Feminist Theory

In September 1791, Charles Maurice de Talleyrand-Perigord laid a report on public instruction before the French Constituent Assembly. It proposed a national system of schools, free and open, and it proposed that girls leave those schools at the age of eight and be taught at home thereafter. Four months later a London publisher issued a book of some three hundred pages, written in roughly six weeks, whose dedication was addressed to Talleyrand by name and asked him to think again.

The book was A Vindication of the Rights of Woman, and its author, Mary Wollstonecraft, was thirty-two and living on what she earned by her pen. She had already published a reply to Edmund Burke. Now she took the argument that the Revolution had made about kings and applied it to husbands: that a claim to rule which rests on nothing but custom and force is not a claim at all.

The move that does all the work

Wollstonecraft's opponents had evidence. Look at the women of the propertied classes, they said. They are vain, they are ignorant of the world, they are skilled at flattery and manipulation and at little else, they faint, they cannot follow an argument for ten minutes together. Jean-Jacques Rousseau had built this into a theory in Book V of Emile, where Sophie is educated for one purpose, to please Emile, and where Rousseau treats her resulting character as an expression of nature.

Wollstonecraft does not deny the observations. She spends whole chapters agreeing with them, and her contempt for the accomplished woman of fashion is sharper than Rousseau's. What she denies is the inference. Those traits, she argues, are exactly what you would predict from the education those women received: taught ornament instead of reasoning, kept from any serious occupation, told that their whole prospect in life depends on being found attractive by one man. Of women so trained she writes that, taught from infancy that beauty is woman's sceptre, the mind shapes itself to the body and, roaming round its gilt cage, only seeks to adorn its prison.

Set out formally, the argument has four steps. First, virtue must rest on understanding, because an action done from habit or fear is not virtuous. Second, the faculty of reason is not sexed; there is no evidence for two kinds of understanding. Third, therefore there is one standard of virtue and one appropriate education, and a system that trains women in manners while training men in judgement is not making women virtuous but making them useful. Fourth, the character now on display is the output of that system and cannot be used as evidence about the input.

Key idea: The traits offered as proof of women's incapacity are the predicted result of the conditions imposed on them, so they cannot also serve as the justification for those conditions. That is a circle, and naming it is Wollstonecraft's permanent contribution.

Notice what she does not claim. She does not claim that men and women are identical; she grants without argument that men are on the whole physically stronger, and she says she does not wish to see the point disputed. She does not even claim to have proved intellectual equality. Her position is that the case has never been tested, and that even if some difference in capacity were eventually shown, the argument for educating a woman's reason would stand, because reason is what makes her a moral agent rather than a pet.

What she was actually asking for

The Vindication asks for national coeducational day schools, for women to be trained for professions including medicine and business, for property and legal standing in marriage, and, in one sentence that she declines to develop, for representation in Parliament. Her model is a wife and mother who is rational, useful, and capable of supporting herself if she has to. She is arguing about the middle ranks, and she says so; the very rich she considers corrupt, and the labouring poor appear in her book mainly as an illustration.

The reception was better than the later legend suggests. The book sold, went into a second edition within the year, and was translated into French and German almost at once. What destroyed its standing was not refutation. In 1798, a year after Wollstonecraft died of septicaemia following childbirth, her widower William Godwin published a memoir that described her love affairs, her illegitimate daughter and her suicide attempts with a candour he thought honest. Reviewers used it to bury the argument along with the author, and for most of the nineteenth century the Vindication was a book that respectable people did not admit to having read.

Mill, and the problem of evidence

The Subjection of Women was written in 1861 and held back until 1869. John Stuart Mill said in his Autobiography that it was, like much of his later work, a joint product with his wife Harriet Taylor Mill, who had died in 1858, and that his stepdaughter Helen Taylor had contributed to it as well. How the labour divided is still argued over. Taylor's own 1851 essay The Enfranchisement of Women is more radical than Mill's book on the question of paid work, which is one reason the argument continues.

Mill opens with a sentence about the legal subordination of one sex to the other being wrong in itself and one of the chief hindrances to human improvement, and then does something more interesting than assert it. He anticipates the reply that women are as they are by nature, and he attacks the evidential basis of that reply directly. What is now called the nature of women, he writes, is an eminently artificial thing. We have never seen a woman except in a condition of dependence; the only experiment that could settle the question, a society in which women were free, has not been run. So the appeal to nature is not a finding. It is a guess dressed as a finding.

He adds a second argument that is easy to miss and hard to answer. If women were genuinely incapable of the work from which law excludes them, the law would be pointless. Nobody legislates to stop people doing what they cannot do. The existence of the prohibition is therefore evidence that somebody expected women to compete, and to compete successfully enough to be worth barring.

The point: Mill converts a question about female nature into a question about the evidence available for any claim about female nature, and shows that under conditions of legal dependence the evidence is worthless in both directions.

The third argument is about what the arrangement does to men. The family as then constituted, Mill says, is a school of despotism, in which the virtues of despotism but also its vices are nourished; a boy grows up believing that he is by birth the superior of half the human race and carries that belief into every other relation. A family justly constituted would instead be the school of the virtues of freedom. This is a claim about character formation, and it is the reason Mill treats marriage law rather than the vote as the deepest question.

He did also treat the vote. In 1866 he presented to the Commons a petition for women's suffrage gathered by Barbara Bodichon and Emily Davies. In May 1867 he moved an amendment to the Representation of the People Bill to substitute the word person for the word man. It was defeated, 196 to 73, which is a larger minority than anyone had expected.

Where the liberal argument runs into trouble

Every position in this course is presented with its best objection, and this one has three that have stuck.

The first comes from Mill's own text. In the second chapter he says that when a woman marries she may in general be understood to have chosen the management of a household and the bringing up of a family as the first call upon her exertions, in the way a man chooses a profession. He means it as a defence of choice. But it concedes the sexual division of labour at the moment it is most in question, and it makes the household a place where the equality he has just argued for is discharged by consent rather than practised. Chapter two of Wollstonecraft has a version of the same problem: her rational woman is still, first, a wife and mother.

The second is structural. The liberal argument works by extending an existing category, the rights-bearing individual, to a group previously excluded. That is why it has been so effective in law. It is also why it has trouble with anything that is not a formal exclusion. When the barrier is a norm about who leaves work at three o'clock, or an economy that treats forty hours of unpaid domestic labour as a private arrangement, there is no statute to strike down. The next lesson is about exactly this gap.

The third is about the standard. Formal equality asks that women be treated the same as men. Later critics, MacKinnon most sharply, ask why the male life pattern is the measure, and point out that a rule of identical treatment for people in unequal positions can preserve the inequality rather than remove it.

Why the tradition is not finished

It would be easy to read the rest of this course as a series of demolitions of the liberal position. That reading does not survive contact with the record. The measurable gains of the last two centuries, married women's property acts, the franchise, access to universities and the professions, equal pay statutes, antidiscrimination law, were won in this idiom and are defended in it now.

Liberal feminism also has live theoretical descendants. Martha Nussbaum, in Women and Human Development (2000), replaces the abstract rights-bearer with a list of central human capabilities, things a person must actually be able to do and be, and argues that a government can be assessed by whether each of its citizens reaches a threshold on every item. That is recognisably in the Wollstonecraft line, and it answers the objection that liberalism ignores bodies and dependency by putting bodily health, bodily integrity and affiliation on the list.

Common misconceptions

• Wollstonecraft argued that women and men are the same. She argued that reason is not sexed and that virtue has one standard. She granted a difference in physical strength and did not claim to have demonstrated equality of intellect, only that the question had never been fairly asked.
• Liberal feminism means moderate feminism. The word names an argumentative structure, individual rights, formal equality, and a state that removes barriers, not a temperature. Wollstonecraft was thought a dangerous radical, and Mill's 1867 amendment was mocked in the Commons.
• Mill wrote The Subjection of Women alone. He credited Harriet Taylor Mill and Helen Taylor. Taylor's 1851 essay goes further than the book on paid employment, and the attribution question is still open.
• The Vindication was ignored when it appeared. It sold and was translated quickly. Its reputation collapsed six years later, for reasons that had nothing to do with its arguments.

What to carry forward

Wollstonecraft's contribution is a method: when a group's observed character is offered as the reason for its treatment, ask what that treatment predicts about character, and refuse to let an effect be counted as a cause. Mill's contribution is the epistemic version of the same point, plus the observation that a prohibition is evidence against the incapacity it claims to reflect, plus the argument that unequal families teach unequal citizens. Both stop short of the household, and both take the existing shape of public life as the destination rather than as something also in need of explanation.

Worth holding on to: the liberal argument is strongest where the barrier is a rule and weakest where it is a norm, and almost every later school in this course can be read as an attempt to reach what a rule-focused argument cannot.

Sources

1. Wollstonecraft, M. (1792). A Vindication of the Rights of Woman. J. Johnson. gutenberg.org
2. Mill, J. S. (1869). The Subjection of Women. Longmans, Green, Reader, and Dyer. gutenberg.org
3. Baehr, A. R. (2021). Liberal feminism. Stanford Encyclopedia of Philosophy. plato.stanford.edu
4. Tomaselli, S. (2021). Mary Wollstonecraft. Stanford Encyclopedia of Philosophy. plato.stanford.edu
5. Nussbaum, M. C. (2000). Women and Human Development: The Capabilities Approach. Cambridge University Press.

Until 25 October 1991, a husband in England and Wales could not be convicted of raping his wife. The rule was not in a statute. It came from a sentence in Sir Matthew Hale's History of the Pleas of the Crown, published in 1736, sixty years after his death, which said that by their mutual matrimonial consent the wife has given herself in this kind to her husband, which she cannot retract. On that day the House of Lords decided R v R and held that the exemption no longer formed part of English law.

Now try the tools from the last lesson on that rule. Wollstonecraft's method asks whether an observed trait is the product of the conditions imposed. Mill's method asks why a prohibition exists at all. Neither quite bites here, because this is not a rule that shut women out of anything public. It is a rule about where the law declined to go. And a legal disability you can point at is exactly what the liberal argument is built to remove.

That is the problem this lesson works. When the barrier is not an exclusion but a boundary, what does a feminist argument look like?

Who drew the line

Carole Pateman put the question in its most compressed form in a 1983 essay: the dichotomy between public and private is, she wrote, ultimately what the feminist movement is about. Her book The Sexual Contract (1988) makes the historical case.

The classic contract theorists tell a story in which free and equal individuals leave the state of nature by agreeing to be governed. Pateman reads the same texts and finds a second agreement embedded in the first. The individuals who contract are heads of households. What they are agreeing about, among other things, is orderly access to women, and the marriage contract is where that agreement is executed. Civil freedom in the public sphere, she argues, is not undermined by the private sphere but constituted by it: the fraternal pact replaces the rule of fathers with the rule of brothers, and it is precisely because a man's authority at home is defined as natural rather than political that his equality with other men in public can be declared complete.

This is a strong claim and it has strong critics, including feminists who think Pateman reads a systematic pact into texts that are merely careless about women. Take the weaker version and it is still hard to dislodge: the boundary between political and non-political is itself a political settlement, and it was not negotiated by the people it placed on the far side of it.

The upshot: asking to be admitted to the public sphere leaves untouched the question of how the public sphere was demarcated, and by whom.

Okin, Rawls, and the household that justice forgot

Susan Moller Okin was a liberal, and her Justice, Gender, and the Family (1989) is the sharpest internal critique the tradition has produced. Her target is A Theory of Justice (1971), the most influential work of liberal political philosophy in the century.

John Rawls lists the monogamous family as part of the basic structure of society, the set of major institutions to which his principles of justice apply. Okin's charge is not that he left it out. It is that having named it, he never applies the principles to it, and that he could not do so without upsetting the rest of the theory. Parties behind the veil of ignorance are described as heads of families, so they do not consider the interests of everyone inside. And Rawls treats the family as the first school of moral development, the place where a sense of justice is acquired. Okin's question is blunt: how does a child learn a sense of justice in an institution to which the principles of justice have never been applied?

The empirical spine of her book is a cycle she calls vulnerability by marriage, and it is worth walking through because it explains why formal equality can be complete and the outcome still lopsided.

• Before any particular marriage, both parties know the pattern: it is usually the wife whose paid work bends around children.
• Anticipating that, women invest less in the training and job tracks that punish interruption, and employers price the expectation in.
• Inside the marriage the wife earns less, so it is rational for the household to prioritise the husband's job, which widens the gap.
• A wife with lower earnings and a broken record has weaker exit options, and weaker exit options mean weaker bargaining power over everything, including how the housework is split and where the family lives.
• If the marriage ends, her standard of living falls further than his, which is visible in divorce statistics, and the expectation that this will happen feeds back into the first step.

No stage of that cycle requires a discriminatory law or a hostile employer. Okin's remedies are legal and liberal: an equal entitlement of both spouses to the earnings of either during the marriage, so that the wage is treated as jointly produced, and workplace and school arrangements that stop assuming an adult at home in the afternoon.

The slogan, and what it actually said

In February 1969, Carol Hanisch wrote a short paper defending consciousness-raising groups against members of the women's movement who thought them therapy rather than politics. Her argument was that women were being told their difficulties were personal failures with personal solutions, and that when a hundred women in a room describe the same difficulty, the personal solution is the wrong frame. When the piece was published in 1970 the editors gave it a title: The Personal Is Political. Hanisch has since pointed out that she did not choose it.

The slogan is routinely read as a demand that nothing be private. That is not what the paper argues. It argues that recurring patterns inside private life have causes outside it, and that a method which pools private experience can find those causes. Consciousness raising was not confession. It was an attempt at evidence gathering by a movement that had no survey data.

MacKinnon: non-intervention is an intervention

The hardest version of the critique comes from Catharine MacKinnon, whose Toward a Feminist Theory of the State (1989) treats the constitutional right to privacy as a problem rather than a resource. Her argument runs in three moves.

First, a right to be let alone is worth most to whoever is already dominant in the space being left alone. Second, when a court says the state may not enter, it is not withdrawing; it is deciding whose account of what happens there will stand, and the answer is usually the account of the party with more power. Third, a right framed as non-interference generates no obligation to supply anything, and the difference shows up fast in money.

The case she points to is Harris v. McRae (1980), which upheld the Hyde Amendment's ban on Medicaid funding for most abortions. The Court reasoned that although the government may not place obstacles in the path of a woman's choice, it need not remove obstacles it did not create, including poverty. A right constructed as freedom from state interference had delivered exactly what its structure predicted.

Why this matters: the claim is not that the state should be everywhere. It is that the state is already everywhere in the household, defining marriage, property, custody, inheritance and consent, so the choice is never between intervention and none, only between one settlement and another.

The reply, and the complication

The liberal answer is not weak, and it has two parts.

The first is that the critique proves the line is movable, not that it is illegitimate. Hale's rule fell in England in 1991; American states abolished their marital exemptions across the 1970s and 1980s; the Violence Against Women Act of 1994 put federal money behind prosecution and shelters. Each of those was a redrawing, achieved with liberal tools, and Okin herself is the proof that you can make the criticism from inside the tradition.

The second is that a protected sphere is not a luxury of the powerful. Lawrence v. Texas (2003) struck down criminal sodomy laws on privacy grounds; access to contraception in the United States arrived the same way. A theory that dissolves the private has to say what protects dissenters, minorities and unpopular households from a state that will not always be sympathetic.

Then the complication, which comes from a tradition you will meet in Module 4. If the remedy for private harm is to send in the police, the remedy is distributed by the same institutions that are already unequal. Black feminist writers, and later Kimberle Crenshaw in her 1991 work on mandatory arrest policies, pointed out that women whose partners are heavily policed, whose immigration status is precarious, or who depend on the household's only income, may face a real cost for calling the state in, and may therefore not call. That is not an argument against entering the private sphere. It is an argument that entering it is not a single act with a single distribution of consequences.

Common misconceptions

• The personal is political means there should be no private life. Hanisch was answering the claim that women's difficulties were individual failures. The argument is about the causes of private patterns, not about abolishing privacy.
• Rawls forgot about the family. He listed it in the basic structure. Okin's objection is that he named it and then did not apply the two principles inside it, while relying on it to produce citizens with a sense of justice.
• The critique of privacy is a demand for more policing. MacKinnon's target is a legal doctrine that treats non-intervention as neutral. What should replace it is contested, and part of the contest comes from feminists who distrust carceral remedies.
• The marital rape exemption was a statute. In England and Wales it was a common law understanding traced to a treatise, which is why a court could end it without legislation.

Where this leaves us

A feminism that works by removing formal exclusions is powerful against rules and nearly silent about boundaries. Pateman argues that the boundary is itself the settlement; Okin shows that a household exempt from justice will reproduce inequality without anyone breaking a law; MacKinnon argues that the state's absence is a decision about whose version of events counts. The liberal reply is that lines can be redrawn and that some protected sphere is necessary, and the record of the last fifty years supports both halves of that reply.

In short: once you see that the private is defined rather than found, the question stops being whether the state should be involved and becomes which arrangement it should enforce, and for whom.

Sources

1. Brake, E., Ferguson, A., and Toop, C. (2024). Feminist perspectives on the family. Stanford Encyclopedia of Philosophy. plato.stanford.edu
2. Wikipedia contributors. (n.d.). R v R. Wikipedia. en.wikipedia.org
3. Okin, S. M. (1989). Justice, Gender, and the Family. Basic Books.
4. Pateman, C. (1988). The Sexual Contract. Stanford University Press.
5. MacKinnon, C. A. (1989). Toward a Feminist Theory of the State, chapters 9 and 10. Harvard University Press.

Gallimard published the second volume of The Second Sex in November 1949. The first volume had sold twenty-two thousand copies in its first week. In 1956 the Vatican put the book on the Index of Prohibited Books, and Simone de Beauvoir later said the reaction from French intellectuals of her own circle was worse than the reaction from Rome.

Volume II opens with a short French sentence that has become the most quoted line in feminist theory. H. M. Parshley, a retired professor of zoology who produced the only English version available from 1953 until 2009, rendered it: one is not born, but rather becomes, a woman. Constance Borde and Sheila Malovany-Chevallier, translating the complete text in 2009, dropped the article: one is not born, but rather becomes, woman. The French has no article there, and the difference is not pedantry. Parshley's line sounds like a person acquiring an identity. Without the article it sounds like a person being placed inside a category, which is nearer to what the following six hundred pages argue.

That is worth knowing for a second reason. Parshley cut roughly a tenth of the book, including large stretches of the material on women's history and many of the individual women Beauvoir cites, and he had no philosophical training, so he flattened technical existentialist vocabulary into ordinary English. Toril Moi documented the damage in 2002. If you have read The Second Sex in an old paperback, you have read an abridgement made by someone who did not know what the words were doing.

Step one: what kind of thing a person is

This lesson walks one argument from its premises to its conclusion, then runs the same materials through a different premise to show where the conclusion comes from.

Beauvoir's starting point is existentialist. A human being has no fixed nature that precedes existence; a person is what they make of what has been made of them. Two terms carry the argument. Transcendence is the movement of projecting oneself toward ends one has chosen, and it is where Beauvoir locates a fully human life. Immanence is life spent in repetition and maintenance: the meal cooked so that it can be eaten and cooked again, the floor swept so that it can be swept again, activity that sustains life without opening anything.

Nothing in that pair is about sex yet. Beauvoir's claim is that a whole civilisation has arranged for one half of humanity to be offered immanence as a destiny.

Step two: the Other, and why this Other is unusual

In the introduction she sets out the asymmetry. A man does not think of himself as a member of a sex; he is simply a human being writing about human beings. A woman is defined in relation to him. He is the Subject and the Absolute; she is the Other. Beauvoir's example is the pair of words on a form: nobody writes that a man has a male body and then reasons from it.

She then asks why this Other has not revolted, and her answer is what distinguishes her account from a straightforward analogy with class or with the situation of Black Americans, both of which she uses. Proletarians were not always proletarians; there was a before, and there is a shared workplace and a shared quarter. Women have no historical beginning to point to, no territory, no separate community. They are distributed one or two at a time among men, tied to fathers and husbands by residence, affection, economic dependence and children. So the word we, which every other subordinated group has been able to say, has been extraordinarily difficult for women to say.

What matters here: the obstacle Beauvoir identifies is not only power but the absence of the social conditions under which a group can constitute itself as a group.

Step three: the situation, and the step summaries leave out

The second volume is a developmental account: childhood, the girl, sexual initiation, marriage, motherhood, social life, old age. It is not a claim that girls are brainwashed. It is a claim that the world a girl enters is arranged so that the paths marked transcendence are obstructed and the paths marked immanence are lit, and that she takes up her situation as anyone must.

Then comes the move most summaries drop. Beauvoir insists that women are not simply victims, because the position of the Other has genuine attractions. Freedom, in this philosophy, is a burden: to choose is to be responsible without guarantee. To be the Other is to be spared that, to be maintained, to be told what one is. She calls the acceptance of this bargain a form of bad faith, and she is unsparing about it while insisting that the bargain is offered under duress and that refusing it costs more than most people can pay. Take out the complicity and you have a sociology of socialisation. With it, you have an existentialist argument in which the person who becomes a woman is also, at every stage, a freedom.

Step four: run it again with one premise changed

Volume I contains long chapters on biology, and readers regularly conclude that Beauvoir must think biology explains subordination. She thinks the opposite, and the reason is a single premise: a fact acquires meaning only inside a project. That the human female gestates is true. Whether that fact is a limit, a resource, or nothing much depends on the technology, the economy and the values of the society in which it occurs. So the body, in her formula, is a situation rather than a destiny.

Now change that premise and hold everything else fixed. Suppose facts do carry their meaning with them, so that a species in which one sex is periodically incapacitated by reproduction will build a hierarchy on it. You get Shulamith Firestone, who opens The Dialectic of Sex (1970) by praising Beauvoir and then concluding that the ground of oppression is biological and must therefore be abolished by technology, not reinterpreted. Module 3 takes that argument seriously on its own terms. For now, notice the mechanism: the two writers agree on the physiology and disagree about whether a fact can mean anything on its own.

The core of it: Beauvoir's conclusion does not come from her data. It comes from her premise that meaning is conferred by projects, and you can see this precisely because Firestone kept the data and changed the premise.

The psychoanalytic rival

Beauvoir devotes a chapter to Freud, and her objection is compact. Penis envy is offered as the explanation of femininity, but it presupposes that the organ is valuable, and the value is exactly what needed explaining. She also complains that Freud built an account of female development by analogy with the male case rather than from any independent study of it.

Psychoanalytic feminism did not stop there; it reconstructed the field to answer Beauvoir's question with different machinery.

Nancy Chodorow, in The Reproduction of Mothering (1978), works from object-relations theory rather than drive theory, and her argument is almost sociological in shape. In societies where women do nearly all the early caregiving, the infant's first and most intense relationship is with a woman. A girl develops a sense of self in continuity with that first figure, which produces a self more comfortable with connection and less sharply bounded. A boy must define himself by differentiating from her, which produces firmer boundaries and a masculinity built on not being that. One consequence is that women grow up with the relational capacities and desires that mothering requires, and men do not. The arrangement produces the people it needs. Chodorow's prescription follows directly: change who does the early caregiving, and the cycle breaks. The empirical status of the account is contested, and she has revised parts of it, but the shape of the argument, an institution that manufactures its own personnel, is worth carrying into later lessons.

Luce Irigaray pushes further back. Speculum of the Other Woman (1974), which cost her her teaching post at Vincennes and her place in Lacan's school, argues that Western thought has operated with one term and its reflection. Woman does not appear in it as a second thing; she appears as lack, mirror, or the matter on which the masculine subject works. If that is right, admitting women to the existing symbolic order is not a remedy, because the order has no place to admit them to. Her response, developed in This Sex Which Is Not One (1977), is to write a feminine speech that does not obey the logic of the single term, and her figure of the two lips that touch each other, neither one nor two, is a deliberate refusal of the arithmetic.

The standard objection is essentialism: that Irigaray grounds a way of speaking in female anatomy and so repeats the move she is attacking. Her defenders answer that she is miming the discourse in order to expose it, quoting the tradition back at itself in a register that makes its assumptions audible. Whether mimicry can be distinguished from the thing it mimics is a fair question, and it is not settled.

What has been pressed against Beauvoir

Three objections have stuck. The first is that transcendence, as she describes it, looks like the shape of a successful male life, so that the remedy is women becoming more like men, and the activities of care and the body are written off as immanence. Care ethics in Module 6 takes this up directly. The second is scope: her material is overwhelmingly drawn from educated European women, and the universality of her claims about the girl and the wife has not survived comparison. The third is grammatical and lethal. If one becomes a woman, who is the one that does the becoming? The sentence seems to require a subject standing outside gender and then entering it, which is precisely what Judith Butler will deny in Module 5.

Common misconceptions

• Beauvoir thought biology was irrelevant. She wrote hundreds of pages about it. Her claim is that biological facts do not carry meanings by themselves, so the same physiology can support opposite social arrangements.
• The Second Sex says women are simply victims. The chapters on the woman in love, the narcissist and the mystic are studies in complicity, and Beauvoir treats accepting the position of the Other as a choice made under pressure rather than a fate.
• Any English translation will do. The 1953 text is abridged and philosophically unreliable. Cite the 2009 Borde and Malovany-Chevallier translation, and say which one you are using.
• Psychoanalytic feminism is Freud with a female author. Chodorow replaces drive theory with object relations and produces a testable social claim; Irigaray attacks the structure of the symbolic order itself. Both begin from Beauvoir's objection to Freud rather than from Freud.

The short version

Beauvoir's argument runs: a person has no fixed nature; a fully human life consists in projects; women are placed in a situation that offers maintenance instead of projects and defines them as the Other of a subject who is taken to be simply human; women take up that situation, sometimes with a complicity that the theory has to include; and the biological facts do not settle any of this, because facts acquire meaning only within projects. Change that last premise and you get Firestone. Answer the same question with object relations and you get Chodorow; answer it with the structure of the symbolic and you get Irigaray.

Bottom line: the sentence about becoming a woman is not a slogan about socialisation. It is the conclusion of an argument whose weakest joint, the one who does the becoming, will be attacked directly in Module 5.

Sources

1. Mussett, S., and Bergoffen, D. (2023). Simone de Beauvoir. Stanford Encyclopedia of Philosophy. plato.stanford.edu
2. Zakin, E. (2011, revised 2023). Psychoanalytic feminism. Stanford Encyclopedia of Philosophy. plato.stanford.edu
3. Beauvoir, S. de. (2009). The Second Sex (C. Borde and S. Malovany-Chevallier, Trans.). Alfred A. Knopf. (Original work published 1949.)
4. Moi, T. (2002). While we wait: The English translation of The Second Sex. Signs: Journal of Women in Culture and Society, 27(4), 1005-1035.
5. Chodorow, N. J. (1978). The Reproduction of Mothering: Psychoanalysis and the Sociology of Gender. University of California Press.

Here is a claim you will hear in seminars, and it is wrong in an instructive way: Engels argued that capitalism created women's oppression, so abolishing capitalism would end it. Almost every word of that sentence misreports the text, and tracing exactly where it goes off is the fastest route into materialist feminism. So this lesson is a debugging exercise. We will run the claim against the source, find the line where it fails, and see what is left standing.

The source is short. Karl Marx died in March 1883. Going through his papers, Friedrich Engels found detailed notebooks on Ancient Society (1877), a study of kinship by the American lawyer and ethnologist Lewis Henry Morgan. Engels wrote The Origin of the Family, Private Property and the State in about two months in 1884 and presented it as the execution of a bequest.

What the text actually says

Engels's story has four stages. In the earliest societies, he argues, production is collective, there is little to accumulate, descent is reckoned through the mother, and household work is a public industry of the group rather than a private service to a man. Then comes the domestication of animals and the accumulation of herds. Herds are property, they grow, and they belong, on Engels's reading of the existing division of labour, to men. Property that grows raises a question no one had needed to answer: who gets it when the owner dies. Descent through the mother now conflicts with the wish to pass wealth to one's own children. So descent is switched to the male line, and the sexual conduct of wives becomes a matter of property law, because an heir must be of undisputed paternity.

That switch is what Engels calls the world historic defeat of the female sex. He is explicit that the monogamous family did not arise from individual sex love; it arose as an instrument for producing certain heirs. And he describes the resulting household in the terms of his own political economy: within the family, he writes, the man is the bourgeois and the wife represents the proletariat.

Remember: the pivot in Engels is inheritance, not the wage. Everything in his mechanism runs through the question of who receives property at death.

Line 1 of the buggy summary: the date

The claim says capitalism. Engels places the defeat at the transition from foraging to herding, which is thousands of years before there is a wage, a factory or a bourgeoisie. His argument is about private property as such. This is not a quibble; it changes what the theory predicts. If the cause were capitalism, capitalist societies should be uniquely patriarchal, which is not what he thinks and not what the record shows.

Line 2: what abolition was supposed to do

Engels does think a socialist transformation is necessary, but read what he says it requires. The first condition for the liberation of the wife, he writes, is to bring the whole female sex back into public industry, and that in turn demands that the individual family cease to be the economic unit of society. Private housekeeping is to be turned into a social industry: care of children and their education a public affair, meals, laundry and cleaning organised collectively. He also wants marriage freed from economic compulsion, with divorce easy, so that a marriage that survives survives on affection.

So even inside the text the answer is not that abolishing capitalism does the job automatically. It is that abolishing private property removes the inheritance motive, and that a specific programme, socialised domestic work plus full participation in production, is required as well.

Line 3: the anthropology

Now the failures that are not misreadings. Engels inherited from Morgan, and behind Morgan from the Swiss jurist Johann Jakob Bachofen, a picture of universal stages: promiscuity, then group marriage, then the pairing family, then monogamy, running alongside savagery, barbarism and civilisation. Twentieth-century fieldwork demolished the scheme. There is no evidence for a universal stage of group marriage; Morgan had inferred it from kinship terminologies rather than observing it. Societies do not march through a single sequence. Foraging societies vary enormously in how much authority women hold.

One confusion in particular is worth killing on sight. Matriarchy means rule by women. Matriliny means descent and inheritance traced through the female line. They are different things, and the second is common while the first has never been securely documented as a society-wide system. Bachofen's mother right was an argument about descent that got read as an argument about rule. The Haudenosaunee case that Morgan knew best is instructive precisely because it is specific: clan matrons held real power, including the nomination and removal of sachems, within an institutional arrangement that has to be described on its own terms rather than filed as a stage of universal history.

Line 4: the hole in the middle

This is the failure that matters most for the rest of the module, and it is not about ethnography. Read the mechanism again. Herds accumulate. The herds belong to men. Men therefore want certain heirs, and women's sexuality is controlled to supply them.

Why do the herds belong to men?

Engels answers by pointing to a division of labour already in place before the surplus appears, and treats it as given. But that division is the thing to be explained. If men already controlled the new wealth because of an existing arrangement between the sexes, then something was already organising relations between the sexes before property could do it, and the theory has quietly assumed what it set out to derive. Gerda Lerner presses this in The Creation of Patriarchy (1986), and Heidi Hartmann will make it the hinge of the next lesson.

The point: Engels explains why property holders wanted control over women's sexuality. He does not explain why the property holders were men, and the gap is exactly the size of a second theory.

Line 5: the prediction, tested

Engels's account produced a falsifiable expectation: put women into paid production, socialise the household, remove property from the equation, and the domestic hierarchy should dissolve. Large parts of that were tried. Soviet and Eastern European economies achieved female labour force participation rates that Western countries did not approach for decades. Alexandra Kollontai, as commissar for social welfare, pushed for communal dining halls, public laundries and maternity provision; the 1918 Family Code stripped marriage of its religious form and made divorce a matter of registration; abortion was legalised in 1920, the first country to do so.

The domestic division of labour did not follow. Time-use studies through the socialist period found women in full-time employment still carrying the overwhelming majority of housework and childcare, the pattern later named the double burden. The communal facilities were underfunded, and the household absorbed the shortfall, as it does everywhere. Meanwhile the more conservative turn of the 1930s recriminalised abortion in the Soviet Union in 1936 and made divorce harder, which tells you something about how secure the gains were.

Marxist parties in the West did not ignore the question either. August Bebel's Woman and Socialism (1879) went through dozens of editions and was among the most widely read books in the German workers' movement; Clara Zetkin built a mass socialist women's movement around it. The charge that the tradition postponed the woman question is a charge about practice, and about the recurring argument that class was the main enemy and everything else a diversion.

What survives

Debugging is not demolition. Three things in Engels have outlasted the framework they arrived in.

First, and most important, the family is treated as a historical institution with an economic function rather than a natural unit outside history. Once you accept that, you can ask what work the household does in an economy, which is the question the whole of Marxist feminism and later social reproduction theory takes up.

Second, the link between property transmission and the control of women's sexual conduct. Stated as a universal law it fails. Stated as a tendency, that where heritable property is significant and lineage matters, restrictions on women's sexual freedom tend to be tighter, it is still used by historians of dowry, legitimacy and honour, and it explains a good deal of variation within societies as well as between them.

Third, the reframing of housework as an economic question. Engels put unpaid domestic labour on the same page as production and asked what it costs and who benefits. The next two lessons are the fight over how to answer that.

Common misconceptions

• Engels blamed capitalism for patriarchy. He dated the decisive change to the accumulation of herds under early private property, millennia before capitalism existed.
• Mother right meant that women ruled. It meant descent and inheritance through the female line. Matriliny is common; society-wide matriarchy has never been securely documented.
• The book is original research by Engels. He worked from Marx's notebooks on Morgan, and the stage theory is Morgan's and Bachofen's. Engels supplied the political economy.
• Socialist regimes never tried the programme. They tried substantial parts of it, and the results are the best available evidence about what employment alone does and does not change.

Putting it together

The buggy summary fails at the date, at the remedy and at the evidence. The text itself fails at the anthropology it borrowed, at the prediction that employment plus socialised services would dissolve domestic hierarchy, and above all at a missing step: it assumes a sexual division of labour in order to explain the sexual division of labour. What it leaves is a way of asking the question, an institution with a function and a history, and a link between property and sexual regulation that still earns its keep as a tendency.

So what?: the hole in the middle of Engels is the reason the next lesson exists. If class relations cannot generate the sexual division of labour on their own, either Marxism needs a second theory alongside it or its categories need rebuilding.

Sources

1. Engels, F. (1884). The Origin of the Family, Private Property and the State. Marxists Internet Archive. marxists.org
2. Engels, F. (1884). Chapter 2: The family. In The Origin of the Family, Private Property and the State. marxists.org
3. Gimenez, M., and others. (2023). Feminist perspectives on class and work. Stanford Encyclopedia of Philosophy. plato.stanford.edu
4. Lerner, G. (1986). The Creation of Patriarchy. Oxford University Press.
5. Sacks, K. (1974). Engels revisited: Women, the organization of production, and private property. In M. Z. Rosaldo and L. Lamphere (Eds.), Woman, Culture, and Society (pp. 207-222). Stanford University Press.

Heidi Hartmann opened a 1979 essay with a sentence that has been quoted ever since: the marriage of Marxism and feminism, she wrote, has been like the marriage of husband and wife depicted in English common law, in that the two are one and that one is Marxism. The essay was called The Unhappy Marriage of Marxism and Feminism: Towards a More Progressive Union. Two years later Lydia Sargent reprinted it in a book alongside a dozen replies, several of them furious, and that volume is where the argument in this lesson was fought.

The question is narrow and consequential. Is women's subordination something that a properly extended theory of capitalism can explain, or does it require a second theory with its own logic, running alongside the first? Everything about what socialists and feminists should do together turns on the answer.

The charge: sex-blind categories

Hartmann's technical complaint is that Marxist categories are sex-blind. Marx can tell you a great deal about the structure of a labour market: that capital seeks cheap and disciplined labour, that it creates a reserve army of the unemployed to hold wages down, that it will deskill work where it can. Those categories describe positions. They do not describe persons. Nothing in the theory of value says which bodies end up in the badly paid, insecure, easily dismissed positions, and Hartmann's point is that the allocation is not random and not explained.

Her positive definition is worth quoting closely because people paraphrase it into something weaker. Patriarchy, for Hartmann, is a set of social relations between men which have a material base and which, though hierarchical, establish interdependence and solidarity among men that enable them to dominate women. Note the three parts. It is a relation among men, not simply a relation between men and women. It has a material base, which she locates in men's control over women's labour power, secured by excluding women from access to productive resources and by restricting their sexuality. And it is hierarchical among men, which is how she explains why a system that benefits men in general is maintained by men who are themselves subordinate to other men.

The evidence: when the two systems fight

A theory of two systems needs a case where they pull in opposite directions, because that is what distinguishes a second system from a department of the first. Hartmann's earlier paper, on job segregation, supplies it.

Consider what capital wanted in nineteenth-century Britain and America: labour as cheap as it could get. Women and children were cheaper. If capital were the only actor, women should have flooded into industrial employment on the same terms as men. What actually happened is that organised male workers fought to keep them out. Craft unions restricted apprenticeships and membership. Unions campaigned for a family wage, a male wage sufficient to support a wife and children, on the explicit argument that a married woman's place was at home. The Mines and Collieries Act 1842 barred women and girls, along with boys under ten, from working underground.

Protective legislation is the sharpest case because it cuts both ways at once. In Muller v. Oregon (1908), the Supreme Court upheld a maximum-hours law that applied only to women, accepting a brief crammed with medical and social evidence about women's physical vulnerability. Real exhaustion was reduced. But the reasoning, that women are a special class whose labour the state may regulate on account of their bodies and their role as mothers, was available for other purposes, and in Adkins v. Children's Hospital (1923) a minimum wage for women in the District of Columbia was struck down. Protection that is granted because of what women are is protection that can be withdrawn, or converted into exclusion, on the same grounds.

Why this matters: Hartmann's argument does not rest on men holding bad attitudes. It rests on a documented record of male workers spending organisational resources, against the immediate interest of their employers, to secure a labour market segregated by sex. If capital did not want that outcome and got it anyway, something other than capital was acting.

The reply: you have conceded too much

Iris Marion Young's response, published in the same 1981 collection, accepts almost all of Hartmann's evidence and rejects the conclusion. Her objection has two parts.

The first is that dual systems theory flatters Marxism. By saying that Marxism handles capitalism adequately and merely needs a companion theory for sex, it leaves Marxist categories exactly where they were and quarantines the feminist insight in a separate box. If the analysis of capitalism is really gender-blind, Young argues, the right response is to rebuild it, not to bolt something onto it.

The second is that patriarchy, treated as a system with its own transhistorical logic, floats free of history. It becomes a constant, present in every society in every period, and a constant explains nothing about variation. Worse, the two systems then need a theory of how they interact, and no such theory was ever produced.

Young's alternative is to make the gender division of labour the fundamental category of the analysis itself, so that capitalism is not a gender-neutral system that patriarchy colonises but a patriarchal capitalism from the beginning. On that reading, the availability of a workforce that can be treated as secondary, marginal, supplementary to a male wage, and therefore paid less and shed first, is not a distortion of capitalist logic but one of its conditions.

A third position, and a fourth

Two other moves are worth having.

Juliet Mitchell had already, in a 1966 essay in New Left Review, refused the choice. Women's condition, she argued, is not one structure but four that vary independently: production, reproduction, sexuality, and the socialisation of children. A society can liberalise one and tighten another, which is why change is uneven and why a single lever never works. That framework predicts something the others do not: reforms that look contradictory are normal, because the structures are not synchronised.

Michele Barrett, in Women's Oppression Today (1980), attacks the word itself. Patriarchy, she argues, has become a term that applies to every society in every period, and a category that never fails to apply cannot discriminate between cases. She prefers to analyse a historically specific object, the family-household system and the ideology of domesticity attached to it, and to show how it was built in a particular economy over a particular century.

Sylvia Walby's Theorizing Patriarchy (1990) is the most serious attempt to answer Barrett without dropping the term. She breaks patriarchy into six structures, paid work, household production, culture, sexuality, violence and the state, and argues that they can vary in strength and combination. That turns patriarchy from a constant into a variable, and lets her describe a twentieth-century shift from private patriarchy, centred on the household and individual men, to public patriarchy, centred on employment and the state, in which women are no longer excluded but segregated and subordinated within public institutions.

What would settle it

Each side has a test it is prepared to accept, which is what makes this a real dispute rather than a clash of vocabularies.

Hartmann's test is conflict: find cases where the interests of capital and the interests of organised men diverge, and see who wins. The family wage campaigns are her exhibit. Young's counter-test is parsimony: show that a single reconstructed theory predicts those same episodes, on the ground that conflict between fractions is ordinary inside any one system, and that a united capitalist interest is a fiction in the first place. Barrett's test is explanatory work: take any statement containing the word patriarchy, delete the word, and ask whether anything has been lost. Walby's answer is that the disaggregated version passes that test because the six structures can be measured moving separately over time.

The family wage itself remains genuinely contested. The economic historian Jane Humphries has argued that a demand for a wage capable of supporting a household was, in conditions of appalling child mortality and no welfare state, a rational strategy for working-class families rather than simply an expression of male interest. That is a live empirical dispute about the same evidence Hartmann uses, and it is a good example of how a theoretical argument can be moved by archival work.

The upshot: nobody now defends the pure form of either position. What survived is Hartmann's demand that a theory say why particular people occupy particular positions, and Young's demand that the answer be built into the analysis rather than appended to it.

Common misconceptions

• Socialist feminism just means being both a socialist and a feminist. It names a specific research problem: whether class and sex are one system or two, and how they interact if two.
• Hartmann says patriarchy is about male attitudes. Her definition is material: relations among men, based on control of women's labour power, maintained through exclusion from resources and restriction of sexuality.
• Protective legislation was simply a gain for women. The reasoning that justified shorter hours in Muller was later used to strike down a women's minimum wage in Adkins. Protection granted on account of what women are can be withdrawn on the same account.
• Rejecting the word patriarchy means denying that women are subordinated. Barrett's objection is that a term applying everywhere cannot explain differences, and she offers a historically specific analysis in its place.

What to remember

Hartmann's charge is that Marxism describes positions and not the people who fill them, and her evidence is that organised men fought capital, at cost, for a labour market divided by sex. Young answers that a second system leaves the first unreformed and turns patriarchy into a timeless constant, and proposes to rebuild the analysis with the gender division of labour at its centre. Mitchell had already split the object into four structures; Barrett doubts the word; Walby saves it by dividing it into six measurable pieces and letting the mix change over time.

Key idea: the productive question is not whether patriarchy exists, but whether the concept does work you could not do without it. Any use of the term in your own writing should be able to answer that.

Sources

1. Wikipedia contributors. (n.d.). Socialist feminism. Wikipedia. en.wikipedia.org
2. Wikipedia contributors. (n.d.). Muller v. Oregon. Wikipedia. en.wikipedia.org
3. Hartmann, H. I. (1979). The unhappy marriage of Marxism and feminism: Towards a more progressive union. Capital and Class, 3(2), 1-33.
4. Hartmann, H. I. (1976). Capitalism, patriarchy, and job segregation by sex. Signs, 1(3, Part 2), 137-169.
5. Young, I. M. (1981). Beyond the unhappy marriage: A critique of dual systems theory. In L. Sargent (Ed.), Women and Revolution (pp. 43-69). South End Press.
6. Walby, S. (1990). Theorizing Patriarchy. Basil Blackwell.

In the summer of 1972, four women met in Padua and founded the International Feminist Collective. Mariarosa Dalla Costa was an Italian academic close to the workerist current that had spent a decade arguing about factory struggle. Selma James was an American who had gone into a factory at fifteen, written a pamphlet called A Woman's Place in 1953 under a pen name, and married the Trinidadian Marxist C. L. R. James. Silvia Federici was teaching in the United States. Brigitte Galtier had come from France.

That year Dalla Costa and James published a slim book, The Power of Women and the Subversion of the Community, containing a demand that struck most people who heard it as either obvious or absurd: wages for housework. Committees followed in Italy, in Britain, and in New York, where Federici set one up in 1974.

The argument they were actually making

Start with the claim about the economy, because the demand only makes sense on top of it. Capital buys labour power by the hour. But labour power is not manufactured anywhere. It is produced daily, in a bed, a kitchen and a bathroom, by someone who feeds a worker, washes his clothes, absorbs his exhaustion and irritation, and returns him to the gate the next morning fit to work; and it is produced generationally, by bearing and raising the next cohort. None of that appears in the wage bill. Dalla Costa's formulation was that the community is not the other side of the factory but an extension of it, a social factory in which the work of reproducing workers is performed without pay.

The wage, on this analysis, does something specific. By paying one member of a household, it makes his work visible as work and hers invisible as love. She then appears not as an unpaid worker but as a dependent, which is a position with no leverage: a worker can strike, a dependent can only be ungrateful.

Federici's 1975 pamphlet Wages Against Housework opens with a line that does the whole job in nine words: they say it is love, we say it is unwaged work.

Bottom line: the demand is not primarily about money. It is a demand that a category of activity be reclassified, and the money is the instrument that forces the reclassification.

The misreading, and Federici's answer to it

The objection came instantly and it was always the same. Paying women to do housework would tie them to the sink, subsidise the sexual division of labour, and turn the state into a husband.

Notice the title of Federici's pamphlet: not wages for housework but wages against it. Her argument is that a demand for a wage is the opposite of an acceptance of the job, because work that is paid can be bargained over, reduced, timed, shared and refused, whereas work that is an expression of your nature can only be done badly or well. To ask for a wage, she writes, is to refuse the definition of the work as a natural attribute of femininity. On her account the campaign's first victory would be that a woman could say the word work about what she was doing at eleven at night, and that this alone changes what is negotiable inside a household.

Whether a demand can be relied on to be read that way, particularly by a government drafting the legislation, is a fair question, and it is the one the critics pressed.

The objection that landed hardest

Angela Davis devoted a chapter of Women, Race and Class (1981) to the demand, and her objection comes from a different vantage point rather than a different premise.

For much of American history, domestic work was not the unpaid thing Black women did at home. It was the paid thing they were confined to doing in other people's homes, for wages that were not covered by the labour protections extended to industrial workers, in a job whose intimacy made it uniquely open to abuse. Framing a wage for housework as liberation reads very differently from inside that history. Davis's conclusion is that the goal should be the reduction and socialisation of housework, through public provision and industrial-scale services, rather than its remuneration inside the individual household. She also points out, with some force, that the campaign never produced a serious answer to who pays.

The other objection was technical and came from Marxists. Through the 1970s a long argument ran over whether domestic labour produces value in Marx's sense. The rough consensus was that it produces use-values and renews labour power, and is therefore necessary to accumulation, but that it does not directly produce surplus value, because nothing in it is exchanged on a market. That conclusion cost the campaign its strongest theoretical claim while leaving the descriptive point standing.

Where Federici took the argument next

Nearly thirty years later Federici published Caliban and the Witch (2004), which reads the origins of capitalism through the same lens. Marx's account of primitive accumulation describes peasants separated from the land by enclosure and driven into wage labour. Federici argues that the process had a second face. Alongside the expropriation of land came the construction of a new sexual division of labour in which women were excluded from waged trades, confined to unpaid reproduction, and stripped of control over their own fertility. The witch hunts of the sixteenth and seventeenth centuries, on her reading, were an instrument of that transformation: an attack on women's autonomy, on the collective life of villages, and on knowledge about contraception and abortion held mainly by women.

The thesis is powerful and the history needs handling with care, which is a good habit to acquire on a book you find persuasive. Current estimates put executions for witchcraft across Europe between about 1450 and 1750 in the range of forty to sixty thousand, not the millions that circulated in popular accounts of the 1970s. Roughly three quarters to four fifths of the accused were women, but the proportion varies sharply by region, and in Iceland and Estonia most of the accused were men. The specific claim that midwives were disproportionately targeted has been challenged directly by historians working from trial records. None of that disposes of Federici's argument about the period, but it means the argument has to be made with the numbers historians actually have.

The theory that came out of it

The durable descendant of this argument is social reproduction theory, and its cleanest statement is Lise Vogel's Marxism and the Oppression of Women (1983). Vogel declines both the two-system and the wage-for-housework routes. Her claim is that capital requires labour power to be renewed daily and replaced generationally; that this renewal happens largely outside the wage relation; and that there is therefore a permanent structural tension inside capitalism between the drive to extract as much labour as cheaply as possible and the need for workers to exist next week and next generation. Whoever performs that renewal is structurally positioned to absorb the tension.

The framework predicts something you can check. When employers or states squeeze the resources available for reproduction, the shortfall does not vanish; it lands on households, and disproportionately on women. Nancy Fraser has developed this as an account of a recurring crisis of care, in which a system free-rides on capacities it does not replenish. Tithi Bhattacharya's 2017 collection brought the framework back into wide use.

What the demand actually won

No state ever paid a wage for housework. But look at what happened to the accounting. The 1985 UN conference in Nairobi and the 1995 Beijing Platform for Action both called on governments to measure unpaid work. Statistical agencies built time-use surveys and satellite accounts to do it. The International Labour Organization's 2018 study put global unpaid care work at 16.4 billion hours a day, with 76.2 per cent of it performed by women, and calculated that valuing those hours at an hourly minimum wage would come to about nine per cent of world output.

A number like that is precisely what an economy that treats domestic work as love cannot produce, because it has no reason to count the hours. The campaign lost its demand and won the argument about the category.

In short: the money never arrived; the measurement did, and the measurement is what makes the work arguable about.

Common misconceptions

• Wages for housework meant paying women to stay home. Federici's pamphlet is titled Wages Against Housework and argues that a wage makes work refusable, because work that is paid can be timed, shared and reduced while work said to be love cannot.
• The campaign was a fringe of the movement with no effect. It lost its central demand and reshaped the terms of the argument, and the international push to measure unpaid work is its clearest descendant.
• Everyone on the left agreed that housework produces surplus value. The domestic labour debate concluded largely the other way: necessary to accumulation, but not directly value-producing, because nothing in it is exchanged.
• Millions were executed as witches. Estimates from trial records give tens of thousands across three centuries. Getting this right matters for the strength of Federici's argument, not against it.

Pulling it together

Four women in Padua argued that the household is a site of production, that the wage hides half of it, and that demanding payment for the hidden half would force it into view as work. Davis answered from the position of women for whom domestic labour had been the paid confinement rather than the unpaid one, and pressed the case for socialising the work instead. The technical debate denied the campaign its strongest theoretical claim. What outlived the demand is the framework: labour power has to be produced, its production happens mostly outside the wage, and every squeeze on the resources for it lands somewhere.

Worth holding on to: a demand can fail as a policy and succeed as an argument, and the test of the second is whether the vocabulary afterwards is different. Nobody in 1970 could have said 16.4 billion hours a day.

Sources

1. International Labour Organization. (2018). Care Work and Care Jobs for the Future of Decent Work. ilo.org
2. Wikipedia contributors. (n.d.). Wages for housework. Wikipedia. en.wikipedia.org
3. Dalla Costa, M., and James, S. (1972). The Power of Women and the Subversion of the Community. Falling Wall Press.
4. Federici, S. (2004). Caliban and the Witch: Women, the Body and Primitive Accumulation. Autonomedia.
5. Davis, A. Y. (1981). The approaching obsolescence of housework: A working-class perspective. In Women, Race and Class. Random House.
6. Vogel, L. (1983). Marxism and the Oppression of Women: Toward a Unitary Theory. Rutgers University Press.

Two books appeared in 1970 within ten weeks of each other. In August, Doubleday published Sexual Politics by Kate Millett, a revision of the doctoral dissertation she had defended at Columbia the previous year. It sold in the tens of thousands and put her on the cover of Time within the month. In October, William Morrow published The Dialectic of Sex by Shulamith Firestone, who was twenty-five and had already helped found three organisations in as many years.

Both are radical feminist in the technical sense: they hold that the subordination of women is not a by-product of class or race but a system in its own right, historically first, and located in domains that both liberal and Marxist analysis had left alone, namely sexuality, reproduction and personal life. Beyond that they are almost opposites, and setting them side by side is the fastest way to see what a theory of this kind has to decide.

The table

Reading the first column

Millett's opening move is a redefinition. Politics, she writes, means power-structured relationships, arrangements whereby one group of persons is controlled by another. Once politics is defined by the structure of the relation rather than by the presence of a state, there is no reason a household or a bed should be outside it, and the phrase sexual politics stops being a metaphor.

She then treats patriarchy as an institution with an ideology, and lists the channels through which it operates: ideological, biological, sociological, economic and educational, force, class, myth and religion, and psychological. Her emphasis falls on the first. A system that had to be held together by violence would be unstable and expensive; this one is held together mainly by consent, produced early, in families and schools, and confirmed by everything a person reads and watches. Force is kept in reserve, and its rarity is a sign of the system's efficiency rather than its mildness.

The second half of the book is the part that made it famous and the part that got it attacked. Millett reads D. H. Lawrence, Henry Miller, Norman Mailer and Jean Genet, quoting sexual passages at length and asking what is being described. Her answer is that the scenes are not descriptions of desire but of hierarchy: who acts, who is acted upon, who is named and who is described in parts.

The Genet chapter is the analytically important one and is usually skipped. Genet's fiction depicts a homosexual subculture in which roles of dominance and submission, coded as masculine and feminine, are assigned among people of the same sex. Millett's point is that the roles therefore cannot be reading off from bodies. They are a script that a hierarchy needs and will assign to whoever is available. That is an anti-essentialist argument, and it is worth registering because radical feminism is routinely assumed to be essentialist.

Key idea: Millett's innovation is treating culture as evidence about power, and her strongest argument is that the masculine and feminine positions are roles in a structure rather than expressions of male and female bodies.

Reading the second column

Firestone begins by praising Marx and Engels for their method and rejecting their premise. Historical materialism was right to look for a material base; it looked in the wrong place. The deepest division is not between classes defined by property but between what she calls sex classes, and its base is reproduction.

Her argument runs from four facts about the human animal. Women are, for substantial periods, dependent during pregnancy and after birth. Human infants are helpless for longer than any other species. That dependence creates an intense and unequal mother-child bond. And the arrangement built on these facts, the biological family, reproduces itself, generating in each generation the psychology that makes it seem natural.

Nothing about this is a complaint about attitudes. Firestone's claim is that the arrangement is grounded in a physical asymmetry, and she draws the conclusion that a materialist should draw: if the base is physical, changing the culture is cosmetic. So the demand is for the seizure of the means of reproduction, which for her means contraception and abortion on demand as a first step and artificial gestation as the eventual one, together with the replacement of the biological family by voluntary households, and the end of the segregation of children into a separate state called childhood.

She is not naive about her own proposal. She is explicit that technology in the hands of the existing order will be used by the existing order, and that the demand only makes sense as part of a revolution. But the premise stands: biology is the base.

What has been pressed against each

Against Firestone, the sharpest objection is the one visible in the table. Beauvoir refused to let a biological fact carry a social meaning by itself. Firestone accepts that it does, and then proposes to abolish the fact. That is not an escape from biological determinism; it is biological determinism with an engineering programme attached, and it concedes the ground on which every conservative argument about women's nature has ever stood.

Her prediction has also been partly tested, which is rare in this field. Reliable contraception, legal abortion in many countries, in vitro fertilisation from 1978, and commercial surrogacy all arrived. The sexual division of labour did not dissolve. What happened instead is closer to what Firestone's own caution predicted: the technologies were absorbed by existing arrangements, with surrogacy in particular generating a market that runs from wealthier commissioning parents to poorer gestating women, frequently across borders. That is a serious result. It suggests that the reproductive asymmetry was not doing the causal work she assigned to it.

Her chapter treating racism through an Oedipal family drama is not defended by anyone now, and it is a useful illustration of what happens when a single-axis theory is asked to explain a second axis. Module 4 is largely about that failure.

Against Millett, the standard complaint is that the literary readings are advocacy: she quotes selectively, ignores irony, and treats a novelist's depiction as an endorsement. Mailer replied at length in The Prisoner of Sex (1971), and some of his specific corrections about Lawrence have been accepted even by readers who think his general position indefensible. The deeper problem is the one Barrett raised in the last lesson: a patriarchy present in every society and every text is a constant, and Millett offers little account of why it takes different forms in different places.

The point: the two books fail in opposite directions. Firestone has a mechanism specific enough to be tested and it did not survive the test. Millett has a description broad enough to fit everything and therefore hard to test at all.

The movement around the books

Neither book is a private intellectual event. New York Radical Women, then Redstockings, then New York Radical Feminists were founded between 1967 and 1969, and Firestone was involved in all three. Anne Koedt's essay The Myth of the Vaginal Orgasm, circulated from 1968, took a claim of clinical sexology and turned it into a political argument about whose pleasure had been defined as normal. Robin Morgan's anthology Sisterhood Is Powerful (1970) put the arguments in front of a mass audience. And Adrienne Rich's 1980 essay on compulsory heterosexuality extended the analysis by asking not why some women are lesbians but why heterosexuality is treated as requiring no explanation at all.

Common misconceptions

• Radical feminism means extreme feminism. The word points at a claim about roots: that sex domination is a system in its own right rather than a side effect of another one. Some radical feminists were and are politically moderate on other questions.
• Radical feminists were all biological essentialists. Millett's Genet chapter argues the opposite, that masculine and feminine are roles assignable to any bodies. Firestone does treat biology as the base, which is why the two are set side by side here.
• Sexual Politics is a book about sexual behaviour. It is a book about power, using sexual writing as its evidence, and its first move is a definition of politics that has nothing to do with bedrooms.
• Firestone wanted technology to solve everything. She insisted that technology controlled by the existing order would serve it, which is closer to what happened than most summaries admit.

The takeaway

Radical feminism made two moves that the rest of this course argues about. It claimed that sex domination is primary rather than derivative, and it moved the site of analysis into sexuality, reproduction and personal life. The first claim has not been established and is directly contested in Module 4. The second has been so completely absorbed that it is now invisible: nobody today needs to be persuaded that reproduction and sexual violence are political subjects, and that is a change these books made.

Remember: Firestone put biology at the base and proposed to abolish it; Millett put ideology at the base and read novels to find it. Their disagreement about where the base lies is the same disagreement, differently placed, that ran through Module 2.

Sources

1. Wikipedia contributors. (n.d.). Sexual Politics. Wikipedia. en.wikipedia.org
2. Wikipedia contributors. (n.d.). The Dialectic of Sex. Wikipedia. en.wikipedia.org
3. Allen, A. (2022). Feminist perspectives on power. Stanford Encyclopedia of Philosophy. plato.stanford.edu
4. Millett, K. (1970). Sexual Politics. Doubleday.
5. Firestone, S. (1970). The Dialectic of Sex: The Case for Feminist Revolution. William Morrow.

In 1979 Yale University Press published a book by a law student arguing that when a supervisor conditions a woman's job on sexual access, he is discriminating against her because of sex within the meaning of Title VII of the Civil Rights Act. Before Sexual Harassment of Working Women appeared, courts had been dismissing such claims as personal matters that happened to occur at work: an unpleasant man, a private quarrel, nothing the statute addressed. Within seven years the Supreme Court had accepted the hostile environment theory the book set out. The doctrinal history belongs to another course; what concerns us is the argument, because Catharine MacKinnon won that case by changing the question the law asks.

This lesson works that change through two examples. The procedure is the same each time: state the facts, run the standard test, run MacKinnon's test, and see where they part.

The standard test

Antidiscrimination law inherited a principle from Aristotle: treat like cases alike and unlike cases differently. Applied to sex, that becomes a question about comparison. Is this woman similarly situated to a man who is being treated better? If yes, she must be treated the same. If she is genuinely different, then differential treatment is permitted and may even be required.

Both branches have been useful, and both, MacKinnon argues, have the same defect. On the sameness branch a woman gets what she can show she deserves by being like a man. On the difference branch she gets special provision because she is not like a man. In each case the man is the measure, and the position of men in the world is treated as the unmarked standard against which claims are assessed. Nobody ever asks whether men are similarly situated to women.

The replacement test

MacKinnon's alternative, set out most compactly in Feminism Unmodified (1987), is to stop asking about difference and start asking about hierarchy. The question becomes: does this practice participate in the subordination of one group by another? Difference is then a symptom rather than a criterion, and the empirical fact that men and women differ in some respect stops being the issue, because differences exist everywhere and only some of them are used to sort people into higher and lower positions.

She calls this the inequality or dominance approach. It is a substantive standard rather than a formal one: it requires a judgement about which practices sustain a hierarchy, which is both its power and, as we will see, the main thing pressed against it.

Case one: a supervisor and a job

Facts. A woman is told, in effect, that keeping her job depends on going to bed with her manager. She refuses and is dismissed.

Run the standard test. Is she similarly situated to a man? The comparison is awkward. The manager is not propositioning male employees, so she is being treated differently; but a court in the 1970s could and did reply that the reason was not her sex as a class but the manager's particular attraction to her as an individual, which is personal rather than categorical. Some courts added that if the manager were bisexual and propositioned everyone, there would be no discrimination at all, which is a reductio that the framework generates on its own.

Run the dominance test. Is this a practice that helps keep women in a subordinate position at work? It requires her to purchase employment with something no male colleague is asked for; it makes her presence in the workplace conditional on sexual availability; and it operates on her as a member of a group, since the availability of the demand depends on her being a woman in a workplace organised so that men supervise women. The answer is yes, and it does not depend on finding a man in an identical position for comparison.

What matters here: the argument that won was not that harassment is bad but that it is discrimination, and that conclusion followed from changing the question from whether she was treated differently to whether the practice enforces a hierarchy.

Case two: change one input

Facts. An employer's disability plan covers every temporary inability to work except pregnancy.

Run the standard test. In General Electric v. Gilbert (1976) the Supreme Court held that this was not sex discrimination, because the plan divided employees into pregnant persons and non-pregnant persons, and the second group contains both men and women. Formally, that is impeccable. There is no man denied pregnancy coverage, so no man is being treated better than a comparable woman.

Run the dominance test. Does excluding the one physical condition unique to women from an otherwise comprehensive benefit contribute to women's subordinate position in employment? It transfers a predictable cost onto one sex, at the point in a career where withdrawal from the workforce is most likely, and it does so in a scheme designed precisely to spread such costs. The answer is yes, immediately.

Same doctrine, opposite results, and nothing changed except the question. Congress reached the second answer by statute in 1978. The point of running the two cases side by side is that a legal test is not a neutral instrument: it decides in advance which facts will count.

Dworkin, and a claim she did not make

Andrea Dworkin came to the same conclusions from a different direction. Her books, Woman Hating (1974), Pornography: Men Possessing Women (1981) and Intercourse (1987), argue that pornography is not a representation of subordination but an instance of it: something is done to the women who are filmed, and something is done by the material to the conditions in which other women live. On that reading the harm is not offence, and the response is not censorship of ideas but a remedy for an injury.

Before going further, correct the claim that follows Dworkin everywhere. She is said to have written that all heterosexual intercourse is rape. She did not, in Intercourse or anywhere else, and she denied it repeatedly and in print. What the book argues is that in a society organised by sexual inequality, the meaning of the act is shaped by that inequality, in the way that the meaning of a gift is shaped by whether the recipient can refuse it. That is a claim about social meaning, not a definition, and it can be argued with. The invented version cannot, which is why it circulated.

MacKinnon's parallel argument concerns consent in rape law. The law asks whether the woman consented, but proves the answer through what the defendant reasonably believed. So the question of what happened to her is answered from his point of view, which she describes as trying the victim's consent from the standpoint of the accused. Whether that describes current law in a given jurisdiction is a question of statute; as a description of the older common law position it is accurate, and it is why reforms in several countries have shifted to affirmative consent standards.

The ordinance, and the judge who agreed

In 1983 the Minneapolis city council commissioned MacKinnon and Dworkin to draft an ordinance, and they produced something unusual: not a criminal ban but a civil rights remedy. Pornography was defined as the graphic sexually explicit subordination of women, and women harmed by it, coerced into it, forced to view it, or assaulted by someone acting on it, could sue. Minneapolis passed it twice and the mayor vetoed it twice. Indianapolis passed a version in 1984.

It was struck down in American Booksellers Association v. Hudnut (1985), and the reasoning is the most interesting thing in the entire dispute. Judge Frank Easterbrook did not say the ordinance rested on a false view of what pornography does. He accepted the premise for the sake of argument and wrote that if pornography shapes how people see and treat women, that only demonstrates the power of pornography as speech. The ordinance failed because it drew a line between depictions of sex that subordinate and depictions that do not, and that is a distinction based on point of view, which the First Amendment forbids the state from making. The Supreme Court affirmed summarily in 1986.

The upshot: the ordinance did not lose because the court rejected the harm. It lost because American constitutional law protects speech precisely when it is effective, and the theory had proved that this speech was effective.

What has been pressed against dominance theory

Three objections, all serious.

The first is about evidence. The causal claim linking pornography consumption to sexual violence is not settled. The 1970 United States Commission on Obscenity and Pornography found no reliable evidence of a causal link; the 1986 Attorney General's Commission, working with a different mandate and heavily criticised methods, concluded the opposite. Laboratory studies find short-term attitudinal effects that are difficult to extrapolate. Anyone who tells you the question is closed, in either direction, is reporting a preference.

The second is about agency. If preferences are formed under conditions of dominance, then a woman's own report that she chose freely can always be reclassified as evidence of how deep the conditioning goes. That move is unfalsifiable, and it takes authority over a woman's account of her own life away from her and gives it to the theorist. Critics inside feminism have pressed this hard, and it is the hinge of the next lesson.

The third is about operation. To ask whether a practice contributes to subordination, someone has to decide which practices do. A formal test at least tells a judge what to look for. A substantive test relocates the judgement without removing it, and Hudnut is what happens when a court is asked to make it.

Common misconceptions

• Dworkin wrote that all heterosexual intercourse is rape. She did not, and said so repeatedly. Intercourse argues that inequality shapes the social meaning of the act, which is a different and arguable claim.
• The MacKinnon and Dworkin ordinance was a criminal obscenity ban. It created a civil right of action for specified harms and deliberately avoided the obscenity framework, which turns on community standards and offence.
• Hudnut rejected the factual claims about pornography's effects. Easterbrook assumed them for the sake of argument and struck the ordinance down as viewpoint discrimination, which made the loss harder to answer rather than easier.
• Dominance theory says men and women are the same. It says the question of difference is the wrong question, and that the right one is whether a practice sustains a hierarchy.

Summing up

Run a fact pattern through the similarly-situated test and you get an answer that depends on finding a comparable man. Run it through the dominance test and you get an answer that depends on whether the practice keeps a group down. Harassment law was built by the second question; the pregnancy cases show the first question producing a result that Congress had to reverse by statute. Dworkin's contribution was to insist that some material is an act rather than an argument, and the ordinance foundered on a constitutional principle that protects speech in proportion to its power.

Bottom line: dominance theory is strongest as a critique of how legal questions are framed and weakest when asked to say, without further argument, which practices count as subordinating. The next lesson is what happened when feminists disagreed about exactly that.

Sources

1. United States Court of Appeals for the Seventh Circuit. (1985). American Booksellers Association, Inc. v. Hudnut, 771 F.2d 323. CourtListener. courtlistener.com
2. Wikipedia contributors. (n.d.). Antipornography Civil Rights Ordinance. Wikipedia. en.wikipedia.org
3. MacKinnon, C. A. (1979). Sexual Harassment of Working Women: A Case of Sex Discrimination. Yale University Press.
4. MacKinnon, C. A. (1987). Feminism Unmodified: Discourses on Life and Law. Harvard University Press.
5. Dworkin, A. (1981). Pornography: Men Possessing Women. Perigee Books.

On 24 April 1982 Barnard College hosted the ninth Scholar and the Feminist conference, on the politics of sexuality. Members of Women Against Pornography and allied groups picketed the entrance in T-shirts reading against sadomasochism, handed out leaflets naming individual participants, and telephoned the college in advance. Barnard's administration impounded the conference booklet, the Diary of a Conference on Sexuality, and later withheld the funding. Participants left with their names on a leaflet and their institutional positions exposed.

That day is the usual marker for the open break in American feminism over sexuality, though the argument had been building since the late 1970s. It is genuinely a dispute rather than a misunderstanding: two positions, each with a serious argument, each drawing on evidence the other accepts as real. This lesson sets them out from their own texts and then asks what would settle which parts.

The first position, in its own terms

The abolitionist argument does not begin from sexual morality, and reading it as prudishness guarantees you will misunderstand it.

It begins from the observation that in the commercial sex trade the sellers are overwhelmingly women and girls, disproportionately poor, disproportionately from state care, disproportionately having entered as minors, and the buyers are overwhelmingly men. That pattern is not incidental. It is what you would expect if what is being sold is sexual access from a group with fewer options to a group with more.

On pornography, the argument from the previous lesson applies: the material is produced by doing something to real people, and it teaches a script about what women are for. Robin Morgan compressed it in 1974 into a sentence that became the movement's slogan: pornography is the theory, and rape the practice. Women Against Pornography, founded in 1979, ran tours of Times Square so that supporters could see the trade rather than argue about it in the abstract.

On prostitution, Kathleen Barry's Female Sexual Slavery (1979) supplied the framework and the Coalition Against Trafficking in Women, founded in 1988, carried it into international politics. The empirical spine is harm data: Melissa Farley's nine-country study, published in 2003, interviewed 854 people in prostitution across five continents and found that 68 per cent met the diagnostic criteria for post-traumatic stress disorder, with rates of physical assault and rape in the trade far above population rates.

The policy conclusion is the demand-side or Nordic approach, first enacted in Sweden's Sex Purchase Act of 1999: criminalise buying, decriminalise selling, fund exit services. Norway and Iceland followed in 2009, Northern Ireland in 2015, France in 2016, Ireland in 2017. The reasoning is that if the trade exists because men are willing to pay, the person to prosecute is the payer.

And crucially: on this account consent is not the right question, because consent obtained by the absence of an alternative is not the free agreement the word usually names. As MacKinnon puts it, the question is not whether she agreed but what her options were.

The second position, in its own terms

The other side is not the view that pornography is harmless or that selling sex is pleasant. It is a set of arguments about law, enforcement, and who gets to describe whose life.

Ellen Willis, in an essay first published in 1979, made the argument that has held up best. The anti-pornography campaign, she wrote, reproduces a nineteenth-century picture in which male desire is dangerous, female desire is faint or absent, and women need protection from sex itself. That picture was the basis of the social purity movements, and it was never good for women who wanted anything the respectable script did not include. Willis's second point was about instruments: laws restricting sexual material are enforced by police, customs officers and courts, and those institutions have their own view of which sexuality is deviant.

That is not a hypothetical. In 1992 the Supreme Court of Canada, in R v Butler, rewrote Canadian obscenity law around harm rather than moral offence, a test that anti-pornography feminists had argued for and that a feminist legal organisation had supported in its intervention. Within months Canadian customs was using the new framework to seize shipments bound for Little Sisters, a gay and lesbian bookshop in Vancouver, which spent the next decade in litigation. The tool built to protect women was pointed at a sexual minority within weeks of being handed over.

Gayle Rubin's Thinking Sex, written for the Barnard conference and published in 1984, gave the position its theory. Societies, Rubin argues, operate a hierarchy of sexual value with a charmed circle at the centre, married, reproductive, monogamous, at home, free of commerce, and an outer limit beyond which practices are treated as pathology. Feminist analysis of gender, she argues, is not automatically an adequate analysis of sexuality, and when it is used as one it tends to relocate the same hierarchy with different practices in the outer ring.

The sex worker rights movement supplied the constituency. COYOTE was founded in San Francisco in 1973, the English Collective of Prostitutes in 1975, and an international committee issued a World Charter for Prostitutes' Rights in Amsterdam in 1985. Its demand was decriminalisation: not state licensing, but removal of the criminal offences that make it dangerous to work with another person, to screen clients, or to report an assault.

So what?: the two positions are not disagreeing about whether harm occurs. They are disagreeing about whether the harm is intrinsic to the transaction or produced by the conditions around it, and about whether the state can be trusted with the remedy.

What the legal experiments show

Two regimes have been running long enough to argue about.

New Zealand's Prostitution Reform Act 2003 removed the offences and set up a statutory review. The committee reported in 2008 that the number of people in the trade had not measurably increased, that most reported being better able to refuse a client and more willing to report violence, and that the worst conditions persisted at the street end where the workers had the fewest options. Sweden's law has been credited with a fall in street prostitution; what happened to indoor and online arrangements is much harder to measure, and sex worker organisations report displacement, less time to screen buyers, and landlords evicting on the strength of the surrounding offences.

Here is a detail routinely lost. The Nordic model decriminalises selling, but it usually leaves in place offences for brothel-keeping and for living on the earnings of prostitution. Two women sharing a flat so as not to see clients alone can therefore still be prosecuted, each as the other's brothel-keeper. Whatever you conclude about the model, the claim that it decriminalises the seller needs that qualification.

On trafficking, an econometric study by Cho, Dreher and Neumayer, published in World Development in 2013, found that countries with legalised prostitution report larger inflows of trafficked people, while the authors themselves stress that reported and actual trafficking are different quantities and that legalisation may raise reporting. That is an honest finding with an honest caveat, and it has been quoted without the caveat by both sides.

Who speaks

One thing that makes the dispute intractable is that both sides can point to women in the trade who agree with them. The English Collective of Prostitutes and international sex worker unions campaign for decriminalisation. SPACE International, founded by women who exited prostitution, campaigns for the Nordic model and objects to being described as unrepresentative. Neither group is a fabrication, and any argument that proceeds by declaring the other side's witnesses to be either brainwashed or unusually lucky has stopped being an argument.

Amnesty International's 2016 policy came down for decriminalisation, drawing formal opposition from a coalition of feminist organisations. Note that decriminalisation is not the same as legalisation: the Netherlands and Germany regulate and license, which creates a compliant sector and an illegal one outside it, and evidence about one regime does not transfer to the other.

Why this matters: when you read a claim about what the evidence shows in this field, check which of three regimes is being described, because criminalisation, legalisation and decriminalisation produce different data and are constantly conflated.

What would settle what

Sort the questions and the argument becomes tractable in parts.

• Does criminalising buyers reduce the total size of the trade, or move it? Answerable in principle, badly measured in practice, because the visible portion is the easiest to count and the one most affected.
• Does decriminalisation increase trafficking? Partially answerable, contaminated by the reporting problem the 2013 study names.
• Do workers report more or less violence under each regime, and are they more or less able to go to the police? The New Zealand review is the best evidence anyone has, and it is one country.
• Is a transaction consensual when the alternative is destitution? No dataset answers this. It is a question about what consent means, and it is where the dispute actually lives.

That last item is the one to notice. Both sides accept that coerced trafficking is a crime that should be prosecuted. The disagreement is about the remainder, and about whether economic pressure is a difference in degree from the pressure that gets anyone into any job, or a difference in kind because of what is being sold.

Common misconceptions

• Sex-positive feminists think pornography is harmless. Willis and Rubin argue about legal instruments and about who enforces them. Willis was explicit that much pornography is contemptible; her point was about what censorship does in practice.
• The Nordic model decriminalises sex workers. It decriminalises selling while usually retaining brothel-keeping offences, so working with another person for safety can remain unlawful.
• Amnesty International supports legalisation. Its 2016 policy supports decriminalisation, which is a different regime from the licensing systems of the Netherlands and Germany.
• Only one side has sex workers on it. Organisations of current workers and organisations of women who have exited both exist and both organise, which is why arguments from authentic voice cannot decide this.

Looking back

The abolitionist position holds that the sale of sexual access is a harm that a market cannot make benign, points at the demographics of who sells and who buys, and reaches for a law aimed at the buyer. The sex worker rights position holds that the harm is concentrated where criminalisation puts it, points at what obscenity and vice enforcement has historically done to sexual minorities and to the women it claimed to protect, and reaches for the removal of criminal law. Everything empirical in the dispute is under-measured; the part that is not empirical is a disagreement about what consent requires.

The core of it: you can hold that a practice is harmful and that criminalising it makes things worse, and most of the serious literature on both sides is an attempt to show that this combination is or is not coherent.

Sources

1. Wikipedia contributors. (n.d.). Feminist sex wars. Wikipedia. en.wikipedia.org
2. Wikipedia contributors. (n.d.). Prostitution Reform Act 2003. Wikipedia. en.wikipedia.org
3. Willis, E. (1983). Feminism, moralism, and pornography. In A. Snitow, C. Stansell, and S. Thompson (Eds.), Powers of Desire: The Politics of Sexuality (pp. 460-467). Monthly Review Press.
4. Rubin, G. (1984). Thinking sex: Notes for a radical theory of the politics of sexuality. In C. S. Vance (Ed.), Pleasure and Danger: Exploring Female Sexuality (pp. 267-319). Routledge and Kegan Paul.
5. Farley, M., and others. (2003). Prostitution and trafficking in nine countries. Journal of Trauma Practice, 2(3-4), 33-74.
6. Cho, S., Dreher, A., and Neumayer, E. (2013). Does legalized prostitution increase human trafficking? World Development, 41, 67-82.

On 29 May 1851, at a women's rights convention in Akron, Ohio, Sojourner Truth stood up and spoke. Two texts of what she said survive, and they are not the same document.

Marius Robinson, who was present and knew her, printed a report in the Anti-Slavery Bugle three weeks later. It contains no refrain, no dialect, and no children. Twelve years afterwards, in 1863, Frances Dana Gage published a version with a repeated question about whether she is a woman, a heavy Southern dialect, and a claim that Truth had borne thirteen children. Truth was born in Ulster County, New York, spoke Dutch as her first language, and had five children. The famous version is the one written by a white abolitionist a decade later, in a dialect its subject did not speak. The Global Women's History course takes the two documents apart side by side; what concerns us here is the argument, which is present in both.

The argument is this. The speakers Truth was answering had been urging that women are delicate and must be helped over ditches and into carriages, and that this delicacy is why they cannot vote. Truth points at herself: nobody has ever helped her into a carriage, she has ploughed and planted and borne the lash, and no one has offered her the protection now being offered as an argument. If she is a woman, the premise is false. If the premise is true, then the word woman in that sentence does not mean all women.

The point: a category that presents itself as universal is often indexed to a particular position, and the fastest way to expose it is to produce a member of the category for whom the description fails.

Between Akron and Boston

Anna Julia Cooper, born into slavery in 1858 and later the fourth African American woman to earn a doctorate, published A Voice from the South in 1892. Her famous formulation is that only the Black woman can say when and where she enters, in the quiet dignity of her womanhood, without violence and without suing or special patronage, and that then and there the whole race enters with her. Note the structure of the claim: it is not that Black women are worse off and deserve more sympathy. It is that their position is diagnostic, and that a settlement adequate to it is adequate to everyone.

The same year, Ida B. Wells published her investigation of lynching, and demonstrated the interlocking claim empirically rather than philosophically. Lynchings were justified by the alleged rape of white women. Wells collected the cases, showed how few involved any such accusation, and showed that the mythology of the endangered white woman and the mythology of the dangerous Black man were the same mechanism seen from two sides. A theory that treated race and sex as separate variables could not describe what she had documented.

Boston, 1977

The Combahee River Collective formed in Boston in 1974 out of the National Black Feminist Organization. It took its name from the raid Harriet Tubman guided up the Combahee River in June 1863, which freed more than seven hundred people in a single night, and the choice was deliberate: a Black woman leading a military operation, not a symbol of suffering.

In April 1977 Barbara Smith, Beverly Smith and Demita Frazier drafted a statement of a few thousand words that is now the most cited document in this literature. Four things in it matter for theory.

First, the analytical claim. The statement commits its authors to struggling against racial, sexual, heterosexual and class oppression together, and describes its particular task as developing an integrated analysis and practice based on the fact that these systems are interlocking. Interlocking is doing work there. An additive model says a Black woman suffers racism plus sexism, two harms in sequence. An interlocking model says the harms are not separable, because the racism is sexualised and the sexism is racialised: the stereotypes, the labour markets and the sexual violence are all specific to the joint position and do not exist in the separated ones.

Second, the political claim, which is the one now most often misquoted. The statement says that the most profound and potentially most radical politics come directly out of our own identity, as opposed to working to end somebody else's oppression. In context this is a claim about where reliable political knowledge and durable motivation come from: analyse the position you actually occupy. It is not a claim that membership of a group settles arguments, and the collective built coalitions constantly. This is the origin of the phrase identity politics, and the current usage has drifted a long way from it.

Third, the universalist conclusion. If Black women were free, the statement says, everyone else would have to be free, since their freedom would require the destruction of all the systems of oppression. This is Cooper's diagnostic move made into a political programme.

Fourth, what it rules out. The statement explicitly rejects biological determinism about men, calling the idea that men are inherently oppressive a dangerous and reactionary position, and explicitly rejects lesbian separatism as a strategy, on the ground that it leaves out too much of what has to be analysed. Both refusals are frequently ignored by people who cite the document.

hooks: who was the movement talking about

bell hooks drafted Ain't I a Woman: Black Women and Feminism at nineteen and published it in 1981. Her target is the account of what women want that had been generalised from a specific stratum.

Take Betty Friedan's problem that has no name: the boredom and waste of an educated woman confined to a suburban house. hooks does not deny that the problem is real. She points out whose problem it is. It belongs to women with houses, husbands earning enough for one income, and educations they were not using. For the women cleaning those houses, paid work was not an unopened door; it was the thing they had always done, for other people, for very little. A movement whose central demand is access to employment is speaking, whether it knows it or not, from inside a particular household.

Feminist Theory: From Margin to Center (1984) turns that into an epistemic claim, and it is the bridge to Module 6. Living at the margin, hooks writes, means seeing both the margin and the centre, because you must understand the centre in order to survive it, while the centre can go a whole life without noticing the margin at all. The margin is therefore a site of deprivation and a site of a wider view at the same time.

hooks also gave the movement a definition that avoids the trap the module is about: feminism is the movement to end sexism, sexist exploitation and oppression. Defined by the practice rather than by a constituency, it does not have to specify in advance which women it is about.

Collins: turning a critique into a research programme

Patricia Hill Collins published Black Feminist Thought in 1990, with a second edition in 2000 and a third in 2022. Its contribution is machinery.

The matrix of domination is her name for the way race, class, gender, sexuality and nation operate as one system rather than several. The part usually left out of summaries is how she says it operates, through four domains of power, and the four are what make the concept usable.

Her 1986 article introduced a second idea: the outsider within. Black women employed as domestic workers spent their days inside white households, seeing the intimate operation of a class and a race from a position of intimacy without membership. That location, she argues, produces a distinctive angle of vision, and she generalises it to Black women entering academic disciplines built without them.

Collins is careful about a trap that this literature walks past frequently. She does not say that being a Black woman gives you Black feminist consciousness. She says that a shared location makes certain things visible and that a standpoint is achieved by working on that material collectively, through struggle and argument. Consciousness is an accomplishment, not a possession, which is why she can also say that people who do not occupy the position can learn the analysis.

Worth holding on to: the difference between an achieved standpoint and an owned identity is the difference between a claim you can argue about and a claim you cannot.

What is contested

Frances Beal's 1969 pamphlet on double jeopardy and Deborah King's 1988 reframing as multiple jeopardy show the tradition working out the arithmetic in public: additive, then multiplicative, then interlocking. Critics ask whether the matrix generates predictions or merely licenses adding a category whenever one is noticed. Collins's four domains are the best answer available, because they specify where to look; whether they yield testable claims is a fair question and one you should ask of any framework, including this one.

Common misconceptions

• The famous Akron speech is a transcript. The version with the refrain and the dialect was published twelve years later by Frances Gage. The contemporary report by Marius Robinson reads quite differently, and Truth did not speak in Southern dialect.
• Identity politics originally meant that identity settles arguments. The Combahee statement used it to mean grounding analysis in the position you occupy, and the collective worked in coalitions throughout.
• Black feminist thought is an additive claim about suffering more. The claim is that the systems interlock, so that the harms at the joint position are specific and do not appear at either single position.
• Collins holds that only Black women can understand this. She distinguishes a shared location from an achieved standpoint, and treats the analysis as learnable.

Recap

Truth's speech exposed a universal category that was quietly particular. Cooper turned that into a diagnostic principle, and Wells demonstrated it with lynching statistics. The Combahee statement named the systems interlocking, coined identity politics in a sense narrower than the current one, and drew the universalist conclusion. hooks showed whose experience the movement's central demands assumed and turned marginality into an epistemic position. Collins built the machinery: a matrix operating through four domains, an outsider-within location, and a standpoint that has to be achieved.

In short: the module's claim is not that some women have it worse. It is that a theory built on one position will produce categories that fail at others, and that the failures are where the theory needs rebuilding.

Sources

1. Combahee River Collective. (1977). The Combahee River Collective Statement. BlackPast. blackpast.org
2. Wikipedia contributors. (n.d.). Combahee River Collective. Wikipedia. en.wikipedia.org
3. Heyes, C. (2023). Identity politics. Stanford Encyclopedia of Philosophy. plato.stanford.edu
4. Collins, P. H. (2000). Black Feminist Thought: Knowledge, Consciousness, and the Politics of Empowerment (2nd ed.). Routledge.
5. Collins, P. H. (1986). Learning from the outsider within: The sociological significance of Black feminist thought. Social Problems, 33(6), S14-S32.
6. hooks, b. (1984). Feminist Theory: From Margin to Center. South End Press.

In 1976, five women sued the General Motors assembly plant in St. Louis. Emma DeGraffenreid was the named plaintiff. The facts were not in dispute. General Motors had hired no Black women before 1964. It had hired Black men, mostly into the foundry. It had hired women, white women, into clerical work, since 1947. Then, in the recession that began in 1970, the plant made layoffs on a last-hired, first-fired seniority rule. That rule removed every Black woman the plant had hired.

The case is DeGraffenreid v. General Motors Assembly Division, 413 F. Supp. 142. It is worth walking through the court's reasoning slowly, because nothing in it is stupid.

The court's steps

Step one. The claim is brought under Title VII of the Civil Rights Act of 1964, which prohibits discrimination because of race, colour, religion, national origin or sex. Those are the available categories, and the plaintiff must fit her claim into them.

Step two. Take the sex claim on its own. Has General Motors discriminated against women? The record shows women hired continuously since 1947. On the face of the hiring data, no.

Step three. Take the race claim on its own. Has General Motors discriminated against Black people? Black men were hired. The court sent that part of the case to be consolidated with a separate pending action about race at the same employer.

Step four. What remains is a claim that the plaintiffs were discriminated against as Black women specifically. The court declined to recognise it, and gave reasons. Allowing it, the opinion says, would let plaintiffs combine statutory remedies to create a new super-remedy going beyond what the drafters intended, and it warned that new protected classes generated by permutation and combination would open a Pandora's box.

Notice what has happened by step four. Each step is defensible. The conclusion is that a group of workers who were the only group excluded from hiring, and who were then removed in a single layoff, have suffered no legally recognisable discrimination. The failure is not in any one step. It is in the requirement that the harm be assigned to exactly one category.

Key idea: a framework can be internally consistent, apply its own rules faithfully, and still be structurally incapable of seeing a harm that actually occurred.

The paper

Thirteen years later, Kimberle Crenshaw published Demarginalizing the Intersection of Race and Sex in the University of Chicago Legal Forum. It opens with DeGraffenreid, and it introduces the image that gave the concept its name.

Think of race and sex as roads meeting at a crossing. Traffic runs in all four directions. An accident at the intersection can be caused by cars coming from any of them, or from several at once. If a Black woman is injured there, the harm may have come from any direction; but the law requires her to establish that it came from one particular street, and if she cannot, she recovers nothing.

Crenshaw then reads two more cases, and their combination is the argument.

• In Moore v. Hughes Helicopters (1983), a Black woman sought to represent a class in a promotion case. Because she had framed her claim on behalf of Black women, the court held she could not adequately represent all women, and it rejected her statistics comparing Black women with other groups. Specifying the intersection cost her the wider class.
• In Payne v. Travenol (1982), two Black women sued on behalf of all Black employees at a pharmaceutical plant. The court let them represent Black women but not Black men. Specifying the intersection bounded the class from the other side.

So: assert the intersection as a distinct claim and it is refused as a super-remedy. Assert it while seeking to represent women and you are told you are unrepresentative of women. Assert it while seeking to represent Black employees and you are told you cannot speak for the men. The doctrine is not hostile in any single instance; it is unable to hold the position at all.

Crenshaw's formulation is that the intersectional experience is greater than the sum of racism and sexism, and that an analysis which does not take the intersection into account cannot address the particular manner in which Black women are subordinated.

The second half of the paper, which people skip

The article is not only about courts. Its subtitle names feminist theory and antiracist politics as targets too, and the argument there is structural rather than moral.

A movement organised around a group tends, Crenshaw argues, to select its agenda from the experiences of members who are otherwise privileged, because those are the members for whom the single axis is the only obstacle. A feminism whose paradigm case is a woman who would have the job but for her sex will produce demands calibrated to that case. An antiracist politics whose paradigm case is a man who would have the job but for his race does the same. Each then produces remedies from which the doubly burdened benefit least, and each can honestly report that it has advanced the interests of its constituency.

Her 1991 sequel, Mapping the Margins, applies this to violence against women and distinguishes three levels. Structural intersectionality is the way a remedy misses because of a person's position: a shelter organised for English-speaking women with independent legal status is not available to a woman without either. Political intersectionality is what happens when a woman's two constituencies make competing demands: reporting domestic violence to the police may be framed as betraying a community already over-policed. Representational intersectionality concerns how such women appear, or fail to appear, in public arguments about the issue.

Her example from immigration law is exact. Amendments in 1986 conditioned permanent residency on remaining married for two years. A woman whose husband was violent had to choose between the marriage and her status. A waiver added in 1990 required documentation of the abuse, of a kind that a woman isolated at home, without English, without her own money and without access to doctors or lawyers, was structurally least able to obtain. The remedy existed and was shaped so that the people it was for could not reach it.

Remember: intersectionality began as a claim about how frameworks fail, in law and in movements, not as a claim about who a person is.

Three things it does not claim

The concept has travelled far enough from its origin that the misreadings need naming.

It is not a ranking of suffering. Nothing in either paper says the more categories you belong to the more authority you have; Crenshaw has repeatedly rejected that reading, and it does not follow from anything in the argument.

It is not a claim that identity determines belief. The cases are about how institutions sort people, not about what those people think. A Black woman who disagrees with Crenshaw is not a counterexample to anything in the paper.

It is not an invitation to add categories without limit. Crenshaw was writing about a specific doctrinal failure with a specific remedy in view. Whether the concept can support an open-ended list is precisely what her critics ask.

The critics

Jennifer Nash's 2008 article in Feminist Review is the most cited internal critique. She raises four problems: intersectionality has no agreed methodology, so it is unclear what doing an intersectional analysis requires; it treats Black women as the prototypical intersectional subject, which reintroduces a paradigm case; it is ambiguous about whether it is a theory of marginalised positions or of all identity, and these have different implications; and it sits awkwardly with accounts of lived experience that do not divide neatly into strands.

Jasbir Puar presses a different objection. Intersectionality, she argues, keeps the categories it set out to complicate: it starts from race and gender as given lines and then examines their crossing, which reinstalls the grid. Her alternative vocabulary of assemblage aims at something that does not begin from fixed identity terms at all. Whether that vocabulary can do the work a court needs is a fair counter-question.

There is also a technical objection. If intersection means interaction rather than addition, quantitative work has to estimate interaction effects, and interaction effects need larger samples than main effects, especially where the intersecting group is small. That is a real constraint on turning the concept into measurement, and it is a live methodological literature rather than a refutation.

And there is the doctrinal question the 1989 paper raised and nobody has closed. Section 14 of the UK Equality Act 2010 created a claim for combined discrimination on two protected characteristics. It was passed and then never brought into force. A statute that recognises the problem and sits unused is a fairly precise measure of how hard the drafting is.

Common misconceptions

• The DeGraffenreid court was hostile to civil rights plaintiffs. It applied the categories the statute gave it, one at a time, and reached an absurd result by careful steps. That is what makes the case useful.
• Intersectionality means the more identities you have, the more authority you hold. Neither paper says this and Crenshaw has rejected it explicitly. The argument is about how institutions sort harms.
• It is a general theory of identity. It began as an analysis of antidiscrimination doctrine and of movement agenda-setting. Whether it generalises is the question Nash raises rather than a settled fact.
• Legal systems have since fixed the problem. The British statutory provision written to fix it has never been commenced, and American courts remain divided on intersectional claims.

What you now know

DeGraffenreid shows a framework applying its own rules correctly and failing to see the only group that was excluded. Crenshaw's 1989 paper names the mechanism, reads two further cases showing the failure from the other two directions, and extends the analysis from courts to movements, where remedies get calibrated to the members for whom one axis is the only obstacle. The 1991 paper distinguishes structural, political and representational versions of the problem and gives the immigration waiver as a case where a remedy was built so that its intended beneficiaries could not reach it. The critics ask for a method, question the prototypical subject, and doubt whether a grid of categories can escape being a grid.

The upshot: read intersectionality as a diagnostic tool with a specific job, and it survives most of the objections; read it as a general theory of persons, and most of the objections land.

Sources

1. Crenshaw, K. (1989). Demarginalizing the intersection of race and sex: A Black feminist critique of antidiscrimination doctrine, feminist theory and antiracist politics. University of Chicago Legal Forum, 1989(1), Article 8, 139-167. chicagounbound.uchicago.edu
2. United States District Court, E.D. Missouri. (1976). DeGraffenreid v. General Motors Assembly Division, 413 F. Supp. 142. CourtListener. courtlistener.com
3. Wikipedia contributors. (n.d.). Intersectionality. Wikipedia. en.wikipedia.org
4. Crenshaw, K. (1991). Mapping the margins: Intersectionality, identity politics, and violence against women of color. Stanford Law Review, 43(6), 1241-1299.
5. Nash, J. C. (2008). Re-thinking intersectionality. Feminist Review, 89, 1-15.

On 17 November 2001, six weeks into the bombing of Afghanistan, Laura Bush delivered the weekly American presidential radio address. It was the first time a first lady had given the whole address alone. The fight against terrorism, she said, is also a fight for the rights and dignity of women.

Something in that sentence is doing work that the sentence does not admit to. Working out exactly what is the business of this lesson, because the answer is not that the speaker was insincere, and it is not that the Taliban's treatment of women was acceptable. Both of those replies dodge the problem. The problem is structural, and three writers have taken it apart.

The composite figure

Chandra Talpade Mohanty published Under Western Eyes in 1984 and in a revised form in 1988. Its subject is a body of feminist scholarship, and its method is to examine what that scholarship produces rather than what it intends.

What it produces, she argues, is a singular figure: the third world woman. She is poor, uneducated, tradition-bound, domestic, family-oriented, religious, and above all a victim. She has no history and no politics of her own. She is defined by what has been done to her. And she exists in the writing only by contrast with an implied other woman, educated, modern, in control of her body and her sexuality, free to decide.

Mohanty identifies three operations that generate her.

• Women are treated as a group already constituted before entering any social relation, with interests that follow from being women rather than from the specific arrangements they live in. Once that is assumed, the differences between an urban professional in Lagos and a landless labourer in Bihar become variations on a single condition.
• Method is used to prove what was assumed. A practice is counted across countries, and the count is taken to show a universal. Her example is veiling. In Iran in 1979 middle-class women who did not habitually cover put on the veil to signal solidarity in a revolutionary movement; under a regime that mandates covering, the same garment on the same body means something close to the opposite. A study that codes veiled as one variable has erased the only thing that mattered.
• A relation of knowledge is set up. One group is the subject who studies, theorises and knows; the other is the material studied. That relation then reappears in politics as one group acting and the other being acted for.

What matters here: Mohanty's charge is not that Western feminists were unkind. It is that a particular way of writing manufactures a subject who requires rescue, and that the manufacturing is invisible to the people doing it.

Nineteen years later she revisited the essay in Signs, and the revision matters as much as the original. Her critique, she wrote, had been taken up in ways that discourage any cross-cultural analysis at all, as though the safe course were to say nothing about anyone else. That was not the argument. She reasserted the need for comparison, for a transnational analysis attentive to capital, and for solidarity that crosses borders, on the grounds that the alternative leaves the field to whoever is already speaking.

Spivak, and what speaking requires

Gayatri Chakravorty Spivak's Can the Subaltern Speak? is a hard essay that is usually reduced to its title, and the reduction gets it backwards.

Her central case is the British abolition of sati in 1829. Two accounts of that event were available at the time and remain available now. The colonial account is what Spivak compresses into a sentence: white men are saving brown women from brown men. The nativist counter-account is that the women wanted to die, that this was their tradition and their choice, and that the British were destroying it.

Her point is what these have in common. Both are arguments made about the woman by parties with institutional standing, and in both she appears as evidence for a position rather than as a party to the argument. There is no third archive containing her own account, because the discursive space in which such an account would be registered as an account did not exist.

That is what the question in the title means. It is not a claim that subaltern people are mute. It is a claim about the conditions of speech: to speak, in the sense that matters politically, requires a hearer positioned to receive what is said as speech rather than as data, symptom or noise. The subaltern position is defined by exclusion from those circuits. Spivak's example is a relative of her own, Bhubaneswari Bhaduri, who hanged herself in Calcutta in 1926 and waited for her menstruation before doing so, apparently so that the death could not be read as the consequence of an illicit pregnancy. It was read that way for decades regardless. She had written her act as legibly as she could, and it was still received as something else.

Spivak also warns against a specific fix. She criticises Foucault and Deleuze for assuming that the oppressed can simply represent themselves and that intellectuals need only step aside, pointing out that representation has two meanings, standing in for someone politically and portraying someone in words, and that stepping aside from the second does not dissolve the first. The intellectual who claims to be a transparent conduit is still doing the selecting.

Her other well-known coinage, strategic essentialism, is the temporary use of a group identity known to be constructed, for a definite political purpose. She has since largely disowned the term, because it was taken as a permit rather than a tactic with an expiry date.

Abu-Lughod, and the garment

Lila Abu-Lughod's 2002 article, which opens with that radio address, presses on the word saving. To save someone is to imply that you are saving her to something, and to something better; the verb carries a destination. It also positions the saver as free and the saved as not, which forecloses the question of what constrains each of them.

Her concrete case is the Afghan chadari, the pleated garment covering the head and body that Western coverage treated as the emblem of the Taliban. It was not a Taliban invention. It was a regional garment, associated with Pashtun respectability, worn long before 1996 and understood by many women who wore it as portable seclusion, a way of moving through public space while keeping the separations their communities expected. When Kabul fell in November 2001, women did not throw it off in the streets, and commentators who had read the cloth as a shackle found this incomprehensible.

Abu-Lughod is careful about what follows. She is not a relativist and says so. Her argument is that the right questions are historical and political: what are the conditions in which these particular women are living, who supplied the weapons, what happened to the schools and the clinics, and what will the intervention that is being justified in their name actually do. She develops the case at book length in 2013.

Saba Mahmood's study of the women's mosque movement in Cairo adds a further turn. Liberal analysis, she argues, recognises agency when it sees resistance to norms and is unable to see it anywhere else. The women she worked with were cultivating piety, docility and modesty deliberately, over years, with effort and skill. That is agency by any ordinary description of the word; it simply is not resistance, and a framework that can only detect one kind will report that these women have none.

Uma Narayan supplies the sharpest test, which she calls death by culture. Dowry murder in India is routinely explained by reference to Indian culture. Domestic homicide in the United States, which she argues occurs at a broadly comparable order of magnitude relative to population, is explained by individual pathology, guns, or the failures of particular men. Same category of harm, two entirely different explanatory registers, and the difference tracks which society is being described rather than anything in the phenomenon.

Bottom line: the critique is not that outsiders may not judge. It is that explanation gets rationed: culture for them, circumstance for us, and the asymmetry does the political work.

The objection this position has to answer

If every cross-cultural judgement reproduces a colonial relation, then nothing can be said about forced marriage, or about a state that bars girls from school, and the critique becomes a shield. States and elites have in fact used it that way, answering criticism by declaring the practice cultural and the critic foreign.

Martha Nussbaum has pressed this case hard, arguing that a refusal of universal standards leaves the worst off with no vocabulary in which to complain, and that the capabilities list in Module 1 is an attempt at a universal minimum that is thin enough to be filled in locally.

The answers offered here are worth stating precisely, because they are not evasions. Mohanty's is that the critique was aimed at a way of producing knowledge, not at knowing; comparison is required, and the requirement is that the categories be built from the situations rather than imposed on them. Abu-Lughod's is that you can hold a view about an injustice and still ask what your intervention will do, and that the two questions are separate. And Spivak's is the least comfortable: that the problem cannot be solved by good intentions, only worked at, by attending to whose speech your institutions are built to register.

Common misconceptions

• Mohanty was attacking Western feminists personally. Her object is a discursive operation visible in texts, which is why her essay proceeds by reading scholarship rather than by assessing motives.
• Can the subaltern speak? means subaltern people are silent. It means that speech requires a position from which it registers as speech, and that the subaltern position is defined by not having one.
• Postcolonial feminism forbids criticism of other societies. Mohanty's 2003 essay explicitly rejects that reading, and Abu-Lughod states that she is not a relativist.
• The Afghan chadari was introduced by the Taliban. It long predates them as a regional garment, which is why the expectation that it would be discarded after 2001 misread what it was.

Where this leaves us

A sentence about fighting for the rights of women can be true in every particular and still do three things at once: produce a composite victim, position the speaker as the one who knows, and supply a justification for something else. Mohanty names the operations that build the composite. Spivak shows that both the colonial and the nativist account of sati speak about a woman rather than with her, and that the fix is not to hand the microphone over and step back. Abu-Lughod asks what saving implies and what the intervention does. Mahmood widens what counts as agency; Narayan shows that explanation itself is distributed unevenly.

So what?: the test to carry away is simple to state and hard to pass. When you read a claim about women somewhere else, ask what the same claim would look like written about your own society by someone with an interest in intervening in it.

Sources

1. McLaren, M., and others. (2023). Feminist perspectives on globalization. Stanford Encyclopedia of Philosophy. plato.stanford.edu
2. Wikipedia contributors. (n.d.). Postcolonial feminism. Wikipedia. en.wikipedia.org
3. Mohanty, C. T. (1988). Under Western eyes: Feminist scholarship and colonial discourses. Feminist Review, 30, 61-88.
4. Mohanty, C. T. (2003). Under Western eyes revisited: Feminist solidarity through anticapitalist struggles. Signs, 28(2), 499-535.
5. Spivak, G. C. (1988). Can the subaltern speak? In C. Nelson and L. Grossberg (Eds.), Marxism and the Interpretation of Culture (pp. 271-313). University of Illinois Press.
6. Abu-Lughod, L. (2002). Do Muslim women really need saving? Anthropological reflections on cultural relativism and its others. American Anthropologist, 104(3), 783-790.

Bodies That Matter opens with a scene of interruption. Judith Butler reports that every attempt to write about the constructed character of gender ran into the same objection from readers, colleagues and audiences: what about the materiality of the body? Butler noticed something about how the question tended to arrive. It came with their first name shortened to Judy, as though the diminutive were part of the correction, a way of calling the theorist back down to the flesh they were accused of having written away.

The objection was not stupid. It was the compressed form of a misreading so widespread that Judith Butler spent a second book and then a new preface trying to undo it. So this lesson runs backwards. Start from the wrong answer, state it in the version its holders would sign, and trace where the reasoning breaks. Once you see the break, the actual claim is narrower and more interesting than its reputation.

The wrong answer, given its best shot

Here is the reading. Gender is a performance. You are not a woman or a man; you do woman or man, through clothes, gait, pitch, the angle at which you hold a cigarette. Because it is something you do, you could do it otherwise. Drag is the proof: the performer wears femininity as a removable costume, which shows that everyone else is wearing one too. The politics follows. Parody your gender, wear the costume ironically, and a system that depends on your sincerity begins to loosen.

This reading has real support in the text. Gender Trouble (1990) discusses drag at length, uses the vocabulary of imitation and parody, and closes on subversion. It is not a lazy misreading. It is the reading the book invites if you take one word in its ordinary English sense.

That word is performative. In its 1999 preface Butler names the resulting picture and rejects it: the idea that one wakes, surveys a wardrobe, selects a gender for the day and returns it at night. The objection is not that this would be frivolous. It is that a gender chosen that way would not be a gender, since nothing in the account explains why the choice would be recognised by anyone else, or what happens to you if you choose wrongly.

Where the fault is: performative does not mean theatrical

The word comes from J. L. Austin, who used it in lectures at Harvard in 1955, published in 1962 as How to Do Things with Words. Austin's target was the assumption that sentences describe. Some do not. When a registrar says the words that make two people married, when an official names a ship, when a person says I promise, no report is being filed. The saying is the doing.

Note the condition Austin attaches. These utterances work only when a great deal is already in place: a convention with an accepted form of words, the right speaker, the right circumstances. Austin calls the failures infelicities. A person with no authority who declares two strangers married has not married them badly; they have not married them at all. A performative is therefore the least free kind of speech there is, since its power comes from an institution standing behind it.

The core of it: performative in Austin's sense means an utterance that acts because a convention licenses it. Reading it as theatrical inverts the concept, turning the most convention-bound speech into the most improvised.

Jacques Derrida sharpened this in a 1972 paper on Austin. What makes a performative work, he argued, is that it is a citation: the form of words has been used before and can be used again, by anyone, in the speaker's absence. He called this iterability. The utterance is effective precisely because it is not original. Butler takes this over wholesale, and it is the hinge of the whole account.

The claim, in four steps

With the vocabulary repaired, here is what gender performativity actually asserts.

1. There is no gender identity sitting behind the acts that are said to express it. Butler puts it as a flat denial: identity is constituted by the very expressions taken to be its results. The model in the background is Nietzsche's remark that there is no doer behind the deed, that the doer is invented and added to the deed.
2. The acts are not chosen. Butler's phrase in Gender Trouble is a stylised repetition of acts: repetition, under constraint, according to forms already laid down, with sanctions for departure. Nobody invents the way a girl is expected to sit.
3. Repetition is what produces the appearance of a stable inner thing. Do something enough times, under enough regulation, and it stops looking like something done and starts looking like something you are. Butler calls this a substance effect, and it is the illusion the theory is built to explain.
4. Because it has to be repeated, it can go wrong. A norm that needed no maintenance would not need to be repeated at all. The room for change is in the gap between one citation and the next, not in a subject who steps outside the whole business and chooses.

Step four is where the misreading does its most expensive damage. On the wardrobe picture, agency belongs to a chooser who exists before the norm. On Butler's picture there is no such chooser, and what looks like a loss is the argument's only source of hope: the norm is fragile in the way anything requiring constant reproduction is fragile.

It is a girl, and everything that follows

The clearest illustration is in Bodies That Matter (1993). A body is delivered. Someone says it is a girl, or now, more often, says it months earlier while reading a screen. Look at what that sentence does. Grammatically it reports. Practically it moves the infant from it to she, and that shift is not a description of a state already obtaining. It starts a series.

The series is what matters. The naming is the first of a very long chain of citations: the form, the clothes bought, the pronoun used by strangers, the school register, the passport, the medical record, the thousand small corrections administered across a life to anyone who drifts. Butler borrows the term interpellation from Louis Althusser, who described ideology as hailing a person the way a policeman calls out in the street, so that turning round is what makes you the one addressed. The girling of the girl is that hail, extended over decades and never quite finished.

Why this matters: the claim is not that a sentence conjures a body out of nothing. It is that classification is an act with consequences, that it has to be renewed, and that the renewal is visible in the ordinary bureaucratic machinery anyone can go and look at.

Reopening the distinction Beauvoir seemed to settle

Module 1 left you with Simone de Beauvoir's formula that one is not born but rather becomes a woman, and Module 3 with Gayle Rubin's sex and gender system of 1975, in which every society takes the raw biological material of sex and works it into gender. Both leave sex intact underneath: sex is nature, gender is what culture makes of it, and the task is to change the second.

Butler asks the awkward question: does sex have a history? Not, does the body exist, but, is the two-part sorting of bodies something we find or something we make while claiming to find it? If male and female are already doing normative work at the moment of description, the supposedly raw material is not available for inspection independently of the culture that sorts it. And then, as Butler puts it, the distinction turns out to be no distinction at all.

This is where the medical record earns its place. Anne Fausto-Sterling, a developmental biologist rather than a philosopher, documented in Sexing the Body (2000) how mid-century clinical protocols for infants with atypical genitalia worked. The decision rules were written in measurements, and the thresholds differed depending on which sex the child was to be assigned. Surgery then produced the anatomy the classification had presupposed. Whatever you conclude about the ethics, the epistemology is hard to dodge: here the description was enforced on the body rather than read off it.

What Butler does not say, and is constantly accused of saying, is that bodies are unreal. The argument is about access: there is no vantage point from which a body is met outside some framework for describing it, and the available frameworks are not innocent. That is a claim about knowledge, not a denial of matter.

Common misconceptions

• Gender is a costume you choose each day. Butler names and rejects this in the 1999 preface. Choice cannot explain why the wrong choice is punished, and punishment is the phenomenon the theory exists to account for.
• Butler says gender is drag. Drag is used to expose an imitative structure that has no original. Bodies That Matter states plainly that drag is not inherently subversive and can just as easily reinforce what it repeats.
• Performative means performed, as on a stage. It is Austin's sense, and Austin set stage utterances aside as a case where the ordinary conditions are suspended.
• Butler denies that bodies exist. The claim concerns the terms on which bodies become knowable and classifiable, not their existence.

Four objections that have cost the position ground

The most damaging came from inside feminist philosophy. Martha Nussbaum, reviewing Butler in The New Republic in February 1999, argued that the politics on offer is symbolic: parodic resignification changes how a norm is inhabited and leaves untouched the wage, the statute, the hospital and the household. She also charged that the prose insulates the argument from testing, which landed harder because a sentence of Butler's had won a bad writing competition run by the journal Philosophy and Literature the year before.

Susan Bordo, sympathetic to construction, made the second objection in Unbearable Weight (1993): the body keeps thinning out into text. Her subject was anorexia and cosmetic surgery, where norms are enforced on flesh that gets hungry and scars, and a purely discursive account cannot say why that flesh matters.

The third comes from Module 2. If gender is produced in discourse, what explains the distribution of unpaid hours in a household, or the fact that the occupations women dominate pay less? A materialist wants a mechanism with money in it, and citation is not obviously one.

The fourth is constructive rather than hostile. Sally Haslanger keeps the construction claim and moves it: what constructs gender is social structure, not discourse, so that to be a woman is to occupy a position subordinated on the basis of presumed female reproductive role. The definition is deliberately political, and it gives feminism a subject matter that survives disagreement about what people call themselves.

Butler's replies are worth knowing. To Nussbaum: construction never meant free construction, and the constraint is the entire point. To Bordo: Bodies That Matter is the answer, and it argues that materiality is not outside power but the effect of power's most sedimented operations. To the materialists: the later work, from Undoing Gender (2004) onward, is about who is legible enough to be housed, treated and mourned, which is not a purely symbolic list.

The short version

The reading that gets Butler wrong takes performative in its everyday theatrical sense and infers a chooser standing behind the gender, free to swap it. Austin's sense inverts that: a performative acts because a convention backs it, which makes gender a compelled citation rather than an outfit. From there the pieces follow. No identity behind the acts. Repetition producing the appearance of substance. The announcement at birth as the first move in a chain maintained by schools, forms and strangers. And the sex and gender distinction reopened, because the sorting of bodies is not obviously something we merely find.

In short: the theory buys its explanation of how gender binds so tightly at the price of making it hard to say what, if anything, is being oppressed. That bill comes due in the next lesson, where two feminisms with incompatible answers to that question have been arguing for fifty years.

Sources

1. Mikkola, M. (2022). Feminist perspectives on sex and gender. Stanford Encyclopedia of Philosophy. plato.stanford.edu
2. Wikipedia contributors. (n.d.). Gender performativity. Wikipedia. en.wikipedia.org
3. Butler, J. (1990). Gender Trouble: Feminism and the Subversion of Identity. Routledge. See also the anniversary edition preface of 1999.
4. Butler, J. (1993). Bodies That Matter: On the Discursive Limits of Sex. Routledge.
5. Austin, J. L. (1962). How to Do Things with Words. Oxford University Press.
6. Fausto-Sterling, A. (2000). Sexing the Body: Gender Politics and the Construction of Sexuality. Basic Books.

Beth Elliott sang at UCLA over the Easter weekend of 1973 before the room voted on whether she should be allowed to finish. She was a folk musician, a former vice-president of the San Francisco chapter of the Daughters of Bilitis, and a member of the committee that had organised the conference she was performing at. A separatist collective had leafleted the venue that evening saying she was a man. Counting roughly thirteen hundred women took over an hour. The vote went about two to one in her favour, she finished, and the next day a keynote speaker denounced her from the platform.

The dispute that evening was not about whether trans women exist or whether they suffer. It turned on a prior question neither side had settled: what makes the category woman the right unit for a political movement, and who decides what falls inside it. Fifty years on the same argument runs in employment tribunals, sports federations, statutes and supreme courts, and it still returns to that question.

The trans feminist case, from Stone, Stryker and Serano

Sandy Stone wrote The Empire Strikes Back: A Posttranssexual Manifesto in 1987 and published it in 1991, as a reply to a book that had named her personally. Her argument turns on evidence any reader of the clinical literature of the 1960s and 1970s can check. To obtain treatment, patients had to satisfy gatekeepers that they matched a diagnostic profile. They learned the profile, passed the textbooks around waiting rooms, and produced the required autobiography. The clinics then wrote those narratives up as findings. Stone's conclusion is not that patients lied but that the standard transsexual life story was in part manufactured by the screening that recorded it, and that the goal it enforced, disappearing perfectly into the target sex, left trans people unable to speak as themselves.

Susan Stryker's 1994 essay in GLQ takes the sharpest available insult and refuses to deny it. Trans bodies had been called unnatural, technologically produced, monstrous. Stryker takes up Frankenstein's creature and speaks from inside it, and the move is not merely rhetorical: the creature's rage is directed at a maker who assembled it and then disowned what he had made. The payoff is a challenge. Every body is produced, by hormones, by surgery, by diet, by the apparatus that assigns and records it. What distinguishes the trans body is not that it was made but that the making is visible.

Julia Serano, a biologist, gave the position its most systematic form in Whipping Girl (2007). Her central move is to split sexism in two. Traditional sexism ranks maleness above femaleness. Oppositional sexism is the separate belief that male and female are rigid, mutually exclusive categories with nothing legitimate between them. The distinction generates a prediction: if both operate, the greatest hostility should fall on those who move from the higher term toward the lower one and breach the boundary as well, so on trans women rather than trans men. Serano calls that combination transmisogyny, and argues it from the treatment of trans women in comedy, journalism and pornography.

Serano also declines two easy positions. She rejects the pure medical model, in which a trans person is a patient with a condition, and she rejects accounts in which gender identity is entirely learned. What she proposes instead she calls intrinsic inclinations: gender identity, gender expression and sexual orientation are separate, largely intrinsic, and not reliably correlated with one another or with anatomy.

Talia Mae Bettcher supplies the argument about violence. In a 2007 paper she describes a double bind. A trans woman who does not disclose is cast as a deceiver, and the deception treated as grounds for rage; one who does disclose is cast as a pretender, a man playing at being a woman. There is no third option, because both readings rest on one premise, that genitals constitute the concealed truth of a person's sex. Bettcher's point is that this premise is not incidental to transphobic violence but is the structure that supplies its justification, which is why the same defence recurs in the trials that follow such killings.

Remember: the trans feminist claim is not that sex is imaginary. It is that the enforcement of gender norms is what feminism opposes, that trans people are among those it falls on hardest, and that a movement against that enforcement cannot coherently perform it.

The gender-critical case, from Raymond, Jeffreys and Stock

Janice Raymond's The Transsexual Empire (1979) makes an argument worth separating from its most notorious sentence. Raymond's target is an institution: a medical and psychiatric complex that meets people in real distress at rigid sex roles and responds with surgery on the individual. On her account this treats a social problem as a private pathology and, by requiring patients to demonstrate conformity to a role in order to qualify, deepens the roles that caused the distress. The book also contains a passage accusing trans women of violating women's bodies by appropriating the female form, which Raymond has restated rather than withdrawn, and it has determined how the book is read ever since.

Sheila Jeffreys in Gender Hurts (2014) draws the conclusion Raymond's premises point at. If gender just is the hierarchy that subordinates women, any politics affirming gender identities entrenches the machinery rather than dismantling it, and the aim should be abolition rather than a longer list.

The most careful contemporary statement is Kathleen Stock's Material Girls (2021), and it goes in steps. First, sex is a real biological category, bimodal rather than a spectrum, with genuine intersex conditions that are rare and do not dissolve the classification. Second, the disadvantages feminism has documented, from unpaid care to sexual violence to the pay gap, track sex rather than declared identity, so an apparatus that stops recording sex loses the ability to measure them. Third, gender identity understood as an inner sense of being a man or a woman is either unverifiable or cashes out in stereotype, which is what feminism set out to attack. Fourth, Stock proposes fictionalism: treating a trans woman as a woman is often a useful and humane social pretence, adoptable widely without settling contested questions about prisons, sport, data or rape crisis provision, which she thinks should be decided by weighing interests case by case. Fifth, and this is in the text, she says trans people should be safe, free from discrimination, and able to live as they wish.

Holly Lawford-Smith's Gender-Critical Feminism (2022) systematises this around females as a sex class, the way a union organises around a job rather than around who feels solidarity with it.

Worth holding on to: the gender-critical claim is not that trans people should be mistreated. It is that sex predicts the harms feminism exists to address, and that replacing it with self-declared identity gives up the measurement.

Laying the two side by side

Read across the rows and something appears that the public argument almost never shows: the two share a great deal. Both hold gender norms to be coercive. Both hold that stereotypes should not determine anyone's life. Both condemn violence against trans people. And for opposite reasons both dislike the born in the wrong body formula, Serano because it flattens wide experience into one medicalised script, Stock because she thinks it false.

What evidence exists, and what it can settle

Much of the public dispute proceeds as though a study could referee it. Mostly it cannot: the disagreement is about which category should organise a politics, and no measurement answers that. Some sub-questions are genuinely empirical.

Take the claim that inclusive access to public facilities increases risk. Amira Hasenbush, Andrew Flores and Jody Herman published a study in 2019 comparing Massachusetts localities that had adopted gender-identity-inclusive public accommodation ordinances with those that had not, and found no relationship between the ordinances and reported criminal incidents in restrooms, locker rooms and changing rooms. That bears on a real claim. It is also one state, one class of space and one outcome, and it says nothing about prisons or sport, where the questions differ and the studies are small and recent.

What matters here: when a dispute mixes conceptual and empirical claims, separate them and say which is which, because a side that wins the concept tends to behave as though the policy followed automatically, and it does not.

Three legal instruments, and what they did not decide

British law has been the main arena, which makes it a useful test case. The Gender Recognition Act 2004 allowed a person holding a gender recognition certificate to be treated as of their acquired gender for legal purposes, subject to exceptions. In Forstater v CGD Europe, the Employment Appeal Tribunal held in June 2021 that gender-critical belief is capable of protection as a philosophical belief under the Equality Act 2010, and said in terms that this does not entitle anyone to express it in ways that harass trans people. In April 2025 the Supreme Court, in For Women Scotland v The Scottish Ministers, held unanimously that man, woman and sex in the Equality Act 2010 refer to biological sex, and that the 2004 Act does not carry across every statute automatically.

None of these decided the underlying question. Forstater concerns what a democratic society must tolerate as a belief, not whether the belief is true. For Women Scotland concerns what Parliament meant in 2010, not what a statute should say. Courts answer the questions put to them, and the question put is rarely the one the argument is really about.

Common misconceptions

• Gender-critical is a polite word for hostility to trans people. It names a position with texts, one of which explicitly holds that trans people should be safe and free from discrimination. Treating it as nothing but animus means never engaging what it claims.
• Trans feminism holds that sex is not real. Serano is a biologist and does not say this. The claim concerns which category should organise a politics, and how sex interacts with the enforcement of norms.
• The Forstater judgment found gender-critical beliefs correct. It applied a legal test about which beliefs a plural society protects, and expressly declined to licence harassment.
• The dispute began with Raymond in 1979. The 1973 conference is six years earlier, and Stock's argument does not rest on Raymond's premises at any point.

Putting it together

Two positions, two libraries, one shared premise that gender norms injure people. They divide over what the injury is a case of. If it is the enforcement of norms on everyone, feminism's unit is the norm and trans people are among its principal casualties. If it is the subordination of females, feminism's unit is sex and losing it means losing the ability to count the harm. Almost everything else, the arguments about facilities, sport, data and law, follows from that choice rather than settling it.

Three questions stay open. Whether gender identity picks out anything existing independently of the norms it is defined against. Whether provisions written in terms of sex can be run without a definition of sex, and what gives when two claims conflict in the same room. And whether the empirical questions can be prised apart from the conceptual ones cleanly enough for evidence to do any work.

Sources

1. Bettcher, T. M. (2025). Feminist perspectives on trans issues. Stanford Encyclopedia of Philosophy. plato.stanford.edu
2. Wikipedia contributors. (n.d.). Feminist views on transgender topics. Wikipedia. en.wikipedia.org
3. Serano, J. (2007). Whipping Girl: A Transsexual Woman on Sexism and the Scapegoating of Femininity. Seal Press.
4. Stock, K. (2021). Material Girls: Why Reality Matters for Feminism. Fleet.
5. Raymond, J. G. (1979). The Transsexual Empire: The Making of the She-Male. Beacon Press.

Amy is eleven, and an interviewer has just read her a story. A man called Heinz has a wife who is dying. A druggist in the same town has a drug that would save her and is charging ten times what it cost him to make. Heinz has raised about half the price and the druggist has refused to sell cheaply or to wait. Should Heinz steal the drug? Amy says he should not, and then she keeps talking.

Jake is also eleven and has heard the same story. He says Heinz should steal it, because a life is worth more than money, and that if the judge understood the situation he would give Heinz the lightest sentence available. One boy in this literature described the problem as being sort of like a maths problem with humans in it.

On the scoring system in use when these interviews were conducted, Jake's answer places him a stage above Amy's. This lesson works out exactly how that happens, then changes one input and watches the result reverse.

Running the scoring by the manual

Lawrence Kohlberg's stage theory came out of a 1958 Chicago doctoral study that followed seventy-two boys, aged ten, thirteen and sixteen, and returned to them over the following years. An all-male sample was unremarkable practice at the time. Six stages fall into three levels: preconventional, where right is what avoids punishment or gets you what you want; conventional, where right is what maintains approval and then what maintains the social order; postconventional, where right is worked out from principles that could be defended to anyone.

The crucial feature of the manual is that it scores the form of the reasoning and not the verdict. Steal or do not steal is irrelevant; what matters is the structure of the justification. So run Jake through it. He ranks two claims, life and property, states that the ranking holds generally, applies it to the case, and anticipates that an impartial judge would follow the same reasoning. That is principle stated, generalised and applied. It scores high.

Now Amy. She talks about what the druggist might do if approached again. She says that if Heinz steals the drug and goes to prison, his wife might get sicker while he is gone and he would not be able to steal a second time. She says Heinz and his wife should talk it through, and that if they explained the situation properly some other way would be found. Pressed, she does not convert any of this into a rule. On the manual, reasoning that leans on relationships and on what other people will do reads as conventional, tied to approval, and it scores a stage lower.

Reading Amy's answer as an answer

Carol Gilligan's move in In a Different Voice (1982) is to stop asking whether Amy reached a stage and start asking what question she thinks she has been given. Jake reads the story as a conflict of claims requiring adjudication; the people are placeholders for the claims. Amy reads it as a tear in a network of relations that has to be repaired without anyone being cut off, and she treats the druggist as a person who could be talked to rather than as a variable.

Once you see that, the scoring result stops being a measurement and starts being a consequence of design. The manual rewards abstraction from particular persons and the impartial application of a rule. Amy's construal cannot be expressed that way, so it registers not as a different answer but as a worse version of Jake's.

The point: Amy's response is not an immature form of Jake's. It is a different reading of what the problem is, and an instrument built around one reading will report the other as deficiency rather than as difference.

Change one input: a real decision instead of a story

Heinz is a hypothetical, and nothing follows from a hypothetical for the person answering it. Gilligan's second study changed that input. She interviewed twenty-nine women, aged from their mid-teens to their thirties, who had been referred by pregnancy counselling and abortion services and who were deciding whether to end a pregnancy. They were interviewed while deciding and again about a year later.

What she reports is a sequence with two turns in it. At first the question is framed around the woman's own survival, what she needs and can manage. Then that framing comes to look selfish to the woman herself, and goodness gets redefined as self-sacrifice, as doing what others need at her own expense. The second turn is the one Gilligan treats as the achievement: the woman puts herself back among the people who count, and the criterion becomes minimising harm across everyone involved, herself included, which is harder than either of the positions it replaces.

The changed input matters in two ways. The reasoning that scored low on the hypothetical is, in a real decision with real costs, plainly the more demanding kind, because it has to be lived with afterwards. And the study has a limit Gilligan does not hide: an all-female sample facing one specific decision cannot tell you what men would have said, so this is a description of a sequence rather than a controlled comparison.

Noddings, and what a relation asks of you

Nel Noddings turned the observation into an ethical theory in Caring (1984), and her structural choice is to make the unit a relation with two occupied positions rather than an agent with a principle. The one-caring attends; the cared-for receives and responds, and the response is part of what makes the caring complete.

Two technical terms carry the account. Engrossment is receiving the other's reality rather than evaluating it, taking in what they are facing before deciding anything about it. Motivational displacement is what follows: your own motive energy runs toward their project rather than yours. Neither is a feeling. Both are things you can do badly, which is what makes this an ethics rather than a mood.

Noddings then draws a consequence most theories would try to avoid, and she draws it deliberately. If obligation runs along relations, it thins as the possibility of a completed relation recedes. She holds that she is not obliged to distant starving children in the way she is obliged to those with whom she can actually stand in a caring relation. Critics read this as a reductio of the whole approach. She reads it as an accurate report of what caring is, and thinks the alternative smuggles the impartial standpoint back in through the door marked care.

Kittay: dependency, and what a theory of justice left out

Eva Feder Kittay writes as a philosopher and as the mother of a daughter, Sesha, who has severe cerebral palsy and profound cognitive disability and who will never be, in the phrase contract theories use, a fully cooperating member of society over a complete life. Love's Labor (1999) makes that phrase the target.

The argument is short and hard to answer. Everyone begins dependent. Most people end dependent. Many are dependent for long stretches in between, and some throughout. A theory that models citizens as roughly equal, independent contractors and then handles dependency as a special case has taken the exception for the rule. Kittay's charge against John Rawls is not that he is indifferent but that the original position is populated by people who do not need anyone.

Her most useful concept is secondary dependency. The person doing the dependency work is not independent either: caring full time for someone who cannot reciprocate makes you reliant on a third party for income and standing. That is a structural fact, not a misfortune, and it explains why care work reliably produces poverty rather than occasionally coinciding with it. Her remedy she calls doulia, after the doula who supports the person caring for a newborn: society owes support to those who care for dependants, so that caring remains possible without impoverishing the carer. Set that beside the wages for housework campaign in Module 2 and you have one structural fact described by two theories that did not much like each other.

What the evidence did to the empirical claim

Gilligan's book carries two claims that are easy to run together. The empirical one is that women and men tend to reason differently. The conceptual one is that a scoring instrument can encode one moral outlook and read the other as immaturity. Only the first is testable, and it has been tested hard.

Lawrence Walker's 1984 review in Child Development found that sex differences in stage scores were uncommon and largely disappeared once education and occupation were held constant: they were tracking who had been to university, not who was female. Sixteen years later Sara Jaffee and Janet Shibley Hyde published a meta-analysis in Psychological Bulletin covering a large body of studies. Women were somewhat more likely to use care reasoning and men justice reasoning, but the effects were small, roughly a fifth to a third of a standard deviation, with heavy overlap, and the orientation used depended strongly on the content of the dilemma.

Gilligan's own position, stated in the book, is that the different voice is characterised by theme rather than by gender, and that the association with women is an observation rather than a claim about nature. Whether the book's rhetoric supports that careful reading is a fair question to put to it.

Why this matters: the strong empirical claim did not survive contact with the data. The conceptual claim did, and it is the part everything since has been built on, because it generalises well beyond moral psychology to any instrument that encodes one way of doing things as the standard.

Okin and Tronto, criticising from inside

Susan Moller Okin makes two moves in Justice, Gender, and the Family (1989). The warning first: a morality of care presented as women's morality risks ratifying the very role assignment that produced the association, converting a skill learned under subordination into a vocation and then into a duty. The demand second: the household is where children first learn what fairness is, so it cannot be exempt from justice. Her charge against Rawls is that leaving the family outside the basic structure leaves the first school of justice unexamined.

Joan Tronto, in Moral Boundaries (1993), relocates the whole thing into political theory. Care, on her account, is a practice with distinct phases: caring about, which is noticing that a need exists; taking care of, which is assuming responsibility for it; care-giving, the hands-on work; and care-receiving, the response of the person cared for, which is the only way to find out whether the care actually worked. She later adds caring with, at the level of a polity.

Her political claim is that the boundary keeping care out of political theory is itself a political achievement: it renders the people doing the work invisible while their work makes everybody else's apparent independence possible. She also presses a warning against her own side. Care detached from justice becomes parochial, attending to whoever is in front of you, which is exactly how distant others get left out. And care may be the morality of the powerless, a competence learned because attending closely to those with power over you is a survival requirement.

Common misconceptions

• Gilligan claimed women are more moral than men. She claimed a different construal of moral problems, and that the standard instrument was built around one of them.
• Care ethics rejects justice. Kittay, Tronto and Virginia Held all treat the two as complementary; the argument is about which one gets treated as the default.
• Care ethics is about having warm feelings. Engrossment and motivational displacement are things you can do badly, and care-receiving supplies a competence test.
• Amy failed the test. She answered a question the manual had no column for, which is a fact about the manual.

Summing up

One eleven-year-old scored lower than another, and unpacking why produced a research programme. The manual was not biased in any crude sense; it was built around one construal of what a moral problem is, and reported everything else as a shortfall. Gilligan named the other construal. Noddings gave it a structure with two positions and accepted the awkward consequence about distance. Kittay showed that the independent citizen of contract theory is a fiction everybody exits at both ends of life. Okin and Tronto pulled the account away from gender and toward politics, one insisting households be measured for justice, the other that care without justice ends up parochial.

The upshot: the surviving claim is about instruments, not about women. Whenever a scale, a test or a promotion criterion reports one group as falling short, the first question is what the instrument was built to reward, and the second is who built it.

Sources

1. Norlock, K., and Pascoe, J. (2025). Feminist ethics. Stanford Encyclopedia of Philosophy. plato.stanford.edu
2. Wikipedia contributors. (n.d.). Ethics of care. Wikipedia. en.wikipedia.org
3. Gilligan, C. (1982). In a Different Voice: Psychological Theory and Women's Development. Harvard University Press.
4. Noddings, N. (1984). Caring: A Feminine Approach to Ethics and Moral Education. University of California Press.
5. Kittay, E. F. (1999). Love's Labor: Essays on Women, Equality, and Dependency. Routledge.

In 1974 Jeanne Altmann published a paper about how to watch animals. It ran in the journal Behaviour, it contained no new findings about any species, and it has since been cited more than twenty thousand times.

What it did was compare sampling methods. Most field observers recorded whatever caught their attention, a practice the paper calls ad libitum sampling, and Altmann showed what that does to a dataset. Conspicuous behaviour gets recorded and quiet behaviour does not. Animals that are large, loud or central get watched; animals at the edge get watched less. The record is then written up as a description of the species. Her alternative, focal-animal sampling, fixes a schedule in advance: this individual, for this many minutes, recording everything, whether or not anything interesting happens.

What changed when the sampling changed

Postwar accounts of baboon and macaque societies had been organised around dominance hierarchies among males, with females appearing as resources over which males competed. That picture was not fabricated. It was what the prevailing method reliably produced, because male competition is exactly the loud, conspicuous behaviour that ad libitum observation over-records.

As the sampling changed and the field's demography changed with it, the picture altered. Sarah Blaffer Hrdy, working with Hanuman langurs at Mount Abu in Rajasthan in the early 1970s, documented infanticide by incoming males and, more importantly, the female counter-strategies around it: mating with several males, which confuses paternity and lowers the risk to infants. Females were running strategies of their own, and the earlier framework had no slot for them.

Here is the question the case forces. The new work was better by the discipline's own criteria: more systematic sampling, more replicable protocols, more predictive power. But it arrived alongside a change in who was doing the observing. Which was responsible, the method or the observers, and what does the answer imply about objectivity? Feminist epistemology has three serious answers, and they disagree.

Three answers, side by side

The first column: the norms were fine, the practice was not

Sandra Harding named these three positions in The Science Question in Feminism (1986), and she meant the first as the modest option. On this view the old primatology was simply bad science. Sampling that over-records the conspicuous is a methodological error identifiable without any feminist theory at all, and the remedy is to do the science properly.

Helen Longino gives this its sophisticated form in Science as Social Knowledge (1990). Objectivity, she argues, is not a virtue an individual can have. It is a property of a community, and it holds only under four conditions: recognised public avenues for criticism, standards the community shares and can appeal to, actual revision of belief in response to criticism, and an authority spread wide enough that dissent is audible. Background assumptions are always doing work in the step from evidence to hypothesis, and the only thing that catches a bad one is a critic who does not share it. Diversity stops being a political extra and becomes a condition of the method working.

The core of it: Longino's argument persuades working scientists more often than the other two, because it does not ask them to give up objectivity. It asks them to notice it was never located where they thought.

The second column: position, and what it lets you see

Nancy Hartsock's 1983 essay adapts a Marxist argument. The proletarian sees production from inside it and therefore sees what the owner need not: a profitable firm can be run without ever understanding where the surplus comes from. Hartsock argues that women's characteristic labour, bodily and repetitive and concerned with maintaining other people, yields an analogous vantage.

Standpoint theory makes three separable claims. That social position shapes what a knower is placed to notice. That the dominant position produces systematically partial accounts, because those who benefit can operate an arrangement without understanding it. And that a standpoint is an achievement rather than a location, worked up through collective struggle and critical reflection, which is why occupying a subordinated position confers none by itself.

Harding adds a pair of terms. Strong objectivity says the usual demand for neutrality is too weak, since it inspects the object of study while leaving the researcher's assumptions outside the frame; a stronger method puts the researcher in the same causal picture. Strong reflexivity is the practice that follows.

The clearest illustration is not a laboratory but national accounts. The rules by which countries measure their product have long counted subsistence production of goods while excluding unpaid domestic services, so a woman spending four hours a day carrying water and preparing food added nothing to the measured national product. Marilyn Waring's If Women Counted (1988) put that in front of statisticians rather than philosophers, and how to value unpaid household work is still argued over in national statistical offices. Nobody made an arithmetic error. The category boundary was set before the counting began.

The third column: Haraway, and the trick of seeing from nowhere

Donna Haraway's 1988 essay in Feminist Studies is the hardest of the three and the most quoted. Her name for the standard conception of objectivity is the god trick: the claim to see everything from nowhere. Vision is always from somewhere, requiring an eye, a position, an apparatus, and each of those is particular.

What she wants is not less objectivity but a different location for it. Situated knowledges are partial, locatable and critical, and their objectivity consists in being accountable for where they were made. She is emphatic that this is not relativism. Relativism, she says, is the mirror twin of totalisation: both are ways of being nowhere while claiming to cover everywhere, and both let the knower off saying where they stand. She also warns against the move her own side finds tempting, romanticising the vision of the less powerful, which swaps a new god trick for the old one. Primate Visions (1989) is the long application.

What each column has to answer

Standpoint theory faces two hard questions. The first is which standpoint, and Module 4 sharpened it: once you accept that women are not one position, the subordinated standpoint splits, and the fragments do not obviously reassemble. The second is Louise Antony's, named in 1993 as the bias paradox. Standpoint theory condemns the bias in dominant accounts while relying on its own preferred position, so it needs a distinction between good bias and bad, and the neutral criterion that would ground it is exactly what the theory said was unavailable.

Feminist empiricism faces a different problem. If the norms of science were adequate all along and the trouble was sloppy practice, why did those norms permit decades of it in a field that thought itself rigorous? Longino's answer, that they were incompletely specified because they said nothing about who must be present to criticise, is a real answer, but it concedes more to the other columns than the modest position wanted to.

Haraway's difficulty is operational. Partial, locatable, accountable knowledges sustained by webs of connection is not a demand a laboratory can implement on a Monday morning.

So what?: the three disagree about the diagnosis and share one commitment. All of them insist that some accounts are better than others, and each is trying to say why without appealing to a view from nowhere. Relativism is the failure state that all three are built to avoid.

The ecofeminist detour

Francoise d'Eaubonne coined the word in 1974, and the strand that belongs here is the one making a claim about knowledge. Carolyn Merchant's The Death of Nature (1980) traces how an organic image of nature as a nurturing mother was displaced during the scientific revolution by a mechanical one, and argues that the new image removed a restraint: you may not easily mine the body of a mother, but you may take a machine apart. Karen Warren supplied the analytic core in 1990, the logic of domination: a chain of value dualisms, culture over nature, mind over body, reason over emotion, in which women sit on the nature side and nature is feminised, so one structure licenses both subordinations. Val Plumwood named the mechanisms in 1993: backgrounding, depending on the subordinated term while denying the dependence, and hyperseparation, exaggerating the difference until the terms look like different kinds.

Vandana Shiva's Staying Alive (1988) is the best known and the most contested. Drawing on the Chipko movement in the Himalayan foothills, she argues that women in subsistence economies hold ecological knowledge that development erases, and connects that knowledge to a feminine principle in nature.

Bina Agarwal's 1992 paper in Feminist Studies is the objection that did the most damage, and it is Module 5's argument applied to a new object. Shiva, she says, treats third world women as one group with a shared consciousness, passes over caste and class, and locates the link in a principle nobody could check. Agarwal's alternative, feminist environmentalism, puts it in material relations: who fetches fuel and water, who holds rights over land, forest and irrigation, and who bears the cost when a commons is enclosed.

Common misconceptions

• Standpoint theory says the oppressed are automatically right. The achievement thesis says the opposite: a standpoint is worked up through struggle and reflection, and a location alone supplies none.
• Situated knowledges is relativism with better vocabulary. Haraway calls relativism the mirror twin of totalisation and rejects both, since neither has to say where it stands.
• Feminist epistemology holds that science is just politics. The primatology case argues that the later work was better science by the field's own standards, which nobody who thinks standards arbitrary could claim.
• Ecofeminism is a single position. Shiva and Agarwal disagree about whether the link between women and environment is spiritual or material, and nearly everything else follows.

Recap

A paper about sampling schedules produced a different account of baboons, and the change coincided with a change in who was watching. Feminist empiricism reads that as a methodological error corrected, with Longino adding that the correction needs critics who do not share the assumption. Standpoint theory reads it as evidence that inquiry begun from the dominant position sees less, and that a better vantage is built rather than occupied. Haraway reads it as the collapse of the view from nowhere, and asks for accountability about location instead. Each has a weakness the other two enjoy pointing out, and none of them is relativism.

Key idea: the practical residue is a question you can ask about any body of evidence. What did the method make easy to see, what did it make hard, and who was in the room when those choices were made?

Sources

1. Anderson, E. (2024). Feminist epistemology and philosophy of science. Stanford Encyclopedia of Philosophy. plato.stanford.edu
2. Grasswick, H. (2018). Feminist social epistemology. Stanford Encyclopedia of Philosophy. plato.stanford.edu
3. Warren, K. J. (2015). Feminist environmental philosophy. Stanford Encyclopedia of Philosophy. plato.stanford.edu
4. Haraway, D. (1988). Situated knowledges: The science question in feminism and the privilege of partial perspective. Feminist Studies, 14(3), 575-599.
5. Harding, S. (1991). Whose Science? Whose Knowledge? Thinking from Women's Lives. Cornell University Press.

An American pharmacist earns close to the same whether she is a woman or a man. The gender earnings gap in that occupation is among the narrowest of any well paid profession. Among lawyers and in corporate finance, drawn from a comparable pool of graduates with comparable degrees, it is among the widest. Something distinguishes pharmacy, and it is not that pharmacists are more enlightened.

Claudia Goldin set out the explanation in her 2014 presidential address to the American Economic Association, and worked out the pharmacy case with Lawrence Katz two years later. Pharmacy standardised. Chains replaced independent ownership, records were computerised, and one pharmacist became substitutable for another mid-shift. Once that was true, an employer had no reason to pay disproportionately more for very long or very specific hours, and the penalty for working part time or taking a career break collapsed. In law and finance the opposite held: pay per hour rises steeply with hours worked and with being the person continuously on the file, so anyone who needs predictable time loses income at a rate far greater than the hours foregone.

Now run the theories of this course over that finding, because a single clean case will not adjudicate between them but it will show what each one has to be able to say.

What each position can and cannot explain about pharmacy

The liberal argument predicts convergence as formal barriers fall, and formally there was no barrier here at all. Women entered pharmacy freely and were paid unequally anyway, until a change in industrial organisation removed the mechanism. That is a gap in a tradition built to identify and strike down exclusions, and it is the gap Module 1 flagged.

The materialist argument does better. Who needs predictable hours, and why, is settled in the household, and the hours penalty is a price attached to that need. But the mechanism Goldin identifies sits in the firm's pay schedule rather than in the family, and a full materialist account has to connect the two rather than assume it.

Dominance theory has the least to say here. There is no obvious sexualised power relation in the pricing of a pharmacist's shift. That is not a refutation of MacKinnon; it is a reminder that a theory built to explain one domain very well can be silent in another.

And the intersectional question arrives immediately. The narrow gap is a gap between women and men within pharmacy. It says nothing about who becomes a pharmacist, and the answer to that question in the United States varies sharply by race and by class.

The point: the pharmacy case is useful precisely because it is small and well measured. It shows that the interesting question is rarely which theory is true, but which theory has anything to say about the mechanism in front of you.

Six disputes that are genuinely open

The rest of this lesson is an inventory. For each, the two positions, then the only thing that matters: what would move it.

One. Does feminism need a stable category of woman? Butler and the anti-essentialists say no, and that the demand for a stable subject has always been the mechanism by which some women get written out. Haslanger says a structural definition can do the work without an essence. Gender-critical writers say a movement without a defined subject cannot bring a claim or count a harm. What would move it is evidence, one way or the other, that a movement organised around a deliberately unstable category can or cannot win a specific measurable thing. There is more of that evidence available than either side usually cites.

Two. Capital, or a second system? Hartmann's dual-systems argument in Module 2 against Young's unified-systems reply. The best available natural experiment is the state socialist record: economies where private capital was abolished, women entered paid work at rates the West did not reach for decades, and the domestic division of labour barely shifted. A dual-systems theorist reads that as decisive. A unified-systems theorist reads it as a change of ownership without a change in how reproduction was organised. That the same evidence supports both readings tells you the dispute is partly about what would count as the system changing, which is a conceptual question wearing empirical clothes.

Three. Consent under unequal power. The sex wars of Module 3 have two live legal experiments attached: New Zealand's decriminalisation of sex work in 2003 and Sweden's criminalisation of purchase in 1999, since adopted by several other countries. Both have been evaluated, and both sides read the evaluations as vindication. The reason is not bad faith. They disagree about which outcome to measure: violence reported by people currently selling sex, or the size of the market. A study cannot referee a disagreement about the dependent variable.

Four. Sex, gender identity, and what institutions should do. The dispute of the previous module. It contains a conceptual question about what gender identity picks out, an empirical question about the effect of specific policies in specific settings, and a political question about who bears the cost when two claims meet in one room. Almost all the public argument runs the three together.

Five. What justice owes to care. The Gilligan debate has largely settled into complementarity, but the hard part has not. Kittay's dependency worker is still poor, and no wealthy country has an answer to what a public institution owes someone who cares full time for a person who cannot reciprocate. This one would move on evidence: several countries have tried carer allowances, credits toward pensions and paid family leave, and the results are measurable.

Six. Objectivity. Antony's bias paradox from the last lesson has not been dissolved. Standpoint theory needs a way to distinguish the bias it condemns from the position it recommends, and no formulation yet commands agreement.

To that list add the oldest one in the course: whether there are universal standards that can be applied across societies without reproducing the relation Mohanty described. Nussbaum's capabilities list is an attempt at a minimum thin enough to be filled in locally. Cases exist on both sides, which is exactly why it stays open.

Three kinds of dispute, and the test that separates them

Look back over that inventory and the items are not the same kind of thing.

Some are empirical and simply under-measured. What a carer's allowance does to poverty among carers is a question with an answer that somebody could go and find.

Some are conceptual, and no measurement will touch them. What the word woman should pick out in a statute is a decision about a category, not a discovery about the world, and treating it as the latter is how people end up citing studies at each other for a decade.

Some are political, meaning that they concern who bears a cost. These do not get settled by argument. They get settled when the distribution of power changes, and the argument afterwards records the result.

The test that tells you which one you are in takes one sentence: name the observation that would change your mind. If you can name it, gather it. If you cannot, stop producing evidence, because you are arguing about a definition or about an allocation, and evidence is decoration in both.

What matters here: the most common failure in this whole field is category confusion between the three, and it is committed by everyone, including the people you agree with.

What has actually been settled

A course built on open questions can leave the impression that nothing closes. Several things have, and each was contested when it was first said.

• Wollstonecraft's circularity: traits produced by a constraint cannot be used to justify that constraint. Nobody now defends the inference she attacked.
• Mill's evidential point: claims about what women are by nature, gathered under conditions of legal dependence, are worthless as evidence in either direction.
• Beauvoir's separation of what is given from what is made, which every later position in this course, including the ones that reject her conclusions, takes for granted.
• The public and private critique: a decision not to regulate an institution is a decision about that institution, not an absence of one.
• Crenshaw's structural point: a remedy built one axis at a time will not reach a person standing at the intersection, whatever the drafters intended.
• Gilligan's instrument point: a scoring system encodes a construal, and will report every other construal as a shortfall.
• Longino's community point: objectivity is not a virtue of individuals, so it cannot be secured by asking individuals to be neutral.

Common misconceptions

• Feminist theory is one position you accept or reject. It is a two-century argument in which the sharpest objections to each position come from inside, which is what most of this course has been.
• The remaining disputes are all about identity. Two of the six above are about labour and money, and one is about the theory of knowledge.
• An old dispute must be unresolvable. Age tells you nothing. Several disputes on this list are old because nobody has done the measurement, not because it cannot be done.
• Getting the theory right settles the policy. Every position here is compatible with more than one policy, and the choice between them is usually the political question, not the theoretical one.

What to carry forward

Seventeen lessons, and the method has been the same throughout: take the strongest version of a position from what its defenders actually wrote, state the objection that has cost it the most ground, and say what remains unsettled. That method is portable, and it is the part of this course that will still be useful when the particular disputes have moved on.

Two habits are worth keeping. When someone tells you what women are like, ask what conditions produced the observation, which is Wollstonecraft's question and Mill's. And when a dispute will not close, ask which of the three kinds it is, because a conceptual disagreement dressed as an empirical one can absorb an entire career.

Bottom line: a pharmacist's pay was equalised not by an argument about equality but by a change in how shifts were priced. Theory told you where to look. It did not do the work, and the theories that admit this are the ones worth keeping.

Sources

1. McAfee, N., and others. (2025). Feminist philosophy. Stanford Encyclopedia of Philosophy. plato.stanford.edu
2. Wikipedia contributors. (n.d.). Feminist theory. Wikipedia. en.wikipedia.org
3. Goldin, C. (2014). A grand gender convergence: Its last chapter. American Economic Review, 104(4), 1091-1119.
4. Goldin, C., and Katz, L. F. (2016). The most egalitarian of all professions: Pharmacy and the evolution of a family-friendly occupation. Journal of Labor Economics, 34(3), 705-746.
5. Haslanger, S. (2012). Resisting Reality: Social Construction and Social Critique. Oxford University Press.