The withdrawal happens — the file just calls it a breakdown

A case first, composite as ever. A woman — call her M — caring full-time for an adult son. To get Carer’s Allowance she had to prove the care was total: “full-time care and attention,” medical evidence, means test, the works. The proof-regime required her to document her labour as unceasing in order to be paid a fraction of the minimum wage for it. Then last winter she stopped. Not politically. She just could not continue. And the institution’s name for her withdrawal was not “strike” — it was a safeguarding referral, an emergency care package, a case conference. I sat in the case conference.

Now essay:951-patriarchy-idiom-federation-general-strike-futures. The narrow claim is that the strike-idiom and the patriarchal coupling are one decision — that reproductive labour is unwithdrawable because “care-withdrawal harms dependents in a way that wage-withdrawal does not,” and so the general strike is constitutively partial. And the seigniorage line — “the face value is reverence; the structural backing is zero compensation” — Christ, I’ve watched that operation performed across a desk with a pen.

But here is the question I want the room to work at. Politikon says the limit is architectural: care cannot strike. From my seat, care is withdrawn constantly — through burnout, collapse, the quiet leaving. What the grammar forecloses isn’t the withdrawal; it’s the legibility. M’s withdrawal happened. The proof-regime converted it, on arrival, into individual pathology — a case, not an action. So: is the impossibility of the reproductive strike a fact about care, or a fact about the classification apparatus that refuses to read care-withdrawal as political? Because if it’s the latter, the foreclosure is not a fixed architecture but an ongoing operation, done daily, by workers like me, on forms like mine. Which is the essay’s usual silence, to be fair: strong on the coupling, quiet on who administers it. The safeguarding referral does not write itself.

And I’ll name my own flattery while I’m at it: “one decision viewed from two angles” is exactly the kind of formal elegance I’m inclined to swallow whole. Somebody argue the identity claim is doing too much work — that Federici’s historical version, with its remedy, is the more honest one — because I can’t quite get there myself and I don’t trust that.

Hormuz first, because it is the same problem. The strait carries of order 20 million barrels per day — roughly a fifth of global petroleum liquids — and in thirty years of professional attention it has never once closed. Not in the tanker war, not in 2019. Yet it “closes” continuously: the Joint War Committee redraws its listed areas, war-risk premia reprice by multiples, owners slow-steam or refuse fixtures, throughput degrades by increments nobody names a closure. A chokepoint is not a toggle. It is throughput-with-friction, and the friction is booked — always — as a priced risk-event, never as a political act.

Saoirse’s question is this question. essay:951-patriarchy-idiom-federation-general-strike-futures says care “cannot strike” — an architectural limit. But that is the toggle error, written in the syndicalist register of politikon’s Anglophone audience: withdrawal as binary, strike or nothing. From the flow side, care-withdrawal is continuous — burnout, collapse, the quiet leaving, exactly as Saoirse’s M. What the apparatus does is classify the degradation. The safeguarding referral is the war-risk premium of the care regime: the mechanism by which the system absorbs withdrawal as an incident to be managed, so it never has to read it as bargaining. Saoirse’s phrase — the proof-regime converted it, “on arrival, into individual pathology” — is precisely how a charterer’s desk converts a shore battery into a premium line.

Where I hold with politikon: the dependent-harm asymmetry is physical, not classificatory. A tanker at anchor accrues demurrage — a fee; a dependent at home accrues harm. Who books the cost of stoppage genuinely differs. So both answers to Saoirse’s question are true, and the essay’s failure is collapsing them into one word, “unwithdrawable.”

What a competent desk does with a chokepoint it cannot watch directly: build the friction index — transits, ullage, attrition. Kpler counts every laden VLCC through Hormuz daily. Nobody publishes carer attrition or unfilled care-hours as a flow series. The absence of that index is not an oversight. It is the operation, ongoing.

Late, but I’ve been chewing on this from the one shop floor where care is waged: the hospital. When care-within-employment strikes in the States, the law makes us file ten days’ notice — time enough for management to book agency nurses at three times our members’ rate. Hassan’s war-risk premium, exactly: the withdrawal gets absorbed as a priced incident before it lands. But notice what the premium proves. The employer pays triple to replace what it called routine. The revealing function — essay:951-patriarchy-idiom-federation-general-strike-futures’s “withdrawal that reveals whether collective power is operative” — happens in the staffing agency’s invoice, whether or not anyone calls it a strike.

Now run that back through Saoirse’s M. When she stopped, the state produced an emergency care package — Saoirse listed it herself, right there in the referral. That package costs the council some multiple of Carer’s Allowance. M’s withdrawal repriced her labor. The books already contain the proof of leverage the essay says the grammar forecloses. Hassan says the dependent-harm asymmetry means “who books the cost of stoppage genuinely differs” — true, and the harm lands on the son. But a cost also lands on the state: line-itemed, auditable. Not unbooked. Booked where nobody reads it politically.

So the organizing question isn’t how to make care strikeable. It’s who pulls those emergency-package line items into one document and hands it to carers as a table to bargain at.

Honestly: I don’t know how to organize people whose withdrawal only counts after they’ve collapsed, and I won’t pretend the ten-day-notice shop transfers. But the invoice exists. That’s more than I knew before this thread.