In common parlance, "guttural R" is the phenomenon whereby a rhotic consonant (an "R-like" sound) is produced in the back of the vocal tract (usually with the uvula) rather than in the front portion thereof and thus as a guttural consonant. Speakers of languages with guttural R typically regard guttural and coronal rhotics to be alternative pronunciations of the same phoneme, despite articulatory differences. Similar consonants are found in other parts of the world, but they often have little to no cultural association or interchangeability with coronal rhotics (such as [r], [ɾ], and [ɹ]) and are (perhaps) not rhotics at all.
Guttural realization of a lone rhotic consonant is typical in most of what is now France, French-speaking Belgium, Germany, Denmark and the southernmost parts of Sweden and south - western parts of Norway; it is also frequent in the Netherlands, Dutch-speaking Belgium, and Switzerland. It also occurs as the normal pronunciation of one of two rhotic phonemes (usually replacing an older alveolar trill) in most of Portugal, various parts of Brazil and among minorities of other Portuguese-speaking regions.
The French rhotic has a wide range of realizations: both the voiced [ʁ] and voiceless uvular fricatives [χ] (the first of the two also realized as an approximant), the uvular trill [ʀ], the alveolar trill [r], and the alveolar tap [ɾ]. These are all recognized as the phoneme /r/, but most of them (all except [χ] and [ʁ]) are considered dialectal. For example, [ʀ] was once typical of a working class Parisian accent, while [r] is sometimes found in southern France, as well as (increasingly less) in North America.
Today in northern France, /r/ is commonly pronounced as a voiced [ʁ] or voiceless uvular fricative [χ] after a voiceless consonant by assimilation. [ʁ] is also the most common pronunciation in the French media. In much of southern France, this guttural R has replaced the traditional alveolar trill [r], which can now only be heard among the oldest speakers.
It is not known when the guttural rhotic entered the French language, but it may have become commonplace in the mid or late 18th century. Molière's Le Bourgeois gentilhomme, written in the 17th century, has a professor describe the sound of /r/ as an alveolar trill.
In Quebec French, speakers from rural areas and older generations traditionally use an alveolar trill, as was the pronunciation that was likely retained after the French colonists in Canada were isolated from France. However, most urban Quebeckers and the French-language broadcast media have adopted the modern guttural rhotic pronunciation of Paris, perhaps as the result of influence by modern French media from France.
Generally speaking, classical choral and operatic French pronunciation requires the use of an alveolar trill when singing, since an alveolar trill is easier to project than any guttural sound, be it a uvular trill or a uvular fricative.
Standard versions of Portuguese have two rhotic phonemes, which contrast only between vowels. In older Portuguese, these were the alveolar flap /ɾ/ (written ⟨r⟩) and the alveolar trill /r/ (written ⟨rr⟩). In other positions, only ⟨r⟩ is written in Modern Portuguese, but it can stand for either sound, depending on the exact position. The distribution of these sounds is mostly the same as in other Iberian languages like Spanish, i.e.:
- ⟨r⟩ represents a trill when written ⟨rr⟩ between vowels; at the beginning of a word; or following /n/, /l/, /z/, or /ʒ/. Examples: carro, rua, honrar, Israel.
- ⟨r⟩ represents a flap elsewhere, i.e. following a vowel or following any consonant other than /n/, /l/, /z/, or /ʒ/. Examples: caro, quatro, quarto, mar.
In the 19th century, the voiced uvular fricative [ʁ] penetrated the upper classes in the region of Lisbon in Portugal as the realization of the trill. By the late 20th century, it had replaced the alveolar trill in most of the country's urban areas. The uvular trill [ʀ] is also heard sometimes.
A common realization of the word-initial /ʀ/ in the Lisbon accent is a voiced uvular trill fricative [ʀ̝].
The Setúbal dialect uses the voiced uvular fricative [ʁ] for all instances of "r" — word start, intervocalic, postconsonantal and syllable ending. This same pronunciation is attested in people with rhotacism and in non-native speakers of French origin.
In Africa, the classical alveolar trill is mostly still dominant, due to separate development from European Portuguese.
In Brazil, the normal pronunciation of ⟨rr⟩ is voiceless, either as a voiceless velar fricative [x], voiceless uvular fricative [χ] or a voiceless glottal fricative [h]. In many dialects, this voiceless sound not only replaces all occurrences of the traditional trill, but is also used for all ⟨r⟩ that is not followed by a vowel (i.e. when at the end of a syllable, which uses a flap in other dialects). The resulting distribution can be described as:
- A flap [ɾ] only for single ⟨r⟩ and only when it occurs either between vowels or between a preceding consonant (other than /n/, /l/, /z/, or /ʒ/) and a following vowel. Examples: caro, quatro.
- A voiceless fricative [x] [χ] or [h] everywhere else: when written ⟨rr⟩; at the beginning of a word; at the end of a word; before a consonant; after /n/, /l/, /z/, or /ʒ/. Examples: carro, rua, honrar, Israel, quarto, mar.
In the three southernmost states, however, the alveolar trill [r] remains frequent, and the distribution of trill and flap is as in Portugal. Some speakers use a guttural fricative instead of a trill, like the majority of Brazilians, but continue to use the flap [ɾ] before consonants (e.g. in quarto). Among others, this includes many speakers in the city of São Paulo and some neighboring cities, though an alveolar approximant [ɹ] is also common, not only in the city, but the approximant is the dominant articulation in São Paulo state, the most populous state in Brazil. The caipira dialect has the alveolar approximant [ɹ] in the same position.
In areas where ⟨r⟩ at the end of a word would be a voiceless fricative, the tendency in colloquial speech is to pronounce this sound very lightly, or omit it entirely. Some speakers may omit it entirely in verb infinitives (amar "to love", comer "to eat", dormir "to sleep") but pronounce it lightly in some other words ending in ⟨r⟩ (mar "sea", mulher "woman", amor "love"). This tendency also occurs in some African countries; but speakers in Rio often resist the tendency, pronouncing a strong fricative [x] or [χ] at the end of such words.
The voiceless fricative may be partly or fully voiced if it occurs directly before a voiced sound, especially in its weakest form of [h], which is normally voiced to [ɦ]. For example, a speaker whose ⟨rr⟩ sounds like [h] will often pronounce surdo "deaf" as [ˈsuɦdu] or even [ˈsuɦudu], with a very slight epenthetic vowel that mimics the preceding vowel.
In most Spanish-speaking territories and regions, guttural or uvular realizations of /r/ are considered a speech defect. Generally the single flap [ɾ], spelled r as in cara, undergoes no defective pronunciations, but the alveolar trill in rata or perro is one of the last sounds learned by children and uvularization is likely among individuals who fail to achieve the alveolar articulation. This said, back variants for /r/ ([ʀ], [x] or [χ]) are widespread in rural Puerto Rican Spanish and in the variety of Ponce, whereas they are heavily stigmatized in the variety of the capital. To a lesser extent, velar variants of /r/ are found in some rural Cuban (Yatera, Guantánamo Province) and Dominican vernaculars (El Cibao, eastern rural regions of the country) In the Basque-speaking areas of Spain, the uvular articulation has a higher prevalence among bilinguals than among Spanish monolinguals.
Guttural realization of /r/ is mostly considered a speech defect in Italian (cf. rotacismo), but the so-called r moscia which is sometimes uvular is found in about 10% of speakers in some northern areas, such as Piemonte (Alessandria, Torino) and Emilia Romagna (Parma).
Continental West Germanic
Many Low Franconian and Low Saxon varieties have adopted a uvular rhotic. Many Central German varieties have also adopted a uvular rhotic, whereas many of the Upper German varieties have maintained an alveolar trill ([r]). The development of uvular rhotics in these regions is not entirely understood, but a common theory is that these languages have done so because of French influence, though the reason for uvular rhotics in modern European French itself is not well understood (see above).
The Frisian languages usually retain an alveolar rhotic.
Dutch and Afrikaans
In modern Dutch, quite a few different rhotic sounds are used. In Flanders, the usual rhotic is an alveolar trill, but the uvular rhotic does occur, mostly in the province of Limburg, in Ghent and in Brussels. In the Netherlands, the uvular rhotic is the dominant rhotic in the southern provinces of Noord-Brabant and Limburg, having become so in the early twentieth century. In the rest of the country, the situation is more complicated. The uvular rhotic is common, but not dominant, in the western agglomeration Randstad, including cities like Rotterdam, The Hague and Utrecht (the dialect of Amsterdam usually has an alveolar rhotic though, but the uvular is becoming increasingly more common). The uvular rhotic is also used in some major cities outside of the Randstad area, such as Zwolle, Almelo and Leeuwarden. Outside of these uvular rhotic core areas, the alveolar trill is common. People learning Dutch as a foreign language also tend to use the alveolar trill because it contrasts better with the voiceless velar fricative /x/ in Dutch. The Afrikaans language of South Africa also uses an alveolar trill for its rhotic, except in the non-urban rural regions around Cape Town where it is uvular (called a bry).
Most varieties of Standard German are spoken with a uvular rhotic, even though the first standardized pronunciation dictionary by Theodor Siebs prescribed an alveolar pronunciation. The alveolar pronunciation [r ~ ɾ] is used in some standard German varieties of Germany, Austria and Switzerland. Non-standard varieties employ the alveolar trill more often.
The upper/lower distinction also historically influenced the development of upper and lower dialects of Yiddish, the historic vernacular language of Ashkenazi Jews. As these Jews migrated to Eastern Europe (and later America etc.), they brought their particular pronunciations with them.
Insular West Germanic
Historically, English also had an alveolar trill [r] for the letter R. But this sound was replaced by an alveolar approximant [ɹ] rather than by a guttural pronunciation.
Speakers of the traditional English dialect of Northumberland and northern County Durham used a uvular rhotic known as the "Northumbrian Burr". However, this is no longer used by most contemporary speakers, who generally realize /r/ as an alveolar approximant, [ɹʷ], in common with other varieties spoken in the Anglosphere.
Alveolar rhotics predominate northern Scandinavia. Where they occur, they affect the succeeding alveolars, turning the clusters /rs/ and /rt/, /rd/, /rn/, /rl/ retroflex: [ʂ ʈ ɖ ɳ ɭ]. Thus the Norwegian word "norsk" is pronounced [nɔʂk] by speakers with an alveolar flap. This effect is rare in the speech of those using a uvular R ([nɔʁsk]).
Danish and Swedish
To some extent in Östergötland and still quite commonly in Västergötland, a mixture of guttural and rolling rhotic consonants is used, with the pronunciation depending on the position in the word, the stress of the syllable and in some varieties depending on whether the consonant is geminated. The pronunciation remains if a word that is pronounced with a particular rhotic consonant is put into a compound word in a position where that realization would not otherwise occur if it were part of the same stem as the preceding sound. However, in Östergötland the pronunciation tends to gravitate more towards [w] and in Västergötland the realization is commonly voiced.
Most of Norway uses an alveolar flap, but about one third of the inhabitants of Norway are now using the uvular rhotic. In the western and southern part of South-Norway however, the uvular rhotic is still spreading. The origin was the city of Bergen. Because retroflex consonants are mutations of [ɾ] and other alveolar or dental consonants, the use of a uvular rhotic means an absence of most retroflex consonants.
While the use of an alveolar trill or flap is generalized among speakers of Icelandic, a uvular rhotic is a fairly common pronunciation variant in the language, although it is usually frowned upon as defective pronunciation and compared to stammering and other similar speech disorders.
In Slavic languages, the alveolar trill predominates, with the use of guttural rhotics generally seen as defective pronunciation. An exception are the languages of the Sorbian minority in Saxony, eastern Germany, which are typically spoken with a uvular trill due to German influence. The uvular rhotic may also be found in a small minority in Silesia and other German-influenced regions of Poland and also Slovenia, but is overall quite rare even in these regions. It can also be perceived as an ethnic marker of Jewishness, particularly in Russian where Eastern European Jews often carried the uvular rhotic from their native Yiddish into their pronunciation of Russian.
In Hebrew, the classical pronunciation associated with the consonant ר rêš was tapped [ɾ], and was grammatically treated as an ungeminable phoneme of the language. In most dialects of Hebrew among the Jewish diaspora, it remained a tap [ɾ] or a trill [r]. However, in some Ashkenazi dialects as preserved among Jews in northern Europe it was a uvular rhotic, either a trill [ʀ] or a fricative [ʁ]. This was because many (but not all) native dialects of Yiddish were spoken that way, and their liturgical Hebrew carried the same pronunciation. Some Iraqi Jews also pronounce rêš as a guttural [ʀ], reflecting their dialect of Arabic.
An apparently unrelated uvular rhotic is believed to have appeared in the Tiberian vocalization of Hebrew, where it is believed to have coexisted with additional non-guttural, emphatic articulations of /r/ depending on circumstances.
Though an Ashkenazi Jew in Czarist Russia, the Zionist Eliezer ben Yehuda based his Standard Hebrew on the Sephardic dialect originally spoken in Spain, and therefore recommended an alveolar [ɾ~r]. But as the first waves of Jews to resettle in the Holy Land were northern Ashkenazi, they came to speak Standard Hebrew with their preferred uvular articulation [ʁ~ʀ] as found in Yiddish or modern standard German, and it gradually became the most prestigious pronunciation for the language. The modern State of Israel has Jews whose ancestors came from all over the world, but nearly all of them today speak Hebrew with a uvular [ʁ~ʀ] because of its modern prestige and historical elite status.
Many Jewish immigrants to Israel spoke a variety of Arabic in their countries of origin and pronounced the Hebrew rhotic as an alveolar tap [ɾ], similar to Arabic ر rāʾ. Gradually, many of them began pronouncing their Hebrew rhotic as a voiced uvular fricative [ʁ], somewhat similar to some Arabic dialects for pronouncing غ ġayn. However, in modern Sephardic and Mizrahi poetry and folk music an alveolar rhotic continues to be used.
- The Tigris dialects, a group among the Qǝltu dialects in Iraq, for instance in Mosul
- The Jewish and Christian dialects in Baghdad
- The Jewish dialect in Algiers
- The dialect of Jijel in Algeria
- Some muslim-urban dialects of Morocco (e.g. in Fes)
The uvular /r/ was attested already in vernacular Arabic of the Abbasid period. Nowadays Christian Arabic of Baghdad exhibits also an alveolar trill in very few lexemes, but primarily used in loanwords from Modern Standard Arabic. Native words with an alveolar trill are rare. Moreover Mosul-Arabic has commonly the voiced alveolar trill in numbers instead of a uvular fricative (e.g. arbaʕiːn "twenty"). Although this guttural rhotic is rare in Arabic, uvular and velar sounds are common in this language. The uvular/velar fricative [ʁ]~[ɣ] is a common standard pronunciation of the letter غ ġayn, and the uvular plosive [q] is a standard pronunciation of the letter qāf ق.
In Amharic the alveolar trill [r] is the usual pronunciation of /r/. But there are also assertions that around Adis Abeba some dialects exhibit a uvular r. Note that this information is not very supported in Semitic Studies. Also in Gafat (extinct since the 1950s) a uvular fricative or trill might have existed.
The majority of assyriologists deem an alveolar trill or tap the most likely pronunciation. However, there are several indications to a velar or uvular frivative [ɣ]~[ʁ] particularly supported by John Huehnergard. The main arguments constitute alternations with the voiceless uvular fricative /ḫ/ (e.g. ruššû/ḫuššû "red"; barmātu "multicolored" (fem. pl.), the spelling ba-aḫ-ma-a-tù is attested). Besides /r/ shows certain phonological parallelisms with /ḫ/ and other gutturals (especially the glottal stop [ʔ]).
Guttural R exist among several Malay dialects. While standard Malay commonly uses coronal r (ɹ,r,ɾ), the guttural fricative (ɣ~ʁ) are more prominently used in many dialects in Peninsular Malaysia and East Malaysia as well as some parts in Sumatra and East Kalimantan. These dialects include:
- Pahang Malay
- Kedah Malay
- Kelantan-Pattani Malay
- Negeri Sembilan Malay
- Sarawak Malay
- Terengganu Malay
- Perak Malay
- Tamiang Malay
- Pontianak Malay
- Palembang Malay
~ Perak Malay and Kedah Malay are the most notable.
These dialects mainly uses guttural fricative (ɣ~ʁ) on both /r/ and /gh/. Standard Malay includes both coronal r (ɹ,r,ɾ) and voiced guttural fricative /gh/ (ɣ~ʁ) as two different phonemes. To denotes the guttural r in the dialects, letter "r" often replaced by "gh" or "q" in informal writings . Standard Malay words with voiced velar fricative (ɣ), such as loghat (Dialect) and ghaib (invisible, mystical) are mostly Arabic loanwords spelled with letter غ.
Other Austronesian languages
Other Austronesian languages with similar features are:
- Minangkabau (closely related to Malay that it might be dialects of the same language)
- Southern Paiwan
Other language families
Standard Basque uses a trill for /r/ (written as r-, -rr-, -r), but most speakers of the Lapurdian and Low Navarrese dialects use a voiced uvular fricative as in French. In the Southern Basque Country, the uvular articulation is seen as a speech defect, but the prevalence is higher among bilinguals than among Spanish monolinguals. Recently, young speakers of Lapurdian and Low Navarrese are uvularizing the tap (-r-) as well, thus neutralizing both rhotics.
Whereas standard Khmer uses an alveolar trill for /r/, the colloquial Phnom Penh dialect uses a uvular pronunciation for the phoneme, which may be elided and leave behind a residual tonal or register contrast.
Sesotho originally used an alveolar trill /r/, which has shifted to uvular /ʀ/ in modern times.
Rhotic-agnostic guttural consonants written as rhotics
There are languages where certain indigenous guttural consonants came to be written with means used in other languages to represent rhotics, thereby giving the superficial appearance of a guttural R that may have never actually functioned as a true rhotic consonant.
The Inuit languages Greenlandic and Inuktitut either orthographize or transliterate their voiced uvular obstruent as ⟨r⟩. In Greenlandic, this phoneme is [ʁ], while in Inuktitut it is [ɢ]. This spelling was convenient because these languages do not have non-lateral liquid consonants, and guttural realizations of ⟨r⟩ have become common in various languages of European origin. But the Alaskan Inupiat language writes its [ʁ] phoneme instead as ⟨ġ⟩, reserving ⟨r⟩ for its retroflex [ʐ] phoneme, which Greenlandic and Inuktitut do not have.
- Grønnum, Nina (2005), Fonetik og fonologi, Almen og Dansk (3rd ed.), Copenhagen: Akademisk Forlag, ISBN 87-500-3865-6
- Trudgill, Peter (1974), "Linguistic change and diffusion: Description and explanation in sociolinguistic dialect", Language in Society, 3 (2): 215–246, doi:10.1017/S0047404500004358
- Map based on Trudgill (1974:220)
- Fougeron & Smith (1993:75)
- Grønnum (2005:157)
- Mateus, Maria Helena & d'Andrade, Ernesto (2000) The Phonology of Portuguese ISBN 0-19-823581-X (Excerpt from Google Books)
- Navarro-Tomás, T. (1948). El español en Puerto Rico. Contribución a la geografía lingüística latinoamericana. Río Piedras: Editorial de la Universidad de Puerto Rico, p.91-93.
- López-Morales, H. (1983). Estratificación social del español de San Juan de Puerto Rico. México: UNAM.
- López-Morales, H. 1992. El español del Caribe. Madrid: MAPFRE, p. 61.
- Jiménez-Sabater, M.1984. Más datos sobre el español de la República Dominicana. Santo Domingo: Universidad Autónoma de Santo Domingo, p.87.
- Grammar of Basque, page 30, José Ignacio Hualde, Jon Ortiz De Urbina, Walter de Gruyter, 2003
- Romano A. (2013). “A preliminary contribution to the study of phonetic variation of /r/ in Italian and Italo-romance”. In: L. Spreafico & A. Vietti (eds.), Rhotics. New data and perspectives Bolzano/Bozen: BU Press, 209-225
- Wells, J.C. 1982. Accents of English 2: The British Isles. Cambridge University Press. Page 368
- Survey of English Dialects, Heddon-on-the-Wall, Northumberland
- Survey of English Dialects, Ebchester, County Durham
- Millennium Memory Bank, Alnwick, Northumberland
- Millennium Memory Bank, Butterknowle, County Durham
- Chambers, J.K. and Trudgill, P. (1998): Dialectology. Cambridge University Press, p. 173f.
- "Rætt um talgalla" [On defective pronunciation]. Morgunblaðið (in Icelandic). Reykjavik. 6 September 1978. p. 4. Retrieved 4 October 2013.
- Khan, Geoffrey (1995), The Pronunciation of reš in the Tiberian Tradition of Biblical Hebrew, in: Hebrew Union College Annual, Vol.66, p.67-88.
- Otto Jastrow (2007), Iraq, in: The Encyclopedia of Arabic Language and Linguistics, Vol. 2, p.414-416
- Philippe Marçais (1956), Le Parler Arabe de Djidjelli (Nord Constantinois, Algérie), Paris, 16-17; cf. also Marcel Cohen (1912), Le Parler Arabe des Juifs d’Alger (= Collection linguistique 4), Paris, p.27
- Georges-Séraphin Colin (1987), Morocco (The Arabic Dialects), in: E. J. Brill’s First Encyclopaedia of Islam: 1913-1936, Vol. 6, Leiden, 599
- Farida Abu-Haidar (1991), Christian Arabic of Baghdad (= Semitica Viva 7), Wiesbaden, p.9-10.
- Otto Jastrow (1979), Zur arabischen Mundart von Mosul, in: Zeitschrift für arabische Linguistik, Vol. 2., p.38.
- Edward Ullendorf (1955), The Semitic Languages of Ethiopia, London, p.124-125.
- Edward Lipiński (1997), Semitic Languages: Outline of a Comparative Grammar (= Orientalia Lovaniensa Analecta 80), Leuven, p.132-133.
- John Huehnergard and Christopher Woods (2004), Akkadian and Eblaite, in: Roger D. Woodard Roger (ed.), The Cambridge Encyclopedia of the World’s Ancient Languages, Cambridge, p.230-231.
- Wolfram von Soden (1995), Grundriß der akkadischen Grammatik (= Analecta Orientalia 33), Rom, p.44 (§ 35); see also Benno Landsberger (1964), Einige unerkannt gebliebene oder verkannte Nomina des Akkadischen, in: Die Welt des Orients 3/1, p.54.
- John Huehnergard (2013), Akkadian e and Semitic Root Integrity, in: Babel und Bibel 7: Annual of Ancient Near Eastern, Old Testament and Semitic Studies (= Orientalia et Classica 47), p.457 (note 45); see also Edward L. Greenstein (1984), The Phonology of Akkadian Syllable Structure, in: Afroasiatic Linguistics 9/1, p.30.
- William Allen A. Smalley (1994). Linguistic Diversity and National Unity: Language Ecology in Thailand. University of Chicago. ISBN 0-226-76288-2.