Academic journal article Erasmus Journal for Philosophy and Economics

Challenging the Majority Rule in Matters of Truth

Academic journal article Erasmus Journal for Philosophy and Economics

Challenging the Majority Rule in Matters of Truth

Article excerpt

Surprises can be useful in epistemology. Epistemology is most helpful when it leads to normative recommendations that are surprising in that they are counterintuitive or in contradiction with established practice (Miriam Solomon 2006, 30).

Seeking and utilising the advice of experts is a very common and useful practice in a complex world; this is especially so given the ever-increasing stream of information, which no single individual can comprehend and process entirely on her own. We regularly ask experts for advice. And in many cases we ask different experts for their independent advice on one and the same issue. If we, for instance, fear a serious disease we may well ask several medical specialists for their diagnoses. But what if the diagnoses given are inconsistent?

How should and how do we actually cope with disagreement among experts? In an explorative paper Alvin Goldman (2001) investigates what good reasons a novice might have for trusting one putative expert more than another. He first presents a (non-exhaustive) list of such reasons. According to the second entry in his list, Goldman's advice to the layman confronted with conflicting expert judgments is to check for "agreement from additional putative experts on one side of the subject in question" (Goldman 2001, 93).

As a matter of fact, this piece of advice is the one Goldman discusses most extensively in his paper. He gives the following formal argument:

Goldman's proposition. If expert judgments are sufficiently reliable and independent of each other, then additional experts confirming some proposition φ add positive credibility to φ (Goldman 2001, 99-101).

The reader may be somewhat disappointed about this result. It does not seem to suit the needs of a layperson confronted with conflicting expert judgments very well. What we would rather have is something like this:

Proposition (*). If expert judgments are sufficiently reliable and independent of each other, thenφ is more probably true than not, if the number of experts confirmingφ exceeds the number of experts confirming-φ.

The idea that the majority of expert judgments may play a decisive role as a basis of informed decisions appears very natural to us. This is somewhat reflected in the vast literature on the aggregation of opinions in psychology and management science (see Budescu 2006; Yaniv 2004 for overviews). There is much evidence to suggest that people predominantly use simple averaging rules to aggregate information from multiple sources. Averaging has recently also become salient in the debate about swarm intelligence and the wisdom of crowds.1 In the case of a categorical binary judgment (e.g., whether or not a surgery is needed to cure a disease), assigning equal weight to the independent judgments of various experts and then averaging them amounts to following the advice of the majority (Yaniv 2004, 76).

In social epistemology, the rediscovery of Condorcet's jury theorem2 eventually made the majority rule a standard reference point in the debate about the demands of epistemic rationality when merging the opinions of independent judges, and in other places such as the recent debate about the so-called discursive dilemma (Kornhauser and Sager 1986, see discussion below).

Thus, there seem to be good reasons to consider the majority rule not only a widely used, but also a well-founded guiding principle in forming an opinion on the basis of diverging expert judgments. However, a claim like (*) is not true in general and Goldman is well advised not to make it! There are important cases in which the majority just cannot decide on matters of truth, even though all judgments are made independently by equally competent jurors with exactly the same opportunity to obtain information on the issue. The object of this paper is to demonstrate the existence of these cases, mainly by presenting a simple counterexample.

Within the theory of judgment aggregation (see List 2012 for an overview) it has been acknowledged that simply following the majority of judgments is not necessarily optimal if truth is the object. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.