User:Womtelo/sandbox/Transphonologization

Transphonologization is a type of sound change whereby a phonemic contrast that used to involve a certain feature X evolves in such a way that the contrast is preserved, yet becomes associated with a different feature Y.

For example, a language contrasting two words *sat vs. *san may evolve historically so that final consonants are dropped, yet the modern language preserves the contrast through the nature of the vowel, as in a pair /sa/ vs. /sɑ̃/. Such a situation would be described by saying that a former contrast between oral and nasal consonants has been transphonologized into a contrast between oral vs. nasal vowels.

The term transphonologization was used initially by André-Georges Haudricourt. It has been frequently mentioned in followers of Panchronic phonology, and beyond.

Other tentative terms have been used to describe the same phenomenon, yet with less success  including rephonologization and cheshirization (see below).

Umlaut
A common example of transphonologization is Germanic umlaut. In many Germanic languages around 500–700 AD, a sound change fronted a back vowel when an or  followed in the next syllable. Typically, the or  was then lost, leading to a situation where a trace of the original  or  remains in the fronted quality of the preceding vowel. Alternatively, a distinction formerly expressed through the presence or absence of an or  suffix was then re-expressed as a distinction between a front or back vowel.
 * Germanic

As a specific instance of this, in prehistoric Old English, a certain class of nouns was marked by an suffix in the (nominative) plural, but had no suffix in the (nominative) singular. A word like "mouse", for example, had a plural  "mice". After umlaut, the plural became pronounced, where the long back vowel was fronted, producing a new subphonemic front-rounded vowel , which serves as a secondary indicator of plurality. Subsequent loss of final, however, made a phoneme and the primary indicator of plurality, leading to a distinction between  "mouse" and  "mice". In this case, the lost sound left a trace in the presence of ; or equivalently, the distinction between singular and plural, formerly expressed through a suffix, has been re-expressed using a different feature, namely the front-back distinction of the main vowel. This distinction survives in the modern forms "mouse" and "mice", although the specifics have been modified by the Great Vowel Shift.

Similar phenomena have been described in languages outside Germanic.
 * outside Germanic
 * For example, 17 languages of northern Vanuatu have gone through a process whereby former *CVCV dissyllables lost their final vowel, yet preserved their contrast through the creation of new vowels: e.g. *mati vs. *mata vs. *matu transphonologized to /mɛt/ vs. /mat/ vs. /mɔt/. This resulted in the expansion of vowel inventories in the region, from an original five-vowel system (*i e a o u) to inventories up to 10 or 11 vowels (depending on the language).

Nasalization of vowels

 * In French, a final sound disappeared, but left its trace in the nasalization of the preceding vowel, as in vin blanc, from historical.
 * In many languages (Sino-Tibetan, Austroasiatic, Oceanic, Celtic…), a vowel was nasalized by the nasal consonant preceding it: this "historical transfer of nasality between consonantal onset and vowel" is a case of transphonologization.

Compensatory lengthening

 * In American English, the words rider and writer are pronounced with a instead of  and  as a result of flapping. The distinction between the two words is preserved by (or transferred to) the length of the vowel (or in this case, diphthong), as vowels are pronounced longer before voiced consonants than before voiceless consonants.

Before disappearing, a sound may trigger or prevent some phonetic change in its vicinity that would not otherwise have occurred, and which may remain long afterward. For example:
 * In the English word night,  the sound (spelled gh) disappeared, but before, or perhaps as it did so (see "compensatory lengthening"), it lengthened the vowel $⟨i⟩$, so that the word is pronounced  "nite" rather than the  "nit" that would otherwise be expected for a closed syllable.

Tonogenesis

 * The existence of contrastive tone in modern languages often originates in transphonologization of earlier contrasts between consonants: e.g. a former contrast of consonant voicing (*pa vs. *ba) transphonologizes to a tonal contrast (*paH vs. *paL)
 * the tone split of Chinese, where the voiced consonants present in Middle Chinese lowered the tone of a syllable and subsequently lost their voicing in many varieties.
 * Floating tones are generally the remains of entire disappeared syllables.

Transphonologization resulting in a new contrast on consonants

 * consonant mutation in Celtic languages (a lost vowel triggered initial consonant lenition, and a lost nasal triggered nasalization).

Other examples

 * In Sanskrit, voiced sibilants *z and *ž of the Proto-Indo-Iranian language were deleted, which lengthened the preceding vowel. Additionally, the elision of the sound *ž resulted in the following voiced dental consonant /d/ to become retroflexed, as seen in the development of Proto-Indo-European *nizdós ("nest") --> Proto-Indo-Iranian *niždás --> Sanskrit nīḍáḥ. In case of a plain *z, the preceding vowel was lengthened without causing the retroflexion of the following consonant as seen in Proto-Indo-European *sízdeti ("sits down") --> Proto-Indo-Iranian *sízdati --> Sanskrit sī́dati.
 * the prevention of sound change by a lost consonant in Lahu;
 * In Estonian and some other Uralic languages, when case endings are elided, the changed root indicates the presence of the case, see consonant gradation.
 * In Turkish, ğ lengthens the preceding vowel, unless when followed by another vowel (which it separates into another syllable) or consonant after or  (which becomes pronounced as  in such a case. In the Eastern and lower Ankara dialects, it is pronounced as.

Other names
Rephonologization has been used by Roman Jakobson (1931 [1972]) to refer to essentially the same process, yet the term didn't catch on, due to its ambiguity. In a 1994 paper, Norman (1994) used it again in the context of a proposed Old Chinese sound change that transferred a distinction formerly expressed through putative pharyngealization of the initial consonant of a syllable to one expressed through presence or absence of a palatal glide before the main vowel of the syllable. However, rephonologization is occasionally used with another meaning, referring to changes such as the Germanic sound shift or the Slavic change from to, where the phonological relationships among sounds change but the number of phonemes stays the same. This can be viewed as a special case of the broader process being described here.

James Matisoff (1991:443) attempted the coinage cheshirization, as a synonym for transphonologization. This term jokingly refers to the Cheshire Cat, a character in the book Alice in Wonderland, who "vanished quite slowly, beginning with the end of the tail, and ending with the grin, which remained some time after the rest of it had gone". The term “cheshirization” has been used by some authors (e.g. McWhorter 2005; Chappell 2006) but never really caught on, due to its opacity.