The Harbarthsljoth is found complete in the Codex Regius, where it follows the Skirnismol, and from the fourth line of stan{z}a 19 to the end of the poem in the Arnamagnæan Codex, of which it occupies the first page and a half.

The poem differs sharply from those which precede it in the Codex Regius, both in metrical form and in spirit. It is, indeed, the most nearly formless of all the Eddic poems. The normal metre is the Malahattr (cf. Introduction, where an example is given). The name of this verse-form means "in the manner of conversation," and the Harbarthsljoth's verse fully justifies the term. The Atli poems exemplify the conventional use of Malahattr, but in the Harbarthsljoth the form is used with extraordinary freedom, and other metrical forms are frequently employed. A few of the speeches of which the poem is composed cannot be twisted into any known Old Norse metre, and appear to be simply prose.

How far this confusion is due to interpolations and faulty transmission of the original poem is uncertain. Finnur Jonsson has attempted a wholesale purification of the poem, but his arbitrary condemnation of words, lines, and entire stanzas as spurious is quite unjustified by any positive evidence. I have accepted Mogk's theory that the author was "a first-rate psychologist, but a poor poet," and have translated the poem as it stands in the manuscripts. I have preserved the metrical confusion of the original by keeping throughout so far as possible to the metres found in the poem; if the rhythm of the translation is often hard to catch, the difficulty is no less with the original Norse.

The poem is simply a contest of abuse, such as the early Norwegian and Icelander delighted in, the opposing figures being Thor and Othin, the latter appearing in the disguise of the ferryman Harbarth. Such billingsgate lent itself readily to changes, interpolations and omissions, and it is little wonder that the poem is chaotic. It consists mainly of boasting and of references, often luckily obscure, to disreputable events in the life of one or the other of the disputants. Some editors have sought to read a complex symbolism into it, particularly by representing it as a contest between the noble or warrior class (Othin) and the peasant (Thor). But it seems a pity to take such a vigorous piece of broad farce too seriously.

Verse-form, substance, and certain linguistic peculiarities, notably the suffixed articles, point to a relatively late date (eleventh century) for the poem in its present form. Probably it had its origin in the early days, but its colloquial nature and its vulgarity made it readily susceptible to changes.

Owing to the chaotic state of the text, and the fact that none of the editors or commentators have succeeded in improving it much, I have not in this case attempted to give all the important emendations and suggestions. The stanza-divisions are largely arbitrary.