Phonological Gain

[zelos]

I was just thinking on a potensial project (not a conlang but related to it) i might go at and came across a thought.

All languages have a bunch of sounds and they can easily change relative to each other, like latin "inpossibilis" eventually becoming "impossibilis" as m and p are more related.

But my question is more of how a language could potensially gain a new sound that it originally didnt have, how would such a process occure? is there any articles?

[Nortaneous]

sound shifts or borrowing from other languages

see: Grimm's law, English /Z/, Finnish /S/, Great Vowel Shift, OE vowel breaking, i-mutation, u-mutation, a-mutation, etc.

[eodrakken]

Allophony followed by sound changes. For example, in English, /t/ is aspirated in initial position, but not when preceded by another consonant. We don't notice the difference because they are allophones.

[stap] <stop> "halt"
[tʰap] <top> "highest part of an object"

Now say English undergoes a sound change where initial [s] is lost. Then you end up with:

[tap] "halt"
[tʰap] "highest part of an object"

And now the aspirated consonant is phonemic, because it allows you to tell the difference between the two words. This is one way that new phonemes develop.

[Colzie]

Allophony followed by sound changes. For example, in English, /t/ is aspirated in initial position, but not when preceded by another consonant. We don't notice the difference because they are allophones.

[stap] <stop> "halt"
[tʰap] <top> "highest part of an object"

Now say English undergoes a sound change where initial [s] is lost. Then you end up with:

[tap] "halt"
[tʰap] "highest part of an object"

And now the aspirated consonant is phonemic, because it allows you to tell the difference between the two words. This is one way that new phonemes develop.

Jamaican English, for instance, loses initial s before a plosive. I think they aspirate by analogy, but the theory is sound.

I find it difficult to imagine other methods than this are often enacted for languages to acquire new sounds...extensive borrowing, maybe, but that's a stretch unless the entire population of speakers have been conquered by a linguistically separate group. Despite the huge influx of words from English, I've yet to hear a German say /dZ/, for example.

Isn't it popular theory that the Southern Bantu languages acquired clicks from their neighbors? Does anyone know how this happened?

[Cedh]

Despite the huge influx of words from English, I've yet to hear a German say /dZ/, for example.

There are a few established words that use /dZ/, e.g. Dschungel, Loggia. But then, quite a number of Germans don't fully voice this affricate (myself included, at least in word-initial position), so it might come out as [tS] instead. (An initial /tS/, however, would normally be aspirated.)

[Whimemsz]

Isn't it popular theory that the Southern Bantu languages acquired clicks from their neighbors? Does anyone know how this happened?

Aside from just borrowing words containing a given sound, speakers of languages in close contact can also have their language undergo sound changes which introduce a sound found in neighboring languages. I have no idea if this is what happened with Southern Bantu languages and the click languages of southern Africa, but it's happened in other places. That's the reason linguistic areas often contain distinctive sounds shared by many languages of various families (cf. the uvular rhotic in many parts of Europe, including France, Germany, Denmark, Belgium, etc.). Languages can also share or borrow sound changes from one another--the sound change /j/ > 0 after a consonant diffused from Cree into Ojibwe-Potawatomi, for instance, and the loss of nasal consonants diffused among the Wakashan languages Makah and Nitinaht (the latter is now pronounced Ditidaht, but it wasn't at the time of first contact).

extensive borrowing, maybe, but that's a stretch unless the entire population of speakers have been conquered by a linguistically separate group

There's no need for anyone to be conquered, just for a number of speakers to be familiar enough with the other language that the borrowings aren't fully nativized. English gained /Z/ through loans from French (although I guess that's a bad example kind of because French was the language of conquerors for a few centuries--nonetheless /Z/ didn't emerge as a separate phoneme until far later, after English kings and nobility all spoke English).

[MadBrain]

Dschinghis Khan?

nasal vowels: vowel + /n/ turns into a nasal vowel, so that /an/ becomes /a~/ but /ana/ stays /ana/. Then the final /a/ in /ana/ is dropped, which makes a /a~/ vs /an/ contrast. Successive loanwords can then build up /a~/ vs /an/ oppositions in non-final parts of the word. Then what was originally /amb,and,ang/ and is now /a~b,a~d,a~g/ can further nasalize into /a~m,a~n,a~N/. This can build up a 4 way /a,an,a~,a~n/ contrast.

One thing I really wonder is, how does a language build up new stop series? In particular, other than that stop~top example, is there another way to build, say, /t_h,t,d/ contrast from /t,d/? How zulu end up with /ph,p,b,b_</ from an original /p,b/? How did Hindi end up with /b_h/, and Sindhi with /d_h/ and /d_</? It's usually not too hard to figure out a path for other sounds (basically successive palatalization for multiple places of articulation and some fricative, plus lenition and fortition for some other sounds) but forming stop series seems to be definitely harder (especially in all those families that have only CV syllable structures).

Subsidiary question: How are "inner" vowel series (typically /y 2/, /1 @/ or /M 7/) formed?

[Makerowner]

I find it difficult to imagine other methods than this are often enacted for languages to acquire new sounds...extensive borrowing, maybe, but that's a stretch unless the entire population of speakers have been conquered by a linguistically separate group.

There are weirder, messier ways of acquiring new phonemes. For example, the "trap-bath split" in Southern England has its origins in allophonic lengthening before voiceless fricatives, but it became phonemic not through a change in the environment, but through uneven spreading through the lexicon. (More common words are more likely to be affected by it.) So today you have pairs like "pass" /pA:s/ vs. "mass" /mas/, "can't" /kA:nt/ vs. "cant" /kant/.

Isn't it popular theory that the Southern Bantu languages acquired clicks from their neighbors? Does anyone know how this happened?

I remember reading that the clicks were used to disguise taboo words, but I don't know where I read that or how well-established that theory is.

[TaylorS]

In some Spanish dialects an aspirated plosive series is developing from /sp/, /st/ and /sk/

[st] --> [ht] --> [t_h]

[Travis B.]

I find it difficult to imagine other methods than this are often enacted for languages to acquire new sounds...extensive borrowing, maybe, but that's a stretch unless the entire population of speakers have been conquered by a linguistically separate group.

There are weirder, messier ways of acquiring new phonemes. For example, the "trap-bath split" in Southern England has its origins in allophonic lengthening before voiceless fricatives, but it became phonemic not through a change in the environment, but through uneven spreading through the lexicon. (More common words are more likely to be affected by it.) So today you have pairs like "pass" /pA:s/ vs. "mass" /mas/, "can't" /kA:nt/ vs. "cant" /kant/.

I have noticed that kind of thing with a lot of the consonant elision and assimilation processes at work right here, where they will apply on a lexical item-to-lexical item, with certain lexical items both inexplicably resisting them while other ones will undergo them even when they "shouldn't" due to stress and metrical-related conditions under which such normally do not occur. Hence they turn out to be in many ways diachronic rather than synchronic-type processes, in that they do not uniformly apply to the normal phonological "pipeline", to borrow a term from computing, even though they are very much active right at the present rather than having only happened in the past and no longer being active.

[zelos]

the gain it from another language doesnt quite answer it in my eyes as it becomes circular. How did they originally gain the sound? and so on. How would any language alone gain any sounds that it originally didnt have?

[Travis B.]

the gain it from another language doesnt quite answer it in my eyes as it becomes circular. How did they originally gain the sound? and so on. How would any language alone gain any sounds that it originally didnt have?

Through retention of phones used in the language such loans originally came from rather than replacing them with native ones due to sufficient familiarity with the source language thereof, that is.

[jal]

How would any language alone gain any sounds that it originally didnt have?

Through sound change. I think that any specific sound can be shown to be able to derive from at least one other sound. The rest of your question is very close to the "how did people start to speak" question, and will have to go unanswered. Listen to babies and toddlers, is my advice.


JAL

[Niedokonany]

Most Slavic languages that have undergone the sound change ɡ > ɣ > ɦ appear to have reacquired /ɡ/ via loanwords. The extent of this varies, some of them have /g/ e.g. in numerous Latin loans like Upper Sorbian biologija or Czech biologie, in others it's still quite marginal (eastern Ukrainian dialects, supposedly).

Loanwords can also break existing patterns of allophone distribution, as a result making the allophones distinct phonemes. Vide: Polish i/y.

How did Hindi end up with /b_h/, and Sindhi with /d_h/ and /d_</?

Well, I think Hindi breathy voiced stops are considered "original", that is they reconstruct them in those places in PIE, at least in some words. As for Sindhi, apparently some glottalicists think the implosives come from PIE glottalic consonants while other linguists derive them from geminates. I don't know anything about Indo-Aryan diachronics, so no idea who's right.

I've also found this and this (tl;dr).

Couldn't breathy voiced stops arise from velarization of some kind where a velar offglide would be debuccalized to [ɦ] > [ʱ]?

[TzirTzi]

I've also found this and this (tl;dr).

Couldn't breathy voiced stops arise from velarization of some kind where a velar offglide would be debuccalized to [ɦ] > [ʱ]?

Thanks for the link, interesting article

/_g/ > /M\/ > /h\/ > /+h\/ seems perfectly possible. Still, there are more straightforward ways to acquire breathy voiced stops, I would think. E.g.: lenition of aspirated stops in specific positions followed by the loss of those conditioning factors; simplification of C+/h/ or C+/h\/ cluster; V+/h/ or V+/h\/ leading to breathy voicing on vowels which is then transferred to consonants.

[zelos]

quite good answers now =)

anywhere where one can find a list in which order sounds usually tend changing as ive read its not entirely random <.<

[Makerowner]

quite good answers now =)

anywhere where one can find a list in which order sounds usually tend changing as ive read its not entirely random <.<

Of course it's not entirely random. A sound change like y>q_h/_l is impossible. There's no simple list of what sound changes occur, but a very general rule (which of course has exceptions) is that sound changes tend to reduce articulatory effort. When the prefix 'in-' is attached to 'possible', to pronounce an [n], you have to time 1) the lowering of the velum, 2) the contact of the tongue-tip with the alveolar ridge, and 3) the closing of the lips for the following /p/ just right; place assimilation produces 'impossible', where only 1) and 3) need to be coordinated.
Of course this is nowhere near the whole story, but it's a start. You can simulate the process by simply saying the words of your conlang over and over, and noticing how you tend to "mispronounce" them. You might find unstressed vowels tending towards [@], intervocalic consonants becoming voiced or fricated, coda consonants being assimilated, etc., all very common processes

[eodrakken]

There's a pretty good summary of some common types of sound changes here:
http://www.compulink.co.uk/~morven/lang ... anges.html

One thing that does seem to be pretty random is *which* plausible changes occur in which languages. For your conlang, that's up to you.

[TaylorS]

In Eridanian /q/ come from /gl/, /gR/, and /Rg/, and /q_h/ from /kl/, /kR/, and /Rk/.

/K/ come from /fl/, /sl/, and /hl/.

/s\/ and /z\/ come from /st/ and /zd/

/r/ comes from /4@`/, intervocalic /nd/, and intervocalic /z/.

[Whimemsz]

The Correspondence Library has a number of different sound changes from different languages and families

[Etherman]

A lot of sound changes can be explained as "feature spreading". That's when a characteristic of a sound spreads to other sounds. For instance you might have the sequence /aka/. Both vowels are voiced by the consonant isn't. The feature of voicing spreads to the consonant, most likely because it's easier to keep the vocal chords vibrating that to vibrate them, halt the vibration, and then start the vibration back up. This results in /aga/ for this example. On the other hand we might end up with /axa/. In this case the feature of voicing didn't spread, but rather the feature of a constant stream of airflow. So rather than stopping the airflow to make the /k/ sound the air was allowed to keep flowing resulting in /x/.

Sometimes the feature that gets spread is nasality. If it spreads to a vowel you get nasalized vowels (as happened in the Slavic languages). It can spread to consonants which then become nasal consonants.

Pretty much any feature can spread. Vowel roundness can be spread even to noncontiguous sounds as happend to the Germanic languages (excluding East Germanic).

Sometimes the opposite can happen. I guess you'd call it feature retraction. Two sounds can share a feature, but one sound loses that feature. In Sanskrit a sequence of two voiced aspirates became a voiced non-aspirate followed by a voiced aspirate. A similar loss of aspiration happened in Greek, but after the voiced aspirates became voiceless aspirates. In many IE languages the feature of aspiration was simply lost everywhere. IE may not have had /b/ but when /b_h/ > /b/ suddenly this new sound was quite common. A similar dissimilatory process took place in a Semitic language (I can't recall which, but it might have been Ge'ez) except the feature of glottalization was lost.

[jal]

A lot of sound changes can be explained as "feature spreading". (...) Sometimes the opposite can happen. I guess you'd call it feature retraction.

I think the common terms are assimilation and dissimilation, though I'm not entirely sure whether these terms are also applied to features. I've heard 'feature spreading' before, but not 'feature retraction'.

Vowel roundness can be spread even to noncontiguous sounds

That would be vowel harmony, which is relatively common in the world's languages.


JAL

[Io]

Bulgarian initially didn't have /f/, it was acquired through extensive borrowing from our Greek neighbours. (Lots and lots of words were coined after Greek words too.)

[zelos]

all of this is very helpful

and as a last question, does anyone know a program that is capable from a root word generate soundchanges onto each other ultimately ending with several different sounding words for various languages?

[sangi39]

In many IE languages the feature of aspiration was simply lost everywhere. IE may not have had /b/ but when /b_h/ > /b/ suddenly this new sound was quite common.

Assuming the classical P, B, Bh system (where P is a voiceless plosive, B is a voiced plosive and Bh is an aspirated voiced plosive). If we use the P, P', B or P, P', Ph, then the situation of the sound changes is different. Greek Ph would derive from Ph. Personally I think there might have been some mid-point between the three theories. P, P' and Ph were in Ph the predominantly heard feature was the aspiration rather than any voicing feature, meaning Ph would be Ph~Bh with a dialectal difference in which one was actually realised, with Bh dialects switching it to B. This would mean Greek Ph came from a Ph varient of PIE Ph~Bh. The loss of aspiration, then, would be the result of a loss of aspiration after a Ph>Bh shift in certain dialects while aspiration in others could simply be lost straight away, allowing for a merger with P and Ph or a shift in the realisation of P while Ph shifted to P. Either way, I'd put a bet on two PIE dialectal plosive inventories:

P-P'-Ph
P-P'-Bh

...with the original system being P-P'-Ph before a dialectal division.