Take 175 mother-father-child triads in New Zealand, hand the parents standard scales measuring two different flavors of sexism, hand the 8-year-olds a stuffed-toy task where a character named Feppy says things like “women need men to look after them” and the kids rate whether that’s sort-of-right or really-right, and see whose attitudes land where. That’s the design behind Intergenerational Links Between Parents’ and Children’s Hostile and Benevolent Sexism, published in Sex Roles by Oksana S. Voznyuk, Nickola C. Overall, and Matthew D. Hammond, with four co-authors, and covered Oct. 17, 2026 by PsyPost. Fathers’ hostile sexism — the “women are trying to control men” kind — predicted kids’ hostile attitudes, sons and daughters both. Mothers’ benevolent sexism — the “women need protecting” kind — predicted kids’ benevolent attitudes, sons and daughters both. Different parent, different flavor, same destination: an 8-year-old with a worked-out opinion about what women are for. John Stuart Mill diagnosed the mechanism behind that second pathway in 1869, in The Subjection of Women, and his argument was specifically about why the polite version is the one that survives.

The Quick Verdict

Question	What we found
What was studied?	Whether mothers’ and fathers’ hostile and benevolent sexism predict their 8-year-old’s agreement with the same attitudes
Who ran it?	Oksana S. Voznyuk, Nickola C. Overall, Matthew D. Hammond, and four co-authors
Where and when?	Sex Roles, DOI 10.1007/s11199-026-01668-9; covered Oct. 17, 2026 by PsyPost
Sample	175 mother-father-child triads in New Zealand; children age 8, 51% girls
Core finding	Fathers’ hostile sexism predicted children’s hostile attitudes regardless of child’s gender; mothers’ benevolent sexism predicted children’s benevolent attitudes regardless of child’s gender
The asymmetry	Mothers’ hostile sexism predicted sons’ hostile attitudes but not daughters’; fathers’ benevolent sexism didn’t predict kids’ benevolent attitudes at all
What Mill already argued	In 1869’s The Subjection of Women, women’s subordination survives specifically because it’s taught in childhood as natural and desirable, “wound … round their whole lives,” not imposed by force the way the openly hostile version could be resisted

The Quick Version

A 2026 study of 175 New Zealand families found that fathers’ hostile sexism and mothers’ benevolent sexism each predict their 8-year-old’s matching attitudes, regardless of whether the kid is a boy or girl. The protective, “women need looking after” kind of sexism transmits just as reliably as the openly hostile kind — it just travels through the other parent. Mill called this in 1869: the subjection that lasts is the one taught as affection, not the one imposed as force.

What the Study Actually Found

Most research on sexist attitudes studies adults — where they came from is usually a guess, backfilled from theory. Voznyuk, Overall, and Hammond’s team went at the question directly: measure both parents’ sexism with validated adult scales, measure the kid’s attitudes with an instrument actually built for 8-year-olds, and see which parent’s which kind of sexism lines up with which kind of kid attitude.

The measurement problem here is harder than it sounds. You can’t hand an 8-year-old a Likert-scale questionnaire about ambivalent sexism and expect a clean answer. So the researchers used a puppet task — a character named Feppy makes statements (“women need men to protect them,” “women try to gain power by getting control over men”), and the child says whether each one is sort-of-right, really-right, sort-of-wrong, or really-wrong. It’s a method built for exactly this population, adapted here for a New Zealand sample (one hostile-sexism item got dropped for not translating well culturally).

The core result split cleanly along the line ambivalent sexism theory predicts, but with a parent-specific twist nobody’s nailed down this precisely before. Hostile sexism — the resentful, “women want to control men” variety — passed from fathers to children, both sons and daughters, at a correlation researchers put around 0.29. Benevolent sexism — “women need protecting,” “women have a purity men should cherish” — passed from mothers to children, also regardless of the child’s gender. Mothers’ hostile sexism showed up too, but only in sons, a pattern the researchers can’t fully explain; one guess offered in the PsyPost coverage is that mothers higher in hostile sexism may parent sons more controllingly, which could carry the attitude through behavior rather than conversation. And fathers’ benevolent sexism, the one you might expect to mirror mothers’, predicted nothing. Benevolent sexism, in this data, is a specifically maternal transmission line.

The researchers frame the stakes plainly: as they put it, “children learn from the attitudes, behaviors, and everyday conversations they see at home,” and by age 8, those messages are already showing up as the kid’s own stated opinions — attitudes that, extended forward, shape “career opportunities and relationships” for however long they stick. Hostile sexism from dad, protective sexism from mom, both landing before the kid is old enough for a bike without training wheels.

What Did Mill Mean By This?

In The Subjection of Women (1869), John Stuart Mill argued that women’s social subordination is unlike every other form of historical subjection in exactly one way: it isn’t maintained by force or law alone, but by being taught to both sexes from infancy as natural, inevitable, and desirable — “wound round” every important relationship in a person’s life before they’re old enough to question it. Men, he wrote, don’t just want obedience from women; they want “a willing slave,” which means the system has to produce agreement, not just compliance, and it does that through upbringing rather than command.

Mill’s argument gets its force from a comparison he draws early in the essay. Every other subjected class in history — slaves, serfs, conquered peoples — knew, at least some of the time, that it was subjected. The arrangement was visibly imposed, which left room for resentment and eventually rebellion. Women’s position, Mill argued, was built differently and on purpose: “All the moralities tell them that it is the duty of women … to make complete abnegation of themselves, and to have no life but in their affections.” Not obedience demanded — affection and duty instilled, starting so early that by adulthood it doesn’t read as instilled at all. It reads as who she naturally is.

That’s the hinge his whole argument turns on, and it’s also the reason Mill spends so much of the book on marriage law rather than on the people calling women inferior to their faces. The loud, contemptuous version of sexism was never his real target, because it announces itself and can be fought in the open. The full text keeps circling back to the quieter mechanism — sentiment, protection, the language of cherishing — because that’s the version built to survive being noticed.

Why Does the Protective Kind Travel Through the Other Parent?

1. Hostile sexism announces itself, which is exactly why it’s easier to resist and easier to inherit from the parent who voices it directly. A father who says, in some form, that women try to control men is handing his kid an explicit grievance — legible enough that a child, boy or girl, can pick it up as a stated belief about the world. Mill’s point about force applies here in miniature: an openly adversarial claim gets heard as a claim, something to agree or disagree with.
2. Benevolent sexism doesn’t announce itself, because it’s delivered as care, and children absorb care differently than they absorb grievance. “Women need protecting” isn’t pitched as an argument. It’s pitched as how you treat someone you love — which is precisely Mill’s claim that this version works by being “wound round” affection rather than stated as doctrine. A mother modeling protectiveness as a loving stance toward women, including herself, teaches the attitude as a feeling before the child could ever articulate it as a position.
3. The gender-specificity of each transmission line matches who’s positioned to model which role. Fathers modeling resentment of women’s supposed power, mothers modeling the “we need shielding” stance both of them may have absorbed from the culture around them — it isn’t that men only do hostility and women only do tenderness. It’s that each parent is transmitting the version of sexism most bound up with the role they themselves occupy inside it.
4. That mothers’ hostile sexism reaches sons but not daughters is the one asymmetry neither the study nor Mill fully accounts for. It’s a real data point, not a clean confirmation of his theory — a reminder that “taught through affection, not force” isn’t a universal law so much as the pattern that explains most, not all, of what moved in this sample.

Why the Study and the Philosophy Rhyme

The honest version of this isn’t “a Victorian philosopher predicted a 2026 parenting study.” It’s narrower, and it holds up better for being narrower: Mill’s specific claim was that the form of subjection built to last is the one taught as natural and loving rather than imposed as rule, because that form reproduces itself through ordinary family life instead of needing enforcement. Voznyuk, Overall, and Hammond weren’t testing moral philosophy. They were testing which parent’s sexism predicts which kind of attitude in an 8-year-old. But benevolent sexism turned out to move through the parent-child bond that runs on affection, modeled by the parent typically coded as the warmer, more protective figure in the household — which is close to exactly the transmission channel Mill named a century and a half before anyone measured it with a puppet.

It connects to the Wollstonecraft piece covered here on competence penalties against women — Wollstonecraft’s argument was that a woman’s reason gets read as a violation of a role she’s supposed to occupy instead, and benevolent sexism is one way that role gets built in the first place, taught at age 8 as “needing protection” long before any adult teammate gets a chance to withhold help from her. It also sits close to the Adorno piece on authoritarianism running in families — a different attitude, a different mechanism (Adorno’s was partly genetic, this one is closer to pure modeling), but the same underlying claim that attitudes toward hierarchy get planted at home well before they’re chosen. And it rhymes with Beauvoir’s account of immanence covered through the Mankeeping Scale: Beauvoir’s “job” of managing a man’s social life and this study’s “women need protecting” belief are both forms of care-coded constraint, dressed as what a good woman does rather than as a limit placed on her.

The Gap Mill Didn’t See Coming

Here’s where the comparison is more interesting than a clean match.

Mill was writing about a single, mostly unified system — one ideology of subjection, taught by a culture more or less in concert, with husbands and the surrounding society reinforcing the same message from every direction. This study found something more specific and arguably stranger: the two flavors of sexism don’t travel together through one parent. They split, cleanly, along father and mother lines, each parent apparently transmitting the half of the ideology closest to the role they occupy within it. Mill had no reason to predict that split, because his framework treats “the subjection of women” as one coherent thing taught by a patriarchal culture as a whole, not two separable attitude-components with two separate parental delivery mechanisms.

The puppet-task methodology also measures something narrower than Mill was diagnosing. He was describing a fully formed worldview, defended with argument, embedded in law and custom and literature. The study measures whether an 8-year-old will say a specific statement is “sort of right” — an early-stage attitude, not yet rationalized, not yet load-bearing in the kid’s sense of self. Whether that early agreement calcifies into the kind of settled belief Mill was arguing against, or softens with more exposure to the world, is exactly what a cross-sectional study of 8-year-olds cannot tell you.

What This Study Doesn’t Show

Be precise about the size of this. The framing invites more confidence than the data earns.

Correlational, Not Causal

The design measures parents’ sexism and children’s attitudes at roughly the same time and tests whether they move together. It doesn’t rule out shared environment, a third factor like community norms, or even some reverse influence of family dynamics on both. The researchers are explicit that this is an association, not proof that a parent’s belief caused the child’s.

A Modest, Single-Country Sample

175 triads from New Zealand is a real, carefully measured sample — but it’s one country, with specific cultural norms around gender that may not generalize to other parts of the world. One hostile-sexism item was dropped from the measure entirely because it didn’t fit the local cultural context, which is itself a reminder that “sexism” doesn’t mean the identical thing in every setting.

Effect Sizes Are Real but Modest

A correlation around 0.29 is a genuine, reportable finding in developmental psychology. It is not a strong, deterministic prediction. Most children in the study were not simply carbon copies of either parent’s attitudes — the transmission pathways identified here explain part of the picture, not all of it.

The Mechanism Is Inferred, Not Observed

Nothing in the study records actual conversations, parenting behavior, or the specific moments where a child might pick up an attitude. The “modeling through affection” story, including the researchers’ own speculation about controlling parenting toward sons, is a plausible interpretation of a correlation, not something the data directly captured.

What Actually Follows From This

1. “Benevolent” sexism isn’t the harmless cousin of hostile sexism — it’s the one more likely to reach the next generation unchallenged, because it doesn’t look like something worth challenging. If a belief sounds protective, it rarely gets flagged as the thing it’s also doing, which is exactly the property Mill identified as what let subjection outlast every argument against it.
2. Fathers aren’t off the hook here, they’re just transmitting the other half. Hostile sexism moved from fathers to children just as reliably as benevolent sexism moved from mothers. Neither parent gets to assume their own sexism, in whichever flavor they happen to hold, is the kind that doesn’t land on their kid.
3. Ask Mill’s question about any belief that frames protection as its whole point: who benefits from the protected person staying in need of protecting? That question doesn’t resolve every case of genuine care versus constraint dressed as care. But it’s a sharper test than “does this sound kind,” which — as both the 1869 text and the 2026 data suggest — is exactly the test benevolent sexism is built to pass.

Mill never met an 8-year-old with a stuffed-toy survey, and Voznyuk, Overall, and Hammond never set out to confirm a dead philosopher’s thesis about marriage law. But a study measuring which parent’s attitudes land in which child, split by exactly the hostile-versus-benevolent seam ambivalent sexism theory already uses, landed on close to the mechanism Mill spent a whole book insisting on: subjection doesn’t need force to survive. It needs to be taught as love before the person on the receiving end is old enough to ask whether it is.

---

This post discusses a 2026 study published in Sex Roles (Voznyuk, Overall, Hammond, and colleagues, DOI: 10.1007/s11199-026-01668-9), reported by PsyPost, and John Stuart Mill’s The Subjection of Women (1869). The study is correlational, based on a single sample of 175 New Zealand family triads, and cannot establish that a parent’s sexist attitudes caused a child’s matching attitudes rather than reflecting shared environment or other factors.