On how the Dunning-Kruger effect complicates normative considerations
Individuals are notoriously self-serving in assessing their competences either in absolute terms or when comparing themselves to others. We are likely to think we are more sociable than others, better than most at judging character and sincerity, or that we perform above average in our workplaces. We often overestimate our levels of knowledge when we objectively know very little. In fact, this bias seems most potent when we are oblivious about some matter. At such times, we may act quite unrestrained in peddling the most absurd notions as facts to others. Our virtual lives of late, cluttered with half-baked claims and notions about the pandemic, offer plentiful evidence for this. What is even more disheartening is that, as the famous Dunning-Kruger effect teaches us, the more incompetent we are, the less likely we are to become aware of our own incompetence. Individuals often fall victim to this effect regardless of their intelligence, social grouping, or their successes at anticipating and counteracting a self-serving bias in some other area.
Despite being familiar in some form for several decades, the Dunning-Kruger effect has not seriously grabbed the attention of normative philosophers. Only epistemologists have seriously considered how it may affect epistemic obligations, for instance, how we should act in circumstances of assumed peer disagreement (Wiland, 2016). We have hardly considered the kinds of moral obligations we might have as individuals, or how we ought to shape public policies and institutions in the face of widespread Dunning-Kruger effects.
Consider, for instance, the decision of a highly educated person whether she should go into politics and compete for public seats. Imagine that this person is educated broadly enough to offer meaningful contributions over a range of public concerns. “But alas”, she reflects, “I don’t know enough about all the relevant laws, or how to draw up or revise a budget. I wasn’t trained for public administration. So I’m hardly competent to take up such a job.” But the educated person fails to consider that if she decides not to pursue the job, a far less competent person, one with far fewer scruples of the aforementioned kind, may attempt to take it up instead. Apart from considering merely whether she is qualified, she must assess whether the Dunning-Kruger effect will generate unwavering confidence in candidates who are far less qualified. So in the face of a lurking threat of social harms arising from incompetence, is the educated person obliged to overcome her reservations?
A further complication arises from the flip side of the Dunning-Kruger effect: in some cases, the truly competent exemplify tendencies to second-guess their competences, even if the area of competence is much more specific than in the previous example. Bertrand Russell noticed both sides of these self-assessment difficulties, when he famously stated that “in the modern world the stupid are cocksure while the intelligent are full of doubt” (1933). Simply, the awareness of the competent that there is still much they don’t know saps their confidence, whereas the incompetent are unperturbed in their lack of awareness of just how incompetent they are. But in that case, the obligation to overcome their reservations may include a psychological hurdle for the competent that makes it particularly demanding.
Assigning the competent with this obligation faces two other crucial difficulties. First, the main lesson of the Dunning-Kruger effect is that those who believe they are competent may very easily turn out to be incompetent. Thus, it’s quite possible that those taking up the obligation to save us from the incompetent, with the best of intentions, are themselves incompetent. When there is no one who could vouch for their competence, self-assessors will often overestimate themselves. Therefore, committing ourselves to beat the incompetent runs at least some risk that we are thereby enabling our own incompetence.
Second, even if we could safely and reliably establish, with the help of others, that we are truly competent, a moral question remains: how much should we be asked to do? How far-reaching is our obligation to clean up after the incompetent, or preventing them from ever making a mess? Surely, if we are competent, we are allowed to appeal to an “agent-centered prerogative […] a modest right of self-interest” (Cohen, 1996), not to invest most of our time, like in the case of taking up a public seat.
Whether moral complications arising from the Dunning-Kruger effect should affect the decisions of individuals, and how, remains an open question that requires serious thought. However, we might think that Dunning-Kruger effects are best neutralized at various levels of institutional structure. Education, for instance, might be attuned to help the most competent in overcoming their imposter syndromes, in steering and reassuring them towards positions of great social importance, and encouraging them to branch out of their epistemic comfort zones. This can, in turn, help the competent in overcoming their psychological barriers when taking up individual moral obligations.
If, however, education fails, the Dunning-Kruger effect stands out as an important consideration in setting up our electoral and governmental institutions. There is no doubt that the effect influences both the incompetent and the competent in their voting behavior, as well as in their decisions to pursue positions of leadership. An institutional arrangement that prevents the incompetent in some way from hijacking important public decisions may very well be the last frontier at which self-serving biases are to be repelled.
