What the Egadi Islands Recovery Programme Shows About Documentation-First Archaeology

In 241 BC, a Roman fleet destroyed a Carthaginian one off Sicily's Egadi Islands in the naval battle that ended the First Punic War. For over two thousand years, the wrecks stayed exactly where they sank. In 2004, a fisherman recovered a bronze warship ram — and what happened next is a better argument for systematic seabed documentation than almost anything that could be said about it directly.

From one object to a mapped site

A single artefact tells you a battle happened somewhere nearby. It doesn't tell you where the rest of the debris field is, how large it is, or what else is sitting on the seabed within it. What turned one lucky find into an ongoing archaeological programme — coordinated by Sicily's Soprintendenza del Mare — was systematic side-scan sonar survey of the surrounding seabed, followed by targeted ROV and diver investigation of every contact the survey identified.

That sequence matters. Side-scan sonar doesn't recover anything by itself — it produces an acoustic image of everything sitting on the seabed across a wide area. Most of what shows up isn't significant. But it converts an open-ended search into a finite list of contacts worth checking, and each subsequent recovery — further rams, helmets, amphorae — has come from a documented, mapped location rather than a fresh blind search.

Why this is a documentation problem, not a diving problem

The technical difficulty of recovering an object from 80 metres of water is real, but it's not the bottleneck. The bottleneck is knowing precisely where to look before committing a dive team or an ROV to a site. A mapped debris field turns "search the strait" into "investigate these forty contacts, in this priority order" — the same shift that makes a site's later condition monitoring meaningful too, since a repeat survey against a real baseline can show what's changed and what hasn't.

This is the same logic that applies to any underwater cultural heritage programme, not just a famous one: the baseline survey isn't a prerequisite step before the interesting work starts. It's most of the interesting work. If there's a site that needs that kind of systematic documentation, we're glad to talk it through.