Exploring material culture and identity in post-medieval Catalonia: a long-term archaeological perspective on greyware pottery production in Quart, Girona, Spain

SUMMARY

Ceramic production has shaped the industrial and human landscapes of the Gavarres Mountains in northeastern Catalonia. This activity remains important in some places, where it transitions between traditional practices and industrial modes of production. Quart, a town known for its greyware pottery production, is an emblematic centre where artisanal expertise in this ancient manufacture has been upheld as an identity-building value despite increasing industrialization. Our research examines the transformation of production processes through landscape archaeology, material science, archival research, and literature review within a diachronic framework starting in the thirteenth century AD. Our aim is to provide an overview of the pottery production in northeastern Catalonia, using Quart as a case study, as a means to integrate the material culture and industrial development of this area into archaeological studies.

Keywords:

• Pottery
• material science
• craftsmanship
• cultural heritage
• diachronic

INTRODUCTION

Northeastern Catalonia has a strong tradition of pottery making, particularly in the territory surrounding the Gavarres Mountains. This area is at the northern limit of the Catalan Coastal Range and provides a singular landscape of great personality, heavily affected by human action but still reflecting much of the activities and ways of life of the human communities that have inhabited them through time. The earliest written evidence about pottery production in this area appeared in the fourteenth century AD, even though archaeological evidence points towards the existence of an already consolidated potting tradition by that time. Today, the towns of Quart and la Bisbal d’Empordà (Girona, Spain) foster this economic activity as a heritage value and a matter of cohesive and identity-building significance. The aim of our paper is to delve into the greyware pottery production modes and outcome, focusing in the town of Quart – which remains one of the oldest and longest-lasting examples to date – as a case study (Fig. 1:1).

FIG 1 Location map of pottery production in the Gavarres Mountains. The locations of greyware and redware workshops as well as clay quarries are indicated (1). Detailed sampling at the town of Quart is provided (2). The size of the sampling circles is proportional to the number of samples included in each location in both cases. Background ortophotograph is provided by the Cartographic Institute of Catalonia (Institut Cartogràfic de Catalunya, ICC).

THE BEGINNING OF A LONG-LASTING TRADITION

On May 7, 1572, King Philip II of Spain granted the potters of Quart the privilege to meet in the parish church and regulate the good order and harmony of a guild devoted to the Saints Justa and Rufina. This guild was founded to protect the interests of a collective of about twelve workshops at that time. The King’s privilege allowed for their freedom of assembly in the absence of royal authority or the constable of Girona. Despite the continuous interference from Gerundian authorities, the potters of Quart managed their guild independently until the late eighteenth century, when the constable was imposed on them. The legal framework established in 1572 allowed the potters of Quart to establish their legal system, regulate access to the economic activity, and exercise corporate solidarity (Soler 2011, 600–601).

However, this privilege was not the foundational milestone of pottery production in the territory, but rather the recognition of a pre-existing and enhanced socio-economic phenomenon. The first written mention of pottery production in Quart dates back to 1312 when pots made in Quart were taxed when entering the city of Girona (Madurell 1968, 77). Based on subsequent events and according to research conducted in recent years (Rosal 2007; Travé, López, and Álvaro 2013; Travé, Creus, and Vicens 2021a), pottery production in Quart was already a consolidated activity at that time. Historical sources relating to the first known potters, however, appear somewhat later, with references to Martí Vicens – already deceased in 1470 – and Jaume Roig, mentioned in 1474 in the will of his widow Margarida (Rosal 2007, 19). The first direct references to potters in business mention Pere Tauler and his two sons in 1484 (Soler 2011, 602).

The case of Quart is not unique in the district, as evidenced by the taxes paid in 1362 for the pots produced in Masarac and Vilarnadal (Rocas, Vicens, and Basart 2016, 39) and mentions of other potters in the area, such as the potter Guillem Terrades, who worked in Castelló d‘Empúries in 1384 (Rosal 2007, 17). It is interesting to note that a widow woman, whose first name is unknown and is referred to by her deceased husband’s surname Sastre, is explicitly mentioned as a female potter in Girona in 1497 (Rosal 2007, 18). References to potters increased in the second half of the sixteenth century, as the production in the neighbouring town of La Bisbal d’Empordà was consolidated. The first potter – Rafael Prats – is mentioned in 1542 (Rosal 2008, 43; Rocas 2017, 13) and, by the late sixteenth century, when the potter’s guild in Quart was founded, la Bisbal d’Empordà already had over twenty active potters. Pottery production in the Gavarres experienced an exponential increase from the mid-seventeenth century onwards, with the emergence of new production centres in Palamós (Rosal 1993), Palafrugell (Rosal 1987), and Pals (Subirana 2017). That marked the beginning of a period of splendour characterized by numerous kilns and brickworks, most of which persisted until the nineteenth century. Despite the challenges faced by this manufacture over the last one hundred years, the production of pottery and tile has remained active in some iconic centres such as Quart and La Bisbal d’Empordà, where some workshops continue to operate to this day.

MATERIAL CULTURE, IDENTITY AND HERITAGE PRESERVATION

The rise of industrialization has mechanized some of the ancient procedures, and consumers’ changing demands have led to transformations in vessel typologies and paste recipes. The development and use of new materials, such as plastic for food containers and multi-purpose homeware, aluminium and stainless steel for kitchenware, and, finally, non-stick coatings made of polytetrafluoroethylene, have put the continuity of ceramic vessels at risk. Despite the recent trend towards finding sustainable materials, there is a growing concern that a longstanding tradition passed down through generations is in danger of disappearing as younger generations progressively lose their interest in artisanal jobs.

The production of greyware pottery at the town of Quart is an example of this phenomenon. Recently, archaeological research into these production practices, particularly in the post-medieval and industrial periods, has provided a theoretical and methodological framework for revisiting traditional craftsmanship from new perspectives (Travé and Vicens 2018). The concept of taskscape (Ingold 1993; Gruppuso and Whitehouse 2020), understood as a landscape shaped by production activities and a community of practice (Brysbaert 2020) is useful for us to approach the topic of traditional pottery production as a landscape-building activity. Our contribution aims to provide an up-to-date overview and discussion of pottery production in Quart and its cultural impact on the surrounding region, based on the wider study of this phenomenon in the Gavarres area.

In this paper we introduce a diachronic view of the greyware pottery from Quart, examining the transformation of production processes by means of landscape archaeology, material science, archival research, and literature review. Our aim is to provide an updated view of pottery production in Quart as a means of integrating the material culture and industrial development of this area, and to discuss some of the challenges of interdisciplinary work combining written, material, graphic and ethnographic sources with archaeological studies.

SITE, MATERIALS AND METHODS

In this section, we provide an overview of the current research on greyware pottery from Quart. We will also introduce the selected materials and the datasets generated for our study. The quantitative and qualitative exploitation of data by means of GIS technology, exploratory statistics and diversity analysis on pottery are the basis of our work. This section outlines the approach taken to gather and analyse our data, and the tools we used to explore and visualize our findings.

LITERATURE REVIEW

There is a significant amount of published literature that provides a synthesis of the existing data found in both public and private historical archives. This information pertains to the existence of potters, workshops, and production activities in Quart and other locations within the Gavarres area. Examples of such archives include the parish archives in La Bisbal d’Empordà and Palamós (Rosal 2008, 1993), the diocesan archive of Girona, and the historical archives of Girona and La Bisbal d‘Empordà (Rosal 2007). Notable works on the potter’s guild and their accounting books have been produced by N. Soler (2011, 2021), whereas anthropological research has compiled oral traditions and memories as well as human landscapes and artifacts. Authors such as Santanach, Rosal, and Suñol (1998) and Sempere (1982, 84–110), among others (Castellanos 1987; Guerrero 1988, 184–204), have also contributed to the study of this phenomenon.

However, the archaeological approach to pottery production has not been as extensively developed. Although there have been proposals for typological classifications of traditional productions such as redware and greyware in the Gerundian area (Castellanos 1987; Sempere 1985, 17–26; Sàez 2009, 97–98), particularly in the cases of Quart and La Bisbal d’Empordà, these classifications are not precise in archaeological terms regarding morphology and dimension in particular. Some pottery assemblages have been published (Albertí and Santanach 1984; Soler 2009; Travé, Creus, and Vicens 2021a; Vicens and Travé 2018), but are still insufficient in number, whereas, in some cases, historical or archaeological perspectives regarding typologies and contexts are not provided (Martín 2017; Paul, Sau, and Solés 2020). Nonetheless, there is some other valuable information about raw material sources in the area (Rocas and Roqué 2015; Subirana 2017) and the location of ancient workshops or kilns (Serra and Ferrer 2002; Romero and Rosal 2014).

Historic sources informing about pottery production are sometimes scarce, lacking details about the processes themselves, the technology employed, or the idiosyncrasies of different centres or workshops. In contrast, abundant material has been recovered from archaeological contexts, surface surveys, and in private or museum collections. Macroscopic and microscopic analysis of these materials provides valuable information on the abovementioned aspects, particularly raw material procurement, clay choice, paste recipes, or shaping and firing techniques. By analysing these materials, we can identify some of the features related to the exploitation of natural resources and their transformation in the Gavarres area. This analysis also provides new data on the relationship between communities and the natural environment, a relationship that has been well documented in medieval times in the area (Mallorquí 2000, 129–138). The production of Quart serves as a case study for this analysis.

POTTERY COLLECTIONS AND DATA GATHERING

The results presented in this paper are built upon the analysis of 550 individuals/samples comprised in three different assemblages. The first assemblage includes 89 samples representing complete or very slightly fragmented individuals that formed the entire collection of greyware pottery kept at the former Pottery Museum of Quart. This collection is representative of the whole range of products made at the town and includes vessels from the 18th to 20th centuries, which are representative of the diachronic evolution of greyware pottery-making at the town. The second assemblage accounts for 195 individuals found in three different archaeological contexts. Some of these were waste pottery reused as vault-building material in Quart (49) and La Bisbal d’Empordà (12), and kept at the Terracotta Museum of La Bisbal d’Empordà. The archaeological context and chronology of the first assemblage is uncertain but, according to the parallels found in the area and the typology of the included individuals, an estimated timespan between the 16th and 18th centuries seems reasonable. The second one is dated to the beginning of the nineteenth century. Notwithstanding some uncertainties in their chronology, these sets are representative of synchronic typologies and production techniques. The presence of vault-building materials in Quart, a well-known production centre, provides reasonable evidence of their origin. Additionally, discovering vault-building pottery in La Bisbal, dating back to a time before the establishment of a consolidated greyware production in this town (which is generally considered to have begun after 1800), raises the possibility of attributing these materials to Quart. Alternatively, it challenges the information provided in written sources, suggesting that production in La Bisbal may have commenced earlier than previously believed. The former hypothesis appears to be the more plausible explanation.

Another set of archaeological individuals was found in Girona (134), dating from the fifteenth century, and studied in previous projects (Travé, López, and Álvaro 2013; Travé, Creus, and Vicens 2021a). An additional group of 266 samples forms a more heterogeneous set of pottery sherds recovered in a wide area across the Gavarres Mountains and coming from surface survey (Fig. 1:1–2).

The assemblage was documented through graphic representation (Fig. 2:1–2) when possible and detailed description (Fig. 2:3). Photographs were taken of the entire assemblage, and pottery profile and elevation drawings were made for 147 individual pieces. These drawings allowed us to measure up to four diameters (rim, base, minimum, and maximum) when feasible. The minimum, medium, and maximum wall thickness and the height of the vessel based on the preserved fragment were measured as well. We also recorded their profile shape and a normalised figure obtained by measuring three or four specific vector points using a predetermined pattern (Fig. 2:4). This simplified expression of the vessel shape, first suggested by Llanos and Vegas (1974), allows for the identification of typologically meaningful clustering when considering other morphometric variables. Macroscopic examination of the entire assemblage enabled us to record their surface finish and colour. In addition, detailed macroscopic analysis of the paste in 400 broken sherds enabled us to record firing atmosphere and paste coarseness. We also noted other macroscopic features of complete individuals, which were visible through cracks or fresh fractures resulting from minor breakage. Furthermore, standard 30 μm thin sections of 50 representative sampled sherds were examined under a polarizing light microscope and visually grouped into petrographic fabrics, based on the nature of their inclusions, clay matrix, and voids (Quinn 2022, 89–97).

FIG 2 Graphic record of the examined vessels through drawing (1) and photograph (2), detailed description within the database (3) and morphometric calculation of variables (4).

During the data gathering phase of our project, we recorded additional rich and varied evidence about vessels still in use or reuse, frequent vestiges of potsherds reused as building materials, and complementary data from written and graphic sources about potters, guilds, and production processes. We also collected information about the various sources of raw clay materials that are currently or were previously in use, as well as their georeferencing in UTM coordinates (ETRS89-31N) through cartographic sources provided by public administrations (the Cartographic and Geological Institute of Catalonia through their tool www.icc.cat/vissir3, and the Spanish Mining Institute through the resource www.ign.es/iberpix/visor). We captured and processed information obtained from both literature review and raw data arising from archaeological material analysis using a relational ontology-mediated database, recording entities and events as Units of Topography and the Actors that performed them when known (Travé, Del Fresno, and Mauri 2020). The theoretical framework and the practical basis of this information system are published in detail elsewhere (Travé et al. 2021b), and we refer to these previous works to avoid redundancy. Although Quart was the site of reference, data capture was not limited to this site, but included the entire Gavarres area.

RESEARCH APPROACH AND DATA ANALYSIS

The typological distribution of vessels and the technological practices in paste preparation, shaping and firing have been investigated by clustering individuals and determining fabrics, types and productions in accordance with their specific features. To achieve typological clustering, Geometric Morphometrics (GM) were applied to the dataset, which is a common practice in archaeology (Okumura and Araujo 2019, 150; Travé 2022 and literature therein). Morphometric data were expressed as tidy datasets (Wickham 2014) with variables recorded in columns and observations in rows. Datasets were then analysed using Exploratory Data Analysis (EDA) including descriptive statistics such as minimum and maximum values, arithmetic and geometric means and standard deviation. Correlation matrices were used to compare the incidence of different variables in type determination. Principal Component Analysis (PCA) was a useful approach to explore the specific features of vessel types, especially since it allows for the visualization of the patterning within the dataset and identification of the existence of clusters depending on the combined influence of many different variables (Baxter 2006) that are determinant in the final clustering. By examining the correlation or covariance between variables, PCA reduces the dimensionality of multivariate datasets while preserving the information about its inner structure and the differences between samples (Aitchinson 1986; Dillon and Goldstein 1984).

Paste recipes, technological practices and the exploitation of raw materials were investigated through a combined macroscopic and microscopic observation of samples. Visual grouping of thin-sections and detailed descriptions (Quinn 2022, 97–124; Whitbread 1989) of inclusions, clay matrix, and voids allowed us to identify different fabrics, which were subsequently interpreted in terms of their raw materials (Quinn 2022, 167–191) and manufacturing technology (Quinn 2022, 212–278). We used geological maps of the area to identify possible sources of compatible raw materials for the production of ceramics assigned to some of the petrographic fabrics.

The combined interpretation of morphological and technological clusters allowed for the definition of productions that were recorded as Units of Topography. We combined this information with data obtained from written and ethnographic sources, as well as literature from the perspective of landscape archaeology. By deeply mapping (Bodenhamer 2015; Earley-Spadoni 2017; Eanes et al. 2019 and related literature) pottery clusters and correlating them with information about potters, pottery workshops, and clay sources within a diachronic timespan, we were able to understand the Gavarres area as a taskscape (Gruppuso and Whitehouse 2020; Ingold 1993). This investigation enabled us to see how pottery production extended beyond being merely an economic activity and became a distinctive trait of human communities. It fostered the interconnection between people and things and defined the specific identity of artisans. As the results will show, this assessment has led us to understand how industrialization processes have been integrated into daily production activities without jeopardizing the handmade character of the greyware pottery production, which is strongly valued nowadays as a heritage legacy.

POTTERY PRODUCTION IN QUART: ARTEFACTS, PROCESSES, AND ENDURANCE

The results of our study provide data for interpreting the phenomenon of pottery-making from different perspectives. We focused on the materiality of vessels, examining their shapes and functions, as well as the technological methods used in their manufacture and the transformations that occurred over time. By conducting both macroscopic and microscopic analyses, we explored their characteristics in greater depth in order to determine the procurement of raw materials and the exploitation of the landscape. Finally, we examined the broader historical and anthropological implications of pottery-making in this context, the potting networks in the area and the cultural exchange between them.

POTTERY PRODUCTION: VESSEL SHAPE, FUNCTION AND STANDARDIZATION

Greyware vessels from Quart are classified according to their specific shape and size, and they represent a wide range of containers for multiple purposes. These vessels have unique names that enrich the particular vocabulary of the Catalan language, which can be challenging when trying to translate them into English accurately. To address this, we will provide both the Catalan name and an approximate English translation for each vessel. The primary variables that characterize these vessels are their top and base diameters, height, and profile, which are strongly correlated (Fig. 3:3). In terms of morphology, vessels can be classified into open, closed, and singular shapes. We also identified four main families of vessels based on their function: large storage containers, vessels for liquids, cooking pots, and specific vessels for domestic and farming activities (Table 1).

FIG 3 Results from Principal Component Analyses. Scatterplots comparing the first two components, with samples identified according to the vessel type (1) and archaeological context (2). Correlation matrix of variables considered within the analysis (3).

TABLE 1 Typological summary and classification of vessels from Quart according to their function and shape. Medium dimensions for each type are provided and standard deviation indicated in brackets. All dimensions are expressed in mm.

Functional family	Vessel name	(Catalan)	Average mean dimensions (and standard deviation)								Morphometric group (PCA)	Number of individuals	Related figures
			Rim diameter		Base diameter		Maximum diameter		Height
Large storage containers	Basin	Conca	434.1	(53.0)	196.2	(46.6)	435.5	(54.2)	186.1	(24.5)	Open shape – group 1	11	Fig. 4:1
	Washbasin	Cossi	711.2	(17.7)	335.7	(144.2)	761.4	(179.5)	680.6	(194.4)	Open shape – group 2	19	Fig. 4:4
	Bunghole jar	Doll	165.5	(34.7)	183.2	(17.4)	412.6	(71.4)	453.7	(92.3)	Close shape – group 3	8	Fig. 4:3
	Jar	Gerra	188.4	(45.6)	162.6	(32.8)	312.4	(61.8)	348.2	(62.2)	Close shape – group 3	13	Fig. 4:2
Vessels for liquids	Oil Measure	Tramostera	373.5		185.8		367.6		331.4		Singleton	1	Fig. 5:2
	Wine measure	Mallal	135.2	(12.1)	198.3	(56.4)	301.3	(72.8)	355.7	(82.3)	Close shape – group 3	3	Fig. 5:1
	Bucket	Poal	111.0	(1.8)	140.5	(0.2)	315.6	(74.8)	343.0	(3.1)	Close shape – group 3	2	Fig. 5:3
	Wine jar	Barral	44.7	(6.2)	134.2	(26.9)	228.4	(36.3)	277.9	(55.0)	Close shape – group 4 (type 1)	15	Fig. 5:4
	Wine jar	Barral	52.2	(3.3)	196.0	(5.9)	334.5	(5.8)	411.7	(10.9)	Close shape – group 4 (type 2)	8	Fig. 5:5
	Water jug	Càntir	–		133.3	(23.5)	236.9	(47.4)	332.9	(54.9)	Close shape – group 5	12	Fig. 5:7
	Cruet	Càntir d‘oli	13.6	(5.9)	146.5	(48.1)	267.7	(94.3)	357.9	(105.6)	Close shape – group 5	3	Fig. 5:6
	Other Jugs	Càntir de ferrer	–		151.9		230.0		330.6		Close shape – group 5	1	Fig. 5:7
	Other Jugs	Càntir de pescador	–		135.2	(18.5)	191.2	(6.3)	208.8	(17.8)	Close shape – group 5	4
Cooking pots	Pitcher	Ansat	156.3	(25.2)	86.3	(19.6)	183.1	(36.2)	194.7	(34.8)	Close shape – group 6	16	Fig. 5:8–9; 11
	Boiling pot	Olla	225.0	(44.1)	102.1	(22.4)	287.9	(57.8)	272.6	(64.6)	Close shape – group 6	4	Fig. 5:10
Specific vessels	Stove	Fogó	327.4		184.2		327.4		189.5		Singleton	1	Fig. 6:1
	Chicken waterer	Abeurador de polls	248.4		219.3		248.4		39.4		Singleton	1	Fig. 3:1
	Funnel	Embut	318.5	(13.2)	148.1	(2.6)	336.4	(38.5)	248.0	(6.7)	Singleton	2	Fig. 6:7
	Heater	Escalfeta	221.9	(45.4)	83.4	(5.1)	224.3	(41.3)	149.0	(23.3)	Singleton	3	Fig. 6:4
	Costrel	Cantimplora	42.8	(5.0)	143.3	(18.8)	162.3	(14.0)	200.1	(19.0)	Close shape – group 4 (outliers)	5	Fig. 6:10
	Bee smoker	Brescadora	–		146.8	(9.4)	179.4	(0.5)	153.6	(12.8)	Singleton (type 1)	2	Fig. 6:6
	Bee smoker	Brescadora	102.8		127.2		184.1		129.2		Singleton (type 2)	1	Fig. 6:6
	Bottle	Ampolla	41.8	(1.2)	95.4	(4.0)	116.5	(12.5)	226.8	(0.6)	Singleton	2	Fig. 6:9
	Barrel	Tina	200.4		337.4		570.5		587.9		Singleton	1	Fig. 6:3
	Waterer	Abeurador	105.3	(7.6)	159.8	(16.2)	200.0	(3.3)	238.5	(48.4)	Close shape – group 5 (outliers)	2	Fig. 6:8
	Brazier	Maridet	88.9	(9.3)	107.3	(26.7)	162.7	(13.9)	117.4	(30.4)	Close shape – group 6 (outliers)	6	Fig. 6:5

Principal Component Analysis (PCA) of morphometric data allows for a more nuanced view of this distribution (Fig. 3:1–2). Basins (Fig. 4:1) and multi-purpose tubs or washbasins (cossis) clearly stand out in the scatterplots from the entire assemblage, and correspond to groups 1 and 2 respectively. The cossi is a large-sized vessel that occasionally reaches one metre in height and top diameter (Fig. 4:4). These vessels are the most emblematic ones made in this town. Some individuals are sumptuously decorated with undulated and straight incised lines or a variable number of pinched coils attached to the vessel walls. Occasionally, the name of the potter or the date of production is incised below the rim. Among the examined collections, the earliest example found is dated to1770, although fragments of these vessels are also found in some archaeological contexts from the 15th and 16th centuries. They were used as home ware for multiple purposes such as storing grain and other products, or for doing the laundry, especially in those cases with a spout at the base.

FIG 4 Typology of large storage containers from Quart. Basins (1), jars (2), bunghole jars (3) and washbasins (4).

Examining the PCA results in detail, a few observations arise regarding the standardization of vessels and their transformation in shape and size over the centuries. The linear distribution of most vessels, parallel to the height vector within the scatterplot – concretely the shapes pitcher (ansat), jar (gerra), bunghole jar (dolls), wine jar (barrals) and water jug (càntir) – is due to the standardized sizes of most of them, usually distributed in two or three regular sizes that fit most individuals. While pitchers were used for food storage that could be warmed up in the same container, jars were related to long-term preservation of specific products such as olives, and other salt-cured food, and bunghole jars were mainly for oil or other liquids. In the case of wine jars, which are classified in two different types according to their size, profile and distinctive features, the standardization of production is clearly visible. Type 1 (Fig. 5:4) comprises small and medium-sized wine jars, occasionally with longer necks. In contrast, type 2 (Fig. 5:5) comprises large vessels with spherical bellies and a spout on the base for easy emptying since they were heavy when filled.

FIG 5 Typology of vessels for liquids and cooking pots. Wine (1) and oil measures (2), buckets (3), wine jars of type 1 (4) and type 2 (5), cruets (6), water jugs (7), pitchers without pouring rim (8), pitchers (9), boiling pots (10) and deformed or non-symetrical vessels (11).

Similarly, cooking pots exhibit a well-standardized range of sizes, particularly in the case of pitchers. These one-handled heating pots frequently – but not always – have a pinched pouring lip, and we can distinguish up to four clear sizes (Table 2) among them. Smaller pitchers (approximately 150 mm high or less) and larger pitchers (approximately 270 mm high) are the less frequent, while small and medium-sized individuals are the most represented. When comparing data from Tables 1 and 2, it can be noticed that the high figures for standard deviation of medium dimensions of the entire assemblage of pitchers are significantly lowered when considering them size by size. The high standardization detected in the case of pitchers is not as evident in other vessel types with high standard deviation values, such as water jugs (Fig. 5:7) and cruets (Fig. 5:6). In this case, the production of Quart is somewhat heterogeneous, and their evolution through the decades is more pronounced when compared to the production of these same vessels at other sites like La Bisbal d’Empordà.

TABLE 2 Size standardization of pitchers. Medium dimensions for each size are provided and standard deviation indicated in brackets. All dimensions are expressed in mm.

Pitcher size	Medium dimensions (and standard deviation)								Number of individuals
	Rim diameter		Base diameter		Maximum diameter		Height
Very small	124.6	(9.1)	66.2	(4.1)	138.9	(8.4)	150.0	(0.0)	2
Small	140.9	(6.8)	76.2	(11.9)	167.7	(7.7)	174.6	(10.1)	7
Medium	177.3	(13.2)	97.1	(11.3)	202.5	(7.8)	219.6	(10.9)	6
Large	200.9		131.4		285.0		275.2		1

The identification of individuals in the scatterplot according to their contexts suggests that medium-sized closed vessels were preferred as waste material to fill the vaults’ spandrels in buildings. The assemblage found as vault-filling material in Quart, highlighted in black in the PCA biplot (Fig. 3:2), is a good example of this practice. Most of the individuals found in these contexts belong to the five main and more common types made in the town: basins, pitchers, buckets, bunghole jars and wine jars. These types are also the earlier vessel types identified and are also present in other contexts on the 15th and 16th centuries (Travé, Creus, and Vicens 2021a). The terminus ante quem for other vessels found in the archaeological contexts of La Bisbal d’Empordà – highlighted in red in the same figure – is 1802, and these might be attributed to the production of Quart according to the features of their shape and paste. However, the origin of a large bunghole jar (doll) found in La Bisbal d’Empordà is uncertain, since it belongs to a context earlier than the origin of greyware production in this town (Fig. 6:3). It shows a slightly different shape with a wider base, when compared to genuine shapes of bunghole jars from Quart. The presence of other large jars from Quart in the reference collection (Fig. 6:2), uncommon in their shape as well, but with very similar paste, might suggest the same origin for this vessel.

FIG 6 Typology of uncommon or singular vessels for specific purpose. Stove (1), large jars (2), uncommon bunghole jar (3), heaters (4), braziers (5), bee-smokers (6), funnel (7), waterers (8), bottles (9) and costrels (10).

More recently, specifically the 19th and 20th centuries, the variety of vessel shapes and functions has expanded to include highly specialized containers for particular purposes. Amongst these are vessels used to contain embers for cooking (stoves) (Fig. 6:1) or other heating purposes (heaters, braziers and bee smokers) (Fig. 6:4–6). Bee smokers are highly specialized vessels and two different types have been identified in the production of Quart. Both of them are ember containers with a nozzle to blow air and an opening to release the smoke. The differences are mainly related to their handling. Type 1 has a handle and a wide opening on the side, while type 2 is held by a cylindrical shaft, and smoke gets released outside through a spout and a circular opening on top. Cooking pots have mostly disappeared from the modern stock, as they were gradually replaced from the sixteenth century onwards by lead-glazed boiling pots and casseroles fired under oxidizing atmosphere. Greyware was then kept for domestic and farming purposes, storage, animal feeding, and also related to the production and trade of wine and oil such as funnels (Fig. 6:7). More recent pottery from the 19th and 20th centuries includes other more specific smaller shapes like feeders, bottles or costrels (Fig. 6:8–10).

TECHNOLOGICAL CHARACTERIZATION: RAW MATERIALS AND POTTING PROCEDURES

Through petrographic characterization, we identified three different clay pastes within the assemblage of vessels and broken sherds. The fabrics corresponding to the production of Quart are determined based on the main features of their non-plastic inclusions, matrix, and voids. The macroscopic and microscopic paste clustering show good correlation (Fig. 7). These fabrics are labelled as MO-028, MO-044, and MO-046, following the correlative numeration in our database that includes fabrics from previous studies conducted elsewhere in Catalonia.

FIG 7 Micrographs and hand-specimen sherds representative of paste fabrics identified within the production of Quart. Different firing variants of the Coarse Sand-Tempered Fabric (MO-028) (1), Very Coarse Quartz and Feldspar Fabric (MO-044) (2), and Quartz and Mica-rich Fine Fabric (MO-046) (3). Significant features are labelled onto the photographs and micrographs. The label “rc” is an abbreviation for “relic coil”.

The first fabric (MO-028) is defined as the Coarse Sand-Tempered Fabric (Fig. 7:1), and it is a genuine production from Quart disseminated throughout the neighbourhood and found at different sites from the thirteenth century onwards (Travé 2018, 58–63, 124–125; Travé, López, and Álvaro 2013). Non-calcareous clay was tempered with a quartz-rich coarse sand in vessels classified under this fabric, which included feldspar, granite, metamorphic, or argillaceous inclusions in smaller amounts. The clay matrix is extremely fine, possibly resulting from levigation before adding temper. Variants of this fabric correspond to the clay matrix features, particularly the firing atmosphere, since the same paste preparation was used for both redware (not included in our study) and greyware pottery. Samples fired under a reducing atmosphere have slight variation in the homogeneity of colour due to the poor control of the kiln structures, and thus affecting the stability of the firing atmosphere. Significantly, the variation in colour correlates with slight differences in the nature of non-plastic inclusions as well. The irregularly fired variant of samples in this fabric includes a more heterogeneous assemblage of inclusions, richer in metamorphic material and feldspar content. In contrast, the uniform highly reducing variant contains predominantly monocrystalline quartz and fewer other inclusions. This fabric includes most of the samples from late medieval and early modern contexts, and most of their microstructures have visible relic coils, demonstrating the traditional way of shaping vessels in this town, at least during the earliest periods, by coil-building the wall vessels and finishing the surface smoothing on the rotating wheel.

The second fabric (MO-044) identified at this town is not very different in terms of the petrology of inclusions. This Very Coarse Quartz and Feldspar Fabric (Fig. 7:2) includes large fragments of granite and other rock fragments within a slightly coarser clay that has not been depurated. A fine silty fraction naturally occurring within the clay provides a rougher texture to the paste that includes large quartz and granite inclusions, as well as fewer metamorphic rock fragments, mainly phyllite. Similarly to the previous fabric, firing conditions differ among samples and cover a wide range between highly reduced and more oxidizing firing atmospheres, although there are no other variations in the nature of the inclusions.

Notably, all samples within this fabric correspond to large basins or tubs (cossis). This suggests that the paste preparation is related to the needs of shaping these large vessels, which require a thicker clay to stand while shaping. Ethnological evidence and the practices of potters today support this idea. These vessels are not modelled at once; there is typically a lapse of a few hours to a day between shaping the base and the final modelling of upper walls and rim to let the base harden a little and ensure the support of the entire vessel’s weight. Slow rotation of the wheel is enough to assemble the coils and smooth the surfaces but does not eliminate the relic microstructure of coils when observed under the microscope. The silty fraction naturally present within the clay, as well as the coarser inclusions, might have helped to gain the required consistency and, therefore, made it unnecessary to remove them deliberately.

Finally, a third fabric (MO-046), which includes water jugs and very late or subactual sherds found in pottery wasters, is defined as a Quartz and Mica-rich Fine Fabric (Fig. 7:3). This fine and non-tempered paste contains small inclusions of quartz and mica – biotite and muscovite mainly – and exhibits frequent textural features, such as clay pellets and lumps that might result from clay mixing (Quinn 2022, 232–235). This has been a common practice throughout the entire twentieth century and still is in traditional pottery production. Potters in Quart distinguish between two types of clay: red, which is more ferruginous, and white, which is more calcareous, and these are combined according to their needs. A third type of clay, named terra de Gornau after the location of the clay source, includes a considerable proportion of organic matter and provides a greasy texture when added to the paste. The occasional presence of pores with an irregular shape and darkened burnt edges proves the addition of this particular clay. Tempering does not seem to be a common practice after the late nineteenth century, according to the material vestiges. The strong alignment of elongate mica inclusions and the macroscopic appearance of the wall surfaces prove that fast wheel throwing is the common method used to elevate the vessel walls.

All three fabrics have a local origin and are compatible with the geological environment of the alluvial fan of the River Onyar (Fig. 8), where most of the exploited clay sources are located (Fig. 8:1). This Holocene sedimentary basin receives materials eroded from the metamorphic and igneous outcrops of the Gavarres Mountains (Fig. 8:2). Firing temperatures are estimated to be below 900 °C for the coarse MO-028 and MO-044 fabrics, based on the optical activity of the matrix (Quinn 2022, 266) and the degree of alteration of certain indicative minerals such as biotite (Quinn 2022, 269; Travé 2021). Samples from fabric MO-046 might have reached higher firing temperatures, as evidenced by the slightly vitrified areas in some cases and the patchy texture of the matrix. The maximum temperature commonly reached in traditional firings in Quart is 960 °C, at which point the vessels exhibit a characteristic egg-yolk colour that precedes the kiln’s final closure prior to oxygen reduction and cooling (Rocas, Vicens, and Basart 2016, 74).

FIG 8 Location of clay sources in Quart (1) and lithological map of the area (2). Background ortophotographs provided by the Cartographic Institute of Catalonia (Institut Cartogràfic de Catalunya, ICC).

ENDURANCE AND CHANGE: GUILD REGULATIONS AND COMMUNITIES OF PRACTICE

According to the obtained results, what we observe in Quart is an increasing complexity of vessel shapes from the 16th to 18th centuries, moving towards standardized modes of production that were tightly regulated by the potters’ guild. While there was considerable transformation in vessel shapes, leading to an increasingly more varied and specific repertoire, a striking feature of this production was the long endurance of paste recipes that did not follow the same patterns of evolution as the vessel shapes did. Since the late Middle Ages, the introduction of lead-glazed redware vessels for cooking purposes forced greyware production to adapt in order to survive, and farming and storage activities have required most of the greyware products since then.

Although greyware vessels were no longer used for cooking, their paste recipes remained unchanged until the nineteenth century, as was revealed by our petrographic analysis. At that time, their paste progressively changed to a fine, non-tempered fabric fired at a higher temperature, which improved the vessels’ strength rather than their toughness. Hence, the technological change came long after the new vessel functions and shapes were well established. However, cossis are an exception to this change, as the coarser pastes were necessary for effective shaping.

The continuity of paste recipes and the strong standardization of shapes probably had to do with the regulations of the guild, which was conservative with regard to the tradition and the potting practices within the town. The requirement to pass an exam to work as a potter in Quart may have discouraged potters from pursuing innovative ideas:

Ítem que no sia licit ni permès a ningún oller ni altre persona obrar ni fer obrar en ses cases obra de terra en lo dit lloch i parròchia de Quart que primer no sie examinat per los dits pabordres ensemps ab quatre altres ollers de dita confraria, i admès per aquells o la major part d’ells, si serà trobat hàbil i sufficient i aço a pena de trenta sous per quiscun i quiscuna vegada que serà trobat fer lo contrari. (ADG: LT 2, 4)

[Item, it is not allowed or permitted for any potter or other person to produce or order others to produce earthenware at their houses in the said place and parish of Quart; unless they are first examined by the said churchwardens together with four other potters of the said guild. They need to be approved by them or the majority of them, if they are found to be capable and sufficient, under penalty of thirty solidi for each and every time that they are found to be doing the opposite.]

Indeed, the regulations of the potter’s guild and their update in 1816 provided a good source for investigating the interconnection between people. This economic activity fostered a strong sense of belonging within those professionals in the town, and solidarity was one of the main reasons why potters defended their interests collectively. The additional regulations introduced in 1816 extended the obligations of potters even further, regarding their collective welfare:

Que cuando muera algun maestro de la citada cofradía con hijos que no se hallen en estado de poder trabajar y desempeñar su oficio o que dexe a su mujer sola o sin hijos de desempeño, que todos los demás maestros hayan de trabajar una fornada o concluir la que haya dexado empezada el difunto, contribuyendo en ello con igualdad a fin que la viuda o hijos del mismo logren este beneficio para ayuda de su subsistencia. (ADG: LT 11b, 3)

[When any master of the aforementioned brotherhood dies leaving children who are not able to work and perform his trade, or who leaves his wife alone or without performing children, all the other masters shall have to work a firing or finish the one that the deceased had started. They will contribute equally to it so that the widow or children may benefit from it to help with their subsistence.]

There is an interesting archaeological element to assess the enduring nature of these pots, not only in their production process but also in their useful life. There is frequent evidence of the repairing of larger vessels, particularly cossis, once they had broken or even when they were in danger of breaking. Many of the samples analysed show evidence of having been repaired with staples (Fig. 9:1) or metallic wires, which are occasionally visible in some vessels (Fig. 9:2–4). In fact, the reuse of vessels, especially cossis, for different purposes than the originally intended is a common feature of the vessels from Quart. Frequently, cossis were embedded in the very structure of buildings and became part of kitchen or farming structures (Fig. 9:6). Alternatively, broken vessels of all kinds were used as building material and are still visible today in many walls throughout the Gavarres area. Occasionally, the particular features of pinched coils allow for the identification of cossis from Quart that have been recycled in this way (Fig. 9:7).

FIG 9 Repairs and endurance of cossis from Quart in the Gavarres area. Archaeological samples with remains of staples (1) and metal wire repairs (2), metal wires visible in cossis still in use (3) or abandoned (4), cossi still kept in Can Punton house (5), embedding of a cossi within the structure of a building (6), and fragments of greyware pottery recycled within the walls of a building in Girona (7).

The long-lasting and durable nature of these vessels and their price meant that they frequently appeared in inventories and wills as part of the family properties that were passed down through generations. An example of this can be seen in the cossi mentioned in the inventory made in 1839 at Can Punton (Verges, Girona, Spain), a house belonging to the family of the Catalan writer Caterina Albert. This cossi is still kept within the abandoned building, which is now publicly owned (Fig. 9:5):

En lo rebost. Primo: una pila de posar oli. Ytem, vuit pots de vidra. Ytem, un cosi de posar murcadas, usat. (AHG: Notarial, Verges, 195, 3-4; transcribed in Silvestre 2022, 359-363)

[In the pantry. First, an oil storage container. Item, eight glass jars. Item, a cosi for storing olive pressing waste, used.]

FURTHER DISCUSSION AND CONCLUDING REMARKS. TOWARDS A NEW PERSPECTIVE OF TRADITION

The Gavarres area is known for its ceramic production, which has become part of its unique character. Until the 1980s, pottery production was a prosperous industry in Quart and La Bisbal d’Empordà. However, almost two generations and forty years later, the latest families of potters are at a crossroads and in need of supportive policies to continue this ancestral occupation or, at least, to vindicate its value as cultural heritage. Nevertheless, some younger potters are strongly committed to continuing this tradition, and various educational and dissemination strategies have been implemented for the general public. Local books and journals have also been published to safeguard this cultural value (Guerrero 1988; Rocas, Vicens, and Basart 2016; Rocas 2017; Santanach, Rosal, and Suñol 1998). In certain areas, such as La Bisbal d’Empordà, diversification strategies have been employed to preserve this activity and ensure its future. This has led to a focus on the artistic value of pottery products, rather than their functionality. Furthermore, the sustainability of pottery as an eco-friendly substitute for plastic and other environmentally harmful materials suggests a second life for these objects.

Changes in the production of pottery in Quart appear to be related to industrialization and the mechanization of pottery wheels during the nineteenth century. Although these technical improvements coincided with changes in the pottery paste used in Quart, they do not appear to be the sole reason for the changes observed. It is possible that the abandonment of ancient clay quarries also played a role in the change in the nature of the pastes used. Despite these changes, contemporary potters in Quart place great emphasis on the traditional hand-made character of their products and see their pottery as a symbol of identity. This idea is reinforced by the continuity in the particular traditional typology established over time. Nevertheless, the transition from coarser to finer pastes means that today’s Quart products are practically indistinguishable from other greyware with a strong traditional component, such as the production of La Bisbal d’Empordà, unless analysed chemically. In the present day, the characterization of pottery from la Bisbal is yet to come. As part of a forthcoming project dedicated to post medieval pottery in a wider area in Catalonia this analysis will be performed.

As archaeologists examining material culture from the post-medieval, modern, and industrial periods, we find several challenges that need addressing in the coming years to better understand the phenomenon of pottery making in Catalonia. One of these challenges is the need for a combined approach between material and non-material heritage from a territorial perspective, allowing exploration of the values of craftsmanship within an industrial world. Another challenge is exploring the connections between the development of the capitalist mode of production, the mechanization of certain tasks, and the desire to preserve ancestral manufacturing techniques and their impact on artisanal products. This topic is worth exploring in order to explain anthropological processes of identity building or the universal character of local or regional elements. The vindication of ‘localness’ and ‘authenticity’ and the relationship of greyware products with an increasing fashion for ‘folk art’ (Trzeciecki 2018, 160) has parallels in other parts of Spain, such as the greyware products from Faro (Asturias) (Ibáñez de Aldecoa 1987) or Quintana Redonda (Soria) (Azcárraga and Rodríguez-Limón 1978), as well as in places as far away as Poland (Trzeciecki 2018).

These questions provide opportunities for debate and future research development in the field of archaeology, particularly when combining the analysis of material culture with ethnoarchaeological approaches. Studying these traditions in our own western countries encourages us to rethink the idea of otherness when artisans and analysts are members of the same communities (Augé 1994, 1996, 45–50). It is important to foster the active role of artisans and help them become subjects who master their self-knowledge, challenging concepts such as backwardness when applied to artisanal manufacturing. This is especially relevant now, as technical advances and technological modernization must adjust to the increasing need for sustainability in industrial production. These challenges provide exciting opportunities for further research in this field.

Abbreviations
ADG	=	Arxiu Diocesà de Girona (Diocesan Archive of Girona)
AHG	=	Arxiu Històric de Girona (Historical Archive of Girona)
LT	=	Llibre dels Terrissers (Book of Potters)

Acknowledgements

Research introduced in this contribution is part of the GREYWARE research project (GREYWARE: Transformaciones en el uso de cerámica utilitaria de cocción reductora: una aproximación diacrónica y social a los modos de producción y consumo [PID2019-103896RJ-I00]). This project is part of the ‘Medieval and Postmedieval Pottery and Production Processes’ research line developed at the Medieval and Post-medieval Archaeology Research Group (2021 SGR 00236 GRC) at the University of Barcelona. The corresponding author wishes to thank Ms Noemí Travé for language edition and review. The collaboration of Ms Elena Vicens in this piece of research is acknowledged as well.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was funded by the Spanish Ministerio de Ciencia e Innovación (MCIN/AEI/10.13039/501100011033).