Villa Magna was built on a ridge. To the far left, one can see the  modern farmhouse. In the center is the site of the “slave barracks,”  and through the trees is the church of S. Pietro in Villamagna.
Villa Magna was built on a ridge. To the far left, one can see the modern farmhouse. In the center is the site of the “slave barracks,” and through the trees is the church of S. Pietro in  Villamagna.

Writing to his tutor Fronto in about AD 141, the future emperor Marcus Aurelius describes his stay at the imperial estate of Villa Magna, near Anagni southeast of Rome:

We are well. I overslept a bit on account of a slight cold, but this seems to have subsided, so at the eleventh hour of the night until the third hour of the day I read from Cato’s De Agricultura, and wrote a little bit, less badly than yesterday, thank god.…So with my throat tended to, I set out for my father and stood by him at the sacrifice.…Then we set ourselves to the task of picking the grapes; we sweated, and rejoiced, and, as another author says, “we left the high-hanging vintage surviving.” …[T]he gong rang, that is, it was announced that my father had gone over to the bath. Having bathed, we therefore dined in the pressing room (we didn’t bathe in the pressing room, but, having washed, we ate there) and we happily heard the peasants bantering. (Fronto, Letters, book IV, letter 6, tr. M. Andrews)

villa_magna_site_plan
This plot of the magnetometer survey of Villa Magna shows the likely extent and layout of the villa. The red areas indicate areas with high value magnetic anomalies and thus likely walls and other features.
Janine Young excavates one of the dolia sunk  into the marble paving of the winery.
Janine Young excavates one of the dolia sunk into the marble paving of the winery.

This passage shows Marcus, then around 21 years old, as a willing if playful student of agricultural technique, performing for his tutor. It also emphasizes the importance of his father’s villa as a wine-producing estate. The ceremonies of the day included a sacrifice, perhaps to Jupiter, the plucking of the first fruits of the Latian vintage, and a banquet in the winery where the emperor, Antoninus Pius, and his guests enjoyed a performance by the slaves who trod the grapes on the high pressing floor, which acted as a sort of stage. Music may have accompanied the treading of the grapes, as we know from many mosaics depicting this sort of scene.

The imperial estate of Villa Magna is located southeast of Rome.
The imperial estate of Villa Magna is located southeast of Rome.

The ruins of the villa today cover around 17 ha in the Valley of the Sacco. Now called Villamagna, it has been known to scholars since the 18th century when the antiquarian Gavin Hamilton visited it and declared that it had too few statues to be worthy of further excavation. No scientific expedition had ever taken place at what was one of the most important imperial villas in Latium until 2005, when a team sponsored by the Penn Museum, the British School at Rome, the international Association for Classical Studies, and the Soprintendenza Archeologica of Lazio arrived at the site. Our aim was to study an imperial estate over the longue durée, exploring how such a property would change over time. The five-year campaign, supported by the 1984 Foundation, the Banc’Anagni, and the town of Anagni, has produced quite extraordinary results, showing the vicissitudes of an elite property over 1,200 years of (almost) uninterrupted occupation.

The project began with a geophysical survey of the site which generated a surprisingly clear vision of the buildings scattered along two low ridges of the Monte Lepini. To the northwest a large peristyle, or colonnaded courtyard, is clearly visible, perhaps the porticus duplex or double porticus characteristic the medieval church of S. Pietro in Villamagna, standing today abandoned and roofless on the hillside. Further south large courtyards seem to succeed each other up the ridge towards the site occupied today by a 19th century farmhouse. On the east side of the modern drive are other buildings, including a curious structure whose small rooms appear to form a checkerboard, with an area of symmetrical rooms resembling bath buildings, and the lines of underground drainage pipes and cisterns. Based on the survey, we decided to work on three sites of the villa: the farmhouse itself, still supported by Roman vaults; the “checkerboard” building; and the area around the church where we imagined the medieval monastery, recorded in parchments in a local archive, to have been located.

The Imperial Winery

this plan of the winery was created when the excavation was  completed in 2010.
This plan of the winery was created when the excavation was completed in 2010.
 a reconstructed view from the emperor’s  couch shows dolia half buried in the floor.
A reconstructed view from the emperor’s couch shows dolia half buried in the floor.

By enormous good luck, the first trench opened in the courtyard of the farmhouse showed that we were within the winery of the villa. The floor was made in the herringbone (opus spicatum) technique characteristic of Roman service floors—but rather than of terracotta, the “bricks” were made of portasanta marble! The floor left space for sunken dolia (singular, dolium), or large fermentation jars, while a platform visible at the start of the excavation proved to be the pressing floor; below it, a marble-lined vat caught the juice from the pressed grapes. By the end of the next two seasons the plan of the winery had become clear. The cella vinaria, or press-room, held 38 dolia, with spaces between them for the transfer of the juice from the vat. In front of it, a large curved room, probably open to the sky, must have been the site of the emperor’s banquet. Sadly, none of its floor survived the later leveling of the area, but the floor preparation, like that in the rest of the building, left no doubt that it was originally marble. We may imagine the emperor and his guests lying on couches in the center of the semi-circular exedra, amusing themselves by watching the treading of the grapes.

The winery was entered by two routes. One was a little service stair, to the east, while another, to the south, was an elegant, wide staircase, clearly the imperial access to the building. The spacing of the latter’s marble stairs, separated by wide landings, allowed the emperor to be carried up on a litter. To the south, a passage led to the stair, past a set of bath buildings—probably those referred to in Marcus’ letters. The collapse of the marbles lining this corridor, and their traces in the mortar on the walls, allowed supervisor Dirk Booms to reconstruct the extraordinarily rich decoration of this space. This gave us a glimpse of the kind of material that would have been destroyed in the medieval lime kilns, which burned marble and limestone to make quicklime for mortar. Not all the statues were destroyed, however; a medieval dump gave us a substantial fragment of a statue depicting Hercules seizing Hippolyta. Only the torso of the Amazon queen survives, with Hercules’ hand clutching her breast. Professor Ann Kuttner, of the University of Pennsylvania, is studying this at present, along with the large number of other Roman statue fragments which must have decorated the villa.

A sketch  reconstruction of  the winery by Dirk  Booms includes  the exedra and the  pressing room.
A sketch reconstruction of the winery by Dirk Booms includes the exedra and the pressing room.

The plan of the whole building is tentatively reconstructed in a 3-D model, which allows us to analyze access and circulation, as well as giving an idea of the scale of the structure, which occupied at least three stories. The central location of the winery, within the grounds of the villa, and the astounding luxury of its decoration, removes it from the sphere of functional buildings and requires us to see it as a ceremonial space. Clearly the emperor was representing himself as a Bonus Agricola, a good farmer, whose careful husbandry at his villa was a metaphor for the care he took of the empire. But there is more going on here than simple ruler self-representation. The villa may have been the site of an annual ceremony, similar to that carried out by the flamen dialis, the high priest of Jupiter, at the vintage festival feast called the vinalia, when the sacrifice of a lamb to Jupiter was followed by the cutting of the first bunch of grapes, signaling the beginning of the Latian harvest. The villa’s construction, which appears from brick stamps to date to the reign of Hadrian, would have had a symbolic importance in displaying the interest of the emperor in wine production for Rome, something that is oddly absent from Hadrian’s villa at Tivoli.

left to right, the “imperial” stair led to the winery. A reconstruction of the marble wall treatment in the corridor leading to the stair is shown  here. A fragmentary statue of Hercules and Hippolyta was excavated from a medieval dump.
Left to right, the “imperial” stair led to the winery. A reconstruction of the marble wall treatment in the corridor leading to the stair is shown here. A fragmentary statue of Hercules and Hippolyta was excavated from a medieval dump.

The Slave Barracks

 the slave barracks were excavated in 2009.
The slave barracks were excavated in 2009.
An “ideal” plan of the barracks is superimposed over the plan of the  excavations.
An “ideal” plan of the barracks is superimposed over the plan of the excavations.

The excavation of the checkerboard building to the north of the winery provided a necessary complement to that prestigious building, allowing us a glimpse of the lives of the people who actually carried out the work. The building has two wings, separated by a narrow alleyway with a drain running down the middle. The south wing consists of a row of ten rooms with, to the south, a portico along a road paved in white paving stones. There were no passageways or doors, however, between the portico and the rest of the building, and it may have simply served as a resting place for people arriving at the villa. A mortar foundation in the alley, suggesting a staircase, as well as various pieces of pavement from the collapse, together seem to show that the building had two stories. On the other side of the alley were two rows of ten rooms each, again, probably from a two-story building. With one exception, the rooms measured 10 x 12 Roman feet (2.95 x 3.54 m) and were paved with beaten earth. In one corner of most rooms were found traces of a dolium, probably used here to contain a grain ration, while hearths and querns, or hand grain mills, were common.

The regimented structure of the building, and its similarity to numerous barracks in legionary fortresses, at first suggested that we were dealing with army barracks, but the general poverty and structure of the rooms, with their individual cooking and storage facilities, convinced us that they more likely constituted the quarters for slave families, or, less possibly, free workers on the estate. Evidence for a large presence of women in the barracks came from the large numbers of hairpins and sewing needles recovered from the drain, as well as the tombs of newborn babies excavated under some of the floors. The skeletons of these children show considerable signs of anemia, which may be evidence for thalassemia, or Mediterranean anemia, in the people living in the barracks. Built in the middle of the 3rd century, the barracks remained in use until the third quarter of the 5th century, when it seems to have been abandoned and collapsed.

While much work has been done on slavery in the Roman world, very little is known of it archaeologically. If, as we believe, we are dealing here with slave barracks, excavation of the structure sheds important light on a period when agricultural slavery is generally assumed to have been on the wane, and allows us to see the continuing investment in production at the villa—probably, in this period, still concentrating on wine.

The Church and the Monastery

The facade of the 6th century church of  S. Pietro in Villamagna is still preserved
The facade of the 6th century church of S. Pietro in Villamagna is still preserved.

The church of S. Pietro in Villamagna was constructed in the 6th century on the remains of an earlier building with the identical plan. Built in the late 4th or 5th century, the earlier building lies at the center of a large court paved with smaller versions of the same white paving stones we find on the road. It was probably built as a church, as its form and burials both inside and outside suggest. In the middle of the 6th century the church was rebuilt in the opus vittatum technique composed of local tuff (volcanic stone) blocks and irregular brick bands, traces of which still survive in the standing walls. A narthex or vestibule was added to the facade, and the whole church continued to serve as a cemetery for the community. The church was the center of the 6th-century settlement, and in front of the narthex we have identified a new winery, constructed in the portico of the large imperial building to the north. Like those of the earlier winery, the dolia of the 6th-century winery were robbed out, leaving large holes where they had been placed, two by two, under the shelter of the portico. The winery is, for the moment, one of the last to have been built in Roman Italy. We do not know who built it, or who the owners of the property were in this period—the Byzantine treasury, the church, and a member of the local elite are all candidates. However, the estate was clearly still in the hands of an owner prepared to undertake serious investment in its structures, both ceremonial and productive. This winery does not seem to have survived the 6th century, and after the construction of a couple of huts for storage or domestic use, the site seems to have been deserted.

A section of the cloister and its cistern at S. Pietro in Villamagna is shown in this sketch reconstruction  by Nick DePace.
A section of the cloister and its cistern at S. Pietro in Villamagna is shown in this sketch reconstruction by Nick DePace.

Around two centuries passed before the next occupants appeared on the site; traces of them are found in the substructures of the ancient winery, and in large deposits of earth for cultivation found over the Roman pavement north of the church and in the ancient winery baths. A little village was forming around the ancient winery and energy was being invested in reviving agriculture on the site. Glazed Forum Ware pottery dates this occupation to the first half of the 9th century. By the end of that century we find the first houses of the village. They were huts constructed, once again, over the remains of the winery baths. This village continued to grow for the next three centuries.

The cemetery for the inhabitants of this village, perhaps as well as for others in the area, again formed near the church, where 499 burials were excavated during the course of the project. These are now being studied by Dr. Francesca Candilio of the University of Rome La Sapienza, assisted by Penn graduate students Rowan Brixey and Samantha Cox. The use of the cemetery covers a period lasting almost a millennium. Together with isotope analysis of the bones, underway at Oxford, the study of the skeletons will provide vital evidence for changes in diet, stature, and general well-being over that time.

A document of AD 976 indicates that four local noblemen gave the church property, together with several estates, to a monastery which henceforward took the name of S. Pietro in Villamagna. The first archaeological traces perhaps associated with this event are the remains of a large buttressed wall, creating an open courtyard north of the church. Pottery from the layers associated with this wall appears to confirm its date in the last quarter of the 10th century.

This reconstruction sketch shows what the barracks may  have looked like when it was in use.
This reconstruction sketch shows what the barracks may have looked like when it was in use.
robbed_out_dolia
The hollows of the robbed-out dolia in the 6th century winery are shown in this paragraph. Note the trace of a dolium base preserved in mortar in the lower row.

Around a century later, the abbey was rich enough to undertake major building campaigns that brought its plan more in line with those of contemporary monasteries. A bell tower was put up in front of the church, and the courtyard was transformed into a square cloister. This cloister is unusual in having the function of a rainwater collection space rather than a garden; the center of the cloister contained four great funnels, which fed rainwater into the large cistern below them.

The water was filtered through the thick layers of pozzolana (local volcanic stone) and gravel which filled the funnels, and drawn out again from the cistern through a well to the side. This remarkable technical solution is for the moment unique among Italian medieval buildings, and suggests the sophistication of the architect employed by the abbey.

Documents of court cases from the 12th and 13th centuries show a continuing quarrel between the abbot of the monastery and the local elite, particularly the lords of the nearby castle at Sgurgola. Through these parchment records, we catch glimpses of the peasants on the abbey lands, who testify to the relatively onerous unpaid labor to which they were subject, as well as to their freedom to leave the estate, taking away the wood from their houses, as long as they continued to provide the work and rent they owed. Although the abbey regularly won these cases, the ascent to the papacy of Boniface VIII, who came from an elite family in Anagni, changed the balance of power. The monastery was suppressed, the abbot and the monks were excommunicated, and the estate was turned over to the cathedral of Anagni.

The Castrum

The 2007 team sits in the winery, in depressions which once held dolia.
The 2007 team sits in the winery, in depressions which once held dolia.

The last chapter in the long occupation of Villamagna is perhaps the least clear. A large defensive wall was built around the church and part of the area once occupied by the monastery, forming a castrum, or fortified settlement. The cemetery seems to have remained in use, but we are unsure about the buildings which lay inside the castrum wall. Our excavations revealed only a substantial lime kiln, which must have been used for the burning of stones from earlier Roman buildings on the estate. We do suspect, from the large number of finds relating to horse equipment, particularly spurs, that the cloister was used as a stable, and that the castrum might have housed a small mounted force. The last document relating to the site in this period is a sentence in a legal document from the cathedral archive for 1498, which records the burning of the site by a neighboring town: Villamagna combusta est.

If a general outline of the life of the estate over these many centuries is now clear, our excavations have, as usual, posed more questions than they have answered. Where was the residence of the emperor? Where were the main baths of the estate located? Was the winery replaced with a larger structure when it went out of use in the mid-3rd century? Who was responsible for the rebuilding of the church and the creation of the 6th-century winery? Who settled the estate in the 9th century? What form did the monastery take in its early phases? And what was its plan in the later ones? Was the castrum an elite residence or simply a fortification? What is obvious, however, is the central role this estate played in the life of the surrounding territory for over a millennium. The imperial investment on the site continued to pay dividends for many centuries.

Read “The Hungry Archaeologist – Anagni