Monday, 3 August 2015


I saw this advert last week:

It's an advert for Ribena. It's got some wordplay around the suffixes -y (which makes nouns into adjectives) and -est (superlative) and the combination of both of these. So in the slogan at the top, we have tastiest and fruitiest, both of which are common enough words, but also blackcurrantiest. That's the kind of word that people might ask 'is that a real word?' and of course it is, because it's made with perfectly legal word-creation processes, but it sticks out because it's new. Blackcurranty isn't normally an adjective, but of course we can make it easily (I just did) by sticking the -y suffix, which turns nouns (like blackcurrant) into adjectives. And then once you've got an adjective that ends in -y, you can make the comparative blackcurrantier and the superlative blackcurrantiest. With a word this long, we'd normally use more and most instead, though, so this word is eye-catching because it sounds unusual.

At the bottom, we have one the fact of it the exact same thing: it says 'You can't get any more Ribenary'. Again, just stick a -y onto a noun (Ribena) to get the adjective. Unfortunately here there is a spelling issue. Ribena, unusually for English words, ends in an 'a'. You can't stick -y onto the end of that, because Ribenay doesn't look at all like the word it's meant to be. So they've spelt it the same way we say it: with an 'r' that wasn't there to start with. This is called 'intrusive r' and all it does is make it easier for us to say two consecutive vowels.

Intrusive 'r' is closely related to 'linking r', which is the one that is there in the spelling but that people who speak dialects like mine wouldn't normally pronounce - like at the end of computer. When that word is followed by a vowel, like in the phrase my computer is switched on, the 'r' comes back again. For me, these two types of 'r' are so similar that I usually have to look them up to check I've got the names right.

Now, I believe that rhotic speakers don't have this intrusive 'r'. Rhotic speakers are the ones who would normally pronounce the 'r' at the end of computer. For those people, linking 'r' isn't really a thing because the 'r' is always there anyway, and the problems we might solve with intrusive 'r' are solved another way. So for those people (and that's most of Scotland, Ireland and North America, for instance), is this advert just weird? Does the word Ribenary work?


  1. Random rhotic Scot here, but I've wondered about these before. It's great to see this in print, and TM too.

    It's unclear to me how we'd (easily) prove whether this is r-sandhi before -y or an example of a post-non-high vowel #ry allomorph of #y, even that might seem irregular/uncommon in the face of so much post-lexical intrusive /r/.

    More to my line of thinking, following Bybee etc, I would just think it is at some stage (maybe now, that it it appearing in print/media) going to be either-or-both, speaker-dependent, including being ambiguously / non-categorically represented by individual speakers (exemplar style with lack of clear abstraction).

    I'm rhotic, as I say, and so it seems obvious that for me, I have an #ry allomorph (not based on personal r-sandhi of course, which I don't have) in my speech - in bananary, as a lexicalised suffixed form. Vanillary too, I think. I find the /r/-less forms possible too, but not quite so highly activated. Once people see these things in print more often (I love the use of both "ribena" and "ribenary" in the same picture), an #ry allomorph becomes more likely / acceptable / intuitive.


  2. Rhotic American (from a historically non-rhotic region, but 100% rhotic myself): The ad is just weird for me, yes. I don't have intrusive r, I don't have a -ry allomorph of -y, and "Ribenary" is impossible. If you add "-y" to "Ribena" you get something that I'd have to spell "Ribena-ey", but in practice I would prefer to avoid trying to say or spell it.