First the link to this week’s complete list as HTML and as PDF.
Previous analyses of archaeological finds and experimental results by e.g. Craddock (1999) and Rovira (2003) have shown that copper smelting in a poorly reducing atmosphere – presumably chosen on purpose to exclude iron – also prevents the reduction and inclusion of tin, leaving it in the slag. As Radivojevic et al. show it seems that one and a half millenia before the first intentional tin bronzes in Mesopotamia European smelters already succeeded in making tin bronze from stannite Cu2FeSnS4. This achievement is all the more impressive for being accomplished from sulphidic ores requiring the oxidizing and reducing two-step process. The previously discussed (Cierny 2005) origin from mushistonite CuSn(OH)6 would have required reduction only.
Jones & Klin’s result seems valid and relevant and it may well yield a significant step toward understanding the genesis of the condition, but looking at the scatter in their figures 1d and 1e it most certainly will not be any help to the search for an early diagnosis, probably the main goal of the study.
Did I mention Cargo Cult last week? I should have held my tongue. The current issue of the respected archaeological journal Antiquity has several articles playing at science and what we get beautifully illustrates Snow’s two cultures.
The first clear indication in Rodriguez & Hastorf is their naming of tools and commercial trade names in place of methods and algorithms. Instead of giving relative errors they list meaningless absolute ones, but those seem to fall into the 30–50 % range. They obviously poured their data into the package and took its output as divine pronouncement, never once stopping to ask by how much they’d have to change their input to achieve correct results and whether such a measurement error is at all feasible. They fail to see that an ellipsis still is an idealized approximation while the circular calculation can be exact given the right way to estimate the average diameter – their method in figure 3a is clearly and obviously wrong. They fail to see one of their equations in table 1 is simply wrong, the elliptical does not become the circular one for a=b=2r, and in their caption they confuse radius and diameter. Their figure 2 shows the approximation of a vessel’s outer volume. It differs from the inner one by the easily determined volume of the vessel itself, mass divided by ceramic density, but that difference is nowhere alluded to by them. The one thing that has merit, though, is their conclusion: a method is inappropriate if employed wrongly by people who haven’t got a clue what it is they’re doing.
As always an automated computer program must only ever be employed by users who understand the algorithm and who are able to achieve the same result manually using only pen and paper.
Motuzaite-Matuzeviciute et al. ask an important question employing a sensible method. Unfortunately their results are meaningless. What they should have done is date other, Near Eastern seeds of uncontroversial and generally accepted age from the same context, at the same time, in the same laboratory, and employing the same method. The reason for this necessity is clearly given in Richard P. Feynman’s famous Caltech commencement address of 1974, in which he also stresses “the idea that we all hope you will have learnt in studying science in school—we never explicitly say what this is, but just hope that you catch on by all the examples of scientific investigation.”. Doing it that way may be fine for science students deeply immersed in their subject for years at university, but it leaves those in the humanities trying to adapt scientific method floundering. Science really has a duty to find a way to get this essence of its elementary foundation across. So far it has failed miserably.
Xella et al. may well be right about the probability of human sacrifice, but Smith et al. offer no proof, not even support for it. The younger group in their figure 4a covers one month, a twelfth of the range, while in 4b it’s three months, or a quarter. If anything 4b shows the older group in Carthage to be over represented. Late Victorian England and Wales seems no more than a badly chosen, inappropriate reference here.