In (Ancient?) Quenya it seems that an initial ñ [ŋ] may have vanished before a vowel or w, but survived in the combination ñg while the combination ñy had already dentalized to ny. Tolkien’s conception of this sound change was very unstable, however. Tolkien vacillated on the Quenya phonetic development of initial ñ [ŋ] a great deal throughout his life, sometimes thinking it changed and sometimes thinking it remained the same.

In the Early Qenya of the 1910s, roots explicitly beginning with Ŋ- developed into an initial n-, and ŊW- became m-. A representative sample includes:

• ᴱ√ŊAHYA > ᴱQ. naya- “it grieves” (QL/65).
• ᴱ√ŊARA > ᴱQ. nara “strife” (QL/64).
• ᴱ✶ŋolđō > ᴱQ. Noldo “Gnome” (GL/41).
• ᴱ√ŊWAÐA > ᴱQ. masta- “tarry, linger” (QL/63).
• ᴱ√ŊWEĐE > ᴱQ. meril “flowers” (GL/46).

Examples in The Etymologies from the 1930s seem to show ñ- > h-, probably first becoming ʒ- [ɣ] in Primitive Elvish as it did medially, so that ñ- > ʒ- > h-. There are, unfortunately, no examples of initial ñw- in The Etymologies, and many of the examples showing ñ- are from deleted roots:

• [deleted root] ᴹ√ÑŌ²/ÑONO > ᴹQ. hu “howl” beside rejected ñōn “growl” (EtyAC/ÑŌ²).
• [deleted root] ᴹ√ÑEL > ᴹQ. helme “friendship” (EtyAC/ÑEl).
• ᴹ√ÑIW > ᴹQ. hiuta- “hint” (EtyAC/ÑIW).
• ᴹ√ÑOL > ᴹQ. holme “odour” (Ety/ÑOL).

The variant ñon in the first example might be a remnant of the Early Qenya development where ñ- was preserved. In the last example, the h- was struck through leaving olme (EtyAC/ÑOL); this might be an early manifestation of the phonetic rules described in Tolkien’s phonetic notes from this conceptual period (see next). In the Comparative Tables of phonetic developments from the 1930s, Tolkien wrote:

ŋ > Q. (ʒ) -; ŋw > ʒw > w; Note Q ŋ > ʒ at same time as g > ʒ. Old ʒ > h (PE19/19).

Tolkien first wrote ŋw > *m in these tables, then ŋw > nw before settling on ŋw > ʒw > w (PE19/19 note #18); these are likely the last remnants of ᴱQ. ŋw > m (see above). This paradigm of ñ- vanishing is also described in the Outline of Phonetic Development (OP1) from the 1930s:

ñ. Initially ñ vanished without trace (as in Kal[aquendian] generally) except that ñy > ny at time that ky > ty and so remained in Q. ny (OP1: PE19/32).

ñw. ñw vanished and w developed as old w (OP1: PE19/35).

Note that nasal ñ when derived from ñg endured in classical PQ, and had a distinct alphabetic sign. But ñ [ŋ] in isolation was not a sound found later in Lindarin or Telerin, and the initial ñ of PQ was usually replaced by n in TQ pronunciation (and hence in the transcription of nouns such as Noldor in the histories). The Noldor, however, who possessed the sound [ŋ], and in certain circumstances used it initially, in their own language, often retained ñ. (OP1: PE19/36).

For the first OP1 quote, Tolkien initially wrote “ñ. This had become denasalized in Eldarin generally at a prehistoric period and fell in with original [ʒ] (OP1: PE19/33 note #18)”. This seems to say that both initial and medial ñ had the same development in CE to ʒ. This is consistent with the sound changes seen in The Etymologies that seems to show ñ- > ʒ- > h- (see above). However, Tolkien updated his ideas so that ñ- vanished instead, surviving only from the combination ñg- and even then eventually becoming n- except in the use of Quenya among the Noldor (in the 1930s, Quenya was not their native language).

The rule whereby initial ñ vanished in Quenya also appears in the second version of the Tengwesta Qenderinwa (TQ2) written around 1950:

In Quenya ʒ vanishes initially without traces. ñ remained somewhat longer and made (with ky > ty) the change to ny. It then vanished leaving initial ñw as w. (TQ2: PE18/104).

In a deleted note at the end of this document, Tolkien seems to be considering a new paradigm in which initial ñ- survives:

Let Q. have free [ŋ]. Let Q. keep ŋ init[ially] and med[ially]. Transcribed ñ but initially and before w, g will do since Q ŋg initially > ŋ (as nd > n) and -ŋgw- > ŋw (as ndy > ny). In spoken Q. of later times except in Valinor (Lindar), initial ŋ > n as noldor, nware, medially ñ > ng. Medially g is often used. Or could g be used = [??], ng = ŋŋ, ŋg as Sigimóre, Siñimóre, Singimóre (TQ2: PE18/107 note #163).

Though rejected in TQ2, this idea of ñ-survival reappears in the Outline of Phonology (OP2) in the 1950s:

ñ initially before vowels remained at first unchanged; medially between vowels it vanished in all the Valinorean or Kalaquenderin dialects ... Initial ñ was preserved: see further under ñg- (OP2: PE19/71-72).

This last sentence was a later addition in ink (PE19/71 note #20). It refers to the following note later in the same document:

The later history of initial ñ whether from old simple ñ or from ñg requires to be noted. It remained in Noldorin pronunciation, but became n- (by sound substitution) in the colloquial TQ of the Vanyar, though the learned among them continued to employ ñ when reading or reciting PQ (OP2: PE19/76).

The paragraph containing this quote itself underwent extensive revision (PE19/77 note #44), to the point where it is difficult to tell whether the survival of simple initial ñ was part of the original composition of OP2 or was a later addition. Tolkien also said initial ñw survived unchanged (PE19/75) and this rule seems to be part of the first draft of OP2. Thus, dating this conceptual shift is difficult. However, there is a note from the mid-1960s that seems to indicate Tolkien reversed himself, deciding again that initial ñ vanished.

loss of g and of ñ (except in initial ñg-) ... (PE17/92).

There are not many examples of this sound change in the 1950s and 60s, but those there are seem to show initial ñ vanishing and ñ > ñw:

• √ÑAR > Q. ar- “by” (PE17/169).
• ✶ñwa-lōth > Q. olos “inflorescence” (PE17/160).

The second example is a bit obscure, but likely first [ŋw-] > [w-] and then the resulting [wo] became [o], though why the a became o isn’t clear. Perhaps it is a reversal of the rule whereby stressed [wŏ] became [wa].

To complicate matters, Tolkien seems to reverse himself again in The Shibboleth of Feanor written in 1968, where he described the origin of Galadriel’s name as follows:

As he gave it in Telerin form it was Alatāriel(lë). The Quenyarized form appears as Altariel, though its true form would have been Ñaltariel. It was euphoniously and correctly rendered in Sindarin Galadriel. The name was derived from the Common Eldarin stem ÑAL “shine by reflection”: *ñalata “radiance, glittering reflection” (from jewels, glass or polished metals, or water) > Quenya ñalta, Telerin alata, Sindarin galad ... (PM/347).

Hat tip to Shihali for pointing this note out to me. This origin of Galadriel’s name by way of Telerin Alatáriel was Tolkien’s preferred derivation of her name in his later writings (PE17/60, UT/266). In theory, Q. Ñaltariel and S. Galadriel could be derived from strengthened *Ñgalatā-rigelle, but this would be unlikely to produce T. Alatáriel. This simplest explanation is that primitive simple ñ- was preserved in Quenya once again, but was lost in Telerin (and possibly ñ- > ñg- > g- in Sindarin, but that is less clear).

To summarize, the conceptual development of these sound changes seems to be:

• 1910s-1920s: ñ- > n- and ñw- > m-.
• early-to-mid 1930s: ñ- > ʒ- > h- (ñw- unclear).
• mid-to-late 1930s: ñ- > (nil) and ñw- > w-.
• 1950s: ñ- and ñw- survived unchanged, becoming n-, nw- in some dialects.

From the 1950s forward, Tolkien’s seems to have vacillated between the last two scenarios, which were probably entangled with his shifting ideas on the development of initial ʒ. One challenge in making this analysis is that primitive strengthened ñg- (which survived and reduced to ñ-) is much more common than primitive initial simple ñ-, so clear examples of its development are hard to find.

Since ñ- from ñg- consistently survives, any ñ-loss must be before initial nasals plus stops reduced to nasals. Since the resulting “w developed as old w (PE19/35)”, ñ-loss must also have been before [w] became [β] initially and between vowels. This means the ñ-loss probably occurred in Ancient Quenya (if it occurred at all).

For the pronunciation of ñ- as n- in various dialects, see the entry on how initial [ŋ] became [n].

Neo-Quenya: For purposes of Neo-Quenya, the words whose etymology depend on initial simple ñ are so rare that the ultimate scenario used only barely matters. My current preference is the scenario of OP2 (initial ñ surviving) because that gives the most probable explanation of Galadriel’s other names: T. Alatáriel and Q. Ñaltariel.

@@@ initial [ŋw] survived according to PE19/38