Bronze Age Metalworking

This is the next in our series of posts on metal-working written by Dr. Kevin Leahy, PAS National Finds Adviser. The articles were first published in The Searcher magazine and are reproduced here with kind permission of Harry Bain, editor for The Searcher.

Typical Late Bronze Age copper ingot from Barham, Kent (KENT-14DDA6). Fragments of dish-shaped ingots are much more common than complete examples. Copyright: Kent County Council, License: CC BY.

In Britain we are highly fortunate in our mineral resources. A lot of geology is packed into these islands, which is why the principles of geology were discovered here. All of the main metal ores are present: non-ferrous metals in the hard rocks of the north and west, and iron in the south and east. Copper ores are found in Wales, Scotland and Ireland, and tin (needed to make bronze) is found in south west England. We have Bronze Age copper mines at Great Orme’s Head in North Wales and metal was transported in the bowl-shaped ingots, pieces of which are found in later Bronze Age hoards.

Copper working started in Mesopotamia in the 6th millennium BC but the first evidence we have for metalworking in Britain dates to around 2500BC when we find gold ornaments and copper daggers in graves with “Beaker” pottery, suggesting metal working was introduced into Britain at the same time as these distinctive pots. Archaeologists saw the arrival of Beakers along with metalworking, barb and tang arrowheads, round barrows and other bits and pieces, as evidence of an invasion but this “Beaker invasion” idea fell out of favour as it came to be seen as “Imperialist”, so everything was explained by cultural influence. However, recent DNA work on Neolithic and Beaker skeletons has shown that there was a massive change in population with the Neolithic people being largely replaced – not what we expected at all.

The first metal objects were made from pure copper which is soft, weak, difficult to cast and certainly no improvement on flint and stone. Later, the addition of arsenic to the copper gave a much stronger alloy. Ötzi, the 5000 year old mummified man found in the Alps, had an axe made from arsenical copper. Arsenic was eventually replaced by tin, about 10% being added to copper to produce bronze, a strong and useful alloy. The addition of tin also made the alloy easier to cast, reducing problems caused by gas and lowering its melting point from 1085ºC to 950ºC, with further additions making its melting point still lower. Another advantage of using alloys was the great improvement to the degree to which the metal could be “work hardened” by hammering.

1. Late Neolithic/Early Bronze Age “Beaker” from Manton Warren, Lincolnshire. Other metal objects sometimes found in Beaker graves are gold basket earrings and copper awls, both of which are present on the PAS database. 2. Copper dagger from Lancashire (LANCUM-3BEC10). Analysis has shown it to have the composition expected of Chalcolithic implements. It dates to around 2350-2150BC. 3. Early Bronze Age flat axe from Lancashire (LANCUM-874565). While probably cast in a flat mould, the cutting edge and butt were hammered to shape and harden them. Date 2250-1900BC. 4. These two Early Bronze Age axes are from Thornton Carr, Lincolnshire, and were found together but show two stages of the development of both axe design and manufacturing technique. The axe on the left is an improvement on the flat axe in that it has flanges running down its sides to prevent it twisting in the haft during use. The right hand axe has flanges but also has a transverse rib to prevent it being forced into the haft and splitting it. 5. Mould for casting Early Bronze Age flat axes from Llangwelyn, Radnorshire. 6 & 7. Stone moulds for a Middle Bronze Age rapier (left) and a spearhead (right).

As well as being worked by hammering, bronze was melted and cast into moulds – indeed all bronze objects start out as castings. It’s like that the earliest moulds for making flat axes were made from stone, although the use of a hollow in the ground cannot be ruled out. These early stone moulds are referred to as “open moulds”, although its highly unlikely that they were open when used as this would allow the top of the axe to oxidise leaving a layer of crud that would have to be scraped off. Covering the top of the mould with a flat piece of wood weighted down with a stone, would ensure that the axe’s upper face remained clean, carbon monoxide from the charring wood preventing oxidation.

The next great development occurred around 1650BC with the appearance of two part moulds which allowed objects with two shaped faces, such as flanged axes, to be cast. The use of clay “cores” made it possible to form hollow castings particularly spearheads and, later, socketed axes. Cores were made from a clay/sand mix and were fixed between the halves of a two part mould before the metal was poured in.

Stone mould continued to be used in the north and west (particularly Ireland) where suitable rocks were available. These needed to stand the heat of the molten metal and to be easily worked so that the matrix could be carved. It is likely that they were heated before the metal was poured in to prevent the mould surface from flaking due to thermal shock and the cold stone “freezing” the metal before the mould was filled. 

11. Mis-cast spearhead from the Guilsfield Hoard. This was meant to be a Late Bronze Age spearhead but the core was left out leaving a solid casting. 12. Two part bronze mould for casting Middle Bronze Age palstaves (SF-2D55E2). The top of the mould is expanded to form a poruing basin for the molten metal. Date 1400-1100BC. 13. Part of a Late Bronze Age leaf-shaped sword which was either mis-cast or broken during use. An attempt was made to repair it by fusing-on a replacement hilt but it hasn’t worked. 14. One half of a bronze mould for casting socketed axes. The decoration on the back of this one is unusual. 15. Casting gates removed from Bronze Age implements (WREX-1BF26F and NARC-85E2F5). They are difficult to date because we don’t know what was being made but they are similar to what would have been produced by some of the moulds pictured here.

Many objects were cast in clay mould which disintegrated  as soon as they were used. Evidence for clay moulds and crucibles is rare even on excavations but it is worth looking out for them – they are highly friable with a black/dark grey inner surface and yellow/orange back.

Clay moulds were made using wooden models or “patterns”, examples of which were found at Tobermore in Co. Londonderry. Somewhat squashed by 3000 years in a peat bog, these patterns were for making moulds for casting two types of spearhead, two socketed axes and a socketed gouge. In use, one side of the pattern would have been covered by a layer of clay and turned over so that the other side could be covered. They would have resembled two-part bronze moulds. The clay would have been allowed to dry and harden a bit before the two halves were separated and the wooden pattern removed. The two halves were put back together with a core set between them. After thorough drying, molten bronze was poured into the mould and the casting made.

A still more sophisticated method of casting involved the use of bronze moulds – a technique which would now be known as “gravity die casting”, in which the two halves of the mould were made not of clay but from bronze. Most bronze moulds are for casting palstaves and socketed axes but we also have moulds for spearheads and a socketed gouge. Pouring molten bronze into a bronze mould isn’t as risky as it might sound as there is little chance of the casting fusing to the mould. There are occasions when the intention was to fuse two pieces of metal together to effect a repair (known as “burning on”) but this is difficult to achieve as the molten metal needs to flow over the solid casting so that it melts. The use of die-casting in the Bronze Age points to the production of axes being on what must have been an industrial scale.

16. Fragments of a clay mould for casting Late Bronze Age swords, and the hilt from an unfinished Ewart Park type sword. 17. Part of a stone mould for casting socketed axes, found at St. Keverne in Cornwall (CORN-031000). The use of stone moulds is a feature of the north and west of Britain. 18. Socketed hammer of typical Bronze Age type (NARC-5B738E). The PAS has recorded over 40 of these hammers. 19. A sheet bronze shield found in Brumby, Lincolnshire. Incredibly, this would have started as a small bronze ingot and was slowly hammered out to shape. It would have provided no defence against Bronze Age weaponry so can only have been used for display. 20. Bronze anvil from Chitterne in Wiltshire (HAMP-7C41DB). 21. This odd looking object from Norfolk (SF-832744) brings together both cast and wrought technology. It is a handle from a Bronze Age cauldron dating to 1200-650BC. 22. Early Bronze Age razor from Caistor in Lincolnshire. The ridges running around the edge were made by hammering the metal to thin, harden and sharpen it as well as having a decorative effect.

The evidence we have for Bronze Age casting includes finds of casting gates through which the metal flowed into the mould. Some of these we can feel confident are Bronze Age but many are impossible to date and could have come from castings of recent date. Defective castings are uncommon (they probably went straight back in the pot) but the fault that we see most often is miscasting where the metal has not full filled the mould. We also see some displaced cores and porosity caused by gas in the metal.

In addition to casting, Bronze Age metalworkers were also producing objects from sheet metal some of which, like shields and buckets, were impressive objects. Nowadays we take sheet metal for granted and can buy metal rolled to any gauge you want. Things were different in the past and if you wanted sheet metal you had to start with an ingot and hammer it to the required thickness, repeatedly stopping and heating the metal to anneal it, preventing it from becoming too hard and cracking. Some Bronze Age hammers and anvils have been found by detectorists and recorded by the PAS.

The work of detectorists and the PAS has revolutionised our knowledge of Bronze Age metalworking with a massive increase in finds and knowledge – we live in exciting times!