Beads and beakers: heirlooms and relics in the British Early Bronze Age.

Woodward, Ann

The idea that artefacts have biographies, with complex life cycles, is not a new one. The book The social life of things (Appadurai 1986a) has been influential amongst archaeologists, but very few detailed analyses of specific biographical trajectories have been undertaken (e.g. Skeates 1995). Using two examples of Early Bronze Age data, this study attempts to develop the line of enquiry and to distinguish between some different types of life cycle. Objects, like people, may have possessed extended biographies. From the moment of production an artefact may have passed through several different spheres of use and meaning (Kopytoff 1986). It may have gone out of use for short or long periods due to deliberate storage, burial or other forms of deposition. Subsequently items may have been recovered and, in whole or in part, returned to circulation within, the human environment. Thus some items may have experienced extremely prolonged, but maybe interrupted, biographical journeys.

At the outset it may be useful to differentiate between valuables and more mundane artefacts. Valuable items, whether defined by their complexity of production, such as a sword or polished stone axe, or by their exotic raw material, e.g. gold, jet or amber, may well have possessed very long life cycles. However, the studies of abrasion, fragmentation and context that might enlighten their life stories have seldom been attempted. Valuables might appear to be the classic candidates for the role of heirlooms in early society. However, the definition of an heirloom is `any piece of personal property that has been in a family for several generations' (OED, author's emphasis). The social value of an object is not related solely to its intrinsic exotic worth, but also to the value that it has accrued over the years in relation to specific human individuals, to families or to other social groupings, or to its significance in systems of reciprocity or exchange. As Appadurai (1986b) has emphasized, certain items become valuables through sequences of exchange and overt display in the social environment.

One way such long-lived valuables might be recognized in the archaeological record is if they occur as fragments. Indeed it has been postulated that some of artefacts in the Balkan Neolithic and Copper Age were manufactured so that they would easily break into recognizable pieces which could be exchanged within social networks (Chapman 2000:70-79 & 104). This phenomenon of fragmentation links the two case studies offered in this paper. Amber spacer beads are fragments from the original composite crescentic necklaces of which they formed components, and large portions of Beakers are pieces from vessels that originally were complete. The amber beads have an intrinsic, exotic value; the Beaker portions may have been less exotic, but appear to have been imbued with a very high value in social terms. It is argued that both functioned as heirlooms. However, this overarching interpretation does carry with it the implication that all heirlooms were subjected to similar patterns of deposition. It is difficult to assemble firm evidence for the processes of temporary curation, long-term storage or interim deposition that might have been involved. However, it can be suggested that intrinsic valuables might have been secreted on or off the body in small bags, pouches or boxes, or worn as bodily ornament. They may have been taken out from time to time for purposes of ritual display, or engagement in cycles of gift giving or exchange. It seems unlikely that large pieces of pottery were preserved in this way, although it is not impossible that part of an antique vessel could have been stored in a cloth or box. One other possible process whereby important vessels, or parts of them, may have been preserved temporarily has been identified and considered in some detail, namely the process of midden formation. Special pots, either complete or broken, may have been deposited within middens, perhaps after feasting episodes, and parts of them retrieved at later dates. A similar idea was suggested by Bruck in relation to the Beaker sherds from Windmill Hill (Bruck 1999).

A detailed study of the formation processes involved in the accumulation of middens in Late Bronze Age Britain has been undertaken by Needham & Spence (1997). They have discussed the interaction of the artefact use cycle and the refuse cycle, which together make the life cycle of artefact biographies (Needham & Spence 1997: figure 1). In basic terms the midden comprises a deposit of provisional refuse (Schiffer 1987: 64:72), which can be used as a resource, both as manure or as a source of raw materials (e.g. flint, stone, potsherds for grog) from discarded or broken artefacts. In social terms, middens may have formed dumps of selected categories of material in specific zones related to living spaces and structures (Moore 1982: figure 4). They also may have been preferred locations for periodic exchange and feasting, or for other acts of conspicuous consumption (Lawson et al. 2000).

Although there is no evidence that large-scale accumulation of middens occurred in southern Britain before the 9th century BC (Needham & Spence 1997: 87), some degree of activity may be adduced for the middle and late Neolithic periods. Humphrey Case (1995: 10-11) has argued that the artefact-rich fillings of many Neolithic pits may comprise deposits of midden material, and that the stone chambers of the West Kennet long barrow were filled up with large quantities of pottery and other debris derived from two different middens. Other part vessels, such as the Beaker portions described below, may also have derived from middens. If these middens were of specific location and ownership, then the resurrected vessel portions may have been heirlooms in the generational sense defined above; but if the source middens were non-specific, then the Beaker portions would not have been heirlooms in this sense, but would have been recognized antique items perhaps better termed relics. `Relic' is used here not in the usual religious sense, applying to the body parts or possessions of holy predecessors, but in one of its more general meanings as `an object invested with interest by reason of its antiquity or associations with the past' (OED). This definition implies remembrance of things past, but in a generalized sense, in contrast to the direct ancestral connotations of the term heirloom.

The details of the discovery in 1994 of an important and highly unusual hoard of Early Bronze Age material on the periphery of a round barrow at Lockington, Leicestershire, England, have recently been published, including stimulating discussion of the metal objects found, two gold armlets and a long copper dagger, and the Early Bronze Age traditions of metalworking to which they belonged (Needham 2000). The other items found with the metal hoard were less spectacular, being the broken portions of two vessels, both of which probably were Beakers. However, the archaeological implications of these ceramic finds are also far-reaching (Woodward 2000a). The Beakers had been deposited as already broken portions of vessels, one inside the other and both covering one of the gold armlets. The rims and necks of both were missing and the sherds were all very abraded. The pots may have functioned as carefully curated family or ancestral property, with the missing portions continuing in circulation as heirlooms, or ground up to form a potent grog for incorporation in newly produced pots. Thus the essence of important pots belonging to significant individuals or families could be preserved and passed down through the generations in a finite and often visible form. The fabric of a pot may have indeed assumed a far greater social and symbolic importance than either its form, or its decoration, sometimes even incorporating the actual remains of a specific human ancestor in the form of calcined bone temper (e.g. at Balneaves, Angus: Russell-White et al. 1992: 299). There is good ethnographic authentication for the symbolic reuse of selected potsherds (e.g. Sterner 1989) and a similar idea has previously been advanced by Brown (1995: 127). The use of pots in this way would also link to another theory, that ceramic vessels of the Late Neolithic and Early Bronze Age period in Britain related to specific individuals rather than to the community at the household, settlement or regional level (Woodward 1995).

In discussion of the Lockington Beakers a list of other potential examples of heirloom vessels was assembled (Woodward 2000a: 58-60). Such pots can only be identified from excavation reports if they were recorded as having been incomplete at the time of deposition. Relevant details would include for instance the observation of ancient breaks, or the fact that significant rim or other portions were missing from pots that had been deposited in an inverted position. Two notable items are the vase a anse found nested, within two other urns in a barrow on Gallibury Down, Isle of Wight (Tomalin 1988: 208-9, figure 4,1), and the nested pair of two Collared Urns, both with significant portions missing, from Long Crichel, Dorset (Green et al. 1982: 50-51). Several recent finds of Beakers appear to have resulted from a similar process of selective deposition. Two pits from separate excavations near Alrewas in Staffordshire contained large portions of Beakers, approximately half of the vessel in each case (Woodward forthcoming).

Independently, and prior to the publication of the Lockington vessels, Bradley had noticed quite different evidence that fragments of Beakers were possibly being employed as heirlooms or relics. He has drawn attention to the contrast between the sequences of ceramic deposition in two different areas of the great henge enclosure at Mount Pleasant in Dorset. These sequences have recently been analysed by Thomas (1996: chapter 7) using data from the full excavation report (Wainwright 1979). Bradley showed that the expected chronological sequence of styles represented in the stratified layers within the eastern ditch terminal of the north entrance of the main henge enclosure was not reflected in the fillings of the ditch of the Site IV timber henge monument (Bradley 2000: figure 37). Here the `orderly pattern is upset. The lower levels of the ditch contain earlier Neolithic pottery as well as Grooved Ware, despite the fact that these styles were in use hundreds of years apart. Beaker pottery is found throughout the filling of the ditch, sometimes in association with the very styles that seem to replace it in the entrance to the monument' (Bradley 2000: 128). It seemed that finds of different original dates were being introduced to the ditch from other locations, and that some of the material, particularly the Beaker pottery, had been circulating for a long time prior to final deposition. In other words, the portions of vessels involved, as suggested by the Lockington examples described above, may have served as heirlooms or relics.

This hypothesis can be investigated further by considering various aspects of the fragmentation of Beakers found at Mount Pleasant. Most Beaker pottery from Site IV was found in segment XIII, the western terminal of the north entrance of the ditch, and within this, the major assemblage derived from layer 5 (Wainwright 1979: figure 13). This was probably associated with the construction of the stone setting within the timber henge (Wainwright 1979: 22). In this layer there were significant quantities of sherds belonging to Beakers of the following styles, as defined by Clarke (1970): All Over Cord, European, Wessex/Middle Rhine, Northern/ Middle Rhine, Northern/North Rhine and Southern (Longworth 1979: 86-8, catalogue items P131-P227). Furthermore a series of degrees of vessel representation, ranging from major near-complete profiles to small individual sherds, were present. Using the data provided by Longworth, and by assessing the sizes of the illustrated items, along with the sherd totals recorded in the published catalogue entries, it has been possible to correlate the degree of vessel representation amongst the vessels belonging to each Beaker style (TABLE 1). The diagnostic material can be divided into three categories: large portions of vessels or `chunks', large sherds and smaller sherds. Higher up in the ditch filling there were more Beaker finds including two further sherds from All Over Cord Beakers (large: P136 and small: P141) as well as a chunk, P161, probably from a Wessex/Middle Rhine Beaker.

Accepting that All Over Cord and European Beakers were earlier than the other styles (Case 1977; 1993), it is useful to compare the results of the size analysis for these two styles against the rest. It reveals that the later Beakers occur usually as large portions, whilst the earlier styles tend to occur as sherds. This would tend to uphold Bradley's theory that some of the Beaker material may have been in circulation as broken vessels for a long period of time, with the earlier portions haying become more fragmented over the years. However, that the early Beakers had sometimes been deposited as larger portions is indicated by the occurrence of two such instances within layer 5 of segment XIII, and also in segments X alpha, layer 5 (P132) and VIII, layer 6 (7131). Large chunks of the later styles also occurred sporadically in other segments, notably a Wessex/Middle Rhine piece (P146, two sherds) in segment IV alpha, layer 8 and a large fragment of Southern Beaker in segment X, layer 5 (P158, one-fifth of the vessel). The overall implication is that large chunks from broken Beaker vessels were circulating or were temporarily deposited for many years, if not generations. Thus they may have functioned as heirlooms or relics before they eventually came to be deposited within the ditch around the focal timber and stone monument. Thomas (1996) has argued that people entered the monument at the northern entrance and immediately turned left to walk between the rings of timber posts. As they did so, some individuals appear to have deposited their ancestral heirlooms or relics, some on the way, as in segments VII, X, Xa and XI (Thomas 1996: figure 7.14), but mostly at the end of their liturgical procession, in the northern terminal of the newly furbished stone and timber edifice.

I have argued elsewhere (Woodward 200Ob: 116-19) that the beads found in the rich graves of Early Bronze Age Britain, many made from exotic substances, often were not the components of necklaces adorning the dead, but were treasured heirlooms. Deriving from ancestral necklaces, they may have been used to perform magical feats in medical, religious or divinatory ceremonies. Both amber and jet display exciting electrical effects when rubbed with fur or textile; fossils and shells of fascinating shape derived from places distant in space and time, and faience was a rare and exotic glassy material of an unusual and penetrating blue colouration. The small numbers of beads found with individual Collared Urn burials, and also in Wessex graves, were demonstrated (Woodward 2000b: figure 63), and the numerical occurrence of all known instances of amber beads, per grave, displays a similar pattern (FIGURE 1).

[FIGURE 1 OMITTED]

Here it will be instructive to take a closer look at one very specific type of bead, the amber spacer plates with their multiple perforations. In southern England these are restricted to graves of the intermediate Wessex 1/Wessex 2 Wilsford Series graves (Beck & Shennan 1991: 76). Only three contexts have produced enough beads, spacers and simple types for it to be postulated that entire necklaces had been deposited in the graves: the Golden Barrow at Upton Lovell (Grinsell 2e) (FIGURE 2), a disc barrow in the Lake group (Wilsford, Grinsell 47/49/50) and the twin disc on Oakley Down (Wimborne St. Giles, Grinsell 8). Of the others containing spacer plates two in particular are notable: firstly the deposit of one complete jet spacer plate, and one or more broken spacers of amber in a ring ditch at Felmersham, Bedfordshire (Radwell I: Hall & Woodward 1977). These items were found in association with an amber pendant, one V-bored amber button and 116 beads of amber and jet. As amber pendants do not occur in crescentic necklaces, these items must represent the remains of at least three different necklaces of classic form. Many of the beads were broken and it seems that what we are dealing with is not a necklace as such (pace Hall & Woodward: plate 2), but a special collection of carefully gathered heirlooms which might have had uses other than the adornment of the body.

[FIGURE 2 OMITTED]

The second is the possible necklace, mainly of simple amber beads, from Barrow VI on Beaulieu Heath (Piggott 1943: 14, plate VII; also see Beck & Shennan 1991: figure 11.2, 1). Associated with the simple beads was a single spacer plate which, from the photograph, can be seen to have been found in a fragmentary condition. Mrs Piggott made the following observation: `From the interesting fact that the space-plate was not functioning as such, there being too few beads for a double row, it appears certain that the necklace was not new when buried.... Probably its rarity had occasioned it to be handed down as an heirloom'. Tomalin (1988) also has suggested that some spacer plates may have circulated as special items for a long time, and Beck & Shennan later noted (1991: 76) that `It would appear that in Britain, at least for spacer plates, there may have been an initial primary phase of circulation while subsequently some pieces remained in circulation for considerable periods and became in effect heirlooms'.

Gerloff argued that crescentic amber necklaces were probably a British design, related to gold lunulae and the jet crescentic necklaces of the Early Bronze Age (Gerloff 1975: 217-19), and the well-known examples from Mycenae and Kakovatos in Greece are thought to have been imported from Britain (Harding & Hughes-Brock 1974). All the amber spacer-plates then known from Wessex and the continent were listed and mapped by Gerloff (1975: 260-63, plate 63), and the continental examples were described and discussed by Hachmann (1957). The specific type of spacer-plate that is common to Wessex and the Mycenean shaft graves is the complex-bored version of `basic pattern'. The borings through these beads include rows of straight string holes separated by V-borings (see FIGURE 2, left). A few examples of this type also occur in Germany and Austria and there are occasional outliers, such as the find from Andrup, Jutland (Gerloff 1975: 218). But in Germany there are many other forms of boring (Hachmann 1957; Gerloff 1975: plate 63) and Hachmann has argued that in this region the; spacers were arranged within necklaces of collier, not crescentic type. In south Germany and Austria, `basic pattern' spacer beads occur in late Early Bronze Age and Middle Bronze Age contexts (of Reinecke's phases A2, B and C), whilst the other complex boring types tend to occur in later Middle Bronze Age Tumulus graves, probably of Reinecke D (Gerloff 1975: 119-220), probably the result of a local development. It will be interesting to study the numerical incidence and fragmentation of the German and Austrian examples.

Using the data provided in Gerloff's (1975) Appendix 8, numbers 11 to 51, it can be concluded that 19 out of the 37 (51%) spacers occur as singletons, and there are only four contexts where the spacers occurred in groups of five or more. The incidence of entire or near-complete necklaces, assessed from the total counts of amber beads recorded by Gerloff, is 25%, a figure which equates well with the percentage of 30% previously established for the British examples. Again, as in Britain and Greece, many of the spacers are fragmentary. The importance of this was noted by Sandars (1959: 293-4) and the number of instances can be calculated from Gerloff's catalogue. Of 12 examples of fragmentary beads recorded in Germany, seven are from sites in Wurttemberg. This region shows the main concentration of the earlier Tumulus contexts (Gerloff 1975:216), and five of the fragmentary spacers from Wurttemberg are in fact those with complex boring in the basic pattern (Gerloff 1975: Appendix 8, 28-33). Thus is seems to have been the `basic pattern' spacers that may have been deposited as heirlooms, and indeed they, like those from Greece, may have originally been imported from Britain.

Sandars (1959: 294) drew attention to Hachmann's observations, who noted that several spacers were fragments and that some of these had been re-bored using a less sophisticated technique. Probably re-use also took place in Greece, where, although the number of spacers and other amber beads from Kakovatos tholos .tomb A suggests that at least part of a crescentic necklace had been deposited, two of the spacers have secondary perforations in a central position (FIGURE 3). Furthermore, Sandars noted (1959: 294) that another amber spacer `was sufficiently rare to be misunderstood, set in gold, and hung as a pendant, to be found in a 7thcentury tomb at Knossos'. Thus we see that amber spacer plates were carefully curated, modified and re-used over long periods of time, functioning as heirlooms or relics in the same way as the Beaker vessels discussed earlier. The beads may have been valued particularly for their raw material. The magical static electrical properties of amber may have caused the beads to have been regarded as highly desirable items for ceremonial purposes, and the spacer plates, with their broad and flat surfaces, would have been especially suited to the production of electrical effects by rubbing. A similar function can be suggested for the finds of single jet spacers: jet displays the same physical properties as amber and the German word for jet is Schwartzbernstein, `black amber'. We may note the single jet spacer found in the bead assemblage for Felmersham, discussed above, and the very pertinent example of a single jet spacer bead purposely laid inside a ring of 24 barbed and tanged flint arrowheads, inside the Armorican tomb of Kerguevarac, Finistere (Piggott 1939: figure 1). Stuart Piggott (1939: 193-4) observed that `it is interesting to see that this isolated fragment of a composite necklace was regarded by its Breton possessor, ignorant of its real purpose, as a valuable and unique object in itself, and was given a position of importance and distinction in the grave in which it was found'.

[FIGURE 3 OMITTED]

The purpose of this short contribution has been to draw attention to the probable importance of heirlooms and relics in the Early Bronze Age. Discussion has centred on two classes of artefact only; no doubt similar arguments could be developed for other categories. The implications of the observations are important. Some of the social implications have been alluded to above. Many items may have been imbued with symbolic significance far beyond their apparent everyday functions, and some classes of artefact may have been much rarer than archaeologists have tended to assume. Such objects may have circulated within and between social groups both over long distances and through many generations, as significant heirlooms or as relics. According to the definitions discussed above the beads may have been heirlooms, passed down through known family or group members, while the portions of decorated Beakers may have been relics, recovered from temporary deposition within a known or discovered location. Such a location may have been a midden or a barrow; in both cases the recovered relics would have been redolent of things past in a general and maybe mystical sense.

There is also a highly significant chronological dimension to this subject. If some items within grave assemblages, hoards and other closed groups were heirlooms or relics, to what extent can the individual items within such groups be treated as contemporary? The chronology of the Bronze Age is based on studies of patterns of association of artefacts in closed contexts such as hoards or graves. It is generally assumed that the date of the individual objects making up each closed group is the same and that a radiocarbon determination from organic material associated with the group will `date' all those objects accurately. However, the different objects may have possessed biographies of varying length and complexity, with some of them having functioned as heirlooms or relics. A radiocarbon determination may date the time of deposition of a grave group but may not reflect the original periods of production and initial use of the various items that make up the group. Thus the very basis of some of our relative and absolute chronologies may need some renewed close attention. It is only through the results of detailed analysis of wear patterns, signs of re-use and fragmentation, such as those advanced here for certain Beakers and classes of amber bead, that the functional and chronological status of closed finds groups can be fully assessed.

TABLE 1. Mount Pleasant: the occurrence of Beaker vessel portions in
segment XIII, layer 5. (Catalogue numbers after Longworth 1979: figures
47-51.)

style large portion
(Clarke) rim,
 no. of rim to base to shoulder
 sherds shoulder neck and base

AOC 5 P134
AOC 8 P139
Eur 8 P164
Eur
W/MR 2 P167
W/MR 9 P166
N/MR 35 P157
N/NR 57 P147
S 2 P160

style large sherd small sherd
(Clarke)
 base shoulder rim wall

AOC P137 P140
AOC P138 P143
Eur P177
Eur P178
W/MR P179 P154
W/MR
N/MR
N/NR
S

Acknowledgements. I am very grateful to Richard Bradley, who suggested that I should write a piece on this topic and commented on a preliminary draft. The final version also benefited from suggestions kindly supplied by Humphrey Case and Stuart Needham, and by two anonymous referees. I also wish to thank Henry Buglass, University of Birmingham, who drew FIGURES 2 & 3.

References

APPADURAL A. (Ed.). 1986a. The social life of things. Commodities in cultural perspective. Cambridge: Cambridge University Press.

1986b. Introduction: commodities and the politics of value, in Appadurai (ed.): 3-63.

BECK, C. & S. SHENNAN. 1991. Amber in prehistoric Britain. Oxford: Oxbow. Monograph 8.

BRADLEY, R. 2000. An archaeology of natural places. London: Routledge.

BROWN, N. 1995. Ardleigh reconsidered: Deverel-Rimbury pottery in Essex, in Kinnes & Varndell (ed.): 123-44.

BRUCK, J. 1999. The nature of the upper secondary fill in the outer ditch, Trench B, in A. Whittle, J. Pollard & C. Grigson, The harmony of symbols. The Windmill Hill causewayed enclosure: 375-80. Oxford: Oxbow.

CASE, H.J. 1977. The Beaker culture in Britain and Ireland, in R.J. Mercer (ed.), Beakers in Britain and Europe: 71-101. Oxford: British Archaeological Reports. International series S26.

1993. Beakers: deconstruction and after, Proceedings of the Prehistoric Society 59: 241-68.

1995. Some Wiltshire Beakers and their contexts, Wiltshire Archaeological Magazine 88: 1-17.

CHAPMAN, J. 2000. Fragmentation in archaeology. London: Routledge.

COATES, G. Forthcoming. Excavations at Whitemoor Haye, Staffordshire, 1997-1998.

CLARKE, D.L. 1970. Beaker pottery of Great Britain and Ireland. Cambridge: Cambridge University Press.

GERLOFF, S. 1975. The Early Bronze Age daggers in Great Britain and a reconsideration of the Wessex Culture. Munich: Prahistorische Bronzefunde. Abteilung VI Band 2.

GREEN, C., F. LYNCH & H. WHITE. 1982. The excavation of two round barrows on Launceston down (Long Crichel 5 and 7), Proceedings of Dorset Natural History and Archaeological Society 104: 39-58.

HACHMANN, R. 1957. Bronzezeitliche bernsteinschieber, Bayerische Vorgechichtsblatter 22, 1-36.

HALL, D.N. & P.J. WOODWARD. 1977. Radwell excavations, 1974-75; the Bronze Age ring ditches, Bedfordshire Archaeological Journal 12: 1-16.

HARDING, A.F. & H. HUGHES-BROCK. 1974. Amber in the Mycenean world, Annual of the British School of Archaeology at Athens 69: 145-72.

HUGHES, G. 2000. The Lockington Gold Hoard. An Early Bronze Age barrow cemetery at Lockington, Leicestershire. Oxford: Oxbow Books.

KINNES, I. & G. VARNDELL (ed.). 1995. `Unbaked Urns of Rudely Shape'. Oxford: Oxbow. Monograph 55.

KOPYTOFF, I. 1986. The cultural biography of things: commodization as a process, in Appadurai (ed.): 64-91.

LAWSON, A.J. 2000. Potterne 1982-5. Animal husbandry in later prehistoric Wessex. Salisbury: Wessex Archaeology. Report 17.

LAWSON, A.J., A. POWELL & R. THOMAS. 2000. Discussion, in Lawson (2000): 266-71.

LONGWORTH, I. 1979. The Neolithic and Bronze Age pottery, in Wainwright (1979): 75-124.

MOORE, H.L. 1982. The interpretation of spatial patterning in settlement residues, in I. Hodder (ed.), Symbolic and structural archaeology. Cambridge: Cambridge University Press.

NEEDHAM, S. 2000. The gold and copper metalwork, in Hughes (2000): 23-46.

NEEDHAM S. & T. SPENCE. 1997. Refuse and the formation of middens, Antiquity 71: 77-90.

PIGGOTT, C.M. 1943. Excavation of fifteen barrows in the New Forest, 1941-2, Proceedings of the Prehistoric Society 9: 1-27.

PIGGOTT, S. 1939. Further Bronze Age `Dagger Graves' in Brittany, Proceedings of the Prehistoric Society 5: 193-5.

RUSSELL-WHITE, C.J., C.E. LOWE & R.P.J. MCCULLAGH. 1992. Excavations at three Early Bronze Age burial monuments in Scotland, Proceedings of the Prehistoric Society 58: 285-324.

SANDARS, N.K. 1959. Amber spacer-beads again, Antiquity 33: 292-5.

SCHIFFER, M.B. 1987. Formation processes of the archaeological record. Albuquerque (NM): University of New Mexico.

SKEATES, R. 1995. Animate objects: a biography of prehistoric `axe-amulets' in the central Mediterranean region, Proceedings of the Prehistoric Society 61: 279-301.

STERNER, J. 1989. Who is signalling whom? Ceramic style, ethnicity and taphonomy among the Sirak Bulahey, Antiquity 63: 451-60.

THOMAS, J. 1996. Time, culture and identity. An interpretive archaeology. London: Routledge.

TOMALIN, D. 1988. Armorican vases a anses and their occurrence in southern Britain, Proceedings of the Prehistoric Society 54: 203-21.

WAINWRIGHT, G.J. 1979. Mount Pleasant, Dorset; Excavations 1970-71. London: Society of Antiquaries.

WOODWARD, A. 1995. Vessel size and social identity in the Bronze Age of southern Britain, in Kinnes & Varndell (ed.): 195-202.

2000a. The prehistoric pottery, in Hughes (2000): 48-61.

2000b. British barrows. A matter of life and death. Stroud: Tempus.

Forthcoming. The prehistoric pottery, in Coates (forthcoming).

ANN WOODWARD, Institute of Archaeology & Antiquity, University of Birmingham, Edgbaston, Birmingham B15 2TT, England.

Received 12 October 2001, accepted 14 March 2002, revised 13 August 2002