Focus on UK Metal Detecting: More on What's this all about?

In the post above, I try to explain with pictures and diagrams what the archaeological record looks like from the point of view of the surface evidence and how archaeologists go about studying and analysing that evidence. In order to keep that post short, I kept the examples down to a few basic ones. This post expands on that by illustrating some results of surface investigations involving the use of metal detectors.

The first is an old one, from the early days of archaeological collaboration with hobbyist metal detectorists, the survey of the site at Cottam B (Anglian and Anglo-Scandinavian Cottam: linking digital publication and archive). The first picture shows the distribution of surface finds in a field overlaid on a simplified plan of the cropmarks:

Cottam B (Internet archaeology)

 In the following plots we see the plots of scatters of individual types of metal objects (the non-metallic material was not collected in this project) and it can be seen that there are relations between them which may be analysed, and if we were to plot other material, slag and other manufacturing debris, pottery of different fabrics, burnt daub etc, then this gives us a basis for interpretations of zonation and maybe chronology of various events on the site. This is all evidence lost if the material is not plotted accurately. 

Cottam B (Internet archaeology)

The two following figures come from the webpage on systematic metal detector survey of Earthsound Archaeological Geophysics from Ireland. The first shows a raw plot of finds, they are zoned, with areas of dense concentrations and 'blank' zones. This pattern may be analysed and determinations made of whether this represents a difference due to differential landuse in the past, or whether this is an indication of the state of preservation of the site.

Detector survey results (Earthsound)

Survey results (Earthsound)

  

The second plot shows the scatter broken down by artefact type, it can be seen that the northern scatter differs from the southern, suggesting that there are archaeological rather than post-depositional reasons for the present distribution of material in the ploughsoil.

The next example is from the webpage of MAP Archaeological practice, and shows the results of a systematic  (gridded) metal detecting survey near Malton. The orange outlines are old evaluation trenches, the shaded area unsurveyed. Again the distribution of material across the field is differentiated, with two linear patterns visible. Here also the metal distribution is shown without reference to that of other material.

Detector survey results (Malton Archaeological Practice)

One of the uses of wide-ranging systematic survey and careful plotting and analysis is well-shown on this picture from the website (What is Battlefield Archaeology?) of the Battlefield Restoration & Archaeological Volunteer Organization (BRAVO) which shows the results of a metal detector survey of case shot at the Parsonage site on the June 1778 Battle of Monmouth battlefield  to show the position and firing range of a particular gun in an historical conflict. The archaeological evidence, the position of the shot identifiable as fired from a single gun here is analysed in conjunction with other evidence to provide information about an event or process in the past (as well as subsequent post-deposition effects and biases introduced by factors such as search method and modern landuse)

(from work by Battlefield Restoration & Archaeological Volunteer Organization - BRAVO)

Quite clearly, even though these objects are found in the ploughsoil, their horizontal position and relationship to other objects (and information) comprises their archaeological context. Removal of the item from the ground without carefully noting the position and precise associations destroys that context. 

The final, for now, example comes from a metal detector survey of a different type, Polish archaeologists searching the April 10th crash site of the Polish Presidential Tupolev by Smolensk airport. The report is restricted, so I am not going to show the actual metal detector survey results here. This plan produced by amateur investigators of the disaster (and there are many) will suffice for discussion.

Fragment which was not buried
by the impact (Smolensk April 10th 2010)

The problem is that many people in Poland believe that the aircraft crashed due to sabotage (I am not one of them). There is a huge number of conspiracy theories about the crash, many of which rely on the small size of the pieces the plane broke into, that some of them are burnt, or have other "odd features". Suddenly everybody has become an internet expert amateur plane crash investigator. My colleagues from the Institute went there to help gather the personal belongings (and smaller portions of the remains) of the people who died which were buried in the sandy soil when the plane hit the ground and shattered. One of the principle tools used were metal detectors. The team found a very large number of small fragments which had not been recovered in the initial pickup. They were handed over to the Russian investigators together with a full report. Here's what the site looked like:

Scatter of debris (Polityka edited from satellite photos)

The plane broke up as it crashed through a forest and bits of it went flying in all directions. I am not going to discuss the report  (and warn at once that no comments of any kind on the Smolensk air crash are going to be published here) but wanted to make one point. The distribution of the burnt fragments about which so much had been made in the right-wing press was shown by the archaeological investigations to have been quite specific. The interpretation of their distribution was that as the plane fragmented, some of the smaller fragments from the parts of the plane that hit the trees and ground first were sucked into the jet engines which were at the moment of impact still working on full throttle and thrown out behind them. The  context of those fragments comprises their horizontal distribution and associations and relationship with every other fragment in that scatter - and it is only from that the interpretation can be reached, a handful of loose burnt fragments from the scatter as a whole tells us nothing, and can lead (as it has) to completely unjustified conclusions.