A Local Design In Late Antiquity. A different view.

M.C. Kosian

In his latest book about the campaigns of the Archaeological Institute of the Free University of Amsterdam in Valesio (Salento, Brindisi district, Italy) prof. Boersma presents the results of the resaerch on the town of Valesio, wich the Free University had studied in several campaigns. A part of this research was for excavation of a Roman bath-building.
In his book prof. Boersma gives a reconstruction of the design of these baths using a orthogonal geometric design1. In this article I want to enter a discussion on Roman design, using these baths as an example, for I have had the opportunity to study these designs closely, for wich I sincerely thank prof. Boersma.
In my opinion several orthogonal reconstructed design-schemes from the Roman period need to be reconsidered. To me these rectangular designs are far too complex for local workers, and not necessary for Roman demands.

Valesio, the baths.

The baths functioned as a posting-station along the Via Traiana 'Calabra', the main-road between Brindisi and Otranto. The station in question was called mutatio Valentia and stood halfway Brindisi and Lecce, in the centre of a large pre-Roman Messapian town which existed until well into the Hellenistic period.


When the station was to be build in the first decades of the fourth century A.D., the design should match several requirements;

To study a Roman design one should look first at the rules given by the Roman architect and engineer Vitruvius. In his fifth book, chapter ten, on Architecture Vitruvius gives the rules according which baths should be build. These rules are based upon a design-scheme of an orthogonal grid of rectangles with a ratio of 1:3.3 Such a grid demands highly skilled architects and construction-workers, because they should be able to create such a modular grid onto the building-site. Comparing the baths of Valesio with this Vitruvian rule shows that this particular proportion was not used by neither the architect nor the construction-workers.
Another design-scheme was the more ancient system of circle- division. In this scheme a circle is divided into certain angles which form the diagonal of a basic rectangle. A circle can be divided into eight 45 degr. angles (4-8 system), into six 60 degr. angles (6-12 system) and into five 72 degr. angles (5-10 system). Of these three systems the 6-12 system is the most simple, based upon a equilateral triangle. This system gives an aesthetically good proportion of 1:tan(60) = 1:1.732. This design-scheme demands far less skilled workers, because the angles can easily be created in the field with ropes. Being build by local craftsmen3 it is highly probable that the baths of Valesio were designed according to this simple, traditional scheme. Yet, being it Roman baths we cannot simply spot the basic equilateral triangle, as we often can in Greek architecture; the architect used in his design the Vitruvian rules for the lay-out of baths. According to these rules the actual bathing-rooms should be on the south-side of the building with windows opening to the southwest, or if that is not possible to the south (Vitr. V.10.1). To match this rules the architect situated the bathing-rooms on the south-side, planned a open courtyard with a well next to them and has his entrance and the stables on the north-side of the building.

The design was probably based upon a rectangle and a square. The basic rectangle forms the central part of the building and is based upon a 60o angle which determine its diagonal. If we assume a thickness of two feet4 the base of the rectangle would measure 47 feet.5 The length of this rectangle was determined by its diagonal = 47 * tan(60) = 81.406 feet = appr. 81' 8". This rectangle was divided lengthwise into halves. A square was drawn on the south of the rectangle of 47' * 47'. This square projected to the east out of the rectangle. The projection was proportioned on the eastern half of the basic rectangle so, that the projection stands to the width of the half rectangle as 1 : tan(60).
In the basic rectangle another rectangle was created on the north-side, also based upon a 60 degr. angle, of 47 / tan(60) = 27.135 feet = appr. 27' 2". This rectangles determined the size of the stables (rooms XVI, XVII). In the eastern half of the basic rectangle another rectangle was formed directly south of the stables. This rectangle had the dimensions (47/2 =) 23' 8" * (23.5 * tan(60) =) 40.703 = appr. 40' 12" (40 3/4'). This rectangle determined the central hall (room III) and its entrance corridor (room II). The wall between the central hall and the corridor was determined by the north-side of the projecting square.
The length of the central hall was divided into two, which determined the walls between both the apodyterium (room IV)6 and the frigidarium (room V), and the courtyard (room IX) and the great room X. The remaining south-east corner of the square determined the first tepidarium (room VI). copying this room to the west (inside the basic rectangle) gave the place of the second tepidarium (room VII). The caldarium (room VIII) was placed in the remaining southwest corner of the square. This determination of the bathing-rooms excluded the wash- basins, but this is in accordance with Vitruvius, who says that the proportions of the bathing-rooms should exclude the basins (Vitr. V.10.4). The basins are added to the bathing- rooms according the given lines of the basic rectangle or, in the case of the frigidarium projecting from the building with the same ratio as the basic rectangle (width room V : length room V incl. basin = 1 : tan(60)).

This was the basic design of the building. To this design two furnaces were added (rooms XII, XIV), one for the tepidaria and one for the caldarium, and two small service- rooms (rooms XIII, XV) were added on the west-side of the building. But as they were on the backside they were not built in proportions (they were not meant to be visible). An entrance portico (area I) was also added, based upon the projection of the square. This portico (and also the eastern stable) bent slightly westward on account of some now unknown structure (the course of the ancient road?). Another not in the design incorporated addition was a small latrine (room XI) in the open corner south of the projecting basin of the frigidarium. This latrine was not part of the building (it has an opening to the south), but was probably built on that place so that it could use the drains coming out of the baths through the east-wall of the frigidarium7. To guarantee an independent water-supply a well was dug out in the courtyard.

We can distinguish three zones in this design:
If we assume these zones as being correct the function of the great room X was public, possibly it was a rest-room or even a dinner room opening to a for guests accessible open courtyard were they could sit in the open, but sheltered from wind (like the peristyles in houses).


Of all the rooms only the caldarium had wall thick enough to support vaultings. The other rooms have had saddle-roofs or even flat roofs. The central hall was probably built proportionally higher than the surrounding bathing-rooms. Except the caldarium all bathing-rooms had saddle-roofs projecting from the central hall. This provided good rain- water discharge, and gave more strength to the building because the ridge-beams were supported by the walls of the central hall. This roof placement also provided large, high placed windows opening to the south for the tepidaria. Over the vaulting of the caldarium a saddle-roof was built east- west, draining the rain-water into the courtyard and to the south of the building. All service-rooms and the stables, and probably even room X have had flat roofs. The southern furnace was vaulted, yet these vaultings were not roofed. The stables were also built proportionally higher than the other rooms, and supported the roof of the entrance portico.


table 1.
table 2.
table 3.


Basic design-scheme of the Baths of Valesio.
Basic rectangle with a ratio of 1:tan(60) combined with a square proportionally projecting from it. A second rectangle with a ratio of 1:tan(60) is constructed to determine the southern walls of the stables.

Design-scheme of the Baths of Valesio.
In the basic rectangle a third rectangle with a ratio of 1:tan(60) is constructed to determine the central hall and its corridor. The basic square is containing the bathing- facilities and the basic rectangle the reception-facilities. Basins and service areas are not added.

Valesio Baths. In blue: Actual state.
In red: Reconstruction based upon the general design-scheme. Basins and service areas added. S: stables; F: frigidarium (with its basin added it forms a rectangle with a ratio of 1:tan(60)); T: tepidaria; C: caldarium; B: basins; P: furnaces; W: well in courtyard; L: latrine. Deviation of portico and eastern stable caused by the unknown structure (road?) is ignored.

Reconstruction of elevation.


*. I want to thank prof. Boersma for giving me the opportunity to study the design of these baths by giving me access to the needed information. Information about these baths as well as about the town of Valesio could be found in:

Field-data: prof. dr. J.S. Boersma
field-drawing: H. Burgers
analysis and drawings: M.C. Kosian
other ideas on classical design by the same author:
Ontwerpsystemen en Vitruvius' terminologie

1. The excavations were carried out by prof. J.S. Boersma of the Free University of Amsterdam (Holland) from the mid 80's till the early 90's.
2. This assumption is based upon the used materials and the construction techniques of the foundations.
3. Assuming that the building-costs should be kept as low as possible, the use of local craftsmen and local materials is far more cheaper than high skilled craftsmen from other places.
4. The used Roman foot that is assumed to be used for the building is 0.296 m., and was divided into 16 fingers. So 1.5 feet = 1' 8" and so on.
5. length room X (east-west) + length room II (east-west) + walls. For the internal measurements of the several room see tables 1 and 2.)
6. The identification of room IV as the apodyterium is based upon its location.
7. The identification of room IV as the apodyterium is based upon its location.
Back to Roman-page