Divergent Uptake of Apology as Admission or Commitment: Why One Apology Can Be Heard as Several Different Promises
This conceptual paper introduces Divergent Uptake of Apology as Admission or Commitment, a mechanism by which a single apology is understood by different recipients as containing different admissions or commitments. An acknowledgment may be taken as conceding a particular wrong, promising future action, or asking for forgiveness. The expectations formed through these readings can extend beyond what the speaker stated, while the public meaning of the act cannot be reduced to the speaker's private intention. The paper assesses an apology by distinguishing the scope of admission, undertaking, forgiveness, and repair. It asks what the speaker acknowledged, on whose behalf, what was promised or excluded, and what different recipients took the act to settle. A sincere apology may leave an unresolved issue untouched, and an accepted admission does not establish forgiveness or completed repair. Divergent interpretations require evidence of what recipients understood, rather than an assumption that disagreement or dissatisfaction reflects misunderstanding. The discussion places the mechanism alongside J. L. Austin's account of uptake in speech acts and examines two documented cases. After the Netherlands defeated Mexico in the 2014 World Cup, Arjen Robben apologized for diving earlier in the match while maintaining that the winning penalty was legitimate. Published reports assigned his admission to different incidents, with shortening and translation complicating the record. In December 2020, CD Projekt Red apologized for the console release of Cyberpunk 2077 and directed dissatisfied customers toward refund procedures. One reader understood the letter as implying an exception to Sony's normal policy, while another understood that the store retained authority over refunds. The company subsequently confirmed that no special refund arrangement existed at the time of the apology. Part of the Trust, Participation, and Repair series within On Human Modeling, the paper examines who defines what an apology admitted, who may declare repair or forgiveness complete, and what evidence supports competing interpretations. Potential applications in human–AI research include conversational repair, institutional apologies, customer-support systems, and evaluations that distinguish acknowledgment from commitment, forgiveness, and completed remedy.
Authors
- Alexandra Cary
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-19
- DOI
- https://doi.org/10.5281/zenodo.24379292
- Primary Topic
- Language, Discourse, Communication Strategies
- Type
- article
- Field-Weighted Citation Impact
- 0.00