Objects of the Past in the Past Investigating the significance of earlier artefacts in later contexts Edited by Access Archaeology Matthew G. Knight, Dot Boughton and Rachel E. Wilkinson eop cha r y olog Ar Acces ess A s Archae Archaeopress Publishing Ltd Summertown Pavilion 18-24 Middle Way Summertown Oxford OX2 7LG www.archaeopress.com ISBN 978-1-78969-248-8 ISBN 978-1-78969-249-5 (e-Pdf) © the individual authors and Archaeopress 2019 Cover image: The Hammer of St Martin (image courtesy of Museum Catharijneconvent, Utrecht / Ruben de Heer) Back cover image: Impression of a medieval silver signet ring, incorporating a Roman carnelian intaglio, from the Evesham Abbey Gardens hoard (Cuming 1876: 116) All rights reserved. No part of this book may be reproduced or transmitted, in any form or by any means, electronic, mechanical, photocopying or otherwise, without the prior written permission of the copyright owners. Contents Contributors����������������������������������������������������������������������������������������������������������������������������������� ix Chapter 1� Objects of the Past in the Past��������������������������������������������������������������������������������������� 1 Matthew G. Knight, Dot Boughton and Rachel E. Wilkinson Chapter 2� Doubtful associations? Assessing Bronze Age ‘multi-period’ hoards from northern England, Scotland and Wales ��������19 Matthew G. Knight Chapter 3� Connecting with the past: Earliest Iron Age multi-period hoards in Wessex �������������������������������������������������������������������������42 Dot Boughton Chapter 4� The Devil or the Divine? Supernatural objects and multi-period hoards in later prehistory ����������������������������������������������60 Alex Davies Chapter 5� Iron Age antiques: Assessing the functions of old objects in Britain from 400 BC to AD 100 �������������������������������������77 Helen Chittock Chapter 6� The Antique Antique? ��������������������������������������������������������������������������������������������������98 Mark Lewis Chapter 7� Rethinking heirlooms in early medieval graves ��������������������������������������������������������115 Brian Costello and Howard Williams Chapter 8� Medieval engagements with the material past: Some evidence from European coin hoards, AD c. 1000–1500 ����������������������������������������������������131 Murray Andrews Chapter 9� Deep Time in the ruins of a Tudor Palace? Fossils from the Palace of Placentia, Greenwich �������������������������������������������������������������������������142 Peter J. Leeming Chapter 10� A shifting chronology of combat damage: Reassessing the evidence for use and reuse on Irish Bronze Age swords�����������������������������������152 David R. Bell i Chapter 2 Doubtful associations? Assessing Bronze Age ‘multi-period’ hoards from northern England, Scotland and Wales Matthew G. Knight The deposition of earlier Bronze Age objects in Late Bronze Age contexts has been recently recognised and recorded for southern England, but the phenomenon is not confined to that region. This paper expands the corpus of such hoards by including a further 11 case studies from northern England, Scotland and Wales. Some associations represent heavily-used or worn material that had been in extended circulation. Other associations contribute to better understanding the prolonged typologies of certain objects and the importance of certain places in the Bronze Age landscape. Overall, this paper suggests we should consider the possibility that some multi-period hoards are genuine, rather than doubtful, associations. Keywords: Bronze Age, deposition, hoards, metalwork, out-of-time Introduction Bronze Age metalwork hoards containing artefacts dating to multiple periods (i.e. ‘multi-period hoards’) have frequently been noted in past surveys but such associations are generally considered doubtful (e.g. Coles 1962: 134; Davis 2012: 52; Rowlands 1976: 70). A recent survey of hoards from southern England identified 41 certain or possible multi-period hoards (Knight forthcoming) and during research for that survey, it became apparent that this phenomenon was not confined to southern England. In this paper, I identify a further 11 multi-period hoards from Scotland, northern England1 and Wales, and provide an analysis of these ‘out-of-time’ objects and their contexts. This offers the opportunity to enhance our understanding of Bronze Age metalwork typologies as well as the contexts and associations of multiperiod hoards; the paper concludes with a comparison of multi-period hoards from northern England, Scotland and Wales with those from southern England. Objects of the past in prehistory It is now widely recognised that prehistoric societies engaged with aspects of their own pasts (e.g. Bradley 2002; Chadwick and Gibson 2013; Jones 2007; Lillios and Tsamis 2010). The enduring nature of material culture, including both monuments and portable artefacts, allows us to interrogate how the past may have been understood by past communities. This has often been investigated through studies of monuments and landscapes that were repeatedly revisited over long periods of time (e.g. Bradley 2002; Gosden and Lock 1998) and there is an increasing appreciation that objects too may have been in circulation or use for extended durations (Davies forthcoming; McLaren 2017; Woodward 2002), or were otherwise rediscovered and reappropriated in later periods (Ferris 2012: 77–93; Hingley 2009; Knight forthcoming). Woodward has convincingly argued that various grave goods in Beaker and Early Bronze Age burials, including beaker pottery, amber beads and other ornaments, show signs of use, wear and curation that mean they may be considered heirlooms passed down over multiple generations (Woodward 2000: 58–60; 2002; Woodward and Hunter 2015: 472ff.). As heirlooms, such objects may have had a mnemonic function, evoking remembrance and establishing links with the past (Lillios 1999; Woodward 2002). By contrast, the various Bronze Age implements found in Iron Age features at the Northern England is defined here as the present English counties north of the River Humber (Lancashire, Greater Manchester, East, West, North and South Yorkshire, Cumbria, Northumberland, Co. Durham and Tyne and Wear). 1 19 Objects Of the Past in the Past hillforts at Breiddin (Powys) and Cadbury Castle (Somerset) are more likely to represent rediscovered deposits that were incorporated into later features given the evidence for multiple re-occupations of these sites (Hingley 2009: 163; Knight forthcoming). In such situations, the re-deposition of the object in certain locations is significant: such depositions may have served to respect (and invisibly mark) ancestral links with the land or commemorate certain locations (Hingley 2009; Knight forthcoming). A Middle Bronze Age palstave was modified with silver adhesions added to the object before it was wrapped in cloth and buried in a Late Iron Age tumulus at Lexden, Colchester (Essex), suggesting this object had been re-appropriated and manipulated a thousand years after its initial production (Foster 1986: 78–80; Hingley 2009: 150, 163). It was buried alongside numerous other ‘rich’ grave goods, including copper alloy anthropomorphic and zoomorphic figurines, a silver Augustus medallion, and pieces of furniture (Foster 1986: 53ff.); the treasured status of the palstave is evidenced by its inclusion alongside other high-status objects. In the Roman period, there is evidence that Neolithic and Bronze Age objects were found, collected and deposited in significant locations, such as temples, perhaps as votive offerings (Adkins and Adkins 1985; Ferris 2012: 77–93). From the above examples, it is clear that old objects in prehistory might be considered in terms of links with the immediate past, such as in the case of heirlooms, or a more distant ‘ancestral’ past, as with rediscovered objects (cf. Caple 2010). These objects may thus have the potential to evoke memories or a sense of time (Rowlands 1993), or alternatively a sense of the mythical and mysterious (Ferris 2012: 84ff.). Given the temporal depth of many landscapes and sites, representing repeated episodes of occupation, interaction and engagement with a wide variety of spaces and places (Gosden and Lock 1998; Ingold 1993), it is inevitable that residual material should be encountered, and it is not suggested here that all old or ancient objects in prehistory should be seen as significant. However, it is the critical appraisal of the treatment of this material and the nature of its redeposition that allows us to infer significance. In other words, interpretations of already old objects in prehistory rely on understanding and appraisal of: firstly, their context referring both to the immediate context (e.g. a pit, a ditch, a burial), and the wider setting (e.g. the overall site or broader landscape); and secondly their condition, pre-depositional treatment and overall potential history of an object (cf. Knight forthcoming). These are crucial aspects to consider when investigating multi-period hoards. Recognising ‘out-of-time’ objects in hoards In his study of Bronze Age metalwork found in Iron Age contexts, Hingley referred to already old objects as ‘out of their time’ (2009: 143), abbreviated here to ‘out-of-time’ objects (following Knight forthcoming). This terminology avoids the loaded implications that more functional attributions such as ‘heirloom’ might imply and allows a neutral starting point (see Knight et al. this volume). The methodology for recognising out-of-time objects in later hoards has been largely maintained here following my original study of southern England to enable comparison between the two studies. This approach relies on the known relative typologies of Bronze Age metal objects, with ‘out-of-time’ objects primarily defined as: any object identified in a context that is two or more metalworking phases later than the expected relative typological sequence Knight forthcoming Established Bronze Age metalwork typo-chronologies have been developed for many of the main objects types, including axeheads (Needham 2018; Schmidt and Burgess 1981), spearheads (Davis 2012; 2015), and dirks, rapiers and swords (Burgess and Gerloff 1981; Colquhoun and Burgess 1988). Now increasingly underpinned by absolute dating techniques (e.g. Needham 2015), it is possible for these artefact types (and others) to be assigned a relative date with some precision. This overall methodology allowed potential case studies to be identified that could then be further interrogated for authenticity. 20 Matthew G. KniGht: DOubtful assOciatiOns? assessinG bROnze aGe ‘Multi-PeRiOD’ hOaRDs The importance of source criticism in this process cannot be overstated (cf. Bradley 1986), with objects studied first-hand where possible and original documentation consulted. A scale of likelihood, comprising Certain, Probable or Possible, was utilised based on: the security of the context and association, influenced by the accuracy of the records kept at the time of discovery; the known object history post-recovery; and the likelihood that incomplete objects definitely represent an older artefact Knight forthcoming This method allowed 45 possible or certain case studies to be recognised and analysed from southern England, 41 of which were considered to be multi-period hoards. Sources consulted included museum catalogues, published corpora, excavation reports and online databases to ensure as many possible instances were recognised. This overall process has been followed here for the regions of northern England, Scotland and Wales, resulting in the identification of multi-period hoards. Key catalogues and corpora from the relevant areas included volumes of the Prähistoriche Bronzefunde (e.g. Davis 2012; 2015; Schmidt and Burgess 1981); Burgess’ (1968) survey of metalwork from northern Britain; Coles’ (1962; 1966; 1971) papers cataloguing Scottish Bronze Age metalwork; museum records at the National Museum of Scotland; and unpublished treasure reports at Amgueddfa Cymru – National Museum Wales. However, the identification process has been adapted here to incorporate some multi-period associations that may have only been one metalworking phase apart or may represent the extended use of older objects. This approach provided a better understanding of typo-chronological aspects and the nature of the association, as well as recognising objects with a prolonged circulation, as will be discussed later in the paper. An advantage of this extended survey is that we can also begin to identify broader trends in object associations. In turn, this might strengthen the possibility that some ‘doubtful’ associations are in fact genuine. The case studies Two certain instances, two probable, and seven possible multi-period hoards were identified during this study (Table 2.1). Multi-period hoards that were deemed too uncertain were largely excluded here, though one example from Greyfriars Church (Dumfries and Galloway), has been noted at the end of the appendix. All multi-period hoards presented here comprise only one or two out-of-time objects alongside a variety of later objects. The out-of-time objects are represented mostly by axeheads, spearheads and dirks or rapiers; other artefact types do occur but they are the exception (Figure 2.1). Most date to the Middle Bronze Age, having been found alongside Late Bronze Age objects, predominantly socketed axeheads. A notable exception is the possible association from Islay which contains an Early Bronze Age halberd alongside Middle and Late Bronze Age objects, typologically spanning the whole Bronze Age. Details of the case studies are presented in the appendix, but aspects of the depositional contexts and condition of the objects are presented here. Distribution and deposition The associations were recovered from a variety of areas across Scotland, Wales and northern England (Figure 2.2). However, as many of the finds derive from nineteenth-century discoveries, details of the exact findspots are scarce. The findspots from Duddingston Loch (Midlothian), Corsbie Moss (Scottish Borders), Wester Galcantray (Highland), Penllyn (Vale of Glamorgan), and Mawr Community (Swansea) can be located, whilst approximate locations can be posited for the Fell Lane (Cumbria), Smalley Bight (West Yorkshire), Callander (Stirling) and Four Mile Bridge (Anglesey) hoards. The remaining findspot locations can only be estimated. Nonetheless the findspot information available allows us to comprehend certain features of depositional location for each association. 21 Site Out-of-Time Object(s) (1) Kincardine, Gr.IV dirk Abernethy, Highland Relative date of Associated objects OoT object(s) MBA c. 1275–1150 BC Likely date of deposition 2 socketed LBA axeheads, Types c. 1150–800 BC Highfield & Gillespie Context Likelihood Key References Found together under a granite boulder. Certain O’Connor and Cowie 1995: 355–357 22 (2) Mawr Community, Swansea Gr.IV rapier frag MBA c. 1275–1150 BC 2 axehead fragments & 2 casting jets LBA c. 1150–1020 BC Found while metaldetecting. Certain Knight and Gwilt 2017 (3) Callander, Stirling Gr.III rapier Gr.IV rapier Type 7C side-looped spearhead MBA c. 1400–1150 BC Socketed axehead, Type Portree LBA c. 1150–1020 BC Found together in the Callander area between 1790 and 1820. No further details. Probable (UA) Burgess 1968: 22, 38–39; Davis 2012: 114, No. 693; Museum Records; Schmidt and Burgess 1981: 186, No. 1064 (4) Duddingston Loch, Midlothian Rapier fragment MBA c. 1400–1150 BC c.44 objects incl. broken swords, spearheads and a vessel ring handle LBA c. 920–800 BC Found while dredging Duddingston Loch in 1778. Probable (UA) Callander 1923: 360–364 (5) Corsbie Moss, Scottish Borders Type 3A side-looped spearhead MBA c. 1500–1400 BC Type Wilburton sword & sword chape (destroyed) LBA c. 1150–1020 BC Found at around the same time during drainage operations in a peat bog. Possible (UA) Davis 2012: 47, No 77; Proceedings of the Society of Antiquaries of Scotland 1921: 17 (6) Fell Lane, Penrith, Cumbria Type 6E side-looped spearhead MBA c. 1500-1275 BC Socketed axehead, Type Yorkshire LBA c. 1020–800 BC Apparently found together while building a house 1883–1893. The socketed axe is now missing. Possible (UA) Bronze Age Card Index; Burgess 1968: 19; Clough 1969: 14 (7) Four Mile Bridge, Anglesey, Gwynedd Type 9A basal-looped spearhead MBA c. 1400–1150 BC Type 11A pegged spearhead LBA c. 1150–1020 BC Found three metres apart on two separate occasions. Possible (UA) Lynch 1991: 236 EBA c. 2300–2150 BC MBA c. 1500–1275 BC 2 socketed axeheads, 1 Type Meldreth LBA c. 1020–800 BC Reportedly a hoard. No further details. Possible (UA; IC) Coles 1962: 134; Proceedings of the Society of Antiquaries of Scotland 1882: 409 (8) Islay, Argyll Halberd and Bute Palstave-adze Type 6B side-looped spearhead Objects Of the Past in the Past Table 2.1: A summary of the multi-period hoards from northern England, Scotland and Wales. Key: E/M/LBA = Early/Middle/Late Bronze Age. IC = Insecure Context referring to lack of discovery detail; UA = Uncertain Association referring to dubious recovery circumstances. Possible (UA; IC) Clark et al. 2017: 27, 36, 51; Walker 1972: 117–118. Supposedly found together in a garden. See details in appendix. LBA c. 1000–800 BC MBA c. 1400–1275 BC Type Lissett short-flanged axehead (11) Wester Galcantray, Highland 2 socketed axeheads, 1 Type Portree Schmidt and Burgess 1981: 136; Varley 1977; Walker 1939: 15 Possible Hoard of axeheads recovered during dredging operations from River Calder. LBA c. 920–800 BC) MBA c. 1400–1150 BC 2 axeheads (short-flanged axehead and palstave) (10) Smalley Bight, Stanley Ferry, West Yorkshire 8 socketed axeheads, Types Everthorpe and Yorkshire Gwilt et al. 2015 M–LBA c. 1275–1020 BC Poss. MBA Knife or rapier frag (9) Penllyn, Vale of Glamorgan 2 palstave butt fragments, spearhead frag & plate frag Found in the same field Possible while metal-detecting in December 2013 and January 2014. Matthew G. KniGht: DOubtful assOciatiOns? assessinG bROnze aGe ‘Multi-PeRiOD’ hOaRDs Figure 2.1: Frequency of different out-of-time object types found in Late Bronze Age multi-period hoards from northern England, Scotland and Wales. The Four Mile Bridge spearheads, for instance, were found near the coast of Holy Island, Anglesey. The Mawr Community hoard, meanwhile, was found overlooking a river valley (Knight and Gwilt 2017), whilst the Wester Galcantray association was recovered from low-lying ground in the Nairn river valley and the Callander hoard likely came from the River Teith valley. Although the exact find location of the Kincardine hoard cannot be identified, it is worth highlighting its deposition under a boulder, close to the River Spey on the northern edge of the Cairngorms, specifically Creag Mheadhonach. The link with multiple natural features was likely significant. Three of the multi-period hoards are from wetland locations: the three objects from Corsbie Moss were recovered from a peat bog; the Duddingston Loch assemblage was recovered from a loch at the base of Arthur’s Seat, the peak of an extinct volcano system, in Edinburgh; and the Smalley Bight hoard was dredged from the gravel beds of the River Calder. The association of finds with rivers, bogs and lakes is well-known in prehistory (Bradley 1998; 2000: 47– 63) and the significance of certain places is explored further below. The condition of the objects In as many cases as possible, the completeness and condition of the out-of-time and associated objects was assessed, with evidence of use-wear and pre- and post-depositional damage recorded. Although this varies, overall most hoards either contained complete or fragmentary out-of-time objects—that is either the objects were 23 Objects Of the Past in the Past undamaged or were represented only by fragments (i.e. where less than 25% of the object survives). The hoards from Penllyn, Mawr Community and Duddingston Loch, for instance, all contained fragments of rapiers alongside other broken and fragmentary later metalwork. Likewise, the two rapiers in the Callander hoard are broken, though about 50% of these objects survives, alongside a complete socketed axehead and spearhead (Figure 2.3). The objects in the Islay hoard show varying degrees of damage, some of which was probably deliberate, but are all largely complete. The earliest object in this association (a halberd) is also one of the most complete. Much of the Fell Lane spearhead survives with a worn blade and broken side-loops, alongside a complete but worn socketed axehead. Similarly, the dirk from Kincardine (Figure 2.4), the earlier axeheads from Smalley Bight, and the spearheads from Corsbie Moss and Four Mile Bridge are both complete alongside other complete objects. Figure 2.2: A map of Britain and Ireland showing the distribution of the case studies described in this paper (numbers correlate with Table 2.1 and the appendix). Case studies are plotted according to the likelihood of truly representing an out-of-time deposition. The locations of multi-period hoards from southern England are also plotted for comparison (information from Knight forthcoming). The Corsbie Moss spearhead and sword were cleaned post-recovery making the interpretation of use-wear problematic, but both appear to have been prepared for use; in the case of the sword this is reinforced by the presence of a chape (lost after recovery), implying deposition in a scabbard (Figure 2.5). Hilt marks are preserved in the patina of the Callander rapiers, suggesting they were hilted at the time of deposition, while chips and nicks along the blades could be linked to their use; the socketed axehead in the same hoard also shows some signs of use-wear. The socketed axeheads in the Kincardine hoard bear evidence of working and sharpening and would certainly have been functional prior to deposition, whilst the earlier dirk in the same hoard is very worn, which could be use-related (O’Connor and Cowie 1995: 355ff.). The Fell Lane spearhead and the older of the two Four Mile Bridge spearheads were seemingly subjected to extensive resharpening and wear over a long period of time (British Museum Card Index; Lynch 1991: 236), apparently having had a long use-life. From the available illustrations and images, it is not possible to determine the extent of use-wear present on the Middle Bronze Age axeheads from Smalley Bight, but at least two of the Late Bronze Age socketed axeheads show signs of sharpening striations, hammering and blunting of the cutting edge, suggesting this was also a hoard of used objects from different periods or accumulated over time. 24 Matthew G. KniGht: DOubtful assOciatiOns? assessinG bROnze aGe ‘Multi-PeRiOD’ hOaRDs Figure 2.3: The Callander hoard. Illustration: Alan Braby © National Museums Scotland 25 Objects Of the Past in the Past Figure 2.4: The Kincardine hoard. Photo: M. Knight, courtesy of the Highland Folk Museum The rapier fragments from Mawr Community and Duddingston Loch are too incomplete and worn to definitively identify indicators of preparation or use. In both cases, the rapiers broke in antiquity and were deposited alongside other broken and worn-out material (Figure 2.6). Some swords and spearheads in the Duddingston Loch assemblage show evidence of use-related damage, as well as evidence of deliberate burning, bending and breaking prior to deposition. By contrast, the halberd from Islay is possibly unfinished, but has bowed and torn edges and a bent tip; it was deposited alongside other damaged objects, including a spearhead and two socketed axeheads. Although this damage is complicated by post-depositional actions such as cleaning, when the hoard was first reported it was recorded that all the objects were ‘much injured’ (Proceedings of the Society of Antiquaries of Scotland 1882: 409) and it seems that at least some of the damage was deliberately inflicted prior to deposition. The multi-period hoards presented here thus represent complex accumulations of objects, treated in a variety of ways and incorporated into a variety of depositional practices. Having recognised this, we can now consider several key aspects that enhance the interpretation of these deposits. 26 Matthew G. KniGht: DOubtful assOciatiOns? assessinG bROnze aGe ‘Multi-PeRiOD’ hOaRDs Figure 2.5: The Corsbie Moss spearhead and sword. Photo: M. Knight © National Museums Scotland Figure 2.6: A selection of worn and fragmentary blades from Duddingston Loch. The Middle Bronze Age rapier is illustrated bottom right. Illustration: Marion O’Neil © National Museums Scotland 27 Objects Of the Past in the Past Multi-period hoards as typological aids Hansen (2016: 194–197) has recently emphasised the importance of recognising the temporal depths of European Bronze Age hoards. He stresses the possibility that metalwork deposits were accumulated over long periods of time and that interpreting the deposition of hoards as a single action in a moment of time is not necessarily accurate (cf. Needham 2007). The Piller hoard (Austria), for instance, contains objects spanning 300 years, as indicated by the relative typologies and the use-wear seen on some objects (Hansen 2016: 195). Obviously, this has bearing on relying on the typology of objects as a method for dating the deposition of hoards and is emphasised by the fact that authors have tended to doubt multi-period associations or regard typological anomalies in hoards as ‘scrap’ (e.g. Britton 1963: 270; Eogan 1983: 188–189, No. 17; Grinsell 1970: 32). This should not be taken as a call to revise typologies (cf. Hingley 2009: 148f.), but rather as a need for caution when considering multi-period hoards. As Hansen puts it: the [traditional] dating of hoards stands tendentiously in a circuitous argument around the duration of certain types of bronzes, which in turn is determined by the time of the hoard’s emplacement as defined by these types Hansen 2016: 196–197 Whilst for some hoards there can be little argument that such anomalous objects occur from a distinctly earlier chronological period, this becomes less certain where hoards contain objects that have a long duration of production or circulation. For instance, there are multiple Ewart Park phase hoards (920– 800 BC) that contain objects from the preceding Wilburton to Blackmoor phases (c. 1150–920 BC). Could these hoards assist us in identifying objects that are late Wilburton–Blackmoor/early Ewart Park developments? This question is particularly highlighted by a recent hoard from Llancarfan, Vale of Glamorgan, Wales. In this hoard a fragment of a St Nazaire sword dating to the Blackmoor phase (c. 1020–920 BC) was associated with material more typical of the succeeding Ewart Park phase, potentially indicating that the fragment may have been quite old when deposited (Gwilt 2006). In this case, however, there has been debate about whether this typologically earlier sword fragment may in fact be used to date the hoard to the earlier metalworking phase and refine our dating of objects conventionally placed in a later metalworking phase (Brandherm and Moskal del-Hoyo 2014: 21–22, footnote 45; Gwilt 2006). The possibility that some multi-period hoards might in fact assist with our understanding of typologies is rarely considered explicitly though. Indeed, some multi-period hoards have been discounted as a result of typological anomalies that are less common. The Callander hoard, comprising two rapiers, a spearhead and a socketed axehead, has previously been considered a doubtful association of objects, due to the lack of additional contextual information and the long typo-chronological span of the objects (c. 300 years) (Coles 1962: 34, 134; though see O’Connor and Cowie 1995: 362, note 5). However, all four objects have a consistent patina and were specifically claimed to have been found together, whilst other finds acquired from Callander at the same time were noted as single finds (see Appendix; Proceedings of the Society of Antiquaries of Scotland 1958: 463). The two rapiers in the hoard date to the Middle Bronze Age (c. 1400–1150 BC), whilst the socketed axehead is a type that was produced and used throughout the Late Bronze Age (c. 1150– 800 BC) (Burgess and Gerloff 1981; Schmidt and Burgess 1981: 186–187). Axeheads like the Callander example have been found in metalwork hoards dating to c. 1150–1020 BC, and indeed Schmidt and Burgess suggest that the Callander hoard could be a genuine association, representing the origins of this type of axehead around 1150 BC (1981: 186–187). Moreover, the Kincardine hoard also includes earlier forms of socketed axeheads alongside a typologically late dirk, which O’Connor and Cowie (1995: 361) use as evidence for the extended use of dirks and rapiers, as well as the earlier production of certain socketed axeheads (Figure 2.7; see also Needham 2017: Appendix 1). This strengthens the likelihood 28 Matthew G. KniGht: DOubtful assOciatiOns? assessinG bROnze aGe ‘Multi-PeRiOD’ hOaRDs Figure 2.7: The expected typological durations of the objects in the Kincardine and Callander hoards (following information in Burgess and Gerloff 1981; Davis 2012; Schmidt and Burgess 1981) Table 2.2: A list of multi-period hoards from Britain containing dirks and rapiers Site No of rapiers/ dirks Associated objects Brading Road, Brighton, East Sussex 1 1 Late-type palstave Callander, Stirling 2 1 MBA spearhead and 2 LBA socketed axeheads Duddingston Loch, Midlothian 1 fragment Blackmoor/Ewart hoard of 44 objects Gorleston-on-Sea, Great Yarmouth, Norfolk 2 fragments 1 MBA palstave fragment & Ewart Park hoard of 122 objects Kincardine, Abernethy, Highland 1 2 socketed axeheads Lanherne, St. Mawgan, Cornwall 1 1 MBA palstave and LBA hoard of unknown number of objects, mostly socketed axeheads Mawr Community, Swansea 1 fragment 2 axehead fragment and 2 casting jets that Callander does indeed represent a genuine association and it is possible that the Kincardine and Callander hoards represent associations of old styles of objects with new forms deposited around 1150– 1100 BC. Across Britain, there are seven instances of dirks or rapiers in Late Bronze Age contexts (Table 2.2). Whilst some of these must inevitably be considered residual fragments (e.g. Duddingston Loch and Mawr Community), others represent situations like Callander. In particular, the fragmentary Middle Bronze Age rapier found with a Late Bronze Age palstave at Brading Road, East Sussex (Curwen 1954: 216f., No. 17) probably represents the extended circulation of the rapier into the Late Bronze Age. To reiterate, this is not a call to redefine established typologies for certain object types, but rather I wish to emphasise that some multi-period hoards might in fact open the possibility for identifying early and late developments in specific object forms, as well as allowing the opportunity to recognise some objects that were in circulation for longer than typical and the reasons for this. Indeed, Needham (2017: 151) suggests that the reason the Callander and Kincardine hoards contain anachronistic objects may be linked with a reduced supply of materials and thus the functional need to keep older objects for longer. This is further likely when we consider that there is limited evidence for the widespread development and adoption of swords in Scotland in the earlier part of the Late Bronze Age; Coles (1966: 114) and Burgess (1968: 23) have both argued that dirks and rapiers may in fact have a prolonged tradition in Scotland compared with southern Britain. It is usual to automatically dismiss collections of objects like Callander based on preconceived notions of typological chronologies, but with a consistent condition of objects and an appreciation of flexibility within these chronologies, we can enhance how we might 29 Objects Of the Past in the Past interpret some multi-period associations and their implications for the social structures that were in place. Of course, this typological flexibility means that Callander and Kincardine actually represent contemporary associations and are not in the truest sense ‘multi-period’. Out-of-time but not out of place We know that some locations held significance to Bronze Age communities for depositing objects. Depositional practices have been linked with certain landscapes and functioning as a socio-political action for managing the inhabited world (e.g. Bradley 1998; 2017; Fontijn 2002; Levy 2010; Needham 1988; 2007). I have previously suggested that the deposition of multi-period hoards in high locations and in or near watery locations may have been part of a method for communities to structure the landscape and establish or legitimise claims to a place (Knight forthcoming). The multi-period hoards presented in this paper follow a similar pattern. Depositions in rivers and river valleys, bogs, lakes, and islands all fit within expected Bronze Age practices. Concentrations of metalwork depositions along river valleys in Britain have been noted in southeast Wales (Gwilt 2004: 121f.), the Thames (York 2002), south-east England (Yates and Bradley 2010a), the eastern England Fenland (Yates and Bradley 2010b), and north-east England (Poyer 2015). The multi-period hoards from Mawr Community (Swansea), Wester Galcantray (Highland), Penllyn (Vale of Glamorgan), Smalley Bight (West Yorkshire), and possibly Callander (Stirling) are thus located in areas one would expect to find hoards and conform to expected depositional practices. Moreover, older objects were being incorporated into contemporary ideologies. Depositions in these locations may have been part of the process for legitimising place, as is often seen in Middle–Late Bronze Age settlements with deposits in liminal locations including doorways and boundary ditches (Brück 2006: 298f.). The inclusion of already old objects in hoards may have assisted in legitimising place as symbols of ancestral claims to the land (cf. Hingley 2009; Knight forthcoming). The same could be argued for the Kincardine dirk and socketed axeheads. The deliberate deposition under a boulder on or near the slopes of the Cairngorms mountain range near a river suggests a place to which significance could have been attributed. Indeed, the depositional location marked by a boulder raises the possibility that this site could potentially be revisited and later objects may have been added. It thus follows that it is important to recognise that hoards and associations may not have been static occasions of deposition, rather they represent accumulations over time (cf. Hansen 2016; Needham 2007: 280–281; see below). This explanation has been proffered for the Corsbie Moss association; whilst some authors have accepted it as genuine (Colquhoun and Burgess 1988: 53)—and indeed the original account suggests the objects were found together or at least at the same time—Davis argued this actually represents a revisited area of deposition (2012: 52). This may be supported by the recovery of an Earliest Iron Age socketed axehead over a decade later from the same area. Meanwhile, all the axeheads in the Smalley Bight hoard reportedly came up at the same time, but as it was dredged from the River Calder, it has tended not to be considered a genuine association (Schmidt and Burgess 1981: 136). This scepticism may result from dredging activity elsewhere that has produced numerous objects from several periods, and from well-known locations that were revisited for depositional activities, such as bogs and rivers. Nonetheless, the multi-period accumulations infer the long-lived importance of places to the societies depositing metalwork, no doubt part of a wider set of beliefs and practices. Places of deposition and the objects deposited may have been remembered and repeatedly or intermittently revisited resulting in the multi-period associations we find today. The earlier rapier in the Duddingston Loch assemblage appears to have less to do with establishing claims to land and more to do with the accumulation of objects over time and ongoing contemporary practices in a significant landscape. This hoard has recently been interpreted as part of a wider practice 30 Matthew G. KniGht: DOubtful assOciatiOns? assessinG bROnze aGe ‘Multi-PeRiOD’ hOaRDs of sacrificing of weapons in watery locations (Mörtz 2018); therefore, the role of the out-of-time object could be linked to the function of the rapier as a weapon. The antiquated rapier in the Duddingston Loch assemblage was no doubt recognised as a weapon of a bygone era, typologically dating to the Middle Bronze Age (c. 1275–1150 BC), and the multiple styles of weapons spanning the Late Bronze Age (c. 1150–800 BC) suggest an accumulation of weapons over an extended period prior to deposition. This need not necessarily have been a conscious process of active collecting, but it is likely each object had been in circulation for variable amounts of time and these were not all the result of a single production event. Thus, each object had its own biography and was individually significant, as well as being significant as part of the overall assemblage. A note of caution must be applied, however, as the collection of metalwork from Duddingston Loch was dispersed upon recovery with some pieces thought to have been given to King George III, and only an unknown proportion of the original collection now survives (Cowie and O’Connor 2007: 319). The seemingly complete condition of some of the objects recorded in drawings, but now mostly lost, led Cowie and O’Connor (2007: 319) to suggest that what may be represented by the assemblage are multiple episodes of deposition in the loch. Nonetheless, the deposition of deliberately destroyed metalwork in Duddingston Loch at the base of Arthur’s Seat is significant, especially as two complete Late Bronze Age swords were deposited on the slopes of Arthur’s Seat itself (Coles 1962: 116). Moreover, at Grosvenor Crescent, about two and a half miles away, a hoard comprising fourteen or fifteen swords, a socketed axe, a ring, a mount and a pin was excavated in 1869, whilst a sword and chape were recovered at Gogarburn, also in Edinburgh (Coles 1962: 118f.). This suggests the Edinburgh landscape may have been an area in which it was significant to deposit martial equipment in a variety of ways and if the out-of-time rapier represents a genuine association it was incorporated into these later practices. Objects past with objects present The possibly extended circulation of the Duddingston rapier raises important questions about how we might recognise the length of time an object was in use and, by extension, the nature of the relationship between the object and its owner(s). In the multi-period hoards, such insights can be gained by analysing the condition of some of the out-of-time objects, and inferring aspects of the potential biographies of those objects. The worn nature of the Fell Lane and Four Mile Bridge spearheads emphasises that these were utilitarian objects that were used and reused over long periods of time. By the time of deposition, a substantial portion of each blade had worn away through use and resharpening, and, in the case of the Fell Lane spearhead, the loops had broken. Other Middle Bronze Age spearheads are known to have received similarly extensive use-lives, such as an example from Merton, Oxfordshire (O’Connor 1979; see also Davis 2012: Pl.36), and at Shrubsoles, Kent, surviving haft remains indicate a Middle Bronze Age spearhead may have been repeatedly re-hafted over several hundred years prior to burial (Taylor 2003: 42–43). If these spearheads were used over long periods of time, for whatever purpose(s), they were likely cared for and curated; in this sense they may have become inalienable objects (Weiner 1992), intrinsically linked with an owner or community, and passed down over time as a treasured object. The eventual deposition of the Fell Lane and Four Mile Bridge spearheads alongside later objects and in revisited landscapes may thus reflect the end of each spearhead’s perceived usefulness or mnemonic role. It is further interesting that the objects associated with each spearhead seem to also have had an extended period of use before deposition. Both the later Four Mile Bridge spearhead and Fell Lane socketed axehead are worn. The same is true of the Kincardine dirk and socketed axeheads. Likewise, in the groups of fragmentary objects from Duddingston Loch and Mawr Community, the out-of-time objects are in a worn-out and broken condition that matches the other objects with which they were 31 Objects Of the Past in the Past deposited. All of the objects in the Islay hoard show some signs of deliberate damage, including plastic deformation (bending and crushing), breaking, and edge damage, suggesting that if this does represent a genuine hoard, the Early, Middle and Late Bronze Age artefacts were all decommissioned, perhaps as part of one event. Where there is no obvious differentiation in the treatment of older objects or later objects, this may infer that, although older, the out-of-time objects warranted no alternative treatment and had been incorporated into contemporary customs and practices. Certainly, in the cases of spearheads and axeheads, the function of these objects probably would have been recognised and even if they were known to be objects with a past, there was no need to treat them differently. This was not always the case though. The hoard from Callander contains a mix of complete later objects and broken out-of-time objects; elsewhere at Yattendon (Berkshire), Shoebury (Essex), and Stoke Ferry (Norfolk), the out-of-time objects were the most complete, suggesting a level of care that was not afforded to typologically later implements which were deliberately broken before deposition (Hawkes 1954: GB.8; Knight forthcoming). Inevitably, that we have been able to recover these hoards archaeologically, means all the objects were eventually deposited and removed from circulation, indicating either that their significance was no longer recognised or that their removal was important for another social strategy. Comparisons with out-of-time objects and multi-period hoards from southern England Finally, it is appropriate to directly compare this survey with the previous survey of out-of-time objects from southern England. Forty-five certain, probable, and possible instances of earlier bronze objects found in later Bronze Age contexts were identified predominantly composed of multi-period hoards, but also four from settlement contexts (Knight forthcoming). The present survey thus brings the overall total of sites for Britain up to 56.2 The most immediate observation is the contrast in numbers of associations involved (see Figure 2.2). Almost four times as many instances of multi-period hoards and out-of-time objects were identifiable in southern England. This is in part skewed by the larger number of hoards from southern England, and particularly those of the carp’s tongue complex in south-east England, which have a greater number of older worn-out objects than the rest of the country (Knight forthcoming). In almost all cases the multi-period hoards comprise only one or two out-of-time objects, and only two periods of time (Davies this volume). Across Britain overall there is a concentration of Bronze Age multi-period hoards in river valleys, which reflects trends in depositional practices generally. None of the hoards from southern England were recovered from lakes or bogs, whereas in Scotland there is at least one from each (Duddingston Loch and Corsbie Moss respectively).3 In southern England, two out-of-time depositions were identified in structured deposits on the Late Bronze Age settlements at Shrubsoles Hill and Iwade, both Kent (Knight forthcoming). The latter is known to have been an island during the Bronze Age and it was posited that the proximity of these two deposits may have been linked to commonly held ideas within an area. With the possible hoard from Four Mile Bridge we might also speculate about the importance of a multiperiod deposit on an island.4 Islands may have been significant places in the Bronze Age (Bradley 2000) and the accumulation of old and new material may have been part of the process of revisiting certain places and areas over time. A similarity across Britain is the predominance of Middle Bronze Age objects alongside Late Bronze Age material. As has been posited in this paper, some of these could represent extensions of the traditional typological span, but this is less clear for those in southern England. The predominance of Middle It is important to recognise though that no attempt has been made in the present paper to identify occupation sites with outof-time objects from northern England, Scotland and Wales, though such sites exist in these regions. One example is a possible Middle Bronze Age gold fragment found in a Late Bronze Age pit at Llanmaes (Vale of Glamorgan, Wales) (Gwilt et al. 2016: 302). 3 Admittedly, this observation in part simply reflects the geography of each region. 4 To this we might also add the Islay hoard, though we must accept that given the size of the island coupled with the lack of details surrounding its recovery, the objects may have come from any number of topographical locations. 2 32 Matthew G. KniGht: DOubtful assOciatiOns? assessinG bROnze aGe ‘Multi-PeRiOD’ hOaRDs Bronze Age artefacts as out-of-time objects no doubt reflects the increased production and deposition of metal objects during that period, increasing the chances that material would either circulate for longer periods or have a higher chance of rediscovery. In the present study, only the Islay hoard contained an Early Bronze Age object, though this association cannot be considered certain. It may be strengthened, however, by the Stoke Ferry hoard, Norfolk, which contained an Early Bronze Age copper halberd alongside broken Late Bronze Age swords and spearheads (Hawkes 1954: GB.8; Lawson 2018: 37–38). If this is a legitimate association, the Stoke Ferry halberd likely represents a rediscovered object and may have been recognised for its age; the Islay halberd could be a similar situation. In southern England it was possible to pick out certain instances where the deposition of out-of-time objects may have been linked to memory creation or, alternatively, forgetting. The destruction and deposition of the Milsoms Corner shield, Somerset, for instance, was linked to ending the ‘life’ of the object and the associations it held with the local community through its age (Knight forthcoming). In northern England, Scotland and Wales such instances could not be conclusively identified and thus the theme of memory has not been explored in this paper. However, it is possible to argue that certain objects, such as the rapier fragments from Mawr Community and Duddingston Loch or the spearheads from Four Mile Bridge and Fell Lane, may represent the retention of some objects over long periods, even to the point that they cease to be functional (e.g. the edge is so worn it can no longer be resharpened). I hesitate to refer to them as heirlooms, as their condition does not suggest the care and veneration expected of heirloom objects (cf. Lillios 1999). However, that such things may have been important to Late Bronze Age communities can be derived from the Earliest Iron Age hoard from Poolewe, where a Late Bronze Age ornament was seemingly retained and worn out over several generations of circulation before deposition with later axeheads (Knight 2019: 13). For the worn and used out-of-time objects in the multi-period hoards under discussion here, we may speculate that these objects acquired certain inalienable qualities that were linked with a known past, accrued over an extended period of possession, circulation and use; in this way depositing objects and hoards may have been mnemonic practices for managing social strategies (cf. Levy 2010: 131ff.). Furthermore, the potential that Smalley Bight, Four Mile Bridge and Corsbie Moss are not true associations, but are instead revisited depositional sites, implies that the locations of deposition were remembered, and material was added over time. Final thoughts So where does this leave us concerning Bronze Age multi-period hoards? A number of case studies previously considered ‘doubtful’ have been highlighted, expanding the known corpus of possible, probable or certain multi-period hoards containing Bronze Age out-of-time metal objects to 52. It should be clear by now that although not a frequent occurrence, such hoards are widely distributed and occurred in a variety of ways and for a variety of reasons that fit within the known Bronze Age hoarding practices. This is not intended to be the final word on the topic, but by illustrating that former assumptions require interrogation, it is hoped that more thorough source criticism will be undertaken in the future to at least explore the possibility that the association of objects that do not fit expected typo-chronological frameworks may in fact be genuine, or at least plausible. The Bronze Age multi-period hoards now known from Britain offer the opportunity to explore theoretical themes, such as mnemonic practices, and can also be used to enhance our understanding of typologies and the relationships between people, objects and the landscapes in which they were deposited. Acknowledgements This paper expands on work I began during my Masters in 2014 on multi-period hoards in southern England and submitted for publication in 2015 (Knight forthcoming). The present paper is the result of numerous fruitful conversations with many colleagues and friends that continue to stimulate my 33 Objects Of the Past in the Past thoughts on this topic. I would like to particularly thank my co-editors for their firm but fair editing that brought this paper down to an (almost) manageable size, as well as Catriona Gibson for offering comments on what she endearingly calls ‘Out-of-Time Objects 2.0’, and Robyn Raxworthy for her continued patience listening to me talk about this topic. Thanks are also due to Rachel Chisholm at the Highland Folk Museum for accommodating my visit to see the Kincardine hoard, to Adam Gwilt for his help with the Welsh hoards, and to Katherine Baxter at Leeds Museums and Galleries and Dave Evans at Wakefield Museums & Castles for providing valuable information on the Smalley Bight hoard. Finally, Trevor Cowie has, as always, been a mine of information relating to the Scottish material and this paper is much stronger for his input and thought-provoking comments. Any remaining errors are my own. Appendix This appendix presents details of the 11 multi-period hoards identified during this research from northern England, Scotland and Wales (summarised in Table 2.1). The importance of source criticism was highlighted in text and this is emphasised by the thorough historiographies of some of the hoards outlined here. The likelihood that each hoard is indeed ‘multi-period’ is presented on a scale of Certain, Probable and Possible and the hoards are listed alphabetically within each certainty. The distribution is presented in Figure 2.2. Certain multi-period hoards (1) Kincardine, Abernethy, Highland (Inverness-shire), SCOTLAND5 Around 1873, a Middle Bronze Age dirk and two Late Bronze Age socketed axeheads were found under a granite boulder at Kincardine, Abernethy (O’Connor and Cowie 1995: 355). Kincardine (or Kinchardine) was a parish in the west of what is now Abernethy and can be identified on OS maps until 1874, though is now known as West Croftmore; the exact findspot location cannot be identified but it was likely in the vicinity. The objects were donated to Am Fasgadh in 1951 and now reside in the Highland Folk Museum, Kingussie (O’Connor and Cowie 1995: 355). The dirk is Burgess and Gerloff’s (1981) Group IV type, broadly dating to the Penard metalworking phase of the Middle Bronze Age (c. 1275–1150 BC), whilst the socketed axeheads are of a type datable to the Wilburton phase (c. 1150–1020 BC). Although this group has traditionally been considered to not be associated based on the typological disparity of the objects (e.g. Coles 1966: 15; Schmidt and Burgess 1981: Nos 1025 and 1141), they are now considered to represent a legitimate association (Needham 2017: Appendix 1; O’Connor and Cowie 1995: 357). (2) Mawr Community, Swansea, WALES Two socketed axehead fragments, two casting jets and a rapier fragment were found while metaldetecting in the Community of Mawr, Swansea, in 2015 (Knight and Gwilt 2017). These objects were dispersed over an area of about three square metres but are considered to be a certain association (Knight and Gwilt 2017). The axeheads and the casting jets date to the Late Bronze Age, whilst the rapier fragment is more typical of the Middle Bronze Age. The modern local authorities for the Scottish sites are given here with historic counties in brackets where appropriate to allow cross-referencing with older sources. 5 34 Matthew G. KniGht: DOubtful assOciatiOns? assessinG bROnze aGe ‘Multi-PeRiOD’ hOaRDs Probable multi-period hoards (3) Callander, Stirling (Perthshire), SCOTLAND Two rapiers, a socketed axehead and a spearhead with asymmetrical side-loops were supposedly found together in the Callander area prior to 1830 and were purchased by what was then the National Museum of Scotland in 1955 (National Museums Scotland Acc. Nos X.DQ 321–324; Proceedings of the Society of Antiquaries of Scotland 1958: 473, fig.7). Museum records note that the artefacts were purchased from George Willis, the founder of Basingstoke Museum, and prior to this, the finds were in the possession of Captain James Richard Hill MacFarlane of Lochhouses, Prestonkirk, and were found between 1790 and 1820. Three other bronze artefacts from Callander were acquired at the same time, but were noted as single finds, including a flanged axehead, a socketed axehead and a spearhead (National Museums Scotland Acc. Nos X.DC 134, X.DE 124 and X.DG 105). This paper argues it is probable the original four objects were found together. (4) Duddingston Loch, Edinburgh, Midlothian, SCOTLAND The group of objects from Duddingston Loch was recovered in 1778 while dredging for shell marl (Callander 1923: 360–364). This assemblage contains in the region of 50 metal objects, though the exact number is unknown, including swords, spearheads, a bucket ring handle and a rapier fragment. Most of the material can be dated to the Wilburton to Ewart Park metalworking phases of the Late Bronze Age (1150–800 BC), including Wilburton swords and spearheads as well as Ewart Park counterparts, though the rapier fragment broadly dates to the Middle Bronze Age, probably towards c. 1100 BC and thus would have been a century old or more at the time of deposition. Assuming a deposition date for the hoard in the early Ewart Park date (c. 900 BC), some of the Wilburton phase metalwork could also have been old when deposited. However, as this was a dredged discovery, it cannot be considered absolutely certain that the rapier does not represent an earlier deposit in Duddingston Loch. Its similarity in condition to the other weapons recovered strengthens the idea that this is a genuine association. Possible multi-period hoards (5) Corsbie Moss, Scottish Borders (Berwickshire), SCOTLAND A spearhead, sword and sword chape were recovered at the same time during drainage operations in a peat bog some time before 1854 (Coles 1962: 24, 107–108, fig.1; Proceedings of the Society of the Antiquaries of London 1856: 121). They were a foot or two below the surface (Colquhoun and Burgess 1988: 51). The chape was destroyed during recovery but was apparently metal. The surviving spearhead dates to the early part of the Middle Bronze Age (following Davis 2012), whilst the sword can be dated to the Wilburton metalworking phase (c. 1150–1020 BC). In 1866, an Earliest Iron Age socketed axehead was recovered from the same area (National Museums Scotland records; Schmidt and Burgess 1981: 242, No. 1589, recorded as ‘Corsbie Tower’), post-dating the sword by 200–400 years and the spearhead by potentially a millennium. Whilst it is possible the spearhead, sword and sword chape were once associated, the later axehead raises the alternative possibility that this was a place where multiple depositions were made over time (cf. Davis 2012: 52). All objects were donated at the same time in 1920 along with two other finds from Corsbie Moss: a Neolithic flint arrowhead and a stone spindle whorl (Proceedings of the Society of Antiquaries of Scotland 1921: 14, 16–17, 19). There is no indication that any of these were associated. (6) Fell Lane, Penrith, Cumbria, ENGLAND At Fell Lane, a Middle Bronze Age spearhead and a Late Bronze Age socketed axehead were supposedly found together, but Clough has since cast doubt on this association as he was informed that they may 35 Objects Of the Past in the Past have been found in the River Lowther in 1931–32 (1969: 14). The Bronze Age Card Index at the British Museum records that the spearhead and axehead were found together while building a house in 1883– 1893; these records are dated to 1923, thus predating the supposed discovery in the River Lowther. Moreover, the private collection in which the objects were held was documented as Lowther Street in Penrith, which may explain why it was suspected these objects came from the River Lowther. Unfortunately, the socketed axehead is now missing and the confusion surrounding the object history has been enough for subsequent authors to doubt the authenticity of the find (e.g. Davis 2012: 101); this association must remain only a possibility. (7) Four Mile Bridge, Anglesey, Gwynedd, WALES A Middle Bronze Age basal-looped spearhead and a Late Bronze Age pegged spearhead were found within three metres of each other, but on two separate occasions (Lynch 1991: 236). The earlier spearhead predates the later one by up to 400 years. (8) Islay, Argyll and Bute, SCOTLAND A group of five objects from Islay in the Inner Hebrides, including an Early Bronze Age halberd, a Middle Bronze Age spearhead and flanged chisel and two Late Bronze Age socketed axeheads, was acquired by National Museums Scotland in 1882 (Proceedings of the Society of Antiquaries of Scotland 1882: 409). However, no details of the circumstances of discovery are preserved. The typologically diverse nature of this hoard, spanning up to 1000 years, led Coles to suggest this probably represented a modern ‘collector’s hoard’ (1966: 117), rather than a genuine association. However, it remains possible these objects were indeed found together due to similarities in patina and condition. (9) Penllyn, Vale of Glamorgan, WALES This hoard comprises fragments of two Late-type palstaves, a spearhead, an uncertain ‘plate’ object and a possible knife or rapier fragment (Gwilt et al. 2015). It was found while metal-detecting in the same field on two occasions in December 2013 and January 2014. Much of the hoard indicates a depositional date early in the Late Bronze Age, probably during the Wilburton metalworking phase at the end of the second millennium BC (c. 1150–1020 BC), but the possible rapier fragment could indicate the inclusion of earlier material in a later hoard (Gwilt et al. 2015). Although the association is secure, the identification of the objects is not. (10) Smalley Bight, Stanley Ferry, West Yorkshire, ENGLAND A group of eleven bronze implements was recovered together while dredging the River Calder just below Smalley Bight farm near Stanley Ferry and presented to Leeds City Museum in 1914 (Walker 1939: 15). This hoard appears to have included Middle Bronze Age flanged axeheads and palstaves alongside Late Bronze Age socketed axeheads, but there is some confusion in the history of the objects post-recovery. Despite reporting eleven implements, Walker only lists ten, which includes ‘seven bronze looped celts… a winged celt without a stop-ridge… a looped palstave… and a bronze object with a bearded man’s head delineated upon it’ (Walker 1939: 15). Of these, he illustrates two: a socketed axehead and a palstave. Later, Varley (1977: 53–54) recorded that part of the hoard had been destroyed when Leeds City Museum was bombed during World War II and lists only six surviving axeheads, including four socketed axeheads, a Middle Bronze Age flanged axehead, and a palstave. However, none of these objects are the two illustrated by Walker. The hoard has now been traced to Wakefield Museums and Castles, where eight axeheads currently survive with the provenance ‘Smalley Bight’, including six Late Bronze Age socketed axeheads and a Middle Bronze Age flanged axehead and a Middle Bronze Age palstave. The surviving objects include the socketed axehead and palstave illustrated by Walker, but not the palstave 36 Matthew G. KniGht: DOubtful assOciatiOns? assessinG bROnze aGe ‘Multi-PeRiOD’ hOaRDs presented by Varley. A ninth fake socketed axehead is currently held at Wakefield Museums and Castles with the provenance of ‘Smalley Bight’, but it is unclear whether this is a replica of an original axehead, or simply a misattribution. Regardless, the Middle Bronze Age flanged axehead and palstave predate the socketed axeheads by several centuries and seem to have been part of the original group as described by Walker. Both the surviving flanged axehead and the palstave are types that can be broadly dated to the Taunton metalworking phase (c. 1400–1275 BC), whilst the socketed axeheads are more typical of the Ewart Park phase (c. 920–800 BC) (following Schmidt and Burgess 1981). (11) Wester Galcantray, Highland (Nairnshire), SCOTLAND A particularly convoluted case study concerns three axeheads from Wester Galcantray (repeatedly misspelled as ‘Golcantry’), Highland (National Museums Scotland Acc. Nos X.DC 25, X.DE 114, X.DE 115); it is worth briefly indulging here to emphasise the necessity and benefit of a thorough analysis of object historiographies. In 1939, Alexander Keiller presented two Late Bronze Age socketed axeheads and an Early–Middle Bronze Age flanged axe to the National Museum of Scotland claiming that they were ‘found at Wester Golcantry ‘in the Tailor’s Garden,’ in 1887’ (Proceedings of the Society of Antiquaries of Scotland 1940: 149). Later Walker reported that all objects came from Miss May Davidson of Clava and Cantray (1972: 117f.). It is still possible to locate the Tailor’s House (now known as Auld Hoose) in Wester Galcantray and the Historic Environment Record records a site visit in 1964 made by the Ordnance Survey during map revision work, during which the owner confirmed that the objects were found in the front garden of that house (Canmore ID 150696). The flanged axehead predates the socketed axeheads by several centuries and this has meant the association has been considered doubtful or not portrayed as an association at all (Clark et al. 2017: 27, 51; Coles 1962: 134). This doubt has been furthered by confusion around the recording of the axeheads in various papers. Whilst Walker is ‘almost certain’ that the axeheads were found together, he lists the wrong flanged axehead (1972: 117). Meanwhile, in his article on Middle Bronze Age metalwork Coles does not refer to any palstave from ‘Wester Golcantry’ but records a flanged axe as ‘Probably nr Clava or Cantray’ (1966: 135), presumably referring to Miss Davidson’s address. The museum registration number provided by Coles is DC 125, which is recorded by the National Museum of Scotland as coming from ‘Wester Golcantry’ so is probably the same axe. The same axehead is recorded as ‘Near Cantray’ in Schmidt and Burgess (1981: 101, No 613), whilst two socketed axes are recorded as ‘Wester Golcantry’ which are possibly associated with a different flanged axehead (Schmidt and Burgess 1981: No 409). This latter axehead has the museum registration number DC 128, and there has clearly been some confusion over which of the axeheads were truly associated. This confusion can be cleared up as the flanged axehead DC 125 and the possibly associated two socketed axeheads are all marked with a consistent numbering scheme: 1939.1024–1026, and a printed note inside one of the socketed axeheads reads: ‘FLINTS, BRONZE PALSTAVE AND SOCKETED AXE-HEADS PROBABLY FROM INVERNESS-SHIRE. Alexander Keiller, 1939.’ Unfortunately, as the artefacts came up in 1887 but did not arrive in the National Museum until 1939, nor do the flanged axehead and the socketed axeheads possess a consistent patina, this association cannot be considered certain. Uncertain hoards The surviving information for several ‘multi-period’ hoards was deemed too uncertain to include here. Of note, however, is the Greyfriars Church hoard from Dumfries and Galloway, which it was possible to investigate during the course of researching this paper and warrants describing here. 6 Record last accessed 30th September 2018 via: https://www.canmore.org.uk/site/15069/wester-galcantray. 37 Objects Of the Past in the Past (no number) Greyfriars Church, Dumfries, Dumfries and Galloway (Dumfries-shire), SCOTLAND This Middle Bronze Age hoard was found in 1866 when excavating the foundations of a church. Originally it contained a spearhead and two axeheads, though was transferred without provenance information from the Crichton Royal Institute to Dumfries Museum. In addition to the Greyfriars hoard, Dumfries Museum received an additional two axeheads and one spearhead also with minimal provenance (Schmidt and Burgess 1981: 92–94, No 529). All six objects have been attributed Greyfriars Church as the original findspot, but it is unclear which three objects constitute the original discovery. The collection of objects now includes an Early Bronze Age spearhead and axehead, as well as a Middle Bronze Age spearhead and three palstaves. Davis (2012: 40–41) argued that the Greyfriars Early Bronze Age spearhead is unlikely to have come from the area. Furthermore, of the three Middle Bronze Age palstaves, one is a South-Western type, which was probably originally from the Welsh Marches (Schmidt and Burgess 1981: 142) and another is an Irish type (Schmidt and Burgess 1981: 168); neither form is common in Scotland. Due to the confusion surrounding the provenance and the eclectic nature of the objects in question, it cannot be guaranteed that the original hoard was indeed multi-period. Bibliography Adkins, L. and R.A. Adkins 1985. Neolithic axes from Roman sites in Britain. Oxford Journal of Archaeology 4(1): 69–75. Bradley, R. 1986. Neolithic axes in Roman Britain: An exercise in archaeological source criticism. Oxford Journal of Archaeology 5(1): 119–120. Bradley, R. 1998 [1990]. The Passage of Arms 2nd Ed. Oxford: Oxbow Books. Bradley, R. 2000. An Archaeology of Natural Places. London: Routledge. Bradley, R. 2002. The Past in Prehistoric Societies. London: Routledge. Bradley, R. 2017. A Geography of Offerings. Deposits of Valuables in the Landscapes of Ancient Europe. Oxford: Oxbow Books. Brandherm, D. and M. Moskal-del Hoyo 2014. Both sides now: The Carp’s Tongue complex revisited. The Antiquaries Journal 94: 1–47. Britton, D. 1963. Traditions of metal-working in the later Neolithic and Early Bronze Age of Britain: Part 1. Proceedings of the Prehistoric Society 10: 258–325. Brück, J. 2006. Fragmentation, personhood, and the social construction of technology in the Middle and Late Bronze Age Britain. Cambridge Archaeological Journal 16: 297–315. Burgess, C.B. 1968. Bronze Age Metalwork in Northern England c.1000 to 700 B.C. Newcastle-upon-Tyne: Oriel Press Limited. Burgess, C.B. and Gerloff, S. 1981. The Dirks and Rapiers of Great Britain and Ireland. Prähistorische Bronzefunde IV, 7. Munich: C.H. Beck. Callander, J.G. 1923. Three Bronze Age hoards recently added to the National Collection, with notes on the hoard from Duddingston Loch. Proceedings of the Society of Antiquaries of Scotland 56 (1921–22): 351–364. Caple, C. 2010. Ancestor artefacts – ancestor materials. Oxford Journal of Archaeology 29(3): 305–318. Chadwick, A.M. and C.D. Gibson (eds) 2013. Memory, Myth and Long-term Landscape Inhabitation. Oxford: Oxbow Books. Clark, G., T. Cowie and S. Kruse 2017. Feats of Clay. Bronze Age Metalworking around the Moray Firth. Inverness: North Kessock and District Local History Society. Clough, T.H.M. 1969. Bronze Age metalwork from Cumbria. Transactions of Cumberland and Westmorland Antiquarian and Archaeological Society 69: 1–39. Coles, J.M. 1962. Scottish Late Bronze Age metalwork: Typology, distributions and chronology. Proceedings of the Society of Antiquaries of Scotland 93 (1959–60): 16–134. Coles, J.M. 1966. Scottish Middle Bronze Age metalwork. Proceedings of the Society of Antiquaries of Scotland 97 (1963–64): 82–156. 38 Matthew G. KniGht: DOubtful assOciatiOns? assessinG bROnze aGe ‘Multi-PeRiOD’ hOaRDs Coles, J.M. 1971. Scottish Early Bronze Age metalwork. Proceedings of the Society of Antiquaries of Scotland 101 (1968–69): 1–110. Colquhoun, I. and C.B. Burgess 1988. The Swords of Britain. Prähistorische Bronzefunde IV, 5. Munich: C.H. Beck. Cowie, T. and B. O’Connor 2007. Late Bronze Age swords from Scotland: Some finds old and new, in C. Burgess, P. Topping and F. Lynch (eds) Beyond Stonehenge. Essays on the Bronze Age in Honour of Colin Burgess: 316–334. Oxford: Oxbow Books. Curwen, E.C. 1954. The Archaeology of Sussex. London: Methuen and Co. Ltd. Davies, A. forthcoming. Forgetting and remembering in the Late Bronze Age and Iron Age of southern Britain: Personhood, identity and ancestors, in C. Gibson, D. Brown, and J. Pyzel (eds) Making and Unmaking Memories. Mundane Mnemonics, Artificial Amnesia and Transformed Traditions. Oxford: Archaeopress. Davis, R. 2012. The Early and Middle Bronze Age Spearheads of Britain. Prähistorische Bronzefunde V, 5. Stuttgart: Franz Steiner Verlag. Davis, R. 2015. The Late Bronze Age Spearheads of Britain. Prähistorische Bronzefunde V, 7. Stuttgart: Franz Steiner Verlag. Eogan, G. 1983. Hoards of the Irish Later Bronze Age. Dublin: University College Dublin. Ferris, I. 2012. Roman Britain Through its Objects. Stroud: Amberley Publishing. Fontijn, D.R. 2002. Sacrificial landscapes: cultural biographies of persons, objects and ‘natural’ places in the Bronze Age of the Southern Netherlands, c.2300–600 BC. Analecta Praehistorica Leidensia, 33/34. University of Leiden. Foster, J. 1986. The Lexden Tumulus: A Reappraisal of an Iron Age Burial from Colchester, Essex (British Archaeological Reports British Series 156). Oxford: British Archaeological Reports. Gosden, C. and G. Lock 1998. Prehistoric Histories. World Archaeology 30(1): 2–12. Grinsell, L.V. 1970. The Archaeology of Exmoor. Newton Abbot: David & Charles. Gwilt, A. 2004. Late Bronze Age Societies (1150–600 BC); Tools and Weapons, in M. Aldhouse-Green and R. Howell (eds) The Gwent County History Volume I; Gwent in Prehistory and Early History: 111–139. Cardiff: University of Wales Press. Gwilt, A. 2006. Llancarfan, Vale of Glamorgan: Late Bronze Age hoard (05.15). Treasure Annual Report 2005/6. London: Department of Culture, Media and Sport: 221. Gwilt, A., M. Lodwick and M. Davis 2015. Treasure Act 1996 (Case 14.09 Wales). A probable Late Bronze Age hoard from Penllyn, Vale of Glamorgan. Unpublished Treasure Report. Gwilt, A., M. Lodwick, J. Deacon, N. Wells, R. Madgwick and T. Young 2016. Ephemeral Abundance at Llanmaes: Exploring the residues and resonances of an Earliest Iron Age midden and its associated archaeological context in the Vale of Glamorgan, in J.T. Koch and B. Cunliffe (eds) Celtic from the West 3: Atlantic Europe in the Metal Ages: Questions of Shared Language: 293–332. Oxford: Oxbow Books. Hansen, S. 2016. A short history of fragments in hoards of the Bronze Age, in H. Baitinger (ed.) Materielle Kultur und Identitätim Spannungsfeld Zwischenmediterraner Weltund Mitteleuropa: 185–205. Mainz: Römisch-Germanisches Zentralmuseum. Hawkes, C.F.C. 1954. Inventaria Archaeologica 1st Set. London: Trustees of the British Museum. Hingley, R. 2009. Esoteric knowledge? Ancient bronze artefacts from Iron Age contexts. Proceedings of the Prehistoric Society 75: 143–165. Ingold, T. 1993. The temporality of the landscape. World Archaeology 25(2): 152–174. Jones, A.M. 2007. Memory and Material Culture. Cambridge: Cambridge University Press. Knight, M.G. 2019. A radiocarbon date for a Llyn Fawr phase hoard from Scotland. PAST. The Newsletter of the Prehistoric Society 91: 13–14. Knight, M.G. forthcoming. Putting ‘out-of-time’ metalwork in its place: commemorating and forgetting traditions through the Bronze Age metalwork of Southern Britain, in C. Gibson, D. Brown, and J. Pyzel (eds) Making and Unmaking Memories. Mundane Mnemonics, Artificial Amnesia and Transformed Traditions. Oxford: Archaeopress. 39 Objects Of the Past in the Past Knight, M.G. and A. Gwilt 2017. Treasure Act 1996 (Case 16.14 Wales) A Late Bronze Age hoard from Mawr Community, Swansea. Unpublished Treasure Report. Lawson, A.J. 2018. Late Bronze Age Hoards: New Light on Old Finds (East Anglian Archaeology 166). Dereham: Historic Environment Service. Levy, J.E. 2010. Memory, landscape and body in Bronze Age Denmark, in K.T. Lillios and V. Tsamis (eds) Material Mnemonics. Everyday Memory in Prehistoric Europe: 123–146. Oxford: Oxbow Books. Lillios, K.T. 1999. Objects of memory: The ethnography and archaeology of heirlooms. Journal of Archaeological Method and Theory 6(3): 235–262. Lillios, K.T. and V. Tsamis (eds) 2010. Material Mnemonics. Everyday Memory in Prehistoric Europe. Oxford: Oxbow Books. Lynch, F. 1991. Prehistoric Anglesey 2nd Ed. Llangefni: The Anglesey Antiquarian Society. McLaren, D. 2016. The provision of amulets and heirlooms in Early Bronze Age children’s burials in Scotland, in F. Hunter and A. Sheridan (eds) Ancient Lives. Object, People and Place in Early Scotland. Essays for David V Clarke on his 70th Birthday: 245–261. Leiden: Sidestone Press. Mörtz, T. 2018. Violence and ritual in Late Bronze Age Britain: Weapon depositions and their interpretation, in C. Horn and K. Kristiansen (eds) Warfare in Bronze Age Society: 168–188. Cambridge: Cambridge University Press. Needham, S.P. 1988. Selective deposition in the British Early Bronze Age. World Archaeology 20(2): 229– 248. Needham, S.P. 2007. Bronze makes a Bronze Age? Considering the systemic of Bronze Age metal use and the implications of selective deposition, in C. Burgess, P. Topping and F. Lynch (eds) Beyond Stonehenge: Essays on the Bronze Age in Honour of Colin Burgess: 278–287. Oxford: Oxbow Books. Needham, S.P. 2015. Appendix 1: A revised classification and chronology for daggers and knives, in A. Woodward and J. Hunter 2015. Ritual in Early Bronze Age Grave Goods. Oxford: Oxbow Books. Needham, S.P. 2017. Assemblage, structure and meaning in bronze metalwork studies: An analysis of the British Penard assemblage. Oxford Journal of Archaeology 36(2): 111–156. Needham, S.P. 2018. The Classification of Chalcolithic and Early Bronze Age Metal Axe-heads from Southern Britain. Oxford: Archaeopress. O’Connor, B. 1979. A Middle Bronze Age spearhead from Merton. Oxoniensia 44: 92–93. O’Connor, B. and T. Cowie 1995. Middle Bronze Age dirks and rapiers from Scotland: Some finds old and new. Proceedings of the Society of Antiquaries of Scotland 125 (1995): 345–367. Poyer, A.J. 2015. The Topographic Setting of Bronze Age Metalwork Deposits in North East England. Unpublished PhD dissertation, Department of Archaeology, University of Sheffield. Proceedings of the Society of the Antiquaries of London. 1856. Proceedings Thursday November 23rd, 1854. Proceedings of the Society of the Antiquaries of London 3 (1853–1856): 121. Proceedings of the Society of Antiquaries of Scotland. 1882. Donations to the museum. Proceedings of the Society of Antiquaries of Scotland, 16 (1881–1882): 401–418. Proceedings of the Society of Antiquaries of Scotland. 1921. Donations to the museum. Proceedings of the Society of Antiquaries of Scotland 5th Ser. 7 (1920–21): 10–24. Proceedings of the Society of Antiquaries of Scotland. 1940. Donations to and purchases for the museum 1939–40. Proceedings of the Society of Antiquaries of Scotland 74 (1939–40): 147–151. Proceedings of the Society of Antiquaries of Scotland. 1958. Donations to and purchases for the museum 1955–6. Proceedings of the Society of Antiquaries of Scotland 89 (1955–56): 458–463. Rowlands, M. 1976. The Organisation of Middle Bronze Age Metalworking (British Archaeological Report British Series 31). Oxford: British Archaeological Reports. Rowlands, M.J. 1993. The role of memory in the transmission of culture. World Archaeology 25(2): 141–151. Schmidt, P.K. and C.B. Burgess 1981. The Axes of Scotland and Northern England. Prähistoriche Bronzefunde IX, 7. Munich: C.H. Beck. 40 Matthew G. KniGht: DOubtful assOciatiOns? assessinG bROnze aGe ‘Multi-PeRiOD’ hOaRDs Taylor, R.J. 2003. The spearhead, in S. Coles, S. Hammond, J. Pine, S. Preston and A. Taylor (eds) Bronze Age, Roman and Saxon Sites on Shrubsoles Hill, Sheppey and at Wises Lane, Borden, Kent: 42–43. Reading: Thames Valley Archaeological Services Ltd. Monograph 4. Varley, R. 1977. Bronze axes from Calderdale. Yorkshire Archaeological Journal 49: 51–56. Walker, I.C. 1972. The counties of Nairnshire, Moray and Banffshire in the Bronze Age, Part II. Proceedings of the Society of Antiquaries of Scotland 104 (1971–1972): 71–120. Walker, J.W. 1939. Wakefield. Its History and People 2nd Ed. Wakefield: The West Yorkshire Printing Company. Weiner, A.B. 1992. Inalienable Possessions: The Paradox of Keeping-while-Giving. Berkeley: University of California Press. Woodward, A. 2000. The prehistoric pottery, in G. Hughes The Lockington Gold Hoard. An Early Bronze Age barrow cemetery at Lockington, Leicestershire: 48–61. Oxford: Oxbow Books. Woodward, A. 2002. Beads and Beakers: Heirlooms and relics in the British Early Bronze Age. Antiquity 76(294): 1040–1047. Woodward, A. and J. Hunter 2015. Ritual in Early Bronze Age Grave Goods. Oxford: Oxbow Books. Yates, D. and R. Bradley 2010a. Still water, hidden depths: The deposition of Bronze Age metalwork in the English Fenland. Antiquity 84: 405–415. Yates, D. and R. Bradley 2010b. The siting of metalwork hoards in the Bronze Age of South-East England. The Antiquaries Journal 90: 41–72. York, J. 2002. The life cycle of Bronze Age metalwork from the Thames. Oxford Journal of Archaeology 21(1): 77–92. 41