User:Urszag/Onbin

Onbin (Japanese onbin (音便) "euphony") is a set of sound changes that occurred in Early Middle Japanese from around the end of the eighth century to the beginning of the tenth century. Onbin changes caused certain consonant-vowel sequences to become replaced with single segments, as in okite > oite or shinite > shinde. The outcomes of onbin changes were dependent or bound moraic phonemes that were not pronounced as independent syllables, but rather in the same syllable as the preceding vowel. Thus, onbin played a role in diversifying the syllable structure of native Japanese words by creating 'heavy' syllables that ended in two vowels or in a vowel followed by a consonant.

Historically, onbin changes did not occur systematically, and some sequences could yield multiple outcomes. The non-deterministic nature of historical onbin changes is exemplified by doublets showing different outcomes of the same original form, such as こみち komichi (without onbin) versus こうじ kōchi (from kouchi, with U-onbin), or あきんど akindo (with N-onbin) versus あきゅうど akyūdo (with u-onbin).

However, in standard modern Japanese, some onbin changes have become grammaticalized in the conjugation of Japanese verbs: as a result, certain verbs systematically display an "onbin stem" before certain suffixes. There is variation over time periods and across dialects in how the onbin stems of verbs are formed and when they are used.

Sound changes
Based on Modern Japanese spelling, four distinct outcomes of onbin can be identified: the two high vowels /i u/, and the two moraic consonants /N Q/. These outcomes can be respectively referred to as u-onbin, i-onbin, N-onbin (Japanese hatsuonbin) and Q-onbin (Japanese sukuonbin). Frellesvig divides /i u/ into two categories based on whether they were originally oral or nasal, and so recognizes six possible outcomes of onbin: /N Q i u ĩ ũ/.

The core sequences affected by onbin were non-word-initial /p k b ɡ m n/ followed by /i u/. Some onbin changes could occur at the end of words, such as /ki ku/ > /i u/ (as in /kisaki/ > /kisai/ 'empress' and /kaku/ > /kau/ 'in this way'), and occasionally /nu/ > /N/ and /mi mu/ > /N/ or /ũ/ (as in /asomi/ > /asoN/ 'courtier'); in contrast, onbin changes involving /bi bu ɡi ɡu/ seem to be attested only in the middle of words.

The segment following the vowel was also relevant: onbin changes could occur before /p t k s b d ɡ z m n/, but not before /r w y/.

(Is that accurate? What about 候う/侍う?)

The core onbin changes can be summarized as follows:

When the consonant in the affected sequence was either nasal /m n/ or prenasalized /b g/, onbin produced nasal outputs (ĩ, ũ, N) that caused following /p t k s/ to be replaced with their prenasalized counterparts, /b d g z/. By Late Middle Japanese, the nasalized vowels /ĩ, ũ/ had merged with oral /i, u/.

In Late Middle Japanese, vowel sequences ending in /u/ (including that derived from earlier /ũ/) underwent coalescence, with the following results:

Sequences involving other consonants or vowels may sporadically show similar changes in specific words:

In inflected forms of consonant-stem verbs (modern godan verbs), original /ti ri/ came to develop onbin variants Q, but Frellesvig argues that this type of onbin stem developed because of morphological analogy, rather than as a result of the same phonetic processes as the other onbin changes. In this context, stems ending in /si/ also developed alternative onbin forms with /i/, but this is absent in modern Japanese.

Verb stems
As a result of onbin changes, some Japanese verbs have a variant stem, called the onbin stem, used before particular suffixes. In contemporary Modern Japanese, this phenomenon affects consonant-stem verbs: that is, verbs that can be analyzed as having an underlying stem that ends in a consonant, from which the dictionary form is derived by adding the suffix -u. (In Japanese grammatical terminology, these are called 五段 godan ("five-grade/pentagrade") verbs, because their inflected forms make use of the five kana composed of the stem-final consonant plus each of the five Japanese vowels, /a e i o u/.) These verbs use the onbin stem before the past, gerund, conditional, and representative suffixes. Onbin stems developed from the original infinitive form, which originally ended with the vowel /i/. Aside from the sequences mentioned above, onbin was extended by analogy to affect /ti ri si/ in forms of verbs with stems ending in the consonants /t r s/.

In standard Japanese, typically verbs with underlying stems ending in /t r p/ have an onbin stem in /Q/, those with underlying stems ending in /b m n/ have an onbin stem in /N/ (with voicing of the suffix), those with underlying stems in /k/ have an onbin stem in /i/, and those with underlying stems in /ɡ/ have an onbin stem in /i/ with voicing of the suffix. (In medial position, Old Japanese /p/ ultimately evolved to /w/ before /a/, and was lost before any other vowel; because of this change, these verbs may analyzed as being /w/-stems instead in modern Japanese.) Verbs with underlying stems ending in /s/ do not show onbin in standard Japanese, but instead use /si/ before the relevant suffixes.

There are a few exceptions. For example:
 * the verb iku 行く shows an onbin stem in /Q/ rather than the expected /i/.
 * the verbs tou 問う 'to ask', kou 請う 'to request' show onbin stems in /u/ rather than the expected /Q/.
 * Also 恋う? たまう?

Some dialects regularly use u-onbin in certain types of verb stems where standard Japanese has Q-onbin or N-onbin.
 * In some dialects, verbs with underlying stems that originally ended in /p/ have u-onbin stems (with monophthongization of /iu eu au ou/ to /juu joo ɔɔ>oo oo/), e.g. Kyoto /koota/ from the verb kaw-. As early as the 17th century, João Rodrigues' Arte da Lingoa de Iapam identified the use of u-onbin in this context as a feature of the Japanese spoken in Kyoto, in contrast to the use of Q in the Kanto dialect.
 * Similarly, in some dialects, verbs with underlying stems ending in the consonants /m b/ show an onbin stem going back to original /ũ/ (e.g. yōde).

The use of onbin stems in inflected verb forms does not appear to have become an established norm in the literary, written Classical Japanese language, although their use in speech may have been more common than in writing.

Adjective endings
Two suffixes used in the inflection of i-adjectives underwent onbin changes:
 * The Old Japanese ending -ki developed by onbin into the Modern Japanese adjectival attributive/conclusive ending -i.
 * taka- 'tall': takai
 * The adjectival infinitive or continuative ending -ku developed an onbin variant -u. However, the form -ku, without onbin, remains the normal form of this ending in Standard Japanese. The use of the u-onbin form (which coalesces with the preceding vowel) is a characteristic of the language spoken in the Kansai (including Kyoto) and Kyushu regions. In Standard Japanese, the onbin variant of this ending is found only in a few isolated words and in the rarely used "super-polite" construction: these forms can be regarded as borrowings from the Kyoto dialect. The use of -u in Kyoto vs. -ku in the Kanto region for the adjectival copula infinitive ending is described in Rodrigues' description of dialect features in Arte.
 * taka-: takaku (Standard Japanese) takoo (Kansai dialect)

Phonetic development
These phonological changes likely had their roots in earlier phonetic variation in the pronunciation of Old Japanese consonants and vowels. Namely, it is thought that the consonants /p k/ could be phonetically realized not only as voiceless stops [p k], but also as voiced stops [b g], voiceless fricatives [ɸ x], or voiced continuants [β ɣ]. The consonants /b g/ were prenasalized and possibly could be either stops [ᵐb ᵑg] or continuants [ᵐβ ᵑɣ]; vowels before /b g m n/ were likely phonetically nasalized. The high vowels /i u/ could possibly be reduced to coarticulations [ʲ ʷ] on the preceding consonant,

There is disagreement about the exact phonetic path by which onbin developed. The development of consonantal onbin (N and Q) can be interpreted as involving vowel deletion. The development of vocalic onbin (I, U, IN UN) can be interpreted as involving vowel deletion, followed by vocalization of the consonants left behind: however, this is debated.

One explanation supposes that the [i] outcome of /gi/ developed via lenition or deletion of intervocalic /g/, parallel to lenition of intervocalic /k/ in /ki/ > [i].

Many accounts postulate that at least some cases of i- and u-onbin developed as a result of consonants being deleted between vowels. If this is the correct explanation for the origin of u-onbin from sequences involving original labial consonants (/p b m/), then the deletion of these consonants must have been preceded by a sound change that turned /pi bi mi/ into /pu bu mu/ in contexts where onbin would occur. Such a change does appear to be attested by spelling variations in texts from the 11th and 12th centuries, although the interpretation of this evidence is debated. The following examples illustrate ways i- and u-onbin might have developed per this approach:
 * /tobite/ [tõᵐbite] > [tõᵐbute] > [tõᵐmute] > [tõute] > [tõũte] > [tõũde]
 * /kakite/ [kakite] > [kagite] > [kaɣite] > [kaite] /kaite/
 * /kagite/ [kãgite] > [kãɣite] > [kãite] > [kãĩde] /kaide/
 * /jomite/ > /jomute/ [jomute] > [jõute] > [jõũde] /joode/

An alternative hypothesis derives the i-onbin outcome of original /gi/ from vocalization of a syllable-final moraic palatal nasal [ɲ], as in [kõᵑgʲite] > [kõɲde] > [kõĩde]. De Chene 1991 argues that the u-onbin outcomes of original /mu mi bu bi gu/ are likewise best explained by an initial development of /mu mi bu bi/ to moraic [m] and of /gu/ to moraic [ŋ], followed by vocalization of moraic [m] and [ŋ] to [u] (in cases where they did not alternatively become /N/). Accordingly, De Chene interprets the use in Heian-era texts of the "mu" kana in the spelling of words like "yomutaru" (from original "yomitaru") as an orthographic device representing moraic /m/. De Chene argues that deriving [u] from vocalization of earlier moraic [m ŋ] helps explain why [u] is found as an outcome of original /ga/ in the word kagahuru > kauburu (ᵑga > ŋ > u) and why word-final /bu bi gu gi/ never show onbin outcomes, but word-final /ki ku/ can.
 * /jomitaru/ > /jomdaru/ > /joudaru/~/joNdaru/
 * /waragutu/ > [waraŋdu] > /waraudu/~/waraNdu/

Instead of vowel or consonant deletion, Frellesvig 2010 prefers to explain onbin as a consequence of phonetic reduction of both the consonant and vowel, resulting in their fusion into a single phonetic segment (e.g. /mi/ > [mʲː], /pi/ > [βʲː]): that could then be phonologically reinterpreted as a single phonemic segment (vocalic or consonantal). The following example illustrates this approach to explaining onbin:
 * /tugite/ [tũᵑɣʲːdʲe] > [tũĩⁿdʲe]

Nasality and nasal spreading
At the end of a morpheme, /Ĩ, Ũ, N/ acted as triggers of nasality spreading, causing following morpheme-initial /p t k s/ to change into prenasalized /b d g z/. Thus, Frellesvig argues that not only /N/ but also the vowels /Ĩ, Ũ/ were distinctively nasalized in morpheme-final position. This type of progressive nasalization across morpheme boundaries ceased to apply as an automatic process in Late Middle Japanese, although it left an effect on the form of inflectional morphemes and lexicalized compound words.

In contrast, Frellesvig analyzes the contrast between /I, U, Q/ and /Ĩ, Ũ, N/ in EMJ as non-distinctive in morpheme-internal position. Only the oral versions occurred before oral consonant phonemes, and only the nasal versions occurred before nasal or prenasalized consonant phonemes. Therefore, they were in complementary distribution, and nasality in this context can be analyzed as a phonological property of the following consonant phoneme that spreads onto the preceding mora: e.g. [pũndʲe] 'brush' and [tũĩ.ⁿdʲe] 'sequence' can be transcribed phonemically as /puCde/ and /tuIde/, respectively (where /C/ represents a moraic consonant unspecified for nasality, and /I/ represents a 'bound' moraic high front vowel unspecified for nasality).

U-onbin
Core sources per Frellesvig:

Other sequences:

I-onbin
Other sequences:

N-onbin (hatsuonbin)
Core sources per Frellesvig:

Other sequences:

Q-onbin (sokuonbin)
Core sources per Frellesvig:

Other sequences: