Moral Responsibility Without Moral Agency (Palgrave Handbook Ch 18)

Nikhil Mahant · 2026 · Palgrave Handbook on the Ethics of AI, ch. 18, pp. 273-285 counterpoint high priority coded

Main argument

Thesis: the 'Agential View' - that for an entity to be morally responsible it must be a moral agent - should be rejected on conceptual-methodological grounds: the Agency Condition 'puts a stricter requirement on moral responsibility than is justified,' loading the concept of moral responsibility with agency features (intentionality, reasons-responsiveness, self-governance) that 'seem unrelated to the notion of moral responsibility'; stipulating 'moral agent' as a technical term is methodologically empty (it makes the View analytic); and theorizing about moral responsibility is 'most optimal' without the Agency Condition. Distinctively CONCEPTUAL rather than metaphysical (contra Semler): the question is not whether responsible entities must possess consciousness etc., but whether responsibility-theory should be built with an agency requirement at all. Upshot for AI: the responsibility-gap debate's shared premise (no moral agency -> no responsibility for the artifact -> gap or human transfer) is unmotivated - responsibility ascriptions to non-agents are conceptually coherent.

Why it matters here

THE CHALLENGE CHAPTER: argues against the 'Agential View' (that moral responsibility requires moral agency) on conceptual-methodological grounds - the assumption underlying BOTH the responsibility-gap literature AND the dissertation's own inference from 'AI is not a moral agent' to 'responsibility stays with humans'. If Mahant is right, denying AI moral agency does NOT settle where responsibility goes - a direct objection the dissertation must answer.

Reading notes

Close read (12pp; Uppsala, Marie Curie project 'AICon'). Distinguishes conceptual from metaphysical routes (contra Semler 2024); engages Hakli & Mäkelä, Coeckelbergh, Matthias, Hieronymi, Fischer. Sections: responsibility-gap motivation; dominant analysis of moral responsibility; agency features; the burden argument; defender's move; two classifications of responsibility kinds.

Mahant, N. (2026). Moral Responsibility Without Moral Agency. In S. S. Gouveia (Ed.), The Palgrave Handbook on the Ethics of AI (ch. 18). Palgrave.

Close reading — 3 coded units

#1 · pp. 273 · definition

“Call this the 'Agential View' of moral responsibility: according to the Agential View, for an entity to be morally responsible, it is necessary that it is a moral agent. The necessity of moral agency for moral responsibility ('Agency Condition') can be a substantial thesis only if it is not taken to be analytically true.”

RL-SYS

#2 · pp. 274–275 · argument

“[Agency features] seem unrelated to the notion of moral responsibility. Thus, the Agency Condition puts a stricter requirement on moral responsibility than is justified. One way to get around this problem is to make the notion of 'moral agent' a technical one by stipulating a definition. However, [...] this alternative is no more methodologically [sound].”

RL-SYS

#3 · pp. 283 · argument

“instead of a metaphysical, it takes a distinctive conceptual approach. While other approaches focus on whether morally responsible entities must (metaphysically) possess the features that they are assumed to possess (e.g., consciousness), I have made a case against Agential View on the ground that it would be the most optimal way to theorize about moral responsibility.”

TU-METAETH

Synthesis-matrix row

Memos (1)

theoretical · unit #1

PRIORITY OBJECTION TO ANSWER: Mahant attacks the inference the dissertation (and Sanwoolu, Hakli & Mäkelä, Coeckelbergh) relies on - from 'AI lacks moral agency' to 'responsibility remains with humans'. If responsibility doesn't require agency, denying AI agency leaves open that the SYSTEM bears (some kind of) responsibility - reopening RL-SYS. Available replies to develop: (a) disambiguate responsibility KINDS (Mahant's own Sect. 7 concedes classifications matter): attributability and liability may not require agency, but ACCOUNTABILITY (answerability - giving reasons, being blamed with uptake, making amends) plausibly does, and the practical stakes (compensation, sanction, repair) run through accountability - so the human-transfer conclusion survives for the responsibility-kind that matters to victims; (b) even granting non-agent responsibility as coherent, it is practically idle without sanctionability - which reconnects to Tasioulas's accountability-capacity point and the Diella case; (c) note convergence: Mahant's view actually HELPS one dissertation thread - it legitimates holding SYSTEMS to normative standards (constrained evaluability) without agency-metaphysics. Engage in the responsibility chapter as the strongest published counterargument; possibly worth direct correspondence (Uppsala, AICon project).