Après nous, le déluge: legislating science

The North Carolina senate tried to pass a bill in June banning state agency researchers from using exponential extrapolations in predictions of sea level, requiring them to just using linear extrapolations. After being generally laughed at, the legislators settled for a compromise: state agencies were forbidden to base any laws or plans on exponential extrapolations for the next three to four years. Now a new report shows that sea levels are rising faster near North Carolina than anywhere else on Earth.

The reason for the bills seems to be partially motivated cognition: the supporters of the bills appear to be coastal municipalities worried that accepting predictions of a serious sea level rise would restrict economic development, send insurance rates skyrocketing and decrease coastal property values. They have made their own predictions that show a far smaller increase, in part based on a paper arguing for slowing sea level rise.

Legislating about science has never been very successful, whether it is defining pi to be exactly three or that pregnancy begins two weeks before conception. Legislating facts has the problem that reality might deviate from the law. Legislating definitions (as in the pregnancy case) leads to semantic confusion. Legislation about how government agencies go about their work, what is funded or how the results are presented is more subtle: there are areas where this makes sense (just as governments decide policing and waste removal policies) and there are areas where this is clearly inappropriate biasing (like suppressing research that produces inconvenient findings). Typically, the more science and its conclusions are micromanaged, the more stupid the results tend to be. The legislation moves from a context of enabling research to biasing the results.

In the end it comes down to the epistemic role of science and science policy, and the ethical importance of having truth-seeking societies.

As has been often argued on this blog, science aims at honest inquiry towards truth. The main reason governments sponsor or perform science is to acquire better information in order to make rational decisions. Science policy that distorts this undermines not just the science, but the ability of the government to make good decisions. This is not just about getting distorted facts, but by reducing the chance that science will act as an outside error signal pointing out when policies are irrational.

In the case of the sea level legislators have no reason to think they know how to extrapolate data well. Even if they think there is a serious disagreement among scientists about the right method and its implications, they are still irrational if they decide on a particular outcome and make that the official input to their future decision-making: a disagreement means that one should be extra careful with jumping to conclusions. Deliberately ignoring certain scenarios means that state decisions will likely now not just diverge from physical reality, but also from the pricing insurance companies and investors use. It is likely that this divergence in policy from reality will at the very least be exploitable by commercial interests that can trade on it. A fool – even a state – and his money are soon parted.

Scientists are not unbiased or free from fashions. I am confident that in a few decades time we will look back at plenty of current climate research and policy and laugh at how biased it was. But the scientific system aims at creating feedback from experiments and peer review that detects errors and in the long run corrects them. Political systems need that feedback too, or they will get hi-jacked by special interests that exploit them. Waiting for state economic bankruptcy or an eventual disaster would likely be a too slow and painful method of correction: since states are tools for making citizen’s lives better and/or protect their rights, it is in everybody’s interest to make them as truth-seeking as possible. And that means keeping science and planning policy open for free inquiry and criticism, producing the best possible inputs to decision-making.

In the end, coastal municipalities that base their plans on a best possible scenario will likely suffer the same fate as the summer resort in Jaws, where the authorities choose to ignore warnings. Reality tends to bite hard if you deliberately ignore it.

  • Facebook
  • Twitter
  • Reddit

4 Responses to Après nous, le déluge: legislating science

  • Dave Frame says:

    Anders wrote: “the scientific system aims at creating feedback from experiments and peer review that detects errors and in the long run corrects them. Political systems need that feedback too, or they will get hi-jacked by special interests that exploit them.”

    And this is the most powerful single argument I’ve seen for liberal democracy: that it self-corrects in the long run, since you can throw the bums out.

    The incentives that regular popularity contest creates can be a barrier to long-term thinking, of course, since it tends to prevent large structural changes that might be good in the long-term but have high adjustment costs (see for instance, Greek fiscal policy circa 1980-2010…). There are ways to bound those incentives (here in NZ we have a fiscal responsibility act which attempted to bound the ranges of fiscal tinkering for short-term political gain (http://www.treasury.govt.nz/publications/research-policy/wp/2001/01-25/04.htm/ – a few other countries have followed suit). As for political systems getting “hi-jacked by special interests that exploit them”… that’s just an endemic part of *any* political system. People compete for power. Best you can do is make it an open, transparent fight with some tight rules (though I appreciate that’s a very value-laden, historically located (Anglophone, liberal, democratic, etc etc) view).

  • Dave Frame says:

    Anders wrote: “In the end, coastal municipalities that base their plans on a best possible scenario will likely suffer the same fate as the summer resort in Jaws, where the authorities choose to ignore warnings. Reality tends to bite hard if you deliberately ignore it.”

    And presumably the same is true of those that scare off investors by basing policy on an equally unlikely worst possible scenario..? Actually I find the idea of legislating regarding the functional form of sea-level rise utterly routine yet quite funny. It’s funny because pretty much everyone I know talks about “x metres by 2100” (ie a point on a graph, rather than a functional form), so if anything the planners would seem to be over-determining the information (which is surprising). It’s utterly routine because administrative law routinely calls on specific ways of treating problems. Some pollutant is to be treated in this manner, some bridge loading in that manner, etc. We don’t usually get excited about the details of that treatment (even where it involves choices like this), especially when the information is as ropey as our understanding of sea-level rise. I think it’s more an indicator of how wildly over-heated the politics of climate change are, than it is an indicator of North Carolina legislators being morons.

    • Anders Sandberg says:

      There is a difference between legally deciding to using some sort of expert consensus on the proper way of treating a technical/scientific problem and deciding using legislator consensus. In the first case presumably the experts contribute relevant domain knowledge, in the second one there is not much of it. In practice one can (and does) move between these poles, of course. But it seems irrational to deliberately steer towards the lack of knowledge pole.

      The real problem with the North Carolina case is that they made a decision *in the light of scientific disagreement* to settle for one allowed method rather than for example mandating that both possible forms of extrapolations should be used in reports to get a sense for the uncertainty.

  • Dave Frame says:

    Well… sure… it seems a bit silly to wed yourself to past change when you have reasons to think future change may be much more severe. But the idea of using linear rather than exponential estimates is reasonable, given (1) how people in the field actually talk about it (2) in the presence of uncertainty about the functional form, etc. Policy makers like nice simple, clear bits of science around which the law can draw its bright lines. Choosing a functional form that you know will be subject to loads of end-point error, but which will doubtless be revised in the coming decades, seems like inviting trouble.

    And while I agree it’s backward-looking (literally) and short-sighted, I don’t think its a particularly egregious example of silly regulatory behaviour with respect to science. It’s not as ignorant and anti-science as Germany’s position on nuclear power, for instance. I suspect the reason North Carolina is copping so much flack on this is probably not unconnected from the fact that the current session is the first time Republicans have won control of the NC legislature in about a zillion years. It’s just science as a front in the American kulturkampf – Germany bans nuclear power: we intellectuals praise them for being “precautionary”; North Carolina uses a blunt instrument for planning around SLR: we excoriate them as anti-science and a laughingstock.*

    [I don’t feel particularly strongly about the details here: Germany could ban invaders from Mars and NC could legislate the details of the return of Quetzacoatl for all I care – but I do get tired of this sort of pattern in discussions of climate change. [A book I was a reading recently* made the point that power elites and intellectual elites have developed hostility towards each other over the last hundred years, a point which I think is true, regrettable, and an occupational pain in the arse for people who want to do nonpartisan climate science/policy work.]]

    *Even though the arguments for the NC position are deeply precautionary: Daivd Rouzer, a sponsor of the bill, said: “If you’re going to use science when you really can’t validate it … you’re going to be implementing policy and rules and regulations that can have a very, very negative impact on the coastal economy of this state.” [http://www.latimes.com/news/nationworld/nation/la-na-sea-level-20120624,0,3935676.story … I feel slightly embarrassed and unclean at citing a paper as crap as the LA Times… but it came up high on the list on google… even if the LA Times is usually more interested in people’s clothes/hairstyles/vehicle choice/net present value than their arguments I guess we probably can trust them to report direct quotes. ?]
    **I think it was Hagen Schulze’s Germany: A New History.


Subscribe Via Email