Home Ellanto

Glaer Volthionnath Gŷr - Iest

(Vǫlsungakviða en Nýja - Upphaf / ‘The New Lay of the Völsungs - Beginning’)

Translated into Sindarin by Ellanto from the first part (Upphaf ‘Beginning’) of Tolkien's Vǫlsungakviða en Nýja ‘The New Lay of the Völsungs’ (The Legend of Sigurd & Gudrún, pp. 59-65).

The ‘The New Lay of the Völsungs’ is an original poem by J.R.R. Tolkien inspired by the motives of the Norse Eddas. Tolkien employs the classic fornyrðislag alliterative metre common in Old Norse poetry: each line contains two lifts (prosodic peaks), and either one or both of the lifts of an odd line alliterate with the first lift of the following even line. Alliteration is the repetition of a consonant or sequence of consonants at the start of a syllable (which is necessarily a stressed syllable when used metrically); all initial vowels alliterate with each other.

In this translation, the original alliterative metre is replicated in Sindarin. This necessitated some deviations from the original text, mostly in the form of rearranging lines but sometimes also in the form of minor rephrasings. A further consequence of the metre is the oftentimes complicated poetic syntax employed in the translation. The back-translation closely follows the syntax used in Sindarin, with one notable difference: the natural adjective-noun order in English is used for the equally natural noun-adjective order in Sindarin, whereas the poetic adjective-noun order in Sindarin is rendered in English as noun-adjective.

The following conventions are used for alliteration in Sindarin:

• Alliteration = phonetic identity of consonants. All vowels alliterate with each other.
• Only stressed syllables count for metrical alliteration.
• Syllabification goes across clitic boundaries but not across word boundaries: e.g. thar eryd alliterates on r, but beraidh eryd alliterates on either b or on a vowel.
• VCCV sequences always syllabify as VC.CV, i.e. idh·rem alliterates on r, not on ðr.

See also the notes at the bottom of the page regarding the adaptation of names, spelling conventions, and grammatical conventions assumed for this work.

Iest (Upphaf - ‘Beginning’)

Notes:

1. The Adaptation of Names

Norse names in the poem are phonetically adapted into Sindarin but left otherwise untranslated. The adaptation is based on the original Old Norse pronunciations rather than on the English forms used in the source text. Forms are minimally adapted to fit with Sindarin phonology and phonotactics, including sequences permissible only in mutated environments. Sequences that are impossible in (West) Sindarin but permissible in Northern Sindarin (e.g. word-initial sp-, st-, sk-) are assumed to be familiar enough to the imaginary Sindarin-speaking translator and audience to be permissible for adapted Norse names.

2. Spelling Conventions

The Sindarin spelling conventions adopted througout this work are mostly as described and used by Tolkien in CEA (PE23/133-144). Notably, word-initial mb-, nd-, ñg-, as the product of nasal mutation of stops or as the product of soft mutation for anciently prenasalised stops, stand for singleton nasals [m, n, ŋ]; these do not represent clusters and can therefore alliterate with nasals. Another convention inspired by CEA is the treatment of gw- > w- mutation as restoration of the primitive ✶w- rather than a deletion of the g. Consequently, the product of soft mutation of e.g. gaear (with primitive initial ✶g-) is written with an apostrophe to denote the deletion (‘aear), whereas the product of soft mutation of e.g. gwaelod (with primitive initial ✶w-) is written without an apostrophe (waelod).

An interpunct · is used to separate a definite article from the following word, e.g. gwae “wind”, e·wae “the wind”. The definite article is never capitalised, and sentence-initial capitalisation is instead carried over to the main word. An apostrophe is used between some proclitics and the following word, e.g. gwanath “death”, pen-wanath “deathless”.

3. Grammar

The present work deviates in some ways from Eldamo's recommendations regarding Sindarin grammar, and as such some of the assumptions made here are noted explicitly below for the readers' convenience.

3.1 Grammatical Mutations

A good understanding of where mutations are expected to occur is crucial for the application of the alliterative metre in Sindarin, and also equally important for appreciating a composition that uses it. As such, it is important to define the mutation environments assumed for the purposes of this work explicitly.

Mutations are assumed to occur in the following environments:

• Definite articles cause mutation of the following element.
• Prepositions, conjunctions, relative pronouns, and similar functional clitics cause mutation of the following element.
• A combination of a noun and an attributive adjective results in the mutation of the second element, whether the word order is noun-adjective (regular order) or adjective-noun (poetic). E.g., ‘a large ship’ would normally be rendered as cair dhaer (mutation applied to the adjective), but it could also be poetically rendered as daer gair
• The direct object of a verb undergoes soft mutation, regardless of its position relative to the verb.
• The verb undergoes soft mutation if it immediately follows its subject, unless the subject is a complex phrase (i.e. the subject phrase contains anything beyond a noun, a definite article, and a possessive pronoun).
• Adverbs mutate the following element if they are procliticised, and undergo mutation if they are encliticised. In practice this means that adverbs participate in mutations unless especially emphasised, in which case they neither mutate nor cause mutation.
• A prepositional phrase is mutated (i.e. the preposition itself undergoes mutation) unless it is sentence-initial or unless it is preceded by a caesura (as in e.g. lines 3 and 4 in stanza 15).

The type of mutation applied in a given case is determined by what the mutation is casued by. Mutations caused by articles and clitics depend on the element that triggered them; other mutations default to soft mutation.

(The above is mostly in line with Tolkien's discussion of mutations in CEA. The treatment of adverbs and prepositional phrases is based on Ellanto's own analysis of the corpus and the origin of Sindarin mutations.)

3.2 Definite Prepositions

This work employs a paradigm of definite prepositions based on the assumption that prepositions whose base forms end in a vowel have special definite forms, whereas prepositions that end in a consonant do not. The sole exception is the dative preposition an, which has a definite form (attested) despite ending in a consonant. Only the following three cases occur in this work:

• mi ‘in’ - def. sg. min (mixed mutation), def. pl. min (nasal mutation).
• po ‘on’ - def. sg. poe (soft mutation), def. pl. poen (nasal mutation).
• an ᴅᴀᴛɪᴠᴇ - def. sg. eni / ‘ni (soft mutation), def. pl. enin / ‘nin (nasal mutation).

3.3 Active Participles

This work assumes that present active participles are formed from the CE suffix ✶-ilā. This does not fit with the majority of attested active participles in Sindarin, which were formed with the -ol suffix, since -ol cannot be derived from ✶-ilā. Instead, the participial suffix is expected to be -el. This, in fact, is attested in Sam's invocation in LotR: tiriel < tiria- (in contrast to the past active participle tíriel < tiri-a in A Elbereth Gilthoniel).

The derivation of participial -el < ✶-ilā predicts a morphological difference in the pluralisation of the active participles of basic and derived verbs. In basic verbs, the e vowel in -el results from the raising of i as a consequence of a-affection; consequently, this vowel is expected to become i if the participle is pluralised, since pluralisation removes the effects of a-affection. On the other hand, in derived verbs the e vowel in -el is the result of the reduction of the ae diphthong that arises from the older ✶ai diphthong, itself the result of suffixing ✶-ilā to -a-; consequently, this vowel is not expected to change if the participle is pluralised. This difference is illustrated below:

• Basic verb: e.g. col- > sg. pres. part. colel < *kolela < ✶kol·ilā, pl. celil < *kœlili < ✶kol·ilī.
• Derived verbs: e.g. danna- > sg. pres. part. dannel < *danθael < ✶danta·ilā, pl. dennel < *denθael < ✶danta·ilī.

3.4 Genitives

There are two types of genitives employed in Sindarin: prepositional genitives and genitive constructs. These are assumed to have distinct syntactic behaviours as follows.

Prepositional genitives are usually formed with the preposition nan (triggering nasal mutation), unless the genitive noun is the object of an action expressed by the head noun, in which case an (triggering mixed mutation) is used, as in e.g. dagnir ‘ni thlûg (st. 14 l. 5).

Prepositional genitives are syntactically flexible and formed of two independently bracketed units. This has the following implications: (1) a prepositional genitive phrase GP can be rendered as either A nan B (e.g. st. 20 l. 5, i·mbelaith nan Amar) or (non-standardly) as nan B A (e.g. st. 7 l. 6, na thessais nestad); (2) the elements of GP can be freely split between lines (e.g. st. 3 ll. 1-2, thamas | na thobas ñgilel); (3) both A and B are full-fledged nominal phrases that can contain articles, adjectives, adjunct prepositional phrases, etc., as required (e.g. in st. 20 ll. 1-2, the second element in [e·linnad] na [lae i·nethl] is itself a genitive construct).

Genitive constructs are syntactically rigid. They are composed of two nouns in apposition with no intervening preposition between them: GP = A B. The elements of the construct are strictly ordered, such that A is the head of the construct and B is the genitive noun. The B element of a construct is normally not mutated, unless it is the object of an action expressed by A, in which case it is lenited (e.g. Dagnir ‘Laurung).

The genitive construct is treated syntactically as a single unit and splitting its elements between lines is seen as unnatural. Only the second element may bear a definite article, which, in turn, implies the definiteness of the entire construct (e.g. eryd heleg ‘mountains of ice’, eryd e·cheleg ‘the mountains of [the] ice’); if the second element is a proper noun, the construct is definite by default (e.g. st. 13 l. 3, naur Surth ‘the fire of Surt’). Normally an adjective will follow the entire construct regardless of which of its elements it refers to (but it must agree in number with the relevant noun).

3.5 Future Tense

In keeping with LVS 14 (PE22/165-8), two types of future formations are assumed in this work: the inflexional future, formed with the suffix -atha, and the auxiliary future, formed with the auxiliary tole followed by an infinitive (without mutation). Similarly to the system presented in LVS 14, but not identically to it, the inflexional future is used for volitional actions, and the auxiliary future is used for non-volitional actions.

The distinction is most clearly illustrated in st. 13 ll. 7-8:

In line 7 the future is used with a generic 3rd person plural subject to imply passive future (cf. i sennui Panthael estathar aen ‘who should be called Fullwise’), ‘shall all be ended?’. In this context, the end of everything may be brought about by the actions of Fenrir (‘the Wolf’), Surt, and Jǫrmungandr (‘the Dragon’), who desire the death of the gods and the destruction of the universe; therefore, it was deemed fitting to render this with the inflexional future, since the action is volitional on their part, even though it is presented as a passive statement ‘all be ended’. (Note: Here ilu is the object of the action, marked by the lack of lenition on the following verb, as verbs only lenite immediately following their subject or some proclitic particle.)

Contrast this with line 8, where the auxiliary future is used because the world fading is inherently non-volitional.

Note: Please do not use this work to train AI.

© 2026, Ellanto.

This work is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-sa/4.0/ or send a letter to Creative Commons, PO Box 1866, Mountain View, CA 94042, USA.