Rebecca Hasselbach-Andee | Language and Cultural Contact: The Case of Ebla

Welcome to all of you. It's very special that finally after such a long time we are able to have a real audience again here in Breasted Hall, but I also of course welcome all those and especially the members of the OI who are joining us tonight online or remotely. It's a distinct pleasure to introduce tonight's speaker, my colleague here in the Oriental Institute, Rebecca Hasselbach-Andy. Rebecca joined us, the University of Chicago and the Oriental Institute in 2005. She came from Harvard at that moment where she had a year earlier in 2004 received her PhD and before that she studied at Johns Hopkins University and in her native Germany Rebecca is a linguist, as you will clearly experience tonight. As such, she speaks many languages, modern and ancient. She speaks Italian, French, modern Hebrew, modern Arabic, and I'm not going to list the ancient languages, but I can absolutely assure you that whatever ancient language she will mention tonight in her lecture, she knows. Rebecca is also one of our most prolific authors here at the OI. She has three books to her name. The most recent one, which came out this year, I think in 21, is a co-authored grammar of Ge'ez, an ancient Ethiopic language. She also is the editor or co-editor of four edited volumes, two of which will come out shortly, and she has written over 40 scholarly articles in our fields and in the fields of the lecture that you will hear tonight by her. Finally, we thought it very appropriate to have Rebecca give the first in-person members lecture since we closed for the pandemic, because she was just recently promoted to full professor here at the University of Chicago. So it gives me great pleasure to announce here her lecture, Language and Cultural Contact in the Third Millennium BCE, the Case of Ebla, and I hope you'll join me in welcoming Rebecca here to the stage. Thanks so much for the kind introduction, and thanks so much for having me here and inviting me to give this members lecture. It's a great honor. So today I would like to talk about a project of mine that is leading us to the third millennium BCE, and more specifically, I'm interested in the complex relationship between northern Syrian city of Ebla and Mesopotamia in the 24th century BCE. My primary interest is on a linguistic level, and what I would like to find out is how exactly the language that is used in the texts found at Ebla relates to Akkadian, the language used in writing in Mesopotamia for about 2,500 years. The situation is quite complex and is not solvable by simply looking at the languages themselves, without recourse to cultural, economic, and political contacts between northern Syria and Mesopotamia during the time. What I would like to present to you today is the beginning of a more synthetic approach that considers the issue of the language of Ebla, commonly called Eblaite, within its ancient Near Eastern context. I will begin with a general introduction to Ebla and the categorizations of its language that have been proposed so far. After that, I will talk about the general issues we face when attempting to figure out what people in ancient societies actually spoke, and then I will give some data points and see how contact with other areas and cultures might help us better understand the situation at Ebla. And lastly, I will provide a proposal of what might be the linguistic situation at Ebla. The site of Tel Mardik, ancient Ebla, which lies about 65 kilometers south of Aleppo, was excavated from 1964 onwards. The first texts were found in 1974, with most tablets that are known today having been excavated a year later. The excavating team found about 17,000 fragments of inscribed clay tablets, which add up to roughly 4,000 texts. These texts are written in a type of cuneiform that resembles the one found in contemporary Mesopotamian sources. The texts mostly come from archives found at the Royal Palace G. Let's see if this actually works. Royal Palace G, right on the Acropolis there. And date to a short time period in the third millennium BCE. More precisely, it is assumed that the archive covers roughly 50 years, from about 2370 to 2320 BCE. Based on this time range, the Ebla archive is contemporary to the late pre-Sargonic periods and early Sargonic period in Mesopotamia. During the period of the archives, Ebla was ruled by three kings, Igris Halep, who ruled about 12 years, Irkab Damu, who ruled about 11 years, and Isa Damu, who ruled about 35 years, who were the last kings of their dynasty. Many of the texts further mention the main officials or viziers of these kings as the persons in charge or the person addressed. The best-known viziers or ministers were Arukum, who served under Irkab Damu for five years. Then we get Ibrium, who served in that role for about 18 years and was appointed by Irkab Damu during the last years of his reign and then continued under Isa Damu. And then we have Ibrium's son and successor, Ibezikur, who served for about 17 years under Isa Damu. It is not clear when writing was introduced to Ebla, but the most common theory is that this happened under Igris Halep and his vizier Arukum, that is not too long before the first written evidence from the archives. The records of the archives were preserved together and mostly in situ because the palace was actually destroyed, probably by a king of Mari during the reign of Isa Damu. The city itself was not destroyed at this time, however. We know that because Sargon of Akkad boasts of having conquered Ebla in the same campaign in which he conquered Mari, which was about 10 years after the destruction of the royal palace at Ebla. Sargon did not destroy the city either, though. Sargon's grandson Naram-Sin takes credit for that. The text of the Ebla archives confirm this historical reconstruction. None of the records in the palace archives of Ebla mentions any Sargonic king or their campaigns against Ebla, which is why the palace of Ebla must have been destroyed before Sargon's and Naram-Sin's campaigns. Ebla was resettled after Naram-Sin's campaign and remained inhabited until what would be roughly the end of the old Babylonian period of Mesopotamia. The final destruction of the city happened around 1600 BCE and was possibly the deed of a coalition of Hittites and Hurrians that were led by the king Homo Sili I. The period that concerns us today is what is called the mature early Syrian period. It's about 1400 to 1300 BCE. During this time, Ebla was an important town of northern Syria. The texts show that it held political and economic relations with southern and northern Mesopotamia, especially with cities such as Mari on the upper Euphrates, Naga, Kish, and Ur. Although it is not specifically mentioned in the text, the archaeological remains also indicate some relations with Egypt. The excavators found stone vases that were made in Egypt and that bear the names of Khephren of the 4th dynasty and Pippi I of the 6th dynasty, and you have the vase of Pippi I on the slide. Artifacts from Cyprus and lapis lazuli from modern-day Afghanistan were found at Ebla as well, and you see the lapis lazuli on the slide as well. As we can see from this evidence, Ebla held an important position on the trade routes between Mesopotamia and the Mediterranean, which is one of the reasons why it was repeatedly targeted by the zirconic kings of Mesopotamia. So let's go over to the textual material. The texts from Ebla cover various genres. The bulk of the material represents administrative texts. Another significant subcorpus comprise chancellery texts and letters. These texts document the diplomatic relationships between the king of Ebla and other rulers. There's further a small group of incantations that seem to have originated in Syrian scribal schools. Lastly, a bilingual Sumerian Eblite lexicalist has been found, which is attested in several versions. This text constitutes the earliest bilingual lexicalist known so far. The diplomatic and administrative texts that were discovered show that the kingdom of Ebla, and seemingly also its language, covered a significantly larger territory than simply the city itself. At the peak of its existence in the 24th century BCE, Ebla's territory included cities such as Alalakh, Karkemish, Hamad, and Tunip, and you see the extent on the map on the slide. As I mentioned before, it is likewise well established that Ebla had extensive trade and political contacts beyond its own territory, reaching to Egypt in the south and Mesopotamia in the east. It seems to have had particularly close ties to Mari and also northern Mesopotamian Kish, since the exchange of people and goods with the cities is well attested. Kish is down there. Ebla was thus not an isolated city-state, nor obscure little town in the middle of nowhere. It was incorporated into a much larger economic and political network, which it partially controlled. It also had substantial political influence over its surroundings and as far as Maring during the period of the archives. Important for our purpose is further that the network Ebla was part of included cultural exchanges as well. It is undisputed that Eblaite borrowed its writing system from Mesopotamia. Another indicator of the close cultural connections between Ebla and Mesopotamia is the fact that some texts at Ebla, including the monolingual lexical lists and some incantations, are in fact written in Sumerian, a language that was undoubtedly not used natively in Syria. The importance of this cultural and economic contact to Mesopotamia cannot be overstated. In my view, it is also important for the evaluation of the linguistic situation in Ebla. With that, let us now turn to the language of Ebla. The investigation of the exact relationship of Eblaite to the other Semitic languages attested during this period, Akkadian, has not yielded clear results so far. Let me briefly explain what has been proposed and what issues there are when trying to come to a conclusion on the subject. Since its discovery, numerous suggestions have been made as to the place of Eblaite in the Semitic language family. The Semitic language family is divided into two main branches, East and West Semitic. East Semitic quotates Akkadian and West Semitic everything else. In West Semitic, we have several other branches. Ethiopian Semitic, which contains the Semitic languages, classical and modern, attested in Ethiopia. We have modern South Arabian, which are languages spoken today in the southern parts of the Arabian Peninsula, that is in Yemen and Oman, and on the island of Socotra. And then we have a branch that is called Central Semitic. Central Semitic is divided into a branch called Old South Arabian, which is a mix of four main languages attested in inscriptions from about 800 BCE to 600 Common Era and that were found in today's Yemen. Then we have Arabic and a branch that is called Northwest Semitic. The latter includes Aramaic, a branch called Canaanite. Hebrew is a Canaanite language, for example, When Ebla'id was discovered in 1974 to 1975, many scholars started working on the textual material immediately and came up with various assessments as to where this newly discovered language might fit into the family tree. Already in 1975, Patinato, the epigrapher of the excavation, based on a small sample of text that was still poorly understood, claimed that Ebla'id was different from Old Akkadian and that it should be classified as Northwest Semitic or, more specifically, Old Canaanite, that is, as a close relative to Hebrew. Although this analysis was initially followed by some scholars, it has subsequently been abandoned with the publication of more material and a better understanding of the writing system and language underlying the texts. Therefore, Ebla'id is not closely related to Hebrew, contrary to what has been picked up by the media in the early years after its discovery. This might suggest that Ebla'id should be considered an East Semitic language, but the classification of Ebla'id as a member of East Semitic was not as straightforward as it might seem. The vocabulary, in particular, undermines the classification of the language as East Semitic, since it exhibits many lexical items that are known from West Semitic languages but not from Akkadian, although it also shares a significant number of lexical isoclases with Akkadian. I will come back to that later. Its phonology and morphology, on the other hand, seems to be close to Akkadian, that is, East Semitic. Because of the seemingly mixed character of the language, some scholars, such as Kaplis and von Soden, suggested that Ebla'id might represent an independent branch of Semitic, an additional third branch besides the two traditional East and West Semitic ones. This approach, again, did not attract much support, despite the fact that it was one of the few that tried to address the seemingly contradictory features of Ebla'id. This only left an affiliation of Ebla'id with East Semitic and, in fact, ever since Krebernik's work on the language, in particular his article from 1996, the subgrouping of Ebla'id as East Semitic has been fairly firmly established. Krebernik was not the first to suggest this type of classification, but he was one of the first to provide sufficient and convincing evidence in its favor, based primarily on Ebla'id morphology. With Ebla'id being understood as East Semitic, the question remains of its exact relationship with Akkadian. In effect, there are two main possibilities. Ebla'id can a. be understood as an independent language of East Semitic or b. as a dialect of Akkadian. Both approaches are found in the literature and it is precisely this part of the classification of Ebla'id that has not yet been settled. Proponents of the idea that Ebla'id is a dialect of Akkadian include scholars such as Dombrovsky, Etzart, Parpola, and Krebernik. Dombrovsky, for example, states that the Ebla'id material reflects the earliest dialect or group of closely related Akkadian dialects. Parpola, although also assuming a close relationship between Ebla'id and Akkadian, proposes that there existed an early form of Semitic that centered around the city of Kish, called Kishite or Kishik, that developed different variations. In the northwest, this Kishite underwent significant influence from West Semitic languages and eventually became what we know as Ebla'id. While in the south, this language was exposed to contact with Sumerian and developed into Old Akkadian. Krebernik, based on an investigation of several linguistic features in the aforementioned article from 1996, equally considers Ebla'id an early Akkadian dialect, although he stresses that this dialect was not simply imported from Mesopotamia, but presumably shows that Akkadian was spoken in a much larger territory in the third millennium BCE than usually assumed. There's a subgroup of scholars among the proponents of analyzing Ebla'id as an Akkadian dialect who suggest that what we are dealing with in this case are not so much distinct dialects but the dialect continuum. This idea ultimately goes back to Garbini, who formulated it in more general terms as a Paleo-Assyrian dialect continuum, of which Akkadian is a geographically peripheral variety. More recently, Rubio has expressed support for a modified version of this idea by proposing that during the Pre-Sargonic period, Akkadian constituted a dialect continuum from Babylonia via Assyria all the way to Ebla'. Lastly, there are those scholars who think that Ebla'id constitutes an independent branch of East Semitic, that is, a language not directly related to Akkadian. These scholars include, for example, Hünergaard and Tropper. Tropper assumes that most differences between Akkadian and Ebla'id go back to innovations in Akkadian. In other words, Ebla'id is a more archaic language than what we know as Akkadian, while Hünergaard, based on several innovative features, suggests that Ebla'id constitutes an innovative branch within East Semitic. Thus, although Tropper and Hünergaard agree in their general assessment that Ebla'id and Akkadian are distinct languages, one sees Akkadian as the more innovative one, while the other assumes that Ebla'id is more innovative. From this overview of competing theories, we can see that there is no consensus regarding the position of Ebla'id within East Semitic. Even scholars who agree in their overall classification can differ quite a bit in the details. As much as the aforementioned scholars vary in their opinions, they have a few points in common as well. Pretty much all of the aforementioned scholars view Ebla'id as being native to the city of Ebla. That is, to put it simply, they assume that the language reflected in the text is the language people natively spoke at Ebla, at least in some capacity. Furthermore, all of them consider Ebla'id to be the product of genetic processes. That means it has developed out of a common East Semitic ancestor along with Akkadian, without any influence from other languages. Only Parpola considers the possibility that language context might have played a role in the development of Ebla'id. These are two very important points. In order to assess the position of a language, we need to know how and in what context it was used. That means we need to know if it was primarily or even only used in writing and or if it was spoken. If it was spoken, we need to understand by whom. That is, whether it was spoken by the majority of the population or only by a minority such as an educated elite. A question that has not been answered satisfactorily so far is whether Ebla'id is indeed the native language of Ebla or whether it was imported together with the writing system. We also need to consider the question of language contact more closely. In the case of Ebla'id, which as I already mentioned has a rather confusing set of mixed features, some that conform to West Semitic and others that conform to East Semitic, this is an important issue. Mixed systems like this can of course develop naturally within a dialect continuum but there is also the possibility that they are caused by an actual mix of languages that are in contact with each other. Before we get to this in more detail, I would like to say a few words about why it can be so difficult to answer these questions though. Let's start with the issue of knowing what people actually speak. At first sight, this might seem fairly straightforward. Native speakers of English in the States most likely write English when they write something for everyday purposes, not let's say Italian, Korean or Killing on. The receipts we get in supermarkets and stores are in English. When I get emails from my colleagues, they are in English. I mention receipts and letters because a lot of the administrative and chancellery texts we have are receipts and letters. So the correlation between written and spoken language seems to be fairly direct, at least from an English speaking point of view and in this country. There are many people, whole societies in fact though, who do not write what they speak. Even with English, the correlation between writing and speaking is more complicated than I just described. Let's take me as an example. Although I obviously speak some degree of English, it is not my native language. By now, after years in this country, I prefer to write in English rather than in German though. Even when I write personal emails, as some of my German colleagues might attest to, I prefer English because my written German is just plain clumsy. That doesn't change the fact that I speak German on a daily basis and I prefer German as a spoken idiom over English. I mean prefer in the sense of it feeling more natural, not that I like it better. So what I write is only partially what I speak. This is also true for some native English speakers who speak different dialects from what is perceived as the written standard. Speakers whose variant is different from the standard language have to adjust how they write and often also how they speak. There are other scenarios in which people in essence write different languages than they speak and that is more common than you would think. A textbook example of such a situation is that of Arabic. In Arab countries, what is spoken is what is referred to as Arabic dialects. What is written and used in mainstream media is a very different version of Arabic called modern standard Arabic or Hussan, which is based on classical Arabic. Nobody speaks modern standard Arabic as a native language. It must be learned similar to a foreign language in school. With social media, blogs and the like, the situation has changed more recently and dialects are also written increasingly. But just a few years ago, the only proper form to write Arabic was modern standard Arabic. Just imagine you speak any form of today's English, but for writing you have to use the style and grammar of Xhosa or for some even Beowulf. Situations where people write a different language than they speak are also known from the ancient Near East. A good example is that of Nabataean and that's one of the languages I actually don't know. The Nabataeans were Arabs who established an important kingdom in northern Arabia and the southern Levant around the 4th century BCE that remained independent until it was annexed into the Roman Empire in 106 Common Era. The capital of this kingdom was the famous site of Petra in today's Jordan. It is well established that the Nabataeans were indeed Arabs who spoke a variant of Arabic. This is evident, for example, in personal names that are attested in the Nabataean textual material and certain traces of Arabic in the same texts. The language they wrote, however, was a version of Aramaic, not Arabic. Why did the Arabs of the Nabataean kingdom write a different language from their native one? This is the result of historical factors. During the Achaemenid Empire, Aramaic was widely used as a lingua franca, that is, as the language of international and often also local correspondence. The Achaemenid Empire, in fact, is another example of a period where at least some of the population, including the ruling elite, the ruling dynasty, which was from Persia, used a completely different language in most of their written correspondence than their own. During the Achaemenid Empire, Aramaic spread widely across the ancient Near East, including to the area of the Nabataean kingdom. The Arabs of that area took over the writing system and with it its most commonly used language, Aramaic. By the time of the Nabataean text, the Achaemenid Empire had lost its influence, however, and the Aramaic of the area had developed its own features that make it distinct from the standardized form of the language that was used during the Achaemenid Empire. Important for our discussion is that the use of Aramaic continued for quite some time among the Nabataeans, despite the fact that they were Arabs and Arabs speaking. Examples of ancient Near Eastern peoples writing the most commonly used culturally or politically dominant language while evidently speaking something else are actually not unusual. When we first have evidence for Akkadian, it is primarily in the form of personal names that appear in texts that are written in Sumerian. These names start to appear around 2600 BCE, only from 2350 BCE on do we find texts that are clearly and syllabically written in Akkadian. The Semitic or Akkadian speaking scribes thus wrote Sumerian for several centuries before switching to primarily writing Akkadian. In the case of the Akkadian and Sumerian interaction, it is likely that at least the Semitic scribes who wrote Sumerian were bilingual. It is not always easy though to find out to which degree scribes knew the language they wrote. What I mean is whether they were fluent bilinguals who used both languages almost to the same degree of fluency if they were partially bilingual, that is they only knew how to express certain topics in both languages, or if they only knew how to write in one of the languages but didn't speak it. The letter for example is most likely the case with the scribes who wrote the canonizing letters from in Egypt that date to the 14th century BCE. These scribes wrote a rather idiosyncratic version of Akkadian that shows a lot of influence from their native Canaanite language. It is likely that the scribes did not speak Akkadian but learned it from non-native speakers just for the purpose of writing international correspondence. To truly understand what is going on with the language, we need to know how fluent a speaker is and in what context the language is used. Unfortunately the texts, ancient texts, do not tell us anything about this. Nowhere does any text mention what languages a scribe knows or whether he is fluent in one but not as fluent in the other. So why is this important? Because depending on how somebody learns a second language and how they use it, we see certain trends in how it affects the language that is learned or spoken and what parts of the grammar are influenced by the dominant language. That is, this influences what type of features might be borrowed from one language into the other. For example, if we primarily get loanwords or if there are changes in the sound system or phonology in the syntax or if we also get changes in the morphology, vice versa. If we see only loanwords in a given language or primarily structural changes, we can deduct what kind of contact existed at least to a certain degree. So to return to our topic of Ebla'it and its classification in East Semitic, as I alluded to already, Ebla'it exhibits a system where parts of its grammar, especially the morphology, corresponds to Akkadian. While the lexicon is mixed, it partially corresponds to Akkadian and partially to West Semitic languages. One possibility is that this system is the result of language contact. Since we do not have native speakers, we can ask about what they speak and whether that differed from what they wrote. The way to figure this out is to look at every aspect of the language, the phonology, morphology, syntax, lexicon and also personal names, in order to see if we can find traces or features that are unusual or that clearly do not fit into the same system. That is, that might be borrowed from a different language. And in the following, I will give you a few examples of Ebla'it features and the mixed character of Ebla'it grammar. I will start with some typical East Semitic innovations that mark this branch as distinct from West Semitic, so that it is clear why Ebla'it is classified as East Semitic. And there are plenty of these features, especially in the pronominal system. So I will only show you a few, don't worry. I will also list the Akkadian, Old Assyrian and Old Babylonian forms for comparison. So in the pronominal system, the pronominal system of East Semitic underwent some important innovations. There are new forms coming in. In particular, Akkadian innovated full paradigms of accusative and dative independent pronouns and corresponding pronominal suffixes. Other Semitic languages only have accusative or bleak pronouns. Ebla'it shares this innovative accusative and dative paradigm. The independent accusative pronouns attested in Ebla'it, which are marked by T as in Akkadian, are given on the slide. So you see, you have Kuwati, which is marked by the T. And the same forms you also have in Old Assyrian, Kuwati, that's the same as Kuwati, just believe me, that just dropped the W. And Kata is also derived from exactly the same form. And the same is true for Niyati. That's actually the same form in all, in Ebla'it and all Akkadian dialects. I should mention that Ebla'it only rarely has Y forms. The pronouns I just listed there are the only variants that occur in Ebla'it text, meaning there are no forms without T. The dative pronouns attested in Ebla'it correspond to the Babylonian, not the Assyrian forms, that is, they have infix S. And you see on the slide that you have Kuwasi. Old Assyrian has Kuwati with a T. But Old Babylonian has Kashim and Kashum, which corresponds to the S sound in Ebla'it. So those are etymologically the same ones. The correspondence with Babylonian in this case is remarkable, since these forms constitute relatively late innovations in Akkadian. One of the most notable Akkadian or East Semitic innovations is the declension of masculine plural attributive adjectives, which, unlike in any West Semitic language, have the ending utum, as in sharu, danutum, strong kings. This innovation causes masculine plural substantives and adjectives to have different pluralizing morphemes, u as in sharu, kings, versus utum as in danutum, strong. Ebla'it has the same ending for these types of adjectives as Akkadian and thus shares this important innovation. There are many more morphological features that Ebla'it shares with Akkadian. These features reflect a very solid picture of Ebla'it as East Semitic and very closely related to Akkadian. There are even some Babylonian type innovations that are not shared with Assyrian, Assyrian being the linguistically more archaic dialect of Akkadian when compared to Babylonian. Let us now have a look at some features that Ebla'it does not share with Akkadian. When we look at the morphology, there are numerous forms that have no correspondence in Akkadian. Some of these are inherited from Proto-Semitic and are thus not indicative for evaluating the position of Ebla'it in East Semitic, but others are clear Ebla'it innovations. Features that are retentions from Proto-Semitic include, for example, the presumed form of the first communosingular interbender pronoun i, which is commonly analyzed as ana and ebla'it, and you see that Akkadian has anaku. In this case, both the Akkadian and Ebla'it forms most likely go back to Proto-Semitic, so each of them would have simply leveled a different form. As for Ebla'it innovations, one of my personal favorites is the second masculine plural independent pronoun antanu, which has no parallels anywhere else in Semitic. This form is based on the second masculine singular pronoun anta, u, which has been pluralized by the morpheme nu, which is also found in the third masculine plural and first communes plural in Ebla'it. So you have antanu in Ebla'it, but as you see in Akkadian, you have attunu, which is a different form. As a particular quirk of Ebla'it, dual pronouns and suffixes in the second and third persons have the consonant m, as in the second communes dual kumain, and third communes dual sumain, while plural second and third person pronouns and suffixes have an n, as in the third masculine pronoun sunu. This type of split in the distribution of consonants between dual and plural forms is otherwise unknown in Semitic, at least as far as I'm aware. Suffice to say that the mix of m and n in the pronominal paradigm of second and third person plural and dual forms is innovative in Ebla'it and unusual for Semitic. There's one more morphological set of innovations I wish to mention, which also constitutes unique Ebla'it innovations, namely the form of t-stem infinitives. The t-stem infinitives attested in Ebla'it have both a prefixed and infixed t, so if you look at the list you see, for example, the gt-taptarism, where you have a t at the beginning and also a t in the middle of the word. In Akkadian, you don't have that, you have only the one in the middle, pitrus, there's no t at the beginning. These forms are primarily attested in the bilingual lexical list, but not exclusively. Hecker suggested that these unique Ebla'it forms can be explained based on verbal nouns that are attested in other Semitic languages that have a prefixed t, such as taf'il in Arabic. Even Akkadian has such forms for some weak verbs, as in the infinitive tadanum, which is the bi-form of nadanum, to give, and all the Syrian tishameh for shitameh, where the sibilant and dental have metathesized, that means they have switched position. The Ebla'it forms are not simply metathesized and then analogically extended by an infixed t, though, since they seem to consistently have a vowel i in the last syllable, that is taptarism, where Akkadian consistently has u as in pitrusum. The source for the prefixed t might thus more likely be verbal nouns of the form taf'il and the like, into which the infixed t was secondarily inserted. In any case, these forms are highly innovative, just trust me. Again, this is far from being a comprehensive list of features, but they are representative. So, what have we seen so far in the morphology of Ebla'it? It is a very coherent system. There are differences to East Semitic or Akkadian, to be sure, but the morphology of Ebla'it is consistent. There are hardly any biforms, and when there are, they are mostly due to text genre or provenance of the text in which they occur. There is more variety in the lexicon, especially when it comes to particles, and here we also see more features that are known from West Semitic languages. So, let me introduce another set of features, namely particles. In the domain of conjunctions, for example, Ebla'it attests to the conjunction up and then, which is known from various West Semitic languages, Hebrew being one example, but not from Akkadian. It has the common Semitic conjunction wa, although in a more archaic form than attested in Akkadian, which is u, just believe me, that derives from wa. It also has the form u for or, which corresponds to Akkadian, although again, there's a biform with a more archaic shape, au, and note that Ebla'it does not seem to have the typical Akkadian form ulu for or, but an innovative form uma instead. The base of all these forms for or, au, is common Semitic, but Akkadian and Ebla'it went separate ways in the way they combined these elements with other particles. For here, Ebla'it has a West Semitic lexeme not shared with Akkadian, pau, which is related to Hebrew po. Some of these differences can be explained as simple archaisms, such as the form au for or. Others are not as clear-cut as in the case of pau. The last lexemes I want to look at are prepositions. Prepositions exhibit the largest degree of variation in the data sets I'm presenting today. If that's too small, don't worry, we'll see details of that in a moment. What we see in this table, or not, are clear doublets. Many prepositions have more than one variant, and these variants alter between those that have parallels in Akkadian and those that do not. Most of these biforms, however, are not used in free variation. They are again dependent on text genre and in some cases place of origin. So here are the more detailed ones. So for in, Ebla'it has two prepositions that are semantically equivalent, in and mean. In is equivalent to pre-Sargonic and Sargonic Akkadian in and later Akkadian ina. This preposition is the normative one in chancellery and economic texts. The preposition mean is exclusively used in royal rituals, which is a separate set of texts whose exact origins are not quite clear. For 204, Ebla'it has two forms, ana and sin. The distribution is the opposite, as in the case of in, in and mean. Ana, which is the form that is also used in Akkadian, is rare in chancellery texts where sin is preferred. It occurs, however, in chancellery texts that come from Mari and other towns, such as Manuwat and Ibal and Hadu, and in letters that are addressed to other cities, such as Kabul and Khamasim. From this we can see that ana was clearly associated with international correspondence, or let's say correspondence that went beyond local Ebla, while sin was preferred in local texts. This is interesting because it shows an awareness of different registers of the language used, a local one and a diplomatic or international one. For from, Ebla'it has two forms as well, wastum, which corresponds to forms of the preposition we find in pre-sargonic texts, for example in those from Mari, and which is the common form in chancellery texts, and a form not otherwise found in East Semitic, minu. Minu is used in administrative and chancellery texts, but less often. This is the only doublet where we cannot clearly distinguish different use patterns, although minu seems to be more restricted in use than wastum, in that it most commonly is followed by place names. The last preposition that is found in Ebla'it but not in Akkadian is bainai, between. This preposition is common in West Semitic languages, such as Hebrew and Arabic, but not in East Semitic. Lexemes such as these would be a nice indicator for West Semitic influence, except that this particular lexeme only occurs once, and that in a lexicalist. Lexicalists can have forms that are not commonly used in administrative and chancellery texts, and should thus be taken with a grain of salt when it comes to evaluating the nature of Ebla'it. So, to summarize, based on the evidence presented, what can we say about Ebla'it? There are a few things that are fairly certain. As mentioned before, Ebla'it shares the most important innovative isoglosses with Akkadian. These clearly mark it as East Semitic, which is a fact that is no longer disputed. It even shares some Babylonian-type innovations, such as dative independent pronouns marked by infixed s. There are, however, numerous differences compared to Akkadian. These start with minor phonological changes, but also include significant innovations, such as the t-stem infinitives with both infixed and prefixed t, which are otherwise unknown in Semitic. It is noteworthy that none of the Ebla'it morphological divergences from Akkadian can be attributed to West Semitic influence with certainty. They either reflect archaisms or unique Ebla'it innovations not shared by any other Semitic language, at least that I'm aware of. Ebla'it also has a few quirks in the morphology that, again, I'm not aware of in other Semitic languages, such as different nasals, n and vs. m in dual and plural pronouns, where other Semitic languages have homogeneous paradigms. Despite these divergences from Akkadian and the notable Ebla'it innovations, Ebla'it morphology shows a very coherent system that hardly has any bi-forms. When bi-forms exist, they're usually connected to the genre of lexical lists or the texts in which they occur originate outside of Ebla, such as Mari or Manuwat. The big question is how to evaluate the divergences from Akkadian and the nature of Ebla'it. In particular, even though the overall morphology of Ebla'it reflects a pretty consistent system, does that necessarily mean that we are dealing with a uniform language, a language that simply developed differently from Akkadian in some areas of its grammar, but is still as a whole a descendant from the same ancestor language, East Semitic? Given the coherence of the morphology, that might be possible. Ebla'it is also sufficiently geographically distant from the core Akkadian area to explain how these divergences could have occurred, but there are problems with such an interpretation. The particles I have talked about do not fit such an analysis. While the morphology is rather consistent, the lexicon as reflected in particle most notably is not. The particles show much greater amounts of doublets and also lexemes that are common in West Semitic. As mentioned before, most of the doublets are not used in free variation, but are clearly connected to either text genre or other factors, such as whether the text in question was intended for local or non-local use. This shows that there was an awareness of different language registers, which lexemes were appropriate for local but not international correspondence, which were fine for literary texts such as rituals, but not for everyday use in chancellor and administrator texts, and so on. The particles also show that there is influence from a non-local language on what is written, as in the example of the preposition ana, which is Akkadian. This complicates matters a bit, and it would be over-simplistic to evaluate Ebla'it as a uniform system. There are other issues as well. A look at the map likewise complicates the whole Akkadian or East Semitic idea. Ebla'it is simply too far west. As mentioned earlier, some scholars have tried to solve the issue by proposing an East Semitic dialect continuum that reaches from Mesopotamia all the way to Ebla', which is a possibility. Ebla'it has a few archaic features compared to Akkadian, which would fit its peripheral position on such a continuum. But it also shares some of the most innovative features of Akkadian, such as the dative pronouns with infix sibilant, only otherwise found in Babylonian, which are not even shared by the geographically and linguistically much closer Assyrian. Some versions of Lexicalist also have the innovative Babylonian infinitive type purus for the d-stem, contrary to the more archaic form powers known from Assyrian. This means there might be more going on than simply a dialect continuum. I think there might be a different solution to the whole issue of Ebla'it, although I need to stress that what I propose is tentative and it needs to be corroborated by a more detailed investigation of all aspects of Ebla'it, grammar and lexicon. So, it is certain that Ebla'it adopted its writing system from Mesopotamia. A very similar type of writing system was also adopted by Mari, although Ebla shows a few divergences from the Mari system. Overall, however, it seems fairly certain that cuneiform writing made its way westward from Mesopotamia via Mari to Ebla. Ebla also has evidence for direct contact with Mesopotamian sources. Some of the lexicalists, for example, have direct parallels in text found at Abu Salabih, a site close to Nippur with the textual finds dating between 2600 to 2500 BCE. A few of the literary texts found at Ebla, written in Sumerian, have likewise direct parallels in text found at In general, incantations at Ebla are primarily written in Sumerian, a language that was clearly imported to Ebla from Mesopotamia together with the writing system, scribal education, and cultural contact. There are other cultural influences from Mesopotamia that are noticeable at Ebla. The northern Mesopotamian calendar is in part used at Ebla, although there's also evidence for a local calendar. We know that singers from Mari, some of them Sumerian, practiced at Ebla. There's evidence for scribes and teachers for mathematics coming from Kish to Ebla. The other way around is attested as well. Two colophones of lexicalists mention trips of young Eblite scribes to Mari. Besides cultural ties, there were, as I mentioned earlier, also close economic ones, especially between Mari and Ebla. It has been well established by now that there was a vibrant trade network reaching from Ebla over northern Syria to Mesopotamia. The cultural and economic ties that have been studied in detail in recent years, they show a very closely connected set of city-states in the second half of the third millennium BCE. These include Ebla, Tel Beda, Mari, Kish, Abu Salabih, and Ur as the most noteworthy ones for our purpose. The contact with Mesopotamia over Mari was constant and vibrant. I think that this cultural contact might help us understand the linguistic situation at Ebla. We know, as I said earlier, that at least one language was imported to Ebla together with a writing system and cultural contact, namely Sumerian. I don't think that anyone has suggested a Sumerian dialect continuum reaching from Ebla to southern Mesopotamia. The use of Sumerian at Ebla is superseded by the use of Semitic, at least in administrative and chancellery texts. This can be explained by a growing tradition of writing Semitic and pre-Sargonic Mesopotamia and adjacent areas, which starts to be evident in the text from Abu Salabih and is clear when we reach the pre-Sargonic texts from Mari and Tel Beda. The texts from Mari, Tel Beda, and Ebla have many similarities in the cuneiform use and Semitic vocabulary, but also show local adaptations. Tel Beda, for example, also uses the preposition sin, two, four, although not as frequently as applied. The ultimate source of this type of writing was Mesopotamia, however, where cuneiform writing was increasingly adapted to writing Semitic. We should not forget that the Ebla archives overlap with the beginning of the Sargonic period, if the current dating is correct, at which time the process was already quite advanced. To make a long story short, I propose, for now, that Sumerian was not the only language that was imported to Ebla. It also adopted the Semitic language that was already in use in writing in Mesopotamia and that spread over northern Syria and the upper Euphrates area, meaning Mari and Tel Beda, through trade and cultural contacts all the way to Ebla. That language was an archaic form of the language of northern Babylonia, and note that this is meant geographically, since all the Mesopotamian sites we can trace in terms of major cultural and scribal contact are in Babylonia and except Abu Salabih in the north. But even Abu Salabih is in terms of writing more closely connected to northern Babylonian traditions. I'm not saying that Eblaite Akkadian is Babylonian. Similar to many contact languages, the Akkadian that was imported to Ebla took on some features from the local substrate language, such as certain phonological and morphological innovations, and in particular lexemes, which is what we would expect in certain types of contact. This explains the rather strange mixed character of Eblaite, looking very much like Akkadian on some level, but completely different on others. I'm not saying that Eblaite was a mixed language in the linguistic sense of the term, just that it had a good deal of substrate influence. What is also implied in this proposal is that we have only very little evidence for Eblaite, if we understand the term as representing the local language. What we have are a few forms and lexemes that might reflect the underlying local language, but what we call Eblaite is more likely Akkadian with substrate influence. This proposal differs from previous approaches that often see Eblaite as a homogenous system native to Ebla. Instead, I wish to open the discussion for taking a broader approach that looks beyond the purely linguistic features and also considers the language within its social and historical background. Thank you for your attention. For over 100 years, the OI has been a leading research center for the study of ancient Middle Eastern civilizations. Join us in uncovering the past and learn about the beginnings of our lives as humans together. Become a member by visiting oi.uchicago.edu slash member