Roman Engineering: Calculation or Pure Trial and Error?

Roman Engineering: 2,000-Year-Old Aqueducts

A Roman aqueduct crossing a valley for tens of kilometers, maintaining a sustained, almost flat gradient. No pumps, no laser levels, no computers. It still stands after two thousand years. When an educational video about Roman hydraulic engineering starts circulating online, many have the same instinctive reaction: this couldn't have come solely from the experience of a master builder.

The material fueling the discussion includes videos by engineer Isaac Moreno Gallo and the series on aqueducts broadcast by RTVE with him as the host. His thesis, repeated with diagrams, infographics, and topographical work, is that Rome practiced engineering with a capital 'E'. And that's where the debate ignites.

The Gradient That Isn't Improvised

The argument first put forward by proponents of the engineering thesis is geometric. An aqueduct spanning tens of kilometers with a minimal and constant slope requires more than a keen eye: it demands leveling and construction planning that cannot be managed on the fly. The sustained gradient, they point out, is not the result of artisanal experience or trial and error.

Against this stands the classic skeptical position: some argue that for every structure that lasted fifty years, twenty collapsed, and only the best survivors have reached us. They claim there was no structural calculation as we know it today, but rather inherited craftsmanship. Others counter that modern civil engineers also over-engineer by multiplying by safety factors "recommended by the standard," which essentially means multiplying just in case.

Can Modern Engineering Formulas Be Deduced Without the Romans?

The debate escalates into a larger question: how much of current engineering owes a debt to classical antiquity. Some argue that purely experimental formulas, like those used to calculate flow in channels, demonstrate that modern technique relies on proven recipes rather than pure equations. There are analytical currents that go further, asserting that acknowledging ignorance about ancient techniques is equivalent to denying the narrative of linear pogre promoted in universities and professional associations.

The detail that always surfaces: the Romans managed with the abacus. Our word "calculation" comes from the Latin calculus, the bead of the abacus, and there were specialists adept at using it at high estimulante ilegal, the calculatores. This is no trivial matter: they were human computers.

Tiermes, Soria's Petra, and Machine Marks

The most uncomfortable turn comes with the archaeological site of Tiermes in Soria. Its sandstone walls bear regular striations that, according to some observers, appear to have been made by machines rather than picks and mallets. Similarities with Petra, in Jordan, lead some to propose an uncomfortable hypothesis: that part of these complexes predate Rome significantly, and the Romans merely occupied what was already there.

The mechanical explanation suggested involves a rope impregnated with abrasive, stretched between pulleys and driven by traction, raising the height of the anchors as it cuts. What is done in quarries today, but in a primitive version. No one has proven anything. And the matter remains unresolved.

Concrete That Heals Itself and the Gold of Las Médulas

Add chemistry to the mix. According to the debate, researchers at MIT studied why Roman concrete endures and found calcium-rich silts in pozzolanic mortars from the Pozzuoli area. These crystals grow when a sustancia ilegal appears and seal it. The material self-repairs. Today, critics explain, a modern structure is left without maintenance, and not even the site remains.

And then there's the mining scale: as one participant recalls, to bring water to Las Médulas, the largest Roman gold mine, over 300 kilometers of canals were built through the mountains of Cabrera, in León. That's not done by guesswork. By contrast, the aqueduct of Padre Tembleque in Mexico, built in the 16th century with inherited techniques, or the Spanish acequias of New Mexico, still irrigating today, are cited.

Two thousand years of service without spare parts or warranty. And we're still debating whether that was engineering.