For the Animals
An Argument Guide
PT

Home / A. Sentience and moral status / Part DIR

A7

"No duties, no rights."

The answer in one sentence

rights protect those who can be harmed, not those who can fulfill obligations. Confusing the two strips protection from babies, people with profound cognitive disabilities and anyone unable to reciprocate. And even those who accept the premise don't reach the conclusion: lacking rights, in the strict legal sense, doesn't authorize causing suffering.

Short answer

The argument swaps two different things. Moral agents are those who can have duties; moral patients are those who can be harmed. Babies, people with advanced dementia and people in a coma are patients without being agents, and they don't lose protection for that. Our whole legal system works this way, with guardians representing those who can't claim anything. Besides, even those who deny animals rights, like Carl Cohen, admit we have duties regarding them.

For children

A newborn baby doesn't know what a rule is, doesn't help anyone and doesn't return favors. Even so, everyone agrees no one may hurt a baby. That shows something important: being protected isn't a prize you earn by fulfilling obligations. It's what is owed to those who can be hurt. Animals work the same way.

Ask a question back

"Does a baby have duties? Then why does it have rights?"

Como pode aparecer

"Animals don't pay taxes or vote."
Everyday conversation
"Rights come with duties."
Everyday conversation
"Would a lion respect your rights?"
Everyday conversation
"When the cow signs the constitution, we'll talk."
Social media
"Are you going to prosecute a shark for murder?"
Rhetorical question
"Rights exist only among beings able to make moral claims on one another." (Cohen)
Academic version
"Morality is an agreement among those who can understand and follow rules."
Contractualist version
"They aren't part of our moral community."
Disguised version

The strongest version

How it is said

"Animals have no duties, don't pay taxes, don't obey any law. So they can't have rights."

Strongest version (steelman). The classic formulation is Carl Cohen's, in a 1986 article in the New England Journal of Medicine defending the use of animals in research:

  1. P1Rights arise, and can only be defended, among beings who make, or can make, moral claims on one another.
  2. P2Animals lack the capacity to understand and apply moral rules.
  3. C1So animals have no rights, and in Cohen's words, "they can have none".
  4. P3Those without rights may be used, subject to any duties of kindness.
  5. C2So using animals is permitted.

Cohen anticipates the objection from humans lacking this capacity and replies that the capacity for moral judgment "is not a test to be administered to human beings one by one", because "the issue is one of kind". That way out is examined in chapter A11.

The same structure appears in contractualism: morality would be an agreement among rational agents, and only parties to the agreement would be protected by it. It's worth noting that John Rawls, although he excluded animals from the domain of justice, held that we have duties of compassion and humanity toward them.

The argument is valid. Its soundness fails at P1, which confuses who makes rules with whom the rules protect, at P2, which is partly false, and at P3, which doesn't follow: lacking legal rights isn't the same as moral permission.

Full answer

  1. a

    Reconstruction and diagnosis

    The argument lives on a conceptual confusion that moral philosophy resolved long ago.

    RoleWho it isWhat it requiresExamples
    Moral agentSomeone who can have duties and be held responsibleUnderstanding rules, deliberating, choosingTypical adults
    Moral patientSomeone who can be harmed or benefitedBeing sentientBabies, people with dementia, people in a coma, animals

    Every moral agent is also a patient, but not every patient is an agent. Argument A7 demands of the patient the credentials of the agent. It's like refusing medical care to anyone who isn't a doctor.

    There's also a leap between C1 and C2. Even if animals had no rights in the strict legal sense, it doesn't follow that we may do whatever we like with them. A consequentialist ethics doesn't even need the language of rights: avoidable suffering is enough to give a moral reason against it.

  2. b

    Where the argument breaks

    1. It proves too much.

    Newborns, people with profound intellectual disabilities and people in a coma have no duties and can't make claims. By P1's criterion, they'd have no rights. Almost no one accepts that, which is enough to reject the premise.

    2. The law itself works this way.

    Guardianship exists precisely to represent those who can't make claims. Future generations, who can't reciprocate anything or take part in any agreement, appear in constitutions and environmental treaties as owed consideration. Inability to complain has never been an obstacle to legal protection.

    3. The "it's a matter of kind" way out is circular.

    Cohen answers the human cases by saying moral capacity characterizes the species, not each individual. But that means judging someone by what other members of their species have, which is exactly the point in dispute. If we applied the same reasoning to intelligence, we'd say a baby can do algebra because adult humans can. Chapter A11 deals with this in detail.

    4. P2 is empirically questionable.

    Animals cooperate, return favors and punish cheating. Vampire bats share food with those who have fed them before, in a pattern of reciprocity documented by Wilkinson (1984). This doesn't make reciprocity a moral criterion, and the book doesn't want to fall into that trap. It only shows that the premise describes animals poorly.

    5. Even the critics admit duties.

    Rawls excluded animals from the domain of justice, but affirmed duties of compassion and humanity toward them. Cohen defends their use in research and still recognizes human obligations of kindness. Conclusion C2, "so we may use them as we please", isn't supported even by the argument's own authors.

    6. The law has already begun to move.

    In January 2022, Ecuador's Constitutional Court ruled, in the case of the monkey Estrellita, that individual animals are subjects of rights protected by the rights of nature, and ordered the state to produce specific legislation. In the United Kingdom, the sentience law came to recognize cephalopods and decapods. None of this required animals to take on duties.

    7. Those who receive care needn't be able to give it.

    Requiring that someone be capable of acting morally in order to be considered is like requiring a patient to be a doctor in order to be treated. The beneficiaries of an activity needn't be able to perform it. See Questão de Ética, vol. I, ch. 10.

  3. c

    Data and sources

    SourceWhat it saysNote
    Cohen, 1986, *NEJM*Rights arise, and can be intelligibly defended, only among beings who actually do, or can, make moral claims against one another; animals "therefore have no rights, and they can have none"The most cited formulation of the argument, accompanied by a defense of animal research
    Cohen, 1986The capacity for moral judgment that distinguishes humans from animals "is not a test to be administered to human beings one by one"; "the issue is one of kind"A reply to species overlap, examined in A11
    Rawls, A Theory of Justice (1971)Excludes animals from the domain of justice, but affirms duties of compassion and humanity toward themEven classical contractualism doesn't authorize cruelty
    Constitutional Court of Ecuador, Estrellita case, 27 Jan. 2022Individual animals are subjects of rights protected by the rights of nature; the state was instructed to create specific legislationThe monkey died in state custody before the final decision, as the case itself records
    Wilkinson, 1984, *Nature*Vampire bats share food reciprocally, favoring those who have fed themShows P2 describes animals poorly, without making reciprocity a moral criterion
    Nussbaum, Frontiers of Justice (2006) and Justice for Animals (2023)Criticizes contractualism for leaving out both animals and people with disabilities, and proposes a capabilities approachChecked in the Harvard University Press catalog (2006)
  4. d

    The strongest counter-reply: the moral community

    The strongest version doesn't say animal suffering is irrelevant. It says rights are a specific instrument, born of reciprocal relationships within a moral community, and that extending them to those who can't take part in that community would be a conceptual mistake. Incapable humans would be included by belonging to the kind, and because excluding them would threaten the stability of the community itself. Animals would remain protected by indirect duties of kindness.

  5. e

    Why the counter-reply doesn't reach the conclusion

    It makes the protection of the vulnerable contingent.

    If a baby is protected because it suits social stability, then in a scenario where that convenience vanished, the protection would fall with it. Moral intuition says the opposite: the baby is protected for its own sake.

    Inclusion "by kind" is the species boundary under another name.

    Asked why the kind matters more than the individual, the answer comes back to "because they're human", which is A10.

    The contract's own device points the other way.

    Mark Rowlands showed that if the veil of ignorance also covered species, the contractors would choose rules protecting all sentient beings.

    Indirect duties would already be enough to condemn the practice.

    If we have duties not to cause unnecessary suffering, industrial livestock farming, with confinement, mutilation and slaughter at scale, doesn't pass even that minimal filter.

    Arguing about the word doesn't change the fact.

    We can debate whether the term "rights" fits. What doesn't change is that someone is being harmed, and that's the relevant moral reason.

Reasoning errors

Logical

Is the form valid?

Category confusion
Treats moral agent and moral patient as the same thing
Proving too much
Excludes babies, people in a coma and people with profound cognitive disabilities
Begging the question
The "moral community" is defined so as to include exactly those one wants to include
Non sequitur
From "has no rights" it concludes "may be used"

Epistemological

Does the evidence support it?

Fragile empirical premise
Ignores cooperation and reciprocity observed in many species
Legalistic fallacy
Confuses what the law recognizes today with what is morally owed

Moral

Is the criterion coherent?

Reciprocity as a condition
Makes the ability to reciprocate the price of protection
Power as legitimacy
In practice, protects those who can demand and leaves unprotected those who can't
Contingent protection
Makes the safety of the vulnerable depend on the convenience of the strong

Biases

Reciprocity heuristic

Our intuitive morality is very sensitive to fair exchange, which makes "rights require duties" sound obvious

Literature on reciprocity as a basic social norm

Social dominance orientation

A preference for hierarchy makes it seem natural to protect only those with the power to demand

Caviola, Everett and Faber, 2019

System justification

The current legal arrangement comes to be read as the correct one

Literature on system justification

Legalistic bias

Confuses "isn't a recognized right" with "shouldn't be"

A common pattern in debates about extending rights

In-group bias

Defines the moral community by one's own group and then uses that definition as an argument

A common basis of speciesism and prejudice among humans

Extrapolations

The principle being tested is: "whoever can't have duties or make claims has no rights".

If it applied to…Babies and small childrenTap to see

They'd lose protection until they learned rules. A newborn doesn't understand norms, takes on no obligations and can't claim anything; by the criterion, they wouldn't hold rights

No legal system in the world accepts this, and Cohen himself has to create an exception to avoid it

If it applied to…People with profound cognitive disabilitiesTap to see

They'd be left out of the moral community. They'll never be able to fulfill duties or take part in agreements

Much of disability rights law developed precisely against this kind of exclusion

If it applied to…People in a comaTap to see

They'd lose rights while unconscious. They make no claims, fulfill no duties, take part in nothing

The law does the opposite: it increases protection and appoints representatives

If it applied to…Future generationsTap to see

We'd have no obligation to them at all. Those who don't exist yet can't reciprocate or demand anything

All climate policy presupposes the opposite

If it applied to…Peoples without the power to retaliateTap to see

They could be dominated. If protection arises from the ability to demand, those who can't demand are left unprotected

This is the historical description of colonialism, and the reason we came to speak of universal human rights

Contradictions

Defends the rights of those who have no duties.

Babies, people in a coma and people with profound cognitive disabilities remain protected, without anyone demanding reciprocity.

Accepts anti-cruelty laws.

Punishing someone who beats a dog only makes sense if something is owed to the animal, not just to its owner.

Recognizes duties to future generations.

People who don't exist yet sign no contract and still appear in environmental policy.

Demands duties only of those they want to exclude.

No one asks a foreign tourist to have duties before recognizing they mustn't be assaulted.

Denies rights and affirms obligations.

It's common for the same person to say "animals have no rights" and then "but you mustn't mistreat them". The second sentence already concedes the essential point.

What the argument gets right

What is true

There's an important truth embedded here. Animals aren't moral agents: they can't be held responsible, tried or punished, and it would be absurd to prosecute a lion for killing a gazelle. Reciprocity also explains well the historical origin of many norms, and the word "right", in the strict legal sense, involves institutions, representation and procedures that need careful design when the holder can't speak for themselves.

But

But none of this supports the conclusion. The asymmetry between us and them is exactly why the responsibility is ours. Only one side of this relationship can deliberate about what it does, and that side is us.

Connections

Usually comes before

A1 Animals aren't rational.Rationality would be the condition for entering the contract
A3 Animals don't have language.Without language no agreement would be possible
A9 Rights are a human invention, only for humans.Prepares the idea of rights as a closed club
A23 Morality exists to protect us from each otherExclusion from the contract becomes a lack of rights
A7this argument

Usually comes after

A11 It's normal for humans to be rational.Cohen's way out for humans without moral capacity
A13 It's natural to put our own first.Swaps contract for emotional ties
A17 Animals are property.From "no rights" to the status of a thing
C2 Lions eat other animals.Reciprocity turned against wild animals
F3 So should we stop lions from hunting?Confusing an agent's duty with a patient's protection

Go deeper in Questão de Ética

Chapters from the open-access collection A Journey Through Animal Ethics that discuss this argument in more detail:

References

Provisional author-date format. The links indicate the sources checked in writing this chapter.

Philosophy

Law

Science

Psychology