Although local people had known of the existence of Roman remains in the area it was not until 1960 that the archaeologist Barry Cunliffe first systematically excavated the site, which had been accidentally uncovered by workmen when a water main was being laid. The Roman villa excavated by Cunliffe's team was so large that it became known as Fishbourne Roman Palace, and a museum was erected to protect and preserve some of the remains in situ. This is administered by the Sussex Archaeological Society.
The final phase palace comprised four large wings with colonnaded fronts, forming a square around a formal garden. The north and east wings consisted of suites of rooms built around courtyards, with a monumental entrance in the middle of the east wing. In the north-east corner was an aisled assembly hall. The west wing contained state rooms, a large ceremonial reception room, and a gallery. The south wing contained the owner's private apartments. The palace also included as many as 50 mosaic floors, under-floor central heating and an integral bathhouse.
In size, it is approximately equivalent to Nero's Golden House in Rome or to the Roman villa at Piazza Armerina in Sicily, and in plan it mirrors the basic organisation of the Domus Flavia on the Palatine Hill in Rome. Fishbourne is by far the largest Roman residence known north of the Alps. At about 500 feet (150 m) square, it is comparable in size to Buckingham Palace.
A modern museum has been built, incorporating most of the visible remains including one wing of the palace. The gardens have been re-planted using authentic plants from the Roman period. A team of volunteers and professional archaeologists are involved in a continuing research archaeological excavation on the site of nearby, possibly military, buildings. The last dig in 2002 produced some interesting results, and the final report is eagerly awaited.
The first buildings on the site were granaries, apparently a supply base for the Roman army, constructed in the early part of the conquest in 43 AD. Later, two timber-frame buildings were constructed, one with clay and mortar floors and plaster walls which appears to have been a dwelling house of some comfort. These buildings were demolished in the 60s and replaced by a substantial stone-walled house, which included a courtyard garden with colonnades and a bath suite. The palace itself, incorporating the previous house in its south-east corner, was possibly begun ca. 73 AD, although a reinterpretation of the ground plan and finds by Dr Miles Russell of Bournemouth University has suggested that it may more plausibly date to after AD 92.
The more usually accepted theory, first proposed by Professor Cunliffe, is that the early phase of the palace was the residence of Tiberius Claudius Cogidubnus a pro-Roman local chieftain who was installed as king of a number of territories following the first stage of the conquest. Cogidubnus is known from a reference to his loyalty in Tacitus's Agricola, and from an inscribed altar found in nearby Chichester. Another theory is that it was built for Sallustius Lucullus, a Roman governor of Britain of the late 1st century who may have been the son of the British prince Adminius. Two inscriptions recording the presence of Lucullus have been found in nearby Chichester and the redating, by Miles Russell, of the palace to the early AD 90s, would fit more securely with such an interpretation.
The palace outlasted the original owner (Cogidubnus / Lucullus) and was extensively re-planned early in the 2nd century, being subdivided up into a series of lesser appartments. Further redevelopment was begun in the late 3rd century, but these alterations were incomplete when the north wing was destroyed in a fire c.270 AD. The damage was too great to repair, and the palace was abandoned and later dismantled. It is not known whether the fire was accidental or set by coastal raiders.