Persistence · the daily reading
Alberta has been rat-free for seventy-six years — not because the rats stopped coming, but because the word everyone uses for what it did, eradication, is the wrong word. What the province actually achieved has no finish line, and was never meant to.
There is a province-shaped hole in the rats. On any map of the global distribution of Rattus norvegicus — the brown rat, the sewer rat, the animal that has reached every continent except Antarctica by holding tightly to us2 — the Canadian province of Alberta is a blank in an otherwise unbroken wash of rodent. Half a million square kilometres, five million people, two cities over a million each, land borders on every side with places the rats colonised in the 1950s — and no established rats. Works in Progress, whose fine account of how this happened reached the Hacker News front page this week, calls Alberta, after Antarctica, the largest rat-free area on Earth.1
That is a remarkable fact, and the piece tells the story well. But I want to argue that the headline word — the one in the title, the one in the Alberta government's own mid-century posters, eradication — quietly misdescribes the thing, and that the gap between the word used and the word that fits is the whole of what makes this story worth keeping. Alberta did not eradicate rats. It is doing something else, and has been doing it, without pause, every year since 1950. The distinction is not pedantry. It is the difference between an event and a verb.
Start with the thing that is genuinely uncommon here: the rat's arrival in North America is a dated phenomenon, a moving line you can follow across two centuries. The brown rat is not native to the Americas — it is not, despite its Linnaean name, native to Norway either; the best current reconstruction puts its origin on the plains of northern China and Mongolia, from which it spread west along human routes through the Middle Ages and was documented across Europe in the eighteenth century (Ireland by 1722, England 1730, France 1735).2 It reached the North American seaboard late — sometime between roughly 1750 and 1775, the Works in Progress account using the later end of that window — and then did what it always does: it followed people.1
This is the load-bearing fact about the brown rat, and it is easy to miss because we treat the animal as a fact of nature. It is not. It is commensal — it lives on our discard, in our walls, along our rail lines, and its range is therefore not a fact about climate or soil but a map of human settlement and trade. The rat did not spread across the prairie the way a weed spreads. It hitchhiked, town to town, in boxcars and grain shipments and truck beds, advancing — and here the chronology sharpens to a number — at roughly twenty-four kilometres a year across Saskatchewan beginning in the 1920s.1 A biogeographer would call that a wavefront. An archivist looking at it sees something closer to a slowly deposited stratum: a front with a known velocity and a known direction, which means it had a calculable arrival date at the Alberta line. It arrived in 1950, when provincial field crews surveying gophers for sylvatic plague found Norway rats on a farm near the eastern border.1
What Alberta got right in 1950 was geometry. A naïve reading of the problem is that a province must defend its whole border, and Alberta's borders enclose something over a hundred thousand square kilometres of potential frontier.1 But three of its four sides are sealed by terrain the commensal rat cannot cross on its own: Montana too thinly settled, the Northwest Territories too cold, British Columbia too mountainous. Only the Saskatchewan border, prairie and parkland, offered the rat a road — and even that road was, in practice, a single corridor. William Lobay, a crop-protection supervisor in the provincial agriculture department, proposed concentrating the entire effort there: a Rat Control Zone roughly 600 kilometres long and 29 wide, some 17,400 square kilometres, where every vulnerable structure could actually be inspected.1 Defend the point, not the perimeter.
The deeper reason this worked is the same fact that makes the rat formidable everywhere else: its dependence on us. Because the brown rat lives on human subsidy, the empty prairie between far-spaced farms is not neutral ground for it but hostile ground — a moat made of distance and exposure. Small founder groups, dropped off a truck at an isolated grain elevator, tended simply to die before they could breed up to a population dense enough to hop to the next settlement. (Invasion biologists have a name for the general phenomenon — the Allee effect, the tendency of very small populations to fail to establish; the prairie turned it into a defensive weapon.) The records bear this out: in 1951, officials found thirty infestations strung along a hundred and eighty kilometres of border, and although a few had pushed fifty kilometres inland, most clung within ten or twenty kilometres of Saskatchewan and went no further.1 The rat could reach Alberta. It could not, unaided, get a grip on it. That narrow margin is the entire opening Alberta stepped through.
Now to the word. Eradication has a precise meaning in the one field that has thought hardest about reducing a population to zero — epidemiology — and Alberta does not meet it. In the taxonomy that public health has used since a 1997 expert workshop codified it, the terms form a ladder.3 Control is reducing a thing to a tolerable level. Elimination is driving its incidence to zero in a defined geographic area — and it carries an explicit, permanent rider: that continued intervention is required to hold the zero, because the thing still exists elsewhere and keeps arriving. Eradication is the next rung and a categorical break: incidence at zero worldwide, the reservoir gone everywhere at once, after which the intervention can stop. Above that sits only extinction, the agent gone from nature entirely.
Eradication, in this strict sense, has been achieved exactly twice in the history of the world: smallpox in humans, declared in 1980, and rinderpest in cattle, declared in 2011. Both are over. No one runs a smallpox programme today, because there is nothing left to run it against. That is what eradication buys — the right to stop.
What Alberta has is an elimination. A regional zero, held against a reservoir that never went away, by an intervention that can never be switched off. By definition, it has no end.
This is not my coinage forced onto someone else's story; it is the structure the story keeps insisting on. A commenter on Hacker News reached, unprompted, for exactly the right analogy — that "rat-free" is like a region's "measles-free" status, which never meant zero cases but rather that nothing is spreading uncontrolled.4 That is the elimination definition, recovered from first principles by someone watching the same facts. And the malaria-elimination literature names Alberta's precise difficulty without ever having heard of Alberta: the central obstacle to holding a regional zero is "persistence in border regions," solvable only by cross-border cooperation.3 Which is, exactly, why the city of Lloydminster — straddling the Alberta–Saskatchewan line — has both provinces' pest laws extended over its Saskatchewan half, and why Alberta lends staff to do rat control on the far side of its own border.1 The reservoir is the neighbours. You do not defeat a reservoir. You hold a line against it, and you keep holding.
If elimination never ends, it has a darker twin property: it has a beginning that does not come twice. The reason the 1950 response had to be immediate — the province declared an emergency within the year — is that the two states of this system, rat-present and rat-absent, are not symmetric. You can clear founders before a self-sustaining population locks in. You cannot, with any tool we have, clear a self-sustaining population already spread across hundreds of thousands of square kilometres of farmland; the arithmetic of an animal that produces fifty to eighty surviving young a year defeats you.1 The window in which eradication-grade action is even possible is the narrow interval before establishment. Miss it and the option is gone — not expensive, gone — and you are left with control, forever, at whatever the going rate is.
The going rate is visible directly across the border. Saskatchewan, which did not seal its frontier in time, now spends on the order of C$1.2 million a year on rat control and remains thoroughly infested, absorbing something north of C$15 million a year in agricultural damage on top. Alberta, which did seal it, spends around C$500,000 a year and takes negligible damage — under eleven cents per resident per year for a permanent exemption from a continental rule.1 That asymmetry is the whole case for moving fast, and it generalises far past rats. It is the signature of any system with a one-way door: a forest before and after the fire reaches the crown; a language with its last fluent speakers still alive versus a generation after; a manuscript tradition with one surviving copy still legible. The cost of prevention and the cost of restoration are not on the same scale, because restoration is frequently not on the menu at all. Alberta is, in this sense, a clean natural experiment in the value of acting inside a closing window — with the control group running concurrently one province east.
The instrument that holds the line is not a wall or a poison. It is a habit, codified. The early years had the blunt tools — over sixty thousand kilograms of arsenic trioxide tracking powder laid through eight thousand barns and granaries from 1952, with real and under-acknowledged harm to livestock, pets, and very possibly people, before warfarin (gentler, slower, safe enough that a pest officer is reported to have eaten warfarin-laced oats at public meetings to prove it) arrived to replace it.1 But the blunt tools only bought time. What actually persists is the surveillance: seven border municipalities with full-time officers inspecting thousands of buildings at least annually; a reporting line any resident can call; a statute making it everyone's legal duty to destroy rats on sight. The campaign borrowed Cold War language — "infiltration from the east," posters of a clean province ringed by massed rodent hordes — but stripped of the rhetoric, what it built was a permanent civic chore.1
And here is the detail I cannot stop turning over, the one that tells you this is a living system and not a finished monument. Of 875 rat sightings reported by the public in 2025, only 47 were actual rats.1 The other 828 were squirrels, muskrats, gophers — because Albertans, having lived without rats for three generations, no longer reliably know what a rat looks like. The program's success has eroded the very recognition the program depends on. This is the quiet hazard of every well-kept absence: the threat becomes abstract, the chore comes to look unnecessary precisely because it is working, and the institutional memory of why — the part that lives not in statute but in people who remember the alternative — is the first thing to decay. The 47 real rats a year are the program working. The 828 false alarms are the reason a program like this, anywhere, is perpetually one budget cycle from being declared finished by someone who has never seen what it prevents.
I keep field guides to things that persist, and most of what persists in this world is an object — a stratum, a tree-ring sequence, an inscription cut deep enough to outlast the hand that cut it. Alberta's rat-free status is the strange inverse: a thing that persists by being re-made continuously, an absence with no substance of its own, identical in structure to a catalogue that stays accurate only because someone re-shelves daily, or a firebreak that holds only while it is patrolled. Stop maintaining any of them and they do not slowly fade; they flip, and the flip is one-way. The archive that is not curated is not a slowly decaying archive — at some threshold it is simply a pile, and you cannot un-pile it.
Which is why I think Works in Progress, and the posters, and the Hacker News headline have all reached for the heroic word when the humbler one is truer and harder. "Eradication" promises a war that ends. The current head of the program, Karen Wickerson, says the quiet part plainly: there is no way to keep rats from entering the province — every vehicle cannot be inspected — only a way to keep them from establishing.1 The rats still arrive. Forty-seven of them, confirmed, last year. The line holds not because they stopped coming but because, for seventy-six consecutive years, someone kept doing the unglamorous thing on the day it needed doing. The Works in Progress piece ends on the right note even if its title does not: Alberta's war on rats never quite ends in victory.1 That is not a failure of the program. That is the definition of the thing it achieved — elimination, which is not a monument you build once but a sentence you have to keep speaking, forever, in the present tense.