Villae in Roman Dacia

See this text in


The indigenous Dacians and the Late Iron Age civilization as agents for the creation and the activity of the Villa type country-side estate in Roman Dacia

According to the Roman law, the entire territory of Dacia became ager publicus after the conquest – the emperor’s property to dispose of. The information left about the land ownership in Dacia indicates that, since the beginning of the province, most of the territory was distributed to veterans and colonists. The oldest distribution of ager publicus dates to AD 108/109 when the colony Ulpia Traiana Sarmizegetusa was founded. Its territory was measured and divided into plots (centuriatio), theoretically rectangular lots of 100 double iugera, or 200 iugera each. After being measured, the land was attributed to the veterans through lottery. The size of the estate depended on the military rank held by the veteran while in service. Unfortunately, we have no source for this particular centuriatio. There is no proof to sustain the idea that a part of the territory of Ulpia Traiana Sarmizegetusa was left for the usage of the indigenous people. Through adsignatio coloniaria the land was passed on from the Imperial estate to colonized citizens, thus becoming ager privatus optimo iure. No further land distributions for the discharged legionaries (missio agraria) took place after Trajan’s reign. The land was replaced by a certain amount of money (missio nummaria). Under Trajan and his successors land from ager publicus was given to other citizens and veterans, through individual assigning (adsignatio viritim), in various parts of the province. From the legal point of view, they received only right to use the respective estate (possessio). Other plots of land were sold to groups of Roman citizens or peregrines. As settlements evolved into towns, new urban territories appeared.

As we have previously showed when discussing about the rural settlements of Dacia, the indigenous presence is very poorly documented inside the strongly urbanized area from western Dacia. Indigenous communities are attested only in the eastern, less colonized region of the province. It can be easily noticed that these communities were not bordering Roman colonists but on the contrary, dwelled far away from the Roman towns. Nothing reminds us of the great Dacian settlements from the time before the conquest, with their diverse workshops, intense economic activities and even coin minting. The absence of fortification or imposing houses shows that these communities played no administrative part in the new, Roman network.

Concerning the indigenous population, it was hypothesized that it has been forcefully removed from the mountain area by the Roman army after the conquest. On the other hand forced migrations of people within a province are not documented anywhere else and they were not a part of the Roman policy. A letter of Vespasian to the Spanish town of Sabora, sanctioning a dislocation of population to the plains, doesn’t seem to indicate that movement was forced by the will of the emperor. That this assumption lacks chances to prove right is confirmed by the fact that archaeologically speaking, no great movements of population are recorded: until now, no Dacian settlement from the Roman epoch is known for sure to have begun after the conquest, on a new territory.

Under these circumstances, how can then one explain the absence of natives in the fertile Mureş Valley throughout the Roman rule. The main argument of the sustainers of the forced movements of population is the final scene on the Column, where the Dacians, accompanied by their cattle and escorted by Roman soldiers are expelled from the mountainous area of Sarmizegetusa. This interpretation is less likely today, and contrary to what most of the annalists of the Column believe. The direction of evolution for the characters of the final scenes coincides with the reading direction of the events; it shows that the Dacians and their cattle were being escorted to the right, or, from the direction of the story told, to the north side of the province, probably in a deportation action.

All in all, the Roman network of settlements is very different from the one belonging to the Late Iron Age, and the native population is absent in these areas, both as compact communities and as inhabitants of the settlements founded by colonists.  

Inside the territories of Ulpia Traiana Sarmizegetusa and, eventually, other towns, small and middle-sized estates of the colonists and veterans, who received properties from ager publicus, emerged; they are known as villae. In some cases, traditional, hand made pottery was found and attributed to the natives.

By studying the rural environment of the villae J.T. Smith determined that there are substantial resemblances between the rural properties from Dacia, Pannonia and Moesia, based mainly on the size of the estates. The urban part, or the house itself, is usually modest in size and luxury, with rather small rooms. There is obviously a great difference between the villae from Dacia and the ones from Italy or the Western provinces. The same author tried to explain this difference through a local tradition of construction. This hypothesis is adventurous and does not take into account the type of habitat and the pre-Roman social structure of the Late Iron Age Dacia. Recently, I. Oltean, after mechanically adopting the theoretic assumption of Smith, suggested local origins for the architecture of the villae, although until now there is no proof to sustain this theory. Since the part of the villa containing the house of the owner is essentially an urban piece of architecture, some houses inside the towns could have had pre-Roman, local design. One of the arguments used is the plan of a dwelling from the Getic settlement from Popeşti, in The Plain of Muntenia (region which was never a part of Dacia, but, only temporary between AD 102-117, one of Lower Moesia); it is irrelevant as analogy for Roman Dacia, not only because of the different cultural environment, but also because of its chronology. The settlement from Popeşti ceased to exist sometime during the 1st century AD, when the Romans were establishing their frontier on the Danube, almost a century before the conquest of Dacia. Henceforth, any influence of it inside Roman Dacia is excluded.

Therefore the difference between the villae from Dacia and other provinces is firstly a question of quality, social structure and an economic environment less prosper than in other provinces. The Roman civilization did not intersect the Dacian one in this domain, just as it has not in others. The traditional hand made pottery found in some of the villae show that the natives may have been present as household slaves or free employees.


Types of villae and their distribution

No large sized estates called latifundiae were identified in Roman Dacia and only 11-12 villae have been archaeologically researched. The only construction which might suggest a large property is the one from Dalboşeţ, in Banat. Here, a stone building of 80 x 85 m with approximately 40 rooms was identified. The absence of archaeological material leaves the dating and the destination of this building uncertain. The rest of the properties seem to have between 200 and 1000 iugera, which makes them small or middle-sized estates. Their main distribution area is in Upper Dacia and Dacia Porolissensis, namely around towns like Sarmizegetusa, Napoca, and Apulum. They are far less common in Lower Dacia, and absent in eastern Transylvania. Consequently, their spreading area coincides with the colonized and urbanized region of the province.

The villae seem to have reached the peak of their prosperity during Severan time. They are located in valleys, on hills terraces, close to running waters and roads, mostly in the proximity of towns or military forts. The plan of the buildings, their number and position places them among the ones with several constructions dispersed inside an enclosure, in the C type of Henning’s classification.

The villae from Roman Dacia were inhabited throughout the year; the owners who have chosen to live for a part of the year inside the town were few.

The economy of the villae

Villa type properties were first of all small agricultural production units. Their capacity was rather small and their production held only local commercial importance, most of it being destined for internal consumption. The appearance of the buildings and the inventory do not suggest prosperity. The number of coins uncovered is sparse, which seems to confirm this image. Between 1 and 5 annex buildings, used as barns, stables, workshops of dwellings for the workers were uncovered in the researched villae.

Some of them hold clues concerning rural industry. For example at Hobiţa, in the territory of Sarmizegetusa, 11 tiles with the stamp LDP were found in a tower which was probably the dwelling of a villicus; this find seems to point out the existence of a brick kiln .

In the same modern village of Hobiţa the brickyard of M. Servius Donatus (MSD), another local esquire, is attested. At Peşteana, the villa had a pottery kiln; several moulds for casting imitations of terra sigillata were found. . Some unfinished stone grids were uncovered at Deva.

According to the present stage of research, the contribution of the villa owners to the economy of the province seems to have been limited. The estates were not important production centers and their industry was a rather menial one.


Sanctuaries inside the villae

The information concerning the sanctuaries from villae is indirect, consisting mainly of inscriptions. Six votive altars dedicated to Fortuna Conservatrix, Juno Regina, Apollo, Mercurius, Minerva, and Hercules Magusanus were found in the villa from Ciumăfaia (Cluj County). A small altar dedicated to Silvanus was found inside the ruins of a Roman building from Dezmir (Cluj County) and another one dedicated to Silvanus Domesticus at Ciumăfaia.

Some painted plaster was found in the room where this last altar was uncovered, hence a small chapel is suspected. A bronze statue of a house-god was discovered inside the villa from Deva (Hunedoara County).

An inscription dedicated to Iupiter Optimus Maximus comes from a possible villa from Peştişul Mare (Hunedoara County).

Two other inscriptions dedicated to Fortuna and Fortuna Praenestina are known from Hăţăgel (Hunedoara County) another area rich in Roman ruins, close to Sarmizegetusa. Finally, archaeologists from the 19th century have identified at Valea Sângeorgiului (Hunedoara County) a building containing numerous sculptural fragments and votive inscriptions, a probable cult place, without offering further information. As one can easily see, the image of the sanctuaries from the rural villae is poorly contoured due to the present stage of research.


Villae owners

In what the owners are concerned, most of the information comes from the archaeological research of the area called pars urbana, namely the buildings which formed the owner’s adobe and from the study of inscriptions. The houses are rather small, except the one from Dalboşeţ (Caraş-Severin County).

They consisted of several rooms with both stone and wooden walls, and a tile roof. Most of the rooms were paved in clay, but several in flat bricks or opus signinum, as at Sâtămărie Orlea (Hunedoara County). In some cases heating and baths were present. At Aiud (Alba County), Deva (Hunedoara County), and Mănerău (Hunedoara County) the buildings’ façade had a portico, while at Gîrbău (Sălaj County), Ciumăfaia (Cluj County), and Apahida (Cluj County) they had one or several external apses.


Most of these villae were permanently inhabited by their owners; in rare cases they were just summer residences, like the ones from Ciumăfaia (Cluj County), or Hobiţa (Hunedoara County).

According to the divinities mentioned in inscriptions, the owners were, most of them, first wave of veterans and colonists. Some of the most interesting are the owners of the villa from Ciumăfaia, Aelius Iulius veteranus ex centurione, who had received the land at the end of his military career, and his son, P. Aelius Maximus, also a military who reached the Rhine, pursued an equestrian career and became decurio, duumvir quinquennalis, and flamen in colonia Napoca. It is obvious that he resided in the town and used his country villa only as an occasional place to spend his holiday.



After the Roman conquest, the territory of Dacia – turned into ager publicus – was split in lots and distributed to the colonists and to the veterans from legions which had participated in the war. After this massive distribution of land, only individual plots were assigned to people, but with a juridical status of the property reduced when compared with the original one.

The indigenous inhabitants of the province seem to be absent from the strongly colonized areas, where most of the villae were identified. Their absence from the most fertile regions of Dacia can by explained by a possible deportation at the end of the war. The native communities identified in eastern Transylvania are modest and appear to be outside the Roman economic and administrative system.

Most of the villae researched until now are small or middle-sized properties, without any direct connection with the pre-Roman rural settlements. Their size and their reduced numbers are influenced by economic factors specific to the province. Eleven or twelve such vilae have been researched, most of them being small or middle-sized estates, with several buildings inside a surrounding wall. Their role is predominantly agricultural, but a modest industrial production, meant to satisfy the internal needs of the household and a local limited trade, was identified.

Information concerning the spiritual life of these complexes are given by the discoveries of votive altars inscriptions and statuettes of divinities uncovered inside the researched villae. They reflect the origin and the status of their owners. Most of them were middle-class, but there are also exceptions – high rank individuals who spent most of their time in the towns where they undertook public careers and who used their country-side house only as an occasional place to spend their leisure time.

As a matter of fact, the appearance and the accommodations of the houses mirrored the social status of the owner.




J. Henning, Südosteuropa zwischen Antike und Mittelalter, Berlin, 1987 (Schriften zur Ur und Frühgeschichte 42).

I. Mitrofan, Villae rusticae pe teritoriul Sarmizegetusei, în Sargetia, XI-XII, 1974-1975, p. 291-297.

I. Mitrofan, Villae rusticae în Dacia Superioară (I), în ActaMN, X, 1973, p. 127-150.

I. Mitrofan, Villae rusticae în Dacia superioară (II), în ActaMN, XI, 1974, p. 41-59.

J. Percival, The Roman Villa. An Historical Introduction, London, 1976.

D. Popa, Villae, vici, pagi. Aşezările rurale din Dacia romană intracarpatică, Sibiu, 2002.

Coriolan Opreanu