Mandated rest days

Why does a rule forbidding useful work one day in seven leave a community better off rather than poorer?

Consider what the rule actually says. Not "rest when you are tired" but: on this day you may not plough, trade, kindle a fire, or carry a load. A prohibition, enforced, falling on a seventh of the year — some fifty working days surrendered annually by a community that can rarely spare fifty of anything. Roman writers mocked it as institutionalised laziness, and by ordinary arithmetic they were right.

So the puzzle is not why rest is pleasant. It is why the rest must be commanded, and why it names a day rather than a quantity. If a day off simply left each household better off, each household could take one, and no law would be needed.

Split the question in two: why rest, and why compel it. The first is easy and uninteresting. Fatigue is real, recovery is real, and the benefits fall mostly on the person resting. On that account the rule is redundant.

So look for something the individual cannot get by choosing. What determines whether a farmer works the marginal day? Partly absolute need — but partly position. His harvest sells into the same market as his neighbour's, his shop competes with the shop opposite, his standing is measured against the village. Where advantage is relative, the day you rest is the day a rival gains ground. Each therefore works.

Now follow that to its equilibrium. If all reason that way, all work — and nobody gains an inch, because relative advantage is precisely what cannot be had by everyone at once. What they bought with the extra day is not standing but exhaustion. Fred Hirsch's Social Limits to Growth named this shape: effort spent on positional goods cancels.

That reframes the prohibition. A rule removing the option to work takes nothing real from anyone, because the prize it forbids you to chase was unattainable in equilibrium anyway. Thomas Schelling made the general point in The Strategy of Conflict: being unable to do something can be a strength, since it frees you from the pressure to do it. The sabbath is that device applied to a population at once.

But this explains a day off, not a shared day — so why not let each household pick its own? Two reasons, and the second is deeper.

The first is verification. A common date is observable: anyone can see whether the shop is shuttered. "Rest as you need to" is unenforceable, and an unenforceable version collapses back into the race, since the person who quietly works is invisible and better off.

The second is that most of what rest is for requires others to be free at the same time. Worship, courts, feasting, teaching, visiting kin, eating with your household — none can be taken alone. A day off nobody shares buys sleep, and none of the goods the institution exists to produce.

There is a rough natural experiment, which I give as recalled and imperfectly detailed. The Soviet Union introduced a continuous working week around 1929: plants never stopped, and workers took staggered rest days by group — arguably more of them than a common Sunday provided. It was disliked and replaced within a few years by a cycle with a common rest day. The complaint that recurs in accounts of it is the one the argument predicts: a family whose members drew different days was never off together. Rest had been preserved and its value destroyed.

Then why a sacred rule rather than a civil one? Because the rule is worth only as much as its unavailability. Any exemption an employer can grant, or a debtor be pressured into waiving, reopens the race — whoever agrees to work outcompetes whoever does not, and the concession spreads until nothing is left. A religious framing supplies a rule nobody may trade away. And notice whom the text protects: Deuteronomy's version extends the rest to male and female servants, the resident foreigner, and the working animals — exactly the parties least able to refuse a bargain. A rule that mattered only to its keeper would have no reason to name them.

This is where the account differs from its neighbours here. Costly religious demands and ritual meaning read a sabbath's expense as information — a visible, burdensome act sorting committed members from free riders. That is compatible with everything above but answers a different question, and there is a clean test between them: a signalling account has no reason to care which day is kept, only that keeping it costs, whereas here the shared date is the entire mechanism.

All of this is a functional reconstruction, and I should be plain about that. The traditions give their own reasons, which are not economic: Exodus grounds the command in creation, Deuteronomy in deliverance from slavery. Why a practice persists is not evidence about why it arose, and we cannot measure whether sabbath-keeping economies outproduced their neighbours.

Three refinements hold this together.

It is not the claim that rest is good and should therefore be required, but the narrower claim that this particular benefit is unreachable by individual choice — which is what makes a prohibition the appropriate instrument rather than advice.

Nor is the sacred framing essential; only the non-negotiability is. A statutory weekend, a union contract, or a custom strong enough to make working shameful does the same job with different backing — which is why the erosion of the weekend by shift work is this problem returning, not a new one.

And the cost is real, which the tradition concedes rather than denies. There are conditions under which a day's stoppage is ruinous, and the rabbinic principle of pikuach nefesh — that saving a life overrides the sabbath — exists because a rule with no exceptions would eventually demand something indefensible. The modern evidence, mostly studies of repealed Sunday-trading restrictions, is mixed and concerns retail hours more than welfare, so I would not lean on it either way.

• Whether fallow-year and jubilee rules have this structure applied to land and debt.

Every term the collection defines is gathered in the glossary.