ISSN 1357-4442Editor: Simon Denison

Issue no 55, October 2000


Roman haymaker

George Lambrick recalls a scrap of metal that turned into a Roman scythe

My favourite find? That must be a late Roman scythe from Farmoor in Oxfordshire. The site lies by the Thames and was dug out in the 1970s to make one of Thames Water's reservoirs where my daughters now go sailing.

The project was important to me because it was where I first became a professional archaeologist in the summer after I left university. I directed the rescue work as a joint project between the Oxford University Archaeological Society (of which I had just been President) and the Oxford Unit, which then offered me a job.

Farmoor was originally thought to be just a minor cropmark site - perhaps a single roundhouse - but during excavation of that site we wandered about the rest of the construction area, doing salvage work amongst the scrapers and bulldozers, delighted when they got bogged down in the mud as we went on working.

It was Mark Robinson, our environmental archaeologist, who spotted the first of a series of Iron Age farmstead sites on the floodplain. They became exposed as layers of alluvium were removed by heavy machinery, and as a result we only found them when they were all but destroyed, their ditches squidged into distorted shapes by the crushing weight of the plant. Occupation layers were rolled up like swiss rolls where the bulldozers had left some spoil.

There was just enough to tell that we were dealing with Iron Age houses with little enclosed yards, but the really telling evidence came from the peat surviving in the bottoms of the otherwise mangled ditches and gullies. Mark's meticulous identification of the preserved seeds, insect remains and snails showed that the farmsteads had been set in extensive grazed grassland and had been subject to flooding, implying they were only used in the summer, in a highly specialised pastoral farming system.

The later, very lowly Roman settlement nearby produced fascinating evidence of horticultural crops and even ornamental box hedges. But the other real interest was the evidence of haymaking from dumps of discarded hay in waterlogged pits.

It seems appropriate that in this context of intensive grassland land management we also found a magnificent 4th century scythe.

I was excavating the end of a Roman trackway where it reached the edge of the gravel terrace and was covered by a layer of peat. Sticking out of the side of the trench at the base of the peat was a small flat piece of iron about 5cm long. A small tug did not free it - in fact it felt rather springy.

Digging into the trench wall is hardly the done thing, but I'm glad I did. It was like pulling handkerchiefs out of a magician's hat - the scrap of metal just went on and on. Five centimetres turned into ten; ten to twenty; twenty to half a metre; the half metre into a whole one, and that into almost two metres, revealing the almost complete scythe. Late Roman scythes are fearsome implements - far longer than the earlier Roman ones or conventional scythes of today, with a large crook at the haft end to act as a counter weight. Anyone who has used a normal scythe would wonder how these massive things could be wielded. This one had already been mended once with riveted plates, and it is easy to imagine how it might have been badly damaged if it were accidentally stuck into the ground.

Was it discarded when it broke again, and then forgotten? It was high quality steel, and would have been valuable as scrap even if it wasn't worth mending again, so is unlikely to have been deliberately left behind. Whatever the reason, I'm glad it did get left - much more interesting than if it had been carefully recycled as scrap, and a real contribution to making us realise the long history of specialised grassland management in lowland river valleys.

George Lambrick is the Director of the Council for British Archaeology

Return to the British Archaeology homepage

Return to the CBA homepage

© Council for British Archaeology and author, 2000