The humanist frame
The argument’s claims stand on their own evidence. The frame is how claims get held, and it is humanism — lowercase, a stance and not a movement: a life stance built out of evidence and reason and meant to be revised when the evidence moves. Not a creed with articles. A method, plus the nerve to live by it before everything is settled. It borrows from Paul Kurtz and the Humanist Manifestos where they hold up, and is not bound by them where they don’t.
That is the right frame for these arguments, because together they ask for a boundary to be redrawn. A stance that could not redraw its own boundary would have nothing to offer here. Three things it supplies directly. Its common decencies are owed rather than earned, which is the floor. Its scope has already been stretched once, from the nation to the planet, so widening is not foreign to it. And its refusal of anyone who claims authority over how another should live is the agency claim, stated as a political commitment.
What the frame does not do is generate the claims. Organized humanism does not entail any particular verdict on wages, property or firms, and its own record shows it: the 1933 Humanist Manifesto called for a cooperative economic order, and later humanists, Kurtz among them, backed away from that and left the question open. A stance that has argued both sides plainly does not settle it. So nothing here rests on humanism having reached these conclusions — the claims have to stand on their own arguments, and they do.
The fit runs the other way. The argument’s principles are ones this frame already holds: that no one may be ruled by another’s say-so, that decency is owed rather than earned, that claims answer to evidence. The case about wage work is a good example of how little exotic machinery it needs. It is not really an argument against markets. It is an argument that one particular contract claims to transfer something that does not move, and its ancestry is the same inalienable-rights reasoning humanism has always used against slavery and against rule by divine right. The argument is at home here. It simply was not carried in by the frame.
The problem has to be said out loud, because it is not small. The tradition’s classical ground is the autonomous rational chooser, and its founding document makes human dignity rest on the claim that humans alone are unfixed while other animals are stuck in their natures. That is the threshold dodge, in the text that named the tradition. So the frame carries the move it is being asked to break.
It breaks on its own terms, though, which is why the frame is worth keeping rather than swapping out. The boundary was drawn with the natural history available at the time and has not been redrawn since, and a stance whose whole claim is that beliefs answer to evidence does not get to hold one belief exempt. Either it revises here or it stops describing itself that way. This is not a concession humanism has generally made — naming it as the frame means arguing for it rather than inheriting agreement, and others working the same seam have taken a different name for it.
Keeping the name does one more thing. Humanism is the most decent version of the move: evidence-loving, anti-authoritarian, on the right side of nearly everything, and carrying it anyway. If the mechanism holds there, it is not a story about bad faith. Which is the whole point — the move does not need anyone to be cruel, only for the stopping-place to be comfortable and well-argued.
Robert Ingersoll, the great orator of American freethought, put the tradition’s best instinct in one line: “Give every other human being every right you claim for yourself.” It is about as generous as a doctrine gets, and the list is sitting in it — human being — doing the stopping without anyone noticing it had been drawn.
The rule: a rule of construction
An arrangement is not made legitimate by who the parties are. It is legitimate or not because of what is being done, and to whom.
Underneath all of them sits one question: is a subject being used as an input? Five checks help answer it, each from a different side. Anyone can run them, and you do not have to accept anything else here to run them.
Is the line drawn by structure, or by membership? The difference is when the question is settled. Membership is settled before anyone looks at the case; structure is settled by looking at it. A list asks whether a being is the kind of thing that counts. A structural line asks whether this being, in this arrangement, has something the arrangement can go against: states it moves toward or away from, a life going somewhere. Structure means nothing more here than what a being actually has. What it has may well put it in a group, but the group comes from looking, not in place of looking. Where the line falls matters less than how it was drawn. Drawn by structure, it moves when the evidence moves, and an unsettled edge is just an open question. Drawn by membership, it is a border, and borders get defended. Arguing about where the capacity to suffer begins costs this model nothing. Turning it into a list costs it its impartiality. That goes for the word sentient too. Used as a membership card, it would be the threshold dodge in new clothes. Here it only names what the looking is for, which is whether there is someone in there for whom things go better or worse.
The rule says legitimacy depends on what is being done and to whom, and that whom can look like the list coming back. It is not, because whom is settled by looking at this being in this arrangement, not by checking a list first. Take a hard case. Whether crabs, lobsters and octopuses can suffer is a question of evidence, and the evidence moved. In Britain it took a 2021 scientific review and then an act of Parliament, in 2022, to add them to the list of sentient beings, one group at a time. Drawn by structure, the line would have moved the day the evidence did, and while it was unsettled the burden would have sat on whoever was boiling the lobster, not on the lobster. An honestly unknown edge is a reason to err on the side of caution, not a gap to use.
In the settled middle, structure and membership look the same; a case that passes every time stops needing the test. But the drawers of any line are members first. Their own case is the pattern the criterion was lifted from, and their own place inside it is never in question. So the real test of a structural line is whether it can outrun the people who drew it: find the property in something unlike them, or press on something that merely looks like them. A line that can only ever find more of its authors was a membership line all along.
Does consent close the case? It does not. Consent fails in two ways. It can be claimed where it is absent: nobody signed the state into being, and everyone is treated as though they did. Or it can be present and empty: the worker really did sign, but what the contract claims to hand over never moves. What it claims to hand over is the worker’s acting, and that stays with the worker. Either way, a contractual agreement is asked to do work it cannot do. It tells you about the procedure, not about what the procedure produced for the people living under it.
Wesley Hohfeld’s vocabulary makes the second failure exact, and it also marks what can and cannot be claimed. In Hohfeld’s scheme, powers and disabilities are what a legal order grants or withholds. A power is the legal ability to change rights and duties, as by signing a contract; a disability is the lack of it. Our own legal order, for example, grants the power to sign an employment contract, one that claims to move the worker’s acting to the employer.
The argument cannot be that nature forbids the transfer; that would be the ventriloquism set aside above. It is that the law grants a power to do something the law itself does not believe happened. The contract says the worker’s acting passes to the employer. When the acting is a crime, the law holds the worker responsible alongside whoever gave the order, and never says the employer acted through a tool. A legal order that will not honor the transfer when it matters has granted a power it cannot make good on. That is the line between delegation and alienation. A delegate acts for you and answers to you, so the acting stays yours. An alienation claims to move the acting itself, and the law’s own account of responsibility says it does not move. This is where consent comes back in. A signature can hand over only what can be handed over, so the worker’s agreement settles nothing about the acting: the contract records a transfer that consent had no power to make.
The sharpest test is the contract that sells the whole person. Robert Nozick accepted it: in Anarchy, State, and Utopia(1974) he held that a free system would let someone sell himself into slavery. Not every libertarian followed him. Murray Rothbard, in The Ethics of Liberty (1982), refused, because a person’s will cannot be handed over: you can promise to obey, but the one obeying is still you, so the contract cannot be enforced. David Ellerman presses the obvious next question. If the will cannot be sold for a lifetime, as the self-sold slave sells it, how can it be sold by the hour? Nothing that voids the slave contract depends on its length. So the defender of the rental has two roads: follow Nozick to voluntary slavery, or follow Rothbard and explain why his reason stops at the factory gate. Kevin Carson has a name for the first road: vulgar libertarianism.
Even the strictest defenders of self-rule knew this. Josiah Warren made the sovereignty of the individual the ground of everything, and still would not let agreement settle a price: a bargain struck on what the buyer’s need and means would bear, rather than on what the thing cost to produce, he called civilized cannibalism — freely agreed, and still a taking of everything above cost. Taken seriously, consent is a test arrangements can fail, not a seal that certifies them as legitimate.
Is standing being earned? Standing here means counting in one’s own right. It cannot be earned. The moment standing depends on what a being contributes or can do, the move is back: the question of what is being done to her gives way again to whether she qualifies, only now the test is contribution instead of species. Dropping species does not help. A test based on responsibility ignores species perfectly well, and hands you the slaughterhouse.
Standing that is not earned is not a thought experiment; people have lived by it. Murray Bookchin, borrowing from the anthropologist Paul Radin, called it the irreducible minimum: in many stateless communities everyone was owed the means of life whatever work they did, and to withhold them was to deny the right to live. Where people tried to rebuild something like it for themselves, the state kept a hand on the tools. Working-class friendly societies were tolerated, and after the Friendly Societies Act of 1793 (Rose’s Act) they were registered and regulated, but watched closely, because their sick pay and unemployment funds could double as strike funds, a threat to employers’ power over wages. Next came the Combination Acts of 1799 and 1800, part of the repressive legislation of Pitt the Younger’s governments, which went further and outlawed workers’ combinations outright. Kevin Carson, drawing on E. P. Thompson, reads this as the state taking away the means people had built to keep one another alive, solvent, and with some power to bargain.
Who wrote the arrangement? If the ones living under it had no hand in writing it, they were inputs, whatever the terms say. The Combination Laws were written about workers by a Parliament workers could not vote for; farm-animal exemptions are written about livestock, with the industry deciding what counts as customary. In both cases the workers and the animals were inputs.
David Graeber questioned policy itself. A policy, he argued, assumes an elite or vanguard standing over the people it governs, one that knows better than those beholden to the policy how their affairs should be run. In turn, the most that joining a policy debate can achieve is to limit the damage. A theory worth having for people who want to run their own affairs starts instead from the conviction that another world is possible, and refuses any trace of vanguardism. Graeber’s objection is the same check aimed at the policymaker: whoever writes an arrangement for others, however well meant, has written it over their heads. It is one more reason this argument refuses the vanguard.
Can they leave? The default is to leave them be: no arrangement at all, no policy over them, each left to their own devices. It is an ought like any other, but it takes nothing from anyone, so it is the cheapest to defend, and every departure from it carries more to justify. A relationship can depart from it and still be sound, if it is mutual, and mutuality does not need a signature. It shows in what a being keeps choosing when leaving is open: the cat who comes and goes, the dog who seeks company, the crows who return to the neighbor who feeds them. The crate, and the policy that permits it, are the opposite case. They exist to remove the exit. The job that is kept because every alternative has been closed off is a milder version of the same thing: he can always find another is only true if he can. A relationship that holds only because leaving has been made impossible is not mutual, whatever else it is. A border is the same question at the scale of a country: a line that decides who may leave or arrive removes an exit, and carries the same burden as any other departure from leaving them be.
The hardest case for this check is the one everyone starts in. A child cannot leave, did not agree, and had no hand in writing the arrangement, and nobody calls a family a crate. The difference is direction. Every child begins as someone else’s plan, brought in without being asked, and the usual story turns that into a debt the child owes. It runs the other way. Seana Shiffrin argues that bringing someone into conditions they could not consent to leaves the one who did it responsible: the imposition is not a gift that calls for gratitude. David Benatar goes further and holds that it would be better never to have been brought in at all; the argument here needs only the weaker claim. That debt has a natural repayment, which is to raise someone who can leave. A family that holds a child only until the child no longer needs it is closing the gap between no exit and exit; the crate exists to keep that gap open. What remains when the child stays because they want to is the mutual kind. A family that keeps the child as an heir who cannot refuse, a carer who cannot go, or a debtor owing a lifetime has made the child an input, however loving the terms. The dependent animal was met earlier: where leaving them be is not possible, the ought is care without use. A child is the case where it becomes possible, and the care is measured by how well it gets there.
What the rule rejects as the answer
The rule of construction, that an arrangement is judged by what is being done and to whom, earns its keep by naming, ahead of time, the things that will show up looking like progress.
Asking a court to admit one being. This is the move in its purest modern form. Win, and the victory certifies the boundary: this one is in, the door still works, and every other being has to queue up and be judged on how closely it resembles whoever got in last. Resemblance is the wrong relation. It makes the line a question of how much like us something is, which is the premise the whole thing exists to refuse.
Sentience written into law as a grant. Granted means held at someone’s discretion, listed species by species, revisable by whoever drew up the list. Britain’s 2022 Act, met earlier, shows the pattern: crabs, lobsters and octopuses waited on a review and a vote, and were admitted one group at a time. Standing held on loan is not standing.
Humane instrumentality. The gentler means to ends, what animal ethics calls welfarism: better conditions, less suffering, careful handling. This is the real competitor, and a strong one — cheap, immediately better for animals alive right now, and it soaks up nearly all the moral energy going while changing nothing about the relation. It is the second failure with the kindness turned up. Naming it here does not mean the suffering it prevents is unimportant. It means you can reduce suffering a great deal and still be treating a subject as an input, and the input relation is what is at issue. This is the old dispute between welfare and abolition, and Gary Francione’s point settles it here: rules for how property may be used leave it property.
Better terms for the rental. David Ellerman calls employment the renting of persons: the firm does not buy the worker, it hires them by the hour, and what it hires is their agency. Higher wages, benefits, even a workers’ seat on the board all make the rental kinder, and all leave it a rental. This is humane instrumentality on the human side. The worker is still the one acting while someone else owns what the acting produces. The contract treats them as a some thing that earns, but the deeds stay with the someone who did them, and no rate of pay transfers ownership of the deeds.
None of these is rejected for being too moderate. Each is rejected for running the move.
Sources and further reading
- Giovanni Pico della Mirandola, Oration on the Dignity of Man (1486). The founding claim that humans alone are unfixed.
- Humanist Manifesto I (1933).
- Paul Kurtz, Eupraxophy: Living Without Religion (Prometheus Books, 1989).
- Robert G. Ingersoll, “The Liberty of Man, Woman and Child” (lecture, 1877). Source of “Give every other human being every right you claim for yourself.”
- Wesley Newcomb Hohfeld, “Some Fundamental Legal Conceptions as Applied in Judicial Reasoning,” Yale Law Journal 23 (1913). Powers, disabilities, and why no signature can transfer what nobody has the power to transfer.
- Voluntary slavery: Robert Nozick, Anarchy, State, and Utopia (Basic Books, 1974), p. 331; Murray N. Rothbard, The Ethics of Liberty (Humanities Press, 1982), ch. 19, on the inalienability of the will; Kevin Carson, whose “vulgar libertarianism” posts on his Mutualist Blog (from the mid-2000s) gave the term its currency.
- Josiah Warren, Equitable Commerce (1846; expanded 1852). Cost the limit of price, and pricing by what need will bear as “civilized cannibalism.”
- Paul Radin, The World of Primitive Man (1953), and Murray Bookchin, The Ecology of Freedom (1982). The irreducible minimum.
- Kevin Carson, “The ‘Progressive’ Welfare State Fantasy,” Center for a Stateless Society (March 20, 2014). On the suppression of working-class friendly societies and mutual aid, and the Combination Laws.
- David Graeber, Fragments of an Anarchist Anthropology (Prickly Paradigm Press, 2004), pp. 9–10. Policy as the negation of politics.
- Seana Valentine Shiffrin, “Wrongful Life, Procreative Responsibility, and the Significance of Harm,” Legal Theory 5 (1999): 117–148. Procreation as an unconsented imposition that leaves the parent responsible, not the child indebted.
- David Benatar, Better Never to Have Been: The Harm of Coming into Existence (Oxford: Clarendon Press, 2006). Coming into existence as always a harm; the antinatalist end of the same line of argument.
- Gary L. Francione, Animals, Property, and the Law (Temple University Press, 1995). The source of the argument that sentient beings cannot coherently be property.
- Jonathan Birch et al., Review of the Evidence of Sentience in Cephalopod Molluscs and Decapod Crustaceans (LSE, 2021); Animal Welfare (Sentience) Act 2022 (UK), which extended recognition to those groups on the strength of it.