1. Gáttir allar
áðr gangi fram,
um skoðask skyli,
um skyggnask skyli,
því at óvíst er at vita
hvar óvinir
sitja á fleti fyrir. (text from Evan's edition)
Here is David A.H. Evans highly informative commentary on strophe 1
in his edition of Hávamál (Viking Society for Northern Research,
University College London, 1986), (and as I have copied the text
below, any mistakes are my own):
'This strophe is quoted near the beginning of Snorri's Prose Edda,
without attribution; see above, p.2. Only the Utrecht ms has line 3;
Worm's ms lacks at vita in 5, and the Uppsala ms has the awkward
Skatnar allir áðr né gangim fram as 1-2 and the pl. fletjum for
fleti in 7. The text in Snorri is evidently somewhat corrupt, though
fletjum is perfectly possible (as in st.35).
1-4 Although the general sense is clear, the construction is
disputed. Some editors take gáttir as acc. object of skoðask um and
skyggnask um, but this is hardly right, since these verbs are
equivalent to skoða (skyggna) um sik and cannot have an object; they
are of the same type as sjásk um, lítask um, leitask fyrir etc., see
Nygaard 2, ch.154. (Skyggnask um occurs in prose, always
intransitively; cp. Fritzner 2 s.v. skygna.) Others understand
gáttir as nom.; this entails taking the infinitives as passives
(with um as the particle). So FJ. It has been denied (e.g. Olson
540, Lindquist 2,1) that refl. with passive sense occurs in the
Poetic Edda, and indeed it is true that in Norse as a whole this
usage is common only in the Latin-influenced 'learned style' and is
otherwise largely confined to a few verbs such as spyrjask, fásk,
byggjask (Nygaard 2, ch.161); yet there are a few Eddaic instances
which come very close to passives (öll muntu lemjask Helg. Hj. 21, á
gengusk eiðar Vsp. 26) and early skaldic verse also supplies
examples (eyðisk land ok láð and tröddusk törgur, both in Eyvindr's
Hákonarmál, cp. FJ 5, 275). This is certainly therefore a
defensible interpretation, but it is perhaps safer to take the
infinitives as intransitive, with gáttir as acc. object of gangi;
for this construction cp. Þorkell ok þeir báðir förunautar gengu út
skyndilega aðrar dyrr en þeir höfðu inn gengit Hkr. ii 166 and other
instances in Nygaard 2, ch.96.
7 sitja...fyrir probably 'are present' (as in 133) rather than
specifically 'lie in ambush' (as von Friesen), though sitja fyrir
can have this sense with a dat. object. CPB 461 insists that gangi
fram must mean 'go to the door' (from inside), as indeed it commonly
does; but this involves the impossible 'lurk round (italics) one's
house' for the last line, and Snorri' use of the strophe shows that
he took it to refer to entry from without.'
My main question about this strophe is about the use of 'um' here.
According to FJ (Finnur Jónsson), the unstressed 'of' must occur
here instead of 'umb' for 'skoðask skyli' or 'skyggnask skyli' to
bear the poetic stress (um, of, and even uf are written variously in
CR, sometimes as meaningless particles, remnants of lost prefixes,
sometimes with one or other meaning). In older texts, such as the
Christian Homily Books, the language is more archaic than in the
text of Hávamál preserved in CR, and both 'umb' and 'of' occur in
places where the later language of CR might have 'um' (although 'of'
and 'uf' also occur). During the earliest Christian period (of which
the Homilies bare witness), it seems that 'um' did not occur at all.
Instead, a distinction between 'umb' (around, about) and 'of' (over;
too much; archaic particle, remant of lost prefixes; etc.) was
maintained. FJ, accordingly, renders 'um' as 'of' here. My question
is: is the sense 'about, around'(umb) absolutely necessary here, or
are the expressions 'of skoðask' and 'of skyggnask' (or inversely,
'skoðask of' and 'skyggnask of') acceptable here? In other words,
which is likely to be the original form here?(keep in mind the CR's
text is from about 1270, much younger than the Christian Homilies,
for example). Another thing to keep in mind is that if the oral
tradition behind the version of Hávamál given in CR was contained
within a single dialect-variety of Norse (and the earliest texts in
Iceland confirm that at least several were present before the so-
called 'dialect levelling' took place), then there would have been a
spoken standard regarding the use of 'of' and 'umb', regardless of
other dialects used them - in other words, it is possible that 'of'
had completely taken over, even in prefixes where 'umb' would seem
to be required, are being only preserved in stressed positions where
the sense 'about, around' was absolutely required. And this brings
us back to FJ's issue about the stress here. So which is it? This is
a tough question, and I do not expect a final answer, but would be
interested in any comments, suggestions, opinions, or views about it.
On a side note, I think both 'á fleti' and 'á fletjum' are fine. If
I had to choose, I might choose the plural, just because it gives
more benches for the 'óvinir' to sit on and otherwise occurs in HM,
as Evans points out.