At present, 53 14C dates from our project are available: 33 from Tel Beth-Shean and 20 from Tel Rehov, referring to various archaeological periods. This collection is one of the largest groups of 14C dates from historical~ anywhere in Israel. There are 32 dates related to the Iron Age: 12 from Beth-Shean and 20 from Tel Rehov. They range from the early 12th century until the 8th century BCE

All dates except nine were measured in the 14C laboratory of the Weizmann Institute in Israel. Most of these dates were obtained since 1998, using liquid scintillation counters. Nine grain samples from one context at Tel Rehov were dated with accelerator mass spectrometry (AMS) at the University of Arizona-NSF 14C laboratory (reported by Professor D Donahue). Other samples from the same context were dated at the Weizmann Institute (see Table 6). All the samples came from well-stratified contexts. Short-lived samples such as cereals, lentil seeds, and olive pits were used preferentially. When such samples were not available, charred timber was dated, following identification by the paleobotanist Dr Uri Baruch.

The standard deviation of most BP dates ranged from 25-50 yr, and in a few cases, 60-70 yr (1σ, 68% certainty). These BP dates were calibrated with the OxCal software (Bronk Ramsey 1999) using the 1998 calibration curve (Stuiver and van der Plicht 1998). Comparisons are made (in the endnotes) with the 1993 calibration curve (Stuiver and Reimer 1993). The differences between the 1993 and 1998 curves are usually negligible, but in two cases, both referring to the 12th century BCE, there is a significant and important difference.

The samples came from a building that contained local Canaanite pottery, Egyptian pottery, a single imported Cypriot White Slip bowl, and other artifacts which can be dated to either the end of the 19th Dynasty or beginning of the 20th Dynasty: around 1200 BCE The University Museum expedition found above this building structural remains attributed to Level VI, which was destroyed at the end of the Egyptian presence, perhaps around 1130 BCE (James 1966: 13-14, Figure 76: 1). We established that these structural remains were in fact renovations of the building excavated by us. Therefore, the latter building belongs to an earlier stage of the Egyptian occupation at Beth-Shean. In the adjacent Area S we found two stratigraphically distinct phases: S4 and S3, both belonging to the time period of the 20th Dynasty. It appears that our phase N4 in Area N corresponds with phase S4 in Area S, and the two should be dated to the early 20th Dynasty, i.e. the early 12th century BCE This date fits all three 14C dates in the one σ group according to the OxCal program (Bronk Ramsey 1995) using the 1998 calibration curve (Stuiver and van der Plicht 1998) (1).

The beams from Area C come from the construction of this level. The first (RT2995) is dated to the 14th-13th centuries BCE and thus points to the use of old olive wood in the construction, as in the case of Stratum S I at Beth-Shean (above). The second sample (RT2997) comes from an elm tree, which has a much shorter life span than an olive tree: its average life span in Israel today is about 50 years (information provided by U Baruch). Our elm tree beam is dated to the late II th early 10th century BCE in the 1σ range, while a lower date in the 10th century is suggested within the 2σ range. This date may therefore provide sound evidence for the 10th century date of construction of this building.

The olive tree wood from Area E (RT2996) gave a date between 980 and 840 BCE, a time range which fits almost exactly the entire Iron Age IIA phase to which Strata V-IV belong. However, such a range does not allow a more precise date within this time range.

The grain in these samples comes from the same building where the elm tree in Table 5 was used for construction. On the basis of the calculated average in Table 6 and the results of the previous paragraph, we thus may conclude that this building was constructed during the 10th century BCE and destroyed during the 9th century BCE, before 830 BCE These conclusions are in accord with the archaeological age assessment based on comparative pottery study (see Mazar 1999a: 37-42 and below).

The flat shape of the calibration curve between 800 and 400 BCE makes 14C dates almost useless for this period. The BP dates of the three samples in conventional 14C years differ by up to 145 years with a standard deviation of up to 40 years. The 1σ calibrated dates diverge: two of them are in accord with the archaeological date, while the third is too low by at least 200 years. The 2σ range of all three provides a time range of 410 years, which includes the 8th century BCE

What are the implications of our dates for the current controversy over the chronology of the Iron Age I-IIA in Israel? This dispute stems from I Finkelstein's suggestion in 1996 that archaeological assemblages traditionally attributed to the 12th-10th centuries BCE should be lowered by 50-80 years (Finkelstein 1996; 1998). In fact, a similar controversy existed in the early 1950s when B Maisler (Mazar) supported a low chronology (ending Tell Abu Hawam III in the late 9th century BCE) while G van Beek, following WF Albright, supported a high chronology (Tell Abu Hawam III in the 10th century BCE; Maisler 1951; van Beek 1955). Finkelstein's suggestion faced strong opposition from other scholars (Mazar 1997b; Ben-Tor and Ben-Ami 1998; Ben-Tor 2000). This controversy has far reaching implications on the correlation of archaeological data with the historical period of the United Monarchy of David and Solomon, as well as on correlations between the Levant, Cyprus and the Aegean in the Iron Age.

Tel Rehov is important for this discussion, since it produced one of the best stratigraphic sequences and abundant pottery assemblages from the Iron Age IIa in Israel (the suggested dates for this period according to Mazar are from around 980 BCE to around 930 BCE) (9). Strata VI, V, and IV at Tel Rehov, indicate a great deal of continuity in the pottery production during this period: red slip and hand burnish techniques are typical, and the pottery forms show only minor changes. The assemblage of these three strata recalls that of Megiddo Strata IVB-VA, Hazor Strata X-VIII, Ta'anach Periods IIA-IIB, Jezreel and other sites which belong to the same archaeological horizon. Finkelstein (1996; 1998) suggested dating all these assemblages to the 9th century BCE (in fact only to part of this century, ending ca. 830 BCE). However, Mazar (following earlier suggestions by Aharoni and Amiran) proposes to allow it a longer time span: from sometime during the first half of the 10th to around 840-830 BCE (Mazar 1997b). This would allow the appearance of the same assemblage in several strata of sites like Hazor and Tel Rehov. In these sites one can observe the continuity in pottery production from the 10th to the 9th century BCE The excavations at Jezreel seem to provide an historical anchor for the end of this assemblage (for more detailed discussion see Mazar 1999a: 37-42), yet not for its beginning, which probably dates back to the 10th century BCE

Finkelstein (in lectures submitted during the year 2000), Sharon and Gilboa (2001) claim that 14C dates from Megiddo and Dor support the low chronology suggested by Finkelstein. Our results are ambiguous. Our dates of timber used in construction at Tel Rehov Stratum V and Beth-Shean Stratum S I tend to show that in both cases the buildings were erected during the 10th century BCE. As mentioned before, however, these dates may be interpreted as providing only a terminus post quem for the construction, and thus it could be claimed that the buildings could have been built in the 9th century BCE. Such a claim, however, seems to be untenable in the case of the elm tree from Tel Rehov. Its radiometric date (RT 2997) indicates that it was cut in the beginning of the 10th century, and it would not be logical to assume that it was used for construction almost 100 years later.

The calibrated average date of the 18 measurements of grain seeds from the destruction of Tel Rehov Stratum IV (Cluster 6 and Table 6) is 900-830 BCE This gives the possible range of dates of this destruction. In the first preliminary report on Tel Rehov the destruction of Stratum IV (= C1 in that report) was assigned with reservations to the second half of the 9th century BCE, though a possibility for an earlier date was not ruled out (Mazar 1999a: 41-42). This conclusion was based on both the 14C dates and on one Cypriot pottery vessel: a bichrome globular jug which is considered in !cyprus to belong to the Cypro Geometric III period, not earlier than 830 BCE. However, it should ( be recalled that the chronology of Iron Age Cyprus depends to a large extent on that of the Levant, and that the attribution of types to chronological periods is based on mere typological considerations, and perhaps needs reevaluation. Thus the date of this destruction could be anywhere in the time range provided by the calibrated average mentioned above.

The chronological debate concerning the 10-9th centuries BCE in Israel is over a time range of between 50 and 100 years. Can 14C dates contribute to a debate over such a narrow time span? The 14C dates discussed above show that modern, sophisticated dating technology, careful selection of well stratified samples and a sufficient number of 14C dates may provide an important contribution, even to chronological debates over such a narrow time span. There are, however, problems, which may hamper the utilization of 14C dates for historical periods. Some of these are:

Debates over the dates of archaeological strata are unavoidable. In spite of the above-mentioned problems, 14C dates are our last resort in establishing a precise as possible absolute chronology for the Southern Levant in the time span between the mid 12th century BCE and the late 8th century BCE. The current debate over the 10th-9th centuries BCE is an excellent case study. Yet it seems that there is a long way to go before the final word will be said in this debate.