are Q voiced?
- In message #10 of elfling, Benct Philip Jonsson alias 'Melroch'
proposes that Quenya stops aren't divided into three categories:
voiced (only after /l, r, n/), unvoiced, and long unvoiced (only after
vowels); but only into two categories: weak and strong (that is, lenis
That is, not like this:
| after vowels | after /l, r, n/
voiced | - | <b, d, g>
unvoiced | <p, t, c> | <p, t, c/q>
long unvoiced | <pp, tt, cc> | -
but like this:
| after vowels | after /l, r, n/
weak | <p, t, c/q> | <b, d, g>
strong | <pp, tt, cc> | <p, t, c/q>
Benct specifies that the difference between the weak and the strong
stops is their length, but that additionally (if I've understood him
correctly), the weak stops are voiced after /l, r, n/. So he suggests
two pronunciations of the weak stops: either voiced or voiceless.
I'd rather suggest that there's but one pronunciation of the weak
stops: voiceless. Like this, Quenya stops would be the same as Finnish
stops (only that standard Finnish has /d/ from original /D/, according
to Harri Perälä and John Cowan in elfscript #1963 and 1968).
It's not uncommon that <b, d, g> represent unvoiced stops. This is
found in many languages, e.g. Icelandic, Danish, Chinese, or southern
German. But do we have any evidence on the pronunciation of <b, d, g>
in Quenya? I've only found one possible evidence: Appendix E of The
Lord of the Rings which says that "NG represents _ng_ in _finger_".
However, this indication is not specific to Quenya, but concerns both
Quenya and Sindarin, and actually most of the pronunciation samples of
app. E are for Sindarin. Is there more evidence?
If there's only two kinds of stops, why are there three kinds of
spellings (<b, d, g> vs. <p, t, c/q> vs. <pp, tt, cc>)?
Theoretically, all weak stops could be spelled with <p, t, c/q> and
all strong stops with <pp, tt, cc>. However, as Benct's pointed out,
this would lead to very odd spellings after /l, r, n/, whereas Tolkien
explicitly wanted to make Elvish spelling "not look uncouth in modern
letters" (app. E).
Theoretically, all weak stops could be spelled with <b, d, g> and all
strong stops with <p, t, c>. However, English people would naturally
voice the <b, d, g>, so this would lead to mispronunciation, whereas
Tolkien explicitly wanted to "represent the original sounds (so far as
they can be determined) with fair accuracy" (immediately before the
So I think that in spite of the three kinds of spellings there are
only two kinds of stops, the three kinds of spelling being but a
consequence of Tolkien's will to have an accurate and not uncouth
j. 'mach' wust
- Tolkien's own pronunciation of Quenya may be helpful. I will use Laurence J.
Krieg's short but insightful article "Tolkien's Pronunciation: Some
Observations" in "An Introduction to Elvish" edited by Jim Allan, pp. 152-159 . It
reproduces among others a phonetic analysis of Namárië (spoken and sung) made at
the Phonetics Laboratory at the University of Michigan, presented in the IPA (or
rather a variant: notably [y] is used where IPA usually has [j]) in a rather
We must account with some influence of English on Tolkien's pronunciation of
Quenya, but it does not affects the distinction between voiced and unvoiced
stops. In fact he regularly pronounces words like _andúne_, _Varda_, _unduláve_,
_sindanóriello_ with [d] and _lumbule_ and _imbe_ with [b].
So his spelling on that point reflects the phonetic realisation rather than
the underlying phonology in the theory of Benct Philip Jonsson.
In addition, it is also interesting to note that unvoiced stops are generally
unaspirated, with only three exceptions which are perhaps to regard as slips:
_únótinar_ with [th] and _caita_ with [kh] in the spoken version, untúpa with
[ph] in the sung version.
Language has both strengthened imagination and been freed by it. Who shall
say whether the free adjective has created images bizarre and beautiful, or the
adjective been freed by strange and beautiful pictures in the mind ? - J.R.R.
Tolkien, A Secret Vice
[Non-text portions of this message have been removed]
- In VT45:14 (s.v. 3EL-) we read :
"*_Helwe_ name of Teler-lord, N _Elwe_, always recorded in Tel. and N
I think there is a typo here (else the editors would have noticed this
slip) and the correct reading should be :
"Q _Elwe_, always recorded in Tel. and N form _Elwe_"
(or, perhaps "N _Elwe_, always recorded in Tel. and Q form _Elwe_")
[The passage as given in VT45:14 is correct; in this instance there
was no slip by the editors, nor a slip by Tolkien. In the Etymologies,
bases beginning in 3- have Qenya derivates in _h-_: thus 3AN- 'male'
> Q _hanu_ 'a male', 3AR- 'have, hold' > Q _harya-_ 'possess', etc.In Noldorin initial 3- disappeared: thus 3AN- > N _anw_ 'a male',
3AR- > N _ardh_ 'realm'.
The point of the passage cited above s.v. 3EL- is that etymologically
the Qenya form of the name of the lord of the Teleri would have
been *_Helwe_, but the only name used in written records (apparently
even records written in Qenya) was the Telerin/Noldorin form _Elwe_.
The asterisk preceding *_Helwe_ does not indicate that it is a
primitive or early form, but rather that it is a form not recorded
- Patrick Wynne wrote:
> [The passage as given in VT45:14 is correct; in this instanceIt is important to note that at least by the sixties, if we judge by
> there was no slip by the editors, nor a slip by Tolkien. In the
> Etymologies, bases beginning in 3- have Qenya derivates in
> _h-_: thus 3AN- 'male' > Q _hanu_ 'a male', 3AR- 'have,
> hold' > Q _harya-_ 'possess', etc. In Noldorin initial 3-
> disappeared: thus 3AN- > N _anw_ 'a male', 3AR- > N _ardh_
"Notes on _Óre_", for example (VT41), Tolkien had already changed
his mind and reversed this change. Thus from 3OR- we have Q.
and T. _óre, ora, órea_ and other derivatives lacking initial consonant,
yet the cognate of Q. T. _óre_ is S. _gûr_. The change of initial 3-
to G- appears in Doriathrin (or Ilkorin?) of the thirties, according to
the Etymologies. Is this a case of Etym.-Doriathrin incorporated into
S. (as opposed to Noldorin)?
-- Jonathan Avidan
[This is an interesting point. The "Notes on _Óre_" date to c. 1968,
and the typescript text on VT41:11 does indeed state that Common
Eldarin 3OR- yielded Q _or-_, T. _or-_, and S. _gor-_. However, the
forms _óre, ora, órea_ cited by Jonathan are A) from a different
(albeit contemporary) set of notes; B) are in fact said to derive from
Common Eldarin root HOR (VT41:13); and C) are evidently Quenya, not
Quenya and Telerin. It is interesting that Tolkien gives _(h)ore_ as
one of the derivatives of HOR in the latter notes, showing that initlal
H- in the root was sometimes retained (dialectally, perhaps).
But Jonathan's basic point stands, that Tolkien clearly changed his mind
in his later writings about what the developments of original initial 3-
should be in the various descendant languages. He was also uncertain
whether this sound should be 3- or H- in the primitive tongue, as
shown by the variation between 3OR- and HOR- in the notes mentioned
above. In "Quendi and Eldar" (c. 1959-60) we can note the root
*HO 'away, from, from among' (XI:368), which is clearly a later conception
of 3O- 'from, away, from among, out of' in the Etymologies. In "Q&E"
this initial *H- was retained in Quenya and Telerin, but lost in Sindarin;
thus *HEK- > Q. _heka!_, T. _heca!_ 'be gone! stand aside', but S. _ego!_
'be off!' (XI:364-65). -- PHW]