What Did You Learn in School Today? | Paul Delnero, Johns Hopkins

It's a great pleasure to introduce tonight's speaker. But before I do that, I want to welcome most of you back to yet another Back to School in Babylonia event. I'm really glad that we have a wonderful turnout and that we also have, again, so many students here. And for all of you watching online, thanks for following along with the series. Matt just mentioned the talk with Anne Godaris. And then in December, we will have another Wednesday lecture with our very own Jana Matuszak. So please come back for that one as well. And now, it's a great pleasure to introduce tonight's speaker, who is really a world-renowned expert of Sumerian everything, especially Sumerian education. And that's why he's here tonight, Paul De Niro. Paul has his PhD from the University of Pennsylvania and is currently Associate Professor of Assyriology, I would say, slash Sumerology at the University of John Hopkins. He has published over 40 articles on all kinds of aspects of Sumerian literature and Sumerian language transmission, transmission of knowledge. And his first book is a two-part thing. There is the published part, Textual Criticism on Sumerian Literature. And then there is the over 2,000-page unpublished part, which we are all very grateful for because it's the first really complete edition of all the Decat texts. Those 10 texts, students studied first when they went into the second phase of education. And his work really revolutionized the way we understand the transmission of Sumerian literature, and also what he will speak on tonight, the way students learned Sumerian literature by memorizing it. And after that, with his second book, which is equally fantastic, he took a break from Sumerian education and went into texts which are practiced as rituals and wrote a whole book about how to do things with tears. And it's a really comprehensive and cultural study of the practice of lamenting to appease the gods in Mesopotamian culture spanning over, I would say, 2,000 years of material, which is super impressive. But I'm very excited that now you're returning to doing something on Babylonian education, and we will also hear about that tonight because the next book project is about the earliest writing exercises and how students acquired the first elements of the Sumerian script. And I'm very excited to learn more about that because this is brand new work. And without further ado, Paul, thanks for coming here tonight, and I'm very much looking forward to your talk. Thank you, Suzanne, for that overly generous introduction. It is really a great pleasure and an honor to be here. I grew up actually not too terribly far from here in one of the suburbs about 20 minutes away, and had never thought in a million years that I would be giving a lecture in this institution, but I remember coming here when I was a little kid to the museum in the building at ISAC and having my eyes made wide open to discover the past and all the fantastic objects that have made me curious enough to spend most of my life up until now studying it, so it's great to be back. Also, to revisit a topic that I haven't worked on really extensively since my dissertation, which Suzanne described in a little bit, so I thought, since I'm returning to an old topic from my dissertation, which in my mind seems like it took place around 1740 BCE, not too long after House F closed, it was appropriate to begin this talk with an anecdote from my time when I first started my studies. At some point early in my time as a graduate student, I asked my dissertation advisor why he became interested in Assyriology. He answered by telling me that when he was younger, he was sitting on a hill in his hometown in Cornwall overlooking a cemetery on what I imagined to have been a dark and misty night, reading Samuel Noah Kramer's translation of the Sumerian text, Lamentation Over the Destruction of Sumer and Ur, and James Pritchard's anthology, Ancient Near Eastern Texts Relating to the Old Testament. As the evocative images of the devastated land of distant days rose off the page and merged with the site of the barren valley of worn and dilapidated graves before him, my advisor knew from that moment on that he wanted to become a Sumerologist. I remember being very moved by this story as I still am today, but also at the same time feeling troubled because I did not have anything even remotely comparable to this wonderful origin story. In fact, at that moment in my life, I had absolutely no idea why I had decided to pursue a career in Assyriology. And if anything, reading the translations of the text in Annet had had the opposite effect on me. Where my advisor discovered poetry that brought the past to life and transformed the world before his very eyes, I found only cryptic and lapidary phrases generously interspersed with dots marking lacuna, an experience that only made the past seem distant and inaccessible and so utterly different than the world in which I imagined myself to live. When I think back on this experience, the cause of my frustration is much clearer to me now than it was back then. I was and still am an avid reader of contemporary fiction and the many novels and stories that have transformed the equivalent of a cemetery in Cornwall into a place of magic and wonder for me had led me to expect any good work of written literature to do the same. The problem is, of course, us and the expectations we bring to the Mesopotamian text we call literary. Because Mesopotamian literary compositions are known to us only from written sources and are composed in an elevated poetic style, they are frequently read like modern works of literature. In other words, a static text that were meant to be read and enjoyed in silent contemplation without the distraction of sights, sounds, smells, and other intrusions from the outside world to break the spell of the words on the page, or in this case, the clay tablet. The same conflation of modern with ancient literature applies equally to conceptions of literacy in ancient Mesopotamia, and in particular, the accessibility and cultural function of writing in relation to Sumerian literary works. This is an example from a recent book by Mark Vandemura, but it's by far not the only example of this. This is from last year, I think, or in the spring, 2023. And here you find, literacy was a restricted skill, even if there were businessmen and generals who were able to write their own correspondence. And certainly only a very small number of people must have engaged with Sumerian literary works after finishing their schooling. Those people formed an intellectual elite whose knowledge must have been considered esoteric to their neighbors. During the past 20 to 30 years, it has become increasingly more common, especially in this country, but in other parts of the world as well, to view Sumerian literary works as texts that were known only to a small group of elites and whose contents had little or no relation to religion as it was experienced and practiced in everyday life. Certainly one of the most influential, if not the first statements to this effect, can be found in A. Leo Oppenheim's famous essay, Why a Mesopotamian Religion Should Not Be Written, first published in 1964, where he writes, and the second group of texts to be examined contains myths and mythologically embellished literary works. To state at the outset my objection to the direct and indiscriminate utilization of such texts, I submit that their contents have already unduly encroached upon our concept of Mesopotamian religion. All these stories about the gods and their doings, about this world and ours and how it came into being, these moralizing as well as entertaining stories geared to emotional responses represent the most obvious and cherished topics for the literary creativeness of a civilization such as that of Mesopotamia. They form something like a fantastic screen, enticing as they are in their immediate appeal, seductive in their far-reaching likeness to stories told all over the ancient Near East and around the Mediterranean, but still a screen which one must penetrate to reach the hard core of evidence that bears directly on the forms of religious experience of a Mesopotamian man. For those familiar with the history of the field, especially at this venerable institution, it is not hard to recognize in this passage a strong critique of Oppenheim's colleague at the time, Thorkild Jakobson, whose highly influential treatments of Mesopotamian religion were based on the exact opposite assessment of Sumerian literature as evidence for reconstructing Mesopotamian religious practices. Be that as it may, Oppenheim's remarks in this passage have since reached a much wider audience to become, in many ways, the consensus view of the relevance of Sumerian literary works to understanding ancient Mesopotamia. And indeed, it is not a distant leap from the perceived elite nature of Sumerian literature to the perception that literacy and access to education and writing was equally limited, as is evident in the following citation from the Sumerologist Piotr Michalowski. I think it will suffice to say that the contents of the text is in many ways irrelevant to the main point of my argument. And the types of texts that he's talking about are scribal exercises, and in particular, exercise copies of Sumerian literary works. It does not really matter what they, the scribal pupils, learned as long as they learned it and it made them different. The conceptual dimensions of literacy literally made them different human beings. They knew of matters that others only imagined existed. They even knew what year it was. In such a context, the perpetuation of a dead literary language, Sumerian, has a definite function which helped to define membership in an exclusive club, the world of bureaucracy. Seen in this way, Mesopotamian mythological narratives of the texts that have been classified as Sumerian literary texts are thus little more than texts written by the elites for the elites for the purpose of forming elites. While this may or may not be true of the Sumerian literary texts that were copied during the course of scribal education, and we'll come back to that later in this talk, it's worth keeping in mind that writing in Mesopotamia was by no means limited to copying religious texts and mythological texts, and this is only just a few of many examples of the different types of media and the different types of inscribed objects that we have from Mesopotamia. Many of these are quite famous, like the object here on the left, which is, of course, the Stele of Hammurabi. You can't see it too well from the slides, but this contains, supposedly, one of the world's oldest known law codes. It's an extension of an inscription that runs also along the backside of the monument. You have a Neo-Assyrian palace relief, which would have been in the throne room of Asar Nazarpal II with an inscription on it. You have the famous Byzantine inscription of Darius and a rock relief in Iran, a very early pictographic administrative text, a text known as the Babylonian Map of the World, which has a drawing of the Babylonian conception of the cosmos, arguably, with the city of Babylon in the middle and mythological regions outside this body of water here, along with a text describing the inhabitants or what one could expect to find in these regions, the famous flood narrative from the Epic of Gilgamesh, a boundary stone from the early dynastic period recording a land transfer, building inscription on a peg, a statue of a ruler carrying a basket with clay for the first brick for building a temple. Both of these objects would not have been visible at all, but would have been driven into the wall of a temple or, in the case of the figurine, buried underneath the foundation and not visible to anyone entering the temple. A model from the city of Mari of a liver used for the purpose of divination with an inscription of the significance of the ominous sign on the liver, and here at the top, a cylinder seal with its impression that has an inscription giving you the name of the ruler, sorry, the owner of the seal and the title. So you can see from this limited overview of the different types of inscriptions that writing was used for many different purposes, many of which, as is clear from these examples, were not to be read in silent contemplation or even to be read at all. A lot of these texts would have been seen, presumably by people who didn't know how to read or were not well-versed enough in cuneiform to decipher the inscriptions, but who nonetheless experienced the inscriptions not only aesthetically, but also probably had some sense of what the content was based on knowledge of the cultural significance of the inscriptions or people telling them from other sources what was there, right? Another thing that's worth keeping in mind when we think about the role of texts like Sumerian literary compositions inscribable education is that writing is a very time-intensive and complex process that begins very early, often, for those of us who have learned to read and write, early in our lives when we were children, or a few people probably can remember things like this, but if you have kids of your own, this will look familiar to you. Writing, for most kids, begins with drawing, in fact, and it's only much later that they begin to learn the process of learning the sounds that go with the letters that they've just learned how to draw, right? And the process of learning how to write and read, of course, gets even more and more complex from there. This is just a slide that tells you some of the basic skills that you have to learn on your way to literacy, vocabulary, language structures, verbal reasoning, literary knowledge, phonological awareness, sight recognition, and all the rest of it. And each of these stages here in the process of learning how to write is represented, for the most part, in some way, shape, or form in what we know of to have been the stages of scribal training in Mesopotamia as well, which we'll talk about in more detail now. So returning to Nupur and scribal education at the early second millennium and the subject of the fantastic exhibit put together by Susana Paulus and her colleagues here on House F, we have a wealth of information about scribal training, how scribes in Mesopotamia learned how to read and write from the very earliest stages of training until the very end, when they learned literary text. So a brief tour of that. This is an often-quoted text. It's almost obligatory when you give a lecture on scribal education to quote it. And this is a text from the very scribal school environment that we're going to be talking about, describing, probably in idealistic terms, a day in the life of a schoolchild, or in this case, a schoolboy. I'm not going to go through all of it, but among the things that he's described as learning and being punished for not doing well enough are speaking Akkadian, so presumably his native language is Akkadian, and he was there to learn Sumerian, how to speak the language, but presumably also to write it, and also handwriting and things like this. So moving away from sources like this to the actual material and other types of textual knowledge we have for scribal education, this is an excerpt from a ruler from the end of the third millennium BCE, a ruler known as Shoghi, who boasts in one of his inscriptions of erecting two centers for scribal learning, one of them in the city of Ur and another one in the city of Nippur for the purpose, among other things, of recording and immortalizing the praise hymns in his honor so that future scribes could continue to copy and learn them and that his name would be perpetuated through them. Nippur, of course, is located here in southern Mesopotamia, one of the two centers that Shoghi boasts of having established a center for scribal education, and indeed, at Nippur, one of the houses that were found in the area called Tablet Hill, where tons and tons of tablets have been found at the end of the 19th century, produced a private house which seems to have been a schoolhouse, which is also the subject of the exhibit here. So what we know about scribal education comes from the types of text that you find at House F and also from other places at Nippur from the same period that probably served a similar purpose. During the very earliest stage of education, they had distinct tablet types that they work on that reflect the pedagogical purpose and aims of these early exercises. These are the so-called school lentils. They're called type four based on a typology that was established by Miguel Seville for the purposes of describing the different types of sources for lexical lists, but it was adopted for describing the shapes of the types of tablets that were used for scribal education. And here you have on both of these texts an extract, a short extract from a text written in two lines in the hand of a more expert scribe, possibly the teacher, imitated directly in the less able and competent hand of a pupil. And the other type of tablet from the earliest stages of education using the same typology, the so-called type two tablets. And this is the front and the back or the obverse and reverse of a typical one of these. And you get the idea pretty easily just from looking at it. This side of the front of the tablet or what we call the obverse contains a short extract from a text, in this case, syllable alphabet B. One of the first exercises that beginning scribes learned how to copy written in the expert hand of the teacher or a more advanced scribe. Next to it on a surface that's been worn off from repeated erasures, the pupil, the scribal pupil's attempt to directly imitate what's in this column. And then flipping the tablet around and looking what's on the back, you'll notice that there's no teacher's copy or expert copy. The pupil has written him or herself. A longer extract from either an earlier part of the same text that's in the model on the obverse or from another list or text entirely that was learned earlier in the sequence. And why this is important is that by correlating or comparing the types of text and extracts that appear in models on the front of these tablets with the type of extracts and lists and texts that occur on the back of the same tablets, you can pretty reliably reconstruct the entire sequence of the lists and texts that were learned by scribes in training. And this is something that was done in 1997 by Nick Valdehus in his dissertation and it's since become the standard reference for understanding what scribes learned in early education in the sequence in which we learned it. We'll talk more about these two things later, but it begins with exercises in sign forms, moves to this text syllable alphabet B. Then after learning how to write the basic signs and some of their pronunciations, they applied this knowledge to write personal names. And then they started learning more advanced vocabulary by learning thematic word lists. And then they moved on to learning more complex signs and more complex values before moving on to the end of the elementary stages of scribal training by copying more practical texts with longer and more complex sentences like model contracts in the Sumerian proverb collections. After completing this part of the curriculum, if they continued with their education, they began to copy and learn Sumerian literary compositions, the type of texts that I was talking about at the beginning of the lecture. In the first group of these are a group of four texts known as the tetrad. We're fairly certain that they learn these next after completing all of these earlier exercises. Again, based on the type of tablets that those four particular texts occur on. So this is a fragment from a type two tablet like the text tablet that we were looking at earlier, which contains on the obverse Lipidish Tarbi, the first text in the so-called tetrad. And when you flip it over and look at the earlier text on the reverse, it's an extract from the list of personal names, a list of personal names, which was learned much earlier in elementary education. So this tells us that this text was still learned early, but after completing more or less the sequence of elementary exercises that made up the beginning curriculum. After this, they learned an even more advanced group of Sumerian literary compositions, a group of 10 texts known as the decad. We know that these texts formed a group and that they were copied together as scribal exercises from a number of sources. This is an image from one of the most important of these sources also on display in the exhibit, the so-called catalog of curricular texts that were copied by scribes. And what it contains is a list of the titles of different Sumerian literary compositions. And the first 10 of them, you can't really read this from the photograph or really from the original without looking closely at it, contains in sequence the titles of the first 10 texts in the decad, so these 10 texts. And we also know from a similar text, which is housed at the Louvre, we don't know exactly where it's come from, but it's a similar catalog, that these first 10 texts were listed in the same sequence in that catalog. So you have at least two texts like this, that list this text together in sequence and group them together. And you also have collective tablets that put different texts from the decad in the sequence in which they occur in that group. So Shulgi A, which is the first text in the decad, is on the obverse when you flip the tablet over, you get the end of the text followed by the beginning of the next text in the decad, the Bidishtar A. And you have similar, oops, things like this. This is a prism that contains Enlil Hymn A, the fifth text in the decad, and this line that's circled here is the first line of the Kesh Temple Hymn, which is the next text in the decad, so another source that connects texts from the decad in sequence. And then you have this text, which is hard to appreciate from this black and white photograph, but it's about the size of a postcard. It contains, on the front and back, over 10 columns of text, close to 1,000 lines, written in a very small microscript. So this is blown up. This is much larger than the original. And it contains the Kesh Temple Hymn, Enki's Journey to Nupur, Anan and Ebik, the Noongal Hymn, and Gilgamesh and Huawach, which are the last five texts of the decad in the sequence in which they were known. So another collective tablet that groups these texts together. Other corroborating evidence comes from House F itself. So when you look at the number of different types of compositions that are found in the house, the texts from the tetrad and the decad form by far the largest number of duplicates in House F. So these texts were copied more frequently there than most of the other types of literary texts that were found there, suggesting that they played an important role in scribal education. And then another group of text that Eleanor Robeson, who studied this material, calls the House F 14, based on the fact that these literary compositions also seem to have been copied fairly frequently at House F. Okay, so moving from that to how they learned, right, this is what they learned, what we just covered there. Going back to modern education and the observation that when children first start to learn, they don't really learn the meaning of writing or how it encodes speech. They start by drawing and the manual execution of letters. This is actually an example from a child trying to learn how to write his or her name for the first time. And you can see there's nothing resembling a letter there. This particular manual gives some advice to parents and to teachers about how to go this. So they tell you if the child has no intent to practice writing and marks go on and off the page and there is no patterns to the marks, then what one should do is look on the bright side of this and point out something promising about it. I like that little dot right there or asking the child to discuss what the scribbles mean as if it were a work of art or something like that, right? And then showing by example through reading yourself and reading to your child to get them to understand that writing is a serious thing and that they need to learn it and all this sort of thing. So moving to that stage of learning how to write in Mesopotamia, this is the slide to show you that cuneiform signs can consist of three basic types of wedges. Some of them have arcane German names and we all know what they mean by them, but this is a Winkelhacken, it's a diagonal wedge, this is a vertical wedge, and this is a horizontal wedge. And before they even learned how to write a single basic cuneiform sign, what they did was copy each of these individual wedge types together on groups of tablets like this. So here is a tablet which we call the first day of school which shows the vertical wedge repeated in sequence. Here's another one with the so-called Winkelhacken repeated on it. And then eventually they move on to the next text in the group, which is syllable alphabet B, which we'll get to in a moment, to illustrate the bridge between the two exercises. This tablet here has a syllable alphabet B on it. This is the edge of the same tablet where the scribe has just repeated the Winkelhacken or the sign element that you see repeated in the slide to the right. Another example of this is that once they learned how to copy each of the individual web, each of the different wedge types individually, they would copy them in a fixed sequence. So you see the vertical, the horizontal, and the Winkelhacken repeated over and over again, right? On the reverse, so this is an earlier exercise of a type two tablet before moving on to the next text in the list, syllable alphabet B on this side. And the relationship between the two is also apparent here. So this is the very beginning of that sign exercise, syllable alphabet B. If you recall, the first wedge in the sequence is the vertical wedge. The first sign in syllable alphabet B starts with a vertical wedge and consists of two more vertical wedges. And eventually with the next new sign that's introduced, they introduce the horizontal wedge right here, repeating the same sequence of signs. And then when you get further down after they practice different uses of the vertical and the horizontal, they learn the diagonal wedge, right? So they're building on this set of skills, just the manual execution of sign parts to learning how to write simple signs that consist of each of these three different elements. Okay, and this is an example of what the entries from this text syllable alphabet B are like from different parts of the list. By reading them out loud, you can get a sense that the emphasis here was less on the meaning and more just on the sign itself. Ba-ba, ba-a, ba-ba-a, ba-mu, ba-ba-mu, ba-ni, ba-ni-ni, ba-ni-a, ba-ni-mu. There's not much to learn here, but there's something very memorable about the sequence. You can memorize this pretty easily and they repeat the same principle throughout the list with ma-ma, ma-a, na-na, na-a, right? So once you have all of that learned, you can pretty much practice this from memory without really having to remember what the text means. You just have to remember what the sign is that begins the new sequence, and that way you can focus more on executing the sign and less on reconstructing the meanings of the signs. This is a group of exercise tablets, type two tablets with the same extract on the obverse and reverse from house F from syllable alphabet B, so it was also copied in this house, presumably by a group of three scribes in this case learning the same text at the same time. And to further illustrate the idea that they were learning how to write as opposed to learning the meanings of the signs, you can get a sense of how unskilled and how unadvanced their hand is at the beginning of their training by comparing what the sign they're imitating looks like compared to how the teacher, the more expert scribe writes it. This is the sign ni, followed by the pupil's attempt to write it directly. You can see this more clearly with close-ups from other extracts on the reverse of a syllable alphabet B. This is the same sign ni. There's a blown-up sign of a well-executed version of that. Here's the beginning scribe's attempt to replicate it, so the angularity is not there. The slant's really down. This is just a horizontal wedge that comes almost vertically across the bottom. If you look at this, I put the way these lines are supposed to look next to it. So this is the za sign, which has four vertical wedges. This is the scribe's attempt to write it. If you didn't know what was supposed to be there, you wouldn't know that it was that sign at all, right? And if you look further down, this is supposed to be the exact same entry, right? And you wouldn't be able to read that either. You can see bits of the mu, right? The wedges that make that up, but a lot of it is missing. And if you look really closely at what's happening here, the scribe not only seems to have failed to execute the za sign correctly, but seems to have anticipated the occurrence of this sign la later in the sequence and written it first, and then written the correct sign za over it, right? And what you gather from this is that this is still about learning the basics and mechanics of the manual execution of sign writing with this list. Similar examples of this come from the next text in the scribal curriculum, a personal name list known as Ananatesh. And this is an excerpt from a literary text that connects syllable alphabet B with a personal name list, Ananatesh. Ananatesh uses a lot of the same basic signs that you find in syllable alphabet B, but uses them to illustrate how these signs are used for writing popular personal names from this period. And one thing that you'll notice when you look at this list, particularly if you know these languages, is that it's bilingual, right? These are Sumerian names, and these down here are Akkadian names. So when we talk about scribal education being primarily about teaching a dead language Sumerian, it should be kept in mind that a lot of these earlier exercises are either not in the language at all because they're focused on writing signs, or they're in fact bilingual, and the writing of personal names is an imminently practical skill. Just about every scribe, regardless of what he or she did after the completion of their training, would involve writing names, right? So when they're beginning to actually learn how the writing system works, and they're continuing to practice how to execute the signs, they're learning a practical skill along with it, and the bilingualism of their education is already apparent in the very earliest stages of their education. This is just to give you some more examples of where they're at with the development of their handwriting. So the copy on the left with the circled sign is from a copy of the text written by a teacher. This is a Nanitesh, and this is from another source where the beginning scribe is attempting to write the same entries. You can see that the E sign here is not well executed, and it just gets worse with some of the other examples. This is the Li sign here. It's got most of the component parts, but the proportions are not right, and in this particular case, you have the Tur sign, which is a fairly difficult sign, and the scribe just went on strike with it, basically. You could either try to write it poorly or just not try to write it at all, and in this case, that was the easy way out and the preferable one, right? Good, so moving from the very beginning of scribal education to the very end and returning to the Sumerian literary compositions, we noted that for very early education, there were two distinct tablet types, these so-called lentils that have short extracts and the type two tablets that we've been looking at. When you get to the more advanced literary compositions like the text from the Deccad, you no longer have those tablet types anymore. You rarely find school lentils with extracts from those compositions and almost never type two tablets either. What you find instead are different tablet types. We call these extract tablets, and just like the name suggests, to illustrate the scale with probably the most famous person in our field, if not the most famous person ever, the curator from the British Museum, Irving Finkel, who is a YouTube version of the talk he gave here in 2015, I've told, has exceeded 35 million views. I don't anticipate that this lecture will come anywhere near that, right? But for those of you playing along at home, you didn't get Irving Finkel tonight. You got me, but you're getting Irving Finkel at least virtually, right? This is just to give you a sense of the size of a type three extract tablet. They're really quite small. They're about as large as his beard, which I think he had just trimmed, right? And then you have another distinct tablet format from this advanced stage of education, which are the multi-column tablets, which generally for shorter compositions contain the entire text as opposed to extracts from them. And what happens when you start doing things like looking at the average number of lines on the extract tablets for the text in the deckhead, what you find is that the number of lines on each of the extract tablets for these individual compositions equates to about 25% of the length of the entire composition. So what that means is they were using the extract tablets to learn four quarters of a longer composition in parts. They would practice copying each quarter, and then once they were done learning each quarter of it, they would practice writing the entire text on the multi-column tablet. And you can sort of get a sense of this by looking at the total number of preserved multi-column tablets or type one sources for each of these texts and multiplying, dividing the number of type three extract tablets for each of these sources by four to get the sense of the number of groups of four for each of these texts. And generally, the number of groups of four sources for each text is about the same as the total number of preserved multi-column tablets. So if you were thinking about a scribe going through each sort of stage in the process of learning an individual literary composition they would practice the text in quarters. Once they'd done that enough to be competent in it, they would produce one single multi-column tablet, okay? Another way we can get at how they were learning these types of texts, so if you recall from the very early stages of scribal education, they were copying directly or they were trying to reproduce the shape of the signs, the executions of the signs from memory. We have further evidence that at this stage of scribal education, the advanced stages when they're learning literary compositions, that they were copying them entirely from memory. It's a bit too complex to get into all of this now, but when I was doing this work 15 or 16 years ago, I was drawing a lot on memory studies of verbatim recall of longer texts, so cognitive psychologists have worked quite a bit on this to get a sense of how the memory works and what types of mistakes people make when they try to recall any type of prose text from memory. There have also been studies of this done in other disciplines, in medieval studies and things like this, trying to find typologies or diagnostic criteria for identifying when a text was copied from memory as opposed to being copied directly from another written source. This is summarizing those criteria drawn from all of these types of studies and others. So one thing that's diagnostic or that occurs frequently in sources copied from memory are conspicuous omissions, additions or substitutions, and also, this is the other key one, elements, lexemes or expressions that are erroneously transferred from one context to a similar context in another part of the text. So basically, when two lines or words in a text are similar and appear in similar contexts, these passages are more likely to be confused with each other than passages that are radically different. So the mind tends to remember things that are very distinct but tends to confuse and misremember things that are very similar is another way of looking at that. So when we look at the types of errors that occur in the sources for Sumerian literary compositions and these are all sources from house F, you can find all of these different types of memory errors. The form on the left is the form in the source from house F and you see it's omitted all of the verbal infixes and elements that are in the middle there. You get entire phrases omitted. Here you get the end of a noun and the prefix from a verb. You find substitutions and you find additions of grammatical elements that make no grammatical or semantic sense when you look at the passage in which they occur. So it's almost as if the scribe were trying to remember what was supposed to be there and either forgetting it entirely or just making something up when he or she couldn't recall what the right line would be. The smoking gun examples of these are the so-called long-distance memory transfer errors and these are confusion of similar lines and forms from entirely different texts. This text here, Sholgi A, is the first text in the Deccad. Lipid Ishtar A is the text that was learned immediately after it but it's an entirely different text. They both contain a line that's almost identical. So you have the same subject, the god an, the same verb which is to make firm and you have something involving a crown being put on the head but the type of crown that's being put on the head is different in each case. Here it's a pure crown, here it's an august true crown and here you also have a modal prefix that makes the verb conditional and here you don't have it at all. These are all sources for Sholgi A that have this line and if you'll note these two sources down here at the bottom, the mistake that they make is to write the line from Lipid Ishtar A line 24 instead. So they've kept all of it the same but put this in the wrong spot and when we go to the same line in Lipid Ishtar A line 24, you'll note that two of the sources from Nippur this time accidentally write the adjective from the earlier similar line in Sholgi A in the same verbal form from the same text and this is unlikely to happen if a scribe is copying directly from another written source because it's coming from memory and the scribe at that particular moment is misremembering and confusing it with a similar attestation somewhere else. Okay, oh and the other thing is because some of these diagnostic memory errors can also occur in sources copied in other ways including directly from another written exemplar, what you need to do is collect all of the variant types that occur within individual sources to get a variant profile. And when you find multiple examples of diagnostic memory errors occurring within the same source, the argument is that that increases the likelihood that that source was copied from memory and that the error that you're identifying as a memory error isn't just an isolated variant that occurred by accident or by some other means, okay? All right, so returning to the question that was posed at the beginning of the talk, why did they learn? And this is another quote from Nick Veldhuis, one of the pioneers in the study of scribal education. The teaching of Sumerian in the Nippur Aduba was not guided by the list of skills a future scribe had to master. The lack of attention to Akkadian and the overdose of highbrow Sumerian point in another direction. It seems that handing down the Sumerian language and tradition as completely as possible was considered to be all important. A pupil of the scribal school was introduced to the technique of writing, but more importantly, he was introduced to the heritage of Sumerian writing and Sumerian poetics. This claim, which was first put forward in 1997 in a groundbreaking reconstruction of elementary education at Nippur is now the consensus view and has been adopted in nearly every treatment of the subject since. Although not stated explicitly, this argument could be understood to imply that literacy was limited and that the ability to read and write was also only available to a small privileged group. This view is based on two assumptions. One is that Sumerian literature was learned primarily as a means of acquiring elite cultural capital. The other is that the acquisition of this knowledge took precedence over learning more basic and practical skills pertaining to writing. However, the number and distribution of the lists and texts from different stages of the Nippur scribal curriculum suggest that the emphasis was very much reversed. The first and only partial edition of Syllable Alphabet B, one of the earliest exercises in the sequence of scribal education, was published in 1959 and was compiled from a selection of 300 sources in the Archeology Museum in Istanbul in a smaller group of additional sources previously published in hand copies. In the time since, over three times as many new sources have been identified, bringing the total to close to 800. By comparison, and you can see this visually on the scribe, this is just a selection of the number of preserved sources for text at different stages of the scribal curriculum with Syllable Alphabet B up to 800 here on the left at the beginning. The average number of sources for the other lists and texts that were copied as scribal exercises is between 50 to 100, with many having closer to 10 and 20, and only a few with more than 100. And if you look at where the literary texts are, they're right down here, right? So what this seems to indicate is that when scribes learned how to write, they spent a lot of time doing this, and very few of them got to the point where they started copying more advanced literary compositions. So if we're going to reduce all of scribal education to learning esoteric knowledge about a dead language like Sumerian, we also have to keep in mind that it looks like, on the basis of the distribution and number of sources, that these exercises were the exception to scribal education and not the rule, and that the norm was to do something more practical and learn basic handwriting, and how to use elementary signs and basic cuneiform signs to write practical words like personal names and vocabulary and this sort of stuff, right? Further evidence for how and why syllable alphabet B was taught can be obtained by examining and comparing the quantity and distribution of extracts from the list on the obverses and reverses of type two sources. And this is an example of how these extracts are divided. This is not the whole list, but they follow a very basic and immediately apparent principle. So they start with the same sign, and then the signs that come after it tend to vary, but also follow a predictable pattern, and they tend to form groups of between nine to 15 entries, and each of these would be an extract from the list. When you get to the end of an extract, you get to the next set, which begins with a new sign and has slightly variant forms of the signs that came after it. The preserved part of the list was divided into 29 sections with six to 14 entries per extract. This is an example of one of these extracts on the obverse of a type two tablet. The model extracts on the obverses consistently contain the same groups of entries, and the number of copies of each group is distributed evenly across the list. So as you can see from the slide, which gives you most of the different sets of extracts, starting with the beginning and going all the way to the end, this is the number of copies of each of these extracts. It's relatively even across the list. There are model extracts for nearly every section of the list, and the extracts for each section begin and end with the same entries. Furthermore, as can be seen from the slide, there are between two to 20 copies of each extract, and the range of the number of copies per extracts is approximately the same for each part of the list. This suggests that the list was taught in its entirety, and that each section of the list was learned with equal frequency. By contrast, when we turn to the reverses of the type two tablets with the same list, Syllable Alphabet B, we find a different pattern. This is an example of the reverse of a tablet with the beginning of Syllable Alphabet B. There are 377 type two sources with excerpts from Syllable Alphabet B on the reverse, and of these, at least 197, or over half, have model extracts from the same list on the obverse. As illustrated by the table on the slide, 151, or 75% of these, have excerpts on the reverses that begin with the very first entry from the list. This is the number of extract tablets that have Syllable Alphabet B on the obverse and reverse that have extracts on the reverse that begin with the first entry. They're repeating the beginning of the text over and over and over again, and only in rare cases do you get extracts on the reverse that begin with later and more difficult sections from the list, and this is showing you here. That it is only with the 13th division, entries 123 to 130, that there are excerpts that begin with later divisions. Moreover, as can be seen from the table from the second division on, there is a general tendency for the excerpts on the reverse to begin with increasingly later divisions the later the division of the model extract on the obverse. Excerpts beginning with entry 41 occur for the first time with model extracts from the 15th division, entries 146 to 147, and excerpts beginning with entry 229 with model extracts from entries 294 to 308, with an equally incremental increase for all of their divisions in between. In addition, in the many sources that have excerpts with the beginning of the list, the length of the excerpt increases with each new division of the model extract on the obverse. Sources with model extracts of the first division have excerpts with 20 to 30 entries, in which the first nine entries are repeated two to three times, but beginning with the fourth division, entries 33 to 40, the lengths of the excerpts increase to an average of 42 entries, and by entries 103 to 116, the excerpts have as many as over 100 entries. So basically what happens is when they're learning the very beginning of the list, they're also copying a lot less of the list on the reverse. But as they get further along in the list and learn more and more advanced signs, they copy later sections of the list and more entries from those later sections. So they're practicing a lot more and the things they're copying are a lot more difficult progressively. These observations about syllable alphabet B and its sources suggest the following conclusions about its pedagogical function. The content and structure of the list provide clear evidence that it was an elementary sign exercise intended to teach the first and most basic aspects of how to write cuneiform. The exceedingly large number of preserved sources in comparison to other lists and texts with over 2 3rds of the known type 2 tablets from Newport containing syllable alphabet B strongly suggest the list was not marginal, but one of the most essential and frequently copied exercises in the scribal curriculum. Since the primary purpose of type 2 tablets was to provide practice visually copying model texts and manually reproducing increasingly longer previously learned excerpts, the pupils who copied syllable alphabet B acquired practical skills in writing and not elite cultural knowledge of the long dead Sumerian language. Moreover, returning to where we started at the beginning of the day in the life of our Mesopotamian student, the emphasis on learning handwriting as opposed to acquiring elite cultural capital from Sumerian literary compositions is also emphasized in the text school days cited earlier in the talk where the student is being chastised for having handwriting that is no good, right? And indeed to cite a Sumerian proverb from the same period as these exercises, a scribe without a hand is like a singer without a voice. Yeah, and I think on that, I think we can conclude.