Dating Mediterranean shipwrecks: the Mazotos ship, radiocarbon dating and the need for independent chronological anchors

Abstract

Studies of ancient Mediterranean trade and economy have made increasing use of sophisticated modelling and network analyses of shipwreck evidence. The dating of most of these wrecks, however, is based solely on assessments of associated ceramic material, especially transport amphorae. The resulting dates are approximate at best, and, as the example of the recently investigated Mazotos ship highlights, sometimes incorrect. Here, the authors describe a widely applicable independent approach based on the integration of tree-ring analysis and radiocarbon dating. Interrogating the subjective assumptions and stepwise logic transfers involved in ceramic-based dating, the authors demonstrate how to produce a more robust and better-defined basis for the analysis of the ancient Mediterranean shipwreck record.

A revised radiocarbon calibration curve 350–250 BCE impacts high-precision dating of the Kyrenia Ship

“…It is evident that the new data offer a revision to the previous IntCal20 dataset. The substantial change in slope at the end of the 5 th century BCE is still evident (as evident also from the dates reported recently on the Mazotos ship: [ 15 ]), but the AMS 14 C values in general are a little older (in 14 C terms) than the IntCal20 curve (tending to shift calibrated calendar ages to slightly more recent ages). The new values also reduce/modify the previous strong dip in 14 C values 350–310 BCE and raise the calibration curve in the period around and following 300 BCE.…”

“…In particular, the new 14 C evidence might suggest a wrecking date sometime in the 280s BCE. Additionally, the new 14 C calibration data fine tune, but overall confirm, the earlier 4 th century BCE date proposed for another Greek commercial ship, the Mazotos ship found off southern Cyprus [ 15 ].…”

“…The revised 14 C calibration data reported in this paper and used for the dating of the Kyrenia Ship are relevant also to other Classical-Hellenistic era ships, including the recently reported dating of the Mazotos ship found off the southern coast of Cyprus [ 15 ]. A combination of tree-ring-sequenced 14 C wiggle-matching and 14 C dates on short-lived materials from the Mazotos ship using IntCal20 placed the last extant tree-ring (a TPQ for construction of the ship) at 442–399 BCE at 95.4% hpd (422–404 BCE at 68.3% hpd) and the date for the ship’s Last Voyage (LV) 393–374 BCE at 95.4% hpd (390–382 BCE at 68.3% hpd).…”

“…Given the considerations discussed above, we can now combine the information available into an overall dating model for the Kyrenia Ship (e.g. [ 15 ]). In essence we have a sequence in time:…”

“…We may reasonably assume that most of the short-lived samples on board as cargo or crew provisions should date from immediately, or shortly, before the last voyage of the ship—although a few of the samples might be older, or even residual material. We can thus model this Phase (2) of dates between a Tau_Boundary and a Boundary in OxCal to create such an exponential distribution with the end of Phase 2 Boundary best describing the wrecking date immediately after the Phase [ 15 ]. This end Boundary can be labelled as ‘LV’ for last voyage.…”