CoChat
Sign Up
Cited by 71Archaeology and ancient environmental studiesBiblical Studies and InterpretationAncient and Medieval Archaeology Studies

14C Dates and the Iron Age Chronology of Israel: A Response

Open access
Mazar, Amihai, Bronk Ramsey, Christopher·2008·Radiocarbon
DOI 10.1017/s0033822200033506
Open this paper in CoChat — free accountOpen PDF

Abstract

Boaretto et al. (2005) published 68 radiocarbon dates relating to 30 samples from 10 Iron Age sites in Israel as part of their Early Iron Age Dating Project. Though the main goal of their paper was an interlaboratory comparison, they also presented results of Bayesian models, calculating the transition from Iron Age I to Iron Age II in Israel to be about 900 BCE instead of the conventional date of about 1000 BCE. Since this date has great importance for all of Eastern Mediterranean archaeology, in this paper we examine the results in light of the dates published in the above-mentioned article. Our paper was revised in light of new data and interpretations published by Sharon et al. (2007). Following a survey of the contexts and specific results at each site, we present several Bayesian models. Model C2 suggests the date range of 961–942 BCE (68% probability) for the transition from Iron Age I to Iron Age II, while Model C3 indicates a somewhat later date of 948–919 BCE (compare the date 992–961 BCE calculated at Tel Rehov for the same transition). In our Model D, we calculated this transition date at Megiddo as taking place between 967–943 BCE. Finally, we calculated the range of dates of major destruction levels marking the end of the Iron Age I, with the following results: Megiddo VIA: 1010–943 BCE; Yoqne'am XVII: 1045–997 BCE; Tell Qasile X: 1039–979 BCE; Tel Hadar: 1043–979 BCE (all in the 68.2% probability range). Figure 4 indicates that the transition between Iron I and II probably occurred between these above-mentioned destruction events and the dates achieved in our Models C2 or C3, namely during the first half of the 10th century BCE. This study emphasizes the sensitivity of Bayesian models to outliers, and for reducing or adding dates from the models. This sensitivity should be taken into account when using Bayesian models for interpreting radiometric dates in relation to subtle chronological questions in historical periods.

AI snapshot

Written from the paper’s full text.

TL;DR

This response reanalyzes radiocarbon dates from the Early Iron Age Dating Project and argues that the Iron I/II transition in Israel was in the mid-10th century BCE, not about 900 BCE. Bayesian models C2 and C3 place the transition at 961–942 BCE and 948–919 BCE (68% ranges), with late Iron I destruction layers supporting a 10th-century transition.

Key findings

  • The authors reject the conclusion of Boaretto et al. (2005) and Sharon et al. (2007) that the Iron I/II transition occurred about 900 BCE; their Model C2 gives 961–942 BCE (68% probability) and Model C3 gives 948–919 BCE (68.2%; 95.4%: 963–913 BCE).
  • Model D for Megiddo is reported in the abstract as placing the transition at 967–943 BCE; the later Model D2 text gives 975–943 BCE (58.2%) or 1000–928 BCE (95.4%).
  • Destruction layers marking the end of Iron Age I are dated in the abstract as Megiddo VIA: 1010–943 BCE; Yoqne'am XVII: 1045–997 BCE; Tell Qasile X: 1039–979 BCE; Tel Hadar IV: 1043–979 BCE (all 68.2% ranges).
  • Several key sites yield dates that contradict the Low Chronology, including short-lived samples from Megiddo VIA and IVB-VA, Tell Qasile X, and Tel Hadar IV; the authors also cite a Tel Rehov transition date of 992–961 BCE (68% probability).

Methods

  • Reanalysis of published radiocarbon dates: Boaretto et al. (2005) published 68 dates from 30 samples at 10 Iron Age sites; the authors add dates from Sharon et al. (2007) and other sources, while excluding some sites and dates (e.g., the 64 Groningen Tel Rehov dates, Tel Dor Weizmann dates, Khirbet en-Nahas, Tel Dan, and others).
  • Bayesian chronological models A–D were built in OxCal, with phases for Late Bronze/Iron I, late Iron I, and Iron IIA, plus a Megiddo-only model; variants removed outliers and excluded charcoal/wood dates.
  • The analysis also combines dates from individual destruction layers and applies χ² tests to test whether they reflect a single event, then compares model transition dates and destruction ranges against the High, Modified Conventional, and Low Chronologies.

Limitations

  • The paper states that calibrated 14C precision is limited, Bayesian models are sensitive to adding or deleting dates, and a single date such as QS2 from Tell Qasile can change results considerably; future publication and analysis may change the picture.
  • The authors note that not all available dates were included, the Early Iron Age Dating Project is still in its second phase, early Iron IIA is hardly represented in the samples, and the models assume a synchronous regional Iron I/II transition, which may bias results.
  • The text also makes evident that outlier removal is judgment-based and affects model agreement and outcomes, and it reports slightly different ranges in places (e.g., Model C2 as 961–942 BCE in the abstract versus 960–940 BCE in the conclusions; Yoqne'am XVII as 1045–997 BCE in the abstract versus 1013–944 BCE in a later table).

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1017/s0033822200033506

Cites

27 of 27 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

23 of 40 cited sources checked · 9 supported · 1 flagged · accuracy not yet measured

  • Premisecitation

    The 2 dates BD1-2 from Bethsaida are not definitely attributed to Iron I or Iron II by Boaretto et al. (2005).

    “We have not included in Model A the 2 dates BD1-2 from Bethsaida, since these are not definitely attributed to Iron I or Iron II by Boaretto et al. (2005)”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Boaretto et al. 2005 source

      The Introduction discusses uncertain Iron Age assignments at Tel Rehov, but not the two Bethsaida dates. Searches of our copy for Bethsaida, BD1, and BD2 found no passage addressing whether those dates were assigned to Iron I or Iron II. (LLM, conf=0.72)

      Experimental — accuracy not yet measured
  • Premisecitation

    Sample MQ1-2 was published by Boaretto et al. (2005) as belonging to Tel Miqne Stratum IV.

    “Sample (=MQ1-2) was published byBoaretto et al. (2005) as belonging to Tel Miqne Stratum IV”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Boaretto et al. 2005 source

      The Materials and Methods passage says the study used samples from Iron Age sites, but the passages retrieved from our copy do not identify sample MQ1-2 or assign it to Tel Miqne Stratum IV. Searches for “MQ1-2,” “Tel Miqne,” and “Stratum IV” did not establish the claimed attribution. (LLM, conf=0.70)

      Experimental — accuracy not yet measured
  • Study design and conductreasoning

    Sample 4501.1-3 is omitted from Table 1 because its date is much too low.

    “We omit sample 4501.1-3(Sharonetal. 2007:24)as it is much too low.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Sharon et al. 2007:24 source

      Table 2 (page not available in our copy) lists sample 4501 from Megiddo K/6, supporting the citation as an identification of the sample. The passages found do not identify measurements 4501.1–3 specifically. In the citing sentence, the decision to omit the sample because it is “much too low” is the citing authors’ own, not a finding attributed to Sharon et al. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
    • Rests on an assumption: The expected date for sample 4501.1-3 is known and is not as low as the measured date.
  • Findingcitation

    Sample 3949 (MG13 and MG14) from Megiddo Level H5 clearly supports a 10th century BCE date, with an average calibrated date of 1000-930 BCE.

    “Sample 3949 (MG13 and MG14) clearly supports a 10th century BCE date (its average calibrated date is 1000-930 BCE; Sharon et al. 2007: 43)”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Sharon et al. 2007 source

      Table 2 identifies sample 3949 as from Megiddo H5, and the calibrated-date listing (entry 67–68, indexed under References; page unavailable) gives a 1005–930 BCE range, close to the claimed 1000–930 BCE. But it also gives wider calibrated ranges extending outside the 10th century, so the claim that the sample clearly establishes a 10th-century date overstates the evidence. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
  • Findingreasoning

    The single date QS2 from Tell Qasile could change the results considerably, especially when only the dates from Boaretto et al. (2005) are included in the analysis.

    “In our case, the single date QS2 from Tell Qasile could change the results considerably, especially when only the dates from Boaretto et al. (2005) are included in the analysis.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Boaretto et al. (2005) source
  • Classified as probably not a substantial claim. 15
  • May combine two claims. 38

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Sharon et al. (2007:8-9) decided to eliminate the highest date in the sample 3932 set as an outlier.

    “Sharon et al. (2007:8-9) discuss the problem of sample 3932 and decided to eliminate the highest date in this set as an outlier.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Sharon et al. 2007:8-9 source

      The Outlier Analysis discusses removing the highest date, 3932aa, as a possibility, but explicitly says the outlier-removal procedure reached an acceptable fit before any sample 3932 measurement was removed. The authors conclude that the prudent course was to leave both measurements in. (LLM, conf=0.90)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured

  • Classified as probably not a substantial claim.

    Premisecitation

    Boaretto et al. (2005) presented Bayesian models calculating the transition from Iron Age I to Iron Age II in Israel to be about 900 BCE instead of the conventional date of about 1000 BCE.

    “they also presented results of Bayesian models, calculating the transition from Iron Age I to Iron Age II in Israel to be about 900 BCE instead of the conventional date of about 1000 BCE.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Boaretto et al. (2005) source

  • Classified as probably not a substantial claim.

    Premisecitation

    Boaretto et al. (2005) published 68 radiocarbon dates relating to 30 samples from 10 Iron Age sites in Israel as part of their Early Iron Age Dating Project.

    “Boaretto et al. (2005) published 68 radiocarbon dates relating to 30 samples from 10 Iron Age sites in Israel as part of their Early Iron Age Dating Project.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Boaretto et al. (2005) source

      The Materials and Methods section describes an Israeli Iron Age radiocarbon study and says this paper presents its intercomparison subset. But the Assessment of Bias reports 90 individual measurements from 22 samples for one comparison, while the broader study involved over 400 analyses from about 20 sites. Our copy does not substantiate the claimed 68 dates, 30 samples, or 10 sites for this publication, or identify the study by the project name given. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingreasoning

    The difference between the results and those of Boaretto et al. (2005) is a minor difference of a mere ~50 yr and may not be fully resolvable by 14C.

    “This sounds like a minor difference of a mere ~50 yr compared to the results ofBoaretto et al. (2005), and may not be fully resolvable by14 C; yet, these years are at the very heart of the chronological and historical debate concerning the 10th century BCE.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Boaretto et al. (2005) source

      The Introduction (page unavailable) says a roughly 60-year chronological difference is close to the radiocarbon precision limit, supporting the caution about resolving a difference of about 50 years. The paper reports its own Iron I/II transition dates as 910–875 BCE, but it does not establish the citing authors’ approximately 50-year difference from their results. In “Can 14 C Dating Resolve the Levantine Iron Age Dilemma?” (page unavailable), it says radiocarbon can resolve the dilemma if enough measurements are available. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Study design and conductown data

    The Bayesian models presented in the second part of the paper are based on the same data and use the same method as Boaretto et al. (2005), yet using additional dates (mainly from Sharon et al. 2007) and removing outliers to the extent needed to achieve good internal consistency.

    “Our Bayesian models presented in the second part of the paper are based on the same data and use the same method asBoaretto et al. (2005), yet using additional dates (mainly fromSharon et al. 2007) and removing outliers to the extent needed to achieve good internal consistency.”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites Boaretto et al. (2005) source
    • cites Sharon et al. 2007 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    In Sharon et al. (2007:27), sample MQ1-2 was marked as 'Stratum Vb (or IV??).'

    “inSharon et al. (2007:27) it was marked as”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Sharon et al. 2007:27 source

      The sample tables in our copy (Tables 2–4; page numbers unavailable) list Tel Miqne samples by numeric identifiers, including samples assigned to Vb. Searches for “MQ1-2” and “Vb (or IV??)” found no passage linking that sample to the quoted designation, so the specific attribution cannot be verified. (LLM, conf=0.72)

      Experimental — accuracy not yet measured

  • Classified as probably not a substantial claim.

    Premisecitation

    Boaretto et al. identify QS1 (T-18161 = 3932a, aa) from Tell Qasile X as an outlier.

    “They identify QS1 (T-18161 = 3932a, aa) from Tell Qasile X as an outlier.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Boaretto et al. 2005 source

      Table 4, in “Can ¹⁴C Dating Resolve the Levantine Iron Age Dilemma?” (page unavailable), explicitly lists T18161aa among four outliers in the combined Rehovot–Tucson analysis. The passages from our copy do not establish that this measurement is QS1 (3932a, aa) from Tell Qasile X, so that identification remains unsupported. (LLM, conf=0.82)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    In Tel Rehov, Stratum VI represents the earlier phase of the period and strata V-IV the later phase.

    “The division of the Iron Age IIA into 2 sub-periods-early and late-as suggested by Herzog and Avitz-Singer (2006; see also Sharon et al. 2007:10-13 and our comment above in relation to Tel Dor) is now substantiated in the excavations of Tel Rehov, where Stratum VI represents the earlier phase of the period and strata V-IV the later phase.”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Herzog and Avitz-Singer 2006 source
    • cites Sharon et al. 2007 source

      In ‘Relative Periodization of the Iron Age I–IIA in Israel’ (page unavailable), the paper tentatively places Tel Rehov stratum VI in the transitional horizon preceding classical Iron IIA. ‘Previous Research’ also calls VI the lowest Iron IIA stratum. Our copy does not identify strata V–IV as the later phase, so it supports only the earlier-phase part of the claim. (LLM, conf=0.83)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Interpretationcitation

    The calibrated date of sample 4286 (1190-1030 BCE at 68% probability) may fit both the high and low chronologies and thus does not help resolve the debate over the earliest Philistine settlement.

    “The calibrated date of this sample(Sharon et al. 2007:24, 41) between 1190-1030 at 68% probability may fit both the excavator's and Mazar's date 1180-1140/30 BCE (during the time of the Egyptian 20th Dynasty) as well as Ussishkin and Finkelstein's lower dates (1140/30-1200 BCE) and thus does not help to resolve this debate.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Sharon et al. 2007:24, 41 source

  • May combine two claims.

    Interpretationreasoning

    The average 2741 ± 18 for sample 3782 given by Sharon et al. (2007:43) must be a mistake, so the average 2645 ± 31 (Sharon et al. 2007:35) is used instead.

    “We use here the average 2645 ± 31(Sharon et al. (2007:35) rather than 2741 ± 18(Sharon et al. 2007:43), which must be a mistake.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Sharon et al. (2007) source

      The Hazor Xa (3782) entry in the paper’s dates table (page unavailable in our copy) gives 2741 ± 18 BP, as the citing sentence says. Our copy does not show the attributed 2645 ± 31 average for that sample; the same entry instead lists 2653 ± 60 as another estimate. The judgment that 2741 ± 18 is a mistake is the citing authors’ own conclusion, not a claim attributed to Sharon et al. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingown data

    The destruction dates of rich late Iron I contexts such as Tell Qasile X, Megiddo VIA, Yoqne'am XVII, and Tel Hadar IV, all in a range between the last decades of the 11th and the mid-10th centuries BCE, support either Model C2 or even an earlier date in the 10th century for the end of Iron I.

    “The destruction dates of rich late Iron I contexts such as Tell Qasile X, Megiddo VIA, Yoqne'am XVII, and Tel Hadar IV, all in a range between the last decades of the 11th and the mid-10th centuries BCE, support either Model C2 or even an earlier date in the 10th century for the end of Iron I.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The transition date from Iron I to Iron II at Megiddo occurred in the range 987-978 (10%), 975-943 (58.2%), or 1000-928 BCE (95.4%).

    “They suggest that the transition date from Iron I to Iron II at Megiddo occurred in the range 987-978 (10%), 975-943 (58.2%), or 1000-928 BCE (95.4%).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The Iron I/Iron II transition occurs between the K4 (=VIA) destruction and the level K3/H6 (=VB) where there are no dates.

  • May combine two claims.

    Findingown data

    Model C2 indicates a 68% probability date range of 960-940 BCE for the transition from Iron I to Iron II, while Model C3 would lower this date to 948-919 BCE.

    “This led to Model C2, which indicates a 68% probability date range of 960-940 BCE for the transition from Iron I to Iron II, while Model C3 would lower this date to 948-919 BCE.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The models assume a synchronous change from Iron I to Iron II across the region.

  • May combine two claims.

    Findingown data

    HZ3 and HZ8 have a poor agreement index, being considerably earlier than the rest of the now much larger group of dates for Iron I.

    “HZ3 and HZ8 as they have a poor agreement index, being considerably earlier than the rest of the now much larger group of dates for Iron I”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the case of Tell Qasile X, the agreement between the results is very poor and only if the 2 earliest and 2 latest dates (QS2, 3, 6, and 11) are rejected do the results pass the χ2 test.

    “In the case of Tell Qasile X, the agreement between the results is very poor and only if we reject the 2 earliest and the 2 latest dates (QS2, 3, 6, and 11) do the results pass the χ 2 test.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The dates from Megiddo K-4 combined together fail the χ2 test, indicating they do not suggest a single age for all samples from the context.

    “Dates from Megiddo K-4 combined together -note the failed χ 2 test indicating that these results do not suggest a single age for all of the samples from the context.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The dates from Yoqneam XVII combined together fail a χ2 test, though not as badly as Megiddo K-4, indicating they do not reflect a single event.

    “Dates from Yoqneam XVII combined together; these results also fail a χ 2 test, though not quite as badly as those for Megiddo K-4, indicating that they do not reflect a single event.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingreasoning

    Models B and C contain much more data and are less dependent on the removal of outliers.

    “Models B and C contain much more data in them and are less dependent on the removal of outliers”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Model C2 gives a date for the transition of 961-942 BCE (68.2%) or 966-929 BCE (95.4%).

    “Model C2 would seem to us the most realistic, while retaining the maximum information with a date for the transition of 961-942 BCE (68.2%) or 966-929 BCE (95.4%).”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The dates from Tables 1 and 2a and MG7 and MG8 might reasonably be supposed to come from any time in Iron I or the very end of the LBA.
    • The authors state this as an assumption: The third independent phase (short-lived material from the end of Iron I) is much shorter than the LBA/Iron I phase.

  • May combine two claims.

    Findingown data

    K9 and QS6 have low agreement indices (0.1% and 25.3%) and are much later than others in the Iron II group.

    “K9 (agreement 0.1%) and QS6 (agreement 25.3%), which are much later than others in the Iron II group.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Findingnone given

    Elsewhere at Tel Rehov, the last phase of Iron I includes substantial buildings.

    “Elsewhere, the last phase of Iron I includes substantial buildings.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    The Yoqne'am XVII dates fit the High Chronology and Middle Chronology, but negate the Low Chronology.

    “These dates fit the HC and MCC, but negate the LC.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The five destruction layers (Megiddo VIA, Yoqne'am XVII, Tell Keisan 9a, Tell Qasile X, and Tel Hadar IV) are reliable, primary contexts and are agreed by all archaeologists of the Levant as representing the end of the Iron Age I.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given

    Tel Hadar IV has direct implications on Aegean chronology.

    “Hadar IV (the latter has direct implications on Aegean chronology).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    All Tell Keisan Stratum 9a samples except K9 fit the High Chronology or Middle Chronology; none fit the Low Chronology.

    “All except K9 fit the HC or MCC; none fit the LC.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The five destruction layers (Megiddo VIA, Yoqne'am XVII, Tell Keisan 9a, Tell Qasile X, and Tel Hadar IV) are reliable, primary contexts and are agreed by all archaeologists of the Levant as representing the end of the Iron Age I.

  • May combine two claims.

    Interpretationnone given

    The Iron Age I dates mentioned in Table 1 either fit the High Chronology or are too early (in the case of Hazor).

    “The Iron Age I dates mentioned above either fit the HC or are too early (in the case of Hazor).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    Figure 3C2 indicates a date from Tell Qasile that may fit only the High Chronology and is somewhat too high even for the Middle Chronology.

    “Figure 3C2 indicates a date that may fit only the HC and is somewhat too high even for the MCC.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The five destruction layers (Megiddo VIA, Yoqne'am XVII, Tell Keisan 9a, Tell Qasile X, and Tel Hadar IV) are reliable, primary contexts and are agreed by all archaeologists of the Levant as representing the end of the Iron Age I.

  • May combine two claims.

    Interpretationreasoning

    Such an earlier date may fit the transition date from Iron I to Iron II as calculated at Tel Rehov (992-961 BCE range, 68% probability) and the approximate date ~980 BCE for this transition as suggested by Mazar (2005).

    “Such an earlier date may fit the transition date from Iron I to Iron II as calculated at Tel Rehov (992-961 BCE range, 68% probability; see Mazar et al. 2005;Bru-Figure 4 Comparison between the average of the destruction layers and Models C2 and C3 (shown in Figure 1).ins et al. 2005a:287, Figure 15.4) and the approximate date ~980 BCE for this transition as suggested by Mazar (2005).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Mazar et al. 2005

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Mazar 2005

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation

    The Iron Age IIA can be divided into early and late phases, a phenomenon confirmed at Tel Rehov.

    “as discussed above, the Iron Age IIA can be divided into early and late phases (Herzog and Avitz-Singer 2006), a phenomenon confirmed at Tel Rehov.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Herzog and Avitz-Singer 2006 source

  • May combine two claims.

    Interpretationreasoning

    The long duration of the period is emphasized by the dates of the last Iron I contexts and by the existence of earlier Iron IIA levels, not represented in the 14 C record.

    “Yet, the long duration of the period is emphasized by the dates of the last Iron I contexts as described in the previous section and by the existence of earlier Iron IIA levels, not represented in the 14 C record.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisenone given

    MG7 and MG8 are defined as 'pre-destruction.'

    “the 2 dates MG7 and MG8, which are defined as”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationnone given

    HB1 and HB2 could fit an Iron I date.

    “though they could fit an Iron I date.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    Both QS1 and QS2 were removed from some models in order to prevent bias.

    “In order to prevent bias, we have removed both QS1 and QS2 from some models (see below).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    The Bayesian models might not be sensitive enough to accommodate such a situation.

    “It might be that the Bayesian models are not sensitive enough to accommodate such a situation.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    If the transition from Iron I/II is a process that takes a generation, then this process was almost certainly underway early in the 10th century BCE, even if it might not have been complete at all sites until the latter half of the 10th century BCE.

    “More critically, however, if the transition from Iron I/II is a process that takes a generation, rather than an abrupt and synchronous event, then this process was almost certainly underway early in the 10th century BCE, even if it might not have been complete at all sites until the latter half of the 10th century BCE.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Sample 3786 from Hazor is an outlier or a mistake made during excavation.

    “we suggest that this sample is an outlier; there might have been a problem such as contamination by a later intrusion, a mistake in the stratigraphic attribution, a mistake in registration, etc.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S4 (based). May combine two claims.

    Interpretationnone given

    The dates of the timbers from Megiddo K-4 are questionable and should not be taken into account in an evaluation of the date of Stratum VIA.

    “while the dates of the timbers are questionable and should not be taken into account in an evaluation of the date of Stratum VIA”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    The charcoal dates in Table 2a are from Iron I, but possibly earlier than the destruction levels.

    “We also class the charcoal dates in Table 2a as being from Iron I, but possibly earlier than the destruction levels as they are dates of charcoal from beams.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Most of the dates from the contexts described in the end of Iron Age I section support the High Chronology or Middle Chronology, and only a few may support the Low Chronology in addition to the High/Middle Chronology.

    “As we have seen, most of the dates support the HC or MCC, and only a few may support the LC in addition to the HC/MCC.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The five destruction layers (Megiddo VIA, Yoqne'am XVII, Tell Keisan 9a, Tell Qasile X, and Tel Hadar IV) are reliable, primary contexts and are agreed by all archaeologists of the Levant as representing the end of the Iron Age I.

  • May combine two claims.

    Interpretationown data

    The single short-lived sample 3796 from Tell Keisan Stratum 9a is calibrated to 1060-930 BCE and could fit the High Chronology and Middle Chronology, but not the Low Chronology.

    “The single short-lived sample ( 3796) is calibrated to 1060-930 BCE and thus could fit the HC and MCC, but not the LC.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The five destruction layers (Megiddo VIA, Yoqne'am XVII, Tell Keisan 9a, Tell Qasile X, and Tel Hadar IV) are reliable, primary contexts and are agreed by all archaeologists of the Levant as representing the end of the Iron Age I.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given

    We are confident that the context of BD1-2 from Bethsaida is Iron IIA.

    “though we are confident that their context is Iron IIA (see above and below).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    The Tel Hadar Stratum IV dates can fit the High Chronology or Middle Chronology and negate the Low Chronology.

    “which can fit the HC or MCC and negate the LC.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The five destruction layers (Megiddo VIA, Yoqne'am XVII, Tell Keisan 9a, Tell Qasile X, and Tel Hadar IV) are reliable, primary contexts and are agreed by all archaeologists of the Levant as representing the end of the Iron Age I.

  • Classified as probably not a substantial claim. May combine two claims.

    Interpretationnone given

    This study emphasizes the sensitivity of Bayesian models to outliers, and for reducing or adding dates from the models.

    “This study emphasizes the sensitivity of Bayesian models to outliers, and for reducing or adding dates from the models.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    Sample 3784 from Hazor is too low for all chronological systems and is suspected to be an outlier like 3786.

    “Sample 3784 (calibrated date 815-790 BCE) is too low for all the chronological systems and is suspected to be an outlier like 3786.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisenone given

    HZ4, HZ5, HZ9, and HZ10 are samples of old wood in secondary use.

    “HZ4, HZ5, HZ9, and HZ10 explained above as samples of old wood in secondary use”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    The date of the destruction of Megiddo Stratum VIA fits the High Chronology and Middle Chronology, but not the Low Chronology.

    “a date that fits the HC and MCC, but not the LC.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The five destruction layers (Megiddo VIA, Yoqne'am XVII, Tell Keisan 9a, Tell Qasile X, and Tel Hadar IV) are reliable, primary contexts and are agreed by all archaeologists of the Levant as representing the end of the Iron Age I.

  • May combine two claims.

    Premisereasoning

    Both samples are of charcoal and thus could be of old wood.

    “Both samples are of charcoal and thus could be of old wood.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Bruins and van der Plicht 2005

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Interpretationcitation

    The Bethsaida sample 4281 date may fit the High Chronology or Middle Chronology, but negates the Low Chronology.

    “may fit the HC or MCC, but negates the LC.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisenone given

    MG14 and BD1-BD2 are dates of olive stones and seeds, respectively.

    “both are dates of olive stones and seeds, respectively”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    The historical date of the destruction at Tell es-Safi attributed to Hazael (about 830-810 BCE) fits the lowest edge of the 68% probability range of SF2 and the upper edge of SF1.

    “This historical date fits the lowest edge of the 68% probability range of SF2 and the upper edge of SF1.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Maeir and Ehrlich 2001

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim. May combine two claims.

    Premisenone given

    HB1 and HB2 are of charcoal.

    “HB1 and HB2 since they are of charcoal”
    Quote in the paperSubstantial claim
50 of 50

  • The paper relies on this without support.

    The five destruction layers (Megiddo VIA, Yoqne'am XVII, Tell Keisan 9a, Tell Qasile X, and Tel Hadar IV) are reliable, primary contexts and are agreed by all archaeologists of the Levant as representing the end of the Iron Age I.

    “5 destruction layers that are reliable, primary contexts and agreed by all archaeologists of the Levant as representing the end of the Iron Age I.”

  • The paper relies on this without support.

    The most acceptable archaeological date for the Negev Highland sites is the 10th century BCE.

    “The most acceptable archaeological date for these sites is the 10th century BCE.”

  • The authors state this as an assumption.

    Samples for 1 context are the same age.

    “We then combined all of the samples for 1 context since these are assumed to be the same age.”

  • The paper relies on this without support.

    Any charcoal or wood date might be significantly older than its context, even though it is not obviously an outlier.

    “Because, in principle, any charcoal or wood date might be significantly older than its context, even though it is not obviously an outlier”

  • The authors state this as an assumption.

    The dates from Tables 1 and 2a and MG7 and MG8 might reasonably be supposed to come from any time in Iron I or the very end of the LBA.

    “These dates might reasonably be supposed to come from any time in Iron I or the very end of the LBA.”

  • The authors state this as an assumption.

    The third independent phase (short-lived material from the end of Iron I) is much shorter than the LBA/Iron I phase.

    “We expect this phase to be much shorter than the LBA/Iron I phase.”

  • The authors state this as an assumption.

    The models assume a synchronous change from Iron I to Iron II across the region.

    “The underlying assumption of the models is that there is a synchronous change from Iron I to Iron II across the region.”
    “All of the generalized models presented here and in Sharon et al. (2007) assume a transition from Iron I/II, which is synchronous across the region.”

  • The authors state this as an assumption.

    The samples from specific destruction layers (Megiddo VIA, Yoqne'am XVII, Tell Qasile X, and Tel Hadar IV) reflect a single event.

    “we tested the assumption that the samples from specific destruction layers (Megiddo VIA, Yoqne'am XVII, Tell Qasile X, and Tel Hadar IV) reflect a single event.”

  • The paper relies on this without support.

    The Iron I/Iron II transition occurs between the K4 (=VIA) destruction and the level K3/H6 (=VB) where there are no dates.

    “The Iron I/Iron II transition occurs between the K4 (=VIA) destruction and the level K3/H6 (=VB) where there are no dates.”

  • The paper relies on this without support.

    The phase K3/H6 (=VB) represents some time in the development of the site.

    “because it clearly represents some time in the development of the site”

  • The paper relies on this without support.

    Stratum VIIA at Megiddo does not date to the 10th/9th centuries BCE.

    “The calibrated date of 10/9th centuries BCE for this sample is an unreasonable date for Stratum VIIA at Megiddo.”

  • The paper relies on this without support.

    The expected date for sample 4501.1-3 is known and is not as low as the measured date.

    “We omit sample 4501.1-3(Sharonetal. 2007:24)as it is much too low.”

  • The paper relies on this without support.

    Short-lived samples from Megiddo K-4 are more reliable for dating than timber samples from the same level.

    “which contradict the dates achieved now by the Iron Age Dating Project on short-lived samples from the same level.”

  • The paper relies on this without support.

    The lowest date in the sample 3932 set is an outlier.

    “It seems more likely that the lowest one should be eliminated.”

  • The authors state this as an assumption.

    The 5 Groningen dates cited in Boaretto et al. (2005: Table 3) were also included in Boaretto et al.'s own model.

    “only the 5 dates measured at Groningen and cited inBoaretto et al. (2005: Table 3) were included in our model, as we assume that they were included in their model as well.”
15 of 15

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Radiocarbon dates for Bronze Age sites in the Levant often conflict with chronological information derived from ceramics and inscriptions, and some samples can be from burned wooden beams cut from trees harvested over 100 years before their destruction.

    “it is a known issue that the radiocarbon dates for Bronze Age sites in the Levant often conflict with the chronological information derived from ceramics and inscriptions and that some samples can be from burned wooden beams cut from trees that were harvested over 100 years before their destruction”

    From The Bronze Age Destruction of Jericho, Archaeology, and the Book of Joshua (2023) · cites this paper as Mazar and Ramsey 2008, pp. 159-80

    Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

    Experimental — accuracy not yet measured

  • Rigorous programs of dating the Iron Age phases have been underway in the last decade.

    “Rigorous programs of dating the Iron Age phases have been underway in the last decade (e.g., Boaretto et al. 2005; Sharon et al. 2007; Mazar and Bronk Ramsey 2008; Finkelstein and Piasetzky 2010a).”

    From Absolute Chronology of Megiddo, Israel, in the Late Bronze and Iron Ages: High-Resolution Radiocarbon Dating (2014) · cites this paper as Mazar and Bronk Ramsey 2008

  • The difference of opinions regarding the debated Iron I/II transition is limited to a few decades.

    “This is essential when dealing with cases such as the debated Iron I/II transition (Mazar and Bronk Ramsey 2008; Finkelstein and Piasetzky 2010a), in which the difference of opinions is limited to a few decades.”

    From Absolute Chronology of Megiddo, Israel, in the Late Bronze and Iron Ages: High-Resolution Radiocarbon Dating (2014) · cites this paper as Mazar and Bronk Ramsey 2008

  • The absolute chronology of the Late Bronze and Iron Age horizons is debated, including the beginning of the LB, the LB I-II, the details of the LB III/Iron I transition and the Iron I/II transition.

    “Yet, the absolute chronology of these horizons is debated, for instance regarding the beginning of the LB (see below), the LB I-II, the details of the LB III/Iron I transition 6 and the Iron I/II transition (e.g., Sharon et al. 2007; Mazar and Bronk Ramsey 2008; Finkelstein and Piasetzky 2010b).”

    From Absolute Chronology of Megiddo, Israel, in the Late Bronze and Iron Ages: High-Resolution Radiocarbon Dating (2014) · cites this paper as Mazar and Bronk Ramsey 2008

  • The disputed transition from the late Iron I to the Iron IIA falls in the range 985-935 BCE, meaning that it cannot be decided according to the Megiddo data.

    “The disputed transition from the late Iron I to the Iron IIA (Sharon et al. 2007; Mazar and Bronk Ramsey 2008; Finkelstein and Piasetzky 2010b) falls in the range 985-935 (Fig. 9), meaning that it cannot be decided according to the Megiddo data.”

    From Absolute Chronology of Megiddo, Israel, in the Late Bronze and Iron Ages: High-Resolution Radiocarbon Dating (2014) · cites this paper as Mazar and Bronk Ramsey 2008

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Abstract omits part of the reported Megiddo transition range

    Major

    The abstract presents 967-943 BCE as the Model D transition range. The Model D discussion instead reports two ranges totaling 68.2% probability, including 987-978 BCE, which lies outside the abstract's range.

  • Conflicting 68.2% date ranges for Yoqne'am XVII

    Major

    The abstract and site discussion give 1045-997 BCE for Yoqne'am XVII, but the later summary of destruction-layer results gives 1013-944 BCE at the same probability level.

  • Hazor 3786 is not later than every Stratum Xa date

    Minor

    The paper says sample 3786's calibrated range is later than those of Stratum Xa, but it gives Stratum Xa sample 3784 an even later range of 815-790 BCE.

  • Two versus three Rosh-Zayit seed samples

    Minor

    The same discussion first says that two seed samples came from Stratum IIa, then refers to three samples numbered 3797-3799 from that phase.

4 of 4

Cites

Flagged first, then by citations

3 of the 30 references couldn’t be matched to a known paper.

Research works printed but not linked: 19

  1. Carmi, 2000, Megiddo III: The 1992–1996 Seasons.
  2. Mazar, 1980, Excavations at Tell Qasile, Part One. The Philistine Sanctuary: Architecture and Cult Objects (Qedem Reports No. 12), 153
  3. Bruins, 2005a, The Bible and Radiocarbon Dating: Archaeology, Text and Science.
  4. Ben-Tor, 2005, Yoqne'am II, the Iron Age and the Persian Period (Qedem Reports No. 6), 421
  5. Bruins, 2005, The Bible and Radiocarbon Dating: Archaeology, Text and Science.
  6. Bruins, 2007, The Synchronisation of Civilizations in the Eastern Mediterranean in the Second Millennium B.C. III., 79
  7. Sharon, 2005, The Bible and Radiocarbon Dating: Archaeology, Text and Science., 65
  8. Cahill, 2006, “I Will Speak the Riddles of Ancient Times,” Archaeological and Historical Essays in Honor of Amihai Mazar on the Occasion of His Sixtieth Birthday.
  9. Finkelstein, 1993, Shiloh, The Archaeology of a Biblical Site.
  10. Finkelstein, 2006, Megiddo IV: The 1998–2002 Seasons., 860
  11. Mazar, 2005, The Bible and Radiocarbon Dating: Archaeology, Text and Science., 195
  12. Maeir, 2001, Excavating Philistine Gath: Have we found Goliath's hometown?, Biblical Archaeology Review, 27, 22
  13. Finkelstein, 2002, The campaign of Shoshenq I to Palestine: a guide to the 10th century BCE polity, Zeitschrift des Deutschen Palästina - Vereins, 118
  14. Levy, 2005, The Bible and Radiocarbon Dating: Archaeology, Text and Science., 448
  15. van der Plicht, 2005, The Bible and Radiocarbon Dating: Archaeology, Text and Science.
  16. Finkelstein, 2005, The Bible and Radiocarbon Dating: Archaeology, Text and Science., 31
  17. Bruins, 2005b, The Bible and Radiocarbon Dating: Archaeology, Text and Science.
  18. Mazar, 2005, The Bible and Radiocarbon Dating: Archaeology, Text and Science., 15
  19. Finkelstein, 2005, The Bible and Radiocarbon Dating: Archaeology, Text and Science., 294

Cited by

Sorted by citations

Related

Authors
Mazar, A., & Bronk Ramsey, C.
Affiliations
Hebrew University of Jerusalem (IL)
Corresponding author
Mazar, Amihai
ORCID
Mazar, Amihai
Venue
Radiocarbon
Publisher
Cambridge University Press (CUP)
Published
2008
Volume
50
Issue
2
Pages
159-180
ISSN
  • 0033-8222 (ISSN-L)
  • 1945-5755
DOI
10.1017/s0033822200033506
OpenAlex
W2162728678
Cited by
71
References
30
Citations by year
2026: 1 · 2025: 1 · 2024: 1 · 2023: 2 · 2022: 2 · 2021: 1 · 2020: 14 · 2019: 3 · 2018: 5 · 2017: 2 · 2016: 1 · 2015: 2 · 2014: 7 · 2013: 3 · 2012: 6
FWCI
5.10
Type
journal article
Language
en
Open access
Licence
www.cambridge.org/core/terms (unspecified, 3121 d)
Locations
  • Radiocarbon (publishedVersion)
  • UA Campus Repository (The University of Arizona) (acceptedVersion)
Topics
  • Physical Sciences › Earth and Planetary Sciences › Paleontology › Archaeology and ancient environmental studies
  • Social Sciences › Arts and Humanities › Religious studies › Biblical Studies and Interpretation
  • Social Sciences › Arts and Humanities › Archeology › Ancient and Medieval Archaeology Studies
Keywords
Iron Age chronology, radiocarbon dating, Megiddo, Tel Rehov, Tell Qasile, 10th century BCE, interlaboratory comparison

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-29.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-29.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review