Chris Salter wrote:

> One of the articles in volume 34(2) Mack et al was the source for an article in
> the Christmas edition of the New Scientist (No 2270/71 p7). This relates to the
> discovery of small steel ingots at Hamwic, Saxon Southampton. This ingot had a
> relatively homogeneous carbon content 1-1.5%, low slag inclusion content, a
feature
> also seen in steel artefacts from Saxon artefacts of Southampton and York.
> They go on to suggest that a two stage process was being used in which cast iron
> was deliberately produced and then decarburized from the liquid state in a
> second process.
>
> I am not totally convinced by this, as the same process can occur naturally
> within the bloomery furnace, all that is required a bloom-master with the skill
> required to control the bloomery temperature, and then to separate out the
> different types of metal from the bloom - a technique used in the tatara
> furnaces of Japan and the products of the Juleff's wind powered furnaces in Sri
> Lanka also produced zoned product. Also, it is noticeable in the hundreds of
> object that I have examined that there is a strong inverse correlation between
> the carbon content and the slag inclusion density.
>
I am in complete agreement with Chris on this matter. I reviewed the
Mack et al paper (twice) for Nature, and commented both times that the
arguments for decarburization at Hamwick were just straws blowing in the
wind. The only arguments advanced in favor were that the steel was
absolutely homogeneous and free of slag. I accept that the steel may
have been molten, but this hardly justifies the assumption that it must
have been produced by decarburization of cast iron.
The artifacts in question were of fairly small cross-section (ca. 2.5 x
2.5 cm) and there are numerous instances of larger pieces of homogeneous
slag-free steel from known bloomeries - in addition to the two mentioned
by Chris, there are many examples from historic or recent African
bloomeries. The claim of Mack et al. cannot be entertained until there
is evidence of (1) cast iron and (2) the process by which it was
decarburized (fining, co-fusion, etc.). Since I made these comments in
the first review, and no such evidence was produced in the second
version of the paper, I presume that Hamwic has yet to yield any such
debris.

I am afraid that this paper is symptomatic of a widespread problem in
British archaeometallurgy, which is its insularity. From Gordon Childe
to the present, few British writers on matters metallurgical bother to
read anything on the topic outside Europe. (Paul Craddock and Chris
Salter are among the exceptions). From Childe's itinerant smiths to Budd
and Taylor's Fairie Queen to Mack et al., there are numerous examples of
faulty inferences that could have been avoided had the authors bothered
to read the world literature - and especially the historic and
ethographic materials from Africa and Asia. I can't go into this further
here, but if you want to see further intemperate musings on the problem
of inference in archaeometallurgy, see my chapter "Science, Speculation
and the Origins of Extractive Metallurgy" in D. Brothwell and A.M.
Pollard (eds.) Handbook of Archaeological Sciences (London: Wiley),
which will be available soon, if it isn't already.

Dave Killick