Learning Curves
The tradition of cast bronze horns developed in Ireland in the Bronze Age. They are masterpieces of early metalwork although much of their casting methods remain speculation. Holger has been working with the experimental archæology project Umha Aois since 2008 to resolve many of their mysteries. No archaeological artifacts of the process itself have been found yet. A definite conclusion about the actual method of casting has to be established and relies on close analysis of the horns (Holmes, 1978 and O'Dwyer, 2004), other archaeological evidence and experimental/practical approaches to the issue. While Peter Holmes maintains a wax-free approach with a two-part mould, a successful experiment to proove this method has yet to be concluded. The lost wax process appears to be possible, but any successful outcomes need to be closely examined to establish if they display the unique characteristic details of the originals. Simon O'Dwyer (Prehistoric Music Ireland) produces satisfactory results using lost wax with contemporary ceramic shell moulds. He has demonstrated clearly the surprising range of styles with which these instruments can be played to produce astonishing sound.
Experimental Archaeology | Outcomes and Findings
Subject: The Dunmanway Horn
Bronze-Age horns from the south-west region of Ireland share many distinctive features. The Dunmanway Horn and the side-blown horn of Mount Edgecumbe (found in a bog near Mallow, Co. Cork in 1750) for example show similar characteristics, dimensions and proportions, including a stepped terminal cone, two rings in the same orientation, six mouth cones and similar bell opening size. The Edgecumbe horn also shows remains of gritty core material, a casting seam on the inside and corresponding discolouration on the outside. The basis for the experiments is the LBA horn (1,000-800BC) found in Dunmanway, Co. Cork in the 1820s. For a while it was in the posession of the reknown folklore collector Thomas Crofton Croker (1798-1854), a native of Cork. The instrument is now in the British Museum, London. It is about 28in long with a mouth diameter of slightly less than 4in. The mouth opening is 5in from the terminal and measures 2in x 1.25in. The finished horn weights 3.25kg or 7lbs.
Aims: Casting Bronze Age Horns
The experimental approach aim to establish the most likely method(s) of producing horns in the Bronze Age. The following problems need to be resolved in the process:
1. establishing a suitable moulding technique for the pattern (wood, wax or clay)
2. ensuring relatively even wall thickness throughout the finished pieces
3. ensuring even flow of metal through the mould cavity, avoiding air locks
4. producing a good surface quality with a minimum of finishing work and fettling
Fiona Coffey produced the first result with very few casting faults during the Umha Aois in 2001 using lost wax. Lost wax in combination with authentic BA clay/sand/dung moulds, refined by Umha Aois over recent years, can produce excellent results in all respects and an even thickness of about 2mm +/- 0.5mm can be achieved in the latest result. Pouring the horn on its side or resting on its end pieces both produce successful casts. The high standard of mould mix reduces the finishing tome to less than 1 hour even with basic hand tools.
The Argument: Lost wax or no wax
The arguments in simplified form: Cave paintings in Spain show that beekeeping in Europe dates back around 8,000 years and the lost wax process was certainly well established for bronze casting in the Iron Age. In the British Isles, however, it is - apparently - only in use since the Iron Age. The process is technically most appropriate to produce the complex form of the horns – if wax was facilitated in the casting process at the time. While it would explain slight difference in thickness, an oval bell mouth and chaplets, but it would not explain the need for a joint line in the core. The two-part clay mould argument put forward by Peter Holmes in 1978 requires no wax, relying on clay/sand alone. Keeping a core over that length steady to such exact tolerances using chaplets, however, is almost impossible. While the process is a viable explanation for the core line, severe miscasts are almost inevitable. No successful horn has ever been cast with this method.
Back to the top
Methods: Experimental Approach
A comparative study of various casting methods need to include:
1. Lost wax around a pre-made core with clay/sand/dung mould
2. Hollow lost wax from a clay or plaster mould, dipped mould as above
3. Cavity two-part mould with solid core / cope and drag (suggested by Peter Holmes)
4. Wooden pattern in combination with clay moulds
5. Stone mould with core (no evidence found and technically difficult and unlikely)
Amongst other indications, the following features of some of the cast side-blown horns could be crucial in determining which of the above processes is the most likely type of manufacture:
- Two horizontal, raised seam lines on the inside of many horns
- Discolouration of the outside along this seam or slight mismatch of mould
- Horizontal toolmarks running inside
- Remains of gritty core material
- Slightly uneven thickness – thicker at the mouth piece
- Slightly oval bell mouth
- Terminal cones are not perfectly shaped but somewhat uneven
- Use of chaplets to suspend the core
- Surface quality, miscasts
All the above except the first point are displayed by the successful horn cast. While this first point was interpreted by Holmes as a clear indication of a cope and drag mould, Jeroen Zuiderwijk demonstrated at the Umha Aois symposium in Doolin 2010 that this can also be a sign of a two-part wax pattern.
The experiments use available materials (clay, sand, dung, charcoal, willow, grass, timber, copper, tin, bone, beeswax, ash and stone) in the Bronze Age and likely methods evident in other objects. Authenticity of these methods is not implied, contemporary tools and utensils are used where appropriate (e.g. kettle instead of clay vessels).The horns require approximately 5-8lbs of metal (average 90%Cu, 10%Sn), an amount easily produced in a pit furnace. However the lack of evidence of sufficiently sized crucibles from prehistoric times still leaves to speculation how these horns were actually poured. Both charcoal-fired and bellows operated pit furnaces and gas furnaces will be used in the reproduction. Clay, sand and dung moulds are best poured at lower temperatures of around 1,100ºC, the moulds fired to 600ºC but cooled to c. 200ºC - pre-heating and higher temperature may both result in inferior surface quality and subsequently more tooling.
Back to the top
Selected Bibliography
Syer Cumming, H.; On Phonic Horns; Journal of the British Archaeological Association, Volume 5; British Archaeological Association; 1850, pp.119ff.
Holmes, Peter: The manufacturing technology of Irish Bronze-Age horns; in: The Origins of Metallurgy in Atlantic Europe (Ryan, M., ed.); Dublin, 1978
O'Dwyer, Simon: Prehistoric Music of Ireland; Stroud, 2004
Gaskell Brown, Cynthia; The Mount Edgecumbe Irish Bronze-Age Horns; The Journal of the Antique Metalware Society, Vol. 16, June 2008; pp.44-51
Back to the top
