Kaizen Asiedu: Residual agency remains under illness and command voices; incentives still require consequences

The Gist

Sick, medicated, and (on her story) commanded by a voice, she was still the one who heard, planned, sent Patrick out, and obeyed. If you reject free will, you still need consequences because people respond to incentives either way. This steelman reconstructs Kaizen Asiedu's strongest case from the video for logical clarity; it is not an endorsement of his conclusions, clinical claims, or any legal outcome.

Conclusion

Even under mental illness, heavy medication, and an alleged command voice, Lindsay remained an agent who could refuse; if free will is denied, incentive-responsive behavior still requires consequences.

Premises

1. Lindsay's judgment was compromised by biology, psychology, and a rapid course of many medications, and she still could have chosen not to kill her kids and did not so choose.
2. On her account she heard a voice commanding her to kill; the hearer is distinct from the voice, so a response choice remained, including refusing the command.
3. Persons are not identical to their thoughts, feelings, or dark voices; they are the agents who experience and respond to conditions and are causes of their actions, not only effects of mind and circumstance.
4. Even in chaotic consciousness a conscience that knows right from wrong remains, harder to hear when sick; when the sick cannot hear it, society's voice must supply incentives that protect everyone.
5. Agency never fully disappears even in severe depression, overmedication, or psychosis; that residual agency is why rehabilitation tries to cultivate the agent rather than abandon the person.
6. Concrete choices in this case included looking up the restaurant distance, sending the husband away so he could not intervene, and carrying out slow successive strangulation.
7. Even if free will does not exist, behavior that responds to incentives exists; removing consequences removes an enforcement mechanism of good behavior, so both sides of the free-will dispute still need consequences.

Assumptions

• The command-voice claim is treated as Lindsay's account as presented, not independently verified.
• Could have chosen otherwise is the author's agency thesis, not a clinical finding.
• The free-will fork is offered so the consequences conclusion does not depend on winning the metaphysics debate.
• Mental illness is acknowledged as real; the target is using it as a full escape hatch from moral responsibility.

Analysis

Overall strength: Moderate. Argument type: Deductive.

Premise Strength

• P1: Lindsay's judgment was compromised... and she still could have chosen not to kill her kids and did not so choose. (Weak) — Presented as near-factual despite being explicitly flagged (A2) as a philosophical thesis rather than a clinical finding; no forensic evaluation, expert testimony, or diagnostic evidence is offered to support the counterfactual capacity claim.
• P2: On her account she heard a voice commanding her to kill; the hearer is distinct from the voice, so a response choice remained. (Weak) — Moves from a conceptual/grammatical distinction to an empirical claim about volitional control without bridging evidence; command hallucinations can involve degraded reality-testing that may functionally collapse this distinction in ways the premise does not address.
• P3: Persons are not identical to their thoughts, feelings, or dark voices; they are agents who are causes of their actions. (Weak) — A substantive, contested position in philosophy of mind and action theory, asserted as a starting point rather than independently justified; not shared by materialist, eliminativist, or standard clinical-psychiatric framings of psychosis.
• P4: A conscience that knows right from wrong remains, harder to hear when sick; society's voice must supply incentives. (Moderate) — The conscience claim is largely unfalsifiable and presupposes what is disputed in severe psychosis, but the practical corollary (society needs its own incentive mechanisms when internal moral cognition is impaired) is a defensible policy point independent of the stronger metaphysical claim.
• P5: Agency never fully disappears even in severe depression, overmedication, or psychosis; hence rehabilitation. (Weak) — Generalizes from a single case to a sweeping claim across a wide range of conditions and severities without citation; risks circularity, using the existence of rehabilitation practice to justify the very agency thesis that motivates rehabilitation.
• P6: Concrete choices included looking up restaurant distance, sending husband away, and slow strangulation. (Moderate) — The most empirically grounded and verifiable premise, and the most diagnostically useful evidence in the argument, but its interpretive use (as proof of intact rational/moral agency rather than organized psychotic behavior) is contested and not adequately defended against the alternative reading.
• P7: Even if free will does not exist, incentive-responsive behavior still requires consequences. (Strong) — Nearly self-evident as a policy-level claim and explicitly constructed to be neutral on the contested metaphysics, making it the argument's most robust component; its main limitation is that it justifies maintaining a general consequence system rather than establishing this specific defendant's individual moral desert.

Potential Fallacies

• Begging the question (circular support) (P2, reinforced by P3) — The claim that a 'hearer' distinct from the 'voice' entails a genuine response-choice treats a conceptual/grammatical separation as if it settles an empirical question about volitional capacity during acute psychosis. Whether that separation translates into functional capacity to refuse is exactly what is in dispute, not something the distinction itself establishes.
• Conflating behavioral organization with preserved moral capacity (P6, feeding into P1 and the conclusion) — Sequential, goal-directed acts (looking up a distance, sending someone away, carrying out a plan) are treated as strong evidence of intact rational agency. But psychiatric and legal literature on psychosis (including closely comparable real cases) shows that highly organized, even meticulous, behavior is fully compatible with profound impairment in the capacity to appreciate wrongfulness — organization is not proof of the specific kind of agency required for full moral or legal…
• Unfalsifiable construct (P4, P5) — Claims that 'a conscience that knows right from wrong remains' and that 'agency never fully disappears' are framed so broadly that no observation could disconfirm them; any behavior can be reinterpreted as consistent with residual conscience or agency, which weakens their evidentiary value even though they may be true in some minimal sense.
• Soft false dichotomy (Overall structure connecting P1-P6 to P7 and the conclusion) — The argument's practical resolution treats the options as either meaningful residual agency (full responsibility) or no free will at all (incentive-based responsibility), without seriously incorporating an intermediate, clinically standard position: substantially diminished but not absent capacity, which many legal systems already recognize as grounds for partial rather than full responsibility.
• Mild straw man (A4 and the general rhetorical framing) — Framing the opposing view as using mental illness as a 'full escape hatch' characterizes a nuanced clinical/legal position (graduated diminished responsibility, insanity standards) in its most extreme and least defensible form, making the argument's own moderate-sounding conclusion look more reasonable by contrast.

Counterarguments

• P2/P3 (High impact) — Forensic psychiatric evidence on command hallucinations shows that severe psychosis can involve degraded ego-boundaries and reality-testing such that the conceptual 'hearer distinct from voice' is not phenomenologically or functionally available in the moment, collapsing the inference to preserved refusal capacity.
• P6 (High impact) — Methodical, sequential planning is well documented in cases of genuine, clinically verified psychosis (including closely comparable real-world cases) that nonetheless met insanity standards; organization of behavior does not establish appreciation of wrongfulness or preserved moral agency.
• Conclusion / overall framing (High impact) — Legal responsibility is normally governed by specific, narrower standards (e.g., capacity to appreciate wrongfulness or conform conduct to law) rather than a general philosophical claim that some minimal agency persists; conflating the two risks applying a standard so permissive that virtually no case of psychosis could ground diminished responsibility.
• P7 (Medium impact) — The incentive-responsiveness argument justifies maintaining consequence-based systems as social policy but does not by itself establish that this particular defendant deserves full blame; general deterrence and individual desert are distinct questions that the fork elides.
• A4 / overall framing (Medium impact) — Characterizing the opposing position as seeking a 'full escape hatch' mischaracterizes more moderate clinical and legal positions that argue for graduated, not total, mitigation of responsibility.

Suggested Improvements

• Evidentiary grounding of the agency claim — Support P1 and P2 with forensic psychiatric evaluation, expert testimony, or clinical literature on command hallucinations and volitional control rather than relying on conceptual/philosophical assertion alone. Would convert the argument's weakest premises from stipulated philosophical claims into case-specific, testable evidence, closing the gap the argument itself acknowledges (A2) between thesis and clinical finding.
• Domain separation — Explicitly distinguish the philosophical/moral claim about residual agency from the legal capacity standard (e.g., M'Naghten or Model Penal Code tests) that actually governs criminal responsibility determinations. Prevents the argument from being read as settling a legal question it does not actually engage, and avoids overreach into domains with their own established evidentiary requirements.
• Engagement with the strongest counterargument — Directly address the well-documented possibility that organized, goal-directed behavior can coexist with genuine incapacity to appreciate wrongfulness, rather than presenting P6 as straightforwardly dispositive of agency. Strengthens the argument's credibility by showing it can survive its most serious objection rather than sidestepping it.
• Graduated responsibility framing — Replace the binary agency-present/absent framing with an explicit continuum model, specifying what degree of residual capacity would or would not suffice for full responsibility. Avoids the false-dichotomy risk and provides a limiting principle, addressing concerns that the current framing could justify punishing virtually any defendant regardless of actual incapacity.
• Separating policy from individual desert — Clarify that P7 justifies retaining consequence-based systems in general, and treat individual culpability in this specific case as a separate question requiring case-specific evidence. Prevents smuggling an individual blame conclusion in through a general deterrence/policy argument, keeping the strongest part of the argument (P7) logically insulated from the weaker parts (P1-P6).

Scenario Tests

• A defendant has documented, multiply-corroborated complete loss of reality-testing during the offense, with no lucid interval, verified by independent psychiatric evaluation. (Challenges) — Forces the argument to specify a limiting principle for how much impairment defeats 'residual agency'; without one, the claim that agency 'never fully disappears' risks being unfalsifiable and inconsistent with a clear case of genuine incapacity.
• The same behavioral facts (researching logistics, isolating a potential intervener) occur but are shown to be driven by a specific, corroborated delusion (e.g., belief that the act would save the children from a worse fate). (Challenges) — Reframes the 'concrete choices' in P6 as products of organized delusional thinking rather than evidence of intact rational agency, undermining the inferential weight placed on planning behavior.
• P7's incentive-responsiveness criterion is applied to behavior modifiable by punishment in populations clearly lacking full moral agency (e.g., young children or non-human animals). (Challenges) — Reveals that 'responds to incentives' is too low a bar to track moral culpability on its own, showing P7 justifies maintaining consequence systems as policy without settling questions of individual desert.
• A case involving a non-psychotic defendant who simply disagreed with a law on moral grounds and claims a 'voice of conscience' compelled disobedience. (Supports) — Here the argument's core insight (agent distinct from thoughts/feelings, and incentives operate regardless of free will) applies cleanly, without the complicating factors of documented psychosis and heavy medication, showing the argument's structure has genuine force outside contested psychiatric contexts.
• The relevant jurisdiction's insanity standard is applied directly to the case facts as commonly reported (rapid polypharmacy, alleged command hallucination, extensive planning). (Neutral) — Outcome would depend heavily on expert testimony not presented in the argument; illustrates that the philosophical argument alone cannot predict or substitute for the actual legal determination, which turns on evidence outside the argument's scope.

Coherence & Relevance

The argument's dilemma structure is internally coherent: if either free will exists (supported, weakly, by P1-P6) or it does not (addressed by P7), consequences are said to follow either way. This makes the practical conclusion resilient to the free-will debate, which is a genuine dialectical strength. However, the coherence is asymmetric — the first horn rests on a chain of contested philosophical claims and a single ambiguous piece of behavioral evidence, while the second horn is comparatively simple and robust. The overall argument therefore reads as a strong policy conclusion (consequences should exist) loosely bundled with a much weaker case-specific claim (this particular defendant retained full-enough agency for full responsibility), and these two levels of confidence should not be treated as equally warranted.

• P1 (Strong) — Directly asserts the conclusion's core claim but does so via assertion rather than clinical or forensic support, leaving the connection evidentially thin despite being conceptually central.
• P2 (Strong) — Central to establishing the residual-choice claim, but the inferential step from conceptual distinctness to empirical capacity is unsupported and is the argument's most exploitable weak point.
• P3 (Moderate) — Provides philosophical scaffolding for P2 but is itself a contested metaphysical position not independently defended, so it reinforces rather than resolves the gap in P2.
• P4 (Moderate) — Connects illness to the need for external incentives, bridging toward P7, but the conscience claim is unfalsifiable and adds limited independent evidentiary weight.
• P5 (Moderate) — Generalizes the case-specific claim to a broader population without citing supporting data, and risks circularity by using rehabilitation practice to justify the agency thesis it is meant to support.
• P6 (Strong) — The most concrete and verifiable support offered, but its interpretation as proof of agency (rather than organized psychotic behavior) is contested and not adequately defended against the alternative reading.
• P7 (Strong) — Cleanly connects to the conclusion's second disjunct and is well-supported independent of the contested metaphysics, though it addresses policy justification for consequences rather than this individual's specific culpability.

View this argument on LogicFirst.ai