In The Decline and Fall of the Roman Empire, Edward Gibbon portrays the pagan emperor Maxentius as a licentious youth and "a tyrant as contemptible as he was odious."

Historians have long assumed that the reviled Roman emperor lived part-time at an 80-acre suburban villa complex until he was killed by his rival Constantine at the Battle of the Milvian Bridge in A.D. 312.

But a University of Colorado-led archaeology team has uncovered evidence that the villa's main hall was never occupied.

Instead, it appears to have been abandoned before completion, said CU archaeologist Diane Conlin, co-director of the Maxentius project, a five-year excavation that began last summer.

"Maxentius builds a lot in Rome during his extremely short reign," Conlin said. "And the pattern - up to our project - is that Constantine either finishes the buildings and takes them over, or he demolishes them and builds something new.

"But this (villa) stands outside that pattern of behavior," she said. "Instead of being finished or demolished, it was abandoned."

Marcus Aurelius Valerius Maxentius ruled Rome from A.D. 306 to 312, a time when the empire was in "a holding pattern at the end of its period of greatness," said CU historian Noel Lenski. Maxentius, son of the emperor Maximian, was in his 20s when he took power.

At the villa complex about two miles south of Rome's center and just outside the city's defensive walls, Maxentius built a chariot course with grandstand seating for 30,000 and monuments to his only son, Romulus. The boy was 9 when he drowned in the Tiber River, the same fate Maxentius met at the Milvian Bridge.

The day after that battle, Maxentius' armor-clad body was fished from the Tiber mud by Constantine's troops. His head was lopped off and displayed as "an emblem of victory and conquest," Conlin said.

Constantine reportedly had a vision of God the night before the battle and converted to Christianity on the spot.

But Maxentius was a pagan. Archaeologists wonder if his villa project on the Via Appia - the first and most well-known ancient Roman road - was an attempt to strengthen ties with Christians, Conlin said.

"One major question that we hope to answer is why Maxentius built this grand villa complex outside the defensive walls of the city when he had full control of pre-existing imperial palaces located in the heart of the capital," she said.

Limited investigations of the villa site were conducted in 1825 and again in the 1960s. Italian archaeologists exposed exterior walls but didn't dig into the interior of the large main hall, or basilica, Conlin said.

CU archaeologists and their colleagues sank two trenches in the main hall during a five-week field season last summer. Two more trenches will be opened in the basilica this summer. Twenty students from CU and Kalamazoo College in Michigan joined in the 2005 work.

After digging through modern garbage deposits and cemented chunks of architectural debris, the team reached the ancient basilica and found "a bare skeleton of brick and concrete," Conlin said.

A finished basilica from the period would feature stone mosaics on the floor and decorative marble slabs on the walls. The researchers found 800 pounds of marble fragments, but no piece was bigger than 12 inches by 6 inches. Instead of mosaic floors, they found only a brick subfloor.

"It seems like no one ever lived there," Lenski said. "It was 85 to 90 percent completed. It was the finishing touches that were left off."

The marble fragments suggest that bigger slabs had been installed but were later hauled away.

"Our preliminary hypothesis is that the building was not finished and was subsequently stripped of all valuable decoration - mainly marble," Conlin said.

The Maxentius project is a collaborative research and teaching program of the University of Colorado, Kalamazoo College and Italy's Comune di Roma.

Conlin and the other project leaders, Anne Haeckl of Kalamazoo College and Gianni Ponti of Rome, began exploring the research potential of the villa site in 2001. After two years of fundraising and negotiations with Italian archaeological ministries, the team conducted a preliminary survey in 2003.

A grant from the Loeb Classical Library Foundation at Harvard University helped fund the 2005 excavation season.

So was Maxentius really as bad a guy as Gibbon and others have portrayed him?

"I don't think he was irredeemably bad," said Lenski, editor of the book The Cambridge Companion to the Age of Constantine.

"But he was a loser," Lenski said. "And when I say that, I don't mean it in the modern sense. I mean he lost to Constantine, so all the information we have about Maxentius comes from sources who were allied with Constantine.

"Nothing survives that is positive or favorable toward him. It's all uniformly negative. . . . In some ways, this villa project is an attempt to add a new voice that might help us see him in a more positive light," Lenski said.