Introduction
A line of scholarship within Roman archaeology has increasingly demonstrated how close attention to the details of small finds can expand our understanding of the ancient Mediterranean world and enrich how we write histories of this period.Footnote 1 Dubbed by one scholar as the “object turn,” this scholarship is one of the products of the wider “material turn.”Footnote 2 These studies show the value of developing histories that begin with and focus on small finds, their materiality, their design, how they were used, and how their practical features intersect with human actions and behaviors.Footnote 3 These theoretically informed object histories focus on and tease out the importance of what things do and are, not simply what they were supposed to represent.Footnote 4 Of particular interest for the present article is how scholars have demonstrated the usefulness of design theory, including the concept of “affordances,” the relationship between an object's properties and the capabilities of an individual that determine how the object could be used.Footnote 5 Examining an object's affordances, interpreted within their contexts, helps us see how design impacted usage by the ancients – including what might be called an object's primary or practical functions (e.g., lighting a lamp for illumination) and secondary uses that derive from those primary functions (lighting a lamp to mark a religious activity).Footnote 6 To illuminate the range of possible uses that are enabled and prompted by affordances, we require additional sources to reconstruct the contexts and particular forms of knowledge in which an object is embedded.Footnote 7 Especially helpful are sources that engage with small objects and are geographically and chronologically proximate to the archaeological finds in question. One such corpus of sources that has been overlooked in the recent wave of object-focused scholarship is rabbinic literature from the Roman era – legal, ethical, and exegetical writings in Hebrew and Aramaic from the first centuries CE.Footnote 8 Roman-era rabbinic literature includes seemingly countless discussions of objects, including their physical details, design, production, and uses.Footnote 9
This article seeks to contribute to object-centered scholarship in Roman archaeology by applying these approaches to finds from Roman Palestine. Most object-centered scholarship has focused on Europe, especially Roman Britain and Italy, with little if any application to archaeological finds from Roman Palestine. In choosing an object, ceramic oil lamps are particularly well suited for case studies for several reasons. They are often readily identifiable – even tiny fragments are frequently diagnostic and can be used to determine a lamp's type. We also have access to a sample of lamps that, in its representativeness, rivals that of any household wares, since a high proportion of lamp finds are published compared to other ceramic vessels. Moreover, with regard to the potential for understanding usage and context, lamps are mentioned frequently in written sources from the era. In addition to metaphorical discussions of light and lamps, there are also many references to usage, often in nitty-gritty detail, especially in early rabbinic texts. This massive corpus of writings, which includes discussions of hundreds of what we may call small finds or artifacts, has been under-utilized by Roman archaeologists in general. Likewise, few specialists in rabbinics engage with archaeology in a significant way, and even fewer move beyond art historical approaches.Footnote 10
This article focuses on the Northern Collar-Neck Lamp from Roman Palestine, bringing in rabbinic sources from the same time and place. It is argued that its design enabled and invited the placement of an additional object on top of its filling hole, as its high collar served as an affordance that invited or inclined one to add a supplemental reservoir, such as a pierced eggshell. This not only created a place for more oil, but also regulated the flow – allowing for only a slow drip so that the fuel would be replenished gradually, extending illumination time. We know of such practices from references in early rabbinic texts, which I argue reflect and engage with material practices that were more widely current in Roman Galilee during the first centuries CE. In making this argument, this article seeks to contribute to scholarship on object-focused histories in Roman archaeology by bringing these approaches to finds and other sources from Roman Palestine and demonstrating the usefulness of early rabbinic sources for understanding materiality, objects, and how objects were used.
Northern Collar-Neck Lamp: physical features
The Northern Collar-Neck Lamp (hereafter, “NCN”; also known as the Ginosar Lamp, Kinneret Type, Lamp with a Collar, Teapot Lamp, and Boot [magaf] Lamp), is shaped like a boot (“biconical” in cross section), with a loop strap handle and a rounded nozzle (Figs. 1–3). It was manufactured on a pottery wheel, usually from clay that is orange-brown in color, and typically measures about 9.1 cm long, 6.6 cm wide, and 4.4 cm high.Footnote 11 Its high collar, which encircles the filling hole and tapers inwards, accounts for about a third of the lamp's overall height.Footnote 12 The collar forms a hemispheric bowl with a large hole in its bottom that opens directly into the lamp's body, which serves as its main reservoir for oil.Footnote 13 The collar was not a later add-on to the lamp or a makeshift appendage, but was rather made on a wheel together with the rest of the lamp and was an integral part of the object's overall design.Footnote 14 The collar itself is similar to those of lamps made in Ephesus that circulated throughout the Mediterranean, including ancient Palestine.Footnote 15 The NCN's body most closely resembles that of the well-known local “Herodian Lamp” type (Fig. 4), which was also wheel-made, with the nozzle fashioned separately.Footnote 16 Notably, Herodian Lamps are frequently found in the same assemblages as NCNs (see below). While the Herodian Lamp has been studied extensively as a source of the material culture of Jewish populations in Roman Palestine, the NCN has received relatively little attention.Footnote 17
Fig. 1. Northern Collar-Neck Lamp from the Sea of Galilee boat, Ginosar. (Sussman Reference Sussman and Wachsmann1990, 98; photo courtesy of the Israel Antiquities Authority.)
Fig. 2. Northern Collar-Neck Lamp from Kfar Kanna. (Abu-Uqsa Reference Abu-Uqsa and Gal2002, 159.)
Fig. 3. Northern Collar-Neck Lamp from Caesarea Maritima. (Sussman Reference Sussman and Patrich2008, 264; with permission by the author.)
Fig. 4. Herodian Lamp from Sepphoris. (Lapp Reference Lapp2016, 247 no. 19; photo: Mariana Saltzberger; courtesy of the Israel Antiquities Authority.)
Distribution and dating
The table below lists all known finds of NCNs, providing the most exhaustive documentation of the lamp to date. It includes the quantities found and contexts when these are known. The find spots have been plotted on the distribution map in Figure 5.
Fig. 5. Distribution of Northern Collar-Neck Lamps. (Map by Roi Sabar.)
While the precise quantities are unclear, it is notable that NCNs were found at some 22 different sites, significantly more than indicated by previous scholarship.Footnote 19 The find spots are overwhelmingly located in the Galilee, an area that was predominantly Jewish and where the Mishnah was edited.Footnote 20 Even if the NCNs were not the most common lamps of their era, their concentrated distribution makes their visibility and familiarity to people who lived in the area more likely, especially compared to the so-called Sabbath Lamp, a wholly unique (and unprovenanced) object that to date has been the artifact most frequently associated with Mishnah Shabbat 2:4 (see below).Footnote 21
With regard to production, Gärtner's suggestion that the NCNs were likely produced in the area of Sepphoris, Nazareth, and Karm er-Ras seems reasonable due to the large concentration of them in this area. This coheres with Aviam's suggestion that NCNs were likely produced in the Galilee and is supported by the petrographic analysis of an NCN from Magdala that indicates that it was manufactured at Shikhin, the site of a lamp workshop, and petrographic analysis of an NCN from Yodefat that suggests that it was produced on site.Footnote 22
My study of the known NCNs also sheds new light on their dating. Sussman's Reference Sussman and Wachsmann1990 publication of the NCN from the Sea of Galilee boat, which dates the lamp to ca. 50 BCE to 50 CE based on its stylistic similarities to the Ephesus lamp, is a touchstone treatment that is frequently cited and followed in lamp studies.Footnote 23 In 2012, Sussman slightly revised the dating, from the end of the 1st c. BCE to 70 CE.Footnote 24 Based on finds at Nazareth, Feig suggested a later date range, from the 1st c. CE to the mid-2nd c. CE, though her dating has been largely overlooked.Footnote 25
My analysis of the known NCNs and their excavated contexts lends support to Sussman's original start date of the mid-1st c. BCE.Footnote 26 That said, based on excavated finds, NCNs continued well beyond Sussman's proposed end dates of ca. 50/70 CE, and align better with Feig's suggestion, as NCNs were found with assemblages and in contexts that extend into the mid-2nd c. CE, or in some cases even later (e.g., Abila, ‘Araba, Caesarea Maritima, Kfar Kana, Magdala, Nazareth, Sea of Galilee, Sepphoris). Sussman herself was aware of the presence of NCNs in 2nd-c. CE contexts, but dismissed this without explanation.Footnote 27 In short, based on excavated finds, I suggest that the NCN dates from the mid-1st c. BCE and remained in use into the mid-2nd c. CE, if not later. As such, these lamps would have been around in the time and place in which the early rabbis, whose traditions are collected in the Mishnah, lived.Footnote 28
Usage and contexts
People frequently find new ways to use objects or modify how they work, often conditioned or prompted by cultural or other considerations that are specific to each user's contexts.Footnote 29 Because the Hebrew Bible prohibits creating a new flame during the Sabbath (Exodus 35:3), a custom developed among some Jews by the 1st c. CE to light lamps on Friday at sunset – just before the start of the Sabbath. This would provide light for the customary Sabbath dinner after dark as well as normal movement and activity around a house on Friday night.Footnote 30 No pre-rabbinic sources provide much detail about how the lighting was performed – how many lamps, which kinds of lamps, the types of oils and wicks used, and so on. Perhaps as early as the 1st c. CE, rabbis – literally “masters” or “teachers” of the laws in the Hebrew Bible and their interpretation – examined this and other customs circulating in Jewish society, elevated them to legal categories, and went about defining and prescribing their fulfilment in great detail.Footnote 31 As they did with many topics, the early rabbis (i.e., “Tannaim”; ca. 1–220 CE) took brief biblical laws and fleshed them out by relating them to objects and their uses – showing their intended audience how these laws ought to be understood and fulfilled in concrete, tangible ways. In doing so, the rabbis projected their expertise and have left us with numerous detailed discussions of objects and materials that circulated in their day and age. The rabbis’ preferences, moreover, often reflect the material contexts of northern Palestine; for example, they prescribed the use of olive oil for fuel and flax for wicks, reflecting the materials that were cultivated, produced, and used locally in Roman Galilee, even if these were less accessible to Jews living elsewhere.Footnote 32
Thus, early rabbinic texts include discussions not only of local objects of the Roman era, but also of how they were used; for example, how people sought to extend the illumination time of lamps. The question of how to extend illumination time automatically, without the need for human intervention, was a curiosity reflected in several ancient sources and aligns with legends of ever-burning lamps.Footnote 33 In any case, extending burn time would be a matter of convenience for many people, regardless of religious observance. It would also be of interest to those who sought to refrain from kindling a new flame during the Sabbath.Footnote 34 Because the most common materials (olive oil and flax wicks) tended to burn for an hour or so, it would be advantageous to find ways to extend burn time, especially to illuminate the customary meal held after dark on Friday night.
What kind of contraptions, devices, setups, and other material practices did people employ to extend illumination time? And which options were too likely to have invited or necessitated human intervention during the Sabbath, which would have been prohibited? The rabbis take up these issues in Mishnah Shabbat 2:4:
לא יקוב אדם שפופרת שלביצה וימלאנה שמן ויתנינה ()[על פי] הנר בשביל שתהא מנטפת [ו]אפילו היא שלחרס. ור' יהודה מתיר. אם חיברה היוצר מתחילה מותר מפני שהוא כלי אחד. לא ימלא אדם [את] הקערה שמן ויתננה בצד הנר ויתן ראש הפתילה בתוכה בשביל שתהא שואבת. ר' יהודה מתיר.Footnote 35
A. One may not pierce the shell of an egg and fill it with oil and place it on the opening (lit. “mouth”) of a lamp so that it may drip, even if [the additional reservoir] is made of clay. But Rabbi Judah permits it.
B. If the potter had connected [an additional clay reservoir] from the outset, it is permitted, because it is [considered] a single vessel.
C. One may not fill a bowl with oil and place it next to the lamp and place the head of the wick into it [i.e., into the bowl], so that it will draw oil. But Rabbi Judah permits.Footnote 36
The passage lays out three material, physical ways in which people endeavored to extend burn time without human intervention during the Sabbath (i.e., after initial set up before the Sabbath). In Lemma A, the end user adds a makeshift, additional reservoir for extra oil, made of an eggshell (prohibited by most rabbis, though Rabbi Judah permits). Lemma B discusses a supplementary reservoir crafted during the original manufacture of the lamp by the professional potter who created the entire object (permitted by all). Lemma C discusses a dish with additional oil, placed next to the lamp, from which the wick draws (prohibited by most rabbis, though Rabbi Judah permits).
It seems that the central issue here is whether the object was created with an additional reservoir at the outset by the manufacturer, or if the additional reservoir was a later add-on or ad-hoc appendage set up by the end user. If the work was done professionally, it probably enhanced the physical integrity and stability of the arrangement, which in turn lowered the likelihood that one would need to adjust the additional reservoir during the Sabbath (which was prohibited) to ensure its continued proper function.Footnote 37 Also relevant for the rabbis would be the original purpose or intent for which the vessel was created, which often provided a template in rabbinic thought for how objects could be used.Footnote 38 The consensus, represented by the anonymous sages, is that one may use the professionally made setup, but not an arrangement constructed by the end user with an eggshell. The text preserves Rabbi Judah's lone dissenting opinion without rationale or explanation, which is common in the Mishnah.
All three setups – the eggshell placed on top, the supplemental reservoir present from the outset, and the dish placed alongside the lamp – are highly material and object-oriented, and deserve attention. Notably, archaeological artifacts have been associated with objects described in (B) and (C), though these identifications are somewhat problematic.Footnote 39 For the present article, I will focus only on the first, (A) – the use of a pierced eggshell. I argue that one could place an eggshell over the filling hole of several different kinds of lamps that circulated during the era, and that this could work well, though the practice was particularly inclined or invited by the affordances of the NCN.Footnote 40
Affordances of collar lamps
“Affordance,” refers to a relationship between a physical object and a person (or other interacting agent).Footnote 41 The concept helps us focus not only on an object's physical properties, but also on the behaviors that those properties enable.Footnote 42 Design theory uses “affordance” to describe perceived functional properties that are made possible and incline people towards specific uses.Footnote 43 Affordances tell us whether a particular action can be undertaken, as well as how efficiently and accurately.Footnote 44 A particular design can afford actions and uses that were unintended by the manufacturer, as humans find ways to give objects new functions.Footnote 45 Indeed, Gärtner and Lapp make the point that the high collar would prevent some spillage of oil from the reservoir – a helpful feature that complements the lamp's portability.Footnote 46
Instructive in this regard is to compare the NCN with other lamps from Roman Galilee. The Palestinian Discus Lamp (“PDL”) was the most widely used lamp in Roman Galilee.Footnote 47 While this lamp exhibited other advantages (e.g., a small filling hole allowed one to fill the lamp to the top and minimize spillage), its wide and shallow discus and small filling hole do not seem to prompt, or even allow for, an eggshell to be balanced on top (Fig. 6). Moreover, that the discus was often decorated may have served as a deterrent against covering it with an egg or other object (Fig. 7). Notably, many of the PDLs had their discus punched out, creating a larger, albeit jagged, filling hole. An eggshell could have been placed over a new, larger filling hole – if one could remove the discus without ruining the lamp, which was difficult to do (Fig. 8).Footnote 48 Other prominent types in the region include the Herodian Lamp (Fig. 4) and Darom Lamp (Fig. 9), whose large filling holes would have supported the circumference of an eggshell. The eggshell, however, would be much more stable if it was supported by a collar surrounding the filling hole; that is, other lamps allowed for placing and stabilizing an eggshell on top, but not nearly as effectively or efficiently as the NCN.Footnote 49 With their high collars, NCNs not only accommodate an eggshell, but also suggest efficiency and even invite one to place an eggshell or something similar in size and shape on top.
Fig. 6. Palestinian Discus Lamp with discus intact, from Sepphoris. (Kelsey Museum of Archaeology; object identifier: 0000.08.9811; see Gazda and Friedland Reference Gazda and Friedland1997, 16 fig. 13, upper left corner.)
Fig. 7. Palestinian Discus Lamp with decorated discus, from Nahal Aviv, Galilee. (Photo: Tal Rogovski, used with permission by Roi Sabar [Reference Sabar2019, 82. Fig. 7:9].)
Fig. 8. Palestinian Discus Lamp with discus removed, from Tiv'on, Galilee. Length: 8.8 cm; width 7.4 cm; height 2.2 cm. (Vitto Reference Vitto2011, 47, fig. 24: 2. Photo: Yael Yolovitch, Courtesy of the Israel Antiquities Authority.)
Fig. 9. Darom Lamp from Ma'yan Barukh, Galilee. (Photo: Yael Yolovitch, Courtesy of the Israel Antiquities Authority.)
Given that NCNs were predominantly found in the time and place in which the early rabbis lived, it seems likely that those rabbis would have been familiar with the NCNs and the various ways that people used them. In this way, we show how NCNs are not just things that tell us about the past; they can also help us to see how things actively shaped the past.Footnote 50 It can be said that the material culture of Roman Palestine shaped and influenced rabbinic traditions and played an often overlooked and undervalued role in the history of Judaism during the Roman era. For Roman archaeologists, these rabbinic sources alert us to usages and material practices that might have otherwise been unknown.
Eggshells
In addition to the lamp, the second object of interest in these material practices is an eggshell. Due to their fragility, however, only a handful of eggshells have survived and been recovered from excavations in the Levant, including very few from Roman Palestine specifically.Footnote 51 They are mentioned with some frequency in early rabbinic texts, and scholars have found that these references are likely to chicken eggs, unless otherwise qualified as eggs of other animals.Footnote 52 In a recent study, Zohar Amar collects and surveys all of the archaeological finds of chicken eggshells from the region. Some 19 different eggshells (whole or fragments) have been found in excavations with dates spanning from the 6th c. BCE (Jerusalem) to the 10th–11th c. CE (Yavneh).Footnote 53 As such, the following should be viewed with the significant caveat that we do not have for eggshells what we have for lamps – namely, we lack a significant number of examples from a particular time and place (e.g., Roman-era Palestine).
Based on the available sources, Amar reasonably concludes that chicken eggs in pre-modern Palestine measured approximately 40–44 cubic centimeters in volume. Thus they were smaller than the eggs that have been produced since the mid-20th c. (46–65 cc, from “small” to “extra-large” eggs), and also smaller than the estimates proposed by traditional rabbinic commentators (50–100 cc; 57 cc is the prevailing opinion), who discussed egg size because of its implications for rabbinic law.Footnote 54 How much additional oil or illumination time would the eggshell provide? The capacity for oil would depend, in part, on how much of the eggshell remained intact after its top was punctured to allow oil to be poured in. Our information on how eggshells were punctured, and/or pierced for that matter, is very sparse. At Sardis (Turkey), broken or pierced eggshells have been found and dated by the excavators to the 1st c. CE. Some have been discovered together with tools that could have been used for creating holes in eggshells: a bronze needle, a bronze pin, and an iron stylus point or pin.Footnote 55 Pliny the Elder mentions using the pointed end of a spoon for puncturing eggshells (HN 28.19).Footnote 56 In Jerusalem, though post-dating both m. Shabbat 2:4 and the NCN lamp, a whole egg was found with a small puncture in its side used to drain its inner liquids and preserve the shell.Footnote 57 In short, we can say that the know-how likely existed to puncture and pierce eggshells – either for pouring oil in or allowing it to drip out – though due to the low survival rates of eggshells in the archaeological record, there is not enough data to know how common these practices were, precisely how they were performed, or how much of the eggshell remained for holding oil.
All that said, even if only half of the eggshell remained intact, it likely would have provided enough extra oil to extend illumination for a meaningful amount of time. Half an eggshell would hold roughly 20–22 cc of oil. For an approximation, we can draw on Dorina Moullou's experimentation with Greek lamps. She writes that flax wicks measuring roughly 5 mm in width – which aligns with the little we know of wicks from Roman Palestine – consume 5–6 ml of oil per hour.Footnote 58 Based on these estimates, it is reasonable to surmise that an eggshell, even if only half full, could potentially provide a useful amount of extra illumination time. For the scenario that the rabbis are addressing, if a lamp was lit at sunset, an extra hour or two would have sufficed to facilitate conducting a meal after dark on Friday night, which had become an aspect of Sabbath observance.Footnote 59 To be sure, further sources on where and how the eggshell was punctured would be needed to provide more specific estimates of the extended illumination time.Footnote 60
Discussion
The rabbis worked out their ideas by relating them to things that they were familiar with, objects and material practices that were part of their immediate surroundings. They seem to presume a shared visual and material culture with their intended audience: they do not introduce objects in detail for readers who may not be familiar with them, but rather presume that their audience already has some knowledge of the items to which they refer. Objects structure human life, as the physical and stylistic characteristics of the material culture within which one is embedded can influence one's thinking in both conscious and subconscious ways.Footnote 61 Here, we may think of Gosden's sensorium or of objectscapes – repertoires of material culture available at a certain site in a certain period.Footnote 62 Mishnah Shabbat 2:4 is one of hundreds (if not thousands) of rabbinic discussions of objects and how they were used (and there are some 80 other discussions of lamps in the Mishnah alone that do not incorporate using an eggshell).Footnote 63 Perhaps the rabbis’ expansive engagement with material culture was prompted or at least enabled by Mediterranean-wide increases in production of goods during the Roman age.Footnote 64 The abundance of objects gave the rabbis physical and mental fodder for interpreting and fleshing out pre-existing Jewish customs and biblical laws that in practice were otherwise vague. In short, we have textual sources on the ideas of individuals (rabbis) who were interested in the details of daily life, as well as archaeological sources on their material milieu – a fortunate pairing that could be useful for Roman archaeologists in writing material histories.Footnote 65
Conclusion
This article has sought to contribute to the growing conversations on object-focused histories among Roman archaeologists by applying these approaches to sources from Roman Palestine. I have argued that the Northern Collar-Neck Lamp's high collar, perhaps designed to reduce spillage, also served as an affordance that enabled and invited an efficient material practice by the user; that is, placing an additional oil reservoir on top, such as a pierced eggshell. Filling an eggshell with oil and letting it drip through the piercing would have prolonged burn time without human intervention. This usage of the lamp, likely not self-evident from examining the finds alone, is brought to light through critical engagement with early rabbinic sources, which have been heretofore under-utilized in recent scholarship on objects in Roman archaeology.
Acknowledgments
I am grateful for the helpful feedback from the anonymous reviewers and Greg Woolf of this journal. This article also benefitted from input and communications from Mordechai Aviam, Carey A. Brown, Adi Erlich, Toby Klein, James R. Strange, and Shulamit Terem. I also benefitted from the Lydie T. Shufro Summer Research Fellowship at the Albright Institute of Archaeological Research in Jerusalem in 2023 and from feedback following presentations of earlier versions of this paper at the Albright Institute (2023), the Theoretical Roman Archaeology Conference (2023), and the Talmudic Literature Conference at Yale University (2023). I thank Tal Rogovski for preparing Figs. 2 and 3, and Roi Sabar for Fig. 7 and for creating the map (Fig. 5). I alone am responsible for all remaining errors.
Competing interests
The author declares none.
Open access
