Technology

28 pages
51 views

Manning, S.W. 2006-2007. Why radiocarbon dating 1200 BCE is difficult: a sidelight on dating the end of the Late Bronze Age and the contrarian contribution. Scripta Mediterranea 27: 53-80.

of 28
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
Manning, S.W. 2006-2007. Why radiocarbon dating 1200 BCE is difficult: a sidelight on dating the end of the Late Bronze Age and the contrarian contribution. Scripta Mediterranea 27: 53-80.
Transcript
  SCRIPTA MEDITERRANEA  , Vol. XXVII, 2006, 00 Sr W. MWhy RadiocaRbon dating 1200 bcEiS difficult: a SidElight on datingthE End of thE latE bRonzE agE andthE contRaRian contRibution Archaeological work employing sophisticated radiocarbon dating (and sometimes other natural science approaches) has made several signicantadvances in the last few years in clarifying and rening, or sometimes complicating/enriching(!), aspects or problems of east Mediterraneanprehistoric chronology (e.g., Cichocki et al. 2004; Levy and Higham 2005;Cessford 2005; Manning et al. 2001a; 2006). Radiocarbon has become anessential element of modern prehistoric chronologies and our consequenthistorical syntheses. With appropriate samples and good methodology,radiocarbon dating has the direct potential to provide independentdates for archaeological contexts, separate from long-standing culturalassumptions, debated proto-historical information, and so on. At the same time, however, this work has been the target for much contrarian aackand discussion. Critics have sought to nd fault with, modify, or dismiss the radiocarbon evidence, analyses thereof, and resultant chronologies—in most (recent) cases with the aim of achieving lower dates than thoseindicated by either radiocarbon or conventional archaeological-historical synthesis, or (usually) both. Although this may at rst sight appear to bean unproductive dialectic with at least one side eectively ignoring the other in all but straw-man terms—and there is undoubted frustration onthe radiocarbon side as work is routinely misrepresented—nonetheless, this situation can in fact be healthy for the wider eld. The contrarian critique can (perhaps inadvertently) usefully lead to stronger and morerobust radiocarbon work, and its tighter integration with archaeological evidence. The outcome is that instead of undermining the radiocarbonwork they wish to aack or dismiss, the contrarians in fact strengthen it in a rather paradoxical and Nietzschean twist. In this paper I look at one example: the aempt to date east Mediterranean archaeological contexts of the end of the Late Bronze Agearound 1200 BCE—traditionally more or less the time of the collapse of the Hiite Empire, the end of the Late Cypriot IIC period on Cyprus, towards the end of the Late Helladic IIIB period in the Aegean, the beginning of the main aested period of the ‘Sea Peoples’ in the eastern Mediterranean, and the ensuing 12 th century BCE so-called ‘crisis years’ (cf. Yakar 2006;Manning et al. 2001b; Warren and Hankey 1989; Sandars 1978; Ward and Joukowsky 1992; Oren 2000; and various papers in this volume). This studyis prompted by the ne example of the contrarian approach to be found in a paper by Hagens (2006). I consider this topic in order to illustrate howimportant an understanding of the natural history of past radiocarbon   SCRIPTA MEDITERRANEA  , Vol. XXVII–XXVIII, 2006–2007, 53–80  Sturt W. Manning 54 variations is to a sophisticated radiocarbon dating programme, andhow sequence analysis in radiocarbon work (of temporally seriated archaeological data based on the excavated stratigraphy) oers a much more robust and powerful means to calendar age determinations thanthe simple calibration of a single radiocarbon age value (whether from asingle date or an average of dates). Indeed, in many instances, selectivecitation of single dates or small groups can easily misrepresent the overallsituation. In the case in point, Hagens achieves his purported criticism ofexisting work, and the suggestion of a lower chronology, by looking at sets of data in isolation, and not as part of sequences. This can be (and in this instance is) misleading. Radiocarbon analysis of archaeological sites is necessarily a holistic study. This paper employs as its example the impossibility of narrowly/successfully dating a context of 1200 BCE  by single-case (or single set) radiocarbon dating. Such a context can only  be successfully dated unambiguously and with precision via a sequence analysis. At the same time, the contrarian aack nicely forces clarication of the situation and so serves us well, since it makes the case it seeks to aack clearer and stronger in the long run. Rr cr  Psses Radiocarbon chronology, and its potential and limitations for a givencalendar time interval, largely depends on the shape of the radiocarbon calibration curve. The current internationally accepted radiocarbon calibration dataset for the Holocene is IntCal04, derived for this timeperiod from known age tree-rings mainly from Germany and Ireland(Reimer   et al.   2004). The previous standard curve was IntCal98 (Stuiver et al. 1998), and was based largely on a similar database of underlyingmeasurements, though some important additions of new data andimprovements exist, for example in the 8 th century BCE. The IntCal04curve is an estimate at ve-year resolution, employing a sophisticated random-walk model which smoothes the inherent noise in the raw calibration datasets on the basis of a moving ve decade window. IntCal98oered ten-year resolution and merely averaged the dates in that intervalto achieve a data point for the calibration curve. It is thus more ‘ragged’(or up and down) than the smoother IntCal04 curve. The two curves are compared for the period 1500–1000 BCE in Figure 1. While largelyvery similar, the slight smoothing of the prominent ups and downs—the ‘wiggles’—in IntCal98 can be observed in IntCal04: the inset shows thecurve data points with 1σ error bars in detail for a sub-period either side of 1200 BCE. The shape of the calibration curve determines dating probabilities for individual radiocarbon ages in any given period. Radiocarbon ageswhich intersect with a steep slope in the radiocarbon calibration curvecan thus yield single, relatively precise, calendar age ranges (for anexample, see Fig. 2). In contrast, radiocarbon ages which intersect withperiods with plateaux or multiple wiggles including similar radiocarbonages, yield either multiple possible calendar age ranges or very wide— non-precise—age ranges (for an example of each, see Figs. 3 and 4). I note here that all calibration and calibration analysis in this paper has been  Why Radiocarbon Dating 1200 BCE is Difcult 55 fig 1. Comparison of the IntCal04 radiocarbon calibration curve (Reimer et al. 2004) with the IntCal98 radiocarbon calibration curve (Stuiver et al. 1998). Inset: detail of the calibration curve data points for the period either side of 1200 BCE. fig. 2. Example calibration of a radiocarbon age (2650 ± 35 BP) which intersects with a steep slope (only) on the radiocarbon calibration curve, and so (with the radiocarbon timescale probability in eect condensed by the curve slope onto a narrow band on the calendarscale) yields a quite precise calendar age range: 831–796 BCE at 1σ and 895–786 BCE at 2σ.OxCal and IntCal04 with curve resolution at 5. The demarcated zones under each (overall)calibration probability distribution here and in the other gures in this paper show (upperone) the 1σ (68.2%) and (the lower one) the 2σ (95.4%) calibrated ranges.  Sturt W. Manning 56 fig. 3. Example calibration of a radiocarbon age (4700 ± 35 BP) which intersects withmultiple discreet areas of the radiocarbon calibration curve because of a series of ‘wiggles’, and so yields three largely equally possible calendar age ranges within a wide overall 260 calendar year range (taking the 2σ limits). OxCal and IntCal04 with curve resolution at 5. fig. 4. Example calibration of a radiocarbon age (4150 ± 35 BP) which intersects with a plateau region of the radiocarbon calibration curve, and so yields a large spread of possiblecalendar age ranges. OxCal and IntCal04 with curve resolution at 5.  Why Radiocarbon Dating 1200 BCE is Difcult 57 performed using the OxCal soware (Bronk Ramsey 1995; 2001;hp:// c14.arch.ox.ac.uk/oxcal/ ), employing version 3.10 as current in 2006. 1   tr  de 1200 bcE Conventionally, the close of the Late Cypriot IIC period, or the LateHelladic IIIB period, has been placed around 1200 BCE, give or take a fewdecades, and the general collapse of Late Bronze Age civilizations in the region has been placed shortly thereaer in the early 12 th century BCE. There is, of course, currently active debate on this point, with, on the one hand, some suggestions for earlier dates and for a more extended process with regard to Greece and Anatolia especially (e.g., Yakar 2006). On the other hand, scholars such as Hagens (2006) wish to argue for the opposite,and thus to reduce the date for the same transition from Late HelladicIIIB to IIIC, or Late Cypriot IIC to IIIA, down to around about 1125 BCE; in other words almost eight decades later. Thus, starting with theconventional view, and simplifying to the ‘textbook’ generalisation, a date of about 1200 BCE is a key watershed marker. Given this, and also giventhe recent proposals for change and/or recent criticism, it is therefore aninteresting question to ask whether we can really date a horizon at 1200 BCE based on radiocarbon evidence? And, in reverse, are aempts (e.g., Hagens 2006) to claim that the radiocarbon evidence support a much laterdate valid?If we consider the time range centred on a calendar date of 1200 BCE, we see that the calendar time range around it, so ca.1300–1100 BCE,given the shape and wiggles of the calibration curve, in eect acts like aplateau in the calibration curve (see Fig. 1, and inset). Thus the correct radiocarbon age for a sample dating about 1200 BCE, such as a radiocarbon measurement of 2960±35 BP, does include 1200 BCE in its calibrated range, but also oers a wide range of other possible dates: 1302–1051 BCE at 2 σ   condence (see Fig. 5). In fact, we can quickly see that no radiocarbon agedetermination (in isolation), even at ‘high precision’ levels, can closely resolve a calendar date of 1200 BCE (see Figs. 6–9). It is an impossibletask, if the dating is approached in isolation. And, in reverse, simulated radiocarbon ages for 1200 BCE give a wide range (Fig. 7, Table 1). Notethat each run of such a simulation produces a slightly dierent set of values (see next section below). Even a hypothetical major focused dating programme measuring 20 good modern (as of 2006) AMS samples froma specic ‘known’ 1200 BCE context (let us assume short-lived seeds all from a context dated exactly to 1200 BCE), which in turn enable us to calibrate a high-precision weighted average with just a ±7 radiocarbon years standard error, nonetheless cannot narrowly resolve 1200 BCE. Instead, such a dataset nds a relatively wide date range covering quite a bit of both the 13 th and 12 th centuries BCE (Figs. 8–9).  Since this paper was delivered and draed, OxCal 4.0 has been made avail - able. The new version has the advantage of making the Bayesian analyses much more fully transparent and numerically explicit.
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x