Heill Konráð!

> We may also have lost some ON inherited

meters, due to the faulty and incomplete preservation of the older
tradition. For example, Snorri quotes 2 lines from a poem called
Heimdallargaldr (níu emk moeðra mögr : níu emk systra sonr), which
meter I cannot identify. Perhaps Snorri could not, either.

Or could it be the last two lines of a 'galdralag' strophe? The metre
fits, it has the characteristic repetition, and that would agree with
the name Heimdallargaldr.

> This syllable loss creates some exceptional effects in the surviving

ON material originating from PN times, such as the verse I quoted in
my last post. Its strange to think that so much of the specifically
ON and charming character of extremely short lines is really just a
result of syllable loss in traditional material, causing speakers to
re-analyze/re-interprete an old meter like ljóðaháttr along new, and
Germanically speaking, radical lines. It works in ON, but its story
must have been a long, slow trip through the jungle ;)

Interesting that all the other Germanic traditions tended to develop
the inherited verse form in the opposite way towards a looser style
with more unstressed syllables. Comapare also the long line of the
Middle English alliterative revival, which even allows an optional
extra beat in the first halfline. Another feature which seems
characteristic of ON verse in contrast to the West Germanic tradition
is the use of stanzas. Old English verse is often impossible to
divide like this because of long sentences, enjambment, and so on, but
Chambers in his study of Widsiþ took the view that the few OE poems
that do show end-stopped lines (with the exception of some "monkish
effisions written under the influence of ecclesiastical Latin") are
probably the most traditional and "purely Germanic" of the lot,
Widsiþ, the Rune Song, the Charms, Deor (the other OE poem with a
refrain):

"...it can hardly be a coincidence that the four poems, the subject
matter of which carries us back to the most primative period, should
agree in the retention of a primative form. They may have been
modernised and interpolated: we know that the Rune Song and the Charms
have been largely rewritten: but all four seem to have their roots in
an age more primitive than that which rejoiced in the rhetorical
interlaced periods of Beowulf and Genesis: an age when the tendency
inherited from the strophic poetry, to make the sentence end with the
line, was still strong."

Wulf and Eadwæcer, with its songlike and ljóðaháttr like features, and
Deor can both be divided into discrete stanza-like sections -- even if
there was no refrain to mark them off. But in both of them the
sections aren't perfectly regular in length, although there are
patterns. Counting halflines here:

Wulf and Eadwæcer: 4+1, 6+1, 8, 6, 2+1, 2+1.
Deor: 12+2, 10+2, 6+2, 4+2, 12+2, 14+2.

Alaric Hall divides another OE poem, the Wife's Lament more regularly
into sections of 10, 24, 24, 24 and 24 halflines. Does the Old Norse
evidence suggest that the regularly strophic 8 (half)line verse arose
out of a less rigid tradition, with sections of only roughly the same
length as in Ynglingatal?

*

> I suspect, but can of course not show, that Go. was very close to ON

in this respect, especially as relates to poetry. ungr vas ek forþum
is somewhat more poetic than ek vas ungr forþum, but still far from
being unnatural. One can say it that way, but it rings differently.

Yes, there's no doubt that Gothic was as free if not freer than Old
Norse in changing word order for style and emphasis. I was just
wondering here whether there would be any preference for putting the
subject pronoun immediately after the first fronted constituent, or
after the finite verb (juggs was ik faúrþis; juggs ik was faúrþis), as
there seems to have been a difference in syntax here between Old
English, on the one hand, and all the other reasonably well attested
early Germanic languages. This is one of the big issues in the debate
among syntacticians about the origins and significance of the verb
second word order in Germanic; so it would be amusing to learn that it
was no issue at all for actual speakers of the language. The two best
Gothic examples I know each follow a different practice: þanuh biþê ût
usiddjêdun eis; iba þank þû faírháitáis -- so, as I say, unless we
find anything else to contradict it, you're probably safe here to
follow the Old Norse word order.

> *einn vrataþa (ek) could occur in ON, and hverfa can mean have some

meaning (in eddic verse, etc.) as in Go., but the fór alliterates
here with forþum, and fôr occurs in Go., meaning the same thing

Interesting... I assumed that the headstave was 'einn', alliterating
with 'ungr', making this Sievers type C.

ungr vas ek forþum
/ x x / x

fór ek einn saman
X x / / x

A Gothic line 'fôr ik áins samana' would I think have the same
structure, but with resolution of -ana in the last dip. I guess
another alternative might be 'jah fôr ik áins' (Sievers type B, with
crossed alliteration--Vowel F, F, Vowel--is that the right name for
it?). That keeps the vocabulary the same but avoids the unattested
idion. But point taken: we certainly don't know enough about Gothic
to rule out the possibility that the idiom *'áins samana' was alive
and well there, either in everyday speech or poetry.

> Another eery similarity is the +/-wisan/wesan - essentially, you can

jsut add it whenever you want to, or if the meter works better with
it. But about -dative mis: hmmmm... Go. has another similarity when
it comes to verbs and the cases they steer - I've never hit the wall
on this issue, as there is hardly a wall to hit. Without any doubt,
verbal case-governance, and the survival of reflexives in -sk, is
one of the most conservative features of ON, which goes amazingly
far when trying to translate to Go. - but again, the devil is in the
details. Here the detail is is Go. can omit the dative, passive
subject with þûhta or not. About the +/-ik/ek and wisan/vesa: these
are implied in the ON, and þóttumsk is somewhat ambiguous here, as
it was early levelled into the first person in ON.

Yes, Old Norse and Gothic agree on most of the fundamental things.
Just a few quirks such as 'hilpan' + gen. in Go. and in OE, versus ON
'hjalpa' + dat. Don't quote me on this, but I think the genitive was
possible in older forms of German, although the dative has taken over
now. The middle voice in Gothic works in very much the same way as in
ON except that the reflexive pronouns aren't assimilated to the verb
endings, and Go. also has the inherited morphological passive which
sometimes encroaches on the role of the ON middle voice. The biggest
devil in the detail here is exactly this question of how the reflexive
works in expressions like "I think myself to be..." There are a
couple of really curious examples in Gothic where, unlike ON, the
reflexive pronoun seems to be added to the end of the *infinitive*
verb, rather than the finite verb that it's subordinated to. This is
the clearest:

wenja...swikunþans wisan uns = EPIZW...PEFANERWSQAI = Lat.
spero...manifestos nos esse `I trust we are made manifest' (2Cor 5:11).

> Do we have any examples of full mis/þis/etc. þûhta?

Oh yes, this is very common (or as common as anything can be in the
limited corpus), except that in main clauses at least the rule seems
to be that oblique experiencer pronouns always follow immediately
after the verb in those instances where there is no Greek model to
immitate. The one example I know of where a dative experiencer
pronoun precedes the verb is in a subordinate clause: in þizei mis
galeikaiþ in siukeim = DIO EUDOKW EN ASQENEIAIS `therefore I take
pleasure in infirmities' (2Cor 12:10).

>> I'm fairly sure the dative pronoun would be left out even though

that's potentially ambiguous with the meaning "to
seem".

> I think so, as to whom is it seeming?

Yeah, to whom it seems or who considers this to be so. One way to
express the idea of thinking something about oneself unambigiously is
to use the passive subjunctive and the relative particle 'ei', e.g.
þugkeiþ im ei in filuwaurdein seinai andhausjaindau DOKOUSI GAR hOTI
EN THi POLULOGIAi AUTWN EISAKOUSQHSONTAI `they think that in their
verbosity they will be heard' (Mt 6:7). So maybe: 'ik þûhta ei ik
áudags sijái' -- hmm, so much for the poetry ;-) Would Old Norse ever
use this sort of construction: mér þykkir at ek sé? I just looked on
Google and found a grand total of two examples of Modern Icelandic
'mér þykir að ég' + subjunctive -- compared to 18 300 examples of 'ég
þykist'!

> On the other hand, do we have examples to the contrary where dative,

passive subject is clearly implied? ON also has some implication
habits: es ek annan fann (where the relative particle es implies
elliptical þá - in fact, one could just as well say 'þás ek annan
fann' - but Go. is too old for this kind of playing around with the
particle, which is still 'î/ei in Go. and is he, where ON shows hann,
replacing PN ez, which just becomes the relative). One side note for
langauge-history interested is that at least one other case of *ez
survived in ON - in fact, its still lurking in Modern Icelandic even:
þareð (eð < et, the neuter of the obsolete *ez 'he' ;) Go. þarei. I
think this one very odd and unusual, as 'þar es' is typical in ON, but
*þar et, lurking in the spoken language somewhere looks to have made a
comeback after es > er rendered the awkward **þarer....

Contrariwise, in Gothic the relative particle can be ommitted with
'þan(ei)' (see Streitberg: Gotische Syntax, para. 359: "...so ist dies
ein wetrvoller Beleg, dass auch im Gotischen wie in den übrigen
germanischen Sprachen die Demonstrativa allein in relativischer
Geltung vorhanden waren"), compare OE þá "when" which is distinguished
from OE þá "then" by the fact that the relative particle doesn't
trigger verb second word order.

> > For the sake of the metre, maybe either: 'áudags ik þûhta' or

change it to a B line with the pronoun and finite verb unstressed at
the beginning: 'ik þûhta áudags wisan'?

> I want the idioms as close as possible, and do not want to change

the position of áudags, as it hurts the poetry, I think. I'll take
the consequences after áudags, if need be: *þûhta mis/ik þûhta/þûhta
ik/ik wisan - whatever is good and natural Go.

In that case, I guess the simplest way would be 'áudags ik þûhta'.
The Gothic Bible has no qualms about adding the dative pronoun where
no such pronoun is found in Greek, so I think this must be
significant, the fact that it was omitted.

> I examined gaman, discovering that OHG OS OE all agree with ON here,

essentially, which lead me to: 1)the Go. meaning 'joy' is unattested
or 2)the ON meaning 'comrade' is unattested, or was lost during the
transaition PN>ON in favour of just 'joy' or 3)the Gothic meaning
developed from an older meaning attested in the other languages, if
the orignal one ever were lost. I elected to solve the problem by
allowing by imaginary translator to stay idiomatic, as usual - in
the worst case scenario, gaman can only mean 'comrade' in Go., in
which case the line makes just as much sense anyway, and it still
carries the same meaning, I think. I think the reader will agree.

This is a good excuse to post a couple of links I came across recently
on the subject of translation between closely related languages.
They're both by Pétur Knútsson.

http://www.hi.is/~peturk/INDICES/Intimations.pdf

See especially the sections comparing the Old English and Old Saxon
versions of Genesis and Halldóra B. Björnsson's Icelandic translation
of Béowulf. We actually have Eysteinn to thank for inadvertantly
sending me in this direction, as I came across it while searching for
stuff on Steinn Steinarr's poem 'Ég geng í hring', of which it has an
English translation.

http://www.hi.is/~peturk/INDICES/index.htm (1984 Um þýðingu Halldóru
B. Björnsson á Bjólfskviðu. Skírnir, tímarit Hins íslenska
bókmenntafélags: 223-244).

This covers some of the same ground, but deals more specifically with
the Icelandic translation of Beowulf, as the title suggests. He gives
examples of Halldóra and the anonymous Old English translator/adapter
of the Old Saxon poem Genesis each adopting exactly this strategy you
descibe with 'gaman' (well, the back up strategy in the event that it
would have been interpreted as a different word by a Gothic
audience!), of matching the sound and the spirit and the poetic style
sometimes at the expense of the precise meaning of a particular word.
It's also interesting to see how uniquely suited Icelandic is to
matching the style of Beowulf, compared to say Modern English where
there have been so many changes in vocabulary and idiom.

LN