There's No Such Thing as a Rodent

Fun fact: There's no such thing as a fish.

This 'QI'-style factum results from the notion that fish do not all belong to the same genus (or 'family', or 'phylum', or some other Biological category). If we take one genetic branch of fish as the real fish, then we miss out a lot of animals which everyone calls 'fish'. But if we broaden our definition and include the other fish, we must include foxes in the same gene-line.

Simply put, fish split early in the genetic line, and now we can't name them according to their genetic classification.

This little factum has taken up my head-space for some decades now, because I can't square it with how we talk about animals in general.

In some parts of Tibet, they sing a ritualized riddle. The question is "What kind of bird gives birth?". The other side sings the reply, "a bat".

We have an easy solution to this odd riddle. Clearly these Tibetans have a word which means something like 'flying thing', which mostly translates fine into 'bird', but not always. We categorize animals more along family lines than form or function, so we should place bats with other mammals.

Solved.

Except of course, for fish, which we do not categorize according to family trees. So now it sounds like anyone could class a bat as a bird. It's a flying thing, which looks kinda bird-like. So, bird.

I want a proper answer to this.

Language isn't logical.

That all fine and well, but this actually speaks in favour of the notion that we could just call a bat a birds.

'I saw a bird', anyone could say. 'Might have been a fruit bat', and you couldn't complain, because 'language is illogical'.

I need a better explanation.

Bats don't look like birds.

I don't know how to test this statement. They look enough like birds that some people class them with birds. But even if we allow the specific example of bats, we have more problems looming.

Dolphins look like fish. Zebra look like horses. But English (and typically European) languages track animal words by the genetic tree quite fiercely, so we cannot simply say 'sure, in Africa, some horses have black and white stripes'.

How could we accept some genetic definitions but not others?

English classifies animals in different ways. Just as we say 'a pride of lions' and not 'a flock of lions', we classify lions as cats due to genetics, while classifying fish by their form. Language has rules, but those rules don't always have great reasons.

But this argument fails too. When someone says they saw a 'flock of lions', they have made a grammatical error, but have not necessarily misunderstood anything about the lions. But someone calling a dolphin a 'fish' has clearly misunderstood something about the individual word 'dolphin' (and probably doesn't understand dolphins).

This explanation fails to show which type of error has been made.

English definitions typically track genetic classifications, but not always.

This answer seems closer than any other, but still doesn't sit right with me. It invites tricks of language. Which species demand technical terms?

Imagine someone catches rodents for a living. We ask if they ever deal with bats.

'Actually', they say, 'bats are not rodents, but chiroptera'.

I want to accept this, but someone could push back.

'Actually', they might argue, 'the word "rodent" includes bats, when you're not doing Biology, as this word simply tracks what people refer to as rodents'.

This seems like weasel-words. We don't need the linguistic baggage of words taking double-definitions when the classifications Biologists use work just fine for natural language.

And why can't we say 'there's no such thing as a rodent'? The same argument applies to fish - specifically that a big patch of what we consider 'rodents' don't have the right genetic relation to each other. We accept the removal of bats from our word 'rodent', but cannot stand to part with tuna.

PS:

My girlfriend just asked me to name my favourite fruit. I said 'tomato', but Serbs don't accept this. They say 'voće' (for 'culinary fruit') and 'plod' (for 'botanical fruit').

Maybe this is the way out of our categorization problems.