Happy new year to all the Lambengolmor!

I am trying to find out what Tolkien meant in one of the paragraphs of
the Gnomish Grammar, where the historical phonology of Gnomish plurals
is discussed (PE11:10), and I would like to know if any of you have a
clearer view of it.

There is a table with the singular and plural forms of the
nominative/inessive, genitive/ablative and dative/allative cases. And
then the origin of all those forms is briefly discussed, comparing them
with their Qenya cognates.

Of the nominative/inessive forms: _-in_ or _-th_, is written:

"_-in_ is a double plural _-î_ = Q _i_ + _-n_, cp. adjectives."
"_-th_ is original and same as Q _-r_"

But then the following note is provided:

"The existence in G. of an _-r_ plural sign in verbs has given rise to
conjecture [...] that G _-th_ does not represent Q _-r_ but that _-r_
is a true plural ending (i.e. _r_ liquid) and _-tt_ = Q _-t_ dual from
_-tt@_ [@ = schwa] a dual ending = _-nt@_. This is possible."

That is the paragraph that I don't entirely understand.

The last part ("_-tt = Q _-t_ dual ...") does not say anything of Gn.
_-th_, the object of the previous part of the note. Had it been "_-th_ =
Q _-t_ dual" instead, I would have understood that those who said that
Gn. _-th_ does not represent Q. _-r_, believed that its origin was the
same as Qenya dual _-t_ (both < _-tt@_).

If that is the implied meaning, _-tt_ could be explained (if it is not
just a mere slip) as a prehistoric form of _-th_, like in _gôtha_
'possess, have, hold' < _iotta_ [semivocalic _i_], or _nith_ 'wax_ <
_nęgitte_ (PE11:42, 60).

But what follows is also somewhat ambiguous: "_-tt@_ a dual ending =
_-nt@_". Both _-tt@_ and _-nt@_ seem to be old dual endings, but what is
their relation, and why is _-nt@_ introduced in the discussion? Perhaps
because they are genetically related (_-tt@_ < _-nt@_ by assimilation,
or the opposite)? Or because _-nt@_ was still active as a Gnomish dual
in some cases? Notice that _-tt@_ could not have that function if it
yielded Gn. plural _-th_, and among the "commonest old duals" below in
PE11:10 there is _hunt_ 'the nose (originally nostrils)', from "old
_-nt_ ending" (cp. PE11:50). It is even possible that the gentive dual
_-wint_ (PE11:11) was formed as a "double dual" = _-wi_ + _-nt_.
However, this would not mean that _-nt@_ had no reflex in Qenya. At
least in verbs and pronouns, _-nt_ was also a dual mark (cf. PE15:46,
s.v. _munt_, _lunt_).

Helios