‘You,’ ‘Us,’ and ‘Them’: Glass and Procedural Knowledge in Cuneiform Cultures


When you want to make a kiln for glassmaking, you search continuously for a propitious day during a favorable month. You lay the foundations of a kiln with four chambers. You make constant offerings and set up purifying divinities so that no impurities may enter: you make lapis lazuli.

These instructions summarize the contents of a corpus of Akkadian glassmaking recipes from more than two and half millennia ago.[1] It was then, in the seventh century BCE, that the king himself claimed to have dedicated clay tablets containing instructions “for your making stones” (colored glasses and frits) to the temple of Nabu and Tašmetu, the patron gods of knowledge:

Written [and checked] according to its original. I am Aššurbanipal, [king of the land of Aššur], on whom the god Nabu and the goddess Tašmetu bestowed wisdom; the one who has sharp eyes. I have written, checked, and collated on tablets the highest level of the scribal arts, such a skill as none amongst the kings my predecessors had learned, everything pertaining to cuneiform signs, the wisdom of Nabu. And deposited (them) for the sake of my life, the preservation of my living, for the displacement of my illness. (For) the foundation of my royal throne, on that day I deposited them in the library of the temple of Nabu, the great lord, my lord, which is in Nineveh. In future days, Oh Nabu, look favorably upon this work and bless my kingship, whenever I call on you, take my hand! Whenever I go to your temple, constantly protect my way. Once this work is placed in your temple and established before you, look favorably (upon me) and remember my good fortune. Tašmetu, great lady, your beloved first-rank wife, who intercedes for me before you—may she sleep well in bed and [daily] ask you for my life without stopping. [Whoever trusts in you], Nabu, should never be ashamed.[2]

This colophon is elaborate, longer than most from the Nineveh libraries. It makes explicit that, according to scribes, the audience for the recipes was supposed to be divine, that we, the uninitiated (la mūdû or “non-knowers”), were never meant to have seen these instructions. The language of the recipes themselves, however, tell a different story. In this ancient corpus of “stone-making” instructions, the grammatical subject is you.

The second-person pronoun populates a long-standing procedural text tradition that spans two millennia of cuneiform cultural history. The earliest surviving procedural texts stem from the early second millennium BCE, or the Old Babylonian Period. During this time, scribes began to instruct “you” (atta in Akkadian). You, in this period, were to calculate the depth of a canal or the area of an irregular triangle. You were to make calculations by following the “procedure” or nēpešum, a noun in Akkadian that derives from the verb epēšu, “to make.” It was during this formative period of intellectual history that the first culinary recipes were produced, Akkadian instructions that detailed how you should make a variety of stews. You texts caught on, becoming a regular feature of Akkadian intellectual history. They endured well into the first millennium BCE, covering a broad range of subjects from medicine, perfumery, and horse-training to mathematical astronomy and the making of cult statues. In the course of two millennia, scribes produced hundreds of cuneiform tablets addressed to you.

Today’s cookbooks and instruction manuals tend to employ imperative rather than second-person verbs. Ultimately, they are the heirs of linguistic innovation. From an ancient scribal perspective, recipes were a textual solution to an epistemological problem: How does one transmit expert knowledge?[3] Or rather, how does one transmit recovered knowledge? At Nineveh, after all, knowledge transmission and knowledge recovery were one and the same thing. In the glassmaking procedures, for example, knowledge is ritually recovered from authorities in the past:

In the process, you set up Kūbu (purifying divinities) within two double hours. You sacrifice a sheep. You make a funerary offering to experts of yesteryear. You collect the ingredients in a (casting)-mould and set it down into an utūnu-kiln…[4]

Language can clue us into the frameworks of technical knowledge in Assyrian scribal cultures. In the passage above, the Akkadian term for “expert” makes no distinction between a glassmaker and a scribe. Both are simply ummânu. Translating Akkadian scholarly texts puts us—puts you, the reader—in a categorically distinct framework of ancient knowledge production practices. Akkadian, however, is only one of the challenges that besets modern translators and historians. The real labor of love requires entering the multivalent world of the cuneiform writing system itself and engaging with “cuneiform cultures.”[5]

Pictured here is one of three Neo-Assyrian tablets housed at the British Museum that preserve instructions for building a glassmaking kiln and making colored glasses. Obverse side. Seventh century BCE. Source: K 00203+ Cuneiform Digital Library Initiative (CDLI), https://cdli.ucla.edu/P393786.

In a given cuneiform culture, the meaning of entire passages of the glassmaking recipes lends itself to multiple interpretations. Take, for example, the instructions for making a glass kiln found in the introductory section of the recipes. They ask you to gather a specific species of wood to light the kiln fire with, pieces of

qurū-wood that have no knots on the surface, bound up with an apu-strap that you cut in the month of Abu, these logs should go into the bottom of the kiln (kūru).[6]

The terms qurū and apu are unique to this corpus of glassmaking recipes. At first glance, this would seem to be an Akkadian lexical challenge, but there’s more going on. And it requires knowledge of orthography and homophony, that is, of scribal hermeneutics. In the excerpt, qurū and apu are homophones (or near homophones) of other terms in the same passage. The wood identified as qurū, for example, is a near homophone of kūru, “kiln-wood.” Similarly, the so-called apu-strap, used to bind the qurū-wood, is a homophone in the month Abu (a well attested month name). Orthographically, the writing of the month Abu, a noun which on its own means “father,” is written using the cuneiform sign 𒉈, the Sumerian term for “fire.” Collectively, these various layers of meaning can be synthesized into a modern sense translation that attempts to capture the type of multivalence that underpins the glass texts. Taking these levels of meaning into account, we might read the same passage as follows:

“kiln”-wood (qurū) logs with no knots on the surface, bound up with the “father”-strap (apu), the logs that you cut in the “month of fire”—also called the month of the father (Abu)—these are the logs that should go into the bottom of the kiln.

By examining the language of the Neo-Assyrian recipes, by honing in on the semantic and orthographic plasticity of cuneiform scholarship, we can move toward explaining how a technological subject like glassmaking could be refashioned into a work of scribal craft. We can explain, moreover, why the king himself (who was not so secretly obsessed with collecting ancient wisdom) felt compelled to dedicate the recipes to the gods of scribal knowledge.


The history of how earlier scholars have interpreted the glassmaking recipes since their discovery in the early twentieth century can tell us a great deal about ourselves. Heinrich Zimmern (1925), Reginald Campbell Thompson (1925), and later A. Leo Oppenheim (1970) were motivated exclusively by the desire to reconstruct Assyrian glassmaking technology from the culture’s ancient recipes.[7] Oppenheim’s work took an experimental turn in the 1960s, when he began a nearly decade-long correspondence with Robert Brill at the Corning Museum of Glass. Their collaboration refined both our philological and our technical understandings of the texts, but these scholars never attempted to explain the ritual and literary elements framing the Assyrian instructions. By the mid–1990s, the archaeologist Roger Moorey concluded rather pessimistically that the recipes were “little more” than literary and lexical texts. He found scant reason to examine the recipes for evidence of technology:

These texts are part of a traditional Mesopotamian literary enterprise: the compilation of lexical lists. These had always included terms for natural stones, so the artificial manufacture of such stones (once it became a regular industrial activity) was of interest. Thus the glass texts, in accord with this long-standing purpose, concern themselves with little more than the appropriate technical vocabulary. As the technical procedures themselves were of little interest, and probably unknown, to the scribes, their descriptions are rudimentary: sufficient for the purposes of lexicography, but inadequate as descriptions of an actual technology. Literary tradition is a notoriously inadequate source for the history of technology.[8]

Moorey’s comments regarding the glass recipes were a response to a long-standing assumption on the part of philologists that such texts could offer an unobstructed window into the craft practices of ancient societies. By contrast, Moorey’s Ancient Mesopotamian Materials and Industries privileged an objects-first approach to the history of technology. Oppenheim and Moorey were on two ends of an either–or spectrum. But today we can take a different approach. We can ask: Could not the glassmaking texts be both technological and scholarly? What would be the benefit of a both–and approach? More broadly, what would an intellectual history of technology in cuneiform cultures look like?

This history might start with the premise that cuneiform recipes, and procedures writ large, are good to think with, particularly when they nudge us to question those culturally inherited intuitions we consider universal. In the study of the material world, these intuitions include the idea that there is a separate non-human world of nature, that materials are composed of atoms or substances, and that “natural” things are ontologically superior to those that are human-made, or “artificial.”[9] There is, moreover, the epistemic intuition that knowledge gained in the arm chair, by reason, number, deduction, and theory (episteme) is distinct and above that gained by making things somatically, is superior to knowledge gained by craft (techne). Cuneiform recipes, and the glassmaking texts in particular compel us to commit to the kiln’s flame the episteme–techne distinction and, with it, the hierarchies and knowledge binaries that were introduced in the Classical world, binaries to which we have long since become accustomed. Expertise, and indeed, knowledge—in Akkadian, nēmequ—during the period of Assyrian scholarship that frames the glassmaking recipes, showed little interest in maintaining such boundaries. All that could be known, “knowledge of everything” (mudu kalama) could be collected, impressed carefully onto clay using a reed stylus, and preserved for perpetuity. All of it. Such was the power of the cuneiform script. It still has something to tell us, if we care to listen.


Recipes, and procedures more broadly, can inspire us to rethink not just knowledge categories (as in nēmequ) but also disciplinary ones. I have often heard historians speak of their discipline as one that engages in a conversation with the past, with its ideas and its human and non-human actors. When describing those historical actors (or “actants” if that’s your style), we appeal to our favorite third-person-plural pronouns. They and them become the linguistic subjects and objects of past conquests, inventions, and declines. You, however, can remain at a safe distance. The Akkadian procedural tradition provides one of the earliest challenges to any unspoken assumptions about historical objectivity. There can be no “view from nowhere” when the texts quite literally make us their subjects. When Akkadian scribes first innovated mathematical and technological instructions during the second millennium BCE, they may have reasoned that most efficient mode of knowledge transmission, recovery, and preservation would be to appeal directly to “you.” And just like that, with a simple pronoun and second-person verbs, scribes opened up a unique line of communication to their past, a conversation between you, us, and them.

Eduardo A. Escobar is a postdoctoral researcher and instructor at the Stevanovich Institute on the Formation of Knowledge, The University of Chicago. His Twitter handle is @ScribeEAE.

  1. A full translation of one lapis lazuli recipe is available at Glass as part of a corpus of glass technological texts in the Open Richly Annotated Cuneiform Corpus (ORACC) consortium at http://oracc.org/. See K 00250 at http://oracc.org/glass/P394484. The digital humanities tools there provide broadly accepted, stable standards for the online publication of digital cuneiform texts. All texts published in Glass will be open access under the Creative Commons Attribution-ShareAlike 3.0 license.  ↩
  2. K 04273, http://oracc.org/glass/P395469. Translation note: Parentheses indicate implied objects in the original text, whereas square brackets indicate breaks and reconstructions of the text, usually based on duplicate manuscripts.  ↩
  3. This question persists in the history of science. A recent essay on this blog by, for example, explores the same question for the field of microbiology. See Charles A. Kollmer, “The Theorist’s Doctrine and the Collector’s Technique: On The Historicity of Expertise in Microbiology,” History of Knowledge, May 25, 2018, https://historyofknowledge.net/2018/05/25/historicity-of-expertise-in-microbiology/.  ↩
  4. K 04266, http://oracc.org//glass/P395468.  ↩
  5. See Karen Radner and Eleanor Robson, The Oxford handbook of Cuneiform Culture (Oxford: Oxford University Press, 2011).  ↩
  6. K.203+ obv. i 18–20 (standard translation), http://oracc.org/glass/P393786.  ↩
  7. See H. Zimmern, “Assyrische chemisch-technische Rezepte, insbesondere für Herstellung farbiger glasierter Ziegel, in Umschrift und Übersetzung,” Zeitschrift für Assyriologie und Vorderasiatische Archäologie 36 (1925):177–208; R. Campbell Thompson On The Chemisty of the Ancient Assyrians (London: Luzac and Co., 1925); and A. Leo Oppenheim et al., Glass and Glassmaking in Ancient Mesopotamia: An Edition of the Cuneiform Texts which Contain Instructions for Glassmakers with a Catalogue of Surviving Objects (London: The Corning Museum of Glass, 1970).  ↩
  8. P. R. S. Moorey, Ancient Mesopotamian Materials and Industries: The Archeological Evidence (Oxford: The Clarendon Press, 1994), 211.  ↩
  9. For a discussion of how cuneiform scientific scholarship developed without appeal to a concept of nature, see Francesca Rochberg, Before Nature: Cuneiform Knowledge and the History of Science  (Chicago: University of Chicago Press, 2016).  ↩
Suggested citation: Eduardo A. Escobar, “‘You,’ ‘Us,’ and ‘Them’: Glass and Procedural Knowledge in Cuneiform Cultures,” History of Knowledge, December 5, 2018, https://historyofknowledge.net/2018/12/05/you-us-and-them-glass-and-procedural-knowledge-in-cuneiform-cultures/.