The engineering problem is gradient. Water moves by gravity, so the channel must fall continuously and very gently — often less than a metre per kilometre — over distances of tens of kilometres, across ground that does not oblige. Getting that right required surveying instruments and a great deal of patience.
Most of an aqueduct is a covered channel underground. The arcades that everyone pictures are the exception, built where a valley had to be crossed and the gradient maintained. Where a valley was too deep, an inverted siphon could carry water down and up again in sealed pipes under pressure.
In the city, water arrived at distribution tanks and was divided between public fountains, baths and private connections — the last taxed, and, Frontinus complains, frequently stolen by tapping the mains. His treatise, written as water commissioner in 97 CE, is the best account of Roman public administration in operation that survives.
What the evidence supports
Most aqueduct length ran underground.
Established by survey of the routes. The visible arcades are the exception, not the norm.
Fraud on the water supply was a recognised administrative problem.
Frontinus describes illegal tapping, undersized official nozzles and collusion by the water staff, in the voice of an official who audited the system.
Lead pipes were used for distribution, and the health consequences are argued.
Lead piping is attested and Vitruvius himself notes that lead workers look unhealthy and recommends terracotta. Whether Roman lead exposure had population-level effects is a live modern debate; water carrying dissolved calcium coats pipes internally, which complicates it.
Examples
The evidence level on each answers how we know what the building looked like, not whether it existed.
- Evidence level: Documented
Near Nîmes, France · 1st century CE
The Pont du Gard
A three-tier bridge carrying the Nîmes aqueduct. Its gradient over the whole 50 km route averages a very small fraction of a per cent. Standing complete across the valley.
- Evidence level: Documented
Rome · Completed 52 CE
The Aqua Claudia
Its arcades stride across the Campagna and enter the city; substantial stretches survive. Substantial arcades survive standing across the Campagna.
- Evidence level: Probable
Segovia, Spain · Probably late 1st or early 2nd century CE
The aqueduct of Segovia
Built of unmortared granite blocks and still standing across the city centre. Its precise date is not fixed by inscription.
Archaeology
Aqueduct routes can be traced on the ground over long distances, and the surviving channels preserve the calcareous deposits that built up in use — which can be measured to estimate flow and years of operation.
Primary sources
Frontinus, On the Water Supply of the City of Rome
The water commissioner's technical and administrative account: routes, capacities, legal regulation and fraud.
Vitruvius, On Architecture 8
Water supply, surveying levels, and the objection to lead piping.
How to read the evidence labels
- Evidence level: Documented
- Supported by a named primary text or by excavated material evidence.
- Evidence level: Probable
- Supported by strong inference that most specialists accept.
- Evidence level: Disputed
- Substantial specialist disagreement exists; competing readings are stated.
- Evidence level: Literary
- Belongs to poetic or mythological construction rather than to history.
- Evidence level: Mythological
- Belongs to shared religious or mythic tradition rather than to any single text.
- Evidence level: Unknown
- The available evidence is insufficient to decide.
These labels describe the status of the evidence, not our confidence in a conclusion. A claim marked Literary is not a claim we doubt; it is a claim about what kind of thing a poem is.