hypercorrection

In linguistics, hypercorrection is defined as usage of pronunciation or linguistic rule that many informed users of a language consider incorrect, but that the speaker or writer uses through misunderstanding of prescriptive rules, often combined with a desire to seem formal or educated.^[1]

Linguistic hypercorrection occurs when a real or imagined grammatical or phonetical rule is applied in a mistaken or non-standard context, so that a desire to be "correct" leads to an incorrect result: Faced with enough exceptions to a rule, the speaker might mistake the exception for the general rule, applying it to situations where it never was meant to occur.

For example, a person might be told that the past tense of "to take" is "took", not "taked", and the past tense of "to shake" is "shook", not "shaked". He might therefore assume that all verbs ending in "-ake" have "-ook" for their past form and apply this to regular verbs where "-aked" would be the correct ending. He would end up saying that he "took the flour and book (rather than baked) a cake with it".

Forms

Hypercorrection may take different forms, such as:

• Grammatic hypercompensation: The effect where an exception to a rule is mistakenly thought to be the rule itself
• Hypercompensation of pronunciation: The effect where a rule of pronunciation for certain words is mistakenly applied to a series of other words
• Hyperforeignism: The effect, where exceptions in grammar or pronunciation in foreign words imported from one language are applied to foreign words from another language. In an extreme form, this can also apply to words that have a foreign origin but have since been assimilated to follow the general rules or even words that are not foreign at all, but just sound that way.
• Hyperurbanism: The applying of forms of grammar and speech associated with refined society to words or sentences that are completely outside the intended scope (defined by Kingsley Amis as an "indulged desire to be posher than posh").
• Phonetic hypercompensation, or overcompensation: The effect that results when a student of a new language has learned that certain phones of his or her original language must usually be replaced by another in the studied language, but has not learned when not to replace them (or has learned, but must consciously remind himself or herself of the exceptions and hence sometimes forgets not to replace).^[2]

English

English has no authoritative body governing "correct" usage. This is unlike some other languages, such as French (Académie française), Italian (Accademia della Crusca), Icelandic (Icelandic Language Institute), Spanish (Real Academia Española) or Maltese (Akkademja tal-Malti). Nonetheless, within groups of users of English, certain usages are considered unduly elaborate adherences to "formal" rules.

Preposition at the end of a clause

Winston Churchill is said to have replied to a hypercorrective memo with the phrase "This is the kind of tedious nonsense up with which I will not put!" (or a similar construction).^[3] That is an example of hypercorrection used as parody: Churchill went beyond creating a grammatically correct sentence, to mock the elaborate refusal to end a clause in a preposition (or insistence on placing the preposition before the relative pronoun); he treated the adverbial particles up and with as prepositions. They are actually part of the phrasal verb put up with, and their placement before put is extremely unusual.

Personal pronouns

Jack Lynch, assistant professor of English at Rutgers University, describes another example of hypercorrection:

We're taught as children, and beginning language learners are told, you don't say 'me and you went to the movies.' It should be 'you and I.' And a lot of people, therefore, internalize the rule that 'you and I' is somehow more proper, and they end up using it in places where they shouldn't -- such as 'he gave it to you and I' when it should be 'he gave it to you and me.'

—^[4]

Hyperforeignism

A hyperforeignism is a special type of hypercorrection resulting from an unsuccessful attempt to apply the reading rules of a foreign language to a loan word (for example, the application of the reading rules of one language to a word borrowed from another), or occasionally to a native English word believed to be a loan word. The result may be "absurd," reflecting "neither the reading rules of English nor those of the language from which the word in question comes."^[5]

Latin and Greek words

Jejune, pronounced /dʒɨˈdʒuːn/ or /dʒiːˈdʒuːn/, is often taken to be a French word and pronounced 'je jeune' although it is in fact Latin in origin.^[6]

The noun octopus is often made plural in English as octopi, originally from the mistaken belief that it is of Latin origin and that all Latin nouns ending in -us take -i to form their plural. However, this is only correct for Latin masculine nouns of the second declension. For Latin fourth-declension nouns, such as manus, the singular and plural forms both end in -us. In fact, octopus is derived from Greek, not Latin. Octopuses is generally considered correct in modern English^[7][8], but the plural in Greek is octopodes.

French words

Non-native French speakers may erroneously omit the last consonant in Vichyssoise /z/,^[9] in the chess term en prise, and in prix fixe. Similarly, in "coup de grâce" some speakers may omit the final consonant /s/, although it is pronounced in French [ku də gʁas] (see also entry Coup de Grâce).^[9]

The word cadre is sometimes pronounced /ˈkɑːdreɪ/ in English, as though it were of Spanish origin. In its French original, the final "e" is silent.^[9]

In French, the final R of the word reservoir is pronounced. A hyperforeign realization may omit it, even though the speaker would normally pronounce a word-final R.^[9]

The "Queen of the hyperforeignisms" is the word lingerie.^[9] Speakers of American English pronounce this /lɑːnʒərˈreɪ/, excessively depressing the first vowel of the French [lɛ̃ʒəʁi] to sound more like a "typical" French nasal vowel, and rhyming the final syllable with English ray, by analogy with the many French loanwords ending in -é, -er, -et, and -ez.

The word "cache" may be pronounced with a final /eɪ/ (as in "cachet"). In French, the final "e" is silent.^[9]

Those who know a little French omit the final z or s in many French proper nouns such as Saint-Saëns, Duras, Boulez, and Berlioz, among many others which do not adhere to standard rules of French pronunciation.

The English word margarine is an interesting case where hypercorrection caused a violation of a standard English reading rule, namely, that g is normally realized as /ɡ/ when it precedes a written a. This word was originally derived from German (Oleo)margarine, which in turn had borrowed it from the French (who had invented it). In both French and German, margarine is pronounced with a hard g as in "go" or "get". Because other words which are pronounced with a soft g in English have a hard g in German, such as intelligent, the dominant pronunciation in English took a soft g for "margarine" to compensate for this.^{[citation needed]}

Spanish words

The digraph ch of Spanish is generally realized /tʃ/, similarly to English. Hyperforeign realizations of many Spanish loanwords or proper names may substitute other sounds. Examples include a French-style /ʃ/ in the surnames Chávez or of Augusto Pinochet ^[9], or a German-influenced {x} in machismo.^[5] This last word may also be pronounced with a [k] on the analogy of English words like "masochism" and "anarchism".

The English word junta derives from Spanish, where the initial consonant is today realized as [x] or [h]. Hyperforeign realizations substitute [j] or [ʒ].^[10]

Italian words

The "g" in Adagio may be realized as /ʒ/, even though the Italian original has an affricate /dʒ/.^[5]

The word bruschetta, particularly in American English is commonly rendered as /bru:'ʃɛtə/ with an English 'sh' sound, probably as a result of Americans' familiarity with words and surnames of German origin containing 'sch,' which would be pronounced this way. An approximation more reflective of Italian phonology would be /bru:'skɛtə/ and the authentic pronunciation in Italian would be /brus'ketta/. A similar problem afflicts the brand name Freschetta, which is routinely pronounced with the 'sh' sound in commercials.

Greek words

Most English speakers pronounce the z in schizophrenia with a /ts/ sound. The "schiz-" prefix is derived from the ancient Greek σχίζειν (skhizein) meaning "to split". The "z" would be pronounced /z/ under English or modern Greek reading-rules, and /dz/ or /zd/ under those of ancient Greek. The word was coined in German, as Schizophrenie, where a /ts/ pronunciation dominates and dominated; but /z/ was the preferred pronunciation for schizophrenia from the word's introduction in English in 1912 until approximately the 1960s. It remains in use in other (rare) words featuring the prefix such as schizocarp and schizogamy. In the 1960s, the /skits/ pronunciation became popular under the influence of German, although, oddly enough, it did not affect the "sch", making the current pronunciation similar to the Italian version schizofrenia (see above Italian words section). As of 2003, the /ts/ pronunciation is the only one given in some major American dictionaries.^[11]

Dutch and Afrikaans words

In Dutch, the combination "sch" is pronounced [sx], except at the end of a word, when it is pronounced [s]. (In Afrikaans, the same combination is sometimes heard as [sk].) However, most English speakers pronounce it as [ʃ] ("sh") following the rules for German, in words such as Rooibosch and veldtschoen.^[12]

Indian languages

The J in the name of the Taj Mahal or raj is often rendered /ʒ/, though a closer approximation to the Hindi sound is /dʒ/.^[5] (J in most other Roman-alphabet spellings of words associated with languages of India is best approximated /dʒ/.)

Another example is the pronunciation of Punjab as /ˈpʊndʒɑːb/ instead of the correct /ˈpʌndʒɑːb/ (help·info)^[13]: the letter u, in the Anglo-Indian spelling convention, usually represents the Hindi neutral vowel, with a sound similar to that of the u in English cup [ʌ]. (The name comes from the Persian panj āb meaning five waters.)

In many words pertaining to Indian religion, an originally short vowel is lengthened in some English pronunciations. Examples include i in Sikh and Shiva and, in American English, u in Buddha and Buddhist. Thus, for example, Sikh may be pronounced to rhyme with "seek", although a rhyme with "sick" would be closer to the original in most English dialects, and would comport better with standard English reading rules.

East Asian languages

Some English-speakers (including the BBC radio news) mispronounce Beijing with /ʒ/, even though the Mandarin Chinese sound represented by the <j> in Pinyin (/tɕ/) is an affricate. The same realization of "J" may afflict mah-jongg.^[5]

Many English speakers pronounce "Genghis Khan" as /ˈɡɛŋɡɪs ˈkɑːn/ with a hard initial g as in "get", in accordance with the usual transliteration systems for Asian languages. In fact the original Mongolian name was something like Tchinggiz, and the spelling "Genghis" was first used by Marco Polo, an Italian writing in French. pronounced /ˈdʒɛŋɡɪs ˈkɑːn/, with a soft g as in "gentle" in accordance with the medieval pronunciation of both those languages, would therefore be closer to Marco Polo's intention as well as to the original name.

Hyperforeignism for comic effect

The silent "t" in "Report" in the title of the parody pundit show The Colbert Report is a hyperforeignism used for comedic effect. It is a play on the host's surname, Colbert (pronounced /koʊlˈbɛər/;^[14]), which is said to be French within the show's fictional back-story, though it is actually Irish.

Hyacinth Bucket, the snobbish main character in the BBC sitcom Keeping Up Appearances, insists her last name is pronounced "Bouquet" /buːˈkeɪ/. According to her husband Richard, Bucket is pronounced as is the water receptacle. However for Hyacinth, it is pronounced "Bouquet," à la française. Series creator Roy Clarke said he got the inspiration for this character trait after meeting someone with the surname "Bottom" who insisted it was pronounced "Botome".

Similarly, some people jokingly give the American retailer Target the pseudo-French pronunciation /tɑrˈʒeɪ/ tar-zhay, as though it were an upscale boutique.

Certain newly-genteel London suburbs were jocularly re-named "Clahm" (Clapham), "Ba-TER-zee-a" (Battersea), "St. Ockwell" (Stockwell), "DAH-zhen-um" (Dagenham), and the like.^[15]

Other languages

This section does not cite any references or sources. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed. (November 2009)

Chinese languages

Modern Cantonese is currently undergoing a phonological shift, one of the changes being the dropping of the initial ng- consonant (pronounced [ŋ]). For instance, the word 牙 (ngaa4, meaning "tooth"), ends up being pronounced aa4 (Note: Cantonese romanization provided using Jyutping). Prescriptivists tend to consider these changes as substandard and denounce them for being "lazy sounds" (懶音). However, in a case of hypercorrection, some speakers have started pronouncing words that should have a null initial using an initial ng-, even though according to historical Chinese phonology, only words with Yang tones (which correspond to tones 4, 5, and 6 in Cantonese) had voiced initials (which includes ng-). Words with Yin tones (1, 2, and 3) historically should have unvoiced or null initials. Because of this hypercorrection, words such as 愛 (oi3, meaning "love"), which has a Yin tone, are pronounced by speakers with an ng- initial, ngoi3.

Speakers of some accents of Mandarin, particularly in the south of China and in Taiwan, pronounce the retroflex initials zh-, ch- and sh- as the alveolar initials z-, c- and s-. Such speakers may hypercorrect by pronouncing words that should start with z-, c- and s- as if they started with their retroflex counterparts.

In Taiwan, under the influence of Taiwanese (Min Nan), many people pronounce the initial f- as h-, and often hypercorrect by pronouncing the initial h- as f-. This is also noticeable in the Hakka population, where many words that begin in h- in Mandarin and Taiwanese begin in f- in Hakka. (Examples: 火, 花)

Bulgarian

In standard Bulgarian and in the eastern dialects, the old yat letter is pronounced as я ("ya") when stressed and the following syllable does not contain the vowels и ("i") or е ("e"), and pronounced as е in all other cases. But in the western dialects it is always pronounced as е. Attempting to speak correctly, some speakers from Western Bulgaria mispronounce many words containing the yat letter - голями ("golyami"), желязни ("zhelyazni"), бяли ("byali"), видяли ("vidyali"), спряни ("spryani"), живяли ("zhivyali") instead of големи ("golemi"), железни ("zhelezni"), бели ("beli"), видели ("videli"), спрени ("spreni"), живели ("zhiveli"). This trend is especially common with past participles such as видяли.

Russian

In pronouncing foreign loanwords, native Russian speakers sometimes palatalize consonants: for instance, pronouncing modern as mod'ern. This partly arises from spelling conventions. In native Russian words, most consonants undergo palatalization before so-called "soft vowels" (or one could say these vowels are written after palatalized consonants).

However, many English and French loanwords in Russian that contain the Russian letter "е" (IPA:/e/, /ɛ/ or /ə/) do not follow this rule, because the nonpalatalized э, that would correctly represent the sound, is only supposed to be written at the beginning of a word or after another vowel (as in Aeroflot).

Examples of hyperforeignisms are found in Russian when loanwords (commonly older loanwords) contain consonants that should be palatalized. Yet some speakers, emphasizing the foreign quality of the word, do not palatalize them. For example: theme (тема), technical (технический), text (текст), museum (музей), gazette (газета) and effect (эффект). Note, тема is perhaps not a very good example of this, as Russian has no "th" sound. The Russian letters allow for either "f" or "t" as the first consonant here. In order not to induce undue difficulty (and inevitably mispronunciation), Russian opts to simplify the loanword to a word that conforms to the Russian alphabet. Indeed, most Russians find "th" a difficult sound, much as foreigners often mispronounce the ы sound in Russian.

West South Slavic languages

The syllables je and ije appear in Croatian, Bosnian, and Montenegrin speech where Serbian has only variation in quality (length of the vowel) of e. Not every Serbian e becomes je or ije like in the other West Balkan countries. Serbian speakers may hypercorrect their dialect by either undersupplying or oversupplying the jes and the ijes.

German

Düsseldorf dialect versus Rhineland dialect

In German, the dialect spoken in the city of Düsseldorf and its surroundings heavily features the front 'ch' sound (aka the "ich sound", [ç]) where standard German calls for the 'sch' [ʃ] sound. Speakers with this accent would say 'Fich' [fɪç] instead of 'Fisch' [fɪʃ] (fish), and 'Tich' [tɪç] instead of 'Tisch' [tɪʃ] (table). This is due to a hypercorrection of the Rhineland accent prevalent in that area of Germany, an accent that often replaces the front 'ch' [ç] sound with the 'sch' [ʃ] sound. Attempting to avoid this error, speakers of the Düsseldorf accent hypercorrect it to an abundance of 'ch' [ç].

Genitive versus dative

Another example is use of the genitive case where the dative case is required. Colloquially, the genitive is often dropped in favor of the dative even if correct grammatical usage demands the genitive. Because language critics deride such substitution, many German speakers use the genitive even with prepositions that actually demand the dative (e.g., entgegen, entlang, gegenüber), seemingly under the false impression that the genitive is always right and the dative is always wrong, or at least that the genitive is a better form of the dative.

Norwegian

The French "Entrecôte" and "Pommes frites" more often than not are pronounced without the final "t" sound. ("Antrekå" or "Angtrekå" and "Pommfri" or "Pomfrii".) Traditionally all french loan words or brands ending in -on, -en or -ant is pronounced -ong, -eng or -ang in Norwegian [and Swedish] (for instance "ballkong", "Citroeng" and "Resturang" or "Ressturang"). This supposedly sounding more "foreign", "french" and posh than an ordinary Norwegian pronunciation of the words. ("ballkån", "Citroenn" and "Restaurant" or "Resstaurant") Citroën is an ironic example of this.

Swedish

An example of hyperforeignism in Swedish is the common use of "chevré" in "chevré[ost]" for "chèvre cheese", which is pronounced quite different from the original French "chèvre". (Possibly by, false analogy with the Swedish "grevé" cheese grevéost.)

Similarly "Entrecôte", which also can often be spelled "Entrecoté", or "Entrêcotè", or some other combination of ^ and ` or ´. More often than not it is pronounced without the ending "t" sound. (Prudery may be a factor here, since the Swedish word "kåt", sounding similar to "côte" ("rib"), means "horny" and the result can be taken to mean "horny in the hallway".)

An example of a hypercorrection is the spelling "åtminstonde" for standard Swedish åtminstone (at least), where the pleonastic "d" can be explained as a hypercorrection among speakers who normally reduce the complex "-nd" to /n:/.

French Bureau meaning desk in Swedish is usually pronounced "Byyrå" with a strong accent on the first syllable, although this is not a comman way to accentuate nouns in Swedish. Presently it is not considered a loan word, but assumedly it could have been pronounced in this odd way to sound more foreign.

Dutch versus West-Flemish

The local dialects of the West-Flanders region do not use the Dutch "ch" /x/ (as in the Scottish 'Loch'). Instead they pronounce both 'g' and 'ch' as a soft 'h', whereas the Standard Dutch way to pronounce it would be 'g'. For example, a West-Flemming would pronounce the phrase 'een gouden hart' (a golden heart) as 'een Houden hart'. Some older people, who grew up speaking nothing but their dialect, are unaware that there is a difference between 'g', 'ch' and 'h' altogether and trying to 'mimic' Dutch, they often overcompensate and pronounce every word they would normally pronounce with a 'h'-sound as a 'g'. This includes, words actually pronounced 'h'. In the example above they would go overboard and pronounce the phrase 'een gouden hart' as 'een gouden Gart'.

In a continuing folk tale an unspecified pastor of some unspecified West Flemish church wants to impress his flock by celebrating mass in flawless 'civilized' ABN Dutch. His 'civilized' Dutch consists out of pronouncing a 'ch' and 'g' correctly as /x/ (instead of the 'h' as West-Flemish dialect does). However to be absolutely sure, he also starts pronouncing the 'h' as /x/ even though he should keep pronouncing it as a 'h'. The effects are hilarious: Instead of praying for "De hele kerk" (the whole church) he ends up praying for "de gele kerk" (the yellow church) and the holy virgin ("de heilige maagd") becomes "de geilige maagd" (The virgin in heat). Finally he ends his sermon in asking what should be "de goede hulp van de Heer" (the good help of the Lord). Instead he asks for "de goede gulp van de geer" : the good trouser opening of the manure.

Latin

In the Middle Ages, the spelling of Latin was simplified in various respects: for example, æ and oe became e, and ch became c. Occasionally these changes were reversed, and e and c were sometimes expanded to æ (or oe) and ch, even when such spelling contradicted Classical Latin. For example, caelum was contracted to celum and re-expanded to coelum. These spellings are often preserved in English derivatives, including et cætera and et coetera (occasionally found as variants for et cetera); foetus (originally fetus); lachrymose, from lachryma (a false Hellenisation, originally lacrima, "a tear"); and schedule, from schedula (originally scedula).

Hebrew and Yiddish

Careful Hebrew speakers are taught to avoid the colloquial pronunciation of בדיוק (bediyyuq, "exactly") as [biˑ.ˈdjuk]. Many speakers accordingly pronounce להיות (lihyot, "to be") as if it were spelled "lehiyyot" ([lɛˑ.hiˑ.ˈjot]), though there is no grammatical justification for doing so.

It is well known that the vowel kamatz gadol, which in the accepted Sephardic pronunciation is rendered as /aː/, becomes /ɔ/ in Ashkenazi Hebrew (and therefore in Yiddish). On the other hand, the vowel kamatz katan, which is visually indistinguishable from kamatz gadol, is rendered as /o/ in both pronunciations. This leads to hypercorrections in both directions.

1. The consistent pronunciation of all forms of kamatz as /a/, disregarding katan and chataf forms, could be seen as a hypercorrection, when Hebrew speakers of Ashkenazic origin attempt to pronounce Sephardic Hebrew (e.g. צָהֳרָיִם, "midday" as "tzaharayim", rather than "tzohorayim" as in standard Israeli pronunciation; the traditional Sephardi pronunciation is "tzahorayim"). This may however be an example of over-simplification rather than hypercorrection.
2. Conversely, many older British Jews consider it more colloquial and "down-home" to say "Shobbes", "cholla" and "motza", though the vowel in these words is in fact a patach, which is rendered as /a/ in both Sephardi and Ashkenazi Hebrew. Other hypercorrections occur when speakers of Israeli (based on Sephardic) Hebrew attempt to pronounce Ashkenazi Hebrew, for example for religious purposes. The month of Shevat (שבט) is mistakenly pronounced "Shvas", as if it were spelled *שְׁבַת. In an attempt to imitate Polish and Lithuanian dialects, kamatz (both gadol and katan), which would normally be pronounced [ɔ], is hypercorrected to the pronunciation of cholam, [ɔj], rendering גדול ("large") as goydl and ברוך ("blessed") as boyrukh.

See also

• Disputed English grammar
• Hypocorrection
• List of English words with disputed usage

Notes

References

• Labov, William. 1966. "Hypercorrection by the Lower Middle Class as a Factor in Linguistic Change". In Sociolinguistics: Proceedings of the UCLA Sociolinguistics Conference, 1964. William Bright, ed. Pp. 84–113. The Hague: Mouton.
• Joshua Blau, On Pseudo-Corrections in Some Semitic Languages. Jerusalem: Israel Academy of Sciences and Humanities 1970.