When the activists responsible for the first modern public-health systems sought to convince their governments to invest in sanitary infrastructure, a favorite tactic was to point to the glorious aqueducts and sewers of ancient Rome.

In stark contrast with European governments of the first half of the 19th century, one writer stressed:

It seems to have been a rule with [the Romans], that from the time when the foundation of a city was laid, to that of the summit of its greatness, no structural operation, public or private, should be permitted to take a shape which might render it a harbour either for disease or crime; and it is to this vigilant forethought that, in the absence of other organising agencies discovered only in our later times, we may attribute the success with which that remarkable people preserved social order, throughout so dense and vast a mass of human beings as the inhabitants of the imperial city in the days of its greatness.

The argument that Rome’s success was a direct result of its commitment to public health popped up first in Victorian London, and almost immediately hopped the Atlantic to be repeated by American sanitarians. While recent paleopathological research into Roman feces suggests that aqueducts weren’t enough to spare Romans from playing host to various gastrointestinal parasites, the ancient capital’s infrastructure was still quite a bit more advanced than Victorian London’s. But it’s not clear that Romans owed their impressive hydraulic technology to some great concern for public health on the part of the government—or, at least not at first.

Rome’s oldest aqueduct was already nearly three centuries old when the Roman emperor Augustus included a brief mention of his improvements to the city’s water system in his autobiographical Res Gestae.

I restored the channels of the aqueducts which in several places were falling into disrepair through age, and doubled the capacity of the aqueduct called the Marcia by turning a new spring into its channel.

As historian Harry B. Evans explained in an article published in the American Journal of Archaeology, Augustus’s changes to Rome’s water supply (executed by his right-hand man Agrippa) were quite a bit more extensive than his own words suggest. In a sweeping, multi-year program starting in 35 B.C.E., Augustus’s administration completely revamped the city’s water infrastructure. The oldest aqueducts were repaired for the first time in a century. Over the course of three decades, four new conduits—the Aqua Julia, Aqua Augusta, Aqua Virgo, and Aqua Alsietina—were built, and the locations of distribution basins and fountains around the city were thoughtfully adjusted to serve Rome’s public spaces as well as its growing residential neighborhoods.

Rome was by far the world’s largest pre-modern city. Counting Romans is an extremely difficult undertaking, but historians generally agree that Rome’s population had surged up to around a million people during Augustus’s lifetime. The insufficiency of Rome’s water supply for its growing population could not have been a surprise to anyone in the city—but the city’s last aqueduct had been completed in 125 B.C.E., almost a century before Augustus asserted his supremacy over the senate.

Why hadn’t the Roman government expanded the capital’s water supply in accordance with its population?

The answer: Rome’s water supply had always been dictated by politics, and it was only under Augustus that the politics of water truly aligned with public interest. Like many of the actions the emperor prioritized early in his career, his improvement of Rome’s water supply implied a pointed political message: “Look how much better things are with me in charge!” In this case, at least, he was right.

In its early days, the city of Rome got its water from privately-dug wells and cisterns that collected rainwater. The Tiber, though conveniently located, was not considered healthy for drinking, in part because the city’s sewers emptied into it.

Rome’s first aqueduct, the Aqua Appia, was built in 312 B.C.E. at the direction of Appius Claudius Caecus. That year, Appius was one of two censors, powerful magistrates who were appointed once every five years. In addition to overseeing the census, the censors also had authority over public finances and public buildings. Generally, this meant minor public-works projects such as fixing up aging civic buildings or, perhaps, building a new temple.

When Appius’s co-censor died soon after entering office, as ancient historian Diodorus Siculus tells us, Appius went rogue.

In the first place he built the Appian aqueduct, as it is called, from a distance of eighty stades to Rome, and spent a large sum of public money for this construction without a decree of the Senate. Next he paved with solid stone the greater part of the Appian Way … he expended the entire revenue of the state but left behind a deathless monument to himself, having been ambitious in the public interest.

Both in and outside of Rome, massive public-works projects in the ancient Mediterranean world were associated with tyranny. Histories of aqueducts in the Greco-Roman world typically begin on the Greek island of Samos with the Tunnel of Eupalinos, an underground aqueduct built by the 6th century B.C.E. tyrant Polycrates. The same Polycrates was also famous for enslaving the entire populations of the islands of Lesbos and Miletus. When Diodorus Siculus wrote that Appius was “ambitious in the public interest” by undertaking enormous public-works projects, what he meant was that Appius was acting like a tyrant—one of the worst things a politician could be accused of in the staunch oligarchy of the Roman Republic.

Building an aqueduct was an opportunity for a politician to immediately win a huge and passionate following among the lower classes that could, theoretically, be used for nefarious purposes. Because politicians could not use public funds to build aqueducts without senatorial permission, the growth of Rome’s water infrastructure was limited by how much the senators trusted each other at any one moment, rather than by how much water the city actually needed.

To make things even more complicated, aqueducts—like most types of infrastructure—require upkeep in order to maintain functionality. Aqueducts fed by springs, like the Aqua Appia, were susceptible to a calcium carbonate buildup known as calcareous sintering. Left unchecked, this incrustation steadily diminished the capacity of water pipes. River-fed aqueducts like the Anio Vetus, on the other hand, tended to develop a buildup of silt, sand, and other debris.

The loss of water volume caused by both sintering and sediment accumulation could easily be kept under control with regular cleanings, but there was no public body in charge of aqueduct maintenance in the Roman Republic—and little reason for the individuals who built aqueducts to make provisions for their future upkeep. By the time an aqueduct had started to need repairs, the person who’d built it had already gotten what he wanted out of the project. Similarly, taking on repairs of an old aqueduct was politically unattractive. Building an entirely new conduit that would serve as a “deathless monument” was a much more appealing proposition.

In Rome, aqueducts required a third type of upkeep that was the result of a social, rather than technical, problem: The flow of water entering the city was constantly diminished by the illegal siphoning of water from the aqueduct channels to private property. While there was a bureaucratic procedure by which—for a fee—private citizens could acquire the legal right to divert and use public water, piracy seems to have been easy enough during the Republic that it drastically affected the flow of clean water into the city.

The events of 144-140 B.C.E. illustrate all these problems at once. As we are told by the later water commissioner Frontinus, “the conduits of the Aqua Appia and the Anio Vetus [which had been completed in 269 B.C.E.] had become leaky by reason of age, and water was also being diverted from them unlawfully by individuals.” Unable to ignore the problem any longer, the senate voted to commission the senator Quintus Marcius Rex to repair and extend the reach of the older conduits. Whether because more water was still needed or because Marcius demanded it, the senate also allowed Marcius to build his own aqueduct, which was known as the Aqua Marcia after its builder.

A year into the project, Frontinus tells us, Marcius ran into a major obstacle.

“On consulting the Sibylline Books for another purpose, the Decemviri are said to have discovered that it was not right for the Marcian water, or rather the [new extension of the] Anio Vetus (for tradition more regularly mentions this) to be brought to the Capitol.”

This strange little anecdote is actually a characteristically Roman bit of political drama. The Sibylline Books were a collection of ancient prophecies pertaining to the Roman state, while the Decemviri were a group of state priests who enjoyed exclusive control over the Sibylline prophecies—and who also happened to be senators themselves. The Roman state was so religious and so legalistic that the mere suggestion of an ill omen was enough to, at the very least, cause severe delays in an opponent’s planned course of action while the matter was investigated. In this case, Marcius had to argue his case in the senate house repeatedly over the course of three years before he was finally allowed to complete his aqueduct.

Egregious as it was, Marcius’s ordeal took place while the senate was still capable of working together in the public interest when absolutely necessary. Within a few decades, however, the senate had grown so viciously divided that the construction and upkeep of aqueducts came to a full halt. Despite rapid population growth, there is no record of a single further material improvement to Rome’s public water supply between the completion of the Aqua Tepula in 125 B.C.E. and the time of Augustus.

Two decades after the start of Augustus’s water program, another adjustment forever changed the political significance of Rome’s aqueducts. In 11 B.C.E., Augustus created something Rome had never seen before: a permanent government office dedicated entirely to the upkeep and improvement of the capital’s water supply. Headed by an appointed commissioner, staffed by public slaves, and subject to the emperor’s wishes, the office of the cura aquarum severed the link between aqueducts and individual political capital once and for all. From this point on, the provision of clean water served a new political purpose: It was a daily reminder to Rome’s citizens of the concrete benefits of its new system of government.