An Early Icelandic Attestation of <i>ÞVeit</i> ‘Thwaite’

The word thwaite in English and its cognates in the Scandinavian languages and Norman French are preserved almost exclusively as placenames: thus the simplex Thwaite is a placename in England, Tveit or Tveita in Norway, Tvede in Denmark, Tveta in Sweden, Twatt in Shetland and Orkney, and (Le) Thuit in Normandy. As second element in compounds, -thwaite occurs frequently in English placenames, for instance Armathwaite and Bassenthwaite, with greatest prevalence in the north but with outliers in Norfolk and Suffolk and elsewhere.1 Correspondingly in Normandy Longtuit and Miltuit are attested, along with numerous other compounds with second element -tuit.2 The contention of this article is that a hitherto unrecognized attestation of the Old Icelandic cognate of ‘thwaite’, þveit, is preserved in a skaldic poem. This example, if valid, would constitute a rare instance of the word being used as a common noun. According to the Old Icelandic Egils saga, the eponymous hero speaks the following verse upon hearing of the death of his protector and friend Arinbjǫrn hersir: Now {the bright ones {of the assembly of Ingvi }} [(lit. ‘assembly-bright ones of Ingvi’) BATTLE > WARRIORS] grow fewer, they who made {the daylight {of the mead-trench}} [DRINKING HORN > SILVER] decrease; where am I to look for generous men, those who hail-showered my {high mountains of the hawk} [ARMS] {with crucible-snow} [SILVER] beyond {the island-studded girdle of the earth} [SEA] for my words?4 The gist of these words, in something approaching the classic ‘Ubi sunt’ topos, is that the ranks of the wealthy protectors to whom the speaker was a client are diminishing. Where, he wonders, is he to look now for such generous benefactors? The two manuscripts of significance for the text of this verse are the mid-fourteenth-century vellum Möðruvallabók (AM 132 fol) (95va) and the late eighteenth-century paper manuscript Handritasafn Landsbókasafns Íslands (Lbs) 1408 4° (116v).5 Lbs 1408 is a member of what Bjarni Einarsson has termed the Eyf3 subgroup of manuscripts, so-called because there are three of them and they all have links with the Icelandic district of Eyjafjörður.6 Although in practice each manuscript presents its own selection of readings, they all represent a mixture of the redactions of Egils saga. Like the other members of this class, Lbs 1408 follows a no longer extant version of the A-redaction that was distinct from that of Möðruvallabók; additionally, in places it incorporates readings from the B and C redactions.7 Bjarni Einarsson’s analysis of the paradosis demonstrates that a few readings from Lbs 1408 and its Eyf3 colleagues uniquely testify to the original saga text.8 In line 4 of the verse Möðruvallabók reads mjaðveitar, whereas Lbs 1408 has mä þveitar. The Möðruvallabók reading is written as a compound, which is clearly correct given that mjað- is the compounding form of mjǫðr (genitive mjaðar) ‘mead’ and cannot stand alone. A literal translation of the compound is ‘of the mead-channel’ (genitive case, from nominative -veit). The Lbs 1408 reading mä þveitar, although set out as two separate words, must also be understood as a compound, i.e. máþveitar, given that the first syllable is the compounding form of már ‘seagull’ (genitive más). The Lbs 1408 reading can be translated as ‘of the seagull’s slice of land’ (genitive case, from -þveit), i.e. ‘of the sea’; the meaning of þveit, here rendered as ‘slice of land’, will be discussed in more detail presently. Whichever of the two readings is chosen, the metrico-syntactic rules of skaldic verse require that they be taken together with the word dag[r] ‘day’ that follows.9 The sole way of achieving that within the rules is by construing dag as defined by the preceding compound noun to mean ‘gold’. This is consistent with the frequent use of expressions for ‘light, radiance’, etc., as a baseword in kennings for ‘gold’.10 The exact identity of the defining element has not been satisfactorily resolved. Two principal lines of explanation have been offered: Finnur Jónsson, in two separate editions, prefers Lbs 1408, using its reading as the basis for his posited máreitar dag ‘day (= light) of the seagull-land [SEA > GOLD]’.11 Here, as already noted, the first syllable, má-, is the compounding form of már ‘seagull’. The second element is reita, which means ‘demarcated area’, e.g., ‘garden bed’. A combination ‘day of the garden bed of the sea-gull’ (‘day of the sea’) conforms to a classic template for gold-kennings, ‘fire of the sea’.12 Finnur offers no explanation, however, for the emendation of þveitar to reitar and, as we shall see, it is unnecessary. The unemended text of the verse as transcribed from Lbs 1408 is printed by Bjarni Einarsson but he does not comment on the interpretation and offers no translation.13 A longer list of scholars has preferred the Möðruvallabók reading. Sveinbjörn Egilsson suggested a gloss præbitor cerevisiæ ‘mead-pourer’, which he explained as ‘convivator, vir hospitalis’ (‘convivial person, hospitable man’). He construed veitar as plural of veitir ‘pourer, supplier’ and took the word dag[r] separately.14 E.A. Kock brought Sveinbjörn’s line of explanation into conformity with the metrico-syntactic rules by re-attaching dag to mjaðveitar. He proposed that mjaðveit ‘mead-pourer’ was a kenning for ‘woman’, on the basis that traditionally it fell to women to pour the drinks at a feast; her dagr ‘day’ is then ‘gold’, on the basis that women wore ornaments of gold.15 A difficulty here, however, is that kennings for gold are nowhere else formed on a template ‘day/fire/light, etc., of the woman’.16 Sigurður Nordal instead interprets mjaðveit as ‘mead-channel’ and thus as a kenning for ‘drinking horn’, whose ‘day’ is ‘gold’ because horns were often adorned with gold mounts.17 While Bjarni Einarsson and Clunies Ross follow this explanation,18 Clunies Ross points out that a template ‘day/fire/light, etc., of the horn’ is also without parallel.19 Where Lbs 1408 includes the verses in Egils saga (verses 123 and 130–32 in Clunies Ross’s enumeration), it is manifestly inferior to Möðruvallabók. Almost all its readings can be regarded as lectiones faciliores, as can be seen from Clunies Ross’s edition. But the present case is different. It is not easy to see how the eighteenth-century Icelandic writer of Lbs 1408 could have come up with the reading þveit—which, as will be discussed presently, is an obscure word—had it not been in the exemplar. In contrast, it is easy to see how the reading of Möðruvallabók could represent a garbled state of the original, since all that is needed is a simple syllabification error, whereby má-þveitar is misinterpreted as máþ-veitar (a natural consequence of the obscurity of the word þveit), with m changed to mi either through minim error or as a scribal emendation. We have already noted that mjaðveitar dag[r] does not conform to any recognized kenning template for ‘gold’, and that is another count against this reading. The probability, then, is that máþveitar represents the original reading and should be adopted in editions. The core sense of þveit appears to be ‘sliver, split-off or cut-off or chopped-off piece’.20 This can be deduced from the Old English cognate verb þwītan ‘to whittle’.21 The Middle English Dictionary definitions of the later form of the verb, þwite, can be summarized as follows: to whittle, trim, cut, hew, carve.22 Similarly the Oxford English Dictionary offers ‘to cut down, whittle, pare, shave; to shape by paring; to cut away’.23 The Germanic strong verb *þveitana- ‘to whittle’ belonged to the same class as *dreiƀana- ‘to drive’ and at one time perhaps possessed an equally extensive array of derived forms. Only vestiges of this survive in the Scandinavian languages. The Old Norse cognate of þwītan, presumably to be reconstructed as *þvíta, with e-grade ablaut, is not attested and all that remains is a set of deverbative nouns plus a causative verb, all based on o-grade and zero-grade forms. The strong masculine noun þveitr (representing the o-grade *þvait-) is glossed by Fritzner as Indsnit frembragt ved Øxe tvers over et Træ som man vil have delt i to Stykker, ‘cut made with an axe across a piece of timber that one intends to divide into two pieces’.24 In skaldic poetry this word is uniquely attested in Óttarr svarti’s Knútsdrapa verse 3. Matthew Townend, in his edition of this poem, translates loosely as ‘a blow, harsh impact’25 but the noun can be rendered in its strict sense. The notion is that Knútr’s victory has made a ‘cut across’ a kingdom that had been united under Edgar, whom the poet pointedly mentions in this stanza as the ancestor of England’s current ruler. Of other derived o-grade forms, the causative verb þveita means ‘hit, strike’ (i.e., ‘cause something to be whittled’).26 The weak feminine agentive noun þveita has two attestations as a heiti for ‘axe’ (i.e., ‘whittler’) in skaldic poetry.27 The weak masculine noun þveiti has the meaning ‘small coin or weight’, with the underlying sense of ‘unit whittled down from a full measure’.28 A generally related sense attaches to the second element of the compound noun ǫngþveiti, which Cleasby–Vigfusson glosses literally as ‘a narrow place, strait’29 and Fritzner more figuratively as slem stilling ‘bad situation’, i.e., ‘predicament’.30 The zero-grade masculine noun þviti is attested in prose solely in Snorra Edda, where it is used as a proper name.31 From the context þviti can be explicated as ‘stone driven into the ground as a tethering stake or picket’.32 The proper name can be explained, like some other proper names in the Edda, as a fossilized form of a common noun. For this an original meaning of ‘slingstone’ has been tentatively proposed by Cleasby–Vigfusson, followed by Townend,33 but a more likely sense, in view of this word’s membership in the *þvíta family, is ‘split stone’. Such a stone would lend itself to being rammed into the ground. The same word appears as a common noun in the Egill attribution Hǫfuðlausn verse 17, being used as second element in the compound hrammþvita (genitive case).34 This can be translated loosely as ‘stone of the hand’, therefore ‘gold’, but here too we should probably regard þviti as a hyponym rather than a synonym of ‘stone’, with the precise sense ‘split or broken stone’. Such a sense would be fitting in a context where the motif of the lord’s breaking up gold arm-rings and other bullion to present to his following is repeatedly invoked. All these derived forms convey the idea of a whittling down or a cutting into, i.e., a reduction or diminution of some kind. To return to the noun þveit, its semantic range evidently included the splitting off of small parcels of land from a larger farm or forest. This is especially evident in placenames: Modern Norwegian tveit signifies ‘clearing, grassy area in forest’, corresponding to Modern Swedish tveta; semantically divergent is Modern Danish tvede ‘tongue of land, peninsula’.35 Modern English dialect thwaite can be defined as ‘piece of ground; especially a piece of ground cleared from forest or reclaimed from waste’.36 The simplex þveit is not attested in Old Icelandic prose.37 Equally, it fails to appear in the Lexicon Poeticum of Sveinbjörn Egilsson or in the revised version of Lexicon Poeticum by Finnur Jónsson,38 nor is it represented in Icelandic placenames or in Modern Icelandic.39 The attestations in Old Norwegian are all from the Diplomatarium Norvegicum, which records numerous examples in Norwegian contexts, mostly if not entirely in the form of place-names.40 The explanation for the absence of this word in Icelandic may be systematic, related to the specific ecology of the island. Whereas in Norway the word tveit predominantly designates a small piece of land within a forest that has been cleared for grazing or cultivation, Iceland had a paucity of such forests to clear. In the kenning ‘slice of land of the seagull’ = ‘sea’ that we see in the Egils saga verse, þveit is an instance of the principle of extreme variation in the formation of skaldic kennings, whereby any synonym, metonym, or hyponym of ‘land’ can be placed within the template. Examples of semantically specialized nouns so used are borg ‘hill’, tún ‘homefield’, and teigr ‘paddock, strip of field’.41 It is thinkable that the poet selected the hyponym þveit because it evoked the concept of whittling or diminution, parallel in semantic terms to the repeated verb þverra ‘diminish/cause to diminish’ in the first line of the verse. At the same time there is also phonetic linkage between the two words: the marked alliteration of þv - in the first line has its echo within the fourth line, in an effect sporadically used by the skalds.42 In rhetorical terms, the entire verse is very highly wrought.43 Standard antithetical motifs to be found in ‘ubi sunt’ compositions such as the Old English poem ‘The Seafarer’—gifts, conviviality, brightness, and comfort versus loss, adverse weather, sea, and mountains—are clustered together through an ingenious use of kenning diction. It is not possible to state for certain whether the verse is authentically a composition by the historical Egill Skalla-Grímsson, whose floruit was in the tenth century. The rare word þveit cannot itself furnish evidence for an early date, since skalds of all periods had a penchant for rare and obscure lexis, in part as a means of extending the range of lexical items available for kennings.44 Kari Ellen Gade has argued that the prominent caesura before the final two syllables in line 4 serves to group this verse with some other likely Egill attributions.45 Such arguments, however, are circular since they depend on largely subjective judgements as to which Egill attributions are likely. A particular difficulty is that a verse attributed to Gísli Súrsson that Gade lists as exhibiting the same structure has strong links with twelfth-century hagiographic and visionary poetry and is likely to have been composed in that period.46 What can be said, nevertheless, is that many skalds of Icelandic origin, including Egill himself, travelled to Norway and some ventured further to Denmark, Sweden, Orkney, and England. In these five regions, as we have seen, the word þveit was in active use in the formation of placenames—active enough that an itinerant skald might have taken note of it. A parallel to such an importation of non-Icelandic vocabulary into skaldic diction would be the word merski ‘marsh’, a loanword from English or Low German listed in the Jarðar heiti,47 a versified mnemonic listing words for the concept ‘land’. We can therefore state conservatively that the attestation of þveit emanates from the classical skaldic era, at some point between Egill’s floruit in the tenth century and the composition of Egils saga in the first half of the thirteenth century.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.