The Economics of Roman Construction in Bracara Augusta (Braga, Portugal): Building Stone Identification and Working Costs

Abstract

Bracara Augusta (Braga, Portugal), one of the most important cities of the NW Iberian Peninsula, was founded by the emperor Augustus approximately between the years 16/15 BC. Throughout the Roman period, it has undergone various monumentalisation programmes according to its status, the first one as a convent capital and a later one as a provincial capital. The investigation of the economic cost of construction in the ancient world allows us to understand Roman architecture and the society of builders, both from economic and social perspectives. In recent years, based on the works of Janet DeLaine and Paolo Barresi, despite the large number of variables and the difficulty of quantifying them, a calculation methodology has been developed to approximate the cost of architecture, based on the information contained in two main documents: the Diocletian’s Edict of Prices (301 AD) and the nineteenth-century architecture and engineering manuals summarised by Pegoretti. Our goal is to understand part of the construction process of Roman architecture in the NW Iberian Peninsula, considering the cost of raw materials, namely stone, and the labour required in its preparation and utilisation. In this study, we present a cost proposal associated with the monumentalisation of two of the city’s important buildings: the forum’s large buildings and the theatre. The construction of the city was made fundamentally using local stone, from different quarries, endowing it with notable buildings and requiring a strong contribution from the local elites.

1. Introduction

According to Ben Russell [1], the study of the economics of the Roman stone trade is important because stone objects are among the most permanent material vestiges of Roman antiquity, with benefits for historians and archaeologists, and their study can provide valuable and relevant information on the Roman construction process and economy: origin, extraction, manufacture, transport, distribution, and installation. Several studies on the constructive processes of ancient architecture have been developed in recent years based on the pioneering work of Janet DeLaine [2], namely concerning the construction costs, with strong economic and social implications for cities’ lives. This work aims to approach the quantification of the cost and time used in the production of stone architectural elements in the public architecture of Bracara Augusta (Figure 1). To do so, we selected two case studies with an Early Empire chronology and similar techniques, where the main raw materials are local stone: a set of monumental architectural elements collected in the city, possibly related to the large public buildings of the forum, and the columnatio of the Roman theatre. In the first case study, we analysed the production of thirteen column bases and three large capitals, with a hypothetical architectural framework. In the second case study, we considered a set of nine pieces discovered during the excavations of the theatre and assumed to be part of the columnatio, analysed according to the graphic restitution of Ricardo Mar [3].

This study of two of the biggest public investments in the city in the Early Empire, the construction of the administrative and religious forum and the theatre, is part of a larger project, which aims to study the economic and social component of the construction of the Roman city and highlights the importance of the interdisciplinary study of building materials [4]. To estimate the economic cost of construction, we applied a calculation methodology developed in recent years, based on Janet DeLaine’s work on the Baths of Caracalla [2], and later improved by other researchers [5,6,7,8,9,10,11,12]. This methodology is based on three fundamental tools: (i) the Edictum Diocletiani et Collegarum de pretiis rerum venalium, dating from the beginning of the fourth century, which informs about the prices of some products and wages [13], the values of which can be extrapolated to previous periods; (ii) Pegoretti’s manual of architecture [14], which allows an estimation of the work effort and execution times in pre-industrial construction processes; and (iii) the related archaeological sources for which there is no written information. It is a challenging investigation, as it involves multiple variables and some of them are difficult to recover through archaeology, such as the use of enslaved people’s labour.

The application of the calculation methodology to these case studies provides interesting results, namely, the cost of work, even if partial, which implies some financial availability and can be compared with the costs of other buildings, and also the origin of the raw material, the skill of local craftsmen, and the architectural setting of some of the pieces considered.

Quantifying manpower and construction costs is a tool to understand past societies.

The methodology needs to be improved, reinforced, and continuously applied to other buildings, allowing, as Janet DeLaine proposes, “to add a new dimension to discussions about the act of building and its socio-economics implications” [15].

2. Materials and Methods

The methodological basis of this research line comes from Janet DeLaine’s study on the construction process of the Baths of Caracalla [2]. This calculation methodology is based on archaeological/historical and epigraphic data and on two fundamental documents: the Edict of Diocletian, which provides prices and wages for more than 1000 items at the beginning of the fourth century [13], and the architecture and engineering manuals of the 19th century, summarised by Pegoretti [14], which provides labour constants for a variety of stone types and stone-carving projects (column shafts, capitals, and simple rectangular blocks) and establishes the necessary times for carrying out the main tasks associated with the architectural process. As stated above, this calculation methodology has been refined in recent years by several authors. Thus, regarding local stone, much of the advances are owed to the work dedicated to the Provincial forum of Tarraco [10], as well as the analysis of the cost of the ornamental program of Cartagena’s Roman theatre [12]. Recently, Javier Domingo [7,8,9,16] sought to validate the results obtained through the calculation methodology, comparing them with data provided by epigraphy and ancient literature, which revealed the cost of part/or all of certain buildings.

We made use of the values set out in the Edict of Diocletian, which can be converted, if necessary, to previous times, extrapolating prices based on the evolution of the price of wheat between the 1st and 4th centuries, as well as the indices and parameters established by Pegoretti [14].1

In this type of exercise, we must not forget the variables that are very difficult to consider and situations that are impossible to detect through archaeology, one of our main sources, such as the weather conditions, the contribution of enslaved people’s labour, occasional offers of evergets, or even the existence of distinct economic realities in the space dominated by the Empire.

Giovanni Pegoretti [14] points out that the tasks linked to the elaboration of a decorative stone element are articulated in four main stages: (i) quarrying, (ii) rough squaring at the quarry (sbozzatura grossolana), (iii) preparatory dressing (apparecchio o taglio rustico), and (iv) finishing, usually already carried out on site. In the case of Attic bases and Doric capitals (extensible to the case of Tuscan models), Barresi, based on Pegoretti’s indications, suggests that the finishing should include the roughing and chiselling of the shapes [5].

It was likely that the stone blocks would arrive at the building sites roughly squared, so it would be the workshops installed in situ that would carry out the most delicate tasks of the work.

In this exercise, the quantities, volumes, and weights are calculated. The estimation of costs involves converting these values into hours of work and monetary costs. The restitution of the total volume of stone used (gross volume) in each of these architectural elements takes into account, however, that parallelepiped blocks were extracted in the quarries that were only squared and semi-elaborated later, until the approximate shape of the pieces was obtained, be they bases, shafts, or capitals [8].

In our case studies, we do not consider the costs of transport or the installation of the elements, for the sake of consistency, since in relation to the city forum, it was only possible to recover the set of architectural elements here under analysis.

In our analysis, we fix certain constants, namely, working days (wd) of 10 h and average daily wages of 2 sestercius (HS) (plus maintenance). To determine the price of the raw materials, we take into account the work of Paolo Barresi [5], who provides us with a table with the conversions of the prices of the marble contemplated in the Edict in the context of the 1st/2nd centuries, which allows to estimate the price of local stone exclusively used in the two case studies considered, assigning to it a value four times lower than that of the cheapest marble mentioned in the Edict, which would be roughly equivalent to 1 HS per cubic foot (ft³) [10].

The description of the architectural elements is based on the work by M. A. Gutiérrez Behemerid [17], on the topic of capitals, and M.I. Escrivà Chover [18], for the analysis of column bases. The column bases, namely those that we assume are associated with the city forum, were organised into types, according to the methodology proposed by Pierre Broise [19], which is based on the various possible combinations of Tuscan capitals, from the equine profiles, and of the bases, according to the tori. The bases are thus distributed in two major types: TSTA (torus, scotia, torus, apophygè) and TSTCI (torus, scotia, torus, cavetto, imoscapus). As for the Tuscan capitals, there is only one specific model, determined by the TC mouldings (torus, cavetto).

The stated dimensions are height (Ht) and diameter (Ø) for bases and capitals, adding width (W) and depth (Dp) for other elements, in centimetres (cm).

The description of granitic rocks is carried out by macroscopic observation of the rock characteristics, such as mineral assemblage, grain size, texture, and colour, to avoid destruction of the architectural elements. This description is compared with the observation of granitic rocks in the field and with the description of granitic rocks in the explanatory notes in 5C and 5-D sheets, the geological map of Barcelos and Braga [20,21], and also with detailed historical and bibliographic research. The analysis of the state of conservation of the granitic rocks is also made by macroscopic observation by analysing the weathering of minerals and the formation of secondary minerals, such as clay minerals, Fe-oxides, and sulphates.

3. The Workshops, Quarries, and Raw Materials

The Roman buildings in Braga were mostly made with local stone materials, mainly granitic rocks. To date, it has not yet been possible to complete the process of identification/location of the quarries that supplied the Roman city. However, the work carried out, based on a macroscopic analysis of the different types of stones present in the Roman buildings studied, combined with a good knowledge of the geology of the region, made it possible to identify the major types of granitic rocks used and the large areas from which they were extracted. This kind of approach incorporates, however, some difficulties, as the same granitic rock pluton can exhibit various facies. Specifically, four major types of granitic rocks were identified: Braga granite, Gonça granite, Póvoa de Lanhoso granitic complex, and Barcelos granitic rocks, differentiated in terms of grain size, texture, and colour (Figure 2 and Figure 3). The substratum of the city of Braga, of which we detected signs of exploration, corresponds to a biotitic monzogranite with rare muscovite, of medium to fine grain, with some feldspar fenocrystals, and has different types of enclaves, with mafic microgranular enclaves as the most abundant but also metasedimentary enclaves [21]. The Gonça monzogranite is two-mica granite and is fine-grained, with a tendency toward a porphyritic texture, showing some chemical weathering [21].

The Póvoa de Lanhoso granitic complex is mainly composed of monzogranites with different facies, is predominantly biotitic with rare muscovite, medium- to fine-grained, and has a porphyritic texture. It also exhibits large greyish microgranular enclaves and, more rarely, metasedimentary enclaves [21]. It is a slightly weathered granite that is usually associated with more sparse fracture networks, which allows the extraction of large blocks but offers greater resistance from the point of view of architectural work, precisely due to the characteristics of its grain. The granitic rocks that occur in Barcelos are predominantly monzogranites, forming a large outcrop, mainly located in the higher parts of the Cávado river basin, on the way out of the city in the direction of Barcelos, occupying the villages of Sequeira, Gondizalves, Pousa, and Areias de Vilar. It is a leucocratic rock, predominantly biotitic, often coarse- to medium-grained but occasionally is fine-grained, with a porphyritic texture [20].

4. Stone Working and Stone Workers

The Romans inherited quarrying techniques and stonework tools from the Greeks [22], which remained unchanged until the mechanisation of the last century. The first task was extraction in open-pit mines or underground galleries. The process consisted of isolating blocs by digging vertical narrow trenches with a specific pick and isolating the stones at their base by the corners. The extracted blocks were then taken to an area where the first carving work was carried out. This presupposed the existence of roads in the quarries themselves and between these and the stonework spaces/workshops. In the next step, the materials were disposed of by water and land, an arduous and costly task. At the building site, it was about finishing and placing the elements in place with lifting devices and levers.

According to Robert Bedon [22], in the different quarries of the Empire, it was possible to find four different statuses of workers: military, free, servile, and penal. In the case of Braga, we know little about the workers who built the city; however, at the time of its foundation ex nihilo, under the Emperor Augustus (27 BC–14 AD) [23] and later during its different major phases of urbanisation/monumentalisation, especially under the Flavians and Antonines (69–192 AD), several construction sites were set up near the great public monuments and the great domus of the city, although archaeology has not yet documented their direct traces. Construction would be one of the most dynamic activities, employing a significant part of the population. The population of builders was mainly made up of the indigenous component and their descendants, both in terms of the commissioners of the works and in terms of artisans and other workers. These workers are mostly anonymous, except for one or two cases related precisely to stone working and revealed by epigraphy. These occurrences are not from the city but from its territory. Reburrinus appears on a possible ara, discovered in the area of Vizela, which indicates his profession: lapidarius [24]. Elpidius, a libertus, possibly a pilgrim, owned a workshop that made stone monuments—officinator—and his name appears on an altar identified in Santa Cruz do Lima—Viana do Castelo [24]. Marcus Pelcius and Dureta, both stone workers, are revealed by an inscription discovered in Afife—Viana do Castelo [24] (Figure 4). Arcius signed a funerary tombstone found in Ponte da Barca and would be a sculptor and lapidarius [24,25].

As for the designation of professions, in Rome, workers of hard materials such as stone were called fabri, who then received a specific epithet depending on their type of work, such as lapidarii, marmorarii, metalarii, quadratarii, structores, and sculptores. It should be noted that, on this issue, Armando Redentor indicates that lapidarii, fabri lapidarii, lapicidae, or marmorarii would be common names for stone workers [24].

Metalarii worked in the extraction of raw materials in the quarries and in the mines, although recent studies, given the ambiguous character of classical texts, suggest that they may not necessarily be the workers but the individuals who exploited these spaces [26]. Roman stone carvers would mostly have been generalists, capable of undertaking a range of work [27]. Quadratarii carved the extracted blocks and serrararii used saws for this purpose. Lapidarii, stone cutters, with greater specialisation than the previous ones, worked any type of stone directly in the quarries or in nearby workshops, giving the desired shape to the stone, producing the ashlars, voussoirs, and architectural elements (sculpted decorations, cornices, and columns) that the works required.

Lapicidae were the stone workers who specialised in engraved the stones. Marmorarii were the stonemasons who worked marble, a noble stone, with greater prestige than simple lapidarii. Sculptors made sculptural details, mouldings or ornamentation, or statuary. Finally, structores or lapidarii structores worked in masonry with small stonework.

Regarding salaries, in the case of free workers, the Edict of Diocletian provides some values for the 4th century. A marmorarius received 60 denarii (D) per day, plus maintenance; a stonemason received 50 D per day, with maintenance. As for the less skilled labourers, perhaps they were paid a similar amount to farm workers, with 25 D a day, with maintenance.2 The figures given in the Edict range from 20 to 60 D per day, with more specialised professions being associated with wages of up to 150 D. Regarding the standard of living, according to the same document, the intermediate value of 60 D allowed the purchase of 7.5 libra of beef, 2 chickens, or even a little more than 1 modius (8.62 litres) of wheat.

5. Results

The results for the two case studies are presented below.

5.1. The Great Architectural Elements of the Forum

5.1.1. Historical Background

Sixteen large architectural elements were collected in the city of Braga, mainly column bases and some capitals, most of which are on display in the gardens of the D. Diogo de Sousa Museum. Their exact provenance has been lost, but given their characteristics, they may have a connection with the administrative and religious forum of the city and the large public buildings that existed there. Curiously, some of these elements are referenced as coming from Largo Paulo Orósio. The forum of Bracara Augusta is still little known, but, according to a sixteenth century map (Braunio’s map) (Figure 5), it was situated in the highest part of the town [28], precisely where the aforementioned Largo Paulo Orósio is located today. Martins and Carvalho [23] indicate that, based on the topography of the site, it is possible to say that the forum would have been of rectangular shape, with its major axis following a NNE–SSW direction, with an approximate width of 294 feet and length of 600 feet. Rui Morais [29], in addition to the architectonic elements analysed here, mentions a set of elements that may come from the forum area: a bronze fragment of an equestrian statue; an inscription of a statue pedestal, dedicated to Agrippa Postumus; and a large quadrangular tombstone engraved with the inscription “Bracara Augusta”. The set to be analysed includes 13 Attic bases and three capitals, one Ionic and two Tuscans. In terms of typologies, all the bases are in the Attic style, without a plinth, with two unequal tori, separated by a scotia delimited, in some cases, by two fillets. The Ionic capital is smooth and features a square abacus, resting directly on the echinus, without a channel, separated from the hypotrachelium by a smooth collar. The pulvini, equally smooth, join in a balteus formed by two vertical stripes. It belongs to type 6 of Gutiérrez Behemerid [17]. The Tuscan capitals have an echinus in the shape of a torus topped by a quadrangular abacus (Figure 6).

5.1.2. Description/Catalogue (Table 1, Figure 7 and Figure 8)

As indicated in the methodology, the bases and capitals were grouped into types. The TSTA type includes five variants (1, 2, 3, 4, and 5) and the TSTCI type has three (1, 2, and 3), ordered by diameter and height in ascending order. With the exception of the TSTA-2, TSTA-3, and TSTCI-2 variants, with three, three, and two specimens, respectively, all of the others are represented by a single element.

The initial set consists of 16 granite elements, including 13 bases and 3 capitals, which are located at the D. Diogo de Sousa Museum of Archaeology, Braga. The bases show eight clearly distinct specimen diameters and heights, which are analysed here.

Figure 7. Monumental bases of column from the forum of Bracara Augusta: (1–8) Attic bases.

Figure 7. Monumental bases of column from the forum of Bracara Augusta: (1–8) Attic bases.

Table 1. Description of the large architectural elements (column bases and capitals) from the forum of Bracara Augusta.

N.º	Object	Museum Number	Type	Findspot	Production Date	Dimensions (cm)	Description
1	Base	1992.0662	TSTA–1	Near Largo Paulo Osório (?), Braga, Portugal	1st/2nd century	Ø. 57.5 Ht. 37	Attic base of column, without plinth, consisting of two semi-circular tori, divided by an arc profile scotia. There are no visible fillets delimiting it. On the top, a fillet and a apophygè.
2	Base	1992.0661	TSTA–2		1st/2nd century	Ø. 63 Ht. 40	Attic base of column, without plinth, similar to above, but with a higher and larger imoscapus. No visible fillets delimiting the scotia.
3	Base	1992.0653	TSTA–3		1st/2nd century	Ø. 67 Ht. 48	Attic base of column, without plinth, similar to above, but with a higher and larger imoscapus. No visible fillets delimiting the scotia.
4	Base	1992.0652	TSTA–4		1st/2nd century	Ø. 86.5 Ht. 50	Attic base of column, without plinth, similar to above, but with larger dimensions. The upper and lower tori are divided by a scotia, placed between two fillets. On the upper torus, there is a fillet and a cavetto.
5	Base	1992.0659	TSTA–5		1st/2nd century	Ø. 94 Ht. 50	Attic base of column, without plinth, similar to above, but with larger dimensions. The upper and lower tori are divided by a scotia, placed between two fillets. On the upper torus, there is a fillet and a cavetto.
6	Base	1992.0650	TSTCI–1		1st century	Ø. 60 Ht. 43	Attic base of column, without plinth. The upper and lower tori are divided by a narrow scotia, placed between two fillets. On the upper torus, there is a cavetto, a fillet, and a smooth imoscapus.
7	Base	CMB 004	TSTCI–2	Dom Afonso Henriques street, 42–46, Braga, Portugal	1st century	Ø. 60 Ht. 45	Attic base of column, without plinth, similar to above, but with different proportions, namely, a more pronounced imoscapus.
8	Base	1992.0657	TSTCI–3	Near Largo Paulo Osório (?), Braga, Portugal	1st/2nd century	Ø. 76 Ht. 42	Attic base of column, without plinth, similar to n.º 6, but with different dimensions.
9	Capital	1992.0660	TC		1st century	Ø. 90 Ht. 50	Tuscan column capital with simple abacus, echinus with torus profile and cavetto shaped necking.
10	Capital	1992.0656	6 (Gutiérrez Behemerid)		1st/2nd century	Ø. 78 Ht. 46	Ionic column capital with square abacus supported directly on the echinus, without channel, separated from the hypotrachelium by a smooth collar. The smooth pulvini join in a balteus formed by two vertical listels.

Table 1. Description of the large architectural elements (column bases and capitals) from the forum of Bracara Augusta.

N.º	Object	Museum Number	Type	Findspot	Production Date	Dimensions (cm)	Description
1	Base	1992.0662	TSTA–1	Near Largo Paulo Osório (?), Braga, Portugal	1st/2nd century	Ø. 57.5 Ht. 37	Attic base of column, without plinth, consisting of two semi-circular tori, divided by an arc profile scotia. There are no visible fillets delimiting it. On the top, a fillet and a apophygè.
2	Base	1992.0661	TSTA–2		1st/2nd century	Ø. 63 Ht. 40	Attic base of column, without plinth, similar to above, but with a higher and larger imoscapus. No visible fillets delimiting the scotia.
3	Base	1992.0653	TSTA–3		1st/2nd century	Ø. 67 Ht. 48	Attic base of column, without plinth, similar to above, but with a higher and larger imoscapus. No visible fillets delimiting the scotia.
4	Base	1992.0652	TSTA–4		1st/2nd century	Ø. 86.5 Ht. 50	Attic base of column, without plinth, similar to above, but with larger dimensions. The upper and lower tori are divided by a scotia, placed between two fillets. On the upper torus, there is a fillet and a cavetto.
5	Base	1992.0659	TSTA–5		1st/2nd century	Ø. 94 Ht. 50	Attic base of column, without plinth, similar to above, but with larger dimensions. The upper and lower tori are divided by a scotia, placed between two fillets. On the upper torus, there is a fillet and a cavetto.
6	Base	1992.0650	TSTCI–1		1st century	Ø. 60 Ht. 43	Attic base of column, without plinth. The upper and lower tori are divided by a narrow scotia, placed between two fillets. On the upper torus, there is a cavetto, a fillet, and a smooth imoscapus.
7	Base	CMB 004	TSTCI–2	Dom Afonso Henriques street, 42–46, Braga, Portugal	1st century	Ø. 60 Ht. 45	Attic base of column, without plinth, similar to above, but with different proportions, namely, a more pronounced imoscapus.
8	Base	1992.0657	TSTCI–3	Near Largo Paulo Osório (?), Braga, Portugal	1st/2nd century	Ø. 76 Ht. 42	Attic base of column, without plinth, similar to n.º 6, but with different dimensions.
9	Capital	1992.0660	TC		1st century	Ø. 90 Ht. 50	Tuscan column capital with simple abacus, echinus with torus profile and cavetto shaped necking.
10	Capital	1992.0656	6 (Gutiérrez Behemerid)		1st/2nd century	Ø. 78 Ht. 46	Ionic column capital with square abacus supported directly on the echinus, without channel, separated from the hypotrachelium by a smooth collar. The smooth pulvini join in a balteus formed by two vertical listels.

Figure 8. Monumental capitals from the forum of Bracara Augusta: (9)—Tuscan capital, (10)—Ionic capital.

Figure 8. Monumental capitals from the forum of Bracara Augusta: (9)—Tuscan capital, (10)—Ionic capital.

5.1.3. Identification and Weathering Degree of the Stone

It is likely that the monumental Roman column bases/capitals of the forum were built with Braga granite, which is a fine- to medium-grained porphyritic biotite monzogranite, with rare muscovite (Figure 9a,b). The monzogranite presents many enclaves, where mafic microgranular enclaves are the most frequent (Figure 9c). The Braga granite also contains metasedimentary enclaves which range in size from a few centimetres to a few meters (Figure 9a). The K-feldspar phenocrysts are frequent and can be several centimetres of length (Figure 9b). These Roman columns show rare signs of weathering, with the formation of Fe-oxides locally (Figure 9b,d). The Braga granite has been extensively used in the past in historical buildings in the city of Braga [30,31,32], mainly due to low transport costs. However, only the petrographic and geochemical signatures will allow the identification of specifically which type of granitic rock was used to build the monumental Roman column bases of the forum in the city of Braga, as well as to identify the weathering degree of the rocks and their conservation state.

5.1.4. Economic Estimate, Production Process, and Cost of Elements (Table 2 and Table 3)

Below is the calculation of the production costs of two of the elements, namely, for the Attic base (1) and the Ionic capital (10), in the four stages. The detailed calculations can be found in Appendix A.

Raw Material Extraction (Expressed in ft³)

To calculate the cost of the raw material used, we considered the gross volume of the material extracted from the quarry, which we multiplied by the estimated price of local stone, i.e., 1 HS/ft³, based on the value proposed by Mar and Pensabene [10] for the upper terrace of the Tarraco provincial forum. It is considered that this value already includes the cost of cutting and extracting the block.

To restore the volume of stone extracted, we added 5 cm to the measurements obtained from each element, on each side of the blocks. The reported values refer to the approximate volumes of blocks extracted from the quarries.

• Base 1
• 0.32 m³ = 12.32 ft³
• 12.32 ft³ × 1 HS = 12.32 HS
• Base 2 = 20.79 HS
• Base 3 = 23.1 HS
• Base 4 = 35.8 HS
• Base 5 = 38.5 HS
• Base 6 = 13.86 HS
• Base 7 = 15.4 HS
• Base 8 = 22.72 HS
• Capital 9 = 19.25 HS

• Capital 10
• 0.66 m³ = 25.41 ft³
• 25.41 ft³ × 1 HS = 25.41 HS

Rustic Work (Expressed in Labour Hours per m³ (lh/m³), in the Quarry)

After the block has been extracted, the rough work is carried out, which consists of assigning the desired size. The volume considered corresponds to one of the extracted blocks, which will lose mass throughout the various stone working stages. We used the variable of 448 h/m³ (h/m³), defined by Pegoretti for common granite [14—Tabella ottava] and a daily wage of 2 HS.

• Base 1
• 0.32 m³ × 448 h/m³ = 143.36 lh
• 143.36 lh/10 h (working days—wd) = 14.34 wd
• 14.34 wd × 2 HS = 28.68 HS

• Base 2 = 48.38 HS
• Base 3 = 53.8 HS
• Base 4 = 83.4 HS
• Base 5 = 89.6 HS
• Base 6 = 32.2 HS
• Base 7 = 35.8 HS
• Base 8 = 52.86 HS
• Capital 9 = 44.8 HS

• Capital 10
• 0.66 m³ × 448 h/m³ = 295.68 lh
• 295.68 lh/10 h = 29.6 wd
• 29.6 wd × 2 HS = 59.2 HS

Semi-Elaboration (Expressed in Labour Hours per m² (h/m²), in the Quarry)

Semi-elaboration is the second stage of the treatment given to the elements, after the initial roughing, which consists of assigning a rough shape according to the desired architectural function: base, shaft, capital, etc. It corresponds to the total surface worked.

For the bases and capitals, we applied the formula [a(1 + ${\displaystyle \frac{0.25}{x}}$)] [14], in which the variable a corresponds to 12 and × to the capital/base diameter in metres.

• Base 1
• 12 × (1 + ${\displaystyle \frac{0.25}{0.575}}$) = 17.22 h/m²
• Worked surface of the base = 1.32 m²
• Thus, 17.22 h/m² × 1.32 = 22.73 lh
• 22.73 lh/10 = 2.3 wd
• 2.3 wd × 2 HS = 4.6 HS|1.15 D

• Base 2 = 5 HS
• Base 3 = 5.6 HS
• Base 4 = 7.6 HS
• Base 5 = 8 HS
• Base 6 = 4.8 HS
• Base 7 = 5.4 HS
• Base 8 = 6 HS
• Capital 9 = 6 HS

• Capital 10
• 12 × (1 + ${\displaystyle \frac{0.25}{0.78}}$) = 15.85 h/m²
• Worked surface of the capital = 2.06 m²
• Thus, 15.85 h/m² × 2.06 = 32.7 lh
• 32.7 lh/10 = 3.3 wd
• 3.3 wd × 2 HS = 6.6 HS|1.65 D

• Finishing

The finishing is the last stage of the treatment given to the objects, in some cases carried out already in situ.

In the case of Attic bases and Doric capitals, a situation that can be extended to Tuscan models, Barresi [5] indicates that this was roughing and chiselling work, for which Pegoretti established a variable of 52 h per m² in the case of granite [14].

The question of the capital is more difficult, since Pegoretti [14] provides us with a numerical value for the finishing of an Ionic specimen in various materials but does not establish this variable for granite. In order to obtain approximate values, we chose a stone with values quite like those of common granite: black Portoro marble, the values of which are 1166.40 h for specimens about 0.45 m high [14], such as the one considered here. This value was later adjusted, based on other tasks carried out in granite mentioned by Pegoretti, to 1036.8 h, proportionally. Then, since this is a simply decorated specimen but made of a material that is difficult to work, we believe that it would be more correct to reduce this value by 1/3, giving a final value of 345.6 h. In the bibliography consulted, we found that Javier Domingo [8] made a similar adjustment in his estimate of the cost of finishing the Corinthian capitals used in the Volubilis Capitol, as they were made of smooth leaves and had little elaborate decoration. In this specific finishing stage, the level of specialisation of the workshop or workshops that made these objects should be considered, taking into account the difficulty and demands that were required of a lapidarius from Braga to transform into granite pieces the designs for materials that were easier to work with, such as marble or limestone.

In the capitals studied, like all those identified in Braga, the stucco that would have finalised the decoration was not preserved.

Based on the formulas established by Pegoretti [14], we have the following:

• Base 1
• Worked surface of the base = 1.32 m²
• Thus, 1.32 m² × 52 h/m² = 68.6 lh
• 68.6 lh/10 = 6.9 wd
• 6.9 wd × 2 HS = 13.8 HS|3.45 D

• Base 2 = 15.8 HS
• Base 3 = 18 HS
• Base 4 = 25.4 HS
• Base 5 = 27.6 HS
• Base 6 = 14.6 HS
• Base 7 = 16.4 HS
• Base 8 = 19.6 HS
• Capital 9 = 18.8 HS

• Capital 10
• 345.6 lh/10 = 34.6 wd
• 34.6 wd × 2 HS = 69.2 HS|17.3 D

Table 2. Dimensions and volumes of architectural elements.

N.º	Object	Dimensions (m)	Surface (m2)	Volume (m3)
1	Attic base	Ht. 0.37 Ø. 0.575	1.32	0.32
2	Attic base	Ht. 0.40 Ø. 0.63	1.52	0.54
3	Attic base	Ht. 0.48 Ø. 0.67	1.73	0.60
4	Attic base	Ht. 0.50 Ø. 0.865	2.45	0.93
5	Attic base	Ht. 0.50 Ø. 0.94	2.66	1
6	Attic base	Ht. 0.43 Ø. 0.60	1.41	0.36
7	Attic base	Ht. 0.45 Ø. 0.60	1.58	0.40
8	Attic base	Ht. 0.42 Ø. 0.76	1.88	0.59
9	Tuscan capital	Ht. 0.50 Ø. 0.60	1.80	0.50
10	Ionic capital	Ht. 0.46 Ø. 0.78	2.06	0.66

Table 2. Dimensions and volumes of architectural elements.

N.º	Object	Dimensions (m)	Surface (m2)	Volume (m3)
1	Attic base	Ht. 0.37 Ø. 0.575	1.32	0.32
2	Attic base	Ht. 0.40 Ø. 0.63	1.52	0.54
3	Attic base	Ht. 0.48 Ø. 0.67	1.73	0.60
4	Attic base	Ht. 0.50 Ø. 0.865	2.45	0.93
5	Attic base	Ht. 0.50 Ø. 0.94	2.66	1
6	Attic base	Ht. 0.43 Ø. 0.60	1.41	0.36
7	Attic base	Ht. 0.45 Ø. 0.60	1.58	0.40
8	Attic base	Ht. 0.42 Ø. 0.76	1.88	0.59
9	Tuscan capital	Ht. 0.50 Ø. 0.60	1.80	0.50
10	Ionic capital	Ht. 0.46 Ø. 0.78	2.06	0.66

Table 3. Summary of the costs distributed among the various stages of the object manufacturing process.

N.º	Object	Material and Extraction (HS)	Rustic Work (HS)	Semi-Elaboration (HS)	Finishing (HS)	Total (HS|D)
1	Attic base	12.32	28.68	4.6	13.8	59.4|14.9
2	Attic base	20.79	48.38	5	15.8	90|22.5
3	Attic base	23.1	53.8	5.6	18	100.5|25.1
4	Attic base	35.8	83.4	7.6	25.4	152.2|38
5	Attic base	38.5	89.6	8	27.6	163.7|40.9
6	Attic base	13.86	32.2	4.8	14.6	65.5|13.4
7	Attic base	15.4	35.8	5.4	16.4	73|18.3
8	Attic base	22.72	52.86	6	19.6	101.2|25.3
9	Tuscan capital	19.25	44.8	6	18.8	88.9|22.2
10	Ionic capital	25.41	59.2	6.6	69.2	160.4|40.2

Table 3. Summary of the costs distributed among the various stages of the object manufacturing process.

N.º	Object	Material and Extraction (HS)	Rustic Work (HS)	Semi-Elaboration (HS)	Finishing (HS)	Total (HS|D)
1	Attic base	12.32	28.68	4.6	13.8	59.4|14.9
2	Attic base	20.79	48.38	5	15.8	90|22.5
3	Attic base	23.1	53.8	5.6	18	100.5|25.1
4	Attic base	35.8	83.4	7.6	25.4	152.2|38
5	Attic base	38.5	89.6	8	27.6	163.7|40.9
6	Attic base	13.86	32.2	4.8	14.6	65.5|13.4
7	Attic base	15.4	35.8	5.4	16.4	73|18.3
8	Attic base	22.72	52.86	6	19.6	101.2|25.3
9	Tuscan capital	19.25	44.8	6	18.8	88.9|22.2
10	Ionic capital	25.41	59.2	6.6	69.2	160.4|40.2

Theoretical Proposal for the Cost of a Monumental Column

Relying on data provided by material culture, including some of the objects considered here, Rui Morais [29] suggested the existence of an octastyle temple in Braga. Based on this theoretical proposal, in collaboration with César Figueiredo, they developed an architectural proposal for the building [29], including one of the bases (Attic base n.º 5) and one of the capitals (Ionic capital n.º 10) of this ensemble, the diameters of which were compatible. The methodology followed consisted of applying the principles developed by Vitruvius [33] in his treatise on architecture, taking into account the measurements of the preserved elements. They thus proposed columns 8.85 metres high and 0.885 metres in diameter at the imoscapus, of which 7.9 metres corresponded to the shaft. Based on these data, we have extended the above exercise to a complete column specimen:

Raw Material Extraction (Expressed in ft³)

• Shaft
• 0.935 m (0.885 + 0.005) × 0.935 m × 7.9 m = 6.9 m³
• 6.9 m³ = 265.65 ft³
• 265.65 ft³ × 1 HS = 265.65 HS|66.4 D

• Rustic Work (Expressed in lh/m³, in the Quarry)
• 6.90 m³ × 448 h/m³ = 3091.2 h
• 3091.2 h/10 = 309.12 wd
• 309.12 wd × 2 HS = 618.4 HS|154.6 D

• Semi-Elaboration (Expressed in lh/m², in the Quarry)
• Shaft: 12 × (2 + ${\displaystyle \frac{0.25}{0.885}}$) = 27.39 h/m²
• Shaft surface: 2πr × shaft height = 2 × 3.1416 × 0.4425 × 7.9 = 21.96 m²
• Thus, 27.39 h/m² × 21.96 m² = 601.48 lh/shaft
• 601.58 lh/10 h = 60.16 wd
• 60.16 wd × 2 HS = 120.32HS|30 D

• Finishing

• 16 lh × 21.96 = 351.36 lh/shaft
• 351.36 lh/10 h = 35.14 wd
• 35.14 wd × 2 HS = 70.28 HS|17.58 D

Counting all of the costs, we propose a value of 1399 HS|350 D for the manufacture of one of the columns of the temple (

Table 4. Summary of the costs of the theoretical column manufacturing process.

Object/Task Cost (HS)	Raw Material Extraction (HS)	Rustic Work (HS)	Semi-Elaboration (HS)	Finishing (HS)	Total (HS|D)
Base n.º 5	38.5	89.6	8	27.6	163.7
Theoretical shaft	265.65	618.4	120.32	70.28	1074.65
Capital n.º 10	25.41	59.2	6.6	69.2	160.4
				Total	1399|350

).

5.2. The Columnatio of the Theatre

The theatre of Bracara Augusta was built at the beginning of the 2nd century, on the western edge of the upper platform of the city, where the forum was based, becoming one of the great works of the Roman period, like the amphitheatre or the public buildings of the forum, involving significant financial and human resources. It is a building with a diameter of 72.63 metres and a height of 13.10 metres (the size of the cavea), which could accommodate an average of 4.000–4.500 spectators. The archaeological work carried out since its discovery in 1999 [3] allowed us to present a proposed plan of the building with a three-level cavea, surmounted by a porticus in summa cavea, and topped with a quadriportic to the west (Figure 10 and Figure 11).

Our focus is the scaenae frons of the building, for which Ricardo Mar proposes a double columnatio of 48 elements (Figure 12). Our analysis is based on the fragments that could be recovered, at the present date, from the original set, namely, an Attic base; two Tuscan bases; two complete column shafts; three fragments of column shafts, with different diameters, certainly corresponding to each of the levels; and two fragments of capitals. Moreover, several fragments of the podium and cornices, with different dimensions, were collected. As such, our quantification of work and cost estimate proposal is supported by the analysis of the remains and elements exhumed in the excavations and the reconstitution of the columnatio.

5.2.1. Description/Catalogue (Table 5, Figure 13 and Figure 14)

Table 5. Description of the granite elements from the columnatio of the Roman theatre.

N.º	Object	Inventory Number	Type	Production Date	Dimensions (cm)	Description
1	Base	EA146	PTSTA	2nd century	Ø. 44 Ht. 30	Attic base of column: plinth, upper and lower torus separated by a scotia between two fillets. On the top, a fillet and an apophygè.
2	Base	EA172	TC		Ø. 35 Ht. 23	Tuscan base of column: no plinth, arc-shaped torus joined with the cylindrical by a kyma shaped molding.
3	Base	EA177	TC		Ø. 34 Ht. 23	Tuscan base of column: no plinth, arc-shaped torus joined with the cylindrical by a kyma shaped moulding
4	1st level shaft	n/a	n/a		Ø 1. 43 Ø 2. 37 Ht. 300	First level shaft: smooth column shaft, polished.
5	Capital	n/a	n/a		Ht. 10 Width (W). 15 (incomplete)	Corinthian capital: fragment of the upper edge of one of the faces of the capital, corresponding to the abacus and a volute.
6	Capital	n/a	n/a		Ht. 7.2 W. 12.8	Ionic capital: small fragment with several broken faces, so that it is not possible to demonstrate its connection to the shaft. Its decoration suggests an Ionic capital. On the front, a listom of eggs stands out, of which two remain, separated by a dart, above a line of pearls.
7	Podium toping element	EA194	n/a		W. 60 (broken) Depth (DP). 44 Ht. 20	Podium topping element: parallelepiped element smoothed on the back and left faces and moulded on the front and right faces, with inverted cavetto on banded moulding, reverse kyma and thin lower listel.
8	Cornice	EA181	TSTCI–3		W. 42 DP. 56 Ht. 20	Cornice fragment: element smoothed on both side and rear faces and moulded on the front face, with reverse kyma.
9	Cornice	EA184	TC		W. 39 (broken) DP. 20 (broken) Ht. 19	Cornice fragment: element fractured on the posterior face, smoothed on the posterior and decorated on the external face, with a convex moulding, a thin listel, and a concave moulding.

n/a: not applicable.

Table 5. Description of the granite elements from the columnatio of the Roman theatre.

N.º	Object	Inventory Number	Type	Production Date	Dimensions (cm)	Description
1	Base	EA146	PTSTA	2nd century	Ø. 44 Ht. 30	Attic base of column: plinth, upper and lower torus separated by a scotia between two fillets. On the top, a fillet and an apophygè.
2	Base	EA172	TC		Ø. 35 Ht. 23	Tuscan base of column: no plinth, arc-shaped torus joined with the cylindrical by a kyma shaped molding.
3	Base	EA177	TC		Ø. 34 Ht. 23	Tuscan base of column: no plinth, arc-shaped torus joined with the cylindrical by a kyma shaped moulding
4	1st level shaft	n/a	n/a		Ø 1. 43 Ø 2. 37 Ht. 300	First level shaft: smooth column shaft, polished.
5	Capital	n/a	n/a		Ht. 10 Width (W). 15 (incomplete)	Corinthian capital: fragment of the upper edge of one of the faces of the capital, corresponding to the abacus and a volute.
6	Capital	n/a	n/a		Ht. 7.2 W. 12.8	Ionic capital: small fragment with several broken faces, so that it is not possible to demonstrate its connection to the shaft. Its decoration suggests an Ionic capital. On the front, a listom of eggs stands out, of which two remain, separated by a dart, above a line of pearls.
7	Podium toping element	EA194	n/a		W. 60 (broken) Depth (DP). 44 Ht. 20	Podium topping element: parallelepiped element smoothed on the back and left faces and moulded on the front and right faces, with inverted cavetto on banded moulding, reverse kyma and thin lower listel.
8	Cornice	EA181	TSTCI–3		W. 42 DP. 56 Ht. 20	Cornice fragment: element smoothed on both side and rear faces and moulded on the front face, with reverse kyma.
9	Cornice	EA184	TC		W. 39 (broken) DP. 20 (broken) Ht. 19	Cornice fragment: element fractured on the posterior face, smoothed on the posterior and decorated on the external face, with a convex moulding, a thin listel, and a concave moulding.

n/a: not applicable.

Figure 13. Elements of the columnatio of the theatre: (1) Attic base, (2) Tuscan base, (3) Tuscan base, (4) Column shaft, (5) fragment of Corinthian capital, (6) fragment of Ionic capital.

Figure 13. Elements of the columnatio of the theatre: (1) Attic base, (2) Tuscan base, (3) Tuscan base, (4) Column shaft, (5) fragment of Corinthian capital, (6) fragment of Ionic capital.

Figure 14. Topping elements from the columnatio of the theatre: (7) Podium toping element, (8,9) Cornice.

Figure 14. Topping elements from the columnatio of the theatre: (7) Podium toping element, (8,9) Cornice.

5.2.2. Identification and Weathering Degree of the Stone

As in the previous case, this set was subjected to non-intrusive analysis. This allowed for the identification of four types of granites, with elements produced in the granites of Braga, Barcelos, Póvoa de Lanhoso, and Gonça (Figure 15).

This seems to indicate the acquisition of materials from different quarries to adequately follow the rhythms of the work or, possibly, due to aesthetic concerns. The choice of finer-grained variants, with some granites from Braga, Barcelos, or Gonça, is justified by the greater flexibility they offer in architectural work.

5.2.3. Economic Estimate, Production Process, and Cost of Elements (Table 6, Table 7 and Table 8)

Raw Material Extraction (Expressed in ft³)

The calculation of the cost of the raw material used considered the gross volume of material extracted from the quarry for the podium, bases, shafts, capitals, and entablatures, which we multiplied by the estimated price of local stone: 1 HS/ft³. This value already includes the cost of cutting and extracting the block.

With regard to the shafts, as suggested by Javier Domingo [8], we added 5 cm to the original block, corresponding to the volume of stone lowered in the semi-elaboration stage of these pieces.

The volumes of stone used in the various elements are thus as follows:

• Bases: 4.72 m³
• Shafts: 21.24 m³
• Capitals: 6.71 m³
• Upper part of the podium: 6.85 m³
• Entablatures: 15.25 m³
• = total of 54.77 m³

• 54.77 m³ = 2109 ft³
• 2109 ft³ × 1 HS = 2109 HS|527 D

• Rustic Work (Expressed in lh/m³, in the Quarry)

The rustic work consists of assigning the desired size to the block extracted from the quarry. In this calculation, we considered the variable of 448 h/m³ defined by Pegoretti [14] for common granite.

• Upper part of the podium:

• 6.85 m³ × 448 h/m³ = 3068.8 lh
• 3068.8 lh/10 h = 307 wd
• 307 wd × 2 HS = 614 HS/153.5 D
• Columns:
• 32.67 m³ × 448 h/m³ = 14,636.2 lh
• 14,636.2 lh/10 h = 1463.6 wd
• 1463.6 wd × 2 HS = 2927 HS/731.8 D
• Entablature
• 15.25 m³ × 448 h/m³ = 6832 lh
• 6832 lh/10 h = 683.2 wd
• 683.2 wd × 2 HS = 1366 HS/341.6 D
• Total
• 54.77 m³ × 448 h/m³ = 24,537 lh
• 24,537 lh/10 h = 2454 wd
• 2454 wd × 2 HS = 4908 HS|1227 D

Semi-Elaboration (Expressed in lh/m², in the Quarry)

After the initial roughing, a rough shape is assigned to the element, based on the architectural function: base, shaft, or capital. It is considered, at this stage, the total surface worked. The work carried out on the bases/capitals is presented separately from that on the shafts.

For shafts, the estimate of labour hours was calculated based on the formula [a × (2 + 0.25/x)] [14], where a is a variable that for common granites corresponds to 12, while x is associated with the diameter of the imoscapus.

• 1st level shafts: 12 × (2 + (0.25)/0.43) = 30.97 h/m²
• Shaft surface: 2πr × shaft height = 2 × 3.1416 × 0.215 × 3 = 4.05 m²
• Thus, 30.97 h/m² × 4.05 m² = 125.43 lh/shaft
• 125.43 lh × 24 = 3010.32 lh
• = 3010.32 lh/10 h = 301.03 wd

• 301.03 wd × 2 HS = 602.06HS|150.5 D
• 2nd level shafts: 12 × (2 + 0.25/0.24) = 36.5 h/m²
• Shaft surface: 2 × 3.1416 × 0.12 × 2.14 = 1.61 m²
• Thus, 36.5 h/m² × 1.61 m² = 58.76 lh/shaft
• 58.76 lh × 24 = 1410.24 lh
• 1410.24 lh/10 h = 141.02 wd
• 141.02 wd × 2 HS = 282.04 HS|70.5 D

For the bases and capitals, we applied the formula [a(1 + 0.25/x)] [14], in which the variable a corresponds to 12 and x to the capital/base diameter.

• 1st level capitals: 12 × (1 + 0.25/0.40) = 19.5 h/m²
• Worked surface of the capital: 0.89 m²
• Thus, 19.5 h/m² × 0.89 m² = 17.36 lh/capital
• 17.36 lh × 24 = 416.64 lh
• 416.64 lh/10 h = 41.66 wd
• 41.66 wd × 2 HS = 83.33HS|20.8 D

• 2nd level capitals: 12 × (1 + 0.25/0.24) = 24.5 h/m²
• Worked surface of the capital: 0.46 m²
• Thus, 24.5 h/m² × 0.46 m² = 11.27 lh/capital
• 11.27 lh × 24 = 270.48 lh
• 270.48 lh/10 h = 27.05 wd
• 27.05 wd × 2 HS = 54.1HS|13.5 D

• 1st level bases: 12 × (1 + 0.25/0.44) = 18.82 h/m²
• Worked surface of the base: 0.70 m²
• Thus, 18.82 h/m² × 0.70 = 13.17 lh/base
• 13.17 h × 24 = 316.08 lh
• 316.08 lh/10 = 31.61 wd
• 31.61 wd × 2HS = 63.22 HS|15.8 D

• 2nd level bases: 12 × (1 + 0.25/0.37) = 20.11 h/m²
• Worked surface of the base: 0.54 m²
• Thus, 20.11 h/m² × 0.54 = 10.86 lh/base
• 10.86 h × 24 = 260.64 lh
• 260.64 lh/10 = 20.06 wd
• 20.06 wd × 2 HS = 40.12 HS|10 D

• Total: 1124.4 HS|281.1 D

• Finishing

As explained in the previous case study, Pegoretti [14] does not provide a numerical value for the finishing stage of a Corinthian granite specimen. So, once again, we resorted to a variable associated with a stone with values very close to common granite: black Portoro marble, i.e., 1.296 h for specimens about 0.50 m high, the case of the capitals of the first level (with 0.40 m of height restituted), and 864 h for specimens 0.33 m high, which corresponds roughly to the case of the capitals of the second level (0.31 m of height restituted). Still on this subject, Begoña Soler Huertas [12], in her study on the cost of marbling the scaenae frons of the theatre in Cartagena, believes that the figures provided by Pegoretti [14] refer to common workshops, the quality of which would not be at all comparable to the workforce responsible for finishing the capitals of that building, which led her to double the value of the variable stated by Pegoretti [14]. In turn, Javier Domingo, in his study of the Volubilis Capitol, decorated with large Corinthian limestone capitals, reduced the variable established by Pegoretti [14] to 1/4. In our specific case, although we are dealing with only two small fragments of capitals, we decided to value the level of specialisation of the workshop or workshops that made them, maintaining the values determined by Pegoretti [14] due to the specific characteristics of granite work.

• So:
• 1st level capitals: 1296 h × 24 = 31,104 h
• 31,104 h/10 h = 3110 wd
• 3110 wd × 2 HS = 6220 HS|1555 D

• 2nd level capitals: 864 h × 24 = 20,736 h
• 20,736 h/10 h = 2073.6 wd
• 2073.6 wd × 2 HS = 4147.2 HS|1036.8 D

• For the finishing of the bases, it is considered that it could be estimated at 1/3 of the value of the capitals:
• 1st level bases: 1555/3 = 518.4 D
• 2nd level bases: 1036.8/3 = 345.6 D

With respect to the shafts, those of the theatre are smooth. In these cases, Pegoretti [14] indicates that finishing a flat-surfaced element in granite required 16 h of labour per m². The surface areas of the shafts on the first (4.05 m²) and second (1.61 m²) levels were then calculated.

• Thus:
• 1st level shafts: 16 lh × 4.05 = 64.8 lh/shaft
• 64.8 lh × 24 = 1555.2 lh
• 1555.2 lh/10 = 155.5 wd
• 155.5 wd × 2 HS = 311 HS|77.8 D

• 2nd level shafts: 16 lh × 1.61 m² = 25.76 lh/shaft
• 25.76 lh × 24 = 618.24 lh
• 618.24 lh/10 = 61.8 wd
• 61.8 wd × 2 HS = 123.6 HS|30.9 D

The elements of entablature that were collected in the excavations of the theatre have a simple decoration. Pegoretti [14] points out that the realisation of simple decorative motifs on a block of granite stone takes 355.20 h/m². The total surface areas of the elements that would cover the podium were calculated at 5 m² (27.52 m × 0.20 m), the entablature of the first level at 18 m² (24.32 m × 0.74 m) and, finally, the entablature of the second level at 10.12 m² (18.4 m × 0.55 m).

• Thus:
• Caping elements of the podium: 5.5 m² × 355.20 h = 1953.6 lh
• 1953.6 lh/10 = 195.4 wd
• 195.4 lh × 2 HS = 390.8 HS|97.7 D

• 1st level entablature: 18 m² × 355.20 h = 6393.6 lh
• 6393.6 lh/10 = 639.4 wd
• 639.4 wd × 2 HS = 1278.8 HS|319.7 D

• 2nd level entablature: 10.12 m² × 355.20 h = 3594.6 lh
• 3594.6 lh/10 = 359.5 wd
• 359.5 wd × 2 HS = 719 HS|179.8 D

• Total: 16,646.8 HS|4161.7 D

Table 6. Dimensions and volumes of the architectonic elements.

N.º	Object	Dimensions (m)	Surface (m2)	Volume (m3)
1	Attic base	Ht. 0.30 Ø. 0.44	0.53	0.06
2	Tuscan base	Ht. 0.23 Ø. 0.35	0.32	0.03
3	Tuscan base	Ht. 0.23 Ø. 0.34	0.31	0.03
4	1st level shaft	Ht. 3 Ø. 0.37/0.43	4.05	0.69
5	Corinthian capital	-	-	-
6	Ionic capital	-	-	-
7	Caping element of the podium	Ht. 0.20 DP. 0.44 W. 0.60 *	0.12	0.05
8	Cornice	Ht. 0.20 DP. 0.45 W. 0.44	0.09	0.04
9	Cornice	Ht. 0.19 DP. 0.20 * W. 0.39 *	0.07	0.01

* Element broken.

Table 6. Dimensions and volumes of the architectonic elements.

N.º	Object	Dimensions (m)	Surface (m2)	Volume (m3)
1	Attic base	Ht. 0.30 Ø. 0.44	0.53	0.06
2	Tuscan base	Ht. 0.23 Ø. 0.35	0.32	0.03
3	Tuscan base	Ht. 0.23 Ø. 0.34	0.31	0.03
4	1st level shaft	Ht. 3 Ø. 0.37/0.43	4.05	0.69
5	Corinthian capital	-	-	-
6	Ionic capital	-	-	-
7	Caping element of the podium	Ht. 0.20 DP. 0.44 W. 0.60 *	0.12	0.05
8	Cornice	Ht. 0.20 DP. 0.45 W. 0.44	0.09	0.04
9	Cornice	Ht. 0.19 DP. 0.20 * W. 0.39 *	0.07	0.01

* Element broken.

Table 7. Approximate volumes of stone used.

Objects	Gross Volume (m3)
Caping element of the podium	6.85
1st level bases	3.02
2nd level bases	1.7
1st level shafts	16.6
2nd level shafts	4.64
1st level capitals	5.18
2nd level capitals	1.53
1st level entablature	9.73
2nd level entablature	5.52
Total	54.77

Table 7. Approximate volumes of stone used.

Objects	Gross Volume (m3)
Caping element of the podium	6.85
1st level bases	3.02
2nd level bases	1.7
1st level shafts	16.6
2nd level shafts	4.64
1st level capitals	5.18
2nd level capitals	1.53
1st level entablature	9.73
2nd level entablature	5.52
Total	54.77

Table 8. Summary of the costs distributed among the various stages of the objects manufacturing process.

Object	Material and Extraction (HS)	Rustic Work (HS)	Semi-Elaboration (HS)	Finishing (HS)	Total (HS|D)
Caping element of the podium	263.6	614	-	390.8	1271|317.7
1st level bases	116	270.6	63.2	2073.6	2523|630.85
2nd level bases	65.44	152	40	1382.4	1640|409.96
1st level shafts	640	1487.36	602	311.2	3041|760.14
2nd level shafts	178.8	415.36	282	123.6	1000|249.94
1st level capitals	199.4	464.12	83.2	6220	6967|1741.68
2nd level capitals	58.92	137.2	54	4147.2	4397|1099.33
1st level entablature	374.6	871.8	-	1278.8	2525|631.3
2nd level entablature	212.52	494.6	-	719.2	1426|356.58
Total	2109	4908	1124.4	16.6468	24.788|6197

Table 8. Summary of the costs distributed among the various stages of the objects manufacturing process.

Object	Material and Extraction (HS)	Rustic Work (HS)	Semi-Elaboration (HS)	Finishing (HS)	Total (HS|D)
Caping element of the podium	263.6	614	-	390.8	1271|317.7
1st level bases	116	270.6	63.2	2073.6	2523|630.85
2nd level bases	65.44	152	40	1382.4	1640|409.96
1st level shafts	640	1487.36	602	311.2	3041|760.14
2nd level shafts	178.8	415.36	282	123.6	1000|249.94
1st level capitals	199.4	464.12	83.2	6220	6967|1741.68
2nd level capitals	58.92	137.2	54	4147.2	4397|1099.33
1st level entablature	374.6	871.8	-	1278.8	2525|631.3
2nd level entablature	212.52	494.6	-	719.2	1426|356.58
Total	2109	4908	1124.4	16.6468	24.788|6197

6. Discussion and Conclusions

The calculation methodology is a tool that aims to approximate the economic cost of historic buildings. With its application, we obtained a cost of 6197 D for the production of the stone elements of the theatre columnatio and the following unit costs for the monumental architectonic elements collected near the city forum: bases between 15 and 41 D, 22 D for Tuscan capitals, and 40 D for the Ionic model. In an extension of the exercise to a theoretical model elaborated by Rui Morais [29], passing through the restitution of one of the colonnades, we proposed a cost of 350 D for one specific column, based on two compatible elements (attic base n.º 5 and ionic capital n.º 10).

The values obtained are intended to be an estimate of the real costs, are based solely on archaeological evidence, and cannot be proven or refuted, so far, through other sources, such as epigraphic or historical ones. In addition, in this type of approach, there are variables that are impossible to assess, so we hope that future excavations of the buildings under analysis, as well as other works of this nature, may optimise or even correct the results obtained.

It should also be noted that these are partial proposals, referring only to part of the construction process of the original buildings, corresponding only to a portion of the total investment, possibly one of the most important, taking into account the weight of the costs with ornamentation, as Ricardo Mar and Patrizio Pensabene [10] observed in the construction of the provincial forum of Tarraco.

In the first case, we focused our analysis only on the monumental architectonic elements that would integrate the colonnades of several buildings, possibly located in the city forum. In the second, our exercise was based on a set of architectonic elements collected in the excavations of the theatre and the restitution proposal of the frons scaenae created by Ricardo Mar.

In the calculation process, we do not consider the costs for the transport of the elements and their placement.

These values can, however, be compared with information gathered from epigraphy or ancient texts corresponding to other places in the Roman world, bearing in mind that, as Javier Domingo [16] has documented, there were different realities in the provinces of the Empire in terms of wages and products.

Starting with colonnades, within the Late Republican/Early Empire interval, Cicero indicates that a 30-foot column in the atrium of a villa would have cost 5000 D [8,12], a value quite far from our proposal. As values closer to ours, we have the example of local limestone columns, 20 feet high, from the agora of Heracleopolis (Ehnasiyyah el-Médineh, Egypt), referenced in a 2nd century papyrus with a cost of 66 D per unit [8,12].

In the case of the theatre, the cost of 24.788 HS/6197 D can be compared with other known data for theatrical architecture. Epigraphy also provides us with some elements of comparison throughout the Empire, the values of which, when necessary, we converted to the reality of the 1st/2nd centuries, in brackets. Thus, at the end of the 1st century, the construction of a cuneus of the imacavea of the theatre of Gerasa (Jerash, Jordan), with a capacity like that of Braga, would have cost 12.000 HS [34]. An inscription from the middle of the 2nd century indicates that the marbling of the scaenae frons of the theatre of Leptis Magna (Libya) would have cost 500.000 HS [12]. Another inscription, dated from 198, reveals that the construction of a proscaenium in the theatre of Beda (Bayda, Libya) cost 50.000 HS (⇔ 1111 HS 1st/2nd century) [34]. The embellishment of the theatre of Narbo (Narbonne, France), at the end of the 2nd century, would have benefited from a donation of 53.000 HS (⇔ 1178 HS 1st/2nd century) [12,34]. There is also a reference to a donation for the construction of the theatre of Calama (Guelma, Algeria), at the end of the 2nd/beginning of the 3rd century, in the amount of 400.000 HS (⇔9000 HS 1st/2nd centuries) [34]. Under the Severi (193–235 AD), the construction of the theatre of Madauros (Mdaurush, Algeria), with a diameter half that of Bracara Augusta, would have cost 375.000 HS (⇔ 8333 HS 1st/2nd century) [34]. We can also compare the data obtained with values estimated elsewhere using the calculation methodology. For example, Begoña Soler [12] estimated the cost of marbling the scaenae frons of the theatre of Cartagena at 80.952 HS. The differences may be explained using a proportionally higher volume of raw stone and also by the cost of importing Luna marble by sea. Paolo Barresi, according to Domingo [9], calculated the cost of the portico in summa cavea of the Flavian amphitheatre of Pozzuoli, formed by 72 columns in Proconesco marble, at 79.992 HS. Ricardo Mar and Patrizio Pensabene [10] estimated the cost of building the upper platform of the Provincial Forum of Tarraco (Tarragona, Spain) at 2.162.580.58 HS, and the same authors, together with Rosario Cebrián [11], calculated the cost of the Augustan forum of Segobriga (Saelices, Spain) at 243.464.40 HS.

The above values can be contextualised by comparing them with some of the wages/salaries at the time. Thus, the salary received by a legionary in Augustus’ time was 225 D per year and under Domitian (81–96 AD), it was 300 D per year [9]. As for salaries, in addition to the daily amounts mentioned in the Edict [13], some values are known from the lex Ursonensis for this city in Domitian’s time, namely for the scribae of the duumvirs, who received 1200 HS; the lictors, with 600 HS; or the viatores, with 400 HS. According to the Edict, and with due extrapolation, a construction worker would earn a maximum of 730 HS, plus maintenance, working every day of the year [16].

This exercise demonstrates some of the possibilities that the application of the calculation methodology can offer to the study of the city. It allows us to bring together economic and social aspects, including economic effort and the labour force; to know the people involved in the construction process, from the commissioners, the financiers, to the workers; and also to try to understand the motivations that led to these constructions.

However, the application of the calculation methodology should be understood as a guidance tool. The results obtained are indicative and should be adjusted with new archaeological work, new developments in construction economics, and with comparative analyses with studies of the same nature.

Beyond purely economic issues, other important aspects stand out, related to the choice of local granite, the transport of materials, although not directly addressed here, the quality of the workshops and stone workers, the source of funding, and the architectural contextualisation of the monumental elements.

In fact, we can question the choice of local granite for the city’s decorative programme, linked to the exploitation of the Braga granite, in parallel with other extraction areas, some of which operated simultaneously, allowing them not to be overloaded. Certainly, the logic of this decision is related to an economy of expenses. The cheapest marble would be associated with a cost four times higher than that of local stone. In addition, long-distance transport would substantially increase the cost of the works.

Still, considering the raw material, we can see that the granitic rock used in the manufacture of the shafts of the theatre’s columnatio comes from the Póvoa de Lanhoso/Donim area, displaying a practically polished finish, almost imitating marble. This option, although completely different from the marble programmes applied in many theatres in Hispania, did not prevent the use of good quality stone and a varied range of granites, offering aesthetically pleasing colour combinations. In addition to the above, the existence of different granites in the same building complexes shows that several quarries were exploited simultaneously to meet the requirements of the urban programme.

The transport of the blocks that generated some of the elements also deserves our attention, in particular because of their probable size and weight. The blocks destined for the large column bases, some of which represent an initial volume of 1 m³, i.e., around 2700 kg, as well as the shafts of the first level of the theatre’s scaenae frons, with each shaft, monolithic, weighing around 1300 kg, must have posed some difficulties for the transporters, first because of the need for resources and their routing to the construction site. On this issue, it is known that the number of draught animals needed depended on the conditions of transport, the terrain, slope, and distance, among others [9]. As for the maximum load that a cart could carry, it is estimated that a pair of oxen could pull a maximum weight of 2.7 tonnes, although some authors propose different values [9]. Of course, this value also varied according to the number of oxen attached to the cart. In the case of the theatre shafts, each ox cart could thus transport two elements per day, a priori with more than one ox. More recently, we have the example of the transport of the monolithic shafts of the frontispiece of the church of Bom Jesus in Braga, extracted from the Donim area, about 15 kilometres away, which required the use of 24 oxen carts, with obvious difficulties on their way to the construction site [35].

The choice of locally available stone is easily understood, although granite is more demanding in terms of labour. The quality of some pieces reveals the early presence of workshops in the city and a labour force displaying a refined technical quality. We realise from these two case studies that the builders of Braga were able, using locally available material, to respond to the great technical and aesthetic demands and standards of that time. Hence, the technical quality of these stone workers, capable of transforming into granite the models designed for easier-to-work rock, such as marble or limestone, should be valued, revealing the great assimilation of Roman building forms but clearly incorporating ancestral local knowledge, inherited from the builders of the Iron Age hill-top forts.

The results of this exercise are a contribution to the knowledge of the costs of monumentalising the city and lead to a reflection on the origin of its financing. The total cost of constructing the city’s forum and associated buildings, as well as the large public entertainment buildings of which the theatre is an example, which we can glimpse through our exercise, must have involved a great financial effort. Between the 1st and 2nd centuries, a monumental forum, a theatre, together with an attached public bath, and an amphitheatre were built in the city [23,36]. Since there is no evidence of imperial funding so far, we can assume that these works were financed entirely by the municipal elites and important families whose enrichment resulted from farming, trade, or craft, as evidenced in Braga by the case of the Lucretii, a successful family of potters who managed to assert themselves socially, politically, and possibly religiously through trade [37,38] (Figure 16), or even in the construction of the theatre of Cartagena, for example [12]. These large public ensembles, with their very own identities, must have had a great impact on the population, conveying a political and propagandistic message of imperial power, and are associated with the growing power of local elites, seeking thereby to reveal their conformity with Rome and obtain the benefits of office or career. In the analysed granite elements of Braga in the present research work, there were observed rare signals of weathering, and petrographic and geochemical analyses should be performed to confirm if it is Braga granite.

A final reflection arises from the study of the characteristics of monumental bases, which resulted in the observation of similarities in some of the pieces studied, and were therefore grouped into types, with a total of eight variants. The connections that can be made, namely through the diameters, are an added value in trying to individualise different buildings.

As we have seen, this approach clearly allows an approximation of costs, but above all, it offers the possibility of increasing our knowledge of past societies.