Promise-Keeping and the Internal Judge

While standard models recognize intrinsic costs of lying, they typically focus on individual decision-making, treating reputational concerns as a trade off between the benefits of lying and the probability of detection by a passive observer. Using a strategic communication game in a controlled experiment, we show that honesty oaths shortcircuit this calculus by internalizing the audience, even when the audience is a strategic counterpart whose beliefs and actions determine payoffs. While the oath dramatically increases truth-telling, those who do break their promise systematically avoid brazen, detectable lies, retreating instead to ambiguity. Crucially, this refusal to be a “brazen renegade” is not a strategic reaction to the receiver: it persists even when the oath is private and compliance cannot be traced to the participant by the experimenter. This inelasticity to external scrutiny challenges standard reputational models: instead, our data are consistent with the oath-taker answering to an internal judge rather than an external one. Furthermore, we show that this internal audience does not strictly require the cognitive amnesia or imperfect recall required by standard self-signaling models; rather, it can also be understood as a present-moment, categorical refusal to generate inescapable evidence of one’s own transgression. Finally, we show that receivers intuit this mechanism, pricing in the oath’s “psychological enforceability” by granting credibility only when the speaker has no room to hide.