Essay Question:

Assess the importance of the discoveries at Byblos for our understanding of the beginnings of the Middle Bronze Age.

Essay written by David Rohl (3rd Year Ancient History/Egyptology).

Submitted to Peter Parr on the 9th November 1989.

Egypt and Palestine in the MBA: The Byblos Connection

It is an unfortunate necessity of life that Levantine archaeologists must depend to such a degree on the chronology of Egypt for their absolute dating of the Palestinian Middle Bronze Age. What is equally unfortunate is the misplaced confidence which those same archaeologists confer on the reliability of the Egyptian dates. This is not, however, entirely the fault of those scholars working outside Egyptology, for they have quite naturally succumbed to the continuous propaganda of their colleagues from across the Sinai peninsular who parade up and down the streets of scholarship dressed in the dazzling cloak of Sothic chronology. However, a number of respected Egyptologist are now beginning to voice their concerns about relying on the two key Sothic anchor points for Egyptian dating and hence the chronology of a major part of the ancient world. Sothic chronology may be, after all, simply another case of 'the Emperor's new clothes'.

This issue can be only briefly addressed in this essay, but the basic criticisms of the method must be aired if only to caution against the current dependency on the Egyptian 12th Dynasty dates by Levantine historians and excavators. On the other hand, the use of relative dating is entirely valid within the framework of certain dynastic sequences. Unfortunately the sequence from the end of the 6th Dynasty to the beginning of the 18th is not entirely devoid of chronological problems and these must have some effect on the question at hand in this essay.

With these cautionary notes in mind we should perhaps begin by stating the basic facts behind the Egyptian/Levantine/Mesopotamian chronological network. These synchronisms are the basic tools used to establish MBA relative and absolute dating and to some extent the seriation of the Middle Bronze artifactual assemblages.

Yantin, prince of Byblos

During the course of Dunand's excavations at Byblos, he uncovered a large fragment of a bas-relief bearing the following hieroglyphic inscription [Montet, Kemi I, p. 90ff]:

... h3ty-' n Kpn Intn whm 'nh r n h3ty-' Ykn m3 hrw ...

... the prince of Byblos, Yantin, repeating life, son of the prince Yakin, justified ...

The term m3 hrw ('justified') indicates that the father of Yantin was deceased, whilst whm 'nh ('repeating life') tells us that Yantin himself was very much alive at the time the inscription was cut. The importance of this inscription lies in the fact that a second column of text is partially extant in which the latter part of the nomen of the Egyptian king Neferhotep I, the 21st king of the 13th Dynasty [Gardiner, 1961, p. 440] can be clearly read. Albright [1965, p. 55] has identified the owner of Tomb IV at Byblos as this same Yantin and he seems to be correct in his attribution: a badly damaged vase inscription from this tomb does indeed appear to bear the name Yantin [Montet, Atlas, pl. CXVII, 787]. On the basis of the Dunand relief, Tomb III has been attributed to Yantin's father, Yakin (probably the Yakin-El of the Carnarvon cylinder [Albright, 1945, p. 11]). Thus artifacts from these two tombs have been assigned a date, on the Egyptian evidence, of c. 1750-1670 derived from the so-called 'Sothic date' of Year 7 of Senuseret III (1872+/-3 – high or 1830+/-3 – low chronologies), some 134 years earlier than Neferhotep I [Kitchen, 1987, pp. 43-50]. Neferhotep's first year (and, by the synchronism, the approximate date of Yantin) is thus set at c. 1738 (high) or 1696 (low).

This approximate date is however at odds with a set of Mesopotamian astronomically-derived dates via a second important synchronism with Yantin. Parrot's excavations at Mari, of the palace of Zimrilim, unearthed a magnificent cuneiform archive, one document of which contained an inventory of gifts from the local Levantine rulers, including a gold vase from 'Yantin-ammu, king of Byblos' [Dossin, 1939, p. 111]. The obvious step was to equate Dunand's Yantin with Parrot's Yantin-ammu and thus provide an important synchronism between Mesopotamia and Egypt. As Zimrilim's palace was destroyed by Hammurabi of Babylon in the latter's 32nd Year, it was possible to link Yantin with another astronomical date – that of Year 8 of Ammizaduga, as calculated from the Venus observations of the Enuma Anu Enlil texts [Langdon & Fotheringham, 1928]. The most probable date thus established for Year 1 Ammizaduga by Huber et al. [1982] is 1702 BC and by retrocalculation [Langdon & Fotheringham, p. 87] an absolute date of 1816 BC for Year 32 of Hammurabi has to be inferred. On these grounds, Yantin is to be dated to c. 1850 [Albright, 1945, p. 10] – a discrepancy of some 112 (high) or 154 (low) years with the Egyptian astronomically derived evidence.

In spite of the fact that recent, more accurate computer programmes have now confirmed that 1702 is the best 2nd Millennium candidate for Year 1 Ammizaduga, a lower alternative of 1582 has been generally accepted (even with the greater number of bad misses which result between the recorded observations and the computer retrocalculations). This gives a date of c. 1696 for Year 32 Hammurabi and c. 1730 for Yantin, a date which agrees well with the high Egyptian chronology date for Yantin of c. 1738.

Having put forward the standard chronological framework within which Tombs III and IV must be placed, I must state that my own personal view of this astronomically based chronology is radically different to that mentioned above and is compatible with a much lower chronological model, based on a revision of Egyptian Third Intermediate Period chronology and supported by the new computer calculations for the various astronomical observations of the ancient Near East [see Mitchell, 1990]. However, in this essay, it will be more convenient to use the standard dating for the period, with the proviso that this is only to be used as an aid to the relative dating of the early MBA.

Tombs I to IV at Byblos and their relative chronology

Reference to the map of the royal cemetery at Byblos (above) shows that there are two groups of tombs. Those of Area B are dated to the Late Bronze Age and include the tomb (V) of Ahiram, presumed by a majority of scholars to be a contemporary of Ramesses II. Tombs I to IV (Area A) are the earliest shaft tombs at the site and appear to run chronologically from west to east; thus Tomb I is the first of the series and Tomb IV the last.

I have already shown that Tomb IV probably belonged to Yantin and is to be dated to c. 1730 in the orthodox chronology. If we assume, with Kitchen [1967, p. 41], that Tomb III belongs to Yakin-El who is otherwise a contemporary of Sehetepibre (4th or 8th king of the 13th Dynasty), then we have a two-generation sequence of funerary artifacts to analyse. We might take one further step and assume that Tombs I and II do not precede III by more than a couple of generations, given their close proximity to III and the absence of other tombs to fill the gap between the two tomb pairings. Tomb I was identified by Montet as that of Abishemu and Tomb II attributed to his son Ibshemuabi, the former datable in Egyptian terms to Amenemhat III (obsidian vase – Montet, 1928, p. 155) and the latter to Amenemhat IV (obsidian box – Montet, 1928, p. 157). This dating for Tombs I and II is consistent with the hypothesis that the Area A necropolis comprises a sequence of four generations of Byblite rulers from Abishemu to Yantin – in Egyptian chronology the interval between Amenemhat III to Neferhotep I: some 76 to 100 years (according to Kitchen, 1987, pp. 49-50) and therefore a reasonable generation of between 20 and 25 years. Thus the burial sequence can be dated from c. 1830 to c. 1730 within the conventional framework.

Having established a general date range for the artifacts from these tombs, we should now review the assemblage and attempt to use the information to link other Palestinian sites to Egypt via Byblos.

Artifacts from the MBA necropolis at Byblos

Pottery Characteristics

Storage jars from all four tombs are typical of the Palestinian MB I ware, with rounded bases, looped handles and plain elongated bodies [Montet Atlas, pl. CXXIV]. These characteristics match closely those of the Tell el-Ajjul Courtyard Cemetery – the type-site for Kenyon's MB I (Albright's MB IIA).

Basin-shaped bowls from Tombs I, II & III [Montet Atlas, pl. CXVI] are identical to types from virtually all the MB sites in Palestine.

Dipper juglets are again exactly like those of the MB II Palestinian assemblages with curved bases, burnished fabric, and with pinched spouts [Montet Atlas, pl. CXXIV]; however types with ring bases were also found in Tombs I & II [Montet Atlas, pl. CXVIII, 800].

What can be inferred from the artifacts of the tombs at Byblos in fact tells us very little. The MB I phase is well established by the reign of Abishemu and therefore, due to the synchronism, with Amenemhat III in Egypt c. 1830; it continues with little change throughout the next century at least until about the point when Egypt's 14th Dynasty comes to power under Nehesi c. 1730. The recent discovery of a large MB I settlement at Tell el-Hebua in western Sinai may shed much further light on this later phase of MB I, as the excavator, Mohamed el-Maksoud, has retrieved two stelae of Nehesy from the site [el-Maksoud, 1983] and it may well prove possible to confirm the continuation of the MB sequence in a 14th Dynasty through to Hyksos 15th Dynasty context.

Similarly, the work of the Austrian Mission at Tell ed-Daba, under the direction of Manfred Bietak has now been able to confirm a number of Egyptian 13th and 14th Dynasty links with the MB archaeological phases revealed at the site [Bietak, Tell el-Daba V]. Figure 2 shows the latest Bietak chart giving the synchronisms adduced by the excavation team at Daba.

It is thus likely to be through the eastern Delta Egypto/Canaanite sites that a date for the beginning of MB I will be determined, rather than the site of Byblos itself, which can only provides samples of the fully developed MB I from its tomb assemblages.

Byblos and the Egyptian 6th Dynasty

If we now turn to the end of the Early Bronze Age at Byblos we again meet a synchronism with Egypt. During his excavations of the temple area, Montet discovered a large jar containing foundation deposits. The considerable quantity of scarabs bearing the prenomen Neferkare found in the jar point to its deposition during the reign of Pepi II and this appears to be confirmed by a total of 36 inscriptions of this king found within the Byblos site. At first it seemed that a number of these were retrieved from the half-metre thick ash layer which marked the end of EB III [Dunand, 1971, p. 20].

The contents of the ‘Montet Jar' was thus thought to be dated sometime before the major series of destructions and conflagrations which coincided with the end of the 6th Dynasty. The actual relative or absolute dating of this event is not, however, very easy to determine, as retrocalculations must be made from the 1872 Year 7 Senuseret III Sothic date. This is relatively simple for the 12th and 11th Dynasties, but only scanty regnal dates are known for the pre-11th Dynasty rulers based at Thebes and Heracleopolis. Kitchen gives a starting date for the 12th Dynasty at 1979 and that for the 11th Dynasty at c. 2122 [Kitchen, 1987, p. 49]. Thus, not only is the 6th Dynasty detached chronologically from the Middle Kingdom, but we also have no stratified artifacts specific to the kings of the 7th to mid-12th Dynasties in the Levant as a chronological guide. We are left with the Montet Jar at one end of the period and the Amenemhat III vase from the tomb of Abishemu at the other. Somewhere in this long synchronous void must fall the transition between Kenyon's EB-MB and the true MB I of the Byblos tombs.

Some of the usual artifactual types which mark the EB-MB in the rest of Palestine were found in the Montet Jar and it has been suggested that the new population groups, which were to occupy Palestine during the next two centuries following the collapse of the EB urban culture, consisted of two major elements – a semitic/Amorite majority along with a non-semitic/Anatolian minority group centred on the coastal region [Saghieh, 1983, p. 125] and resident, in particular, at Byblos itself.

A recent detailed study, made by Saghieh [1983], of Third Millennium Byblos has, however, revealed that we must make an important amendation to our understanding of the EB III destruction and its relationship to Egyptian chronology. She has shown that it was the JII town, following the EB III destructions, which appears to have been strongly influenced by the Indo-European newcomers and not the KIV town which predated the destruction by fire. This so-called non-semitic group, who were skilled in metal-working and who have been equated with a proto-Hurrian element [Schaeffer, 1949, pp. 49-120], may have later given rise to the tradition that the major Hyksos rulers (apparently bearing Hurrian/Indo-European names) originated from Phoenicia.

Characteristic of this northern group are the swollen-headed toggle pins [Montet Atlas, pl. LXIX] and the metal torques [Montet Atlas, pl. LXX] found in quantity in the Montet Jar foundation deposit - both of which are typical of the EB-MB but not attested in either the EB or MB.

Saghieh's JII level thus appears to mark the arrival of a new group around the time of the long reign of Pepi II, as evidenced by the appearance of this king's name associated with the construction of megaron type temples by these 'Torque Wearers'. If we regard the new group as the agents of destruction of the KIV city, then the discovery of Pepi's name in the ash layer by Montet may be at odds with this interpretation, allowing insufficient time for Egypt to establish new relations with the new group. If, however, the Pepi finds were intrusive to the destruction layer by the simple fact that they were depositions, then we can adopt the more attractive historical model proposed by Schaeffer [1948]. He argues that the EB III destructions were caused by a natural disaster rather than by military invasion. It would thus be logical to date these earthquakes and conflagrations to sometime before Pepi's reign, enabling us to see the arrival of the new population groups as a reaction to the catastrophic events, perhaps at the beginning of the 6th Dynasty. In other words they were symptomatic of the effect rather than the cause of the EB final catastrophe. The artifacts which distinguished this group from the Amorites were also found by Schaeffer at Ugarit, pointing to a seaborne or at least northern link with their country of origin [Dunand, 1971, p. 22].

Summary

The combined evidence seems to pinpoint the beginning of the EB-MB phase to shortly before the reign of Pepi II in Egypt and the arrival of the JII population at Byblos.

There is currently, however, no precise way to determine the beginning of the true MBI (Albright's MB IIA) period. It would be quite reasonable on historical grounds to link the latter to the beginning of the 12th Dynasty, when Egyptian rulers re-established trading links with Palestine and the Syrian coast; but the first truly datable synchronism does not occur until the reign of Amenemhat III of Egypt and Abishemu of Byblos.

It looks as if we must await the publication of Bietak's Tell el-Daba V volume on stratigraphy and chronology in order to shed further light on this difficult question and perhaps even then it will require a further detailed assessment of the Tell el-Hebua excavations before a precise picture emerges to clear away the haze of this little understood period.