In both (Old) Sindarin and (Old) Noldorin, aspirates [pʰ, tʰ, kʰ] became spirants, probably [ɸ, θ, x] as suggested by David Salo (GS/§4.80). Later [ɸ] became [f] and initial [x-] became [h-], partially obscuring the earlier sound changes. This phonetic development applied both to original primitive aspirates and to aspirates that arose from other sources such as voiceless stops that aspirated after consonants. Tolkien explicitly listed these phonetic developments in the Comparative Table of phonetic changes from the 1930s both initially and medially (PE19/18, 23).

For combinations of aspirated voiceless stops, such as [ttʰ] or [ktʰ], both consonants became spirants: [θθ] or [xθ]. This was explicitly described by Tolkien in several places for “dynamic long” pairs of aspirates:

The geminate or dynamic long pph, tth, kkh being weakened in aspiration became identical with pp, tt, kk. (The latter were, probably, in all Eldarin weakly aspirated; but in Noldorin the aspiration became strong so that pp, pph etc. coalesced in pph etc. and eventually (in Exilic) produced spirantal combinations) (Outline of Phonetic Development [OP1], 1940s, PE19/43).

Lengthening produced the following modifications: (i) p, t, k > pp, tt, kk. In the case of (rare) ph and of th, kh the original lengthening may have been pph, tth, kkh, but in Eldarin the aspiration was lost - or rather pp, pph became identical, either as pp in Quenya or as pph as in Old Noldorin (second version of Tengwesta Qenderinwa [TQ2], circa 1950, PE18/91).

Examples of the spirantalization of aspirates are extremely numerous in both Sindarin and Noldorin:

[pʰ] (> [ɸ]) > [f]:	✶phelgā > S. and N. fela “mine, cave” (PE17/118, Ety/PHÉLEG).
[ppʰ] (> [ɸɸ]) > [ff]:	✶et-pel > eppel > S. and N. ephel “outer fence” (PE17/65, WR/137).
[tʰ] > [θ]:	√THIL > S. and N. Ithil “moon” (SA/sil, Ety/THIL).
[ttʰ] > [θθ]:	ᴹ✶kwetta > S. and N. peth “word” (PE17/46, Ety/KWET).
[kʰ] > [x]:	√KARAK [> *karka > *karkʰa] > S. and N. carch “fang, tooth” (SA/carak, Ety/KARAK).
[kkʰ] > [xx]:	✶rokkō > S. and N. roch “horse” (Let/282.1706, Ety/ROK).

The combinations [ptʰ] and [ktʰ] become spirantal as well ([ɸθ], [xθ]), but the first spirant in each pair [x], [ɸ] became a vowel before [θ] with various later effects. See the entry on the vocalization of such spirants for examples and more details.

Note that there are an analogous set of sound changes in Quenya and Telerin, but these were a parallel set of changes rather than ones that occurred at some common stage, since the details differed in each child language. For example, Quenya and Telerin preserved primitive long voiceless stops such as tt: Q. atto, T. atta “daddy” vs. S. atheg (VT48/6).

Conceptual Development: These changes are not relevant to Primitive Elvish as Tolkien conceived of it in the 1910s and 1920s, because Tolkien did not introduce aspirates into Eldarin phonetic developments until the 1930s. Prior to that point, the analogous set of phonemes in Primitive Elvish were themselves already voiceless spirants (PE12/15).

GS/§4.80 GS/§4.81 @@@