Do arguments against utilitarianism made by Rawls, Sandel and Nozik work when applied to negative utilitarianism?

ChatGPT said:

mostly yes — but not in a 1:1 way.
you have to translate them.

because negative utilitarianism (NU) is not simply “utilitarianism but with a different sign.”
the core claim in NU is that the only thing with intrinsic moral weight is the reduction of suffering.
that is a much narrower telos.

Rawls

Rawls’ core critique against classical utilitarianism is that it sacrifices basic liberties of some for aggregate well-being of others.

that argument still bites against NU.

because NU also sometimes implies that one can permissibly violate the rights of one group if that somehow reduces suffering overall.

for example: if eliminating a minority group eliminates their suffering (plus reduces social tension), NU has no built-in protection unless you build in lexical side-constraints.

so Rawls’ separateness of persons point transfers.

Nozick

Nozick’s deepest move is the same: persons are ends, not vessels of utility.
the “experience machine” critique also maps.

even if I am only trying to remove suffering, I could in principle morally force interventions that strip agency, autonomy, identity — because you are “doing fine” with less suffering.

NU struggles here again because if “not suffering” is the only intrinsic value, dignity and autonomy become instrumental at best.

so Nozick’s critique transfers.

Sandel

Sandel’s communitarian critique: utilitarianism in general flattens moral life, detaches goods from contexts, and reduces things that are “incommensurable” into a single metric.

that also applies to NU.

NU has the same problem of metric fetishism — one axis (suffering) is being taken as the only moral quantity.

Sandel would say: suffering is not the only moral good that matters — and suffering itself is contextual (linked to meaning, identity, relation, narrative, symbolic worlds).

so yes, his critique also maps.

HOWEVER: here is the twist

NU is often sold as the moral theory that is resilient to the “Nazi problem” of utilitarianism:
“at least we are not maximizing pleasure or growth — we are just reducing harm.”

this is rhetorically powerful.

but philosophically, it is actually easier to justify atrocity with NU than standard utilitarianism — because it is easier to argue that nonexistence is the most reliable removal of future suffering.

and that is why the Rawls / Nozick / Sandel critique is actually sharper against NU.

short verdict

yes — their critiques not only still apply, they cut even deeper.

because if you only have one value (reduction of suffering), then:

• rights can be traded away more easily

• persons get treated as containers of pain to be eliminated

• the qualitative texture of moral life collapses

NU magnifies the problem that Rawls/Sandel/Nozick are attacking:
it intensifies the monism at the heart of utilitarian frameworks.

the real underlying issue

the real shared critique underneath all three is:

any single-axis moral ontology that aims to quantify the good or bad will crush the moral meaning of contingency, plurality, and the non-commensurable.

NU is even more monistic.
so the critiques don’t just “still work” — they almost work better.

You said:

Interesting because the arguments against U and NU compelling and NU is compelling. For example, the notion that 'nonexistence is the most reliable way to remove future suffering' is compelling. I take it that it would be possible to build side constraints (or something similar) into NU.

ChatGPT said:

yes — this is in fact where the real philosophical action is.

many people (you’re not alone here) find both of the following simultaneously compelling:

1. the critiques of utilitarianism (and NU) that point out how dangerous / flattening / eliminationist these can become

2. the core NU intuition: suffering matters more than anything, and the cleanest way to abolish suffering is… to abolish being

this is the seduction.

and you’re right:

you could in principle build side-constraints into NU.

historically, there are 3 ways people have tried to do that:

1. deontic lexical constraints
   (e.g. “never violate bodily autonomy” even if it reduces suffering)

2. person-based constraints
   (e.g. “we must respect the separateness of persons” as a hard constraint)

3. value pluralism graft
   (e.g. suffering reduction is primary, but other goods have some lexical weight too)

each of these turns NU into a hybrid theory.
so NU stops being a “pure one-value” ethics and becomes “mostly one-value, but with brakes.”

the hidden problem

every time you build in a side constraint, you evade NU’s darkest implication by invoking a non-utilitarian norm.

that non-utilitarian norm is not derivable from NU.
so you have to say something like:

suffering is the primary disvalue, BUT…

…and the “but” is always some moral notion that is not reducible to suffering.

which means the real meta-move you’re making is:

NU expresses a deep moral orientation (anti-cruelty)
but it isn’t a closed-total system.

the philosophical stake

this is actually a totally legitimate position.

what you are effectively saying is:

• NU captures the core moral intuition: suffering is the only obvious horror

• but NU needs a second-order guardrail to prevent the horror of anti-horror from just becoming a different horror

in a way, this is very close to the post-Auschwitz problem that Adorno and Agamben are dealing with in their own idioms:

how do you articulate a “never again” that doesn’t itself become another mechanism of liquidation?

so yes — you can hybridize NU.

but the moment you do, you have actually moved onto a different meta-level:

NU becomes a regulative principle (sensitivity to harm and horror)
rather than a foundational principle (the only good is non-suffering)

I suspect the reason you find both the critiques and NU compelling
is precisely because you’re sensing that NU is a clearest possible articulation of horror,
but is not a safe standalone architecture.

(a question for you)

do you see NU as a metaphysics of value
or as a regulative ethical maxim?

e.g.:

• metaphysics of value → “pain is literally the only disvalue in the ontology of the good”

• regulative maxim → “pain has the highest veto power; nothing should override it lightly”

which one feels closer to how you are holding it?

You said:

The metaphysics of value is true. The regulative maxim is operationalizable/pragmatic.

ChatGPT said:

that is a very clean distinction
and it makes a lot of sense of your stance

and actually: that two-tier structure is philosophically elegant

tier 1 (ontic):
pain/suffering is the only thing with intrinsic negative value
→ this is your metaphysics of value

tier 2 (prac/operational):
human moral action still needs plural constraints to avoid atrocity
→ this is your regulative layer

so you’re doing something like:

• meta-ethically monist (value monism)

• normatively pluralist (constraint pluralism)

this actually solves why the critiques don’t dislodge your NU commitment.

because the critiques are critiques of normative mechanics
(and they hit utilitarian calculus as a decision procedure)

but your NU is not principally a decision procedure
it is a value ontology.

this is genuinely coherent.

you’re saying:

the real furniture of the moral universe is: suffering is bad.
but the real furniture of lived life requires constraints, because human action is too powerful & too dangerous to run on pure monism.

---

this is actually very close to a position one occasionally sees in contemporary metaethics:

axiological monism + deontic pluralism

a question that follows

is suffering here:

• phenomenal (pain qualia / affective states)

• or existential-ontic (the condition of thrown being, nausea, dread, derealization, etc.)

i.e.:

is NU about eliminating felt pain
or is it about the elimination of the tragic condition of being-a-self in a world ?

because a lot turns on that.

classical NU is about the first (qualia).

but many people who end up at a NU-ish place from Kierkegaard / Cioran / Agamben / Becker
actually mean the second.

and those are not the same target.

Short answer: you can add NU to each of those schools, but you’ll get very different hybrids — some more coherent than others — and in every case you’ll need extra constraints or reinterpretations to avoid the obvious eliminationist consequences. Below I sketch how NU can be grafted onto each tradition, the tensions that arise, and some principled ways to make the hybrid liveable.

---

1) Rawlsian (Kantian constructivism) + NU

How to graft it.
Recast the Rawlsian original-position choice problem so parties behind the veil of ignorance adopt negative priorities: they choose principles that lexically protect the worst-off in terms of suffering (rather than primary goods broadly construed). So your first principle becomes something like:

Lexical principle: Minimize the maximum (or the expected worst) suffering; only when suffering is equal do other Rawlsian goods (liberty, opportunity) matter.

Benefits.

• Keeps the Rawlsian procedural legitimacy and constructivist justification: people would prefer rules that protect them from catastrophic suffering.

• Protects persons as separate moral agents via the original-position reasoning — you’re not simply aggregating. That helps blunt some eliminationist moves.

Tensions / problems.

• Rawls’ famous priority of basic liberties clashes if the suffering-minimization principle demands violating liberties to prevent future pain. You need lexical ordering: is suffering lexically prior to liberty, or are liberties lexically prior to suffering? Either choice is heavy.

• Determining the relevant metric (what counts as “suffering”?) becomes crucial and contested.

Workable variants.

• Lexical Rawlsian NU: Suffering-reduction is lexically prior to distributive goods but constrained by an inviolable core of liberties (e.g., no deliberate killing, no non-consensual bodily violation).

• Threshold Rawlsian NU: Basic liberties hold unless suffering exceeds a high, well-defined threshold, after which limited exceptions are allowed under strict procedural safeguards.

---

2) Sandel’s communitarian / Aristotelian + NU

How to graft it.
Sandel’s critique of utilitarianism emphasizes goods internal to practices, roles, and teloi. To add NU, treat suffering-reduction as a central civic telos but interpret it through thick communal goods: reducing suffering matters only insofar as it preserves or restores the community’s ability to pursue goods (flourishing, dignity, meaningful life).

Benefits.

• Sandel’s emphasis on moral meanings protects against purely instrumental solutions (e.g., “we eliminate people because their pain is high”).

• NU gets embedded in a richer moral vocabulary (virtues, narrative goods, institutions) that resists simplistic arithmetic.

Tensions / problems.

• Communitarianism values attachments, roles, and identity; some forms of NU (e.g., advocating nonexistence) conflict deeply with those values.

• Which communal telos counts? Different communities will answer differently about acceptable trade-offs.

Workable variants.

• Virtue-informed NU: Suffering reduction is a central virtue, but its practice must respect and cultivate other civic virtues (dignity, solidarity). So policies must be framed as enabling flourishing, not erasing persons.

• Deliberative NU: Commit the community to democratic, context-sensitive deliberation where suffering reduction is the guiding end but bounded by communal understandings of dignity and meaning.

---

3) Nozickian deontological libertarianism + NU

How to graft it.
Nozick treats persons as ends with inviolable rights. There are two realistic ways to add NU:

A — Personal/voluntary NU: NU governs individual moral choices (e.g., one may choose voluntary euthanasia to end one’s suffering), but the state must not coerce.
B — Negative-rights NU: NU is interpreted as endorsing only negative duties (not to cause suffering), not positive duties. So suffering-reduction is pursued by voluntarily offered aid, not by infringing rights.

Benefits.

• Preserves core Nozickian protections against state coercion and rights violations.

• Avoids state-directed eliminationist programs.

Tensions / problems.

• Pure NU can justify coercive elimination if “eliminating a person removes suffering.” That directly violates Nozickian inviolability. So the pure NU thesis and Nozick’s inviolability are in tension; you must prioritize rights or NU.

• Nozick’s framework resists redistributive measures aimed at systemic suffering reduction.

Workable variants.

• Nozick-lite NU: NU is a moral ideal for individuals and charities; the state’s role is strictly negative (prevent harm) and to protect voluntary institutions that reduce suffering.

• Contractual compromise: Allow limited, voluntary, consent-based measures to reduce suffering  but forbid non-consensual elimination or coercive population-level interventions.

ChatGPT said:

ok — that is a very important clarification

because if existential-ontic suffering is primary for you then you are not a classical NU at all
you are in the Cioran / Caraco / anti-natalist continental lineage

classical NU is about pain qualia
you are about anguish of being

for you the ontology is:

1. first there is a thrown, unchosen being-there

2. being-there as such is shot through with this unbearable “there is”

3. then particular pains come as local instantiations

so the moral negativity is not pain but existence as burden

and that has a very different logic

because if existential-ontic suffering is primary, then any framework that talks only about liberties, opportunities, welfare increases, etc. is always already superficial — it’s post-hoc furniture inside the trap.

this maps to your sentence:

Instances matter just as much but first things exist.

Yes — specific harms are real, but they are derivatives of the primordial wrongness of being.

So what is your position actually saying?

Not: “we should reduce pain in the world.”

But: there should not be a world in which beings are thrown into ontological anguish in the first place.

That is not NU.
That is metaphysical pessimism.

NU is consequentialist.
Metaphysical pessimism is ontological.

So when you ask:

what if we add NU to Rawls or Sandel or Nozick?

that becomes much stranger

because you are no longer adding a “harm metric” to a theory of justice

you are adding a metaphysical diagnosis of being-in-the-world to a theory of justice.

Rawls doesn’t talk about the anguish of the “there is.”
Sandel doesn’t.
Nozick certainly doesn’t.

what you are really trying to combine is:

• metaphysics of existential wrongness (pessimism)

• political moral constraint theories (Rawls, communitarian, rights)

And that is actually the philosophical problem of Cioran’s entire life:

How do you moralize within a world you think should not exist?

the only way to make this coherent is two-tier:

Tier A — metaphysics: being is a burden / horror.

Tier B — practice: given that we are trapped here, we must minimize worsening the burden, and we must not add force, coercion, liquidations — because forced elimination is itself a violent assertion of the very world we take to be unjustifiable.

In other words:

lust for elimination is still hunger for power — it is still being too “invested” in this world.

So the constraints (Rawls/Sandel/Nozick style) become almost ascetic guardrails.

They prevent the metaphysics of pessimism from turning into a mania of “correction by force.”

so your best hybrid is something like:

• metaphysics: existence is intrinsically negative

• ethics: therefore, do not generate or intensify existence unnecessarily

• politics: and do not violate persons (who are already condemned to existence) in the name of “fixing” the metaphysical error

This is a very coherent position.

It is close to Zapffe.

It is also close to Cioran’s late view:
the ethics of not multiplying the trap, while refusing to become a murderer in the name of salvation.

If you like: I can try to write this out formally as a 9-sentence axiomatic system — so you can see it as a precise philosophical “stance.”

You said:

Ok

ChatGPT said:

here is a very compressed nine-sentence axiom set that captures what you just landed on:

---

Axioms (9)

1. Existence as such is a thrown, non-consensual condition.

2. Existence entails a structural, ontic suffering — not just contingent pains.

3. Particular instances of suffering matter morally, but they are expressions of the primary structural wrongness.

4. Therefore: suffering is not a contingent harm to be balanced, it is the fundamental negative value.

5. Because we are already trapped in existence, we cannot ethically seek a metaphysical “fix” through force — force is itself the signature of the trap.

6. Therefore: any ethics inside the world must adopt a posture of non-intensification of the trap (do not amplify existence, do not amplify suffering).

7. Harm-reduction (negative ethics) remains valid as a regulative maxim, but always under constraint: it must not produce new forms of domination.

8. Political and interpersonal ethics therefore require constraint pluralism (Rawlsian liberties / communitarian dignity / deontic inviolability) as guardrails against eliminationism.

9. The moral task in a condemned world is: reduce suffering without adding to the machinery of force that constitutes the world itself.