A. Ljóðaháttr guidelines

Some thoughts on editions of poems in ljóðaháttr meter

I. The meter

As far as the meter is concerned, ljóðaháttr is difficult, and there is (still) no scholarly consensus on whether the third, full line has two or three lifts. The first two half-lines are patterned according to fornyrðislag, but the first half-line is sometimes hypometrical (= kviðuháttr: Hm 3/1 Eldz er þǫrf) or hypermetrical with three lifts (Hm 89/1 Bróðurbani sínom) while the second half-line is often hypermetrical and contains three lifts rather than two (HM 6/5 kømr heimisgarða til). Furthermore, in both of these lines, resolution is allowed on the second dip, which is otherwise prohibited in the other ON meters (but allowed in WGmc alliterative poetry): Hm 85/1 Brestanda boga; Hm 13/2 sá er yfir ǫlðrom þrumir. The third, full line appears (at least in my opinion) to contain either two (Hm 143/5 ec reist siálfr sumar; Hm 146/3 oc mannzcis maðr) or three (HM 4/6 orðz oc endrþǫgo; Hm 137/15 fold scal við flóði taca) lifts, but it is also patterned loosely on fornyrðislag (usually hypermetrical). Lines composed in ljóðaháttr is characterized by extended dips internally and in ancacrusis (i.e., the dips can contain more syllables than what we would expect from fornyrðislag): Hm 120/ 6 teygðo þér at gamanrúnom; Hm 45/4 fagrt scaltu við þann mæla). Another complicating factor are all the pronouns (ek, þú etc.) that appear in abundance, often cliticized on to finite verbs (heyrðu, skaltu etc.: Hm 119/3 nióta mundu, ef þú nemr; HM 130/8 fǫgro scaldu heita; Hm 138/1 Veit ec, at ec hecc). So: two lifts (or sometimes three) in the first line (e.g. Type B: Hm 145/1 Svá Þundr um reist; Hm 89/1 Bróðurbana sínom), two or three lifts in the second line (e.g. Type B: Hm 145/2 fyr þióða rǫc; Type BC Hm 146/7 oc sútom gorvǫllom), two or three lifts in the third, full line (e.g. Type B: Hm 146/3 oc mannzcis mǫgr; Type DB Hm 138/6 siálfr siálfom mér).

Other rules that govern the composition of poetry in other meters and are violated in ljóðaháttr are those governing word order (e.g. Kuhn’s laws) and resolution (which can be suspended or occur on the second lift). Hence it is clear that, under these circumstances, it is very difficult to establish rules for the edition of poems in this meter.

However, there are certain metrical principles that are clearly discernible.

1. First of all, the rules governing alliteration are rarely violated (one or two alliterating staves in the first line, one in the second, two in the third line, although there are occasionally three: Hm 43/6 vinar vinr vera). When they are, something is usually corrupt in the textual transmission. For example, Hm 125/5-6: þrimr orðom senna | scallattu þér við verra mann is metrical if þér and við are switched around scallattu við þér verra mann (Type BA), which is also corroborated by the syntax.

2. Although the meter is characterized by extended dips, a` la OS and OHG alliterative poetry, there are extremely few violations of the metrical patterns at the end of the half-lines in lines 1 and 2 (except that one needs to account for resolution in metrical position two). Hence the extrametrical syllables that are found in the internal dips and in anacrusis are not found at the right edge of the vers. That is why I emended skaltu to skalt and veittu to veitt in Hsv.

3. The third, full line always ends in a short monosyllable or a disyllable with a short stem. This is a ‘golden’ rule (just take a look at Hm in NK). If this rule is violated, something is more likely than not corrupt in the textual transmission. From Hsv I gave the example syni sínum (where other mss have sínum syni), and, as I recall, once 1199^x has fyrðum where some of the others have the less common firum (which is metrically correct). I think that, if the meter can be restored by using variants from the other mss that are metrically correct, we must do so. However, if none of the mss yields a metrically ‘correct’ reading, we should not emend conjecturally, but point out the deficiency in the Notes.

2. Normalizations and emendations

I have already mentioned the extended dips that occur internally and in anacrusis in ljóðaháttr. Skj B and Skald attempts to make poems in this meter correspond more closely to fornyrðislag, and they do that in a number of ways. First of all, they tend to delete the negation eigi and replace it by the cliticized verbal negation -at. Secondly, they silently omit pronouns (cliticized and non-cliticized) that occur in the extended dips. In my opinion, we should not do that, and I have not done so in the Hsv edition. We cannot exclude the possibility that these adverbs and pronouns were added during the scribal transmission, but we don’t know that. However, when such a cliticized pronoun is located at the right edge of a line of Types B, E, or D4 (or an extended version of these types), I would emend by deleting the superfluous syllable (hence veittu > veitt, skaltu > skalt). This is not really an emendation, because we routinely do that in other, more regularized meters.

Secondly, in some cases an alternative form of a word will render the line metrical. For example, 1199^x uses both aldri and aldrigi (skaltu aldri) in the second line, with alliteration on <a>. In this case the line could be taken as a line of Type A, but the placement of the alliterating stave violates the rules of alliteration (alliteration must fall on the first and not on the second lift). If we go with the extended form of the adverb (skaltu aldrigi) the line is metrical (Type C). I suspect that the short form is used because it creates a trochaic line, written by a scribe who was unfamiliar with the rules of alliteration. In this case, I would use the longer form.

Thirdly (and very frustrating), is the ms. transmission, as I mentioned in my email message. We (or our contributors) have dated the poems to C13th, and we settled on post 1250 normalizations. However, the mss use C14th and post C14th syntax and orthography. I think we must either normalize to C13th whole-scale in accordance with our current practice in Vol. 7, or reproduce a C14th version of the poems as they are rendered in the mss and according to our C14th forms of orthographic and syntactic guidelines. It is my understanding that we have decided on the former. If so, I do not see any reasons for noting such normalizations by italics or ‘*’ in the Text, or by justifying them in the Variant readings (‘þótt: so X, þó 1199^x’; or ‘þótt: þó 1199^x’). The instances I have seen usually involve omissions of at (svát, þótt, etc.), or omission of the rel. particle (er). These are not emendations, but standard normalizations. I think we can put our section on C14th normalizations to good use here, by pointing out that such orthographic and syntactic features that occur in the later mss (and are noted in that section) will be silently emended in Hsv and Sól?

References