For the second week running, causation is in the news. Last week it was the shootings in Tucson, Arizona, raising the question of the role of violent right-wing rhetoric. This week it’s the floods, with this morning’s Age portraying Victoria’s premier and governor at odds over the causal effect of climate change.
In each case the lesson is the same: causation is an inherently fuzzy concept, and the naive expectation that there are hard and fast answers to be found, out there “in the world”, is simply not justified.
Immediately after the Arizona murders it was suggested that climate change provides an analogy with how to think about the causal process. One might say that violent rhetoric made certain events more likely without drawing a direct link, just as changing climate increases the probability of extreme weather events. Right on cue, Australia has provided some fine examples of the latter.
Graham Readfearn in yesterday’s Crikey explained the reasons why climate change needs to be given some place in the story of the Queensland floods. But when his sources say that “not enough studies had been done to have confidence about the role of climate change in single extreme weather events”, they are asking for the impossible.
Science could conceivably show the absence of a connection between the floods and climate change (for example, by showing the planet isn’t really warming at all, as a dwindling band of denialists still expect).
But no single event can ever be shown to be “caused” by a long-term trend: not because the evidence isn’t there, but because the question is unanswerable in principle. Causation just doesn’t work like that.
As Queensland, so Victoria. Premier Ted Baillieu, in saying “I don’t think we are in any position to make a comment” on the link between floods and climate change, may just — with good reason — have been shying away from a direct attribution of causation.
But he was made to look foolish with his comment that “where long-standing infrastructure has been in place, it’s interesting that engineers managed to set the levels of those bridges and those levees and those ramps at the appropriate height” — only for the levees at Kerang this morning to be “leaking very badly” as the Loddon River reached “a level beyond what the flood defences are designed to withstand.”
Baillieu, of course, has a strong denialist wing in his own party to deal with; they are already likely to be annoyed with his promise to meet Labor’s targets on greenhouse gas reductions and by his environment minister’s support for a price on carbon. Governor David de Kretser may not have appreciated just how delicate his new premier’s position is.
It’s unlikely that the governor was deliberately courting controversy with the view that increased flooding around the world is not entirely unrelated to a warmer climate. And unless we deny the existence of climate change entirely — a position that currently lacks any shred of intellectual credibility — it seems we have to admit that this is what the science tells us to expect with increasing frequency, and for our own good we really should start doing something about it.
The last sentence of my story last week on Arizona can be repeated with only minimal change: “Whether or not we blame climate change for the floods in eastern Australia, we can certainly blame it for making exactly that sort of thing more likely.”