclude that during the Bronze Age, Ireland had an import trade in tin, and probably therefore an export trade in copper or some other product. This is the earliest evidence of Irish commerce. Bronze cannot have been the material of ordinary industry, nor, unless the inhabitants were very unwarlike, can bronze have been the material of ordinary weapons of war. It is a very durable material, almost unaffected by the action of the elements during centuries. Numerous as the finds of bronze tools and weapons have been in Ireland, they should have been immeasurably more numerous if tools and weapons of bronze had been in every man's hands throughout the Bronze Age, which, according to Coffey, lasted from about 1800 B.C. to about 350 B.C. In fact, Sir Robert Kane, in his work on "The Industrial Resources of Ireland," in a footnote regarding the once extensive copper mines of the Danes' Island on the Waterford coast, supplies an interesting proof of what otherwise we should reasonably expect to be true, that the ordinary working population of the Bronze Age continued to use the implements of the preceding Stone Age.3 Weapons and tools of bronze must therefore have been in the hands chiefly of a more opulent class than the general population. Gold was also used for ornaments, and Ireland is noted for the abundance of its gold ornaments dating from the Bronze Age. Native Irish gold was worked from very remote times, but it is also certain that in the early Christian period gold was brought to Ireland by Oriental merchants in exchange for other products of the country. Sickles of bronze bear witness to the tillage of the soil for corn during this period. It will be seen that there was a mixture of various peoples in Ireland at the time. From this we might expect that there were various degrees of civilisation, and so the remains of Bronze Age sepulchres indicate. The simpler and ruder forms of these are found all over the country. The highly elaborate sepulchres of the region of the lower Boyne, its tributary the Blackwater, and the lower Liffey, are indicative of a relatively high civilisation in those parts, the ancient territory of Bregia. Along with these we may take into account an old Gaelic tradition. It tells that when the Gaels came to Ireland many of the fertile plains had still to be cleared of forest, but there was one plain, Magh n-Ealta, stretching northward from Dublin, which was called the Ancient Plain and was already clear of forest before they arrived. Its name is interpreted as meaning "the plain of the flocks of birds," by which we may understand that it was frequented by the various kinds of gregarious birds which we see in our own time hovering around the plough, rooks, jackdaws, starlings and seagulls. It is worth noting that towards the opposite border of the same region of Bregia there is another plain of the same name, still represented in the name of Moynalty village, about four miles north of Kells and on the Moynalty river, which is a tributary of the Meath Blackwater.

I shall here mention an additional indication that the Gaels were not in occupation of Ireland during the Bronze Age. In ancient Gaelic tradition, the great chambered tumuli of the Boyne are taken to be the tombs or the dwellings of an earlier race.

We pass on now to consider some of the evidence supplied by our ancient literature regarding the population which inhabited Ireland before the coming of the Gaels, that is, according to the conclusions I have already drawn, before the Iron Age. The Gaels occupied Ireland as a conquering and dominant people. During the early centuries of their occupation, whatever language or languages had been spoken in Ireland before them completely disappeared as languages, leaving no doubt some traces behind in the names of places, etc., and probably also influencing to some degree the Gaelic language itself. But for a long time there was nothing like a complete fusion of the old and the new population. The older population remained, not as a mere promiscuous swarm of subject folk, but preserving in a large measure its ancient organisation and sub-divisions. This state of things continued during the early centuries of Christianity in Ireland.

Most of the manuscript evidence concerning these ancient communities is still awaiting collection, publication, and study. Some of it is to be found here and there in the old genealogical tracts, which are still unpublished, and some in the annals. There is a good deal of very ancient material on the subject quoted in the introductory part of the great Book of Genealogies by Dubhaltach Mac Fir-Bhisigh. There is one particular tract dealing specially with the names and topography of these ancient subject communities. It exists in a number of MSS., and has been printed by Craigie in the Revue Celtique from a single MS. of the Edinburgh collection. From internal evidence I think that this tract is of not later date than the eighth century. I mention these facts to show how much has still to be done before we can claim a near approach to full and accurate knowledge of the existing evidence.

There are, however, some larger divisions of the ancient population, spread over wide areas and comprising in each instance several of the smaller named groups; and about these larger divisions there is sufficient information to warrant the essaying of some account of them. Chief among these may be reckoned the Picts. The tract just mentioned shows that there were subject communities of the Picts around Cruachain, the seat of the Connacht kings, and all over Mid-Ulster, from Meath to Loch Foyle.

Along the lower part of the Shannon, in the counties of Galway, Tipperary and Limerick, there was an ancient population known as Fir Iboth, or by the adjectival name Ibdaig. These names contain the Irish equivalent of the name by which the western islands of Scotland were known to Greek and Latin writers of the first and second centuries of the Christian era, i.e., Ebudae. The modern name Hebrides originated in a mistaken writing of this name, and it is curious that the most celebrated island of the group got its English name, "Iona," in the same way. Ptolemy makes these islands belong to Ireland not to Britain. Solinus says the inhabitants in his time grow no crops and live on fish and milk. It is possible that an ancient branch of this population preserved their identity by forming, so to speak, a fisherman caste on the banks of the Shannon. There is evidence that something like the Hindu caste system, in so far as it is linked with the occupations of the people, existed among the descendants of the Pre-Celtic population in Ireland. One of these subject communities is known by the variant names Tuath Semon, Semonrige, Semrige, and Semaine. Each of these names contains the Irish word seim, meaning a rivet, and may be translated the Rivet-folk. This people dwelt in the Desi territory of Munster, where those copper-mines are found which were worked in the Bronze Age by miners using tools of stone and wood. Taking the facts together, it seems reasonable to infer that the Semonrige tribe were the descendants of the ancient copper-smiths of the district, and that they obtained their name from the commodity in which they paid their tribute to the dominant Celts, for the name is Celtic. It should be well noted here that these Irish metal-workers are presented to us in early Irish records as descendants of the pre-Gaelic population; whereas, as we have seen, the current theory in British archæology assumes that the occupation of working bronze was distinctive of the Gaels themselves and was introduced by them.

Another copper-producing district is that of Béarra in West Munster, bordering on Berehaven. Here in ancient times dwelt another "rent-paying" community bearing the significant name of Ceardraighe, "the Smith Folk." There was also either a branch of this folk or another community of the same name situate around the ancient seat of the Munster kings, Teamhair Luachra, a suitable locality in which to find constant employment for a caste of workers in bronze.

According to the tract on the Rent-paying Communities, all over the parts of Munster which, in historical time, were regarded as being specifically Ivernian, including large districts in the present counties of Tipperary, Limerick, Cork and Kerry, there was distributed one of these subject communities which bore the name Tuath Cathbarr, i.e., "the people of helmets." Since there is no record and no likelihood that this subject people were a fighting caste, as undoubtedly some of the subject-communities were in other parts of Ireland, we may infer that they got their name from being employed in the manufacture of battle-gear.

I come now to the most celebrated of all the pre-Celtic folks that inhabited Ireland, the Fir Bolg. In including these among the industrial castes of ancient Ireland, I claim the support of the oldest written traditions, which clearly tell that the Fir Bolg, or "Men of Bags," obtained that name from an industrial connection with leathern bags. The story of the origin of the name, as found in the Book of Invasions, Keating's History, etc., is no doubt well-known. They migrated, we are told, from Ireland to Greece (Greece in ancient Irish writings means the Eastern Empire). There, being outlanders, according to the ideas of our forefathers, they did not obtain the local franchises and became a serf people. Their occupation was to carry sand and earth in leathern bags and spread a soil over rocky places, as is still done in parts of Ireland, to make fertile land. From this occupation, they were named Fir Bolg. They afterwards used the hides in which they worked to construct ships in the ancient fashion, and in these ships they escaped back to Ireland and liberty.

Quite a different version of the story is found in the Book of Lecan, a book which contains a great miscellany, awaiting most desirable publication, of excerpts from older writings, especially excerpts of material which does not accord with what one may call the received teachings of later times on matters of Irish legend and tradition. This particular passage contains what is doubtless the oldest extant account of the Fir Bolg. Its language, in my opinion, is of not later date than the eighth century. Like the accepted story, it says that they were a branch of the race of Nemed, but unlike the accepted story, it does not say that they left Ireland in a body and came back to it in a body after many years. On the contrary, it tells us that they continued to inhabit Ireland all the time, but carried on a particular trade with the eastern world. The manner of their trade was this. They put Irish earth into leathern bags and exported it to the east, where they sold it to the Greeks to be spread on the ground around their cities as a protection against venomous reptiles. From this trade they got the name of Bagmen.

Dubhaltach Mac Fir Bhisigh, in the unpublished introduction to his Book of Genealogies, tells us that Fir Bolg was the specific name of a particular section of the pre-Gaelic population, but became extended in common usage so as to be applied to the whole of that population. Of this statement we have abundant corroboration, with details enabling us to locate the abode of various sections of the Bag-folk properly so called. One section, called Bolgraighe, was the principal Rent-paying community of the ancient Tir Conaill, a territory of much smaller extent than the Tir Conaill of later times. Another section inhabited the district of Sliabh Badbgna (Slieve Baune) in the east of County Roscommon, where, I have been told, popular tradition still recognises their descendants. Another section dwelt in the district of Cong in the south of County Mayo, another in Sliabh Eachtgha (Aughty) in the south of County Galway.

The manufacture of bags from hide or leather was no doubt not a highly esteemed occupation, and it was probably out of contempt that the name Fir Bolg was extended to the whole conquered population by the Celtic ascendancy. The subject communities produced not only skilled artisans but men of great piety and learning in early Christian times. Saint Mo-Chuarog, for example, who is called Sapiens, "the Learned," and who introduced a reform into the Irish chronography of his time, was a member of the Rivet-folk, the Seamonraighe of the Déisi. But the general attitude of the Gaels towards the older population was undoubtedly disdainful. The passage quoted by Dubhaltach from "an ancient book" is familiar to many in O'Curry's translation:

"Every one who is black-haired, who is a tattler, guileful, tale-telling, noisy, contemptible; every wretched, mean, strolling, unsteady, harsh and inhospitable 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 Galians, of the Liogairne, and of the Fir Domhnann in Eirinn. But the descendants of the Fir Bolg are the most numerous of all these."

This is fine old ascendancy talk, the sort of language that has served in many ages to justify the oppression of liberty; and there is plenty of evidence that the older population was in some instances subjected to very harsh treatment—in some instances, not in all, nor were the ancient communities always spoken of in such terms of contempt.

Among them, besides industrial groups or castes, there were also others which appear to have followed the profession of arms. Cú Chulainn, according to one tradition preserved by Dubhaltach, belonged to a non-Gaelic tribe called Tuath Tabhairn, and it will be remembered that he is once described as "a small dark man." "Thou little elf!" his charioteer used to call him, to provoke him to do his utmost in the fight. His rival, Fear Diadh, was a noble of the Fir Domhnann from Connacht, and the Fir Domhnann still existed as a subject community in the times to which the tract on the Rent-paying Folks has relation. They are located in a stretch of country comprising the greater part of the counties of Mayo and Sligo. In the eastern Midlands, from the Shannon to the Irish Sea, the same tract places another of these ancient tribes named the Luaighni—a name still preserved in that of the barony of Lune in Meath. These are represented as forming the chief fighting force of the kings of North Leinster in the heroic period. When Conchobhar sets out to exact reparation for the Táin and the invasion of Ulster, he is met by the forces of the Luaighni at Rosnaree on the Boyne, his heroes one after another are worsted in the fight, his army almost routed, and it is only when their king has fallen in single combat that the Luaighni abandon the field. In the curious story of the revolution brought about by the revolt of the Rent-paying tribes against the oppressive rule of the Gaelic nobility, it is the chief of the Luaighni, Cairbre of the Cat's Head, who becomes king of Ireland for twenty years.

Still more remarkable is the tribute of the ancient saga to the valour and discipline of the Galians. In the ninth century the Galians are still described by the poet Mael Muru as one of the outstanding sections of the population who are not Gaels. The tract on the Rent-paying Folks divides them into three tuatha and gives the location of each. They inhabited the northern parts of old Leinster, in the present counties of Wicklow, Kildare, and King's County. The story of the Táin tells how the Galians excelled all the other troops that joined Medb on her march from Cruachain for the invasion of Ulster. "This enterprise," said the warlike queen, "will be a barren one for all of us, except for one force alone, the Galians of Leinster." "Why blamest thou these men?" said her consort. "Blame them we do not," replied Medb. "What good service then have they done that they are praised above the rest?" said Ailill. "There is reason to praise them," said Medb. "They are splendid soldiers. When the rest are beginning to make their pens and pitch their camps, the Galians have already finished setting up their booths and huts. When the rest are still building booths and huts, the Galians have finished preparing their food and drink. While the others are getting ready their food and drink, the Galians have done eating and feasting, and their harps are playing for them. When all the others have finished eating and feasting, by that time the Galians are asleep. And even as their servants and thralls are distinguished above the servants and thralls of the Men of Erin, so shall their heroes and champions be distinguished above the heroes and champions of the men of Erin on this hosting. It is folly then for the rest to go, for the Galians will enjoy the victory." And in fear and jealousy the queen declared that nothing would please her but to fall upon the Galians and destroy them. Her husband expostulated. "Shame on thy speech!" he said, "a woman's counsel, for no better reason than because they pitch their tents and make their pens so promptly and unwearily." And Fergus interposing swore that he and his Ulstermen would stand by the Galians to the death. The Galians, he said, are but one division in eighteen of our army. Even so, we shall take care that they shall be no danger to us. And he took and divided the forces of the Galians among the rest so that not five of them were in one place together.

Of this Galian stock came Fionn and Oisin and Oscar and all their kindred, according to some accounts. They were of the sept Ui Tairsigh, one of the three folks who, says Mael Muru, are not of the Gaedhil. This sept dwelt at Drumcree in the barony of Delvin in Westmeath. Their name and existence as a sept is probably not so ancient as the time of Fionn, but we may suppose that in their own time they claimed descent from the family of Fionn, from Clann Bhaoisgne.

Other possible instances of occupation-castes are found in the names Céchtraighe "plough-folk," Corbraighe and Corbetrighe "chariot-folk" (Carbantorigion, the name of a town of the Selgovae in southern Scotland), Gruthraighe "curd-folk," Lusraighe "herb-folk," Medraighe "weight or balance-folk," Rosraighe "linseed-folk," Rothraighe "wheel-folk," Sciathraighe "shield-folk."

The tinker clans of recent times in Ireland and Scotland may well be survivals of some of these ancient industrial communities.

It is certain that ancient tribes remained in every part of Ireland after their conquest by the Gaels, and retained in some measure during the early Christian period in Ireland their ancient organisation, often under their own ancient lines of chiefs. This is matter of strictly historical record, and if any similar records had existed and were still extant in Britain, we should hear less of the cheap and easy history of successive populations, each of them completely exterminating those that inhabited the land before them. Writers on history would not find themselves flatly contradicting ethnologists on the strength of their own gratuitous assumptions, when ethnologists say that the modern English race is largely composed of descendants of the primitive inhabitants.

On this subject of primitive races, there is one point which, in passing, I desire to bring out. One of the founders of the modern study of ethnology, Quatrefages, has given a good illustration of a sort of scientific method akin to some that we have had already under consideration. A glance at the map showed him that Ireland represented a north-western limit of the likely spread of the human race in remote times. The migratory movements of antiquity were thought to have, generally speaking, a western trend in Europe. Ireland besides was an island, which in the distant past must have been reached through Britain. Conclusion: Ireland was the place in which to look for primitive European types, and in Ireland the surest place to find the primitive types must be the extreme north-western part. Accordingly, M. Quatrefages packed his portmanteau in Paris and labelled it for Belmullet. This kind of scientific quest is usually successful. It succeeds after the manner of the schoolboy who, before entering into the intricacies of a question in algebra, takes the precaution of providing himself with the answer from the end of the book. M. Quatrefages found the Mayo seaboard swarming with a primitive race of men. I do not propose to examine his discoveries in detail. Anyone who is curious about them is referred to the late Dr. Hogan's little book on "The Irish People," which is the source of my information. In a paper contributed by me to the Royal Irish Academy's "Clare Island Survey," on the Place-names and Family-names of Clare Island, I showed that nearly half of the families now living there could be traced to an earlier home in distant parts of Ireland. I pointed out that in remote ages, the parts of the sea that adjoin the land and the parts of the land that adjoin the sea must have afforded the freest highway for movements of population. It must have been so in the glacial period and during its decline, when the scanty population must have lived a life like that of the modern Eskimos who travel long journeys in their canoes and change their habitation at will. It must have been so in the barren period that succeeded the age of ice, when animal and vegetable food was much more abundant on the sea-shore than inland. And it must have been so in the succeeding forest period, when the inland regions became difficult to traverse. In fact, until men became tillers of the ground and road-makers, the sea-edge was their grand highway. Hence it is that the population of the seaboard is always the most mixed and variable. The place to look for the least movement and least variation is inland, especially in deeply wooded, swampy or mountain areas, which offer the least attraction to newcomers and from which an older population is hardest to dislodge. And this, I think, is also the lesson of ethnological research conducted without foregone conclusions. In all western Europe, there is no region that contains a larger proportion of a late-coming population than the Orkneys, Shetlands and Hebrides and distant Iceland, the uttermost extremes of the north-west.

The ancient legends of Ireland tell of certain peoples which are not represented by territorial groups in the historical record. Most conspicuous among these are the Tuatha Dé Danann and the Fomori ("Fomorians"). The late D'Arbois de Jubainville showed very clearly that these two peoples belonged to pagan mythology. His work on the subject can be read in the English translation by Mr. Best, "The Irish Mythological Cycle." I cannot now attempt to go over the ground it covers, even in summary, but shall content myself by adding a few cogent proofs to those which it supplies. About the year 1000 the poet Eochaidh O'Flainn wrote a poem on the Tuatha de Danann. He began by setting himself the question, were these folks human or were they demons. He answers that they were mortal men of Adam's race, and we are even told by what deaths they died. The very fact that the question had to be asked is conclusive as to the popular belief. But the poet was not satisfied with having brushed this popular belief, a survival of paganism, to one side. In his concluding verses he protests "I do not worship them, I worship the one true God." So that as late as the year 1000 people in Ireland still spoke of the Tuatha De Danann as objects of heathen worship.

An older writer, quoted in the Book of Lecan, tells a plainer tale. He does not admit the truth of the ancient mythology, and says that the Tuatha De Danann were a remnant of the fallen angels. They assume, he says, bodies of airy substance so as to become visible to men, the better to tempt them. They come at the call of sorcerers and those who practise malevolent incantations by walking in circles lefthandwise. They used to be worshipped, and it was they who invented the spells sung by smiths and druids and wise-women and pilots and cupbearers. From them druidism came in Ireland.

The poet-historians did not succeed in killing off the Tuatha De Danann. In 1088 the annalist Tigernach died, and in 1084, four years before his death, his chronicle contains an account of a pestilence which visited Ireland at that time. The cause of this pestilence, says the chronicler, was revealed in that year to a certain man, Gilla Lugán, who was in the habit of frequenting a fairy mound at Hallowtide, the old heathen festival of Samhain. There in the year 1084, Oengus appeared to him and told him that the plague was brought to Ireland by legions of evil spirits from the islands of the northern ocean, who spread it over the country with their fiery breath. And Gilla Lugán himself, says the chronicler, afterwards saw one of these demon legions on the rath of Mullaghmast, and in whatsoever direction their fiery breath came on the land, there the plague broke out among the people.

In Agallamh na Seanorach, the rulers of the Tuatha Dé Danann are still alive in St. Patrick's time, and inhabit the hills associated with their memory. One of them has recently come to life once more in Dublin, Finnbheara of Cnoc Meadha. From the hills at Tourmakeady you can see Cnoc Meadha, a low round hill, on the eastern horizon. It was pointed out to me by a man who knew all about it. That is where Finn Bheara lives, he said. He is the king of the Good People. He is not always there. When Finn Bheara is living in Cnoc Meadha, it is a good year for the country. When he goes away, it is a bad year.

A poem in Duanaire Finn tells how Oengus aided the Fiana in their hostilities with king Cormac, and, like the gods in the Homeric poems, remained invisible while he fought on their behalf.

The passage already cited from the Book of Lecan tells how the Tuatha De Danann arrived in Ireland. They came, it says, without ships or boats and first alighted on Sliabh an Iarainn, in the heart of the country.

The mythology of the Irish Celts was not originally shaped in Ireland. They brought it along with them from central Europe, and just as the ancient scriptures of the Hindus bear traces of having been originally composed in a climate very different from that of Hindustan, so I think the Irish mythology shows some traces of its continental origin. The Fomori of Irish tradition were not inhabitants of Ireland. They always appear as invaders. They come from the north, from the unknown places of the northern ocean. The demons who brought the pestilence to Ireland in 1084 were Fomorians. They are always enemies of the people of Ireland. They were enemies to Parthalon's people, and after them to Nemed's people, the Fir Bolg, and after them to the Gaels. They were a malevolent race of immortals. In the popular view, among heathens, a people expected to be defended by the gods of its own worship. If a hostile people had other gods, these were expected to fight on the other side. Hence there was a natural tendency to regard a double set of immortals, one party being foreign and malevolent, the other domestic and benevolent. But the Irish people, before the Norse invasions, knew no human enemies in the northern ocean. Accordingly, I think that the Fomorians originally belonged to the continental geography of Celtic mythology, and that the sea from which they came was not the ocean to the north of Ireland but the Baltic and the North Sea, and that their islands were originally perhaps Britain and Ireland and the islands of the Baltic and the Scandinavian peninsula itself, which was thought to be an island when it first became known to the Greeks. The Fomorians would be perhaps in part identical with, in part associated with, the gods of the peoples dwelling on the shores of those northern seas before the Celtic expansion northward and north-westward.

We have glanced at the process by which one of our poet-historians endeavoured to transform popular tradition into a kind of history more acceptable to his own school. Christian learning brought into Ireland a double stream of history, derived from the Old Testament and from the Greek and Latin historians. The two streams had already been mingled in one by early Christian historians like Eusebius and Orosius. The works of these writers were well-known in early Christian Ireland. The Chronicle of Eusebius, a history of the ancient kingdoms of the world, written in parallel columns, a column to each kingdom, was known through the Latin translation by St. Jerome and its continuation by Prosper of Aquitaine in the fifth century. It became the basis of the writing of Irish history, and was continued in Ireland, with an Irish section added, down to the early years of the seventh century. By adopting this basis and model, the early Christian historians of Ireland brought themselves inevitably face to face with the task of linking and fitting the old Gaelic tradition to this existing framework of Biblical and Greco-Latin history.

We cannot doubt that the Celts, like the Greeks, Persians, Egyptians, Northmen and other ancient peoples, had what is called a cosmogony of their own, an account of the beginning of the world. Cæsar tells us that the Druids expounded the nature of the gods and also of the material universe. This cosmogony could find no place in the new scheme, and it disappeared, leaving perhaps a few traces in the genealogies. In like manner, other parts of the popular tradition and native lore required to be transformed and recast to find a place in the accepted scheme of world history. That is why the Tuatha De Danann became mortals in the teaching of the learned while they remained and still remain immortal in the traditions that come down from heathen times.

The native tradition had its own account of the origin of the Celtic people. That account, as we shall see, was not such as could be adopted into the Christian world-history received from Eusebius and St. Jerome. It was completely rejected by the Irish historians, as completely as modern Irish people reject the substituted account when they say that their ancestors were Celtic.

To provide a theory of the origin of the Gaels more in keeping with the received world-history, a search was made through the Latin historical and geographical writings that were used in the Christian schools of Ireland and suitable discoveries were made. The most serviceable material for the purpose was found in the world-history of Orosius, a Spanish historian who wrote in Latin about the year 400. Quotations from Orosius by name and word for word show that his book was well-known in the Irish schools. It had the advantage of combining a geography of the world with a history of the world.

In those times, the ordinary Latin name for the people of Ireland was Scotti, Scots. It is the name used for them by Orosius, and also by St. Patrick, and it was accepted by all the early Irish writers who wrote in Latin. But this name Scotti does not appear in Latin before the fourth century and gave no direct clue to trace the origin of the Gaels. In the historical and geographical Latin writings to hand, the people's name that most nearly resembled Scotti was Scythi, Scythians. Accordingly, we are told that the Gaelic people were of Scythian origin.