Why we’re epistemically flexible but stubbornly self-regarding

A deep-rooted psychological asymmetry

Dan Williams recently published a great essay on two types of pessimism about humanity: moral or motivational pessimism (especially about overcoming selfishness) and epistemic pessimism – pessimism about our ability to acquire and deploy knowledge. He argues that Thomas Sowell unhelpfully bundles them, and that the case for motivational pessimism is considerably stronger (with some nuances).

I share Dan’s views and highly recommend reading his essay. It got me thinking about why the motivational and the epistemic cases come apart. In my view, a root cause is that our inborn motivations are far more fixed than our inborn epistemic intuitions.

We can, of course, change our motivations to some degree, but there are clear limits. It’s effectively impossible for people to become completely impartial between themselves and unrelated others. My sense is that even in communes organised around sharing, people remain intuitively selfish in much of their everyday behaviour.

By contrast, our epistemic intuitions are more flexible. Take, for instance, the belief that objects like stones, trees, and tables are solid – a view that science has, in a sense, overturned. We can easily accept that verdict, even though those objects will continue to appear solid to us. We can replace our intuitive worldview with one that’s grounded in scientific evidence. In fact, we stake our lives on this revised worldview every time we board a plane or take a prescription drug.

Of course, many beliefs are distorted by cognitive biases and motivated reasoning, but their resistance to change is often overstated. Our beliefs form an interconnected web, so biased or self-serving beliefs will typically conflict with other things we believe. Moreover, we argue with others, who can reveal inconsistencies we refuse to see. This means that distorted beliefs can be corrected, and frequently are.

And we don’t just change our beliefs, but also our methods for acquiring them. While we’re naturally attracted to striking anecdotes, we’ve learnt that experiments and statistical inference are more reliable.

Even people with limited abilities can form fairly accurate beliefs thanks to a shortcut: deference to epistemic superiors. Instead of trying to figure things out themselves, they can listen to those with expertise. There’s no similarly effective shortcut for improving the motivations of unusually selfish people.

The asymmetry between the motivational and the epistemic cases ultimately has evolutionary roots. Since the genes of unselfish individuals were less likely to spread, there was strong pressure to lock selfish motivations in place (as Dan notes). There was no comparable pressure to lock in specific beliefs, such as the belief that tables are solid. On the contrary, there was pressure towards a degree of epistemic flexibility, since so much of what we need to know can only be learnt from experience.