INTENTIONAL BREAKAGE IN THE BRITISH NEOLITHIC: SOME COMMENTS AND EXAMPLES

Transcription

1 Hugo Anderson-Whymark INTENTIONAL BREAKAGE IN THE BRITISH NEOLITHIC: SOME COMMENTS AND EXAMPLES Hugo Anderson-Whymark 1 ABSTRACT The act of intentionally breaking lithic artefacts results in distinctive fracture patterns that have been well documented by replication studies. These attributes have allowed the identification of intentional breakage in numerous lithic industries, but to date there has been little consideration of intentional breakage in the British Neolithic, despite the use of this technique for the production of tools (e.g. chisel arrowheads) and their destruction before burial. This paper explores the role of intentional breakage in the British Neolithic and argues for more systematic recording of patterns of breakage. Full reference: Anderson-Whymark, H Intentional breakage in the British Neolithic: some comments and examples. Lithics: the Journal of the Lithic Studies Society 32: Keywords: Intentional breakage; British Neolithic; arrowhead manufacture; chisel arrowheads. INTRODUCTION The intentional breakage of flakes, primarily for the production of tools or usable flake segments, is well documented in diverse lithic industries. For example, among Mousterian and Late Upper Palaeolithic flakes (Bordes 1953; Bergman et al. 1987); Near Eastern Neolithic and Chalcolithic industries, particularly in relation to the production of sickle elements and burins (Rosen 1997); and, closer to home, in the manufacture of gunflints (Skertchly 1879). In the British Neolithic, however, the role of intentional breakage has been neglected, although it is both a feature of tool production (e.g. chisel arrowheads) and destructive deposition practices (Thomas 1999, 66). This paper highlights the attributes of intentional breakage, in the hope of increasing the identification of this practice, and reviews the current evidence for intentional breakage in the British Neolithic using examples from the Middle and Upper Thames Valley. PREVIOUS RESEARCH AND REPLICATION STUDIES The majority of the previous work undertaken was based upon the investigation of Palaeolithic assemblages and has been comprehensively reviewed by Bergman et al. (1987) and as such does not require duplication in this short article. Replication studies have played an important role in the investigation of intentional breakage and studies by Bosinski and Hahn (1972), Owen (1982) and Roche and Tixier (1982) have all highlighted traits and patterns of breakage. The most extensive replication study of over 300 experimentally broken surfaces was undertaken by Bergman et al. (1987). The study provides a broad framework for the identification of intentional breaks; results which have been borne out by the current author s own experimental replication. The characteristics of intentional breakage form two distinct groups which have been defined as contact features and features resulting from flexion (Bergman et al. 1987). The contact features result from direct percussion on the surface of the flake and include the presence of a cone or bulb of percussion, incipient cones or crushing at the point of percussion (ibid. 24). Notably many of these features are very slight and can only be identified by careful examination. Features resulting from flexion (fracture initiated by bending the flint) include wedge-shaped fracture lines, lips on the edge of the breaks and conchoidal fracture marks (ibid. 26; see Figure 2 for examples). The wedge-shaped fracture lines always develop on the opposite surface from the point of impact and on occasion, lead to the detachment of wedgeshaped elements (see Figure 1, 8 9). The development of a lip is always on the opposite surface to the wedge-shaped fracture lines, and relates to the surface of impact. The study determined that contact features can be used confidently to identify intentional breakage, but it is noteworthy that 25% of Leafield Road, Oxford, OX4 2PL. hugo@flintwork.co.uk 16

2 Lithics 32 Figure 1. Archaeological examples of intentional breakage in the British Neolithic. intentionally broken flakes only bore attributes from flexion. Bergman et al. (1987, 28) consider the features resulting from flexion to be less reliable in the identification of intentional breakage, as during some flexion features appeared on accidental breaks. However, these accidental breaks developed during attempts to manufacture intentional breaks, indicating that the flakes were subject to a considerable impact prior to a break at an unintentional location (ibid. 27). The current author s replication of 60 intentional breaks only produced wedge-shaped fracture lines on intentional breaks, and the experiments demonstrated the considerable force required for these wedge-shaped lines to develop. Therefore, whilst only contact features can positively identify an intentional break, the presence of wedge-shaped fracture lines can in the current author s opinion be taken to suggest that the flake has been subject to a considerable impact, most probably resulting from an intentional delivery. BREAKAGE IN BRITISH NEOLITHIC ASSEMBLAGES Intentional breakage in the British Neolithic will be considered in three main categories: 1. The use of intentional breakage for the production of blanks for retouched tools. 2. The production of blade and flake segments for use in an unaltered state. 3. The intentional breakage of retouched tools and utilised flakes. In general, the level of breakage encountered in Neolithic assemblages is usually somewhere between 35% and 40% of the total number of flints in an assemblage. For example, out of 11,000 flints from a preserved earlier Neolithic land surface at Dorney Lake (Area 6), Buckinghamshire (formerly known as Eton Rowing Course) 35% were broken, while a further 9,800 flints in the same area, but 17

3 Hugo Anderson-Whymark preserved in surface middens and deposits within tree-throw holes averaged 38% (Lamdin-Whymark forthcoming). Levels of breakage commonly encountered in the later Neolithic are perhaps marginally higher than that of the earlier period, and particularly high levels of breakage are seen in association with Grooved Ware. Indeed, at Barrow Hills, Radley, Grooved Ware associated pit 3196 (1117 flints) and ring ditch 801 (519 flints) included 58% and 49% broken pieces respectively (Bradley 1999, 49 and 87). displayed fractures resulting from percussion or flexion. The proximal pieces represent the removal of the bulb, while the distal ends are considered to represent mis-shapen and abandoned arrowhead blanks. A selection of the debitage, and a complete chisel arrowhead from the same feature, are shown in Figure 2. Blank production using intentional breakage The intentional breakage of flakes for the production of tools is most commonly associated with transverse arrowheads. However, it is not uncommon for the bulbar end of serrated flakes and scrapers to have been removed by an intentional blow, presumably to facilitate hafting or prehension (see Figure 1, 7). Chisel and petit-tranchet transverse arrowhead forms are almost entirely produced on blanks created by the intentional breakage of a flake into two or more fragments. The intentional breakage of a flake removes the bulb of percussion, leaving a blank requiring limited retouch to acquire the desired form. The process of retouching very often removes the majority, or all evidence of the snap itself making it impossible to consider the method employed to break the flake. Given the simple method of blank production it is not entirely unsurprising that debitage from chisel arrowhead production is rarely identified in flint assemblages. A range of debitage resulting from the production of transverse arrowheads was, however, recovered from excavations on the Maidenhead to Windsor Flood Alleviation Scheme. A small assemblage of 64 flints recovered from a Plain Bowl-associated deposit within a tree-throw hole at Marsh Lane West on the Flood Alleviation Scheme (Lamdin-Whymark forthcoming), represents the first evidence of debitage from transverse arrowhead manufacture. The assemblage included a large number of broken flints amounting to 60% of the assemblage. Among these broken flints several proximal and distal flake fragments Figure 2. A selection of debitage from the manufacture of transverse arrowheads and a complete example of a chisel arrowhead from a tree-throw hole on Marsh Lane West. A second example of an unfinished transverse arrowhead was recovered from a second early Neolithic, Plain Bowl-associated tree-throw hole on Marsh Lane East, Site 1 (Lamdin- Whymark forthcoming). The tree-throw hole contained 360 flints, 42% of which were broken, including a small number with intentional breaks. A refit was identified between one of these broken flakes and the unfinished chisel arrowhead (Figure 3). The break demonstrated features resulting from flexion including a lip and wedge-shaped fractures, which indicated that the flake was broken by pressure from the dorsal surface. At the distal end a small area of invasive retouch has been applied to the dorsal surface, but the blank appears to have been abandoned due to the break fracturing at the wrong angle and failing to form the desired shape. Segmented flake and blade tools The intentional reduction of flakes and blades into usable segments is poorly documented in British Neolithic assemblages, although a few possible examples may be highlighted (Figure 18

4 Lithics 32 Figure 4: Traits of intentional fracture. 1. Direct percussion, note the slight negative bulb of percussion, lip and wedge-shaped fracture. 2. Anvil struck, note the bulbs of percussion on each side of the flake (one positive and one negative). 1, 1 4). The identification of these tools is however universally problematic, even within industries which accept the presence of segmented flake tools. For example in the Near Eastern Neolithic and early Bronze Age where the use of blade fragments as sickle blades is commonly accepted (Rosen 1997, 49). The identification of some of these elements is still often difficult and Rosen dwells on the problem of distinguishing a broken sickle element from a snapped sickle element if it is the appropriate length for hafting (c 5 cm; ibid.). In this debate Rosen raises the problematic issue of the inconsistency in categorisation between technological attributes and typological categories (ibid.). The presence of intentional breaks technologically confirms the presence of a segmented blade or flake, but it is the regular typological form which allows categorisation as a sickle element. The question over the existence of segmented flake tools in the British Neolithic is therefore not easily resolved as it is essential to demonstrate both intentional breakage and the presence of regular typological forms of segmented flaked. However, few regularly segmented flakes have been identified in the British Neolithic to date and the occasional wedge-shaped flake fragments that are recorded, appear to represent no more than ad hoc tools. It is also noteworthy that the technique of segmenting flakes against an anvil, replicated using stone tools in Figure 4, has not been observed in any of the archaeological examples of intentional fracture from Neolithic contexts so far examined by the author. The intentional breakage of tools Intentional breaks have been observed on a wide variety of retouched artefacts, but scrapers are by far the most frequently broken tool type and many of these examples are associated with interesting deposition patterns. At Blewbury, Oxfordshire, a scraper was described as having been snapped in two and a half placed in each of one of a pair of pits (Halpin 1984; Thomas 1999, 66). The breakage of this tool, although technically not described in a fashion to allow definitive identification as an intentional break, and the artefacts mode of deposition have, however, been cited as a carefully controlled practice associated with Neolithic pit deposition practices (Thomas 1999, 66). At Barrow Hills, Radley (ring ditch 801) a scraper was recovered which had been snapped in half and then in half again and while again the break is not described it is noted that the scraper was perhaps deliberately snapped (Bradley 1999, 49). Similarly, at Dorney Lake, a quarter of a scraper was recovered a pit (13650), but it was clear the tool had been snapped half by flexion, burnt and then quartered by another intentional blow which left clear contact features (Lamdin-Whymark forthcoming). Gaffney and Tingle reviewed the evidence for breakage across 725 scrapers collected during field survey on the Berkshire Downs (1989). In total 60 were broken, representing 8.3% of the total; a figure over double the 4% breakage 19

5 Hugo Anderson-Whymark recorded among flakes (a figure which included breaks resulting from plough damage). The high level of breakage recorded and a pattern of breakage often propagate from the retouched edge led the authors to suggest that the damage may have resulted from the use of the scraper as the working edge of a hafted implement, perhaps functioning as a substitute for larger core tools (ibid, 47 48). This conclusion can be questioned as it implies that the tools were used for tasks for which they were inappropriate and liable to break, however, hafting may have an important role to play in the pattern of breakage observed, but perhaps for a different reason. If we are to consider that a tool has been adequately hafted for use, at the point it is considered worn out it will need to be removed from the haft to allow for the insertion of a new tool. The force needed to remove a tool from its haft will depend on how well the tool is fixed and in certain cases a substantial blow, perhaps with enough force to break the flint, may be needed to facilitate its removal. An explanation of breakage during removal from a haft need not only apply to scrapers, as it is likely that numerous tools and flakes were hafted for use, and may have been broken in such a fashion. The practice of deliberate breakage is, however, not solely confined to scrapers. A later Neolithic pit in Oxford contained two halves of a well-utilised flake were refitted on an intentional break (Lamdin-Whymark 2005). Furthermore, within the micro-debitage from this pit small wedge-shaped elements resulting from flexion were recovered (Figure 1, 8 9), indicating the breakage of other flints. CONCLUSIONS Replication studies have provided a significant body of data to aid the identification of intentional breakage. These studies have, however, not been extensively applied to lithic assemblages from the British Neolithic and currently patterns of breakage are not systematically recorded. There is, however, little question of the use of intentional breakage for the production of transverse arrowheads in the British Neolithic, although the subject of a snapped flake industry requires considerably more investigation to demonstrate distinctive typological forms. The intentional breakage of tools, particularly scrapers, is of considerable interest and requires further research to investigate if the patterns of breakage are to deposition in specific contexts, e.g. pits, or if it reflects methods of hafting and/or technique of extracting the tools from their haft. ILLUSTRATION CATALOGUE 1. Oxford Chemistry Building, Oxford. (Lamdin Whymark 2005). Context 3443, Later Neolithic pit. Flake with distal break from direct percussion to dorsal surface; note the prominent bulb of percussion. 2. Taplow Court, Bucks. Allen et al. 2009). Context 219,?Neolithic. Medial segment of bladelike flake with evidence of direct percussion from the dorsal surface. 3. Cotswold Community, Gloucestershire. (Lamdin-Whymark 2010a; 2010b). Small find 2076, context 16007,?Neolithic. Medial segment of a flake with evidence of direct percussion. 4. Taplow Court, Bucks. Allen et al. 2009). Context 819, Earlier Neolithic tree-throw hole. Medial segment of flake with evidence of direct percussion. 5. Taplow Court, Bucks. Allen et al. 2009). Context 476, Later Neolithic tree-throw hole. End scraper with a break from direct percussion to the dorsal surface. 6. Taplow Court, Bucks. (Allen et al. 2009). Context 138, Residual, Neolithic/Bronze Age. Distal fragment of a scraper broken by direct percussion to the ventral surface. 7. Cotswold Community, Gloucestershire (Lamdin-Whymark 2010a; 2010b). Small find 2139, context Later Neolithic Durrington Walls Style Grooved Ware associated pit. Side scraper; note the removal of the bulb by direct percussion to the ventral surface (see additional fracture/fissure) prior to the application of retouch. The distal break, formed by direct percussion to the dorsal surface, post dates the use of the artefact. 8. & 9. Oxford Chemistry Building, Oxford. (Lamdin-Whymark 2005). Context 3443, Later Neolithic pit. Two wedge-shaped elements resulting from direct percussion to the ventral surface of a single ridged flake. ACKNOWLEDGMENTS The author would like to thank Eton College, the Environment Agency (through Phil Catherall) and Oxford Archaeology for permission to use their data in advance of full publication. REFERENCES 20

6 Lithics 32 Allen, T. G., Hayden, C., and Lamdin-Whymark,H, From Bronze Age enclosure to Saxon settlement: archaeological excavations at Taplow Hillfort, Buckinghamshire. Thames Valley Landscapes Monograph 30. Oxford Archaeology, Oxford. Bergman, C. A., Barton, R. N. E., Collcutt, S. N. and Morris, G Intentional Breakage in a Late Upper Palaeolithic assemblage from Southern England. In M. H. Newcomer and G. de G. Sieveking. (eds) The human uses of flint and chert: proceedings of the Fourth International Flint Symposium held at Brighton Polytechnic, April 1983: Cambridge University Press, Cambridge. Bordes, F Notules de typologie Palaeolithique: I. Outils mousteriens a fracture volontaire. Bulletin de la Societe Prehistorique Francaise 50: Bosinski, G. and Hahn, J Der Magdalenien- Fundplatz Andernach (Martinsberg). Beitrage zum Palaolithikum in Rheinland. Rheinische Ausgrabungen 11: Bradley, P Worked Flint. In A. Barclay and C. Halpin. (eds) Excavations at Barrow Hills, Radey, Oxfordshire. Volume 1: The Neolithic and Bronze Age monument complex. Oxford Archaeology, Oxford. Gaffney, V. and Tingle, M The Maddle Farm Project: An Integrated survey of Prehistoric and Roman Landscapes on the Berkshire Downs. British Archaeological Reports. British Series 200. Archaeopress, Oxford. Halpin, C Blewbury. Oxford Archaeological Unit Newsletter 11: 1 2. Inizan, M.-L. Roche, H. and Tixier, J Technology of knapped stone. C.R.E.P., Meudon. Lamdin-Whymark, H, 2005 The flint. In, P. Bradley, B. Charles, A, Hardy and D. Poore, Prehistoric and Roman activity and a civil war ditch: Excavations at the Oxford Chemistry Research Laboratory, 2-4 South Parks Road, Oxford. Oxoniensia 70, Lamdin-Whymark, H, 2010a Worked flint. In K. Powell and A. Smith, Evolution of a farming community in the Upper Thames Valley: excavation of a prehistoric and post-roman landscape at Cotswold Community, Gloucestershire and Wiltshire. Volume 2: Specialist reports. Thames Valley Landscapes Monograph 32. Oxford Archaeology, Oxford. Lamdin-Whymark, H, 2010b Worked flint. In K. Powell, A. Smith and G. Laws, Evolution of a farming community in the Upper Thames Valley: excavation of a prehistoric and post-roman landscape at Cotswold Community, Gloucestershire and Wiltshire. Volume 1: site narrative and overview. Thames Valley Landscapes Monograph 31. Oxford Archaeology, Oxford. Lamdin-Whymark, H. Forthcoming. The struck flint. In T. G. Allen, A. Barclay and A. Cromarty (eds) The Archaeology of the Eton Rowing Lake and the Maidenhead to Windsor flood alleviation scheme: Volume 1: Neolithic to Middle Bronze Age archaeology. Thames Valley Landscapes Volume. Oxford Archaeology, Oxford.. Owen, L Analysis of experimental breaks on flint flakes and blades. Studia Praehistorica Belgica NFWO Contact Groep (Prehistorie-Prehistoire) Group Contact FNRS: Roche, H. and Tixier, J Les Accidents de Taille. Studia Praehistorica Belgica NFWO Contact Groep (Prehistorie- Prehistoire) Group Contact FNRS: Rosen, S. A Lithics After the Stone Age: A handbook of stone tools from the Levant. Altamira Press, London. Skertchly, S. B. J On the manufacture of gun-flints: the methods of excavating for flint, the age of Palaeolithic man, and the connexion between Neolithic art and the gun-flint trade. H.M.S.O., London. Thomas, J Understanding the Neolithic. Routledge, London. 21