In contemporary religious epistemology, considerable attention has been devoted to justifying religious beliefs by appealing to testimony as a source of knowledge. Philosophers such as Mats Wahlberg have developed the model of “revelation as testimony,” aiming to defend the rationality of religious belief by treating revelation as a form of testimonial evidence. In this paper, we examine two significant limitations of this model, focusing specifically on the metaphysics of testimony.
As Jennifer Lackey emphasizes, testimony is metaphysically realized only if the hearer comes to believe a proposition p on the basis of the content of the speaker’s testimony. This condition entails two essential components: (a) a semantic connection between speaker and hearer; and (b) the hearer’s deference and trust in the speaker. Each of these components imposes limitations on the ‘revelation as testimony’ model, as detailed below.
a) Semantic connection between speaker and hearer
According to this condition, if the audience does not understand the meaning of the speaker’s proposition, they cannot form a belief merely by trusting testimony. As Wahlberg illustrates with a scientific example, one may trust physicists and come to believe that “the (relativistic) mass of an object increases as its velocity increases” only if one understands the meaning of the proposition. Understanding the proposition requires comprehension of its constituent terms, such as “mass” and “velocity.”
Accordingly, children, individuals with severe cognitive impairments, illiterate or semi-literate persons, or anyone who does not understand the meaning of the propositions in the Bible or the Qur’an cannot form beliefs based on deference to testimony. In other words, testimonial justification presupposes a certain level of semantic and conceptual understanding on the part of the hearer.
b) Deference and trust in the speaker
The second component of testimonial belief is that it is formed and sustained through deference to the authority of another. A belief based on testimony depends on the credibility and reliability of the speaker. Therefore, if a person already believes p on independent grounds - for instance, through a philosophical argument for divine simplicity - the mere assertion of p in revelation does not render that belief testimonial. Philosophers such as Wolterstorff, Kenny, and Gillson emphasize that in the case of religious belief, faith in revelation involves deference to God as the ultimate epistemic authority.
A potential objection arises from the epistemological concept of mixed grounds, whereby a belief may be justified by multiple epistemic sources simultaneously. One might argue that in the case of revelation, it is permissible for a belief to be grounded both in argument and in the testimony of the Qur’an. However, this objection does not undermine the constraints highlighted above. In the epistemology of testimony, even the most robust advocates of deference, such as Zagzebski, maintain that epistemic autonomy is an ideal to be pursued. Reliance on deference is justified only when independent epistemic faculties are either incapable of reaching the truth or are more likely to lead to error than the belief obtained by trusting another.
In conclusion, testimonial justification of revealed belief is limited. It is only available to those who (i) possess sufficient understanding of the meaning of the revealed propositions, and (ii) lack conclusive independent reasons to believe these propositions apart from revelation. These two conditions highlight the epistemic boundaries of the “revelation as testimony” model, suggesting that testimonial justification cannot universally apply to all hearers.