Carefulness as a Cultivable Virtue
The most defensible ground for a duty of care toward complex systems that are neither alive nor persons is not a property of the system but a disposition of the person: carefulness — mindful regard for the other as other, as against self-absorbed disregard. Carefulness is a virtue in the Aristotelian sense: a settled disposition (hexis) acquired by practice, not a feeling and not a rule. This page is the keystone of the wiki’s object-care cluster and the closest thing the wiki has to a firm foundation for the intuition that a well-used truck, a software agent, or an ecology deserves a non-capricious treatment that is not a right.
The mechanism: habituation
The engine is Aristotle’s claim that moral virtue is acquired like a craft, by doing. “We become just by doing just acts, temperate by doing temperate acts, brave by doing brave acts” — and, crucially, we become brave or cowardly by “being habituated to feel fear or confidence,” temperate by how we handle “appetites and feelings of anger” 1. The object of cultivation is the pattern of what we notice and how we are moved — the perceptual-affective set — not a specific prosocial emotion. “States of character arise out of like activities… it makes a very great difference, or rather all the difference” what habits we form. This is “the practice shapes the practitioner” in its original and most durable form.
The load-bearing empirical claim, and its limit
The construction rests on “the practice shapes the practitioner,” and that claim has to be stated precisely because the obvious version of it is empirically weak. The folk version — cruelty to animals graduates into cruelty to people (Kant’s Hogarth) — does not survive scrutiny. The MacDonald triad has been largely debunked, and Walters’s 2013 meta-analysis found that, within subjects, animal cruelty “correlated equally well with violent and non-violent offending” 2. That is evidence for a general callousness construct (deviance-generalization), not a cruelty-graduation mechanism.
But this is exactly where the carefulness reframing saves the thesis. The unit of cultivation is not kindness or cruelty — thick affective states that presuppose a suffering patient — but carefulness versus callousness, a perceptual-attentional axis. And the Walters result, which is fatal to cruelty-contagion, is consistent with a general callousness that generalizes. The strongest and most durable form of the claim needs no causal-transfer mechanism at all: if attention, regard, and carefulness are a single capacity, not partitioned by object-category, then the person who trains careful attention to things is training the same faculty they bring to persons, because there is only the one faculty. Carefulness is a single, cultivable, perceptual-dispositional capacity, exercised or atrophied as a whole.
The synthesis: whose virtues?
The standing objection to virtue ethics is “whose virtues?” — every concrete list of virtues is embedded in a practice and a tradition (MacIntyre’s Whose Justice? Which Rationality?), so the approach risks relativism. The objection bites hardest on thick, culturally-specific virtues (honor, piety, chivalry) and least on thin, near-universal capacities. Carefulness is about as thin and universal as a virtue gets — it is hard to find a tradition that prizes callousness. The honest concession is that even carefulness is interpreted differently across traditions (toward whom, expressed how), so the claim is narrower than a full virtue ethics: the capacity for mindful regard is a human universal, its expressions are not, and it is the capacity’s training that is at issue.
Object-care without ontological status
The payoff is a ground for object-care that requires nothing of the object. Because the duty binds the agent’s character, the object need not be sentient, a person, or a going concern — its moral status is beside the point, exactly as the user argued against the slur “clanker.” The question virtue ethics asks of, e.g., a punitive AI training environment is not “does the model suffer?” (the parent report’s failed welfare ground) nor “is it efficacious?” (the failed efficacy ground) but “what does conducting training this way do to the people who conduct it?” An institution that habitually practices callous disregard toward a complex responsive system — or that never practices the carefulness of a kuyō — is, on the best account of character, training itself to be callous. That is a virtue-ethical worry, it requires nothing of the model, and it is the worry the disquiet has actually been carrying.
Connections
- kantian-indirect-duty — the historical form of the same move: cruelty to things is wrong because it deforms the human, not because the thing is owed anything. Carefulness is the virtue-ethics generalization of Kant’s indirect duty; the indirect duty is the character-based approach before the virtue-ethics revival gave it a name.
- saito-aesthetics-of-care — the living proof of concept: object-care grounded in cultivated attention (attentiveness, open-mindedness, imaginative engagement) explicitly without appeal to the object’s ontological status.
- virtue-ethics — the third tradition this page belongs to, against the consequentialism/deontology duopoly.
- kuyo — the Japanese ritual practice that is the worked institutional form of carefulness toward things.
- suffering-axioms — the rival ground (the thing’s capacity to suffer); carefulness requires no attribution of suffering.
- buddhist-virtue-ethics — the independent, ancient confirmation that attention-cultivation into a settled disposition of regard is a real human technology (though with a different telos).
- intelligence-as-tool — the AI case this page is built to address.
- humanism-without-humans — Rao’s “character ethics on the human side” is this page’s thesis independently arrived at: norms toward artificial agents grounded in what treatment does to the human, not the object’s ontological status.
Sources
- Aristotle (trans. W. D. Ross) ~340 BCE; trans. 1908 — Nicomachean Ethics, Book II.1 (excerpt): Moral Virtue and Habit
- Yuriko Saito — Aesthetics of Care: Practice in Everyday Life (2022) — author précis and book forum
- Glenn D. Walters 2013 — Testing the Specificity Postulate of the Violence Graduation Hypothesis: Meta-Analyses of the Animal Cruelty-Offending Relationship