On Common Knowledge and Ad Populum: Acceptance as Grounds for Acceptability

Starting points for argumentation

All reasoning, including the reasoning used in argument, has to start from somewhere. Although it may be possible, in principle, to offer support for every claim, in any particular case this strategy cannot be used without hopeless regress. Thus, not every claim used in reasoning can owe its acceptability to some set of reasons offered in its support. Instead, in the context of any given argument or piece of reasoning, some claims must be accepted—if only as starting places—on some other basis. These claims can be called the basic premises of an argument.¹

Dialogic approaches to argumentation typically take as the starting place of argumentation the discussants' shared commitments. Alternately, rhetorical approaches standardly take as the starting place of argumentation the audience's existing commitment set or, more broadly, whatever an audience is willing to accept. With each approach the idea seems to be that the effectiveness of persuasion depends on the commitment of the audience to the starting points of argumentation. In a dialogic context, if there are no points of agreement between a proponent and opponent, there is nothing for arguers to "take hold of " when designing and deploying their arguments and no space in which argumentation can take place. In reasoning more generally, if no claims are initially admitted, there is nothing from which inferences can be drawn, for inference can only generate claims on the basis of other claims. Further, there may even be nothing with which to draw inferences, for if no inferential rules are initially accepted, there will be no inferential moves that can be made even given some initial data set. In these approaches the fact of acceptance (or agreement) seems to give a prima facie acceptability to a set of claims (which I will call an initial commitment set) used as a starting place for argumentation.

As a starting place for argumentation, an initial commitment set can include both good and bad information. More generally, we tend to hold that most of our commitments are fallible—they are subject to defeat as refuting evidence comes to light. As such, some claims in an initial commitment set might be unacceptable according to any relevant standard of acceptance. The hope is that the projects of inquiry, argumentation, critical examination, and rational assessment will help sort things out by weeding out the bad claims in the initial commitment set. So, that some claim is accepted by an arguer is a reason for it to be a starting place in argumentation, although it is not on its own a reason (even a prima facie reason) for its acceptability. Rather, the acceptability of a claim is determined by how well it survives the process of argumentation, not where it stands at the beginning.

In this article I explore the role acceptance can play in establishing the acceptability of a claim by examining the relationship between appeals to common knowledge and appeals to popular opinion. Typically, that a claim is common knowledge is taken as grounds for its acceptability, whereas appeals to popular opinion are seen as fallacious attempts to support a claim. Against this I argue that appeals to common knowledge generally provide no better evidence for a claim than appeals to popular opinion and, as such, that appeals to common knowledge ought to be just as successful—or unsuccessful—as appeals ad populum.

I begin by describing a standard account of appeals to common knowledge and popular opinion that should be familiar to anyone who has taught or studied reasoning skills. I proceed to set out an alternative to this standard view (largely due to Douglas Walton) on which some appeals to popularity can provide defeasible yet presumptive support to a claim sufficient to shift a burden of proof on the balance of considerations. In general, I hold that the standard account is correct and that where ad populum appeals succeed in providing good reasons, they do so because they have been reconstructed as having another argument form that introduces independent reasons for accepting the conclusion. Each of these accounts helps to...