One interesting feature of Sindarin is that its nasal mutations are distinct from those of Welsh. In particular, the nasal mutations of p, t, c are ph, th, ch rather than voiceless nasals “mh, nh, ngh” as in Welsh. This is especially peculiar given that Sindarin mostly underwent the same medial phonetic developments as Welsh, whereby (for example) nt > nth > nnh (long voiceless nasal) > nn.

To explain this, David Salo proposed a special Sindarin development occurring only at morpheme boundaries, where nasals were lost before other (single) consonants (GS/§4.187). This would mean that the historical development of the definite plural of têw “letter” would be in·tîw > in·thîw > i·thîw, with the nasal loss explaining why the th- remained rather than further developing into a voiceless nasal. This nasal loss occurs for the plural definite article in with most plural forms, whether or not they undergo nasal mutation:

• i chîn “the children” (hên-plural)” (WJ/403).
• i mbair “the houses (mâr-plural [mb-])” (SD/128).
• i melegyrn “the great trees (belegorn-plural)” (WJ/185).
• i mírdain “the jewelsmiths (mírdan-plural)” (S/286).
• i Negyth “the Dwarves (Nogoth-plural)” (WJ/338).
• i ngaurhoth “the wolf-horde (gaur-class-plural [ng-])” (WJ/160).
• i ngelaidh “the trees (galadh-plural)” (PE17/60) [actually written (i)Ngelelaið, probably a mistake].
• i Pheriain “the Halflings (Perian-plural)” (LotR/859).
• i sedryn “the faithful (sadron-plural)” (UT/153).
• i thiw “the signs (têw-plural)” (LotR/305).
• i thuin “the pines (thôn-plural)” (PE17/81).

The full form generally appears before vowels and consonant clusters that resist mutation:

• in Edenedair “the Fathers of Men (Adanadar-plural)” (MR/373).
• in Edhil “the Eldar (Edhel-plural)” (S/286).
• in Drúedain “the Woses (Drúadan-plural)” (UT/319).
• in Gwanûr “the twins (gwanunig-dual)” (LotR/1054).

The lack of nasal loss before clusters can be explained by the fact that the nasal preceded two consonants rather than one; such consonant clusters resist nasal mutation for similar reasons.

The mechanics of the nasal loss is mostly straightforward. Before other nasals, the n simply assimilated to the following nasal and shortened. Such shortening occurred at the end of words, so it makes sense it would occur at the beginning of word as well, for example: in·m > im·m > i·m. Before s (and probably also l), the nasal assimilated to following sound and again shortened as was the cases with ss (and ll) at the end of words, for example: in·s > is·s > i·s.

Voiceless stops and spirants require a more specialized explanation. David Salo proposed that the result was [in·t] > [in·θ] > [iθ·θ] > [i·θ] (GS/§4.112, §4.184), with a backwards assimilation of the nasal to a voiceless spirant rather than the forward assimilation of the spirant to a (voiceless) nasal as occurred medially: [-nt-] > [-nθ-] > [-n̥n̥-] > [-nn-]. In 1969 notes, Tolkien himself described a mechanism where the nasal was lost via nasalization of the preceding vowel, which then later lost its nasality: [in·t] > [in·θ] > [ĩ·θ] > [ī·θ] > [i·θ] (PE23/138 note #12); see below for discussion.

There is evidence of similar sound changes in compounds. Some of the best examples are found among the kings of the realms of Arnor and later Dúnedain chieftains, whose names mostly seem to be compounds with an initial elements of ara(n) “king”. For example the 1st chieftain of the Dúnedain was named Aranarth, probably aran + arth “realm” also seen in Arthedain “*Realm of the Edain”. Some of these names seem to involve nasal losses before voiceless spirants:

• Araphor = (perhaps) “*King Fist” = aran + paur.
• Araphant = (perhaps) “*Full King” = aran + pant.
• Arathorn = “Steadfast King” aran + thorn (PE17/113).

Only the last etymology is confirmed, but all three seem to support nasal loss before voiceless spirants at morpheme boundaries rather than (for example): aran+pant > aramphant > **arammant.

An even clearer example appeared in The Etymologies from the 1930s: ON. inpanta > N. in-fant > ifant “aged” = în “year” + pant “full” (Ety/GENG-WĀ, YEN). The word ifant reappears untranslated in Sindarin texts from the 1950s as well (WJ/192). Another possible example is (hypothetical) S. ephed as a derivative of en-pet “*say again” in notes from around 1959 (PE17/167). The example ephed is less clear since ph often represents long [ff] in Sindarin as opposed to simple f, but ph can also represent [f] where it is the result of nasal mutation.

Nasal Losses in Common Eldarin Article (CEA): There is a somewhat detailed discussion of the development of nasals before voiceless spirants in notes on the Common Eldarin Article (CEA) from 1969, where Tolkien said:

The first change was one common to T. and S. and so very ancient. After the loosening of the aspirates to spirants a preceding nasal was lost, and the vowel before the nasal was nasalized and lengthened. The nasality was later lost in both T. and S. (PE23/137).

In T. [and] S. a nasal before f, þ, χ was lost with nasality ... hence T imph, inþ, iñχ > ī̜f, ī̜þ, ī̜χ (in case of article unstressed ī̜ > i) ... Hence [nasal] mutations in S. of ph, th, kh ... > f, þ, χ (PE23/138 note #12).

In this document, nasal loss before voiceless spirants was very early and only applied to voiceless spirants derived from aspirates. The nasal remained before voiceless spirants that were the result of nasal + voiceless stop combinations:

The spirantalizing of stops after a nasal was however very much later: a purely S. change only affecting the medial stops and not those that had become final after the loss of all original CE final vowels (-mp, -nt, -ñk remained); the nasal was therefore not phonetically lost before the new spirant (PE23/137).

This is consistent with the nasal mutations seen in CEA: thoron “eagle” → i·theryn and falas “beach” → i·felais versus tawar “woodland” → in·þewair and parth “field” → in·pherth (PE23/139).

However, this aspect of the CEA system is not compatible with nasal mutations of voiceless stops seen before 1969, including examples from The Lord of the Rings such as têw “letter” → i thiw (LotR/305) and Perian “Halfing” → i Pheriain (LotR/953). Thus it is likely that prior to CEA, the loss of nasals occurred for nasals + voiceless spirant regardless of whether they were the result of ancient nasals + aspirates or ancient nasals + voiceless stops. But this requires that the sound change was conditioned somehow on the environment, such as at morpheme boundaries as suggested by David Salo (see above). Alternately, the nasal loss may have occurred before the main word stress (or a stressed second element of a compound). For example, long voiced nasals nn and ññ shortened before the main stress in the system described in CEA:

This simplification of long (“double”) consonants [including nn, ññ] was not actually confined to final position. In the spoken language they were reduced initially and medially after an unstressed vowel followed by the main word-stress (PE23/140).

Neo-Sindarin: For purposes of Neo-Sindarin, I would ignore the changes to nasal losses before voiceless spirants introduced in CEA and stick to the pre-1969 system which seem to require specialized nasal losses at morpheme boundaries as originally theorized by David Salo.