R is a common letter in many languages, but it represents vastly different sounds. Sometimes even within the same language. As Wikipedia says: Being “R-like” is an elusive and ambiguous concept phonetically. In fact, we might not even consider the R-sound used in, say, Spain, related to the R-sound used in France/Germany/Denmark, if they were not spelled the same. However, it was the will of history that they are, and in this post, I will present the three main families of R-sounds in European languages: English, Spanish, and French.
I am trying to keep it jargon free; hence the names above. They are a simplification, and in fact many languages use the three types. For example, the “Spanish” type is also Italian, Finnish, Icelandic, Lithuanian, Polish, Czech, etc. Some languages, such as Dutch and Portuguese also use more than one type, even within a single dialect. In this post I will cover the three “mainstream” R families plus a lot of other interesting R-facts, including that R is sometimes a vowel!
(R is incidentally also a very nice statistics program, but that is not the topic of this post!)
Since you are reading this in English, I think you are familiar with this type of R:
Consonants do not usually stand alone, so what you hear is the R mixed with vowels: [ra ara].
There is some variation in where the English type R is used. In non-rhotic dialects, Rs that are not immediately followed by a vowel disappear. A famous example is the sentence “Park the car in Harvard Yard”. In a non-rhotic dialect, in Boston for example, it would be pronounced “Pahk the car in Hahvahd Yahd”. Dialects with intrusive R on the other hand insert them where they didn’t appear originally: “I saw a film” becomes “I sawr a film” (oh boy).
The “Spanish” R covers two types, depending on whether you tap the tongue once behind your upper front teeth or roll it multiple times:
Some languages use only one of these two. In (standard) Spanish, they are both used: For single R (as in “pero”, meaning “but”) and double R (“perro”, “dog”), respectively. (However, I once met a Carmen who told me she liked her name pronounced with a roll consisting of three taps.)
This type of R is also where “R-ness” becomes strange: The single tap has a D-like quality to it. For example, the tt in “better” in English is also made by tapping the tongue on the gum ridge behind the front teeth!
French R, German R, Danish R. Also known as guttural (throaty) R. Like the Spanish R, it can also be done with or without a trill:
These sounds are produced while touching the back of the tongue to the uvula — the dangly thing at the back of the throat. (Danes are in fact lazy and only approach the back of the tongue to the uvula, but the sound is almost the same.)
In the southern parts of Norway and Sweden, close to Denmark, this type of R is also used.
I have not included Portuguese as a type of R, because this beautiful language in fact uses a large array of sounds to pronounce the letter R, depending on dialect and context. As Wikipedia says:
The actual uvular pronunciation [χ ʁ ʀ] is common in Portugal, although the older trill [r] is also heard. In Brazil, the total inventory of /ʁ/ allophones is rather long, or up to [r ç x ɣ χ ʁ ʀ ħ h ɦ]
The pronunciation of an R at the beginning of a word in Brazil can be especially counter-intuitive (at least if you’re not from Brazil!). As I learned from my Portuguese teacher (from Brasília), the word real (meaning both real and royal, and being the Brazilian currency), is pronounced heau(!) An initial R is a breathy H-sound, and a final L is a U-like sound.
R as Vowel
In the section on the English R, I mentioned non-rhotic dialects (typical of England) where Rs that are not immediately followed by a vowel disappear. But it might be more correct to say that they are vowels. Sometimes they prolong existing vowels, as in “park the car”, where the R prolongs the existing A-sound: “paak the caa”. Other times, you can hear the R as a separate A-like sound: beer [bia], your [yua/yoa], poor [pua/poa]. It also merges with E: reader [reeda], bother [botha]. Technically, the sound is a schwa, a “lazy”, neutral vowel made while relaxing the mouth. I guess whoever invented the word “gangsta” also spoke a non-rhotic dialect.
I think this phenomenon is especially interesting because everybody knows that R is a consonant, but in practice non-rhotic speakers pronounce something like half of all Rs as a vowel!
The phenomenon is also part of Danish. For example, in “stor” [stoa], meaning “big”, or “biler” [bila], meaning “cars”. The length of this vowel even distinguishes meanings: svar [svaa], svare [svaaa], svarer [svaaaa], svarere [svaaaaa]. These four words mean “(an) answer”, “(to) answer”, “(an) answerer”, and “(some) answerers”. Even natives have trouble distinguishing these…
I will round off with the constructed language Lojban because it has an interesting approach to the letter R.
You may have heard of the constructed language Esperanto. Unlike Lojban, the older, and relatively successful Esperanto is Eurocentric (and uses the Spanish type R). Lojban is made to make sense in the whole world. Its vocabulary is based on the most spoken languages worldwide: Mandarin, English, Hindi, Spanish, Russian, and Arabic. Its inventory of sounds is also sparse, which means that (small) variations in pronunciation of letters between cultures do not disturb the meaning. Especially for R, any of the rhotic (R-like) consonant sounds are allowed, so you can just choose the one that is already in your native language, or the one that is the easiest to learn!
Its grammar is also fascinating. For example, it can use any subject-verb-object order: SVO, SOV, OVS, OSV, VSO, VOS. It also has a complete BNF! (For those who know what that is.) But I will save that for some other time!