Chapter 8: Animals: Husbandry, Transport, and Power

Era span: ~12,000 BCE onward · Difficulty: mid
Requires: Ch 3, Ch 4, Ch 7
Unlocks: Ch 9, Ch 14, Ch 16, Ch 24

Domesticated animals are living engines, nutrient pumps, fiber factories, and walking larders. An ox delivers sustained draft of ~0.5 kW (roughly six humans' continuous work); a horse ~0.7–1 kW sustained, and several times that in short bursts. Every machine chapter ahead assumes this biological prime mover as its default until steam arrives (Ch 23).

8.1 Species Selection and Roles

Match species to land: sheep/goats on hills, cattle on meadow, pigs on forest mast, chickens everywhere.

8.2 Breeding Management

Controlled breeding is applied genetics before genetics:

• Keep detailed parentage records once writing exists (Ch 11) — selection without records plateaus fast.
• Select replacements from mothers who conceived early, birthed unassisted, mothered well, and produced heavily.
• One ram/bull services 30–50 ewes/cows: sire choice is half the flock's genetic future — never breed the average male.
• Castrate surplus males early (oxen, barrows): docility converts directly into usable work and manageable herds.
• Avoid inbreeding: rotate sires with neighbors — the first cooperative institution most villages form (Ch 47).

Culling ledger (annual): tag every birth (dam/sire/date); score mothers on fertility, ease, milking, temperament; keep daughters of the top third, eat/sell the rest; swap sires between villages every 2 years. A flock selected this way visibly changes in five years; an unrecorded one drifts for fifty.

8.3 Feed Arithmetic and the Hay Problem

The binding constraint on animal power is winter feed:

• A cow eats ~10–12 kg dry matter daily; an ox working hard needs grain supplementation on top of roughage.
• Haymaking (cutting grass at flowering, drying to <18 % moisture, stacking under roof) is the single invention that makes large herds survivable at northern latitudes. Scythe (Ch 15) and rake work; schedule hay around weather like a military campaign.
• Root crops (turnips, mangels) extend winter fodder and slot neatly into rotation (Ch 7) — historically the pairing that ended seasonal livestock slaughter.

Key threshold: if you can carry 100 % of herd numbers through worst-case winter, you stop slaughtering stock every autumn and start accumulating capital animals — herd size becomes compound interest.

Hay campaign (treat as battle): cut at early flower (peak feed value), tedder-turn twice, bale/stack below 18 % (a twisted wisp that crackles, not bends), roof that night — rain on curing hay leaches half its value. One wet week without hay is a dead ox by February; plan acres accordingly (~1 ha meadow per cow overwintered, region-dependent).

8.4 Harness: The Difference Between Pulling and Choking

Dead end avoided: the throat-and-girth strap. Strapped across the soft tissue of the throat, a horse presses its own windpipe as it leans into the load and gives up much of its usable pull. The often-quoted figure of ~25 % is probably too pessimistic (§8.8), but the direction is not in doubt: much of antiquity ran on poorly harnessed horses.

Correct sequence: 1. Ox yoke (wooden beam on shoulder humps): correct for cattle from the start; zebu/humped breeds suit it best. 2. Breast-strap harness for light horsework: the load bears on the chest below the windpipe — a large improvement, though still short of a collar. 3. Padded collar harness with hames and traces: load transfers through shoulders to traces, bypassing the windway entirely; horses deliver near-full strength and can haul plows, wagons, and canal barges. Build collars from leather (Ch 3), straw-stuffed, fitted individually.

Whippletree/eveners distribute pull among team members so each animal bears an equal share — mechanical fairness as engineering (Ch 15).

Fitting check: two fingers between collar and neck all around; hames seated, traces level to load; sores = refit same day (a galled shoulder idles a team for weeks). Train in halter → long-rein → light drag → full load over weeks — a rushed breaker makes a ruined puller.

8.5 Working Animals in Production Systems

• Plowing with ox pair: ~0.4 ha/day with moldboard (Ch 7). Horses plow faster but need better feed; oxen plow cheaper — choose by fodder economics, not prestige.
• Carting: two-wheel carts for farms, four-wheel wagons for trade; spoked wheels cut rotating weight dramatically (Ch 15). Paved or drained roads repay themselves in broken axle savings (Ch 24).
• Riding/post networks multiply administrative reach — the physical layer of governance (Ch 47).
• Manure: a stalled ox returns ~20 kg/day of fertility. Barn design routes it to a covered heap; well-rotted manure is worth more than the traction sometimes. The animal is a nutrient pump moving fertility from pasture to plow.
• Dairy as preservation: fresh milk keeps for hours to days; butter, ghee (clarified butter), yoghurt and other soured milks, and cheese turn it into weeks to years of storable fat and protein. Cheese concentrates milk roughly tenfold — about 10 L of milk per kilogram of hard cheese — by curdling with rennet (from calves' stomachs) or acid, then cutting, salting, pressing, and ageing. Raw milk carries brucellosis and tuberculosis (§8.6), so milk for drinking is boiled or pasteurised (Ch 30 §30.5). Long-aged hard cheeses carry less risk than fresh soft ones, but not none.
• Bees: a managed hive yields honey (a stable sweetener and preservative — Ch 4 §4.4) and beeswax (sealing, lubrication, candles, writing tablets, and the wax models of lost-wax casting — Ch 10 §10.3), and its foragers pollinate orchards and legumes. Langstroth's movable-frame hive (1852), built around the 6–9 mm "bee space" that bees leave open rather than fill, lets combs be inspected and honey taken without destroying the colony.

Jump: skip donkey-era pack networks entirely where rivers allow — towpaths plus collar-harnessed horses let one horse move the load of a few hundred packhorses, anticipating canal freight economics in Ch 24.

8.6 Health Basics

Safety warning: livestock share disease with handlers — anthrax, brucellosis, rabies, and parasites move from hide, milk, birth fluids, and bites to humans without warning. Quarantine arrivals 3 weeks, wear cover for births and carcasses, boil suspect milk, wash and disinfect after handling sick stock, and cull and burn anthrax-suspect cases rather than opening them; vaccinate and dose only by weight and schedule.

Preventive doctrine beats veterinary heroics:

1. Clean water, dry bedding, ventilation — most "mystery" illness is housing.
2. Hoof care (trimming, and shoeing horses on abrasive terrain) prevents lameness, the top career-ender of draft animals.
3. Quarantine new arrivals 3 weeks; parasite rotation of grazing paddocks breaks worm cycles without drugs.
4. Basic kit: castration tools, stitching awl and thread, tourniquet; know your limits and cull early — a chronically sick animal is a feed leak and infection source.

Hoof round (6–8 weeks): pick daily, trim flare, shoe worn/soft hooves on stone roads (hot-fit, clinched, never quicked); rest the footsore immediately. Lameness ignored becomes a carcass — the cheapest veterinary act is the timely rest day.

8.7 Domestication Dates and Geographies

Every domestication traded some wild autonomy for productivity under human selection — faster maturity, tamer temperament, higher yield — with genetic bottlenecks (founder effects) that modern breeding works to widen again via rare-breed conservation.

Zoonotic ledger: close animal contact helped seed humanity's epidemic pathogen pool, but the date, host, and pathway differ by pathogen and remain subjects of molecular-clock and archaeological debate. Measles diverged from the cattle virus rinderpest, with molecular-clock estimates now reaching back to around the 6th century BCE (Düx et al., 2020); smallpox's livestock-origin story remains unproven. Herds nevertheless fed cities while transmitting anthrax, brucellosis, rabies, parasitic infections, and other hazards—the sanitation chapter (Ch 30) inherits this bill.

8.8 The Harness Numbers Debate

Textbooks long cited throat-girth harness delivering ~25 % of potential pull versus near-full efficiency for the padded collar. Calibrate honestly:

• Those figures descend from Lefebvre des Noëttes' reconstruction experiments (1931), which drove a century of textbook repetition.
• Later scholarship (Spruytte's rebuilds; John Langdon's Horses, Oxen and Technological Innovation, 1986) showed des Noëttes' ancient-harness reconstruction was likely too crude — the real gap was smaller, though still real — and that medieval ox→horse adoption tracked feed prices and road quality as much as collar engineering.
• What stands: breast-straps appear in China by ~5th c. BCE; padded collars, attested earlier in China, spread across Europe roughly 800–1000 CE (the Bayeux Tapestry shows a collared horse drawing a harrow). Direction of the effect is solid; magnitudes are contested. This book flags such contested numbers explicitly rather than hiding them — that is what an objective register does.

8.9 Milestone: Part I Complete

With draft, manure, rotation, storage, and shelter all standing, the village carries itself through worst-case winter with margin — population growth becomes plannable, and one in five hands frees for smithing, scribing, and engineering. Parts II–VI spend that surplus; Part I earned it.