The same report says that at least three people have been infected after exposure on a dairy farm in the state, and that thirty-three dairy herds have recently tested positive. Health officials say farm workers who tend the animals are considered at higher risk. The Yakima Health District, with the state's health department, is watching those infected and those close to the dead. The Centers for Disease Control and Prevention state that the disease is not known to pass from person to person, that the risk to the public remains generally low, and that the virus examined in Washington showed no change that would let it spread more easily. Which farm it was, what protective coverings the workers wore, whether the three sick worked on the same farm: on these the report is silent. Yet these are the first things anyone would ask who wished to know the matter itself, and not the rumour of it.

Let us first define, as geometers do. A fence is a line that men draw on the ground to separate what is theirs from what is another's. A species is the form that nature has given to living things. The crowd confuses the two, and supposes that the cow, the bird and the man, because they are shut in different enclosures, are governed by different laws of the body as well. The very name of the disease refutes the error: it is called the flu of birds, and yet it has entered the cow and passed from the herd to the herdsman. The name was given by men; the disease does not read its name.

The demonstration proceeds by steps. Given: many herds are infected. Given: men move bodily among these herds every day. Given: contagion passes wherever body is brought near to body. It follows that the fence which separates the cow from the neighbour does not separate the cow from the one who milks her, for the milker stands inside the fence. Therefore the boundary between beast and mankind is set neither in timber nor in stone, but in that part of nature which our eyes cannot see. Where the eyes fail, measurement must take their place. Nor is it enough to count the dead; we must know, as Tacitus advises, "not only the chances and outcomes of events, which are mostly fortuitous, but their reason and causes as well."

It is good to remember Hipparchus, who, by comparing older observations of the stars with his own, detected a motion so slow that no one could perceive it within a single lifetime. The same must be done with this disease. When the Centers say they have found no change, they speak truly of what has been measured today; they give no promise about tomorrow. A low risk is a ratio, not a wall: a fraction that grows as the number of infected herds grows. Those govern rightly, then, who neither panic nor sleep, but keep tables, as astronomers keep tables of the sky.

The remedy is neither new nor hard. First, those who tend the herds stand on the frontier, like soldiers on the bank of the Danube; to protect them, test them and watch them is the city's duty, not its favour. Next, the disease must be sought out in the animals before it reaches man, as a fort sends its scouts ahead. Last, honour is owed to those who measure, whose labour goes unseen so long as it is done well. The crowd sees the fence and believes itself safe; the philosopher cannot see the contagion, and so she measures it. Nature does not acknowledge our boundaries; she acknowledges only those who observe her with care.