Rhotacism or rhotacization is a sound change that converts one consonant to a rhotic consonant in a certain environment. The most common may be of to. When a dialect or member of a language family resists the change and keeps a sound, this is sometimes known as zetacism. The term comes from the Greek letterrho, denoting.
Albanian
The southern, Tosk dialects, the base of Standard Albanian, changed to, but the northern, Gheg dialects did not:
In Aramaic, Proto-Semiticn changed to r in a few words:
bar "son" as compared to Hebrew ben
trên and tartên "two" as compared to Demotic Arabic tnēn and tintēn, from Proto-Semitic *ṯnaimi and *ṯnataimi. Compare also Aramic tinyânâ "the second one", without the shift.
l changed to the tapped r between vowels in Basque. It can be observed in words borrowed from Latin; for example, Latin caelum became zeru in Basque. The original l is preserved in the Souletin dialect: caelum > celu > zelü.
Western dialects of Finnish are characterised by the pronunciation or of the consonant written d in Standard Finnish kahden kesken- kahren kesken. The reconstructed older pronunciation is.
Goidelic languages
In Manx, Scottish Gaelic and some dialects of Irish, a cluster developed into, often with nasalization of the following vowel, as in Scottish Gaelic cnoc .
All surviving Germanic languages, which are members of the North and West Germanic families, changed to, implying a more approximant-like rhotic consonant in Proto-Germanic. Some languages later changed all forms to r, but Gothic, an extinct East Germanic language, did not undergo rhotacism.
Note that the Modern German forms have levelled the rhotic consonant to forms that did not originally have it. However, the original sound can still be seen in some nouns such as ', "being" and ', "loss". Because of the presence of words that did not undergo rhotacisation from the same root as those that did, the result of the process remains visible in a few modern English word pairs:
is and are
the comparative and superlative suffixes -er and -est and derived words such as more and most, better and best, etc
rise and rear
loss and forlorn
English
Intervocalic and are commonly lenited to in most accents of North American and Australian English and some accents of Irish English and English English, a process known as tapping or less accurately as flapping: got a lot of becomes. Contrast is usually maintained with, and the sound is rarely perceived as /r/.
Reflecting a highly-regular change in pre-Classical Latin, intervocalic s in Old Latin, which is assumed to have been pronounced, invariably became r. Intervocalic s in Classical Latin suggests either borrowing or reduction of an earlier ss after a long vowel or a diphthong. The s was preserved initially and finally and in consonant clusters. Old Latin became in Late Latin by analogy with the rhotacised forms in other cases such as genitive, dative and accusative honoris, honori, honorem. The d and the l changed to r before another d or l and so the same consonant did not appear twice in a row :
In Neapolitan, rhotacism affects words that etymologically contained intervocalic or initial.
LAT. DENTE > Neap. dente "tooth"
LAT. PEDE > Neap. pere "foot"
Portuguese and Galician
In Galician-Portuguese, rhotacism occurred from to, mainly in consonant clusters ending in such as in the wordsobrigado, "thank you" ; praia, "beach"; prato, "plate" or "dish"; branco, "white"; prazer/pracer, "pleasure"; praça/praza, "square". Compare Spanish obligado, playa, plato, blanco, placer, plaza from Latin obligatus, plagia, platus, blancus, placere, platea. In contemporary Brazilian Portuguese, rhotacism of in the syllable coda is characteristic of the Caipira dialect. Further rhotacism in the nationwide vernacular includes planta, "plant", as, lava, "lava", as , lagarto, "lizard", as and advogado, "lawyer", as. The nonstandard patterns are largely marginalised, and rhotacism is regarded as a sign of speech-language pathology or illiteracy.
Rhotacism, in Romanesco, shifts l to r before a consonant, like certain Andalusian dialects of Spanish. Thus, Latin altus is alto in Italian but becomes arto in Romanesco. Rhotacism used to happen when l was preceded by a consonant, as in the wordingrese, but modern speech has lost that characteristic. Another change related to r was the shortening of the geminatedrr, which is not rhotacism. Italian errore, guerra and marrone "error", "war", "brown" become erore, guera and marone.
In Romanian, rhotacism shifted intervocalic l to r and n to r. Thus, Latin caelum became Romanian cer, Latin fenestra Romanian fereastră and Latin felicitas Romanian fericire. Some northern Romanian dialects and Istro-Romanian also changed all intervocalic to in words of Latin origin. For example, Latin bonus became Istro-Romanian bur: compare to standard Daco-Romanian bun.
Rhotacism is particularly widespread in the island of Sicily, but it is almost completely absent in the Sicilian varieties of the mainland. It affects intervocalic and initial /d/: cura from Latin caudam, pit is eri from Latin pedem, reci from Latin decem'.
Spanish
In Andalusian Spanish, particularly in Seville, at the end of a syllable before another consonant, l is replaced with r: Huerva for Huelva. The reverse occurs in Caribbean Spanish: Puelto Rico for Puerto Rico.
The beginning of the change is attested in the Freising manuscripts from the 10th century AD, which show both the archaism and the innovation. The shift is also found in individual lexical items in Bulgarian dialects, дорде 'until' and Macedonian, сеѓере. However, the results of the sound change have largely been reversed by lexical replacement in dialects in Serbia and Bosnia from the 14th century. Dialects in Croatia and Slovenia have preserved more of the lexical items with the change and have even extended grammatical markers in -r from many sources that formally merged with the rhotic forms that arose from the sound change: Slovene dialect nocor 'tonight' on the model of večer 'evening'. The reversal of the change is evident in dialects in Serbia in which the -r- formant is systematically removed: Serbian veče 'evening'.
