Guide to Sound Changes

A forum for guides, lessons and sharing of useful information.
xijlwya
sinic
sinic
Posts: 263
Joined: 01 Sep 2010 15:55
Contact:

Guide to Sound Changes

Post by xijlwya »

Hi everyone! This is a introduction to sound changes as demanded here by Ossicone, Eldin and others.

Table of Contents
Last edited by xijlwya on 29 Jul 2012 16:02, edited 3 times in total.
xijlwya
sinic
sinic
Posts: 263
Joined: 01 Sep 2010 15:55
Contact:

Before Getting into the Changes

Post by xijlwya »

0. Before getting into the Changes

Be sure to have yourself familiarized with IPA and the basics of phonetics. I recommend reading this. We will need that stuff to understand certain sound changes. I'll try to not overburden this with the masses of strange vocabulary coming from phonetics - but probably that will be impossible in some places.
Note that I largely follow the book Historical Linguistics by Trask, which contains two chapters on sound changes (chapter 3 and 4).

1. Were does Sound Change come from?

If we take the representation of a word in IPA, for example panic (something you don't right now), which is /ˈpæn.ɪk/ we see the sequence of sounds that make up the word. These are represented as being separate. First comes [p] then [æ] then [n] and so on. This is not entirely realistic. It is a model so to speak, which has is flaws when it is compared to reality.
When we speak, the sounds of a word influence each other, because our speech organs are continously changing form one sound to another, rather than "popping" from one into another. For example, when pronouncing the [p], the opening of the mouth after the release of pressure is more open because of the following [æ] than it would if an [ɪ] followed (compare the two vowels in their degree of openness!). So - if I were prefectionist, I could jot down the pronounciation of panic as [pʰ̞æ̃n.ɪk] , because of the sounds influencing each other (don't worry if this is not understandable to you - I just wanted to illustrate that the phonemic representation in lacking some information!).
xijlwya
sinic
sinic
Posts: 263
Joined: 01 Sep 2010 15:55
Contact:

Assimilation / Dissimilation

Post by xijlwya »

2.1 Assimilation / Dissimilation

Assimilation means as much as "to become more alike", whereas dissimilation means "to become less alike". Both describe how sounds in a word influence each other and thus change each other. Especially assimilation is very common in all languages and an important process in sound changes.

2.1.1 Assimiliation

When there are two sounds near (not necessarily neighboring) each other (which is said to be happening from time to time ;)), they represent two different positions of tongue and manner of airflow. By putting for example the places of articulation closer together, pronounciation effort can be reduced. Using less energy for pronouncing sounds is generally an underlying principle of sound change! So, because we are lazy even in our unconcious speech behaviour, we tend to "slur" sounds into each other - which is called assimilation.
We can now distinguish between several different kinds of assimilation.
First of all, there is total assimilation where one sound becomes essentially another. As an example Trask states Latin /nokte/ becoming Italian /notte/ (double t in IPA states gemination, I don't know if that is a general convention) and thereby totally assimilating the [k] into the geminated [t].
Second, as a contrast to this, there is partial assimilation. Actually, most assimilations are partial. As an example you could consider the Japanese word <konbanwa> which is pronounced /kom.baŋ.wa/ - this is of course no diachronic (i.e historical) change! - You can see, that the written <n> turns out to be pronounced as [m]. This is due to the [ b] that follows: Both [m] and [ b] have the same point of articulation - bilabial. The alveolar sound [n], which was to be expected from the spelling is partially assimilated to be more alike to the [ b].
Another distinction possible is between contact and distant assimilation. We talk about contact assimilation when the two sounds
influenced are neighboring each other - just as in the examples above. Distant assimilation, as opposed to that, is when two sounds not neighboring
each other influence each other. There is an example in Trask from Latin: There is a preform of "five" which is supposed to be *[peŋkʷe] which has
then changed to *[kʷeŋkʷe] resulting in the word quinque (after another vowel change). In that example, the [kʷ] in the middle of the word had influenced the [p] in the beginning to totally assimilate.
Regarding the sequence in which the sounds occur in the word (e.g. first [p] then [kʷ]), we can distinguish another category: anticipatory or regressive assimilation and perseverative or progressive assimilation. The example from Latin above was a regressive assimilation, because the assimilated sound was before the one causing the assimilation, so you could look at it like the [kʷ] in the middle of the word "reaches forward" through the word - which is not in the sense of the reading progression - that's why it's called regressive. These notions may very well be confused, even in Linguistic literature. Trask advises to use "anticipatory" and "perseverative" respectively.
Another possibility we haven't considered yet is that the two sounds may change both as well. This is called mutual assimilation.

Okay, that was a lot! But let's break it down to the basics: We have two dimensions of assimilation now. One being the distance and one being the direction. The distance can either be contact, with the sounds next to each other or distant; the direction can either be anticipatory (from right to left), perseveratory (from left to right) or mutual (in both directions at the same time).

2.1.2 Dissimilation

Dissimilation is the opposed process to assimilation. That means, two sounds become more distinct from each other. This seems to contradict the principle I stated before - by dissimilating two sounds, the effort to pronounce them rises. So when does dissimilation occur? First of all, dissimilation is by far more seldom than assimilation. It occurs when there is a repetion of sounds in a word. Trask mentions the Italian word <colonello> and the Spanish equivalent <coronelo>, where the [l] is dissimilated to [r], probably to avoid a repetion. Intrestingly, English uses the Italian orthography and the Spanish-like dissimilated pronounciation (cp. <colonel> -> /ˈkɜː.nl̩/ or /ˈkɜ˞ː.nl̩/).
Also, when two sounds of the same manner of articulation occur in succession, they may as well be dissimilated.
xijlwya
sinic
sinic
Posts: 263
Joined: 01 Sep 2010 15:55
Contact:

Lenition/Fortition

Post by xijlwya »

2.2 Lenition / Fortition

Lenition and fortition, or weakening and strengthening use a kind of arbitrary scale to say what is strong and what is weak. I just took the scale from trasks work. Generally, plosives seem to be the strongest obstructions to the airflow, which is why they are strong. Vowels seems to be the weakest obstruction of the airflow, hence they are weak. Now, all other manners of articulation are ordered in that scale.

2.2.1 Lenition

Lenition, or weakening means that a given consonant becomes weaker. To say what is weaker and what is stronger, we have to introduce some relations between consonants. Trask introduces the following:
  • 1. geminate > simplex: degemination
    • Latin cuppa "cup" -> Spanish copa "wine glass"
      Latin siccu "dry" -> Spanish secu
    2. plosive > frictaive > approximant: spirantisation
    • Latin habebat "he had" -> Italian aveva
      Latin faba "bean" -> Italian fava
    3. plosive > liquid: "liquidisation"
    • Standart English water /ˈwɔː.tər/ -> General American /ˈwɔː.ɾɚ/
    4. oral plosive > glottal plosive: debuccalisation (probably from latin "bucca" meaning cheek)
    • Standart English water -> Scottish /ˈwɔː.ʔər/
    5. non-nasal > nasal: nasalisation
    • Latin sabanu "covering" -> Basque zamau "table cloth"
    6. voiceless > voiced: voicing
    • Latin strata "road" -> Italian strada
      Latin lacu "lake" -> Italian lago
Note: ">" means "is stronger than" and the term "simplex" is used in a rather unusual manner here - it probably just means that a geminated consonant is reduced to a non-geminated one)

Generally speaking, the weaker variant demands less effort than the stronger variant regarding the energy used to produce the sound. With the principle mentioned under assimilation, which states that speakers tend to reduce the effort, sound changes are likely to occur going from left to right in the above relations. Such changes are called lenitions, because the sounds becomes "weaker". So, I can conclude: Lenition is when consonants change like in the list above.
Lenition is likely to occur in patterns where there are two vowels surrounding a single consonant. Lenition can be seen as the consonant becoming more "vowel-like". I think it is obvious that, to reduce effort, it is plausible for the consonant between to vowels to become more alike the vowels. This change can also be seen as an assimilation!
Lenition is not confined to occur in only one of the above scales at a time. And Trask states that the list may indeed by incomplete.

Another special kind of lenition is deletion where the affected consonant disappears entirely. Especially the [h] is very likely to be deleted entirely because it is the weakest consonant.
  • Latin homo "man" -> French homme /Om/
    Old English heafod -> English head
It is important to add that lenition is by no means restricted to occur between two vowels! It may as well occur word intial or final as well.

2.2.2 Fortition

Fortition is basically an inversion of lenition. Just as assimilation and dissimilation, lenition and fortition have asymmetric frequencies: lenition is by far more likely to occur. Still, fortition is not rare (whatever that means...).
Again, fortition seems to contradict the principle of reducing speech effort. But if only lenition would take place, languages would sooner or later all degenerate to zero - leaving no phonemes to speak - which is rather unlikely to happen.
Unfortunately, I haven't found a statisfactory explanation of this yet - Trask states, that languages preserve themselves mostly by borrowing.

Examples following
Alomar
cuneiform
cuneiform
Posts: 113
Joined: 13 Apr 2012 16:02
Location: Chicago
Contact:

Re: Guide to Sound Changes

Post by Alomar »

Aw man! I really would have loved/would still love more of this.
Native: :usa:
Conversational: :deu:
Learning: :fra: :ita: :grc:
Check out my Mychai Blog
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Guide to Sound Changes

Post by eldin raigmore »

Alomar wrote:Aw man! I really would have loved/would still love more of this.
Me too. I know way too little about this.
Iyionaku
mayan
mayan
Posts: 2102
Joined: 25 May 2014 14:17

Re: Guide to Sound Changes

Post by Iyionaku »

Thank you! That helped a lot. All I knew about sound shifts was the High German consonant shift.
Wipe the glass. This is the usual way to start, even in the days, day and night, only a happy one.
thetha
mayan
mayan
Posts: 1545
Joined: 29 Apr 2011 01:43

Re: Guide to Sound Changes

Post by thetha »

I could try to continue where the OP left off if they don't come back soon. If everyone's okay with that, anyway.
User avatar
Thrice Xandvii
runic
runic
Posts: 2698
Joined: 25 Nov 2012 10:13
Location: Carnassus

Re: Guide to Sound Changes

Post by Thrice Xandvii »

That'd be awesome, looks like its been a bit since the original activity... about 2 years ago!
Image
Clio
sinic
sinic
Posts: 228
Joined: 27 Dec 2012 23:45

Re: Guide to Sound Changes

Post by Clio »

Teddy wrote:I could try to continue where the OP left off if they don't come back soon. If everyone's okay with that, anyway.
Do you intend also to follow Trask, or some other resource(s)?

Either way, I'm much more than okay with your continuing this.
Niûro nCora
Getic: longum Getico murmur in ore fuit
scratchpad
User avatar
DesEsseintes
mongolian
mongolian
Posts: 4331
Joined: 31 Mar 2013 13:16

Re: Guide to Sound Changes

Post by DesEsseintes »

Teddy wrote:I could try to continue where the OP left off if they don't come back soon. If everyone's okay with that, anyway.
Yes, do! Please do! [:D]
thetha
mayan
mayan
Posts: 1545
Joined: 29 Apr 2011 01:43

Re: Guide to Sound Changes

Post by thetha »

Clio wrote:
Teddy wrote:I could try to continue where the OP left off if they don't come back soon. If everyone's okay with that, anyway.
Do you intend also to follow Trask, or some other resource(s)?

Either way, I'm much more than okay with your continuing this.
I don't have any books on historical linguistics, but I'm confident that I can provide sufficient examples for every attested type of change. I think that's all that's necessary; when we have questions about the viability of sound changes in conlangs we're just asking how the sound changes we want compare to the tendencies in existing languages. Thus all that is needed is to give a descriptive list of examples and glean some common trends for them. I'm sure I can do that! [:)]
User avatar
atman
sinic
sinic
Posts: 407
Joined: 05 Dec 2012 17:04

Re: Guide to Sound Changes

Post by atman »

Teddy wrote: I don't have any books on historical linguistics, but I'm confident that I can provide sufficient examples for every attested type of change. I think that's all that's necessary; when we have questions about the viability of sound changes in conlangs we're just asking how the sound changes we want compare to the tendencies in existing languages. Thus all that is needed is to give a descriptive list of examples and glean some common trends for them. I'm sure I can do that!
Very good! And thanks for doing what I'd never find the time to do, even if I do have several books on historical linguistics (all about the same family, however).
Երկնէր երկին, երկնէր երկիր, երկնէր և ծովն ծիրանի.
xijlwya
sinic
sinic
Posts: 263
Joined: 01 Sep 2010 15:55
Contact:

Re: Guide to Sound Changes

Post by xijlwya »

Hi guys.

Finally, someone apprechiates the work put into this [:P] .
Unfortunately, I don't see me continiuing this guide because the topic is pretty much off my head since I wrote the last lines of this guide. I would love to see it continued by someone else though.
Teddy wrote:I could try to continue where the OP left off if they don't come back soon. If everyone's okay with that, anyway.
That being said, I'm very ok with that :mrgreen: .
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Guide to Sound Changes

Post by eldin raigmore »

Thanks, xijlwya.
Prinsessa
runic
runic
Posts: 2647
Joined: 07 Nov 2011 14:42

Re: Guide to Sound Changes

Post by Prinsessa »

While it may perhaps be possible to refer to it as an assimilation still, many proto-indo-europeanists tend to consider the possibility of the case of quinque to have been one of analogy with the initial consonant of the neighbouring numeral quattuor. The same seems to have happened in the opposite direction in Germanic, where four took on the initial consonant of five (earlier *p-).

Nonetheless, it could have been a normal assimilation in both branches (as well as Celtic and probably others), or indeed an assimilation fueled by analogy.

Latin does seem to display similar processes elsewhere too, though, if one is to believe the suggestion that coquō (cook) should originally have had an initial *p- and would be cognate to a group of semantically corresponding words with initial p- in other IE languages. I don't know if that can be proven, but I suppose it does not seem unlikely. Among these cognates are Greek and Sanskrit words, neither language of which did any numeral dance.

Food for thought!
thetha
mayan
mayan
Posts: 1545
Joined: 29 Apr 2011 01:43

Re: Guide to Sound Changes

Post by thetha »

Lol I forgot that I made that promise. I'll just say that it may or may not happen at this point if I remember that this thread exists.
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Guide to Sound Changes

Post by eldin raigmore »

Skógvur wrote:While it may perhaps be possible to refer to it as an assimilation still, many proto-indo-europeanists tend to consider the possibility of the case of quinque to have been one of analogy with the initial consonant of the neighbouring numeral quattuor. The same seems to have happened in the opposite direction in Germanic, where four took on the initial consonant of five (earlier *p-). …. (and more) ….
Good points (IMG and IMO), Skógvur. Thanks for raising those possibilities
thetha
mayan
mayan
Posts: 1545
Joined: 29 Apr 2011 01:43

Guide to Sound Changes

Post by thetha »

xijlwya wrote: 2.2.2 Fortition

Fortition is basically an inversion of lenition. Just as assimilation and dissimilation, lenition and fortition have asymmetric frequencies: lenition is by far more likely to occur. Still, fortition is not rare (whatever that means...).
Again, fortition seems to contradict the principle of reducing speech effort. But if only lenition would take place, languages would sooner or later all degenerate to zero - leaving no phonemes to speak - which is rather unlikely to happen.
Unfortunately, I haven't found a statisfactory explanation of this yet - Trask states, that languages preserve themselves mostly by borrowing.

Examples following
Plenty of examples of fortition can be found in the Austronesian family. The most striking of these is the fortition of intervocalic glides among the languages of Sarawak and to maybe a lesser degree those of Sabah. observe these correspondences for the word for "two":

PMP (Proto-Malayo-Polynesian) *duha
No fortition:
+Indonesian: dua
+Malagasy: roa
Fortition:
+Kadazan: duvo
+Long Terawan: ləbbeh
+Tunjung: rəgaʔ
+Kiput: dufih
+Chamorro: hugwa

In all of these languages the intervocalic *h was lost, and then a *w was inserted in between the vowels for epenthesis. This *w then became the wide variety of consonants you can see here. A similar change seems to have happened in Spanish, where original /j/ is now pronounced /ʒ dʒ ʃ/ etc. by various regions. I suppose fortition can be roughly said to be a change where a sound decreases in sonority, but this definition might include things we wouldn't call fortition. I'm not sure yet.

I'll probably be using a lot of Austronesian examples because that's what I know off the top of my head, but I'll try to keep some variety as well in my future posts.
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Guide to Sound Changes

Post by eldin raigmore »

In Italian, what were originally two-stop clusters, have been replaced by a geminated stop; e.g. viz. cf. English "doctor" to Italian "dottore".
Is that not simplification by eliding one stop "followed by" fortition of the remaining stop?
Post Reply