Talk:Assimilation (phonology)

This  level-5 vital article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects:

Linguistics: Phonetics This article is within the scope of WikiProject Linguistics, a collaborative effort to improve the coverage of linguistics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.LinguisticsWikipedia:WikiProject LinguisticsTemplate:WikiProject LinguisticsLinguistics articles ??? This article has not yet received a rating on the project's importance scale. This article is supported by Phonetics Task Force.

I'm very confused by this:

/b/ becoming labialised as in "boot"

/b/ is already a labial stop, right? See Labial consonant.

Sorry if I am doing this talk page thing wrong, I'm a pretty new user.

Practically rewrote the whole thing (haven't finished yet and I'm sure it needs more proofreading and rechecking for clarity). The original definitions and discussions, while well-meaning, seemed awfully thin (there are brief definitions etc. under "sound change") and inaccurate in detail, and the examples struck me as not presented in an illuminating way. Especially the Russian ones, thoroughly opaque unless you already understand the principles. Alsihler 22:22, 14 December 2006 (UTC)[reply]

This all sounds more than a little condescending ("well-meaning", "thoroughly opaque"), but I can't seem to convince myself that the result is any better than the (now obviously no longer existent) original. At least we learn that there were examples of Russian (presumably regressive) assimiliation, which is, as the original author was, no doubt, aware, one of the "classic" cases of phonetic assimilation. Do I also understand correctly that you haven't managed to finish your rewrite of this article, having started in December, 2006? Maelli (talk) 03:37, 14 March 2012 (UTC)[reply]

This article isn't very clear, even for someone who has heard of assimilation before. I would suggest less long-winded sentences, more sub-sections, and possibly adding the examples to the theoretical explanations, to back them up. Zigzig20s 14:51, 13 January 2007 (UTC)[reply]

Surely assimilation isn't restricted to sound changes which occur at word boundaries? [User: STB, 08 June 2011)

I know of three types of assimilations, which would make it easy to read (three sections). This is probably restrictive, there are possibly more types of assimilations, and hopefully someone can expand my notes. I will, however, type what I know here - hopefully this can be used as a basis for a clean-up.

1 - regressive assimilation : a phoneme is influenced by another phoneme that comes before it

JohndanR sez: You're saying the same thing as (2), putting it in the passive voice. Why not just say "A phoneme influences a preceding phoneme."?

I recall "retrogressive assimilation" is the more common term. JohndanR (talk) 00:54, 11 January 2016 (UTC)[reply]

2 - progressive assimilation : a phoneme influences another phoneme that comes after it

3 - coalescence : fusion rather than influence. In this case :

• /t/ becomes /p/ before /p, b, m/
• /t/ becomes /k/ before /k, g/
• /d/ becomes /b/ before /p, b, m/
• /d/ becomes /g/ before /k, g/
• /n/ becomes /m/ before /p, b, m/
• /n/ becomes /ɳ/ before /k, g/
• /s/ becomes /ʃ/ before /ʃ, j/
• /z/ becomes /ʒ/ before /ʃ, j/

eg : goodbye /gʊbbaɪ/, that pen /ɤappen/ (sic - sorry not the right symbol for the 'bad' sound)

I was also wondering if anyone knew which phonetician discovered this theory. It should also be noted that I don't always agree with it - for instance, in "that pen" I use a glottal stop and say, "tha' pen"...(then again I don't speak 'RP') Zigzig20s 15:09, 13 January 2007 (UTC)[reply]

Hi, the article says:

• Assimilation is a typical sound change process by which the phonetics of a speech segment becomes more like that of another segment in a word (or at a word boundary), so that a change of phoneme occurs.
• All become /m/ before /p/, /b/, and /m/ [...]

Shouldn't this be changed so that it mentions allophones too? I mean, "All become [m] before /p/, /b/, and /m/" looks better IMHO. --Kjoonlee 01:21, 15 January 2007 (UTC)[reply]

Well, there's assimilation in pronunciation, which can be one of the causes of allophony, and there's assimilation in diachronic change, which can actually be a source of new phonemes ... --User:Eritain on 23 June 2007

Aahh... I'd never thought of it that way, but it makes sense. Thank you :) --Kjoonlee 08:01, 23 June 2007 (UTC)[reply]

I interpret the text as assimilation inevitably conveys a change of phonemes. This is wrong. In Swedish, for instance, the /ø/ phoneme is realized as [ø] in all positions but in front of /r/, where it (in most varieties) is realized as the more open [œ]. The word höra (eng. to hear) is pronounced as [hœ:ra]. The alternative pronounciation, [hœ:ra], doesn't convey a different meaning from [hœ:ra]. The two different sounds are instances of the same phoneme. Therefore I will change the text. --130.237.171.141 (talk) 10:37, 5 August 2008 (UTC)[reply]

How is that an example of assimilation? —Angr 10:45, 5 August 2008 (UTC)[reply]

A pattern in the distribution of sounds, such as described for Swedish, is not assimilation; it's just part of the way the language is at a given moment. A change over time in the distribution of sounds might be assimilation, depending on what the change is. You'd have to compare modern Swedish to an earlier form of the language to see assimilation. The essential distinction between assimilation and other sound change is that a change of phoneme occurs.

The lead paragraph of this article doesn't make the "over time" element of the definition clear, and in fact to find it stated in a lead I had to follow the link to "sound change", and then follow the link from that lead to "language change". Pi zero (talk) 12:17, 5 August 2008 (UTC)[reply]

But assimilation need not be diachronic. There is also synchronic assimilation, as in [tɛn̪ θɪŋz] ten things, which incidentally is also an example showing that the result of assimilation need not be a separate phoneme. —Angr 12:20, 5 August 2008 (UTC)[reply]

I'm wrong. Curiously, I'm also partly right (not about the same thing, of course).

On one hand, there is a serious problem here, that may span this article and several others in its neighborhood: even without the removed passage (twice removed, at last count), the lead sentence here does in fact say that assimilation is strictly diachronic — or, more charitably, one could argue that it says this article is strictly about diachronic assimilation — although, as I said, you have to thread through several articles to realize that's what it's saying. Repairing the problem, at least technically, might require only a tweak somewhere, like not saying that assimilation is sound change, or not saying that sound change is language change; but even if so, working out what to tweak seems to require a strategic understanding of a whole network of articles.

On the other hand, I was having defective deja vu. I had an encounter early this year with another mutli-article muddle involving the Sound change article — but not involving this article, which was what threw me off. The phrase that's now been removed here was, I see, added by an unregistered (though apparently not entirely uninformed) user early in 2007. It seems to have stood up for a year and a half, though, which I suspect may be because a whole series of articles in this neighborhood are confusing, notably in that their subject boundaries are difficult to work out. The muddle I was involved in was over the difference between sound change and phonological change, which apparently is a matter of whether phonemes are involved (hence my neuronal misfire — sorry about that). Pi zero (talk) 14:03, 5 August 2008 (UTC)[reply]

Angr. To answer your question above: The /r/-sounds (and supradentals/retroflexes) in Swedish requires a lowering of the yaw since /r/ involves an apical articulation, the yaw lowering means that the preceeding /ø/ is realised as the more open [œ]. Most of the other Swedish alveolars/dentals are laminals which require the mouth to be more close and therafore the preceeding /ø/ is still pronounced as [ø]. However this rule doesn't apply to all varieties of Swedish, but it in most varieties it is an important phonological rule.

Well, unless there's independent evidence that Swedish /r/ is phonologically [+low] or something, I think it's a very tenuous example of assimilation. I don't see why an apical pronunciation of /r/ should entail lowering of the jaw (not "yaw") anyway. Are other vowels lowered before /r/, or just /ø/? —Angr 15:42, 14 August 2008 (UTC)[reply]

Angr: Yes! The unrounded counterpart, /ɛ/, is realized as the more open [æ] before /r/ (or retroflexes). By the way... In the south of Sweden /r/ is realized as an voiced uvular fricative (or trill). If the mentioned rule above only could be explained as a co-articulatory effect the phenomenon would not show up these parts (since the uvular articulation wouldn't entail a lowering of the yaw). However, the rule does show up even there. Therefore it is a assimilation (part of the Swedish phonology), if my hypothesis of /r/ entails yaw lowering is correct. To my knowledge, there is no investigation about laminal articulation in Swedsih coronals (I could check more thoroughly though). Most litterature says that Swedish coronals are more dental (fronted) than in e.g German and English. I the book "Talkommunikation" by Robert McAllister, the feature high/low is unspecified for /r/ and retrofelxes.

My impression however is that typical German and English foreign accent is caracterized by that lack of laminals (sounding more like the retroflexes (or supradentals)). Finally, this discussion (about the Swedish rule) might not be completly relevant to this article since it treats assimilation in general. I think we can agree on the fact that a synchronous assimilation means a sound change that can be a change of allophones or even phonemes. —Preceding unsigned comment added by 130.237.171.141 (talk) 10:25, 15 August 2008 (UTC)[reply]

There's also vowel harmony, consonant harmony, and consonant gradation which are all complex assimilatory processes involving phonemes. —Preceding unsigned comment added by Nevarren (talk • contribs) 11:15, 23 March 2010 (UTC)[reply]

The word bank, pronounced [bæŋk], is given as an example illustrating anticipatory assimilation of a nasal consonant to a stop consonant. This would mean that there is a phonological process realizing /bænk/ as [bæŋk]. But do we have a good reason to assume a phonemic /bænk/ giving rise to an uttered [bæŋk]? For whatever reason, the orthography is bank, but orthographical rules can be somewhat arbitrary and be guided by non-phonological considerations. The orthography of the word might as well have been bangk. In the financial meaning, the word comes from French banque /bɑ̃k/, and in the meaning "incline" from proto-Germanic *bangkon. A better example is unkind [ʌŋˈkaɪnd] – or it would be if that word is actually pronounced that way, but I'm not sure it is.  --Lambiam 16:23, 30 January 2011 (UTC)[reply]

There's a section where the article reads that 'Some authorities distinguish between partial and complete assimilation'. If there are linguists out there who don't, then this is news to me. Is everyone cool with this sentence being made more passive? Something like 'Assimilation can be complete or partial...'. 63.224.17.43 (talk) 12:53, 21 September 2012 (UTC)[reply]

While cupboard could be a good example of assimilation and degemination absorbed historically (diachronic restructuring), it makes no sense synchronically, and will surely confuse/puzzle readers not steeped in phonology. Does anyone object to it being removed? 47.32.20.133 (talk) 14:56, 22 June 2018 (UTC)[reply]

Yes, I would object. This article covers both synchronic and diachronic assimilation, right in the lead section.  — SMcCandlish ☏ ¢ 😼  16:20, 22 June 2018 (UTC)[reply]

That depends on what "covered" means. They're both there, but presented with no explanation of the distinction, and in the third paragraph the opportunity to illustrate is totally missed. Read the text as though you know nothing of phonology and phonetics, you're not consciously aware of structure vs. surface phonetics, etc. Aside from misuse of phonemic / / where phonetic [ ] are called for ([ˈhæmbæɡ] is a pronunciation; /ˈhæmbæɡ/ would be a structure), there's no explanation in the lead section of the distinction between synchronic phonetic assimilation and diachronic restructuring, leaving the uninformed reader to figure out for him/herself the chasm in status between the handbag and cupboard examples -- or not figure it out. (Somewhat less urgently, handbag is a complex example /ndb/ → [mb] involving not only assimilation -- same for handkerchief further on -- whereas examples such as in California, incomprehensible illustrate assimilation simply and clearly.) There's also the opportunity to pull up an example from below to illustrate: historical Latin /kt/ > Italian /tt/ → [tt] (e.g. otto) and synchronic /kt/ → [tt] (ictus).47.32.20.133 (talk) 19:58, 22 June 2018 (UTC)[reply]

I don't think your points is wrong at all. However, the solution to inclarity and insufficient information isn't deletion of more information and of examples. As for which examples are clearer and simpler, a potential solution is moving them. If both the lead examples are technically more complex than we'd like, use them as complex examples (with explanations of the complexity) in the body, and recycle some of the simplest ones in the lead (or add some new, similar ones). We don't need to actually lose anything. And it will actually be helpful to retain (with more material) the complex ones, as an illustration of how assimilation can interface with other linguistic processes.  — SMcCandlish ☏ ¢ 😼  20:29, 22 June 2018 (UTC)[reply]

We may be saying the same thing to some extent, although very differently. My point is that a principle should be introduced as clearly as possible, illustrated with examples that are as transparent as possible. Neither is the case with handbag or cupboard in the intro. 47.32.20.133 (talk) 21:38, 22 June 2018 (UTC)[reply]

Right. I'm just urging a shuffling of examples and addition of some if necessary, without removals, as to the total content.  — SMcCandlish ☏ ¢ 😼  23:13, 22 June 2018 (UTC)[reply]

In the example of Regressive Assimilation, we see this:

• 読んだ: yom (/jo̞m/) + da (/da̠/) → yonda(/jo̞nda̠/)

But gramamtically it is

• 読んだ: yom (/jo̞m/) + ta (/ta̠/) → yonda(/jo̞nda̠/)

So, why is this an example of Regressive Assimilation, as in my understanding this better characterises Reciprocal Assimilation. Is there something that I'm missing here? Because I'm confused. CJLippert (talk) 19:52, 3 October 2019 (UTC)[reply]

This article was the subject of a Wiki Education Foundation-supported course assignment, between 17 October 2022 and 30 November 2022. Further details are available on the course page. Student editor(s): Holymohly (article contribs).

— Assignment last updated by Holymohly (talk) 18:59, 11 November 2022 (UTC)[reply]

Hello! I am the student editing this article - I'm going to be adding a new section titled "How Assimilation Has Altered English Pronunciation", with a subsection that specifically focuses on the affrication of /tr/ and /dr/, which is a more recent development in English. I will be adding this section tonight - feel free to give criticism/make edits! Holymohly (talk) 04:36, 28 November 2022 (UTC)[reply]

I just made the update - I decided to change the section title to "Influence on Language". Any edits/suggestions welcome! Holymohly (talk) 06:04, 28 November 2022 (UTC)[reply]

This article was the subject of a Wiki Education Foundation-supported course assignment, between 11 January 2023 and 27 April 2023. Further details are available on the course page. Student editor(s): Calvacr, PraiseGod.A (article contribs).

— Assignment last updated by PraiseGod.A (talk) 15:36, 30 March 2023 (UTC)[reply]