The Architectural Terracotta marks of Bracara Augusta (Braga, Portugal): a first typology classification
Abstract
Architectural Terracotta (ATC) is one of the most common materials in excavations from the Roman period. These ceramic building materials are an essential component of construction. Some of these pieces show potter´s marks, of different categories, that allow access to the production world of these materials. This investigation is a first typological classification of the 1216 marks from ATC materials, collected from 41 archaeological sites in <i>Bracara Augusta</i> (Braga, Portugal). Most of the marks were collected from the <i>domus</i> of Carvalheiras, one of the most emblematic archaeological sites of the city, currently under a musealization process. With this work it was possible to correlate the studied marks with specific terracotta types (shapes), context distribution and associated chronologies. The results suggested an organized and dynamic production, and an open-market, supported by numerous <i>officinae</i>, certainly of different sizes. Some of them were located near the housing area and reveal the presence of a large number of workers, including women and children. Further approaches on mineralogical, chemical and technological characterization of ATC, linked with stratigraphy, are under development.
Full text
heritage Article The Architectural Terracotta Marks of Bracara Augusta (Braga, Portugal): A First Typology Classification Jorge Ribeiro 1,* , Filipe Antunes 2,† and Ana Fragata 3 Citation: Ribeiro, J.; Antunes, F.; Fragata, A. The Architectural Terracotta Marks of Bracara Augusta (Braga, Portugal): A First Typology Classification. Heritage 2021,4, 4126–4147. https://doi.org/10.3390/ heritage4040227 Academic Editor: Ákos Török Received: 31 August 2021 Accepted: 23 October 2021 Published: 31 October 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1Lab2PT/IN2PAST, Institute of Social Sciences, University of Minho, Campus de Gualtar, 4710-057 Braga, Portugal 2Museum of Archaeology D. Diogo de Sousa, Rua dos Bombeiros Voluntários, 4700-025 Braga, Portugal 3GeoBioTec, Geosciences Department, University of Aveiro, Campus de Santiago, 3810-193 Aveiro, Portugal; [email protected] *Correspondence: jribeir[email protected] † This author had passed away. Abstract: Architectural Terracotta (ATC) is one of the most common materials in excavations from the Roman period. These ceramic building materials are an essential component of construction. Some of these pieces show potter ´ s marks, of different categories, that allow access to the production world of these materials. This investigation is a first typological classification of the 1216 marks from ATC materials, collected from 41 archaeological sites in Bracara Augusta (Braga, Portugal). Most of the marks were collected from the domus of Carvalheiras, one of the most emblematic archaeological sites of the city, currently under a musealization process. With this work it was possible to correlate the studied marks with specific terracotta types (shapes), context distribution and associated chronologies. The results suggested an organized and dynamic production, and an open-market, supported by numerous officinae, certainly of different sizes. Some of them were located near the housing area and reveal the presence of a large number of workers, including women and children. Further approaches on mineralogical, chemical and technological characterization of ATC, linked with stratigraphy, are under development. Keywords: Hispania; Roman Western provinces; Bracara Augusta; tilery; tile; brick; ATC; marks 1. Introduction French archaeologists that study the Roman and Medieval periods use the term “ATC: Architectural Terracotta (ATC)”, to designate roof tiles (tegulae and imbrices), bricks, pipes, circular, semicircular and quarter-circle bricks, and other elements associated with the building of thermal spaces [ 1 ]. These ceramic building materials, such as tiles and bricks (Figure 1), had an important role in the construction of Roman cities, being used in multiple contexts, particularly in roofs and thermal buildings. However, unlike other types of ceramics, they have been unexplored in academic works. Bracara Augusta (Braga, Portugal) was an economically and politically important city in the Iberian Peninsula, founded by the emperor Augustus in the late first century BC. (Figure 2). It was the capital of the conventus bracaraugustanus and, in the late third century AD, became the capital of the new province of Gallaecia, under emperor Diocletian. During the fifth and the sixth centuries the city was the capital of the Suebi Kingdom, which was absorbed and annexed by the Visigoth Kingdom in AD 585. From then on, and until the end of the seventh century, the city lost some political importance, but not religious, with the leading role being driven by the Church [2–5]. Heritage 2021,4, 4126–4147. https://doi.org/10.3390/heritage4040227 https://www.mdpi.com/journal/heritage
Heritage 2021,44127 Heritage2021,4 2 Figure1.BrickandtiletypologybyBrodribb. BracaraAugusta(Braga,Portugal)wasaneconomicallyandpoliticallyimportantcity intheIberianPeninsula,foundedbytheemperorAugustusinthelatefirstcenturyBC. (Figure2).Itwasthecapitaloftheconventusbracaraugustanusand,inthelatethirdcen‐ turyAD,becamethecapitalofthenewprovinceofGallaecia,underemperorDiocletian. DuringthefifthandthesixthcenturiesthecitywasthecapitaloftheSuebiKingdom, whichwasabsorbedandannexedbytheVisigothKingdominAD585.Fromthenon, anduntiltheendoftheseventhcentury,thecitylostsomepoliticalimportance,butnot religious,withtheleadingrolebeingdrivenbytheChurch[2–5]. Figure 1. Brick and tile typology by Brodribb.
Heritage 2021,44128 Heritage2021,4 3 Figure2.LocationofBracaraAugustainHispaniaandplanofthecity/low‐imperial(aroundthecity grid)andmedievalwalls,withthecastletothenortheast(ModifiedfromMagalhães[6]) (©UAUM). TheproductionofceramicbuildingmaterialsinBracaraAugustaseeminglyhadan importantsocialandeconomicimpact.Someofthesepiecesshowedmarksthat,ifstud‐ iedinmoredetail,couldprovideabetterunderstandingoftheactorsinvolvedinATC productionandsomeaspectsrelatedtotheirorganization.InBraga,threetypesofmarks weregloballyidentified:stamps,finger‐mademarksandgraffiti(alphabeticalandnu‐ meralmarks).Thisstudypresentsafirsttypologicalclassificationofthesemarks,taking intoaccountarchaeologicalandhistoricalevidences.Thus,1216markswerecatalogued, from41archaeologicalsitesofthecityandfrombuildingsthathaddifferentfunctionali‐ ties,fromthefoundationofthecityinthefirstcenturyBC,untiltheseventhcenturyAD. Inthisarticle,weconsideredothersimilarstudiescarriedoutinothercontextsofthe Romanworld,namelyinFrance[1,7],wherevariousinvestigationsfocusingonthedif‐ ferenttypeofmarksunderanalysisprovideelementsofcomparisonfortheWestern RomanWorld.Weapproachedseveralquestionsrelatedtothevariousmarks,especially regardingtheirfunction,positionandorientation,inordertoidentifydifferentofficinae andtheirroleintheorganizationofproductionandthetechniquesusedinthemanu‐ factureofmaterials.AfirstreportonthestudyofthesemarkswascarriedoutbyFilipe Antunes,aformertechnicianattheMuseumofArchaeologyD.DiogodeSousa(MDDS), inBraga.Allmarkswereanalysedandclassified,fromhistorical,criticalandtechnolog‐ icalperspectives. 2.TheLaterarii´sMarks,TheirInterestandStudy ATCmaterialsoccurinlargequantitiesinthearchaeologicalrecordandourap‐ proachresultsfromthevaluableinformationthatcanbeextractedfromthismaterial, relatedtotheirgreattypologicalvariabilityandmarks. Inpreviousstudies,GoulpeauandLeNy[7]investigatedthefingerprintmarkson Gallo‐Romanconstructionbakedclaymaterialsandfoundthatalmostalltypesofele‐ mentsfromarchaeologicalcontextshavemarksintentionallyplacedbythelaterarii, perhapswiththeexceptionoftubulilaterici.Therearefourmajorfamiliesofmarks[8]. Finger‐mademarks,calledsignaturesinEnglisharchaeologicalstudies[7],areusually themostused,correspondingtosimpleormorecomplexlines,drawnonthesurfaceof thepiece.Graffiti,showthegreatestvariability,fromsimpleletters,combinationsoflet‐ ters(representinginitialsofnames),completesentencesandalsonumeralsymbols. Stamps(signacula),averyraretypeofmark,aremadewithamatrixinmetal,ceramicor wood,withthenameofthepotterorowner[9];andincisions,aslinesrepresentingRo‐ mannumbers,areusuallymadeononeofthesidesofthepieces. Figure 2. Location of Bracara Augusta in Hispania and plan of the city/low-imperial (around the city grid) and medieval walls, with the castle to the northeast (Modified from Magalhães [ 6 ]) ( © UAUM). The production of ceramic building materials in Bracara Augusta seemingly had an important social and economic impact. Some of these pieces showed marks that, if studied in more detail, could provide a better understanding of the actors involved in ATC production and some aspects related to their organization. In Braga, three types of marks were globally identified: stamps, finger-made marks and graffiti (alphabetical and numeral marks). This study presents a first typological classification of these marks, taking into account archaeological and historical evidences. Thus, 1216 marks were catalogued, from 41 archaeological sites of the city and from buildings that had different functionalities, from the foundation of the city in the first century BC, until the seventh century AD. In this article, we considered other similar studies carried out in other contexts of the Roman world, namely in France [ 1 , 7 ], where various investigations focusing on the different type of marks under analysis provide elements of comparison for the Western Roman World. We approached several questions related to the various marks, especially regarding their function, position and orientation, in order to identify different officinae and their role in the organization of production and the techniques used in the manufacture of materials. A first report on the study of these marks was carried out by Filipe Antunes, a former technician at the Museum of Archaeology D. Diogo de Sousa (MDDS), in Braga. All marks were analysed and classified, from historical, critical and technological perspectives. 2. The Laterarii´s Marks, Their Interest and Study ATC materials occur in large quantities in the archaeological record and our approach results from the valuable information that can be extracted from this material, related to their great typological variability and marks. In previous studies, Goulpeau and Le Ny [ 7 ] investigated the fingerprint marks on Gallo-Roman construction baked clay materials and found that almost all types of elements from archaeological contexts have marks intentionally placed by the laterarii, perhaps with the exception of tubuli laterici. There are four major families of marks [ 8 ]. Finger-made marks, called signatures in English archaeological studies [ 7 ], are usually the most used, corresponding to simple or more complex lines, drawn on the surface of the piece. Graffiti, show the greatest variability, from simple letters, combinations of letters (representing initials of names), complete sentences and also numeral symbols. Stamps (signacula), a very rare type of mark, are made with a matrix in metal, ceramic or wood, with the name of the potter or owner [ 9 ]; and incisions, as lines representing Roman numbers, are usually made on one of the sides of the pieces. These marks can have different meanings: the signature and graffiti can be related to the officinae, the stamps with the distribution of the materials, and incisions can be related to counting pieces. As stated by Charlier, this was the only way to explain that some pieces
Heritage 2021,44129 had three different types of marks, as was the case of a tegula found in Switzerland with a finger-made mark, a stamp and an incision [8]. The main objective of the study of marks present in ATC was to investigate the laterarii society in terms of manufacturing processes, work organization, such as the number of workers or teams of workers involved, the number of officinae, or even the circulation of production. Additionally, important information on the potter’s manufacturing techniques, age or gender, could be based on the size of the fingers, and accidental marks could be associated with women and child labor. The study of signatures has benefited from numerous contributions, particularly in the work of Swiss and French archaeologists in the first half of the 20th century [ 10 – 12 ]. Only 20 years later, in 1973, new developments emerged, with Neumann’s publication on Vindobona building materials [ 13 ], and above all Brodribb’s intense work on Beauport Park (GB) materials [ 14 ], updated in 1987 [ 15 ]. Two years later, in 1989, an article dedicated to finger-made marks and incisions in Roman Gaul was published [ 7 ]. For the issue of graffiti, the contribution of Charlier is significant, who in 1999, investigated the marks in the context of socio-legal conditions of the officinae work [ 8 ] (work in which the author addressed the various types of marks), as well as on the practice of writing in the tiled areas of Roman Gaul [ 16 ]. Stamps are the most studied type of marks. Among the studies concerning the reality of the different Roman provinces, we can mention the publication by De Poorter and Claeys [ 17 ], on the marks from Belgium, and the work of Rico dedicated to Betica production sites, their surroundings, and stamps [ 18 ]. The Ferdière investigation [ 1 ] was dedicated to the potteries in Lugdunensis and Aquitania, where the author elaborated a reflection on this type of mark. In the context of the South Mediterranean Provinces, we should give a special mention to the work of Mills, on the trade of ceramic building materials from Carthage and Beirut [19]. Concerning Hispania and, in particular, the territory that became Portugal, this topic was relatively unexplored [ 20 – 25 ]. In the specific case of Braga, it is important to stress the work of Morais [ 9 ] on the marks and graffiti identified in the city, which, although of a more generic nature, provided data for several publications under the scope of ceramic building materials [ 26 , 27 ]. From the same territory, the Gallaecia, it is also important to point out the work on the ceramic construction material of the Roman villa of Toralla [ 28 ], the marks on the building material of the Roman camp of Ciadella [ 29 ], and additional synthesis work on the building ceramics of this province [30]. 3. Marks Meaning, Criteria and Marking Frequency The main idea underlying this study was to approach the evidence of a relationship between marks and economic and social aspects. The researchers that have been investigating this theme are quite unanimous in considering that, although it is a field to be explored, the function(s) of these marks will certainly be related to the internal organization of the officinae [ 1 , 7 ], and thus, are a way to better understand the potteries’ workers, mainly potters, molders and owners of the workshops. It should be noted that the marks reveal little or no information about some workers, namely those linked to minor tasks. One of the most consensual interpretations on the meanings of the marks—which could be composed of simple signs, a single letter, or several—may be associated to the day laborer, with implications on the daily production of each worker, and their payment or not. Therefore, the smaller officinae did not need many marks, as each individual mark would be associated with a worker. In this regard, a very interesting aspect could be found in the Edict of Diocletian, which mentions the payment according to a quantity of materials produced and to the task they did. Several investigations, some more extensive than others, lead some authors such as Goulpeau and Le Ny to adopt a more cautious approach, stating that “there will not be a fixed practice, corresponding to the use of these marks for each workshop according to its own functioning internal logic” [7].
Heritage 2021,44130 Regardless of the issues mentioned above, the analysis of these specific marks in some ceramic building materials suggest that the two main marking criteria seemed to be the speed, by using mainly simple lines, and easiness, through the use of curvilinear and rectilinear motifs [ 7 ]. The morphology of some marks does not seem to meet these standards, by using a combination of various types which may disclose the originality of some workers. The finger-made marks were the first marks to be placed on the pieces, still on the worktable, right after the molding process and while the clay was still in a fresh state [ 8 ]. They were usually done with the fingers, or with the branch of a bush. In the different studied contexts, there was the use of single fingers from both right and left hands. The marks would, by rule, be created on the face of the piece facing the molder, in its working position. On the other hand, stamps were applied at a later stage in the process, when the clay was still drying and showing a consistency neither too soft nor too hard. Previous studies disclosed that the frequency of the marks varied according to the type of material and the size of the production [ 7 ]. Thus, the production of tegulae, in part because of its multiple uses, has always been more important than the production of imbrices. Another paradigmatic example were the pilae from hipocaustum, which normally consisted of a set of 11 lateres. On the other hand, circular bricks, compared to the previous ones, were produced in smaller quantities. Additionally, it seemed that not all elements were marked. 4. Sample This sampling resulted from the materials collected from the archaeological excavations that have been developed in Braga since the 1970s, in the context of the rescue project of Bracara Augusta, by the Archeology Unit of the University of Minho (UAUM) [ 31 ]. It should be noted that the marks considered in this study come from the places of use and not from manufacturing contexts, even though four kilns have been documented in or around the city. The only example of a fully excavated kiln, the activity of which was exclusively for firing tegulae, was identified in 2009, during the construction works of the new Hospital of Braga (excavation under the responsibility of archaeologist David Mendes). Moreover, other officinae were documented, including those located in the archaeological area of Casa do Poço [ 32 , 33 ], and the kilns discovered in Rua dos Falcões (numbers 8–10—Irmandade de Sta. Cruz), an intervention by the archaeologist Armandino Cunha from the Archeology Office of Braga City Council (unpublished). In addition, remains of other roman kilns were also found in the Avenida da Imaculada Conceição (Oficinas da Livraria Cruz), an intervention directed by UAUM (unpublished), and in Rua Santos da Cunha, this one identified in the context of the opening of the street in 1955 [34–36] (Figure 3). The dataset gathered from these sites was very significant and diversified, amounting to 1216 marks from more than 40 contexts/archaeological zones (Table 1), with all types of bricks showing marks, except tubuli (Table 2). These marks belong essentially to the categories of finger-made marks (line, curve and derivatives), graffiti (almost all of them traced with the finger; one letter or more and numerals), and stamps. However, this sampling showed some limitations, since it dealt with different contexts of use, corresponding to different officinae and chronologies, between the first century and the seventh century AD. Comparing this with other studies, in a work carried out in the Armorica region, in High Britain, there were 1049 finger-made marks, ready to be analysed [ 7 ], and in England, Brodribb counted 186 marked elements, within hypocaust bricks and rectangular bricks, based on a total of 1167 valid elements [14].
Heritage 2021,44131 Heritage2021,4 6 Figure3.Modeloftilekiln,byFilipeAntunes(©MDDS). Thedatasetgatheredfromthesesiteswasverysignificantanddiversified,amount‐ ingto1216marksfrommorethan40contexts/archaeologicalzones(Table1),withall typesofbricksshowingmarks,excepttubuli(Table2).Thesemarksbelongessentiallyto thecategoriesoffinger‐mademarks(line,curveandderivatives),graffiti(almostallof themtracedwiththefinger;oneletterormoreandnumerals),andstamps.However,this samplingshowedsomelimitations,sinceitdealtwithdifferentcontextsofuse,corre‐ spondingtodifferentofficinaeandchronologies,betweenthefirstcenturyandthesev‐ enthcenturyAD.Comparingthiswithotherstudies,inaworkcarriedoutintheAr‐ moricaregion,inHighBritain,therewere1049finger‐mademarks,readytobeanalysed [7],andinEngland,Brodribbcounted186markedelements,withinhypocaustbricksand rectangularbricks,basedonatotalof1167validelements[14]. However,despitethelargequantityofanalysedsamples,itshouldbeconsidered thatpiecesfromtwogroups,inparticular,werepresentinverylargenumbers:thether‐ maeofAltodaCividadeandthearchaeologicalzoneofthedomusofCarvalheiras,re‐ sultingfromtheextensivearchaeologicalinterventionsdevelopedhere. Table1.BracaraAugusta:listofmarkscollected. ArchaeologicalSite Numberof Identified Marks % Falperra231.89 Dume151.23 Braga(withoutcontext)988 Paço10.08 Muralha20.16 PraiadasSapatas171.4 GualdimPais(1989)20.08 GualdimPais(2000)10.16 “25Abril”90.74 Figure 3. Model of tile kiln, by Filipe Antunes (©MDDS). Table 1. Bracara Augusta: list of marks collected. Archaeological Site Number of Identified Marks % Falperra 23 1.89 Dume 15 1.23 Braga (without context) 98 8 Paço 1 0.08 Muralha 2 0.16 Praia das Sapatas 17 1.4 Gualdim Pais (1989) 2 0.08 Gualdim Pais (2000) 1 0.16 “25 Abril” 9 0.74 Carvalheiras 209 17.19 Granjinhos 45 3.7 Rua A. Henriques (1993) 6 0.49 Cavalariças 43 3.54 Fujacal 17 1.4 Avenida Central necropolis 16 1.32 Rodovia necropolis 2 0.16 Carlos Amarante necropolis 11 0.9 Cangosta da Palha necropolis 31 2.55 Colina da Cividade 36 2.96
Heritage 2021,44132 Table 1. Cont. Archaeological Site Number of Identified Marks % Termas 340 27.96 CaféAvenida 17 1.4 Rua S. Geraldo, 34 25 2.06 Largo S. João do Souto 4 0.33 Hospital 3 0.25 Ex Albergue Distrital 20 1.64 Casa da Bica 26 2.14 Tanque de Água 2 0.16 Rua do Anjo, 55 22 1.81 Misericórdia (Mis 96) 10 0.82 Rua de S. Geraldo (1994) 14 1.15 Cardoso da Saudade (1993) 18 1.48 Frigideiras 9 0.74 Rua Frei Caetano Brandão, 183/185 e rua Santo António das Travessas 20/26 (1998/2001) 14 1.15 Seminário de Santiago (1996) 38 3.13 Jardim da Misericórdia (1998) 11 0.9 Misericórdia A (1999) 11 0.9 Jardim da Misericórdia (1996) 25 2.06 Fujacal (1997/1998) 15 1.23 Sé4 0.33 Rua de S. Sebastião 2 0.16 Fonte do Ídolo 1 0.08 Rua de S. Geraldo, 27/31 1 0.08 Total 1216 -
Heritage 2021,44133 Table 2. Bracara Augusta: list of marks collected, according to ATC typology. Typology Number of Marks % Tegula 290 23.85 Imbrex 8 0.66 Vault brick 46 3.78 Latere bessalis 116 9.54 Latere pedalis 20 1.64 Latere sesquipedalis 4 0.33 Latere bipedalis 17 1.4 Latere longum semi bessalis 2 0.16 Latere longum semi pedalis 190 15.63 Latere longum bessalis 4 0.33 Latere lydion 466 38.32 Beveled brick 1 0.08 Circular brick 4 0.33 Other shapes/ indeterminates 48 3.95 Total 1216 - However, despite the large quantity of analysed samples, it should be considered that pieces from two groups, in particular, were present in very large numbers: the thermae of Alto da Cividade and the archaeological zone of the domus of Carvalheiras, resulting from the extensive archaeological interventions developed here. 5. Methods The approach to the finger-made marks in this study followed the methodology proposed by Goulpeau and Le Ny [ 7 ]. Their investigation on these marks in the context of Gaul was fundamental to issues related to the marks’ frequencies, functionality and typological organization. Among other aspects, the authors standardized the description of the finger-made marks, based on a specific terminology, exact form and size of each mark. They proposed a rigorous and innovative reference typology, organized into large families of marks and their subtypes, which we have adapted to our study context. Regarding the other types of marks, namely incisions, stamps and graffiti, we focused on some studies developed by Brodribb [ 14 , 15 ], Le Ny [ 37 ], Charlier [ 8 ] and Ferdière [ 1 ], dealing, respectively, with the study of materials from Britannia and Gaul, which provides us with elements related to the characteristics of production, the marks ´ function, and even classification hypotheses, in comparable contexts to the Western Roman Empire. In the present investigation a protocol based on the one proposed by the ATC Network (Réseau TCA) [ 38 ], a French working group dedicated to the study of ceramic building materials, in which we take part, was developed, considering the materials collection, classification and storing specificities. This protocol is complementary to the above methodology, allowing us to establish the bases of the work upstream of the study of the marks themselves, and valuing the importance of the ceramic building materials. One of the main transversal findings in common to all authors that have been investigating this theme, is the need to organize the main forms into large types, or in other words into typologies. Considering the exact shape of each mark and its multiple variations, we tried to define a set of large families into which they could be grouped (see detailed description of the tasks in Figure 4).
Heritage 2021,44134 Heritage2021,4 9 Figure4.Maintasksofourstudy. 6.Results Wehaveestablishedsevendistinctgroups:curvilinear—1,waves—2,loops—3, dashes—4,complexes—5(finger‐mademarks),finger‐lineletters—6,finger‐linenum‐ bers—7(graffiti).Additionally,wecreatedagroupfor(rare)stamps—8andanotherone foraccidentalmarks—9(Figure5).Thegraffitigroup,whichincludedfinger‐madeletters andnumerals,with512elements,wastheonethatassumedthegreatestrepresentation (42.1%ofthetotal)followed,indecreasingorder,bythecurvilinearone(240 marks—19.74%),thedashes(217marks—17.85%)andtheloops(142marks—11.68%) (Table3). Figure 4. Main tasks of our study. This approach presents some limitations, namely in the specific case of the fingermade marks and the graffiti, which are those that appear in larger numbers, due to the fragmented state of many marks, as well as to the great similarity of some motifs, which complicates the typological attribution to one or another family. Furthermore, the materials from the oldest excavations have not always been preserved, and their drawings do not have the necessary graphic quality. This work should be seen as a first approach and its progression will have to go through updating of the typology and the study of the marks from the various referenced archaeological sites, namely those that are supported by a secure stratigraphy. 6. Results We have established seven distinct groups: curvilinear—1, waves—2, loops—3, dashes—4, complexes—5 (finger-made marks), finger-line letters—6, finger-line numbers— 7 (graffiti). Additionally, we created a group for (rare) stamps—8 and another one for accidental marks—9 (Figure 5). The graffiti group, which included finger-made letters and numerals, with 512 elements, was the one that assumed the greatest representation (42.1% of the total) followed, in decreasing order, by the curvilinear one (240 marks—19.74%), the dashes (217 marks—17.85%) and the loops (142 marks—11.68%) (Table 3).
Heritage 2021,44141 Heritage2021,4 16 Figure11.DomusofCarvalheiras:markswithhigherfrequency(adaptedfromF.Antunessketch). Figure 11. Domus of Carvalheiras: marks with higher frequency (adapted from F. Antunes sketch).
Heritage 2021,44142 Heritage2021,4 17 Figure12.DomusofCarvalheiras:someoftheidentifiedvariantsofATCmarks(adaptedfromF. Antunessketch).. Figure 12. Domus of Carvalheiras: some of the identified variants of ATC marks (adapted from F. Antunes sketch).
Heritage 2021,44143 Heritage2021,4 18 Figure13.DomusofCarvalheiras:ATCmarksincapitalletters(adaptedfromF.Antunessketch).(N/A:notapplicable). 7.Conclusions BracaraAugustawasanimportantRomancityinHispaniaandtheanalysisofthe datasetsofdocumentedbrickmarksfoundthere,invariousarchaeologicalsites,pro‐ videdsignificantresultsandinsightsintothecity´sbrick/tileworkersandrevealedpart oftheinnerlifeoftheofficinae.Inparticular,thedomusofCarvalheiraswasanimportant casestudydueitssizeanddiversityofavailabledata.Inthispaper,weapproachedthe ATCmarkmeanings,themarkingcriteria,theproportionofmarkedmaterials,andwe proposedatypologyessay.Theanalysiswasbasedonavastdatasetof1216elements, withinwhichwehaveindividualized847distinctmarks,whichweclassifiedintonine types.Morethan90%ofthemarkscollectedbelongtothegroupoffinger‐mademarks (geometricshapes,finger‐linelettersandfinger‐linenumbers).ThetypesofATCthatare mostmarkedarethelaterelydionandthetegula,withrespectively466and290examples. Thedocumentedevidenceseemedtohighlightanintenseactivityinthecity,withthe collaborationofmanyworkers,atdifferentperiods,afactprovenbythestratigraphy documentedinseveralofthearchaeologicalsitesmentioned,suchasthedomusofCar‐ valheiras.Similarly,theidentificationofvariantsinmanyofthedocumentedtypesof markscouldemphasizethepresenceofseveralpottersinthesameworkshop.Certain elementsrevealedanorganizedproduction,withsomeworkshopslocated,atleast,at Figure 13. Domus of Carvalheiras: ATC marks in capital letters (adapted from F. Antunes sketch). (N/A: not applicable).
Heritage 2021,44144 Table 6. Domus of Carvalheiras: frequency of identified typologies of marks. Mark Typology Frequency % Subtotal % Curvilinear 64 30.62 Finger-made marks: 57.89 Waves 11 5.26 Loops 7 3.35 Dashes 31 14.83 Complexes 8 3.83 Finger-line letters 68 32.54 Graffiti: 37.32 Finger-line numbers 10 4.78 Indeterminates 10 4.78 7. Conclusions Bracara Augusta was an important Roman city in Hispania and the analysis of the datasets of documented brick marks found there, in various archaeological sites, provided significant results and insights into the city ´ s brick/ tile workers and revealed part of the inner life of the officinae. In particular, the domus of Carvalheiras was an important case study due its size and diversity of available data. In this paper, we approached the ATC mark meanings, the marking criteria, the proportion of marked materials, and we proposed a typology essay. The analysis was based on a vast dataset of 1216 elements, within which we have individualized 847 distinct marks, which we classified into nine types. More than 90% of the marks collected belong to the group of finger-made marks (geometric shapes, finger-line letters and finger-line numbers). The types of ATC that are most marked are the latere lydion and the tegula, with respectively 466 and 290 examples. The documented evidence seemed to highlight an intense activity in the city, with the collaboration of many workers, at different periods, a fact proven by the stratigraphy documented in several of the archaeological sites mentioned, such as the domus of Carvalheiras. Similarly, the identification of variants in many of the documented types of marks could emphasize the presence of several potters in the same workshop. Certain elements revealed an organized production, with some workshops located, at least, at times, near residential areas, where men, women and also children seemed to work. The case of stamps was quite interesting, on the one hand, there were very few examples, with only two pieces found, on the other hand, they were usually associated with open and dynamic markets [ 7 ]. Bracara Augusta, partly due to its geostrategic position, was shown from very early times to be a privileged market, supplying the domestic consumption of the city and its area of influence, as proven by the study of its trade based on ceramic materials [ 9 ]. In the case of ATC materials, and reinforcing what was said above, we believe that a similar situation could have happened, although this aspect needs to be further explored. Some of the marks were associated with the workers that made them, identified by their names and they could be the owners of the workshops, the production foremen or even “simple” workers. The specific functionality of the marks and the marking criteria still need further investigation, but some of the cases recorded in Braga seemed to lift the veil on the internal life of the officinae, informing about the different responsibilities in production between the workers and the owners. The main limitations of this study are related to the fragmentary nature of some marks, which makes their reading, interpretation and inclusion in one or another category difficult. In the same way, the pieces collected in the oldest excavations do not present a secure stratigraphic framework. It should also be noted that the majority of the marks identified in the city belong to the group of the finger-made marks and graffiti, precisely those which offer greater difficulty in terms of understanding their functionality. Future work on the mineralogical and chemical characterization of ATC, under development and already applied by our team in the context of ancient mortars [ 42 , 43 ], could give more information on the raw materials used, namely their provenance, as well as the
Heritage 2021,44145 manufacturing process, workshop organization and production distribution. These characterization analyses could be linked with stratigraphy, allowing a deeper contextualization, which might lead to the development of further chronotypologies. Additionally, it would be important to consider the development of an adequate analysis platform, defining, for example, the technical aspects (movements performed, fingers used, etc.) and metrological aspects and extending the comparative analysis to other areas of marked ATC. Another possible increment to this investigation could be the update of the initial database, with materials collected from excavations in recent years, that could result in the identification of groups of marks with greater frequencies and in the updating of known marks. Author Contributions: Conceptualization, J.R. and A.F.; methodology, J.R. and A.F.; validation, J.R. and A.F.; formal Analysis, J.R. and A.F.; investigation, J.R., A.F. and F.A.; writing—original draft preparation, J.R. and A.F.; writing—review and editing, J.R. and A.F. All authors have read and agreed to the published version of the manuscript. Funding: This research was supported by the Landscape, Heritage and Territory Laboratory (Lab2PT), Ref. UIDB/04509/2020, financed by national funds (PIDDAC) through the FCT/MCTES, and the GeoBioSciences, GeoTechnologies and GeoEngineering Research Centre (GeoBioTec), Ref. UIDB/04035/2020, funded by FCT and FEDER funds through the Operational Program Competitiveness Factors COMPETE and by national funds (OE), through FCT in the scope of the framework contract foreseen in the numbers 4, 5 and 6 of the article 23, of the Decree-Law 57/2016, of 29 August, changed by Law 57/2017, of 19 July. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Acknowledgments: The authors wish to thank and remember Filipe Antunes, whose interest on brick marks at the Museum of Archaeology D. Diogo de Sousa, in Braga (MDDS) allowed the development of this approach. The collection and synthesis work (unpublished) developed by Filipe Antunes, former technician at the MDDS, who collected, inventoried and drew the different marks in materials collected from the excavations in Braga, until the beginning of the 2000s, outlined a first typology approach and constituted the basis of this study. The sampling was carried out at the MDDS. The authors are grateful to Isabel Silva (Director of MDDS), Maria JoséSousa, Clara Lobo, Vitor Hugo Torres, Isabel Marques, Manuel Santos, Amélia Marques and all the staff of MDDS, for their help in the access to the materials of interest to this study. Manuela Martins, Luis Fontes, and other archaeologists at the Archaeology Unit of the University of Minho also provided access and context to the materials. Conflicts of Interest: The authors declare no conflict of interest. Abbreviations ATC architectural terracotta GeoBioTec GeoBioSciences, GeoTechnologies and GeoEngineering Research Centre IND indeterminate IN2PAST Associate Laboratory for Research and Innovation in Heritage, Arts, Sustainability and Territory Lab2PT Landscape, Heritage and Territory Laboratory LSP longum semi pedalis MDDS Museum of Archaeology D. Diogo de Sousa N/A not applicable TCA terres cuites architecturales UAUM Archaeology Unit of the University of Minho
Heritage 2021,44146 References 1. Ferdière, A. La production de terres cuites architecturales en Gaule et dans l’Occident romain, àla lumière de l’exemple de la Lyonnaise et des cités du nord-est de l’Aquitaine : Un artisanat rural de caractère domanial? Rev. Archéologique Du Cent. De La Fr. 2012,51, 17–187. 2. Martins, M. Bracara Augusta: A Roman town in the Atlantic area. In Early Roman Towns in Hispania Tarraconensis. Journal of Roman Archaeology Supplementary Series 62; Abad Casal, L., Keay, S., Ramallo Asencio, S., Eds.; JRA: Portsmouth, RI, USA, 2006; pp. 213–222. 3. Martins, M.; Braga, C.; Magalhães, F.; Ribeiro, J. Constructing identities within the periphery of the Roman Empire: North-west Hispania. In Rome and Barbaricum. Contributions to the Archaeology and History of Interaction in European Protohistory; Curcã, R.G., Rubel, A., Symonds, R., Voß, H.U., Eds.; Archaeopress: Oxford, UK, 2020; pp. 135–154. 4. Martins, M.; Carvalho, H. Bracara Augusta and the changing rural landscape. In Proceedings of the International Colloquium “Changing Landscapes. The impact of Roman towns in the Western Mediterranean”, Castelo de Vide, Marvão, Portugal, 15–17 May 2008; Corsi, C., Vermeulen, F., Eds.; Ante Quem: Bologna, Italy, 2010; pp. 281–298. 5. Martins, M.; Fontes, L. Bracara Augusta. Balanço de 30 anos de investigação na capital da Galécia Romana. In Simulacra Romae II : Rome , les Capitales de Province (Capita Provinciarium) et la Création d’un Espace Commun Européen : Une Approche Archéologique; González Villaescusa, R., de Arbulo, J.R., Eds.; SociétéArchéologique Champenoise: Reims, France, 2010; pp. 111–124. Available online: http://data.cervantesvirtual.com/expression/581249 (accessed on 8 April 2021). 6. Magalhães, F. The Roman domus on the NO Peninsula. Architecture, Construction and Sociability (In Portuguese). Ph.D. Thesis, University of Minho, Braga, Portugal, 2019. Available online: http://repositorium.sdum.uminho.pt/handle/1822/64109 (accessed on 28 April 2021). 7. Goulpeau, L.; Le Ny, F. Les marques digitées apposées sur les matériaux de construction gallo-romains en argile cuite. Rev. Archéologique de L’ouest 1989,6, 105–137. [CrossRef] 8. Charlier, F. Les conditions socio-juridiques du travail dans les tuileries d’après les marques sur les matériaux en Gaule et dans les autres provinces occidentales romaines. In Le Travail. Recherches Historiques; Annequin, J., Geny, E., Smadja, E., Eds.; Presses Universitaires Franc-Comtoises: Paris, France, 1999; pp. 163–204. Available online: https://www.persee.fr/doc/ista_0000-0000_ 1999_act_698_1_2467 (accessed on 14 May 2021). 9. Morais, R. Autarcia e Comércio em Bracara Augusta. Contributo Para o Estudo Económico da Cidade No Período Alto-Imperial (Bracara Augusta: Escavações Arqueológicas 2); Unidade de Arqueologia da Universidade do Minho: Braga, Portugal, 2005; ISBN 972-9382-14. 10. Jahn, V. Die römischen Dachziegel von Windish. Anz. Für Schweiz. Alterhemskunde 1909,11, 110–129. 11. Bernardin, J. Les signes sur tuiles romaines. La Physiophile (SociétéD’études D’histoire Nat. De Montceau-Les-Mines) 1954 ,41–42, 65–66. 12. Chauffin, J. Les tuiles gallo-romaines du Bas-Dauphiné.Gallia 1956,XIV, 81–88. [CrossRef] 13. Neumann, A. Ziegel aus Vindobona (Der Römische Limes in Österreich, XXVII); Osterreischischen Akademie der Wissenschaffen: Wien, Austria, 1973. 14. Brodribb, G. Markings on tile and brick. In Roman Brick and Tile, Studies in Manufacture, Distribution and Use in the Western Empire (BAR International, Series 68); McWhirr, A., Ed.; BAR Publishing: Oxford, UK, 1979; pp. 211–220. [CrossRef] 15. Brodribb, G. Roman Brick and Tile; A. Sutton Publishing: Gloucester, UK, 1987; ISBN 0-86299-363-6. 16. Charlier, F. La pratique de l’écriture dans les tuileries gallo-romaines. Gallia 2004,61, 67–102. [CrossRef] 17. De Poorter, A.; Claeys, P.J. Les Sigles sur Matériaux de Construction Romains en Terre Cuite de Belgique (Acta Archaeologica Lovaniensia, Monographiae I); Katholoeke Universiteit Leuven: Leuven, Belgium, 1989; ISBN 90-6831-184-0. 18. Rico, C. La production de briques et de tuiles en Bétique romaine. L’exemple de la vallée du Guadalquivir. In Proceedings of the International Conference La Brique Antique et Médiévale: Production et Commercialisation d’un Matériau, Saint-Cloud, France, 16–18 November 1995; Boucheron, P., Thèbert, Y., Broise, H., Eds.; École Française de Rome: Rome, Italy, 2000; pp. 177–192. 19. Mills, P. The Ancient Mediterranean Trade in Ceramic Building Materials: A Case Study in Carthage and Beirut. Roman and Late Antique Mediterranean Pottery 2; Archaeopress: Oxford, UK, 2013. [CrossRef] 20. Santos, J.N. dos. Marcas “dedadas” nas telhas romanas do Castro de Guifões. Stud. Gen. 1962,IX, 235–282. 21. Étienne, R.; Fabre, G.; Lévêque, P.; Lévêque, M. Fouilles de Conimbriga, Vol. II—Épigraphie et Sculpture; De Boccard: Paris, France, 1976; 274p. 22. Curado, F.P. Tégula epigrafada de Póvoa do Concelho (Trancoso). Ficheiro Épigráfico 1988,27, 11–13. 23. Correia, V.H.; Fernandes, L.D.S.; Ruivo, J.S. Os proprietários de oficinas de cerâmica de construção de Conimbriga e da Lusitânia ocidental: Continuidade e ruptura. In L’artisanat Romain : Évolutions, Continuités et Ruptures Italie et Provinces Occidentales (Monogr. Instrumentum, 20), Proceeding of the 2nd Coll. D’Erpeldange, Luxemburg, 26–28 October 2001; Polfer, M., Ed.; Éditions Monique Mergoil: Montagnac, France, 2001; pp. 151–172, ISBN 2-907303-61-9. 24. Fernandes, L.S.; Ferreira, R. Marcas de oficina em tijolos romanos de SEILIVM.Conimbriga 2002,XLI, 257–267. [CrossRef] 25. Sousa, L.; Nunes, M.; Gonçalves, C. Tegulae com marcas de oleiro e pegadas de animais no concelho de Lousada. Oppidum 2007 ,2, 57–74. 26. Morais, R.; Ribeiro, J. Produções cerâmicas de Bracara Augusta. In Hornos, Talleres y Focos de Producción Alfarera en Hispania; Bernal, D., Juan, L.C., Bustamante, M., Díaz, J.J., Sáez, A.M., Eds.; UCA: Cádiz, Spain, 2012; pp. 193–208.
Heritage 2021,44147 27. Ribeiro, J. Arquitectura romana em Bracara Augusta. Uma Análise das Técnicas Edilícias (Colecção Teses Universitárias, nº 5); CITCEM: Porto, Portugal, 2013. Available online: https://ler.letras.up.pt/uploads/ficheiros/15505.pdf (accessed on 14 May 2021). 28. Vieito Covela, S.; Pérez Losada, F.; Fernández Fernández, A. El material cerámico de construcción de la villa y necrópolis de Toralla. Tipología y usos. In Las Villae Tardorromanas em el Occidente del Imperio. Arquitectura y función, Proceedings of the IV Coloquio Internacional de Arqueología en Gijón, Gijón, Spain, 26–28 April 2006; Carmen Fernández Ochoa, Virginia García Entero; Fernando Gil Sendino; Ediciones Trea: Gijón, Spain, 2008; pp. 755–766. Available online: https://issuu.com/martinxvm/docs/ construcciontoralla (accessed on 18 September 2021). 29. Rodríguez Nóvoa, A.A.; Valle Abad, P.; Fernández Fernández, A. Una nueva marca sobre tégula de la Cohors Prima Celtiberorum procedente del campamento de Ciadella (Sobrado dos Monxes, A Coruña). Ex. Off. Hisp.-Boletín De La SECAH 2021,12, 65–68. 30. Pérez Losada, F. Contribuciónóestúdio da cerámica de construcción na Galicia romana. In Galicia: Da Romanidade áXermanización: Problemas Históricos e Culturais, Proceedings of the Encontro Científico en homenaxe a Fermín Bouza Brey (1901–1973), Santiago de Compostela, Spain, 19-23 October 1992; Museo do Pobo Galego: Santiago de Compostela, Spain, 1993; pp. 241–262. Available online: https://dialnet.unirioja.es/servlet/articulo?codigo=7580812 (accessed on 18 September 2021). 31. Martins, M.; Carvalho, H. Roman city of Bracara Augusta (Hispania Citerior Tarraconensis): Urbanism and territory occupation. Agri. Centuriat: Int. J. Landsc. Archaeol. 2017,14, 79–98. 32. Sousa, J.J.; Ponte, M.S. Novos elementos para a arqueologia bracarense. In Actas das I Jornadas Arqueológicas; Associação dos Arqueólogos Portugueses: Braga, Portugal, 2017; pp. 389–411. 33. Sousa, J.J. Subsídios para a carta arqueológica de Braga. Studia Archaeol. 1973,23, 20. 34. Sousa, J.J. Subsídios para o estudo da arqueologia Bracarense. Lucerna 1966,5, 589–599. 35. Sousa, J.J. Inventário de materiais para a arqueologia bracarense: Lucernas. Bracara Augusta 1966,20, 165–178. 36. Sousa, J.J. Novo molde de lucernas aparecido em Braga. Trab. De Antropol. E Etnogr. 1969,XXI, 309–311. 37. Le Ny, F. La Production des Matériaux en Terre Cuite en Gaule Romaine. Ph.D. Thesis, Universitéde Rennes I, Rennes, France, 1992. 38. Réseau, T.C.A. Groupe Méthodologie, 2015, Protocole de Prélèvement des Terres Cuites Architecturales (TCA). Available online: https: //reseautca.hypotheses.org/files/2021/01/PROTOCOLE-PRELEVEMENT-DES-TERRES-CUITES-ARCHITECTURALES.pdf (accessed on 8 May 2021). 39. Morais, R. Transition from the Pre-Roman World to the Roman World in the Northwest of the Iberian Peninsula: The Example of Bracara Augusta. In Proceedings of the IX International Research Conference (IRC 2020), New York, NY, USA, 30–31 January 2020; International Scholary and Scientific Research & Innovation: New York, NY, USA, 2020; pp. 2020–2031. Available online: https://hdl.handle.net/10216/130140 (accessed on 10 May 2021). 40. Marado, L.M.; Ribeiro, J. Biological Profile Estimation Based on Footprints and Shoeprints from Bracara Augusta Figlinae (Brick Workshops). Heritage 2018,1, 33–44. [CrossRef] 41. Triães, R.; Correia, V.H.; Coroado, J. A utilização dos materiais cerâmicos de construção em Conimbriga. Conimbriga 2002 ,XLI, 153–164. [CrossRef] 42. Fragata, A.; Ribeiro, J.; Candeias, C.; Velosa, A.; Rocha, F. Archaeological and Chemical Investigation on the High Imperial Mosaic Floor Mortars of the Domus Integrated in the Museum of Archaeology D. Diogo de Sousa, Braga, Portugal. Appl. Sci. 2021 ,11, 8267. [CrossRef] 43. Fragata, A.; Candeias, C.; Ribeiro, J.; Braga, C.; Fontes, L.; Velosa, A.; Rocha, F. Archaeological and Chemical Investigation on Mortars and Bricks from a Necropolis in Braga, Northwest of Portugal. Materials 2021,14, 6290. [CrossRef]