Mitigating epistemic risk

I just read a nice paper by Gabriele Contessa on the mitigation of inductive risk, cleverly titled On the Mitigation of Inductive Risk.1 His primary question is whether responsibly applying values in science should left be left to individuals or whether there ought to be community-level processes. He offers a number of arguments against the individualistic approach and briefly sketches what a socialized approach might look like.

I agree with Contessa’s negative arguments, and he introduces some helpful labels. For example: The “problem of precautionary cascades” is when individuals along the way each silently apply the same level of erring on the side of caution, so that the final outcome is exponentially more cautious than the individuals themselves would endorse.

His positive suggestion is to “involve experts from a number of different relevant disciplines as well as the relevant stakeholders in the assessment and management of inductive risk at the advisory stage.” A limitation of this strategy is that it is costly and slow. So deliberative procedures involving communication across disciplines and the community don’t make sense for every question. Sometimes the values involved are anodyne or obvious. Sometimes the evidence is so overwhelming that any plausible values would underwrite the same judgment.2 The social rigamarole is inappropriate if it’s one of those times.

Sadly, explicitly identifying that the values are anodyne or that the evidence is overwhelming is a separate judgement. It thus involves separate risks. Contessa calls this “the problem of second-order inductive risk”, but we can also push it to the third order and beyond. The regress must stop somewhere, in judgements which aren’t the result of collective deliberation and community consultation.

As such, the community-level approach to mitigating inductive risk requires more than just deliberative procedures for the hard cases. It also requires some guidance for identifying the hard cases, lest that fall back to just personal judgement.

  1. It connects to issues I’ve discussed in some of my published and unpublished work.
  2. I’ve blogged about this before.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.