Skip to content
Read • Question • ConnectAnthony’s notes & researchReturn with a new question
Anthony’s NotesA personal field guide
From the notebook

Can a Feeling Be Wrong - Stoics vs Hume and Hutcheson

The claim under test: “Morality itself ultimately depends on certain affections and sentiments.” What Seneca and Epictetus would say to it, and where the real fault line between Stoicism and Scottish sentimentalism actually runs — which is not where the textbook contrast puts it.

The divide in one line

Stoic: your feelings answer to reality — they contain claims, claims can be false, so a feeling can be refuted.
Hume: your feelings are the reality — there is nothing behind them to answer to, so a feeling can be changed but never refuted.

Not reason versus feeling. Whether there is a fact of the matter your sentiments can be wrong about. Everything else follows from that.


1. The question that isolates the disagreement

Not “is feeling important?” — both sides say yes. The question is: can a feeling be mistaken? Not unhelpful, not unpopular. Wrong — mistaken about something.

Test case: you feel disgust at an interracial couple.

  • Stoic: you are making an error. Something in the world is a certain way, your reaction says it is another way, and the reaction fails to match — the same shape as thinking a stick in water is bent. You can be corrected because there is something to be corrected against.
  • Hume: there is nothing your feeling failed to match. Disgust is not a report about the couple; it is a fact about you. The inventory is: a couple, and a person having a reaction. You can call the reaction ugly, inconsistent, socially disastrous. You cannot call it false, because it was not claiming anything.

Why Hume ends up there. Facts do not tell you what to do. “This causes suffering” is a fact; “don’t do it” is a push. Only feeling supplies the push, so feeling is the source — and a source cannot be checked against itself. → Hume Is-Ought Problem

Why the Stoics do not. They think the world has a built-in direction, so “this is bad” is a real claim about that direction and can be true or false. And they think seeing it clearly already moves you — no separate push required.

The practical consequence

If Hume is right, moral disagreement is not two people looking at the same fact with one of them getting it wrong. It is two people built differently, and no amount of looking harder settles it. → Moral Realism vs. Anti-Realism · Subjective Morality


2. Three contrasts that dissolve on inspection

The textbook version of this comparison is mostly wrong

Each of these is a live misconception, and dropping them is what makes the real divide visible.

“Stoics are rationalists; sentimentalists care about feeling.”
Both sides are theorists of the affections. The Stoics have an elaborate taxonomy — the four generic passions, and the three eupatheiai (joy, caution, wish) that the sage does have. Apatheia means freedom from the four disturbances, not absence of affect. → Stoicism and Emotion - Margaret R Graver

“Reason versus passion.”
The orthodox Stoic soul is a unity. Chrysippus denies there is a separate irrational part to be at war with: passion is reason gone wrong, not something else overpowering reason. So the Stoics do not accept the battlefield Hume describes when he says reason is and ought to be the slave of the passions — they think the metaphor has no referent. (Posidonius Platonized back toward a divided soul, which is why Galen uses him against Chrysippus in On the Doctrines of Hippocrates and Plato — but Seneca and Epictetus are Chrysippean.)

“Sentimentalists are subjectivists.”
Not Hutcheson. For him the moral sense is a real perceptual faculty, implanted by God, reliably approving benevolence. That is a teleological guarantee, which makes him structurally closer to Stoic oikeiōsis than to Hume. Hutcheson and James Moor published a translation of Marcus Aurelius in 1742: the Scots were reading the Stoics as allies, not opponents.


3. The real divide: a Euthyphro question

Is an affective response correct because it tracks something, or is the response what makes the thing so?

StoicHume
Status of the responseAnswerable to a standard outside itselfConstitutive of the moral fact
The standardNature is rationally ordered; some things really are kata phusinThere is none; vice and virtue are not in the object
Analogy Hume reaches for—Secondary qualities — colours and sounds, perceptions in the mind, not real qualities in bodies (Treatise 3.1.1)
Verdict on a feelingDefendant before the tribunal of right reasonNot a candidate for truth at all

Hutcheson is the hard case because he wants both: the moral sense operates like a secondary-quality perception, and it was designed to approve of what deserves approval. That is a live tension in him, and it is why Hume — who drops the theology — ends up somewhere Hutcheson did not intend.

Filed under the cognitivism question → Moral Cognitivism vs. Non-Cognitivism · truth-apt · Emotivism


4. The second divide, downstream of the first: whose standpoint?

  • Stoics build from the agent asking what should I do? Virtue is the perfection of that agent’s practical reason and the sole good; everything else — health, reputation, your child’s survival — is a preferred indifferent.
  • Hume and Smith build from the spectator asking what should I approve? Moral qualities are whatever excites approbation in a disinterested observer; Smith then does the work of making that observer impartial and internalized.

This is a genuinely different question, and it explains the concrete disagreements:

  • Hume attacks the “monkish virtues” — celibacy, fasting, penance, self-denial — in EPM 9, on the grounds that they serve no purpose; Stoic self-hardening reads to him as adjacent to that.
  • Smith’s treatment in Theory of Moral Sentiments VII.ii.1 is more sympathetic but lands the same objection: Stoic detachment asks us to feel less for our own children than natural affection dictates, and that under-response is itself a moral failing on a spectator standard.
  • The two verdicts on the same man: Epictetus’ father at Discourses 1.11 is convicted by the Stoics for acting wrongly. On a Humean reading, a father who felt too little would already be defective before he did anything.

Cross-vault → Stoic Compassion and Adam Smith's Sympathy · Sympathy vs Empathy · Adam Smith


5. What Seneca and Epictetus would say to the claim itself

They concede the phenomenon and reject the promotion of it to a foundation.

They deny that affections are a non-rational faculty

A passion is not a brute feeling that reason then rides. It is a judgment — assent to an impression that something is good or bad. Chrysippus’ definition (pathos as krisis) stands behind Enchiridion 5: people are disturbed not by things but by their opinions about things.

So when the sentimentalist says morality rests on affections, the Stoic hears: morality rests on assents to evaluative impressions — his own position, badly described. The question is not whether feeling is involved but whether the judgment embedded in the feeling is true.

They concede the starting point, not the criterion

Seneca grants what the sentimentalist is pointing at. In De Ira 2.2–4 he isolates the primus motus — the involuntary shock, the shiver at a shout, the flush of anger — and insists it is not yet a passion, because passion requires the mind’s assent. And the whole developmental story (oikeiōsis) begins in natural affection. Nature’s impulses are the pointer; they are not the standard. On this view the sentimentalist has correctly identified the propatheia and then promoted it to a foundation.

Epictetus’ test case — the passage to have on hand

Discourses 1.11: the father who ran out of the room and abandoned his sick daughter, then defended himself by saying he acted out of natural affection (philostorgia). Epictetus takes the case apart precisely to show that affection cannot be self-certifying: the man’s feeling was real, his action was wrong, and only judgment about what is genuinely good could tell him so.

He then generalizes with the image of a measure — we do not settle weights by how heavy things feel. Whenever feelings conflict, something outside the feelings has to adjudicate; that something is the kanōn, the criterion. (Compare Discourses 2.11: philosophy begins in the recognition that conflicting opinions require a standard.)

Seneca’s sharpest line against sentimentalism

De Clementia 2.5–6. Pity (misericordia) is a defect of soul — a sickness that another’s grief produces in the beholder. The wise man relieves suffering without contracting it; clemency is the rational disposition, pity the disease that mimics it.

This is a head-on collision with the Hume–Smith sympathy tradition, which grounds approbation in exactly that shared feeling. Seneca’s argument: a helper governed by sympathy will help whoever moves him most, which is not the same as helping whoever is owed help.

The structural objection

Enchiridion 1: the good lies in what is up to us — judgment, impulse, desire, aversion. Affections that arise unbidden and vary with temperament, digestion and proximity are partly the province of fortune. If morality depends on them, morality is hostage to fortune, and virtue is no longer the one thing that cannot be taken away. Seneca makes the same point from the other end at Ep. 66: virtue admits of no increase or decrease. A morality grounded in sentiment does. → Dichotomy of Control


6. Where the sentimentalist still lands a punch

Hume’s real weapon is not “feelings matter.” It is the motivation argument: reason alone is inert — “reason of itself is utterly impotent” in producing action (Treatise 3.1.1.6, restated at EPM Appendix I).

The Stoic reply is not to deny motivational internalism but to reject the Humean psychology that makes it a problem. An impulse (hormē) just is assent to an impression presenting an action as appropriate — the impulsive impression, phantasia hormētikē. Judgment on this account is intrinsically motivating, so no separate conative faculty is needed.

That makes Stoicism one of the strongest historical counterexamples to the Humean theory of motivation, which is why it keeps resurfacing in the internalism debate. → Internalism and externalism

The cost of that reply

It is load-bearing on Stoic physics. It works because value judgments are true or false against a rationally ordered nature. Strip out the teleological cosmology and you are left with a cognitivist psychology that no longer has anything to be cognitive about.

That is precisely the gap the neo-Aristotelian naturalists are trying to fill without the cosmology — keeping the Stoic-shaped answer (responses answerable to facts about the kind of creature we are) while dropping the logos. → Philippa Foot · Moral Naturalism · framework of virtue


7. Plain-language version

Both think feelings are central. That is not the disagreement. The disagreement is whether a feeling can be mistaken.

  • A Stoic says: your anger contains a claim — this is an outrage. Claims can be false. So your anger can be wrong the way a belief can be wrong, and you can be argued out of it.
  • Hume says: your anger is not claiming anything. It is a reaction. There is no fact it is failing to match, so there is nothing to correct. You can change it, but not by refuting it.

And Stoics do not want you emotionless. The ideal person feels joy, love, goodwill, care — corrected feelings, not shrunken ones. What drops away is only the feelings built on false valuations: panic about money, rage about insults.

The second difference: where the same story leads. Both trace where our feelings come from — parents, teachers, poets, the crowd, what everyone praises. Nearly the same account.

  • Stoic: that is how we got corrupted. Real values exist underneath; we were taught wrong.
  • Hume: that is morality. There is no layer underneath. Shared human reactions, shaped by shared life, are the whole thing.

That second contrast is developed from Graver’s Chapters 6–7 in → Graver Audit - Correcting the Stoic Side of the Hume Comparison


Provenance and verification status

Drafted from a Claude conversation, 2026-08-22 (share link in Source). Verification is uneven and worth knowing before quoting:

  • The Hume passages in §3–§6 and the correction apparatus in the companion note were checked against the text in-session; see the verdicts in Hume's Corrections of Sentiment - The General Point of View.
  • The Stoic passage citations in §5 and the passage index are unverified against the primary texts. They are plausible and standard, but treat locators as to be checked before they go into anything public.
  • The Hutcheson/Moor 1742 Marcus Aurelius translation and the TMS VII.ii.1 and EPM 9 references are likewise uncollated.
  • Nothing here is sourced from Graver — she never mentions Hume, Hutcheson, or the Scottish Enlightenment. See the audit note.

Source

Claude conversation, 22 August 2026 — https://claude.ai/share/54448ade-4c4e-46e8-ad5f-77a2817d3316