User:Kwékwlos/Sound changes from Indo-European to Japonic

This page will show many intricate details on how Proto-Indo-European evolved into Proto-Japonic. Please note that all of this is very speculative, so be warned.

Proto-Indo-European to Early Proto-Indo-Iranian (3000 BCE to 2000 BCE)

 * 1) Fronting of palatovelar consonants into affricates (*ḱ > *ć, *ǵ > *ȷ́, *ǵʰ > *ȷ́ʰ).
 * > >  >  >  2.4 LH > 2.3 LL
 * > >  >  >  2.I H- (irregular accent)
 * > >  >  >  2.4 LH > 2.3 LL
 * 1) Labiovelar consonants become plain velar consonants (*kʷ > *k, *gʷ > *g, *gʷʰ > *gʰ).
 * > >  >  >  >  2.I H-
 * > >  >  >  3.I H-
 * > >  >  (accent shift) >  (irregular change) >  2.1=1.1 HH=H
 * 1) Change of *l and *l̥ to *r and *r̥.
 * > >  >  >  >  3.1 HHH
 * > >  >  >  2.4 LH > 2.3 LL
 * 1) Palatalization of s before *r/r̥ (also includes those from *l/*l̥), *u/*w, *k, and *i/*y as *š.
 * > >  >  (irregular change) >  2.4 LH > 2.1 HH
 * > >  >  >  2.4 LH > 2.3 LL
 * > >  >  (dual form) >  >  2.3 LL (irregular assimilation)
 * > >  >  >  >  2.5 LF (irregular labial assimilation)
 * 1) Bartholomae's law (voicing assimilation after aspirated consonants)
 * > >  >  >  3.II L-
 * 1) Palatalization of velars before *e, *ē, or *i (*k > *kʲ, *g > *gʲ, *gʷʰ > *gʲʰ)
 * > >  >  >  >  2.I H-
 * > >  >  >  3.I H-
 * > >  >  (remodeling) >  >  2.2 HL > 2.1 HH
 * > >  >  >  2.II L-
 * 1) Brugmann's law (open-syllabic *o > *ō)
 * > >  >  >  1.2 H(L)
 * > >  >  (addition of prefix) >  >,  4.11 LLHL > 2.1 HH
 * 1) *h₂ before a vowel and after a voiceless stop creates a voiceless aspirated consonant (*ph₂ > pʰ, *th₂ > *tʰ, *kh₂ > *kʰ)
 * > >  >  2.4 LH > 2.3 LL
 * 1) Merger of the three laryngeals *h₁, *h₂, *h₃ (after vowel coloring) as *H.
 * > >  >  >  2.I H- (irregular accent)
 * > >  >  2.II L- (accent neutralization due to clause position)
 * > >  >  2.x
 * 1) A laryngeal between two consonants vocalizes as *i, or disappears.
 * > >  (irregular change) >  >  2.x
 * > >  >  >  3.II L-
 * 1) Merger of *e, *o as *a, *ē and *ō as *ā.
 * > >  2.II L- (accent neutralization due to clause position)
 * > >  (accent shift) >  (irregular change) >  2.1=1.1 HH=H
 * > >  (accent shift) >  >  2.4a/2.5a LH/LF
 * > >  (irregular change) >  >  >  1.3 L

Early Proto-Indo-Iranian to Proto-Japonic-Nuristani (2000 BCE to 1500 BCE)

 * 1) Aspirated consonants lose their aspiration (*bʰ > *b, *dʰ > *d, *gʰ > *g, *gʲʰ > *gʲ, *ȷ́ʰ > *ȷ́, *pʰ > *p, *tʰ > *t, *kʰ > *k).
 * > >  >  >  >  1.3 L
 * > >  >  >  >  2.II L-
 * > >  >  (with preposed genitive) >
 * > >  >  >  2.4 LH > 2.3 LL
 * > >  >  >  2.4 LH > 2.3 LL
 * 1) Initial *r or *Hr becomes *ř.
 * > >  (addition of prefix) >  >  (metathesis) >  3.4 LLL > 2.2 HL
 * > >  >  >  3.1 HHH
 * 1) Loss of the laryngeal *H between two vowels, probably conditioned by accent. If a sequence of two distinct vowels results, a glide is inserted, otherwise they merge.
 * > >  >  >  >  1.3 L
 * > >  >  >  2.4 LH > 2.1 HH
 * + >  >  (glide dissimilation) >  >  2.II L-
 * > >  >  >  2.3/2.4 LL/LH
 * 1) Reduction of syllable-initial *an as *n̥ followed by shortening of syllable-initial *ān as *an.
 * > >  (remodeling) >  >  3.7 LHL > 2.1 HH
 * > >  >  >  2.4 LH > 2.3 LL
 * + >  (irregular long vowel) >  >  >  3.5b LLH

Proto-Japonic-Nuristani to Pre-Proto-Japonic (1500 BCE to 1000 BCE)

 * 1) The syllabic consonants *m̥, *n̥, *r̥ become *am, *un, *ur syllable-finally; else they are retained.
 * > >  >  >  3.5b/3.6b LLH/LHH
 * > >  >  >  1.2 H(L)
 * > >  >  2.4/2.5 LH/LF
 * 1) Single *ā becomes *o, except when unaccented, is followed by *u, or in word-final position, where it becomes *a instead.
 * > >  >  >  1.1 H(H)
 * > >  (addition of prefix) >  >  >,  4.11 LLHL > 2.1 HH
 * > >  >  >  2.4 LH > 2.3 LL
 * > >  >  >  1.2 H(L)
 * > >  >  >  1.1 H(H)
 * > >  (accent shift) >  >  2.4a/2.5a LH/LF
 * 1) *wa becomes *wu when not word-initial or word-final (although it may be a later development).
 * > >  >  >  > *sitwá >  2.4 LH > 2.3 LL
 * > >  (accent shift) >  >  2.4a/2.5a LH/LF
 * 1) Secondary palatovelars lose their palatalization (*kʲ > *k, *gʲ > *g).
 * > >  >  >  3.I H-
 * > >  >  >  3.1 HHH
 * > >  >  (accent shift) >  (irregular change) >  2.1=1.1 HH=H
 * 1) Simplification of *tˢt and *dᶻd as *tt, *dd
 * > >  >  >  2.4b LH
 * > >  >  >  >  3.II L-
 * > >  >  >  >  2.II L-
 * 1) Loss of *s before a voiceless stop (*sp > *p, *st > *t, *sk > *k)
 * > >  >  >  >  2.II L-
 * > >  >  >  2.5a LF
 * > >  >  >  2.4 LH > 2.3 LL
 * > >  (remodeling) >  >  >  2.II L- (irregular accent)
 * 1) Voiced consonants are devoiced (*b > *p, *d > *t, *g > *k, *ȷ́ > *ć).
 * > >  >  2.4b LH
 * > >  (remodeling) >  >  >  2.2 HL > 2.1 HH
 * > >  >  >  (with preposed genitive) >
 * > >  >  >  2.4 LH > 2.3 LL
 * 1) Loss of *ř (could be a later development).
 * > >  (addition of prefix) >  (metathesis) >  >  3.4 LLL > 2.2 HL
 * > >  >  >  3.1 HHH
 * 1) *kš becomes *ć word-initially or *ćć word-medially before high vowels.
 * > >  (remodeling) >  >  3.7 LHL > 2.1 HH
 * 1) Phonemic split of *a into *a and *ë under complex, unclear conditions.
 * > >  >  >  2.4 LH > 2.3 LL
 * > >  (remodeling) >  >  >  3.7 LHL > 2.1 HH
 * > >  >  >  2.4 LH > 2.3 LL
 * > >  >  >  2.4 LH > 2.3 LL

Pre-Proto-Japonic to Early Proto-Japonic (1000 BCE to 700 BCE)

 * 1) Loss of final consonants.
 * > >  >  >  2.4 LH > 2.3 LL
 * > >  >  >  1.2 H(L)
 * > >  >  >  2.4/2.5 LH/LF
 * 1) Loss of laryngeals, lengthening the preceding vowel and causing glottalization of following *w as *wH from metathesis.
 * > >  (remodeling) >  >  >  2.2 HL > 2.1 HH
 * > >  >  >  2.4a LH
 * > >  (addition of prefix) >  >  4.11 LLHL > 2.1 HH
 * 1) Simplification of consonant clusters and gemination of intervocalic *-y-. Generally, the first consonant is deleted, except for the ones listed below. This is believed to have led to the split correspondence in Ryukyuan between B and C of tone class 2.4/2.5 words, as words containing geminates or consonant clusters belong to class B (2.4b/2.5b).
 * 2) *ćr > *ć, but it rhotacized a following nasal.
 * > >  >  >  3.7 LHL
 * > >  >  2.4b LH
 * 1) *ćš > *ćć > *ć (also *ćš > *ć word-initially).
 * > >  (remodeling) >  >  >  2.4 LH > 2.3 LL
 * > >  >  >  2.2 HL > 2.1 HH
 * 1) *ćt > *ćć > *ć.
 * > >  >  >  2.4/2.5 LH/LF (fossilized locative)
 * 1) *kr > *k, but *k...kr > *k...r.
 * > >  >  2.I H-
 * > >  >  2.2 HL > 2.1 HH
 * > >  (accent shift) >  (irregular change) >  >  2.1=1.1 HH=H
 * 1) *kš > *kk > *k, except word-initially.
 * > >  (remodeling) >  >  >  2.4 LH > 2.3 LL
 * 1) *mr > *m or *nr > *r, probably a dialectal feature, but the regular reflex seems to be *nr > *r causing a pitch rise on the following vowel and creating various accent classes.
 * > >  >  >  2.2 HL > 2.1 HH
 * > >  (remodeling) >  (irregular change) >  >,  2.2 HL > 2.1 HH
 * 1) *ntr > *nr > *n or *r.
 * > >  >  >  2.x
 * > >  >  2.2 HL > 2.1 HH
 * > >  (remodeling) >  >  2.2 HL > 2.1 HH
 * 1) *pr > *p.
 * > >  >  1.3 L
 * 1) *sk, *šk > *ss, *šš > *s, *š.
 * > >  >  2.II L-
 * 1) *tr > *r in all positions except word-initially, where it becomes *t. Note that dialectal reflexes also give *ć > *s or in one known case, *y.
 * > >  >  2.4 LH > 2.3 LL
 * > >  >  2.4 LH > 2.3 LL
 * > >  >  >  >  3.7 LHL > 2.3 LL
 * + >  >  >  3.II L-
 * 1) *ts > *t. This occurred before *ć > *c (pronounced *ts).
 * > >  >  2.4 LH
 * > >  (remodeling) >  >  >  1.3 L
 * 1) *ć becomes *c.
 * > >  >  2.4 LH > 2.3 LL
 * 1) Rhotacization of *s and *š as *r, except if there is a sequence of two such consonants.
 * > >  (irregular change) >  >  >  2.2 HL
 * > >  >  2.4 LH > 2.1 HH (irregular accent?)
 * > >  (accent shift) >  >  2.4a LH
 * 1) Rhotacization of unaccented *t as *r, except word-initially or after *k, *t, or *n̥.
 * > >  (addition of prefix) >  3.5 LLH > 3.4 LLL
 * > >  >  >  2.4 LH > 2.3 LL
 * > >  >  1.1 H(H)
 * 1) *kš > *kus word-initially.
 * > >  (remodeling) >  >  3.II L-

Early Proto-Japonic to Middle Proto-Japonic (700 BCE to 300 BCE)

 * 1) *ay > *i when word-initial or following *sw (also when following *t if unaccented), otherwise *a. *aw > *a, which causes class 2.4 and 2.5 nouns to have class B.
 * > >  >  >  1.3 L
 * > >  2.4 LH > 2.3 LL
 * > >  2.4b LH
 * > >  (addition of prefix) >  >  2.5b LF
 * 1) Arisaka's law (spreading of remaining *ë to adjacent vowels except *i). *n̥ and *r̥ = *ën and *ër spread *ë to *u only.
 * > >  >  >  >  2.4 LH > 2.3 LL (irregular accent?)
 * > >  >  2.4 LH > 2.3 LL
 * > >  (remodeling) >  2.4a LH
 * 1) Syllabic consonants vocalize or receive an epenthetic high vowel (which feeds into Arisaka's law if *u) except for *n̥ word-initially. *r̥t > *r only after *p. The sequence *rn (likely a retroflex nasal) that results from it is secondary; old *rn having become *n during the simplification.
 * > >  2.4 LH > 2.3 LL
 * > >   >  2.4a LH
 * > >  >  2.4 LH > 2.3 LL
 * 1) *wH (from *Hw and *wH) > *mp.
 * > >  (remodeling) >  >  >  2.2 HL > 2.1 HH
 * 1) Long vowels shorten, except for *ū, which becomes a short close central vowel (written as *ū) that becomes *i or *u, probably the earliest difference between Japanese and Ryukyuan.
 * > >  >,  2.4 LH > 2.3 LL
 * 1) *ë is raised to u, near *m, *n, and sometimes *r.
 * > >  >  >  2.II L-
 * > >  >  >  2.I H- (irregular accent)
 * > >  >  2.1=1.1 HH=H
 * > >  >  2.II L-
 * 1) Remaining *ë becomes *ə.
 * 2) *c becomes *s.
 * > >  >  2.4 LH > 2.3 LL
 * > >  2.4 LH > 2.3 LL
 * 1) *o is raised to *u.
 * > >  >  >  1.1 H(H)
 * > >  >  >  1.1 H(H)
 * 1) *u lowers to *o (different from the old *o) near *m and *n, except when *y precedes it.
 * > >  >  2.4 LH > 2.3 LL
 * > >  >  (remodeling) >  3.3 HLL
 * > >   >  2.4a LH
 * > >  (remodeling) >  >  2.4 LH > 2.3 LL
 * 1) *w is lost after consonants. *iwa > *iya and following that, word-final *iya > *i.
 * > >  >  >  1.3 L
 * > >  >  2.4 LH > 2.3 LL
 * > >  >  >  >  2.4 LH > 2.1 HH
 * > >  >  >  2.3/2.4a LL/LH
 * 1) *i preceding *n dialectally lowers into *e, in part due to the Japanese reflex retaining the vowel.
 * > >  (dual form) >  >  2.3 LL (irregular assimilation)
 * > >  (addition of prefix) >  >  3.4 LLL > 2.2 HL
 * > >  >  2.5 LF (irregular accent)
 * 1) *rn becomes *r
 * > >  >  2.4 LH > 2.3 LL
 * > >   >  2.4a LH
 * 1) *aya contracts to *a and shifts the accent to the right, unless it is syllable-final and is accented on the suffix, where it is retained.
 * > >  >  >  3.II L-
 * > >  2.4 LH > 2.3 LL
 * > >  >  2.3/2.4a LL/LH
 * 1) *awa > *uwa.
 * > >  >  >  1.3 L
 * 1) Fortition of *w as *p after *a or *i with some exceptions involving *i, which lowers to *e.
 * > >  >  >  2.4 LH > 2.1 HH
 * 1) The sequence *i...ə́ becomes *é...e with a leftward shift in accent.
 * > >  >  >  1.1 H(H)
 * > >  >  >  1.x
 * 1) Syllable-final *ya after consonants becomes *e (> Old Japanese e2).
 * > >  >  >  2.4b LH
 * 1) Accented word-final *nt becomes *t in some cases, probably dialectal.
 * > >  >  2.4 LH > 2.3 LL
 * > >  >  2.4b LH
 * 1) In numerous cases, a word-final high pitch following a nasal consonant becomes low, causing the creation of accent classes 2.3 and 3.4.
 * 2) Adjectives with word-final high pitch also undergo the same process, regardless of the environment, also affecting nouns derived from it.

Middle Proto-Japonic to Late Proto-Japonic (300 BCE to 300 CE)

 * 1) *n̥m > *n̥w, in some cases changing to *y conditioned by accent.
 * > >  >  1.3 L
 * > >  >  1.1/1.2 H(H)/H(L)
 * 1) *n̥ vocalizes into a high vowel.
 * > >  >  >  1.3 L
 * > >  >  >  >  3.7 LHL > 2.1 HH
 * 1) Lenition of unaccented *p as *w or *w̃.
 * > >  >  >  3.4/3.5 LLL/LLH
 * 1) *p is lost (probably via lenition first) if the following vowel is unaccented and is followed by a syllable containing *p or *t.
 * > >  2.4a LH
 * > >  2.4 LH > 2.3 LL
 * > >  >  >  2.3/2.4 LL/LH
 * + >  >  >  3.I H-
 * 1) Assimilation of non-initial *n...r as *r...r.
 * > >  >  2.I H-
 * > >  >  2.II L-
 * 1) Loss of unaccented *r in labial environments, otherwise an epenthetic *p is inserted. This creates a new set of diphthongs.
 * + >  >  >  3.I H-
 * > >  >  2.1/2.2 HH/HL
 * 1) Loss of *a before a sequence *wi, with a later addition of epenthetic *i alongside a high pitch.
 * > >  >  >,  2.4 LH > 2.1 HH
 * 1) Loss of unaccented initial *i or *u, in some cases they are retained with a high pitch.
 * > >  >  2.2 HL
 * > >  >  >  2.4 LH > 2.3 LL
 * 1) *n > *m near single *o.
 * > >  >  >  2.2 HL
 * > >  >  >  2.4 LH > 2.3 LL
 * 1) *wi > dialectal *i (probably via earlier *yi) or *yu.
 * + >  >  >,  2.I H-
 * > >  >  2.3/2.4 LL/LH
 * 1) Syncope of unaccented syllables containing *m, *n, *r, and *y, which may cause a 1.2 or 2.5 accent pattern via nasalization.
 * > >  >  1.1 H(H)
 * > >  >  >  1.2 H(L)
 * > >  >  >  3.4/3.5 LLL/LLH
 * > >  >  >  2.5a LF
 * > >  >  >  >  2.5b LF
 * 1) *w̃ > *m and denasalization of vowels (but may cause nasal spreading).
 * > >  >  3.4/3.5 LLL/LLH
 * > >  >  >  1.2 H(L)
 * > >  >  >  2.5b LF