> ok brotnaði sverðit undir hjöltunum

http://www.vikingsword.com/petersen/ptsn158x.html
http://www.vikingsword.com/petersen/

efra(øfra)/eptra hjaltit (the upper/further-back hjalt)
= the knob, boss or pommel on the end of the sword

fremra hjaltit (the further forward hjalt)
= the cross-piece or guard between the hilt and the blade

meðalkafli (between piece, middle bit)
= the grip or hilt; the bit you would hold between the two 'hjölt'

brandr = swordblade

oddr = point
blóð-refill = point (of a sword)

According to Fritzner, the plural 'hjölt' can either mean (1) the
cross-piece alone; or (2) the pommel and the cross-piece (the singulat
'hjalt' referring to one or the other of these). He cites the example
from Njáls saga under the first definition. But Cleasby/Vigfússon and
Zoega only have the second definition, pommel and cross-piece, and CV
has a different version of this line from Njáls saga, with the
singular noun:

sverðit brotnaði undir hjaltinu

http://lexicon.ff.cuni.cz/png/oi_cleasbyvigfusson/b0265.png

Either way, I would have thought the most obvious interpretation is
that the blade has broken just below the guard. The adverbs visualise
the sword from the point of view of someone holding it (eptra,
fremra), or else pointing downwards (efra). The latter way of
visualising it suggests that 'undir' means that the break is occuring
on the blade itself just "below" the hilt. Some similar examples:

brast sverðit undir hjaltinu ok fór brandrinn grenjandi niðr í ána
"the sword broke below the guard and the blade went howling into the
river"

brast sverðit þegar sundr undir hjöltunum
"the sword broke at once below the guard"

ok þá brast sverðit við hjaltit
"and then the sword broke at the guard"

I wonder if there's any evidence from archeology or reconstructions
that this was a weak point, or if it was just a storytelling convention.

LN