Once upon a Glass—Cycles, Recycles and Reuses of a Never-Ending Material

Abstract

Glass can be considered a locus of meaning, a material which has been the repository of traditional knowledge and technological expertise for at least three millennia. The history of glass speaks of know-how, technological transitions, and contaminations of recipes for its manufacture, which have changed across the world over the centuries. As the amount of recovered glass from archaeological contexts is much lower compared to ceramic and metal finds, research has often considered glass as a rare material. Furthermore, glass production, in ancient times as in the present day, requires the use of selected raw materials and noticeable amounts of fuel, making reuse and recycling practices necessary to foster sustainability, from both an economical and an environmental perspective. Latin authors, such as Juvenal and Martial, reported buyers of broken glass in Imperial Rome, presumably destined for recycling. Archaeometry has also provided data that allow, today, to clarify different aspects related to production cycles, uses and reuses of a material that, starting from the Roman age, became as common as modern plastics. From beakers and goblets reused with different purposes to mosaic tesserae detached for making new mosaics or to be refused and employed as “pigments” for colouring glass, this paper aims to provide an overview of reuse and recycling practices of ancient glass through a discussion of selected case studies from Roman to Middle Ages, showing how the cycle of this material can be framed as an actual example of sustainable circular economy in the past.

1. Introduction

Following decades of public awareness and information campaigns, glass is now considered a highly reusable and recyclable material: it does not emit potentially harmful substances to humans and the environment, it does not absorb odours and it does not alter the flavour of its contents. But when did it all start? The sustainability of glass dates back a long time before the birth of the ecological awareness that today guides our decisions in terms of recycling and reuse. The reasons behind these practices have mainly been related to practical issues, linked to the immediate availability and greater cost-effectiveness of the material to be re-fused. The process of recycling, however, can have been impacted by a variety of economic and/or social factors in the past. Because modern recycling is largely based on economic concepts of value, global trade and energy expenditure, these concepts are frequently applied incorrectly to the past.

Despite the recent increase in papers dealing with recycling in the past [1,2,3,4,5], there is little discussion of the motivating factors for and the implications of recycling [6]. As Sainsbury and colleagues pointed out, the recycling process should be viewed more appropriately as a contribution of meaning to objects and materials, involving four important parameters in the analysis: time, form, function and property [7]. With its modern baggage, the word “recycling” frequently causes misunderstandings in the ancient interpretation, as recycling is quite often conceived as a task related solely to economic and environmental concerns. Given these considerations, it may be more appropriate to refer to “mutability” when talking about recycling in the past, a concept which has recently been introduced by Sainsbury and colleagues to describe “a more useful umbrella term for a wide range of activities involving changes to both form and substance” [7]. In line with the definitions proposed by Sainsbury and co-workers, this paper will use the term “recycling” when the original object has been completely remade through a melting step, and none of the original form remains. The term “reuse” will be employed to highlight a change in the purpose or cultural significance of the object.

The suitability of glass for recycling was widely exploited in the past, and research has been conducted to attest to the various ways in which recycling can be detected in the archaeological record, with a focus on the first millennium AD [2]. It is widely recognised that the invention of blowing, between the 1st century BC and the 1st century AD, marked an epochal change in glass manufacturing and use. Glass changed from an aristocratic material used to make pieces not intended for widespread consumption into a material similar to plastics today, as costs and manufacturing time were reduced and glass objects became commonplace items [8]. Following the introduction of the glass-blowing technique, a decrease in the price of glass on the market occurred due to the faster and easier production of vessels [8,9,10,11]. Glass was mass-produced at a small number of specialised glassmaking sites, which supplied numerous secondary workshops where raw and broken glass was re-melted and shaped. The act of recycling was an integral part of the chaîne opératoire inside these workshops, to be carefully considered to fully understand the range of glass forms, colours and compositions found in archaeological records [2,5,12,13,14,15,16]. The re-introduction of glass waste and cullet into the production cycle is due to the material’s unique properties, which allow it to remelt at lower temperatures than those required for the production of glass from raw materials. Therefore, even small kilns could be used to obtain new material to be reworked with significant time and cost savings compared to primary production. Furthermore, the recovery of objects reused with functions different from those for which they were created, as well as the ancient literary references to the practices of glass repair, demonstrate that the “life” of the product did not end when it was damaged or broken, but it continued afterwards as well. The concept is not extraneous to the archaeological field, where the notion of the biography of objects was discussed for the first time by Kopytoff [17]. Kopytoff suggested that, when studying objects, it is important to consider not only the original purpose for which an artefact was made, but also the various ways that it may have been used throughout its lifetime, and the various meanings that may have been attached to it culturally during this period. Further development of this idea is provided by [18], reflecting upon the need to apply the concept of use-life to objects from our material culture, for a more comprehensive understanding of the relationship between things and men. This concept has so far been studied especially in the field of ancient metallurgy [19,20].

This paper, based on a contribution presented at the 2022—AIAr (Associazione Italiana di Archeometria) Thematic Conference “Sustainability in Cultural Heritage” (29th June–1st July 2022, Padua, Italy), aims to highlight how the mutability of objects from the past can be discussed as a sustainability-related issue in the context of ancient productions, with a focus on glass between the Roman and Middle Ages.

2. Materials and Methods

No previously unpublished materials were studied or analysed for this paper. The article revisits data from glass assemblages that have already been studied and published, to frame them in the ongoing discussion about the sustainability of ancient productions. Therefore, all details on the selected materials for each case study covered in the article as well as for analytical protocol and the instrumental conditions can be found in the reference papers mentioned in the text.

3. Results and Discussion

Archaeological evidence and archaeometry have shown that, in the Roman age, the majority of raw glass was produced along the Levantine coast and in Northern Egypt [5,21,22,23,24]. As the cost of transport was significant, recycling and the secondary production of glass into smaller, localised workshops—where raw glass was eventually mixed with colourants/decolourants/opacifiers and then shaped into objects—was economically convenient. Broken glass had its own economic value in ancient times, as attested by written sources. Allusions to the trading of broken glass for sulphur to be used as an adhesive can be found, for instance, in Martial, Statius and Juvenal [25,26,27].

Archaeological evidence also supports the trade of raw glass in cullets. One of the most well-known cases is the merchant ship Iulia Felix, which went down off the coast of Grado in the first half of the third century AD: a wooden barrel was found in the ship’s hold and contained about 140 kg of glass fragments from plates, cups and bottles [28,29]. The Iulia Felix’s cargo shows a peculiarity: glass is mostly colourless and naturally coloured, suggesting that even the choice of broken glass was very accurate and, presumably, intended to minimise issues with unwanted colouring during the re-fusion of the glass. Another intriguing fact is revealed by the archaeometric data collected on a variety of glass finds from the cargo [13,30]. Plots in Figure 1 highlight a clear distinction between the chemical compositions of the antimony-decoloured glass and the naturally coloured glass; antimony can be used to create colourless glass because it was a decolouring agent used exclusively in Roman times, so its concentration can be used to create a “recycling index” for glass [30]. Archaeometric data demonstrated that the Iulia Felix’s cargo contained recycled glass which was made with 54% antimony-decoloured glass, indicating that recycling has likely been practised for some time. Glass cullets to be recycled have also been recovered in a number of excavations, both in urban contexts and near military camps [2,3,27,31].

Along with recycling, the reuse of glass objects is attested in Roman times. Referring to the definition given above, reuse occurs when the object loses its initial purpose. From this perspective, cases such as that of the jug recovered in Usk (Wales), whose damaged rim was cut and smoothed without affecting its original usage, is not to be intended as an example of reuse [32]. Differently, cases such as the bottoms of tableware cups and bowls carefully cut out, like attested in the necropolis of Zadar, Croatia, and reused as lids for other containers [33,34], are examples of reuses.

Following the widespread use of glass in the Roman world, the current state of research indicates that it is likely that glass experienced its greater mutability between the late antique and the medieval periods, in terms of both recycling and reuse with the loss of the original function of the objects. The case of Nogara (province of Verona, North-eastern Italy), for instance, has provided valuable insights into the complexities of the glass industry between the 10th and 11th centuries AD [35]. Archaeometric data provided evidence for the occurrence of glass recycling; samples with intermediate chemical compositions between natron and soda plant ash glass were identified, indicating a proclivity for recycling earlier glass, and suggesting a gradual change from natron-based to soda ash-based production technology (Figure 2).

Figure 2. Samples from Nogara with an intermediate natron soda ash as flux, demonstrating the recycling of glass (following [35], modified). The samples from Nogara are subdivided by type: biconical cup (), goblet disc foot () beaker edge with decoration (), beaker edge without decoration (), beaker side (), waste glass and mosaic tesserae (), bottle ring foot (). Mean chemical compositions and standard deviations of main compositional natron groups, identified in literature, also shown: “Roman” glass (); Levantine I (), Levantine II (), Egypt I (), Egypt II () and HIMT () (data from [22,23,36,37,38,39,40]). Squared areas: compositional areas of glasses of compositional Group 2.2 of [23]; Carolingian glass [41]; 8th–12th-century natron glass from various Italian sites [42,43,44,45,46,47] and those obtained with soda plant ash from Grado, Vicenza [48] and Venice [49].

Figure 2. Samples from Nogara with an intermediate natron soda ash as flux, demonstrating the recycling of glass (following [35], modified). The samples from Nogara are subdivided by type: biconical cup (), goblet disc foot () beaker edge with decoration (), beaker edge without decoration (), beaker side (), waste glass and mosaic tesserae (), bottle ring foot (). Mean chemical compositions and standard deviations of main compositional natron groups, identified in literature, also shown: “Roman” glass (); Levantine I (), Levantine II (), Egypt I (), Egypt II () and HIMT () (data from [22,23,36,37,38,39,40]). Squared areas: compositional areas of glasses of compositional Group 2.2 of [23]; Carolingian glass [41]; 8th–12th-century natron glass from various Italian sites [42,43,44,45,46,47] and those obtained with soda plant ash from Grado, Vicenza [48] and Venice [49].

Another type of recycling was also attested in Nogara: SEM and LA-ICP-MS analyses demonstrated that blue and reticello decorations found on drinking beakers were obtained by melting earlier mosaic glass tesserae, as shown by the amounts of Co, Cu, Sn, Sb and Pb and the presence of calcium antimonate crystals in the matrix (Figure 3).

Mosaic glass tesserae are a fascinating case study in the recycling and reuse of glass in antiquity. The use of small cubes made of coloured opaque glass is attested since the late Hellenistic period, with specific reference to 2nd–1st century BC multi-coloured bowls to obtain a distinctive three-dimensional decorative pattern [50]. However, coloured and opaque glass reaches its maximum expression in the making of mosaics for architectural decorations. Mosaic tesserae are made of a vitreous matrix within which crystalline phases act as opacifiers and pigments are dispersed [51]. Several cases documented in the literature [52,53,54,55,56,57,58,59,60] demonstrate how this particular category of glass artefacts has experimented with a variety of recycling and, more importantly, reuse practices. The first and most obvious instance of the recycling of mosaic tesserae is their removal from structures devoid of their original ornamentation for the purpose of reusing them in new apparatuses. There is archaeological and archaeometric evidence in favour of this widespread practice. The study of the mosaics located in the Sacellum of Saint Prosdocimus (basilica of S. Giustina, Padua, Italy) is an example [54,55,56]. The decorative mosaic’s construction site dates back to the 6th century AD; however, an analysis of about 200 tesserae revealed that not all of them were made in the same historical period. Some tesserae, in fact, have a vitreous matrix and opacifying agents that are typical of the Roman era and have simply been reused in their original state. Other tesserae, on the other hand, have a late-antique vitreous matrix but Roman-era opacifiers and are, thus, recycled. Others are “new” tesserae, made ad hoc for the Sacellum, a very complex situation that nevertheless brings out a sort of “sustainable” use of glass. This practice of reusing glass mosaic tesserae has been documented in other geographical areas and historical periods as well [57,61,62]. Recent studies attest to the reuse of tesserae “taken” from pre-existing monuments in the making of the mosaic decorations of the monuments of Syria’s first Islamic Caliphate, a practice described by textual sources and partly confirmed by archaeometric analyses. Compositional analyses revealed the presence of both Levantine and Egyptian compositional categories: Apollonia-type and Foy-2 base glass indicate a continuity with the production of mosaic glass tesserae in the late antique Levant; Egypt I glass clearly distinguishes Umayyad tesserae, providing tangible evidence of other legacies [63]. A recent study on a prominent assemblage of glass tesserae from the Great Mosque of Damascus has further demonstrated Egyptian glass’s dominance among the coloured tesserae, with other base glass types in circulation prior to the 8th century [64]. It is difficult to say whether these tesserae were simply reused or if the glass was recycled and transformed into tesserae during construction. Parallels can be drawn with 8th-century glass vessels, where there is a significant decline in Apollonia-type Levantine I, while Foy 2.1 is missing; it appears more likely that the tesserae represent reused material collected from older mosaics or some storage facilities [63]. Al-Maqdisī, al-Ya’qūbī, ibn Zabāla, al-Dinawārī and ibn Rusta report mosaic cubes sent from the Byzantine emperor to Umayyad caliphs, and al-Tabarī also states that the emperor ordered searches for these cubes in ruined cities [63,65]. It should also be noted that the rise of the Umayyad caliphate occurred during a period when the production of mosaics appears to have declined significantly in the Mediterranean basin: between the 7th and 8th centuries, when the number of new buildings adorned with mosaics dropped from more than 50 to 20 and then 9 exemplars [66]. It is possible that the decline in mosaic production resulted in a decrease in demand for tesserae, and thus the use of recovered materials from existing, abandoned monuments was encouraged. In San Vincenzo al Volturno (Isernia, Southern Italy), a case of recycling of opaque coloured tesserae is documented [14]. Combined archaeological and archaeometric studies have provided sound evidence of specific procedures related to colouring window glass. The Monastic Complex of San Vincenzo al Volturno is among the most representative cases. Built in the early 8th century on a late Roman villa, the monastery underwent significant expansion in the 9th century thanks to Abbot Joshua. It became one of the largest monastic complexes in Carolingian Europe, with the installation of temporary workshops for making construction and materials. On October 881, the monastery was sacked and fired by the Saracens. Archaeological surveys underpinned evidence of dedicated glass workshops in the monastic complex: fragments of crucibles with glass adhering to the internal surface, processing waste and fragments of blow pipes have been unearthed. Moreover, a large amount of opaque coloured glass mosaic tiles and coloured and translucent window glass has been recovered. Data from archaeometric analyses demonstrated that glass tesserae from dismantled mosaics were recycled for making coloured window glass, according to the practice described by Theophilus Presbyter in De Diversis Artibus. In fact, the text makes specific mention of using mosaic tesserae as pigments for colouring windows. However, it remains unclear, at the current state of knowledge, how widespread this practice was. Unanswered questions regarding whether this practice implied specific cultural meanings, or was exclusively related to economic reasons and the unavailability of materials due to the inflection of commercial traffic that affected the Mediterranean basin between the 8th and 9th centuries [67]. The recycling of mosaic tesserae to be used as a pigment has recently been documented in the production of Viking glass beads in Ribe, Denmark. Here, evidence for the existence of a workshop established in the c. 700 AD has been underpinned, manufacturing monochrome and polychrome beads related to the previous tradition in Western Europe, but developing distinct designs [68]. After the medieval period, coloured glass continued to be used as a pigment. This is the case, for instance, with the smalt found in European paintings as early as the 15th century and used until about the 19th century [69]. Smalt is a finely ground potash-based glass containing cobalt as a colouring agent; conchoidal fracture fragments of translucent blue colour inside the pictorial layer allow us to identify this pigment from an archeometric perspective, as the example provided in Figure 4. Here, the detection of elements such as silicon, potassium and cobalt has been highlighted by spot measurements analyses performed using SEM-EDS, allowing these inclusions to be classified as smalt [69].

In the making of pictorial layers, colourless powdered glass was extensively used all over Europe during the 15th and 16th centuries [70]. The technological intention behind the addition of glass to paint layers seems to be linked to two key purposes: to facilitate the grinding of specific pigments—such as orpiment—and to accelerate the drying of oil paints, even if glass was not capable of performing this function. It is interesting to note that artists used whatever colourless glass was locally available, as preliminary compositional studies confirm the general geographical distribution found in studies of vessel and window glass. In Italian paintings, soda ash glass is almost exclusively used, whereas in German and Dutch paintings, soda ash glass, wood ash and wood ash–lime glass are present, similar to what has been discovered in archaeological sites [71].

4. Conclusions

The discussion of selected case studies from different geographical areas and historical periods provides further evidence of how recycling and reuse of glass were well-known and established practices in the ancient world.

The analysis of the case studies selected for this paper allows us to reflect on the need to start re-considering recycling in the past under new perspectives, going beyond technological purposes and the economics of material production and processing. The reuse of objects has been the subject of sporadic and not systematic investigations in the field of ancient glass studies, and more attention should be given to a better understanding of the underlying reasons. On the one hand, the reuse of objects or parts of them with a new function may find a logical explanation in the necessity to meet a practical need—as, for example, the bottoms of containers adapted to lids—the reuse and recycling of mosaic tesserae is, on the other hand, considerably more problematic in terms of interpretation. To the current state of knowledge, we do not know how widespread this practice was geographically or chronologically, or whether it had special ideological and/or symbolic meanings.

This paper intends to pave the way for new, more in-depth discussions on the just-introduced notion of “mutability”, challenging not only the utilitarian but also the symbolic justifications for the recycling and reuse of glass products in ancient times for a better understanding of the societies that used them.