Introduction
Across the region once ruled by the Roman Empire, one of the most ubiquitous and readily identifiable artifacts unearthed by archaeologists is the red gloss ceramic known today as terra sigillata (sometimes still referred to as “Samian ware” in Britain).Footnote 1 While, traditionally, scholars have looked, in particular, at questions related to the production and exchange of terra sigillata, more recently, an ever-growing corpus of studies have investigated its use in diverse contexts across the Roman provinces.Footnote 2 Although it is clear that different social groups could use terra sigillata in different ways across the empire, relatively few studies have investigated the greater role that the introduction of terra sigillata may have had in terms of larger social transformations over time following the Roman conquest. In this regard, the region of Mediterranean Gaul is particularly interesting for understanding the social uses of terra sigillata compared with earlier ceramics, as the indigenous peoples of the region had already begun importing large quantities of fineware for local use prior to the Roman conquest, in this case, black gloss ceramics mainly produced in the Italian peninsula.Footnote 3 This article examines the differential uses of various types of fineware, ranging from earlier black gloss ware to later terra sigillata, as found both in discard assemblages within settlements and in the funerary record in the Roman province of Gallia Narbonensis (modern Mediterranean France). The analysis focuses specifically on the region of eastern Languedoc, especially the Gallic settlement of Lattara (modern Lattes). Between ca. 200 and 50 BCE, black gloss vessels appear to have been used extensively in daily life and were frequently discarded, while they also played an important role in large-scale ritual meals held at the site of certain tombs in the late 2nd c. BCE and early 1st c. BCE. Major transformations in social practices involving fineware ceramics then occurred after ca. 50 BCE, whereby fineware vessels were discarded far less frequently and appeared in fewer numbers in tombs. By the late 1st c. BCE, sigillata vessels appear to have played a much reduced role in funerary rituals held at tombs compared with earlier black gloss vessels. In addition, the appearance of smaller Gallic terra sigillata vessels in the 1st c. CE seems to signal a shift to more individualized patterns of dining in daily life and to reinforce a new importance for the individual in dining and other facets of social life that first emerges toward the end of the 1st c. BCE.
Investigating terra sigillata consumption
Easily identifiable by its shiny red gloss, terra sigillata is found in large quantities at sites throughout the Roman Empire and even beyond its borders, in places such as central Europe (Fig. 1).Footnote 4 Although earlier finewares such as Italic black gloss did circulate in considerable numbers across parts of the Mediterranean, the scale and distribution of terra sigillata appears to have been largely unprecedented in the ancient world.Footnote 5 Scholars estimate that, at their height in the later 1st c. CE, the workshops at La Graufesenque alone were producing around 15 million sigillata vessels per year.Footnote 6 This large-scale production of terra sigillata in workshops throughout the provinces of the empire, especially in Gaul, seems to have been enabled by significant social transformations, linked to new forms of social organization, property relations, and economic investment.Footnote 7
Fineware ceramics mentioned in the text. (©DocProtoMidi.)
At the same time, scholars have also pointed out that the appearance of terra sigillata represents significant changes for provincial societies not only in terms of economic production and exchange, but in consumer practices as well.Footnote 8 Greg Woolf, most notably, has referred to this development as part of a veritable “consumer revolution” for provincial societies, arguing that mass-produced and widely circulated Roman material culture, including sigillata, was not just “richer and more various but … also ordered by a new regime of values and tastes.”Footnote 9 Accordingly, an increasing number of studies have investigated the use of terra sigillata in local contexts across the empire and its implications for understanding local lifeways.Footnote 10
From these studies, it is apparent that local provincial societies generally drew upon a fairly limited number of terra sigillata vessel forms (Fig. 2), reflecting the types produced in large numbers at the workshops in southern Gaul and elsewhere, but that discrete local practices are often discernible behind this standardized façade.Footnote 11 The majority of these consumption studies, however, have been largely synchronic in scope, with relatively little diachronic analysis of the ways in which the appearance of terra sigillata vessels was related to larger social transformations. In a recent important work on the active role of terra sigillata in effecting historical events, Astrid Van Oyen draws attention to the tendency of scholarly work on terra sigillata to avoid engaging with the larger social implications that its appearance may have had, arguing that, “discussions of [terra sigillata] consumption patterns have little to say about issues of cultural change, which have been hard to ignore in the provincial Roman archaeology of the last decades.”Footnote 12 Somewhat related to this, many studies have tended to isolate an investigation of local uses of terra sigillata from a broader analysis of the uses of ceramic assemblages as a whole, which encompass not just finewares but also the more common cooking and storage vessels often found in association with sigillata in archaeological contexts.Footnote 13
The most common forms for black gloss (above) and Gallic sigillata (below) mentioned in the text. (Drawings by the author.)
An important exception to these trends is the work of Martin Pitts, who has identified, in particular, a notable shift in consumption patterns in Britain from the Late Iron Age through the period of Roman rule in terms of ceramics use, including sigillata.Footnote 14 In particular, Pitts has noted a decline in more communal forms of consumption, focusing on locally made drinking vessels, and a corresponding greater emphasis on more individualized dining, involving sigillata dishes for eating, something that first became evident in military and colonial contexts in Britain and then became more generalized during the Flavian Period.Footnote 15 As mentioned earlier, however, unlike in Britain, black gloss fineware, imported mainly from Italy, had already circulated widely in Mediterranean Gaul in the 2nd and 1st c. BCE, and it becomes important in this case to identify whether patterns of use in black gloss vessels were similar or different to those for later sigillata vessels. Disentangling these issues has important implications for understanding both the impact of the Roman conquest on dining practices – as opposed to more locally driven transformations – and the potential agentive role that ceramics such as terra sigillata, as standardized and widely circulated objects, may have had in transforming provincial societies. In particular, this latter point touches upon a long-standing debate about the potential for objects to “act” on humans (so-called material agency), albeit without the intentionality and reflexivity associated with human agency.Footnote 16
Comparing ceramic uses in different contexts
In order to identify potential differences in the usage of black gloss and sigillata vessels, this article focuses on two important archaeological contexts: everyday use, in which ceramics were employed over time and then discarded; and funerary contexts, in which ceramic vessels were utilized in rituals and then left in the tomb. For the discard assemblages, unless otherwise stated, calculations of the frequency of wares and types are based upon the number of identified rims, with amphorae and dolia excluded from the counts. Rim sherds from the same vessel are taken to represent only one vessel. Although the use of estimated vessel equivalents (EVEs) to calculate ceramic frequencies has become the norm in the United Kingdom, few, if any, excavations in France regularly record ceramics in those terms.Footnote 17 As such, I have relied on the standard norm for the region to calculate the minimum number of vessels.
Here, it is important to note that although discerning differences between the uses of fineware vessels and other ceramics in domestic and funerary contexts can be an important means of identifying local consumer choices and the impact that the appearance of terra sigillata vessels had on daily lifeways, a comparison of the two contexts is not always straightforward. In particular, the means by which ceramic vessels came to be deposited in the archaeological record are notably different. In the case of grave goods found in tombs, people made active choices about what to place with the deceased. By contrast, ceramics discovered in discard assemblages – generally speaking, the most frequent context in which archaeologists find ceramics – reflect the gradual accumulation of broken pottery that was thrown away after use over the course of generations. Ethnoarchaeological studies have identified a number of different ways that vessels are broken, although more focus has often been placed on cooking-ware vessels, which, because of how they are used, tend to break more frequently than fineware.Footnote 18 These reasons for breakage can include: people accidentally dropping vessels while using or transporting them, vessels wearing down and breaking after repeated use (especially cooking vessels), animals knocking over vessels, children playing with vessels and breaking them, vessels being set down too hard, etc.Footnote 19 In addition, studies in historical archaeology of ceramic assemblages thrown into privies have suggested that ceramic assemblages were often discarded as a lot at the end of the life cycle of a building, often in association with changes in the people using the building, and by extension, the ceramics as well.Footnote 20
In some cases, pottery in the ancient world was likely thrown away in trash middens or deposits relatively near where it was used.Footnote 21 In other cases, it is possible that trash eventually accumulated in larger deposits outside a settlement, but could then be brought back into the site as part of fill for building projects.Footnote 22 In this way, the frequencies of specific ceramic wares and vessel forms in discard contexts are not a perfect snapshot of the exact numbers of different vessels that would have been present in a specific household at any given time.Footnote 23 In particular, vessels that break more easily and/or are used more frequently, with more wear on them (such as cooking pots) generally come to be overrepresented in ceramic assemblages, particularly over longer periods of time.Footnote 24 Although ethnoarchaeological examples are seldom a perfect match for Roman Mediterranean France – in part because they lack large quantities of fineware – it is nonetheless clear that fineware vessels utilized for more ceremonial purposes in ethnoarchaeological contexts tended to have a much longer use life, sometimes decades long. A survey of the time lag between the manufacture of historical-period glazed earthenwares and their actual deposition in the archaeological record suggests that fineware ceramics could be used for 15–25 years (or about a generation) before being thrown out.Footnote 25 By contrast, ethnoarchaeological studies suggest that cooking vessels tend to be discarded after shorter periods of use, sometimes as a little as one year.Footnote 26 In a study of time lag associated with sigillata vessels found in tombs from Roman Britain, Colin Wallace notes that in some cases of southern Gallic terra sigillata, the vessels could be used for 30 to 60 years before being deposited in a tomb; for sigillata from central and eastern Gaul, this time lag could be as much as a century and a half.Footnote 27
Background to ancient Mediterranean Gaul
The ceramic data used for this study are drawn from the modern region of eastern Languedoc, today in the French administrative région of Occitanie (Fig. 3). This article draws especially upon the ceramic data from the Iron Age and Roman-period settlement at Lattara and its associated Roman-period necropolis, which were selected due to the author’s familiarity with the site, as well as the large quantity of ceramic data available for analysis (Fig. 4). This work builds upon an earlier analysis by the author of changing patterns in black ware and terra sigillata consumption at Lattara from the discard assemblages only. These results have been updated from more recent excavations, and the important funerary context has been added to the analysis of fineware use.Footnote 28
Map of Iron Age and Roman Mediterranean Gaul indicating the location of sites mentioned in the text. (Map by the author.)
Map of the excavations of Lattara (Lattes, France). The numbers indicate excavation zones from which the discard ceramic assemblage data are derived. (Map modified by author, courtesy of Lattes excavations.)
During the Iron Age (ca. 750–125 BCE), the Gallic peoples of the region began to concentrate in densely occupied, fortified settlements.Footnote 29 Lattara, which was founded around 500 BCE, was a typical indigenous settlement for the time, and throughout its occupation history, it was both a thriving settlement, with a population of at least several thousand people, and an important port of trade.Footnote 30 During the Iron Age, the settlement consisted of a series of closely spaced, elongated blocks of rooms separated by narrow streets. These rooms, which could be joined internally into units of up to three or four rooms, were used for various social activities, including sleeping, socializing, artisanal activities, and food preparation and cooking.Footnote 31 This style of housing continued at Lattara until the end of the 1st c. BCE. Ceramic fragments are often found scattered along and in the dirt floors of rooms, as well as in the thicker layers of rubble laid down when the houses were periodically remodeled. These modifications to the rooms often involved tearing down some or all of the mudbrick walls and raising the floor level of the room using this rubble, or possibly rubble brought in from elsewhere at the site.
Following the initial Roman conquest of the region between 125 and 121 BCE, the importance of the port continued. The core area of the Iron Age settlement, however, was gradually abandoned during the course of the 1st c. CE, such that occupation was much less dense by the second half of that century, although the port continued to be used until sometime in the 3rd c. CE.Footnote 32 The ceramics discarded from the period ca. 25 BCE–75 CE at the site come from rubble layers associated with the reorganization and eventual abandonment of the core area of the settlement in the second half of the 1st c. CE.
To date, no Iron Age necropolis has been discovered at Lattara, although an important Roman-period necropolis has been found that was in use from the end of the 1st c. BCE to the end of the 1st c. CE. The necropolis was uncovered in the 1960s, not far from the settlement, on the other side of the eastern branch of the Lez River (Fig. 4).Footnote 33 Since no comprehensive data exist from Iron Age funerary contexts at Lattara, data for the use of black gloss vessels in a funerary context from the 2nd to the 1st c. BCE come from isolated burials in the countryside and concentrations of burials around the ancient settlement of Nemausus (modern Nîmes) and Ugernum (Beaucaire) (Fig. 3).Footnote 34 A necropolis containing 12 tombs spanning the period from the early 1st c. BCE to the early 1st c. CE has also been discovered at the site of Le Paradis at Aramon.Footnote 35
Italic black gloss at Gallic settlements and cemeteries: 225–50 BCE
Although black gloss ceramic vessels – mainly in the form of bowls – were present before the end of the 3rd c. BCE at Lattara, as was the case elsewhere in the region, they were far less frequent than they would become in the 2nd–1st c. BCE. Given the high numbers of large, locally made, non-wheel-thrown bowls in the discard assemblages, it is likely that these black gloss vessels were used less frequently, perhaps in more ritual contexts. By the second half of the 2nd c. BCE, however, black gloss fineware, especially Campanian A, became quite common at Lattara and elsewhere in the region, with vessels appearing in discard assemblages as well as in funerary contexts. Interestingly, however, there are clear differences between the two contexts in terms of the use and function of these vessels. Whereas multi-functional, medium-sized black gloss bowls predominated in the discard assemblages, larger black gloss bowls and platters were favored in tombs, playing an important role in what appear to have been on-site funerary meals. This practice of commemorating the dead through funerary meals at the tomb site occurs most frequently in the period associated with the conquest and pacification of Mediterranean Gaul, between ca. 125 and 70 BCE, with these funerary practices drastically giving way to new ones in the second half of the 1st c. BCE.
For the period 225–125 BCE, Campanian A comprised approximately 35% (n = 1,872) of the ceramic discard assemblages from Lattara (based upon the number of identifiable rims, excluding amphorae and dolia), with black gloss finewares making up about 38% (n = 2,016) overall (Fig. 5). For the period between 125 and 50 BCE, this percentage of black gloss increased to 46% (n = 2,227) for Campanian A and 48% (n = 2,312) overall. Between 225 and 125 BCE, the most common form in discard contexts within the settlement at Lattara was the bowl CAMP-A 27a-b, which comprised approximately 20% (n = 1,030) of all identifiable rims. A recent residue analysis from a later tomb, dating to the beginning of the 1st c. BCE, from eastern Languedoc (Saint-Pastour at Vergèze) revealed chemical traces of wine in two similar black gloss forms of bowls (CAMP-A 27c and CAMP-A 28ab), suggesting the use of bowls like this in ritual feasting involving wine.Footnote 36 It should be noted, however, that traces of animal fat were found in the vessels as well, indicating that they could also have been used for food (whether solid or liquid) and suggesting a polyvalent function for these types of medium-sized bowls.
Frequency by percentage for discard assemblages at Lattara (above) and for the funerary context for eastern Languedoc (below) amongst all ceramics (excluding amphorae and dolia). The numbers above the bars represent total number of identifiable rims (for the discard assemblages) and total number of vessels (for the funerary context). (Prepared by the author.)
A much smaller number of other forms were present as well, including the smaller bowl 28ab, the larger bowl 33b, the two-handled drinking cup 42Bc and the relatively deep dish 36 (Fig. 6). Perhaps as a result of the increase in the availability of black gloss bowls, both wheel-thrown beige ware bowls and locally made non-wheel-thrown bowls appear to have declined in importance by this time. Although non-wheel-thrown bowls still comprised 14% (n = 744) of all identifiable rims at Lattara, this was down from 27% (n = 972) for the period 300–225 BCE. By the end of the 2nd c. BCE, an increasingly diverse range of Campanian A vessel forms were appearing in significant numbers at Lattara, including several types of larger bowls (especially 27Bb, 33b, and 31b) and the dish/shallow bowl 36 (Fig. 7). By now, non-wheel-thrown bowls were far less prominent in the assemblages, making up only around 5% (n = 222) of ceramics for the period 125–50 BCE. However, non-wheel-thrown cooking pots continued to be important.
Table indicating the most common forms for the period 225–125 BCE by percentage amongst all ceramics (excluding amphorae and dolia). The numbers in parentheses represent the total number of identifiable rims (for the discard assemblages) and total number of vessels (for the funerary context). (Photos ©DocProtoMidi.)
Table indicating the most common forms for the period 125–50 BCE by percentage amongst all ceramics (excluding amphorae and dolia). The numbers in parentheses represent the total number of identifiable rims (for the discard assemblages) and total number of vessels (for the funerary context). (Photos ©DocProtoMidi.)
Especially given the decline in importance of the larger non-wheel-thrown bowls and the increase in numbers of large black gloss bowls and plates – as opposed to just smaller bowls – it appears that Campanian A ware, in particular, became, by the end of the 2nd c. BCE, the main table ware for serving and consuming food. It should also be noted that, during this time period, it seems – based upon the distribution of black gloss vessels throughout the settlement – that families and social groups living at the site all had relatively equal access to black gloss.Footnote 37 In all this, it appears that Lattara was fairly typical for the region, with similar trends occurring across eastern Languedoc, although the status of Lattara as a key trading port likely resulted in somewhat greater access to imported black gloss than for settlements in the hinterland.Footnote 38
Shifting from the discard ceramic assemblages at Lattara to the funerary context, it should first be noted that for the period 225–50 BCE, few cemeteries associated with indigenous sites have been uncovered for Mediterranean Gaul.Footnote 39 This curious absence in the archaeological record may in part be due to relative invisibility; if the cemeteries were as unostentatious as those from the 4th and 3rd c. BCE, such as the examples at Sizen-Vigne and Ambrussum, they may simply be extremely difficult to detect.Footnote 40 However, for the period ca. 225–125 BCE, archaeologists have discovered five isolated cremation burials in eastern Languedoc that now contained, for the first time, intact black gloss vessels among the grave offerings interred with the deceased.Footnote 41 These five tombs each contained between four and nine ceramic vessels overall, and between two and five Campanian A vessels. Although the sample size is small, there is a trend toward favoring bowls over dishes, although two of the tombs contained one or two CAMP-36 dishes. Given how frequently the form CAMP-A 27a-b appears in the discard context from Lattara (representing 20% overall of the ceramic assemblages for this period), in the funerary context it appears to be no more frequent than any other form, with only one example found. In fact, the small cup CAMP-A 28ab appears more frequently (three instances in the five different tombs), even though it appears far less often in the discard contexts from Lattara, representing only about 2% (n = 88) of the ceramic assemblage for this period. It is worth noting that this relative paucity of CAMP-A 28ab in the discard context compared with CAMP-A 27a-b, with almost all vessels represented by rims rather than intact forms, is not due to issues of identification; on the contrary, the form CAMP-A 28ab has one of the more distinctive rims from amongst the Campanian A vessels.
This distinction in the preference for certain black gloss vessel forms in a funerary context continues for the period 125–50 BCE. Archaeologists have identified over 25 tombs for this period, especially in the vicinity of the Gallic settlements of Nemausus (modern Nîmes) and in three concentrations of tombs from Beaucaire at the sites of Colombes, Marronniers, and Sizen (Fig. 8).Footnote 42 One notable tomb was also found in the vicinity of Lattara at La Céreirède (SP4002), dating to ca. 100–75 BCE.Footnote 43 A striking aspect of all these tombs is the significant number of beige ware pitchers, with all but three of these tombs containing at least one pitcher and sometimes as many as six, the average being around three pitchers per tomb; this represents a far more frequent occurrence than in the discard assemblages. The pitchers likely indicate that some kind of beverage was served, probably either wine and/or beer or mead, at a funerary meal. Supporting this suggestion, in seven instances, the tombs also contained at least one (and sometimes two) Italic wine amphora, although this does not necessarily mean that for those tombs with no wine amphora such vessels were not present at the funerary meal. Indeed, the prominence of pitchers and drinking bowls for all the tombs of this period make it quite likely that wine or some other alcoholic beverage was served in all instances. Also related to the consumption of beverages, with only one exception (Tomb 17 from Marronniers at Beaucaire), every tomb contained at least one and up to six black gloss bowls (with the average being two) that were smaller than 16 cm in diameter and realistically could have been used for drinking liquids.
Tomb from the site of Mas de Jallon, near Beaucaire, ca. 100–75 BCE, with grave goods on the right (excluding amphorae). (Drawings by the author, photo ©DocProtoMidi.)
In regard to black gloss vessels, an immediate difference in comparison with the discard context at Lattara for the same period is a greater presence of large dishes and platters, most noticeably forms CAMP-A 5, 5/7, 6, and 36. In contrast to the discard contexts, where these forms made up 12% (n = 559) of the overall ceramic assemblage, large black gloss dishes/platters comprised 21% (n = 64) of the aggregate ceramic assemblages for all tombs from the period 125–50 BCE. The ceramic assemblages, along with faunal remains in the tomb, seem to point to the importance of some kind of funerary meal for the deceased.Footnote 44 Interestingly, the ashes of the deceased seem to have been less of a focus in the arrangement of the tomb than the ceramic vessels; in fact, during this period, generally only a symbolic handful of the ashes was scattered across the tomb.Footnote 45 This is particularly noticeable in the early 1st-c. BCE tomb from La Céreirède at Lattara. Here, a large stone chest – which likely attests to the high social importance of the individual involved – was placed prominently in the tomb. However, it contained only a symbolic sprinkling of the ashes of the deceased, along with scattered faunal remains and three drinking goblets.Footnote 46
Throughout the 2nd c. BCE and into the 1st c. BCE, then, black gloss vessels appear to have played an important role in both quotidian and more ritual consumption practices. Although discard and funerary contexts contained similar forms of black gloss, black gloss bowls of the type CAMP-A 27a-b were more often represented in discard assemblages than in tombs, suggesting that they were likely used more frequently for daily practices and thus had a higher breakage rate, perhaps because of their versatility as a vessel that could be used for both liquids and solid food. By contrast, larger dishes and platters seem to have played a more important role in serving food for funerary meals held at the grave site and appear to have been broken less frequently in daily life, suggesting perhaps a more restricted, specialized use for them. This practice of holding on-site funerary meals, implicating relatively large numbers of people, appears to correspond to a period of disruption in Mediterranean Gaul, beginning with the Roman conquest of 125–121 BCE and continuing with the suppression of numerous revolts, including a revolt of the Volcae Arecomici of eastern Languedoc in the 70s BCE. Perhaps not coincidentally, many (although not all) of these tombs contained weapons, and it is certainly possible that the exceptional rites conducted at these tombs, involving the use of black gloss platters and large bowls for serving food – something that seemingly was not bestowed on all deceased members of the community – was reserved for those with a special status, including perhaps warriors who fell in battle against the Romans.Footnote 47
The end of black gloss ceramics: 50 BCE–25 CE
In the second half of the 1st c. BCE, there appears to be a notable disruption of the finewares present in Mediterranean Gaul. In part, this seeming decline in the amount of fineware present is related to the end of black gloss production by the midpoint of the 1st c. BCE. More generally, however, it perhaps also reflects growing economic stagnation or even impoverishment of local populations following the suppression of revolts in the first half of the century. There were also significant changes in funerary practices by the end of the 1st c. BCE in terms of rites, with the end of large on-site feasts for certain celebrated members of the community and, instead, the use of a much smaller number of ceramic vessels to hold small food offerings for the deceased. As part of this change, weapons were no longer placed in tombs.
Given that the ceramics found in the occupation layers of the settlement reflect the accumulation of broken and discarded pottery vessels over generations of use, it is perhaps unsurprising that even after 50 BCE – when workshops in the Italian peninsula were ceasing the production of black gloss and had begun experimenting with other wares – black gloss still appears in large numbers in the floor and rubble layers at Lattara. In fact, for occupation layers of the settlement as a whole for the period 50 BCE–25 CE, Campanian A still comprised approximately 21% (n = 366) of all identifiable rims at Lattara, illustrating the time lag associated with the eventual deposition of ceramics into the archaeological record.Footnote 48 In addition, the disturbance of earlier stratigraphic levels by extensive building projects within the settlement by the end of the 1st c. BCE likely contributed to the presence of higher quantities of vestigial black gloss ceramics in latter occupation layers.Footnote 49
Turning to the funerary record, for the period ca. 60–40 BCE, the general preference for larger serving bowls and dishes remained similar to what we saw for the first half of the 1st c. BCE. These larger dishes and serving bowls were paired with between one and several drinking bowls (most prominently, the two-handled bowl CAMP-B 127). A particularly good example is Tomb SP347 from the Mas des Abeilles at Nîmes, which is the latest tomb for the period 60–40 BCE in the region of eastern Languedoc to show the importance of an on-site funerary meal through a relatively high number (seven in this case) of large dishes and serving bowls, along with several drinking bowls (two examples of CAMP-B 127), now paired with two thin-walled drinking goblets (a new development), three smaller bowls that could have served different functions, and six large pitchers.
However, in the second half of the 1st c. BCE, funerary practices involving fineware ceramics begin to change, involving a certain simplification of funerary rituals.Footnote 50 Only two tombs after ca. 40 BCE contained Campanian A vessels: Tomb 18 from Marronniers and Tomb 1 from Sizen, both just outside the settlement of Ugernum and both dating to ca. 25–1 BCE (based upon the presence of Italic sigillata in the two tombs).Footnote 51 In place of Campanian A in tombs, several new types of fineware appeared, including Campanian C vessels, locally made imitations of black gloss vessels (Céramique dérivée de la campanienne A et C), locally made imitations of Italic sigillata vessels (so-called presigillata), and, eventually, Italic sigillata after ca. 25 BCE (Fig. 9).Footnote 52 All of this seemingly suggests a situation in which Campanian A vessels were no longer widely available and locals were instead using alternatives, although some older black gloss vessels may still have been passed down for a generation or two until the end of the 1st c. BCE.
Table indicating the most common forms for the period 50 BCE–25 CE by percentage amongst all ceramics (excluding amphorae and dolia). The numbers in parentheses represent the total number of identifiable rims (for the discard assemblages) and total number of vessels (for the funerary context). (Photos ©DocProtoMidi).
Overall, for the second half of the 1st c. BCE, there were now far fewer ceramic vessels in the tombs throughout eastern Languedoc: 5.35 ceramic vessels on average for the period 50 BCE–25 CE, compared with 12.4 ceramic vessels for the period 125–50 BCE. In addition, especially after ca. 25 BCE, there is a decrease in the average diameter of plates and bowls compared with that for the period 125–50 BCE (Fig. 10). This decrease in the average diameter of vessels in funerary contexts for this period in part reflects the smaller size of the sigillata vessels compared with earlier wares (with a few exceptions, such as the three platters found in Tomb 4 from Marronniers at Beaucaire, each of which measured 39 cm in diameter).Footnote 53 There is an especially notable decline in the number of dishes measuring between 24 and 26 cm, which were replaced by smaller dishes of between 15 and 17 cm.Footnote 54 Italic terra sigillata begins to appear in the tombs after ca. 25 BCE, but never in numbers as large as for the black gloss, comprising around 15% (n = 15) of all the ceramic vessels found in tombs dating to ca. 25 BCE –25 CE (Fig. 5). Likewise, unlike the earlier tombs from ca. 125–50 BCE, which could contain up to 23 Campanian A vessels (on average 4.76 Campanian A vessels per tomb), the tombs from ca. 50 BCE–25 CE never contained more than three Italic terra sigillata vessels (on average 1.5 Italic terra sigillata vessels per tomb). As with discard contexts, smaller dishes and cups were common, and no examples of molded Italic sigillata vessels have been found in tombs.
Histogram showing the changing diameters of vessels from funerary contexts. (Prepared by the author.)
These trends are particularly noticeable at the one late 1st-c. BCE necropolis known for eastern Languedoc, at the site of Le Paradis at Aramon, near Beaucaire. Here, archaeologists discovered 12 tombs dating between 75 BCE and 25 CE. By 25 BCE, there was a decline in the number of ceramic vessels placed in the tomb, with the disappearance of large serving bowls and dishes, as well as coarse ware vessels for preparing food, suggesting an end to food being served to celebrants at the site of the tomb. Instead, each tomb generally contained one small dish, as well as a drinking goblet and small cup.Footnote 55 These smaller vessels often had faunal remains associated with them, suggesting food offerings to the dead, but overall, there seems to have been less emphasis on an actual large-scale funerary feast at the site of the tomb than had previously been the case. However, in contrast to later tombs in the region for the 1st c. CE, the tombs still contained only a symbolic handful of the deceased’s ashes.Footnote 56
Overall, what is striking is not so much the continued prevalence of Campanian A in the discard assemblages at Lattara and elsewhere, but the fact that, unlike in the tombs, Campanian C and local black gloss imitations do not appear frequently in the discard assemblages from the period 50 BCE–25 CE. For the occupation levels of the period 50 BCE–25 CE, local imitations of Campanian A and C, as well as actual Campanian C ware, only comprised around 2.5% (n = 42) of all identifiable rims. For the period 25–100 CE, when we would expect these wares to start entering the archaeological record after a period of use, they comprise less than 1% (n = 6) of all identifiable rims, suggesting they had limited impact on daily use at Lattara.Footnote 57 In the specific case of a dirt floor layer from the end of the 1st c. BCE, in Room 6c, dating to after ca. 25 BCE, from the ongoing excavations of the Roman port at Lattara, amongst all identifiable rims (n = 37) from the floor, there was only one example of a CAMP-C rim (or 2.7% of the assemblage, roughly equivalent to the overall percentage for all discard assemblages at Lattara), compared with four CAMP-A rims. Only five sherds of local imitations of black gloss and two sherds of presigillata were found, compared with 13 total sherds for CAMP-A (out of a total of 159 sherds from the unit of stratigraphy). The majority of vessels from this room consisted of cooking vessels (either wheel-thrown or non-wheel-thrown), suggesting that those were far more frequently broken in daily life than fineware. Although Campanian C vessels and local imitations of black gloss were being placed in tombs for this period, they do not seem to have been broken and discarded in the same way as Campanian A had been previously, and it is possible that local peoples had more restricted access to finewares after ca. 50 BCE than in earlier periods.
At the same time, Italic terra sigillata only comprises 1.8% (n = 32) of all identifiable rims in the discard assemblages at Lattara for the period 50 BCE–25 CE and then 3.5% (n = 32) for the period 25–100 CE. Nonetheless, it is notable that the appearance of regionally produced wheel-thrown cooking ceramics does make a much more noticeable impact on the discard assemblages at Lattara, largely replacing locally made non-wheel-thrown wares by the end of the 1st c. BCE. While a relatively wide range of different forms of small plates and cups in Italic sigillata are present at Lattara, no single form is typical. Furthermore, very few examples of molded Italic sigillata or more expensive platters have been found, suggesting perhaps a situation in which the local population was unable to afford large quantities of expensive fineware forms. Interestingly, large sigillata dishes and platters were evidently available in at least parts of Roman Mediterranean Gaul, based upon an important deposit of Italic sigillata vessels and other ceramics from the site of Le Tassigny in the provincial capital of Narbo Martius and dating to the end of the 1st c. BCE.Footnote 58 Apparently thrown out in one lot, the deposit contained the remains of 37 large Italic sigillata platters measuring between 24 and 42 cm in diameter, with an average diameter of 32.6 cm. In addition to 48 Italic sigillata bowls with diameters of between 8 and 16 cm, there were also another 37 smaller Italic sigillata dishes measuring between 13 and 18 cm in diameter, with an average diameter of 16.1 cm. If large sigillata platters such as those from Le Tassigny were present at Lattara, however, they do not appear in the archaeological record, suggesting, at most, infrequent use in daily life. Likewise, with a few exceptions, such as Tomb 4 from Marronniers and an early 1st-c. CE tomb from the necropolis at Lattara, those who participated in the funerary rites apparently did not see a need for large platters – or lacked the financial means to purchase these costlier vessels – favoring smaller dishes, instead, for small servings of food. In all this, then, the decline in the quantity and frequency of use of fineware ceramics in both discard assemblages and funerary contexts for the period ca. 50 BCE–25 CE represents significant changes in the practices of local peoples.
The rise of Gallic terra sigillata: 25–75 CE
The 1st c. CE saw a continuation of the trends that began after ca. 50 BCE, with fineware less common in the discard assemblages at Lattara and of lesser importance in funerary contexts, where small numbers of fineware vessels were used for food offerings for the deceased and, above all, for ritual libations at the tomb. In addition, the continuing trend in the reduced size of sigillata vessels in the 1st c. CE compared with earlier black gloss vessels likely created a greater emphasis on individual dining practices in daily life.
Although Gallic terra sigillata – mainly from the workshops at La Graufesenque – does appear in significant numbers in the discard assemblages from Lattara for the 1st c. CE, comprising as it does just under 20% (n = 178) of all identifiable rims, it never reached the numbers of Campanian A vessels at the height of their popularity, when black gloss wares overall comprised around half (49%) of all identifiable rims for the period 125–50 BCE (excluding amphorae and dolia). This appears to be fairly typical for, if not somewhat higher than, the average percentage of Gallic terra sigillata for the region in the first half of the 1st c. CE.Footnote 59 The most common type of Gallic sigillata found in the 1st-c. CE discard assemblages from Lattara was the small cup Drag. 24/25 (24% of all identifiable sigillata forms, n = 42), followed by several types of smaller dishes, especially Drag. 15/17 and Drag. 18 (with dishes overall comprising 33% of the sigillata assemblages, n = 59), and the cup Drag. 27 (12%, n = 22) (Figs. 11 and 12), generally mirroring many other sites in the region and across the western Roman Empire.Footnote 60 The greater proportion overall of cups compared with dishes in the discard assemblages from Lattara seems to reflect a greater demand for cups across Gaul, as opposed to a greater use of dishes in southern Spain and Italy.Footnote 61 Molded decorated forms, especially the molded bowl Drag. 29 (11%, n = 19), were present but not particularly numerous, something that is also typical across the western Roman Empire before ca. 60 CE.Footnote 62
Table indicating the most common forms for the period 25–75 CE by percentage amongst all ceramics (excluding amphorae and dolia). The numbers in parentheses represent the total number of identifiable rims (for the discard assemblages) and total number of vessels (for the funerary context). (Photos ©DocProtoMidi.)
Overall frequency by percentage of certain Gallic sigillata forms amongst all Gallic sigillata. (Prepared by the author.)
The necropolis at Lattara, which was in use from the end of the 1st c. BCE to the second half of the 1st c. CE, exhibits a markedly different use of terra sigillata vessels compared with both contemporary discard contexts and the use of black gloss vessels in the 2nd- and early 1st-c. BCE tombs. In contrast to those earlier tombs, almost all the examples at Lattara contained a cinerary urn for holding all the ashes of the deceased (rather than a small handful, as was previously the case), seemingly indicating a greater symbolic value placed upon the deceased at the tomb than what was seen before. Indeed, this is a trend that is noticeable not just at Lattara but throughout the region for the 1st c. CE.Footnote 63 The 30 tombstones recovered from the necropolis show a mixture of local Celtic names (at least five instances among the deceased listed) and Latin names (of which at least 13 apparently belonged to Roman citizens), suggesting an important indigenous population at Lattara in the 1st c. CE – a small percentage of whom gained Roman citizenship – alongside a smaller number of Italic settlers.Footnote 64 It is also important to note that, here, the ceramic vessels used for analysis of changing funerary practices at Lattara were all secondary grave goods, placed in the tomb along with the ashes of the deceased, rather than being put on the funeral pyre first.
Of the 169 total tombs recorded from the necropolis, 155 contained at least one ceramic vessel. Similar to before, for the period 50 BCE–25 CE, there were relatively fewer ceramic vessels in the tombs than for the period 125–50 BCE, with an average of 4.8 ceramic vessels per tomb. Of the 155 tombs containing ceramics, a little fewer than half (46%, n = 71) contained sigillata (mainly from southern Gaul, although a few forms were Italic). Amongst all ceramic vessels from the necropolis, Gallic sigillata only made up around 13.2%, which is significantly lower than its occurrence in discard contexts from Lattara for the 1st c. CE. By contrast, around 75% (n = 117) of tombs with ceramics contained at least one beige ware pitcher. The vast majority of tombs also contained a wheel-thrown coarse ware cooking pot that functioned as the cinerary urn, and a mortar and lamp.
Furthermore, when sigillata did appear in tombs, it was generally present in only small quantities. One tomb (Tomb NL3-T6) from the second half of the 1st c. CE did contain eight different Gallic sigillata vessels: three Drag. 27b cups, one Drag. 33a2 bowl, three dishes (one Ri1 and two Drag. 18b), and a mortar (Ri12). This tomb, however, seems to be an interesting exception; otherwise, the tombs that contained sigillata vessels never contained more than four of them, for an overall average of 1.6 sigillata vessels per tomb. Amongst sigillata forms, dishes (especially Drag. 15, 17, and 18) seem to be overrepresented in the necropolis compared with the discard contexts at Lattara, with approximately half (53%, n = 47) of all sigillata vessels in the necropolis being dishes, versus approximately 33% for all discard contexts at Lattara (Fig. 12). After dishes, the next most common forms in the necropolis were the small cups Drag. 24/25 (14% of all sigillata forms, n = 12) and Drag. 27 (11%, n = 10), as was similar in the discard contexts. The bowl Drag. 33 seems to be underrepresented in the necropolis compared with the discard contexts, while the cup Ri8 is somewhat overrepresented in the necropolis (Fig. 12). As with the discard contexts at Lattara, molded vessels were quite rare, with only four instances of them found in the necropolis. The lack of decorated bowls is also typical for cemeteries elsewhere in the western Roman Empire, implying that they were not used in funerary rituals associated with tombs.Footnote 65
Although dishes were the most common sigillata vessel form present in the tombs, no more than three dishes were ever placed in the same tomb. In 45% of the instances in which a tomb contained at least one sigillata dish, a thin-walled drinking goblet was also present. This trend of a paired plate and drinking cup was also noted for the earlier necropolis of Aramon. In four cases, the small cup Drag. 24/25 appeared with a sigillata dish, but interestingly, in only one instance – that of the rather unusual tomb NL3-T6 mentioned earlier – did the cup Drag. 27 appear with sigillata dishes. Furthermore, Drag. 27 cups (unlike the generally smaller Drag. 24/25) were never paired with a thin-walled drinking goblet, which may perhaps suggest that, at least in the context of the tombs, these Drag. 27 cups were meant to hold a beverage.Footnote 66
Following a trend that began with the appearance of Italic sigillata vessels in the last quarter of the 1st c. BCE, for the 1st c. CE, the sigillata vessels were on average much smaller in diameter than black gloss vessels from the 2nd–1st c. BCE (Fig. 10). Although there is clear evidence that potters produced sets of sigillata vessels in a range of standardized sizes, with these sizes fluctuating over time, overall, it is apparent that sigillata vessels tended to be smaller on average than earlier fineware vessels.Footnote 67 Although one Italic sigillata plate found in a tomb did measure 29 cm in diameter, the average diameter for sigillata dishes in the necropolis was 16 cm, compared with 25.25 cm for the earlier black gloss plates and bowls. Along with this trend of smaller vessels, the fact that sigillata appeared in smaller quantities in the tombs, compared with the earlier black gloss, suggests perhaps that sigillata was not used for a large, on-site funerary meal to honor the deceased (Fig. 13). Although food remains, including animal bones, were found in association with sigillata vessels in the tombs, this could represent a symbolic meal, or offerings to the dead, rather than the remnants of a larger meal shared amongst the living.Footnote 68 Cooking vessels (with the exception of the urn used as a repository for the ashes of the deceased) were relatively rare for the necropolis, in contrast to the tombs from the late 2nd and early 1st c. BCE. Indeed, of the ceramic vessels that occur more frequently in tombs than in discard contexts, all seem to be associated with drinking or pouring liquids. These overrepresented vessels in tombs include beige ware pitchers and thin-walled drinking goblets. In addition, if mortars were used for grinding and preparing spices and herbs for mixing with alcoholic beverages such as wine – as has been suggested by researchers – their overrepresentation in tombs could also be connected with liquids.Footnote 69 In 1st-c. CE Roman Mediterranean Gaul more generally, the overall trend seems to be that families held a funerary meal at the site of the pyre, followed by rituals involving libations at the tomb itself.Footnote 70
Typical tomb contents from the Roman-period necropolis at Lattara with funerary inscription. (©DocProtoMidi.)
Compared with the ubiquity of black gloss in both domestic and funerary contexts for the 2nd and early 1st c. BCE, the lower frequency of sigillata from the 1st-c. CE necropolis at Lattara is interesting. Overall, the presence or absence of sigillata in the tombs at Lattara seems to be linked to whether the choice was made to place food in the tomb, something that appears to have occurred only about half the time. There is also some evidence – although the sample size makes it difficult to comment definitively – that sigillata was somewhat more correlated with more ostentatious goods in the tomb, as well as seemingly occurring more frequently in tombs that could be identified as female on the basis of either the tomb inscription or the identified biological sex of the deceased individual’s remains.
In any case, the trends noted for the necropolis at Lattara seem to be typical for the region. For instance, at the 1st-c. CE necropolis at Navitau, just downhill from the settlement of Sextantio (only about 9 km north of Lattara), archaeologists found six tombs, of which three contained one or two sigillata vessels.Footnote 71 A somewhat different picture, however, emerges at the Roman-period necropolis from Valladas at Saint-Paul-Trois-Châteaux (the ancient settlement of Augusta Tricastinorum), around 110 km northeast of Lattara but still in Mediterranean Gaul.Footnote 72 As at Lattara, molded sigillata was strikingly absent from the tombs at Valladas, and the sigillata vessels tended to be in pairs comprising a plate and small cup (most often Drag. 24/25, but also Drag. 28, Ri8, and Ri9). However, sigillata vessels occurred in almost every tomb (20 out of 23) at Valladas that could be securely dated to the period 15–70 CE. Furthermore, within these tombs, sigillata vessels were far more numerous than at Lattara, with an average of 4.55 sigillata vessels per tomb compared with 1.6. The pairing of a sigillata plate with a cup thus often occurred multiple times in the same tomb (e.g., four plates and four cups). Similar trends, where graves contained suites of terra sigillata recurring forms, have been noted elsewhere in the western Roman Empire but slightly later, during the Flavian period, at important settlements such as Verulamium (St. Albans) and Ulpia Noviomagus (Nijmegen). Pitts has interpreted this as the signaling in tombs of elite dinner parties with standardized suites of terra sigillata vessels, and the examples from Lattara along with the cemetery at Valladas likely represent the early emergence of this ethos.Footnote 73 In this, it is probably significant that the settlement associated with the necropolis at Valladas – Augusta Tricastinorum – functioned as a local civitas capital, enjoying a higher social status than Lattara. Furthermore, this also appears to point to the relatively low economic status of those buried in the necropolis at Lattara, something that, more generally, seems to be indicated at the settlement overall by the low numbers of decorated terra sigillata forms and the absence of costly platters.Footnote 74
Discussion
Based upon all this, a picture emerges of a time between ca. 200 and 50 BCE during which large quantities of black gloss vessels were frequently used, broken, and thrown out in daily life. Within the same period, black gloss vessels – especially large dishes/platters and bowls – appear to have played an important role in funerary meals held at tomb sites for a select number of individuals in local Gallic society around the time of the Roman conquest and immediately following it. These larger black gloss dishes and platters do not appear nearly as frequently in the discard assemblages, suggesting perhaps that they were used less often (and thus broken less frequently) in daily life than more quotidian forms such as the smaller bowl CAMP-A 27a-b that could be utilized for both solid food and liquids. This period, 200–50 BCE, was followed by a time in the second half of the 1st c. BCE when fineware ceramics, including locally made black gloss imitations, Campanian C, and Italic terra sigillata, while present, seem to have been far less frequently discarded in daily life at Lattara, and also appeared in fewer numbers in the funerary context.
This rather dramatic disparity in the numbers of fineware vessels in the archaeological record before and after ca. 50 BCE arguably suggests that there were both differences in the quantity of fineware vessels available to local peoples and differences in terms of how those vessels were used in daily activities and eventually discarded. If the availability of fineware ceramics, whether imported or locally made, was much more restricted after ca. 50 BCE than access to earlier black gloss vessels between ca. 200 and 50 BCE, it is quite possible that local peoples used fineware more carefully and less frequently in the second half of the 1st c. BCE, leading to a significant reduction in the deposition of Campanian C, local black gloss imitations, and Italic sigillata into the archaeological record at settlements such as Lattara. In addition, studies have indicated that vessels often have a higher chance of breaking during feasting than in normal daily life (indeed, common sense suggests that ceramic vessels can break more readily when alcohol is involved).Footnote 75 In this light, the greater discard rates of Campanian A fineware may also be linked, in part, to its greater use in feasting practices. Certainly, there is general agreement among scholars that ritual feasting involving wine was an important social practice in late 2nd-c. and early 1st-c. BCE Gaul, and it is not coincidental that Campanian A vessels were imported into Gaul alongside large quantities of Italian wine amphorae.Footnote 76 The decline in fineware in the discard assemblages at Lattara after ca. 50 BCE and the corresponding decline in evidence of large-scale communal meals at the site of certain tombs in the region seem to support the trend noted by Martin Pitts for Roman Britain of an eventual move away from large-scale communal feasting to more individualized dining practices.Footnote 77
More generally, it is important to set these changes in finewares within the larger social context of Roman-occupied Mediterranean Gaul. The increasing use of black gloss vessels in the region came at a time after the Second Punic War when the Roman Republic was increasingly involved there, with a number of destruction layers present at indigenous settlements in western Languedoc and western Provence.Footnote 78 The use of black gloss for meals and larger feasts within Lattara and other settlements in the region could conceivably reflect a moment of increasing social stress in local society, with feasting perhaps being an important aspect of tying communities together, while at the same time offering possibilities for more intense social competition. In the decades following the initial Roman conquest of the region in 125–121 BCE, a number of large-scale revolts broke out amongst local peoples in reaction to oppressive policies of Roman taxation and grain levies, which at times appear even to have led to famine.Footnote 79 In eastern Languedoc specifically, there was a revolt among the Volcae Arecomici in the mid-70s BCE.Footnote 80 Following the suppression of these uprisings by the Roman armies, Roman provincial rulers often punished rebellious peoples by confiscating their land and turning it over to loyal Gauls or incoming Italic settlers. Archaeologically, at Lattara the period after 50 BCE was characterized by the gradual abandonment and tearing down of the traditional elongated blocks of houses in the settlement, as well as the dismantling of its ramparts.Footnote 81 The significant decline in the amount of fineware entering the archaeological record between ca. 50 and 1 BCE – something that is evident across the region – could conceivably reflect, in part, a certain level of economic stagnation and the eventual impoverishment of the local population at Lattara and elsewhere, a situation that would only improve in the first half of the 1st c. CE.
In any case, it was in the context of these disruptions in the use of finewares at Lattara that there appear to have been important changes in funerary practices and perhaps more banal dining practices as well. In the funerary context, fineware vessels in the second half of the 1st c. BCE at sites like Aramon no longer seem to have functioned as vessels for a large-scale meal shared by many people, but were rather utilized for smaller food offerings for the dead. Later, at Lattara in the 1st c. CE, sigillata vessels were not systematically present among the grave goods, unlike vessels for pouring and consuming liquids, suggesting that even the practice of leaving food offerings at the tomb was not always followed. In some cases, locals in the late 1st c. BCE may have attempted to carry on older practices with new fineware vessels, as suggested by Tomb 4 from Marroniers at Beaucaire (dated to 20–1 BCE), which contained three very large (39 cm in diameter) Italic sigillata plates (type 11.1).Footnote 82 Several interrelated factors likely contributed to this rather dramatic change in funerary practices, including the arrival of Italic settlers with different funerary customs and the emulation of those customs, at least to some extent, by local peoples, as well as the impoverishment of local populations and changing conceptions about the individual and society.
The small size of both the Italic and especially the Gallic sigillata forms available at Lattara also likely significantly restricted the practical possibilities for food consumption. The larger black gloss bowls and dishes could have been used for serving and eating both solid food and soups and stews. Moreover, their large size meant that they could hold enough food for serving from one to several people. By contrast, the smaller and shallower sigillata dishes are not as suitable for stews and can hold less food (Fig. 14). A comparative, ethnoarchaeological study of vessel size and use has concluded that, generally speaking, across the world, individualized dining vessels tend to measure between 10 and 23 cm in diameter (with an average of 14 cm), whereas family dining vessels that are shared amongst several individuals measure between 8.4 and 95 cm in diameter (with an average of 24.6 cm).Footnote 83 The smaller sigillata dishes (with an average diameter in the Lattara necropolis of 16 cm) correspond well with individualized dining vessels on this scale, whereas the black gloss dishes and platters (e.g., CAMP-A 5, 6, 5/7, and 36) and larger bowls (of the type CAMP-A 27Bb) (with an average diameter in funerary contexts for black gloss dishes/platters of 25.25 cm) generally fall within the range of family vessels.Footnote 84 Based upon experience of dining with various reconstructions of black gloss and sigillata vessels, the author can attest that an average-size sigillata Drag. 18 dish can hold a sufficient (if somewhat on the small side) meal for a grown adult, while the black gloss dish CAMP-A 36 can hold enough food to serve several adults (realistically, two or three) (Fig. 14). The most commonly represented sigillata forms, the small cups Drag. 24/25 and Drag. 27, which Dannell associates with the paradixi and acetabula, respectively, mentioned on firing lists from La Graufesenque, are seemingly best suited for small side dishes or sauces, but certainly not the main meal.Footnote 85 Perhaps the most versatile Gallic sigillata vessels is the bowl Drag. 33, which like its predecessor, the black gloss bowl CAMP-A 27a-b, could conceivably have been used for drinking liquids and for eating both solid food and stews, and it is notable that the form Drag. 33 appears less often in the funerary record.Footnote 86
Modern reproduction by the author of black gloss and Gallic sigillata, showing relative size and possible uses of the vessels. ø = diameter; h = height.
Overall, the prevalence of small Gallic terra sigillata dishes and cups thus seems to indicate a shift to more individualized patterns of dining, in which vessels could hold small portions of food or sauces, with food perhaps served from large, wheel-thrown coarse ware bowls and pots. Certainly, this new emphasis on individualized dining practices has been noted by scholars for other areas of the Roman Empire as well.Footnote 87 Interestingly, the emphasis on individual servings seems to be conceptually mirrored in a new emphasis, in tombs of the 1st c. CE, on the deceased individual, based upon the new importance placed on collecting the individual’s ashes in a cinerary urn that was now much more centrally located within the tomb.Footnote 88 The social disruption associated with the final repression of indigenous resistance to Roman rule may have led to a breakdown of traditional social groups and the emergence of more individualized ways of being – a development that is certainly seen in colonial situations more generally across time and space.Footnote 89 This shift is also evident in the transformations in houses at Lattara and in the surrounding region at the end of the 1st c. BCE, with the disappearance of elongated blocks of rooms housing multiple families and the likely emergence of new social units living in courtyard houses.Footnote 90
Conclusion
As mentioned, relatively few studies have investigated the active role that the dramatic appearance of Gallic terra sigillata across western Europe may have had, if any, in the social practices of local provincial societies compared with earlier practices going back to the pre-Roman Iron Age. Certainly, the mere presence of terra sigillata is a poor proxy for social transformations, whereas an identification of changes in reoccurring associations of certain types of ceramic vessels is far more informative. Taking this approach, however, the evidence presented here suggests that, at least in the case of Mediterranean Gaul, the widespread appearance of Gallic terra sigillata in fact played little active role in transforming provincial societies. On the contrary, the major social transformations evident in the archaeological record and related to fineware uses occurred between ca. 50 and 1 BCE, when only relatively small quantities of Italic terra sigillata were present at indigenous settlements. These fineware-related changes included shifts in their breakage and deposition into the archaeological record, perhaps related to diminished access to finewares brought on by economic stagnation and impoverishment of local populations in the wake of Roman suppression of indigenous revolts. Furthermore, there were dramatic changes in funerary practices during this time, involving a move away from large-scale communal feasts at the grave site and a new emphasis on employing only a small number of fineware vessels containing food offerings for the deceased. Although the influx of Italic settlers into Mediterranean Gaul may have influenced the changing social practices, given the continued presence of indigenous populations at settlements like Lattara, much of the impetus for these changes likely came from social transformations within local societies as they adapted to colonial rule. The small size of Gallic sigillata dishes and cups likely played an important role in structuring and reifying in material daily practices the new social order that was already emerging in the conquered province by the late 1st c. BCE, several generations before the widespread appearance of Gallic sigillata. This new social order of Roman Gaul in the 1st c. CE seems to have placed a much greater emphasis on the individual, both in daily practices such as food consumption and in funerary rituals. In all this, although the physical characteristics of material objects like ceramics certainly influenced the practical possibilities of certain social actions (e.g., small vessels meant smaller portions of food that could be practically consumed at one time), it was above all the recurring sets of social relations and interactions of people – unfolding in specific historical contexts – that engendered distinct patterns of material remains in the archaeological record. In all this, ceramic fineware – far more than a mere chronological marker – reveals important insights into the changing lived experiences of local peoples.
Lastly, this article has also hopefully demonstrated the usefulness of comparing ceramic data from discard contexts and funerary assemblages, despite the obvious differences by which the archaeological record was formed in each instance. Indeed, regardless of these differences, it is striking that the relative prevalence of any type of fineware vessel appears to be similar in both the discard and funerary contexts. Thus, when Campanian A was frequent in the discard assemblages, it appeared more frequently in the funerary record as well, and when sigillata appeared less frequently in discard assemblages, it also appeared less frequently in the tombs. Of course, more work is still needed – perhaps through the creative use of ethnoarchaeology involving reproductions of black gloss and sigillata – to understand more critically under what circumstances fineware actually came to be discarded and how often. In all this, the results presented here also hopefully underscore the need, as always, for the contextual analysis of ceramic data that takes into account the complex site formation processes that could have led to their creation.
Acknowledgments
I would like to thank the three anonymous reviewers for their very helpful and insightful comments and critiques. I am also grateful to my colleague in anthropology at Gettysburg College, Kirby Farah, who read a draft of the article and offered helpful feedback. I am also grateful for the help and insights – and all the conversations about ceramics – from Derek Rosenberry, previously with Gettysburg College and now a Preparator at The Mint Museum in Charlotte, North Carolina, who graciously helped to fire the reconstructed ceramics used in Fig. 14.
Open access
