“So, you believe in God?”

Just because I doubted an evolutionary explanation….

“So,” [pause]

“You believe in God?”

I was making my usual Platonic point about the poverty of the senses – about how abstract notions such as ‘oneness’, ‘similarity’, ‘dissimilarity’ and so on, couldn’t be derived from sense experience.  I might even have used the phrase: ‘the poverty of the manifest image’. I like that phrase.

The conversation was going well. We touched on evolutionary biology and the evolutionary advantages of grouping concrete objects into abstract ‘kinds’ of objects.

“So be it”, says I,  “but that doesn’t help to explain how we were able to invent ‘kinds’ in the first place. You can’t point to a kind of thing. Evolutionary biology can explain how we point to things themselves, so long as pointing is evolutionarily advantageous.”.

I begin to warm to my subject: “we don’t see a species, class or kind”.  What I’m trying to get at is that our ability to classify things is a precondition of our having any experience at all, not something we learn from repeat experiences. No doubt classifying is an evolutionarily useful ability, but advantageousness doesn’t explain how we can do it at all.

And then the god-bomb went off. I might have seen it coming, or heard it ticking. I should have. I mean, you can’t go around pointing out how mysterious abstract thought is without inviting a deus ex machina.

Nobody likes inexplicables, and I did make abstract thought sound like a divine gift.

But there is – isn’t there? – something epistemologically puzzling, mysterious even, about how we come to classify particular things into abstract kinds.

Would it help solve the mystery if we introduced a bit of guesswork into the way we develop into our ability to classify things into kinds? Maybe we can make headway if we say that we mostly pretend that things are similar, and leave it to scientists to find out for sure.

Perhaps we know, roughly, that some things look a bit like other things, and at that point we take a conceptual risk and say they are ‘the same kind of thing’.  That is, broadly speaking, what science does.

Naturally this process involves some gerrymandering. I mean it’s not as if we ever see two things that are exactly alike. So of course, we imaginatively chop reality into bits that seem more manageable for our purposes, and quietly ignore the bits that don’t fit.

But you can see, now, why I kept quiet about this argumentative move. I was sitting around a table with some pretty predatory conversationalists.

One whiff that science could explain abstract thought, and I would have been mincemeat.

Or would I? How might this scientific gerrymandering work? You’d still already have to have some idea of what ‘similarity’ is before the very first time you tried out your gerrymandering on the world.

This proto-scientist still begs the question: where did we get ‘similarity’ and ‘dissimilarity’ from?  Normally, we learn about our environment from repeat exposure. But testing whether things are as similar as they appear must beg the question that we know what ‘similar’ is.

So ‘similarity’ must be something we bring to the world when we look at it. Without a pre-existing sense of what ‘similarity’ or ‘difference’ is, our image of the world would be a horrible undifferentiated mess. We need to start with some intellectual resources, to have some idea of what to look for, in order even to begin to make sense of it all.

I can see how all that sounds dangerously unnatural.

It sounds as though I deny that the evolutionary history of ‘perception’ can explain how we perceive in the extraordinarily differentiated way that we do. Well yes, in a sense I do deny that – evolutionary explanations are nice, but it’s good to notice when they don’t satisfy us.

The evolution of perception would be something like a list of all the trials and errors we’ve had as a species (or as an individual) in the ways we’ve invented of classifying the world.  I claim that before that list can even get started, we need to explain how we got the tools to classify anything in the first place.

I’m certainly not saying that we are doing anything cosmic or divine when we apply rudimentary ideas of ‘similarity’ in the act of perceiving something.

All I want to say is that having some sort of an understanding of similarity is logically prior to perceiving it.

  1. The astute may already have noticed that this line of thought is essentially Neo-Kantian (albeit with suitable disclaimers about the idiosyncrasies of the above).  I am content with that label – Who could disagree with the Kantian dictum?:

Thoughts without content are empty, intuition without understanding is blind

2.  Conversations are imprecise.  I did say: “we don’t see species, classes or kinds”, but perhaps I shouldn’t have.  I am, of course, perfectly happy to say that we can see similarities between objects, and if we can do that, then perhaps we can see that an object might belong to a particular class of objects.  (But that still sounds slightly different from saying that we see the ‘class’).

I am only claiming that seeing a similarity is not an ability that is derivable from sense-experience.  I’m not saying that it doesn’t occur in sense experience.  That really would be odd.

And the last thing I want is to cross swords with J.L. Austin.  Poor Mr Maclagan is immortalised for having implied that we cannot see a resemblance, but that we ‘apprehend’ or ‘intuit’ it with some other mental faculty.  Austin replied:

I have heard it said that it is odd to talk of ‘smelling a resemblance’: and certainly it is as well to consider other senses than that of sight.  But it is not odd to talk of smelling two similar smells, or two smells which are sensibly alike (though the plain man might well ask, how could they be alike except sensibly?).  And if I were forced to say either ‘I smell the resemblance’ or ‘I intuite it’, I know which I should choose.  The plain dog would, I am sure, say it smelled resemblances: but no doubt your philosophical dog would persuade itself that it ‘inhaled’ them.

I am, most definitely, the plain dog.