• from Garfinkel and Ganor (2008:2-3)

Khirbet Qeiyafa is a 2.3 hectare site, surrounded by massive fortifications of megalithic stones that remain standing to a height of 2–3 meters. The site is located in the western part of the high Shephelah (Israel map grid 14603 12267), atop a hill that border the Elah Valley on the north. This is a key strategic location in the biblical kingdom of Judah, on the main road from Philistia and the Coastal Plain to Jerusalem and Hebron in the hill country. Two kilometers to the west lies Tell Zakariyeh, commonly identified as biblical Azekah and two and a half kilometers to the southeast is Khirbet Shuwayka, commonly identified with biblical Socoh.

A number of European explorers visited Khirbet Qeiyafa during the 19th century: V. Guerin, C.R. Conder and H.H. Kitchener. During the 20th century, the site was neglected; it is not referred to in the works of the leading scholars in the field of biblical historical geography, such as W.F. Albright, B. Mazar, Y. Aharoni and Z. Kallai. In the last 20 years, archaeological surveys of the Shephelah region, conducted by Y. Dagan and Z. Greenhut, revisited the site (for further information and references see Garfinkel and Ganor 2008).

The site aroused interest in 2005 when Saar Ganor noted impressive Iron Age structures under later remains. The authors examined the site in 2007 and in 2008 conducted a six-week excavation season on behalf of the Institute of Archaeology of the Hebrew University of Jerusalem. During these two seasons nearly 600 sq. m of an Iron Age IIA city were unearthed in the western part of the site, in Area B where a four chamber gate, casemate city wall and two buildings were found (Fig. 1). A unique find here was an ostracon, a five-line inscription written in ink on a pottery sherd. The Iron Age IIA remains were founded on bedrock and covered by a thin Hellenistic layer. No superimposed Iron Age living floors or walls were found at the site, indicating a single phase of Iron Age settlement lasting for a short period of time, probably not more than 20 years.

The expedition undertook additional field work during the first week of November 2008. A second gate has been identified in the eastern part of the city (Area C). The front side of the gate is composed of two monumental blocks of stone, one on each side (Fig. 2). Each stone has an estimated weight of 10 tons. This is the most massive gate ever found in any biblical city to date. The enormous efforts invested in the gate’s construction far exceed technical requirements and was clearly intended as a statement of power and authority. The eastern gate, facing Jerusalem, was the main entrance to the city.

Khirbet Qeiyafa is the only site in the Kingdoms of Judah and Israel with two gates. Even cities three or four times its size, such as Lachish and Megiddo, have only a single gate. This unique feature provides a clear indication of the site’s identity as biblical Sha’arayim, a place name that means “two gates” in Hebrew.

There has been broad speculation over the years regarding the biblical name of Khirbet Qeiyafa, none based upon solid evidence. The name Sha’arayim has been proposed twice: once during the 2008 excavation season by David Adams of Concordia Seminary in St. Louis; and in midSeptember by Anson Rainey of Tel Aviv University. On those occasions, the existence of a second gate was not yet known.

The dating of the Iron IIA city of Khirbet Qeiyafa is based on relative and absolute chronology. The relative chronology is early Iron IIA, as indicated by the pottery. The characteristic bell-shaped Philistine bowls, known in large quantities from Tell Qasile Stratum X (Mazar 1985, Fig. 34:1–10) and Tell Miqne Stratum IV (Ortiz 2000) are entirely absent. Thus, the site cannot be dated to the late Iron Age I. Currently there is a debate over whether Iron Age I ended ca. 1000 BC or 920/900 BC (Mazar and Bronk Ramsey 2008; Finkelstein 1996; Sharon et al. 2007). On the basis of our current knowledge, Khirbet Qeiyafa cannot be dated to the 11th century BC.

Absolute dating at Khirbet Qeiyafa is based on four burnt olive pits, measured at Oxford University. The results of the first two samples were sent to us on 6 October 2008 and the other two samples on 7 November 2008:

• OxA-19425 Qeiyafa 5 2851±31
• OxA-19426 Qeiyafa 6 2837±29
• OxA-19588 Qeiyafa 7 2799±31
• OxA-19589 Qeiyafa 1 2883±29

The average of these four measurements, as provided by Christopher Bronk Ramsey of Oxford University is: 2844±15 (Fig. 3). After calibration the dating is 1026–975 BCE (59.6%) or 1051–969 BCE (77.8%). As Khirbet Qeiyafa is an Iron Age IIA site, we are left with a dating post-1000 BCE, that is, 1000–975 BCE (59.6%) or 1000–969 BCE (77.8%). These dates fit the time of King David (ca. 1000–965 BCE) and are too early for King Solomon (ca. 965–930 BCE).

• from Garfinkel and Ganor (2008:3-4)

Sha ray ’a im is mentioned three times in the Bible:

1. In the city list of the tribe of Judah it appears after Socoh and Azekah (Josh 15:36). As mentioned above, Socoh is located 2.5 km to the southeast of Khirbet Qeiyafa and Azekah 2 km to the west of Khirbet Qeiyafa. Thus, the geographical location of Sha’arayim in the Elah Valley region suits its location in the biblical list.
2. After David killed Goliath, in the Elah Valley, between Socoh and Azekah, the Philistines escaped through the “road of Sha’arayim” (1 Sam 17:52). In this case, again, Sha’arayim is mentioned in closed proximity to the Elah Valley, Socoh and Azekah.
3. In the city list of the tribe of Simeon, Sha’arayim is mentioned as one of the cities “until the reign of David” (1 Ch 4:31–32).

One may claim that there were two cities called Sha’arayim, one near the Elah Valley and the other in the Negev. However, the appearance of Sha’arayim in this list is quite problematic. There are three parallel texts for the city list of the Negev and the tribe of Simeon: Joshua 15, Joshua 19 and 1 Chronicles 4. In the same location in each list a different name appears: Shilhim in Josh 15:32, Sharuhen in Josh 19:6, and Sha’arayim in 1 Chr 4:31. These changes are usually interpreted as scribe errors since all three names are begin with the letter shin. Sha’arayim does not appear in the two other lists, so there was, in fact, no second city with this name in the Negev.

Generally neglected by scholars is the association of the closing words: “until the reign of David” only with the Sha’arayim version and immediately adjacent to this name. Apparently, the text of 1 Chr 4:31–32: “Sha’arayim. These were their towns until the reign of David” is a remnant of a list, now lost, that included cities from the period before David’s regency.

There has been speculation over the years regarding the location of biblical Sha’arayim (for a survey of views, see Rainey 1982), though lacking a solid base. Very important to us is the association of Sha’arayim with King David twice in the biblical tradition. Sha’arayim is not mentioned in conjunction with any other later First Temple period tradition. This observation is consistent with the archaeological and radiometric data that indicate a single-phase settlement in the early 10th century BCE at Khirbet Qeiyafa.

• from Thomas (2014:38-40)

Compared to Megiddo, which has remained central to the archaeology of ancient Israel and this debate in particular for a very long time, Khirbet Qeiyafa is a relative newcomer but its interpretation has come to attract much interest and it has taken up an important place in this debate. Khirbet Qeiyafa is a fortified site upon a hill to the north of the Valley of Elah between the lowland coastal Shephelah plain and the Judean hill country, by a road leading up towards Jerusalem. The site’s excavators Garfinkel and Ganor uncovered a settlement with a single Iron Age phase featuring no sub-strata or modifications, apparently dating to the early Iron IIA. This included buildings, a casemate wall and two four chambered gates⁷⁷. Along with the apparent early Iron IIA pottery, the site has produced several radiocarbon results. As the site appeared to have only one short period these results were taken as roughly contemporary and were therefore averaged to produce an approximate date for the site in the late 11th-early 10th century BC.⁷⁸ The excavators have identified the site as Judean, regarding its city planning, material and material culture as indicative of such a site, which they feel is not reflected in northern and non-Israelite sites. They argue that its layout with a casemate wall and abutting houses is an early precursor to the layout of later southern sites such as nearby Beth-Shemesh⁷⁹ . To this end, the excavators have stated that their dating of the site, being Iron IIA with radiocarbon dates possibly pointing to the start of occupation in the 11th century BC, contradicts the Low Chronology and points to a possible transition from Iron I to IIA that conforms more to the High Chronology. Further they state that though the Low Chronology has led to suggestions that fortified sites previously assigned to the 10th century BC should be down dated and that there were no southern cities with actual fortification walls until the 9th century BC, Khirbet Qeiyafa with its two gates and casemate wall contradicts this view⁸⁰ . They conclude that such a construction effort could only be linked to a strong central authority, namely a state like the one in Jerusalem with its conflicts with nearby Gath, and not just some local polity⁸¹ .

The relative dating of Khirbet Qeiyafa based upon the classification of its pottery has been disputed. Singer-Avitz initially proposed that though there is difficulty in distinguishing Iron I and IIA pottery assemblages due to continuity in forms, the site should in fact be considered an Iron I site due to certain forms which are only otherwise known in Iron I strata at other sites⁸² . Garfinkel with Kang retorted that Singer-Avitz had not followed her own criteria in offering examples of what she saw as exclusively Iron I pottery forms that did, in their opinion, appear in Iron IIA strata at other sites as well and confirmed their dating of Khirbet Qeiyafa to the early Iron IIA⁸³ . They point specifically to the characteristic hand-burnished red-slipped pottery typical of the Iron IIA which occurred though only in rather small amounts at Khirbet Qeiyafa⁸⁴ . Singer-Avitz again insisted that the site was Iron I in date, responding to Garfinkel and Kang’s criticisms of her identification of certain pottery forms as belonging to only the Iron I. She argued that Garfinkel and Kang had misidentified some pottery as having equivalents in Iron I and that they had overlooked some instances where pottery that they claimed was exclusively Iron IIA had in fact been found in Iron I contexts⁸⁵ . With regards to the hand-burnished red-slipped pottery, Singer-Avitz noted that it had also been found in small amounts in a few other Iron I layers⁸⁶.

Finkelstein has preferred to associate the site with the emerging northern kingdom rather than with the early state of Judah and Jerusalem in the southern highlands. This affiliation is based primarily upon the existence of Iron I casemate defences in what appeared to be the heart of the early northern kingdom, around Gibeon and Bethel, as opposed to Iron IIB sites like nearby Beth-Shemesh claimed as later developments of Khirbet Qeiyafa’s planning by the excavators⁸⁷. Finkelstein doubts an association with Judah due to the lack of comparable fortified sites nearer to its heartland around Jerusalem. Further, Finkelstein argues that present indications of population of the highlands in the Iron I period based upon the number and size of settlements in the two respective areas render it unlikely that Judah had the manpower to build a site like Khirbet Qeiyafa, as the population there seems to have been rather sparse while in the north it was much larger⁸⁸.

Finkelstein has also presented a different understanding of the excavators’ radiocarbon results. He rejected the excavators’ practice of averaging the Iron Age samples because he did not accept that they could be certain that these samples all came from the same moment in the settlement’s lifespan because they were not from a single event like a clear destruction layer. Therefore, he has accepted only that the samples indicated a probable lifespan of the settlement. As he additionally accepts Singer-Avitz’s aforementioned assignment of the site to Iron I, he therefore dates the site based only on the highest and lowest dates, to between about 1050 and 915 BC, in line with the Low Chronology⁸⁹. The excavators have criticised this approach as they state that under Finkelstein’s long expanse of occupation about six generations would have occupied the site without having made any changes to the site’s buildings whatsoever, which they judge as impossible. They are therefore confident in averaging the radiocarbon dates and their resultant earlier date for the site⁹⁰.

Only one phase of occupation was discerned at Khirbet Qeiyafa during Iron Age IIA - a period assignment that was derived from radiocarbon, architectural style (casemate gates), and pottery. Earlier Iron I remains were not found below it and later Iron II remains were not found above it (Garfinḳel & Ganor, 2009:4). The occupation was short-lived. Garfinḳel & Ganor (2009:75) noted that in Area B it seems that the settlement existed for only a short period (20–30 years), as no evidence was found indicating the elevation of floors or the construction of one wall over another. Occupation ended in a massive destruction which contained fallen objects, broken pottery and items of value in the rubble. Garfinḳel & Ganor (2009:85) described the destruction layer in Area B as follows:

Settlement on the site probably ended in massive destruction. This is indicated by the large quantities of restorable vessels found in the various rooms (figs. 5.51–5.58). Such fragments were found not only on the floors, but among the debris that accumulated between the walls. These vessels probably fell from shelves, from roofs or from higher floors. Valuable objects, such as an iron blade (fig. 9.36) and an elaborate basalt bowl (fig. 5.57), also indicate that the site was not left in a peaceful manner. However, one typical destruction characteristic is missing — there is no evidence of burning, aside from one room in the eastern part of the site, which has not been fully excavated yet. There could be various reasons for the absence of burning; the stones from which the houses are constructed may have prevented a serious fire, or perhaps the city was destroyed during winter rains. Signs of burning may be found in future in other parts of the site.

An opinion was not offered as to the cause of destruction - e.g. due to military conquest or seismic activity. Pottery and Radiocarbon suggested a date of occupation around 1000 BCE¹ and during the reign of King David - which is compatible with High Chronology or Modified Conventional Chronology of Iron Age IIA but not with Low Chronology. Garfinḳel & Ganor (2009:89, 91, 94) dated construction of the Iron IIA gate in Area B as follows:

The dating of the initial construction of this gate to the Iron Age is based on three aspects:

1. The houses and the casemate city wall abut the gate and form a single architectural stratum. The gate cannot be later than the houses and the city wall.


2. Pottery and ¹⁴C results from the houses and the second casemate clearly date these structures to the Iron Age IIA.


3. Under the thick Hellenistic sediment in the passageway, a thin Iron Age layer was found, which was preserved in the eastern part of the area. The northern drain was found below this layer. This layer includes only Iron Age pottery. Radiometric dating of an olive pit from this layer gave a date in the tenth century BCE calibrated (“Qeiyafa 7”, tables 3.2–3.3).

Collapsed stones were also found inside the casemates at Area B. Garfinḳel & Ganor (2009:95) described these as follows:

Inside the casemates were found thousands of collapsed stones (figs. 5.74–5.75). These stones came from the upper sections, which collapsed either during the destruction of the city or later, when it was standing in ruins. It is estimated that c. 200,000 tons of stone were needed for the construction of the city wall.

Collapsed stones were also found outside of and downslope from the gate in Area C.

Effect	Location	Image(s)	Description
Broken Pottery	Area B Fig. 5.41 Area B: detailed plan of the Iron Age remains at the end of the 2008 season. Garfinḳel & Ganor (2009) Fig. 5.42 Area B: schematic plan of the Iron Age remains at the end of the 2008 season. Garfinḳel & Ganor (2009)	Fig. 5.51 Area B: close-up of destruction debris with restorable cooking pot and a lamp on the floor of Building II. Garfinḳel & Ganor (2009) Fig. 5.52 Area B: close-up of destruction debris with fragments of a baking tray in Room B of Building I. Garfinḳel & Ganor (2009) Fig. 5.53 Area B: Destruction debris in Locus B169, Room A in Building I. Garfinḳel & Ganor (2009) Fig. 5.54 Area B: close-up of destruction debris in Locus B269. Garfinḳel & Ganor (2009) Fig. 5.55 Area B: close-up of destruction debris in Locus B134. Garfinḳel & Ganor (2009) Fig. 5.56 Area B: close-up of destruction debris in Locus B130. Garfinḳel & Ganor (2009) Fig. 5.57 Area B: close-up of destruction debris, a basalt bowl and a pottery fragment in Locus B127, Building II. Garfinḳel & Ganor (2009) Fig. 5.58 Area B: close-up of destruction debris in Locus B122. Garfinḳel & Ganor (2009) Fig. 5.90 Area B: destruction debris on the floor of Room B in Building I. Two storage jars and a chalice were later restored from this pottery concentration. Garfinḳel & Ganor (2009) Fig. 5.91 Area B: a juglet found in the destruction debris of Room B in Building I. Garfinḳel & Ganor (2009)	Settlement on the site probably ended in massive destruction. This is indicated by the large quantities of restorable vessels found in the various rooms (figs. 5.51–5.58). Such fragments were found not only on the floors, but among the debris that accumulated between the walls. These vessels probably fell from shelves, from roofs or from higher floors. - Garfinḳel & Ganor (2009:85)
Objects of value found in rubble	Area B Fig. 5.41 Area B: detailed plan of the Iron Age remains at the end of the 2008 season. Garfinḳel & Ganor (2009) Fig. 5.42 Area B: schematic plan of the Iron Age remains at the end of the 2008 season. Garfinḳel & Ganor (2009)	Fig. 5.57 Area B: close-up of destruction debris, a basalt bowl and a pottery fragment in Locus B127, Building II. Garfinḳel & Ganor (2009)	Settlement on the site probably ended in massive destruction. This is indicated by the large quantities of restorable vessels found in the various rooms (figs. 5.51–5.58). Such fragments were found not only on the floors, but among the debris that accumulated between the walls. These vessels probably fell from shelves, from roofs or from higher floors. Valuable objects, such as an iron blade (fig. 9.36) and an elaborate basalt bowl (fig. 5.57), also indicate that the site was not left in a peaceful manner. - Garfinḳel & Ganor (2009:85)
Collapsed Stones	Areas Band C Fig. 5.41 Area B: detailed plan of the Iron Age remains at the end of the 2008 season. Garfinḳel & Ganor (2009) Fig. 5.42 Area B: schematic plan of the Iron Age remains at the end of the 2008 season. Garfinḳel & Ganor (2009) Cover Image Site Plan of Khirbet Qeiyafa Garfinkel et al. (2016)	Fig. 5.48 Area B: the northern section of Square N32 inside a casemate of the city wall. Most of the fill (c. 3 m.) is made of fallen stones of various sizes. Garfinḳel & Ganor (2009) Fig. 5.74 Area B: the appearance of the first Iron Age casemate. Fig. 5.22 presents the same location after a few more days of excavation. Note the Hellenistic wall constructed only on the outer wall of the casemate. Garfinḳel & Ganor (2009) Fig. 5.75 Area B: collapsed stones in the first casemate (view to the north). Each casemate was filled with thousands of stones like these, apparently collapsed from the upper parts of the wall. Garfinḳel & Ganor (2009) Fig. 5.113 Area C: the southern gate before clearing collapsed stones (view to the west). The northeastern corner of the gate was recognizable prior to excavation. Garfinḳel & Ganor (2009)	Inside the casemates [of Area B] were found thousands of collapsed stones (figs. 5.74–5.75). These stones came from the upper sections, which collapsed either during the destruction of the city or later, when it was standing in ruins. - Garfinḳel & Ganor (2009:85)

• Modified by JW from Fig. 5.41 of Garfinḳel & Ganor (2009)

• Earthquake Archeological Effects chart
  Earthquake Archeological Effects (EAE)
  
  Rodríguez-Pascua et al (2013: 221-224)
  of Rodríguez-Pascua et al (2013: 221-224)

Effect	Location	Image(s)	Description	Intensity
Broken Pottery	Area B Fig. 5.41 Area B: detailed plan of the Iron Age remains at the end of the 2008 season. Garfinḳel & Ganor (2009) Fig. 5.42 Area B: schematic plan of the Iron Age remains at the end of the 2008 season. Garfinḳel & Ganor (2009)	Fig. 5.51 Area B: close-up of destruction debris with restorable cooking pot and a lamp on the floor of Building II. Garfinḳel & Ganor (2009) Fig. 5.52 Area B: close-up of destruction debris with fragments of a baking tray in Room B of Building I. Garfinḳel & Ganor (2009) Fig. 5.53 Area B: Destruction debris in Locus B169, Room A in Building I. Garfinḳel & Ganor (2009) Fig. 5.54 Area B: close-up of destruction debris in Locus B269. Garfinḳel & Ganor (2009) Fig. 5.55 Area B: close-up of destruction debris in Locus B134. Garfinḳel & Ganor (2009) Fig. 5.56 Area B: close-up of destruction debris in Locus B130. Garfinḳel & Ganor (2009) Fig. 5.57 Area B: close-up of destruction debris, a basalt bowl and a pottery fragment in Locus B127, Building II. Garfinḳel & Ganor (2009) Fig. 5.58 Area B: close-up of destruction debris in Locus B122. Garfinḳel & Ganor (2009) Fig. 5.90 Area B: destruction debris on the floor of Room B in Building I. Two storage jars and a chalice were later restored from this pottery concentration. Garfinḳel & Ganor (2009) Fig. 5.91 Area B: a juglet found in the destruction debris of Room B in Building I. Garfinḳel & Ganor (2009)	Settlement on the site probably ended in massive destruction. This is indicated by the large quantities of restorable vessels found in the various rooms (figs. 5.51–5.58). Such fragments were found not only on the floors, but among the debris that accumulated between the walls. These vessels probably fell from shelves, from roofs or from higher floors. - Garfinḳel & Ganor (2009:85)	VII+
Displaced or Collapsed Walls suggested by Objects of value found in rubble	Area B Fig. 5.41 Area B: detailed plan of the Iron Age remains at the end of the 2008 season. Garfinḳel & Ganor (2009) Fig. 5.42 Area B: schematic plan of the Iron Age remains at the end of the 2008 season. Garfinḳel & Ganor (2009)	Fig. 5.57 Area B: close-up of destruction debris, a basalt bowl and a pottery fragment in Locus B127, Building II. Garfinḳel & Ganor (2009)	Settlement on the site probably ended in massive destruction. This is indicated by the large quantities of restorable vessels found in the various rooms (figs. 5.51–5.58). Such fragments were found not only on the floors, but among the debris that accumulated between the walls. These vessels probably fell from shelves, from roofs or from higher floors. Valuable objects, such as an iron blade (fig. 9.36) and an elaborate basalt bowl (fig. 5.57), also indicate that the site was not left in a peaceful manner. - Garfinḳel & Ganor (2009:85)	VII+ or VIII+
Displaced or Collapsed Walls ? - suggested by Collapsed Stones	Areas Band C Fig. 5.41 Area B: detailed plan of the Iron Age remains at the end of the 2008 season. Garfinḳel & Ganor (2009) Fig. 5.42 Area B: schematic plan of the Iron Age remains at the end of the 2008 season. Garfinḳel & Ganor (2009) Cover Image Site Plan of Khirbet Qeiyafa Garfinkel et al. (2016)	Fig. 5.48 Area B: the northern section of Square N32 inside a casemate of the city wall. Most of the fill (c. 3 m.) is made of fallen stones of various sizes. Garfinḳel & Ganor (2009) Fig. 5.74 Area B: the appearance of the first Iron Age casemate. Fig. 5.22 presents the same location after a few more days of excavation. Note the Hellenistic wall constructed only on the outer wall of the casemate. Garfinḳel & Ganor (2009) Fig. 5.75 Area B: collapsed stones in the first casemate (view to the north). Each casemate was filled with thousands of stones like these, apparently collapsed from the upper parts of the wall. Garfinḳel & Ganor (2009) Fig. 5.113 Area C: the southern gate before clearing collapsed stones (view to the west). The northeastern corner of the gate was recognizable prior to excavation. Garfinḳel & Ganor (2009)	Inside the casemates [of Area B] were found thousands of collapsed stones (figs. 5.74–5.75). These stones came from the upper sections, which collapsed either during the destruction of the city or later, when it was standing in ruins. - Garfinḳel & Ganor (2009:85)	VII+ ? or VIII+ ?

The archeoseismic evidence requires a minimum Intensity of VII (7) or VIII (8) when using the Earthquake Archeological Effects chart of Rodríguez-Pascua et al (2013: 221-224).