Some speakers who use a non-standard accent or dialect of a language, occasionally desire to "adjust" their speech to the standard.

I'm interested in knowing if there is a word for when this fails based on an inference error.

An example (or so I've heard) of this would be some communities in New York who have dropped the "r" at the end of words -- maya (mayor), heata (heater), etc (avoiding IPA here). When these speakers decide to use the "standard" pronunciation, they realize they "have to put the Rs back". The mistaken inference I'm referring to would cause a word like Brenda to be pronouced "Brender" as it could (incorrectly) sound like its "r" had been dropped. A famous example of this is the Billy Joel song, Scenes From an Italian Restaurant, a story about Brender and Eddie.

Recently I've been studying Mandarin in Taipei, one of my teachers is occasionally making a similar adjustment that has caused me to learn not just a non-standard pronunciation, but to actually pronounce it in way that no one would say, except for someone making this inference error.

The most recent example is learning to pronounce "always - zǒng" as "zhǒng". The issue is that some Mandarin speaking communities often pronounce the "zh" as "z", and since she pronounced in "zǒng" in this case, she mistakenly thought she was pronoucning it wrong (an incorrect inference in this case), and reversed the "error", and taught me "zhǒng".

So in order to deal with my frustration, at least I can know if this has a name in the linguistics world. If it doesn't we could name it after her :)

  • 2
    Not heard of a special term for this but 'overcorrection' immediately springs to mind as a possible variant. [also, your question reminded me of somewhat similar but still a bit different effect, where people pronounce a sound not present in the actual word or phrase. Example: "I saw it" pronounced as "I sore it" - with an 'r'. Can't recall the term for this either].
    – tum_
    Jun 28, 2016 at 11:01
  • @A.Toumantsev yes I thought of that too, and it might be a reenforcing separate issue. This has been formalized in French. They "insert" a consonant in inversion-based questions between words that are bounded by vowels "Y a-t-il un problème?", "A-t-on"... Of course the French had to think of something to address the emigration of so many of their consonants. I guess they just made some up out of thin air, lol.
    – pixelearth
    Jun 28, 2016 at 11:07
  • Yes, I do have some elementary French, so I'm aware of that. I'm now trying to think if we have anything similar in Russian... [Spotting such things in your native language is usually much harder...]
    – tum_
    Jun 28, 2016 at 11:23
  • @pixelearth I don't think the French consonant insertion is related to mispronounciations or overcorrection errors which would mean they pronounce it somewhat "wrong", but it's a simple and generalisable phonological rule to facilitate pronounciation, as "Y a-il un problème" with two subsequent vowels contradicts unmarked (possibily even well-formed) expressions in French. I assume this happens for the same reason as liaison between vowels (the word-final -s in les cheveux is silent, while the -s in les élèves is pronounced), which has nothing to do with inference errors either. Jun 28, 2016 at 13:58
  • @lemontree agreed, which is why I referred to it as a "separate issue".
    – pixelearth
    Jun 28, 2016 at 14:50

3 Answers 3


The closest term to what you need is hypercorrection which is sometimes called hyperurbanism:

In linguistics or usage, hypercorrection is a non-standard usage that results from the over-application of a perceived rule of grammar or a usage prescription. A speaker or writer who produces a hypercorrection generally believes that the form is correct through misunderstanding of these rules, often combined with a desire to appear formal or educated.

The Wikipedia article I gave a link to has examples of hypercorrection from different languages.

  • 4
    Very good. And this is something I'm seeing in the reports of our Spanish team every day: "Some English-Spanish cognates primarily differ by beginning with "s" vs. "es," such as the English word "spectacular" and the Spanish word "espectacular." A native Spanish speaker may conscientiously hypercorrect for the word "establish" by writing or saying "stablish," which is archaic."
    – tum_
    Jun 28, 2016 at 13:19
  • 1
    @tum_ "Lucy, you got some 'splaining to do!"
    – iacobo
    Sep 7, 2019 at 16:03
  • Very common for Common Czech speakers when using Standard Czech. Jan 19, 2021 at 11:04

Your "Brender" is not an example of hypercorrection but of the standard non-rhoticity in Billy Joel's (Bronx?) dialect. This is a very interesting phenomenon with lots of related concepts, including linking/"intrusive" R, sandhi, and epenthesis.

I can't comment on the Chinese question, except to tell you it is almost certainly an example of something different--probably something closer to "hyper-correction"--than this. So it doesn't make sense to yoke these two examples together.

  • I'm confused. This article seems to support my opinion pretty clearly. Billy Joel is from Oyster Bay NY, and the regional pronunciation is as you say, predominantly non-rhotic. There is no final "r" in Brenda to begin with. So that fact that this phantom "r" is brought out in an intervocalic context, just proves it is a hyper correction. But this specific hypercorrection doesn't need an intervocalic context to be brought out. I've heard "pizza" sound like "pizzer". Anyway they both seem to me to be interesting examples of hypercorrection.
    – pixelearth
    Jun 30, 2016 at 17:23
  • 3
    @pixelearth The article discusses how the linking-r is characteristic of non-rhotic dialects, i.e., not an extrinsic feature aimed at hypercorrection or anything else. See this part: en.wikipedia.org/wiki/…
    – SAH
    Jun 30, 2016 at 18:23
  • I think we're just interpreting the information differently. The WP article calls this "an overgeneralizing reinterpretation", which I feel in this case is tantamount to saying "fossilized hypercorrection". This in itself is just as valid as any of the reasons that language changes over time. Curiously, one of the "further reading" references is "Halle, Morris; Idsardi, William (1997). "r, hypercorrection, and the Elsewhere Condition". I don't know if I care enough to look up that publication, though.
    – pixelearth
    Jul 1, 2016 at 3:35
  • 1
    @pixelearth Another example of a dialect in which this phenomenon takes place (again, as an attendant feature of non-rhoticity) is British Received Pronunciation, i.e., the language of the British Upper Class. In that case it is surely not hypercorrection, fossilized or otherwise. While I don't know enough to say definitively that it isn't a sign of hypercorrection in Billy Joel's case, that explanation fails a good test: Occom's razor.
    – SAH
    Jul 1, 2016 at 18:32

When this is in fact what is happening, this is often referred to as ʜʏᴘᴇʀᴄᴏʀʀᴇᴄᴛɪᴏɴ, as noted in another post here. However, what may be perceived as a failed attempt to mimic a prestige variety of a language is often, in fact, merely a feature of a variety of the language that the observer is unfamiliar with. The ignorance in such cases is on the part of the observer, not the observee.

The widely observed, widely studied phenomenon of intrusive /r/ in prevocalic environments is also widely misunderstood by non-linguists. Prevocalic intrusive /r/ occurs exclusively in non-rhotic varieties of English where it is used to separate a non-high vowel from a following vowel of any description. In rhotic accents a glottal stop is likely to be used in such environments. Non-rhotic varieties of English include the prestige varieties dominant in the English speaking countries in the southern hemisphere, and England itself as well as several regional varieties in North America. Intrusive /r/ is a well-known characteristic of modern so-called Received Pronunciation, the prestige variety in the south of the UK. The idea that intrusive /r/ here is not a naturally occurring part of the language but is due to these speakers wanting to emulate the (unfairly) socially stigmatised speakers of less prestige varieties would, of course, be ridiculous. However, only slightly more so than the idea that this feature of non-rhotic English spoken in North America is due to these speakers 'aspiring', but failing, to emulate their rhotic neighbours.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.