SOURCE: “Orpheus and Orfeo: The Dead and the Taken,” Mediem Aevum, Vol. XXXIII, No. 2, 1964, pp. 102-11.

[In the following essay, Allen argues that fairy tradition is central to Sir Orfeo and that, because the tale originated from the substitution of fairy elements in place of Greek myths, the search for its “elusive Celtic source” should be abandoned.]

The genius of the poet of Sir Orfeo lies not in creation but in adaptation, in refashioning for his own purposes material which he shares with countless generations of popular story-tellers. In its chivalric splendour and in its concern with love, honour and loyalty, the poem mirrors his tastes and preoccupations and those of his age, but for its rich and curious detail and for its very subject it draws upon beliefs which were to live on almost unchanged for centuries in the more distant parts of the British Isles.¹ All this is, of course, common knowledge. What is less generally realized is that the fairy faith might include more than the opinion that there existed a dreaded but picturesque race of immortal beings. Inseparable from it were ideas about man himself—about the relationship of soul and body, about the nature of death and immortality—which to us are so alien as to be almost incomprehensible; and it is from them, and not from its author, that our poem derives contradictions and absurdities which must have troubled many readers. Heurodis is carried off by the fairy host while to her watching attendants she appears to be peacefully asleep in her own orchard; her going, nevertheless, is presented not as a dream, but as an objective event:

As ich lay þis vndertide
And slepe vnder our orchardside
Þer come to me to fair kniȝtes
Wele y-armed al to riȝtes,
And bad me comen an heiȝing
And speke wiþ her lord þe king:
And ich answerd at wordes bold,
Y no durst nouȝt, no y nold.
Þai pricked oȝain as þai miȝt drive.
Þo com her king, also blive …
Wold ich, nold ich, he me nam,
And made me wiþ him ride …
And seþþen me brouȝt oȝain hom
Into our owhen orchard …

(ll. 133-163)

Ought we to see in the supposed inconsistency a deliberate heightening of the vividness of Heurodis’ words?² So subtle an explanation is, I think, unnecessary. In modern tales of the sidhe many of the actors consort with the fairy host while to their families they seem to be safe in their beds—a survival of the wide-spread primitive conception that during life, and especially during sleep or unconsciousness, the ‘soul’ or ‘self’ is able to wander from the body on adventures of its own.³ The poet has been neither careless nor ingenious: he has simply preserved a feature which is as traditional as most of the rest of his material.

But for the study of Sir Orfeo the interest of this less familiar side of the fairy creed extends beyond its usefulness in resolving isolated difficulties. Behind one neglected and misunderstood episode there lies a belief which, as I hope to show, suggests a new approach to a more general and more important problem: the origin of the entire poem. It is this episode, and the wider implications of what it contains, that I propose to consider. The scene is in the castle of the fairy king:

Þan he (Orfeo) gan bihold about al
And seiȝe liggeand wiþin þe wal
Of folk þat were þider y-brouȝt
And þouȝt dede, and nare nouȝt.
Sum stode wiþouten hade,
And sum non armes nade,
And sum þurch þe body hadde wounde
And sum lay wode, y-bounde,
And sum armed on hors sete,
And sum astrangled as þai ete,
And sum were in water adreynt,
And sum wiþ fire al forschreynt.
Wiues þer lay on child-bedde,
Sum ded and sum awedde,
And wonder fele þer lay bisides,
Riȝt as þai slepe her undertides.
Eche was þus in þis warld y-nome,
Wiþ fairi þider y-come.

(ll. 387-404)

Against the glittering magnificence of the background nothing could be more hideously unexpected than this assembly of maimed and suffering figures. Their presence is usually dismissed as a distorted and incongruous reminiscence of the classical Hades,⁴ but this temptingly obvious explanation ignores one inconvenient fact: that in ll. 389-90 the poem states explicitly that they are not dead:

of folk þat were þider y-brouȝt
And þouȝt dede, and nare nouȝt.

How can the decapitated and the asphyxiated, the burnt and the drowned be said still to live? To us these lines may seem at the worst nonsensical, at the best a clumsy attempt to reconcile the irreconcilable mythologies of Hades and fairyland;⁵ but less than a hundred years ago an Irish or a Scottish countryman would have recognized in them the echo of his own convictions:

Very few die at all, most are taken.
When a man dies, he does not die at all, but the daoine maithe take him away.
No-one dies, but the daoine maithe
take him away, and leave something else in his place.
Not one in twenty dies a true death, they all pass into another life.(6)

These speakers believed that many, perhaps most, of those who were thought to die were in reality no more dead than Orfeo's stolen Heurodis. In their last agony they too had been carried off, body and soul, by the triumphant sidhe, and a lifeless changeling, a cunningly fashioned image of wood or straw, left in the place of each.⁷ Inside the hollow hill the ‘dead’ and the taken endured the same captivity, and tale after tale demonstrated that both could be won back in the same ways, and with the same ease.⁸ Certainly fairyland was not the home of the dead, the Celtic Hades, as it has been so often and so misleadingly called: with or without a semblance of dying, its mortal inhabitants had passed unchanged, in earthly flesh and blood, from one world to another.⁹

Inside the fairy castle Heurodis and her companions remain exactly as they were at the moment when they were abducted, stretched in sleep, or frozen in grotesque attitudes of apparent death. The catalogue of 11. 391 ff. is only too likely to have been varied, elaborated and deformed by any reciter or scribe into whose hands it has happened to fall (in the Ashmole manuscript it is among the most mangled portions of the whole text); the greater part of it, nevertheless, is still explicable in terms of fairy superstition. One group of captives lie ‘wode, y-bounde’ (1. 394): similarly, in the Ireland of the nineteenth century, delirium was still looked upon as an infallible sign that the sick man's spirit (or, more accurately, his taise) was already with the daoine maithe, who would soon, at his death, claim his body as well.¹⁰ With the rest, a different principle is at work. Their lives have been cut short suddenly, by violence, accident or misadventure; and here we may discern a reflection of the assurance which, for later generations, lightened the peculiar pathos of untimely and unexpected death: that so to die was not to perish, but to join the fairy host.¹¹ For some parts of the list a more specific justification can also be found. Those who died in child-birth and those who died by drowning were invariably considered to be taken,¹² and are placed in fairyland also by our poem (11. 397 and 399). The mutilated of 11. 391-2 may have suffered the bloody violence with which, earlier in the poem, Heurodis herself is threatened;¹³ the wounded of 1. 393—and possibly the horsemen of 1. 395—may have met the same fate as Thomas Reid, who died in the Battle of Pinkye, but whom Bessie Dunlop saw among the sidhe.¹⁴ For the rest of the catalogue (11. 396 and 398) there are to my knowledge no exact parallels, but since almost any unnatural death was attributed to the good people, it is impossible to say whether we owe these remaining lines to a single narrator's whim or to authentic popular tradition.¹⁵

If my arguments have so far depended almost solely upon modern evidence, my excuse must be that the early material at our disposal is inevitably limited and scattered. Until the nineteenth century a belief which was essentially the property of the illiterate stood little chance of surviving in any written record. Normally it would find expression not in statement but in narrative, in tales of the unmasking of changelings, or of the recovery from the sid of those mourned as dead, but until folklore became a respectable study, these unpolished accounts of supposedly everyday events would be preserved only by accident, or through the curiosity or credulity of a few educated men. The proceedings of the witchcraft trials and the Secret Commonvealth of Robert Kirk bear witness that many Scots of the sixteenth and seventeenth centuries were persuaded that in fairyland the dead still lived,¹⁶ while earlier, and well before the date of our poem, the same conviction may be traced in a small number of stories which have been handed down to us by Latin writers. In recent times this belief was both distinctively Celtic and distinctively Gaelic, but at one time it was probably current also in Brittany.¹⁷ Of the tales I am about to quote, one is undoubtedly Irish, and another is set in Britain the Less. The first comes from a description of the Wonders of Ireland which, round about the year 1076, Patrick, Bishop of Dublin, composed in halting Latin verse.¹⁸ In it the sidhe appear in Latin and Christian guise as daemones; otherwise, the passage speaks for itself with admirable clarity:

Haec res mira solet numeros celebrantibus addi.
Vir bonus et verax aliquid mirabile vidit.
Quondam namque die volucres in flumine cernens
Proitiens lapidem percussit vulnere cignum
Prendere quam cupiens tunc protinus ille cucurrit.
Sed properante viro mire est ibi femina visa,
Quam stupido viso aspiciens haec querit ab illa
Unde fuit, quid ei accidit, aut quo tempore venit.
Haec: infirma fuit, inquit ei, et tunc proxima morti
Atque putata meis sum, quod defuncta videbar,
Demonibus sed rapta fui cum carne repente.
Hanc vix crediblem rem tunc audivit ab illa,
Quam secum ducens saciavit veste ciboque
Tradidit atque suis credentibus esse sepultam
Qui quod erat factum vix credere iam potuerunt.

The history of the miles Britanniæ Minoris from Walter Map's De Nugis Curialium is already well-known to students of Sir Orfeo, but for the wrong reasons. Its confusion of the dead and the fatæ has won it fame as the ancestor of the mediæval poem, as a ‘Celtic tale which had already been contaminated with the classical legend of Orpheus’.¹⁹ A glance at the passage in question will show how ill-founded this theory is. In Dist. II xiii Map gives a brief summary:

… miles quidam uxorem suam sepellisse reuera mortuam, et a chorea redibuisse raptam, et postmodum ex eo filios et nepotes suscepisse, et perdurare sobolem in diem istum, et eos qui traxerunt inde originem in multitudinem factos, qui omnes ideo ‘filii mortuae’ dicuntur.

In Dist. IV viii the story is told in full:

Miles quidam Britanniæ Minoris uxorem suam amissam diuque ploratam a morte sua in magno feminarum cetu de nocte reperit in conualle solitudinis amplissime. Miratur et metuit, et cum rediuiuam uideat quam sepelierat, non credit oculis, dubius quid a fatis agatur. Certo proponit animo rapere, ut de rapta uere gaudeat, si uere uidet, uel a fantasmate fallatur, ne possit a desistendo timidiatis argui. Rapit eam igitur, et gauisus est eius per multos annos coniugio, tam ioconde, tam celebriter, ut prioribus …

This tale resembles our poem only in the common-place motifs of the fairy dance and the company of women, and for it no more complicated explanation is needed than that which Hartland²⁰ gave long ago: that Map had, not surprisingly, misunderstood an account of the recovery from the sidhe of a woman who, like so many others, was not dead, but taken.

Our remaining examples are provided, more unexpectedly, by Petronius Arbiter and Thomas of Chantimpré, and are set, disconcertingly enough, in Italy and in Flanders. In every other respect, however, they might be from the lips of a Gaelic story-teller; and in them we must have tales of Celtic origin with which, in ways too fortuitous to be discovered, Italian and Flemish narrators have become acquainted, and to which they have given a local colouring. In Chap. 63 of the Cena Trimalchionis Petronius reproduces a changeling story. In it the sidhe (or, in their pagan Latin dress, the strigæ) replace the corpse by an image some time after the victim's apparent death, just as in modern tales the exchange may be postponed until the wake, or the progress to the church yard:²¹

Cum adhuc capillatus essem, nam a puero vitam Chiam gessi, ipsimi nostri delicatus decessit, mehercules margaritum, caccitus et omnium numerum. Cum ergo illum mater misella plangeret, et nos tum plures in tristimonio essemus, subito strigae coeperunt; putares canem leporem persequi. Habebamus tunc hominem Cappadocem, longum, valde audaculum et qui valebat: poterat bovem iratum tollere. Hic audacter stricto gladio extra ostium procucurrit, involuta sinistra manu curiose, et mulierem tanquam hoc loco—salvum sit, quod tango—mediam traiecit. Audimus gemitum, et (plane non mentiar) ipsas non vidimus. Baro autem noster introversus se proiecit in lectum, et corpus totum lividum habebat quasi flagellis caesus, quia scilicet illum tetigerat mala manus. Nos cluso ostio redimus iterum ad officium, sed dum mater amplexerat corpus filii sui, tangit et videt manuciolum de stramentis factum. Non cor habebat, non intestina, non quicquam: scilicet iam puerum strigae involaverant, et supposuerant stramenticium vavatonem.

In Chap. LVII of the second book of his Bonum Universale Thomas of Chantimpré enumerates the outrages of the Dusii, spirits who inhabit mountains and groves; and it is here that he inserts the tales which merit our attention.²² They have more pretensions to plot than any of the others we have examined: they combine the two ever-popular topics of the changeling and of the winning-back of the dead, and their climax is of a kind particularly favoured by later story-tellers, a dramatic disclosure at the wake. In ‘Guerthenae’ in Brabant a young girl dies …

… iuvenis amator puellæ de villa eadem in crepusculo noctis pergebat ad aliam, et dum per dumeta transiret, audivit vocem quasi feminæ lamentantis, Sollicitus ergo discurrens et quærens, auditam invenit puellam quam mortuam æstimabat, cui et dixit: ‘mortuam te plangunt tui, et huic unde venisti?’ ‘Ecce’, ait, ‘vir ante me vadit qui deducit me’. Stupefactus ad hoc iuvenis, cum neminem alium nisi solam puellam videret, audacter rapit eam, et in domo extra villam protinus occultavit.

He goes to the girl's father, and asks for her hand in marriage. Amazed, the father consents, if he can bring her back to life and health.

Mox iuvenis, cum relevasset linteum quo cooperta putabatur, figmentum mirabile, quale a nullo hominum fieri potuit, invenerunt. Dicitur autem ab illis qui figmenta huiusmodi diabolica inspexerunt, ea esse interius putrido ligno similia, levi exterius pellicula superducta. Hinc reducta puella est, et patri reddita, sanaque post dies aliquot dictum iuvenem maritum accept, et usque ad tempora nostra incolumis perduravit.

The next section contains an almost identical story:

Simili prope modo, cum quidam in confinio Flandriæ sororem languidam et mortuam putatam, sub eadem die, antequam sepeliretur figmentum, inter arundineta iuxta maris littora reperisset, reduxit eam ad propria, et ingressus domum ubi ab amicis quasi mortua plangebatur, discoopertum figmentum extracto gladio in frusta conscidit, horrentibus cunctis atque clamantibus, cur in funus sororis tanta crudelitate sæviret. Et mox subridens, ‘crudelis’ inquit ‘videor in sororem, non est illud sororis corpus, sed figmentum et illusio dæmonum’. Et hæc dicens secum cunctos accepit, et duxit ad domum propriam, et eis sororem reductam ostendit. Et hæc usque ad tempora nostra permansit.

The conclusions which emerge from these tales are both unambiguous and reassuring. They are centuries older than the main body of our evidence, yet in them we find the same endlessly repeated events used to clothe the same constant articles of faith: the survival of the dead, the hope of their return, and (since the victims include a woman of child-bearing age, a little boy, and an unmarried girl) the association of abduction and untimely death. There could be no clearer confirmation of much of my interpretation, and no better illustration of the unchangingness of popular tradition and of popular belief.

About the origin of Sir Orfeo all that is certain is that in it we have a remote and Celtic descendant of the legend of Orpheus and Eurydice; beyond that, everything is a matter of speculation, particularly the exact nature of the imaginative process by which that legend was remoulded. (From the remoulding of the legend itself must, of course, be distinguished the accretion to it, in its new form, of motifs from fairy superstition, which is not here our concern.) Since Kittredge's time most scholars have followed him in assuming that the events of the classical myth were combined by an Irish or a Breton story-teller with those of an already current native tale; but none known to us, whether the celebrated Tochmarc Etaine or any other, is in fact sufficiently like the mediæval poem to be justly regarded as its Celtic parent.²³ How then did the story of Orpheus become the story of Orfeo? To that question I should like to propose an answer which, if it is accepted, will enable us to abandon for ever the search for Sir Orfeo's elusive Celtic source. The poem as we know it owes its existence to an act of substitution, the substitution of taking for death and of the sid for Hades: by that change the myth enters into fairy tradition, and upon it the presence of all Sir Orfeo's strange and marvellous detail depends. The reasons for that substitution must by now be quite evident. To a Gaelic or a Breton narrator the fable of Orpheus’ quest for his lost wife would seem at once familiar and oddly unfamiliar: the tales to which he was accustomed described the recovery not of the dead, but of the abducted, of those wrongly imagined to have died. In recounting his newly-acquired material, he would naturally restore to it what was for him its proper condition. Instead of a ‘blending’ or a ‘fusing’ (to borrow the terms usually employed) of two independent tales, there would be a reinterpretation of the ancient legend by one who believed, as his countrymen did for centuries, that death might be no more than an illusion and a deceit.

The resulting story may already have had the famous happy ending of Sir Orfeo, for both Euripides and Isocrates²⁴ allow Eurydice to be reunited with her husband for once and for all, and there are indications that this account may have survived into the Middle Ages, alongside the better-loved and more pathetic Vergilian version.²⁵ It would, however, disagree with the extant poem in one important particular: in it the heroine would appear to die. The eventual loss of this feature can be explained quite simply. In popular tradition tales of the rescue or the retum of the ‘dead’ and of the taken were both equally frequent, and the boundary between them was extremely ill-defined. The supposed death of the victim, which alone distinguished the first type of narrative from the second, did not always occupy a prominent position (in one of the tales in Wentz's collection it is added as a mere afterthought),²⁶ and might be present in some versions of a story and absent in others. J. M. Synge and Hans Hartmann both retell the same tale: every night a woman would come back from the sidhe to her home for food, and during these visits she disclosed to her husband that he might win her back by dragging her from her horse as the fairy procession rode past a certain spot. In Synge's version the heroine is believed to be dead and buried, in Hartmann's, there is no word of her ‘death’.²⁷ For those who held the fairy faith, it little mattered whether counterfeit death or undisguised abduction befell a man or woman. Against this background the change which concerns us ceases to be at all surprising. As our story passed from narrator to narrator, the ‘death’ of the hero's wife might easily come to receive less and less attention. In time it might, with equal ease, disappear altogether, and there would be left the straightforward account of rescue from the sid which forms the basis of Sir Orfeo.

The often-praised lightness and charm of Sir Orfeo are deceptive: the distinctive flavour of the poem comes not from them, but from a unique and haunting combination of fourteenth-century graces and immemorial fears. The world of the poem, for all its outward elegance, is still the primitive world of popular belief, a world in which men are forever surrounded and threatened by cruel and capricious beings:

‘Loke, dame, tomorwe þatow be
Riȝt here vnder þis ympe-tre,
And þan þou schalt wiþ us go
And live wiþ ous euermo.
And ȝif þou makest ys ylet
Whar þou be, þou worst yfet,
And totore þine limes al
Þat noþing help þe no schal,
And þei þou best so totorn
Zete þou worst wiþ ous yborn.’

(11. 165-174)

Upon this sense of danger and unease the poem relies for its strength, for in its subject it has none of the grave yet passionate dignity of the antique myth—to win back the dead requires a love strong enough to overcome the laws of nature, to win back the taken requires only cunning and resolution. The gulf which separates the two tales is best exemplified in the scenes which are enacted before Pluto and before the fairy king. By his music Orpheus forces the inexorable powers of death to feel pity for human grief and human longing:

Tandem ‘vincemur’ arbiter
Umbrarum miserans ait
‘Donamus comitem viro
Emptam carmine coniugem’.

(Boethius De Consolatione Philosophiæ III, xii)

In Sir Orfeo resource has taken the place of feeling, and the recovery of Heurodis turns upon a short battle of wits between her captor and her husband. The happy ending of the mediæval poem (if it is indeed an innovation) is only one manifestation of an all-pervasive difference in quality. In spirit the story of Orpheus and the story of Orfeo have very little in common, and between them lies a belief in which death itself loses its bitterness and finality, and is swallowed up in enchantment.

Notes

1. The poem is an early representative of the fairy abduction story which later is particularly characteristic of popular Celtic tradition, and of which a work such as Sean O'Suilleabhain's Handbook of Irish Folklore (Dublin 1942) will provide countless examples. The sidhe of Sir Orfeo are admirably typical of their kind in their love of fine clothes and white horses [ll. 145 ff. compare L. C. Wimberly Folklore in the English and Scottish Ballads (Chicago 1928) p. 188; J. G. Campbell Superstitions of the Highlands and Islands of Scotland (Glasgow 1900) p. 148; P. Kennedy Legendary Fictions of the Irish Celts (London 1860), p. 113], and in their human past times of hunting, dancing and fighting [ll. 282 ff., compare O'Suilleabhain, op. cit. p. 460 ff.; Wimberley, op. cit., p. 194 ff.; J. Crofton Croker Fairy Legends of the South of Ireland (London 1870) p. 218]. On the perils of sleeping outside on May Day see O'Suilleabhain, op. cit. p. 473, and Hans Hartmann Über Krankheit, Tod, und Jenseitsvorstellungen in Irland (Halle 1942) p. 139 ff.; on the perils of sleeping beneath a tree see F. J. Child English and Scottish Ballads (New York 1956) I 340, and II 505, and G. L. Kittredge American Journal of Philology VII 190. The hollow hill and the green plain of fairyland are traditional [see H. R. Patch The Other World (Harvard 1950) p. 46 ff.] as are the brilliance and splendour of the fairy castle [see Child, loc. cit.; Kennedy, op. cit., p. 116; Robert Kirk The Secret Commonwealth (Edinburgh 1815) p. 5]; but, despite the parallels adduced by A. J. Bliss in his edition of Sir Orfeo (Oxford 1954) p. xxxviii, its crystalline wall, pillars of gold, and glowing carbuncles do not necessarily derive from Celtic visions of the Other World, since they may be found in descriptions of any magnificent building, from the heavenly Jerusalem to the palace of Prester John. (See Patch, op. cit. p.149 f. and 203 f.)

2. As is suggested by Constance Davies MLR LVI 161.

3. See H. Hartmann op. cit., p. 118 and 147 ff. The Irish term for this wandering soul is taise, for which English has no equivalent. The taise is not a disembodied spirit: it is both visible and tangible, and is in all respects indistinguishable from the man himself.

4. See G. V. Smithers Mæ XXII 85 f., and Constance Davies, loc. cit. 164.

5. As is suggested by Constance Davies id.

6. Quoted from material in the possession of the Irish Folklore Commission, Dublin, by kind permission of Professor J. Delargy. These remarks were recorded in this century by collectors working in Gaelic-speaking Ireland.

7. See H. Hartmann op. cit., p. 168.

8. See H. Hartmann op. cit. p. 176.

9. On this point see H. Hartmann op. cit. p. 162 and 176, and Elisabeth Hartmann Die Trollvorstellungen in dem Sagen und Märchen der Skandinavischen Volker (Stuttgart-Berlin 1936) p. 86 f. In modern Scotland and Ireland the dead and the inhabitants of fairyland were, admittedly, sometimes confused, but this confusion is directly and explicitly contradicted in the remarks just quoted, and is explained by H. Hartmann (op. cit. p. 160 f.) as a corruption of the belief I have been describing. Some mediæval and renaissance authors identify Pluto and Proserpina with the fairy king and queen (Chaucer Merchant's Tale ll. 983 ff., Dunbar Golden Targe, ll. 124 f., Henryson Orpheus and Eurydice 11. 110 ff., T. Campion, in Elizabethan Lyrics ed. N. Ault (London 1949) p. 151): this identification probably depends mainly upon the learned equation of Diana in her multiplicity of aspects with the fairy queen on one hand and with Proserpina in the other. (See Prudentius Contra Symmachum I 367; Servius, comm. on Ecl. iii 27; Albericus I, Theogony II xii ii.) L. H. Loomis has demonstrated that Chaucer probably also had in mind Sir Orfeo itself (Studies in Philology XXXVIII 14 ff.); and in a later article I hope to show that Henryson too was influenced by our poem.

10. See H. Hartmann op. cit. p. 147.

11. See H. Hartmann loc. cit. and Reidar Th. Christiansen Folkliv 1938 p. 330 ff.

12. See H. Hartmann op. cit. p. 125 and 144. For the belief that the drowned are taken there is one very early scrap of evidence. In the Bodleian MS of the Dinnsenchas (Rawlinson B 506, ed. by Whitely Stokes in Folklore III 467 ff.) the scribe has added to the story of Tuag, who was drowned as she was being carried away to become Mannanan's bride, the note no comad e Mannanan fesin nodos-berid (or perhaps it was M. himself carrying her off). In making this addition, he may have been inspired by the conviction that the drowned had become the captives of the beings who ruled over the water where they had lost their lives.

13. For modern parallels see Bealoideas VII p. 85 f., and J. M. Synge The Aran Islands (Dublin 1911) p. 60.

14. See the trial of Bessie Dunlop for witchcraft in R. Pitcairn Ancient Criminal Trials of Scotland. (Edin. 1833) I 49 ff.

15. Death by fire and death by water conventionally go together: see Floris and Blancheflour 1. 383 ff. and Richard Coeur de Lion l. 1635 ff. The inclusion in the catalogue of the drowned (which certainly does come from popular superstition) may therefore have suggested to a narrator the addition of the burned to the list.

16. See Pitcairn op. cit. p. 49 ff. and 161 ff., and Kirk op. cit. p. 5 and 12.

17. See H. Hartmann op. cit. p. 161.

18. Ed. by Mommsen in Monumenta Germanica Historica, Auctorum Antiquissimorum XIII 218 ff. Patrick was almost certainly translating into Latin a vernacular list of marvels: see A. Gwynn The Writings of Bishop Patrick (Dublin 1955) p. 126 ff.

19. Bliss op. cit. p. xxxiii, and Smithers MÆ XXII 86 ff.

20. E. S. Hartland Science of Fairytales (London 1925) p. 343.

21. For the delay in the exchange compare H. Hartmann op. cit. p. 169 ff. In Chap. 38 of the Cena Trimalchionis Petronius refers to another tale of Celtic provenance: that of the leprechaun's cap and the crock of gold. For a discussion of both passages see E. Hartmann Zeitschrift für Folkskunde, neue Folge, Band 7 (1937) 308 ff.

22. Thomas of Chantimpré Bonum Universale de Apibus II lvii § 20 and 21, p. 552 f. in Douai ed. of 1605. For modern stories with a similar climax see H. Hartmann op. cit. p. 175 ff.

23. Kittredge loc. cit. p. 176-202; Smithers loc. cit. p. 86 ff.; J. Burke Severs Studies in Medieval Literature in Honour of A. C. Baugh (Philadelphia 1961) 194 ff. Bliss's rejection of Tochmarc Etaine as a source for Sir Orfeo (op. cit. p. xxxiii ff.) seems to me to be quite convincing. They resemble each other closely only in that in each the intended victim is warned, and a guard is set to prevent her abduction. A motif of this kind cannot be used to prove anything about the relationship between two tales, since it would appear that it was not proper to any one story. It is found elsewhere, in places as diverse as Map's De Nugis Curialium (see Dis. II, cap. xxix, where Map uses it to embroider a narrative taken from Pseudo-Turpin chap. 7), and a modern Irish story (see Bealoideas VII 85 f.), and was probably a story-teller's device which could be employed in any appropriate situation. The game of chess or cards (which also occurs in the Tochmarc Etaine) is a good example of the same kind of ‘free’ motif in Irish story-telling: see Miles Dillon, Irish Sagas (Dublin 1959) p. 19. With the relationship between Sir Orfeo and another suggested source, Map's tale of the miles Britanniae Minoris (R. S. Loomis, MLN LI 28 ff., C. Davies MLR XXXI 354 ff.) I have already dealt.

24. Euripides Alcestis 357 ff.; Isocrates (ed. F. Blass Leipzig 1898) Oratio xi 7.

25. A twelfth-century sequence speaks unambiguously and without qualification of Orpheus’ recovery of his bride: see G. M. Dreves Analecta Hymnica VIII (Leipzig 1890) 33. Two manuscript illustrations known to me show Orpheus and Eurydice walking away from the mouth of Hell together: see F. Saxl and H. Meier Catalogue of Astrological and Mythological MSS of the Latin Middle Ages III (London 1953) pl. xl, and J. van den Gheyn ed. Epitre D'Othea, 100 miniatures reproduced from the MS of Jean Mielot (Brussels 1913), pl. 70. P. Dronke discusses the whole problem thoroughly in Classica et Medievalia XXIII 198-215.

26. Y. E. Wentz The Fairy Faith in Celtic Countries (Rennes 1909) p. 66.

27. Synge op. cit., p. 210, Hartmann op. cit., p. 177. The two accounts differ in their degree of alaboration and in their endings, but are undoubtedly versions of the same story.