Not betting on ‘ameliorative analysis’

NOTE HXA7241 2021-04-11T07:28Z

A look at the limits of doubt leads to a diagnosis of ‘ameliorative analysis’.

When posting a comment at Daily Nous, it seems that one cannot simply, bluntly, say TW are not W ‒ it perhaps falls short of civility/whatever. So you have to say something more like: ‘there are strong arguments to suppose TW are not W’. But this seems strange … where is the doubt in the assertion? Think of it like this: is it something we could bet on? Could we place a bet, like at ‘The Long Now Foundation’ for 10 years in the future, to wait and see which answer the experts have decided is correct?

Imagine one day discovering a strange new flying creature ‒ is it bird or mammal or whatever? It has an X, but also a Y, or we cannot quite make out the shape and what is what … Well, we can expect that we just look more closely, with scientific know-how, and most likely resolve it comfortably into known categories. The doubt here flows from its contingency ‒ it is new, and could happen to be anything, then we examine it and settle things. It seems we could make a bet on this.

But what if we already know everything about this creature, in terms of all our categories? Then where is the doubt? There seems none left. So why are we unsure? Can I not just say, pointing: that is one of them, and it must be clear. How can there be a dispute? You might say ‘look, it has these other features’, but so what? I am talking about these features ‒ your claim does not deny them ‒ it has both. We have to at least be able to recognise when we are pointing at the same thing.

This seems like betting on a race after it has been run. How can that make sense? Maybe you might suggest that the uncertainty is that we do not how the judges are going to decide. … So we bet because we do not know what counts as winning? But then how could we bet at all? Knowing what counts as winning is part of what a bet is.


This all seems closely related to ‘ameliorative analysis’/‘conceptual engineering’ confusion/word-play.

In truth, the question is foreclosed at the start. The answer is already right there in the question. Yet somehow we are confused. It is like we know everything in a scene, but then are distracted, and when we look back, we have forgotten our place, or what was what.

How can this happen? Is it a similar mechanism to a paradox? A paradox cannot physically exist ‒ it is purely an artefact of misunderstanding. We find that our synchronisation with the world has momentarily drifted, diverged. But, importantly, not entirely: some part has, and some part is still closely attached. We cannot have lost grip entirely, since then we would not be talking about the same thing, and not know we were not, and not realise anything was wrong. There must be some rightness in our misunderstanding.

But here it is a levels issue: we talk about a thing, but at times we jump up a level and are talking about, talking about, a thing. With that extra mechanism there is always something to keep hold of. Even when we know nothing about something, we still handle it as ‘that particular unknown thing’. We must always know that about it, in a sort-of adjacent way. We always have the box for the thing, even knowing nothing about the thing itself.

Could you bet on darts being a sport or a game? Yes: you could bet on darts being classified by the Olympics as a sport. Note that this is the opposite way around to the strange flying creature. There we did not know the thing, but we knew the classifications; here we know the thing, but do not know the classification. The classification is maybe not really an expected or ordinary kind of unknown, but it has an indeterminacy, it is really ‘something we will be doing’ with darts, and we have not yet decided, and there is uncertainty in, what we will do.

So you can have darts, and several other things, and then make a new box to put them together in, and call the box ‘quasi-sports’. But you could not have some things, including darts, put them in box, and call the box ‘darts’ ‒ you used one label for two things, an inconsistency. ‘Darts’ cannot be both the game, and the group it is in.

And this looks like the diagnosis of the malady of ‘ameliorative analysis’. It says, eg: this is a W (the thing), these are things Ws do (the box), and now we will call the things that behave like that ‘W’s. It is the same faultiness as the last darts example: reusing a label and confusing matters with that inconsistency. You cannot take the thing, and the group built from it, and name them both the same ‒ it is just a trivial error.