Grimm's Teutonic Mythology
Chapter 17: Wights and Elves
Apart from deified and semi-divine natures there stands a whole
order of other beings distinguished mainly by the fact that, while those have
issued from men or seek human fellowship, these form a seperate community, one
might say a kingdom of their own, and are only induced by accident or stress
of circumstances to have dealings with men. They have in them some admixture
of the superhuman, which approximates them to gods; they have power to hurt
man, being no match for him in bodily strength. Their figure is much below the
stature of man, or else mis-shapen. They almost all have the faculty of making
themselves invisible. (1) And here
again the females are of a broader and nobler cast, with attributes resembling
those of goddesses and wise-women; the male spirits are more distinctly marked
off, both from gods and from heroes. (2)
The two most general designations for them form the title of
this chapter; they are what we should call spirits nowadays. But the word spirit
(geist, ghost), (3) like the Greek daimwn, is too comprehensive; it would
include, for instance, the half-goddesses discussed in the preceeding chapter.
The Lat. genius would more nearly hit the mark. (see Suppl.).
The term wiht seems remarkable in more than one respect, for
its variable gender and for the abstract meanings developed from it. The Gothic
vaíhts, gen. vaíhtáis, is feminine, and Ulphilas hardly
ever uses it in a concrete sense; in Luke 1, 1 he translates by it pragma,
and much oftener, when combined with a negative, ouden (Gramm. 3, 8. 734). This, however, does not exclude the possibility of vaíhts
having at other times denoted to the Goths a spirit regarded as female; and
in 1 Thess. 5, 22 the sentence apo pantoj eidouj
ponhtou apecesqe is rendered: af allamma vaíhtê ubiláizô
afhabáiþ izvis, where the Vulg. has: ab omni specie mala abstinete
vos; the use of the pl. 'vaíhteis ubilôs' of itself suggests the
notion of spirits. The other Teutonic tongues equally use the word to intensify
and make a substantive of the negative, and even let it swallow up at last the
proper particle of negation; (4) but
in all of them it retains its personal meaning too. The OHG. writers waver between
the neut. and masc.; the Gothic fem. is unknown to them. Otfried has a neut.
wiht, with the collective pl. wihtir, (5) and likewise a neut. pl. wihti, which implies a sing. wihti; thus, armu wihtir,
iv. 6, 23; armu wihti, ii. 16, 117; krumbu wihti, iii. 9, 5; meaning 'poor,
crooked creatures,' so that wiht (derivable from wîhan facere, creare)
seems altogether synonymous with being, creature, person, and can be used of
men or spirits: 'in demo mere sint wunderlîchiu wihtir, diu heizent sirenae,'
Hoffm. Fundgr. 19, 17. In MHG. sometimes neut.: unreinez wiht, Diut. 1, 13;
Athis H. 28; trügehaftez wiht, Barl. 367, 11; vil tumbez wiht, 11, 21;
sometimes masc.: bœser wiht, Barl. 220, 15; unrehter bœsewiht, MS. 2, 147ª,
Geo. 3508; kleiner wiht, Altd. bl. 1, 254; der wiht, Geo. 3513-36; der tumbe
wiht, Fragm. 42ª; and often of indeterminable gender: bœse wiht, Trist. 8417;
helle wiht, Geo. 3531; but either way as much aplicable to men as to spirits.
Ghostly wights are the 'minuti dii' of the Romans (Plaut. Casina, ii. 5, 24).
In Mod. Germ. we make wicht masc., and use it slightingly of a pitiful hapless
being, fellow, often with a qualifying epithet: 'elender wicht, bösewicht
(villian).' If the diminutive form be added, which intensifies the notion of
littleness, it can only be used of spirits: wichtlein, wichtelmann; (6) MHG. diu wihtel, (7) MS. 1, 157ª;
bœsez wihtel, Elfenm. cxviii.; kleinez wihtelîn, LS. 1, 378, 380, Wolfdietr.
788, 799; OHG. wihtelîn penates; wihtelen vel helbe (i.e. elbe), Lemures,
dæmones, Gl. Florian. The dernea wihti, occulti genii, in Hel. 31, 20.
92, 2 are deceitful demonic beings, as 'thie derno' 164, 19 means the devil
himself; lêtha wihti, 76, 15; wrêda wihti 76, 1. In Lower Saxony
wicht is said, quite in a good sense, of little children: in the Münster
country 'dat wicht' holds especially true of girls, about Osnabrück the
sing. wicht only of girls, the pl. wichter of girls and boys; 'innocent wichte'
are spoken of in Sastrow, 1, 351. The Mid. Nethl. has a neut. wicht like the
H. German: quade wicht, clene wicht (child). Huyd. op St. 3, 6. 370; arem wiht,
Reinh. 1027; so the Mod. Dutch wicht, pl. wichteren: arm wicht, aardig wicht,
in a kindly sense. The AS. language agrees with the Gothic as to the fem. gender:
wiht, gen. wihte, nom. pl. wihta; later wuht, wuhte, wuhta; seo wiht, Cod. Exon.
418, 8. 419, 3. 5. 420, 4. 10. The meaning can be either concrete: yfel wiht
(phantasma), leás wiht (diabolus), Cædm. 310, 16; sœwiht (animal
marinum), Beda, 1, 1; or entirely abstract = thing, affair. The Engl. wight
has the sense of our wicht. The ON. vœtt and vœttr [[supernatural being, spirit]],
which are likewise fem., have preserved in its integrity the notion of a demonic
spiritual being (Sæm. 145ª): allar vœttir, genii quicuque, Sæm.
93b; hollar vœttir, genii benigni, Sæm. 240b; ragvœttir or meinvœttir,
genii noxii, (8) landvœttir, genii
tutelares, Fornm. sög. 3, 105. Isl. sög. 1, 198, etc. In the Färöes
they say: 'feâr tû têar til mainvittis (go to the devil)!'
Lyngbye, p. 548. The Danish vette is a female spirit, a wood-nymph, meinvette
an evil spirit, Thiele 3, 98. The Swedish tongue, in addition to vätt (genius)
and a synonymous neut. vättr, has a wikt formed after the German, Ihre,
p. 1075. Neither is the abstract sense wanting in any of these dialects.
This transition of the meaning of wight into that of thing on
the one hand, and of devil on the other, agrees with some other phenomena of
language. We also address little children as 'thing,' and the child in the märchen
(No. 105) cries to the lizard: 'ding, eat the crumbs, too!' Wicht, ding, wint,
teufel, vâlant (Gramm. 3, 734. 736) all help to clinch a denial. O. French
males choses, male genii, Ren. 30085. Mid. Latin bonœ res = boni genii, Vinc.
Bellov. iii. 3, 27 (see Suppl.).
We at once perceive a more decided colouring in the OHG. and
MHG. alp (genius), AS. œlf, ON. âlfr [[elf]]; a Goth. albs may safely
be conjectured. Together with this masc., the OHG. may also have had a neut.
alp, pl. elpir, as we know the MHG. had a pl. elber; and from the MHG. dat.
fem. elbe (MS. 1, 50b) we must certianly infer a nom. diu elbe, OHG. alpia,
elpia, Goth. albi, gen. albjôs, for otherwise such a derivative could
not occur. Formed by a still commoner suffix, there was no doubt an OHG. elpinna,
MHG. elbinne, the form selected by Albrecht of Halberstadt, and still appearing
in his poem as remodelled by Wikram; (9) AS. elfen, gen. elfenne. Of the nom. pl. masc. I can only feel sure in the ON.,
where it is âlfar [[pl. of álfr - elves]], and would imply a Goth.
albôs, OHG. alpâ, MHG. albe, AS. ælfas; on the other hand
an OHG. elpî (Goth. albeis) is suggested by the MHG. pl. elbe (Amgb. 2b,
unless this comes from the fem. elbe above) and by the AS. pl. ylfe, gen. pl.
ylfa (Beow. 223). (10) The Engl.
forms elf, elves, the Swed. elf, pl. maasc. elfvar (fem. elfvor), the Dan. elv,
pl. elve, are quite in rule; the Dan. compounds ellefolk, ellekoner, elleskudt,
ellevild have undergone assimilation. With us the word alp still survivies in
the sense of night-hag, night-mare, in addition to which our writers of the
last century introduced the Engl. elf, a form untrue to our dialect; before
that, we find everywhere the correct pl. elbe or elben. (11) H. Sachs uses ölp: 'du ölp! du dölp!' (i. 5, 525b), and ölperisch
(iv. 3, 95c); conf. ölpern and ölpetrütsch, alberdrütsch,
drelpetrütsch (Schm. 1, 48); elpentrötsch and tölpentrötsch,
trilpentrisch (Schmid's Swab. dict. 162); and in Hersfeld, hilpentrisch. The
words mean an awkward silly fellow, one whom the elves have been at, and the
same thing is expressed by the simple elbisch, Fundgr. 365. In Gloss. Jun. 340
we read elvesce wehte, elvish wights.
On the nature of Elves I resort for advise to the ON. authorities, before all others. It has been remarked already (p. 25), that the Elder Edda several times couples œsir and âlfar together, as though they were a compendium of all higher beings, and that the AS. ês and ylfe stand together in exactly the same way. This apparently concedes more of a divinity to elves than to men. Sometimes there come in, as a third member, the vanir (Sæm. 83b), a race distinct from the æsir, but admitted to certain relations with them by marriage and by covenants. The Hrafnagaldr opens with the words: Alföðr orkar (works), âlfar skilja, vanir vita,' Sæm. 88ª; Allfather, i.e., the âs, has power, âlfar have skill (understanding), and vanir knowledge. The Alvîsmâl enumerates the dissimilar names given to heavenly bodies, elements and plants by various languages (supra, p. 332); in doing so, it mentions œsir, âlfar, vanir, and in addition also goð, menn, ginregin, iötnar, dvergar and denizens of hel (hades). Here the most remarkable point for us is, that âlfar and dvergar (dwarfs) are two different things. The same distinction is made between âlfar and dvergar, Sæm. 8b; between dvergar and döckâlfar, Sæm. 92b; between three kinds of norns, the âs-kungar, âlf-kungar and dœtr Dvalins, Sæm. 188ª, namely, those descended from âses, from elves and from dwarfs; and our MHG. poets, as we see by Wikram's Albrecht, 6, 9, continued to separate elbe from getwerc. (12) Some kinship however seems to exist between them, if only because among proper names of dwarfs we find an Alfr and a Vindâlfr, Sæm. 2. 3. Loki, elsewhere called an âs, and reckoned among âses, but really of iötun origin, is nevertheless addressed as âlfr, Sæm. 110b; nay, Völundr, a godlike hero, is called 'âlfa lioði,' alforum socius, and 'vîsi âlfa,' alforum princeps, Sæm. 135ª,b. I explain this not historically (by a Finnish descent), but mythically: German legend likewise makes Wielant king Elberich's companion and fellow smith in Mount Gloggensachsen (otherwise Göugelsahs, Caucasus?). Thus we see the word âlfr shrink and stretch by turns.
1. But so have the gods (p. 325), goddesses (p. 268) and wise women (p. 419). Back
2. Celtic tradition, which runs particularly rich on this subject, I draw from the following works: Fairy Legends and Traditions of the South of Ireland, by Crofton Croker, Lond. 1825; 2nd ed., parts 1, 2, 3, Lond. 1828. The Fairy Mythology, by Th. Keightley, vols. 1, 2, Lond. 1828. Barzas-Breiz, chants populaires de la Bretagne, par Th. de la Villemarqué, 2e éd., 2 vol., Paris 1840. Back
3. OHG. keist, AS. gâst, OS. gêst (see root in Gramm. 2, 46); Goth. ahma, OHG. âtum for ahadum, conn. with Goth. aha (mens), ahjan (meminisse, cogitare), as man (homo), manniska, and manni, minni belong to munan, minnen (pp. 59. 344. 433). Back
4. Aught = â-wiht, any wight or whit; naught = n'â-wiht, no wight, no whit.---Trans. Back
5. So: thiu diufilir, iii. 14, 53, by the side of ther diufal, iii. 14, 108. Back
6. In Hesse wichtelmänner is the expression in vogue, except on the Diemel in Saxon Hesse, where they say 'gute holden.' Back
7. Swer weiz und doch niht wizzen wil, ...................Whoso knows, yet will not know,
der slæt sich mit sîn selbes hant; ...................Smites himself with his own hand;
des wîsheit aht ich zeime spil, ..................His wisdom I value no more than a play
daz man diu wihtel hât genannt: ..................That they call 'the little wights':
er lât uns schouwen wunders vil, ....................He lets us witness much of wonder,
der ir dâ waltet. ..........................Who governs them.
The passage shows that in the 13th cent. there was a kind of puppet-show in which ghostly
beings were set before the eyes of spectators. 'Der ir waltet,' he that wields them, means the
showman who puts the figures in motion. A full confirmation in the Wachtelmäre, line 40:
'rihtet zu mit den snüeren (strings) die tatermanne!' Another passage on the wihtel-spil in
Haupt's Zeitschr. 2, 60: 'spilt mit dem wihtelin ûf dem tisch umb guoten win.' Back
8. Biörn supposes a masc. (fem.?) meinvættr and a neut. meinvætti; no doubt mein is noxa, malum; nevertheless I call attention to the Zendic mainyus, dæmon, and agramainyus, dæmon malus. Back
9. Wikram 1,9. 6, 9 (ed. 1631, p. 11ª 199b). The first passage, in all the editions I have compared (ed. 1545, p. 3ª), has a faulty reading: 'auch viel ewinnen und freyen,' rhyming with 'zweyen.' Albrecht surely wrote 'vil elbinnen und feien.' I can make nothing of 'freien' but at least a very daring allusion to Frigg and Frea (p. 301); and 'froie' = fräulein, as the weasel is called in Reinh. clxxii., can have nothing to say here. Back
10. Taking AS. y [as a modified a, œ, ea] as in yldra, ylfet, yrfe, OHG. eldiro, elpiz, erpi. At the same time, as y can also be a modified o (orf, yrfe = pecus), or a modified u (wulf, wylfen), I will not pass over a MHG. ulf, pl. ülve, which seems to mean much the same as alp, and may be akin to an AS. ylf: 'von den ülven entbunden werden,' MS. 1, 81ª; 'ülfheit ein suht ob allen sühten,' MS. 2, 135ª; 'der sich ülfet in der jugent,' Helbl. 2, 426; and conf. the ölp quoted from H. Sachs. Shakspeare occasionally couples elves and goblins with similar beings called ouphes (Nares sub v.). It speaks for the identity of the two forms, that one Swedish folk-song (Arwidsson 2, 278) has Ulfver where another (2, 276) has Elfver. Back
11. Besold. sub v. elbe; Ettner's Hebamme, p. 910, alpen or elben. Back
12. In Norway popular
belief keeps alfer and dverge apart, Faye p. 49. Back