Commentators on recent articles on climate change in this journal have argued that the scientific study of climate change is useless and/or untrustworthy. Useless because all we should and do care about is the weather, and it is not possible to attribute any particular weather event, no matter how unusual, to climate change. Untrustworthy because the basis for identifying what counts as the accepted science is expert peer review and this process is corrupt or unreliable.
So far as I can recall these are recent claims. The relationship between weather and climate used to be thought straightforward, and peer reviewed publication followed by peer reviewed criticism was accepted as the basis of progress in science. But now both have become hot issues.
Happily light can be shed on these hot issues by considering the comparable relationships and processes in other less politically charged fields of human endeavour. Indeed, cricket can teach us much of what we need to know about these matters.
Consider first the relationship between the weather and the climate. It is just like the relationship between a batswoman’s performance in any particular innings, and her form. (See below for more on why I discuss batswomen rather than batsmen.)
We don’t get in a muddle about the relationship between performance and form in cricket, even though the relationship between the two is not easy to define.
For example, no one would think of saying that a batswoman was in continuing good form if she got out for a duck in each of her last ten innings, nor in continuing poor form if she had recently been the top scorer for her team. But neither would we say that she had fallen into poor form when bowled first ball in one innings, if previous to this she scored highly in a run of innings.
We are equally relaxed in saying that her getting out for a duck is explained by her being in poor form, if this has been typical of her recent performances, or that her scoring a double century in an innings is explained by her being in good form, rather than mere good luck, if she is happily enjoying a run of good innings.
Clearly the relationship between performance and form is a complex one but it is not one which causes us any particular difficulty.
When we talk about a batswoman’s form, we have two distinct but related concepts in mind. First, her longish term average performance, such as her average score in the last ten innings. Second, the collection of skills which we take her to exercise when batting, which explains her (current) usual performance at the crease.
In the first sense, the relationship between performance and form is simply arithmetic and definitional. Recent average performance is form. If performance changes, by definition so does form. A one-off higher or lower-than-usual score does not shift the average much, but a few such aberrations shifts the average significantly, and so we revise our estimate of her form.
In the second sense, the relationship is causal rather than definitional: form causes (and thus explains) performance. But the causal relationship is, shall we say, probabilistic rather than necessary or certain. In her best form she may unluckily be dismissed early in her innings by an unplayable ball. But if she gets out to a simply played delivery, and especially if she has been making a habit of this, we explain this by her being in poor form.
Just so with the climate and the weather.
By definition, the climate is the long term average of the weather. But we also think of the climate as the collection of natural forces (sea surface temperatures, wind circulation patterns, humidity etc) which causes (and thus explains) the weather we have.
If over a period of some years the weather changes from what it has previously been, becoming hotter, say, or wetter, or even more varied, by definition the climate has changed. We should have no more trouble in saying this than we do in saying that if our batswoman’s performance changes, then so does her form.
Similarly if the weather changes in some particular way that shows a definite trend, we should not have any trouble in saying this is caused by a change in the climate.
Of course, just as our batswoman can get out early to an unplayable ball without being judged to be in poor form, so too can an unusual weather event be put down to chance rather than a change in the climate. But when the average starts to shift, we will happily use that changed average in cricket to explain changed future performance, and we should have no more difficulty in explaining a trend in the weather by reference to a change in the climate.
In short, what we learn from cricket here is that we have a pretty good understanding of the relationship between observed performance and theorised underling causal abilities, even when the relationship between the two allows a probability of unexpected events, and we can apply this understanding to the relationship between the weather and the climate. Thus we should have no more reason to reject a change in the climate as an explanation for changed weather than we have to reject a change in form as an explanation for changed batting performance.
Cricket is no less helpful in thinking about the role of peer review in science. Peer review plays the same role in science as the selectors play in cricket.
The role of the selectors is to consider candidates for the team, and reject those whom they think will be most easily beaten by the opposition teams to be played. Selection certainly does not guarantee that the chosen players will be successful, but no one expects this. If someone were to say that probably every team member chosen by the selectors will be beaten by an opposition team at some stage, we would all respond with a ‘so what’ rather than shock. What would shock us would be evidence that the selectors used some criterion other than their assessment of the players’ likely success in choosing the team.
We would defend the role of the selectors not on the basis that their choosing a player was sufficient to ensure that they would be successful, but rather that choosing some players rather than others is necessary to having a team which might win and indeed to having a team at all. For if there were no selectors, anyone could roll up to the ground and claim to be playing for Australia. Perhaps the women’s team, tired of being sent to the Bankstown oval, might run on the SCG to face the glory of English manhood in the battle for the ashes – or if the English had given up on selectors too, perhaps our women’s team would face the Barmy Army. Indeed, why not go further and dispense with the teams entirely and just have the crowd bring their own bats and balls and play as they wish?
Clearly doing away with the selectors could lead to a lot of fun, but it will not allow us to decide who gets to keep the ashes.
We have no difficulty in appreciating the role of the selectors as necessary but not sufficient for choosing a successful national cricket team – male or female – even though we might disagree with their decisions on occasions. Certainly we do not take as a reason to give up on the process the claim that on average less than half the teams chosen by selectors will win the game for which they have been chosen – even though this is true. What would persuade us to change the selection system would be evidence that, by its very nature, it was not directed at choosing the best team.
We should have no more difficulty in appreciating the role of peer review. It is certainly not sufficient to ensure the publication of only those papers which will survive criticism for a long time (and thus be practically accepted as true), but it is necessary to ensuring that the contest between claims is restricted to those which at least have a chance of success. What would persuade us to change the system of peer review would be evidence that, by its very nature, it was not directed to choosing the papers which are most likely to withstand criticism. But there is no evidence of this in the complaints about the system of scientific selection recently aired here.
NOTE: Originally published in On Line Opinion, 7 May, 2011