Support Provided ByLearn More
Ancient WorldsAncient Worlds

Watering Ancient Rome

Peter Aicher, an Associate Professor of Classics at the University of Southern Maine, has spent years studying the graceful arches and ingenious plumbing of Ancient Rome's vast water distrubution system. In this interview, he describes how the Romans developed this elaborate system, which included aqueducts 60 miles long.


Pont du Gard Roman aqueduct

Receive emails about upcoming NOVA programs and related content, as well as featured reporting about current events through a science lens.

Roman architects and hydraulic engineers built the magnificent water-transporting bridge known as Pont du Gard 2,000 years ago.
© Samo Trebizan/iStockphoto

An Empire Built on Water

Support Provided ByLearn More

NOVA: I thought we'd begin by looking at the big picture. What would Rome and the Roman Empire have been like without their aqueducts? What did these water bridges mean to their civilization?

Peter Aicher: The Romans could not have built cities as big as they did without aqueducts—and some of their cities wouldn't have existed at all. Romans sometimes built cities on dry plains. They'd find a spring in the mountains and take that water into the city, which would not have been possible without the transported water. With the water, they could have their baths, their fountains, and their drinking water.

It also would be impossible to imagine Rome, which had about 1,000,000 people at its peak, without its large aqueducts. The Romans could have obtained their water from the river, wells, and springs, but these sources would have become polluted in a large city.

Their society would have been very different without imported water. There would not have been a bath culture. Also, the city wouldn't have been nearly as clean—visitors to Rome at the time were amazed at the cleanliness. The hidden half of the water system—sewers—took aqueduct overflow and flushed the refuse into the river, which damaged the river, but kept the city of Rome clean.

Trevi Fountain in Rome
Rome's magnificent Trevi Fountain, built in the 1700s, still relies on sections of an ancient aqueduct for its water.
© Aleksandar Kolundzija/iStockphoto

How did they find these water sources? Was there a science to locating these springs and picking other water sources?

It was a quasi-science. There were rules of thumb and long traditions of places to look for water. Vitruvius wrote 10 books on architecture and he wrote about finding water. Some of his advice is obvious, like looking for mist or a lake and checking the water quality. He recommends talking to the local inhabitants around a water source to see what their complexions are like. If they looked healthy, that might mean that their water source was pure and worth tapping.

"When Rome was besieged by the Barbarian invasions, one of the first things they did was to cut the aqueducts into the city."

The Romans carefully checked the source of water. Did it look attractive? Was it brackish? Streams and lakes could be a problem. Frontinus, who wrote a handbook on the Roman aqueduct system in about A.D. 100, describes how the water would arrive in town muddy after a storm in the countryside. You'd suddenly see this muddy water coming out of the fountains. To some degree, they could clean that water up. Or they could divert it to industrial uses or irrigation, where cleanliness was not as important.

The really prized waters in Rome came from springs. They were harder to find because the springs often lay underground. There are certain techniques the Romans used for figuring out where to dig a well to find the spring. For instance, green grass in a dry season or a certain manner of vegetation could be indicators. It was difficult; springs do not always bubble up to the surface. Sometimes they would dig down to the water table and build an underground tunnel to begin the aqueduct.

So they were actually building tunnels?

Yes. A tunnel might be five, 10, or even 20 feet down. They couldn't pump the water up, because it wasn't practical. So they would dig a tunnel right from the beginning. Some of these aqueducts are almost entirely underground.

tunnels of aqueduct
Romans used tunnels to get to prized underground springs and bring water through hills and mountains.
© WGBH Educational Foundation

Were there advantages or disadvantages to building tunnels?

There were definitely advantages. Tunnels don't disturb surface activities such as farming or traffic, and that's a plus. Think of what highways today can do to a neighborhood. Tunnels also were less vulnerable to wind erosion, the weather, and earthquakes.

One of the first aqueducts in Rome was built entirely underground for security reasons. Underground, it wasn't vulnerable to enemies. When you build arches, it's like advertising to the enemy: Here is our aqueduct.

As Rome grew and the borders spread 1,000 miles away, that was no longer a concern. They weren't worried about the enemy getting to them.

When Rome was besieged by the Barbarian invasions, one of the first things they did was to cut the aqueducts into the city. So you see how it hangs together. The security of Rome and the building of a stable empire allowed the stupendous archways, these arcades, to be safely built.


The arcades are what come to mind as one of the signatures of the Roman Empire. Why did the Romans seem to have a love affair with arcades?

The aqueducts were largely a gravity system. They had to keep the water at a certain level because if they lost that level, it was hard to get it back up again. So to maintain a slope towards the city and to bring it in at a high enough level, you had to keep the channel at a certain height. So when that channel came to a dip in the landscape they built an arcade or a bridge to take the water over it.

In Rome, the long dramatic arcades occur in the five or six miles right outside of town. They built them there because the land dips down before rising again to the hills in Rome.

If they had to build an aqueduct only five feet high, they would build a wall. But above that, they used the archway. This saved material. And again, arcades are not as disruptive to the landscape. A wall five miles long does damage to the transportation on the surface and creates a water barrier.

Also—and I think this is important—the arch and the arcade, which is a series of arches, is beautiful. I think the Romans were as struck as we are by the beauty of the arcades curving over the landscape. In fact, some of the best villas were built to look out over the aqueducts. That was no accident. It's like landscape art.

Segovia Arcades
The arcades make up only a small percentage of the Roman aqueducts, but they are among its most memorable components. Pictured is an aqueduct in the city of Segovia, near Madrid, Spain.
© jorgefelix/iStockphoto

They obviously were beautiful things. But how good were they at delivering drinkable water? Did they import water that you or I would want to drink?

On that score, I'd follow the ancient sources. They praise some water sources; others they really panned. Focusing on Rome, which I know best, there were some real stinkers.

Just like Romans today, they were connoisseurs. They actually ranked them. At the top of their list was the Aqua Marcia, one of the long aqueducts that came from the springs in the mountains and traveled 60 miles. In fact, a newer aqueduct of the same name delivers Rome's most prized water today from springs in that same area. Spring water was generally cooler than stream or lake water, and cleaner, too.

Another source that was highly prized was an underground aqueduct, the Aqua Virgo, which is delivered to the Trevi fountain today. Today these waters would be undrinkable if they weren't treated because the city has spread out to include Virgo's springs. Water from other aqueducts would come in muddy. Frontinus tells us of an aqueduct that tapped a lake north of Rome. He says it was a real stinker.

"Admittedly, they had nothing so convenient as tap water in a fourth-floor apartment."

Now we use chemicals and water treatment plants to clean our waters. Did the Romans have any mechanisms for cleaning their water?

They didn't use chemicals, but they had other ways. First, they used settling basins. It was like a pool. The basins would slow the water down. As it slowed, the impurities or the load, as it's called, dropped out of it. That would remove some of the sand and other impurities.

We also purify water by aerating it. The water in the aqueducts was exposed to air throughout its journey, although I don't know if the Romans knew this improved the quality of their water.

Instead of a settling basin, one of the aqueducts had zigzags built into it. We figure that these zigzags caused the water to slow down, which would unload impurities. Occasionally the Romans would have to shut the water off. Someone would climb into a tunnel from the surface through a well hole. There would be little hand- and footholds carved into the walls of these shafts. Sometimes they'd go down 30, 50 feet. Slaves would shovel out impurities that would be hauled to the top in buckets.

Remaining mysteries

Are there still mysteries left about the aqueducts and the water distribution system that archeologists, people like yourself, are trying to understand?

There are a number of mysteries that remain. We don't know how the Romans engineered some of the tunnels or how they went about surveying a tunnel that went under a deep mountain, for example. There's still work being done on that. We also don't know, especially in Rome, much about how the distribution system worked on the local level.

What do we know about the local water distribution system?

We know that the water came into the city on a gravity system in the open air, like a stream. When it got to the city, it changed into a closed system. It did that by going into a large tank or water tower, called a castellum, typically placed at a high spot in the city.

From the tanks, water would be transported through an underground system of pipes beneath the street. Water could be delivered up again to a height equal to the water level in the tank. It generally went to public fountains, baths or drinking basins, since only the very wealthy had their own private delivery pipes.

We don't know much about the system of piping itself, at least in Rome. We do know that most of the pipes were made of lead in that city. This varied, however, depending on the locale. In Germany, for instance, where there was a lot of wood, pipes were made out of wood. Elsewhere they might be terra-cotta.

Did they have no sense, then, for the dangers of lead?

Actually, they did. At least Vitruvius did. He makes his point by saying, "Hey, look at the people who make these lead pipes!" Apparently, these workers weren't in the best of health.

What do you think of the theory that the Roman Empire collapsed because the Romans suffered from lead poisoning?

Not much. The Romans did use lead in their pipes. However, two things about the Roman water supply mitigated the unhealthy effects of lead. The first is that the water in the Roman aqueducts rarely stopped running. They had shut-off valves, but they didn't use them much. The water was meant to move. It would flow into a fountain or a basin. Overflow would pour into the gutter and then flush the city.

Today, if you have lead pipes, they tell you to let the water run for awhile before you drink it. That prevents water from sitting in the lead pipes and becoming contaminated. That flushing out happened naturally in the Roman system.

Secondly, a lot of the water, especially in Rome, was hard water. It had lots of minerals in it that would coat their pipes. We often use filtration systems to take some of the minerals out. The Romans didn't have that, so these minerals would encrust and coat the inside of the pipe. That layer of minerals served as a buffer. In fact, the aqueduct channels would gradually accumulate these deposits. Periodically, they would have to chip out all the encrustations.

NOVA bath
The NOVA bath, built in Turkey, is similar to private baths built in the homes of wealthy Romans.
© WGBH Educational Foundation

Would the bath that NOVA built have been supplied by aqueduct water?

It's very possible, although a smallish bath like the one NOVA built could be run with water from a nearby stream or well. The huge baths, such as the famous ones in Rome, could not have been operated without the aqueducts. Some of the baths were supplied with water from private branch lines, sub-aqueducts really, that detached from the main aqueduct.

The water distribution system sounds extremely sophisticated, especially considering this was 2,000 years ago. Did the water cost people or was it a part of what the Roman state gave gratis to its people?

Both. Rome may be something of an anomaly, but it had so much water that many private users would get a grant from the emperor during imperial times. They didn't have to pay; it would be an act of patronage. However, others, such as industrial users, would have to pay.

Most people would get their water from street basins, where the water was free. Pompeii probably has the best preserved distribution system. Its basins are fairly regularly spaced, so that most people didn't have to walk more than about 150 feet to get water. Admittedly, they had nothing so convenient as tap water in a fourth-floor apartment.

Editor's Notes

This feature originally appeared on the site for the NOVA program Secrets of the Lost Empires.

National corporate funding for NOVA is provided by Draper. Major funding for NOVA is provided by the David H. Koch Fund for Science, the NOVA Science Trust, the Corporation for Public Broadcasting, and PBS viewers.