Sunday, July 31, 2011

Science Doesn't Say

How many times have you heard someone say that "science tells us" - or that it shows, reveals, says, proves, or makes clear?

It's very common. But it's misleading.

Scientists never talk like this while they're doing science, which suggests that there's something wrong with it. Rather, we say: "Our experiment was inspired by the fact that X, which was shown last year by Y et al".

Y et al aren't just some bunch of famous smart guys who came up with an idea and told everyone, and everyone believed them, because scientists respect authority - which is what "Science Says" means.

No, Y et al is a paper, or other report, and when we say that it shows something, we mean it quite literally. Scientific data is like a photograph or, more accurately perhaps, a window, through which we can just see X.

'Science' is nothing special. It's just looking at stuff.

Indeed, there are scientific papers where the key result is literally a photo, usually taken down a microscope or through a telescope, but still. This paper is a great example. The key result was that the little yellow thing in the third image grew some extra sprouts from day 0 to day 1. It takes some knowledge of the context to understand why that's so interesting, but the actual result is right there.

However, even where the result isn't literally a picture, it is still a window.

This line shows the chemical composition of a particular part of someone's brain. Each of the peaks on the curve corresponds to a particular chemical, and the height of the peak tells us how much of that chemical there is.

There's nothing mysterious about why particular chemicals cause particular peaks. It's well understood. (Conceptually, it's like each molecule is a bell, of a particular size and shape, and they make different sounds when you shake them around. The line is what you get when you shake the piece of brain up, and record how much of each sound you hear back.)

Getting this data is a high tech process requiring special equipment, but all that's just background detail when you actually come to do it. Just as a photographer doesn't need to worry about the mechanics of their camera, and you don't need to worry about how your eye gathers and focusses light as you're reading this.

There is an element of authority and trust in science, but not in any special sense. To take published evidence at face value, you do need to trust that the authors haven't manipulated it, and to trust that they gathered it in the way they described.

But the same goes for any other kind of evidence. Any photograph could be Photoshopped, or the caption could be misleading. Anything you read could be made up. In everyday life, we don't worry about this unless there's a particular reason to.

A scientific journal is just a newspaper with access to better equipment.

There's a view in which "Science" is a kind of oracle that hands down judgements from on high, with scientists as priests who record and proclaim the revelations. This leads to no end of problems.

It easily leads to the view that science is somehow especially hard to understand, or even that we can't understand it, so there's no point in trying. It can lead to the idea that science can't be very interesting compared to the real world. It leads to questions what good it can do, or whether science can ever answer 'the big issues'.

When you realize that science is just looking at stuff, you see that those concerns, far from being valid, don't even make sense.

No comments:

Post a Comment