Proinsias MacCana (1970)

SOURCE: Introduction to Celtic Mythology, The Hamlyn Publishing Group Limited, 1970, pp. 11-18.

[In the following excerpt, MacCana explores the combination of unity and diversity found in the various branches of Celtic mythology.]

The Sources

The earliest sources [of Celtic Mythology] are those relating to the Celts of the continent—mainly Gaul—and of Romanised Britain. Unfortunately they have serious shortcomings. Gaulish literature, being purely oral, disappeared with the Gaulish language: we have it on Caesar's authority that the druids of Gaul considered it improper to commit their learning to writing, and on this point he is substantially borne out by the Irish evidence. As a result, since mythology implies narrative of some sort or other, Gaulish mythology, properly speaking, is lost beyond recovery. There remains, of course, a considerable body of residual evidence, but, since by its very nature it is allusive rather than descriptive, or else is reported at second hand, the modern student is frequently in the uncomfortable position of working from the ambiguous to the unknown.

The evidence is of three types: dedicatory inscriptions such as occur throughout the territories occupied by the Romans, plastic representations of Celtic divinities, and observations by classical authors. In the first two categories the great bulk of the material belongs to the Roman period, and consequently it raises difficult problems of interpretation. For example, Gaulish sculpture developed under Greco-Roman influence and it is no easy task to determine precisely to what extent this influence may have affected the motifs of the sculpture as well as its form. As for the classical authors, it is a matter of scholarly opinion how much value should be placed upon their testimony. Most of them derive their information from earlier sources: even Caesar, who had a better opportunity than most to become acquainted with the Gaulish situation, is far from relying on his own experience and observation. And no doubt all of them were influenced to a greater or lesser degree by the forms and concepts of classical religion and mythology. These considerations have led some scholars to reject the classical evidence out of hand, which is probably an excess of scepticism. It should not be forgotten that a number of observations by the same classical authors on matters of Celtic custom and social organisation are corroborated by Irish literature: so closely in fact that certain early Irish tales might almost have been written to illustrate these comments on the continental Celts, and this few scholars would entertain as a serious possibility. The classical evidence therefore merits consideration, but it must be treated with extreme caution.

By way of contrast, the recorded testimony of Irish literature is later by a millenium or more, but, as we have seen, it has a conservative quality which more than outweighs the disparity in date. (The Irish language, despite the later date of its documents, seems in many respects to be more conservative than Gaulish, and the same may well hold true for the mythologies.) The writing down of Irish oral tradition had already commenced by the end of the sixth century, but time and the Viking raiders proved a ruthless combination and only a few manuscript fragments survive from the period before c. 1100. Then comes the first of a number of great manuscript compilations which between them preserve a wealth of varied material relating to the Irish past. These manuscripts are themselves relatively late, but they have been compiled from earlier sources and many of the individual items which they contain may be dated on linguistic grounds centuries earlier than their extant transcription. But, irrespective of their date of composition, it is beyond question that these texts contain a vast amount of pre-Christian matter.

Among the tales which formed an important part of the filidh 's repertoire there are some which concern themselves explicitly with the supernatural world, and for that reason modern scholars sometimes refer to them as the Mythological Cycle. But this is a rather misleading title since in point of fact most early Irish narrative is mythological to a greater or lesser degree. There is much to be said for the native system of classification which groups the individual titles not by cycle but by theme: plunderings, cattle-raids, wooings, battles, voyages, adventures, elopements, etc. But for the sake of brevity the remaining tales may be divided into three broad categories: miscellaneous tales assigned to the reigns of various kings, historic and prehistoric (though this distinction has little relevance to the historicity of their content), the cycle of the Ulaidh or 'Ulstermen' with Conchobhar mac Nessa their king and Cú Chulainn their youthful hero, and finally the cycle of Fionn mac Cumhaill and the roving bands of warriors known as fiana.

The Ulster cycle was the literature of greatest prestige in the early period; it is heroic literature par excellence and it concerns itself with the activities and virtues that typify heroic society everywhere. By contrast, the Fionn cycle (or fianaigheacht, as it is often called) was more popular among the lower orders of society and correspondingly less highly esteemed by the filidh, and it is in fact only from the twelfth century, a watershed in Irish history and culture, that it bulks large in the literary record. Nevertheless, its roots lie deep in the pagan past. The great delight of the fiana, and their principal activity, is hunting, and this fact alone gives the cycle a quite different temper to that of the Ulster tales. It is predominantly a literature of the open air that ranges far and wide throughout the changing landscape of Ireland, and in due course it becomes a convenient vehicle for numerous nature lyrics.

To this varied collection of tales one must add the pseudo-historical material, and in particular the Leabhar Gabhala, 'The Book of Invasions', and the Dinnshenchas, 'The History of Places'. The former is a twelfth-century compilation which purports to describe the several invasions of Ireland from the time of the Deluge (and even before it!). It is weak on history but relatively strong on myth. The Dinnshenchas, which also belongs to the twelfth century in its definitive form, is a massive collection of onomastic lore 'explaining' the names of well-known places throughout Ireland. Marie-Louise Sjoestedt has characterised the two rather neatly: Leabhar Gabhala is the mythological pre-history of the country and the Dinnshenchas its mythological geography.

There is enough evidence to indicate that Wales also inherited a rich mythological tradition, but, unfortunately, it is poorly documented. Like Ireland, Wales has its great manuscript compilations, the earliest of them from about the end of the twelfth century, but they do not preserve such a wealth of material from the early period as do their Irish counterparts. This is especially true of prose literature, and the earliest surviving tales, Culhwch and Olwen and The Four Branches, were probably first written in the eleventh century. The four tales, or 'branches', of the Mabinogi constitute one of the most important sources for British mythology. They abound in mythological themes and motifs and their dramatis personae are the ancient gods of Britain. Nevertheless, they represent the mere debris of a tradition recast in a loose narrative framework by a talented author who was less interested in preserving sources than in producing an effective piece of literature. There is also a considerable volume of mythological matter scattered throughout the remainder of medieval literature, but clearly any semblance of an integrated mythological tradition had passed away long before the extant literature was recorded. What remains is an imbroglio of anecdotes, allusions, motifs and characters which under close scrutiny gradually reveal the outlines of a number of familiar mythological paradigms within a British setting.

The Welsh evidence derives a special interest from its close association with the great continental cycle of Arthurian romance. Welsh together with Breton literary tradition provided the many Celtic elements incorporated in the Arthurian romances of Chrêtien de Troyes and his fellows, and not a little of the enduring fascination of these stories is due to their essentially mythological character. The original Arthur may well have been a historical person, but the King Arthur of medieval romance and his knightly entourage are much larger than life and share many of the mythological traits of the Irish hero Fionn mac Cumhaill and his fiana.

The Diversity of Celtic Mythology

To speak of 'Celtic mythology' is not to imply a close unity, but merely to recognise a tangible relationship based upon common inheritance. What we know of the mythology of the continental Celts hardly suggests a sustained correspondence with that of Ireland and Wales, and this cannot be due entirely to the unequal documentation. Even among the insular Celts the differences are, at first glance, much more evident than the underlying similarities. Nor is this very surprising, for a number of reasons: the several peoples in question do not derive from a single community of continental Celts; over the last two thousand years or more they have evolved somewhat differently in their sd'cial and cultural organisation; in the case of Britain and Gaul, but not of Ireland, they have been conditioned to the physical presence of Rome over a period of centuries; and, finally, it can safely be presumed that all have assimilated much of the religious thought and usage of the pre-Celtic inhabitants of their several areas.

These considerations go far towards explaining the wide discrepancies between the three visible branches of Celtic mythology. But by themselves they are not sufficient to account for the lack of unity and order which is so evident within each separate branch. Instead, it has been argued that this incoherence simply reflects the decentralised structure of Celtic society, in which each tribe functioned as an independent political unit, the inference being that political autonomy was coupled with religious autonomy and that each tribe had its own special gods, which might, or might not, be common to neighbouring tribes. It may be that this is in fact one of the causes of what has been described as 'the local and anarchical character' of Celtic mythology, though its effects may well be exaggerated by a defective documentation dating in all cases from a period of drastic readjustment, when native religious usage was exposed to the influence of systems of greater sophistication and prestige.

The Celts being notoriously rich in paradox, it is perhaps not surprising to find that this local independence, which is such a feature of their political organisation, is in some respects counterbalanced by a highly developed sense of cultural affinity among the learned classes. Nowadays we know that what gave the Celts such unity as they possessed was not common racial origins but a common culture and environment. The classical ethnographers identified them—not infallibly it may be said—by their language, their shared characteristics, and their mode of life, as well as by their geographical location, and one can still sense something of this cultural coherence in the remarkable analogies, both of ideas and their expression, in the traditional literatures of Ireland and Wales. What is even more to the point, the druidic order existed throughout the Celtic world and its organisation appears to have been essentially the same in all areas. The cult of the centre to which its members attached such importance epitomises their professional solidarity and their assiduous fostering of an ideological unity transcending the political divisions within each nation or agglomeration of tribes. This is a persistent trait and nowhere is it evidenced more clearly than in post-Norman Ireland where the filidh conserved an astonishing cultural unity in a world of political strife and instability.

This faculty for combining unity with diversity, centripetal with centrifugal forces, is no less evident in the mythology. Here the externals present a bewildering variety. The nomenclature continually renews itself even when the underlying concepts remain undisturbed. The myths proliferate in endless narrative variants but their themes are constant and, so far as one can judge, in large measure common to the whole Celtic world. For instance, the theme of divine sovereignty which is such a permanent and such a fundamental element of Irish tradition was also familiar in Britain and in Brittany, though most of the literature to which it gave rise there is known only from occasional allusions. It is this underlying homogeneity that justifies us in speaking of one Celtic mythology rather than of several.

Jeffrey Gantz (1981)

SOURCE: Introduction to Early Irish Myths and Sagas, translated by Jeffrey Gantz, Penguin Books, 1981, pp. 1-27.

[In the following excerpt, Gantz offers an overview of Celtic culture, history, society, religion, and literature.]

One day, in winter, Derdriu's foster-father was outside, in the snow, flaying a weaned calf for her. Derdriu saw a raven drinking the blood on the snow, and she said to Lebarcham 'I could love a man with those three colours: hair like a raven, cheeks like blood and body like snow.'

'The Exile of the Sons of Uisliu' (p. 260)

This passage, from one of the finest stories ever written in Ireland, evinces much of what Irish literature is: romantic, idealistic, stylized and yet vividly, even appallingly, concrete. Most of all, it exemplifies the tension between reality and fantasy that characterizes all Celtic art. In Ireland, this art has taken many forms: illumination (the books of Durrow and Kells), metal work (the Ardagh Chalice and the Tara Brooch), sculpture (the stone crosses at Moone and Clonmacnois), architecture (the Rock of Cashel and the various round towers), music (Turlough O'Carolan). But this tension manifests itself particularly in the literature of Ireland, and most particularly in the myths/sagas—no more precise description is possible, at least for the moment—that survive in Irish manuscripts dating back to the twelfth century.

There are many reasons why this should be so. To begin with, these stories originated in the mists of Irish prehistory (some elements must predate the arrival of the Celts in Ireland), and they developed through the course of centuries until reaching their present manuscript state; consequently, they manage to be both archaic and contemporary. Their setting is both historical Ireland (itself an elusive entity) and the mythic otherworld of the Side (Ireland's 'faery people', who live in burial mounds called 'side' and exhibit magical powers), and it is not always easy to tell one from the other. Many of the characters are partially euhemerized gods—that is, they are gods in the process of becoming ordinary mortals—so that, again, it is not easy to tell divine from human.

At bottom, this tension between reality and fantasy is not accidental to the circumstances of literary transmission and formation but rather an innate characteristic, a gift of the Celts. The world of the Irish story is graphic: blood spurts not only from the calf flayed for Derdriu but also from the lips of Anlúan as his head is thrown across a table (in 'The Tale of Macc Da Thó's Pig'); the 'hero' of 'Bricriu's Feast' is tossed from the balcony of his house on to a garbage heap; the warriors of Ulaid (the Irish name for Ulster) are all but roasted in an iron house (in 'The Intoxication of the Ulaid'). Yet this story-world is also magically bright and achingly beautiful. Two pairs of lovers—Mider and Étaín (in 'The Wooing of Étaín), and Óengus and Cáer Ibormeith (in 'The Dream of Óengus')—turn into swans. The hero of 'The Destruction of Da Derga's Hostel' can dispatch several hundred foes without even reaching for his weapons; Macc Da Thó's pig is so large that forty oxen can be laid across it. Myth obtrudes upon reality at every turn. In 'The Destruction of Da Derga's Hostel', a bird descends through a skylight, sheds his bird outfit and sleeps with the woman Mess Búachalla, thus fathering the story's hero, Conare Már; in 'The Wooing of Étaín', Mider's wife, Fúamnach, turns her rival Étaín into a scarlet fly; in 'The Wasting Sickness of Cú Chulaind', Cú Chulaind is horsewhipped and then healed by two women from the otherworld (shades of the German women in Fellini's Casanova). In these Irish stories, then, the pride and energy of reality are allied with the magic and beauty of fantasy—and the result is infused with a rare degree of idealism. In the otherworld of 'The Wooing of Étaín', not only are bodies white as snow and cheeks red as foxglove, but there is no 'mine' or 'yours'.

The Celts

The traditions of these early Irish stories originated with the Celts, an Indo-European group who are the ancestors of the Irish, the Scots, the Welsh, the Cornish, the Bretons and the people of the Isle of Man. When and where this group first appeared is, rather fittingly, an elusive, even controversial, question. The conservative view, and perhaps the most prevalent, is that the Celts surfaced with the beginning of the Iron Age in Europe, roughly 1000 B.C.; and this is certainly the earliest period in which the archaeological testimony affords positive proof. Myles Dillon and Nora Chadwick, however, propose to date the first Celtic settlements of the British Isles to the early Bronze Age (circa 1800 B.C.) and to identify the Beaker Folk as Celts.¹ Leon E. Stover and Bruce Kraig go further still: comparing the Classical descriptions of the Iron Age Celts with what they infer from burials at Stonehenge and Únĕtice (a cemetery near Prague), they propose to classify 'the Wessex and Únĕtician warriors as formative Celts' and conclude by claiming that the Celts 'emerged as a dominant people in Europe by the beginning of the third millennium B.C.'² The controversy is largely semantic. Wessex as presented by Stover and Kraig does look like an early form of what is described by Posidonius and Caesar, but then so does the heroic society of Homer's Iliad, and of course there is no linguistic evidence at all. Presumably, from the beginning of the third millennium on there developed, in Europe and subsequently in Britain and Ireland, heroic societies that gradually became, both culturally and linguistically, Celtic.

In any event, by the beginning or the early part of the first millennium B.C., the Celts clearly had emerged, not as a subset of their Slavic or Germanic or Italic neighbours but as a discrete Indo-European ethnic and cultural group; moreover, during the course of that millennium, they became the dominant people in non-Mediterranean Europe. From their homeland (probably in Bohemia), they expanded westward into France and Spain and, eventually, Britain and Ireland; southward into Italy; and eastward into Turkey, where they became the Galatians of St Paul. These early Celts took with them not only their chariots and their iron swords but also a distinctive geometric/linear art, called Hallstatt (after an important cemetery in Austria). By 500 B.C., a new art form had sprung up, this called La Tene (after a site in Switzerland); much less restrained than its predecessor, La Tene is a kind of baroque development, all curves and spirals and luxuriant plant and animal outgrowths. At this time, too, the Celts began to come under notice of the Classical authors: Herodotos, writing in the mid-fifth century, described the Keltoi as tall (by Mediterranean standards) and with light skin and hair and eyes, boastful and vainglorious but demonic in battle, childlike and ostentatious but hospitable, fond of hunting and feasting and music and poetry and glittering jewellery and bright colours; and his impressions were confirmed by subsequent accounts, particularly those attributed to Posidonius in the first century B.C.³

With their energy and warlike temperament, the Celts were able to expand quickly; by 390 B.C., they had sacked Rome, and by 279 B.C., Delphi. Many tribes settled in France, where the Romans called them Gauls, but their numbers also included the Boii (Bologna, Bohemia), the Belgae (Belgium) and the Helvetii (Switzerland); moreover, their settlements included Lutetia Parisiorum (Paris), Lugudunum (Lyon), Vindobona (Vienna) and Mediolanum (Milan), and they also named the Sequana (Seine) and the Danuvia (Danube). Unfortunately, Celtic tribal free-spiritedness was no match for Roman civic organization. Caesar's defeat of Vercingetorix, at Alesia in 52 B.C., signalled the decline of the Celts' hegemony in Europe; thereafter, they were overrun and assimilated. As a distinct entity, Celtic language and culture disappeared in Europe (though of course their influence persisted); in Great Britain, the Celtic tribes were driven back into Scotland, Wales and Cornwall (from where they eventually reclaimed Brittany) by the numerous incursions of Romans, Angles/Saxons and Normans.

Ireland was a different story. By virtue of its westerly and isolated geographic position, this island remained free of Roman colonization; thus, Irish society did not change appreciably until the advent of Christianity (in the fifth century) and the arrival of Viking raiders (some time thereafter). Consequently, the culture of the Iron Age Celts survived in Ireland long after it had been extinguished elsewhere. It is this conservatism that makes the early Irish tales, quite apart from their literary value, such a valuable repository of information about the Celtic people.

The Irish

As elusive as the date of the Celts' emergence in Europe is the date of their arrival in Ireland. Such megalithic tombs as Knowth, Dowth and New Grange, which now appear to date from the middle of the fourth millennium, testify to the presence of an indigenous, pre-Celtic culture; but how soon afterwards Celts—even formative Celts—appeared is open to controversy. If the Bell-Beaker people are viewed as proto-Celts, then one might say that they—assuming they reached Ireland as well as Britain—represent the beginnings of Celtic culture in Ireland; against this, archaeological evidence of large-scale immigration to Ireland between 2000 and 600 B.C. is wanting. If the indigenous population evolved into a Celtic one at the behest of a small number of aristocratic invaders, however, no such large-scale immigration would have been necessary. In any event, we know that Celts of the Hallstatt type reached Ireland by the middle of the sixth century and that Celts continued to migrate to Ireland and Britain until the time of the Belgic invasion in the first century B.C.

How and in what form they arrived is even more uncertain. According to Lebor Gabála (The Book of Invasions), our earliest copy of which dates from the twelfth century, Ireland was subjected to six invasions, those of Cessair, Partholón, Nemed, the Fir Bolg, the Túatha Dé Danand and the sons of Míl Espane. Irish history being what it is, the particulars of the Lebor Gabála account are open to question; what matters is that Ireland was, or was felt to have been, settled by a succession of different tribes. That these people actually arrived in separate waves—as opposed to filtering in more or less continuously—is moot; but the early tales do reflect the existence of different ethnic groups.

The Ireland of these tales is apportioned into four provinces, called, perversely, cóiceda, or 'fifths': Ulaid (Ulster), Connachta (Connaught), Lagin (Leinster) and Mumu (Munster). The fifth province was probably Mide (Meath), though there is also a tradition, probably artificial, that Mumu was once two provinces. Either this fifth province was original and disappeared (while the word cóiced persisted), or else the original four provinces became five after the emergence of a new power centre. Mide, which encompassed both Bruig na Bóinde (New Grange) and Temuir (Tara), is the setting for the early mythological tales, and this argues for its status as an original province. On the other hand, Mide was also the territory of the Uí Néill, who by the fourth century had supplanted the Ulaid as the dominant power in Ireland; this argues for its being a later addition. Moreover, the name Mide, which means 'middle', looks palpably artificial—of course, the entire province setup may be artificial.

In any case, there are, in the stories of this volume, four centres of action. Mide, with its numerous burial mounds, is the setting for the early mythological tales. It is peopled by the Túatha Dé Danand (the People of the Goddess Danu), who, though presented by Lebor Gabála as a wave of invaders, appear in these tales as the denizens of the otherworld, the Síde. They interact freely with the ordinary people of the mythological stories, and they also appear in some of the more historical tales. Ulaid, with its capital of Emuin Machae (near present-day Armagh), is the primary setting for the historical (insofar as any of the Irish tales are historical) sagas of the Ulster Cycle; its king is Conchubur son of Ness, but its champion is the mythic hero Cú Chulaind. The arch-enemies of the Ulaid (province names apply to the people as well) are the Connachta, who have their capital at Crúachu, in the west of Ireland. These people may well have originally occupied Mide, for their queen, Medb, is often identified as the daughter of the king of Temuir, and she may once have been a fertility goddess. It also seems more logical that Ulaid's foe should have been centred in adjacent Mide rather than in the distant west; and this in fact would have been true if the Ulster Cycle tales reflect the historical conflict between the Ulaid and the emerging Uí Néill of Mide. The tradition that the Connachta were the enemies of the Ulaid coupled with the fact that Connachta was now the name of Ireland's western province would have given the storytellers sufficient reason to move Medb and her husband, Ailill, from Temuir to Crúachu. Finally, there are the people of Mumu; they play a more peripheral role in the Ulster Cycle, but the king Cú Ruí son of Dáre does figure prominently in several tales.

When the events related in these stories might have taken place is yet another mystery. The Alexandrian geographer Ptolemy, who wrote in the second century A.D. but is believed to have drawn upon sources at least two hundred years older, provides evidence that Ireland was then Celtic-speaking; however, few of his names—and they are restricted both in number and in location—suggest those of our stories, so that one might suppose the people of these stories (insofar as they were real) had not yet appeared. At the other end, the milieu of the tales predates the advent of Christianity, while the circumstances of the Ulster Cycle must predate the Uí Néill appropriation of Emuin Machae. Kenneth Jackson has placed the formation of the Ulster tradition somewhere between the second century B.C. and the fourth century A.D., which seems entirely reasonable.

What Irish life was like during this period is, fortunately, not such a difficult question. On the one hand, we have the evidence of the Classical authors, Posidonius (via Diodorus Siculus and Strabo) and Caesar—evidence that was taken from Gaul and Britain but must surely have been valid for the Irish Celts as well. On the other, we have not only the evidence of the stories but also that of the Irish annals and genealogies and law tracts.

What emerges from the collation of this evidence is a culture of extraordinary vitality and beauty. Irish society exhibited the same tripartism that Georges Dumézil perceived elsewhere in the Indo-European world: a warrior class headed by a king; a priestly class (the druids); and a class of farmers and free men. The king of a túath, or tribe, was often subject to an over-king, to whom he gave assurances of allegiance and from whom he received some kind of support; the over-king, in turn, might have been subject to his provincial king. (The idea of a high king or king of Ireland is probably a fiction, fabricated by later peopl'es—notably the Uí Néill—to provide a historical justification for their claim to rule Ireland and perpetuated by the romanticism of subsequent tradition.) Kingship seems originally to have been sacral—indeed, the 'kings' in the mythological tales are barely euhemerized gods. In some traditions, the tribal king was ritually married to the tribal goddess (Medb, for example); in others, he had a sympathetic relationship with the land: if he were healthy and virile, the land would be fertile, while if he were blemished or impotent, the land would become barren. (This Wasteland idea is not, of course, exclusively Celtic.)

In 'The Destruction of Da Derga's Hostel', a druid partakes of the flesh and broth of a slaughtered bull and then lapses into a deep sleep, wherein he is expected to see the form of the new king. In later Irish history, however, the king was chosen from an extended family unit; and his position, continually contested by other family members (just as in fifteenth-century England), was far from secure. Curiously, the kings of the Irish stories are not battle leaders: either they betray vestiges of divinity (Cú Ruí, for example) or they have a young champion as heir and rival. Examples of this second pattern—which reflects the relationship of Agamemnon to Achilles and anticipates those of Arthur to Lancelot and Mark to Tristan—are legion: Mider temporarily loses Étaín to his foster-son Óengus; Conchubur loses Derdriu to the young warrior Noísiu and relinquishes supremacy in battle to Cú Chulaind; Cet rather than King Ailill is the champion of the Connachta.

The second class of Irish society, the priests, is more controversial. Popular notions of white-robed druids overseeing human sacrifices, cutting mistletoe with golden sickles and chanting spells over magic cauldrons persist—and not without reason. But Strabo points out that the druids

concern themselves with questions of ethics in addition to their study of natural phenomena. And because they are considered the most just of all, they possess the power to decide judicial matters, both those dealing with individuals and those involving the common good. Thus they have been known to control the course of wars, and to check armies about to join battle, and especially to judge cases of homicide. When there is a large number of these last, they suppose there will be a large return from the land as well. And both they and others maintain that the soul and the cosmos are immortal, though at some time in the future fire and water will prevail over them.⁴

Diodorus, moreover, makes mention of

certain philosophoi and religious interpreters, men highly honoured, whom they call Druids … It is their custom not to make any sacrifice without one of these philosophoi, since they believe that offerings should be rendered to the gods through the agency of those well acquainted with the divine nature (on speaking terms, one might say), and that requests for favours should likewise be made by these same men. In matters of war too the philosophoi are readily obeyed, they and the singing bards, and this by enemies as well as their own people. Often, in fact, when battle lines are drawn and armies close ground with swords and spears poised, they will step out into the middle and halt both sides, as if enchanting wild beasts. Thus even among the most savage barbarians, the spirit yields to the arts, and Ares reveres the Muses.⁵

Valuable as they are, these Classical accounts, at second hand and biased, should not be accepted at face value: the druids were, most probably, neither human-sacrificing savages nor great moral philosophers. Certainly, there is no evidence of either role in the Irish tales. In the mythological stories, druids are magicians: in 'The Wooing of Etain', Fuamnach, who has been reared by the druid Bresal, is able to turn her rival, Étaín, into first a pool of water and then a scarlet fly; in 'The Destruction of Da Derga's Hostel', Ingcdl's druids bring about Conare's death by making him thirsty. The druids of the Ulster Cycle, however, are little more than wise old men (reminiscent of Nestor), though they claim some power of prophecy. Cathub and Senchae are greatly revered for their sagacity and for their peacemaking ('Bricriu's Feast' and 'The Intoxication of the Ulaid' fully confirm Diodorus's account of druidic intervention between combatants), but they display neither magical powers nor moral philosophy. It seems that the process of becoming a druid was a protracted one—according to Caesar, it could take twenty years—and involved the study of myth/history, law, science, religion and philosophy. Since the Celts in general and the druids in particular were averse to writing their knowledge down (out of fear that it might be corrupted if outsiders found it, but doubtless also because of the druids' desire to preserve their privileged status), all this material had to be memorized. In short, the druids appear to have been the caretakers of whatever knowledge—from magic to science—their people possessed.

The third class of people were free men who farmed and herded. As the clients of a chieftain or other land-owner, they received rent of the land, perhaps some stock, and some protection from enemies; in return, they surrendered a portion of what the land yielded and did some kind of service for their landlord. The upper class of these tenant farmers took possession of the rented stock after seven years; the lower classes did not and were in effect serfs. At the bottom of the social scale were the slaves; these were often people captured from neighbouring tribes, but they do not appear to have been numerous.

Irish society, especially that of the historical tales, was an aristocratic one. The strongholds of the Ulster Cycle—Crúachu and Emuin Machae—are not cities but rather compounds where the king lives with his household and where he regales his chieftains with feasts and entertainments: poets, singers, musicians, jugglers. These strongholds may also have been centres for rounding up stock in autumn and for the holding of annual fairs, such as the one described at the beginning of 'The Wasting Sickness of Cú Chulaind': 'Each year the Ulaid held an assembly: the three days before Samuin and the three days after Samuin and Samuin itself. They would gather at Mag Muirthemni, and during these seven days there would be nothing but meetings and games and amusements and entertainments and eating and feasting.' And drinking. Such a lifestyle dictated an expansionist policy towards one's neighbours, since, in order to distribute wealth to their clients, kings and chieftains first had to accumulate it. Even in the mythological stories, the importance of land and possessions is patent: in 'The Wooing of Étaín', Oengus asserts his right to land from his father, the Dagdae, and it is the wealth of Bruig na Bóinde that enables him to compensate his foster-father, Mider, when the latter is injured.

The Irish year was divided into two parts: winter and summer. The first day of November, called Samuin, was both the first day of winter and the first day of the new year; the feast has since given rise to Hallowe'en/All Saints' Day and contributed the bonfire to Guy Fawkes celebrations. Samuin was a day of changes, of births and deaths; it was an open door between the real world and the otherworld. Oengus (in 'The Wooing of Étaín') dispossesses Elcmar of Bruig na Bóinde at Samuin, and he finds his beloved (in 'The Dream of Oengus') at Samuin. It is at Samuin that Da Derga's hostel is destroyed and Conare Mar is slain (the death of a king at Samuin is so common as to suggest regeneration myths and ritual slaying); it is at Samuin that, in 'The Wasting Sickness of Cú Chulaind', beautiful birds appear at Mag Muirthemni and Cú Chulaind is entranced by Fand; it is at Samuin that, in 'The Intoxication of the Ulaid', the Ulaid charge off to the southwest of Ireland and are nearly burnt inside an iron house. Proinsias Mac Cana has called Samuin 'a partial return to primordial chaos … the appropriate setting for myths which symbolise the dissolution of established order as a prelude to its recreation in a new period of time';⁶ and there can be no doubt that Samuin was the most important day of re-creation and rebirth in Ireland.

The first day of May, called Beltene, marked the beginning of summer; this feast has since given rise to May Eve/Walpurgisnacht and May Day. Beltene was a less important day, and, consequently, less information about it has survived; the name seems to mean 'fire of Bel' (Bel presumably being the Irish descendant of the continental god Belenos) or 'bright fire', and there is a tradition that cattle were driven between two fires on this day so that the smoke would purify them. In any case, the rites of Beltene were probably directed towards ensuring the fertility of land and stock. The Welsh hero Pryderi is born on the first of May, and this fact coupled with the unusual circumstances of his birth (the concurrent birth of colts, the otherworld visitor) suggests that Beltene was also a day when the real and the fantastic merged.

The beginnings of spring and autumn were also celebrated, but even less is known about these holidays. Imbolg, which fell on the first of February, seems to have been the beginning of the lambing season; it is also associated with the goddess Brigit (Briganti in Britain), whose successor, Saint Brighid, has her feast day, significantly, on the first of February. Lugnasad, which fell on the first of August, was named after the god Lug and seems to have been a harvest festival; if so, it was probably a late addition, since harvest time (that is, the end of the grazing season) in a pastoral (as opposed to an agrarian) community would have fallen closer to Samuin. In any case, the opening sentences of 'The Wasting Sickness of Cú Chulaind' show that the annual autumn round-up and assembly of the Ulaid took place at Samuin.

For Celtic and Irish religion, there is a wealth of evidence: the testimony of the Classical writers, especially Caesar; that of Gaulish sculpture and inscriptions; and that of the surviving Welsh and Irish myths. The resultant picture, however, is far from clear. Caesar identifies a Gaulish pantheon headed by Mercury and including Apollo, Mars, Jupiter and Minerva; corroborating evidence is so absent, however, that one has to suspect he is simply pinning Roman tails on a Celtic donkey.⁷ It is the Gaulish sculptures and inscriptions (we have no stories, unfortunately) that attest to the true nature of Celtic religion: no pantheon, but rather localized deities with localized functions; and this accords with what we know of the Celts politically, for they had little tolerance for centralized authority, even their own. The more widespread and possibly more important deities include Lugos (Mercury in Caesar, Lug in Ireland, Lieu in Wales; he gave his name to Lyon, Leiden and Liegnitz (Legnica), as well as to the Irish autumn festival of Lugnasad); Belenos, whose name means 'bright' and who might have been a rough counterpart to Apollo; Maponos (Mabon in Wales, the Macc Oc in Ireland; his name means 'great son'); Ogmios, whom Lucian describes as the Gaulish Herakles and as a god of eloquence;⁸ Cernunnos, whose name means 'horned' and who presumably is the horned figure on the Gundestrup cauldron; and Epona, a goddess whose name means 'great horse'. Much attention has been given to the trio of Esus, Taranis and Teutates in Lucan⁹ and to the sacrifices with which they allegedly were appeased (hanging, burning and drowning, respectively), but their true importance is uncertain. Evidence as to how these and other Celtic gods (who are literally too numerous to mention) related to each other—the kind of testimony we find in Greek mythology—is totally lacking.

The evidence of the Irish tales, our third and final source, is abundant, but it has suffered from faulty transmission, political distortion, historical overlays and church censorship; the result is no clearer than that from the continent. The Ireland of the tales comprises two worlds, 'real' and 'other'; but the line between them is not well demarcated. Even the location of the otherworld—which should not be confused with the Classical underworld—is uncertain: sometimes it is to the west, over the sea; sometimes it is in the southwest of Ireland (where it may be called the 'House of Dond', Dond being a chthonic deity); but usually it is found in the great pre-Celtic burial mounds of the Side, of which the most important in the tales is Bruig na Bóinde, today's New Grange. The Irish otherworld is, not surprisingly, a stylized, idealized version of the real one: everyone is beautiful, and there is an abundance of beautiful things, and the joys of life are endless—hunting, feasting, carousing, perhaps even love. Paradoxically (of course), though this otherworld makes the real one seem a shadow by comparison, it is the Side who are the shadows, for they have no physical strength for fighting; just as Pwyll, in 'Pwyll Lord of Dyved', is asked to fight on behalf of the otherworld ruler Arawn, so Cú Chulaind, in 'The Wasting Sickness of Cú Chulaind', is asked to fight on behalf of the otherworld ruler Labraid Lúathlám. The Side are distinguished primarily by their power of transformation: they move invisibly, or they turn themselves (and others) into birds and animals. But they exert no moral authority, and, while they can injure and heal, they do not have that power over life and death characteristic of the Greek Olympians. Often they seem just like ordinary humans.

Relatively few of the names from Gaulish inscriptions reappear in Ireland—given the decentralized nature of Gaulish religion, this is not surprising. Lug is the major figure in 'The Second Battle of Mag Tured', but in the stories included in this volume he appears prominently only as the father of Cú Chulaind. The Macc Óc is a central character in both 'The Wooing of Étaín' and 'The Dream of Óengus', but he has been so thoroughly euhemerized that there is no trace of the Gaulish Maponos; and such names as the Dagdae, Mider, Bóand, Étaín, Cáer Ibormeith, Medb and Cú Ruí have no apparent continental counterparts. Many of the quasidivine figures in these tales are associated with animals or with natural features. The name Bóand, for example, means 'white cow'; but Bóand is also the Irish name of the river Boyne. At the outset of 'The Wooing of Étaín', Bóand sleeps with the Dagdae, whose other name, Echu, means 'horse'; Frank O'Connor saw this 'love affair' between a horse god and a cow goddess as a reconciliation between Bronze Age invaders and the indigenous Neolithic civilization, which gives some idea of how old these stories might be.¹⁰ Like Rhiannon in 'Pwyll Lord of Dyved', Macha of 'The Labour Pains of the Ulaid' is a euhemerized horse goddess; and the same may be conjectured of Étaín, whose epithet Échrade means 'horse troop'. A number of the Síde appear as birds: Mider and Étaín leave Temuir as swans, and Óengus (Mider's foster-son) and Cáer Ibormeith return to Bruig na Bóinde as sẘans; Conare Már's unnamed father discloses himself to Mess Búachalla in the form of a bird; and Fand and Lí Ban first present themselves to Cú Chulaind as birds.

Strabo's testimony, the evidence of lavish grave goods buried with the wealthy, and the identification of the Boyne burial mounds as the dwelling place of the Side all suggest that the Irish did believe in a life after death. But the Irish otherworld was not simply an anticipated joyful afterlife; it was also—even primarily—an alternative to reality, a world that the hero might enter upon the invitation of a king or a beautiful woman. Inasmuch as this otherworld, no matter how beautiful, is not quite human (there is, for example, no winter), the hero never stays; but the alternative—and thus the tension—is always present.

Finally, there is the language, as beautiful and elusive as any aspect of Irish culture. Just as the Celts were a distinct Indo-European entity, so their languages formed an independent branch of the Indo-European language tree; nonetheless, Celtic is more like Italic (that is, the Romance languages) than it is like any of the other Indo-European language groups, and many place and personal names in Gaulish are very similar to those in Latin. For example, the Gaulish suffix -rix (as in Vercingetorix) is the counterpart of the Latin word rex, both meaning 'king'.

In the British Isles, the Celtic languages divided into two groups, one spoken primarily in Britain (and comprising Welsh and, eventually, Cornish and Breton), the other spoken primarily in Ireland (and comprising Irish Gaelic and, eventually, Scottish Gaelic and Manx Gaelic). The most obvious (though not necessarily the most important or fundamental) difference between the two groups is that Indo-European q^u became p in the British languages (the word for 'four' was petwar) and c in the Irish group ('four' was cethair).

At the time our stories are purported to have taken place—which is to say any time before the fourth century—the Irish language probably looked a good deal like Gaulish and not so very different from Latin. By the time these stories were being written down, however—and this could have begun as early as the seventh century—drastic changes had taken place: many final syllables had dropped away, many medial vowels had disappeared and many medial consonants had been simplified or lightened. Thus, the word for 'horse', equus in Latin, had become ech in Ireland at this time. The language of the tales, then, is quite different from that of the time they describe; and this makes the correlation of the stories' proper names with those in earlier sources (such as Ptolemy's geography) even more difficult. Although the syntax of the new language was straightforward, the morphology was not: regular verb conjugations often looked wildly irregular, and word roots occasionally disappeared altogether. The principles of phonetic change were aesthetic rather than semantic; the resultant language was soft and subtle, verb poor but noun-and-adjective rich, static and yet vital.

Irish Storytelling

Irish literature—meaning whatever was written down in Irish—of this time encompassed a broad area, including history, genealogy and law tracts; but it is poetry and narrative prose that are relevant to the early Irish myths and sagas. The earliest poetry was alliterative and syllabic, with end-rhyme appearing later. In Welsh literature, there are epics told entirely through the medium of verse—the Gododdin, for example; in Ireland, however, the storytelling medium is invariably prose. Some of the very archaic poetry is essential to the tales in which it appears; thus, the rhetorics in the early part of 'The Cattle Raid of Cúailnge' help to clarify the relationship among Ailill, Medb and Fergus.¹¹ The poetry in 'The Exile of the Sons of Uisliu', on the other hand, reinforces the narrative, adds detail—mostly descriptive—and provides weight; but it could be omitted without loss of sense. Conceivably these myths/sagas were at one time recited entirely in verse; what remains, however, is largely decorative.

The earliest form of transmission must have been oral. Storytelling was a favourite entertainment among the Celts, and one version of 'The Voyage of Bran' states that Mongan (an Ulaid king who died about A.D. 625) was told a story by his filí (a kind of poet) every winter night from Samuin to Beltene. Presumably, the storytellers did not memorize entire tales—rather, they memorized the outlines and filled in the details extemporaneously. Eventually, perhaps as early as the seventh century, the tales began to be transcribed; and thereby two processes, rather opposite in effect, were initiated. In many cases, tales are reworked and acquire a literary veneer; this is certainly true of the Book of Leinster opening to 'The Cattle Raid of Cúailnge', and it would seem to apply to 'The Cattle Raid of Fróech' and to the concluding section of 'The Wasting Sickness of Cú Chulaind'. But these same tales have also deteriorated considerably by the time they reach our earliest (twelfth-century) surviving manuscripts. This deterioration is not likely to have originated with the storytellers themselves, for a long tale would naturally be prolonged over several evenings (which would be in the storyteller's interest, since during that time he would be enjoying his host's hospitality); and in any case, as James Delargy has pointed out, no audience would 'have listened very long to the story-teller if he were to recite tales in the form in which they have come down to us'.¹² The people who wrote these stories down, however, were—for the most part—not literary artists; and of course, they lacked the incentive of an appreciative (and remunerative) audience. Banquet-hall transcription cannot have been easy, and the scribe doubtless grew weary before the storyteller did; consequently, it is not surprising that spelling is erratic, that inconsistencies abound (this could also result from a storyteller's attempting to conflate multiple traditions) and that many tales deteriorate after a promising beginning. Some formulaic passages, such as in 'The Destruction of Da Derga's Hostel', are represented simply by 'et reliqua'. As manuscripts were recopied, moreover, additional errors inevitably appeared. Some areas are manifestly corrupt, and in the case of the archaic poetic sections it seems doubtful whether the scribes understood what they were writing. All this is hardly surprising—just consider the problems attendant upon the texts of Shakespeare's plays, only four hundred years old—but it should be remembered that what survives in the manuscripts, however beautiful, is far from representative of these stories at their best.

The Irish Manuscripts

The language of these tales varies considerably as to date; but at its oldest, and allowing for some degree of deliberate archaism, it appears to go back to the eighth century; one may assume the tales were being written down at least then, if not earlier. Unfortunately, Scandinavian raiders were legion in Ireland at this time, and they tended to destroy whatever was not worth taking away; consequently, very few manuscripts predating A.D. 1000 have survived. Among the missing is the Book of Druimm Snechtai, which belonged to the first part of the eighth century and included 'The Wooing of Étaín', 'The Destruction of Da Derga's Hostel' and 'The Birth of Cú Chulaind'.

Of the manuscripts that have survived, the two earliest and most important for these tales belong to the twelfth century. Lebor na huidre (The Book of the Dun Cow) is so called after a famous cow belonging to St Ciaran of Clonmacnois; the chief scribe, a monk named Mael Muire, was slain by raiders in the Clonmacnois cathedral in 1106. Unfortunately, the manuscript is only a fragment: though sixty-seven leaves of eight-by-eleven vellum remain, at least as much has been lost. Lebor na huidre comprises thirty-seven stories, most of them myths/sagas, and includes substantially complete versions of 'The Destruction of Da Derga's Hostel', 'The Birth of Cú Chulaind', 'The Wasting Sickness of Cú Chulaind' and 'Bricriu's Feast' as well as an incomplete 'Wooing of Étaín' and acephalous accounts of 'The Intoxication of the Ulaid' and 'The Cattle Raid of Cúailnge'.

The second manuscript, which is generally known as the Book of Leinster, is much larger, having 187 nine-by-thirteen leaves; it dates to about 1160 and includes in its varied contents complete versions of 'The Cattle Raid of Fróech', 'The Labour Pains of the Ulaid', 'The Tale of Macc Da Thó's Pig' and 'The Exile of the Sons of Uisliu' as well as an unfinished and rather different 'Intoxication of the Ulaid' and a complete, more polished 'Cattle Raid of Cúialnge'. Two later manuscripts also contribute to this volume: the Yellow Book of Lecan, which offers complete accounts of 'The Wooing of Étaín' and 'The Death of Aífe's Only Son' and dates to the fourteenth century; and Egerton 1782, which includes 'The Dream of Oengus' and has the date 1419 written on it.

These manuscripts do not, of course, date the stories they contain. Our earliest complete version of 'The Wooing of Étaín' appears in the fourteenth-century Yellow Book of Lecan, yet we have a partial account in the twelfth-century Lebor na huidre, and we know from the contents list of the Book of Druimm Snechtai that the tale was in written form by the early eighth century. What we do not know—and probably never will—is whether the Druimm Snechtai version was very different from the one in the Yellow Book of Lecan, whether the tale assumed written form earlier than in the eighth century, and what the tale was like before it was first written down. Even the surviving manuscripts, which we are fortunate to have, are far from ideal: obscure words abound, some passages seem obviously corrupt, and there are lacunae and entire missing leaves.

The Irish Material

Convention and tradition have classified the early Irish tales into four groups, called cycles: (1) the Mythological Cycle, whose protagonists are the Síde and whose tales are set primarily among the burial mounds of the Boyne Valley; (2) the Ulster Cycle, which details the (purportedly historical) exploits of the Ulaid, a few centuries before or after the birth of Christ; (3) the Kings Cycle, which focuses on the activities of the 'historical' kings; (4) the Find Cycle, which describes the adventures of Find mac Cumaill and his fíana and which did not achieve widespread popularity until the twelfth century. Although these categories are useful, it should be remembered that they are also modern (no particular arrangement is apparent in the manuscripts, while it seems that the storytellers grouped tales by type—births, deaths, cattle raids, destructions, visions, wooings, etc.—for ease in remembering) and artificial. Characters from one cycle often turn up in another: the Síde-woman Bóand is introduced as Fróech's aunt in the Ulster Cycle's 'Cattle Raid of Fróech'; the otherworld-figure Manandán appears in the Ulster Cycle's 'Wasting Sickness of Cú Chulaind' and in the Kings Cycle's 'Adventures of Cormac'; Ulaid warriors join the invaders in the Mythological Cycle's 'Destruction of Da Derga's Hostel'; Ailill and Medb, king and queen of Connachta, take part in the Mythological Cycle's 'Dream of Óengus'. Also, one should not suppose that the Mythological Cycle is populated exclusively by deities or that the other cycles are inhabited exclusively by mortals: many of the 'humans' are barely euhemerized gods, many of the 'gods' behave much like humans, and the two groups are often difficult to distinguish.

The material of these tales encompasses both impacted myth and corrupted history. Although Irish mythology does evince the tripartism detected by Georges Dumézil in other Indo-European cultures ('The Second Battle of Mag Tured' is on one level an explanation of how the priests and warriors—Dumézil's first two functions—wrested the secrets of agriculture from the third function, the farmers), its fundamental orientation seems more seasonal than societal, for the mythic subtexts of the tales focus on themes of dying kings and alternating lovers. (This strong pre-Indo-European element in Irish mythology probably derives both from the Celts' innate conservatism and from the fringe position of Ireland in the geography of the Indo-European world.) These themes are stated most clearly in 'The Wooing of Étaín' and 'The Exile of the Sons of Uisliu'. In the former story, Bóand passes from her husband, Elcmar, to the Dagdae (also called Echu) and then returns to Elcmar; Étaín goes from Mider to Óengus and back to Mider, from Echu Airem to Ailill Angubae and back to Echu, and from Echu Airem to Mider and back (in some versions) to Echu. In the latter tale, Derdriu passes from an old king, Conchubur, to a young hero, Noísiu, and back to Conchubur after Noísiu's death; when Conchubur threatens to send her to Noísiu's murderer, she kills herself. Sometimes, the woman's father substitutes for the dying king (this variant appears in the Greek tales of Jason and Medea and Theseus and Ariadne): Óengus has to win Étáin away from her father in 'The Wooing of Étaín' and Cáer Ibormeith away from hers in 'The Dream of Óengus'; Fróech has to win Findabair from Ailill and Medb—but primarily, and significantly, from Ailill—in 'The Cattle Raid of Fróech', while Cú Chulaind has to win Emer from Forgall in 'The Wooing of Emer'. Sometimes, the dying king is absent, and the regeneration theme is embodied in the wooing of a mortal hero by a beautiful otherworld woman (whom he often loses or leaves): Cáer Ibormeith seeks out Óengus in 'The Dream of Óengus', Macha comes to Crunniuc in 'The Labour Pains of the Ulaid', Fand appears to Cú Chulaind in 'The Wasting Sickness of Cú Chulaind'. (This variant persists even into the Find Cycle, where Níam's wooing of Oisín becomes the basis of Yeats's 'The Wanderings of Oisín'.) And sometimes the theme treats only of the dying king: in 'The Destruction of Da Derga's Hostel', Conare Már is slain, at Samuin, in the hostel of a chthonic red god; in 'The Intoxication of the Ulaid', Cú Chulaind is nearly burnt, also at Samuin, in an iron house in the southwest of Ireland (where the House of Dond, an Irish underworld deity, was located). Centuries of historical appropriation and Christian censorship notwithstanding, these regeneration themes are never far from the narrative surface; and in their ubiquitousness is apparent their power.

As history, the early Irish tales verge upon wishful thinking, if not outright propaganda. The Ulster Cycle, however, does appear to preserve genuine traditions of a continuing conflict between the Ulaid (who appear to have concentrated in the area round present-day Armagh) and the Uí Néill (who were probably centred at Temuir, though for reasons suggested earlier they have been moved to present-day Connaught by the storytellers); in any case, it is a valuable repository of information about the Ireland of prehistory—what Kenneth Jackson has called 'a window on the Iron Age'¹³—with its extensive descriptions of fighting (chariots are still the norm) and feasting (an abundance of strong words and strong drink) and dress (opulent, at least for the aristocracy) and its detailing of such institutions as fosterhood, clientship and the taking of sureties. The important but not very surprising conclusion generated by this information is that the Irish society represented by the Ulster Cycle is still very similar to the Gaulish civilization described by Caesar; and there are good reasons to think it not very different from the Celtic world of an even earlier period.

What is surprising, though, is that these tales—which betray a natural and unmistakable bias towards the Ulaid and against the Connachta—do not more consistently depict Ulster society at its zenith. Cú Chulaind is the only true hero in the Ulster Cycle, and his deeds are more often superhuman than heroic; Conchubur, as early as 'The Boyhood Deeds of Cú Chulaind', serves notice that he will be largely a roi fainéant; and among the Ulaid warriors there is, 'The Cattle Raid of Cúailnge' excepted, more talk than action. Odder still, in many of the best-known and most important tales, there are clear instances of parody. In 'The Death of Aífe's Only Son', the Ulaid are awestruck by the feats of a seven-year-old boy; in 'The Tale of Macc Da Thó's Pig', Ulaid and Connachta are reduced to fighting over a dog (at least, in 'The Cattle Raid of Cúailnge', the bone of contention is a bull), and the Ulaid are ridiculed and put to shame by the Connachta champion; in 'The Intoxication of the Ulaid', Cú Chulaind loses his way and leads the Ulaid on a drunken spree across Ireland, while the two druids guarding Cú Ruí's stronghold bicker and quarrel; and in 'Bricriu's Feast', the wives of the Ulaid warriors squabble over precedence in entering the drinking hall, while Bricriu is accidentally flung out of his house and on to a garbage dump. Conchubur's treachery (equivalent to Arthur's murdering Lancelot) in 'The Exile of the Sons of Uisliu' eliminates any doubt: the society of the Ulster Cycle, for all the splendour that attaches to it, is a society in decline.

Notes

¹ Myles Dillon and Nora Chadwick, The Celtic Realms (New York: New American Library, 1967), pp. 1-2, 214.

² Leon E. Stover and Bruce Kraig, Stonehenge: The Indo-European Heritage (Chicago: Nelson-Hall, 1978), p. 141.

³ Herodotos, 2:33.

⁴ Strabo, Geographia, 4.4.4 (translation by Timothy Gantz).

⁵ Diodorus Siculus, 5: 31.2, 4-5 (translation by Timothy Gantz).

⁶ Proinsias Mac Cana, Celtic Mythology (London: Hamlyn, 1970), p. 127.

⁷ Julius Caesar, De bello gallico, 6.17.

⁸ Lucian, Herakles, 1.1.

⁹ Lucan, De bello civili, 1.444-6.

¹⁰ Frank O'Connor, The Backward Look: A Survey of Irish Literature (London: Macmillan, 1967), p. 242.

¹¹ A rhetoric is a dense, archaic poetic passage.

¹² James Delargy, The Gaelic Story-teller (London: G. Cumberlege, 1947), p. 32.

¹³ K. H. Jackson, The Oldest Irish Tradition: A Window on the Iron Age (Cambridge: Cambridge University Press, 1964).

Jaan Puhvel (1987)

SOURCE: "Celtic Myth," in Comparative Mythology, The Johns Hopkins University Press, 1987, pp. 166-88.

[In the following excerpt, Puhvel compares early Celtic mythology—from Julius Caesar's writings on the gods of Gaul, to Ireland's mythic history and the stories of the Ulster and Fenian Cycles—with the mythologies of other Indo-European cultures.]

"The Dead prevailed in testimony over the Living, since preference was given to the written word." This denunciation of the dead letter as against vivifying live speech occurs in an Irish tale about the falsification of the inscribed name on a famous sword, a forgery that impressed the judges more than the rightful owner's oral testimony. It is a late reflex of the well-known druidic aversion to writing mentioned by Caesar in De bello Gallico, a priestly attitude not unlike that of the Vedic brahmins. Both cultures had early access to writing (India from age-old contacts with Mesopotamia, Gaul since the founding of Massilia [= Marseilles] and other Greek colonies on the "French Riviera" about 600 B.C.E.) but chose not to make large-scale use of it until a later period (Epic India and the coming of Christianity respectively; in the wake of Saint Patrick, Irish monks would scribble furiously all over western Europe, as if to make up for centuries of self-imposed Celtic illiteracy). This tabu against writing may be based on dogma that lingered in the farthest east-west reaches of the Indo-European continuum, as did the archaic items of vocabulary discussed in chapter 3 (Vedic rāj-: Gaulish rīg- 'king', etc.). Caesar reports druidic teachings of metempsychosis, and India was of course the prime and expanding locus of transmigrational lore. Perhaps the transmission of the sacred text was held to be analogous to the progress of the individual and collective soul, with each successive generation rejuvenating and reincarnating the Word. Such sanctification of orality contrasts with the codificational trends first in evidence in the ancient Near East. Likewise the evidence for open-air divine service makes an impression of expansiveness, as opposed to an ever more centripetal focusing on some boxed-in holy of holies. In this latter respect the northern peoples (Celtic, Germanic, Baltic, Slavic) hewed to a seemingly old pattern, while Italy and Greece had already been strongly internalized by their Mediterranean-Near Eastern exposure.

Chapter 3 outlined the main phases of Celtic protohistory. It was the first great Indo-European migration traceable by historical as well as archeological means, but owing to its own essential illiteracy it remains hazily penumbral. Although the Celts managed to disperse themselves as far afield as Galatia (the old Hittite heartland in central Asia Minor), they were essentially ensconced in central and western Europe, with evanescent excrescences into Italy and Spain and more durable migrations into the British Isles. Pressed by Romanization and Germanic migrations, they were gradually squeezed to the northwestern fringes of Europe. The Celts are thus truly "marginal"; there were never other early Indo-Europeans to the west or north of them, and they were shielded by geography and choice against the impact of Mediterranean civilization. If sedentariness had been their virtue, coupled with early literacy, they would have been an extraordinary source for Western Indo-European archaism. As it is, one must arduously forage for scraps of their early traditions.

In moving westward through the Indo-European continuum, it appears that the social class divisions of priests, warriors, and cultivators, so pronounced in Vedic India in conjunction with pantheonic structures, and subsequently rigidified, played a much less pervasive part in Iranian tradition and were hardly in evidence in Greece and Rome. Despite some echoes here and there in religious, legendary, ideal, or real patterns, neither Greek (Dorian, Ionian, later Athenian) "tribal" divisions nor the Roman patrician: plebeian dichotomy preserved much of the prehistoric setup. Not so among the Celts, where Caesar separated the Gauls into druides (priests), milites (warriors), and miserrima plebs (wretched masses), or early Ireland, where the druids (cf. Vedic brahman-), the flith ('dominion'; cf. Vedic ksatram), and the bo airig ('cow freemen'; cf. Sanskrit irya-ka- and Avestan vastryofsuyant-) are uncannily reminiscent of the Indo-Iranian kind. Unlike Germanic society, where the system had been broken by the virtual disappearance of the priestly class, the tripartite societal pattern lingered on in its insular Celtic holding tank, and by the end of the first millennium it experienced a remarkable recrudescence in Anglo-Saxon England. About 900, in his own Old English version of Boethius's Consolation of Philosophy, King Alfred the Great divided a ruler's subjects into gebeomen, fyriòmen, and weorcmen, which Aelfric and Wulfstan echoed a century later by oratores, bellatores, and laboratores. Just as the Celtic-Germanic symbiosis in England had saved the tradition, the Norse superstrate on both sides of the Channel (culminating in the Norman invasion of 1066) facilitated its spread to the north of France during the eleventh century, whence it expanded to form the basis of the medieval European and later three estates, which in France lasted as the basis of society down to the revolution of 1789. This is another illustration of the insidious power of Celtic persistence, not unlike its substratal ability to overhaul a seemingly triumphant Latin into subsequently unintelligible French (see chap. 3).

Our written knowledge of the Gaulish pantheon begins with Caesar (De bello Gallico 6.16-17), who credits the Gauls with plenty of religious practices and enumerates their gods in Roman interpretation, with brief characterizations: "Of gods they worship Mercury most" as the omnium inventorem artium 'inventor of all skills', next Apollo (dispelling disease), Mars (in charge of war), Jupiter (holding sway in the heavens), and Minerva (originator of arts and crafts). This is a strangely unhelpful grab bag, notable for the low ranking of "Jupiter" and the preeminence of "Mercury," whose functions seem to overlap with Minerva's at the end; but Caesar was not doing systematic ethnography; he was merely padding an otherwise meager annual military report to the Senate with curious detail about the natives of Gaul. Caesar's account acquires some systematic meaning only by comparison with subsequent sources. A century or so later the poet Lucan (Pharsalia 3.399-425) depicted a blood-spattered sacred grove at Massilia and (1.444-46) described human sacrifice among the Gauls:

           immitis placatur sanguine diro
Teutates, horrensque feris altaribus Esus,
et Taranis Scythicae non mitior ara Dianae

          harsh Teutates is cruelly propitiated with blood,
and dread Esus on savage platforms,
and the altar of Taranis, a match for that of
  the Scythian Diana¹

Later commentators on Lucan supplied their own Roman interpretations for Teutates, Esus, and Taranis, either Mars, Mercurius, and Jupiter or Mercurius, Mars, and Dispater (the ruler of the Otherworld, whom according to Caesar the Gauls claimed as their ancestor). Caesar also mentioned huge wickerwork dummies inside which human victims were burned, which chimes with the information of Lucan's commentators that Teutates' victims were drowned in casks, those of Esus were hanged and lacerated, and those of Taranis were burned. At this point the names begin to be informative:

Teutātes (Toutates, Totates, Tutates) is derived from ^* tewta 'people' (Old Irish tūath, Oscan touto, Gothic thiuda) and thus resembles in meaning the Umbrian Vofione and the Roman Quirinus.

Ēsus may mean simply 'Lord' and be cognate with Latin erus 'master' (with rhotacism); yet there is also the variant form Aesus, pointing to an original diphthong.

Taranis is cognate with the u-stem ^* taranu- seen in Old Irish torann, Welsh taran 'thunder' (Taran is a man's name in the Welsh Mabinogi, and there was a Christianized Saint Taran in Brittany). The Celtic taranis metathetic for tanar- (= Germanic ^* thunar- 'thunder'), as seen in the inscription J(ovi) O(ptimo) M(aximo) Tanaro (Chester, England, 154 C.E.; Corpus Inscriptionum Latinarum 7.168).

People's god, lord, thunderer—this descriptively named triad does not sound very compatible with the Roman Jupiter, Mars, Quirinus, which explains the wretched results of the Roman interpretation. Teutates may somehow fit the third-estate or "collective" slot of Quirinus, but the relatively low estate of Caesar's "Jupiter" shows that Taranis is no head of the pantheon, rather a thunder-god with either wheel (= thunderball) or spiral (= thunderbolt) as emblem, of the order of the Norse Thor (< ^*Thunaraz). Equating Esus with Mars makes no particular sense either, once it is clear that Taranis is no match for Jupiter. But the alternative interpretation of Esus as "Mercury" is interesting on several counts:

1. In Tacitus's Roman interpretation of the Germanic pantheon, too, "they worship Mercury most," and there "Mercury" reflects with certainty the great god ^* Wo oanaz, the Norse Odin (see chap. 11).
2. Esus is homologous with the Norse Odin in receiving human sacrifices by hanging, whereas the victims of Teutates were drowned, in the manner of offerings to the Germanic fertility deities, for example, Tacitus's Nerthus (see chap. 11) or the man being plunged head-long into a vat in a cultic scene depicted in high relief on the inside of the Gundestrup cauldron, a Celtic silver-plated copper vessel found in a Danish peatbog (first century B.C.E.).
3. Granted "Mercury's" pride of place on Caesar's list, it is possible that "Mercury" = Esus is the high god of the Gaulish pantheon, above Taranis and Teutates.

Esus has mythologically significant iconography. The "Paris Altar," datable to the time of Christ and found in 1711 under the choir of the cathedral of Notre Dame de Paris (now in the Musee des Thermes at the site of the Roman baths of Lutetia Parisiorum), depicts on two sides Taranis (inscription Jovis) and Volcanus, and on a third the bearded, ax-wielding god Esus cutting a tree. The fourth panel shows a similar leafy tree protruding from behind a bull on whom three birds are perched, surmounted by the inscription Tarvos Trigaranus 'Bull with Three Cranes'. On an analogous relief from Trier (now in the Landesmuseum there) a woodcutter is felling a tree with a bull's head and three birds on top of its foliage. At the basis of this depiction may lie some lost myth reminiscent of Indra's killing of Triśiras, as told in Yajurvedic texts and especially in book 5 of the Mahābhārata, how Indra struck down the adversary with his bolt, but then enlisted the services of a woodcutter who completed the slaying by severing the three heads with his ax, whereupon birds (woodcock, partridge, and sparrow) escaped through the three necks. Tricephalic representations are common in Celtic iconography, but in the absence of texts, the full meaning of the Gaulish scene is bound to remain opaque.

The figure of Esus can perhaps be brought together with the theonym Lugus, which lurks in toponyms (Lugu-dunum 'Lugus town' > Lyons in France, Leiden in Holland, etc.), and with the Irish Lug (discussed further below) who was the all-around man of skill (samildanach; cf. "Mercury" as omnium inventor artium). Lug was the father of Cūichulainn in Irish saga, even as Odin was the sire of prominent Norse heroes; likewise Esugenus 'Begotten of Esus' was a Gaulish name of nobility. When Lugudunum = Lyons became the capital of Gaul, Augustus fixed his own imperial feast day there on I August, which was the date of the festival of Lugnasad in Irish lore and presumably had been the holy day of the town's eponym; the emperor "usurped" the antecedent cult, much as later Notre Dame de Paris rose on the emplacement of a sanctuary of Esus and Taranis.

Esus-Lugus, Taranis, and Teutates as a triad receiving human sacrifices may thus roughly match the Scandinavian set of Odin, Thor, and Freyr in pagan Sweden, who were given human victims at Uppsala up to the Christianization in the eleventh century (see chap. II). They, like Jupiter-Mars-Quirinus, were a stylized Western Indo-European embodiment of the erstwhile tripartite pantheon, thus a match for the Eastern structure first glimpsed at Mitanni (Mitra-Varuna, Indra, Nasatya; see chaps. 3 and 4). Around their axis swarmed the rest of the pantheon. "Mars" takes on more sense in two directions. There is epigraphic attestation of Mars Teutates (e.g., Marti Toutati) who, like the rural Roman Mars, seems more a patron of peace and fertility than a war-god, warding off plague, invasion, and blight alike. At the same time, "Mars" and "Mercury" alternated in Lucan's scholiasts as interpretations of Teutates and Esus, and in inscriptions both take similar epithets (e.g., Vellaunus, of unknown meaning). Thus "Mars" may hide yet another divine figure, structurally akin to Esus-Lugus and perhaps partially syncretized with him. One thinks here of the faded Vedic Mitra, Roman Diuis Fidius, and Scandinavian Tyr by the side of Varuna, Jupiter, and Odin, thus the god of covenants paired with the magical lord of oaths and bonds but subsequently eclipsed and almost absorbed by him (Mitra practically reduced to Varuna's dvandvamate, Diuis Fidius as an "allomorph" of Jupiter, Tyr "militarized" along with Odin in Scandinavia). "Mars" was likewise the Roman interpretation of the Germanic god ^* Tiwz (> Old Norse Tyr), as "Mercury" was of ^*Wōôanaz; a Latin inscription from northern England (third century C.E.) mentions Mars Thingsus, thus the Germanic god Tiw, the protector of the judicial assembly held on Tuesdays, even as the Scandinavian Tyr originally patronized the thing (Latin Martis dies came out in Germanic interpretation as Tuesday; see chap. II). The Gaulish name of such a "Mars" is not known, but he appears in Ireland as Nūadu beside Lug (see further below).

The rest of the Gaulish divine inventory can be characterized more in list fashion. "Apollo" does duty for both Borvo or Bormo (god of thermal healing waters; cf. the French name Bourbon) and Belenus 'Bright' or Grannus 'Sun' as solar deity (cf. Old Irish Beltene 'Brightfire', name of the Mayday feast, and grian 'sun'). "Vulcanus" is attested in Gaul, but his native name is not; it may have been based on the word for 'smith' (as his Irish counterpart, the divine smith Goibniu, reflected gobae 'smith'; cf. MacGowan 'Smithson'). "Hercules" characterized the Gaulish Ogmios, according to Lucian (Herakles, chap. I) an aged, baldish, sun-blackened champion who literally held men spellbound by the ears with golden-tongued eloquence, matching the Irish 'sun-faced' strongman Ogma, who was credited with devising the Ogam notch script; elements of his myth may have resembled those of the Indo-European warrior hero embodied by both the Greek Herakles and the Norse Starkaðr (the former met up with Geras 'Old Age' and was known as Hercules Musarum; the latter was both a senex and a great skaldic poet; cf. chap. 13). "Dispater" (Dis, the Latin name for Hades, from Dīves 'Rich' translating Greek Plouton = "Pluto") cropped up in Lucan's scholiasts as the alternative cover for Taranis but more plausibly reflects a death-god. The Gaulish deity Sucellus 'Good Striker' or Silvanus 'Woodsman', depicted with mallet and cask, sometimes with wolfskin and dog, may be a representation of this "ancestor" figure, who can perhaps be thought of as a "first colonizer" on the lines of the Vedic Yama or as the otherworldly replica of the ruling deity, in the manner of Zeus Katakhthonios or the Norse Odin, who extended his dominion to include Valholl. In any event belief in an afterlife was strong, for Gauls would lend money on terms of postmortem reimbursement.

"Minerva," with the epithet Belisama 'Brightest', is the cover term for a great goddess. Powerful female types stand out in Celtic mythical lore at both the divine and the saga levels. The transfunctional goddess has here come into her own. "Minerva" had a temple with "eternal flame" in third-century C.E. Britain and is identifiable with the British Celtic theonym Brigantia, formally identical with the Sanskrit feminine adjective brhati 'great, lofty' and with the Irish Brigit, the later saint with her feast day of Imbolc (I February) and her monastery with perpetual fire at Kildare. (Unlike the usual overlay, e.g., with the Virgin Mary superimposed on the sanctuary of Aphrodite at Cypriot Paphos, the Celtic deity was simply Christianized, name and all.) Triplicity or triunity is in evidence among the Celtic mythical females: Brigit herself had two synonymous "sisters," there was the triad of Irish Machas, and Gaul had the triple Matres or Matrae or Mātrōnae. Just as the Greek three-by-three Muses did not perturb Homer's Muse, the Matronae did not preclude a single great Matrona, embodied in a river (Matrona > Marne), whereas in Ireland the mother-goddess was the land itself (Ēriu; cf. the Indic Sarasvatf [river] vs. the Iranian Harahvaiti [land]). Matrona was the mother of the 'Divine Son', Maponos, matching Modron and her son Mabon in Welsh saga and the river-goddess Bō and (> Boyne) and her 'Young Son' (Mac Ōc = Oengus) in Irish lore. In addition the Gaulish male deities had "consorts," somewhat as in Roman or Indic religion (Rosmerta 'Foresighted' for "Mercurius," Nemetona 'Shriner' for "Mars," Damona 'Cow' for Borvo, Sirona 'Star' for Grannus, Nantosuelta 'River-?' for Sucellus, and so forth).

There were the taurine Tarvos Trigaranus and an antlered god Cernunnos 'Horned One', the latter sitting cross-legged in the inside of the Gundestrup cauldron with a ram-headed snake in his left hand and torques in his right hand and around his neck, surrounded by a wolf, a stag, and a bull. But females were more prominent among the theriomorphic type, though their animal nature tended to be mitigated: Epona 'Mare' (surnamed Regina 'Queen') had shed her presumable original hippomorphism and was depicted riding side-saddle, much like her Welsh saga version Rhiannon (< ^*Rigantō na 'Queen') in the Mabinogi. Like many Celtic deities, she was not free of an otherworldly tinge; horses appear to have been important to that place, as they were in Greek and Hittite myth (see chap. 8). Dea Artio was a bear-goddess (cf. the Greek Artemis), and Dea Arduinna rode on boarback. Cathubodua (Corpus Inscriptionum Latinarum 12.2571) is matched by the ornithomorphous Irish Bodb (catha) '[Battle] Crow' and such other bird-shaped battle-goddesses as Morrigan 'Great Queen'. A kind of animal totemism is also discernible (names like Boduognatus 'Crowson'), reaching over into Ireland (Ciichulainn [ciu 'dog'], Oisīn [oss 'stag']), although it need be no stranger than Leo, Wolf, or Bjorn as men's names in the Latin and Germanic orbits.

The bulk of the Gauls seem to have belonged linguistically to the branch later known as "British" (still surviving in Welsh and Breton), whereas the Irish (and Scottish) represent the "Gaelic" variety (the differences resemble those between the Oscan-Umbrian and Latin types of Italic, e.g., in the treatment of original labiovelars [qu sounds], which developed into labials [p sounds] in British and Oscan but were retained in Gaelic and Latin: Welsh pump, Oscan pompe, but Irish cōic, Latin quinque 'five'). Apart from names, the records of insular myth postdate Christianization. Ireland has the edge here for several reasons. It was a backwater of tradition, escaping Roman invasion and sheltered from Christianity until the mission of Saint Patrick (who died in 461). On the other hand, its monastic converts fast developed a zealous literacy that was pious and antiquarian-minded at the same time, leaving a record the Welsh material from the High Middle Ages (such as the Mabinogi) can only remotely emulate. The separate developments were conditioned by geographic isolation and linguistic split alike (Irish and Welsh were no longer mutually intelligible).

Ireland had its own "mythic geography," made up of the five cōiced 'fifths', comprising Ulster in the north, Connaught in the west, Munster in the south(west), and Leinster in the (south)east, all in relation to Mide 'Middle' (county Meath, the site of Tara, seat of the ard-rī 'high king'). This is an inherited structure with a clear parallel in ancient India, where the 'five tribes' (páñca krstáyah or carsanáyah) typically express the ārya or even human totality (páñca equaling viśve 'all'; cf. the Hittite pankus 'plenary assembly', and English fist> ^*pnk^Wstis 'a fivefold', namely the sum of the five fingers [< ^* pénk^we penk^wros]). Presumably in India, too, the ethnocentric seat of power and the four outlying cardinal divisions formed the basis of the quinary system (cf. the Chinese "Kingdom of the Middle"). The fivefold division of the community survives elsewhere, too, especially with tribes that have recently emerged from the obscurity of prehistory—for example, the five demes of Sparta or the Philistine pentapolis of Gaza, Askalon, Ashdod, Gath, and Ekron. The partition of the Irish year into quarters, punctuated by the great feasts of Samain (1 November, New Year, Day of the Dead, later All Saints' Day), Imbolc (1 February), Beltene (Mayday), and Lugnasad (1 August), may be at least proto-Celtic, for Samonios is attested also on the Gaulish "Calendar of Coligny"; in this system the "markers" fall in the interspaces of the solstices and equinoxes, thus in the "center" of each quarter.

Central to the Irish myth tradition are the epic reflections of the old Celtic gods, the Tūatha Dé Danann 'People of the Goddess Danu'. Traditions about them can be culled from various antiquarian texts, such as the Dindshenchas ('Place Traditions') or Cōir Anmann ('Fitness of Names', dealing with onomastic etymology), but especially in the Lebor Gabāla Ērenn ('Book of Conquests of Ireland'), a legendary history compiled in the twelfth century, and the Cath Maige Tuired 'Battle of Mag Tuired'. The monkish redactors have fitted the Tiūatha into a whole string of legendary invasions of Ireland, starting biblically with a certain Cessair, a granddaughter of Noah, and antedating the Flood. The only survivor of the flood was Cessair's husband Fintan, who led a subsequent ichthyo- or ornithomorphic open-ended life as an observer of Irish history. The next settler was Partholōn with his followers, a culture hero who cleared the land and instituted custom and also had a first conflict with the Fomoire, monstrous archenemies each with a single arm and leg who loom large in the sequel. Partholōn and his folk perished in a plague and were succeeded by Nemed and his people. After their leader's death the Fomoire imposed on them crushing tribute, leading to revolt, decimation, and emigration of the remnants to the south and the north. The descendants of the first group ultimately returned as the Fir Bolg, who are credited with the division of Ireland into cōiceds and a golden age reminiscent of Hesiodic and Avestan descriptions (soft showers, plentiful harvest, no falsehood). The northern emigrants, equally descended from Nemed's people, returned later as the Tūatha and defeated the Fir Bolg in the First Battle of Mag Tuired. Besides the mother figure Danu (or Anu), they comprised her brother Dagda 'Good God' (also known as Ollathair 'Allfather'), the latter's daughter Brigit, Ogma the champion, Goibniu the smith, Dian Cēcht the healer, and Nechtan (onomastic match of the Latin Neptūnus, whose story is elaborated in chap. 16). Their king was Nūadu, who lost his arm in the battle against the Fir Bolg. They brought along four magical objects: the spear of Lug, the sword of Nūadu, the cauldron of Dagda, and the Great Fāl. The first two were talismanic weapons that ensured victory, the cauldron guaranteed a full meal at all times, and the Great Fal would cry out whenever mounted by the true king of Ireland (matching in kind the Scottish Stone of Scone under the coronation chair of the British sovereign).

At this point the Fomoire reappear. They seem to be a shadow presence all along—the perennial foe and yet linked intimately to the society they antagonize; in chapter 7 we were reminded of them in the context of the ambivalent relations of Iran and Turan. When Nūadu abdicates the kingship because of his physical handicap, Bres ('Beautiful'; son of a Fomoire father and a Tūatha mother) is elected to succeed him on the urgings of the women; here one is reminded of the Etruscan period in Roman legendry down to the Republican War, when even Brutus the Liberator was (conversely) a Tarquin on his mother's side (even as in Iran Khosraw's mother was the daughter of the king of Turan). Bres turns out to be a poor choice, for he favors the Fomoire, forces menial tasks on the Tūatha champions, and lacks the generosity essential to a successful Irish king. Culinary abundance and other entertainment decline under his niggardly regime, and he draws the fatal ancient equivalent of a poor press, namely lampooning by the poets of the Tūatha. When his resignation is demanded, Bres with his twisted loyalties musters his Fomoire allies for assistance.

The stage is thus set for a showdown. The Tūatha regroup against their own failed king. Nūadu has a new prosthetic silver arm made by Dian Cecht and the latter's son Miach and is hence rehabilitated to resume the kingship with the epithet Argatlam 'Silver Hand'. At this point the young Lug arrives at Tara (he too was part Fomoire, daughter's son of Balor, the Fomoire champion) and impresses everyone with his range of expertise, to the point that Nūadu cedes his leadership to him. Seven years of preparations culminated in the great Second Battle of Mag Tuired. During the slaughter Dian Cecht and his three children used magic arts (singing spells over corpses thrown into a healing well) to revive the dead of the Tūatha, thus furthering their victory. Lug circumambulated the enemy host on one foot and with one eye closed, a magic circuit that mimed the single-leggedness of the foes in general and Balor's "evil eye" in particular. The latter shared with Ciichulainn and assorted Norse berserks the contortionist trick of sucking in one eye and expanding the other to a monstrous, paralyzing gaze, only in Balor's instance it took four men to raise the resulting heavy lid. At that instant Lug shot a sling stone through the opening and forced the eye backward through Balor's head, so that its evil was vented instead at the Fomoire themselves. Thus died Balor (having earlier slain Nūadu), and the Tuiatha triumphed. This marked the end of the Fomoire, while Bres was reprieved in return for becoming a governmental advisor in matters of agriculture. But the peace era of the Tūatha was not to last either, for a final invasion by the "Sons of Mil," that is, the Irish under their king Eremon, forced them to retire to the side, the fairy mounds, and thus perpetuate their influence from a subterranean spirit world. This is a unique tandem approach by levels to ethnogenesis and Christianization. The gods once held sway over the land, and in a sense they still do, despite yielding the surface of the Hibernian soil to the folk and to a new cult and being retired to what is sometimes called the "lower mythology." In subtlety the Irish solution outshines downright demonization of the onetime pagan gods in Christianized cultures.

Sound method requires that this material be confronted first with other Celtic, in this instance Welsh, traditions before resorting to extra-Celtic comparisons. Our main sources are the Four Branches of the Mabinogi and such other tales as Lludd and Lleuelys and Culhwch and Olwen. Far more than Irish tradition, this literature is caught up in the more cosmopolitan world of medieval romances and international folktales, nor did its authors have the same feeling for native myth that still guided the Irish compilers. The frequently fantastic narrative rarely yields systematic matter for comparison but can nevertheless be probed for incidental relevant lore.

Clear parallels to the Tūatha occur in the Welsh material. The Fourth Branch concerns the family of Don (cf. Danu) with her brother Math (cf. Dagda), ruler of Gwynedd (North Wales). Math could live only if his feet were held in a maiden's lap, except in time of war. Through the machinations of two sons of Don the incumbent foot holder was disqualified for loss of maidenhead. At length Math interviewed a new candidate, Don's daughter Aranrhod, who had to step over a magic wand as proof of her virginity, but in doing so dropped a male child (the sea-god Dylan) and a second object that was put in a chest. The latter turned out to be another son whom Aranrhod would not name until tricked to do so, and he became Lleu Llawgyffes 'Lleu the Deft-Handed', a match for the Irish Lug Samildanach 'Lug the Many-Skilled'. Aranrhod likewise swore to deny him weapons and spouse but was also induced unwittingly to countermand this ban. Lleu eventually became lord of Gwynedd. Gofannon 'Smith' (cf. Goibniu) and Amaethon 'Plowman' (both with the "augmentative" ^*-no- suffix) were further sons of Don. Amaethon is mentioned in Culhwch and Olwen, which is also the attested source for Lludd Llawereint, the Welsh match for Nūadu Argatlam (Lludd < Nudd by alliterative assimilation to the epithet Llaw-ereint, literally 'Hand [of] Silver'). The tale of Lludd and Lleuelys tells of Lludd, king of Britain (popular etymon of London, known in Welsh tradition as Caer Lludd 'Lludd-town'; cf. names like Ludgate), and his brother Lleuelys who had married his way to the throne of France. Three calamities afflicted Britain under Lludd: the advent of a people called the Coraniaid, armed with such advanced eavesdropping capabilities that no secrets were safe, a terrible dragon fight every Mayday, attended by a blood-curdling cry that debilitated people, animals, and nature at large, and an ongoing mysterious disappearance of provender from the royal larder. Brother Lleuelys (in whose name we recognize Lleu), "a man of great and wise counsel," supplied the remedy: a poison specifically calibrated to exterminate the Coraniaid, a means of talismanic entombment of the dragons for ever after, and the arrest of the magical purloiner of food and drink, who not only made restitution but reformed to become the king's faithful retainer (in modern times he might have become his castle security consultant). Thus Lleuelys helps Lludd overcome a tripartite set of afflictions, the kind we have seen from India to Iran to Rome, consisting of miscarriage of language or violation of verbal sanctity (botched formulas, lies, spells, oath breaking, eavesdropping), breach of communal peace (invasion, armed attack, mayhem, unsettling screams), and loss of physical resources (blight, plague, famine, theft). The regime of Bres in Ireland, relieved by the intervention of Lug taking over from Nuadu, was likewise characterized by breakdown of verbal trust (Bres plotting with the Fomoire against the Tiiatha), debilitation of warriors (Ogma reduced to carrying firewood), and shortage of victuals (unfair tribute of crops and produce to the Fomoire, the stingy foodways of Bres himself). The formula seems to be inversely replicated in the series of invasions, where Partholōn's and Nemed's people were decimated by plague and famine, the Fir Bolg and the Tūatha were conquered by force of arms, and according to druidic prophecy, the Sons of Mfl will perish by fire and water after a breakdown of law and social order. In the principal Old Irish legal corpus, the Senchus Mor, disease/famine, warfare, and breach of covenant are singled out as the threefold causes of calamity, even as in Plato's Republic (413b-14a) Socrates lists theft, spells, and violence as reasons for alienation from truth. Underlying the stories of Nūadu (genitive Niuadat) and Lug and that of Lludd and Lleuelys we may thus discern a Celtic myth of Lugus bringing relief to Nodons; the latter is attested in dedications from Lydney (cf. Lludd!) in Gloucestershire bordering South Wales (Deo Nodonti) and seems to mean 'Fisher' (cf. Gothic nuta 'fisherman', from ^* nuddn[s]), the probable ancestor of the Arthurian "Fisher King" of the Grail legend, whose maiming resulted in the Waste Land.

Other Branches of the Mabinogi are less amenable to comparison, but the mythic background is still palpable. In the First Branch, Pwyll, lord of Dyfed (South Wales), whose name means 'Wisdom', also mysteriously acquires condominium in the Otherworld (Annwn; cf. the Arthurian Avalon) before meeting and marrying the Lady on the White Horse, Rhiannon 'Queen' (who was compared above to the Gaulish horse-goddess Epona Regina). Their son Pryderi was kidnapped at birth and was found and reared by Teyrnon (< ^* Tigernonos 'Lord'), the events being linked with the birth of a horse. The Second Branch deals with the brood of Llyr 'Sea', notably the giant Brān ("The Blessed", Bendigeid-fran), Manawydan, and their sister Branwen, and the misadventures resulting from the latter's unhappy marriage to the king of Ireland. Here the Irish connection is felt in more than the plot, for the main characters match in name the Irish god Manannān Mac Lir ('Son of the Sea') and such Otherworld-farers as the hero of the "Voyage of Bran" (cf. Saint Brendan). Sea voyages and Otherworld notions are often interlinked in Celtic lore, implying Elysian isles or sunken paradises in the West (Mag Mell 'Field of Pleasure', Tir na n-ag 'Land of the Young', Land of Lyonnesse, City of Is, etc.), but in fairytale fashion the Otherworld can also be entered by a kind of on-the-spot enchantment—through the looking glass, so to speak—without any visible or distant boundaries. In the Second Branch battle between the Welsh and the Irish, the latter made good field use of their "Cauldron of Rebirth" to revive the casualties, a close parallel to Dian Cēcht's restorative well in the Second Battle of Mag Tuired. Only seven Welshmen survived, including Manawydan; Brān's severed head was buried as a talisman at London, facing the Continent, where it has warded off invasion of Britain at least since 1066. The Third Branch ties together the first two (Manawydan marrying Pwyll's widow Rhiannon, whose son Pryderi is now lord of Dyfed; together they have to fight off an evil spell cast on the land by a certain Llwyd), while in the Fourth Branch, in addition to the family saga of Dōn, Pryderi's death is told, tied to a gift of horses, a last leitmotif alluding back to the mythology of his mother Rhiannon. In this way the Mabinogi still yields vague echoes of Celtic myth, in confirmation of the Irish and even Gaulish material.

Essaying extra-Celtic comparisons for the insular tradition, we find the Irish pair Nuadu and Lug (or their Welsh counterparts Lludd and Lleu) homologous with Scaevola and Cocles of the Roman Republican War, marked by loss of a hand and one-eyed battle magic respectively. On the divine level this means comparability with the Germanic pair ^* T-Jwz and ^* Wooanaz, in Old Norse terms Tyr and Odin, the former having given up his right hand in order to bind the wolf, the latter sacrificing an eye to gain magical powers. The ascendancy of Lug over Nūadu parallels the eclipsing of Tyr by Odin in Norse tradition. Ogma seems to embody the warrior level, with his club matching Thor's hammer (and Feridun's gurz), while the third estate is reflected by Bres, the agricultural expert and popular favorite elected king by the women's vote. That he has to be put in his place by a violent conflict indicates that the Second Battle of Mag Tuired contains elements of the Indo-European myth of the "war of the divine classes" (Indra vs. Aśvins, Romans vs. Sabines, Asir vs. Vanir, discussed in chaps. 4, 9, and 11). Yet while the Fomoire in addition to Bres himself seem to be to a degree "Sabinic" (soil oriented, intermarrying with the Tiiatha, etc.), as a group they are not integrated by the conflict but rather are eliminated from the scene. Hence more than a divine skirmish and conciliation took place at Mag Tuired. The Fomoire also have a definite demonic tinge, that of a monstrous and oppressive archenemy who is yet too close for comfort and sooner or later requires a showdown with eschatological overtones. The Indic and Iranian gods versus demons, the Norse gods versus giants, the Roman patriots versus Tarquins, and such modern mutations as communism versus capitalism are all fair parallels (cf. the "end struggle" of the Internationale). Dian Cecht's resurrective battlefield well has its analogue in ancient India, where Kavya Uganas, serving as sorcerer of the demons, kept reviving the dead in the ongoing conflict against the gods. But while Ragnarök and communism hold out a postconflict millennium, India eschewed eschatology, contenting itself with a mere epic resolution in the Mahābhārata (see chap. 5). In Ireland too the epicized triumph of the Tūatha proves transitory. Even as the Pāndavas move on to posthumous heaven, the Tiiatha are retired to the mounds in favor of a new set of men. But Ireland is not yet through with mythological surprises. The king of the Sons of Mil, Eremon, is etymologically the equivalent of the Gaulish Ariomanus, reflecting the same personified ^* aryomn 'Aryanness' as is seen in the Vedic Aryaman and the Iranian Airyaman. In addition, very specific traits connect Eremon with both of the latter. The dossier of Eremon in the Lebor Gabála involves his role as builder of causeways and royal roads. In the Historia Britonum of Nennius, the Book of Leinster, the Book of Lecan, and some other sources, Eremon arranged a protection against poisoned enemy arrows that consisted of pouring cow's milk into furrows on the battlefield. He also provided wives to his allies and arranged for hereditary succession in favor of the Irish, his own people. All three features are distinctly "Aryamanic" in Indo-Iranian terms: Aryaman is connected with roads and pathways from the Rig-Veda onward. Airyaman invented the gaomaeza ritual of decontamination and healing, consisting of filling furrows with bovine excretions, specifically urine. Aryaman, Airyaman, and Vidura (the epic transposition of Aryaman; see chap. 5) are all connected with marriage rituals. The Indo-Iranian ^* Aryaman was clearly a satellite of ^* Mitra, a kind of hypostasis of the Mitra type proper, an abstraction expressing the selfsense of the community and championing the operation of communal welfare and health, especially in terms of marital compacts and rituals of healing. In Celtic tradition, a linear downward projection has replaced the "timeless" level of coexistence of figures such as Mitra and Aryaman. Eremon is in fact a diachronic hypostasis of Nūadu as king of the "next" layer of settlers, appropriately closer to mankind proper, even as Aryaman was in contrast to Mitra. Thus Celtic theology seems to have perpetuated, at the far end of the Indo-European continuum, a match for the Mitra-Aryaman structure at the other extremity—a notable instance of the "archaism of the fringe."

The perceptive mythological ear may pick up further Celtic: Indic echoes. Aranrhod's triple withholding of name, weapons, and spouse from the young Lleu re-calls Devavrata's renunciation of kingship and marriage in the Mahābhārata in return for a new name (Bhīsma; see chap. 5); since Bhīsma incarnates Dyaus, something of the Indo-European sky-god ^* Dyews may still lurk in the Celtic Lugus-Lug-Lleu (he was worshiped on mountaintops, for example, at the Lugnasad in Ireland, and with the colossal statue erected by the Gaulish Arverni on the peak of Puy de Dome in the Massif Central range in the Auvergne). Students of ancient India are reminded of the Vedic three-stepping world measurer Visnu in the triple-leaping Irish Saint Moling, and of the all-encompassing Krsna-Visnu of the Bhagavadgītā when the bard of the Sons of Mīl, Amairgin, takes possession of the soil of Ireland with a lay in which he equates everything with himself. The death-god Donn, first of the Tulatha to die and the post-mortem receiver of mankind, has traits in common with the Vedic Yama. The nine or so forms of marriage listed in Irish and Welsh traditions resemble the Laws of Manu by several varieties of family-centered and wealth-related "honorable" union at the beginning and stealthy rape or "mockery" at the end, separated by free-will and abduction marriages closely resembling the Indic gandharva and raksasa; in this instance the Indic and Celtic multiplicity contrasts with the reductionistic tripartite stylization seen in Roman law (chap. 9).

In the Lebor Gabála the female figures (Danu, Brigit) are in the background; this is a man's world, perhaps owing to skewing by the monkish compilers. Important women are more to the fore in the Welsh tales (Modron, Rhiannon, Aranrhod), but they hardly do full justice to the notable goddesses of the Gaulish pantheon either. To find proper insular reflections of the female component in Celtic myth one must turn to the Irish sagas and to folklore. Danu-Anu, Brigit, Eriu, Boand, the three Machas, Bodb, Morrigan have flitted past in the pages above. They, augmented by Medb, reflect aspects of the Indo-European transfunctional goddess, but with specific Celtic emphases. Sacrality, "untaintedness" is not much in evidence. Instead there is embodiment of "sovereignty" in the secular sense, and nary a virgin in sight. There is hypertrophy of the "amazonal" and fertility components, but these have often been tucked under the umbrella of "sovereignty" to yield domineering warrior-harridans and fountains of sexuality alike. Some figures were a bit "provincial" (B and local to Ulster, Medb in Connacht with a "double" in Leinster, Brigit in Leinster, Anu in Munster with her two breasts embodied by the twin hills of county Kerry known as Da Chich Anann 'Paps of Anu'; see fig. 9), but names like Medb (< ^* Medui, Bodb (< ^* Bodua), and Brigit (< ^*Brigantia) are paralleled in Gaul and Britain; they all reflect a pan-Celtic type. Those that have more "myth" to their name, such as Medb, the Machas, and Boand, will have their stories told in chapters 14, 15, and 16.

The world of Irish saga in its imaginative and anarchic fluidity is even harder to explore for comparative mythology than the lucubrations of the "Mythological Cycle" centered on the Lebor Gabála. But while the latter may with reservations be called a "chronicleepic" on the lines of Livy, the sagas of the "Ulster Cycle," with their alternating prose narrative and direct-speech verse, and the story and ballad versions of the "Fenian Cycle" can be characterized as remnants of heroic epic art. Folktale motifs proliferate, stock themes abound,² and yet the sagas have an ancient flavor that makes one suspect their core of a mythical residue. What they lack in direct mythological concern is compensated for by sheer untouched native archaism and by the absence or mitigation of any clerical distortions.

The "Ulster Cycle" purports to be "tribal," involving the Ulaid (Ulstermen) with their king Conchobor Mac Nessa residing at Emain Macha ('Twins of Macha'), near the later Christian capital of Ard Macha ('Hill of Macha', Armagh). It is a reflection of Celtic Iron Age warrior society, untainted by Roman, let alone Christian, influences. It bears notable resemblance to the heroic late Bronze Age traditions depicted in the Homeric poems. Conchobor's courtiers included the druid Cathbad, the conciliatory sage Sencha, the oversexed champion Fergus Mac Roich with his outsized phallus, and the trickster and troublemaker Bricriu. The old king grew himself a young wife, Deirdre ("of the sorrows"), whose elopement with Noise brought death to Noise and his brothers and desperate suicide upon herself; the story is of a Celtic triangle type also seen in the Fenian (Finn: Grāinne: Finn's nephew Diarmaid) and Arthurian traditions (Mark: Yseult: Mark's nephew Tristan; Arthur: Guinevere: Lancelot). Underlying it is the willful, strong female divine figure of Celtic myth, epitomized in saga form by Queen Medb, who takes and discards mates at will but whose power play no longer matches her sex drive in courtly intrigue, so that she trails "fate" and disaster in her wake. The chief driving forces in this company were aristocratic honor and the craving for fame, and the main pastimes included feasting, feuding for primacy over the "hero's portion," and intertribal fighting, especially with Connacht under Medb over a great bull whom the latter coveted (Tāin Bō Cuailnge 'Cattle Raid of Cooley'). Into this circle was born Conchobor's nephew Sētanta, sired by Lug and synchronized with the birth of twin foals (cf. Pryderi in the Mabinogi). Tutored and trained by the best teachers (Amairgin, Cathbad, Sencha, Fergus), Sētanta took on the name Cūchulainn ('Culann's dog') in atonement for one of his "boyhood deeds" (killing the hound of the smith Culann). Like Achilles, he alacritously chose a short life of fame and in short order developed into a full-blown berserk, complete with contortions (riastrad: eye trick, mouth stretched from ear to ear, hair standing on end as if by static electricity, magic halo [lūan lāith 'warrior's moon'] over his head), but also subject to the multiple geis or tabu that shadows the Irish hero. He killed a murderous triple adversary (three sons of Nechta Scēne) but thereupon in his frenzy attacked Emain Macha itself (shades of Horatius slaying the three Curiatii then turning on his own sister). Conchobor confronted him not with champions but with a phalanx of naked women; when this vanguard embarrassed his youthful modesty, the men grabbed him and cooled him off in successive tubs of cold water (one burst, one boiled, the third merely warmed). Now a full-fledged champion, he had to ward off single-handed Medb's cattle raid into Ulster, until the other men could recover from their annual "childbed sickness" (a couvadelike travesty of travail induced by a curse of the Third Macha; cf. chap. 15). Later he enrolled for some adult education from the divine instructresses Scāthach 'Modest', her daughter Uathach 'Terrible', and Aife 'Fair', who sound suspiciously trifunctional on the pattern of the Greek Euphrosyne, Aglaia, and Thaleia (it must be an ancient set, for 'modesty' is not otherwise a notable component of Celtic transfunctional triads). Uathach advised Cūchulainn to wring from her mother Scāthach at swordpoint a triple set of boons: instruction in warfare, Uathach's hand in wedlock, and a foretelling of his future, for Scāthach was also a prophetess. Instead the hero ended up begetting upon Aife a son Conlae, whom he later unwittingly slew in combat, in enactment of the Rustam: Sohrab pattern of filicide (cf. chap. 7). His own end came by being caught between two geasa when his enemies served up roast dog on his passage: he could neither pass up a meal nor eat dog, which might be his totem. A statue of him collapsing, with the death crow perched on his shoulder, adorns the main post office in Dublin. His head was retrieved by his brother-in-arms, Conall Cernach, chiming with the well-documented Celtic cult of human heads (cf., e.g., Brān's head buried at London).

Even this quick glance at some salient points of the "Ulster Cycle" points up the scattered but palpable mythical elements in the heroic narrative. The same is true of the "Fenian Cycle," but with a difference. It is not "localized," though mostly southern (Leinster and Munster), not centered at some heroic court; in fact it is almost extrasocietal with its free-roaming band (fiana) of fighters and hunters beholden only to themselves, an epic-romantic Irish body of tradition about ancient initiational warrior brotherhoods still marked by a free-will autonomy resembling that of Indra and the Maruts. First codified in the Acallam na Senōrach 'Parley of Oldsters' in the twelfth century and fueled by James Macpherson's Ossianic fancies in the eighteenth, it has proved durable with storytellers and balladeers until modern times, especially in the southwest (Munster). Much nature lyricism enters into the epic material, Finn-Fingal himself (with his son Oisīn-Ossian and grandson Oscar) being not only a warrior but a seer (fili) and a bard (fāith, cognate with Latin vates). He had his "boyhood deeds" like Cūchulainn, and like Conchobor he suffered marital infidelity at the hands of his willful wife and his nephew Diarmaid, the fellow with the irresistible 'love spot' (ball seirc); when the latter was mortally gored during a boar hunt in Sligo, Finn begrudged him his magic healing waters, thus declining to save him (even as Kay Us withheld his elixir from the dying Sohrab in the Shāh Nāma). Despite the folktale aura, Finn's divine origins are probable (^* Vindos 'White', cf. the Welsh Gwyn son of Nudd as a magical hunter and fighter, and theophorous continental place-names like Vindobona 'Vienna'); he may in fact be an allonymous saga version of Lug, and so may Cūchulainn, the "son of Lug." Lugus-Lug-Lleu is, as we have seen, homologous with the Germanic ^* Wōðanaz, the patron of the Germanic warrior bands, who had part demonic (giant) ancestry, even as Lug was part Fomoire. It would be nice to report that ^* Lugos means 'Bright', as is sometimes alleged, beside ^* Vindos 'White', but more probably it matches Gaulish loûgos 'raven and thus provides another analogue to ^* Wōðanaz with his ravens. But the intra-Irish similarities of Lug and Finn are themselves patent; just as Lug the spear wielder killed Balor with his evil eye, the eight-year-old Finn with the help of a magic spear vanquished a one-eyed fire-breathing marauder who would burn down Tara on every Samain; thus he secured his position within the brotherhood, but he had to deal even later with an antagonist Goll ('One-Eyed') Mac Morna (also called Aed 'Fire'), who before his birth had killed his father Cumaill. Lug was of consummate skill, while Finn had his "thumb of knowledge" that imparted preternatural insight (even as Odin had his magic of runes). Altogether Finn strengthens the case for a Celtic magician-god ^* Lugos, in many respects (part-demonic ancestry, secret knowledge, one-eyedness, magical warrior character, perhaps raven connection) similar to the Germanic ^* Wōðanaz; his saga reflections are not only Lug and Lleu but Cūichulainn and Finn, and even Arthur (the boar hunter of Culhwch and Olwen), who resembles Finn in many respects.

The "Historical Cycles" deal with royal persons who sometimes interact with figures such as Medb (e.g., Ailill, Lugaid, Conn, Art, and Cormac); these are dealt with in chapter 14. In other instances, such as Suibhne Geilt 'Sweeney the Madman', the saga itself is of direct relevance to Indo-European warrior myth (see chap. 13).

Notes

¹ That is, the Artemis Taurica of the Crimea, to whom strangers were sacrificed, as in Euripides' play Iphigenia in Tauris.

² Some of these have parallels in India, such as the magically potent "act of truth" in Irish lore and the analogous satyakriya, resorted to by Damayanti and Slta (for the latter, see chap. 5), or even the romantic commonplace of the "love sight unseen" (sercc ēcmaise: adrstakema) afflicting the likes of Nala and Damayanti, Oengus, and Findabair (in the Tain Bo Froich).