Scribes: The Diplomats of the Amarna-Age

The Canaanite Letters reflect the important role the scribes played in diplomacy.^[1] They were not just passive writers; they imprinted their own professional acumen and ingenuity into these letters.

Scribes inflected their letters with rhetorical strategies, and even scribal cues that worked together as attention grabbing strategies. These would only have been noticed by other cuneiform specialists, and helped paint the intended picture for those receiving the letters. In a similar way to how we encode layers of meaning into our own texts today—through the words we use, how we spell them, how we arrange them, and through the fonts and other design features that we use—ancient writers also did what it took to get their messages across to their audiences.

The letters do not mention the scribes by name, nor do they tell us any biographical details about where they were born, where they lived, or about their families. However, comparing the differences in the letters helps us to identify different scribal communities and even individual scribes.

The Hidden Hand of the Scribe

While the Amarna Letters were composed by scribes, they were written as though they contained actual speeches from one king to another. We will take, for example, the introductory sentence of a letter ‘Abdi-Heba of Jerusalem sent to the pharaoh. To understand how the scribe used the cuneiform script, I transcribe the text here,^[2] using Assyriological conventions:

¹ [a]-na ^I⸢LUGAL⸣ EN-ia ⸢qí⸣-⸢bí⸣-⸢ma⸣ ² um-ma ^IÌR-ḫé-ba ÌR-ka-ma ³ a-na 2 GÌR.MEŠ EN-ia LUGAL-ri ⁴ 7-ta-a-an ù 7-ta-a-an am-qut-mi^[3]

^{EA 286: 1-4} Speak [t]o the king, my lord, a message from IR₃(=‘Abdi)-Hebat, your servant. I fall at the two feet of my lord, the king, seven times and seven times.^[4]

In this short passage, it is worth drawing attention to the complexity of the cuneiform writing system. The scribe here uses three different types of cuneiform signs:

Logograms/Sumerograms —signs that represent Sumerian words to be read as Akkadian words. For example, the Sumerian word lugal “king” is used to represent the Akkadian word šarrum, which also means “king.” These are transcribed with capital letters.

Syllabic—signs representing syllables, and vowels (e.g., tu, ta, ti), transcribed with lower case letters.

Determinatives—signs that were semantic classifiers (e.g., to mark a personal name or title), transcribed as superscriptions.^[5]

The title “my lord, the king,” is a standard reference to the pharaoh in the Canaanite Letters. The letters written by peer, or high-status polities tend to identify the pharaoh by name. Ancient scribes had to learn the appropriate greetings and expressions of politeness expected in diplomatic letters for their respective patrons.^[6]

The description of ‘Abdi-Ḫeba’s act of prostration at the pharaoh’s feet—“seven times and seven times”—symbolizes ‘Abdi-Ḫeba’s total deference to his overlord. Obviously, ‘Abdi-Heba in Jerusalem is not physically bowing before the pharaoh in Akhetaten; the reference was protocol language, common in the Canaanite letters, and learning how to do so correctly was an important part of scribal training. Again, scribes writing for higher status rulers used different formulae in this part of their letters. Thus, scribes encoded the hierarchy of the letter sender and recipient in the first few lines of a letter.

‘Abdi-Ḫeba sent at least 6 letters to the pharaoh (EA 285–291), all written by the same scribe (see part 8 for more on this scribe). Because the letters are written in the first person—as though this ruler is speaking directly to the pharaoh—the scribe seems to be invisible. This is the convention in nearly all of the Levantine Letters. Yet the shadow of the Jerusalem scribe, a skilled letter writer, and a master of the cuneiform script left its imprint on the letters.

Moreover, the scribe left a second, personal message to the scribes working in Egypt at the end of four of their letters; this addendum was not meant to be read to pharaoh.

Gatekeepers of Amarna Diplomacy

Access to scribes who knew cuneiform script gave elites power and enabled them to be players in diplomatic discussions about trade, military intervention, and gift-giving. The Amarna Letters are testaments to the creativity of cuneiform scribes all over the ancient Middle East who had the difficult job of writing to scribes who were far away, spoke different languages, and used cuneiform in different ways.

Egypt and the Levant were regions that were constantly in contact, and already during the Amarna Period, both the Canaanite alphabet and the Egyptian hieroglyphic and hieratic scripts^[7] were being used in Canaan.^[8] Also, objects inscribed with Egyptian writing, too, moved through the Levant, even in the earlier Middle Bronze Age period. For example, inscribed scarabs and scaraboids (small amulets shaped like scarab beetles, which were sometimes inscribed), with Egyptian writing were very popular and were even used in local Canaanite funerary cult.

Yet, there is no evidence that the scribes who were writing the Canaanite Amarna Letters used Egyptian writing. We only have evidence that they used cuneiform, as did most other scribal communities in the ancient Middle East in the late second millennium.

The Spread of Cuneiform: From Mesopotamia to Syria to Canaan

While cuneiform originated in Mesopotamia, it was used in Syria by the mid-3^rd millennium B.C.E. Thousands of cuneiform tablets were discovered at the site of ancient Ebla (Tell Mardikh), also the site of the earliest royal archive (dated to about the 24^th–23^rd centuries B.C.E.).^[9] Archaeologists even found a room with designated shelves used to store cuneiform tablets, whose organization has been reconstructed.^[10]

Ebla is also important because it demonstrates that by this early period, scribes in Syria were already adapting cuneiform for their own purposes, such as writing in “Eblite,” the local West Semitic language. Ebla was eventually destroyed (most likely by Sargon of Akkad), leaving behind its well-baked tablets as a testament to the sophistication of its scribal community.

Cuneiform texts were also found in Mari, once a rival of Ebla, but which continued into the first half of the second millennium B.C.E. until it was destroyed by Hammurabi of Babylon (1759 B.C.E.). As Mari had extensive influence, it might have been a source for the spread of cuneiform south into Canaan.

Hazor is the southern Levantine site which contains the most cuneiform tablets dating to the second millennium B.C.E. It was in contact with Mari as well as other more northern cities in the Middle Bronze Age.

While the number of cuneiform texts from the southern Levant is small, compared to other regions, the cumulative evidence suggests that cuneiform was used in this region in the Middle Bronze Age; moreover, the texts that have been dated to this period are widely distributed, geographically, and they represent different genres and uses of cuneiform: administration, legal texts, liver models, and even a fragment of Gilgamesh, likely used in higher levels of scribal training.^[11] By the Amarna Period, cuneiform was not a borrowed script in Canaan but a part of the fabric of local scribal communities.

Scribal Training

Scribal training involved much more than just memorizing cuneiform signs: Scribes were taught tablet making, tablet layout and design, and how to transform clay into cuneiform writing.

Some scribes were native Akkadian speakers (such as the scribes working for the Kassite or Assyrian kings). Most of the scribes, however, were non-native Akkadian users. Their writing was influenced by their own languages and thus often reflect regional differences in how scribes used cuneiform, based on the training of the scribes and language used at the local court. Therefore, it was important that scribes developed universal conventions for communicating to each other, especially for high-stakes letters, like those in the Amarna corpus.

Similarities in the letter introductions, the formulaic structure and even the organization of the letter reflect this process and the challenge that faced scribes: how to communicate in cuneiform to another scribe, across the world, in a way that would yield positive results—this was especially difficult given that Akkadian was typically not the native language of the scribes receiving or sending the letters.

Canaano-Akkadian

The Amarna Letters from larger kingdoms (see parts 2 and 3) tend to use more standard Akkadian forms that reflect the main Akkadian dialect of this period, known as Middle Babylonian.

The Egyptian scribes themselves wrote in Egypto-Akkadian, when communicating with elites outside of Egypt.^[12] The term “Egypto-Akkadian” simply means that they wrote in a way that reflects the influence of their own language, but also the conventions of the scribes working in Egypt, for the pharaoh. Their letters tend to be highly formulaic, and to employ protocol language, such as epithets that reflect the religious changes taking place in Egypt.

The Canaanite Letters show even more influence from local languages. They have elements from both Akkadian and the scribes’ local Canaanite dialects, periodically peppered here and there with Egyptian words. They largely use an Akkadian-based vocabulary and are written in standard form.

At the same time, the scribes’ writing practices are greatly influenced by their own speech practices—the letters mostly follow the word order and rules of Canaanite grammar. They even use occasional Canaanite words— sometimes along with a wedge-shaped scribal mark, known as “gloss mark” (or Glossenkeil in German)—or even entire Canaanite phrases.^[13] Their letters are so distinctive that scholars have used them to study later Iron Age Canaanite dialects, like Hebrew, Phoenician, and Moabite.

For example, the letters contain morphemes that are known from later Canaanite dialects. In the following passage, the scribe uses Akkadian words but encodes the verbs using Canaanite affixes. And, the scribe employs a Canaanite word, known as a “gloss” that is flagged with a scribal wedge mark.

Canaanite Verbal Forms

EA 297 is a short letter, written by a scribe working in the area of Gezer for an elite named Yapa’u. The letter employs a metaphor that compares the local ruler to a brick that supports a foundation. The local elite thereby expresses his support for pharaoh’s rule and objectives in the region. The letter has a two-part structure: an introduction (lines 1-7) and a longer body message (lines 8-21).

First we will look at the language of the letter, as an example of how Canaano-Akkadian worked. Then we will look at the figurative language in the letter, and try to puzzle out of this was the “language” of the elite or the scribe.

The introduction is written using more standard Akkadian. The scribe also uses a very conventionalized letter-heading and prostration formula:

^{EA 297:1–7} Speak to the king, my lord, my god, my Sun god, a message from Yapa’u, your servant, the dust of your two feet. I fall at the feet of the king, my lord, my god, my Sun god, seven times (and) seven times.

However, the language and orthography change in the letter’s main message. The scribe uses Canaanite forms, built on an Akkadian vocabulary. For example, in the passage below, the scribe uses Canaanite grammatical features: the word order follows that in Canaanite languages, and the verbs are marked following Canaanite morphology.

⁸ mi-im-ma ša qa-˹ba˺ ⁹ LUGAL EN-ia a-na ia-ši ¹⁰ iš-te-mé-šu ma-gal ¹¹ SIG5-iš

^8–11a Whatever the king, my lord, ordered for me, I have obeyed it very carefully.

The verb written qa-˹ba˺ is a 3ms form of Akkadian qabû (“to speak”), but marked as a Canaanite qatal form; the verb written iš-te-mé-šu is a 1cs form of šemû (“to hear”), a cognate in Akkadian and Canaanite. The form, however, is a 1cs yaqtul perfective form, and it is followed by a 3ms pronominal suffix.

The verb below, in₄-né-ep-ša-ti₇, is also built from an Akkadian verb (a N-stem of epēšu “to do”), yet the scribe adds a 1cs suffix, from Cananaite (-ti), and uses it as a passive form (“to become”), though this verb has a different meaning in other Akkadian texts:[14]

^11b ša-ni-tam ù ¹² in₄-né-ep-ša-ti₇ ¹³ ki-ma ri-qí URUDU : (scribal mark) sí-ri ¹⁴ ḫu-bu-ul-li ¹⁵ ˹iš˺-tu qa-at ¹⁶ ˹LÚ˺.MEŠ ^KUR ˹Su˺-te^{MEŠ 17} ù a-nu-ma iš-te-mé ¹⁸ sa-ri ša LUGAL DÙG.GA-ta ¹⁹ ù it-ta-ṣa-at ²⁰ a-na ia-ši ù pa-ši-iḫ ²¹ lìb-bi-ia ma-gal

^11b–21 Something else: I had become like a {Can. damaged} copper pot (Can. gloss: pot) because of the power of the Suteans. But, now, I have heard the sweet breath of the king. It came forth to me, so my heart is very much at rest.

A Canaanite Gloss

The scribe also employs a Canaanite gloss in his description of how the local ruler has been abused like a “damaged copper pot” after the sumerogram. Here, we can ask: is this expression a reflection of the words of the ruler who sent the letter, or the scribe?

The scribe uses a similar analogy in a letter for a different ruler, in EA 292, a letter written for Baʿlu-šipṭi. The language is nearly identical, just in a different context:

^{EA 292:41–49} Something else: Take note of the deeds of Piʾya, son of {f}Guʾlat[i], against Gezer, a maidservant of the king, my lord. How many days has he been plundering her so that she has become like a {Can. damaged} pot on account of him.^[15] From the mountains, men are ransomed for 30 (shekels of) silver, but from Piʾya, for 100 (shekels of) silver!

This additional suggests that this particular expression was known and used by the scribe. The gloss may even have been learned as a part of their scribal training, using lexical lists. Juan Vita has argued that glosses that were use several times and/or that were used for very common words, are likely a reflection of the scribe’s training using multilingual teaching texts.^[16]

The letter, however, ends alluding to the life-giving power of pharaoh, and to his magical breath. This is a common trope in the Canaanite Letters, which derives from concurrent Egyptian epithets of the pharaoh. As seen by this example, the scribes interwove influences from their cuneiform training, their own languages, the in-put of their employers, and also Egyptian elements, into their letters.

A Scribal Code

Canaano-Akkadian was not a spoken language.^[17] It was a written scribal code that is better studied as a reflection of the scribes’ cuneiform “literacy” practices than as a window into their “speech” practices.^[18] The term “code” is an accurate description for the different linguistic and extra-linguistic aspects of this system and to describe how it actually worked. Canaano-Akkadian was a way to communicate that involved language, spelling practices, scribal marks, and a range of conventions about how to properly write letters to Egypt.

While past work privileged the linguistic forms in the letters, today, more holistic approaches consider all different aspects of the tablets as important data about the scribes, such as their orthographic practices and use or omission of expected formulae, as well as the extra-linguistic features used by scribes. The study of the letters even factors in the process of tablet making and inscription. Scribes were trained in the selection of clay, how to form a tablet, and in the appropriate design and layout of the tablets (such as whether or not to use scribal marks).[^19]

The Canaanite Amarna Letters as Scribal Artifacts

Approaching the letters as scribal artifacts—texts made by scribes for scribes—also draws attention to the relationship between the content and communicative strategies in the letters, asking the question: Where and why did scribes use Canaanite elements in their letters? And, did the scribes use Canaanite elements in particular contexts?

The scribes tend to use Canaanite elements in the main messages of the letters, and in important sections of the letters. They also used Canaanite elements and/or more clusters of Canaanite forms, in concert with other communicative strategies. For example, as we saw in the examples above (in EA 297), the introductions of the Canaanite Amarna Letters are mostly written in standard Akkadian, sticking, for the most part, to the conventions. In the body of the letters, however, we see the scribe’s innovative use of cuneiform and their linguistic dexterity. Certain scribes use more Canaanite elements than others. And the scribes use these elements from their own languages alongside other communicative strategies, for example, within poetic passages that seem to reflect local literary culture, or in passages with parallelism.

There is compelling evidence in the letters that some scribes were mindful of their language choices, and that they used their local languages to highlight important passages. For example, in the letters of the Jerusalem scribe (seen in a later section), the scribe rarely uses Canaanite forms, or the local Canaano-Akkadian forms. Yet when these forms occur, the scribe uses them alongside parallelism, and in passages that have a strong rhetorical force. Together, these different elements enriched the letters and added texture that would have caught the attention of pharaoh’s scribes.

When we look at the complexity of the letters, we see that the scribes were not incompetent communicators, as scholars once proposed.^[20] They were skilled interlocutors for Canaanite elites. Only scribes who were familiar with Canaanite and Akkadian could have written these letters.

Importantly, only scribes who knew cuneiform and how it was used by Canaanite scribes could have read and fully understood the Canaanite Amarna Letters. These were texts written by scribes for other scribes. This means that scribes controlled the flow of written information between Canaanite elites and Egypt. They played a vital role in diplomacy.

Scribal Communities and Their Networks

The scribes were not monolithic; there was no single “Canaanite” scribal community or way of using cuneiform.^[21] Some of the differences map onto regional divides. For example, scribes in areas more frequently in contact with Egypt, or in the northern Levant, show more influence of standard Akkadian.

Previously, scholars assumed that the tablets were formed and inscribed at the main capital cities of the elites mentioned in the letters. However, a comprehensive petrographic analysis of the clays of the Amarna Letters determined that many of the Canaanite Letters were formed from clays from different locations than the cities of the letter senders.^[22]

And, over 20 letters originated at sites with an Egyptian military and administrative presence in the Levant.^[23] Scribes did not typically carry clay with them. This means that certain letters were actually written at Egyptian bases in Canaan when scribes, and the rulers that they represented, traveled to meet Egyptian officials face-to-face.

Some scribes therefore played a more active diplomatic role and even traveled with local elites to meet Egyptian officials, or to neighboring sites when elites periodically traveled and/or took refuge with nearby allies. For example, Rib-Hadda of Byblos took refuge in Beirut (see part 7). The clays of his letters suggest that he brought along one of his scribes.^[24]

During these trips, they came into contact with scribes from other sites. We might think of such scribal “contact” as a contributing factor to how scribes learned from each other; this would explain the existence of similarities between the letters of different scribes in Canaan.

Postscript Messages: Messages Between Scribes and Diplomats

It is tempting to imagine that while the scribes and other officials were stationed far away from home, they exchanged stories and passed on useful information about their professional work, or even just said hello. We have no records of this. However, a few letters have secondary messages that are addressed to the people who worked for the pharaoh and Levantine elites.

A scribe working for Aziru of Amurru (see part 4) included a second message (lns. 36–44), known as a “postscript” or a “piggy-back letter,” to the official message from Ba‘luya and Beti-ili, to their ruler, Aziru, in Egypt (EA 170: 1–35). The secondary message is addressed from Amur-Baʿalu to four members of Aziru’s entourage (Rabi-ʾilu, ʿAbdi-URAŠ, Binana and Rabiṣidqi).

^{EA 170: 36–44} To Rabi-ʾilu and ʿAbdi-URAŠ, to Bin-ʿAnu and Rabi-ṣidqi, a message from ʾAmur-Baʿlu/Haddu. May it be well for you. Do not worry, and do not become angry! Here, it is very well with your households. Wish ʿAnatu well.

Who is Amur-Baʿalu? It seems likely that this person was the scribe who wrote this tablet, and that he included a private greeting to his colleagues who were with Aziru at the time.

In-House Scribes Versus Contractors

Although the Canaanite Amarna Letters are all written in cuneiform, they reflect different levels of mastery of this script and different communicative aims. We can use paleography (the way the scribes formed cuneiform signs), coupled with the linguistic features, and the scribe’s spelling practices, to create profiles of individual scribes and whole communities.

A great example of this holistic approach is Juan Vita’s 2015 analysis of the paleography of the letters. Vita connected letters to specific scribes, using the form of the cuneiform signs and other features in the letters.^[25] He also used the petrographic evidence from the clays of the letters to build a profile of the scribal communities behind the letters. Putting together this varied evidence paints a nuanced portrait of the scribes in Canaan in the Amarna Period.

Some of the letters are simple and short and very formulaic; other letters are linguistically and rhetorically complex. For example, the scribe who worked for Abimilki of Tyre created a series of poems in a letter (EA 147).

^{EA 147:5–15} My lord is the sun that comes forth over the land (Akk. gloss: the lands) day after day like the mark of the sun, his favorable father (i.e., Amenhotep III), the one who gives life by means of his sweet breath and returns by means of his {Can. north wind}, the one for whom the entire land is stationed in peace by means of the power of a right arm (Egy. gloss: strength), the one who gave his thunder in the heavens like the Storm god, so that the entire land {Can. became afraid} of his thunder.

As seen from these examples, the scribe used both Canaanite and Egyptian in the letter, clustered in these poetic sections. The scribe essentially deconstructed the form of letters, and inserted several poetic sections, drawing attention to the pharaoh’s power as a solar deity. Here, we see, perhaps the clearest sign of a scribe in Canaan versed in Egyptian religious and scribal developments.

Sometimes a group of scribes only worked for a particular royal family and shared a similar training in how to write in cuneiform and in how to formulate letters (see part 7 for the example from Byblos). Others worked at several different sites, and for different elites as needed. Some even worked a circuit, moving through a particular region.

For example, the scribes working in the area of the Jezreel Valley wrote for several rulers based at sites along this strategically important axis of trade and agriculture.^[26] The similar and very formulaic nature of their messages suggest that they may have been working for Egypt, that is, they were dispatched to local elites who did not have a scribe of their own. Similarly, a scribe working in the area around the ancient site of Tel Gezer wrote several letters for different rulers, including a local queen.^[27]

Making the Amarna Age Possible

In sum, while the Canaanite Letters tell scholars much about the Egyptian-Levantine diplomacy, from the perspective of rulers and elite officials, the letters are also witnesses to the labor and skill of the people who wrote them: scribes who specialized in writing cuneiform letters. The scribes were more than just note takers or letter writers. They were key participants in Amarna Age diplomacy. They made the Amarna “system” possible.