Sindarin Phonetic Development (Part 59)

Sindarin Phonetic Development (Part 59)


S. short [u] often became [o]; [ŭ|uC{uw}|u{mnŋ}] > [o|uC{uw}|u{mnŋ}]

In Sindarin and Noldorin a short [u] was generally lowered to [o]. This sound change was discussed by Bertrand Bellet in his essay on Vowel Affection in Sindarin and Noldorin (VASN). He proposed that it was a general change inhibited by a few special circumstances (see below). One complication in analyzing this sound change is that often [u] > [o] was the result of a-affection. There are nevertheless many examples of [u] > [o] that cannot be explained this way:

  • nukotto/a > S. Nogoth “Dwarf” (WJ/413).
  • urkō > S. orch “Orc” (WJ/390).
  • ᴹ√NUS > nusina > ON. nuhina > nohen > N. noen “wise, sensible” (EtyAC/NUS).
  • ᴹ√RUS > ON. russe > N. rhoss “polished metal” (Ety/RUS).
  • ᴹ√SUK > N. sogo “to drink”, present tense sôg (Ety/RUS).

This last example is of interest because the (Noldorin) past tense of this verb was sunc, where [u] > [o] does not occur. It is representative of one of the conditions where the sound change [u] > [o] is inhibited. These conditions were described in detail by Bertrand Bellet (VASN), and I agree with most of his conclusions (but see below).

[u] was preserved before nasals: There are numerous examples where [u] was preserved if it appeared before a nasal, either [m], [n] or [ŋ]:

Apparent exceptions to this rule can be explained by a-affection:

[u] was preserved before syllables with [u] or [w]: Where the syllable containing [u] was followed by another syllable containing a [u] or a [w], then the [u] was preserved in both syllables:

  • ngurū > S. guru “death” (PE17/87).
  • uruku > S. urug “bogey” (WJ/390).
  • ᴹ✶tyulussē > N. tulus “poplar-tree” ( Ety/TYUL).
  • ᴹ√KUR [> kurwe] > curw > N. curu “cunning” vs. ᴹQ. kurwe “craft” (Ety/KUR).

Apparent exceptions to this rule might also be explained by a-affection:

  • (U)RUÞ [> urutha > urotha] > S. oroth “rage” vs. Q. ursa (PE17/188).
  • ᴹ√TULUK [> tuluka > tuloka] > N. tolog “stalwart, trusty” vs. ᴹQ. tulka “firm, strong, immovable, steadfast” (Ety/TULUK).

Bertrand Bellet further suggested that the lowering of [u] might be inhibited by a following syllable containing [i], but he provided no examples. He also noted several Noldorin counterexamples involving the infinitive suffix -ie or -i:

  • ᴹ√DUL [> dulie > dolie] > dœlio > N. delio “conceal” (Ety/DUL).
  • ᴹ✶or-tur- [> orthuri- > othori- > othœri-] N. ortheri “conceal” (Ety/TUR).

Although obscured by the later sound changes of [o] > [œ] > [e] by way of i-affection, these forms can only be explained by a lowering of [u] to [o]. Until a clear example emerges that supports an alternate scenario, I’m going to assume that the presence of a following syllable containing [i] did not block the lowering of [u] to [o].

Long [ū] was preserved: Where the [ū] was long it did not become [o]. This is true of original primitive [ū], and also of cases where primitive [ō] became [ū]:

Other aberrations: There are a few apparent exceptions to the rules above with a final [u] that can be explained by the presence of a primitive nasal [m] that later spirantalized and then vanished after [u]. In these examples, the [m] would have survived long enough to prevent [u] > [o]:

  • ᴹ√ERE > N. eru “desert” vs. ᴹQ. erume (Ety/ERE), likely from primitive *✶erumē.
  • ᴹ√TELU > N. telu “dome” vs. ᴹQ. telume (Ety/ERE), like from primitive *✶telumē.

There are a few more aberrant forms that are harder to explain:

In the first example, perhaps the u was originally long, or it was reformed by analogy with S. rûth “rage”. In the second example, Bertrand Bellet suggested that ✶guruk might actually be the stem form and the original primitive form was *✶gurukā, so that the [u] > [o] via a-affection. The last example from the Etymologies defies any easy explanation, especially since Tolkien gave this verb as nod- elsewhere (Ety/WŌ); perhaps it is simply a mistake, as suggested by Bertrand Bellet.

Conceptual Development: Isolated short u was quite common in the Gnomish of the 1910s, so it seems unlikely this general sound change was an aspect of Tolkien’s earliest conceptions of the languages. Roman Rausch suggested that [u] > [o] in final closed symbols in Gnomish in his Historical Phonology of Goldogrin (HPG/§1.5):

  • G. culu “gold” vs. genitive culon, dative culor (GG/14).

There is a hint of this sound change in the Early Noldorin Grammar from the 1920s:

(2) o (from ŏ or ŭ) mutates to ui, u (unaccented occasionally > wi) (PE13/122).

However, this sound change is not entirely consistent with the attested examples in the early Noldorin Word-lists and Noldorin Dictionary, for example: ᴱN. grug “angry, grieved” (PE13/145) or ᴱN. lhub “fat” (PE13/149). Roman Rausch suggested in his Historical Phonologies of Ilkorin, Telerin and Noldorin around 1923 that in fact sometimes in Early Noldorin the opposite sound change of [o] > [u] occurred before nasals (HPITN/§4.2.2):

  • ᴱN. gunn “dragon” vs. ᴱQ. kondo (PE13/162).

This is also supported by a note in the Early Noldorin Grammar:

(4) u (from ŏ + nasal, ŭ) mutates to ui (PE13/122).