A Tale of Two Cities

Introduction

The northern Levant passed one of the most fascinating periods of its history under the last pharaohs of the 18^th Dynasty. This period and its events remain debated in scholarship because the written documents available are substantial enough to raise interest, and at the same time scarce and limited enough to also raise questions. While it is clear that the Egyptians and the Mittani had divided the northern Levant between themselves since the time of Thutmose III and that the Hittites fought both of them to establish their control of the region, questions remain unanswered over how the events developed, over who controlled what, and over when they controlled it.

This article aims at answering one of these questions. It is clear that the Egyptians chose both Ṣumur on the ʿAkkār Plain and Kumidi in the Biqāʿ Valley to be their “capitals” in the northern Levant. It is also clear that, from these centers, they administered the region under their control. But the reasons behind the Egyptian choice of establishing more than a centre remain somewhat unclear.^[1] In the next pages, I first present a methodology of analysis of the data at our disposal, I then reconstruct the diachronic succession of the events related to the two centres, and finally I argue that Kumidi became an Egyptian centre only after Ṣumur had fallen in the hands of the rulers of the small state of Amurru, probably to replace it in its administrative capacities.

“Focal Points” and the “Proximity Principle”

The corpus of letters found in the Egyptian capital of Akhetaten – modern Tall al-ʿAmārna (Tell el-Amarna) – is probably the most important source of information on the Egyptian Administration in the Levant during the 18^th Dynasty. The letters found in the capital of Amenhotep IV/Akhenaten are administrative briefs sent to Egypt from across the Near East to discuss the short-term necessities and circumstances of a handful of regional kingdoms and dozens of small and belligerent Levantine cities. The letters, sent over a timespan of about 30 years, were never meant to form a coherent narrative and the succession of the events they expose is difficult to reconstruct.^[2] In the past decades this succession has become a disputed and thorny problem, which at its centre both the diachronic sequence of the letters themselves and the placement of the events they narrate in the wider Near Eastern chronology. The several approaches and several types of evidence that have been put forward over the years to tackle these issues can be combined into a single methodology, which will be used in this paper.^[3]

The first element is Campbell’s idea of “Focal Points,”^[4] which gives indications in terms of contemporaneity. According to Campbell, it is possible to consider two or more letters as contemporary based on the presence of a “Focal Point,” a character or an event that they have in common. For example, both EA 86 and EA 116, written by the ruler of Gubla (Byblos) Rib-Addi, mention Abdi-Aširta of Amurru, who is thus a Focal Point between them.

The second element is the idea that the larger corpora (i.e. sets of letters from the same city) within the ʿAmārna archive make it possible to reconstruct a somewhat clear succession of the events they mention. The tens of letters written by Rib-Addi are especially useful in this sense, as they mention, and at times describe, events regarding northern Lebanon over a period of more than a decade.^[5] For example, Rib-Addi describes in his letters Abdi-Aširta’s fall, his probable death, and the rise on the throne of Amurru of his son Aziru. Since Amurru is ruled by Abdi-Aširta in EA 86, and by Aziru in EA 116, EA 116 must have been written after EA 86.

The third element is the idea that Focal Points between letters written in Gubla and letters written elsewhere offer clues about the diachronic succession of events taking place elsewhere in the Levant. Continuing with the same example, since EA 59, written by the people of the city of Tunip, mentions the actions of Aziru of Amurru, the letter and the events it describes must have been written later than EA 86.

The fourth element is the idea that external sources, such as the historical prologues in Hittite treaties,^[6] provide a wider context for the contemporary wars fought in the Levant between Egypt, Mittani, and Ḫatti. Focal Points between these sources and the ʿAmārna Letters allow us to place the events mentioned in the letters within the known chronology of the Near East. The invasion of Nuḫašše, for example, is mentioned both in EA 165, written by Aziru, and CTH 51, the treaty between the King of Mittani and Ḫatti, and thus acts as a Focal Point. The events mentioned in CTH 51, therefore, likely happened later than the events described in EA 86.

Since the intertwined history of Ṣumur and Kumidi provides several useful Focal Points, it is possible to discuss the dates of the known events during the period covered by the ʿAmārna Letters. In addition, their regional proximity to both Gubla and the “international border” with Ḫatti and Mittani makes it possible to both follow the diachronic succession of the events in the city’s letters and provide contexts for the historical, political, and military events of the Near East. In this respect, I will rely heavily on Violetta Cordani’s doctoral research.^[7] This is because of her convincing argument that the First Syrian War of Suppiluliuma I developed during 4/5 years^[8] rather than the 1 year, as is usually assumed.^[9] A thorough discussion of this issue would be out of place here, but a quick recapitulation of the arguments is necessary. The main source on this war is the historical prologue of the treaty between Suppiluliuma of Ḫatti and Šattiwaza of Mittani (the already mentioned CTH 51), in which Suppiluliuma states to have campaigned from the region of Išuwa in northern Syria to the region of Nuḫašše in the Orontes Valley, plundering “all of these lands in one year.”^[10] According to Cordani, CTH 51 follows closely the canons of Hittite annalistic texts, and in these documents every campaigns ends with the mention of the booty that the king either brought to Ḫattusa, or sent there with a portion of his army.^[11] In her view, since the prologue is divided into four paragraphs, and since the end of each paragraph is marked by the mention of the booty that the king is said to have brought to Ḫatti, the First Syrian War lasted four to five years, and the sentence above should be considered an aggrandizement of the kings’ actions rather than accepted at face value.^[12] This is particularly convincing in light of the distance the army is said to have travelled: Išuwa and Nuḫašše are 800 km apart, and although an army of the time might have walked such a distance in a single year, to do so the campaign would have consisted of series of razzias without any attempt at empire building rather than a carefully structured plan of conquest that is described in the text.^[13]

The concept behind the “Proximity Factors” of the two cities is more straightforwardly explained. The letters reflect a system of semi-independent city-states, each governed by its own ruler – the ḫazannu – and supervised by an Egyptian commissioner – the rābiṣu – located in one of the many centres the Egyptians had established along the Levant. The Egyptian commissioner exacted the taxes from the locals, regularly travelled between his seat and the Egyptian capital, and communicated to the local rulers the wishes and the commands coming from the Pharaonic court. Two decades ago Goren, Finkelstein, and Naʾaman have demonstrated that by establishing the source of a tablet’s clay one can clarify the provenance of the letter.^[14] They were able to petrographically and chemically analyse 300 tablets of the corpus and localize the origin of the clay from which they were made by comparing the tablets’ clay with the supposed geological environments of the tablets’ origin.^[15] They discovered that the clay of 46 letters (15 % of the examined tablets) matched the geological environments of the Egyptian centres in the Levant rather than that of senders’ cities. More specifically, the rulers of Gezer, Lȧḵiš (Lachish), and Gath^[16] wrote from Azzati/Ġazza (Gaza), those of ʿAkkā (Akko), Šamḫuna, and Urusalim (Jerusalem)^[17] from Bẹṯ Šəʾȧn (Beth Shean), those of Ḫašabu, Ḫasi, Guddašuna, Enišasi, and Tobiḫi^[18] likely from Kumidi,^[19] Rib-Addi of Gubla (Byblos) and Aziru of Amurru from Ṣumur (Fig. 1).^[20]

Fig. 1

Location of the cities governed by rulers writing from Egyptian centres (map by author).

To explain this fact, Goren, Finkelstein and Naʾaman theorized that the Canaanite mayors would have arrived quite often at the Egyptian administrative centres, and would have sent their letters from there.^[21] If one accepts this logical explanation, the distribution of these cities clearly defines a pattern that I have defined as the “Proximity Principle:”^[22] barring exceptional circumstances, a local king would have related to the closest Egyptian centre and the closest Egyptian commissioner rather than one farther away. Therefore, by identifying which cities interacted with which commissioner, each of whom was located at a particular site, it is possible to map a centre’s “Proximity Factor”, i.e. its area of influence. This principle seems to apply to most letters. Yet, there are rare cases (the exceptional circumstances above) in which commissioners “outreached” their apparent boundaries relating to cities not in their immediate vicinity,^[23] and when rulers “circumvented” their commissioner relating to different Egyptian centres. Overall, the Proximity Principle provides a valuable tool to analyse the Egyptian presence in general, and the interrelations between Ṣumur and Kumidi in particular. For example, had Ṣumur and Kumidi been Egyptian centres at the same time, their Proximity Factors would have rarely – if ever – overlapped, whereas if in fact Kumidi was established as an Egyptian centre to replace Ṣumur in controlling the northern Levant, Ṣumur’s and Kumidi’s Proximity Factors would have overlapped.

I will use the two methods discussed above to argue that Kumidi was established after Ṣumur had been conquered by Abdi-Aširta of Amurru, and that eventually it replaced the coastal site as the main Egyptian centre in the northern Levant. I will first analyse and explain the events connected to each city as they can be understood from the documents at our disposal, and then gather and “calculate” the Proximity Factors of the two cities.

Ṣumur in the ʿAmārna Letters: History and Chronology

Located near the mouth of the Eleuthero river in the ʿAkkār Plain, the city of Ṣumur^[24] controlled the main access point to Syria from the Mediterranean, and its possession was cardinal to access and control coastal Lebanon, the Biqāʿ, and the Orontes Valley.^[25] As it is only during the late reign of Amenhotep III that Egyptian direct control of the city is described in written documents, it is unclear when the city was chosen for this role.^[26] Considering its strategic position, it is not surprising that the Egyptians chose to make this city their main centre in the northern Levant (provided with a commissioner, a garrison, and a palace),^[27] nor is it surprising that it remained a desired place to control for all other regional players, including the kingdom of Amurru.

Fig. 2

3D rendering of the Northern Lebanese coast in the ʿAmārna Age (after Kilani 2020: 157, fig. 5.4, courtesy of Marwan Kilani).

The mountainous region of Amurru, located in the mountains of Lebanon south of the gate of Ḥims (Homs), was inhabited by semi-nomadic people, and appears to not have had a capital^[28] or a recognized ruler at the beginning of the time period covered by the ʿAmārna Letters (Fig. 2). Under Abdi-Aširta and his successors, however, it developed into a large kingdom and regional power capable of scheming against the Egyptian administration.^[29] According to Rib-Addi, who chronicled these events in a series of letters addressed to the Egyptian court,^[30] the process started with the conquest of the large coastal cities of Ardata and Irqata, continued with the subjugation of a series of small centres in Gubla’s political sphere,^[31] and ended with an attack on Ṣumur itself. It was only after this last conquest that Abdi-Aširta wrote the first of his letters^[32] introducing himself as a loyal subject of the pharaoh, justifying his intervention in the city,^[33] and requesting to be recognized as the guardian of the Egyptian interests in the area.^[34] Contrary to what could be expected, Abdi-Aširta’s actions did not meet immediate response, and he appears to have used that time to continue diminishing Gubla’s control in the area.^[35] It appears that the Pharaoh tacitly accepted this situation.^[36] After all, the presence of a local garrison in an Egyptian centre was a common practice, and in Egyptian eyes Ṣumur might have been just another centre garrisoned by a local ruler.^[37] At some point, however, the Egyptians decided to put an end to Abdi-Aširta’s activities, sending two contingents to capture and (likely) kill him.^[38] The reasons behind this change of mind are not clear. One possibility is that Amurru had grown to such an extent that the Egyptians felt they could no longer control it; another is that the Egyptian court believed Rib-Addi’s accusations that Abdi-Aširta was seeking an alliance with the other Great Kingdom of the region, Mittani.^[39] It might also have been a combination of the two factors. Whatever the reasons, the death of Abdi-Aširta allowed the Egyptians to buy some time, but ultimately his demise was not enough to bring Amurru’s political activities to an end.

Soon, Amurru started expanding again under the leadership of the sons of Abdi-Aširta. First they conquered Irqata and Ardata. Then they were able to lay siege to Ṣumur and conquer it. It is unclear how long this siege took, but since Rib-Addi wrote at least 5 letters – 8 according to Cordani^[40] – in the time of the siege, it would not be surprising if it took between one and two years. After the conquest, one of Abdi-Aširta’s sons called Aziru achieved prominence, and in a letter to Amenhotep IV/Akhenaten (EA 165) asked to be recognized as the protector of Ṣumur. After this moment, Ṣumur does not seem to have held the same status in the Egyptian administration anymore, as no commissioner appears in later letters. This might be due to a silence of the sources, because some time after Ṣumur’s conquest Rib-Addi was exiled from Gubla,^[41] and his successor Ili-rapi appears to have been less inclined to chronicle the events for the Egyptian court.^[42] Considering the fact that several letters found in Kumidi were written by Aziru and Ili-rapi, however, it seems more likely that the Egyptians did not attempt to re-establish their direct control over the city.^[43]

The succession of these events is well established, and therefore its discussion has been relatively straightforward. Their connection with the wider chronology of the period, by contrast, is more complex, as it hinges on two “Focal Points,” which will be used also for the discussion of Kumidi: the succession on the throne of Amurru, and the second capture of Ṣumur. Both the contemporaneity of Abdi-Aširta with Amenhotep III and that of Aziru with Amenhotep IV/Akhenaten are widely recognized. It is also often argued that there was an “interregnum” under the “sons of Abdi-Aširta” before Aziru took power, either between Amenhotep III and Amenhotep IV/Akhenaten, or solely under the latter.^[44] This is due to the fact that in many letters Rib-Addi says to Amenhotep IV/Akhenaten that it was his father who killed Abdi-Aširta, and that he should do the same to Aziru.^[45] The death of Abdi-Aširta appears to have happened in the last years of the reign of Amenhotep III, possibly between Years 34 and 36 according to both Cordani and Stavi.^[46] The second Focal Point is the second capture of Ṣumur by the sons of Abdi-Aširta, an event that directly ties the succession of events of Rib-Addi’s archive with the First Syrian War of Suppiluliuma I. This connection lies in one of the few letters written by Aziru: EA 165. In this letter, the Amurru leader mentions the presence of the Hittite army in Nuḫašše (the region of the Middle Orontes),^[47] and since chemical and petrographical analyses of this tablet have shown that it had been composed in Ṣumur,^[48] the city must have been in Amurru’s control when he wrote the letter. It therefore follows that Ṣumur was conquered by Aziru before the arrival of the Hittite army in Nuḫašše. According to Cordani’s interpretation of CTH 51, the Hittite army arrived in Nuḫašše during the third year of the war.^[49] Since the war started between Year 4 and Year 5 of Amenhotep IV/Akhenaten,^[50] it arrived in Nuḫašše between Year 7 and Year 8. Thus, Aziru was likely in control of Ṣumur in Year 7 of the Pharaoh.

In conclusion, it appears that the Egyptian centre of Ṣumur was lost for the first time to Abdi-Aširta of Amurru between Year 34 and 36 of Amenhotep III. Reconquered, it was lost again to Aziru likely in Year 7 or 9 of Amenhotep IV/Akhenaten.

Kumidi in the ʿAmārna Letters: Succession or Diarchy?

Compared to the history of Ṣumur, that of Kumidi is less straightforwardly described. The city lacks someone like Rib-Addi to chronicle its events, and the picture that emerges from the documents at our disposal is complex and contradictory. Kumidi appears in EA 198 as a regular Levantine city with a ruler, Arašša, who first asks the Egyptian dignitary Ḫamašše for support in proving his loyalty to the Pharaoh, and then asks to be granted “life”, the right to rule over his own city. However, Kumidi appears also as a regular Egyptian centre, with a commissioner named Paḫuru, a garrison, and a Proximity Factor that covers the Biqāʿ, the Bāšān, and coastal Lebanon.^[51] This situation is unparalleled, and the presence of both a local ruler and an Egyptian commissioner in the same city raises the question of Kumidi’s form of government: did Paḫuru and Arašša share control of the city, or was the local king removed by the Egyptians?

Hachmann and Naʾaman have answered this question in opposing ways.^[52]

Hachmann proposed that Kumidi became an Egyptian garrison town only late in the reign of Amenhotep IV/Akhenaten, while Naʾaman argued in favour of a diarchy, suggesting that Kumidi had been joinlty controlled by the Egyptians and the locals from long before the beginning of the ʿAmārna Period. Hachmann identified the Ḫamašše mentioned in EA 198 with the ambassador of the same name mentioned in EA 11, 27, and 29. Naʾaman saw these two persons as mere homonyms, because in the ʿAmārna corpus commissioners are often called to witness a local rulers’s loyalty, and naming an ambassador is unparallelled. Hachmann proposed a change in foreign policy in the Late 18^th Dynasty, while Naʾaman advanced the notion that the passive approach to foreign policy of Amenhotep IV/Akhenaten makes any change to the pre-existing Egyptian structure in the Levant unlikely.

I believe there are many reasons to discount Naʾaman’s ideas. First and foremost, there is nothing in the documents at our disposal to suggest that Ḫamašše was a commissioner. Even assuming that he was one, there is nothing to suggest that he was Kumidi’s commissioner, or that Kumidi had been Egyptian for some time. On the basis of the (lack of) information provided by EA 198, Ḫamašše might just as easily have been a commissioner in the close Egyptian centres of Bẹṯ Šəʾȧn or Ṣumur. Furthermore, Naʾaman’s idea does not fit easily with the information at our disposal. On the one hand, because it requires that either Arašša or one of his successors remained in power throughout the reign of Amenhotep IV/Akhenaten without ever being mentioned again, as there is no trace of rulers in Kumidi in later letters. It has already been said, for example, that it was common for local rulers to send troops to guard an Egyptian centre, and the troops guarding Kumidi were sent by Biryawaza of Dimašqa (Damascus).^[53] If Naʾaman were right, there would have already been a ruler in Kumidi before Biryawaza, one who would have contributed with its own army to the security of his own city. The fact that both the ruler and the army go completely unnoticed in Biryawaza’s letter would need an explanation that is not offered. On the other hand, even though Ḫamašše and Paḫuru supposedly governed the same region, there is a considerable difference in the quantity of mentions between the two, one that appears to be difficult to argue solely on the basis of the chances of discovery: the latter is only mentioned in EA 198, while the former is mentioned in 10 different letters written across northern Lebanon and southern Syria. Furthermore, the notion of Amenhotep IV/Akhenaten’s passive approach to foreign policy, commonly held in the 1980s, has subsequently been seriously questioned.^[54] Finally, if “unparalled in the sources” is an argument to dismiss Hachmann’s theory, it should dismiss Naʾaman’s as well. Since there are no rulers of Azzati (Gaza), Bẹṯ Šəʾȧn, or Ṣumur in the documents at our disposal, a diarchy between a commissioner and a local ruler would be just as unique as an ambassador called to witness a ruler’s loyalty. In other words, not only does Naʾaman’s “diarchy” theory find no support in the data at our disposal, but it would also require convoluted explanations for the lack of documentation in the late ʿAmārna Period, for the absence of the ruler of Kumidi in the dealings with Dimašqa, and for the lack of letters regarding the commissioner called Ḫamašše.

Overall, Hachmann’s idea appears to be preferable. To begin with, it fits nicely with the available evidence, starting from the passage in which Arašša declares his loyalty to his overlord: May the king, my lord, ask his commissioners whether I am a loyal servant of the king, my lord. May the king, my lord, ask Ḫamašše whether I am a loyal servant of the king, my lord.^[55] By mentioning first “commissioners” and then Ḫamašše, Arašša appears to voluntarily single out Ḫamašše from the group of “commissioners”. Since Kumidi lies on the main inland road through the Biqāʿ to Waššukkanni and Bābilu (Babylon), it is reasonable to assume that an Egyptian ambassador would have stopped in Kumidi during his travels, and would thus have been known to the local ruler. It would therefore be unsurprising if Arašša had called on a known ambassador to support his plea of loyalty, and it would have been just as normal if he had separated him from the “commissioners” group. Moreover, Hachmann’s interpretation does not require additional explanations to fit with the available evidence. If the two Ḫamašše(s) were one and the same person there would be no reason to assume a displacement of a number of the commissioner’s letters, nor to explain Arašša (and his army’s) absence in Biryawaza’s letter.

The written evidence, therefore, points decisively to the idea that the Egyptian presence in Kumidi did not begin until late in the period covered by the ʿAmārna Letters. Another way to approach the problem is to look for archaeological evidence of a prolongued Egyptian presence in Kumidi during the 18^th Dynasty. While excavations at other settlements defined as Egyptian in the ʿAmārna Letters, such as Yapu (Jaffa) and Bẹṯ Šəʾȧn,^[56] have revealed locally-produced Egyptian pottery (and especially the diagnostic flowerpots),^[57] there is little archaeological evidence to support a long Egyptian presence in Kumidi,^[58] and the interpretation of what exists is widely disputed: the Egyptian stone vessel bearing the name and title of the Egyptian dignitary Rawoser (HAtj-a Ra-wsr), for example, was once used to support the notion of a long Egyptian presence in the city,^[59] whereas it is now commonly believed to be a result of trade.^[60]

In conclusion, Arašša’s removal from power and the transformation of Kumidi into an Egyptian centre explain his absence in later letters. This scenario fits nicely with the available evidence and should therefore be seen as the most likely possibility.

Kumidi in the ʿAmārna Letters: Establishing a Terminus Post Quem

Given the fact that a diarchy appears to be unlikely, the question becomes one of time-frame: when did Arašša’s rule over Kumidi end, and when was the city turned into an Egyptian administrative centre? Unfortunately, a precise answer to these questions seems hardly possible on the basis of the current evidence. Contrary to Ṣumur, Kumidi lacks a chronicler, like Rib-Addi, and there are no letters to document when and under what circumstances the city was taken over by the Egyptian administration. We must, therefore, rely on what we know to reconstruct what we do not know, finding the termini post and ante quem for the end of Arašša’s rule and the establishment of the Egyptian presence.

Since Arašša was clearly still ruling over Kumidi when he wrote EA 198, and since he does not appear anywhere else, this letter must be considered the terminus post quem. Beside its historical implications, EA 198 is a standard letter, with a normal opening passage^[61] and the usual request to be granted “life”. The only Focal Point in the letter is the character of Ḫamašše. As mentioned above, this is due to the fact that an ambassador of the same name is mentioned in the letters written by the Great Kings of Mittani (EA 27: 37–40, 52–54; 29: 25–29) and Bābilu (EA 11: 9–22). Dating these letters might therefore also suggest a date for EA 198. EA 27 and 29 were written by the Mittanian king Tušratta and can be dated with relative certainty, as the latter had a hieratic docket registering the date and place of arrival in Egypt, and the former is directly connected to the latter. According to the docket, EA 27 was delivered to Thebes by two Hurrian emissaries named Pirissi and Tulubri in the first month of Peret of the second regnal year of Amenhotep IV/Akhenaten.^[62] From an internal reference, it appears that EA 29 should be dated to some years later. Since in line 113 Tušratta says that two of his emissaries have been kept in Egypt for the preceding 4 years, the letter should be dated to Year 6 if the emissaries in question were Pirissi and Tulubri, and to Year 4 (at least) if they were not.^[63] EA 11, written by the Babylonian king Burna-Buriaš, is more difficult to date, as its registration docket is unreadable,^[64] and the letter does not contain Focal Points. However, if an argument from silence can be made, what the letter does not mention might be revealing. In EA 9 (19–35), Burna-Buriaš desperately tried to veto any contact between Egypt and the nascent kingdom of Aššur (testified by EA 15), claiming the Assyrian king as his own vassal. Since the contacts continued unimpeded (as testified by EA 16), either Burna-Buriaš found his peace with the subject or EA 11 was written before EA 9. Considering that Aššur could only develop its foreign relations after the Hittite conquest of Mittani^[65] in the first decade of Amenhotep IV/Akhenaten,^[66] it would appear that EA 11 is to be dated roughly to the same period as EA 27 and 29, if not somewhat later.

As there can be no certainty over the identity of Ḫamašše in light of the discussion above, it is worth considering 3 different scenarios:

1. the Ḫamašše of EA 27 and EA 198 are two different people, as Naʾaman sustained

2. the two Ḫamašše are the same person, and he was a commissioner

3. the two Ḫamašše are the same person, and he was the ambassador

In the unlikely scenario A, EA 198 would have no Focal Points, and there would be no way to date it. The only possible indication would therefore be the terminus ante quem discussed below. In scenario B, also unlikely, EA 198 should be dated at the end of the reign of Amenhotep III, because by Year 2 of his successor Ḫamašše was already an ambassador. Year 2 would therefore be the terminus ante quem of EA 198. In light of the discussion above, scenario C appears to be the most likely, and in that case, since it can be established that Ḫamašše worked as an ambassador at least between Year 2 and 6 of Amenhotep IV/Akhenaten, it would follow that also EA 198 is to be dated in the same period, if not somewhat earlier.^[67]

Kumidi in the ʿAmārna Letters: Establishing a Terminus Ante Quem

The establishment of the terminus ante quem is a more complex affair, since the diachronic discussion needs to be connected to both the conquests of Ṣumur, and to the First Syrian War of the king of Suppiluliuma of Ḫatti. The earliest letters related to this discussion are EA 116, 122, and 123, all written by Rib-Addi of Gubla. Their mention of a commissioner of Kumidi predate (with different degrees of certainty) the second conquest of Ṣumur, and therefore they offer a first terminus ante quem for Kumidi’s takeover on part of the Egyptians in Year 7 of the reign of Amenhotep IV/Akhenaten, and might be dated as early as Year 3 of the same king.

The letter that can be dated with the highest degree of certainty is EA 116, in which Rib-Addi mentions a “commissioner of Kumidi” in a list of Egyptian forces, whom he thinks should be used to impede Aziru’s conquest of Ṣumur.^[68] Clearly, therefore, there already was a commissioner in Kumidi before Ṣumur was lost to the Egyptians for the second time. Since it was argued before that Aziru conquered Ṣumur in Year 7 of Amenhotep IV/Akhenaten, EA 116 must have been written before or during Year 7. A somewhat earlier date for this letter can be suggested on other – less cogent – arguments. The first one is the mention of the crowning of Amenhotep IV/Akhenaten.^[69] According to both Miller and Cordani,^[70] this mention means that the letter was written in the first few years of his reign. The second one is the ironic comparison of the sons of Abdi-Aširta to the Kings of Mittani, Kūš (Cush), and Ḫatti.^[71] According to Cordani,^[72] the inclusion of Ḫatti in the list is indicative for two reasons: first, since in similar earlier lists Rib-Addi does not mention Ḫatti,^[73] its inclusion likely means that the First Syrian war had already started; second, since Rib-Addi is arrogant about Ḫatti’s presence, and not as nervous as he appears to have been when he wrote EA 126 (written at around the same time as EA 165),^[74] the First Syrian War had not yet arrived near Gubla. For all of these reasons, EA 116 should be dated to the first few years of both Amenhotep IV/Akhenaten (likely Year 5 or Year 6), and of the First Syrian War (likely the first or second year).

EA 122 and 123 might be the earliest mentions of the activities of Paḫuru, but are almost self-contained and of a difficult chronological collocation. In them, Paḫuru is said to have sent Suteans to Gubla to kidnap 3 men, killing a Šardana mercenary in the process.^[75] The only Focal Point is a short mention of the sons of Abdi-Aširta at the end of EA 123, a passage where Aziru is not singled out.^[76] This would date them to before the siege of Ṣumur (sometimes at the beginning of the reign of Amenhotep IV/Akhenaten), when Aziru had not yet risen to prominence among his brothers. A date between Year 3 and Year 5 would appear to be likely, but impossible to prove. From his actions it is possible that Paḫuru would have already been a commissioner in Kumidi, and this might indicate that Kumidi had – at this early stage – an influence over Gubla, possibly because Ṣumur had yet to be re-built after having been re-conquered.

In conclusion, a “hard” terminus ante quem for the establishment of Kumidi as an Egyptian centre can be set in Year 7 of Amenhotep IV/Akhenaten, the latest possible date for EA 116, which mentions the commissioner of Kumidi. However, EA 116 might be dated as early as Year 5, and EA 122 and 123 as early as Year 3. A “soft” terminus ante quem for the transformation of Kumidi would therefore be Year 3 of Amenhotep IV/Akhenaten.

Kumidi in the ʿAmārna and Kumidi Corpora: Chronology of the Later Letters

If EA 116, 122, and 123 represent the earliest attestations of the commissioner in Kumidi, all letters that testify to the activities of the centre need to be considered in this analysis. A total of thirty-three letters will be discussed in the following pages: fourteen mention either Paḫuru or a “commissioner of Kumidi,” ^[77] ten were likely written in the centre itself,^[78] and nine were found – or are thought to have been illegally dug – in Kāmid al-Lōz.^[79]

Paḫuru and the Egyptian centre in Kumidi appear in fourteen letters: eight from Gubla,^[80] three from the Orontes Valley, ^[81] two from the Bāšān,^[82] and one from the Damascene.^[83]

1. EA 117 was apparently written right after EA 116. In this letter, Aziru is starting to acquire his leadership role among his brothers, Ṣumur is under siege, and Rib-Addi again requests that Paḫuru and the commissioner of Ṣumur, Yanḫamu, are sent against Amurru.^[84]

2. EA 129 pertains to the same period, but it is clearly written after Ṣumur had been conquered.

3. EA 131 mentions Paḫuru as being a “loyal servant” to the king, and asks again that troops are sent to Gubla to wage war against Amurru.

4. EA 132 is likely the last to have been written by Rib-Addi: Ṣumur was taken, and Aziru is said to be planning an attack on Kumidi.^[85]

5. EA 139 was written by Rib-Addi’s successor Ili-rapi, and it is thus particularly late. In it, Paḫuru is said to be inciting Gubla’s men against Aziru and wanting to wage war against him.^[86]

6. EA 207 and 208 are too short and too fragmentary to be dated: in both Paḫuru is mentioned in relation to local squabbles, and his role is comparable to that of any other commissioner. Neither sender could be identified.

7. EA 57, written by the ruler of Qaṭna, Akizzi, is exceedingly deteriorated, and thus any chronological indication would be at best an informed guess.

8. EA 189 and EA 197 describe the same event from opposite sides: the war in the Biqāʿ and the Bāšān between Aitakkama of Qādeš and Biryawaza of Dimašqa. According to Cordani, since in EA 162 the Pharaoh orders Aziru to interrupt all relation with Aitakkama, “a man with whom the king is angry,”^[87] the relation must have deteriorated after this letter and, therefore, EA 189 and 197 should be considered to be later than EA 162, which Cordani dates to Amenhotep IV/Akhenaten’s 14^th Year.^[88]

9. EA 190, sent to a ruler whose name cannot be read, orders him to guard and defend both Paḫuru and Kumidi, defined as “the king’s garrison city”.^[89] There is little to be said regarding the date, but since there is mention of the travel of the Egyptian army, it might be dated in the later years of Amenhotep IV/Akhenaten, a time when also the Egyptian campaign of the ʿAmārna Letters should be dated.^[90]

On the basis of the letters mentioning Paḫuru it seems possible to argue that he acted as commissioner in Kumidi for several years as the first mention of his activities (EA 116) can be dated to Year 7 of Amenhotep IV/Akhenaten, (but EA 122 might be dated to Year 3), and he is mentioned in letters written until Year 14 of the same king.

The second group is composed by ten letters which were likely written in Kumidi in the same period of time:^[91]

1. EA 174, 175, 176, 363 were written respectively by Biʾiri of Ḫašabu, Ilu-Dayyani of Ḫasi, an unnamed king, and Andi-Resa, ruler of Enišasi. They are all similar in content, and they were all written by the same scribe, likely at the same moment. Since they all mention the expedition of Aitakkama of Qādeš in the Biqāʿ (also discussed in EA 189), they are likely to be dated to Year 14 of Amenhotep IV/Akhenaten, like EA 162 mentioned above.

2. EA 177, written by the ruler of Guddašuna, Yamiʾuta, is too fragmentary and offers no Focal Points.

3. EA 178, written by Ḫibiya, ruler of an unnamed city, offers no Focal Points, and it is therefore impossible to date.

4. EA 179 was written by a deposed ruler of the city of Tobiḫi. There are no explicit Focal Points in the letter, so a date cannot be specified with any confidence. However, since in the letter the author says that Tobiḫi has become “like a city of the lands of Amurru”,^[92] it might be dated at a later stage when Aziru was expanding its influence.

5. EA 185 and 186 were written by a ruler of Ḫasi whose name is not Ilu-Dayyani, but Mayarzana, meaning that these two letters cannot be dated to the same time as EA 175. Besides this basic point, however, these letters are “an extreme example”^[93] of undatable documents, as they are lengthy and in good state but contain no Focal Points. The two letters discuss the same issue (a series of conquests of Aman-ḫatpi, ruler of Tušulti) and are intimately connected, but no character or event can be linked with any certainty to any character or event in the other letters.

6. EA 187 was written by a ruler of Enišasi who is not Andi-Resa of EA 363, but an individual named Sadeya. This likely means that the two letters cannot be dated at the same time. Unfortunately, there are no Focal Points to date this letter.

Few of the letters likely written in Kumidi can be dated with any certainty: EA 174, 175, 176, and 363 can be dated after the 14^th Year of Amenhotep IV/Akhenaten. Unfortunately, however, EA 177, 178, 179, 185, 186, and 187 cannot be dated on the basis of the events they mention.

Finally, an even later (but unquantifiable) temporal horizon is offered by the letters that have been found in Kāmid al-Lōz, ancient Kumidi (Table 1). As they have never been published in an unitary way, they are here discussed following Pruzsinszky’s nomenclature.^[94] A general argument to date these letters resides in EA 198: since a commissioner was responsible for the transmission of orders to the local rulers, and since the king of Kumidi would have had no reasons to intercept and store letters from the Pharaoh to another king, it is most likely that they are to be dated to a post-Arašša period.^[95] Setting aside KMD 3, 4, and 7, which are too damaged to offer any data, the other letters all offer a reason to be dated at the end of the ʿAmārna Period.

KMD 1 and 2 are nearly identical letters, sent by an unnamed Pharaoh to Zalaia of Dimašqa and Abdi-Milki of Sazaena.^[96] Although the two toponyms are spelled differently, Turri has identified the Abdi-Milki that in EA 203 is ruler of ^uruša-az-hi-mi with the Abdi-Milki that in KMD 2 is ruler of ^uruša-sa₃-i₁₅-na, explaining the difference in spelling with an alternance between nunation and mimation.^[97] Zalaia of Dimašqa instead is otherwise unknown, since in the ʿAmārna corpus the king of Dimašqa is called Biryawaza.^[98] However, in EA 194 Biryawaza mentions both his father Šuttarna and his grandfather Hastar, who presumably ruled before him. As Naʾaman argued,^[99] this would make Zalaia either a successor of Biryawaza and therefore date these letters at the end of the period covered by the ʿAmārna Letters or a predecessor of Hastar and therefore date these letters to long before the ʿAmārna Period. Yet, in the ʿAmārna corpus both Biryawaza and Abdi-Milki reply to the same Egyptian letter and therefore are likely contemporary.^[100] At the same time, KMD 1 and 2 are nearly identical and therefore also likely contemporary. As a consequence, if Turri is right in his identification, Abdi-Milki was contemporary with both Biryawaza and Zalaia. To account for the lifespan of Abdi-Milki, KMD 1 and 2 must either be dated to the ʿAmārna Period, shortly precede it, or shortly follow it. However, since it is unlikely that Abdi-Milki would have reigned long enough to know four generations of kings of Dimašqa, KMD 1 and 2 should both be dated after the ʿAmārna Period. The unsigned letter KMD 5 discusses the weapons of a “bi-ri-di-ya” that are located in Kumidi. It is generally held^[101] that this man might be identified with the ruler of Magidda (Megiddo) of the same name that appears in 5 letters of the ʿAmārna corpus.^[102] Among these letters is EA 247, in which he replies in affirmative to the request for food for the Egyptian campaign likely connected to EA 190. Unfortunately, KMD 5 does not provide any clear-cut evidence that the two persons are one and the same.^[103] However, a plausible – but at the same time unprovable – possibility is that the ruler of Magidda had departed with the Egyptian army and had died somewhere in the Biqāʿ leaving his weapons at the Egyptian centre of Kumidi.^[104] This would date KMD 5 to the late ʿAmārna Period. KMD 6, 8 and 9 are sent by Aziru and by Rib-Addi’s successor Ili-rapi. The presence of Ili-rapi dates them to the second decade of Amenhotep IV/Akhenaten, but their date cannot be established with any more precision. Of the nine Kumidi letters, therefore, six can be dated to the last decade of Amenhotep IV/Akhenaten, whereas three cannot be dated at all because they are exceedingly fragmentary.

In conclusion, all of the letters that are to be related to the Egyptian centre in Kumidi, either because they mention the commissioner Paḫuru or because they were written or found in Kāmid al-Lōz, are to be dated to the last decade of Amenhotep IV/Akhenaten.

Combining the Data: The Chronology of Ṣumur and Kumidi

The findings of this discussion, which has examined the chronology of the two centres of Ṣumur and Kumidi, can now be summarized. Ṣumur appears to have been conquered twice by the army of the kingdom of Amurru, first at the end of the reign of Amenhotep III (possibly between years 34–36), and then in the 7^th Year of Amenhotep IV/Akhenaten. After this conquest, there are no letters that seem to be connected to the presence of an Egyptian commissioner in the city. The history of Kumidi is a bit more complex. Its reconstruction has been based on the assumptions that there was no “diarchy” and that Arašša was not in power anymore when the city became an Egyptian centre.

A terminus post quem could not be precisely identified. EA 198, the only point of reference, has a single Focal Point in the Egyptian ambassador Ḫamašše and, although he was certainly active between Year 2 and Year 6 of Amenhotep IV/Akhenaten, his ambassadorial capacity likely started before and might have continued after these two dates. Thanks to the many connections with the lengthy and detailed archive of Rib-Addi, however, the terminus ante quem is more easily and reliably fixed to Year 7 of Amenhotep IV/Akhenaten, and possibly as early as Year 2 or 3 of the same king. Before this date, there are no letters to support the notion of an Egyptian presence in Kumidi. Moreover, of the 39 documents that pertain to the ʿAmārna and Kumidi corpora and that are related to the Egyptian presence in Kumidi, none can be dated earlier than Year 2 or 3 of the same Pharaoh: 28 can be dated, with varying degrees of certainty, to the first and second decade of the reign of Amenhotep IV/Akhenaten, while 11^[105] could not be dated at all.

Therefore, it is reasonable to argue that Kumidi was established as an Egyptian centre at the beginning of the reign of Amenhotep IV/Akhenaten, after the first loss of Ṣumur.

Fig. 3

Chronological distribution of the letters discussed; the outlines indicate possible ranges of time (graphic representation by author).

Figs. 4

Proximity factors of Ṣumur and Kumidi (map by author).

Combining the Data: Proximity Factors of Ṣumur and Kumidi

According to the Proximity Principle discussed at the beginning,^[106] it is possible to reconstruct to a certain extent the areas of influence of each Egyptian centre by considering what cities interacted with each seat of power.

Ṣumur’s Proximity Factor can be reconstructed by considering who interacted directly with the centre – Gubla and Amurru – and who interacted with the five Egyptian persons that either certainly or likely held the office of commissioner in Ṣumur:

1. Paḫamata is mentioned as the commissioner of Ṣumur at the time of its first conquest in five letters,^[107] which come from Amurru (2), Gubla (2), and Siduna (Sidon/Ṣayda) (1).

2. Amanappa appears in ten letters,^[108] all written by Rib-Addi, and all connected to the period of the first conquest of Ṣumur.

3. Yanḫamu, as I have argued elsewhere,^[109] appears to have worked in Ṣumur only for a limited period of his career, between the first and the second conquest of Ṣumur. In this role, he is mentioned in nine letters, from Amurru, Gubla and Biruta (Bayrūt/Beirut).^[110]

4. Pawura is mentioned as a commissioner of Ṣumur in six letters from Rib-Addi,^[111] all written after his death by the hands of the sons of Abdi-Aširta, likely during the second conquest of Ṣumur.^[112] According to all commentators^[113] these letters from Gubla are among the last letters written by Rib-Addi: Ṣumur had already fallen to Aziru, and Gubla was in danger.

5. Hapi, Pahamata’s son, who was repeatedly accused of having handed Ṣumur over to Aziru, appears in seven letters,^[114] six from Gubla and one from Ṣurru (Tyre).

In light of these elements, it can be said that commissioners of Ṣumur were in contact with Gubla, Amurru, Siduna, Biruta, and Ṣurru. The centre of Ṣumur, therefore, appears to have covered the ʿAkkār Plain and coastal Lebanon, as far south as Siduna.

Kumidi’s Proximity Factor can be reconstructed by considering who interacted with the centre, either by writing from the centre itself^[115] or by being the addressee of a letter found there,^[116] and who interacted with its commissioner Paḫuru.^[117] Much of the data has already been exposed in the chronological discussions of the letters: Paḫuru dealt with Qaṭna, ʿAštȧrot, Qādeš and Dimašqa; the kings who wrote from Kumidi were the rulers of Ḫašabu, Guddašuna, Tobiḫi, Ḫasi, Enišasi, and two unnamed ones, all located in the Biqāʿ; the letters found in Kumidi connect the city with Amurru, Gubla, Dimašqa, Sašimi, and (possibly) Magidda. However, the letter supposedly from Magidda (KMD 5) is not relevant to the discussion. The unsigned letter asks for the restitution of objects belonging to a man called Biridya, like the ruler of Magidda in the ʿAmārna Letters. If the two Biridya are the same person, then retrieving his objects would count as an exceptional circumstance. If the two Biridya were different people, there would be nothing to tie the unsigned letter to a specific city. In either case, KMD 5 would have little value for the Proximity Factor of Kumidi.

Considering these elements, Kumidi’s Proximity Factor appears to have covered not only the Bāšān and Biqāʿ Valley, but also reached areas that one would either assume for reasons of regional proximity to be related to Ṣumur (such as the Orontes Valley and the cities of Qaṭna and Qādeš) or that are known to have been connected to Ṣumur (such as Gubla and Amurru). In this regard, it is important to consider the dates of the letters that connect Gubla and Amurru with Kumidi. Setting aside the letters in which the commissioners of the two centres appear together (EA 116–117), they are all to be dated to periods in which Ṣumur is not an Egyptian centre, either before its reconstruction (EA 122–123) or after its destruction (KMD 6, 8, 9). In other words, in the periods when Ṣumur was not an Egyptian centre, the rulers of Gubla and Amurru had to interact with the nearest Egyptian centre, the one that held the reins of the Egyptian administration in the north: Kumidi.

Conclusions

Five conclusions can be reached on the basis of the previous analysis of the letters from Tall al-ʿAmārna and Kāmid al-Lōz:

1. Ṣumur was conquered by Amurru twice, the first time by Abdi-Aširta sometime at the end of the reign of Amenhotep III, and the second time by Aziru in Year 7 of Amenhotep IV/Akhenaten.

2. There are no letters connected to the commissioners of Ṣumur that can be dated to the second decade of Amenhotep IV/Akhenaten.

3. It is extremely unlikely that Kumidi had been an Egyptian centre before the period covered by the ʿAmārna Letters, as the city in EA 198 appears to be governed by Arašša and a “diarchy” between the Egyptian commissioner and a local ruler cannot easily fit with the pieces of information provided by the documents at our disposal.

4. Kumidi was established as an Egyptian centre at the beginning of the reign of Amenhotep IV/Akhenaten, before Year 7 and possibly before Year 3.

5. In the second decade of Amenhotep IV/Akhenaten, the commissioner in Kumidi had influence on the kingdoms of Amurru and Gubla, replacing Ṣumur as the Egyptian administrative centre in the northern Levant.

These five elements coherently support the idea that Kumidi was established as an Egyptian centre only after Abdi-Aširta’s conquest of Ṣumur. Moreover, the fact that Kumidi extended its influence also on areas formerly controlled by the coastal city further supports the notion that measures of administrative re-adjustments were taken, and that the city in the Biqāʿ eventually replaced the coastal one as the main administrative centre in the northern Levant.