One of the supposed traits of older types of British Received Pronunciation is that /r/ can be a tapped sound (for those reading this week, this sound is similar to the ‘tt’ in American ‘butter’). In ‘traditional’ RP, this typically occurs in between vowels, as in words like ‘very’ and ‘terrible,’ resulting in the (wrong) impression that these words are pronounced ‘veddy’ and ‘teddible.’
It’s clear that this was, in fact, something that at least some RP speakers did who were born before World War I or so (for reasons I’ll mention below, it would be near-impossible to get any more specific). This is obvious in the speech of, say, Noël Coward (born in 1899):
But a number of factors make it difficult to assess the scope of r-tapping in older types of RP. First, it strikes me that among British actors of a certain generation, this could often be a very deliberate theatrical choice. Rex Harrison taps his r’s quite frequently in the film adaptation of My Fair Lady, yet hearing interviews with him from around that time, this feature was not present (or rarely present) in his own accent.
This real life/stage life dichotomy is a problem that goes beyond one feature. For example, in older RP, the ‘o’ in ‘code’ is pronounced much as it is in General American English, with a fairly back-starting diphthong (transcribed as [oʊ]). You can hear this sound in this old recording of John Gielgud reciting a monologue from Othello. Yet in this interview from the 1960’s, it’s clear that in his personal speech, Gielgud used a central-starting ‘o’ sound typical of more contemporary RP.
The other complicating factor with people like Harrison and Gielgud, though, is that I would bet many one-time RP r-tappers dropped the feature as the decades when on. We know that Queen Elizabeth II ‘contemporized’ her accent in the late 20th-Century (thanks to this study). It would be unsurprising if r-tapping weakened in the speech of individuals as it became less common.
So what’s the bottom line with RP r-tapping? Unforunately, it started declining in frequency just at the dawn of recorded speech. So whether it was ever truly widespread as a naturally occurring feature is a bit hard to tell. Thoughts?