Talk:Metathesis (linguistics)

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Linguistics / Phonetics   
WikiProject icon 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.
 ???  This article has not yet received a rating on the project's quality scale.
 ???  This article has not yet received a rating on the project's importance scale.
Taskforce icon
This article is supported by the Phonetics Task Force.

American Sign Language section?[edit]

I'm thinking of adding a section on metathesis in Sign Language (specifically ASL) but I don't think I have the photos which would be beneficial, if not crucial, to understanding the explanation. I guess even if I had the pictures, I've never tried uploading and linking to photos in wiki. Any help or any comments? --Alex DG 04:12, 19 December 2006 (UTC)

There is a curious note under the American Sign Langauge section (about the information not being cited) that doesn't appear in any of the other language examples that aren't cited. Why is that? Also, The introduction still talks about metathesis being a process affecting sounds and syllables. Although American sign language is assumed to have syllables it certainly doesn't have 'sounds'. I would suggest we reword the intro to refer to "phonological units" or "sub-lexical units" rather than sounds or syllables.Alex DG (talk) 17:39, 25 January 2011 (UTC)

I know this is an old comment, but I'd like to cover this for anyone else who has such a question.
While some people use the tag indiscriminately, the {{cite}} tag often carries with it the implication that the information given is thought to possibly be inaccurate. In fact, in its original usage, the tag was used to indicate that information should be removed if a citation was not found.
It's also quite likely that the person tagging it knew something about sign language but not about the other languages, and thus only felt comfortable tagging the sign language section. (Although the possibility also exists that, since metathesis was defined as having to do with sounds, the tagger understandably questioned the possibility of metathesis in a silent language.) — trlkly 12:54, 25 January 2014 (UTC)

Other stuff[edit]

Spelling changes do not cause phonological changes; the US pronounciation is not a result of a spelling change in "jewelry/jewellery". I pronounce it "julrey", with only two syllables, which I suspect is the most common US prononunciation. (And I pronounce "jewel", normally, with a single long diphthong, in contrast to the usual two-syllable British pronunciation.) I think the spelling change reflects the three -> two syllable change, and not the metathesis "julary". Hence, I'm deleting entirely the comment about spelling. --Tb 00:45 27 Jul 2003 (UTC)

It struck me also that the pronunciation "julary" is the result of an analogy that likens the word to the many English words that end in -ary or -ery, rather than the result of spelling. -- IHCOYC 15:20 27 Jul 2003 (UTC)

Definitely. Metathesis is very frequently associated with analogy to other words. --Tb 20:43 27 Jul 2003 (UTC)

What phonetic transcription is being used in this article? It's not X-SAMPA—I've never heard anyone pronounce "jewelry" with a doubled /o/ sound. Gwalla | Talk 17:51, 9 Aug 2004 (UTC)

It's a proposed representation for English that I created after reading complaints about X-SAMPA being unintelligible, IPA untypeable, and English "phonetic spellings" hard to interpret. See Wikipedia:English phonetic spelling. There seems to be no great haste to adopt it. Smerdis of Tlön 18:05, 9 Aug 2004 (UTC)
If you're going to use a nonstandard phonemic representation, please say so somewhere. Frankly, I'd prefer IPA, X-SAMPA, or a set of symbols like those used in dictionary pronunciation keys to a system that uses digraphs confusingly ("ie" for IPA /i/ or English "long e"?). I agree about English "phonetic spellings" being awful—they're frequently confusing and occasionally misleading. Gwalla | Talk 21:38, 9 Aug 2004 (UTC)
I put in a tag to the page explaining them (EPS) but it seems it was not conspicuous enough. AAR, I have reverted it back to English phonetic spelling like it originally stood; that's probably good enough to get the point across. Smerdis of Tlön 03:02, 10 Aug 2004 (UTC)

Would anyone else agree that the pronunciation 'Febyuary' for February isn't metathesis, but a contraction of the 'r' sound? It doesn't seem to exhibit the same reordering of the other examples. Xyzzyva 19:01, Dec 6, 2004 (UTC)

I tend to agree; it's syncope instead. -- Smerdis of Tlön 20:01, 6 Dec 2004 (UTC)

I think aksian vs. ask and hwat vs. what also belong here, but I'm not sure of their ancient forms. Adam78 13:36, 14 October 2005 (UTC)

hwæt > what is not metathesis. This simply gives two examples of the spelling of aspirated w: hw in OE, wh in Modern English, which is usually pronounced as regular initial w. --Rich 19:46, 23 March 2006 (UTC)

I just added the "wymon for women" example. I think it's a good common example, but as the vowel is also shifted if someone fells strongly elsewise please feel free to remove it. --Zombiejesus 03:23, 27 January 2006 (UTC)

What do you mean 'wymon'? Is that a representation of pronunciation? If so, we need some more information to decipher it.

stragety for strategy. witness heqs 18:47, 8 July 2006 (UTC)

Are "realator" and "nucular" really examples of metathesis? I think they're more accurately be described as examples of epenthesis, but certainly not metathesis.

Is "liberry" really metathesis? Seems to me more like an elision of the first r sound.

Could someone confirm whether the standard English pronunciation of 'iron' as if it were 'iern' is an example of metathesis (and if so, provide an IPA rendering of the latter)? Stevvers 18:43, 20 January 2007 (UTC)

I can't provide a source but I think it probably went something like this:
  • /ˈaɪron/→ After the Great English vowel shift.
  • /ˈaɪrən/→ Vowel reduction.
  • /ˈaɪərn/→ Metathesis.
  • [ˈaɪɚn]→ While the schwa and /r/ may be seperate phonemically, phonetically they are combined into a rhotic schwa. Ƶ§œš¹ [aɪm ˈfɻɛ̃ⁿdˡi] 21:26, 20 January 2007 (UTC)
Thanks. I've just checked the OED, which claims that "apparently" the sequence was /ˈaɪrən//ˈaɪərən//ˈaɪər(ə)n/ via syncopation, rather than metathesis. (The modern pronunciation that they give is the nonrhotic (British) version: /ˈaɪən/.) Stevvers 00:24, 22 January 2007 (UTC)
That's really odd. I was under the impression that schwa was only epenthisized before an /r/ in the coda, not syllable initial ones. But the OED is the OED. Ƶ§œš¹ [aɪm ˈfɻɛ̃ⁿdˡi] 04:46, 22 January 2007 (UTC)
How is /ˈkʌmftɚbl̩/ for comfortable metathesis, based on the description here? It looks just like elision of the 'or' to me. Skittle 17:12, 2 February 2007 (UTC)
The pentultimate vowel is ɚ, so a rhotic rather than a plain shwa, so there's metathesis of r and .
What I however am having problems understanding are these two:
--Tropylium 12:33, 20 June 2007 (UTC)


I think the mention of the meaning of metathesis in chemistry is redundant due to the fact that it is mentioned in the disambiguation. The statement in the article ought to be deleted.

The first sentence of the linked scientific paper is plausibly useful -- and on-point: "Derived from the Greek words meta (change) and thesis (position), metathesis is the exchange of parts of two substances. In the reaction, AB + CD → AC + BD, B has changed position with C." In this context, the otherwise abstract concepts of metathesis are especially well-visualized in the unique diplomas which were crafted for King Carl XIV Gustaf to present to the three chemists who shared the Nobel Prize in 2005.
Ooperhoofd 13:21, 20 June 2007 (UTC)

Israel and vegetable[edit]

Hello, All.

The article gives these as examples of metathesis in English:

I think they should not be included. I see no transposition of phonemes, only the shortening or omission of sounds.

The first change in Israel is the omission of a sound for the initial I. The several typical variant pronunciations for the rest of the word seem to hing on

(1) whether the speaker believes that the a should be pronounced
or [ɑ] or [a] (with a [j] or [i] sound to soften the transition to [ɛ] or [ə], for the e),
or [i] (the [i] linking sound, but missing the [ɑ] or [a] that might precede it);
(2) whether the speaker believes that the e should be pronounced [ə] or [ɛ];
and (3) whether the speaker believes that the letters ae should be treated as a single unit pronounced [eɪ], as in Gaelic.

No transposition.

In [ˈvɛtʃtəbəl] and [ˈvɛdʒtəbəl], there is omission of the sound of the medial e, along with, in one case, devoicing of [dʒ] into [tʃ]. Again, no transposition.

I will remove those words from the list.

President Lethe 03:52, 19 September 2007 (UTC)


How sure are we about this one? I could envision a series such as this: miraculo > *miragulo > *miraglo > *miragro > miraglo with plain ol' liquid dissimilation. Step 3 could come about if it was reached after gl- >> ll had started. --Tropylium (talk) 17:53, 27 June 2008 (UTC)

Greek and Latin[edit]

Can anyone recommend good references for metathesis in Greek and Latin?--Meieimatai 04:46, 7 July 2008 (UTC)

Murciélago (Spanish)[edit]

In the section about Spanish, it was written:

  • murciégalo for murciélago

implying that "murciégalo" (the non-standard variant) is a metathesis of "murciélago" (the standard variant, "blessed" by the Spanish Royal Academy). In fact, this is not true. In this case, it is the standard variant that is a methatesis of the variant contemporarily considered as non-standard. To see this, look up "murciélago" in the Spanish Royal Academy dictionary and check its etymology. --Antonielly (talk) 13:27, 5 March 2009 (UTC)

Poor example[edit]

The example in the introduction (pronouncing 'comfortable' as if it was spelt 'comfterble') is surely an example of elision and not metathesis — in fact, it is even used as an example in the elision article. —Stephen Morley (talk) 14:44, 23 April 2009 (UTC)

We've gone over this before. comfortablecomfterble --Trɔpʏliʊmblah 11:40, 25 April 2009 (UTC)
We're talking about /ˈkʌmfətəbəl/ versus /ˈkʌmftəbəl/ — the difference I see is an elided schwa. There are much better examples we can use — the asterisk example further down the article is an indisputable case of metathesis. —Stephen Morley (talk) 14:35, 25 April 2009 (UTC)
And in rhotic accents (as noted!), it's from /-fɚtə-/ to /-ftɚ-/. But you're right, a better example would work for non-rhotic accents too. A medial case appears to be more typical than the ks/sk change (cf. the other languages) — I'll go with "foilage" since there's no contending definition. Strictly speaking it's not a simple phoneme reversal tho, but there's no need to get pedantic about it there.--Trɔpʏliʊmblah 09:33, 28 April 2009 (UTC)
If there's agreement about this can we take it out? As noted above this seems to be an example of a schwa elision and calling it metathesis is dubious at best. Fysidiko (talk) 15:17, 2 May 2010 (UTC)


"What is sound before and after metathesis depends on assumption of language ancestry if protowords cannot be attested." Any ideas what that gibberish means? Wikipedia needs to be in plain english. Futhermore, are all those examples from other cultures strictly necessary? This is the English Language Wiki, after all. This entry sounds like a final PhD dissertation on The Analysis of Metathesis in Protolanguages of Eastern Senegal.Jubilee♫clipman 14:22, 17 September 2009 (UTC)

Say that in Examplese there is a word "wurkis" and in the related Forinstancian a corresponding word "wukris": it is clear that metathesis has occur'd, but it is not immediately clear in which language.
The other language examples are, I think, useful for demonstrating that metathesis can have different degrees of systematicality. By the English examples alone we'd risk giving the impression that metathesis only occurs as a sporadic speech error. --Trɔpʏliʊmblah 22:06, 27 September 2009 (UTC)

Third, Thirteen and Three[edit]

This triplet is given as an example of metathesis in English. I doubt it, because this particular example could be simply a manifestation of phonological evolution in English. Compare Latin: tres (three), tertius (third). Has Latin undergone a similar metathesis? I have seen that the words *treyes (three) and *trtyos (third) are posited for the Proto-Indo-European with the syllabic -r- in the second word. Then historical evolution explains why we have tres but tertius (not tretius) in Latin, and likewise, in English. We want a specific example on metathesis in English, not one which illustrates the evolution of non-vocalic syllablic sounds r, l, m and n. —Preceding unsigned comment added by (talk) 22:40, 5 July 2010 (UTC)

Pronunciation of metathesis[edit]


Data didn't mispronounce it and nor is the IPA pronunciation given the 'correct' (i.e. only) pronunciation, by a long chalk. Stressing of syllables is especially variable in English speakers even moving from dialect to dialect throughout the UK. Considered across all of the English speaking populations this is even more the case. Indeed, the listed pronunciation is one which seems least like to me (a native English speaker, travelled widely), as it happens.

/'mɛtɘθiːsɘz/ and /mɛtɘ'θiːsɪs/ for example are two pronunciations which I can attest and which are both far more likely than a 'schwa'd initial ee combined with a strongly stressed second syllable, in my opinion.

In any case, there is no way it can be argued that it was mispronounced, as is currently stated, when used in ST:TNG. —Preceding unsigned comment added by (talk) 15:19, 15 May 2011 (UTC)

The pronunciations métathesis and metathésis are correct based on English rules of accent, which place accent on the first syllable of each part of a compound word. Here the parts are meta and thesis, and each of these two pronunciations accent one of these parts. But historically the accenting of the word metathesis has been based on Latin rules of accent, which place accent on the third syllable from the end unless the second syllable from the end is long (heavy). So based on English rules of accent mètathésis is the correct pronunciation, but based on Latin rules metáthesis is correct. — Eru·tuon 02:53, 11 June 2011 (UTC)

/ˈkʌmftərbəl/ Not an Example of Metathesis[edit]

I can't cite a source or I'd just edit the article, but I'm quite certain that /ˈkʌmftəɹbl/ (which the article gives as /ˈkʌmftərbəl/, a pronunciation I'm pretty sure does not occur) is not an example of metathesis. It represents an intrusive /ɹ/ as a hypercorrection of the non-rhotic contraction /ˈkʌmftəbl/. The /ɹ/ does not exchange positions with any other phoneme. Expected metathetic forms for "comfortable" would be */ˈkʌmfrətəbl/ or */ˈkʌmfətəɹbl/, neither of which exists, as far as I know. Jdcrutch (talk) 19:37, 10 June 2011 (UTC)

I think the article's analysis is metathesis of the sequence /ər/ (then deletion of /ə/): /-fəɹtəbəl/ → */-ftəɹəbəl/ → /-ftəɹbəl/. /əl/ isn't that odd; it's just another way of phonemically analyzing syllabic /l̩/. — Eru·tuon 22:10, 10 June 2011 (UTC)
That's how I pronounce it, and my dialect is rhotic. /ˈkʌmfr̩təbl̩/ → /ˈkʌmftr̩bl̩/ may not be the most obvious example of metathesis, but in a rhotic dialect I don't know how else you'd characterize it. — kwami (talk) 02:32, 11 June 2011 (UTC)
If I had to hazard a guess, I'd say that since the dialect of origin is non-rhotic, the careful pronunciation of the word should be /'kʌmfɜtɨbl/ . . . so, were I English, my not-so-careful pronunciation might be /'kʌmfətəbl/, which could then be skinnied down to /'kʌmftəbl/. I think this might actually be an instance of elision and Cheshirization. Uuords (talk) 09:52, 17 December 2013 (UTC)


This is more likely lenition: /ˈaɪ.r̩n/ > /ˈaɪrn/, if underlying /rə/ is assumed /ˈaɪ.rən/ > /ˈaɪɚn/. ᛭ LokiClock (talk) 14:28, 29 August 2011 (UTC)

I don't understand. Is the change of syllabic r to non-syllabic r (or non-syllabic r to r-coloring) an example of lenition? Such a sound change isn't listed in the Lenition article.
The metathesis explanation is supported by the OED, which lists the phonetic series /iːrən/ > /aɪrən/ > /aɪərən/ > /aɪər(ə)n/ > /aɪə(r)n/ > /aɪən/, of which the second and last forms are the ones given in this article. — Eru·tuon 16:44, 29 August 2011 (UTC)
Lenition is my characterization of /ˈaɪ.r̩n/ > /ˈaɪrn/, but now I realize that's not the change it's talking about. My suspicion was for equating /ɚ/ with /ər/. /aɪən/ solves it, going with rhotic & non-rhotic saying that these /r̩, ə/ come from /ər/. Would you note OED's explanation here or on Wiktionary? ᛭ LokiClock (talk) 07:55, 30 August 2011 (UTC)

Natural/Conscious etc[edit]

I think not enough explanation is given on the reasons metathesis takes place - i.e. a clear distinction needs to be made in cases of natural progression in the development of a language (i.e. like the Spanish milagro (miracle) or Argelia (Algeria)), and completely intentional examples such as verlan. It could even be mentioned that spoonerisms are an example of metathesis (originally unintentional; subsequently - intentionally comic), rather than simply having it in the see also section. BigSteve (talk) 12:05, 21 May 2012 (UTC)

Can I have a crips please?.:)1812ahill (talk) 20:12, 27 January 2013 (UTC)