The universal taxi

I was listening to the radio this morning and they were talking about linguistics. I feel very conflicted about this because I love hearing real, proper linguistics on the radio! It's so rare! but the linguist in question has expressed anti-trans attitudes in the recent past and so I can't call myself a fan. But there we go; they at least weren't discussing such issues so they didn't express them in the course of this conversation.

They were talking about demonstratives, this and that. It was a nice discussion, lots of information, and fun facts about proximal and distal demonstratives (this here vs that there) and how these change in context, and also how many languages have a three-way system with a medium-distance and a far-distance one.

And finally they talked about how pronouns are universal: all languages have them. This is indeed super cool, and basically indicates that they are a Very Important and Useful language feature. Any language that didn't have them for some reason would likely innovate them pretty sharpish, and home signs include them, for instance.

Then someone wrote in to say that 'Interestingly, one word that is universal is 'taxi''. This is interesting, but it's not the same thing. Taxi is 'universal' because it's been borrowed into lots of languages. I'd actually be surprised if it's truly universal in the sense that every single language has this word. There are presumably a few where the concept hasn't been needed and lots more where in fact there's a different word meaning 'private car and driver hired for single journeys', or whatever it is that taxi means. A quick google looks like in quite a lot of languages familiar to English speakers, the word for 'taxi' is something like taxi, which does give an impression of universality from this vantage point. And the 'universal' part is the specific form of the word ('taxi'), with a certain meaning.

Pronouns are universal in a different way. The words are not the same in every language (me, your, we, etc). They haven't been borrowed. There is a lot of historical relatedness, to be sure; it's no coincidence that we looks like German wir. But the very fact that there are pronouns is what's universal. This is fundamentally more interesting than that a particular word has been borrowed a lot. (By the way, here is a study that says that the most universal word is 'huh'.)

Not all moths

I host a quiz night in our local micropub (it's cosy). Recently, in the 'insects' round, I asked whether insects are warm- or cold-blooded. They are cold-blooded, of course (this is not a technical term: they are actually ectothermic, which means they don't regulate their body temperature internally). One quizzer challenged me, because there are in fact three species of moth (out of maybe 10 million species of insect - we don't know exactly) that are warm-blooded.

Leaving aside the fact that rounding this off to, say, 3 significant figures is 0.00%, we have a question of genericity. Now, either my statement 'insects are cold-blooded' is an absolute statement meaning 'all insects are cold-blooded', in which case a single warm-blooded insect is enough to prove it wrong, or it's a generic statement: 'insects in general are cold-blooded'. I meant the latter, of course, and in the context of a quiz question where two options are given, this should be clear. One of the things Steven Pinker said, actually, was that to avoid the hedging ('almost', 'in general', etc) you find in bad prose, you should allow your reader to assume the generic interpretation. In academic writing there is a place for the precision afforded by hedging, but for much other writing I agree.

There's limits though. While researching my quiz, I read the supposed fact that 'babies are born with blue eyes'. That, I thought, was astonishing. It turned out that what the author meant was 'white babies who will have blue, green, hazel or grey eyes', not 'babies in general' - there is a very high proportion of babies in the world who have dark brown eyes, and are not born with blue eyes. If you're going to make generic statements you do have to be clear about what the universe of discourse is and your generic statement has to actually apply to the majority of things in it. (In this case, the author had made the very easy mistake of forgetting that not everyone is exactly like them.)

Another of my quizzers challenged another question. In the picture round I had asked for the name of the species of fish pictured. One was a goldfish, and the team had written 'carp'. I didn't allow this, and the challenger wanted to know why, when a goldfish is a carp. It's true that goldfish are carp, but not all carp are goldfish. 'Goldfish' is therefore a hyponym of 'carp' (and 'carp' is a hypernym of 'goldfish').

I am clearly not strict enough with my quizzers. If I keep blogging about their complaints perhaps they'll stop.