In the last day I had the opportunity to take a more careful look at E1a
and E1b on PE16:98-104. In november I started to analyze "Earendel"
as we knew it from MC:216, mainly as an "exercise". I'm quite
satisfied with the results, as the publication of PE16 with its
glossaries, word-lists, declensions and conjugations has given more
tools for the understanding of the late 20s.
However, there's a point I'd like to discuss here with all of you. On
the analysis of _kalmainen_ in line 7 on PE16:102 we read: "the fact
that the word is translated as plural 'in the lights' suggests an
interpretation of its construction as _kalma-i-nen_, with a variant
plural suffix _-i_ used instead of _-lí_ for the sake of the metre.
And these considerations suggest that _-nen_ had already emerged in
Tolkien's conception as a suffix of the instrumental case. In the third
of the Qenya Declensions Tolkien notes the archaic and rare use of the
form _kiryainen_ as an instrumental plural to a variant singular
_kiryain_." I'm of the opinion that the first assumption (_kalma-i-
nen_) is improbable, because we have no evidence of such a possibility
in earlier and later texts: in EQG there is _-inen_ (a partitive ending with
no number indications); in QD2 [PE16:112] we find the sg. _-inen_ and
the pl _-línen_; in QD3 [PE16:113-115] we find the sg _-inen_ and the
pl _-línen: and the alternative/archaic forms sg._-in_ and pl _-inen_;
in the _Entu, Ensi, Enta_ Declension (VT36:8) we find both the sg.
forms _-in_ and _inen_ (they seem alternative); while in the later
(first half of the 1930s, before the Etym.) Bodleian Declension (VT28:8)
we find exclusively a sg. _-in_ and a pl. _-inen_ (and we see instead
the emergence of a "second plural" in _-li_). The evidence suggests the
slow emergence of a sg./pl. pair _in/inen_ (that fits well in scheme
of a recurring _-n_ plural ending in cases; for a discussion of a
supposed development, see VT28:22), which finally became the only one
in the Bodleian Declension (hereinafter BD), rather than an emergence of
_-nen_, which is clearly a later idea. Then, _kalmainen_ should only be
either a QD3 archaic plural, according with Tolkien's own translation
'lights', or a singular Qenya word translated as an English plural.
Metrical evidence suggests that Tolkien used the _poetic_ (PE16:115)
form _kalainen_ to maintain the metre (and in that sense he was allowed
to do so, and _kalmainen_ is clearly described as an _archaic and poetic
form_); while in line 8 the same metre "required" him to use the
"normal/common" instrumental plural _-línen_.
The emergence of coexisting instrumental sg. endings _in/inen_ both in
QD3 and _Entu, Ensi, Enta_ (hereinafter ED) that led to exclusive _-in_
(with _-inen_ plural as in QD3 and a _-línen_ whose interpretation
of "emergence of the second plural" however is not better supported by
evidence than "normal vs. poetic plural", even if more probable
given the whole external evolution of Qenya declensions) in the later
BD, suggests that QD3 and ED are close in date. Convergences between
QD3 and ED are not limited to the instrumental.
 In ED line 8 (always supposed to be the "partitive" case in
relation with BD _-ika_) has _-iko_ like the partitive _-iko_ in QD3;
while in QD2 the partitive ending is _-ita/-ika_ (which reappears
 QD3 along with the "short poetic instrumental" _in_ mentions
also the "short poetic ablative" _-al, -el, -ul, &c._ and the "short
poetic inessive" _-as, -es, -os, -is_, which are found also in ED, the
inessive/locative in line 9 (_entosse, (-os)_, etc.), and the ablative
in line 10 (_entollo, entul_, etc.).
 QD3 mentions also a "poetic partitive of obscure origin [
_-hta_", which is found also in in ED in line 12 (_entuhta_ etc.).
 ED line 13 seems to have a relationship with QD3 "adjectival".
 QD3 dual forms (PE16:114) seem also to converge with the roman-
numbered lines in ED: _kiryant/assant; kiryat/assat, kiryu/assau (older
assu), kiryur/assor (assur), kiryaimet/assaimet, [kiryant]as/assanta
(as), kiryallot > [kiryall]ut/assallo(ut), kiryandos/assandos_.
Its difficult to say which one is older, and they were probably
composed at the same time; but ED has been amended after QD3,
making it slightly older. However, the only secure assertion is
that they are contemporary. 3 particulars can be observed.
[A] _assu_ and _assur_ have been chosen as definitive (by underlining
them) in a second time, while in QD3 they are first written.
[B] _-as_ endings and _-ut_ endings seem to have been written directly
in ED (in the photocopy theres no difference, and the fact that on the
4th row the two forms has been written in full, suggest an original
writing of both forms), while in QD3 they are clearly a later emendations.
For a better understanding, its necessary to have a closer look at ED Ms.
[Marq.3/8/2:5 verso], and maybe someone who has access to the original
version can confirm or deny this interpretation, by checking if a different
pen or ink has been used.
[C] in QD3 one theres no evidence for the late evolving inessive _-nye_
in ED; in ED originally there was written _entasse/entas_ (VT36:26) and
only in a second time, after the supposed composition of QD3, the 9th
line was modified to _entasse(-as)/<entas>entanye_.
-- Ugo Truffelli