Sacred Texts  Sagas and Legends  Celtic  Index  Previous  Next 



In olde days of the King Artour,
Of which that Bretons spoken gret honour,
All was this lond fulfilled of faërie;
The elf-qrene with hir jolts companie
Danced full oft in many a grene mede.
ACCORDING to a well-known law of our nature, effects suggest causes; and another law, perhaps equally general, impels us to ascribe to the actual and efficient cause the attribute of intelligence. The mind of the deepest philosopher is thus acted upon equally with that of the peasant or the savage; the only difference lies in the nature of the intelligent cause at which they respectively stop. The one pursues the chain of cause and effect, and traces out its various links till he arrives at the great intelligent cause of all, however he may designate him; the other, when unusual phenomena excite his attention, ascribes their production to the immediate agency of some of the inferior beings recognised by his legendary creed.
The action of this latter principle must forcibly strike the minds of those who disdain not to bestow a portion of their attention on the popular legends and traditions of different countries. Every extraordinary appearance is found to have its extraordinary cause assigned; a cause always connected with the history or religion, ancient or modern, of the country, and not unfrequently varying with a change of faith. [a]
The noises and eruptions of Aetna and Stromboli were, in ancient times, ascribed to Typhon or Vulcan, and at this day the popular belief connects them with the infernal regions. The sounds resembling the clanking of chains, hammering of iron, and blowing of bellows, once to be heard in the island of Barrie, were made by the fiends whom Merlin had set to work to frame the wail of brass to surround Caermarthen. [b] The marks which natural causes have impressed on the solid and unyielding granite rock were produced, according to the popular creed, by the contact of the hero, the saint, or the god: masses of stone, resembling domestic implements in form, were the toys, or the corresponding implements of the heroes and giants of old. Grecian imagination ascribed to the galaxy or milky way an origin in the teeming breast of the queen of heaven: marks appeared in the petals of flowers on the occasion of a youth's or a hero's untimely death: the rose derived its present hue from the blood of Venus, as she hurried barefoot through the woods and lawns; while the professors of Islam, less fancifully, refer the origin of this flower to the moisture that exuded from the sacred person of their prophet. Under a purer form of religion, the cruciform stripes which mark the back and shoulders of the patient ass first appeared, according to the popular tradition, when the Son of God condescended to enter the Holy City, mounted on that animal; and a fish only to be found in the sea [the Haddock] stills bears the impress of the finger and thumb of the apostle, who drew him out of the waters of Lake Tiberias to take the tribute-money that lay in his mouth. The repetition of the voice among the hills is, in Norway and Sweden, ascribed to the Dwarfs mocking the human speaker, while the more elegant fancy of Greece gave birth to Echo, a nymph who pined for love, and who still fondly repeats the accents that she hears. The magic scenery occasionally presented on the waters of the Straits of Messina is produced by the power of the Fata Morgana; the gossamers that float through the haze of an autumnal morning, are woven, by the ingenious dwarfs; the verdant circlets in the mead are traced beneath the light steps of the dancing elves; and St. Cuthbert forges and fashions the beads that bear his name, and lie scattered along the shore of Lindisfarne. [c]
In accordance with these laws, we find in most countries a popular belief in different classes of beings distinct from men, and from the higher orders of divinities. These beings are usually believed to inhabit, in the caverns of earth, or the depths of the waters, a region of their own. They generally excel mankind in power and in knowledge, and like them are subject to the inevitable laws of death, though after a more prolonged period of existence.
How these classes were first called into existence it is not easy to say; but if as some assert, all the ancient systems of heathen religion were devised by philosophers for the instruction of rude tribes by appeals to their senses, we might suppose that the minds which peopled the skies with their thousands and tens of thousands of divinities gave birth also to the inhabitants of the field and flood, and that the numerous tales of their exploits and adventures are the production of poetic fiction or rude invention. It may further be observed, that not unfrequently a change of religious faith has invested with dark and malignant attributes beings once the objects of love, confidence, and veneration. [d]
It is not our intention in the following pages to treat of the awful or lovely deities of Olympus, Valhalla, or Meru. Our subject is less aspiring; and we confine ourselves to those beings who are our fellow-inhabitants of earth, whose manners we aim to describe, and whose deeds we propose to record. We write of FAIRIES, FAYS, ELVES, aut alio quo nomine gaudent.
Like every other word in extensive use, whose derivation is not historically certain, the word Fairy has obtained various and opposite etymons. Meyric Casaubon, and those who like him deduce everything from a classic source, however unlikely, derive Fairy from Φήρ, a Homeric name of the Centaurs; [e] or think that fée, whence Fairy, is the last syllable of nympha. Sir W. Ouseley derives it from the Hebrew לאפ (peer), to adorn; Skinner, from the Anglo-Saxon ranan, to fare, to go; others from Feres, companions, or think that Fairy-folk is quasi Fair-folk. Finally, it has been queried if it be not Celtic. [f]
But no theory is so plausible, or is supported by such names, as that which deduces the English Fairy from the Persian Peri. It is said that the Paynim foe, whom the warriors of the Cross encountered in Palestine, spoke only Arabic; the alphabet of which language, it is well known, possesses no p, and therefore organically substitutes an f in such foreign words as contain the former letter; consequently Peri became, in the mouth of an Arab, Feri, whence the crusaders and pilgrims, who carried back to Europe the marvellous tales of Asia, introduced into the West the Arabo-Persian word Fairy. It is further added, that the Morgain or Morgana, so celebrated in old romance, is Merjan Peri, equally celebrated all over the East.
All that is wanting to this so very plausible theory is something like proof, and some slight agreement with the ordinary rules of etymology. Had Feërie, or Fairy, originally signified the individual in the French and English, the only languages in which the word occurs, we might feel disposed to acquiesce in it. But they do not: and even if they did, how should we deduce from them the Italian Fata, and the Spanish Fads or Hada, (words which unquestionably stand for the same imaginary being,) unless on the principle by which Menage must have deduced Lutin from Lemur--the first letter being the same in both? As to the fair Merjan Peri (D'Herbelot calls her Merjan Banou--D'Herbelot titre Mergian says, "C'est du nom de cette Fée quo nos anciens romans ont formé celui de Morgante la Déconnue." He here confounds Morgana with Urganda, and he has been followed in his mistake. D'Herbelot also thinks it possible that Féerie may come from Peri; but he regards the common derivation from Fata as much more probable. Cambrian etymologists, by the way, say that Morgain is Mor Gwynn, the White Maid.), we fancy a little too much importance has been attached to her. Her name, as far as we can leant, only occurs in the Cahermân Nâmeh, a Turkish romance, though perhaps translated from the Persian.
The foregoing etymologies, it is to be observed, are all the conjectures of English scholars; for the English is the only language in which the name of the individual, Fairy, has the canine letter to afford any foundation for them.
Leaving, then, these sports of fancy, we will discuss the true origin of the words used in the Romanic languages to express the being which we name Fairy of Romance. These are Faée, Fée, French; Fada, Provençal (whence Hada, Spanish); and Fata, Italian.
The root is evidently, we think, the Latin fatum. In the fourth century of our aera we find this word made plural, and even feminine, and used as the equivalent of Parcae. On the reverse of a gold medal of the Emperor Diocletian are three female figures, with the legend Fatis victricibus; a cippus, found at Valencia in Spain, has on one of its sides Fat. Q. Fabius ex voto, and on the other, three female figures, with the attributes of the Moerae or Parcae. [these two instances are given by Mdlle. Amélie Bosquet (La Normandie Romanesque, etc. p. 91.) from Dom Martin, Rel. des Gautois, ii. ch. 23 and 24.] In this last place the gender is uncertain, but the figures would lead us to suppose it feminine. On the other hand, Ausonius [Gryphus ternaril numeri] has tres Charites,, tria Fata; and Procopius [Do Bell. Got. i. 25] names a building at the Roman Forum τά τρία, adding οΰτω γάρ 'Ρωμας τάς μοίρας νενομίκαδι καλείν . The Fate or Fata, then, being persons, and their name coinciding so exactly with the modem terms, and. it being observed that the Moerae were, at the birth of Meleager, just as the Fees were at that of Ogier le Danois, and. other heroes of romance and tale, their identity has been at once asserted, and this is now, we believe, the most prevalent theory. To this it may be added, that in Gervase of Tilbury, and other writers of the thirteenth century, the Fada or Fée seems to be regarded as a being different from human kind. [g]
On the other hand, in a passage presently to be quoted from a celebrated old romance, we shall meet a definition of the word Fée, which expressly asserts that such a being was nothing more than a woman skilled in magic; and such, on examination, we shall find to have been all the Fées of the romances of chivalry and of the popular tales; in effect, that fée is a participle, and the words dame or femme is--to be understood.
In the middle ages there was in use a Latin verb, fatare, [h] derived from fatum or fata, and signifying to enchant. This verb was adopted by the Italian, Provençal [i] and Spanish languages; in French it became, according to the analogy of that tongue, faer, féer. Of this verb the past participle is faé, fé; hence in the romances we continually meet with les chevalier, faés, les dames faées, Oberon la faé, le cheval étoit faé, la clef était fée, and such like. We have further, we think, demonstrated [see our Virgil, Excurs. ix.] that it was the practice of the Latin language to elide accented syllables, especially in the past participle of verbs of the first conjugation, and. that this practice had been transmitted to the Italian, whence fatato-a would form fato-a, and una donna fatata might thus become una fata. Whether the same was the case in the Provençal we cannot affirm, as our knowledge of that dialect is very slight; but, judging from analogy, we would. say it was, for in Spanish Hadada and Hada are synonymous. In the Neapolitan Pentamerone Fata and Maga are the same, and a Fata sends the heroine of it to a sister of hers, pure fatata.
Ariosto says of Medea--
E perchè per virtu d' erbe e d'incanti
Delle Fate una ed immortal fatta era.
I Cinque Canti, ii. 106.
The same poet, however, elsewhere says--
Queste che or Fate e dagli antichi foro
Gia dette Ninfe e Dee con piu bel nome.
--Ibid. i. 9.
Nascemmo ad un punto che d'ogni altro male
Siamo capaci fuorchè della morto.
--Orl. Fur. xliii. 48.
which last, however, is not decisive. Bojardo also calls the water-nymphs Fate; and our old translators of the Classics named them fairies. From all this can only, we apprehend, be collected, that the ideas of the Italian poets, and others, were somewhat vague on the subject.
From the verb faer, féer, to enchant, illude, the French made a substantive faerie, féerie, [j] illusion, enchantment, the meaning of which was afterwards extended, particularly after it had been adopted into the English language.
We find the word Faerie, in fact, to be employed in four different senses, which we will now arrange and exemplify.
1. Illusion, enchantment.
Plusieurs parlent de Guenart,
Du Loup, de l'Asne, de Renart,
De faeries et de songes,
De phantosmes et do mensonges.
Gul. Giar. ap. Ducange
Where we must observe, as Sir Walter Scott seems not to have been aware of it, that the four last substantives bear the same relation to each other as those in the two first verses do.
Me bifel a ferly
Of faerie, me thought
Vision of Piers Plowman, v. 11.
Maius that sit with so benigne a chere,
Hire to behold it seemed faerie.
Chaucer, Marchante's Tale.
It (the horse of brass) was of fairie as the pople semed,
Diversè folk diversely han demed.
--Squier's Tale.
The Emperor said on high,
Certes it is a faerie,
Or elles a vanité
With phantasme and faerie,
Thus she bleredè his eye.
--Libeaus Disconus.
The God of her has made an end,
And fro this worldes faerie
Hath taken her into companie.--
Gower, Constance.
Mr. Ritson professes not to understand the meaning of faerie in this last passage. Mr. Ritson should, as Sir Hugh Evans says, have 'prayed his pible petter;' where, among other things that might have been of service to him, he would have learned that 'man walketh in a vain shew,' that 'all is vanity,' and that 'the fashion of this world passeth away;' and then he would have found no difficulty in comprehending the pious language of 'moral Gower,' in his allusion to the transitory and deceptive vanities of the world.
2. From the sense of illusion simply, the transition was easy to that of the land of illusions, the abode of the Faés, who produced them; and Faerie next came to signify the country of the Fays. Analogy also was here aiding; for as a Nonnerie was a place inhabited by Nonnes, a Jewerie a place inhabited by Jews, so a Faerie was naturally a place inhabited by Fays. Its termination, too, corresponded with a usual one in the names of countries: Tartarie, for instance, and 'the regne of Feminie.'
Here beside an elfish knight
Hath taken my lord in fight,
And hath him led with him away
Into the Faerie, sir, parmafay.--Sir Guy.

La puissance qu'il avoit aur toutes faeries du monde.
Huon de Bordeaux.
En effect, s'il me falloit retourner en faerie, je ne sçauroye ou prendre mon chemin.
--Ogier le Dannoys.
That Gawain with his oldè curtesie,
Though he were come agen out of faerie.
Squier's Tale.
He (Arthur) is a king y-crowned in Faerie,
With sceptre and pall, and with his regalty
Shailè resort, as lord and sovereigne,
Out of Faerie and reignè in Bretaine,
And repair again the ouldè Roundè Table.
Lydgate, Fall of Princes, bk. viii. c. 24.
3. From the country the appellation passed to the inhabitants in their collective capacity, and the Faerie now signified the people of Fairy-land. [k]
Of the fourth kind of Spritis called the Phairie.
K. James, Demonologie, 1.3.
Full often time he, Pluto, and his quene
Proserpina, and alle hir faerie,
Disporten hem, and maken melodie
About that well
--Marchante's Tale
The feasts that underground the Faerie did him make,
And there how he enjoyed the Lady of the Lake.
Drayton, Poly-Olb., Song IV.
4. Lastly, the word came to signify the individual denizen of Fairy-land, and was equally applied to the full-sized fairy knights and ladies of romance, and to the pygmy elves that haunt the woods and dells. At what precise period it got this its last, and subsequently most usual sense, we are unable to say positively; but it was probably posterior to Chaucer, in whom it never occurs, and certainly anterior to Spenser, to whom, however, it seems chiefly indebted for its future general currency. [l] It was employed during the sixteenth century [m] for the Fays of romance, and also, especially by translators, for the Elves, as corresponding to the Latin Nympha.
They believed that king Arthur was not dead, but carried awaie by the Fairies into some pleasant place, where he should remaine for a time, and then returne again and reign in as great authority as ever.
Hollingshed, bk. v. c. 14. Printed 1577.
Semicaper Pan
Nunc tenet, at quodam tenuerunt tempore nymphae,
- Ovid, Met. xiv. 520.
The halfe-goate Pan that howre
Possessed it, but heretofore it was the Faries' bower.
Golding, 1567.
Haec nemora indigenae fauni nymphaeque tenebant,
Gensque virum truncis et duro robore nata.
Virgil, Aeneis, viii. 314.
With nymphis and faunis apoun every side,
Qwhilk Farefolkis or than Elfis clepen we.
Gawin Dowglas.
The weeds (quoth he) sometime both farms and nymphs, and gods of ground,
And Fairy-queens did keep, and under them a nation rough.
Phaer, 1562.
Inter Hamadryadsa celeberrima Nonacrinas
Naias una fuit .
--Ovid, Met.
l. i. 690.
Of all the nymphes of Nonacris and Fairie ferre and neere,
In beautie and in personage this ladie had no peere.
Pan ibi dum teneris jactat sua carmina nymphis.
Ov. Ib .xi. 153.
There Pan among the Faire-elves, that daunced round togither.
Solaque Naiadum celeri non nota Dianae--Ov. Ib. iv. 304.
Of all the water-fayries, she alonely was unknowne.
To swift Diana.
Nymphis latura coronas
--Ov. Ib. ix. 337
Was to the fairies of the lake fresh garlands for to bear.
Thus we have endeavoured to trace out the origin, and mark the progress of the word Fairy, through its varying significations, and trust that the subject will now appear placed in a clear and intelligible light.
After the appearance of the Faerie Queene, all distinctions were confounded, the name and attributes of the real Fays or Fairies of romance were completely transferred to the little beings who, according to the popular belief, made 'the green sour ringlets whereof the ewe not bites.' The change thus operated by the poets established itself firmly among the people; a strong proof, if this idea be correct, of the power of the poetry of a nation in altering the phraseology of even the lowest classes [n] of its society.
Shakspeare must be regarded as a principal agent in the revolution; yet even he uses Fairy once in the proper sense of Fay; a sense it seems to have nearly lost, till it was again brought into use by the translators of the French Contes des Fées in the last century.
To this great Fairy I'll commend thy acts.
Anthony and Cleopatra, act iv. sc. 8.
And Milton speaks
Of Faery damsels met in forests wide
By knights of Logres or of Lyones,
Lancelot, or Pelleas, or Pellinore.
Yet he elsewhere mentions the
Faery elves,
Whose midnight revels by a forest side
Or fountain some belated peasant sees.
Finally, Randolph, in his Amyntas, employs it, for perhaps the last time, in its second sense, Fairy-land:
I do think
There will be of Jocastus' brood in Fairy.
Act i. sc. 3.
We must not here omit to mention that the Germans, along with the French romances, early adopted the name of the Fees. They called them Feen and Feinen. [o] In the Tristram of Gottfried von Strazburg we are told that Duke Gylan had a syren-like little dog,
Dez sie dem Herzoge gesandt '
Twas sent unto the duke, pardé,
Uz Avalun, der .Feinen land,
From Avalun, the Fays' countrie,
Von omen Gottinne.--V. 1673.
By a gentle goddess.
In the old German romance of Isotte and Blanscheflur, the hunter who sees Isotte asleep says, I doubt
Des the menachlich sei,
   If she human be,
Sic ist schoner denn eine Feine.
She is fairer than a Pay.
Von Fleische noch von Beine
   Of flesh or bone, I say,
Kunte nit gewerden
   Never could have birth
So schones auf den erden.
   A thing so fair on earth.
Our subject naturally divides itself into two principal branches, corresponding to the different classes of beings to which the name Fairy has been applied. The first, beings of the human race, but endowed with powers beyond those usually allotted to men, whom we shall term FAYs, or FAIRIES OF ROMANCE. The second, those little beings of the popular creeds, whose descent we propose to trace from the cunning and ingenious Duergar or dwarfs of northern mythology, and whom we shall denominate ELVES or POPULAR FAIRIES.
It cannot be expected that our classifications should vie in accuracy and determinateness with those of natural science. The human imagination, of which these beings are the offspring, works not, at least that we can discover, like nature, by fixed and invariable laws; and it would be hard indeed to exact from the Fairy historian the rigid distinction of classes and orders which we expect from the botanist or chemist. The various species so run into and are confounded with one another; the actions and attributes of one kind are so frequently ascribed to another, that scarcely have we begun to erect our system, when we find the foundation crumbling under our feet. Indeed it could not well be otherwise, when we recollect that all these beings once formed parts of ancient and exploded systems of religion, and that it is chiefly in the traditions of the peasantry that their memorial has been preserved.
We will now proceed to consider the Fairies of romance; and as they are indebted, though not for their name, yet perhaps for some of their attributes, to the Peries of Persia, we will commence with that country. We will thence pursue our course through Arabia, till we arrive at the middle-age romance of Europe, and the gorgeous realms of Fairy-land; and thence, casting a glance at the Faerie Queene, advance to the mountains and forests of the North, there to trace the origin of the light-hearted, night-tripping elves.

[a] the mark on Adam's Peak in Ceylon is, by the Buddhists, ascribed to Buddha; by the Mohammedans, to Adam. It reminds one of the story of the lady and the vicar, viewing the moon through a telescope; they saw in it, as they thought, two figures inclined toward each other: "Methinks," says the lady," they are two fond lovers, meeting to pour forth their vows by earth-light." "Not at all," says the vicar, taking his turn at the glass; "they are the steeples of two neighbouring churches"

[b] Faerie Queene, III. c. iii. st. 8, 9, 10, 11. Drayton, Poly-Olbion, Song VI. We fear, however, that there is only poetic authority for this belief. Mr. Todd merely quotes Warton, who says that Spenser borrowed it from Giraldus Cambrensis, who picked it up among the romantic traditions propagated by the Welsh bards. The reader will be, perhaps, surprised to hear that Giraldus says nothing of the demons. He mentions the sounds, and endeavours to explain them by natural causes. Hollingshed indeed (l. i. c. 24.) says, "whereof the superstitious sort do gather many toys."

[c] for a well-chosen collection of examples, see the very learned and philosophical preface of the late Mr. Price to his edition of Warton's History of English Poetry, p.28 et seq.

[d] in the Middle Ages the gods of the heathens were all held to be devils.

[e] Φήρ is the Ionic form of θήρ, and is nearly related to the German thier, beast, animal. The Scandinavian dyr, and the Anglo-Saxon deon, have the same signification; and it is curious to observe the restricted sense which this last has gotten in the English deer.

[f] preface to Warton, p. 44; and Breton philologists furnish us with an etymon; not, indeed, of Fairy, but of Fada. "Fada, fata, etc.," says M. de Cambry (Monumens Celtiques), "come from the Breton mat or mad, in construction fat, good; whence the English, maid."

[g] see below, France. It is also remarked that in some of the tales of the Pentamerone, the number of the Fate is three; but to this it may be replied, that in Italy every thing took a classic tinge, and that the Fate of those tales are only Maghe; so in the Amadigi of Bernardo Tasso we meet with La Fata Urganda. In Spain and France the number would rather seem to have been seven. Cervantes speaks of "los siete castillos do las siete fadas;" in the Rom. de Ia Infantina it is said, "siete fadas me fadaron, en brazos de una ama mia," and the Fées are seven in La Belle au Bois dormant. In the romance, however, of Guillaume au Court-nez, the Fées who carry the sleeping Renoart out of the boat are three in number--See Grimm Deutsche Mythologie, p. 383.

[h] a MS. Of the 13th century, quoted by Grimm (ut sup. p. 405), thus relates the origin of Aquisgrani (Aix la Chapelle): Aquisgrani dicitur Ays, et dicitur eo, quod Karolus tenebrat ibi quandum mulierem fatatam, sive quandum fatum. Quae alio nominee nimpha vel dea vel adriades (l. dryas) appellatur, et ad hanc consuetudinem habetat et eam cognoscebat; et ita erat, quod ipso accedente ad eam vivebat ipsa, ipso Karola recedente moriebatur. Contigit dum quadam vice ad ipsam accessisset ut cum ea delectaretur, radius solis intravit os ejus, et tune Karolus vidit granum auri lingue ejus affixum, quod facit abscindi et contingenti (l. in continenti) mortua est, nec postea revixit.

[i] "Aissim fadaro tres serors / En aquella ora qu' ieu sui natz / Que totz temps fos enamoratz."--Folquet de Romans. (Thus three sisters fated in the hour that I was born, that I should be at all times in love.): "Aissi fuy de nueitz fadatz sobr' un puegau."--Guilh. De Poitou. (Thus was I fated by night on a hill.)--Grimm, ut sup. p. 383 

[j] following the analogy of the Gotho-German tongues, zauberei, Germ. trylleri, Dan. trolleri, Swed. illusion, enchantment. The Italian word is fattucchiera.

[k] here too there is perhaps an analogy with cavalry, infantry, squierie, and similar collective terms.

[l] the Faerie Queene was published some years before the Midsummer Night's Dream. Warton,(Obs. on the Faerie Queene) observes: "It appears from Marston's Satires, printed 1598, that the Faerie Queene occasioned many publications in which Faeries were the principal actors. Go buy some ballad of the FAERY KING.--Ad Lectorem.:
Out steps some Faery with quick motion, / And tells him wonders of some flowerie vale--/ Awakes, straight rubs his eyes, and prints his tale.--B. III. Sat. 6."

[m] it is in this century that we first meet with Faery as a dissyllable, and with a plural. It is then used in its fourth and last sense.

[n] the Fata Morgana of the Straits of Messina is an example; for the name of Morgana, whencesoever derived, was probably brought into Italy by the poets.

[o] Dobenek, des deutschen Mittelalters und Volksglauben. Berlin, 1816.

Next: Persian Romance