The Nonrandom Walk of Knowledge

Research output: Contribution to journalArticlepeer-review

Abstract

A person's epistemic goals sometimes clash with pragmatic ones. At times, rational agents will degrade the quality of their epistemic process in order to satisfy a goal that is knowledge-independent (for example, to gain status or at least keep the peace with friends.) This is particularly so when the epistemic quest concerns an abstract political or economic theory, where evidence is likely to be softer and open to interpretation. Before wide-scale adoption of the Internet, people sought out or stumbled upon evidence related to a proposition in a more random way. And it was difficult to aggregate the evidence of friends and other similar people to the exclusion of others, even if one had wanted to. Today, by contrast, the searchable Internet allows people to simultaneously pursue social and epistemic goals. This essay shows that the selection effect caused by a merging of social and epistemic activities will cause both polarization in beliefs and devaluation of expert testimony. This will occur even if agents are rational Bayesians and have moderate credences before talking to their peers. What appears to be rampant dogmatism could be just as well explained by the nonrandom walk in evidence-gathering. This explanation better matches the empirical evidence on how people behave on social media platforms. It also helps clarify why media outlets (not just the Internet platforms) might have their own pragmatic reasons to compromise their epistemic goals in today's competitive and polarized information market. Yet, it also makes policy intervention much more difficult, since we are unlikely to neatly separate individuals' epistemic goals from their social ones.

Original languageEnglish (US)
Pages (from-to)249-264
Number of pages16
JournalSocial Philosophy and Policy
Volume37
Issue number2
DOIs
StatePublished - Dec 1 2020

Keywords

  • Bayesian updating
  • collective epistemology
  • Internet
  • polarization
  • pragmatic reasons
  • selection effects

ASJC Scopus subject areas

  • Philosophy
  • Social Sciences(all)

Fingerprint

Dive into the research topics of 'The Nonrandom Walk of Knowledge'. Together they form a unique fingerprint.

Cite this