Friday 9 May 2008

Hedges



This is a classic Guardian ad. In the first eight seconds we conclude from the evidence the man is running away. After the next eight seconds the conclusion is that he is stealing a bag. In the last part we see that he saves the man from some falling masonry.

Now let us imagine that the middle part were the only part we had seen. Scientifically, it would be imprecise to say 'The man is stealing the bag'. This is because there may be (and indeed is) an alternative explanation.

It so happens that you know there is an alternative explanation because you have seen the ad through to its end. However, in scientific life there isn't the luxury of having all the points of view - they are still to come. You must still acknowledge that. Therefore, any conclusions made must nod towards the contingency of knowledge, they must leave room for uncertainty, they must anticipate the reader's possible rejection; in short, they must always hedge.

To infuse your writing with tentativeness and possibility, you will need to salt and pepper your language with words from the hedging arsenal, things like 'may', 'might', 'could', 'suggest', 'indicate', 'seems possible', 'implies' 'if', 'possibly' and so on.

In the Guardian Points of View ad, instead of "The man is stealing the bag" good scientific writing would say "The man may be stealing the bag", "The man could be stealing the bag", "The evidence indicates the man is stealing the bag" etc.

Hedging makes scientific writing fuzzier, but it makes it a whole lot more precise. You are saying. 'Hey there reader, based on the info I have it is likely that this is what is going on but just want to let you know that there may be other causes so please don't kick my arse if you later find out I'm wrong.'

(Hedging also acts as rather nice padding for criticisms. It may feel weird taking down guys whose job it is to think about this sort of stuff. So, instead of coming out with a categorical 'You're wrong', the introduction of a little 'may' leaves room for future discussion and fewer enemies.)

No comments: