Another interesting datum pertaining to lenition or
lack thereof after S./N. _morn_ in compounds occurs
in Tolkien's early "Noldorin Word-list", published in
_Parma Eldalamberon_ #13. Under the entry for
_magli_ 'a bear' (lit. 'honey-eater') we find _Gormagli_
'Great Bear', a name given to Túrin by the Gnomes of
Nargothrond, who also called him _Mornvegil_ 'Black
Sword', = Q. _Mormakil_ (PE13:149). This same word-
list also gives N. _morn_ 'black, dark' (pl. _myrn_) and
N. _megil_ 'sword' (pl. _megiliath_) (p. 150).
The change of _megil_ > _-vegil_ in _Mornvegil_ is
particularly interesting given that the closely
contemporary Early Noldorin Grammar (also given in
Parma 13) states that N, M, L, R, S were _not_ affected
by mutation! Perhaps _Mornvegil_ in the early word-
list represents a dissimilative change of M > V for
easier pronunciation, rather than actual lenition or
mutation? The name _Gormagli_ (rather than **_Gor-
vagli_) in the same entry as _Mornvegil_ might argue
in favor of this point of view (the Qenya form is given
as _Oromatsile_, with Q. _ile_ 'honey').
What we really need in order to clarify what is going
on in N. _Mornvenniath_ vs. later S. _Mormegil_ etc.
is an analysis of attested _Noldorin_ mutations as
evidenced in the _Etymologies_ and closely contemporary
texts. So far as I know, no such study has been done --
the existing writings on mutation lump Noldorin and
Sindarin together, with predictably contradictory
results. Perhaps a member of this list will feel inspired
to compile and publish a study based on Noldorin alone.
-- Patrick H. Wynne