The Chiming of Crack'd Bells: Recent Approaches to the Study of Artefacts in Archaeology 9781407313214, 9781407342856

Table of contents :
Front Cover
Title Page
Copyright
Table of Contents
Introduction
Pots as Things: Value, Meaning and Medieval Pottery in Relational Perspective
Vehicles for Thought: Terrets in the British Late Iron Age
(Ad)Dressing the Anglo-Saxon Body: Corporeal Meanings and Artefacts in Early England
All Form One and One Form All: the relationship between pre-burial function and the form of early Anglo-Saxon cremation urns
Plates and other vessels from early modern and recent graves
Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire
Not so Much a Pot, More an Expensive Luxury: Commercial Archaeology and the Slow Death of Pottery Analysis
The Organisation of Late Bronze Age–Early Iron Age Society in the Peak District National Park

Citation preview

BAR S2677 2014 BLINKHORN & CUMBERPATCH (Eds)

The Chiming of Crack’d Bells: Recent Approaches to the Study of Artefacts in Archaeology Edited by

Paul Blinkhorn Christopher Cumberpatch

THE CHIMING OF CRACK’D BELLS

B A R

2677 Blinkhorn and Cumberpatch cover.indd 1

BAR International Series 2677 2014 15/10/2014 09:39:30

The Chiming of Crack’d Bells: Recent Approaches to the Study of Artefacts in Archaeology Edited by

Paul Blinkhorn Christopher Cumberpatch

BAR International Series 2677 2014

ISBN 9781407313214 paperback ISBN 9781407342856 e-format DOI https://doi.org/10.30861/9781407313214 A catalogue record for this book is available from the British Library

BAR

PUBLISHING

Contents

Introduction...........................................................................................................................................................................1 Chris Cumberpatch and Paul Blinkhorn Pots as Things: Value, Meaning and Medieval Pottery in Relational Perspective...............................................................3 Ben Jervis Vehicles for Thought: Terrets in the British Late Iron Age.................................................................................................17 Anna Lewis (Ad)Dressing the Anglo-Saxon Body: Corporeal Meanings and Artefacts in Early England............................................27 T. F. Martin All Form One and One Form All: the relationship between pre-burial function and the form of early Anglo-Saxon cremation urns.................................................................................................................................39 Gareth Perry Plates and other vessels from early modern and recent graves...........................................................................................65 Beth Richardson Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire................73 Christopher Cumberpatch Not so Much a Pot, More an Expensive Luxury: Commercial Archaeology and the Slow Death of Pottery Analysis..........................................................................................................................................99 Paul Blinkhorn The Organisation of Late Bronze Age–Early Iron Age Society in the Peak District National Park.................................105 Kevin Cootes

i

Introduction Chris Cumberpatch and Paul Blinkhorn

The book is the long-delayed follow up to Not so much a pot, more a way of life (Cumberpatch and Blinkhorn 1997). Like its predecessor it is based on a session at a TAG conference, in this case that held at Liverpool in 2012, although with some differences in the line-up of papers. Even after thirty-four years, TAG remains the most stimulating and vital event on Britain’s annual archaeology conference circuit and we are pleased to be able to celebrate it in this volume.

we that hope will be shared by those who buy or borrow the book. In addition to presenting new approaches to archaeological data, we hope that the book reflects the centrality of research to the discipline. Research, we assert, lies at the core of archaeology. The commitment to research, to the accumulation of data, to its analysis and to its theoretically-informed interpretation is one of the primary constituents of archaeology as an enterprise and a discipline. Whether undertaken within the commercial sphere, as a voluntary activity or within the university sector, it is research, the practical dimension of curiosity, that unites archaeologists. Without a commitment to robust and intellectually defensible research we will be left with nothing but a handful of speculative tales about Kings, Celts and Warriors, subject to the distortions of politicians and sectarians. For this reason we regret the disavowal of research by some within commercial archaeology and are profoundly disturbed by the marketisation of the university sector which seems to be the inevitable result of the Browne Report, even setting aside the scandalous waste of resources represented by the astronomical salaries paid to bureaucrats and administrators whose contribution to the culture of research is at best marginal and at worst actively counter-productive.

The diversity of papers presented at the conference session is reflected in those included in this volume and we make no apologies for the range of subjects represented. We would have liked more and regret the pressures of work on individual authors which precluded them from submitting papers to the book. There are real dangers for an academic discipline when its internal distinctions begin to harden into divisions whether of methodology, the objects of enquiry or philosophy. In the case of archaeology this is a particular danger given the supposed distinctions between the commercial, academic and popular or democratic approaches to the subject which provide ready-made fault lines along which the discipline can easily split. When these fault lines harden and divisiveness becomes reified, then the discipline itself is in peril as it becomes easy prey for those for whom intellectual curiosity and the process of open-minded enquiry into the material world represent a threat rather than an opportunity. We hope that those who pick this book up at a conference bookstall or who read a synopsis on an internet site or even find it in one of those imperilled institutions known as libraries will find papers outside their own particular area of specialisation of interest and perhaps even of relevance. Reading the papers prior to publication has been a pleasure and one

Before this introduction becomes the occasion for the presentation of a political manifesto by the editors, we would like to thank the contributors to the volume whose patience with a long-drawn out editorial process must to be acknowledged, to those who presented papers in Liverpool but who were unable to contribute to the final volume and to the organisers of the conference itself.

1

Pots as Things: Value, Meaning and Medieval Pottery in Relational Perspective Ben Jervis 1 Biographical approaches to material culture are now commonplace within archaeological discourse, most notably in relation to the study of material culture (see, for example, Gosden and Marshall 1999; Joy 2010, Morris 2011; Martin 2012). Such approaches have demonstrated that artefacts gain and lose meaning throughout their lives and that meaning is multi-faceted; essentially that the same objects can have a plurality of meanings at any one time. Less well articulated are the processes through which objects become meaningful, particularly in relation to the location of the agency for its emergence. In this contribution I seek to examine the emergence of meaning and value in relation to a form of material culture often taken to be meaning- and value-less, medieval pottery.1

being the result of the coming together of a range of ‘actors’ in a particular place at a particular time. Therefore, whilst the physical properties of an object limit the activities which it can afford, these are not the only limiting factor and by becoming entangled in multiple webs of interaction a single object can be enacted in multiple ways both throughout its life but also at the same time. Therefore, the trajectories of meanings that artefacts take through their production and use need not be linear, but rather can be considered as a series of sporadic cycles of re-forming and re-making. The meanings, value and properties of even the most solid object are fluid, emerging relationally as they fall in and out of networks of interaction. Moving away from pottery this is well illustrated through a recent study of sheep during the 2001 Foot and Mouth disease epidemic (Mol and Law 2008). This study considers that the same sheep can become enacted in multiple ways, by being simultaneously enrolled in multiple networks of interaction. For example, sheep were enacted individually as veterinary sheep, collectively as epidemiological sheep, as economic entities and as part of the flock, which in turn constitute the farm. These processes of enacting had multiple effects acting, for example, upon the cost of sheep, creating scientific models and inciting emotional responses within farmers. A sheep then is not just a sheep but a multi-faceted thing, which becomes meaningful and effective through its enrolment in courses of action. This same logic can be transferred to any thing, be they animate or inanimate – they are not inherently meaningful, but rather enacted, in multiple ways, as meaningful objects by being entangled within courses of action. For example, Goldberg (2008, 135) demonstrates silver spoons could function in a variety of ways in the medieval period, with them gaining social value by being enacted in particular ways, to become symbols of wealth and status.

Meaning in Motion Objects have a set of material properties which remain (relatively) stable, and influence, to some degree the ways in which people interact with them (Schiffer and Skibo 1997). Pottery in particular is a highly stable substance. Although use may lead to the abrasion or chemical attrition of surfaces (Skibo 1992; Biddulph 2008; Perry 2012a; Jervis 2011), it is unlikely that a pot will ever completely ‘wear out’. Rather, a pot is most likely to lose its physical stability through breakage, the loss of form, and thus intended, or perhaps conventional, utility. Furthermore, pottery can be repaired (Marter-Brown and Seager-Smith 2012; Perry 2012b; Kyle 2012), garnishing ceramic vessels with a uniquely durable set of physical properties, related both to the chemistry of its material and the durability of form. It is these material properties which underpin, in part, its utility to the archaeologist. Whilst an objects physical properties contribute to the ways in which it can become meaningful, we can contrast these physical, material, properties and what might be termed ‘performance characteristics’ (Schiffer 1999, 19) affordances (Knappett 2005, 47-9) or emergent properties (Conneller 2011, 120). These are the properties of objects which are fluid, coming about relationally as the physical properties of objects are interpreted or promote action in a particular way. A ceramic vessel only affords drinking if a substance to drink is present and the form is recognised as being appropriate for such an activity, for example. As such the utility of the form as a drinking vessel is emergent,

The use of the word thing in the title of this paper is a deliberate attempt at making a distinction between the enacted and non-enacted pot. Following Bill Brown’s (2001) ‘thing theory’, ‘things’ can be considered to be in the background, part of a malaise of ‘thingness’ which surrounds us. The opposite is the object, the enacted, meaningful thing. The categories of thing and object can be considered to be relational – being achieved and maintained through interaction, with, furthermore, the same item simultaneously being enacted as multiple objects and remaining a thing depending upon our perspective (see also Knappett 2011, 176). To take a simple example, a mug may be enacted as a personal drinking vessel to the consumer, yet this mundane interaction may sub-consciously wash over anyone else, having no effect, with the mug remaining a thing. An object then can be

 ([email protected]) School of History, Archaeology and Religion, Cardiff University, John Percival Building, Colum Drive, Cardiff, CF10 3EU 1

3

Ben Jervis considered to have been enacted, to have gained meaning through interaction and the ability to effect through interaction. The key point to take from this discussion is that meaning is not inherent within an object, but rather it emerges through action, in a multi-faceted and relational manner. Furthermore, meaning is a broad term and could equally be substituted for value, utility or identity, all of which are relational concepts, underpinned in part by the physical properties of a thing but also by the others, both human or otherwise, with which it becomes entangled.

dissolving and being re-made; the world is in motion and objects play a key role in determining the trajectories it takes. Individuals experience different ‘social realities’ depending upon the connections they make, and, thus the ways in which they navigate, negotiate and experience the world. Through becoming entangled in multiple lives objects develop multi-faceted meanings, being enacted in multiple ways and having varying effects on individual and collective identities, defined here as the way in which we relate ourselves to human and non-human others.

The astute reader will have, by now, identified an underlying influence in the discussion thus far. The ideas presented here can be seen as grounded in the ActorNetwork Theory of John Law (1992) and Bruno Latour (2005), but also in related relational approaches which have been developed in sister disciplines, most notably human geography (Thrift 2002; Anderson and Harrison 2010), as well as in archaeology (e.g. Shanks 2007; Herva 2009; Knappett 2011; Hicks 2011). The concept of motion is central to such approaches, which demand us to acknowledge that meaning is not fixed, that ‘the social’ is constantly formed and re-made through action (rather than determining action) and that agency is neither a property of humans or things.

In summary the meaning of things is neither fixed nor arbitrary. Meaning is an emergent property of an object, coming about and being maintained through action. Meaning is multi-faceted and the same ‘thing’ can be enacted as multiple ‘objects’ simultaneously, bringing about varying effects in the form of divergent trajectories of experience and multiple conceptualisations of identity. Meaning and Medieval Pottery One reason behind this lack of interpretive study is that pottery is taken to be an inherently low value, and therefore unimportant, object by the majority of scholars. This is undoubtedly a fair point and is attested to in historical documents (and its absence from them) such as manorial records and port books (le Patourel 1969; Courtney 1999, 103; Brown 2011, 193). The study of 19th century pottery shows a strong relationship between the price of pottery and the contexts in which it was used (Spencer-Wood and Heberling 1987, 69) and therefore the monetary value of pottery can be considered an emergent property of ceramics which determine the courses of action in which they can become enrolled at a given point in their biography. Clearly value changes throughout the life of an object, and we need to understand how pottery ‘achieved’ its low relative value and how its value influenced the course its biography could take and thus the effects of its use. Dyer (in press), for example, demonstrates how materials and manufacturing techniques underpin differences in the recorded monetary values of furniture found in medieval houses. To understand the value of pottery we must start at its conception. Clay is an abundant and easily accessible resource. It is a malleable material which is easy to form and crucially pottery can be mass produced, in a way which metal and glass vessels cannot, with several tens or even hundreds of pots being fired in a single kiln. Pottery lacked exclusivity by virtue of the fact that it was abundant and simple rules of supply and demand determined that it was therefore of low economic value. The effect of constant and repeated engagement with clay and pottery, often in marginal locations due both to the dangers of ‘fire’ technologies and the need to access raw materials, led to the emergence of a distinctive professional identity and impacted both the way in which potters related to others in medieval society and the ways in which society related to them, as a marginal social group (Blinkhorn this vol). Already then we see that pottery acquired low economic value and this directly impacted upon the trajectories of action in which potters could follow through their lives

Agency is a contentious issue and is notoriously difficult to define (Knappett and Malfouris 2008; Robb 2010). Here I will define it as the ability to cause effect. Whilst everything and everyone has the potential to cause effect, this is only realised through the coming together of entities. As such, agency can be seen as spun through action (Watmore 1999, 27) distributed between people and things and therefore being an unstable, relationally achieved and fluid property of assemblages (or what Latour (2005) may term ‘hybrids’) of human and nonhuman actors. Agency is different to intentionality, which can be considered an emergent property of a person, created in specific situations in which a person comes to be inclined to act in a conscious and deliberate manner. We can then identify that the agency for the affordances, performance characteristics or emergent properties of an object is not located within the human producer or user, or in the object itself, but rather in their coming together. In other words, objects come to be indentified, meaningful and valuable through action and only retain an identity, meaning or value through repeated action, be it with the same or a similar object. That is not to say that the meaning of things is contextual. Such phrasing suggests that within a specific social context meaning is fixed and pre-determined. But social contexts are neither fixed nor pre-determined. Following Latour (2005) ‘the social’ emerges through action. Whilst social networks can become durable, principally through the enrolment of physically durable entities (objects) into social discourse (see Mol and Law 1995), action unfolds context, rather than taking place within it (Gregson and Rose 2001, 441). Therefore ‘the social’ can be conceptualised as a bundle of connections, which are constantly forming, 4

Pots as Things: Value, Meaning and Medieval Pottery in Relational Perspective and impacted the way in which they and their wares were perceived within medieval society. It may be possible to criticise pottery studies for fetishising ceramic vessels, thereby elevating them to a level of importance which they never achieved in the medieval period. Such a criticism however would misunderstand the distinction between the enacting of pottery as a functional object in the medieval period and the enacting of sherds or vessels as research tools in the present day. Clearly, the pots become entangled in very different courses of action and therefore develop very different meanings. Furthermore, simply because an object was unimportant in the collective consciousness, it does not means that its production, exchange, use and deposition did not have social effects, the study of which can only enrich our understanding of the medieval period.

pre-conceptions about the social and economic value of pottery to medieval people and the second aim of this paper is to critically consider value, through the study of pottery alongside that of medieval documents. Fifty Sherds of Grey: Dominance and Submission in the Medieval Dining Hall The first case study is concerned with exploring the relationship between pottery and identity within a medieval castle. Within the socially competitive world of medieval England the dining hall became a location in which things could become meaningful objects, enrolled in the construction and transmission of identities and forcing others to relate to their surroundings in particular ways. Items which, through their biographies, had come to seen as high status contributed to an atmosphere of opulence (e.g. Pluskowski 2004, 300), whilst rules of etiquette managed the ways in which people could interact (Phillips 2005, 146). Furthermore, by being enrolled in courses of action, objects such as pottery could, temporarily, become enacted as tools for the negotiation and transmission of identities of dominance, power and wealth. In order to explore the role of pottery in this process it is first necessary to briefly reflect on the nature of identity in relational perspective, and the ways in which ceramics have contributed to the study of medieval identities.

The publication of ‘Not so Much a Pot, More a Way of Life’ (Blinkhorn and Cumberpatch 1997), the pre-cursor to this volume, marked a watershed in the realisation of the interpretive value of medieval pottery. Prior to the mid 1990s the focus of ceramic studies was largely centred on reconstructing patterns of production and exchange as well, of course, as building chronologies (see Davey 1988; Brown 1988 for a critique). There were, of course, exceptions, most notably Richard Hodge’s (1981) processual examination of the early medieval economy and Julian Richards’ (1987) exceptional study of AngloSaxon cremation urns, both of which sought to ground the study of ceramics within solid, theoretically informed, interpretations. Two papers published in Medieval Ceramics in 1988 (Davey 1988; Brown 1988) sought to stimulate a deeper realisation of medieval pottery’s interpretive value, but it was not until the mid 1990s and the publication of the aforementioned volume, as well as a series of papers in the journal Medieval Ceramics which took a more interpretive perspective (Brown 1997; Courtney 1997; Guttierez 1997; Blinkhorn 1999) that the full potential of the material started to be realised. It is easy to criticise, but of course the characterisation, dating and development of analytical techniques were important steps in building a situation in which the meaning of pottery to medieval populations could be examined. The introduction of themes such as identity, meaning and a greater understanding of the role pottery played in the development of medieval social contexts, has opened up a broader field of research, with considerably greater relevance to our understanding of medieval society (see for example Jefferies 2001; Bryant 2004; Mellor 2004; Jervis 2005; Peters and Verhaege 2008; Spavold 2010; Vroom 2011; Jervis 2011; Perry 2012a; Brown 2012; Kyle 2012; Naum 2012). In the last 15 years pottery has moved, in some spheres at least, from being a tool to being a rich source of information which can inform, not only our understanding of the medieval economy, but also the ways in which people experienced and understood the world around them in the medieval period. In particular, the study of pottery has become aligned with the study of identity within medieval archaeology and one aim of this paper is to consider the implications of a relational approach to this theme. Analysis remains hindered however by

Pottery and Identity Identity is a theme which has long been prevalent within medieval archaeology. In particular the early relationship between history and archaeology led to archaeologists seeking to identify signatures of ‘known’ groups, in particular in relation to ethnicity (e.g. Myres 1969; 1977; see Curta 2007 and Hinton 2011 for summaries and critique and Jervis (2014) for an extended discussion of these themes). The advent of post-processual archaeology in the 1990s saw greater diversity in the use of material culture, in the broadest sense, to identify elements of identity such as gender (Gilchrist 1994; Hadley 2004) and to examine the ways in which material culture could be used to construct identities of status (de Clerq et al 2007) or resistance (Smith 2009). For the sake of brevity I will focus here on the ways in which pottery has contributed to the development of medieval identity studies. Blinkhorn’s (1997) study of habitus, identity and Anglo-Saxon pottery shifted the ceramicists’ gaze, from focussing on specific types of pottery as reflections of group identities, to move towards an emphasis on the relationship between practice and identity. Working in the Roman period, Pitts (2007) has critiqued this relationship, emphasising the importance of not replacing objects with practices as symbols of identity, but rather to consider practice as the mechanism through which identity emerges as effect (Pitts 2007, 701). In the context of the preceding discussion, identity can be considered an emergent of people, which emerges through the formation and maintenance of relationships with their material 5

Ben Jervis surroundings. Davey (2000), for example, explores how pottery production practices were one component of the negotiation of a hybrid Manx identity in the Middle Ages, rather than standing for such an identity. Naum (2012) too sees the production of pottery in medieval Denmark as a medium through which identities could be re-formed, through the undertaking of habitual action, cuing memory and contributing to the building of a sense of home. Similar conclusions are reached by Kyle (2012) in her study of pottery in early medieval Ireland and Vroom (2011) in her study of ceramic decoration in the Byzantine world; that practices allow people to relate to their surroundings in particular ways, thus making identities, rather than replacing objects as static reflections of them.

of relationships, and therefore are fluid and multi-faceted, rather than static and socially pre-conditioned. The agency for identity formation is therefore not located within people, nor do objects have the inherent agency to stand for identities, rather this agency is spun and maintained through action. A sense of home, and thus a feeling of belonging, for example, is created through continued interaction with objects and spaces which have become familiar, with these objects being enacted as icons of memory and objects of identity, thanks to their entwinement in the life history of an individual. As we navigate our lives, engaging with the material world, things are enacted as objects and, through repeated, meaning becomes sedimented within them as they are enacted as icons of memory; individual or collective meaning is not inherent within things but which is achieved and maintained through action. Identity is an effect of these entangled biographies and trajectories of action.

It has been demonstrated therefore that pottery can be used in sophisticated ways to examine the ways in which identities emerged in the medieval period. However by defining identity as a relational phenomenon, it becomes necessary to de-centre our approach to the concept, to move away from seeing all identities as conscious, intentional constructions, but rather as being an effect of the agency spun and distributed through interactions between people and the material world. It becomes necessary to ask how objects become enrolled in identity formation and question the effect of these processes in terms of the making of things into meaningful objects.

That is not to say that people are incapable of consciously using objects to construct, or rather transmit, ideas of identity in particular situations. In order to maintain such an identity vigilance is required, with the associations behind it being continually recognised and reproduced as markers of difference (Saldenha 2010, 287). Therefore the agency to construct an identity can be seen as the product of particular courses of action, in which the intentionality, or desire, to express such an identity is formed and the things, which can be enrolled in this process and thus enacted as symbols of identity, are accessible. Perhaps the most obvious examples of such identities are the stressing of ancestral links through the construction of ethnicity in burial practice (Halsall 1996) or the expression of wealth through the building of a noble identity (de Clerq et al 2007). As has been shown particularly through the study of ethnicity, such constructed identities are complex. It is not enough to identify simple distinctions, rather we need to consider that these transmissions of identity were themselves courses of action, in which some people could manage their relationships with the material world, and thus their identities, whilst having a profound impact on the identities of others, for example by limiting the ways in which they could relate to their surroundings. In this way it becomes apparent that whilst, at certain times, the identity to construct an identity may come about, the agency for identity emergence is not a human property, but rather an emergent property of fluid assemblages.

If identity is taken to emerge relationally, as an emergent property of people, it is not, by definition, inherent within people. As such the agency for its formation cannot be seen as located within a person, but rather spun through action and interaction with the world around them. We may even go as far as to suggest that identities are so fluid that they are in a constant state of ‘becoming’, with identity not being a possession of an individual or group, but rather a process formed through persistent action and interaction at a certain time, in a certain place (Geenhough 2010, 38). In other words, we don’t define how things make us feel about ourselves; rather action enacts ourselves and our surroundings in multiple ways, which is in part determined by our past experiences and the biographies of things. Such an approach finds parallels in recent considerations of ‘relational personhood’, in which the individual is considered not to reside within the body, but rather the person is a bundle of relationships between the body and the material world (see for example Fowler 2001; Jones 2005; Kirk 2006). Therefore individual identities do not exist a priori, but rather persons are formed and re-formed as they engage with their surroundings through practice. Identities are therefore not products of habitus, but are created with it (Chapman and Gaydarska 2011, 37; Jones 2005, 216; Cresse 2012 382; Knapp and Van Dommellen 2008, 22). Whilst is dangerous to try to impose a relational conceptualisation of personhood onto the medieval mind, such an approach offers a useful analytical tool, in which we can see individuals not as the product of a particular social context, but rather as constitutive of a world in which individuals and ‘the social’ alike are constructed

Dining, Pottery and Identity: The Case of Barnard Castle The archaeology of castles has been dominated in recent years by debates over function (Platt 2007; Creighton and Liddiard 2008). However, studies of lived spaces have allowed a greater understanding to develop of the varying ways in which castles were experienced by their occupants (Gilchrist 1999; Hicks 2009; Creighton and Liddiard 2008), demonstrating them to have been dynamic spaces, which have great potential for the study of identity through the study of the relationships between people, 6

Pots as Things: Value, Meaning and Medieval Pottery in Relational Perspective

Fig 1. Pottery from Barnard Castle. A: Location of Barnard Castle. B: Jugs from the Lord’s Table Group. C: Jars from the Inner Ward Group. Redrawn by the author from Austin (2007b) space and things. Barnard Castle (County Durham), a castle under the tenure of the powerful de Balloil family in the 13th-14th centuries and subsequently a possession of the Earl of Warwick, excavated between 1974 and 1981 (see Austin 2007a; 2007b) offers an opportunity to further explore this potential.

of identity formation, the emergence of meaning and the transformation of meaningless spaces within the castle into meaningful places. One effect of the varying technologies and raw materials employed in the production of objects was to influence their value, based upon the effort and time required to manufacture them and also the scale at which they could be produced. This value, not arbitrarily applied but rather intrinsically bound up within the objects biography, is an emergent property, but one which impacts the social relationships into which an item can be drawn. The assemblage from the Lord’s table is the most varied in terms of materials and would appear the most valuable. Within the socially competitive medieval world these objects had the potential to be enacted as prestige or exclusive objects, with the act of dining producing the agency for this potential to be realised. These objects gained meaning as exclusive objects through use and also became enacted as tools for the transmission and negotiation of identity. The decorated pottery used at the Lord’s table then was not intrinsically of high status (see also Courtney 1999, 104), rather it became enacted as a status object through being enrolled in the process of formal dining. The Lord could display his wealth through the use of

Amongst the assemblage of material culture two distinct groups of serving waste were identified by the excavator, based both on their composition and the area from which they were excavated (figure 1). The first, containing glass, highly decorated ceramics and a metal ewer is believed to relate to the Lord’s table (Foreman et al 2007, 498-9). The second, from the lower ward, consists of a group of much plainer pottery (ibid). Immediately it is possible to argue that these assemblages reflect the relative status of these various consumers within the castle. Although I will argue that this is a correct interpretation, the ‘common sense’ foundations upon which such an assumption is built, that these objects are of inherently varying status (and indeed that these areas of the castle are inherently meaningful), are weak (see also Courtney 1997, 101-2). Rather we need to examine how these objects came to be enacted as indicators of status and what the effects of their use were in terms 7

Ben Jervis these objects (primarily those of metal and glass, rather than ceramic pots), whilst rules of etiquette surrounding their use enacted them as tools of dominance and submission.

lower social status and therefore their presence relates to the social status of the individuals who were using them. Certainly in terms of economic value these locally produced vessels, which have minimal capital investment behind them, were likely to be the cheapest vessels available and therefore the most accessible. But we should question how people came to be related to these vessels and what the effect of these relationships was (see also Courtney 1997, 99). These people were excluded from fine dining and therefore from particular spaces within the castle. These vessels are likely to have been goods associated with the castle, rather than personal possessions and therefore their use in particular spaces had the effect of marginalising and creating stark divides within the castle. The frequency with which people interacted with these vessels impacted upon how they related to these objects and the ways in which these interactions contributed to their sense of self. For the members of the castle elite these pots were most certainly things, invisible non-objects however for servants and lower status members of the castle these objects were intimately bound to their experiences of everyday life, they contributed to their professional identities, as servants or cooks, perhaps even by extension making them non-people in the eyes of the elite who were masked from the goings on in the background of the castle. Rather than reflecting social difference therefore these vessels were bound up in the emergence and durability of divides within the castle, with the biographies of the vessels (which determined their price and availability, for example) being tightly bound up with that of their users and further contributing to both the lower status of this pottery but also cementing the identities, experiences and relative status of those who were using them. Further processes of identity emergence, particularly in relation to gender and age amongst the staff of the castle, are also likely to have developed through the use of these vessels, but further research into the demographics of the castle staff is required to build any conclusions.

Rules, it can be considered, are not inherently meaningful, but must be continuously enacted to retain their value (see Latour 2010). In doing so they limit courses of action and allow social structures and identities to be re-made and retained; through continued referencing of rules a perpetual snowball effect occurs in which the agency for continuity is relentlessly re-spun until, for whatever reason, they cease to be enacted. Courtesy texts outline etiquette and considerations such as the order in which drink was taken, which often reflects social hierarchy but in doing so re-enforced social relationships of dominance and status, whilst gestures surrounding serving embodied identities of submission on the part of servants (see Phillips 2005), with the vessel in some instances acting as the medium through which this relationship was negotiated (for example through the serving of drink or the washing of hands). These written or unwritten rules narrowed the ways in which people related to others, ensuring that through action they knew their place. The agency to form and re-make hierarchy through dining was not deferred by humans to the rules, nor did it reside within people, rather it was formed and enacted through the coming together of people, spaces and things and the continued referencing (and thus enacting) of these documents, which could be considered to ‘black box’ behaviour, providing a means through which durability could be infused into the fluid and dynamic ‘social’. These objects then had multiple effects and became meaningful in multiple ways. Participation in dining promoted a feeling of exclusivity and inclusivity, whilst those who could not participate became excluded, directly impacting their identities by limiting the social relationships in which they could participate. Etiquette controlled dining, and was made meaningful through continued adherence through action, forming the agency for these objects and the spaces in which they were used to become prestigious. Furthermore they were enacted as tools of submission, inclusion and exclusion. Control of the dining space therefore equipped an individual to build and maintain their own identity of power, whilst limiting or channelling the ways in which others could relate to the world around them, and thus develop individual senses of identity, and, similarly, limiting the ability of these people to construct identities, effectively causing dining to be an area in which social hierarchy could not only be displayed, but also maintained and enacted. The things which were bound up in these courses of action, including pottery, gained social as well as economic value, as they became enacted as exclusive objects which could be enrolled in processes of identity transmission and impacted identity formation.

This brief analysis has actually shown that some preconceptions about the relative value of medieval pottery are not wholly incorrect. What I have tried to question is the ways in which these pre-conceptions were reached and to consider the effects of interactions with a low value commodity on the lives of medieval people. I have argued that pottery is not inherently low value, but achieves a lower value throughout its biography due to the nature of the resources required to produce it. I have then considered how within contexts of use these vessels become enacted in various ways, by being drawn into social assemblages which include people, rules and spaces. Within the Lord’s hall, through the enacting of rules of etiquette for example, the vessels had multiple effects on the identities of people. On the one hand they allowed the lord to develop and further his own identity, whilst the effect on others was to contribute to senses of dominance and submission, inclusion or exclusion and servitude, in various forms from the sense of diners being drawn into general relationships of service to the lord and servants embodying their subservient position through enacting gestures. Similarly within the constable’s quarters the use of plain pottery had implications for how people

To quote from the original pottery report “in contrast pottery from the Middle Ward is very plain, being mostly either oxidised or reduced jugs with little or no decoration. If this was the area of the constable’s lodgings… then this would seem appropriate” (Freeman et al 2007, 499). The implication here is that plainer vessels were of inherently 8

Pots as Things: Value, Meaning and Medieval Pottery in Relational Perspective related to their surroundings, through limitations placed upon their access to and the nature of their relationships with more economically valuable forms of material culture. Within the town ward collective consumption can be argued to have bound up jugs in the weaving of collective and individual memory and identity. The meaning of pottery then is not inherent within it, nor is it fixed. Rather ceramic use impacted in various ways on the identities of people, creating vessels, people and spaces with multi-faceted meanings or senses of identity. In terms of ‘things’ and ‘objects’, pots became objects in different ways for different people, and it was within these processes of becoming which the agency and effect of things were established.

instances; the death intestate of a peasant, following a peasant fleeing the manor or in the case of a peasant being convicted of a felony. In all cases it was the lords right to seize the goods belonging to the peasant. Not all goods present within the peasant home belonged to the peasant and in Worcestershire a distinction can be shown between the Principialia; household and agricultural equipment belonging to the manor, and the goods of the peasant (see Field 1965). These inventories, largely drawn from the east midlands, do not appear to make such a distinction and can be argued to have been intended as a complete record of the material culture of the peasant home. The majority of records in the 14 inventories considered in Briggs’ (in press) study relate to animals, furnishings and agricultural equipment (figure 2). It is noticeable that there are few mentions of clothing or dress accessories, which are likely to have been the kind of personal possession taken with a fleeing peasant. A number of vessels are mentioned in the inventories. Whilst statistically wooden vessels are most common, the majority of these records come from a single inventory; however it should be considered that a large number of the vessels used within the peasant home are likely to have been of wood. What is perhaps surprising is the large number of metal vessels, largely of bronze, present at a time when we generally consider these to have still been of high value, and thus potentially high status, partly due to their occurrence in the archaeological record (Egan 2005, 198; 201) (figure 3). Two things need to be considered here; firstly that metal vessels are likely to have been recycled and secondly that if these items formed part of the principalia although being used by peasants, they did not belong to them. There are only two mentions to earthenware pots however, one quantified and valued at 1d and the other included within a record of ‘various vessels of wood, earthenware and metal’.

Negotiating Economic Value: Pots, Documents and the Non-Object The case study from Barnard Castle has highlighted the issues surrounding making assumptions about the social and economic value of pottery in medieval society. Perhaps the best way to examine the ways in which pottery was perceived and valued by medieval communities however is to examine textual references. As already mentioned, pottery ubiquitously has a low value in relation to other objects when it appears in documents such as port books or manorial records. Rather than simply take documents as a reflection of the relative values of objects however, I want to consider how the value of pottery was continually negotiated through the creation of documents to see if what insights this can provide into contemporary perceptions of the material. The documents under discussion are inventories, present in manorial rolls of the 14th-16th centuries, which document the possessions of peasants (Briggs in press). These inventories appear to have been produced in 3

Fig 2. The Occurrence of Object Types in Selected Peasant Inventories. Data from Briggs (In Press).

9

Ben Jervis

Fig 3.Vessels Occurring in Selected Peasant Inventories by Material (n=42). Data from Briggs (In Press). The picture of peasant material culture presented in the inventories varies greatly from that apparent from archaeological assemblages from medieval village sites. A survey of assemblages from 17 sites (appendix 1), believed to be associated with the rank and file members of the rural population, rather than manor houses, was undertaken by the author as part of a project to examine the consumption patterns of medieval rural communities. As is to be expected, pottery is by far the most frequent class of material culture from these excavations, with jars, jugs and bowls dominating the assemblages, generally being of relatively local origin. The largest class of non ceramic

finds were dress accessories (principally associated with belts). A total of 49 metal vessels, principally of bronze were present (an average of 2.9 per site). It should be noted that ceramics have a shorter use life than metal vessels and therefore several pots may have been used in the life of a single metal vessel (figure 4). However, the evidence does suggest that ceramics were used in greater quantities than vessels of any other material, except, perhaps, for wood. For example, a house at Dinna Clerks (Devon), which burnt down, preserving artefacts in situ within the house, contained 12 vessels, of which the majority are jar forms (Beresford 1979). These two records tell very

Fig 4. Occurrence of Non-Ceramic Small Finds in Selected Rural Assemblages (see Appendix 1 for references).

10

Pots as Things: Value, Meaning and Medieval Pottery in Relational Perspective different stories about the material culture of the rural home and in particular the absence of pottery perhaps tells us something about the value and perception of it held by medieval people.

communities (see also Courtney 1999, 102). From a ceramic perspective this is well shown by the evidence from the Flemish fishing village of Raversijde, where imported maiolica appears to have become enrolled in specific consumption practices, becoming meaningful in specific ways to this community (Peters and Verhaege 2008, 116). These documents therefore record a particular value, which emerged through a particular course of action, which produced the document as an artefact. This is well illustrated by changes in the ways in which objects are referenced in late medieval wills, with documents referencing an increasing range of objects as the result of changing attitudes to death, inheritance and belief (Cohn 2012, 990), with these developments spilling over to effect a change in the formation processes behind these documents and the agency for remembrance coming to be spun through the enacting of these documents. A useful work to consult in considering agency and documents is Bruno Latour’s (2010) The Making of Law. In this book Latour argues that documents are not the law, rather the law is an unstable bundle of references between documents, places, people and things. The role of documents is a particular one; they act to ‘black-box’ decisions, meaning that a referential starting point is created through the creation of the document, which introduces durability into regulatory systems and by extension into ‘the social’. Crucially, Latour argues that the law does not pre-form ‘the social’ and these documents do not stand for it, nor does ‘the social’ exist at some level beyond the law. Rather, they are mutually constitutive, with the law being one sub-set of the wider bundle of connections which form the unstable and changing social. It is fruitful to consider inventories in the same way, with the value system which they document also being a referential bundle of connections within the wider bundle which is ‘the social’. These documents are not values, but rather ‘black-box’ past evaluations of objects, providing points of reference which can be enacted in processes of inventorying, bringing durability to value systems. Furthermore these processes cannot be divided from the biographies of the things being recorded nor from the social, which these documents are both a product of and constitutive of, by being the product of a process of assembly in which people and things come together with social effects. Therefore, rather than objects being intrinsically valuable, the process of inventorying created a context in which these documents were enacted as references of value and things were enacted as objects of worth. Although value was in part determined by the materials and biography of objects, it also emerged from and was maintained through the process of inventorying.

The simple conclusion to draw here could be that inventories did not record low value items such as pottery and that items such as dress accessories may not feature as they were worn by the peasants themselves. However if medieval archaeology wishes to be a truly interdisciplinary, historical archaeology, it is necessary to problematise the relationship between history and archaeology, to move away from common sense interpretations, to consider that documents and objects were formed and used within the same material settings and therefore that both were bound up in courses of action and had the capacity to contribute to the emergence of agency and therefore to cause effects. Andrén (1998, 32) demonstrates that the relationship between medieval history and archaeology has largely been one of complementarily, with, essentially, one discipline making up for the deficiencies of the other. Furthermore, as in the late 1980s and early 1990s medieval archaeology reached a degree of maturity it sought to break away from its role as ‘the handmaiden of history’ seeking to address themes purely on the basis of archaeological evidence, using specifically archaeological approaches (Andrén 1998, 33; Rahtz 1983; Austin 1990). A more constructive approach however may be to consider text as a form of material culture (see Christopherson 1979, 6; Moreland 1992). This is fundamentally different to seeing material culture as a text to be read, as it sees documents as a particular form of object, with particular material properties, which provide them with the potential to effect in particular ways when enrolled into courses of action. Clearly artefacts, in the archaeological sense, and texts are not identical, but neither are all artefacts, physically a glass window is fundamentally different to a linen tablecloth or ceramic pot. By considering texts, just like pots, to be made as things and re-made as objects through the creation of associations through enrolment in courses of action we can move beyond simply identifying the discrepancies between historical and archaeological sources, to examine how relationships between these different forms of artefact contributed to the emergence of ‘the social’ of medieval England. Documents such as inventories are often taken as reflections of medieval value systems, being seen as, if not genuine representations of market value at least as indices of relative value. Whilst they do have this utility, they should not be seen as determining the value of things, or as a passive representation of unchanging value, as we have seen the value of things is fluid, emerging and changing through the biography of an object. This is well demonstrated by Goldberg’s (2008, 130) comparative study of rural and urban inventories, in which inventories are used not only to show different spending habits between urban and rural populations, but also to consider how the goods present acquired index the emergence of differing systems of value and served to build contrasts in the material lives, and thus identities, of medieval

In summary then, I do not see value as residing within a thing, nor inventories as static records of value. Rather I consider that through the process of inventorying value emerges, with reference to past valuations ‘black-boxed’ in these documents. Durability of this value system is only achieved through the enacting of these documents, as well perhaps as memories of the value of things in the marketplace, with this process serving to enact objects in particular ways; as valuable or value-less. This may 11

Ben Jervis Conclusion: Durable Ephemerality

impact upon the trajectory an objects’ biography can take, by limiting the courses of action in which it can become entangled, as seen, for example, at Barnard Castle.

This paper has in some ways simply served to re-enforce some of our pre-conceptions about the low value and status of pottery in medieval society. However, I have attempted to show that value is an emergent property of things – something which is accrued, lost or altered throughout an items life. Social value is fleeting, produced in the moment. The management of interactions with pottery could contribute to processes of identity creation, whilst effecting the emergence of identities, as people experienced vessels in different ways with varying effects. The temporary nature of this social value, and its separation from economic value, is shown in the way in which pottery was not enacted as valuable in processes of inventorying, a process which maintained value systems through processes of referencing, rather than simply reflecting them. The value of pottery was ephemeral, indeed as a container or a cheap tool it was perhaps ephemeral itself – a disposable and easily replaceable thing of little consequence. It is easy to write pottery off as having no significance to medieval people. However to equate the low value of pottery with a lack of meaning and effect is to ignore the transient, emergent qualities of this most ubiquitous of things. Rather than re-stating clichés the aim of research must be to re-create those courses of action and consider their effects. In doing so we can move beyond seeing pottery as reflecting identities and documents as reflecting the way of the world, to think about the ways in which people, places and objects were assembled and maintained, allowing us to engage with the plurality and transience of the medieval experience.

What then, are the implications of this thinking for our understanding of pottery’s role in the medieval period? On the whole, pottery – the most abundant artefact from rural excavations, appears to have been excluded from this process of inventorying. Whilst pots developed a social value and were instigated in the spinning of the agency to, for example, bring about and maintain social differentiation at the table, they were not enacted as objects of worth in the process of inventorying. Pots remained as things, or perhaps consciously became non-objects, as the process of inventorying served to maintain their place within the medieval value system, a place which was achieved, if we can call it that, through its biography. The ‘thing-ness’ of pottery appears more permanent than its ‘object-ness’. The large quantity of pottery recovered from excavations perhaps attests to its ephemeral nature as an object, as something temporarily enacted as a socially valuable actor but did not retain this value, and, through its biography and the courses of action in which it became entangled, maintained a somewhat liminal position in the medieval consciousness. Therefore, whilst inventories have something to tell us about the value of things, by considering how valuations were enacted we can consider how the agency for the emergence of value and the durability of value systems was spun through these human-object-document relationships. Rather than the documents reflecting how things were, they were enrolled in the process of making them that way.

12

Pots as Things: Value, Meaning and Medieval Pottery in Relational Perspective Bibliography

Christopherson, A. 1979. Arkeologi: Mer enn et hull I jorden? Synspunkter på forholdet mellom arkeologi og hisotire Meta 1, 4-8. Cohn, S. 2012 Renaissance Attachment to Things: Material Culture in Last Wills and Testaments The Economic History Review 65(3), 984-1004. Conneller, C. 2011 An Archaeology of Materials. Substantial Transformations in Early Prehistoric Europe, Routledge. Courtney, P. 1997 Ceramics and the History of Consumption: Pitfalls and Prospects Medieval Ceramics 21, 95-108. Creighton, O. and Liddiard, R. 2008 Fighting Yesterday’s Battle: Beyond War or Status in Castle Studies Medieval Archaeology 52, 161-9. Creese, J. 2012 The Domestication of Personhood: A View From the Northern Iriquoian Longhouse Cambridge Archaeological Journal 22(3), 365-86. Curta, F. 2007 Some Remarks on Ethnicity in Medieval Archaeology Early Medieval Europe 15(2), 159-85. Davey, P. 1988 Theory and Practice in Medieval Ceramic Studies Medieval Ceramics 12, 3-14. Davey, P. 2000 Identity and Ethnicity: A Ceramic CaseStudy from the Isle of Man Medieval Ceramics 24, 319. De Clerq, W., Dumolyn, J. and Haemers, J. 2007 “Vivre Noblement”: Material culture and Elite Identity in Late Medieval Flanders Journal of Interdisciplinary History 38(1), 1-31 Driscoll, S. 1988 The Relationship Between History and Archaeology: Artefacts, Documents and Power In S. Driscoll and M. Nieke (Eds) Power and Politics in Early Medieval Britain and Ireland, Edinburgh University Press. Dyer, C. In Press Furnishings of Medieval Peasant Houses: Investment, Consumption and Life Style In: A. Furió and F. Garcia-Oliver (Eds) Pautas de Consumo y Niveles de Vida en el Mundo Rural Medieval, University of Valencia Press. Egan, G. 2005 Urban and Rural Finds: Material Culture of Country and Town c.1050-1500 In C. Dyer and K. Giles (Eds) Town and Country in The Middle Ages. Contrasts, Contacts and Interconnections, 1100-1500 Society for Medieval Archaeology Monograph 22, 197-210. Field, R. 1965 Worcestershire Peasant Buildings, Household Goods and Farming Equipment in the Later Middle Ages Medieval Archaeology 9, 105-45. Freeman, C., Bown, L. & Austin, D. 2007 Pottery In D. Austin (Ed) Acts of Perception: A Study of Barnard Castle in Teesdale Volume II The Architectural and Archaeological Society of Durham and Northumberland Research Report 6, 352-499. Fowler, C. 2001 Personhood and Social Relations in the British Neolithic With a Study from the Isle of Man Journal of Material Culture 6(2), 137-63. Gilchrist, R. 1994 Gender and Material Culture: The Archaeology of Religious Women, Routledge.

Anderson and P. Harrison (eds.). 2010 Taking Place: NonRepresentational Theories and Geography, Ashgate . Andrén, A. 1998 Between Artifacts and Texts. Historical Archaeology in Global Perspective, Plenum Press. Austin, D. 1990 The ‘Proper Study’ of Medieval Archaeology In D.Austin and L. Alcock (Eds) From the Baltic to the Black Sea. Studies in Medieval Archaeology, Routledge, 9-42. Austin, D. 2007a Acts of Perception: A Study of Barnard Castle in Teesdale Volume I The Architectural and Archaeological Society of Durham and Northumberland Research Report 6. Austin, D. 2007b. Acts of Perception: A Study of Barnard Castle in Teesdale Volume II The Architectural and Archaeological Society of Durham and Northumberland Research Report 6. Beresford, G. 1979 Three Deserted Medieval Settlements on Dartmoor: A Report on the Late E. Marie Minter’s Excavations Medieval Archaeology 23, 98-158. Biddulph, E. 2008 Form and Function: The Experimental Use of Roman Samian Ware Cups Oxford Journal of Archaeology 27(1), 91-100. Blinkhorn, P. 1997 Habitus, Social Identity and AngloSaxon Pottery In: P. Blinkhorn and C. Cumberpatch (Eds) Not So Much a Pot, More a Way of Life, Oxbow, 113-124. Blinkhorn, P. 1999. The Trials of Being a Utensil: Pottery Function at the Medieval Hamlet of West Cotton, Northamptonshire Medieval Ceramics 22-23, 37-46. Blinkhorn, P and Cumberpatch, C (Eds) Not So Much a Pot, More a Way of Life, Oxbow. Briggs, C. 2011 Manorial Court Roll Inventories as Evidence of English Peasant Consumption and Living Standards, c.1270-c.1420 In: A. Furió and F. GarciaOliver (Eds) Pautas de Consumo y Niveles de Vida en el Mundo Rural Medieval, University of Valencia Press. Brown, B. 2001 Thing Theory Critical Inquiry 28(1), 1-22. Brown, D. 1988 Pottery and Archaeology Medieval Ceramics 8, 15-22. Brown, D. 1997 Pots from Houses Medieval Ceramics 21, 83-94. Brown, D. 2011, Imported Pottery and the South Coast Ports of Medieval England In A. Boucquet-Liénard and B Fajal (Eds) À Propo[t]s de l’Usage, de la Production et de la Circulation des Terres Cuites dans l’Europe du Nord-Ouest Autour des VIVe-XVIe Sèicles, CRAHAM 185-96 Brown, D. 2012 When is a Pot Still a Pot? In B. Jervis and A. Kyle (Eds) Make-do and Mend: Archaeologies of Compromise, Repair and Reuse, BAR Int. Ser. 2408, 107-14. Bryant, V. 2004 Death and Desire. Factors Affecting the Consumption of Pottery in Medieval Worcestershire Medieval Ceramics 28, 117-124. Chapman, J and Gaydarska, B 2011 Can we Reconcile Individualisation with Relational Peronhood? A Case Study from the Early Neolithic Documenta Praehistorica 38, 21-44. 13

Ben Jervis Kirk, T. 2006 Materiality, Personhood and Monumentality in Early Neolithic Britain Cambridge Archaeological Journal 16(3), 333-47. Knapp, B and Van Dommelen, P 2008 Past Practices: Re-thinking Individuals and Agents in Archaeology Cambridge Archaeological Journal 18(1), 15-34. Knappett, C. 2005 Thinking Through Material Culture, University of Pennsylvania Press. Knappet, C. 2008 The Neglected Networks of Material Agency: Artefacts, Pictures and Texts In C. Knappett and L. Malafouris (Eds) Material Agency. Towards a Non-Anthropocentric Approach, Springer, 139-56. Knappett, C. 2011 An Archaeology of Interaction. Network Perspectives on Material Culture & Society, Oxford University Press. Knappett, C. and Malafouris, L. (Eds) 2008 Material Agency. Towards a Non-Anthropocentric Approach, Springer. Kyle, A. 2012. More Than Just a Quick Fix? Repair Holes on Early Medieval Souterrain Ware In B. Jervis and A. Kyle (Eds) Make Do and Mend: Archaeologies of Compromise, Repair and Reuse, BAR Int Ser 2408, 91-96. Latour, B. 2005 Reassembling the Social. An Introduction to Actor Network Theory, Oxford University Press. Latour, B. 2010 The Making of Law. An Ethnography of the Conseil D’État Polity. Law, J. 1992 Notes on the Theory of the Actor Network: Ordering, Strategy and Heterogeneity, http://comp. lancs.ac.uk/sociology/soc054jl.html. Law, J. and Mol, A. 1995 Notes on Materiality and Sociality The Sociological Review 43(2), 274-94. Le Patourel, J. 1969 Documentary Evidence and the Medieval Pottery Industry Medieval Archaeology 12, 101-26. Marter-Brown, K. and Seager-Smith, R. 2012 What Did the Apocrypha Know? Glued Pottery Vessels from Springhead and Other Romano-British Sites in South and Eastern England In B. Jervis and A. Kyle (Eds) Make-do and Mend: Archaeologies of Compromise, Repair and Reuse, BAR Int. Ser. 2408, 5-12. Martin, T. 2012 Riveting Biographies: The Theoretical Implications of Early Anglo-Saxon Brooch Repair, Customisation and Use-Adaptation In B. Jervis and A. Kyle (Eds) Make-do and Mend: Archaeologies of Compromise, Repair and Reuse, BAR Int. Ser. 2408, 53-66. Mellor, M. 2004 Changing Rooms: Fixtures, Fittings and Movable Goods in European Lifestyles Medieval Ceramics 28, 125-38. Mol, A. and Law, J 2008 The Actor-Enacted: Cumbrian Sheep in 2001 In C. Knappett and L. Malafouris (Eds) Material Agency. Towards a Non-Anthropocentric Approach, Springer, 57-78. Moreland, J. 1992 Restoring the Dialectic: Settlement Patterns and Documents in Medieval Central Italy in A. Knapp (Ed) Archaeology, Annals and Ethnohistory, Cambridge University Press, 112-29.

Gilchrist, R. 1999 Gender and Archaeology: Contesting the Past, Routledge. Gregson, N. and Rose, G. 2001 Taking Butler Elsewhere: Performativities, Spatialities and Subjectives Environment and Planning D 18, 433-52. Goldberg, P 2008. The Fashioning of Bourgeois Domesticity in Later Medieval England: A Material Culture Perspctive In M. Kowaleski and P. Goldberg (Eds) Medieval Domesticity. Home, Housing and Household in Medieval England, Cambridge University Press, 124-44. Gosden, C. and Marshall, Y. 1999 The Cultural Biography of Objects World Archaeology 31(2), 169-78. Greenhough, B. 2010 Vitalist Geographies: Life and the More than Human In B. Anderson and P. Harrison (eds), Taking Place: Non-Representational Theories and Geography, Ashgate, 37-54. Gutierrez, A. 1997 Cheapish and Spanish . Meaning and Design on Imported Spanish Pottery Medieval Ceramics 21, 73-82. Hadley, D. 2004 Negotiating Gender, Family and Status in Anglo-Saxon Burial Practices., c. 600-950 In L. Brubaker and J. Smith (Eds) Gender in the early medieval world. East and west, 300-900, Cambridge University Press, 301-23. Halsall, G. 1996 Female Status and Power in Early Merovingian Central Austrasia: The Burial Evidence Early Medieval Europe 5(1), 1-24. Herva, VP. 2009 Living (With) Things: Relational Ontology and Material Culture in Early Modern Northern Finland Cambridge Archaeological Journal 19(3), 388-97. Hicks, D. 2011 The Material-Culture Turn. Event and Effect In D. Hicks and M. Beaudry (Eds) The Oxford Handbook of Material Culture Studies, Oxford University Press, 25-98. Hicks, L. 2009 Magnificent Entrances and Undignified Exits: Chronicling the Symbolism of Castle Space in Normandy Journal of Medieval History 35(1), 52-69. Hinton, D. 2009 Medieval Identity Issues In R. Gilchrist and A. Reynolds (Eds) Reflections: 50 Years of Medieval Archaeology, 1957-2007, Society for Medieval Archaeology Monograph 30, 453-64. Hodges, R. 1982 Dark Age Economics, Duckworth. Jeffries, N. 2001 Historically Visible but Archaeologically Invisible? The Huguenots of Spitalfields Medieval Ceramics 25, 54-64. Jervis, B. 2008 Pottery and Identity in Late Saxon Sussex Medieval Ceramics 29, 1-8. Jervis, B. 2011 A Patchwork of People, Pots and Places. Material Engagements and the Construction of ‘the Social’ in Hamwic (Anglo-Saxon Southampton), UK Journal of Social Archaeology 11(3), 239-65. Jervis, B. 2014 Pottery and Social Life in Medieval England: Towards a Relational Approach, Oxbow. Jones, A. 2005 Lives in Fragments? Personhood and the European Neolithic Journal of Social Archaeology 5(2), 193-224. Joy, J. 2010 Iron Age Mirrors. A Biographical Approach, BAR British Series 518. 14

Pots as Things: Value, Meaning and Medieval Pottery in Relational Perspective Accessories from English Rural Settlement Sites Journal of Material Culture 14(3), 309-332. Spavold, J. 2010 Faith Made Manifest: An Interpretation of the Decoration on Cistercian Ware Medieval Ceramics 31, 33-48. Spencer-Wood, S. and Heberling, S. 1987 Consumer Choices in White Ceramics. A Comparison of Eleven Early Nineteenth Century Sites In S. Spencer-Wood (Ed) Consumer Choice in Historical Archaeology, Plenum Press, 55-84. Thrift, N. 2002 Non-Representational Theory. Space, Politics, Affect, Routledge. Vroom, J. 2011 The Morea and its Links with Southern Italy After AD1204: Ceramics and Identity Archeologia Medievale 38, 409-30. Whatmore, S. 1999 Hybrid Geographies: Rethinking the ‘Human’ in Human Geography In D. Massey, J. Allen and P. Sarre (Eds) Human Geography Today, Polity Press, 22-39.

Morris, J. 2011, Associated Bone Groups in Southern England and Yorkshire, c.4000BC to c.AD1500, BAR British Series 535. Myres, JNL, 1969 Anglo-Saxon Pottery and the Settlement of England, Oxford University Press. Myres, JNL. 1977 A Corpus of Anglo-Saxon Pottery of the Pagan Period, Cambridge University Press. Naum, M. 2011 Ambiguous Pots: Everyday Practice, Migration and Materiality. The Case of Medieval Baltic Ware on the Island of Bornholm (Denmark) Journal of Social Archaeology 12(1), 92-119. Peters, M. and Verhaege, F. 2008 Medieval Coastal Fishing Communities in Coastal Flanders (Belgium) and Their Relation to Western Mediterranean Commodities Medieval Ceramics 30, 103-118. Perry, G. 2012a. Beer, Butter and Burial. The Pre-Burial Origins of Cremation Urns From the Early AngloSaxon Cemetery of Cleatham, North Lincolnshire Medieval Ceramics 32, 9-22. Perry, G. 2012b. A Hole for the Soul? Possible Functions of Post-Firing Perforations and Lead Plugs in Early Anglo-Saxon Cremation Urns In B. Jervis and A. Kyle (Eds) Make-do and Mend: Archaeologies of Compromise, Repair and Reuse, BAR Int. Ser. 2408, 43-52. Phillips, K. 2005‘he Invisible Man: Body and Ritual in a Fifteenth-Century Noble Household Journal of Medieval History 31, 143-62. Pitts, M. 2007 The Emperor’s New Clothes? The Utility of Identity in Roman Archaeology American Journal of Archaeology 111(4), 693-713. Platt, C. 2007 Revisionism in Castle Studies – A Caution Medieval Archaeology 51, 83-102. Pluskowski, A. 2004 Narwhals or Unicorns? Exotic Animals as Material Culture in Medieval Europe European Journal of Archaeology 7(3), 291-313. Rahtz, P. 1983 New Approaches to Medieval Archaeology Part 1 In D. Hinton (Ed) 25 Years of Medieval Archaeology, University of Sheffield, 12-23 Richards, J. 1987 The Significance of the Form and Decoration of Anglo-Saxon Cremation Urns, BAR Brit Ser 166. Robb, J. 2010 Beyond Agency World Archaeology 42(4), 493-520. Saldenha, A. 2010 Politics and Difference In B. Anderson and P. Harrison (Eds) Taking Place: NonRepresentational Theories and Geography, Ashgate, 283-302. Schiffer, M. 1999 The Material Life of Human Beings, Routledge. Schiffer, M. and Skibo, J. 1997 The Explanation of Artefact Variability American Antiquity 62(1), 27-50. Shanks, M. 2007 Symmetrical Archaeology World Archaeology 39(4), 589-96. Skibo, J. 1992 Pottery Function: A Use Alteration Perspective, Plenum Press. Smith, S. 2009 Materializing Resistant Identities Among the English Peasantry. An Examination of Dress

Appendix 1: Reports from which artefact data was collected Beresford, G. 2009 Caldecote: The Development and Desertion of a Hertfordshire Village, Society for Medieval Archaeology Monograph 28. Biddle, M. 1962 The Deserted Medieval Village of Seacourt, Berkshire Oxoniensia 26, 70-201. Chambers, R. 1974 A Deserted Medieval Farmstead at Sadler’s Wood, Lewkner Oxoniensia 38 146-67. Chapman, A. 2010 West Cotton, Raunds. A Study of Medieval Settlement Dynamics AD450-1450. Excavation of a Deserted Medieval Hamlet in Northamptonshire, 1985-9 Oxbow. Chapman, A., Atkins, R and Lloyd, R. 2003 A Medieval Manorial Farm at Lime Street, Irthlingborough, Northamptonshire Northampshire Archaeology 31, 71-104. Chapman, A and Audouy, M. 2008 Raunds: The Origins and Growth of a Midland Village, AD450-1500. Excavations in North Raunds, Northamptonshire, 1977-87, Oxbow. Crick, J. 1999 Evidence for Medieval Settlement from Marston Moretaine Bedfordshire Archaeology 23, 118-34. Hardy, A. 2001 The Excavation of a Medieval Cottage and Associated Agricultural Features at Manor Farm, Old Grimsbury, Banbury Oxoniensia 65, 345-80. Ivens, R., Busby, P., Sheperd, N, Hurman, B and Mills, J. 1995 Tattenhoe & Westbury. Two Deserted Medieval Settlements in Milton Keynes, Buckinghamshire Archaeology Society Monograph 8. Johnston, A., Bellamy, B and Foster, P. 2001 Excavations at Southwick, Northamptonshire, 1996 Northamptonshire Archaeology 29, 129-60. King, N., Roberts, J and Zeepvat, R. 1994 Caldecote: Excavations and Fieldwork 1966-91, Buckinghamshire Archaeological Society Monograph 9

15

Ben Jervis Cemetery at Loughton, Milton Keynes Records of Buckinghamshire 43, 77-126. Wardill, R., Slowikowski, A and Meckseper, C. 2012 Excavations at Walton Manor, Walton, Milton Keynes Records of Buckinghamshire 52, 115-34.

Mynard, D and Zeepvat, R. 1991 Great Linford: A Medieval Village at Milton Keynes, Buckinghamshire Archaeological Society Monograph 3. Newton, A and Sparrow, P. 2009 Medieval Archaeology at 139, 141 and 143 Buckingham Road, Bletchley, Milton Keynes Records of Buckinghamshire 49, 141-62. Phillips, M. 2012 Medieval Settlement Remains, Part of a Medieval Cemetery and Later Structural Remains Associated with Chicheley Hall Records of Buckinghamshire 52, 135-68. Pine, J. 2003 Excavation of a Medieval Settlement, Late Saxon Features and a Bronze Age Cremation

16

Vehicles for Thought: Terrets in the British Late Iron Age Anna Lewis

Chariot terrets, sometimes called rein-guides, are among the most numerous copper-alloy artefacts recorded from Iron Age Britain. They are known from graves, hoards, votive deposits and settlements, and stray finds occur in huge number, with over 600 recorded on the Portable Antiquities Scheme database from England and Wales alone. Despite the frequency with which they are encountered, and the diversity of their contexts, previous studies of terrets have been limited in scope. Where terrets have been studied in detail, this has generally been with the aim of creating typologies within broader catalogues of Iron Age metalwork (such as Spratling 1972, MacGregor 1976 and Palk 1992), where the material is presented in isolation from any theoretical framework. Previous studies have tended to focus on the British series of “D-shaped” terrets, with little attention paid to other terret types that developed in Britain during and after the Roman invasion.

passed over the yoke through the terrets to the driver’s hands. In Britain, sets of five terrets were most commonly used: four smaller terrets to guide the reins, and one larger terret to hold the straps that secured the yoke to the pole. (Palk 1992, 88-89; Macdonald 2007, 8.)

Terrets are of significance to Iron Age studies partly because of their sheer number: it seems that in Iron Age society they were relatively common artefacts, to which we can imagine many people would have been exposed. The range of contexts in which terrets are found demonstrates their potential to shed light on diverse aspects of life in the British Iron Age.

From the third century BC until perhaps as late as the third century AD, the “D-shaped” series of terrets developed in Britain. These are mostly made of copper alloy, but are sometimes copper-coated around an iron core. In its simplest form, the D-shaped terret consists of a subcircular ring with a straight attachment bar at the base, defined at each end by a moulded terminal; typically, the ring is widest at the base and becomes slimmer towards the top (Figure 1). Variations on the simple form include moulded embellishments around the ring in the shape of ribs, lips, wings, rounded knobs or flat platforms, which

Terrets can, in theory, be any shape or size. It is essential that a terret contains a space large enough to admit the reins, and a means by which it can be attached to the yoke, but all other elements are decorative. On Continental Europe, from the middle La Tène period to the Roman era, terrets vary widely in form – from the two hefty, cubeshaped terrets pictured in Jacobsthal (1944, nos. 163, 175), decorated on each external side with a bulbous, exaggerated human face, to numerous examples of a simple ring set on top of a smaller attachment loop or shank (e.g. Pič 1906, pl. XXVI, 1).

The aim of this paper is twofold. Firstly, it is argued that our understanding of terrets would be enriched by the application of aspects of object theory; in particular, it is suggested that this approach can help us to examine the expression of individual and communal identities through use and deposition of terrets in Late Iron Age Britain. Secondly, it is argued that despite the absence of explicit theoretical frameworks, previous studies of terrets have made implicit assumptions about British and Roman identities by excluding those terret forms thought to be Roman. The traditionally perceived distinction between British and Roman terret forms is assessed and challenged. The paper begins with a brief introduction to terrets and terret forms known in Britain from the Iron Age and Romano-British periods. It moves on to apply aspects of object theory to the study of terrets. Finally, the paper addresses the traditional distinction between native and Roman terret types, using the “massive terret” as a case study. Terrets: their forms and functions When pairs of horses or ponies were used to drive carts or chariots in Iron Age Britain, terrets were necessary to prevent the reins becoming tangled. The terrets sat along the yoke; the reins were attached to the horse bits, and

Fig 1. Simple terret from Shropshire (PAS database: LVPL-2F6BC0, © Portable Antiquities Scheme / British Museum) 17

Anna Lewis may incorporate decorative materials such as enamel or coral. The shape of the terminals and attachment bar also varies, as well as the shape of the ring itself. Perhaps the most distinctive form, the “crescentic” terret, has a broad, flat ring which increases in width towards the top, and which is decorated in enamel on both faces with typical La Tène motifs such as tendrils and scrolls. A few isolated artefacts resembling the crescentic terret are known from Europe and North Africa – one comes from Eauze in Gers, France (Michon 1925), and another from Faiyum in Egypt (Norman 1904) – but otherwise the D-shaped series is almost entirely restricted to Britain. Other terret forms, echoing those found on the Continent, are known here. One of the most common forms to be found in Britain is the “dropped-bar” terret, consisting of a roughly circular main ring with a smaller rounded or rectangular attachment loop beneath (Figure 2). Similar is the “skirted” terret, in which the small attachment loop is sheltered by a skirt or canopy. Although the great majority of these artefacts are stray finds, not from datable contexts, they are generally considered to be Roman. The British “massive” terret – consisting of a heavy-set ovular ring with a slim attachment bar or pin concealed within the deep, hollow base, from which a stubby central protrusion rises (Figure 3) – is conventionally treated as a postConquest development of the native British terret series, although its affinities with Continental types are wellknown (MacGregor 1976, 48; Palk 1992, 53-55).

Fig 3. Massive terret from Shropshire (PAS database: LANCUM-1F06E0, © Portable Antiquities Scheme / British Museum) The traditionally accepted view of terrets in Britain, then, is that the artefacts fall into two camps: firstly, the native D-shaped series, which begins perhaps in the third century BC, and which continues to develop during the Roman conquest. The second camp includes various Roman types, introduced to Britain as Rome spreads its hold and influence across the country. The massive terret falls somewhere between the two. * Terrets show considerable stylistic variation over space and time, so they have long been viewed as artefacts that can help archaeologists to determine the date and origin of associated finds. Horse-trappings, E. T. Leeds wrote in 1933 (113), “keep such close step with the changes in decorative style observable on other classes of objects, that they can be employed as criteria for relative chronology with almost the same confidence as other things upon which reliance is more normally placed.” To this end, several attempts have been made to draw up a thorough typo-chronology of terrets, beginning with Leeds’s own effort in 1933, which has been modified over the years by Mansel Spratling (1972), Morna MacGregor (1976) and Natalie Palk (1992). Although there are differences between the various typologies – certain authors choose to group some forms together into one type, while others split the same forms into sub- and sub-sub-types – all follow the same basic structure, and all agree that the earliest terrets in Britain are probably the simple forms and the lipped and multilobed forms, well-known from the Arras cemeteries of East Yorkshire (Stead 1979, fig.17.2-4). Recent modelling of radiocarbon dates for chariot burials in East Yorkshire estimates that the chariot burial rite began in 255-195 cal BC (95% probability), probably between 225-200 cal BC, and lasted for up to 70 years (Jay et al 2012).

Fig 2. Dropped-bar terret from Cheshire (PAS database: LVPL1965, © Portable Antiquities Scheme / British Museum)

18

Vehicles for Thought: Terrets in the British Late Iron Age Further evidence for the early origin of these forms comes from Gussage All Saints. Charcoal from pit 209 at Gussage produced a calibrated date range from the fourth to the first centuries BC (Garrow et al 2009); the deposit contained mould fragments for the manufacture of simple, ribbed and lipped terrets, as well as for simple terrets embellished with complex mouldings along the spine and on each face. Assessing the densely compacted fill of pit 209 and the lack of evidence for silting, Duncan Garrow and Chris Gosden (2012, 271-276) argue that the material is most likely to have been both created and deposited over a short length of time, probably during the second century BC.

material record. Lynn Meskell (2001, 24) has argued that archaeology is especially suitable for the study of identity – that “the very materiality of our field – the historical depth of monuments and objects, their visibility in museums, their iconic value” – presents us with, in effect, the stuff of people’s lives. As will be discussed in the following section, the relationship between people’s lives or identities and the material objects with which those lives are shared can be close and complex. We saw above that terrets have often been used as tools with which to attempt identification of the age or provenance of an assemblage. This approach has resulted in various efforts to link particular terret forms with particular peoples or tribes in Iron Age Britain, but usually without deeper consideration of the interplay between the identities of those peoples and their material culture. For example, in her publication of the Stanwick (Melsonby) hoard from North Yorkshire, MacGregor (1962) attempts to link items in the assemblage to certain tribes, and to certain regional “schools” of metal-working. Despite the location of the assemblage in north-east England, she attributes much of the material to the Dobunni – believed to have inhabited south-west England – and notes south-western “Belgic” and south-eastern “Icenic” influence. She suggests that the southern character of the assemblage could be due to the movement northward of Dobunic and Belgic peoples around the time of the Roman invasion.

The typologies move on through winged forms to crescentic, knobbed and platform-decorated forms, which are generally considered to have emerged around or perhaps slightly before the time of the Roman invasion. Leeds (1933, 120-121) notes the resemblance between the enamel quatrefoil created by the accidental flattening-out of one of the wing-pairs of a Polden Hills terret into a platform, and the enamel quatrefoils found on the deliberately flat platforms of the subsequent form, demonstrating a likely continuity of design. Apart from Spratling’s typology, which due to its southern British geographical remit includes no massive examples, the typologies all end with the massive terret. This artefact type is thought to be post-Conquest in Britain, and to have been current until around the third century AD (Leeds 1933, 123; MacGregor 1976, 49; Palk 1992, 58). The existing typologies do not cover any terret forms found in Britain from outside the D-shaped series, with the exception of the massive type, which MacGregor (1976, 48) considers to be a post-Conquest product of British craftspeople “working under Continental influence”.

This interpretation poses many questions that MacGregor addresses either fleetingly or not at all. If the material can be attributed to specific tribes, why has the work of several different tribes all been found in one deposit? To which “tribe” does the deposit belong? Does the assemblage reflect an individual, or a larger group or community? What is the significance of the “Roman” influence on some of the pieces? Why is mostly horse-gear included in the assemblage? Why was the assemblage deposited at all? Was there an intended audience or recipient for the deposition? Why was this site chosen for the deposition? – what was the site’s significance for exiles or migrants? MacGregor’s decision to attribute the hoard to peoples from southern England may partly be due to the fact that, when she was writing, the areas between north-east and south-west England had yielded relatively little Late Iron Age material (Colin Haselgrove pers. comm.); and admittedly, the first publication of an assemblage is not necessarily the place to raise these interpretive questions. Nonetheless, questions like these need to be asked. If we attempt to interrogate the material in this way, we can seek out its significance to the individuals and communities who used it.

Terrets and identity The concept of identity has become a central concern in interpretive archaeology over recent decades, particularly among advocates of post-processualism, who have argued that processual archaeology’s focus on systems and the long-term caused it to lose sight of the past individual (Thomas 2000, 13). In Iron Age archaeology, the old culture-historic model – of a Britain inhabited by various discrete, named tribes, each with their own definable territory and their own esoteric material culture – is no longer widely accepted; instead, attempts to identify past peoples acknowledge the plurality of social identities. Fraser Hunter (2006, 286-288) states that we need to be aware of the potential of material culture to express “the existence of different identities at different scales”: even within one small region, some artefacts may reveal “a local level of cultural tradition”, while others may “mark affiliation to a wider world”.

The tendency of previous studies to focus on British D-shaped terrets and to ignore other forms involves some implicit assumptions about Iron Age British identity in general. The entire D-shaped series – including forms that appear around the time of the Roman Conquest – is traditionally treated as “native”, with the implication that the artefacts were made and used by native British

Identities are not only informed by locality or geographical reach, of course, but by gender, sexuality, age, health, religious beliefs and so on. All of these aspects of people’s lives can, to a greater or lesser extent, be reflected in the 19

Anna Lewis people. All other terret forms are treated as “Roman”, and are marginalised in discussions of Late Iron Age material culture in Britain. Questions of whether the non-D-shaped terrets were used by British peoples, the contexts in which British peoples may or may not have used them, and any distinctions that may or may not have been drawn by British peoples between D-shaped and other terret forms are left un-raised.

consume in due quantity and quality becomes a mark of inferiority and demerit” (Veblen 2007, 53). It is, of course, not only wealth that can be demonstrated through the consumption or display of particular objects. Objects can demonstrate aspects of a person’s identity such as religious belief, age and gender, either explicitly (the kara or steel bracelet worn by Sikhs as a reminder of their devotion to the Guru), or implicitly (a tobacco pipe, in present-day British society, could belong to a young woman, but would be far more likely to belong to an elderly man). As identified by Morin, this association between people and artefacts works in different directions. A kind of object can, through use, develop an association with a particular person or kind of person over time and gain iconic status as a result; consequently, people may acquire a particular item deliberately to project some desired aspect of their identity to wider society.

A theoretical approach to the study of terrets that draws on aspects of object theory can help address these issues. To begin with, it is useful to look at some of the ways in which objects and people inter-relate, in order to consider how material artefacts can convey information about the people who use them and the societies in which those people live. Terrets and object theory In archaeology, the study of material and social culture goes hand-in-hand: through material objects we hope to reach past people, and through study of past people we hope to understand the objects they have left behind. To view terrets simply as dating tools is to divorce them from the social contexts in which they were made and used: to divorce them from the people who made and used them. This is a restrictive approach, as it prevents us from recognising that the distinction between artefact and person is not absolute.

Although both Morin’s and Veblen’s discussions are concerned with modern western society, the idea that status and wealth are expressed through material consumption often arises in Iron Age archaeology. Iron Age artefacts that appear to have been expensive in terms of material or labour are frequently described as “status”, “élite” or “prestige” goods. Where they are found in association with human remains, the artefacts are typically used as the basis for an interpretation of the human as someone of high status. For instance, Mike Parker Pearson (1999, 56-57) considers that the “rare and elaborate” artefacts contained in certain East Yorkshire chariot burials indicate “royal status”.

Human-made objects such as terrets cannot exist without people; but what is more, it is extremely rare for people to live without the company of objects. In many societies, people are marked or defined by the objects they keep: a person’s sense of self can be deeply connected to his or her use of material culture. Describing France in the 1960s, Violette Morin claimed that certain kinds of object had become so closely associated with certain kinds of people that they had, in effect, traded places. It was not always the case, she argued, that intellectuals wore glasses, and that the bourgeoisie bought fridges: rather, it was the case that glasses signified the intellectual, and fridges signified the bourgeoisie. These objects had gained human characteristics, to such an extent that the objects were able to impart their characteristics onto people. She used the term “l’objet biographique”, or “biographical object”, to describe an object that was able to convey a person’s life and character. (Morin 1969.)

The concept that social status, material wealth and conspicuous consumption go hand in hand appears viable when looking at Iron Age Britain, where there can be marked differences in the quantity and apparent value of artefacts associated with individuals. At the Kirkburn cemetery in East Yorkshire, for example, grave K5 contained a chariot and a set of associated horse-trappings, plus an extremely rare iron mail coat, while in comparison grave K6 contained a stud, a hollow ring, a jet bead and an amber bead (Stead 1991). Nonetheless, it needs to be noted that this concept – of the link between personal wealth and conspicuous consumption – is not universal. Alfred Gell (1986) has written about the modern-day Muria people of central India, some of whom have found themselves suddenly wealthy as a result of advances in agricultural technology. Rather than display their wealth, which would risk alienating them from their peers, the newly rich among the Muria maintain the appearance of living in poverty, while accumulating valuable savings. Gell’s example of the Muria also illustrates the gap that can exist between the impression people wish to convey through their choices of consumption, and the reality of their lives.

In the situation described by Morin, a person’s possessions can be considered to indicate (or even to determine) that person’s position in society. If this is possible, then it is not surprising that in some circumstances a person may deliberately obtain and use a certain object in order to convey their status, or wealth, or some aspect of their character. This practice of “conspicuous consumption” was described in the context of the late nineteenth-century USA by Thorstein Veblen, who argued that people who are highest on the social scale will seek out the highest quality and most expensive goods primarily for personal comfort, but also for “honorific” purposes, because “the failure to

*

20

Vehicles for Thought: Terrets in the British Late Iron Age Turning to terrets, we can see evidence for these theoretical points. In chariot burial contexts, we see terrets that are personally associated with individuals: for instance, the coral-studded set of terrets from the female chariot burial at Wetwang Slack, excavated in 2001 (Stacey 2004). We cannot know whether the interred individual made use of the terrets in life, but it is likely that those who conducted the burial wished to convey an association between the individual and the grave-goods.

Terrets occur with some frequency at sites of apparent votive deposition. At Bury Hill, for instance, six D-shaped terrets were recovered along with various other copperalloy items of horse gear from a single deposit of soil thrown into the tops of two adjacent pits (Cunliffe and Poole 2000, 47). Radiocarbon testing on samples of animal bone from these pits produced a calibrated date range from the fourth to the second centuries BC (Garrow et al 2009, 113-114). Cunliffe and Poole draw on the deposition of the terrets to interpret elements of religious or cosmological belief at the Bury Hill settlement site: they note the layers of periodic fill and silting in the pits, and the emphasis on horse-trappings, and suggest that this “gives some hint of a periodicity within the belief system and underlines the significance attached at this time to the chariot” (Cunliffe and Poole 2000, 80). They note that an exceptionally high proportion of animal bone at the site was horse bone – 48.2% – and posit that the horse and chariot may have played significant rôles in both the social system and belief system at Bury Hill (ibid, 79).

On a symbolic level, these terrets with their rare and exotic decorations may have been intended to represent wealth and status, in line with Parker-Pearson’s (1999) interpretation of valuable grave-goods. Chariot-gear generally may be seen as evidence of a community with long-distance trading contacts and social or economic influence, as Melanie Giles (2008, 407) has pointed out. However, the Wetwang terrets are not only meaningful in themselves: their very deposition is significant. At least one of the coral studs appears to have been replaced with a stud made from enamel, which may indicate that the terrets had a working life, in the course of which they underwent damage and repair. That they were used but eventually buried invites several possible interpretations: they could have been the possessions of the dead woman, and their “lives” could have ended along with hers; they could have been heirlooms of some sort, unsuitable for general circulation after the end of the family’s line; they could have been “sacrificed” to facilitate the woman’s transition to an afterlife; and so on. Nor do we have to see the terrets as representing only the individual with which they were buried: to bury such high-value artefacts, removing them from circulation, could be seen as an ostentatious gesture on the part of the deceased woman’s family or community as a whole – an announcement that whoever buried the woman was sufficiently wealthy and secure to be able to dispose of these valuable assets. Perhaps the woman was not wealthy or prominent, but happened to die at a time of political tension, so a grand gesture was made of her interment. These interpretations are all speculative, but they illustrate how individual, familial or communal identities can be expressed through the use and deposition of terrets.

Taking the terrets together with other items of horse-gear recovered from the site (ibid., 47-51) and the exceptional proportion of horse-bone, Cunliffe and Poole have been able to sketch an impression of a society characterised by an “élite warrior-dominated group” (ibid., 81), for whom the horse and chariot possessed a significance at once practical, symbolic and cosmological. In this case the terrets have been treated as artefacts capable of disclosing elements of the social culture of the people who used them during the Iron Age: as vehicles for thought, ideas and identities. * It is quite possible, then, to see individual and communal identities and behaviours expressed through the use and deposition of terrets in Iron Age Britain. Indeed, we cannot encounter terrets without also encountering the activities and ideas of the people who used them in the past. The same applies, however, to those non-D-shaped terrets so often excluded from discussions of Iron Age Britain, and it is to those terrets that I would now like to turn. “Roman” and “British” terret types: a re-consideration

We see evidence for social or communal identities in the deposition of terrets in other contexts, such as votive hoards and ritual pits. Veblen (2007, 60) claims that the display of wealth through consumption is characterised by “waste”, or expenditure that does not serve human life or well-being on the whole. This idea is echoed by Jean Baudrillard (1998, 53), who has argued that every human society is geared towards structural excess; in a comment particularly relevant to Iron Age Britain, Baudrillard suggests that in some societies “excess” may be the portion of societal wealth set aside for the gods, and that in all societies excess is the prerogative of a privileged minority, intended to reproduce the privilege brought about by caste or class.

The previous section of this paper argued that human-made objects are never truly separable from the people who made and used them: any consideration of a human-made object must engage with those people whose selves, or identities, are bound up in the artefact, and with the choices those people made in the artefact’s acquisition, use and disposal. When looking at any item of material culture, we need to ask who used it, in what situation, and why. When looking at a society undergoing colonisation, such as Late Iron Age Britain, the identification of “native” and “colonial” material and social culture is a primary issue. Were particular kinds of material culture used by native people, or colonisers, or both? In the case of terrets, 21

Anna Lewis as we saw earlier, the traditional distinction between supposedly British and Roman terret types implies some major assumptions about cultural identities in Late Iron Age Britain. The D-shaped series is treated as native, and as belonging within the La Tène or Celtic tradition of material culture; other forms are treated as Roman, and consequently excluded from discussions of Late Iron Age material culture in Britain. (There is some ambiguity even in this distinction: while the British Museum’s “Technologies of Enchantment” database of Celtic art includes some examples of “Roman”-style terrets, it excludes others; Garrow and Gosden’s (2012) publication on the Technologies of Enchantment project does not explain this.)

The type seems to develop into the kind found in the Mainz area of the Rhineland, two examples of which are illustrated by Joseph Déchelette (1914, fig. 510). These examples each have a circular ring with a spacious central cavity, mounted upon a sturdy neck which in turn is mounted upon a rather perfunctory skirt. Gripping the base of the ring and extending up into the centre, almost to the top, is an ornate stalk. The stalks appear to be detachable, and have been found – without rings attached – at the oppidum at Stradonice, Bohemia (ibid.; Pič 1906, pl. XXIII no. 14), a site assigned by Déchelette (1914) to the first century BC, and also in late La Tène contexts at Karlstein, Bavaria (Reinecke 1965, nos. 1166, 1167). Elsewhere on Continetal Europe, terrets develop into the form found in the La Tène III chariot burial at Armentières (Mariën 1961, 49). M. E. Mariën (ibid., fig.68) illustrates one of the four terrets from Armentières: it consists of a latitudinal ovular ring set upon an angular neck. The neck sits on top of a broken skirt, which appears to extend sideways, but not to the front or back. The ring itself is deep, and funnels inwards to a recessed ovular cavity at its centre. Extremely similar terrets to those from Armentières are known from the oppidum at Manching (Jacobi 1974, taf.52), which was abandoned by around 50 BC. Some of the Manching terrets, dated by Jacobi to the Late La Tène period (ibid., 200; taf.52, nos. 809, 814-16), make clear the relationship between the type and the British massive terret: the resemblance is particularly notable at the bottom of the ring, just above the neck, where the two sides of the ring sweep towards each other in a motion that, if continued, would extend upwards into a form of the moulded central protrusion found in the British massive terret (Figure 5).

However, an assessment of Iron Age terret forms from across Europe demonstrates that the generally accepted distinction between British and Roman terrets is not clearcut. As a case study, I would like to take the “British” terret form most obviously linked to Continental terret types: the massive terret. Two pieces of yoke from Italy, illustrated in Lindenschmit (1858, heft.II, taf.5), between them bear three harness ornaments that appear near-identical to the British massive terret. These ornaments are described by Lindenschmit (ibid.) simply as “Roman”, although MacGregor (1976, 48) has described the massive type as, “devised by British craftsmen, working under Continental influence some time between the late first and the midthird centuries AD.” The earliest known relatives of the massive terret on mainland Europe are four terrets from a chariot burial at Nanterre, northern France, dated to La Tène II (Duval 1961, 68), or around 270/250-150/120 BC (Figure 4). These terrets each have a sub-circular keeled ring, and at the base a cube-like structure inside which the attachment bar is concealed. A decorative protrusion rises from the base into the centre, terminating about halfway towards the top of the ring.

From the above discussion, we can see that on the Continent the massive terret sequence seems to begin around the third to second centuries BC, with the terrets from Nanterre: terrets with a plain, keeled ring, a concealed attachment

Fig 5. Terret from Manching, Germany (Jacobi 1974, tafel 52.814)

Fig 4. Terret from Nanterre, France (Duval 1949, figure 4.3)

22

Vehicles for Thought: Terrets in the British Late Iron Age bar and a central protrusion. The type develops by the first century BC into the type with long central protrusion found at Mainz and, in a different direction, to the sort with ovular keeled ring and angular skirted plinth found at Armentières and Manching. By the time the massive terret surfaces in Britain in the first century AD, it has developed a thick, smooth ring and distinctive stubby central protrusion, whilst retaining the attachment bar concealed within the base. At some point in its journey from northern France to northern Britain, the massive terret has developed from a Continental La Tène artefact to one variously described as Roman (Lindenschmit 1858), as British (Kilbride-Jones 1935), as a British artefact influenced by Continental design (MacGregor 1976, 48), and as a Roman artefact influenced by European – especially German – design (Laing and Laing1986). Of course, it is quite plausible that La Tène-style terrets were encountered, imitated and modified by Romans on mainland Europe, who then brought Romanised forms to Britain: forms which were subsequently developed by British craftspeople. Hybridisation of material culture clearly occurred during the Roman Conquest; Davis and Gwilt’s (2008) work on the Seven Sisters hoard from Neath shows us that, from the first century AD onwards, terrets and other items of horse-gear continued to be made in Britain in the La Tène stylistic tradition, but sometimes involved Roman techniques of manufacture such as the inclusion of zinc in the copper alloy. It is not straight-forward to designate artefacts of this nature as either Roman or British. The somewhat tortuous evolution of the massive terret demonstrates especially well, however, the difficulty of assigning particular terret types – or, indeed, terret features – to Roman or native manufacturers, or to particular dates, as certain elements of terret design seem to have been taken up and abandoned, periodically, by craftspeople working in both the Roman and La Tène traditions.

Fig 6. Terret from Laumersheim, Germany (Kimmig 1944-50, tafel 3)

In keeping with this point, we can turn briefly to the terrets from Laumersheim, among the earliest examples known in Europe (Figure 6). These items – with smooth, circular rings and boxy skirts – have been dated to the fourth or fifth centuries BC (Kimmig 1944-50), but are similar to the type found at the later site of Stradonice (Pič 1906, pl. XXVI, 1) with plain, circular ring and plain skirt concealing an attachment loop or shank. In Britain the type seems to be related to the skirted terret, generally considered Roman. We can also briefly consider three double-ringed terrets from La Courte, Belgium, each with two small adjacent cavities set within recesses inside a latitudinal ovular ring (Mariën 1961, fig.21). The ring sits on top of a skirt which extends downwards, at both front and back. At either side of the neck the skirt extends outwards into flat arms. Underneath the skirt are the remains of an iron shank. (Figure 7.) Mariën (ibid., 49) dates the double-ringed terrets from La Courte to Continental La Tène III – from around 150/120 BC until Roman conquest – on the basis of their stylistic similarity to the harness rings of Armentières.

Fig 7. Terret from La Courte, Belgium (Mariën 1961, figure 21)

23

Anna Lewis they did on mainland Europe (Furger-Gunti 1991). As we are told by Tacitus, they were used to fight against the Romans in the uprisings of Boudicca and Calgacus (Church and Brodribb trans. 2003; Birley trans. 1999). Adoption of Roman horse-gear could be seen as evidence for the breakdown of a distinct British identity and for the hybridisation of elements of British and Roman identities, which would be particularly interesting in the context of warfare. If, however, many of these supposedly Roman artefacts were closely related to Continental La Tène designs, examples of which could perfectly well have been in circulation in Britain before the Roman invasion (indeed, from the Parc-y-Merich (Sheppard 1941) and Isleham (Britton 1960) hoards, examples are known of droppedbar and skirted terrets that pre-date the Roman invasion by centuries), then their association with invading Roman forces may not have been notable. To exaggerate the point: if, on the one hand, the adoption by British peoples of La Tène-influenced chariot fittings introduced by the Romans could be interpreted as an identification, or collaboration, with Roman culture, then adoption of the very same chariot fittings could equally be interpreted as solidarity with oppressed Roman subjects on the Continent. Both arguments are far too simplistic, but they serve to illustrate the ambiguity of the division that has traditionally been made between native and Roman terrets in Britain.

Fig 8. Skirted terret from Newport (PAS database, NMGW-7D0AA5, © Portable Antiquities Scheme / British Museum)

Conclusion Terrets were common items in Late Iron Age Britain, and had both practical and ceremonial functions. Their recovery from burials and from contexts of apparent votive or ritual deposition suggests that they were used in practices underlain by complex cosmological or sociological beliefs. We need to treat terrets not merely as a series of types, useful primarily for the dating of sites or other artefacts, but as material manifestations of social culture, of identities and ideas.

The type clearly resembles those terrets with skirts and arms that are increasingly well-known in Britain (e.g. NMGW-7D0AA5 on the PAS database: Figure 8), and which are generally thought of as Roman. The double-ring is unusual in Britain, but not unheard of: a double-ringed, skirted example is known from Chinnor in Oxfordshire (BERK-41B625 on the PAS database). As with massive terrets, it is plausible – indeed probable – that artefact types based upon La Tène design could have been modified by the Romans and then introduced into Britain. However, it is note-worthy that most of the terret features considered in Britain to be Roman – such as skirts, arms, attachment loops and dropped bars, doubleloops and shanks – may in fact have their origins in La Tène harness gear. This has certain implications if we are to use material culture as a basis for interpretation of sociocultural issues, behaviours and identities in Late Iron Age and Roman Britain.

It is also important that the British terret series is not viewed in isolation. Those terret forms traditionally described as “Roman” appear to be widespread in Britain post-Conquest, but may not have been perceived by native British peoples as Roman, or as particularly distinct from other terret forms. A thorough investigation into the contexts in which “Roman” and “British” forms are found will help us to understand which forms were used when, where, by whom, and why; but any future work on terrets in Late Iron Age Britain needs to address all known kinds, and needs to consider the people who used them. A division between “Roman” and “British” forms ignores questions of social culture in order to focus on stylistic attributes, and is in this sense superficial.

For example, taking a classic post-colonial theoretical approach to the interpretation of Romano-British material culture, it could easily be concluded that the adoption by native Britons of Roman-style horse-gear indicated interaction between British and Roman peoples, and the assimilation of aspects of Roman culture into a sphere – chariotry – that had once been resolutely British. War chariots apparently survived in Britain far longer than

Acknowledgements I would like to thank Professor Colin Haselgrove for his helpful comments on an earlier draft of this paper. 24

Vehicles for Thought: Terrets in the British Late Iron Age Bibliography

Laing, L. and Laing, J. 1986. Scottish and Irish metalwork and the Conspiratio Barbarica. Proceedings of the Society of Antiquities of Scotland 116, 211-221 Leeds, E. 1933. Celtic Ornament in the British Isles down to A.D. 700. Clarendon Press Lindenschmit, L. 1858. Die Alterthümer unserer heidnischen Vorzeit. Verlag von Victor von Zabern Macdonald, P. 2007. Llyn Cerrig Bach: A study of the copper alloy artefacts from the insular La Tène assemblage. University of Wales Press MacGregor, M. 1962. The Early Iron Age metalwork hoard from Stanwick, N. R. Yorks. Proceedings of the Prehistoric Society 28, 17-57 MacGregor, M. 1976. Early Celtic Art in Northern Britain. Leicester University Press Mariën, M. 1961. La Période de la Tène en Belgique: Le groupe de la Haine. Musées Royeaux d’Art et d’Histoire Meskell, L. 2001. Archaeologies of identity. In Hodder, I. (ed.), Archaeological Theory Today. Polity Press, 187213 Michon, É. 1925. Anneau émaillé trouvé à Eauze (Gers). Antiquaries Journal 5, 145-147 Morin, V. 1969. L’objet biographique. Communications 13, 131-139 Norman, P. 1904. [Untitled.] Proceedings of the Society of Antiquaries of London 20, 57-60 Palk, N. 1992. Metal Horse Harness of the British and Irish Iron Ages. Unpublished thesis (PhD), University of Oxford Parker Pearson, M. 1999. Food, sex and death: Cosmologies in the British Iron Age with particular reference to East Yorkshire. Cambridge Archaeological Journal 9(1), 4360 Pič, J. 1906. Le Hradischt de Stradonitz en Bohême. Trans. (into French from Czech) Déchelette, J. Karl W. Hiersemann Reinecke, P. 1965. Mainzer Aufsätze zur Chronologie der Bronze-Und Eisenzeit. Rudolf Habelt Verlag Sheppard, T. 1960. The Parc-y-Meirch hoard, St. George Parish, Denbighshire. Archaeologia Cambrensis 96, 1-10 Spratling, M. 1972. Southern British Decorated Bronzes of the Late pre-Roman Iron Age. Unpublished thesis (PhD), University of London. Spratling, M. 1979. The debris of metalworking. In Wainwright, G., Gussage All Saints: An Iron Age settlement in Dorset. Department of the Environment Archaeological Reports No.10, 125-149 Stead, I. 1979. The Arras Culture. The Yorkshire Philosophical Society Stead, I. 1991. Iron Age Cemeteries in East Yorkshire. English Heritage Thomas, J. 2000. Introduction: The polarities of postprocessualism. In Thomas, J. (ed.), Interpretive Archaeology: A reader. Leicester University Press, 1-18 Veblen, T. 2007 (first published 1899). The Theory of the Leisure Class. Oxford University Press

Baudrillard, J. 1970. The Consumer Society. Trans. Turner, C. 1998. Sage Birley, A. trans. 1999. Tactitus: Agricola – Germany. Oxford University Press Britton, D. 1960. The Isleham hoard, Cambridgeshire. Antiquity 34, 279-282 Church, A. and Brodribb, W. trans. 2003. Tacitus: The Annals & The Histories. Modern Library Cunliffe, B. and Poole, C. 2000. The Danebury Environs Programme: The prehistory of a Wessex landscape. Vol. 2 - Part 2: Bury Hill, Upper Clatford, Hants., 1990. English Heritage and Oxford University Committee for Archaeology Monograph No. 49 Davis, M. and Gwilt, A. 2008. Material, style and identity in first century AD metalwork. In Garrow, D. et al (eds), Rethinking Celtic Art. Oxbow, 146-184 Déchelette, J. 1914. Manuel d’archéologie préhistorique celtique et gallo-romaine. 2, Archéologie celtique ou protohistorique. Pt.3, Second age du fer ou époque de La Tène. Picard Duval, P.-M. 1961. Paris Antique: des origines au troisième siècle. Hermann Furger-Gunti, A. 1991. The Celtic war chariot. In Moscati, S. et al (eds), The Celts. Thames and Hudson, 356-370 Garrow, D. et al 2009. Dating Celtic Art: A major radiocarbon dating programme of Iron Age and early Roman metalwork in Britain. Archaeological Journal 166, 79-123 Garrow, D. and Gosden, C. 2012. Technologies of Enchantment? Exploring Celtic Art: 400 BC to AD 100. Oxford University Press Gell, A. 1986. Newcomers to the world of goods: Consumption among the Muria Gonds. In Appadurai, A. (ed.) The Social Life of Things. Cambridge University Press, 110-138 Giles, M. 2008. Seeing red: The aesthetics of martial objects in the British and Irish Iron Age. In Garrow, D. et al (eds), Rethinking Celtic Art. Oxbow, 59-77 Hunter, F. 2006. Artefacts, regions, and identities in the Northern British Iron Age. In Haselgrove, C. and Moore, T. (eds), The Later Iron Age in Britain and Beyond. Oxbow, 286-296 Jacobi, G. 1974. Werkzeug und Gerät aus den Oppidum von Manching: Band 5. Franzsteiner Verlag GMBH Jacobsthal, P. 1944. Early Celtic Art. Oxford University Press Jay, M. et al. 2012. Chariots and Context: New radiocarbon dates from Wetwang and the chronology of Iron Age burials and brooches in East Yorkshire. Oxford Journal of Archaeology, 31(2), 161-189 Kilbride-Jones, H.E. 1935. An Aberdeenshire Iron Age miscellany: (1) Stone circle at Foulanton; (2) Bronze terret from Rhynie, and distribution of the type. Proceedings of the Society of Antiqities of Scotland 69, 445-454 Kimmig, W. 1944-50. Ein wagengrab der frühen Latènzeit von Laummersheim. Germania 28

25

Anna Lewis Online sources British Museum. “Technologies of Enchantment” database of early Celtic art available at http://www. britishmuseum.org/research/projects/technologies_of_ enchantment/the_celtic_art_database.aspx [Accessed 18th February 2013] Portable Antiquities Scheme. Online database available at http://finds.org.uk/database [Accessed 15th February 2013] Stacey, R. 2004. Evidence for the use of birch-bark tar from Iron Age Britain. Past 47. Available at: http:// www.le.ac.uk/has/ps/past/past47.html [Accessed 15th February 2013]

26

(Ad)Dressing the Anglo-Saxon Body: Corporeal Meanings and Artefacts in Early England T. F. Martin 1 Early Anglo-Saxon dress, particularly that of women, could be an elaborate affair, often comprising several layers of undergarments, dresses, cloaks, mantles and headdresses, all held in place with elaborate brooches, girdles and belts, complemented by various pendants, purses and other accessories, and festooned with strings of beads. For all the many typological, chronological and technical studies of Anglo-Saxon dress objects, there has been almost no work looking at their use in structuring perceptions of feminine and masculine bodies. In this paper, I consider how different styles of dress current in the fifth and sixth centuries AD created different bodies, and how these manners of dressing guided movement, posture, gesture as well as emphasised different anatomical aspects of the male and female bodies differentiated by age and perhaps even ethnic identity. The core idea of this paper is that by bringing the body into the framework of artefactual analysis, we gain a more holistic comprehension of how objects and people work together to create identity. Dr Toby Martin is a British Academy postdoctoral research fellow at the Institute of Archaeology, University of Oxford. His research focuses on the role of women’s dress in the emergence of regional identity groups and kingdoms in Europe in the 4th to 7th centuries.

For most of the history of early Anglo-Saxon archaeology, its practitioners have not had to worry too much about anything beyond the artefact. From the 19th century through to surprisingly late in the 20th century, the study of objects for their own sake was sufficient to establish the chronologies, typologies and distributions upon which scholarship depended. The historical record, whose word was for the most part uncritically accepted, provided authoritative context (e.g. Leeds 1912; Myres 1969; see Moreland 2001).1 Criticism of this necessary if occasionally circumscribed work in the 1980s and 1990s led to the relative neglect of specialist artefact research. There was perhaps a tacit understanding that with chronologies and typologies now in place, archaeologists could progress with more interpretative aspects. Accordingly, the focus switched from the objects to their wider context and artefact studies became, for want of a better word, unfashionable. For instance, pursuing work on landscape and settlement archaeology was a serious concern given how cemeteryfocused early Anglo-Saxon archaeology had become. Advances in osteological methods, alongside the rise of the modern excavation report, permitted multivariate analyses of cemetery data. These tended to draw attention away from the technicalities of objects and led to an almost exclusive focus on grave contexts (Pader 1982; Lucy 1998; Stoodley 1999). Recently however, unprecedented numbers of metal-detected finds recorded by the Portable Antiquities Scheme (henceforth PAS, founded in 1997) have revealed the gaps that have grown in our knowledge. These recent finds have necessitated the creation of new typologies, the overhauling of old ones and the refining of chronologies (Suzuki 2009; Martin 2011; Kershaw 2013; Felder forthcoming; Weetch forthcoming). Because these metal-detected finds lack archaeological contexts,

the pendulum is beginning to swing back and attention is returning to the artefact. More crucially, new methodology and theory has expanded the role of artefact specialists beyond technical knowledge of the corpus, typology and chronology. It is now the challenge of specialists to coax these objects from the dry pages of catalogues, databases, PAS records and grey literature back into their human, lived context. Dress, Bodies and Identity Focus on the object, frozen in stasis in the catalogue or site report, has resulted in an over-emphasis on its visual properties, often at the expense of its tactility or kinetic potential. After all, artefacts exist only to interact with the human body (Hamilakis, Pluciennik and Tarlow 2002, 9). The potential of looking at body-object interaction is therefore enormous (Mauss 1973, 75) and for many types of artefact, largely unexplored. For instance, the habitual performance of everyday dress is a primary factor in the construction, or more specifically the embodiment, of identity. Looking at how objects guide and influence bodily movement and gesture permits insight into how our perception of the social world is essentially a material, intercorporeal, or body-to-body, experience (Mathews 2005, 2). Dress offers a uniquely privileged view of human-object interaction. Clothing shapes the body, emphasising particular aspects while hiding others. It can attribute meaning to the anatomy and also guides or even restricts movement. The intimacy between bodies and dress led to Terence Turner’s idea of clothing as a “social skin”: the culturally intelligible facade of a biological substrate (Turner 1980). For Marcel Mauss, who originated the idea of culturally variable body “techniques”, bodies quite explicitly constituted a symbolic language (Mauss 1973, 76). He suggested that the way in which we learn to move provides an instrument for creating social difference. In

( [email protected]) Institute of Archaeology 36 Beaumont Street Oxford OX1 2PG 1

27

T. F. Martin these views, the body is inscriptive and its morphology and movements are encoded cultural symbols. This concept, sometimes termed the ‘body as project’, can imply that the physical body is coached in certain movements and has adornments gradually added to it, all in an effort to express the inner identity of the individual. Hence, it relies on the pre-existence of a primary, interior self, a premise that has led to considerable criticism (Meskell 1996, 7-8; Gilchrist 2000, 91; Joyce 2005, 151). The idea originates from the Cartesian duality of body and mind and as such, it is an historically specific product of the Enlightenment (Meskell 1996, 3; Thomas 2002). Hence, there is a need to accept that the formation of identity is simultaneous with bodily performance. We must not only acknowledge the communicative and symbolic acts of the body and its adornments, we must also recognise the primacy of corporeality as the site of all lived experience (Meskell 2000; Fisher and Loren 2003, 225-6).

fabrics are occasionally found on the corrosive products of these metallic items and microscopic examination of them, alongside the positions of dress-fasteners, such as brooches, clasps, pins or buckles on the body has allowed specialists to establish the basic range of garments worn by the early Anglo-Saxons (Owen-Crocker 2004; Walton Rogers 2007). This work permits us to look beyond the artefact to its wider context as part of dress. Therefore, it is not only possible to look at different styles of brooch, but also the different styles of dress they fastened. In short, we can relate brooch form to brooch function. There are, of course, some problems with reconstructing everyday dress from mortuary costumes. The dead do not dress themselves. These were intentional selections of clothing assembled specifically for a ritual event. However, we have no reason to believe that such costumes were not also those worn in everyday life. For instance, continental sculpture depicting Germanic dress replicates the clothing evidenced in graves (Owen Crocker 2004, 445, 49-50). Additionally, most brooches recovered from graves show signs of wear and more than twenty per cent of some brooch types show substantial repairs or other adaptations demonstrating frequent usage (Martin 2012, 56). Nevertheless, not all objects placed in graves were necessarily the possessions of the deceased. For instance, it seems likely that some objects may have been funerary gifts (King 2004). Likewise, there are reasons to believe that not all objects deposited with children were necessarily their possessions (Crawford 1993, 85). However, for the most part, there is little reason to believe that dress objects were not buried with the individuals that had worn them in life. Indeed, perhaps the personal nature and intimately corporeal associations of these objects demanded their removal from circulation through mortuary deposition in the first place.

In accordance with the development of body theory in the social sciences, archaeology has progressed from seeing dress as a straightforward reflection of identity. Dress constructs social perceptions of the body and is therefore intimately involved in the experience and embodiment of identity; especially those forms of identity in which the body is paramount, such as gender (Joyce 2005, 142-3; Perry and Joyce 2001, 68). Judith Butler (1991; 1993), suggests that sex and gender are performances rather than personally held qualities. Identity does not pre-exist, awaiting expression through clothing and other bodily performance, but “dress is an embodying activity” (Fisher and Loren 2003, 228, see also Nordbladh and Yates 1990, 223 and Yates 1994, 51). Dress, therefore, is not a skin of pre-formed, normative symbols draped passively over the body. Some of the key generative structuring principles of identity, or the repeated, habitual actions from which identity emerges, lie in how people choose to wear garments or ornaments, how this material culture shapes and gives meaning to the body and how clothing influences or inspires particular movements and gestures. This paper will explore how dress was involved in the embodiment and enactment of identity in the early AngloSaxon period, not only through passive, symbolic display, but also through active performance, movement and the experience of dressed corporeality. This is a key means through which a focus on artefacts can reveal the much wider aspects of the human, social, and above all bodily world that not only had this material culture deposited in graves, but also experienced the world through them on a daily basis.

Due to the range and quantity of metal dress-accessories deposited in women’s graves, our knowledge of feminine dress is substantial. There is less certainty about the clothing of younger girls, but from the age of about ten to twelve, women were frequently buried wearing peplos dresses. This was a tubular garment, generally made of a woollen twill, pulled up over the body and fastened over the shoulders by a pair of brooches (Stoodley 1999, 117; Gowland 2006, 148, see Fig. 1). Although there were variations on this basic garment, the peplos seems to have been a standard dress for most adult women in the early medieval Germanic world. Its origins, however, are more ancient and widespread. The garment is known from classical Greek art, Iron Age Europe and even Asia (Walton Rogers 2007, 148-9). In early Anglo-Saxon England, it seems that between eighty and ninety-five per cent of adult women wore this garment in the grave (Walton Rogers 2007, 144). In the east of England, the presence of little clasps worn at the wrists indicates that the peplos dress was often layered over the top of a sleeved, generally linen, garment (Walton Rogers 2007, 154). Additionally, from about the age of eighteen, a small number of these women began to be buried wearing cloaks fastened at the

Early Anglo-Saxon Dress Although archaeologists and metal-detectorists have recovered early Anglo-Saxon brooches, wrist-clasps, beads, buckles and other dress fittings in their thousands, substantial fragments of textile do not survive from this period. Hence, much of what we know about early Anglo-Saxon costume comes from dress ornaments recovered from graves. The impressions of long-decayed 28

(Ad)Dressing the Anglo-Saxon Body: Corporeal Meanings and Artefacts in Early England

1. Anglo-Saxon feminine garments. From left to right: peplos, single-fastened cloak, dualFig 1. AFigure nglo-Saxon feminine garments. From left to right: peplos, single-fastened cloak, dual-fastened cloak. Drawing by Lynne Martin fastened cloak. Drawing by Lynne Martin.

throat of mid-chest by one or two large bow brooches costume was relatively simple and conventional, remaining respectively (Martin 2011, 280-3, see Fig. 1). This is virtually unchanged for the entire first millennium AD where the linking of brooch styles and brooch functions (Walton Rogers 2007, 199). Whether men regularly wore becomes important. While smaller dress-fasteners such cloaks is open to question and this matter will be subject to as disc, saucer, small long, annular or openwork brooches some discussion below. A sleeved garment worn by some fastened the peplos dress, large bow brooches, such as men, popular at least by the seventh-century, was known cruciform and great square-headed brooches tended to as the ‘warrior jacket’. It often features on the decorative fasten cloaks (see Fig. 2). Head-coverings did not require metalwork of this century (such as the Sutton Hoo helmet brooches, so we know slightly less about them, but a thin plaques) and is possibly evident from a handful of early gauzy fabric occasionally survives on the external surface Anglo-Saxon graves (Walton Rogers 2007, 210). of brooches, which suggests head-coverings were more Certain types of feminine jewellery were characteristic of common than we imagine (Walton Rogers 2007, 157). particular geographical regions. For instance, cruciform Other accessories included festoons of glass and amber brooches are restricted to the east and northeast of England beads strung between brooches, as well as items worn (these are the cloak-fastening brooches mentioned above, at the girdle including purses, keys and decorative items see Fig. 2b), while saucer brooches are only common in known as girdle-hangers, which also tended to be worn the southwest (Fig. 2d). Kent meanwhile has its own only by women over the age of about eighteen (Stoodley Figure 2. The main types of early Anglo-Saxon brooch. (a) Great square-headed brooch from distinctive assemblages including button and small square1999, 117). Lakenheath, Suffolk. (b) Cruciform brooch from West Stow, Suffolk. (c) Small Great distribution headed brooches (Fig.long 2e).brooch Thesefrom localised Men’s dress, on the other hand, featured few if any metal patterns demonstrate the important role feminine dress Chesterford, Essex. (d) Saucer brooch from Welford, Berkshire (PAS find number BERK-A6EC93). dress-accessories. At the most, a simple iron or copperplayed in regional identity, but the supposed reflection of (e)sufficed. Button brooch from the Isleof of their Wightclothing (PAS find number (f) Disc brooch from alloy buckle However, traces Angles,IOW-F0C2A5). Saxons and Jutes in these distributions is a longoccasionally survive on more substantial iron objects in standing and contentious issue (Leeds 1945; Lucy and the grave, such as weaponry. Fortunately, continental Reynolds 2002, 10). Most would now accept, however, that depictions of Germanic dress from the late Roman period the contention arises more from conflicting understandings help bolster this evidence. In early Anglo-Saxon England, Page 1 of of 3 early medieval ethnicity than it does from the existence men generally wore tunics and trousers (Owen Crocker of regional, named identities per se. Most research now 2004, 111-118; Walton Rogers 2007, 199-205). Although suggests that Anglian or Saxon ethnicity during this period men sometimes fastened them with a buckle, male graves was restricted to an elite tier of society and was probably frequently preserve no dress-accessories whatsoever. This relatively fluid and negotiable, depending more upon to

29

Figure 1. Anglo-Saxon feminine garments. From left to right: peplos, single-fastened cloak, dualT. F. Martin

fastened cloak. Drawing by Lynne Martin.

Figure 2. The main types of early Anglo-Saxon brooch. (a) Great square-headed brooch from Fig 2. The main types of early Anglo-Saxon brooch. (a) Great square-headed brooch from Lakenheath, Suffolk. (b)

Cruciform brooch from West Stow, Suffolk. (c) Small long brooch from Great Chesterford, Essex. (d) Saucer brooch Lakenheath, Suffolk. (b) Cruciform brooch from West Stow, Suffolk. (c) Small long brooch from Great from Welford, Berkshire (PAS find number BERK-A6EC93). (e) Button brooch from the Isle of Wight (PAS find number

IOW-F0C2A5). (f) D(d) isc brooch Butlerfrom ’s Field , Gloucestershire . (g) A(PAS nnular find brooch from Empingham II, Rutland. Chesterford, Essex. Saucerfrom brooch Welford, Berkshire number BERK-A6EC93). (h) Penannular brooch from Broughton Lodge, Nottinghamshire. All drawings by T. F. Martin.

(e) Button brooch from the Isle of Wight (PAS find number IOW-F0C2A5). (f) Disc brooch from whom one owed allegiance than one’s actual lineage or origins (Pohl 2002; Curta 2007; Moreland 2010, 159192). Ethnicity is unlikely to have been perceived as a primordial quality shared by all individuals originating from a certain region of northern Germany. All thePage same,1 women in eastern, so-called Anglian, regions of England may have substantiated claims to origins or allegiances though their manner of dress. After all, if migration does not explain the dramatic changes in dress between the Roman and Anglo-Saxon periods, there must have been some, probably political, motivation to assume Germanic costume. Brooches were not badges that explicitly stated origins and allegiances. Rather, they were part of a whole package of material symbols, including bodily habits, which made some women appear different to most of the community.

major social purpose is to screen and control the display of the anatomical body. As such, clothing can act to accentuate or diminish the importance of anatomical features through exposure or concealment. This is not a straightforward process. For instance, the ofnecessarily 3 covering of male and female genitalia, observed as a bare minimum in the dress of most societies, accentuates, just as much as it attempts to diminish, the social significance of this anatomy. Clothing also adapts the shape and size of bodies, whether through the enlargement of the body with a cloak, the constriction of a waist with a corset, or the elevation of height through high-heeled shoes. How the individual uses clothing to perform these morphological transformations, exposures and concealments is just as important to identity formation as the specific style of brooch they wear, or the weave and colour of a textile.

The above discussion serves to outline the range and complexity, as well as the archaeological and historical significance, of the identities that early Anglo-Saxon dress embodied. However, my principal concern in what follows is not to delve too far into what these identities actually meant, but to illustrate that dress is not capable of constructing identity all by itself. I am to go beyond the categorisation of clothing styles and interrogate the meaningful relationships between bodies and dress.

The costumes of some early Anglo-Saxon women were substantial. As described above, the standard adult feminine costume was a peplos dress worn over another garment, often evidenced as a sleeved under-dress by decorative copper-alloy clasps that fastened the cuffs. Over the age of eighteen, however, some women donned a cloak fastened by one or two substantial brooches. These maximally dressed women are of particular interest to this discussion. On average, about one in every six women above this age wore these cloak-fastening cruciform brooches as a mortuary costume in eastern England (Martin 2011, 234). It is not unusual to find some of these individuals wearing two cloaks, which were worn over a peplos dress and a sleeved undergarment, making a costume of at least

Shaping, Concealing and Exposing the Body Although clothing is foremost a practical measure for protecting against the sun, rain, cold and the wind, its

30

(Ad)Dressing the Anglo-Saxon Body: Corporeal Meanings and Artefacts in Early England four layers, not counting other garments that may not have required brooches for fastening and are therefore not evidenced in graves, such as headscarves or hoods (Walton Rogers 2007, 196-7). These costumes could be fastened with up to five or so substantial brooches and were often festooned with beads, pendants and gathered in girdles from which may have hung yet more accessories. Not only would these maximally dressed women have cut an imposing figure, it is also important to note that their clothing almost entirely screened the anatomical body away from public view. The transformation of the anatomical body into a clothed, ‘social skin’, however, does not diminish its importance. On the contrary, the mystification of the bodies of some Anglo-Saxon women may have actually accentuated their significance.

fastening, or were fastened with non-durable materials that would not preserve in graves (e.g. wooden buttons, toggles or stitching). Yet the evidence from textile analysis only offers five examples of men interred with weaves comparable to those from women’s cloaks (Walton Rogers 2007, 207). In sum, and despite these cautions, we do not have any evidence from early Anglo-Saxon England to suggest that cloaks had the same importance for men as they did for women, whose corpses they regularly clothed and the fastenings of which were among the highest achievements of early Anglo-Saxon craft. Men’s bodies were more prominently on display, whilst at the same time less elaborated with clothing. Depictions of the body during this and slightly later periods supports this trend toward the exposure of male bodies and the concealment of female ones. The famous Finglesham buckle (Hawkes and Grainger 2006, 413), for instance, depicts a warrior wearing nothing but a belt. That this was an everyday costume seems highly unlikely, but it may represent a character from myth, or perhaps a costume assumed for ritual purposes, perhaps even warfare. Although they are from a slightly later period, a goldfoil figure (‘goldgubber’) from Sorte Muld on Bornholm, Denmark depicts an apparently bearded individual, naked but for a necklace (Hedeager 2011, 132). Other gold-foil figures similarly represent apparently nude individuals wearing just a belt or a necklace, one of which, from the obvious depiction of a phallus, is unambiguously male (Watt 1999, 180). A bare-chested man is also depicted fighting a bear on the Torslunda plates from Sweden. Gold bracteates, also from Scandinavia, occasionally depict probable male figures (some seem to have beards) wearing nothing but short skirts (see Hedeager 2011, 206 for examples). Closer to home in Anglo-Saxon England, a small number of tiny figurines have recently been discovered (see Brundle 2013), which also offer examples of exposed male bodies. Although some are clearly male, others are potentially female or perhaps even intentionally transgendered. The most famous of them from Carlton Colville, Suffolk, is male and is clearly clothed, yet the morphology of their body, including an obvious phallus, shows through. Although depictions of women are in any case rare in the art of this broad period, the few Scandinavian examples known from gold-foil figures, the characters depicted on gilt silver pendants from Oland, and the embroidered bird-woman figure on a textile from Oseberg, Norway (Hedeager 2011, 77) all show the feminine form swathed in clothing.

For men, however, the situation was quite different. Masculine clothing displayed an anatomical form very close to the one it covered. Tunics are generally thought to be, at the most, knee-length (e.g. Walton Rogers 2007, 206), while trousers seem to have been an exclusively male garment. Although Gale Owen-Crocker (2004, 50) has put forward the possibility of peplos dresses that fell just below the knee, we must still consider that women probably wore them over longer undergarments and often under substantial cloaks. Whether or not men wore cloaks is a problematic question (Owen Crocker 2004, 104-11; Walton Rogers 2007, 206-9). Considering the English climate, it seems highly unlikely that cloaks were a strictly feminine garment. Yet, male graves provide very little evidence for them. A small number of individuals buried with brooches have been determined by osteological examination to be biological males. These include an individual with iron penannular brooches at Abingdon in Oxfordshire, three from Mucking in Essex with shoulder brooches and a further example from Collingbourne Ducis in Wiltshire wearing a Romano-British disc brooch in a Roman fashion (Stoodley 1999, 34; Walton Rogers 2007, 206-7). A few biologically male burials with cruciform brooches are also known from Norton in Tees, Empingham II in Rutland and Castledyke South in Lincolnshire (Martin 2011, 217). Some of these examples include other items generally found with women, which, rather than suggesting they might be transgender individuals (see Knüsel and Ripley 2000), makes it more likely that they represent occasional errors in skeletal sexing. Such errors have a frequency of up to ten per cent (Henderson 1989), which these exceptions do not exceed. There were some reservations about the osteological sexing at Norton whose male and female population had “less clear-cut sexual distinctions” (Sherlock and Welch 1992, 107; Stoodley 1999, 10). At Empingham II, poor skeletal survival alongside inconsistencies recording and archiving the site may have been responsible for some error (Timby 1996, 16). Some of these male burials are exceptional for other reasons: iron penannular brooches and Romano-British disc brooches are unusual occurrences in any AngloSaxon context and may represent surviving RomanoBritish fashions. Despite these reservations, it is perfectly possible that men wore cloaks that did not require

In sum, there was a sharp contrast in shaping, concealing and exposing male and female bodies, reflected in both everyday, practical dress and art. Particular women in the grave, and depictions of particular men, whether they were real or mythical, seem to accentuate this difference. Mary Douglas conceived the body as a symbol of society at large and consequently suggested that its formal control, including the manner in which it is covered and clothed, is a matter of boundary maintenance and a means of controlling what might be otherwise dangerous or polluting 31

T. F. Martin (Douglas 1966, 138). Dressing differently is therefore not only a method of maintaining boundaries between social groups, but it is also a means of managing the cultural unease around the sheer physicality of bodies and their biological functions. Concealing particular bodies under swathes of textile does not so much obscure them as betray a heightened social concern. Although the anatomical masculine body was a prominent cultural symbol, it is that which Anglo-Saxon society never depicted, but hid away from view under layers of textile and festoons of jewellery, which might have been their major anxiety. Even though the material wealth of these women suggests they were of an elite tier in society, Douglas’s (2003, 78) summation “bodily control is a symbol of social control” hints at the possible existence of underlying patriarchy seeking to control and subdue the female Anglo-Saxon body.

maintained these boundaries in the social memory, which was perhaps part of the motivation for furnished burial in the first place. Configuring the Body The idea of clothing as a ‘social skin’ implies that garments and other adornments do not just conceal the body. Dress also inscribes the body’s surface and demarcates its anatomical elements. Focal points on the body are signified by concealment, exposure or ornamentation. Alternatively, bodily foci can arise from habits of dressing. In the case of most early Anglo-Saxon women’s clothing, something of the female body’s schema can be represented in a series of prevalent, recurring constellations (Fig. 3). The peplos fasteners obviously emphasise the shoulders, but when a cloak is worn over the top of this, attention is drawn to a central point at the throat or two points on the chest. Further, sleeve-clasps highlight the wrists, while a host of objects sometimes suspended from a girdle or belt emphasises the waist. Considering the large size and elaborate nature of early Anglo-Saxon jewellery, these items go far beyond practical necessity, while some, such as girdle-hangers and beads, lack a practical function altogether. In some cases, the oversized nature of some brooches even compromised their practical function (Hines 1997, 293). Because these configurations occur only on women, their gendered nature is obvious: men never have these locations highlighted in graves or pictorial sources. Hence, these constellations highlighting wrists, shoulders, the throat, and the breast were likely to have become synonymous with femininity. They also changed with age. Women over the age of twelve accentuated the shoulders and then a subset of these individuals, after the age of about eighteen, wore cloaks that accentuated the throat or breast. This is also about the age when some women began to accentuate the waist with girdle items. Whatever their specific meanings, these configurations may well betray how the coincidental structures of clothing and its methods of practical fastening gave rise to the creation of gendered bodily differences.

Related to this idea of boundary maintenance is Roberta Gilchrist’s suggestion that the prohibition of an externally acknowledged sexuality among celibate later medieval nuns acted to construct a psychical interior, an internal and private experience of sexuality (Gilchrist 2000, 91). While this situation is obviously historically specific, it successfully highlights the differences between external, public sexuality and the personal experience of the body. This comparison also exists between aspects of male and female bodies in the early Anglo-Saxon period. Concealing particular women’s bodies under layers of textile mystified and privatised their sexuality. Quite the opposite was true for Anglo-Saxon men and this divergence would doubtlessly have affected these individual’s experienced sexuality. Although these divergent treatments of the male and female body tempt us to construct dichotomies relating male/female to public/private, such structural tropes are well explored in the anthropological and archaeological literature and turn out to be rarely as polarised as they might initially appear (Rosaldo 1974, 23-35; Reiter 1975; Strathern 1980; Gilchrist 1997, 55). In the case of early Anglo-Saxon dress, although some women’s bodies were concealed and private, they also bore elaborate decoration, undoubtedly for public display. Hence, we cannot confine them to a private, domestic sphere. In fact, if we envisage these magnificently dressed women as embodiments of regional identities, quite the opposite is true. These women’s bodies, one way or another, were symbols of social groups. Gilchrist’s internalised identities (above) also relate to Rissman’s (1988) observations of Harappan burial. Rissman observed how similar internal and uncontested identities constructed through clothing become publicly contestable through their display at funerals (cited in Joyce 2005, 149). The same may have been true for the Anglo-Saxon burial rite. Additionally, whoever was responsible for dressing the corpse would have literally transgressed these layers of privacy, transforming the deceased into a contestable, public object or even an embodied ideal of the social group to which she belonged. Simultaneously, however, the public display of the deceased at the funeral in their full regalia actually

Artistic representation may also contribute to this discussion. Depictions of men frequently decorate feminine jewellery as part of Salin’s Style I, an art style found throughout much of Europe during this period (Salin 1904; Haseloff 1981). Intriguingly, we therefore have two layers of symbolism: depictions of the male body that ornament and attribute meaning to the female one (Martin 2013, 10-11). In most northern European art, and particularly that of Scandinavia, attention focuses on the male head: moustaches, mouths, eyes, brows and cheeks, but only rarely is the torso depicted. The male head, therefore, seems to have been a particularly prevalent icon (see also Danielsson 2002, 181). There is not a single depiction of a female face, to my knowledge, from early Anglo-Saxon England, although they do feature in the slightly later Scandinavian art I have already described. Masculine bodily symbolism potentially also features on the zoomorphic terminals of cruciform brooches, 32

Butlers Field, Gloucestershire. (g) Annular brooch from Empingham II, Rutland. (h) Penannular brooch from Broughton Lodge, Nottinghamshire. All drawings by T. F. Martin.

(Ad)Dressing the Anglo-Saxon Body: Corporeal Meanings and Artefacts in Early England

Figure 3. Early Anglo-Saxon feminine body locations highlighted by jewellery. (a) Peplos fasteners

Fig 3. Early Anglo-Saxon feminine body locations highlighted by jewellery. (a) Peplos fasteners and wrist-clasps. wrist-clasps. (b) Single-fastened wrist-clasps. Dual-fastened cloak and (b)and Single -fastened cloak and wrist-claspscloak . (c) Dand ual-fastened cloak(c) and wrist-clasps. D rawing bywrist-clasps. T. F. Martin.

Drawing by T. F. Martin. which may depict phallic imagery (Martin 2013, 9). This decoration occasionally also features on girdle-hangers. The possible phallic 25 symbolism of some pot-stamps, made from pressing cruciform brooch terminals into the wet clay, has also been observed (Ravn 2003, 122). Additionally, there20is the obvious phallus on the figurine from Carlton Colville, mentioned above. These gendered focal points, including feminine shoulders, chests and 15 masculine head and phallus, provide waists, alongside the some insight into the construction of sexuality through male and female bodily symbolism. There is no precedent 10 for suggesting that the cultural significance of all this symbolism necessarily pre-existed the clothing. Rather, it may have been constructed through manners of dressing 5 and artistic representation.

hand side’, sums up the negative, dangerous and polluting symbolism of that which is not ‘right’. Hence, laterality is not just a means of differentiating the sides of the body; it is often fundamental to systems of organising the whole world.

Early Anglo-Saxon feminine dress occasionally displays laterality. For instance, larger brooches used to fasten a cloak at the throat (generally cruciform brooches and great square-headed brooches) show a very slight preference for being worn with their head-plate to the wearer’s right-hand side and the foot to their left (Fig. 4). It may be significant that large bow brooches depicted on Scandinavian goldfoil figures also share this orientation (e.g. Watt 1999, 176). Because the pin runs from the head to the foot and pushing the pin through the fabric in this direction is Laterality is crucial to an easier movement for a right-handed individual, this 0 body symbolism (Hertz 1973; Mauss 1973, 78). For instance, the hand on which one might might be expected. Foot points upward Nonetheless, this habit may have Foot points Foot points to left Foot points to right wear a ring, the side on which one wears a sword, or the generated subtle, lateralised bodily meanings. We may Downward shoulder shoulder shoulder over which one drapes a cloak all contain clues find more significance in Gale Owen Crocker’s (2004, 48) as to theFigure values 4. attributed to the ‘right’ or the ‘left’, which observation that whensingle-fastened brooches were worn singly to fasten Rates of directional brooch use among 54 women wearing cloaks drawn are often seen as strong or weak, sacred or profane, clean a variation of the peplos that was fastened only on one from (Needham the following cemeteries: et they al 1987), Empingham or polluting 1973). Therefore,Morning clothingThorpe and (Green shoulder, were generally wornIIon(Timby the left 1996), side. While jewellery, along with bodily performance, help transform such a garment elaborates the left shoulder with jewellery, Castledyke South (Drinkall and Foreman 1998), Sewerby (Hirst 1985), Bergh Apton (Green and symmetrical bodies into asymmetrical ones. From the it also exposes the right hand side of the torso, with the body, these symbols radiate outward and laterality is peplos passing underneath the right arm. Not only does this a fundamental means of organising space from the show a much more obvious lateralisation of the dressed perspective of the embodied individual (e.g. Cunningham body, it represents a more demonstrably conscious decision 1973). In English, the right-hand side is abstracted even to a cultural trend. During the fourth and fifth Page 2toofconform 3 further to mean moral virtue (‘righteousness’), foremost century, the Roman crossbow brooch, a likely precedent (‘right-hand man’), or even, quite simply, ‘correct’. Our of the Germanic and Anglo-Saxon cruciform brooch, was term ‘sinister’, deriving from the Latin for ‘on the lefta highly prominent symbol of masculine, perhaps even

33

Figure 3. Early Anglo-Saxon feminine body locations highlighted by jewellery. (a) Peplos fasteners and wrist-clasps. (b) Single-fastened cloak and wrist-clasps. (c) Dual-fastened cloak and wrist-clasps. Drawing by T. F. Martin.

T. F. Martin

25 20 15 10 5 0 Foot points Downward

Foot points to left Foot points to right Foot points upward shoulder shoulder

of directional brooch women wearing single-fastened cloaks drawn Fig 4.Figure Rates 4. of Rates directional brooch use amonguse 54 among women 54 wearing single -fastened cloaks drawn from the following cemeteries: Morning Thorpe (Green et al 1987), Empingham II (Timby 1996), Castledyke South (Drinkall and Foreman from the following cemeteries: Morning Thorpe (Green et al 1987), Empingham II (Timby 1996), 1998), Sewerby (Hirst 1985), Bergh Apton (Green and Rogerson 1978), Great Chesterford (Evison 1994), Finglesham Castledyke South (H (Drinkall and Foreman 1998), (Hirst 1985), Bergh Apton (Green and awkes and Grainger 2006) and Sewerby Westgarth Gardens (West 1988).

official, power (Janes 1996; Butcher and Bayley 2004, (Mauss 1973, 86; see also Douglas 1966; 1970). The social 214). The crossbow brooch, like the emperor’s cloakof movement is of course what defines a movement Page 2 ofcontrol 3 clasp, was always worn on the right shoulder (Janes 1996, or gesture as unnerving, impolite, awkward or threatening in 127; Walton Rogers 2007, 175). Elite circles in late and the first place. Being eminently communicative, gesture is post-Roman northwest European world would almost fundamental to the embodiment of identity (Mathews 2005). certainly have been acutely aware of this protocol, given its For instance, the idea that gender is less something that prominence in statuary, ivories and mosaics. Additionally, people ‘are’, but more something that people ‘do’ (Barrett associations between masculinity and crossbow brooches, 1988) leads us to expect gender identity to include ways of worn on the right, are evident in late fourth-century moving or particular gestures or postures that were citational, Roman Britain at the Lankhills cemetery in Winchester, so public and repetitive (Perry and Joyce 2001, 66-67). this may well have been a costume encountered regularly Of course, we can only speculate over the situations under in late Roman Britain (Clarke 1979, 151, 170, 263). In which early Anglo-Saxons learned particular movements. the Roman world, men as well as women commonly wore Potentially, this was an extended part of the gift exchange brooches (Bayley and Butcher 2004, 214). The prevalence ritual when objects such as elaborate jewellery were of brooches and cloaks among senior, elite women in the received. Certain acts of labour, such as weaving, were post-Roman period is therefore extremely interesting. The domestic tasks performed in front of other household Anglo-Saxon tendency of wearing a brooch on the left members, who may have observed gestures, movements shoulder was possibly part of this conscious subversion and techniques in order to emulate them. Given the close and transformation of masculine Roman power into a links between some forms of masculinity and warfare Germanic feminine status symbol. (Härke 1990), particular masculine deportments would have Moving the Body been learnt during training for combat or even warfare itself. Sources of idealised movements or gestures, whether or not Bodily movements and gesture are intimately linked they had a practical equivalent, include the figurines and with dress and ornamentation. The movement of a hand, gold foil figures already discussed (Brundle 2013, see also for instance, might be emphasised by the presence of Joyce 2005, 146; Perry and Joyce 2001, 68). rings, bracelets or wrist-clasps. All of these items go towards making that gesture an effective and recognised Movement is often an underrated aspect of dressed behavioural code. Wearing particular objects or garments performance in archaeology, but there are elements of might also make one choose to move differently, depending Anglo-Saxon costume that suggest movement really was a on practicality, or perhaps the desire to display these conscious concern. The glitter of bronze, gold and silver objects to maximal effect. Marcel Mauss established the jewellery and the manner in which much of it was decorated idea that particular manners of moving are socially learnt in relief, would have created flashing, reflective surfaces as and saw them as a means of controlling potential disorder the individual moved. More obvious, however, would have

34

Rogerson 1978), Great Chesterford (Evison 1994), Finglesham (Hawkes and Grainger 2006) and (Ad)Dressing the Anglo-SGardens axon Body : Corporeal Westgarth (West 1988). Meanings and Artefacts in Early England

5. A set of three spangles on a dress pin from Carlton Scroop, Lincolnshire Figure 5. A Figure set of three spangles on a dress pin from Carlton Scroop, Lincolnshire (LCNGR:1995.3509). Photo by T. F. Martin reproduced with permission with from permission Lincoln Collection . (LCNGR:1995.3509). Photo by, T. F. Martin, reproduced from Lincoln Collection. been the noises made by a slightly rarer article of feminine garb: what we refer to as spangles, and Germans refer to as Klapperschmuck (Fig. 5). These two names summarise the properties of these small tags: they not only spangled and shone, but they also made a jangling noise that accompanied movement and gesture, chorused by the rattling of beads and the chiming of metal objects such as keys, pendants and brooches. Small iron bells very occasionally occur in AngloSaxon graves (e.g. Butler’s Field, Gloucestershire, grave 148) and although these are perhaps unlikely constituents of everyday costume, their possible uses and associations with movement merit at least a mention. In sum, AngloSaxon women’s costumes were made with movement in mind. The high rates of breakage among brooches (Martin 2012), suggests these items really were moved around in on a daily basis and were not just brought out for occasional, sedate use.

however, was not the only slightly unpractical garment. The unwieldy nature of many vastly oversized brooches has already been mentioned. Re-enactors have also reported that sleeve-clasps are particularly uncomfortable and completely unpractical for weaving, given their tendency to become tangled in the threads of a loom (Owen Crocker 2004, 59). Of course, we must bear in mind that this complaint may have more to with the way in which these modern re-enactors have learnt to use their bodies wearing modern clothes. Additionally, the whole point of a clasp is that it allows a sleeve to be undone and rolled up, perhaps precisely with such activities on mind.

Interpretations of movement in masculine attire are more speculative. We might assume that the wearing of trousers and tunics, as opposed to dresses and other drapery, allowed for and perhaps even encouraged more active or vigorous movements, with every gesture being Returning to the cloaks worn by older women, we can highly visible compared to the concealing nature of some recall that there were two major types: one fastened at the feminine costumes. Masculine identity was closely linked throat by a single brooch and one fastened on the chest by to the weapons they were buried with and presumably two brooches (Fig. 1). The second of these, though not owned and used (Harke 1990; Dickinson and Harke 1992, necessarily as restrictive as it appears, certainly screens 55-60). Not only would the use of weaponry in combat much of the arms. The covering of parts of the anatomy has have required a highly specialised training in particular considerable implications for thinking about gesture. Overt bodily techniques, but we might also imagine that items gesticulation was slightly limited in this garment and was such as spears and shields were held at rest according therefore perhaps deemed inappropriate for these women. to particular conventions. Indeed, Sue Brunning has On the other hand, gesticulation with the arms is in some suggested that some 6th- and 7th-century sword hilts show ways emphasised by a billowing of the cloak, or a reaching a pattern of wear consistent with the repeated resting of round its edges to articulate a gesture visibly. Additionally, Page 3 ofa3hand on hilt when the weapon was sheathed at the hip this garment was not suitable for manual labour, or any (Brunning 2013), a posture perhaps as affected then as it particularly energetic activity. The dual-fastened cloak, seems to us now.

35

T. F. Martin Occasionally, osteological analyses can also reveal interesting examples of pathologies that would have caused an individual to move differently. The most common among these is osteoarthritis, which can cause anything from mild discomfort to excruciating pain and would certainly have affected the way in which an individual walked or used their hands, for instance. Additionally, looking at musculoskeletal markers can provide information on the build and particular muscular developments of an individual, providing clues concerning repeated movements. A particularly poignant example of pathology, however, can be found in grave 15 at Quarrington in Lincolnshire (Dickinson 2004, 30). The woman in this grave showed severe indicators of muscle wasting, perhaps brought on by suffering from tuberculosis for a number of years indicated by lesions in most of her vertebrae. A further lesion on a rib indicated a lung infection still active when she died at the age of about eighteen. The specialist’s report suggests that the severity of atrophy on all limbs indicated very lengthy periods of bed rest, if not paralysis. However, this grave was also among the richest in the cemetery and the individual was one of the maximally dressed and cloaked women this paper has touched on several times. This example sits alongside Heinrich Härke’s (1990, 36) observation that weapons were occasionally interred with men not fit to wield them, including small children, individuals with malunited fractures, and even one instance of a man with severe spina bifida. It is therefore as interesting to think how this woman from Quarrington fitted into a world in which the body and its movements were fundamental to social interaction as it is reassuring to learn that her physical disabilities did not exclude her from the considerable social statuses her jewellery and clothing indicate.

their root, based on the traditional necessities of typology, chronology, distributions and the archaeological contexts of objects in graves. Debts to these more traditional concerns should not be underrated and neither should that which is beholden to innovations in identity and body theory. The role of the artefact specialist is to both address and exceed the technical concerns that previously narrowed their contributions to the wider discipline. The archaeology of early Anglo-Saxon costume demonstrates how the body’s symbolic meanings emerge from the performance of dress. Whether garments were consciously impractical, such as the wearing of enormous items of jewellery, or unconsciously practical, such as the direction from which one pins a cloak, they constituted some of the generative structuring principles of bodily meanings in early Anglo-Saxon England. Without these considerations, we lack a holistic understanding of the social construction of identity through its embodiment in human corporeality. None of the observations outlined above offer straightforward interpretations of meaning from bodily conventions. For instance, the gendered concealment and exposure of the anatomical form through dress appeals to conclusions based on the public masculine body and the private feminine body, but they do not stand up to critical examination. Similarly, even though we can detect lateralised codes of behaviour, it is difficult to get at the cosmological significance of the left and the right in the Anglo-Saxon mind. Nevertheless, the cultural prevalence of the male face as a cultural icon on the one hand and the obvious social concern for particular women’s bodies swathed in layers of textile on the other provokes further interpretation. If we follow Mary Douglas’s idea of the body as a microcosm of society (Douglas 1970), perhaps it is possible to pursue a metaphor of the early AngloSaxon household symbolised by male ‘heads’ and female ‘bodies’. Through this trope, male and female bodies are merged and scaled up to represent the household, which, as the basic economic and political unit of early AngloSaxon society, becomes further enlarged as a miniaturised model of society. Such an interpretation does not force a prioritisation of either gender in terms of power, nor does it rely on the flawed public/private dichotomy. Rather, it invokes a potential model through which early AngloSaxons understood and experienced the world through a corporeal metaphor.

Summary and Conclusion Above all, approaching dress-related artefacts through their relationships with the body brings us closer to understanding identity as an embodied performance. As such, my discussion here encourages artefact-based studies to reach beyond the confines of the catalogue and to explore the human, essentially social anthropological, context. My approach shows how bodily interactions with material culture create social meanings and how comprehensions of the human body depend on such performances. Nevertheless, the observations here were, at

36

(Ad)Dressing the Anglo-Saxon Body: Corporeal Meanings and Artefacts in Early England Bibliography

Gilchrist, R. 2000. ‘Unsexing the body: the interior sexuality of religious medieval women’ in Schmidt, R.A. and Voss, B.L. (eds.) Archaeologies of Sexuality, 89-103. Routledge. Green, B. and Rogerson, A. 1978. The Anglo-Saxon Cemetery at Bergh Apton, Norfolk. Norfolk Archaeological Unit: East Anglian Archaeology Research Report 7. Green, B., Rogerson, A. and White, S.G. 1987. The AngloSaxon Cemetery at Morning Thorpe, Norfolk. Norfolk Archaeological Unit: East Anglian Archaeology Research Report 36. Hamilakis, Y., Pluciennik, M. and Tarlow, S. 2002. ‘Introduction’, in Hamilakis, Y., Pluciennik, M. and Tarlow, S. (eds.) Thinking Through the Body: Archaeologies of Corporeality, 1-21. Kluwer/Plenum. Härke, H. 1990. ‘“Warrior graves”? The background of the Anglo-Saxon weapon burial rite’, Past and Present 126, 22-43. Haseloff, G., 1981. Die Germanischen Tierornamentik der Völkerwanderungszeit. Studien zur Salin’s Styl I. De Gruyter: Vorgeschichtliche Forschungen 17. Hawkes, S.C. and Grainger, G. 2006. The Anglo-Saxon Cemetery at Finglesham, Kent. Oxford Institute of Arcaeology: Oxford University School of Archaeology Monograph 64. Hedeager, L. 2011. Iron Age Myth and Materiality. Routledge. Henderson, J. 1989. ‘Pagan Saxon cemeteries: a study of the problems of sexing by grave goods and bones’ in Roberts, C.A., Lee, F. and Bintliff, J. (eds.) Burial Archaeology. Current Research, Methods and Developments, 77–83. British Archaeological Reports (British Series) 211. Hertz, R. 1973. ‘The pre-eminence of the right hand: a study in religious polarity’ in Needham, R. (ed.) Right and Left: Essays on Dual Symbolic Classification, 3-31. University of Chicago Press. Hines, J. 1997. A New Corpus of Anglo-Saxon Great Square-Headed Brooches. Boydell Press. Hirst, S. 1985. An Anglo-Saxon Saxon Inhumation Cemetery at Sewerby East Yorkshire. Department of Archaeology, University of York: York University Archaeological Publications 4. Janes, D. 1996. ‘The golden clasp of the Late Roman state’ Early Medieval Europe 6(2), 127-153. Joyce, R.A. 2005. ‘Archaeology of the body’ Annual Review of Anthropology 34, 139-158. Kershaw, J. 2013. Viking Identities: Scandinavian Jewellery in England. Oxford University Press. King, J.M. 2004. ‘Grave goods as gifts in early Saxon burials (ca. AD 450-600)’ Journal of Social Archaeology 4, 214-238. Knüsel, C. and Ripley, K. 2000. ‘The berdache, or manwoman in Anglo-Saxon England and early medieval Europe’ in Frazer, W.O. and Tyrell, A. (eds.) Social Identity in Early Medieval Britain, 157–191. Leicester University Press.

Barrett, J.C. 1988. ‘Fields of discourse: reconstructing a social archaeology’. Critique of Anthropology, 7(3), 5-16. Bayley, J. and Butcher, S. 2004. Roman Brooches in Britain: A Technological and Typological Study Based on the Richborough Collection. Society of Antiquaries of London: Reports of the Research Committee of the Society of Antiquaries of London 68. Brundle, L. 2013 (in press). ‘The body on display: exploring the role and use of figurines in early AngloSaxon England’. Journal of Social Archaeology. Brunning, S. 2013. The “Living” Sword in Early Medieval Northern Europe: An Interdisciplinary Study. PhD thesis, University College London. Butler, J. 1990. Gender Trouble: Feminism and the Subversion of Identity. Routledge. Butler, J. 1993. Bodies that Matter: On the Discursive Limits of ‘Sex’. Routledge. Clarke, G. 1979. Pre-Roman and Roman Winchester, Part II: The Roman Cemetery at Lankhills. Clarendon Press. Crawford, S. 1993. ‘Children, death and the afterlife in Anglo-Saxon England’ Anglo-Saxon Studies in Archaeology and History 6, 83-91. Cunningham, C.E. 1973. ‘Order in the Antoni house’ in Needham, R. (ed.) Right and Left: Essays on Dual Symbolic Classification, 204–238. University of Chicago Press. Curta, F. 2007. ‘Some remarks on ethnicity in medieval archaeology’ Early Medieval Europe 15(2), 159-185. Dickinson, T.M. 2004. ‘An early Anglo-Saxon cemetery at Quarrington, near Sleaford, Lincolnshire: report on excavations, 2000-2001’ Lincolnshire History and Archaeology 39, 24–45. Dickinson, T.M. and Härke, H. 1992. ‘Early Anglo-Saxon shields’ Archaeologia, 110, 1-94. Douglas, M. 1970. Purity and Danger: An Analysis of Concepts of Pollution and Taboo. Penguin Books. Douglas, M. 2003. Natural Symbols: Explorations in Cosmology (reprint of 2nd edition). Routledge. Drinkall, G. and Foreman, M. 1998. The Anglo-Saxon Cemetery at Castledyke South, Barton-on-Humber. Sheffield Academic Press: Sheffield Excavation Reports 6. Evison, V.I. 1994. An Anglo-Saxon Cemetery at Great Chesterford, Essex. Council for British Archaeology: CBA Research Report 91. Felder, K. forthcoming. A Key to Early Anglo-Saxon Identities? Girdle-hangers in 5th and 6th Century England. DPhil thesis, University of Cambridge. Fisher, G. and Loren, D.D. 2003. ‘Introduction to special section Embodying Identity in Archaeology’, Cambridge Archaeological Journal, 13(2), 225-30. Gilchrist, R. 1997. ‘Ambivalent bodies: gender and medieval archaeology’ in Moore, J. and Scott, E. (eds.) Invisible People and Processes: Writing Gender and Childhood into European Archaeology, 42-58. Leicester University Press.

37

T. F. Martin Leeds, E.T. 1912. ‘The distribution of the Anglo-Saxon saucer brooch in relation to the Battle of Bedford’ Archaeologia 63, 159-202. Leeds, E.T. 1945. ‘The distribution of the Angles and Saxons archaeologically considered’ Archaeologica 91, 1-105. Lucy, S.J. 1998. The Early Anglo-Saxon Cemeteries of East Yorkshire: An Analysis and Re-Interpretation. British Archaeological Reports (British Series) 272. Lucy, S. and Reynolds, A. 2002. ‘Burial in early medieval England and Wales: past, present and future’ in Lucy, S. and Reynolds, A. (eds.) Burial in Early Medieval England and Wales, 1-23. Society for Medieval Archaeology: SMA Monograph Series 17. Martin, T.F. 2011. Identity and the Cruciform Brooch in Early Anglo-Saxon England: An Investigation of Style, Mortuary Context, and Use. PhD thesis, University of Sheffield, http://etheses.whiterose.ac.uk/2369 (accessed 8/2/2013). Martin, T.F. 2012. ‘Riveting biographies: the theoretical implications of early Anglo-Saxon brooch repair, customisation and use-adaptation’ in Jervis, B. and Kyle, A. (eds.) Make-do and Mend: Archaeologies of Compromise, Repair and Re-Use, 53-65. British Archaeological Reports (International Series) 2408. Martin, T.F. 2013. ‘Women, knowledge and power: the iconography of early Anglo-Saxon cruciform brooches’ Anglo-Saxon Studies in Archaeology and History 18, 1-17. Mathews, S. 2005. ‘The materiality of gesture: Intimacy, emotion and technique in the archaeological study of bodily communication’. Published online by ‘Semioticon: the Open Semiotics Resource Centre’, http://www.semioticon.com/virtuals/archaeology/ materiality.pdf (accessed 8/2/2013). Mauss, M. 1973. ‘Techniques of the body’ Economy and Society 2, 70-88. Meskell, L. 1996. ‘The somatization of archaeology: institutions, discourses, corporeality’ Norwegian Archaeological Review 29(1), 1-16. Meskell, L. 2000. ‘Writing the body in archaeology’ in Rautman, A.E. (ed.) Reading the Body: Representations and Remains in the Archaeological Record, 13-21. University of Pennsylvania Press. Moreland, J. 2001. Archaeology and Text. Duckworth. Moreland, J., 2010. Archaeology, Theory and the Middle Ages. Duckworth. Myres, J.N.L. 1969. Anglo-Saxon Pottery and the Settlement of England. Clarendon Press. Nordbladh, J. and Yates, T. 1990 ‘This perfect body, this virgin text: between sex and gender in archaeology’ in Bapty, I. and Yates, T. (eds.) Archaeology After Structuralism, 222-237. Routledge. Owen-Crocker, G. 2004. Dress in Anglo-Saxon England: Revised and Enlarged Edition. Boydell Press. Pader, E.-J. 1982. Symbolism, Social Relations and the Interpretation of Mortuary Remains. British Archaeological Reports (International Series) 130.

Perry, E.M. and Joyce, R.A. 2001. ‘Providing a past for “Bodies That Matter”: Judith Butler’s impact on the archaeology of gender’ International Journal of Sexuality and Gender Studies 6(1/2), 63-76. Pohl, W. 2002. ‘Ethnicity, theory and tradition: a response’ in Gillett, A. (ed.) On Barbarian Identity: Critical Approaches to Ethnicity in the Early Middle Ages, 221–239. Brepols. Ravn, M. 2003. Death Ritual and Germanic Social Structure (c. AD 200-600). British Archaeological Reports (International Series) 1164. Reiter, R.R. 1975. ‘Men and women in the south of France: public and private domains’ in Reiter, R.R. (ed.) Toward an Anthropology of Women, 252–282. Monthly Review Press. Rissman, P. 1988. ‘Public displays and private values: a guide to buried wealth in Harappan archaeology’ World Archaeology 20, 209-228. Rosaldo, M.Z. 1974. ‘Woman, culture, and society: a theoretical overview’ in Rosaldo, M.Z. and Lamphere, L. (eds.) Woman, Culture and Society, 17–42. Stanford University Press. Salin, B., 1904. Die Altgermanische Thierornamentik. Wahström and Widstrand. Sherlock, S.J. and Welch, M. 1992. The Anglo-Saxon Cemetery at Norton, Cleveland. Council for British Archaeology: CBA Research Report 82. Stoodley, N. 1999. The Spindle and the Spear: A Critical Enquiry into the Construction and Meaning of Gender in the Early Anglo-Saxon Burial Rite. British Archaeology Reports (British Series) 288. Strathern, M. 1980. ‘No nature, no culture: the Hagen case’ in MacCormack, C. and Strathern, M. (eds.) Nature, Culture and Gender, 174–222. Cambridge University Press. Suzuki, S. 2008. Anglo-Saxon Button Brooches. Typology, Geneaology, Chronology. Boydell Press. Timby, J.R. 1996. The Anglo-Saxon Cemetery at Empingham II, Rutland. Oxbow. Turner, T.S. 1980. ‘The social skin’, in Cherfas, J. and Lewin, R. (eds.) Not Work Alone: A Cross-Cultural View of Activities Superfluous to Survival, 112-140. Sage. Walton Rogers, P. 2007. Cloth and Clothing in AngloSaxon England AD 450-700. Council for British Archaeology: CBA Research Report 145. Watt, M. 1999. ‘Kings or gods? Iconographic evidence from Scandinavia gold foil figures’ Anglo-Saxon Studies in Archaeology and History 10, 173-183. Weetch, R. forthcoming. A cultural biography of brooch use in England and Northwest Europe, AD 700-1100. PhD thesis, University of Reading. West, S.E. 1988. Westgarth Gardens Anglo-Saxon Cemetery Suffolk: Catalogue. Suffolk County Planning Department: East Anglian Archaeology Research Report 38. Yates, T. 1994. ‘Frameworks for an archaeology of the body’ in Tilley, C. (ed.) Interpretative Archaeology, 31-72. Berg.

38

All Form One and One Form All: the relationship between pre-burial function and the form of early Anglo-Saxon cremation urns. Gareth Perry 1 Abstract1

The Use-Alteration of Anglo-Saxon Cremation Urns

There is an enduring assumption that early Anglo-Saxon cremation urns were produced for the funeral and that they served no domestic purpose. Recently, however, evidence has emerged which demonstrates that urns are likely to have taken part in the production and consumption of fermentation produce before their burial. These findings require a reexamination of urn form and how it might relate to pre-burial function; such an re-examination forms the basis for this discussion. By drawing on a range of ethnographic studies of pottery producing and using societies, this paper demonstrates that function is primarily the means by which producers and users of pottery classify their vessels. The cognitive decisions made by these societies in identifying and naming their vessel types are explored, illustrating that that perceptions of proportion, in particular morphological ratios and vessel size, are of major concern when distinguishing between functional classes. It demonstrates that characteristics frequently used to define Anglo-Saxon types, such as whether a pot is ‘shouldered’, ‘baggy’, or ‘bichonical’, are in fact minor variants of much wider themes and as such they should not form the basis of Anglo-Saxon pottery taxonomies. With this in mind a new classification is developed, revealing that the early Anglo-Saxons had clear perceptions of the acceptable size, shape and volume of their vessels, and that the different Anglo-Saxon vessel forms can be directly related to their preburial functions.

In order to contextualise this study it is essential to briefly review the primary evidence for pre-burial use of urns, namely use-alteration analysis (for a full discussion see Perry 2011). Use-alteration analysis seeks to identify ‘the chemical or physical changes that occur to the surface or subsurface of ceramics as a result of use’ (Skibo 1992, 45). Changes may include the build up of soot deposits, absorption of phosphorous and fatty acids, accumulation of mineral salts within the fabric, and the discolouration and breakdown of a vessel’s surface (Hally 1983, 4; Skibo 1992, 106). Examination of 958 urns from Cleatham and 504 from Elsham revealed that at least 71% showed signs of having being used prior to burial. More specifically, around a third of complete vessels exhibited use-alteration characteristics indicative of having contained lactic acid fermented foodstuffs, such as dairy products and brewed beverages, particularly beer (Perry 2011; 2012; 2013) (Figure 1). As one of the most basic relationships in ceramic studies is the link between form and function, this evidence for pre-burial use impels us to consider how different forms might relate to pre-burial function.

Introduction The pre-burial origins of early Anglo-Saxon cremation urns have long been debated. For many years the general consensus has been that urns were produced especially for the funeral (e.g. Laing and Laing 1979; Leahy 2007b; Richards 1987; Wilson 1965). However recent re-evaluation of the evidence contradicts these earlier conclusions (Perry 2011; 2012; 2013). A programme of use-alteration analysis on pottery recovered from the cemeteries of Cleatham and Elsham (North Lincolnshire) and 80 non-funerary find sites that surround them demonstrates that these urns took part in domestic activities before their burial. In particular, they seem to have been involved in the production and consumption of fermented produce. We can no longer, therefore, view ceramic assemblages from early Anglo-Saxon cemeteries as simply funerary in nature; they must be seen as domestic assemblages preserved in a funerary context. Given the broad range of urn form-types, there is a distinct possibility that different forms had specific pre-burial functions; this paper considers these forms in the context of their pre-burial use.

Fig 1. Internal

pitting of

Elsham Urn EL76NT,

a use-

alteration characteristic which indicates that a vessel has held lactic acid fermented produce.

Notably,

this

form of attrition is most common on decorated early

Anglo-Saxon

pottery,

whilst

sooting

commonly on undecorated pottery (see

 University of Sheffield

1

39

occurs

more

Perry 2011). Photograph taken by author.

Gareth Perry ‘Form’ the Beginning Virtually any review of Anglo-Saxon pottery studies must begin with the work of J.N.L. Myres (1969; 1977). Myres’s primary motivation was to harness the supposed potential of pottery for tracing Germanic migration into England, by drawing decorative parallels between Continental vessels and types found in England (Leahy 2007a; Richards 1987). While his German counterparts used form as the primary indicator of vessel date, Myres (1969, 24) dissented, suggesting that such an approach could ‘prove fatal’ when classifying pottery produced in fifth- and sixthcentury England. Myres’s own assumptions, however, proved equally fatal. He argued that the migrant potters ‘were a chance assortment of uprooted amateurs’ who would be ‘unlikely to maintain a typological exactitude of form’, and who applied decoration ‘to any shape that might emerge from their unskilled efforts’ (Myres 1969, 22-5). He believed that whilst decorative development could be traced throughout the fifth and sixth centuries, it would be foolish to expect potters to produce ‘clear-cut ... well-defined ceramic types’ in an ‘intelligible sequence of evolving forms’ (Myres 1969, 22-25). Yet, as Julian Richards (1987, 26) highlights, it is unclear why crossing the Channel should have rendered potters incapable of producing particular vessel types, or why form should stagnate whilst decoration flourished. Despite his dismissive attitude, Myres’s (1977) Corpus reveals that he did, in fact, attempt to place vessels into a number of morphological groups. It is worth considering how these groups were developed and used to classify pottery.

Fig 2. Earlier classifications are based on the contour and position of the vessel shoulder. Myres records vessels 3258 and 1403 as ‘sub-biconical’ and ‘shouldered’, respectively, yet they are exactly the same shape (see scaled overlay, top right). Hamerow’s classification describes vessels 138.3 and 139.2 as globular jars, yet they are not the same form, whist Myres describes both 2319 and 632 as ‘biconical’ urns, again they are not the same form. Urn 3402 is classified as ‘sub-biconical’ and 1571 as ‘shouldered’ (redrawn from Myres 1977 Figures 94, 98, 105, 128, 151, 208, and 225 and Hamerow 1993, Figures 138.3 and 139.2).

It was in his 1969 volume Anglo-Saxon Pottery and the Settlement of England that Myres developed a typology of form, initially to classify undecorated pottery, but later applied to decorated pottery. Based on parallels between English and Continental vessels, Myres organised and dated undecorated pots into the following types: biconical (fifth-century), hollow-necked (‘early’), sub-biconical (thought to have a ‘long life’), shouldered (c.AD 500), bowls (‘early’), globular (fifth- and early sixth-century), and low-bulbous forms (mid to late seventh-century) (Myres 1969, 25-8). Founded entirely upon morphology, this approach is at odds with his primary system of classifying and dating, clearly contradicting his doubt regarding the existence of a ‘sequence of evolving forms’ (Myres 1969, 22-5).

demonstrated: biconical urns have a ‘less pronounced carination at or not far away from the mid-point of the profile’; hollow necked urns are a ‘variety of the biconical type’; and sub-biconicals ‘derive’ from the biconical and hollow necked types (Myres 1969, 25-8). In the absence of a well-defined system the whole approach becomes highly subjective, with subsequent scholars forced to decide for themselves what might constitute, for example, a ‘typical’ sub-biconical urn (Richards 1987, 27).

Commenting on these form-types, Myres (1977, 1-2) reported that there were an ‘infinite series of shapes’ and that ‘each of the main [form] groups merge imperceptibly into others’. As a consequence, many urns could easily be placed into more than one group. For example, Myres classified urn 3258 as a sub-biconical urn with an everted rim, but 1403 as a shouldered urn with hollow neck and everted rim (Figure 2). Yet these urns are entirely the same shape; thus, the reader must ask where sub-biconical urns end and shouldered urns begin. These problems reveal more about Myres’s categorisation than the pottery itself. The lack of clarity in his taxonomy can be readily

The vagaries inherent in Myres’s typology provided the stimulus for many subsequent studies into early AngloSaxon pottery forms. For example, although they based their taxonomies on Myresian types, Kenneth Fennel (1964), Catherine Hills (1976, cited in Richards 1987, 30) and Helena Hamerow proceeded to produce classifications based on well-defined criteria. Fennel (1964, 225-7, 2637) suggested that vessels from Loveden Hill (Lincolnshire) should be classified as follows: biconical ‘when the 40

All Form One and One Form All change of diameter is at about the mid point of the height’; shouldered, when the change is ‘above’ this point; and baggy, when ‘below’. Hills provided similar criteria in her analysis of urns from Spong Hill (Norfolk) – if a vessel’s maximum diameter was above its mid-height then it should be classified as “shouldered”, but “baggy” if it was below (Hills 1976, cited in Richards 1987, 30).

work (e.g. urns 632 and 2319, Figure 2) and as Hamerow’s forms are based on his typology it is no surprise to find the problem replicated. Perhaps the most rigorous and enlightening studies of Anglo-Saxon vessel form were those undertaken by Richards (1982; 1987). Rather than attempting to (re) define Myres’s types, Richards used statistical methods to investigate the relationship between form, the decoration of cremation urns, and the individual contained therein. Using Principal Component Analysis (PCA) on urns from Spong Hill and Mucking, he identifed three morphological ratios which accounted for 93% of the variability in form. Ratio 1, the ‘width ratio’ of a vessel, accounted for 79% of variability and was described as follows:

Fennel’s classifications were further defined by morphological ratios. He determined each urn’s ‘mouth’ and ‘height’ ratio (ratio of mouth diameter-to-maximum diameter and the ratio of height-to-maximum diameter, respectively) then used these ratios to define ‘typical’ types. He classified jars, for example, as vessels with a height ratio of 0.9 or above, whilst a wide-mouthed bowl was a vessel with a height ratio less than or equal to 0.8 and a mouth ratio greater than or equal to 0.7 (Fennel 1964, 225-7, 2637). Whilst these definitions provide future analysts with a clear set of characteristics with which to classify pottery, it is unclear how and why Fennel arrived at these ‘cutoff’ values (Richards 1987, 28). Despite this, his approach revealed some interesting relationships, including links between particular forms and types of decoration (Fennel 1964, 267, 294). Unfortunately, Fennel did not pursue this avenue of research, proceeding instead (á la Myres) to draw Continental parallels, applying and charting the progress of Germanic settlement in Lincolnshire.

Maximum diameter Height

(Ratio 1)

‘Wider’ vessels had a large Ratio 1 value, and narrow vessels a smaller value. The second most significant characteristic described the ‘shoulderedness’ of a vessel. It accounted for 9% of the variability and was expressed by Ratio 2: Height of maximum diameter

Whilst Fennel, Hills and Myres concentrated on pottery recovered from cemeteries, Helena Hamerow’s (1993) classification focused on forms from the domestic sphere, specifically the settlement at Mucking (Essex). Using proportions such as the ratio of height-to-rim diameter, as well as the profile and location of maximum diameter, she developed a dendrogram that allowed hierarchical categorisation of form. This divided pots into jar and bowl forms, with bowls being defined as vessels with a rim diameter greater than their height, and jars as those with a rim diameter less than their height (Hamerow 1993, 40). Jars were subcategorised as biconical, globular, shouldered, straight-sided ovoid and low bulbous, whilst bowls were split into simple and complex forms. Unlike Fennel and Hills, Hamerow did not attempt to define any of these terms, providing instead examples of typical types with reference to particular vessels, the dendrogram, and a type-series summary. For example, globular jars were described as ‘[r]estricted, usually necked, complex profile; max girth lies roughly at the centre point; rim diameter min 57 [mm], max 360 [mm], average 151 [mm]’ (Hamerow 1993, 40-1).

Height

(Ratio 2)

‘Shouldered’ vessels had a large ratio value whilst those with a low value were said to be ‘baggy’. Finally, Ratio 3 described the level of neck restriction and accounted for just 5% of variability: (Maximum diameter - Rim diameter) (Height - Height of maximum diameter) (Ratio 3) Vessels with a low Ratio 3 value were said to be ‘open mouthed’, whilst ‘restricted’ vessels had a higher value. Despite accounting for only 5% of the variability in form, it was this ratio which demonstrated ‘[t]he greatest number of significant associations between particular shapes and other attributes of the cremation’ (Richards 1982, 3644). It appears, then, that a vessel’s accessibility was an important concern in the selection of an appropriate urn for burial. This is significant because ethnographic studies demonstrate that the neck restriction is often a major concern in determining the functional suitability of pottery vessels (Rice 2005, 241) and although Fennel calculated a similar characteristic (‘mouth’ ratio), neck restriction has never been at the forefront of studies of early Anglo-Saxon vessel form.

Although Hamerow’s typology provides analysts with a clearer indication of what constitutes a typical form-type, it is still problematic. Consider, for example, the globular jar outline presented above; with such a range of rim diameters one would expect that the smallest derive from smaller versions of the largest jars, yet as vessels 138.3 and 139.2 demonstrate (Figure 2), this is not the case. Both pots are classified as globular jars, yet they are clearly not the same form. The same problem is present in Myres’s

Richards’s ratio-based approach revealed that the decoration of cremation urns, and the ages and genders

41

Gareth Perry of the individuals they contained, could be directly related to form. For instance, diagonal lines occurred less frequently on urns with high shoulders (high Ratio 2), but more commonly on low and mid-shouldered urns (low Ratio 2) (Richards 1982, 43-4). Frequently, taller vessels were found to contain the remains of adults, while children were more commonly found in shorter vessels. Males were concentrated in vessels with large maximum diameters, while women were more frequently in vessels with above average rim diameters (Richards 1987, 1349). Whilst infants and children were buried in small pots (i.e. lacking in height), infants were distinguishable within this group by being more commonly buried in open mouthed pots (low ratio 3). In contrast, older adults were commonly associated with narrow pots with restricted necks (low Ratio 1 and high Ratio 3 value), and brooches, which are not associated with any particular age or gender, were significantly linked to vessels with high shoulders (irrespective of size). There appears, then, to be a direct relationship between specific skeletal groups, grave goods and the overall vessel form (Richards 1987, 150-4). Unfortunately, Richards did not provide ‘typical’ examples of types, hampering attempts by future analysts to deduce how these ratios combined into actual vessels forms. Nevertheless, such findings provide a window onto Anglo-Saxon perceptions of vessel form and are clearly at odds with Myres’s suggestion that studies of form would prove ‘fatal’ to attempts to understand the pottery of this period.

combination of these ratios which guide a person’s perception of a ‘typical’ vessel type. With insights into the uses of early Anglo-Saxon cremation urns prior to burial (Perry 2011; 2012; 2013), the time is ripe for a new approach to the study of form. Accordingly, this paper develops and interprets a new taxonomy, based on the specialist uses of pots suggested by use-alteration analysis. Central to the new taxonomy, discussed below, is the concept that typology should not be focused on the shoulder. Analogies with ethnographic studies have been drawn upon to support this. Whilst the potential dangers of ethnographic analogy are acknowledged, the same concerns seem to cross cultures in native classifications (see below), and they therefore provide a justifiable theoretical basis on which to being to re-classify AngloSaxon pottery. Through the eyes of the living Let us now consider the ethnographic evidence for ways in which communities use and classify pots. Figure 3 (vessels 1-6) shows water jars produced by six potter-communities from villages in the Luo region of Kenya (Herbich 1987, Figure 2). If a Myers-based system was used to classify these vessels, with the emphasis being placed on whether pots are rounded, biconical or shouldered, this would ignore their common features, divorcing them from one another and from their intended uses. For example, vessels 1-4 would be classified as globular, vessel 5 as rounded, and 6 as shouldered. Thus, despite their similar forms and the fact that they have exactly the same function, a Myresbased system would classify these six vessels as three different types.

Finally, it is worth mentioning that in his study of Cleatham urns, Leahy (2007a, 73-5) considered whether Richards’s Ratios 1-3 and an additional Ratio 4 (2 x shoulder radius ÷ maximum diameter), which he suggested expressed the overall form as a single value (globular vessels have a value of 1.00, baggy vessels greater then 1.00, and biconicals, very low values), could be used as chronological indicators. His study revealed that average ratio values, and thus vessel shapes, varied little over time.

Ingrid Herbich (1987, 195-6) termed these localised variations in Luo pottery ‘micro-styles’. She stressed that micro-styles should not be seen as adaptations of a ‘common ideal form’; rather, they are ‘the result of different local conceptual traditions’ which develop out of the learning patterns, motor habits, social relationships and habitus of the potters who produce them (Dietler and Herbich 1989; 1994, 464; Herbich 1987, 195-6). Micro-style variation is not restricted to the Luo’s water pots (Figure 3 – vessels 7-8), nor the Luo themselves; for example, Figure 4 shows the shape of cooking pots produced by three communities in the Kalinga region of the Philippines and the shapes of water carrying jars produced in Guatemala.

As this review demonstrates, almost all approaches to Anglo-Saxon pottery form begin by classifying the position and shape of the shoulder. This places considerable weight on a single characteristic, detracting from the importance of overall vessel form. Moreover, despite continual criticism of Myres’s methodology, analysts repeatedly base their types on his and, accordingly, their taxonomies often suffer from the same inherent problems. To what extent, then, do these classificatory systems have real significance? This question is especially pertinent when we consider that the systems were constructed entirely by analysts, with little evidence to suggest that they had any relevance to early Anglo-Saxon society. It is notable that the most enlightening analyses of form are those undertaken by Richards (1982; 1987) who did not attempt to recycle Myres’s typology, but started from scratch, allowing measurements and morphological ratios to direct his research. This is a particularly significant observation, as ethnographic studies demonstrate that it is often the

The above examples demonstrate that classifying according to the shape and position of the shoulder divorces vessels with exactly the same function from one another. Conversely, this approach may also draw together vessels with entirely different functions. For example, cooking and water pots produced by potters in the Kalinga village of Dalupa are both globular in profile (Stark 1999). Reconsidering early Anglo-Saxon pottery in light of this ethnographic evidence suggests that, although Myres saw urns 2319, 3402, 632 and 1571 (Figure 2) as three different forms, what we actually have are just two 42

All Form One and One Form All forms, one with a wide mouth and a width considerably greater than height (large Ratio 1 and low Ratio 3) (632 and 1571) and the other with a narrow restricted neck, where the width and height are approximately equal (high Ratio 3 and Ratio 1 ≈1.0) (2319 and 3402). That they are ‘shouldered’, ‘biconical’ or ‘sub-biconical’ is of secondary concern; this is purely micro-style variation. Methods for classifying early Anglo-Saxon pottery form clearly require a critical overhaul. Crucially, if we are to fully understand pottery form, we must begin by producing a typology that is not centred on the contour and position of the shoulder. As Kluchohn (1960, cited in Rice 2005, 276) explains, in order for a typology to be regarded as more than just another set of groupings it must be theoretically based, with all of the analyst’s types reflecting as closely as possible the ideas that potters had in mind when they made their vessels. We cannot, of course, ask AngloSaxon potters how their pots were classified, but we can explore how they might have classified vessels through ethnographic analogy, specifically the naming systems (folk classifications) employed by non-industrial potteryproducing and -using communities (Rice 2005, 277-82). Prudence Rice (2005, Chapter 9) examined in detail how archaeological approaches to typology compare to folk classifications. Analysing an extensive body of ethnographic data, she demonstrated that pottery users and makers assign names to vessels according to intended function; they do not tend to focus on features such as body contour (Rice 2005, 278; see also Longacre 1981, 53-4; Linares de Sapir 1969, 8; Birmingham 1975, 384; DeBoer and Lathrap 1979, 105-10). Yet how are we to uncover the intentions of Anglo-Saxon communities when designing their vessel forms? The prospect may look bleak. However, where detailed studies of ethnotaxonomy

Fig 3. ‘Micro-style’ variation amongst Luo (Kenya) pottery: water jars (1-6) and beer serving vessels (7-8) (redrawn from Herbich 1987, Figure 2 and Dietler and Herbich 1998, Figure 10.2).

Fig 4. ‘Micro-style’ variation. Kalinga (Philippines) cooking pots (left) and Guatemalan water carrying jars (right) (redrawn from Stark 1999, Figure 3.4 and Reina and Hill 1978, Map 10). 43

Gareth Perry have been undertaken, they suggest that classifications are reproducible without prior knowledge of vessels’ functions or producers. Specifically, these groupings may correlate with ‘precise measurements or ratios of vessel sizes and proportions’ (Rice 2005, 279).

We can see, then, that function and shape (in the form of ratios) are significant concerns in the folk classification of pottery. The successful use of ratios in determining functionally significant ethnographic types is particularly encouraging for attempts to re-categorise early AngloSaxon vessel form. A range of ratios that have previously been employed to describe the forms of Anglo-Saxon pottery has already been presented, the following discussion now examines whether these ratios may be used to identify and distinguish between early AngloSaxon vessel types.

In his ethnographic study of the classificatory system employed by pottery-producers and -users in Mexico, Willet Kempton (1981) revealed that whilst informants named types according to function, ratios were an extremely significant concern in their identification of these types. For instance, when asked to provide verbal definitions of specific vessel-types informants were reluctant (or unable) to articulate what made an olla an olla, for example, or a jarro a jarro. Instead, they referred to typical examples from their own homes or sketched one. If verbal definitions were attempted they inevitably focused on function or the presence of added functional elements such as spouts or handles (Kempton 1981, 35-7). For example, according to informants, the only difference between a jarra (pouring vessel) and a jarro (drinking vessel) is that a jarra possesses a spout but a jarro does not. However, when one informant was presented with a range of spouted and un-spouted vessels she categorised one as a ‘jarro with a spout’. When questioned about this inconsistency she revealed that it was too fat to be a jarra (Kempton 1981, 36-9). This ‘fatness’, or ‘width-to-height ratio’, requirement was never mentioned by any of his informants, but it was pivotal in their classification of all vessel forms. William Longacre (1981) and the linguist William Labov (1973) also highlighted the importance of the width-to-height ratio in identifying and naming vessel types in societies as disparate as the tribes of the Kalinga (Philippines) and English-speaking students in American universities. Longacre demonstrated, by calculating and comparing the width-to-height ratios of different types of Kalinga cooking pots, that it was possible to ‘replicate the native system’ revealing ‘functionally significant categories’ (Longacre 1981, 54, emphasis added). He concluded that by using ratios a non-Kalinga might ‘discover’ the native types without prior knowledge of their indigenous systems (Longacre 1981, 54).

Painting by Numbers: Anglo-Saxon Urn Ratios In order to test whether Fennel’s (1964) and Richards’s (1982; 1987) ratios can be used to distinguish between functionally significant types, and thus form the basis of a new classificatory system, their ratios are applied to vessels from known classification systems, namely the ethnographic groups described above. The measurements used in this part of the analysis are taken from published illustrations of ethnographic vessels. Richards’s Ratios 1 and 3 and Fennel’s ‘mouth-to-width’ ratio (hereafter Ratio 5) are calculated and compared. No consideration is given to Ratios 2 and 4 as these are concerned with the shape and position of the shoulder (see above). Table 1 provides the ratios obtained for the Luo water jars and beer drinking pots shown in Figure 3. The mean Ratio 1 values of these two form-types are sufficiently different to allow them to be distinguished numerically; the mean Ratio 3 and 5 values also differentiate these two types. Ratio-based distinctions are again revealed by considering Gamo pottery (Figure 6 and Table 1). Here we see that the narrow-necked beer jar (batsa C) has a Ratio 1 value of 0.81, whilst the communal eating and storing bowl (shele F) has a Ratio 1 value of 1.17; again Ratio 5 helps to delimit these types more precisely. Of concern, however, are the values obtained for Ratio 3. Richards (1982, 44) suggests that this ratio describes how restricted the neck/ mouth is, yet the batsa (C) is clearly restricted, whilst the shele (F) is open-mouthed, despite both having exactly the same Ratio 3 value. Is this simply an anomaly, caused by using non-Anglo-Saxon types, or is it a major flaw in the application of Ratio 3?

We must note, however, a potential pitfall of a ratio-based approach. Although Longacre (1981) demonstrated that it is possible to determine and distinguish between vessel types mathematically, both Kempton (1981) and Labov (1973) highlight the danger of rigidly applying mathematical definitions. For example, Labov (1973, 352-9) asked his informants to name a range of open mouthed, handled vessels, with varying width-to-height ratios. All agreed that the vessels were cups when the ratio was 1 or 1.2, but as this ratio increased the numbers identifying the vessels as bowls also increased (Figure 5). Significantly, when the ratio was c.2, informants were equally likely to name the vessel as either a cup or a bowl, clearly demonstrating that rigid numerical boundaries cannot be applied to classify vessel types.

In order to investigate this problem further, let us consider the ratio values of Anglo-Saxon vessels (Table 2 and Figure 2). Urns 2319, 3402, 1885 and 1636 all have restricted necks and are of similar height, width and rim diameter. The only feature dividing them is the position and contour of their shoulders. Importantly, these four vessels all have very similar Ratio 1 and Ratio 5 values and these are clearly distinguishable from the wide, squat, open-mouthed urns – numbers 632, 1609, and 1571. Consideration of Ratio 3 reiterates the problem highlighted. Vessels 3402 and 1609 have almost identical Ratio 3 values, yet one is open-mouthed and the other restricted. This problem with Ratio 3 is very concerning, as Richards stated that it was the most significant ratio in 44

All Form One and One Form All

Fig 5. The significance of the width-to-height ratio in the classification of ceramics and the dangers of rigid numerical classification. As the ratio increases (i.e. the vessels get wider) the number of informants identifying vessels as bowls also increases. Note that when the ratio is c.2, informants are equally as likely to identify a vessel as either a cup or bowl (redrawn from Labov 1973, Figures 5 and 6). terms of identifying correlations with other aspects of the Anglo-Saxon cremation rite (Richards 1982; 1987, 33-44). Clearly, an apparent relationship between, for example, the gender of an individual and the Ratio 3 value of the urn in which they were buried cannot be considered real if both restricted and unrestricted vessels share the same value. The most effective indicators of Anglo-Saxon vessel form are therefore Ratios 1 and 5. It is encouraging that both Longacre and Kempton found that the ratio of width-toheight (Ratio 1) was the most significant in distinguishing between functional classes in their ethnographic studies of pottery classification (see above); it is also this ratio which, according to Richards (1987, 33-44), accounts for the greatest variation in Anglo-Saxon vessel form. Furthermore, as Rice (2005, 212, 241) notes, the characteristic most often modified or adapted to meet distinct functional requirements is the orifice (also see Henrickson and McDonald (1983) for a detailed discussion on how orifice morphology affects functionality); our Ratio 5 (ratio of mouth-to-maximum diameter) describes this characteristic. Whilst ratios are useful in describing vessel form, we cannot rely solely on them to distinguish between AngloSaxon types. Indeed, numerous ethnographic studies (e.g. DeBeor and Lathrap 1979, 105-10; Longacre 1981; Linares de Sapir 1969, 8) reveal that size may distinguish different functional classes which share the same form. It is worth considering how size might affect functionality and thus the classification of Anglo-Saxon types. For

Fig 6. Gamo pottery. The tsua (A), is a drinking vessel, the otto (B) is a multipurpose vessel used in cooking, storage and as a beer storage/fermenting jar. The batsa (C) is a dedicated beer storage/fermenting jar, the tsaro (D), is a multifunctional cooking and storage jar, whilst the shele (F) is a bowl (redrawn from Arthur 2006, 36-7, Figure 2.7). 45

Gareth Perry example, the Gamo’s beer mug (tsua), beer fermenting jar (batsa) and multifunctional cooking, storing, transporting and beer fermenting vessel, the large jar (otto) (Figure 6 and Table 3), are all of the same form, consequently their ratio values are almost identical; it is their size that separates them functionally (Arthur 2006, Table 2.2; 2002). As the otto is of a similar size to the batsa it is often used to ferment beer. In contrast it would be impossible to ferment large quantities of beer in a single tsua, and it would be extremely difficult to lift a full batsa or otto to the lips in order to take a drink.

taller than the 1A urns. Each of these sub-groups is divided according to size, with larger urns being denoted by i and smaller vessels by ii and iii (Figure 7). To distinguish between size groups, histograms of maximum height have been produced for each of the main types (Figure 8). In order to prevent the use of unsubstantiated cut-off values to define types, no numerical limits have been identified; instead, an overall impression of each group is provided by calculation of the mean and range of each measurement and ratio (Table 4). To help determine whether vessels of different sizes might belong to the same functional class, estimated volumes (Table 4) have been calculated for each urn, following the method described by Blinkhorn (1999).

A New Typology Having determined that specific ratios and vessel size can reveal native classifications of pottery, these characteristics are now applied to determining Anglo-Saxon types. It should be noted that this part of the analysis makes a distinction between decorated and undecorated pottery. Criticism has previously been levelled at studies that have made arbitrary distinctions between plain and decorated pottery (e.g. Blinkhorn (1997) criticised Myres’s prejudice towards the study of plain undecorated pottery). Nevertheless, findings from use-alteration analysis, published elsewhere, reveal that plain and decorated vessels fulfilled different functions (Perry 2011). As function is at the core of this discussion, such a distinction is necessary.

The New Typology Six main form groups are identified. These groups are divided into size and morphological sub-groups, resulting in a total of 30 types (Figure 7 and Table 4). The number of urns attributed to each group is given in Table 5. Group 1 (Figure 7) Group 1 urns are characterised by restricted mouths, and widths that are approximately equal to their heights. Although there is considerable variation in the position and contour of the shoulder, the same basic form runs throughout each of these vessels – a voluminous body with a narrowing neck, leading to a narrow mouth. On account of minor ratio-based variation Group 1 is separable into three sub-groups, each sub-group being divided into three further groups based on size.

Methodology Measurements of height, rim and maximum diameter of the decorated Cleatham urns have been obtained from the online Cleatham archive (Leahy 2007c), whilst measurements of the Elsham urns have been determined by measuring scale drawings and the urns themselves. These measurements are used to calculate the various ratio values. As ratio values cannot be determined from incomplete pots only vessels with complete profiles are considered – a total of 317 from Cleatham and 173 from Elsham. At no point has consideration been given to whether urns are shouldered, biconical, or globular (this is micro-style variation); they are purely sorted and classified according to ratio-based characteristics. For example, vessels with tall bodies (low Ratio 1) and restricted mouths (low Ratio 5) are separated from those with tall bodies and wide mouths, whilst tall wide-mouthed vessels are separated from those with wide mouths and squat, wide bodies (large Ratio 1 and 5 values).

1A A tri-modal distribution of heights is observable in this group, equating to three broad sizes – small, medium and large – with average volumetric capacities of c.7 litres, c.4.5 litres and c.1 litre respectively (Table 4). The average ratio values of these three sizes all fall within a very narrow range – Ratio 1: 1.06-1.11 and Ratio 5: 0.43-0.53 – but these should not be taken as boundaries, as they are merely the range of average ratio values of the three sizes. 1B Group 1B urns have a slightly squatter appearance than the other vessels in Group 1, a point confirmed by their slightly larger Ratio 1 values; the range of average Ratio 5 values, however, are almost identical to those of the Group 1A (Table 4). At both cemeteries three sizes are identifiable, on account of a tri-modal distribution of heights, translating to average capacities of c.5 litres, 3.5 litres and 1.5 litres, with a consistent c.3cm difference between the mean heights of the 1Bi and 1Bii (Table 4 and Figure 8). This neatly demonstrates that the potters were working towards a mental template of acceptable size and form.

Each identified form-group has been assigned a numerical identifier. For example, Group 1urns have restricted necks and a height approximately equal to width (Ratio 1 ≈ 1.0 and low Ratio 5) and Group 2 urns are squat, wide vessels with wide mouths (large Ratio 1 and large Ratio 5). These groups have been further divided according to minor ratiobased variation. For instance, urns belonging to Group 1 have restricted necks, and heights approximately equal to width (Ratio 1 ≈ 1.0). Those attributed to 1B appear ‘squatter’ than the 1A urns, whilst 1D urns appear slightly 46

All Form One and One Form All

Fig 7. A new typology of Anglo-Saxon vessel form (urn images redrawn from the Elsham archive, held in North Lincolnshire Museum, and the online Cleatham archive – Leahy 2007c). See Appendix A for a range of urns attributed to these groups. 47

Gareth Perry 1D

Group 4 (Figure 7)

Group 1D urns have approximately the same Ratio 1 values as those in 1A; they are separable, however, on account of their slightly less restricted mouths (a marginally greater Ratio 5 value, although the mean rim diameters are the same as 1B and 1A urns). Group 1D urns do not follow a bi- or even tri-modal distribution of height; instead, at both cemeteries, urns attributed to 1D are normally distributed with modal heights ranging from 200-220mm (Figure 8). As this pattern is repeated at both cemeteries we are once more given the impression that the potters were working within a framework of mental templates.

Like Group 1A, urns in Group 4 have heights approximately equal to their widths. What separates them from Group 1A, is their slightly wider mouths. Group 4 is divided into two types, 4A and 4B, and again these appear in smaller and larger sizes (Figure 7). 4A As one would expect, the four sizes of urns in this group have very similar Ratio 1 and 5 values. All four sizes are found at Elsham and Cleatham and the ratio-based, size and volumetric characteristics of each type are very similar at both cemeteries (Figure 8 and Tables 4 and 5).

Group 2 (Figure 7)

4B

Urns in Group 2 are characterised by their squat, wide bodies (large Ratio 1), and unrestricted necks (large Ratio 5). Based on size and ratio values this group separates into two subgroups. Urns belonging to 2A appear squatter than those of 2B. Like the Group 1 urns, these vessels divide into small, medium and large. With a c. 3cm difference between Elsham’s 2Ai and 2Aii (no 2Aii urns were identified at Cleatham), and as the mean volumes of 2Ai urns are almost identical at both cemeteries (Table 4) we are once more given the impression that potters were working towards acceptable norms. There are very strong links between Group 2 and Group 3 urns, and it is quite possible that the slightly squatter character of Group 2 is just micro-style variation, and that Groups 2 and 3 are essentially the same types of pots.

Only eight urns, all from Cleatham, are attributable to Group 4B. The mean Ratio 1 value is almost identical to that of 4Ai and they are, on average, the same size as the 4Aiii vessels. They are separable, however, on account of their wider mouths; indeed, their Ratio 5 values are more akin to those of 3Bii and 3Biii, whilst the average rim diameter is equal to that of the wide mouthed 3Bii (Table 4). Group 5 (Figure 7) Urns belonging to Group 5 are distinguishable by their wide, tall bodies and relatively unrestricted necks. The major feature of these vessels is their height to width ratio; indeed they have the appearance of being the tallest form in the typology. Again we see two sub-groups, 5A and 5B; these are separable on account of their apparent neck restrictions (Ratio 5 values).

Group 3 (Figure 7) Like urns in Group 2, urns belonging to Group 3 are characterised by open mouths and squat, wide bodies, but they are distinguished from Group 2 by slightly smaller Ratio 1 values, being less squat. Urns in Group 3 can be sub-divided into small, medium, large, and perhaps even extra-large.

5A Group 5A urns have what appear to be the most restricted necks within this group. However, when their rim diameters are compared with those of those in 5B we see that there is very little difference between the two (Table 4).

3A Group 3Ai urns are the largest of the Group 3 urns. They have slightly lower Ratio 1 and 5 values than those of 3Bi, appearing slightly more restricted in the neck and marginally narrower in width. Despite this, 3Ai and 3Bi urns have almost identical average rim diameters (Table 4). Again these vessels appear to have been produced to relatively standard sizes as the average heights, widths, volumes and Ratio 1 and 5 values of the 3Ai urns are virtually identical at both cemeteries (Table 4).

5B Group 5B urns are separable from those in 5A on account of their larger Ratio 5 values (i.e. they have less restricted to necks). Like the other form groups, we see small, medium and large sizes and once more there are similar volumes for these types at both cemeteries and c.3cm difference between the mean heights of the large (5Bi) and medium (5Bii) sizes (Table 4).

3B

Group 6A (Figure 7)

Group 3B is divided into three sized-based groups, separated on account of a tri-modal distribution of heights. At both Cleatham and Elsham a c.3cm separation exists between the mean heights of 3Bi, 3Bii and 3Biii, equating to three average capacities of c.5 litres, c.3 litres and c.1.5 litres (Table 4 and Figure 8).

Very few urns belong to Group 6 and all derive from Cleatham. These vessels are characterised by large, squat bodies with slightly restricted necks. Indeed, in terms of their Ratio 1 and 5 values, they are somewhere between Groups 1 and 2 (Figure 8). No size-based divisions are

48

All Form One and One Form All

Fig 8. Histograms of vessel height in Group 1 and 4 and the average calculated volumes for form types. Note that the frequency peaks occur at approximately points for urns found at both Elsham and Cleatham and that the average volumes of each type are almost identical at both cemeteries. observable; indeed the range of heights is rather narrow, with just 6cm between the maximum and minimum vessel heights. However, the capacity ranges from 6.1-11.0 litres, with a mean of 8.2 litres (Table 4).

diameters, heights (c.140mm), widths and volumes (c.1.5 litres) (Figure 7 and Table 4). Whilst ‘miniatures’ are not a separate group in their own right, it is worth highlighting the similarities between these small pots at this point in the discussion, before considering how various vessels in the typology might have been used.

‘Miniatures’ Group Approaching the form of early Anglo-Saxon pottery in the way described above reveals that potters seem to have produced miniature versions of all the major types – 1Aiii, 1Biii, 1Diii, 2Aiii, 3Biii, 4Aiv, 5Biii, for example – and it is notable that all these ‘miniatures’ have very similar rim

Ungrouped Vessels In total there are 32 Cleatham and nineteen Elsham urns that do not fall into the groups defined above. This does not mean that they are radically different from other 49

Gareth Perry urns, rather they are ‘on the edge’ of the main groups and they cannot be confidently assigned to a particular formtype. For example, EL76NW (Figure 9) has ratio values consistent with Group 4 (more specifically, 4Aiv), yet its rim diameter makes it appear slightly too narrow to be placed in this group. Despite this, it is clear that this vessel is part of the ‘miniatures group’. Larger urns also fall into the ungrouped category. For example, Cleatham urn 697 is almost bottle-like in form, with a similar rim diameter to vessels in Group 1 (100mm), similar volume to urns in Groups 1Aii and 1Di (4.3 litres), and a height of 270mm, which places it alongside vessels in Group 1Ai. Clearly it has great affinity with Group 1 urns, yet it is slightly too narrow to be placed in any of the Group 1 sub-groups. The overriding characteristic of ungrouped urns, then, is that they almost conform to one of the defined groups, but they do not quite possess all of the required attributes.

appears, then, that potters had clear understandings of the range of acceptable vessel types and sizes, and that they produced vessels according to mental templates. Although this typology is based upon material from Cleatham and Elsham, it is applicable beyond the limits of North Lincolnshire. The urns shown in Figure 2 are drawn from a wide geographical area, yet each fits into the typology (Table 6). This supports Richards’s (1987, 93-9) observation that there is a remarkable level of consistency in the dimensions and ratio values of urns recovered from cemeteries across the country, further promoting the idea that potters were producing vessels according to culturally acceptable standards. It is clear that most vessel types are present at both Cleatham and Elsham, and that vessels from other cemeteries fit into the typology, but there are some disparities which merit discussion. For example, Group 6A and 2Ai urns are present at Cleatham but not at Elsham. Group 1D urns are common at Elsham but not at Cleatham, whilst Group 1B urns are common at Cleatham but not at Elsham (Table 5). This might simply result from levels of preservation or sample size, but it does raise several questions. Were different communities of potters producing different forms, and are some of these forms just micro-style variants as opposed to being different functional classes? We can begin to answer these questions by considering how each form might have functioned in the domestic sphere.

Summary Contrary to Myres’s (1969, 22-5) assertion that it would be foolhardy to expect early Anglo-Saxon potters to have shaped pots based on pre-defined forms, the evidence discussed above demonstrates that there is, in fact, is a remarkable level of consistency in Anglo-Saxon vessel forms. This exists within groups, between groups, and, in particular, in the range of types present in each cemetery assemblage. The Ratio 1 and 5 values of the largest and smallest versions of individual types are virtually indistinguishable, clearly demonstrating that potters had a definite perception of the relationship between height, width, rim diameter and maximum diameter, and understood how this translated into vessels of varying size. This is further supported by other characteristics: the presence of numerous bi/tri-modal distributions; consistent c.3cm differences between the mean heights of larger and smaller versions of the same types; and similarities in mean heights, widths, rim diameters and approximate volumes of vessel types from both cemeteries (Table 4). It

The Form and the Function One of the most basic relationships elucidated in ceramic studies is the link between form and function (Rice 2005, 207-11, 224-5). Each task in which a pot is employed places specific demands on it and each category of use requires a different combination of attributes and properties of form. This includes: the amount of produce to be held; whether the contents will be solid, liquid, hot or cold; the stability of the vessel; how often the contents will be accessed and moved in and out; the duration of each use episode; whether the pot will be tended whilst in use; and whether utensils will enter it. Although use-alteration evidence demonstrates that cremation urns were employed in fermentation processes before burial (Figure 1 and Perry 2011), given the range of forms it seems safe to assume that not all vessels functioned in the same manner. We must therefore attempt to discover the roles that they may have played in the production and consumption of this produce. To do so, we may draw ethnographic analogies with specific morphological characteristics and combine this with use-alteration data. Plotting the frequency of internal pitting (Figure 1) on each of the form-types identified in this study reveals that each form was involved in the production and consumption of fermented produce (Figure 10). Might it be possible, then, to identify vessels with different functions, such as short-term drinking containers or longer-term fermenting vessels, based on the severity of attrition? Unfortunately,

Fig 9. Ungrouped urns (images redrawn from the Elsham archive, held in North Lincolnshire Museum, and the online Cleatham archive – Leahy 2007c) 50

All Form One and One Form All

Fig 10. The frequency of occurrence of internal pitting (see Figure 1) on each of the form types identified in the new typology (see Figure 7). Note that pitting was absent from some types at one cemetery, yet present in that type at the other (e.g. 1Bi and 3Bi). Functional Properties of Anglo-Saxon Types

this seems unlikely. In his study of use-alteration on Gamo pottery, Arthur (2002, 348; Table. 1; 2003) demonstrated that all vessels involved in the production and consumption of lactic acid fermented produce suffered from internal pitting, even drinking vessels, and that the levels of attrition could not be considered as markers of the frequency or duration of use. Our best chance of understanding how these vessels might have been used, then, is by considering the properties of individual forms.

Considering the groups in the typology alongside functional and use-alteration data reveals a relatively narrow range of functions, and suggests that some types may have acted as functional equivalents. For example, groups 1Ai, 1Aii, 1Bi, 1Bii, 1Di and 1Dii (Figure 7) have very similar Ratio 1 and 5 values and mean rim diameters (Table 4). The mean volumes of 1Aii, 1Bi, and 1Di are all within a litre of one-another, as are the volumes of 1Dii and 1Bii (Table 4). A consideration of the functional characteristics of these six types adds weight to the suggestion that they were functional equivalents. Vessels with restricted necks are typically used to hold liquids (Table 7). These narrowmouthed vessels inhibit access to the contents but prevent spillage and evaporation, and also facilitate pouring. In this typology, the largest Group 1 vessels (1Ai) may represent longer-term storage, but their volumes prevent them from being so bulky that they could not be moved when full; perhaps they were used when larger amounts of produce were being prepared/served/stored. The smaller

Henrickson and McDonald (1983) undertook an extensive survey of the relationship between vessel form and function. They considered pottery produced and used by 24 societies ‘with widely diverse economic and sociopolitical systems, ranging from isolated hunter/gatherer/horticulturists to peasant towns’, recording properties of vessels and their primary functions (Henrickson and McDonald 1983, 631). Table 7 provides an overview of their findings and it forms a basis for interpretation of the types identified in the new typology.

51

Gareth Perry Group 1 vessels (1Aii, 1Bi and 1Di and the smaller 1Bii, and 1Dii) would be even easier to move. It is notable that the small diameter mouths of all urns in Group 1 (c.12cm) would also facilitate closure by a skin or textile cover. Interestingly, McKinley (1994, 103) observed that whilst some urns from Spong Hill had ceramic lids, others appear to have been sealed with perishable materials such as leather or cloth. Although there are no vessels with lids from either Cleatham or Elsham, a survey of published cemetery reports demonstrates that ceramic lids are more commonly found with vessels in Group 1 (Table 8). These observations correspond with Henrickson and McDonald’s (1983, 632-3) findings that narrow-mouthed vessels are more frequently associated with lids, and that lidded vessels are commonly associated with the storage of liquid.

Finally, the volumes and dimensions of urns belonging to the ‘miniatures group’ reveal that they held small amounts of produce, implying that they were portable, easy to fill, and could be comfortably held in the hand, making them suitable for personal consumption. As discussed, use-alteration analysis suggests that these cremation urns were involved in lactic acid fermentations such as dairying and brewing before their burial. Having explored the functional properties of their forms we can now consider them in the context of their pre-burial uses. Whilst it is acknowledged that no residue analysis has been undertaken on these urns, and we cannot therefore categorically identify their pre-burial uses, given the range of drink-related paraphernalia found in early Anglo-Saxon inhumation and cremation graves (e.g. glass beakers, bowls and cups, drinking horns, bronze-bound wooden buckets and stoups, wheel-thrown vessels thought to be associated with the consumption of wine, and spoon-shaped metal strainers thought to have been used to separate solids from liquids in the consumption of fermented drinks (Arnold 1988; Cook 2009; Evison 1979; Harden 1978; Hills et al.1984; Hirst and Clarke 2009; McKinley 1994; Ravn 2003; Stoodly 1999)), it seems that the production and consumption of fermented beverages is an extremely likely candidate for causing this attrition. As little is known about early medieval brewing techniques it is difficult to identify the functions that particular vessels might have served. However, as the process appears to have been very similar across time and cultures, we can use ethnographic and historical sources to identify the basic procedures and range of vessels typically involved. In doing so, we can explore the roles that each Anglo-Saxon type may have fulfilled in the production and consumption of fermented beverages.

Given the possibility that some forms operated as functional equivalents, it is interesting to note that the Elsham-using communities produced few 1B vessels, yet they used 1D urns extensively (Table 5). Cleatham shows the opposite; 1B was common, but 1D was relatively rare. It is possible that Cleatham’s and Elsham’s burial communities were producing what were essentially micro-style variants of functionally equivalent pots. Similarly, urns in Groups 2 and 3 may have been functional equivalents; both types are ‘squat’, with unrestricted mouths (Figure 7), meaning that they offer stability and it would be easy to stir their contents and scoop or ladle them out. Although stirring, dipping, or ladling may have been viable means of manipulating and removing produce from 4Ai pots, their slightly larger heights (compared to 3Ai and 3Bi), along with the reduction in mean rim diameter (Figure 7 and Table 4), result in a slight decrease in accessibility. On the other hand, these characteristics mean that they would be easier to pour from, allow the affixation of a lid, and that there would be less chance of spillage when moving the vessel. The dimensions of smaller Group 4 urns, particularly 4Aii and 4Aiii might allow them to be used in a similar manner to Group 1 urns, a point supported by the fact that the mean heights, maximum diameters and volumes of the 4Aii urns are almost identical to those of 1Di and 1Aii (Table 4). Moving onto Group 5, their functional properties are very similar to those of Group 4, with their slightly taller bodies making their contents even less accessible. This increase in height also makes them, according to Henrickson and McDonald’s (1983, 632-3) functional properties of pottery, appropriate for longerterm storage of produce.

‘Make us a Brew’ First a mash is produced by mixing boiled water with malted or un-malted grains. Mashing produces a saccharified liquid called wort; this sugar is used later in the brewing process by the microbes that cause fermentation. Once the wort has cooled the spent grain may or may not be removed from it. The bacteria and fungi that cause fermentation may be added to the wort. This is not a necessary ingredient, however, as levels of airborne microbes and fungi are often sufficient to begin spontaneous fermentation. Similarly, if fermentation takes place in vessels made of porous materials (e.g. wood or low-fired ceramic), which have previously been used to make fermented produce then the process will be initiated by starter cultures present in the vessel wall. At this stage the beverage is left to ferment (Arthur 2003, 519; Clark 1983, 100; Corran 1975, 12-19; Dietler and Herbich 2006, 400-2; Garine 2001, 194-5; Stone 2006, 15-16).

Vessels belonging to Group 6 have similar functional properties to those in Group 1. Indeed, the Ratio 5 values of Group 6 urns resemble those in Group 1A and 1B, whilst their average volume is identical to that of Cleatham’s 1Ai urns (Table 4). It is possible that Group 6 urns might actually be a very large version of Group 1B. The slightly wider appearance of urns in this group might simply result from these large vessels requiring extra stability; their wider bodies would certainly provide a lower centre of gravity.

Fermentation is a relatively short process. The whole procedure, from cooling the wort to the final fermented product being drawn off, is normally completed within two to eight days. If hops are not added to the wort the 52

All Form One and One Form All resultant beverage (ale) has poor keeping qualities and has to be consumed within a few days (Clark 1983, 24; Dietler and Herbich 2006, 40; Stone 2006, 16). In contrast, hopped beer has greater keeping qualities, but as hops were not introduced to England until c.AD 1400 (Corran 1975, 42-4), it is highly likely that early Anglo-Saxon fermented drinks were un-hopped and therefore had to be hastily consumed.

(1983, 632-3) observations that vessels used to store liquid often have narrow-mouths and lids. Significantly, lids do not merely keep the liquid inside, they also assist fermentation. Yeasts only produce alcohol if forced to respire in anaerobic, acidic conditions, and lids restrict the flow of oxygen into the vessel, helping to maintain a ‘carbon dioxide barrier on the surface of the liquid’ (Dietler and Herbich 2006, 402). Anglo-Saxon vessels belonging to Groups 1Ai, 1Aii, 1Bi, 1Bii, 1Di, 1Dii, 4Aii and 6A are particularly suited for such tasks. All have narrow mouths and necks that would allow closure by a skin, bowl or pot sherd; significantly, these types are most commonly associated with ceramic lids (Table 8). Also notable are the average volumes of these vessels, ranging from c. 4.5 to 8.0 litres. Un-hopped beers have poor keeping qualities (see above), and if beer was made by individual AngloSaxon households for their own consumption, to prevent wastage by spoiling, relatively small qualities would presumably be the norm.

With an understanding of the brewing process, we can now consider the properties of vessels typically used in each step, and how these may relate to the Anglo-Saxon vessel types identified here. For analogous material we must once again turn to ethnographic studies, since written records that might definitively reveal specific uses for vesseltypes during the early medieval period (and, indeed, later) are few. Studies of the Muzey and Duupa (Cameroon) (Garine 2001; de Garine 2011), the Luo (Kenya) (Dietler and Herbich 2006), and the Gamo (Ethiopia) (Arnold 2002; 2003) are considered in the following discussion, and are supplemented by the scanty written evidence from the Roman, Saxon and medieval England.

In a typical brewing process, the final steps are distribution and consumption. The properties of vessels used in these stages reflect the processes themselves. For example, the Luo drink unfiltered beer through straws, while portions of filtered beer are drawn by dipping small containers into the liquid (Figure 12). Dipping requires two vessel types; the first must have a wide unrestricted mouth, whilst the second must be small enough to pass through the mouth of the former. If dipping was a mode of Anglo-Saxon serving the best suited vessels would be a combination of Groups 2 and 3 – particularly 2Ai, 3Ai and 3Bi – and the ‘miniatures group’. Certainly, the mean heights and diameters of urns in the ‘miniatures group’ suggest that many could pass through the mouths of 2Ai, 2Bi, 3Ai and 3Bi urns. Alternatively, distribution might be achieved by pouring quantities of beer from small narrow-mouthed vessels into individual bowls or cups; this is the mode of distribution practiced by the Muzey of Cameroon (de Garine 2011). In the Anglo-Saxon period this might have been achieved by pouring liquid from small narrow-mouthed vessels like 1Aii, 1Aiii, 1Bii, 1Bii, 1Di, 1Dii into vessels belonging to the ‘miniatures group’.

Figures 11 and 12 depict the vessels and procedures used by the Gamo (Ethiopia) and Luo (Kenya) in production and consumption of beer. In the first stages, when the grain is soaked to release sugars and begin germination, the vessels used are generally broad and wide-mouthed. The accessibility of these vessels allows for easy mixing, stirring and removal of the grain/flour. Vessels belonging to Groups 2 and 3 in the new Anglo-Saxon typology would be particularly suitable for such procedures. Once steeped, the grains/flour may be treated in one of the following ways: dried, allowed to germinate, malted, used to produce wort, or remain in the liquid as it advances to the fermentation stage. For instance, after soaking, the Muzey filter the grain from the liquid and the resultant wort advances to the final ferment (de Garine 2011, 136). The Gamo and Luo, on the other hand, allow the flour to remain in the liquid until the fermentation process is complete. Irrespective of when the grain/flour is separated from the liquid, a method of filtration is required; these filtration devices are many and varied (e.g. sieves, woven filters and plant material (de Garine 2011; Deitler and Herbich 2006; Nordland 1969). It is not immediately obvious how filtration might have been accomplished in the Anglo-Saxon period, but when used in conjunction with a fibrous medium such as hay or straw, those vessels with post-firing-perforations in their lower walls and bases could have served as adequate filters (Perry 2012).

The information discussed so far, much of it from ethnographic sources, can be supplemented by Roman, later Anglo-Saxon and medieval documentary evidence. Whilst these sources mention the names of vessels that were used in brewing, there are, unfortunately, no records that outline the actual process. Furthermore, the accounts that name specific vessels largely derive from high-status and ecclesiastical sources (Corran 1975, 25-40), and do not necessarily represent the smallscale household-production that we might expect in the early Anglo-Saxon period. Nonetheless it is useful to consider some of these sources.

Whether or not the grain/flour remains in the liquid throughout fermentation, ethnographic studies reveal that there are many cross-cultural similarities in the forms of fermentation vessels. For example, the vessels in which the Gamo and Luo ferment are of large volume, with narrow mouths and necks, and are sealable by either a skin lid or by placing a bowl or base of another vessel over the mouth (Arthur 2003; Deitler and Herbich 2006). Such characteristics accord with Henrickson and McDonald’s

In the first century AD Tacitus observed that the northern Europeans produced a ‘liquor for drinking ... made out of barley or other grain, and fermented into a certain resemblance to wine’ (Germania XXIII, xxxv; Church and Brodribb 1942). Similarly, Pliny the Elder recorded that the 53

Gareth Perry

 

Water  boiled in  otto 

Flour placed in a  shelle and the hot  water is poured  over and mixed in;  the mixture is left to  cool 

Flour paste and water  placed in to a beer jar  (batsa), or a large storage  jar (otto) and fermented  for five days before  consumption.  Jars are  sealed with pot sherds, or  by placing bowls over the  mouths

Drinks  consumed  from tsua or  gourd cups 

Fig 11. Flowchart of Gamo (Ethiopia) beer production (see Figure 6 for the vessels involved). Process derived from Arthur (2003).

Fig 12. Flowchart of Luo (Kenya) beer production. Processes derived from Dietler and Herbich (2006) (vessels redrawn from Herbich and Dietler (1991, Abb. 3)). 54

All Form One and One Form All ‘people of the Western world have ... intoxicating drinks, made from corn steeped in water’ (Historia XIV, xxix; Bostock and Riley 1855). Both accounts suggest that on the Continent the grain was whole when soaked. In a late ninth-/ early tenth-century remedy a reference was made to ‘new ale before it be strained’ (Leechbook I, li; Cockayne 1865, 125); this implies that at some stage in the brewing process the solid matter was separated from the liquid. As discussed by Perry (2012), cremation urns with post-firing perforations in their lower walls were probably the means of separation.

Whilst function lies at heart of indigenous taxonomies it has been shown that distinctions between functional classes are made according to ratio and size-based characteristics. By ignoring whether vessels are biconical or globular and arranging them according to their size and morphological ratios, a new taxonomy of Anglo-Saxon vessel-form has been developed. The taxonomy comprises six broad groups, each of which is sub-divided according to minor differences in ratio values and size. Classifying the pottery in this way reveals that potters had clear concepts of acceptable forms and that they attempted to manufacture them according to relatively standardised increments of size, shape and volume.

In a thirteenth-century treatise from Hertfordshire, Walter de Biblesworth wrote that the ‘vat’ in which the barley was steeped should be ‘large and broad’ (Corran 1975, 25). The requirement of a ‘broad’ soaking vessel correlates with the findings of the ethnographic studies discussed and supports the suggestion that vessels belonging to Groups 2 and 3 of our typology might have functioned as such. Medieval household accounts provide insights into the frequency of production. At Bolton Priory (Yorkshire) ale was brewed, on average, about every five days in 1307-8, whilst in the brewery of Katherine de Norwich in the years 1336-7, manufacture was undertaken about every ten days (Stone 2006, 16). The frequency of brewing correlates with what we know of unhopped beers; that they have poor keeping qualities (see above). When brewing was undertaken in the medieval home we would expect, then, that the quantities were small. The records of an inquest, held in Basingstoke in 1381, into the death of a six year-old girl, show this to be the case. The child died after falling into a 2 ½ gallon (c. 11.4 litres) pan of hot wort (TNA JUST 2/155, membrane 7 dorse – pres. com. Chris Woolgar). It is significant that the mean volumes of types identified, in our typology of vessel-form, as being most likely to have produced and subsequently fermented the wort are similarly small, for example 2Ai and 3Bi = c. 6 litres (c. 1 1/3 gallons),3Ai and 1Ai = 8 litres, 9(c.1. 3/4 gallons), 1Aii, 1Bi, 1Di = 4-5 litres (c. 1 gallon) (Table 4). Taking the ethnographic, archaeological and historical evidence of the vessels involved in brewing, their functional properties, and the brewing process, it is possible to suggest a hypothetical early Anglo-Saxon procedure, with the vessels used at each stage placed in context (Figure 13).

In considering the functional properties of vessel-form alongside ethnographic accounts of brewing and written sources from the Roman, late Saxon and medieval periods, it has been possible to place the early Anglo-Saxon forms in the context of the processes suggested by use-alteration analysis. Characteristics shared by a number of form-types suggest that many represent functional equivalents. As some types are more common at Cleatham than Elsham, and vice versa, we can suggest that different communities were producing functional equivalents according to localised traditions; this accords with Herbich’s (1987) idea of ‘microstyle’ variation. Although the typology is based on vessels from just two cemeteries in North Lincolnshire it is clearly applicable to pottery excavated from cemeteries throughout early Anglo-Saxon England. Urns from large cemeteries such as Spong Hill (Norfolk), Mucking (Essex) and Sancton (Yorkshire) should now be fully classified according to this typology so as to refine the typology and identify any further regional differences (e.g. is Group 6A just a larger version of 1B?). Classification of pottery excavated from fifth- and sixth-century cemeteries on the Continent is also advocated. Indeed, as Myres’s study of decorative styles employed by potters on the Continent and those in sixth-century England identified ‘an almost purely English development’, in panelstyled stamp decoration (Myres 1969, 58), there is every possibility that ‘purely English’ forms also existed. Finally, it is worth highlighting that later Anglo-Saxon law codes and heroic poems such as Beowulf reveal the drinkserving role that women fulfilled in the household (e.g. Bradley 1982, 427-8, 462, 464, lines 612-624, 1925-30, 2018-21; Laws 14 and 16 Law of Æthelbert (AD 560-616), Attenborough 1922, 7). As Richards (1987, 139) identified that women buried more frequently in urns with above average rim diameters – Groups 2 and 3 in the present typology, which are particularly suited to the initial soaking of grains and/or the final distribution of the beverage – and older adults were more commonly found in tall, restricted necked vessels (Group 1 in this typology) it is possible that age and gendered vessel-choices reflected particular roles that the deceased played in life or might be expected to fulfil in an afterlife. To investigate this possibility, the ages and genders of the individuals who were found in the urns from Elsham and Cleatham now need to be considered alongside the new taxonomy of form.

Discussion and Conclusions This paper has reviewed the previous approaches to AngloSaxon vessel form, demonstrating that previous typologies are based on arbitrarily chosen characteristics and that there is no evidence that they had any relevance to the people who created and used the pottery. Despite regularly criticising Myres’s classificatory system, analysts have repeatedly attempted to recycle, justify and redefine his types. As a consequence our understanding of Anglo-Saxon vessel form has advanced very little in the last half century. By considering the forms of pottery created by modern pre-industrial societies it has been demonstrated that the contour of the vessel shoulder (e.g. biconical, shouldered and globular) – the core of Myresian taxonomic systems – should be considered as micro-style variation of wider functional classes.

55

Gareth Perry Grain soaked in a wide mouthed vessel (for  example 2Bi, 2Bii, 3Aii, 3Bii) for 2 to 3 days  to allow germination and the release of  sugars. 

Water boiled in a plain, undecorated vessel.

Soaked grain added to the boiled water.  The resulting mixture, the mash, is  left to cool.  

Method 1

Method 2 

The spent grain is filtered from the liquid  by using a filtering medium, such as straw,  and a perforated vessel.

The mash is poured into  fermenting jars (e.g. 1Ai,  1Aii, 1Bi, 1Bii, 1Di, 1Dii,  4Ai, 4Aii, 5Ai, 6A) and  extra water added.  The  jars are sealed with  either a ceramic, skin or  textile lid.  The mixture is  left to ferment for c. 2 to  5 days. 

The liquid, the wort, is poured into  fermenting jars (e.g. 1Ai, 1Aii, 1Bi, 1Bii,  1Di, 1Dii, 4Ai, 4Aii, 5Ai, 6A).  The jars are  sealed with wither a ceramic, skin or  textile lid.  The wort is left to ferment for  c. 2 to 5 days. 

Fermented drink  poured into  smaller narrow‐ mouthed vessels  (e.g. 1Aii, 1Bii  1Dii, 1Dii, 4Aii,  4Aiii, 5Aii). 

Fermented drink  poured into  wide‐mouthed  serving vessel. 

Consumption by  pouring liquid  into small cup‐ type miniatures  (e.g. 2Biii, 3Biii,  4Aiv).   

Consumption by dipping small cup‐type vessels/bowls (Miniatures  Group) into a wide‐mouthed serving vessel (e.g. 2Ai, 2Bi, 2Bii, 3Ai,  3Aii, 3Bi, 3Bii) or by ladling the liquid from these wide mouthed‐ vessels into the cup‐type vessels/bowls (Miniatures Group).  

After fermentation the  grains are separated by  pouring the liquid  through a perforated  vessel which contains a  filtrating medium such as  straw.  The liquid passes  into a large wide‐ mouthed serving vessel  (e.g. 2Ai, 2Bi, 2Bii, 3Ai,  3Aii, 3Bi, 3Bii). 

Fig 13. A hypothetical early Anglo-Saxon brewing process, based on ethnographic, archaeological and historical evidence (see Figure 7 for the relevant vessel-types). 56

All Form One and One Form All Acknowledgments

Leahy, Chris Woolgar, and Chris Cumberpatch and Paul Blinkhorn for editing this volume, arranging the TAG session and inviting me to contribute. Thanks also the Arts and Humanities Research Council for funding this research, part of a PhD undertaken at the University of Sheffield.

I am grateful to the following people for the assistance in the preparation on this paper: Dawn Hadley, Peter Day, Vicky Crewe and Jane Young, Rose Nicholson of North Lincolnshire Museum (Scunthorpe) for providing access the urns and excavation archives, Freda Berisford, Kevin Max Dia (cm)

Vessel

Height (cm)

Height of Max Dia (cm)

Rim Dia (cm)

Ratio 1

Ratio 3

Ratio 5

Figure

Luo Water Jars 1 2 3 4 5 6 Average

52 49 53 53 46 47

52 53 58 66 47 48

23 23 26 27 17 21

34 26 29 26 25 27

1.00 0.92 0.91 0.80 0.98 0.98 0.93

0.62 0.77 0.75 0.69 0.70 0.74 0.71

0.65 0.53 0.55 0.49 0.54 0.57 0.56

3 3 3 3 3 3

30 33

14 17

38 40

1.17 1.09 1.13

-0.19 -0.25 -0.22

1.09 1.11 1.10

3 3

38 25

31 37

0.81 1.17

0.81 0.81

0.52 0.69

6 6

Luo Beer Drinking Vessels 7 8 Average

35 36

Gamo Beer Jar and Serving Bowl C F

60 54

74 46

Table 1: Ratio values calculated for ethnographic pottery forms shown in Figures 3 and 6. Max Dia (cm)

Vessel 2319 3402 1885 1636 632 1609 1571

56 57 58 61 64 61 60

Height of Height (cm) Max Dia (cm) 60 32 60 27 55 26 60 33 46 25 46 25 45 28

Rim Dia (cm) 26 30 25 29 42 44 45

Ratio 1

Ratio 3

0.93 0.95 1.05 1.02 1.39 1.33 1.33

Ratio 5

1.07 0.82 1.14 1.19 1.05 0.81 0.88

0.46 0.53 0.43 0.48 0.66 0.72 0.75

Table 2: The ratio values calculated for urns shown in Figures 2 (data derived from Myres 1977, Figures 105, 128, 151, 208 and 225). Vessel

Max Dia (cm) A B C D

15 45 60 36

Height (cm)

Rim Dia (cm)

18 63 74 46

Ratio 1

8 28 31 20

Table 3: Ratio values of Gamo pottery (see Figure 6). 57

0.83 0.71 0.81 0.78

Ratio 5 0.53 0.62 0.52 0.56

1Ai 1Aii 1Aiii 1Bi 1Bii 1Biii 1Di 1Dii 1Diii 2Ai 2Aii 2Aiii 2Bi 3Ai 3Aiii 3Bi 3Bii 3Biii 4Ai 4Aii 4Aiii 4Aiv 4Biii 4Biv 5Ai 5Aii 5Bi 5Bii 5Biii 6A

Form

Height (mm) Cleatham 260 (244‐277) 218 (180‐244) ‐ 207 (170‐221) 184 (171‐200) 145 (140‐150) 218 (203‐230) 194 (179‐205) ‐ 188 (177‐198) ‐ ‐ 171 (152‐184) 219 (203‐236) 124 (109‐135) 191 (168‐207) 161 (145‐178) 134 (112‐148) 232 (211‐245) 203 (187‐220) 181 (163‐198) 141 (118‐174) 177 (160‐185) 163 (162‐163) 279 (255‐3.10) ‐ 234 (215‐273) 202 140 (115‐166) 227 (196‐255)

58

Width (mm) Cleatham 289 (250‐316) 236 (215‐262) ‐ 256 (230‐284) 215 (200‐236) 164 (150‐170) 230 (217‐242) 191 (178‐200) ‐ 276 (254‐302) ‐ ‐ 238 (220‐256) 282 (248‐310) 162 (145‐176) 248 (240‐258) 211 (190‐232) 178 (160‐190) 268 (246‐284) 231 (216‐248) 212 (182‐240) 164 (136‐200) 207 (200‐215) 191 (182‐200) 276 (262‐310) ‐ 234 (204‐260) 190 147 (120‐168) 292 (270‐322) Elsham 265 (229‐306) 230 (196‐254) 162 (154‐167) 256 (240‐275) 234 (219‐257) 172 218 (194‐236) 202 (188‐217) 173 (167‐178) 269 (258‐290) 249 (236‐264) 196 (194‐197) ‐ 275 (260‐292) ‐ 241 (233‐251) 215 (194‐233) ‐ 279 (254‐294) 224 (210‐242) 185 (168‐207) 145 (139‐153) ‐ ‐ 240 (219‐254) 169 (156‐182) 208 (192‐222) 166 (153‐181) 153 ‐

Rim Dia (mm) Cleatham 129 (100‐160) 101 (86‐122) ‐ 124 (98‐140) 111 (86‐142) 86 (72‐101) 126 (116‐140) 111 (102‐122) ‐ 190 (170‐204) ‐ ‐ 154 (138‐180) 197 (170‐224) 102 (60‐120) 193 (165‐214) 160 (126‐186) 128 (112‐137) 166 (152‐180) 148 (130‐170) 137 (108‐159) 110 (91‐130) 165 (160‐182) 141 (136‐146) 169 (156‐188) ‐ 162 (142‐188) 138 108 (90‐122) 163 (146‐178) Elsham 119 (100‐139) 116 (97‐133) 85 (75‐96) 126 (114‐136) 115 (100‐126) 86 116 (96‐142) 110 (89‐125) 100 (94‐106) 177 (160‐206) 190 (174‐201) 115 (111‐118) ‐ 190 (183‐197) ‐ 182 (144‐208) 149 (139‐158) 121 171 (160‐188) 153 (139‐171) 120 (108‐139) 111 (99‐125) ‐ ‐ 161 (139‐189) 107 (94‐125) 150 (144‐156) 118 (108‐131) 111 ‐

Volume (l) Cleatham 8.1 (5.3‐10.4) 4.6 (3.4‐6.0) ‐ 5.2 (4.0‐6.2) 3.1 (2.5‐3.8) 1.3 (1.2‐1.4) 4.2 (3.4‐5.2) 2.7 (2.2‐3.4) ‐ 6.0 (5.1‐7.2) ‐ ‐ 3.7 (2.8‐4.6) 7.8 (5.5‐10.4) 1.1 (0.9‐1.5) 5.2 (4.4‐6.3) 3.1 (2.3‐3.7) 1.7 (1.3‐1.8) 6.7 (5.5‐7.7) 4.4 (3.3‐5.9) 3.3 (1.9‐4.3) 1.2 (0.8‐1.8) 3.6 (2.7‐4.1) 2.4 (2.2‐2.6) 9.2 (7.1‐12.0) ‐ 5.3 (4.0‐6.5) 2.50 1.2 (0.6‐1.8) 8.2 (6.1‐11.0) Elsham 6.7 (4.5‐9.2) 4.4 (2.7‐5.6) 1.2 (1.1‐1.2) 5.4 (5.2‐5.5) 3.8 (2.9‐4.3) 1.70 4.3 (3.2‐5.3) 3.0 (2.4‐3.8) 2.0 (1.9‐2.1) 6.3 (5.4‐7.8) 4.8 (3.4‐5.5) 2.2 (2.0‐2.3) ‐ 7.7 (7.1‐8.7) ‐ 5.4 (5.1‐5.7) 3.3 (2.4‐4.0) 1.50 8.1 (6.1‐10.2) 4.4 (3.3‐5.2) 2.3 (1.8‐3.2) 1.0 (0.9‐1.2) ‐ ‐ 5.9 (4.4‐6.8) 1.9 (1.7‐2.3) 3.7 (3.1‐4.3) 2.1 (1.8‐2.5) 1.20 ‐

Ratio 1 Cleatham 1.11 (0.99‐1.26) 1.08 (0.96‐1.23) ‐ 1.24 (1.10‐1.44) 1.17 (1.03‐1.31) 1.13 (1.00‐1.21) 1.06 (1.01‐1.13) 0.99 (0.96‐1.0) ‐ 1.47 (1.41‐1.57) ‐ ‐ 1.39 (1.27‐1.55) 1.29 (1.20‐1.40) 1.31 (1.25‐1.38) 1.30 (1.16‐1.52) 1.31 (1.22‐1.86) 1.32 (1.25‐1.43) 1.15 (1.10‐1.22) 1.14 (1.07‐1.26) 1.17 (1.06‐1.28) 1.17 (1.09‐1.27) 1.17 (1.11‐1.28) 1.18 (1.12‐1.23) 0.99 (0.95‐1.05) ‐ 1.00 (0.92‐1.08) 0.94 1.05 (0.96‐1.14) 1.29 (1.14‐1.44) Elsham 1.08 (0.95‐1.18) 1.06 (0.95‐1.13) 1.10 (1.08‐1.13) 1.21 (1.13‐1.31) 1.28 (1.13‐1.41) 1.27 0.95 (0.85‐1.09) 1.02 (0.96‐1.09) 1.06 (1.02‐1.10) 1.37 (1.28‐1.44) 1.46 (1.38‐1.52) 1.35 (1.33‐1.37) ‐ 1.22 (1.17‐1.30) ‐ 1.24 (1.14‐1.30) 1.30 (1.28‐1.36) ‐ 1.10 (1.05‐1.14) 1.13 (1.02‐1.21) 1.14 (1.05‐1.24) 1.20 (1.11‐1.29) ‐ ‐ 1.0 (0.97‐1.04) 0.99 (0.96‐1.03) 0.99 (0.96‐1.03) 0.89 (0.74‐1.01) 1.08 ‐

Table 4: Form-group characteristics – mean and (range) values (see Figure 7 for examples of each type)

Elsham 246 (224‐292) 217 (192‐240) 147 (142‐152) 211 (206‐219) 183 (161‐200) 136 229 (215‐254) 197 (186‐210) 164 (160‐171) 196 (183‐206) 171 (156‐181) 145 (144‐146) ‐ 226 (219‐238) ‐ 194 (182‐204) 166 (143‐182) ‐ 253 (243‐261) 198 (178‐215) 163 (142‐179) 120 (118‐125) ‐ ‐ 239 (226‐253) 170 (151‐185) 209 (200‐215) 187 (169‐207) 142 ‐

Ratio 5 Cleatham 0.45 (0.36‐0.57) 0.43 (0.36‐0.50) ‐ 0.49 (0.38‐0.55) 0.52 (0.42‐0.61) 0.52 (0.43‐0.60) 0.55 (0.50‐0.61) 0.58 (0.54‐.061) ‐ 0.69 (0.63‐0.80) ‐ ‐ 0.65 (0.58‐0.74) 0.70 (0.63‐0.77)  0.68 (0.60‐0.76) 0.78 (0.68‐0.87) 0.76 (0.61‐0.87) 0.72 (0.61‐0.83) 0.62 (0.58‐0.69) 0.64 (0.56‐0.78) 0.64 (0.59‐0.75) 0.67 (0.60‐0.82) 0.80 (0.76‐0.85) 0.74 (0.73‐0.75) 0.61 (0.56‐0.66) ‐ 0.70 (0.60‐0.79) 0.73 0.74 (0.62‐0.92) 0.56 (0.53‐0.60)

Elsham 0.45 (0.36‐0.55) 0.51 (0.42‐0.58) 0.53 (0.47‐0.62) 0.49 (0.49‐0.59) 0.49 (0.39‐0.57) 0.50 0.53 (0.49‐0.62) 0.55 (0.45‐0.64) 0.58 (0.57‐0.59) 0.66 (0.62‐0.71) 0.77 (0.73‐0.81) 0.59 (0.56‐0.61) ‐ 0.69 (0.63‐0.76) ‐ 0.76 (0.60‐0.86) 0.70 (0.65‐0.73) ‐ 0.61 (0.56‐0.66) 0.68 (0.63‐0.75) 0.65 (0.58‐0.77) 0.76 (0.71‐0.82) ‐ ‐ 0.67 (0.61‐0.74) 0.63 (0.61‐0.69) 0.72 (0.70‐0.75) 0.71 (0.68‐0.72) 0.73 ‐

Gareth Perry

All Form One and One Form All

Form

Cleatham

%Cleatham

Elsham

%Elsham

1Ai 1Aii 1Aiii 1Bi 1Bii 1Biii 1D 1Dii 1Diii 2Ai

24 15 0 22 18 4 12 7 0 8

8% 5% 0% 8% 6% 1% 4% 2% 0% 3%

20 17 4 5 6 1 17 11 3 4

13% 11% 3% 3% 4% 1% 11% 7% 2% 3%

2Aii 2Aiii 2Bi 3Ai 3Aiii 3Bi 3Bii 3Biii 4Ai 4Aii 4Aiii 4Aiv 4Biii 4Biv 5Ai 5Aii 5Bi 5Bii 5Biii 6A Ungrouped

0 0 19 12 7 12 15 7 8 24 11 13 6 2 6 0 12 1 10 10 32 317

0% 0% 7% 4% 2% 4% 5% 2% 3% 8% 4% 5% 2% 1% 2% 0% 4% 0% 4% 4% 10% 100%

4 2 0 4 0 5 5 1 5 8 11 3 0 0 5 5 3 4 1 0 19 173

3% 1% 0% 3% 0% 3% 3% 1% 3% 5% 7% 2% 0% 0% 3% 3% 2% 3% 1% 0% 11% 100%

Table 5: The frequencies of occurrence of each form type at the cemeteries of Elsham and Cleatham. Note that these frequencies only relate to decorated vessels with reconstructible profiles.

59

Gareth Perry Form Group 4Ai 4Aii 1Di 5Bi 1Di 2Ai 2Ai 1Di 1Di 2Ai 1Di

Urn Number 1403 3258 3402 138.3 139.2 632 1571 1636 2319 1609 1885

Cemetery Loveden Hill (Lincs) Loveden Hill (Lincs) Sancton (Yorks) Mucking (Essex) Mucking (Essex) South Elkington (Lincs) Caistor-by-Norwich (Norfolk) Caistor-by-Norwich (Norfolk) Sancton (Yorks) Caistor-by-Norwich (Norfolk) Caistor-by-Norwich (Norfolk)

Table 6: The typology of form developed in this chapter is applicable to pottery obtained from cemeteries across England (see Figure 2 for relevant urns).

Vessel Characteristic

Properties - Easy access - Facilitates filling

Orifice – unrestricted

- Aides the use of the hands/utensils when stirring or extracting the contents - Extraction of liquids by dipping/ladling - Extraction of dry goods by scooping - Difficult to extract contents

Orifice – restricted

Lids Tall Shallow and wide mouthed

Large

- Often contain liquids as they facilitate pouring, prevent spillage and inhibit evaporation - Allow closure with wooden/ceramic/skin/textile lids - May be associated with longer term storage due to restrictions on access - Commonly associated with liquids - Prevent spillage, contamination and evaporation - Restricted access - Longer term storage (weeks or months) - Immediate access to contents - Low centre of gravity providing stability - Longer term storage - Difficult to manoeuvre when full - Communal consumption - Might indicate shorter term storage

Small

- Easy to manoeuvre when full - Personal consumption

Table 7: The functional attributes of vessel characteristics (based on Henrickson and McDonald 1983)

60

All Form One and One Form All

Cemetery

County

Urn Number

Fig.

Reference

Baston

Lincolnshire

42

10

Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Spong Hill Lackford

Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Norfolk Suffolk

36 70 51 52 52 53 51 73 74 74 80 81 82 88 88 89 89 90 90 91 92 101 100 105 104 119 15

Hills et al. 1987 Hills et al. 1987 Hills et al. 1987 Hills et al. 1987 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Hills and Penn 1981 Lethbridge 1951

1Bi 1D 1Ai 1Ai 1Bii 1Bi 1Ai 1Ai 4Aiii 4Aiii 4Aii 1Bii ? ?1Bii 1Biii 1Ai 1Bii 1Ai 1Bi 1Bi ?1Ai 4Aiv 1Ai 4Aiii 1Bi 1Bii 1Ai

Lackford

Suffolk

15

Lethbridge 1951

?5Aii

Lackford Newark

Suffolk Nottinghamshire

1360 1085 2483 2586 2531 2642 2056 1835 1875A 1875B 1936 1806 1963 1991 1784 1778 2090 2099 1772 2035 1753 1791 2048 2111 1892B 2011 49,18(6) 48,2494(HG, 13) 50,17(A,8) 3556

15 276

Lethbridge 1951 Myres 1977

?1Ai ?

Mayes and Dean 1976 Hills 1977 Hills 1977

Table 8: The forms of urns found with lids

61

Form Group 1Bi

Gareth Perry

Appendix A:

Cleatham urns attributed to each of the form groups. Images of these urns are viewable on the Archaeology Data Service Website (Leahy 2007c): http://archaeologydataservice.ac.uk/archives/view/cleatham_cba_2007/ doi:10.5284/1000011

Form Group 1Ai 1Aii 1Bi 1Bii 1Biii 1Di 2Ai 2Bi 3Ai 3Aiii 3Bi 3Bi 3Biii 4Ai 4Aii 4Aiii 4Aiv 4Biii 4Biv 5Ai 5Bi 5Bii 5Biii 6A

Urn Number 58,60,71,137,242,265,285,336,371,373,396,479,492,498,509,566,768, 788,791,865,887,900,911,1000 415, 422, 344, 520, 536, 548, 577, 261, 594, 699, 728, 861, 890, 988, 1004 12, 56, 136, 200, 204, 255, 283, 364, 458, 513, 523, 573, 582, 598, 773, 889, 944, 951, 991, 1104, 1105 129, 171, 172, 175, 196, 214, 286, 361, 404, 562, 211, 279, 333, 585, 640, 934, 977, 957 183, 318, 394, 528 52, 98, 293, 328, 353, 471, 544, 624, 693, 730, 807, 870, 830, 845, 851, 1070, 829, 600, 739 83, 212, 270, 330, 596, 689, 815 370, 109, 115, 176, 308, 454, 460, 488, 521, 572, 587, 636, 641, 704, 871, 956, 978, 391, 1031, 1103 89, 91, 291, 306, 326, 384, 386, 398, 742, 895, 897 121, 90, 124, 304, 470, 431, 943 63, 64, 68, 233, 262, 327, 1026, 483, 938, 763, 1076, 1068, 64, 222, 332, 808, 80, 178, 316, 374, 459, 468, 534, 864, 875, 1097, 967, 1100 4, 590, 616, 123, 358, 397, 789 383, 567, 599, 65, 465, 237, 487, 690 141, 69, 113, 192, 194, 258, 284, 294, 325, 331, 351, 388, 519, 622, 654, 705, 707, 762, 856, 873, 961, 990, 953, 1027 73, 257, 464, 632, 785, 796, 402, 1059, 1058, 1005 82, 72, 244, 251, 403, 489, 760, 941, 188, 199, 193, 1094, 813 575, 228, 296, 349, 400, 452, 463 412, 744 101, 77, 210, 213, 425, 954 119, 153, 198, 202, 232, 239, 259,273, 322, 429, 173, 1095 236 57, 29, 169, 877, 340, 382, 390, 1015, 484, 692 145, 219, 532, 565, 583, 639, 759, 777, 793, 1136

62

All Form One and One Form All Bibliography

Document Technique Économique, Culturel. JuanLes-Pins: APDCA, 459-472. Deitler, M. and Herbich, I. 1998. Habitus, techniques, style: an integrated approach to the social understanding of material culture and boundaries. In Stark, M.T. (ed.) The Archaeology of Social Boundaries. Washington: Smithsonian Institution Press, 233-263. Deitler, M. and Herbich, I. 2006. Liquid material culture: following the flow of beer among the Luo of Kenya. In Witzka, H-P. (ed.) Grundlegungen: Beiträischen und Afrikanischen Archäologie für Manfred K.H. Eggert. Tübingen: Franke, 395-407. Evison, V. 1979. A Corpus of Wheel-thrown Pottery in Anglo-Saxon Graves. Leeds: Maney. Fennell, K.R. 1964. The Anglo-Saxon Cemetery at Loveden Hill. Lincolnshire, and its Significance in the Relation to the Dark Age Settlement of the East Midlands. Unpublished doctoral thesis, University of Nottingham. Garine, E. 2001. An ethnographic account of the many roles of millet beer in the culture of the Duupa agriculturalists, (Poli Mountains) Northern Cameroon. In de Garine, I. and de Garine, V. (eds) Drinking: Anthropological Approaches. New York: Berghan Books, 191-204. Hally, D. 1983. Use alteration of pottery vessel surfaces: an important source of evidence for the identification of vessel function. North American Archaeologist 4 (1), 3-36. Hamerow, H. 1993 Excavations at Mucking: Vol.2. London: English Heritage. Harden, M. 1978. Anglo-Saxon and later medieval glass in Britain: some recent developments. Medieval Archaeology 22, 1-24. Henrickson, E.F. and McDonald, M.M.A. 1983. Ceramic form and function: an ethnographic search and an archaeological application. American Anthropologist 85 (3), 630-643. Herbich, I. 1987. Learning patterns, potter interaction and ceramic style among the Luo of Kenya. The African Archaeological Review 5, 193-204. Hills, C. 1977. The Anglo-Saxon Cemetery at Spong Hill, North Elmham. Part I. Norwich: East Anglian Archaeology. Hills, C. and Penn, K. 1981. The Anglo-Saxon Cemetery at Spong Hill, North Elmham. Part II: Catalogue of Cremations. Norwich: East Anglian Archaeology. Hills, C., Penn, K. and Rickett, R. 1984. The AngloSaxon Cemetery at Spong Hill, North Elmham. Part III: Catalogue of Inhumations. Norwich: East Anglian Archaeology. Hills, C., Penn, K. and Rickett, R. 1994. The AngloSaxon Cemetery at Spong Hill, North Elmham. Part V: Catalogue of Cremations. EAA Report 67. Norfolk: East Anglian Archaeology. Hirst, S. and Clark, D. 2009. Excavations at Mucking: Volume 3, the Anglo-Saxon Cemeteries (2 vols). London: Museum of London Archaeology.

Arnold, C.J. 1988. An Archaeology of the Early AngloSaxon Kingdoms. London: Routledge. Arthur, J.W. 2002. Pottery use-alteration as an indicator of social status: an ethnoarchaeological study of the Gamo of Ethiopia. Journal of Archaeological Method and Theory 9 (4), 331-355. Arthur, J.W. 2003. Brewing beer: status, wealth and ceramic use alteration among the Gamo of southwestern Ethiopia. World Archaeology 34 (3), 516-528. Arthur, J.W. 2006. Living with Pottery: Ethnoarchaeology Among the Gamo of Southwest Ethiopia. Salt Lake City: University of Utah Press. Attenborough, F.L. 1922. The Laws of the Earliest English Kings. Cambridge: Cambridge University Press. Birmingham, J. 1975. Traditional potters of Kathmandu Valley: an ethnoarchaeological study. Man 10 (3), 370-386. Blinkhorn, P. 1997. Habitus, social identity and AngloSaxon pottery. In Blinkhorn, P. and Cumberpatch, C. (eds) Not So Much a Pot, More a Way of Life. Oxford: Oxbow, 113-124. Blinkhorn, P. 1999. The trials of being a utensil: pottery function at the medieval hamlet of West Cotton, Northamptonshire. Medieval Ceramics 22-23, 37-46. Bostock, J. and Riley, H.T. (trans.) 1855. The Natural History, Pliny the Elder. London: Taylor and Francis. Bradley, S.A.J. 1982. Anglo-Saxon Poetry: An Anthology of Old English Poems in Prose Translation with Introduction and Headnotes. London: Dent. Church, A. and Brodribb, W.J. (trans.) 1942. Complete Works of Tacitus. Tacitus. New York: Random House. Clark, P. 1983. The English Ale House. London: Longman. Cockayne, O. 1865. Leechdoms, Wortcunning and Starcraft of Early England Volume II. London: Longman. Cook, J. 2009. Discussion. In Hirst, S. and Clark, D. 2009. Excavations at Mucking: Volume 3, the AngloSaxon Cemeteries (2 vols). London: Museum of London Archaeology, 55-6. Corran, H.S. 1975. A History of Brewing. London: David and Charles. DeBoer, W.R. and Lathrap, D.W. 1979. The making and breaking of Shipibo-Conobo ceramics. In Kramer, C. (ed.) Ethnoarchaeology: Implications of Ethnography for Archaeology. New York: University of Columbia Press, 102-138. De Garine, I. 2011. Drinking, ritual and conviviality in Northern Cameroon. In Schiefenhovel, W. and McBeth, H. (eds) Liquid Bread: Beer and Brewing in Cross-Cultural Perspective. New York: Berghahn Books, 133-46 Dietler, M. and Herbich, I. 1989. The technology of Luo pottery production and the definition of ceramic style. World Archaeology 21(1), 148-64. Deitler, M. and Herbich, I. 1994. Ceramics and ethnic identity: ethnoarchaeological observations on the distribution of pottery styles and the relationships between the social contexts of production and consumption. Terre Cuite et Société. La Céramique, 63

Gareth Perry Kempton, W. 1981. The Folk Classification of Ceramics: a Study of Cognitive Prototypes. New York: Academic Press. Laing, L. and Laing, J. 1979. Anglo-Saxon England. London: Book Club Associates. Labov, W. 1973. The boundaries of words and their meanings. In Bailey, C-J. N. and Shuy, R.W. (eds) New Ways of Analysing Variation in English. Washington: Georgetown University Press, 340-73. Leahy, K. 2007a. ‘Interrupting the Pots’: The Excavations of Cleatham Anglo-Saxon Cemetery. CBA Research Report 155. Oxford: CBA. Leahy, K. 2007b. The Anglo-Saxon Kingdom of Lindsey. Stroud: Tempus Publishing. Leahy, K. 2007c. The Excavation of the Cleatham AngloSaxon Cemetery, North Lincolnshire. Available at: http://ads.ahds.ac.uk/catalogue/archive/cleatham_ cba_2007/index.cfm?CFID=573996&CFTOK EN=39545028. Linares de Sapir, O. 1969. Diola Pottery of the Fogny and the Kasa. Expedition 11 (3), 2-11. Longacre, W. 1981. Kalinga pottery: an ethnoarchaeological study. In Hodder, I., Isacc, G. and Hammond, N. (eds) Pattern of the Past: Studies in Honour of David Clarke. Cambridge: Cambridge University Press, 49-66. Mayes, P. and Dean, M.J. 1976. An Anglo-Saxon Cemetery at Baston, Lincolnshire. Occasional Papers in Lincolnshire History and Archaeology. No. 3. Sleaford: The Society for Lincolnshire History and Archaeology. McKinley, J. 1994. Spong Hill: Part VIII: The Cremations. East Anglian Archaeology Report No. 69. Norfolk: Norfolk Archaeological Unit. Myres, J.N.L. 1969. Anglo-Saxon Pottery and the Settlement of England. Cambridge: Cambridge University Press. Myres, J.N.L. 1977. A Corpus of Anglo-Saxon Pottery of the Pagan Period, 2 vols. Cambridge: Cambridge University Press. Perry, G.J. 2011. Beer, butter and burial: the pre-burial origins of cremation urns form the early Anglo-Saxon cemetery of Cleatham, North Lincolnshire. Medieval Ceramics 32, 9-21.

Perry, G.J. 2012. A hole for the soul? Possible functions of post-firing perforations and lead plugs in early AngloSaxon cremation urns. In Jervis, B, and Kyle, A. (eds) Make-Do and Mend: Archaeologies of Compromise, Repair and Reuse. BAR International Series 2408. Oxford: BAR, 43-52. Perry, G. J. 2013 United in Death: the pre-burial origins of Anglo-Saxon cremation urns. Unpublished PhD thesis, University of Sheffield. Ravn, M. 2003. Death Ritual and Germanic Social Structure AD 200-600. BAR International Series 1164. Oxford: BAR. Reina, R.E. and Hill, R.M.1978. The Traditional Pottery of Guatemala. Austin: University of Texas Press. Rice, P. 2005. Pottery Analysis: a Sourcebook. Chicago: University of Chicago Press. Richards, J.D. 1982. Anglo-Saxon pot shapes: cognitive dimensions. Science and Archaeology 24, 33-46. Richards, J.D. 1987. The Significance of Form and Decoration of Anglo-Saxon Cremation Urns. BAR British Series 166. Oxford: BAR. Skibo, J.M. 1992. Pottery Function: a Use-Alteration Perspective. New York: Plenum Press. Stark, M. 1999. Social Dimensions of Technical Choice in Kalinga Ceramic Traditions. In Chilton, E.S. (ed.) Material Meanings Critical Approaches to the Interpretation of Material Culture. Salt Lake City: University of Utah Press, 24-43. Stone, D.J. 2006. The consumption of field crops in late Medieval England. In Woolgarth, M., Serjeantson, D. and Waldron, T. (eds) Food in Medieval England: Diet and Nutrition. Oxford: Oxford University Press, 1126. Stoodley, N. 1999. The Spindle and the Spear. A Critical Enquiry into the Construction and Meaning of Gender in the Early Anglo-Saxon Burial Rite. BAR British Series 288. Oxford: BAR. Wilson, D. 1965. The Anglo-Saxons. London: Thames and Hudson.

64

Plates and other vessels from early modern and recent graves Beth Richardson 1

Early modern grave goods are extremely rare but recent excavations by Museum of London Archaeology (MOLA) of eighteenth- and nineteenth-century cemeteries in London have produced seven burials with ceramic and pewter plates, dishes and bowls placed on the chest, stomach or pelvis of adults and children. This paper describes these burials and the few other examples currently known from Britain and elsewhere, and explores superstitious and spiritual beliefs and practical reasons behind the practice. Some of these vessels probably contained salt which, as a preservative, symbolised eternal life and was also thought to ward off the devil and postpone decomposition (eg Bickerstaffe 1784, Brand 1813, Noël Hume 1974, Tarlow 2011). Others may have contained bread or bread and salt for the practice of sin-eating, although this is considered less likely, or simply have been personal possessions.1 The London burials The earliest London example of the practice is a tin-glazed earthenware plate with ‘chinaman and rocks’ motif dated c 1670-90, found inverted on the pelvis of an adult skeleton in the churchyard of St Martin Vintry, City of London. This was one of hundreds of early modern coffin contents from bombed churches and churchyards recorded in the years after the Second World War by Ivor Noël Hume and others but the only one to have contained a plate (Noël Hume 1974, 169-72). The circumstances of the find were more or less forgotten although the plate, a particularly fine example of its type, was displayed and subsequently stored in the Museum of London’s reserve collection (Fig 1).

Fig 1. Museum of London et al, 2013, 91). At New Bunhill Fields two plates were buried with Mrs Elizabeth Bailey who died aged 25 in 1833: a transfer-printed ‘willow pattern’ plate on her upper left arm and a New Hall porcelain plate on her pelvis (Figs 5, 6). From the same burial ground a transferprinted ‘willow pattern’ tea plate was found in the coffin of Lucy Emily Williamson Morris who died in 1848 aged 10 weeks, and a thin and corroded pewter plate was found face down on the left side of the pelvis of an adolescent of indeterminate sex (Fig 7); (Pearce & Richardson in Connell & Miles 2012, 50-3).

Further plates, dishes and bowls were found during recent large-scale excavations of several eighteenth- and nineteenth-century burial grounds. The earliest are from the extramural burial ground for St James, Piccadilly where a small salt-glazed stoneware dish dated c 17201780 was found on the pelvis of an adult of undetermined sex, an early to mid 18th-century tin-glazed earthenware plate on the pelvis of an adult female (Fig 2), and a thin pewter plate, in very poor condition, by the left arm of an adult female (Fig 3); (Davis et al 2010, 26). The others are early to mid nineteenth-century, from excavations at St Mary & St Michael’s Catholic burial ground in Shadwell, and New Bunhill Fields non-denominational burial ground in Southwark. At St Mary & St Michael’s a white-earthenware tea plate with impressed blue shelledged decoration dated c. 1840-70 was found in the coffin of a child aged 1-5 years (Fig 4); Richardson in Henderson

Other British and continental burials Plates, bowls and other vessels buried as grave goods are known from Leicester, Barton-upon-Humber (Lincolnshire), Wetwang (East Yorkshire), and Norwich. The earliest of these, a corroded and perforated pewter plate dug up with a corpse in St Mary’s churchyard, Leicester in the mid to late eighteenth century, is the first recorded example and may have been a century or so earlier. It was thought sufficiently unusual to be the subject of a note and further correspondence in the Gentleman’s Magazine quoted below (Bickerstaffe 1784, 258; 1785, 603; see also Brand 1813, 235, Noël Hume 1974, 170). The other plates and bowls were found far more recently. In the large-scale excavations at St Peter’s Church, Barton-upon-Humber a white china dinner plate was found beneath the feet of an early nineteenth-century adult

 Senior Finds Specialist Museum of London Archaeology 1

65

Beth Richardson

Fig 2. Maggie Cox/ Museum of London Archaeology

Fig 3. Maggie Cox/

MOLA

66

Plates and other vessels from early modern and recent graves

Fig 6. Maggie Cox/

MOLA

MOLA

Fig 4. Maggie Cox/

Fig 7. Maggie Cox/

67

MOLA

MOLA

Fig 5. Maggie Cox/

Beth Richardson male burial, and a blue and white transfer-decorated china bowl was found below the coffin and beneath the skull of an adult female (Waldron 2007, 26-8). During excavations in the churchyard of St Nicholas’s Church Wetwang a late eighteenth-century Late Blackware cup was found between the feet of an eighteenth- or nineteenth-century burial (Cumberpatch 2010). Another cup, a transferprinted teacup with a peacock design, was found on the ribcage of a probable female aged 17-18 in an eighteenth and nineteenth century Baptist cemetery at Priory Yard, Norwich (Cedifor Caffell, A, & Clarke, R, 2011, 252-3). In Ireland the remains of fourteen metal mugs, thought to be soup mugs, were excavated with mid nineteenth-century burials at Kilkenny Workhouse (information from Siobhán Scully, Margaret Gowen and Co. Ltd) These mugs may simply have been personal possessions, perhaps used to administer last drinks or medication, as indeed may some of the other cups, plates and bowls found with burials.

Both Fremmer, who described a twentieth-century rural Jamaican practice of burying a plate of salt and coffee with the dead (Fremmer, 61), and Parrington & Roberts have suggested the possibility of a fusion between West African and British or European burial practices in the Americas. Discussion The London pewter plates are plain and undecorated, wafer-thin and corroded to the point of disintegration, as was the Leicester plate which was ‘corroded and perforated’ when found. While conditions within the coffin and any salt present would cause corrosion the overall thinness of the metal suggests that the plates were inexpensive; like burial clothes they may have been specifically produced for wake and/or burial purposes. The quality of the ceramic plates, dishes and other vessels may correspond to the social status of the people they were buried with, although also chosen as suitable for purpose. We do not know the identity of the St Martin Vintry burial but the late seventeenth-century ‘chinaman and rock’ plate is a good quality item. The mid-eighteenth-century tin-glazed earthen ware plate and salt-glazed earthenware saucer in the St James Piccadilly Churchyard are good quality but fairly commonplace tableware; although there was a workhouse in a corner of the cemetery they are likely to have come from the graves of fairly well-todo parishioners who were living in this area at the time. In contrast the people buried in St Mary & St Michael’s Catholic burial ground (Irish who had recently fled the Famines) and those in New Bunhill Fields nondenominational burial ground were poor, and the plates and saucers in these cemeteries, although valuable to their owners, are inexpensive everyday items, as are the plates, saucers and cups in the other late eighteenth- and nineteenth-century English and American cemeteries. It must be significant too that most of the burial finds are flat-based plates, dishes, saucers and small tea plates (the latter found with babies or small children). Deeper bowls and cups are far less common, and the two cups may, as suggested be personal possessions, possibly used to administer water or medication, although this is difficult to prove.

The only currently known examples of these burials from continental Europe are from Switzerland. A plate and a bowl were found during recent excavations of 1600 burials at Spitalfriedhof St Johann, Basel. Both were broken and found inverted in similar positions on the chest, one in the coffin of a young adult female, Maria Agnes Haubensack buried in 1854, and the other held in position under the left arm of a young adult male, Eduard Beck, who died in 1853 (Kramis et al, 2009). There is however a description of the practice in early twentieth-century Malta: ‘amongst our lower and middle classes the practice still lingers of placing a dish of salt on the stomach of the corpse whether male or female. The popular belief …is that the salt prevents early decomposition’ (Cremona, 1923, 352). The Americas Several eighteenth and nineteenth-century burials with plates and saucers are known from Jamaica and North America. A plate and a saucer found with two eighteenthcentury colonial English burials in Jamaica are almost certainly an extension of the British tradition. Here a feather-edged Creamware plate dated c 1770-80 was found resting against the pelvis of an adult male in a coffin buried on a sugar plantation in Trelawny, north-west Jamaica and a white salt-glazed stoneware saucer dating from the second or third quarter of the eighteenth century was found in the family mausoleum of the Ricketts family of Westmoreland, south-western Jamaica. (Fremmer, 1973, 58-9). In North America several plates and saucers have been found from excavations of nineteenth- and twentiethcentury African-American burial grounds. An adult burial and a child burial were found with plates placed on their stomachs at the early nineteenth-century First African Baptist Church burial grounds, Philadelphia (Parrington & Roberts 1990, 150-1), a saucer was found on the chest of an adult female at the late nineteenth- and early twentieth- century Elko Switch cemetery near Huntsville Alabama (Shogren et al 1989) and another beneath the left pelvis of a female at Cedar Grove Cemetery in Arkansas (Rose 1985, 96; Niquette, Stallings & Mauck, 1994).

It seems most likely that most of the plates and other vessels contained salt. The origins of this common practice are obscure. No instances are recorded from an extensive survey of burial practice in Britain in the later medieval period (Gilchrist & Sloane 2005; Gilchrist, 2008), although neither ceramic nor pewter plates or other vessels would have been as readily available as in the early modern period. It is possible that wooden plates or vessels were used or salt placed directly on the body. Salt may in any case have been a much less expendable commodity in medieval times. The first known written reference to the practice is the correspondence in the Gentleman’s Magazine during 1784

68

Plates and other vessels from early modern and recent graves and 1785 (Bickerstaffe 1784) concerning a burial found with a pewter plate in a Leicester graveyard:

Professor Evans of the Presbyterian College, Carmarthen, [who] actually saw a sin-eater about the year 1825, who was then living near Llanwenog, Cardiganshire...only when a death took place did they seek (the sin-eater) out, and when his purpose was accomplished they burned the wooden bowl and platter from which he had eaten the food handed across, or placed on the corpse for his consumption’ (Puckle, 1926)

Some years ago, in St Mary’s church, Leicester, a [corroded, and perforated] pewter plate was dug up … conjectured to have been laid, charged with salt, on the body of a deceased, and forgotten to be withdrawn. I saw and examined it myself. [W BICKERSTAFFE]

In an example from late nineteenth-century Ely the bread was placed directly on the body so again there would have been no archaeological evidence (Tarlow, 2011 citing Porter 1969).

Your correspondents … seem to have both adopted a wrong opinion, in relation to the plate in the coffin. Instead of a pewter plate, well filled with coarse salt, laid on the corpse, as is still the custom in several counties of England, it is most probable, that it was a patten placed on the breast of the deceased, to shew he had been a priest. For the pewter plate and salt, laid on a corpse with the intent to prevent air getting into the bowels, and so swelling up the belly as to occasion either a bursting, or at least a difficulty in closing the coffin, are both removed before the coffin is shut up, and never buried with the corpse.

It is also possible that some of the vessels found with burials - particularly perhaps drinking vessels such as the mug from Wetwang Church graveyard and the tea cup from Baptist cemetery at Priory Yard Norwich - may have been treasured possessions or the last item used by the deceased. Hole (1940, 53) tells how: ‘a woman in a Lincolnshire parish laid the broken pieces of her husband’s favourite jug and mug on his grave because she was “that moidered wi’ crying that I clean forgot to put ‘em along of him in t’coffin”’. It is of course a practice which has continued and may be increasing today, as with the many grave goods of Alan Clark MP (d 1999; Clark 2002).

*C*C* A few years later the antiquarian John Brand recorded the custom in his Popular Antiquities (Brand 1795 etc, 234-5). In his native county of Northumberland ‘a pewter plate containing a little salt [was placed] on the corpse’, in Ireland the plate of salt was placed over the heart and in Scotland unmixed [portions of] salt and earth were placed on the breast in a wooden platter. The salt was said to be ‘an emblem of the immortal spirit’, a preservative (quoting the Gentleman’s Magazine correspondence) and a deterrent to the Devil (‘the Devil loveth no salt on his meat’). One of the last British records, also from the north, is that by the Coroner for the Eastern division of Cumberland in a letter of 1920 to Sir James Frazer:

Some conclusions

The tradition of sin-eating also seems to have involved placing a plate on a corpse. A Mr Moggridge of Swansea told the Cambrian Archaeological Association in 1852

Many if not all of the pewter and ceramic plates, dishes and other vessels found with burials must have contained salt. Widespread references but uncommon finds must reflect the fact that they were normally removed before burial. The Gentleman’s Magazine correspondence refers to the plate being laid on the body and forgotten to be withdrawn or removed and never buried with the corpse. It is also likely that the custom had been common but, like other magical beliefs, declined after 1700 (Thomas 1971) to become marginalised in the provinces and among poorer and less-educated people. One of the nineteenthcentury burials found with plates from London was that of a recent Irish immigrant and it may well be that the families of the other eighteenth- and nineteenth-century London (and Jamaican) burials also originated from areas where the custom was still practised. London’s nineteenth century burials with plates, dishes and other vessels were those of poor people in Catholic or non-denominational graveyards.

… when a person dies his friends sent for the sineater of the district…placed a plate of salt on the breast of the defunct, and upon the salt a piece of bread. Then he muttered some incantation over the bread, which he ate, thereby eating the sins of the deceased. (Hull, 1928, 225)

It is interesting that the custom seems to have been tolerated by the Catholic and Anglican Churches and by other nondenominational religions. This may reflect the Christian symbolism and reputed preservative qualities of the salt, although it is also possible that the officiating clergy were simply unaware of the coffin contents.

But there are also references to the removal and burning of the vessels. In nineteenth century Wales again:

Contra the assertion of the Cumberland Coroner in 1920, there seems to be no archaeological bias towards adult male burials, although the sample is small. In London the plates,

Another custom, almost universal amongst the poorer classes in the country, is the placing of a saucer of salt on a dead man’s breast I have never noticed it on a woman’s. An undertaker told me it was done for hygienic reasons but I doubt this (Frazer 1920).

69

Beth Richardson London Burials Catalogue

dish and tea-plates were found with three adult women, two children (a female baby and a child of undetermined sex) and three burials of unknown gender. The pattern is similar for the rest of the burials in this study; only three plates were found with adult male burials (at Spitalfriedhof St Johann, Basel, Trelawny sugar plantation, Jamaica and St Peter’s Barton-upon-Humber, although this last example is unusual as the plate was beneath the feet). Most of the plates and other vessels were found on the chest, the stomach or the pelvis; others have obviously been displaced during decomposition. Three vessels were found inverted and at least one of these must have been intentionally placed (a plate under Edward Beck’s left arm in the Basel burial), possibly to make direct contact between and the contents of the vessels and the body. This is an initial survey listing only twenty-one burials, almost half from London. Certain patterns have emerged however, and a large body of documentary evidence suggests that most of the plates and other vessels placed with the bodies would have contained salt. A fuller survey of post-medieval burial finds may reveal more examples found with named individuals with family origins from specific areas of Britain or the Continent and there will almost certainly also be ‘plate burials’ of an earlier date. Many people from MOLA were involved in the excavation of these burials and the conservation and post-excavation work on the plates and other ceramic and pewter finds. I should like to thank them all and particularly Sarah Ritchie who excavated the Marshall Street cemetery and sent me her preliminary research (including the quotations from Hole’s English Folklore and Hull’s Folklore of the British Isles), Jacqui Pearce and Nigel Jeffries who identified and discussed the ceramic plates and dishes and Maggie Cox who took the photographs. I should also like to thank Chris Green who read and contributed to my text. Chris Cumberpatch, Malin Holst and Gerhard Hotz corresponded with me and I am grateful for their comments. Finally it was the pioneer Anglo-American archaeologist Ivor Noël Hume who fifty years ago made the connection between a single plate in a coffin and an ancient tradition – and, as in so many other instances, led the way.

70

•

St Martin Vintry churchyard, City of London. Tinglazed earthenware plate with ‘Chinaman and rocks’ motif typical of the late seventeenth century inverted on the pelvis of an adult skeleton (Noël Hume 1974, 169-72).

•

St James, Piccadilly extramural burial ground, Marshall Street, W1. Tinglazed earthenware plate dated c. 1700-60 face-up on pelvis of adult female. (Davis et al 2010, 26).

•

St James, Piccadilly extramural burial ground, Marshall Street, W1. small salt-glazed stoneware dish dated c. 1720-80 face-up just below the pelvis of an adult of undetermined sex (Davis et al 2010, 26).

•

St James, Piccadilly extramural burial ground, Marshall Street, W1. Pewter plate by left arm of adult female (probably displaced) (Davis et al 2010, 26).

•

St Mary and St Michael’s Catholic burial ground, Lukin Street, E1. Staffordshire white-ware tea plate with impressed blue shell-edged decoration on the rim dated c. 1840-70 in coffin of child aged 1-5 years (Richardson in Henderson et al, 2013, 91).

•

New Bunhill Fields burial ground, Southwark (SE1?). Two plates in the coffin of Mrs Elizabeth Bailey who died in 1833 aged 25. (1) transfer-printed ‘willow pattern’ plate found face-up on upper left arm, (probably displaced) and (2) New Hall porcelain plate with moulded basket-weave border and overglaze painted roses in centre and around rim found faceup on pelvis (Pearce and Richardson in Connell and Miles 2012, 50-3).

•

New Bunhill Fields burial ground, Southwark (SE1?). Transfer-printed ‘willow pattern’ tea plate in coffin of Lucy Emily Williamson Morris who died in 1848 aged 10 weeks (Pearce & Richardson in Connell & Miles 2012, 50-3).

•

New Bunhill Fields burial ground, Southwark (SE1?). Pewter plate found face down on left side of sub-adult pelvis (probably displaced) (Pearce & Richardson in Connell & Miles 2012, 50-3).

Plates and other vessels from early modern and recent graves Bibliography

Hole, C, 1940 English Folklore, London Hull, E, 1928 Folklore of the British Isles, London King, C, & Sayer, S, 2011 The Archaeology of PostMedieval Religion, 249-271 Kramis, Link, F, Hotz, G 2009 Spitalfriedhof: Von der Drehscheibe in den Brustkorb – Eine ungewöhnliche Absolutdatierung von neuzeitlicher Keramik, 30-5, Basel Niquette, C M, Ross-Stallings, N & Mauck, J G, 1994 Dead Men Do Tell Tales: Comments on the Kentucky Transportation Cabinet’s Grave Relocation Procedures Noël Hume, I 1974 All the Best Rubbish, London Parrington, M and Roberts, D G. 1990, Demographic, Cultural, and Bioanthropological Aspects of a Nineteenth Century Free Black Population in Philadelphia, Pennsylvania, in J.E. Buikstra (ed.) In A Life in Science: Papers in Honor of J. Lawrence Angel, Center for American Archaeology Scientific Papers 6, Kampsville, Illinois, 138-170 Porter, E., 1969 Cambridgeshire Customs and Folklore, London Puckle, B S, 1926 Funeral Customs: their origin and development, London Rose, J C (ed.) 1985 Gone to a Better Land, Arkansas Archaeological Survey Research Series No. 25, Fayetteville, Arkansas Shogren, M G, K R Turner, & J C Perroni 1989, Elko Switch Cemetery: An Archaeological Perspective. Alabama State Museum of Natural History, Division of Archaeology, Report of Investigations 58, Tuscaloosa, Alabama Tarlow, S, 2011 Ritual, Belief and the Dead in Early Modern Britain and Ireland, Cambridge Thomas, K, Religion and the Decline of Magic 1971 Waldron, A, 2007 St Peter’s, Barton-upon-Humber, Lincolnshire. A parish church and its community. Vol 2 The human remains Oxford

Bickerstaffe, W, 1784, correspondence with C*C* in Gentlemans Mag 54 (1784), 258 and 55 (1785), 760 Brand, J, 1795 Observations on the Popular Antiquities in Great Britain, ed H Ellis; revised edition Bohn, 1849 Cedifor Caffell, A, & Clarke, R, 2011, The General Baptists of Priory Yard, Norwich in King, C, & Sayer, S, (eds) The Archaeology of Post-Medieval Religion, 249-271 Clark, A, 2002 Alan Clark; the Last Diaries. In and out of the wilderness (ed I Trewin) Connell B, & Miles A, 2012 ‘He being dead yet speaketh’: excavations at three post-medieval burial grounds in Tower Hamlets, east London, 2004-10, MoLA Monog. 64, London Cremona, A, 1923 Maltese Death, Mourning and Funeral Customs, Folklore 34, 352 Cumberpatch, C.G. 2010 An eighteenth century Late Blackware cup from Wetwang cemetery, East Yorkshire Unpublished archive report. Humber Sites and Monuments Record Davis, S, Henderson, M, and Ritchie, S 2010, Marshall Street Baths and Dufours Place cleansing depot London W1; post-excavation assessment and updated project design, unpub MOLA report Frazer, J, 1920 Beliefs regarding death in Cumberland, Folklore 31, 154 Fremmer, R 1973 Dishes in Colonial Graves: Evidence from Jamaica, Historical Archaeology 7, 58-62 Gilchrist, R, 2008 Magic for the Dead? The archaeology of magic in later medieval Britain, Medieval Archaeology 52, 119-59 Gilchrist, R & Sloane, B, 2005 Requiem: the medieval monastic cemetery in Britain. Museum of London Archaeology Service Henderson, M, Miles, A, Walker D 2013 ‘He being dead yet speaketh’ Excavations at three post-medieval burial grounds in Tower Hamlets, east London, 2004-10 Museum of London Archaeology Monograph 64

71

Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire Christopher Cumberpatch Introduction

by Staffordshire potters in both innovation and production throughout the early modern and recent periods.

In previous papers I have discussed aspects of the medieval and post-medieval pottery industries in Yorkshire and neighbouring areas (Cumberpatch 1997, 2003). In this paper I shall extend this discussion to consider some aspects of the Yorkshire pottery industry in the late 17th and 18th centuries; in archaeological terms the late postmedieval and early modern periods. Such a general review is needed because of the increasing amount of archaeological and historical work being undertaken on sites of this date and because of the increasing levels of interest in historical archaeology both within the profession and amongst the growing numbers of actively engaged amateurs and volunteers.

The geographical focus of the paper is on southern Yorkshire (the modern counties of South and West Yorkshire) where the data are abundant and where research has been most closely focussed but complementary evidence will be cited where appropriate. The focus of the first part of the paper will be on the potteries themselves but in the latter part attention will be given to consumer sites although many of these, particularly those located in the major industrial cities, pose significant methodological problems, some of which have been briefly considered elsewhere (Cumberpatch 2005a). Typology and chronology

The archaeology and social history of the 17th and 18th pottery industry and its products are characterised by a high degree of inter-disciplinary fragmentation with contributions from collectors, historians, archaeologists and archaeological scientists appearing in semi-isolation from each other. Furthermore there is the issue of the enduring focus on Staffordshire as the heart of the British pottery industry, particularly within archaeology. Even while numerous excavations and historical accounts have made clear the extent to which pottery manufacture was dispersed across Britain, it is still common to read of, for example, ‘Staffordshire Slipwares’. Clearly no one can deny the importance and influence of the Staffordshire pottery industry in the 18th and 19th centuries but an exclusive focus on the pottery industry of that county distorts the bigger picture and clouds understanding of the geographical, chronological and sociological complexity of the industry, including the role of potters who moved from Staffordshire to other areas of the country taking practices and techniques with them.

Before looking at the archaeological and historical data pertaining to the production and consumption of pottery there are a number of terms which require definition. The principal ones relate to the three classes of pottery which can be identified in 18th century assemblages; vernacular tableware, formal tableware and utilitarian ware. The term vernacular tableware covers the wares which emerged out of the post-medieval tradition (Brears 1971, Cumberpatch 2003) and were widely manufactured throughout the 18th century. They include Slipware, Late Blackware, Slip-Coated ware, Mottled ware and early (18th century) Brown Salt Glazed Stoneware. As the term vernacular tableware implies, the vessels involved are generally small in size and seem to have been designed for use at the table rather than as cooking or storage vessels (considered below as utilitarian ware). Vernacular tablewares were manufactured by potters working in ‘country potteries’ using primarily local resources and seem to have been distributed though local markets or by itinerant pedlars and packmen (Spavold and Brown 2005; Chapter 7, Griffin 2012: xiii-xv). While some of these wares, notably the Slipwares, were highly decorated the range of colours was generally limited to yellow, black, red-brown, dark brown and honey-coloured; a colour palette referred to as ‘earth-toned’ by Yentsch (1991) whose approach while interesting, is marred by her failure to distinguish the medieval potting tradition from the post-medieval tradition and by the rigidity of her High Structuralist interpretative framework.

In view of this there seems to be a place for broad regional overviews of the subject which start from a specifically archaeological perspective but which can provide a broader framework within which the material in its considerable diversity can be considered (cf. Barker 2004). This review, although regional in focus is intended to illustrate the wider point that the pottery industry was a national one which made a distinctive contribution to both the economic and the physical landscape as well as to what might be termed the domestic landscape through its creation of consumer goods. Such goods played important practical and symbolic roles within households at a time of significant social and economic change. It is probable that this case study could be paralleled in other areas of Britain without detracting from the important role played

The formal tableware category includes White Salt Glazed Stoneware, Creamware, and Pearlware with Whiteware belonging to the mid to late 19th century and as such outside the scope of this paper. A great deal has been written on the history of the formal tablewares and their relationship to 73

Christopher Cumberpatch rather than c.1485 as was hitherto thought to be the case (Moorhouse 1983, Cumberpatch 2003).

Tin Glazed Earthenware and imported Chinese porcelain. Such accounts have tended to focus on the details of their production and the social history of their use as part of the ‘civilising process’ of the 18th century (including Elias 1994, Yentsch 1991, Kowaleski-Wallace 1997, Richards 1999) which is characterised in archaeology as the rise of the ‘Georgian order’ (Johnson 1996). A significant body of literature documents the history and products of the factories involved in social historical and typological terms. For the study area defined above the most important are Lawrence (1974), Cox and Cox (2001), Cox 2013, Griffin (2001, 2005, 2012) and Edwards and Hampson (2005) although the field as whole is far too large to be reviewed here.

The date of c.1720 for the beginning of the early modern period is based on the development of the White Salt Glazed Stoneware body, the first truly practical alternative to Chinese porcelain suitable for indigenous mass production (Edwards and Hampson 2005). The subsequent development of lead-glazed refined earthenwares (Creamware and Pearlware), the invention of the techniques of transfer printing and the development of practical hard and soft paste porcelain bodies have all served to establish the 18th century as critical in the history of pottery manufacture in Britain (Richards 1999, Barker 2004, Godden 2004). The importance of these wares as elements within the social and economic history of Britain and their attractiveness to collectors has enshrined their pre-eminent role in a range of narratives including those of the industrial revolution and the revolution in manners and behaviour which characterise the 18th century. It is not my intention to provide a critique of these narratives (which have stimulated the creation of a body of high quality work) but rather to set them in the wider context of early modern pottery production and consumption and to draw out the implications of this contextualisation for archaeology.

Utilitarian ware includes Brown Glazed Coarseware, Brown Glazed Fineware, Yellow Glazed Coarseware and later (19th century) Stonewares. Although widely regarded with some distain by ceramicists from all traditions, the prominent place of such vessels (pancheons, bowls, storage jars and cisterns) in virtually all archaeological assemblages from the 17th century onwards means that they represent a significant part of social and economic practice over a long period of time. This gives them an inherent archaeological significance which should not be dismissed simply because of their ubiquity and mundane role within the household. It should also be noted that the continuation of production into the mid 20th century (Anderson 1964) means that they provide a tangible link with an industry that stretches back to the 17th century and even slightly earlier. The deployment during the late 1980s and 1990s of Brown and Yellow Glazed pancheons in pubs and restaurants as part of efforts to create a fauxrural and specifically faux-Irish ambience (perhaps better termed ‘Oirish’ to denote the essentially sentimental and ahistorical nature of the practice) has yet to be tackled in detail by critics of contemporary interior design and of the corporate creation of fashionable public ambience.

Economic structures and practice Two apparently distinct forms of economic organisation can be distinguished within the 18th and 19th century pottery industry. Following earlier usage, these may be called country potteries and factories. The broad distinction is one that is familiar in archaeological narratives and similar distinctions have been employed by a number of writers concerned with the organisation of manufacturing crafts and industries in complex agrarian societies (e.g. Peacock 1982, van der Leeuw 1976, 1984, Cumberpatch 1991, Andrews 1997). In the context of the post-medieval and early modern periods in Britain, country potteries have been defined as small craft potteries in which the local clays were transformed into useful earthenware vessels for the use of surrounding communities (Brears 1971: 9)

Turning to chronology, a tripartite periodisation will be employed in the paper in an effort to break away from the crude and fundamentally misleading bipartite medieval : post-medieval distinction which offers nothing of value to an understanding of a long and complex time period. The chronology used can be summarised thus:

Brears’ seminal discussion begins in the mid-16th century and traces the history of the country potteries throughout the country until their decline in the earlier part of the 19th century. Brears’ definition focuses on the use of local raw materials and the importance of local markets but we can add other criteria to these to characterise a particular form of early industrial organisation which places pottery manufacture alongside other forms of craft and industrial production which contributed to the complexity of the ‘long’ industrial revolution. The sparse survival of such potteries into the mid 20th century as producers of utilitarian wares allowed the creation of a unique film record by John Anderson involving Isaac Button, one of the last of the ‘country potters’ working professionally (Anderson 1964).

Post-medieval: c.1450 – c.1720 Early modern: c.1720 – c.1840 Recent: c. 1840 – present day The early dating of the start of the post-medieval period (c.1450) reflects the inadequacy for archaeology of the conventional range of dates for the end of the medieval period which range from 1485 (Battle of Bosworth) to 1530 (the final stages of the Tudor ecclesiastical reforms). It reflects the recent establishment of the origins of Cistercian ware by Anne Irving (unpublished) at c.1450

74

Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire In earlier publications (e.g. Cumberpatch 2010a) I have described the country potteries as family-run enterprises which existed as part of complex structures of economic organisation, sometimes with the aid of investment from outside the family or kin group. This model sees the country potters combining pottery manufacture with farming or small-holding, thereby gaining greater security of income and spreading economic risk across a range of unrelated activities. This was not an uncommon form of pre- or proto-industrial organisation and has been proposed for other industries within the study areas, as will be discussed below. It might be described as a distinctive ‘mode of production’, although this usage differs significantly from the traditional Marxist usage in that it does not elevate the principle to a dominant one in the structuring of society or the social formation; indeed its application to early modern society reveals the extent to which distinct modes of production may co-exist within an economically liberal society given the existence of accommodative social conditions and circumstances.

that were later dominated by wealthy ‘merchant princes’ (cf. Hunt 2004), the earlier stages of the development of industrial production were organised largely through existing structures of production by individuals who were part of the artisan communities and not, as Smail puts it solely by ‘far-sighted, risk-taking entrepreneurs’ (Smail 1992). As noted above I have, in earlier papers, suggested that the country potteries were part of a dual economy similar to that identified in the edge tool industries. In the course of this paper I shall suggest that this was not in fact the case and that while such a system may have operated in some parts of the UK and even in parts of Yorkshire and neighbouring areas, it does not seem to have been universal or even common. In spite of the superficial similarities it no longer seems possible to apply a model drawn from the edged tool trades to pottery manufacture (contra Cumberpatch 2010a). The textile industry may be a better parallel although analogy is perhaps not the best way to proceed and to understand the pottery industry it is necessary to look at actual cases in some detail and to review the data critically.

David Hey has documented the existence of a ‘dual economy’ in the nail-making and edged tool trades of Hallamshire with farmers and small-holders working in the metal trades during the agricultural off-season (1969, 1972, cf. Thirsk 1984:209). Battye has provided further documentation of this form of economic organisation in Eckington and Norton near Sheffield based on an analysis of probate inventories. She describes the situation that existed in Norton from the mid 16th to mid 18th centuries thus;

Table 1 lists the country potteries identified to date within the study area. It has been compiled from published sources with additional details based on the author’s own observations. In some cases, particularly where unpublished excavations are concerned, details are relatively sparse but after the excavation of several sites in the study area the broader picture is becoming clear although there is still a need for further work, particularly on material held in museum archives.

The 61 tool makers of Norton may have been cutlers, scythe smiths, scythe grinders or sickle makers, but for many of these men the making of those implements was not their sole means of making a living. The wills and inventories examined and analysed indicate that in 45% of cases there is evidence of a dual occupation (Battye 2003:65)

The pottery factories which produced the formal tablewares seem, from a modern perspective, to be more conventionally ‘industrial’ in character. They were organised around a principle of wage labour with capital supplied, in large part, by external investors or sleeping partners in the enterprise. In practical terms the factories were able to deploy capital to import raw materials for the manufacture of fine stonewares and refined earthenwares and to develop the more complex means required for their production. This was the context within which the ‘heroic entrepreneurs’ (Wedgwood, Spode, Minton and their contemporaries) of the 18th and 19th century pottery industry operated and within which the classic cream, white and transfer printed tablewares typical of the ‘Georgian Order’ were developed and marketed throughout the world (Barker 2004). This is also the aspect of the 18th and 19th century pottery industry that has come to dominate historical and antiquarian accounts of the period and, to some extent, the study of the ceramics themselves.

Battye suggests that there was no significant variation in the numbers of families for whom the dual economy was a way of life over her study period. This was not restricted to the makers of edged tools; Dual occupation was a way of life throughout the period of study from 1534 – 1750, not just for men involved in making scythes, sickles, shears and other metal goods for which Eckington became known, but also for the many other craftsmen whose names and occupations appear in the will and inventory lists (1999:321) A similar structure is alluded to by Riden in his account of Chesterfield between the end of the 15th and the mid-17th century (1984:152 – 3).

Potters and potteries: The organisation of production The seventeenth century

In West Yorkshire John Smail has examined the emergence of ‘artisan entrepreneurs’ in the textile trade and has concluded that, even in the West Yorkshire textile towns

The range of pottery vessels in use during the 17th century in Yorkshire is, thanks to excavations of Civil War period 75

Christopher Cumberpatch Site

Date range

Products and summary details

Reference

Potterton, Barwick-inElmet, Leeds

C16th

Cistercian ware and utilitarian ware

Mayes and Pirie 1966

Coxwold

C16th / C17th

C16th wasters and possible kiln furniture in Yorkshire Museum; documentary reference to a C17th potter

Wilson 1988; Appendix 10

Daisy Farm, Lastingham

C16th – C17th

Wasters resembling material from Stearsby; documentary reference to a potter in 1592

Wilson 1988; Appendix 11

Stearsby

LC16th- EC17th

Ryedale ware; jugs, jars, cisterns in the late medieval/ early post-medieval tradition

Wilson 1988:122-130

Wrenthorpe (Potovens), Wakfield

MC16th – EC18th

Cistercian ware, Blackware, Early Yellow ware, Yellow ware

Brears 1967, Bartlett 1971, Moorhouse and Roberts 1992

Causeway Pottery, Cliviger Valley, Lancs

C17th

Cups, mugs, tygs, jars

Fletcher pers. comm., Ellis nd, Burnley Express and News, July 1955

Thornhill Hall, Dewsbury

?16th to 17th century

Unknown; glazed stones suggest the presence of a kiln

Cumberpatch 2000

Pule Hill, Northowram

Mid-17th century

Slipware, Yellow ware, Blackware type

Cumberpatch 2000, Lawrence 1974, Bryant, unpublished

Bate Ain, Northowram

Late 17th & 18th century

Blackware type, Yellow ware, Mottled type ware, Slipware

Cumberpatch 2000, Lawrence 1974

East Gilling

LC17th

Documentary reference 91686); pottery, wasters and burnt material

Wilson 1988: Appendix 10

Yearsley

C17th – C18th

Documentary reference and oral tradition; a branch of the Wedgewood family founded a pottery; location unknown

Wilson 1988: Appendix 10

Holbeck, Leeds

EC18th

Unknown

Lawrence 1974; 62-3

Sheffield Manor, Sheffield

c.1709 – 1738 & later

Mottled ware

Beswick, pers. comm., 1978, Cumberpatch 2010, 2011, 2012, Hadley and Harlan 2011

Midhope / Midhopestones, Sheffield

1720 – c. 1845

Slipware, Brown Glazed Coarsewares, Redware

Lawrence 1974, Ashurst 1987, 2002, Kenworthy 1928

Swinton, Rotherham

1745 - 68

Slipwares; flatware, hollow ware

Cox and Cox 2001

Brown Moor pottery, Lazencroft, Leeds

1739 – c.1777

Slipwares and others

Lonton 2008, Griffin 2012

Barwick-in-Elmet, Leeds

c. 1747 - 1788

Slipware, Slip Coated ware, Mottled ware

Cumberpatch 2007, JB Archaeological Services/BEHS 2007

Silkstone, Barnsley

c.1754 - 1802

Slipwares, Slip Coated ware, Manganese Mottled wares, Brown Glazed Coarsewares

Brears 1971, Lawrence 1974, Cumberpatch 2004, White 2007, 2012

Soil Hill Pottery, Halifax

c.1760 - 1965

Mainly utilitarian ware but also some novelty wares; knife boxes, candle holders, puzzle jugs etc; later exclusively utilitarian wares

Anderson 1964

Howcans, Nortowram

LC18th

Founded by George and Abraham Halliday after the closure of the Bate Hayne pottery; possible slipware manufacture

Lawrence 1974; 185

Bolsterstone (Bate Green), Pot House Farm

c. 1778 – 1796

Slipware, Black Glazed Coarsewares

Lawrene 1974, Ashurst 1987, 2002, Griffin 2012

Scales Pottery, Ilkley

MC18th

Wasters recovered from site but no evidence of a kiln or workshop; Slipware, Late Blackware, Slip Coated ware, BGCW,

Young and Vince 2010

Warren Vale, Rawmarsh

LC18th – EC19th

Brown Salt Glazed stoneware ink and other moulded bottles

Milefanti and Brears 1971

Table 1. The principal late 17th and 18th century country potteries in South and West Yorkshire and neighbouring areas discussed in the text.

76

Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire sites such as Pontefract Castle (Cumberpatch 2002) and Sandal Castle (Brears 1983), relatively well known (Cumberpatch 2003). In contrast we know of relatively few potteries dating to this period, at least when compared to the medieval and early modern periods. There seems to be some indication that the pottery industry became centralised during the 16th and 17th centuries in the form of ‘potting villages’ such as Wrenthorpe near Wakefield and Ticknall in South Derbyshire (Brears 1967, Bartlett 1971, Woodrow 1971, Moorhouse and Roberts 1992, Spavold and Brown 2005, Boyle 2002-3, Irving unpublished). How far these sites reflect the wider situation is unclear and the effects of differential survival and a lack of research are both issues that must be taken into account in this context.

appears to be of 18th century type (predominantly Late Blackware and Slip Coated ware but also Brown Salt Glazed Stoneware), presumably reflecting the later use of the site rather than its early years. Neither site has been fully published in spite of their potential to inform our understanding of the 17th century pottery industry. More dispersed sites of this date have been identified in the uplands of North Yorkshire (Wilson 1988) but in most cases little is known of their history as little excavation has taken place. The notes presented by Wilson (1988: Appendix 10) suggest that 17th century potteries existed at East Gilling, Yearsley, Coxwold, Daisy Farm, Lastingham and probably Stearsby and that they were the successors to a number of potteries dating to the medieval and early post-medieval periods. Although work is in progress on an assemblage from Yearsley (Young, pers comm.), further work on this important group of sites is highly desirable given the poor state of our knowledge of medieval and later pottery in North Yorkshire.

Pottery manufacture in both Wrenthorpe and Ticknall survived into the 18th century but appears to have suffered a progressive decline through the century with that of Wrenthorpe considerably earlier and more rapid than Ticknall (Brears 1967:1-7), although the reasons for this are unclear.

In Lancashire there is some evidence of two potteries in the Cliviger Valley (Fletcher pers. comm.) perhaps located on marginal land capable of supporting only extensive agriculture. One of these, the Causeway Pottery seems to have originated in the late 17th century although the date is based upon the survival of a single dated cup (1671) and it is entirely possible that the pottery has earlier origins. The second pottery dates to the later 18th century. As in the case of North Yorkshire, further work on these sites would be highly desirable.

The products of these potteries during the 17th century were primarily Blackwares, Yellow wares and larger Coarse Blackware vessels which overlap in large part with the heterogeneous Midlands Purple ware group. Redwares, both plain and slip-trailed, appear during the mid 17th century and persist into the early 18th century (Brears 1967:22-27) but as yet there is no evidence within the study area for the production of press-moulded slipwares with layered and feathered slip decoration prior to the 18th century even though production was underway in Staffordshire before the mid 17th century (Barker 1993).

In summary, the 17th century saw pottery production in both nucleated potting villages and also at more dispersed sites, some at least of which lay on ground less suitable for agriculture and which would have been ideally located so as to form part of a ‘dual economy’ as described by Hey and Battye. The range of vessel types was relatively limited in comparison to the developments that were to characterise the 18th century and black or dark brown, yellow and orange vessels were the norm. We need to know more about the production of the larger utilitarian wares in the 17th century and in particular more about the later Midlands Purple wares, the development of Brown Glazed Coarseware and its relationship to the early postmedieval Late Humberware or Green Glazed Sandy ware. Even without this information however, it is clear that the use of multi-flued updraft kilns was standard and that saggars were in regular use. As discussed elsewhere (Cumberpatch 2003), this technology had changed little since the late medieval period and similar kilns were to remain in use in the country potteries throughout the 18th century.

In addition to Wrenthorpe, two sites in West Yorkshire are of particular interest; Pule (or Puel) Hill and Bate Hayne (or Ain) near Northowram and Halifax. The pottery at Pule Hill was established by the Halliday family, reputedly gypsies originally from Scotland, most probably during the 1640s (Brears 1971:140-4, 223, Lawrence 1974:183185, Griffin 2012:295). The site was not wholly suitable for a pottery and in the late 17th century the Hallidays moved to Bate Hayne, a more sheltered and lower-lying site (Lawrence 1974:184-5) where a new pottery was built. The family remained involved in the pottery industry and in the late 18th century moved again, this time to Howcans where they continued to make pottery and, later, firebricks and chimney pots until 1916. In spite of recent research by John Griffin (2012:295) no further documentary evidence pertaining to the sites has come to light. Excavations at the Pule Hill site by G.F. Bryant in 1966 revealed the well-preserved base of a multi-flue kiln and part of the surrounding structure (Bryant 1967, pers comm., Brears 1971;140). Pottery associated with the kiln is currently held in the Tolson Museum, Huddersfield and consists of Blackware and Yellow ware along with saggars, kiln furniture and debris and is of typical 17th century type (Cumberpatch 2000). A considerably larger quantity of material survives from an excavation at Bate Hayne and

The eighteenth century In contrast to the relatively sparse information pertaining to the 17th century pottery industry, the 18th century offers an embarrassment of riches. In this section brief accounts of individual country potteries will be used to illustrate 77

Christopher Cumberpatch specific aspects of their organisation, their relationship to the factories and the range of wares produced.

vessels. A number of distinct fabric sub-groups within the Mottled ware category were identified amongst the Manor assemblages and were given the preliminary names of Mottled wares 1 to 4 (Cumberpatch 2010b, 2011a, 2012a).

Sheffield Manor The earliest of the 18th century country potteries so far identified in South Yorkshire is that established on the site of Sheffield Manor in 1709 (Hadley and Harlan 2011: 6-9). A circular multi-flue kiln, excavated in 1971, was built into the ruins of the earlier Long Gallery very shortly after its partial demolition (Hadley and Harlan 2011: Figures 15, 16 and 17). The excavation remains unpublished but short accounts of the kiln and its products have appeared (Beswick 1978, Crossley 1990: 254) while the results of the excavations have recently been the subject of summary reports (Hadley and Harlan 2011, Crewe 2012). Recent excavations (2009 – 2011) involved the re-excavation of the earlier trenches and the recovery of a considerable additional assemblage of pottery, including production waste (Cumberpatch 2010b, 2011a, 2012a).

Mottled ware 1 is a hard, fine textured light buff coloured fabric with few inclusions visible to the naked eye. At a low magnification (X10) it has a fine granular texture and contains sparse fine black inclusions and, in some cases, fine voids with black margins. There was no sign of the fine white inclusions noted in the case of Mottled wares from Silkstone (described below) which superficially resemble the Manor products (Cumberpatch 2004a:4). Nor were there any signs of the streaks of red clay seen in examples of Mottled ware from Bolsterstone and Midhope, as reported by Hurley (2009:67). Mottled ware 2 has a much darker and more opaque brown glaze and the fabric is much lighter, almost white in colour. Low-power magnification shows only sparse inclusions but these include possible fine grains of muscovite, fine non-crystalline pale orange-red grains and slightly larger non-crystalline whitish grains. White-firing Coal Measures clays were exploited from the later medieval period onwards in South Yorkshire (Cumberpatch 2004b) and it is probable that such deposits were the source of the clays used at the Manor pottery. The medieval Coal Measures wares are distinguished by their coarse texture and while the evidence of the other 18th century country potteries demonstrates that potters had few problems obtaining fine clays, it is probable that levigation would have been required in order to obtain acceptably fine clays. Levigation tanks are typically located close to the site of the pottery and a convenient water source but, to the best of the author’s knowledge, none have yet been identified on the Manor site.

Documentary evidence is sparse but the name of John Fox is closely associated with the pottery, not least because his initials appear on the base of a Dutch oven now in Weston Park Museum, Sheffield (Hadley and Harlan 2011:Figures 11 and 12). The Fox family were involved with at least one other pottery, that at Midhope / Midhopestones to the north of Sheffield which was started in the 1720s by Fox’s mother, Mary Blackburn and her second husband (Harlan 2010) but the family, originally from Fulwood Hall in Sheffield, is best known for its connection with the Bolsterstone glasshouse from its establishment in the mid 17th century until its closure in 1758 (Ashurst nd, 25). John Fox seems to have combined running the Manor pottery and the Bolsterstone glasshouse for a number of years (Ashurst nd.:27) which implies that his interests were much broader than those of a simple potter/farmer and that he had an understanding of pyrotechnologies that went beyond a single craft. Further work on the documentary sources would be valuable in determining the nature of Fox’s funding for the Manor pottery and, in particular, the role of venture capital in its establishment. Details of the workforce are perhaps unlikely to have survived but one must wonder how far Fox was actually involved in the daily activity of the pottery given his role at the glasshouse.

Mottled ware 3 was characterised by its fine orange to red fabric, usually combined with a white slip coating internally and externally, presumably to ensure the appearance after firing of the desired mottled finish. The fabric varied in character between a very hard, smooth dense texture and a rather more granular texture more reminiscent of Mottled ware 1. The considerable difference between this and other fabrics from the site probably indicates the exploitation of a different clay source.

The waste from the Manor pottery includes a large quantity of wasters and saggar fragments which indicates that the products of the pottery included mugs, tankards, porringers, posset pots, chamber pots, dishes and wheelthrown plates in addition to more unusual items such as the Dutch oven. The Manor pottery is unusual amongst known potteries of this date in that production appears to have concentrated on a single ware type, Mottled ware. The name is derived from the mottled appearance of the lead glaze which is coloured with fine grains of manganese or iron to give a distinctive speckled appearance. The density of the colourant can vary considerably between batches of glaze and this explains the high degree of variation between individual vessels and groups of

Mottled ware 4 was close in character to Mottled ware 1 with a finely granular texture, fine red and cream-coloured inclusions and prominent fine linear voids. Like Mottled ware 3, type 4 was typically white slipped although in this case the necessity was not obvious as there seemed no reason why the light-firing fabric should have interfered with the appearance of the glaze. Future work on the assemblages from the site should include a programme of petrographic and chemical analysis designed to characterise the fabrics and differentiate them from other sources of Mottled ware (cf. White 2012). This will build on the work of Markus (nd. 25-6) which involved the

78

Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire examination of a small number of petrographic samples and a description of the fabrics.

the oven and associated buildings on the site. Gough may also have been responsible for the construction of three houses close to the pottery and was one of four individuals later responsible for the building of the first school in the Midhope township. In June 1762 Gough and his daughters assigned the lease to John Taylor but apparently as a result of debt, Taylor’s lease was transferred to John Whiteley of Woodseats in July 1765. The name of William Gough will occur again in connection with the Brown Moor pottery near Leeds (below).

The shape and size of the saggars clearly reflected the range of vessels manufactured in the pottery with taller examples (up to 175mm) for hollow wares and the flatter examples for individual plates. The latter included examples with overfired plates still in situ. The presence of fragments of stone coated in glaze suggested that the kiln was refurbished on a number of occasions and its internal fittings replaced with new elements. Given the duration of activity on the site (c.1709 – c.1738) this is to be expected. It is unclear from the available evidence whether the surviving parts of the kiln included evidence of rebuilding but with a lifespan of some thirty years, it is unlikely that such evidence would be absent.

Kenworthy raises the question of whether the John Taylor associated with the Midhope Pottery was the same individual as the John Taylor of Silkstone (mentioned below) but notes that if this were so, then Taylor would have been only nineteen years old when he took over the Midhope Pottery “which renders such identity somewhat doubtful from a legal point of view” (Kenworthy 1928:57).

Midhope / Midhopestones

In 1774 the Midhope Pottery was assessed as worth four shillings and sixpence per year in the Waldershelf Land Tax Bill. John Whiteley’s widow sold the lease to Edward Appleyard of Sheffield Park in 1793 and Joseph Lindley (also involved with the establishment of the pottery at Bate Green, Bolsterstone described below), seems to have occupied and operated the site from the late 1780s until his death in 1801.

The extant pottery assemblage from Midhope was collected in the early 20th century by the antiquarian Joseph Kenworthy from the site of the pottery or potteries prior to the inundation of the area by the Midhope reservoir. The assemblage is now held in Weston Park Museum, Sheffield. Kenworthy published an account of the Midhope Pottery (1928) and it is from this volume that the following account has been summarised (see also Griffin 2012:146-147). Further work on the Midhope and Midhopestones potteries is desirable as it seems that there were at least two in relatively close proximity and is currently being undertaken by members of the Bolsterstone Archaeology Group. It is hoped that this will clarify some of the outstanding issues surrounding these potteries.

In 1788 the Hunshelf Overseers of the Poor paid Joshua Lindley the sum of two guineas to take the son of Widow Horsfield as an apprentice pot-maker. The close relationship between pottery and glass making and particularly between Bolsterstone and Midhope becomes clear at this point. Kenworthy writes … Richard Ellis of Midhope Hall, who lived 1717 – 1809 and a person named Smith, manufactured glass at Bate Green [Bolsterstone], but they made nothing of it and the Glass House was finally converted into a Pot House by Richard Ellis in 1778, he having bought it from the Trustees of William Parkin in September, 1777 …

According to Kenworthy the first evidence of a pottery in the Midhope area dates to 1720 when William Bosville leased an area of land known as Nether Mill Green to George Walker of Hunshelf and Robert Blackburn of Langsett for the purpose of erecting a pot house. The lease was for ninety-nine years at a rent of five shillings per annum. Both Blackburn and Walker are described by Kenworthy as ‘gentlemen’ and it seems from his text (which is written in a self-consciously mannered and ‘archaic’ style) that these were the terms employed in the original lease. Blackburn was the second husband of John Fox’s mother, Mary which provides a family link between the Manor and Midhope potteries. The involvement of a limited number of families in the Yorkshire pottery industry is something that will emerge as a theme in the remainder of this paper.

It was then taken over … by Joshua Lindley, who came from Midhope, and it is possible that this person was employed at Midhope Pottery, and lived on Bate Green [Bolsterstone] when Richard Ellis decided to turn the Glass House into a Pot House, probably at Lindley’s suggestion (1928:60). After Lindley’s death the vacant pottery was advertised on two occasions (1802 and 1803; Griffin 2012;146) and in 1803 it was leased to Thomas Fawley. By 1828 it was occupied by Matthew Thickett an ‘earthenware manufacturer’. The site seems to have been sold in 1828 to John Haigh when a potter named William Stead was in occupation and by 1835 it was occupied by James Barraclough. Pottery production continued until sometime between 1835 and 1845 when the site, described as ‘formerly used as a pottery’ was sold to Luke Moorhouse.

Blackburn also had an interest in the glass house at Bolsterstone with which Fox was involved (Ashurst 1987:149 – 153) and this is one of several links between 17th and 18th century glass manufacture and pottery manufacture. On Blackburn’s death in 1727, his wife sublet the lease on the Midhope Pottery to William Gough, a potter who had been employed by Blackburn to build 79

Christopher Cumberpatch It is clear from even a brief examination of the assemblage held by the Sheffield Museums Trust that it represents only a selection of the material originally in Kenworthy’s possession. The range of vessel types in the archive is extremely limited with press-moulded, slip-decorated dishes predominating. Both plain and pie-crust rims are present and the decoration seems to be somewhat more elaborate than that from Silkstone but it should be noted that the assemblage cannot be taken as representative of the output of the potteries, given the circumstances of its recovery. Mottled wares are notable by their absence and there was only one wheel-thrown plate although the assemblage also included a shallow saggar. The large size and good state of preservation of the individual sherds and vessels would seem to suggest that they were recovered from a waster dump rather than a working area. Hurley’s analysis of the Midhope assemblage (2009:62-5) suggested that four fabric types were represented in the Kenworthy collection but that one of these was a minor type associated only with a Slipware cup.

(1928: 58, Fig. No. 29). The second was a cylindrical bottle in a similar fabric with the inscription ‘RB 1779’ (British Museum cat. no. D110). The latter was originally made for Richard Bradley and was bought from his granddaughter in 1898. Excavations undertaken in 2002 by English Heritage on the site of the Silkstone glasshouse and the later pottery demonstrated the survival of strata relating to glassworking in the latter half of the 17th century and pottery manufacture in the early part of the 18th century (Cumberpatch 2004a, Dungworth and Cromwell 2006, White 2007). The dating evidence derived from the historical documents appears to be contradicted by the evidence of clay tobacco pipes from the excavation which suggest that pottery manufacture predates the documentary references by some twenty to fifty years; this conundrum remains to be resolved. Context 0009 which produced a large quantity of pottery production waste was also associated with good quality clay tobacco pipes dating to between c.1700 – 1730 (Higgins 2006:167-169). This would make the Silkstone pottery contemporary with the Sheffield Manor pottery and also contemporary with the early blast furnace at nearby Rockley Smithies. Further work on the site, designed to resolve this problem and to locate the kiln and workshops would be highly desirable but at present it seems unlikely that this will be possible.

Silkstone According to Lawrence (1974) and Griffin (2012;146), the first mention of a pottery at Silkstone occurs in deeds dated 1754 when potovens, a house and a cornmill are recorded as being owned by James Scott and occupied by John Bailey, Ralph Taylor, Joseph Goldthorpe and Michael Taylor. The inventory of John Scott, dating to 1746, makes no mention of the pothouse, which tends to suggest that Lawrence’s earliest date (1754) lies close to the establishment of the pottery. The estate is mentioned in 1775 when it was bought by Richard Fenton and the same tenants, minus Michael Taylor, are again referred to. Land Tax assessments refer to John Taylor, a potter, in 1767 and he is mentioned as a tenant from 1781 to 1812. The 1802 militia list mentions a second William Taylor and Edward Taylor (ages twenty-five and thirty-four) as potters. William Taylor, described as a potter in the 1780s, ran the works from 1812 until 1821. A review of the documentary evidence pertaining to the site by the late Dr. Denis Ashurst revealed a number of conveyances and other documents which refer to the pottery and to potters during the period between 1754 and 1775 and mention specifically potovens in addition to the glasshouse which preceded the pottery on the site (Cumberpatch 2004a; 15, Dungworth and Cromwell 2006:162).

The evidence from the excavation indicates that the Silkstone potters were producing a wide range of vernacular tablewares including Slipware (with both trailed and combed or feathered decoration), Slip Coated ware, Late Blackware and Mottled ware (Cumberpatch 2004a: Figs. 2 – 11, Dungworth and Cromwell 2006: Figs. 18 & 19) with smaller quantities of an unusual white bodied ware bearing either clear or black glaze. Eight distinct fabrics were identified at X10 magnification (Cumberpatch 2004a; 3-5, 2006a:Table 8) and there were clear indications that specific fabrics were associated with specific vessel types (Cumberpatch 2004a:3-4, White 2007: Table 2). Subsequent petrographic and chemical analysis confirmed this association with White concluding that the potters manufacturing the different types of slip and glaze decorated wares at Silkstone made use of a number of clays and clay mixtures, and these appear to be related to specific manufacturing methods (White 2007:19).

Peter Brears (1971) has reproduced an illustration of the pottery, dated to 1806, taken from J.C. Nattes Views from Nature and currently held by the Cannon Hall Museum and Art Gallery, Barnsley. The clay preparation area, the workshop and a tall multi-flued kiln with its surrounding hovel can all be clearly seen in the picture (Cumberpatch 2004a: Fig 24, Dungworth and Cromwell 2006: Fig. 2).

White identified two fabrics as particularly associated with wheel-thrown forms and a different set associated with the press-moulded slipwares. One of the latter had clearly been prepared by mixing two different types of clay, one of which was employed in the manufacture of the wheel-thrown wares. A similar mixture had been used for the saggars and kiln furniture. There was also a correlation between the fabric types and the techniques used to obtain the mottled finish. Vessels in White’s Fine Quartz and Chert fabric group were associated with glazes coloured with significant amounts of manganese

Lawrence identified only two extant pieces from the pottery. The first was a large dark red earthenware flagon with dark brown glaze and slip trailed inscription ‘PE 1777’ which is also described and illustrated by Kenworthy 80

Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire oxide. In contrast, glazes on sherds in the Well-Sorted Quartz fabric group were coloured with higher iron oxide and lower manganese oxide concentrations. Both types of glaze produced similar effects and the choice of which to use was presumably made by the potter or potters although the rationale behind these decisions is unclear. White’s plausible interpretation of these results is that the variations in fabric type and glaze composition may relate to different potters employing their own preferred techniques and recipes to obtain specific and similar outcomes. She refers back to the historical information and suggests that they may be related to the changes in the ownership of the site and the potters who worked there. This use of scientific analysis to identify changes in working practices which might be related to individual potters (even if it is not possible to associate a particular practice with a specific named individual) is one that could with advantage be applied to other country potteries within the study area, following White’s example (cf. Allday and Millard 2009:605). Subsequent comparison of two of the Silkstone fabrics with samples from Stoke-on-Trent and Bristol have demonstrated that it is also possible to use petrographic analysis to distinguish the products of different country potteries in the manner familiar from numerous studies of prehistoric and medieval pottery (White 2012).

Documentary evidence, summarised by Ashurst (1987: 204-6) links the site with the potter Joseph Lindley who took out a lease on the glasshouse site in 1778 and established a pottery there which remained active until about 1796. Lindley was also associated with the Midhope/Midhopestones potteries discussed above and Ashurst describes Bolsterstone as an ‘outlier’ of the Midhope group of potteries. Trailed and combed slipware was prominent amongst the Bolsterstone material and the vessels had a fine sandy textured buff to pale orange fabric with some streaking (Ashurst 1987: Figs 24 & 25). This was similar in general appearance to the Silkstone material, although no detailed analysis designed to assess the extent of the similarity of the fabrics has yet been carried out and any direct linkage would seem to be unlikely. The range of fabrics was limited in comparison to Silkstone (Hurley 2009:602) but further work on the assemblage would be highly desirable. Decoration included both bi-coloured and tricoloured trailed and feathered slip designs. The majority of sherds appeared to be from press-moulded dishes, often with pie-crust rims, with relatively few examples of slipdecorated hollow wares. Late Blackware of 18th century type was present in significant quantities (Ashurst 1987: Fig 23). Utilitarian wares included Brown Glazed and Yellow Glazed Coarse wares although whether these were manufactured on the site or had been brought in from elsewhere is unclear. The assemblage also included a number of saggars, easily distinguishable from the glass crucibles from the site by their distinctive coarse gritty fabric (Ashurst 1987: Fig 25; 7 – 9). Mottled wares were generally rare and appeared to be limited to larger utilitarian vessels rather than the mugs that were such a prominent feature of the Silkstone and Sheffield Manor assemblages. Elsewhere this type of pottery has been classified by the author as Coarse Mottled ware. Ashurst has noted that the waster dumps associated with the pottery were used as the foundations for a series of houses built in the early 20th century some 100m north of the site. This observation emphasises the necessity for development control archaeologists to consider a broad area around possible or actual potteries when seeking to mitigate the negative impacts of development on archaeological assets.

The range of vessel forms manufactured at Silkstone was difficult to determine with any accuracy as a result of the high degree of fragmentation in the assemblage and the very limited scale of the excavation. Types identified with certainty included press-moulded dishes, wheel-thrown plates, mugs, mugs/jugs, tall, narrow hollow ware vessels and thin-walled hollow ware vessels (Cumberpatch 2004a: Figures 2 and 3, 2006a: Figures 18 and 19). In most cases this range of vessels has counterparts amongst the other country potteries although the thin-walled vessels with their characteristic overhanging rims (Cumberpatch 2004a: Figure 3; 15) seem to be a rare form. Bolsterstone Excavations at Pot House Farm or Bate Green, Bolsterstone, near Stocksbridge on the northern outskirts of Sheffield have focussed primarily on the nationally important 17th and 18th century glasshouse (Ashurst 1987, 2002, nd.) and the later pottery remains little known outside archaeological circles (see, for example, Griffin 2012:148). The extant excavated pottery assemblages, currently held on behalf of the community by Sheffield Museums Trust, represent the accumulation of a number of different pieces of work undertaken by the late Dr Denis Ashurst and the South Yorkshire Archaeology Unit (1987, 2002). The pottery assemblages have yet to be fully published (see Ashurst 1987: 204 – 211, Figures 23 to 25 for a selection of the material recovered from the main phase of excavation) but it appears that the potters produced a wide range of different wares, consistent with the evidence from the other country potteries discussed here with the exception of Sheffield Manor.

The Swinton (Rockingham) Pottery The Swinton (later Rockingham) Pottery was founded in the mid 18th century by Joseph Flint who is recorded as having paid rent for a brickyard and a sum of 15s annually for the right to dig for clay on Swinton Common (Cox and Cox 2001:20). Details of Flint’s business activities are obscure but Alwyn and Angela Cox who have researched the site in detail (1983, 2001) believe him to have been a man of some enterprise who anticipated the potential offered by the exploitation of local coal and clay. In 1752-3 he appears to have expanded his existing business interests to encompass a ‘New Tileyard’ and in 1753 he bought coal for a ‘Pot house and Tyle yard’. In 1754-5 he seems to have sold both businesses as separate concerns, 81

Christopher Cumberpatch the pot house to Edward Butler and the tile yard to John Bette. Other records show that the pot house was in need of repair in 1756 suggesting that it had been in existence for some years and the later owners of the Rockingham pottery, the Bramelds, believed it to have been founded as early as 1745 (Cox and Cox 1983, 2001:18-26).

allowed him to exploit a range of local and regional business opportunities. Once again, however, we know nothing of the workforce. Letters accompanying some of the invoices submitted to the Wentworth estate show that Malpass maintained an active interest in the pottery business but this is far from proof that he was involved in the day-to-day manufacture of pots or that he had any particular skill as a potter.

There are no accurate records of the types of pots that Flint was producing and the Cox’s believe that it was limited to coarse earthenware. They suggest pancheons, jugs, porringers in Brown Glazed Coarseware but as these wares continued to form part of the output of the Rockingham pottery into the 19th century, examples cannot be identified with any accuracy.

The products of the Malpass period are known both from the evidence of invoices but also from finds of slipware with the embossed initials WM internally (recalling the initials of John Fox on the Dutch oven from the Manor pottery). The invoices refer to utilitarian wares; pitchers, porringers, pudding pots and butter pots for the kitchen and dairy. Sherds recovered from the site of the Swinton pottery are more informative than the invoices and include press-moulded dishes decorated with trailed and feathered slip decoration. Cox and Cox have noted the high quality of the slipwares produced during this period of the operation of the Swinton Pottery (2001:24-5, Plate 1) and have sought a connection with the Staffordshire industry. This exists in the form of the Malkin family who moved from Burslem in Staffordshire to South Yorkshire in the 1760s. Samuel Malkin is buried in Wath-upon-Dearne churchyard (1769) and the gravestone records that he was ‘late of Boslem’. The name of William Malkin, one of two sons whose names are followed by the profession ‘Potter’ in parish registers, occurs on an invoice dated 13th June 1775, indicating a continued involvement with the Swinton pottery (Cox and Cox 2001:26).

Edward Butler and his wife took over the Swinton Pothouse sometime before Martinmas (11th November) 1755. At this time Butler was about fifty-one years old and had previously held thirty-six acres of land in Swinton, although we do not know what he did with this land. The quality of products may have improved after the Butlers took over the pottery but it is not clear where the technical skills required for this improvement came from (Cox and Cox 2001:22). An invoice sent to the Marquis of Rockingham on 15th December 1757 shows that the Butlers were supplying the Wentworth household with utilitarian earthenwares and the Cox’s suggest that a case can be made for the manufacture of Brown Salt Glazed Stonewares at Swinton although this remains uncertain (2001:22). Edward Butler died in early 1763 and in his will described himself as a ‘pott Maker’ and implied that he and his wife were closely involved in the work of the pottery. Elizabeth Butler continued to run the pottery for a further two years until 1765 but we know nothing of the workforce.

In 1768 William Malpass went into partnership with William Fenney and this event marked the transformation of the pottery from a country pottery producing utilitarian ware and vernacular tableware to a factory producing formal tableware. William Fenney’s career prior to this event is of considerable interest. Born around 1700 into a family with strong links to the glass-making industry (Ashurst nd:27-8, Cox and Cox 1983:233), he married Mary Fox, a member of the family which ran the Bolsterstone glasshouse from the 1650s until its closure in 1758 prior to the redevelopment of the site as a pottery (discussed above) and which also counted amongst its members John Fox of the Manor pottery. At the time of his marriage his position at the glasshouse was that of works manager (Cox and Cox 2001:27).

In 1765 the pottery passed into the hands of a family friend and close associate of the Butlers, William Malpass. Although seemingly of humble origins, Malpass developed diverse business interests (Cox and Cox 1983:233). He rented several parcels of land in Swinton and Kilnhurst and had a house, stables and wharf on the banks of the Don Navigation, also in Kilnhurst. From 1755 he operated a limekiln close to the wharf, no doubt using the canal to distribute the lime. From around 1770 until his death in 1807 Malpass ran an inn at Kilnhurst and was also associated with the Tinsley to Doncaster turnpike scheme. He appears to have taken over John Bette’s tile yard in Swinton in 1761-2 and there is a long-standing tradition that he was involved in the foundation of the Kilnhurst Old Pottery in c.1746. John Griffin has reviewed this tradition and has found no definite evidence to support it, beyond a series of assertions in the secondary literature (Griffin 2012:73-4). Cox and Cox describe Malpass as ‘very much an entrepreneur; a man of enterprise, willing to take on almost anything that brought him extra income’. It would seem that in this case a country pottery was being run by someone with diverse local business interests and presumably either the capital or creditworthiness which

Fenney appears to have had a career marked by disagreements. He left the Bolsterstone glasshouse in 1738 after a dispute over plans for expansion with his mother-in-law, Mary Blackburn, to set up the Catcliffe glassworks. This enterprise seems to have proved too much for his financial resources and in 1759 the enterprise was taken over by the May family (Ashurst nd.:48-9). Fenney, however, was clearly not impoverished by this development and in spite of being a skilled glassmaker he joined John Platt (described by Cox and Cox as ‘a masonarchitect’) in 1765 in establishing the Rotherham Old Pottery. Once again the partnership did not last and after 82

Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire

Plate 1. The Soil Hill pottery in 2010 (photograph: author)

Plate 2. The remains of the kiln at the Soil Hill Pottery showing two of the flues and part of the interior of the kiln (photograph: author) 83

Christopher Cumberpatch buying himself out of the contract for the sum of £120 in 1767, Fenney joined William Malpass at Swinton in 1768. The Rotherham Old Pottery was already producing formal tablewares (Creamwares and tortoiseshell coloured wares) and was in every sense a factory which employed twentytwo men and forty boys. Cox and Cox suggest that some of the Swinton workers were recruited from Staffordshire while others may have been ‘poached’ from Rotherham. Fenney’s modernising aims at Bolsterstone may have alienated his mother-in-law but in the field of pottery manufacture he seems to have found a congenial partner in Malpass. Two new ceramic bodies were introduced; White Salt Glazed Stoneware and cream-coloured earthenware or Creamware. This radical change of direction, from a country pottery to a more technologically sophisticated factory, involved alterations to the facilities, the enlargement of the site and the regular procurement of raw materials from non-local sources. Fenney appears to have provided the money for these developments. This seems to be the only example of a country pottery that was redeveloped to become a factory, at least in southern Yorkshire and it would appear that this was largely the result of the entrepreneurial flair shown by both Fenney and Malpass. In view of this it is more than a little ironic that the eventual closure of the Rockingham Works (as the enterprise was known after 1826) in 1842 seems to have been due in large part to the lack of such flair on the part of the Bramelds and their financial dependency on the Fitzwilliam family as well as an unfavourable economic climate and the failure of overseas enterprises (Cox and Cox 2001: chapter 5).

2007;51) that the pottery from Barwick could have come from Lazencroft and been dumped on the site in order to raise the ground level and deal with localised flooding or waterlogging. Such use of pottery waste is extremely common but without further investigations in both Barwick and Lazencroft the suggestion is difficult to substantiate although petrographic and chemical analysis might clarify the issue. The range of pottery from Barwick includes the range of types expected from an 18th century country pottery. Four fabrics were identified and, as at Silkstone, there appeared to be a correlation between the type of fabric and the type of vessel. Fabrics 1 and 2 were both pale buff in colour with sparse, fine black and white inclusions with Fabric 2 distinguished by the presence of red streaks suggesting the mixing of different clays for a specific purpose, as seen at Silkstone. Both of these fabrics seem to have been used preferentially for Slipware dishes and Slip Coated hollow ware vessels. In contrast Fabrics 3 and 4 were both associated with the Late Blackware group and were both red in colour. Fabric 3 resembled the hard, fine dark red fabric typical of 17th century Blackware with sparse fine white inclusions and occasional rounded red and black grains. Fabric 4 was a much brighter red fabric with sparse fine red and black grit and occasional large round white inclusions. This apparently deliberate selection of specific clays for specific purposes suggests the knowledgeable exploitation of particular clay sources by the potters. The apparent regularity of the phenomenon suggests that it was a regular part of the practical knowledge or habitus of 18th century potters and implies a good knowledge of the local area and its soil types.

Barwick-in-Elmet Excavations undertaken in 2006 in Barwick-in-Elmet by the Barwick-in-Elmet Historical Society revealed part of a dump containing the typical waste products of an 18th century country pottery (Cumberpatch 2007a).

The range of vessel and ware types included a broad range of vernacular tablewares; Slipware, Slip Coated ware, Late Blackware and Mottled ware. The latter consisted of three sub-types although at least one (Type 3) was probably not made at the site (Cumberpatch 2007a: 16-22). Vessel types were broadly standard and included slipware dishes with trailed, banded, feathered and joggled designs internally, slip trailed hollow wares, wheel thrown dishes with a mottled glaze finish and a variety of hollow wares in Slip Coated ware and Late Blackware. The fragmentation of the assemblage meant that less information was recovered on the range of vessel forms than was desirable although the broad picture was generally clear.

Historical and documentary research (Teal, in BEHS 2007) has indicated that the site was owned by William Gough and his son, Godfrey, between 1747 and 1788. William was listed in the Barwick parish register as a potter from Brown Moor or Lazencroft (adjacent localities in the parish of Barwick) in the 1740s/1750s but there is no evidence that he occupied the site, preferring instead to rent it out to tenants. His son Godfrey occupied the site from c.1758 and appears to have brought up his family there. A number of entries in the parish register refer to him as a ‘potmaker’ although he may initially have made his living as a yeoman farmer (Teal in BEHS 2007:49). He seems to have remained in Barwick until 1788 after which he moved to Manchester, transferring the property to John Walton. There are no further references to the manufacture of pottery on the site and it seems likely that it ended with Godfrey Gough.

Brown Moor Pottery, Lazencroft Excavations by the Leeds Archaeological Fieldwork Society (LAFS) between 2006 and 2009 at Lazencroft Farm near Leeds revealed a considerable quantity of pottery wasters and the base of a hovel and kiln some 9 metres in diameter (Lonton 2006, Allday and Millard 2009). Documentary research has produced a lease dated 26th March 1739 between Edward Gascoigne of Parlington and William Gough of Lazencroft which made reference to

Evidence of pottery manufacture of 18th century date has also been identified at Lazencroft near Leeds, as discussed below. The suggestion has been made (BEHS 84

Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire ‘pott ovens, workhouses, warehouses and appurtenances’ (Allday and Millard 2009:597). A bankruptcy document dating to 1782 relating to Thomas Medhurst who appears to have purchased the lease from Gough referred to ‘four cottages, with pot ovens and other suitable erections for carrying on the business of a pottery’. The excavation has been documented by Allday and Millard (2009) and Lonton (2006) and while a full report has yet to be completed (Allday and Millard 2009:611) it is clear from the interim reports that a variety of wares were produced including Slipware and Mottled ware. No analysis of the fabrics is yet available and it will be of considerable interest to see whether any distinctions that may exist between the fabrics can be correlated with distinctions in the range of vessels manufactured. Given the evidence cited above for such practices, it seems likely that this will be the case, particularly given that the character of the Slipware vessels seems to reflect the presence of a number of potters working on the site (Allday and Millard 2009:605).

clear that those who founded and financed the potteries were seeking craftsmen from Staffordshire and that there were a series of relationships between those involved in both the establishment of the potteries and those who worked in them, the latter presumably supplying the technical expertise which the local entrepreneurs lacked. Other country potteries in Yorkshire The sites discussed above are those for which a useful amount of evidence is available but it is unlikely that they represent all of the country potteries operating in the study area in the 17th and 18th centuries. Further documentary work and a review of museum collections would probably reveal the existence of more such potteries and it is likely that commercial archaeological investigation will uncover hitherto unsuspected examples such as that recently identified at Scales in North Yorkshire (Young and Vince 2010). A number of such possible sites are included in Table 1 and Griffin (2012) lists more, including many known to have produced utilitarian earthenware, a subject discussed below.

Allday, Millard and Lonton have drawn attention to the fact that an individual named William Gough was also associated with the pottery at Midhope (discussed above) around 1727 and it seems likely that the two are one and the same individual. Barwick parish records indicate that three generations of Goughs worked in the area from the 1730s to the 1770s along with Samuel Malkin, Cornelius Toft and other potters.

One issue which requires further work is the manufacture of Brown Salt Glazed Stoneware. To date only one site within the study area with definite evidence for such manufacture is known. This is Warren Vale in Rawmarsh (Rotherham). A brief note in Post-Medieval Archaeology (Milefanti and Brears 1971) records the excavation of two pottery kilns following building work on the cottage at No. 9, Warren Vale. No documentary evidence was found for this pottery, but the artefactual assemblage suggested a late 18th or early 19th century date. Mould fragments suggested that the potters had been engaged in the manufacture of small ink bottles using open single piece moulds and small bottles using two-piece moulds. In the opinion of the authors this represented the earliest evidence for the nonfactory use of moulds for slip casting known at that time.

Lonton has suggested that Gough left Midhope and took the lease which enabled him to develop the Brown Moor Pottery from Edward Gascoigne because both families were Catholics who suffered from religious intolerance during the 18th century and who might therefore have sought mutual support as landlord and tenant in a generally hostile world. William Gough died in 1774 at the age of 87 and the latest reference to the pottery dates to 1782 (Leeds Mercury) when Thomas Medhurst’s estate, including the Brown Moor pottery, was advertised for sale.

There is little other evidence for stoneware manufacture although Llewellyn Jewitt claimed that stonewares were manufactured by the Butlers at Swinton and Cox and Cox refer to sherds from excavations on the site which appear to date to this period. The stoneware industry was, of course, a feature of north-east Derbyshire and Nottinghamshire in the 18th and 19th centuries with a number of places, notably Chesterfield, Alfreton and Nottingham having extremely important stoneware industries making tablewares and, later, cooking wares in vast quantities (Walter 1999).

Vessel types known at present include slipware cups, pressmoulded plates/dishes, tankards, posset pots and chamber pots. Mottled wares include tankards, porringers, jugs and large, flat-sided cooking pots. Some of the slipware plates/ dishes bear the initials S M which have been linked with the name of Samuel Malkin; parish records refer to Samuel Malkin and Thomas Malkin, potters in 1744 and 1745 who may have been the grandsons of the more famous Samuel Malkin of Burslem.

Utilitarian wares Throughout the period discussed here, the production of vernacular and formal tablewares was paralleled by the manufacture of utilitarian wares. Until the 19th century and the introduction of the coal-fired domestic range with the consequent expansion of brown salt glazed stoneware production into the realm of utilitarian ware, the sector was dominated by three types of pottery; Brown Glazed Coarseware, Brown Glazed Fineware and Yellow Glazed

It is of particular interest to note that a Samuel Malkin was also associated with the Swinton pottery in the 1760s along with his sons, Samuel and William (Cox and Cox 2001:26). The elder Samuel died in 1769 but William is named in a document of 1775 as still working at the Swinton pottery. There is clearly a need for further documentary research to disentangle the complicated family and commercial relationships involved here, but even at this stage it seems 85

Christopher Cumberpatch Coarseware. The origins of these wares can be traced back to the 17th century and, in the case of jars and jugs, perhaps even to the late medieval period although colours and, to some extent, shapes changed dramatically during the early post-medieval period. Brown and Yellow Glazed Coarseware remained in production into the mid 20th century and are still made in some craft potteries although Brown Glazed Fineware, (generally smaller hollow ware vessels) seem to have been eclipsed by the production of 19th century Brown Salt Glazed Stoneware and by the development of kitchenware vessels in refined earthenware bodies (Whiteware, Cane Coloured ware, Banded ware etc). A full study of this important industry is long overdue as it is clear that within the study area (and almost certainly across most of the country) potteries making utilitarian earthenwares were common. Griffin (2012) lists numerous potteries in which potters were manufacturing ‘coarse earthenware’, black and brown glazed earthenware and so on. Many are known from newspaper advertisements, bankruptcy proceedings and similar sources but few have been investigated archaeologically. A site was discovered close to the site of the medieval pottery in Rawmarsh near Rotherham (Cumberpatch 2004b) but although it produced clear evidence of the production of Brown Glazed Coarseware the developer declined to pay for the analysis of the material and no further work on the assemblage was possible.

the edged tool trades and as proposed by myself elsewhere for the vernacular tableware industry. Although it is far from clear from the published documentary data precisely how the country potteries operated, it is known that in many cases the named individuals who appear in the records as ‘potter’ or ‘pot-maker’ were in fact the owners or part-owners of a pottery (Griffin 2012:xii) rather than the people who actually made and fired the pots. How many of the actual workers maintained small-holdings or kept animals to supply household needs is unclear but there appears to be little direct evidence that the potteries described above operated on a part-year basis in spite of the obvious problems inherent in potting year-round in the British climate. The general impression is of an industry that was sufficiently profitable to attract the interest of entrepreneurs such as Joseph Flint, William Malpass and William Fenney who maintained a range of business interests in addition to pottery. Others, including John Fox and Robert Blackburn, appear to have had interests in other pyrotechnological industries, specifically glass manufacture. At Silkstone the pottery seems to have been one of several enterprises owned by James Scott and while it is quite possible that his tenant(s) could have been farmer-potters, the references in contemporary records to several individuals specifically as potters or pot makers may indicate that this was their primary business. Table 2 summarises the ownership of the sites discussed above as it is understood at present.

A better outcome was achieved in respect of the site of the Attercliffe New Pottery (also known as the Darnall Road Pottery) in Darnall, Sheffield where an industrial pottery of mid to late 19th century date was partially excavated by ARCUS (University of Sheffield) in 2006/2007 (ARCUS 2006, Cumberpatch 2007b, Griffin 2012:55). The ceramic assemblage from the site included items associated with pottery production; wasters, ring stilts, rods and clay pads (Cumberpatch 2007b). It appeared that Unglazed Red Earthenware vessels (predominantly horticultural wares) and Brown Glazed Coarseware and possibly Yellow Glazed Coarseware were all manufactured on the site. The fact that such wares were produced on an industrial scale is testament to the economic significance of the utilitarian ware industry.

The one exception to the picture of an industry characterised by inter-linked family and business connections would seem to be the Halliday family of the Pule Hill, Bate Hayne and Howcans potteries who seem to have pursued their own trajectory and to have maintained a family business from the mid-17th century until 1916. Their status as gypsies of Scottish origin may have set them apart socially from other Yorkshire potters either because of prejudice or because of a strong internal dynamic that kept the business ‘in the family’ when other potters were seeking outside investment or were employed by individuals with wider financial and business interests. Further documentary work on the Halliday family would be of considerable interest, as would the opportunity to publish the pottery assemblages from Pule Hill and Bate Hayne.

The best example of the traditional production of utilitarian wares is John Anderson’s film of Isaac Button working at the Soil Hill Pottery in 1964. Comparison of Button’s working methods with the traces that can be seen on archaeological examples indicates the limited degree to which the industry had, in terms of practice, changed since the early 18th century. The Soil Hill Pottery remains standing, albeit in a poor condition (Plates 1 and 2). Although it enjoys limited statutory protection, its current condition gives grave cause for concern.

In the case of the Goughs, Catholics who appear to have sought a co-religionist landlord to mitigate the effects of sectarian bigotry, there seems to have been no sense of independence in the way that appears to characterise the Hallidays. The recruitment of Samuel Malkin (who does not seem to have been a Catholic) appears to be primarily a business decision unaffected by sectarian concerns on either side and in this regard they fit easily within the broader picture of pottery production during the 18th century.

The organisation and structures of production

The terms of William Gough’s lease, as cited by Allday and Millard, make no mention of agricultural buildings but do include reference to workhouses and warehouses which may imply that the pottery was substantial enough

The evidence presented above does not support the description of the country potteries as part of a dual economy as defined and described by Hey and Battye for 86

Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire Site

Date range

Owners and others associated with the potteries

Reference

Northowram, Pule Hill

Mid-17th century

Halliday family (see below) who moved their operations to Bate Ain in mid C17th)

Cumberpatch 2000, Lawrence 1974, Bryant, unpublished, Griffin 2012

Northowram Bate Hayne (or Ain)

Later 17th to late 18th century

The Halliday family moved from Pule Hill in mid C17th and worked it until the late C18th, four brothers, Richard, Isaac, George and Abraham being the last to do so

Cumberpatch 2000, Lawrence 1974, Griffin 2012

Northowram Howcans

LC18th - 1916

George & Abraham Halliday left Bate Hayne in the late C18th and established a new pottery at Howcans; run as a family concern until the death of William Halliday in 1916

Griffin 2012

Sheffield Manor

Post c.1708 – 1738 & later

John Fox (son of Mary Blackburn)

Beswick, pers. comm., 1978, Harlan 2010, Hadley and Harlan 2011

Midhope / Midhopestones

1720 – c. 1845

George Walker and Robert & Mary Blackburn (1720), William Gough (1727), John Taylor (1762), John Whitley (1765), Joshua Lindley (1788), Thomas Fawley 1803, William Appleyard 1816, various until 1845

Lawrence 1974, Ashurst 1987, 2002, Kenworthy 1928, Griffin 2012

Swinton

1745 – 68

Joseph Flint (1745-55); Edward & Elizabeth Butler (175565); William Malpass (1765-8); William Malpass & William Fenney (1768-77); Samuel Malkin and sons involved from the 1760s onwards

Cox and Cox 2001; Chapter 1

Brown Moor pottery, Lazencroft

1739 – c.1777, 1782

William Gough I (1739), William Gough II (1774), Samuel Malkin, Thomas Malkin (1744, 1745) Cornelius Toft

Lonton 2008, Griffin 2012;266-7

Barwick-in-Elmet

c. 1747 - 1788

William and Godfrey Gough (1758 – 1788)

Cumberpatch 2007, JB Archaeological Services/BEHS 2007

Silkstone

c.1754 - 1802

James Scott (owner) w/ John Bailey, Ralph Taylor, Joseph Goldthorpe and Michael Taylor (tenants), 1745 – 17??; Richard Fenton (owner) w/ John Bailey, Ralph Taylor and Joseph Goldthorpe, tenants (1775); John Taylor (1767), William Taylor (1812-1821)

Brears 1971, Lawrence 1974, Cumberpatch 2004c, White 2007, Griffin 2012

Bolsterstone (Bate Green)

c. 1778 – 1796

Joseph Lindley (c.1778 – 1796)

Ashurst 1987, 2002

Godfrey Gough was the son of William Gough I and brother of William Gough II. William Gough I held a lease on the Midhope Pottery from 1727, moving to Lazencroft to found the Brown Moor Pottery in 1739.

Table 2. Summary of the ownership details of the potteries discussed in the text.

to include facilities which could have included drying sheds, an essential requirement for year-round potting in an inclement climate. The recruitment of potters from Staffordshire, also attested in the case of the pottery factories (including, perhaps, William Hawley, first of several generations of distinguished South Yorkshire potters; Griffin 2012:59), seems also to have been the case in some at least of the country potteries. Samuel Malkin, Cornelius Toft and others would presumably have had no reason to leave Staffordshire for Yorkshire if the only role on offer was that of a farmer-potter. Griffin has published census figures for the mid and late 19th century which show a limited number of pottery workers with an origin outside South Yorkshire and from the Staffordshire pottery towns specifically (2012:xii). The figures for Mexborough and Swinton for 1851 show thirty-seven individuals from Staffordshire and a further eight from towns with a pottery industry other than those in Yorkshire and Staffordshire. These figures relate to a period during which the Yorkshire pottery industry was well established and was presumably producing its own skilled workforce via apprenticeships and so may have had

less need to recruit workers from Staffordshire. It would be of considerable interest to investigate documentary sources relating to the 18th century in order to determine the numbers of migrants from Staffordshire who arrived in Yorkshire during this period in order to work in the pottery industry. Taken together with the archaeological evidence in the form of the pottery assemblages, this might allow the definition and delineation of groups of people, some of them inter-related, who were responsible for the establishment and maintenance of traditions of practice over several generations. In broader terms, the picture presented above is not consistent with that of a small-scale industry supplying a shrinking market at a time when new technologies were making rapid inroads into a traditional consumer base. That local entrepreneurs were willing and able to invest in country potteries suggests that these enterprises were seen as viable business opportunities, the equal of the newer fineware potteries, glassworks, inns, lime works and so on. The case of William Fenney may be slightly unusual but the evidence of his dispute with his mother-in-law over the 87

Christopher Cumberpatch modernisation of the Bolsterstone glassworks suggests that he was a man of vision and perhaps an early adopter of new technology. While investment in a country pottery may have seemed to be a viable and entirely sensible decision, with the benefit of hindsight it is clear that investment in a pottery producing fine stoneware and refined earthenware was the one with more long-term stability and Fenney may have been aware of this, given his interest first in the Rotherham old Pottery and later in the transformation of Swinton. Having said this, the later history of the Rockingham Pottery under the Bramelds testifies to the risks inherent in a fixation on new technology and new products at the expense of well-established product lines and technologies.

which might incorporate the documentary as well as the archaeological data. Etienne Wenger has suggested that Communities of practice are formed by people who engage in a process of collective learning in a shared domain of human endeavour (Wenger: nd). Wenger’s main focus appears to be on contemporary situations but the idea has a good deal to commend it in terms of offering a framework of explanation which goes beyond simply using untheorised terms such as ‘fashion’ which confuse or ignore relations of production and consumption and which offer nothing in the way of explanation (Cumberpatch 2003). While not invalidating the useful heuristic framework provided by the notion of modes of production, it appears to offer a means of understanding both the kind of socially-driven technological change that characterises the 18th century pottery industry as well as providing a framework within which the stability of structures of production can be conceptualised. It may offer a way of combining the archaeological data pertaining to the production of the pottery with the documentary evidence pertaining to the personal and financial organisation of the industry in a manner which does not reduce one to a mere illustration of the other. More importantly it offers a theoretical structure within which the definition of a mode of production can be combined with a practical means of encompassing change as well as stability within a model of social and economic practice. This is clearly what is required when dealing with a period of rapid and fundamental change in society and economy as the 18th century. As a community of practice, the 18th century vernacular tableware industry is characterised by the traits summarised in Table 3 and this aspect may be discussed at greater length in a future article.

If correct, this interpretation of the data has wider implications. Potters working in the vernacular tradition were not the last representatives of a dying tradition but were actively supplying a steady demand for a type of pottery that still held an important place in many, if not most, households throughout the 18th century. While it is quite possible that some potteries were primarily family concerns possibly run in conjunction with small-holdings it is also clear that local entrepreneurs and investors saw this branch of the industry as a viable and potentially profitable investment. In this respect the manufacture of vernacular pottery seems to have more in common with the glass industry with which it shared such close connections than it does with the edged tool makers of Hallamshire documented by Hey and Battye. While Yorkshire may not have produced a rival to Thomas Toft in terms of individual craftsmanship, a similar range of techniques and skills were widely employed to produce a broadly similar range of wares which, to judge from the archaeological evidence, appear to have been used extensively in both rural and urban homes. In fact, the situation is one in which we see potters across the study area connected by traditions of practice and by family and financial relationships in a manner which suggests that the most productive way to conceive of them and to understand them is through the idea of communities of practice.

Pottery factories in South and West Yorkshire The industrial production of pottery in the study area has been the subject of a great deal of detailed study in recent years to the extent that we now have both general surveys and specific studies of individual potteries. Publications by Lawrence (1974), Griffin (2001, 2005, 2012), Cox and Cox (1983, 2001), Cox (2013) and Edwards and Hampson (2005) have rendered any detailed account of individual factories superfluous in the present context although some brief notes are necessary before moving onto a consideration of the role of consumption in structuring the

I have written elsewhere about the importance of Bourdieu’s concept of habitus in understanding the production and consumption of medieval and postmedieval pottery (Cumberpatch 1997, 2003) and while this is also applicable to the 18th century vernacular pottery industry, the idea of communities of practice may offer a compatible theoretical and practical framework for further investigation of the character and structure of the industry Standardisation of technology

Wheel-thrown and moulded vessels; common methods of decoration; similar kiln and other structures

Standardisation of products

Limited range of ware types: Slipware, Slip Coated ware, Mottled ware, late Blackware

Internal relationships organised around familial and professional structures

Involvement of individuals with a background in pyrotechnology (glass and pottery industry); recruitment of workers with existing skills

Table 3. Eighteenth century vernacular tableware manufacture; traits and characteristics

88

Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire production and distribution of pottery in the period under discussion.

inventive manufacturers such as Malpass and Fenney and their later successors the Brameld brothers who ran the Rockingham Pottery until they bankrupted themselves in 1842, in part because of their quest for novelty and technical sophistication (Cox and Cox 2001:Chapter 5).

While the influence of the Staffordshire pottery industry as a source of both technical knowledge and marketing skill must not be underestimated it should also be recognised that many of the same physical conditions existed in Yorkshire and the north-east of England as they did is Staffordshire; coal, water, canals (later railways), major ports and a local tradition of pottery manufacture going back into the medieval period are found on both sides of the Pennines. Staffordshire’s role in technological and stylistic developments from the 18th century onwards is undeniable, but throughout the 18th century and into the late 19th century, communities in Yorkshire and the northeast were able to obtain earthenwares and high quality tablewares from local producers who were, in many cases, also significant exporters. Potteries such as Rockingham supplied aristocratic and royal clients and, technologically, were at the leading edge of innovation even if their finances were less than secure (Cox and Cox 2001, Griffin 2001). Foreign markets were also of considerable importance. The Don Pottery during its ownership by the Green family (1801 – c.1830) exported to Russia, France, Belgium, Germany and South America (Griffin 2001:38) and the customs records from the port of Hull record the export of considerable cargoes of pottery, one example mentioned by Griffin consisting of 258 crates of earthenware. Pots from the Don Pottery carried backstamps recording the names of wholesalers and dealers, a number of them seemingly of foreign origin and also included a mark in Greek (Griffin 2001:204;7; see also Cumberpatch, 2008a, for examples from Upper Allen Street in Sheffield).

It is to the 18th century commodity culture and the consumption of pottery in Yorkshire that we must now turn. Ware classes and the contexts of use and consumption In an earlier paper (Cumberpatch 2003) I described how the end of the medieval potting tradition was characterised by the appearance of new types of pottery distinguished by a novel range of phenomenological variables, including colour, texture and shape which were strikingly different to those of medieval pottery. I suggested that the reasons for these changes should be sought not in the area of technological competence and practice but rather in terms of social change as manifested and expressed in the range and style of everyday material culture. That this change seems to date from the mid 15th century (Irving unpublished) rather than the late 15th century (Cumberpatch 2003) with the appearance of Cistercian wares and so predates the traditional end of the medieval period is a matter that requires further investigation but presents no significant theoretical problem. For the purposes of this paper, the existence of the distinctive post-medieval pottery tradition spanning the mid 15th to the early 18th century will be taken as given. The significance of this observation is the evident continuation of this tradition, as represented by utilitarian ware and vernacular tableware, into and throughout the 18th century. As I have argued above this was a vibrant tradition in its own right which was robust enough to be of interest to local entrepreneurs.

In Sheffield there is some evidence in the form of maker’s marks (albeit not yet fully tabulated or reviewed) from sites excavated in the city that Staffordshire marks only become common in the later 19th century. While some earlier ones are known it seems that during the later 18th and early 19th century local factories; the Bramelds of Rockingham, the Twigg’s of Kilnhurst, the Barkers of the Don Pottery and others account for the majority of marked pieces. This seems to extend to the highest levels of society and may also be reflected in the historical record. In 1753 the 2nd Marquis of Rockingham re-equipped his house with tableware after his marriage, ordering porcelain from dealers in London. From 1768 until the mid 19th century he and his successors were able to purchase the highest quality earthenware and porcelain from the Rockingham works at Swinton, from where it was also widely exported (Cox and Cox 2001).

Much of the literature upon which we depend for the interpretation of early modern and recent pottery assemblages from archaeological investigations is derived from traditions of connoisseurship and collecting on the one hand and social history on the other. Both these traditions provide a valuable complement to archaeological data but as neither draws primarily on such data both follow individual paths which are not entirely congruent with archaeological discourse. Social history, as represented by the work of, amongst many others, Porter (1991, 2000), Brewer and Porter (1993), Kowaleski-Wallace (1997) Richards (1999), Berg (1999), Berg and Clifford (1999) and Vickery (2009) provides a wider social context within which we can place our assemblages of ceramics. A substantial parallel body of work documents in detail the products of individual potteries, the work of particular potters, designers and entrepreneurs with particular reference to the formal tablewares and ornamental wares of the 18th and 19th centuries (including Lawrence 1974, Godden 1991, 2004, Edwards 1994, Coysh and Henrywood 1997, Cox and Cox 2001, Edwards and Hampson 2005 and Griffin 2001,

In his survey of the commodity culture of Victorian England (1851 – 1914) Thomas Richards (1990) has suggested that a commodity culture emerged in England during the mid to late 19th century. In suggesting this, he appears to fail to take into account the ceaseless marketing of new forms, colours and types of pottery from the early 18th century onwards, not only by individuals such as Wedgwood but also by other lesser known but equally 89

Christopher Cumberpatch 2005, 2012) often encompassing valuable accounts of the social history of the pots, potters and potteries. These two traditions of research, which have contributed so much to the study of formal tableware and the practices of 18th century polite society, appear to have far less to say about what went on in the dairy, the kitchen or the alehouse, at least in terms of the use and character of material culture. Likewise such accounts have tended to sidestep the question of production in the country potteries in favour of a focus on the factories in which the new formal tablewares were manufactured.

bought in quantities significant enough for production to be a profitable enterprise throughout the 18th century. Such wares were also designed for use at the table, although perhaps not the same table as that which saw the deployment of the formal white and cream tableware. It would seem that on the basis of colour and function, the black, brown and mottled earthenware mugs, tankards and porringers were a development of the multi-handled Blackware tygs and drinking vessels of the 17th century. The 18th century vessels were better suited to the practice of individual eating and drinking than were some of the earlier vessels whose large size and multiple handles seem intended to facilitate the passing of vessels from hand to hand, perhaps as part of communal drinking rituals. In this regard it would seem that the tendency towards individualisation which is seen in terms of the formal white and cream tablewares of the 18th century may also be reflected in the vernacular tablewares even though they were used in social contexts other than those in which the white and cream wares were used.

There is little doubt that the rise of an ideology of sensibility and civility during the 18th century, together with the technical and entreprenurial skill of those individuals whose names have become linked with the 18th century pottery industry was instrumental in establishing a tradition of white and decorated tablewares which persists to this day. It is widely accepted that this was one aspect of what Matthew Johnson and others have called the ‘rise of the Georgian Order’ (1996) and appears to go along with a concern for whiteness in other aspects of material culture – the bleaching of cloth being another example (Tarlow 2007:165-171). The mass production and marketing of White Salt Glazed Stoneware and the later Creamware, Pearlware and English porcelain brought such wares into the reach of the emerging middle and lower middle classes and clearly represented the material culture sought after by those with social aspirations (Richards 1999: Chapter 4). Hard, durable and relatively cheap, the new types of pottery rapidly replaced Tin Glazed Earthenware and competed successfully with Chinese porcelain. Archaeologically, the ubiquity of these wares seems to attest to the rapidity of their adoption; they appear not only on urban sites but also widely on rural sites and while the chronological resolution is generally too poor to determine relative rates of adoption, there is nothing to suggest that there was any great time-lag between adoption in rural areas and in towns.

Alongside the drinking vessels, porringers and bowls we have the slipware and mottled ware plates and dishes with bright yellow and red-brown decoration in the case of press-moulded slipwares and a honey-coloured mottled finish in the case of the wheel-thrown plates. While this can be described as a tableware tradition, it was one which involved a simpler range of vessel types, without the elaborate diversification into many specialised vessel forms that we see in the sets of formal tableware (e.g. Griffin 2001: 56-97). The obvious question to be asked is how the vernacular tablewares fit into our picture of 18th century society given that the narrative is dominated by the model of middle class civility and social aspiration and its material culture correlates. How can we accommodate the persistence of what would appear to be a development of a continuing tradition of practice alongside the emergent one. The answer is presumably that we are dealing with different spheres of consumption within which different classes of object were deemed more or less appropriate and used accordingly. Given that the colours of the vernacular tablewares resemble those of the utilitarian earthenwares more closely than they do the formal tablewares it seems probable that they were used in less formal situations within the household. But is this a class-based distinction and if so is it present within households or between households of different social class and aspiration or both? Is it a gender-based distinction as Yentsch (1991) has suggested? Can the vernacular tableware be seen as subverting the emergent Georgian Order by offering a subtext to the apparently dominant narrative of whiteness, differentiation and precision in function, order and civility? If so then where, conceptually speaking, was this sub-text located and what was its relationship in terms of practice, to the new social practices that emerged and flourished in the course of the 18th century?

In terms of the domestic wares generally, the persistence of an essentially post-medieval tradition of utilitarian vessels designed for processing and storage of food (bowls, pancheons, jars, cisterns etc) contrasts with the deployment of fine white and cream coloured wares which mark the commitment of the emerging middle and yeoman classes to the values of civility and sensibility. Only in the 19th century do kitchenwares in refined earthenware and stoneware bodies appear in any quantity and even then traditional pancheons persist into the 20th century. The distinction between the world of the formal dining table and the rituals of tea and coffee drinking (with all that they represent in social terms) and the kitchen and allied spaces is perhaps to be expected given what else we know of society at this time. While Lorna Weatherill has pointed out ‘the household did not project its image through its cooking equipment’ (1996) this observation does not take account of the fact that in addition to the utilitarian wares, vernacular tablewares in black, brown, yellow and bright variegated colours remained in production and were 90

Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire These questions cannot be answered here but it is suggested that, given the historical evidence, the co-existence of the two ceramic traditions reflects a dynamic relationship between the domestic and public spheres within the home during the 18th century while class, income, gender structure and the nature of aspiration all contributed to variability within and between families and households (Vickery 2009). Some brief examples may illustrate the point.

extract usable information from the majority of sites (Cumberpatch unpublished 1, 2, Cumberpatch 2005a). Similar, although less acute problems affect many urban assemblages but some are of interest in the present context. Tables 4, 5 and 6 summarise data from a variety of sites across the study area. In each case medieval, postmedieval pottery and mid to late 19th century pottery has been omitted from the total and data have been extracted from only those phases dating to the 18th and early 19th centuries. The percentages cited in Table 6 are based upon the estimated (maximum) number of vessels figure (ENV) calculated as a proportion of the total from the phase under consideration. Full details of the assemblages can be found in the references cited below.

In his autobiography Benjamin Franklin describes how his life changed in response to his wife’s aspirations for her husband: It was lucky for me that I had [a wife] as much dispos’d to industry and frugality as myself. … We kept no idle servants, our table was plain and simple, our furniture of the cheapest. For instance, my breakfast was a long time bread and milk (no tea), and I ate it out of a twopenny earthen porringer, with a pewter spoon. But mark how luxury will enter families, and make a progress, in spite of principle: being call’d one morning to breakfast, I found it in a China bowl, with a spoon of silver! They had been bought for me without my knowledge by my wife, and had cost her the enormous sum of three-andtwenty shillings, for which she had no other excuse or apology to make, but that she thought her husband deserv’d a silver spoon and China bowl as well as any of his neighbors. This was the first appearance of plate and China in our house, which afterward, in a course of years, as our wealth increas’d, augmented gradually to several hundred pounds in value (Franklin 2005:51).

Excavations on the site of Scholes Lodge Farm near Leeds (a few kilometres north of Lazencroft and west of Barwick-in-Elmet) revealed an 18th century farm building (probably a house later used as a barn) overlying an earlier farm of medieval date (Cumberpatch 2007c, Lightfoot, McCluskey and Cumberpatch 2008). The 18th and early 19th century phases of activity (Phases 2b and 2c; Tables 4 and 5) on the site produced mixed groups of pottery which included residual medieval material alongside both formal and vernacular tableware as well as the ubiquitous utilitarian ware. The formal tableware included Tin Glazed Earthenware, White Salt Glazed Stoneware, Creamware and Pearlware while the vernacular tableware included Slipware, Mottled ware and Late Blackware. Formal tableware was prominent in both phases with the numbers and diversity of types somewhat higher in Phase 2c than in 2b (although residuality in the case of the Tin Glazed Earthenwares and White Salt Glazed Stoneware cannot be ruled out) with Creamware prominent in both phases and Pearlware increasing in importance in the later phase. Numbers of vernacular table ware vessels were lower in Phase 2c than in 2b but the proportions of the different wares remain similar apart from a fall-off in the number of Mottled ware vessels. It would perhaps be unwise to read too much into the details of the figures but the broad pattern seems to be of a farming household that was able to acquire both formal and vernacular tablewares and seems to have been able to follow the fashionable shift from Creamware to Pearlware in its various forms (plain, transfer printed and edged) in the later 18th and early 19th centuries.

Although involving an imported Chinese porcelain bowl rather than a fine stoneware or refined earthenware vessel, this change would seem to reflect the way in which vernacular tableware occupied a lower status in relation to the refined earthenwares which ultimately replaced them. To place the cost of the China bowl in perspective, Spavold and Brown quote prices of between a farthing and twopence for Ticknall pots in the seventeenth century (depending on size) and note that these prices remained stable for several hundred years (2005:153-4). Whether Franklin was really happy with his earthenware porringer or whether he was striving to present an image of a man of simple tastes only urged on to greater self-indulgence by his wife is unclear but the anecdote serves to exemplify the social distinction between the two types of pottery.

A broadly similar pattern can be seen in the data from the excavation of stone-built estate worker’s cottages at Alderley Sandhills, Cheshire (Cumberpatch 2004c). Here, an American-style quadrant method of excavation imposed artificial problems on the analysis and blurred the distinction between contexts and phases but the evidence from Quadrant B, the location of the earliest assemblage of pottery from the site (late 18th and 19th centuries) seems to indicate a similar pattern of pottery acquisition to that seen at Scholes Lodge Farm. Vernacular tablewares were represented by Late Blackware, Slipware, Redware and Mottled ware which together constituted 8.8% of the total

Excavated sites in Yorkshire have tended to produce mixed pottery assemblages in which vernacular and formal tableware occur side-by-side irrespective of the original contexts or chronologies of use. In the case of sites in Sheffield the assemblages can best be described as haphazard, semi-homogenised, aggregated samples and without detailed work on the taphonomic and siteformation issues (deemed to be prohibited by the dictates of commercial archaeological practice) it is difficult to 91

Christopher Cumberpatch Type

Estimated (maximum) number of vessels

Type

Estimated (maximum) number of vessels

Brown Glazed Coarseware

8

Brown Glazed Coarseware

1

Brown Glazed Coarseware type

1

Brown Glazed Coarseware type

2

Yellow Glazed Coarseware

1

Yellow Glazed Coarseware

5

Brown Glazed Fine ware

2

Brown Glazed Fine ware

2 1

White Salt Glazed stoneware

4

Tin Glazed Earthenware

Creamware

10

White Salt Glazed Stoneware

1

Pearlware

1

Creamware

35

TP Pearlware

1

Pearlware

2

Fine Redware

3

Edged ware

1

TP Pearlware

12

Redware

20

Fine Redware

2

Slipware Type 1

1

Cane Coloured ware

1

Late Blackware

15

Slip Banded CC ware

1

Mottled ware type

1

Whiteware

1

Mottled ware

9

Slip Coated ware

1

Redware

10

Slipware

2

Slipware type 1

1

Late Blackware

12

Mottled ware

2

Slipware

2

Brown Salt Glazed stoneware

1

Table 4. Eighteenth century pottery from Scholes Lodge Farm Phase 2b (Residual medieval and early postmedieval pottery omitted from data)

Table 5. Eighteenth century pottery from Scholes Lodge Farm Phase 2c (Residual medieval and early postmedieval pottery omitted from data)

while formal tablewares constituted 42.7% of the total (the remaining material being utilitarian ware and 19th century table and kitchen wares). While the proportion of vernacular tableware was low, it was nevertheless present alongside the more expensive wares.

households represented by these data. At Garden Street and Blonk Street in Sheffield (Cumberpatch 2007e and 2008b) the representation of formal tableware was higher than that of vernacular tableware. In both cases this is slightly curious. The Blonk Street site lies close the location of the markets and the shambles, a situation not dissimilar to that of Church Walk in Doncaster. Here the issue of site formation may be an issue as might the fact that the site may be relatively late in date, as indicated by the higher proportions of transfer printed Pearlware in the assemblage. Garden Street lies in an area known as The Crofts. By 1830 this was a notorious slum but in the 18th century it was a densely built up area of craft and industry and included a number of substantial dwellings built by local manufacturers adjacent to their works (Belford 2001: 106-7) and this may account for the higher proportion of formal than of vernacular tableware. A site on Tenter Street (Cumberpatch 2005b) at the eastern edge of The Crofts produced an assemblage closer in character to those from Wakefield and Bradford with vernacular tableware forming a substantially higher proportion of the total than formal tableware. Further work on the assemblages from The Crofts, including the full analysis of an assemblage from Edward Street, might allow a finer-grained analysis of the data in spite of the problems that affect most of the excavated assemblages from Sheffield (Cumberpatch 2005a).

At Ash House Farm near Sheffield the situation was almost the reverse of that seen at Alderley Sandhills with formal tableware constituting just 4.6% of the total while vernacular tableware formed 70.5% (Cumberpatch 2006). Urban sites in South and West Yorkshire generally show a consistent pattern with formal tablewares constituting a lower proportion of the total than vernacular tablewares while utilitarian wares vary widely in proportion (Table 6). Some caution is required in interpreting these data. Two particular cases may be noted. The site at Church Walk in Doncaster (Cumberpatch 2007d) lay in the midst of an area of industry, notably tanning and carcase rendering and this may account for the very low representation of formal tableware in the assemblage. The Trinity Walk, Wakefield site (Cumberpatch 2011b) was not phased stratigraphically and the phasing used in the pottery report was based upon the ceramic evidence alone, an unsatisfactory state of affairs. These caveats having been noted, the pattern seen on the sites from Wakefield and Bradford is similar with all showing a higher proportion of vernacular tableware than formal tableware even while the proportion of the latter was high enough to suggest that it was a regular presence in the

92

Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire Phase

Utililarian ware

Vernacular tableware

Formal tableware

Notes

Scholes Lodge Farm

2b

10.4%

43.4%

16.5%

Lightfoot, McCluskey & Cumberpatch 2008, Cumberpatch 2007c

Scholes Lodge Farm

2c

10.4%

29.1%

56.2%

Site Rural sites

Alderley Sandhills Ash House Farm, Sheffield

48.5%

8.8%

42.7%

Cumberpatch 2004c

1

24.2%

70.5%

4.6%

Cumberpatch 2006b

4

12.2%

34.1%

3.5%

Horizon EMOD01: Cumberpatch et al 2007d

20%

36.5%

11.1%

Cumberpatch 2007g

11.1%

45.6%

25.0%

Cumberpatch 2009

Urban sites Church Walk, Doncaster Westgate Wakefield (AOC) Westgate Wakefield (BUFAU)

5

Trinity Walk Wakefield

C3 & C4

24.3%

40.6%

17.7%

Ceramic phasing: Cumberpatch 2011b:Appendix 1

Broadway Centre, Bradford

4&5

8%

38.1%

14.5%

AS WYAS 2006, Cumberpatch 2007f; see report for discussion of phasing

Blonk St Sheffield

1, 2, 3, 4

18%

31.3%

43.8%

Cumberpatch 2008b

Garden Street, Sheffield

Early

35.4%

26.9%

35.1%

Cumberpatch 2007e

30.6%

37.1%

20.7%

Cumberpatch 2005b

7.7%

24.4%

64%

Cumberpatch 2012c

13.9%

6.1%

71.1%

Cumberpatch 2012b

Tenter St Sheffield Other sites Sale Old Hall Flower Estate, Wincobank, Sheffield

All but latest features

Table 6. Proportions of different ware classes on sites in South and West Yorkshire with additional examples from Cheshire and Greater Manchester

Conclusion

Two assemblages from more well-to-do households, Sale Old Hall in Greater Manchester and Wincobank Hall in Sheffield (the Flower Estate) show a significant contrast with both the urban and rural sites (Cumberpatch 2012b, c). Both produced assemblages dominated by formal tablewares with low representations of utilitarian and vernacular tableware. While this suggests differential uptake of formal tableware it is notable that vernacular tableware was nevertheless still present and was presumably in use in some contexts within almost all households.

In this article I have tried, through a review of the current literature, to define and delineate the scope of an archaeological approach to aspects of the pottery industry of the 17th and 18th centuries in southern Yorkshire. My concentration on the country potteries as opposed to the factories reflects the fact that it is in this area that archaeology can make a significant contribution to our understanding of the pottery industry and can provide data and interpretations that complement the extensive historical and documentary research already carried out into the production of formal tableware by historians and collectors. It is clear that a considerable amount more work remains to be done on consumer as well as production sites although the recent history of archaeology in Sheffield does not inspire confidence in commercial excavation as the means to undertake this work. It is also unfortunate that Yorkshire is one of the few areas of the country to be excluded from recent efforts by English Heritage to define regional research agendas which might serve to outline the scope and potential of new work and to stimulate university-based research using the vast amount of data generated by commercial archaeology. Given this situation it is presumably the responsibility of those of us working in this supposedly backward province to make good the lack although there is no clear mechanism by which this might be done.

This brief survey of data selected on a haphazard basis is not intended to contradict or subvert the established accounts of 18th century pottery use but to draw attention to what seems to be a neglected phenomenon. Taken together, the evidence reviewed above pertaining to both production and consumption would seem to indicate a dynamic market for vernacular tableware throughout the 18th century. How far we can understand its eventual disappearance in terms of simple competition from the larger pottery factories with their larger workforces and economies of scale and how far we need to look more closely at changing patterns of consumption is unclear at present and must be investigated further in the future.

93

Christopher Cumberpatch Possible future areas of investigation include:

archaeology firms apparently in order to cut costs. The very limited cost savings resulting from this largely unacknowledged decline in the quality of archaeological fieldwork are far outweighed by the loss of detail in the grey literature reports and the fundamental compromise of the cardinal principle of preservation by record.

Pottery production: •

Petrographic and chemical analysis to reveal more about the deliberate manipulation of fabrics/clay bodies to produce intended outcomes;

•

Detailed analysis of production technology to characterise inputs and outputs;

•

Documentary research to illuminate the relationships between potters and entrepreneurs, the employment of local and immigrant workers and the networks of association and familial connection within the pottery and related industries

The re-establishment of minimum standards across the industry is an essential pre-requisite for moving on from mere description to interpretation and understanding and to re-establishing archaeology as a research-based discipline and not merely an adjunct of the financial sector of the wider economy. Acknowledgements This paper draws on the results of work I have undertaken for commercial archaeology units and trusts and for voluntary sector projects funded mainly by the Heritage Lottery Fund. The additional research necessary to move beyond the confines of single sites has been undertaken in my own time and at my own expense. The critical reader can be assured that no shareholder dividends, managerial bonus payments or corporate profits have been damaged in the preparation of this paper.

Circulation, distribution and marketing: Both documentary research and scientific analysis have a role to play here with the work of Spavold and Brown at Ticknall providing a model for the analysis of documentary resources and the modelling of the market system. Similarly White’s comparison of the petrographic and chemical characteristics of wares from Silkstone, Bristol and Stoke-on-Trent indicates that the further and more extensive use of these techniques when applied to existing archives will have a significant part to play in the study of the early modern pottery industry.

Thanks are due to the following individuals and organisations for their assistance with aspects of this paper: Paul Blinkhorn, Vicky Crewe (Sheffield University), Mark Fletcher, John Griffin, Dawn Hadley (Sheffield University), George Haggarty, John Hudson, Graham Oliver, Geoff Preece, Peter Robinson (Doncaster Museum and Art Gallery)

Consumption: In order to understand the consumption of pottery, we need not only to make greater use of documentary data (including probate inventories) but also to undertake contextually sensitive analysis and reanalysis of excavation records with the explicit aim of resolving issues around site formation processes, particularly on urban sites. This will require a return to traditional methods of post-excavation analysis including the use of the Harris Matrix and the practical principles set out in MAP II (English Heritage 1991) which have been abandoned by some commercial

Archaeological Services WYAS, Sheffield University, Hope Historical Society, Castleton Historical Society, English Heritage, Barwick-in-Elmet Historical Society The responsibility for the interpretation of the data remains my own as does responsibility for any errors or omissions in the text.

94

Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire Bibliography

Berg, M and Clifford, H. 1999 Consumers and luxury: Consumer culture in Europe 1650 – 1850. Manchester University Press Beswick, P. 1978 The Sheffield Manor products from the Peacock Inn In: P. Borne, T. Courtney and P. Dixon The Peacock Inn, Chesterfield. Derbyshire Archaeological Journal 98; 47. Boyle, A. 2002-2003 The Cistercian ware products of Ticknall, south Derbyshire Medieval Ceramics 26/27; 113 – 118. Brears, P.C.D. 1967 Excavations at Potovens, near, Wakefield Post-medieval Archaeology 1; 3 - 43 Brears, P.C.D. 1971 The English Country Pottery David and Charles. Brewer, J and Porter, R. 1993 Consumption and the world of goods Routledge Bryant. G.F. 1967 Yorkshire, West Riding, Halifax PostMedieval Archaeology 1; 116-118 Cox, A. 2013 The Newhill Pottery; Its sale and connections with the Kilnhurst and Don Potteries The Northern Ceramic Society Newsletter 171, 35-43 Cox, A. and Cox, A. 1983 Recent excavations at the Swinton Pottery: White Salt-Glazed Stoneware and Creamware pre-1785 Transactions of the English Ceramic Circle 11; 232-254 Cox, A and Cox A. 2001 Rockingham 1745 – 1842 Antique Collectors Club Coysh, A.W. and Henrywood, R.K. 1997 The dictionary of blue and white printed pottery 1780 – 1880. Volumes 1 & 2. Antique Collectors Club Crewe, V. 2012 ‘Ancient luxury and modern filth’ new insights into 19th century life at Sheffield Manor Lodge Post-Medieval Archaeology 46-2, 333-140 Crossley, D. 1990 Post-medieval archaeology in Britain Leicester University Press Cumberpatch, C.G. 1991 The production and circulation of Late Iron Age slip decorated pottery in Central Europe unpublished PhD Thesis, University of Sheffield Cumberpatch, C.G. 1997 Towards a phenomenological approach to the study of medieval pottery. In: C.G. Cumberpatch and P.W. Blinkhorn (Eds.) Not so much a pot, more a way of life. Oxbow Monograph 83. Oxbow Books. Cumberpatch, C.G. 2000 Medieval and post-medieval pottery in the Tolson Memorial Museum, Huddersfield: A rapid assessment with notes on pottery studies in West Yorkshire. Unpublished archive report for the West Yorkshire Archaeology Service. Cumberpatch, C.G. 2002 The pottery In: I. Roberts Pontefract Castle Archaeological Excavations 1982 –86. Yorkshire Archaeology 8. West Yorkshire Archaeology Service / English Heritage; 169-226. Cumberpatch, C.G. 2003 The transformation of tradition; the origins of the post-medieval ceramic tradition in Yorkshire. Assemblage 7 http://www.shef.ac.uk/ assem/issue7/cumberpatch.html

Allday, K and Millard, J. 2009 A newly discovered Leeds slipware pottery Transactions of the English Ceramic Circle vol. 20, part 3 Anderson, J. 1963 – 1965 Isaac Button – Country Potter Yorkshire Film Archive Andrews, K. 1997 From ceramic finishes to modes of production: Iron Age finewares from central France In: C.G. Cumberpatch and P.W. Blinkhorn (Eds.) Not so much a pot, more a way of life. Oxbow Monograph 83. Oxbow Books. ARCUS 2006 ARCUS 1008.1 Archaeological desk-based assessment, Darnall Road, Sheffield Unpublished archive report, ARCUS / University of Sheffield AS WYAS 2006 Broadway Centre, Bradford, West Yorkshire: Archaeological evaluation and excavation. Unpublished archive report by Archaeological Services WYAS Ashurst, D. 1987 Excavations at the 17th-18th century Glasshouse at Bolsterstone and the 18h century Bolsterstone Pothouse, Stocksbridge, Yorkshire. Postmedieval Archaeology 21; 147-226. Ashurst, D. 2002 Pothouse Farm (Bolsterstone Glasshouse) in: D. Saich (Ed.) Archaeology in South Yorkshire 1998-1999. South Yorkshire Archaeology Service. Ashurst, D. nd The history of South Yorkshire glass J.R. Collis Publications Barker, D. 1993 Slipware Shire Publications Barker, D. 2004 The industrialization of the Staffordshire potteries In: D. Barker and D. Cranstone (Eds.) The archaeology of industrialization The Society for PostMedieval Archaeology monograph 2. Maney Bartlett, K. 1971 Excavations at Potovens, near Wakefield 1968 Post-Medieval Archaeology 5; 1-33 Battye, K.M. 1999 Probate records as a source for the study of metal-working in Eckington 1534 – 1750 Derbyshire Archaeological Journal 119; 297-328 Battye, K.M. 2003 Probate records as a source for the study of the scythe makers and other metal workers of Norton 1533 – 1750. Transactions of the Hunter Archaeological Society 22; 49 – 69. BEHS 2007 36 The Boyle, Barwick-in-Elmet, West Yorkshire: Archaeological investigation and interpretation. Unpublished archive report by JB Archaeological Services on behalf of the Barwick-inElmet Historical Society. Belford, P. 2001 Work, space and power in an English industrial slum: ‘the Crofts’, Sheffield 1750 – 1850 In: A. Mayne and T. Murray (Eds) The archaeology of urban landscapes: Explorations in slumland New directions in archaeology Cambridge University Press Berg, M. 1999 New commodities, luxuries and their consumers in eighteenth century England In: M. Berg and H. Clifford (Eds.) Consumers and luxury: Consumer culture in Europe 1650 – 1850. Manchester University Press

95

Christopher Cumberpatch Cumberpatch, C. G. 2010a The manufacture of vernacular tablewares in Yorkshire: a preliminary discussion In: C. Casswell and P. Daniel ‘Stone Was the One Crop That never Failed’: The Archaeology of a TransPennine Pipeline. Excavations between Pannal and Nether Kellet 2006-2007 BAR British Series 526. Oxford. 125-8 Cumberpatch, C.G. 2010b Pottery from excavations at the Manor Lodge, Sheffield (1107b) Unpublished archive report for Sheffield University Cumberpatch, C.G. 2011a Pottery from excavations at Sheffield Manor Lodge (SML10) Unpublished archive report for Sheffield University Cumberpatch, C.G. 2011b Pottery from excavations in Trinity Walk, Wakefield (Sites 1124b and 1124c) Unpublished archive report for Wessex Archaeology Cumberpatch, C.G. 2012a Pottery from excavations at Sheffield Manor Lodge, 2011 (SML11) Unpublished archive report for Sheffield University Cumberpatch, C.G. 2012b Early modern pottery from the Flower Estate, Wincobank, Sheffield (ARCUS 753d) Unpublished archive report for Wessex Archaeology Cumberpatch, C.G. 2012c Early modern and recent pottery from excavations on the site of Sale Old Hall, Greater Manchester Unpublished archive report for Northern Archaeological Associates Cumberpatch, C.G. unpublished 1 Pottery from the T.C. Harrison site, London Road, Sheffield (Site 760a – d) Unpublished archive report for ARCUS (University of Sheffield) Cumberpatch, C.G. unpublished 2 Eighteenth and nineteenth century pottery from the site of the Suffolk Works, Suffolk Road, Sheffield Unpublished archive report for ARCUS (University of Sheffield) Dungworth, D and Cromwell, T. 2006 Glass and pottery manufacture at Silkstone, Yorkshire Post-Medieval Archaeology 40/1, 160 – 190. Elias, N. 1994 The civilising process Blackwell Edwards, D. 1994 Black Basalt: Wedgwood and contemporary manufacturers Antique Collectors Club Edwards, D and Hampson, R. 2005 White Salt Glazed Stoneware of the British Isles Antique Collectors Club English Heritage 1991 Management of Archaeological Projects (2nd Edition) English Heritage Franklin, B. 2005 The autobiography of Benjamin Franklin Digireads.com Publishing Godden, G. 1991 Encyclopaedia of British pottery and porcelain marks Barrie and Jenkins Godden, G. 2004 Godden’s Guide to English Blue and White Porcelain Antique Collectors Club Griffin, J.D. 2001 The Don Pottery 1801 – 1893. Doncaster Museum Service Griffin, J.D. 2005 The Leeds Pottery 1770 – 1881. Leeds Art Collections Fund Griffin, J. 2012 The Yorkshire Potteries Leeds Arts Collection Fund Hadley, D. and Harlan, D. 2011 Excavations at Sheffield Manor Lodge 1968 – 1980 The University of Sheffield

Cumberpatch, C.G. 2004a Pottery from excavations at Silkstone, Barnsley, South Yorkshire. English Heritage Centre for Archaeology Report number 50/2004. Cumberpatch, C.G. 2004b Medieval and post-medieval pottery production in the Rotherham area http://ads. ahds.ac.uk/catalogue/specColl/ceramics_eh_2003/ Cumberpatch, C.G. 2004c The pottery from excavations at Alderley Sandhills (ASP03): Archive report Unpublished archive report for University of Manchester / Manchester Museum Cumberpatch, C.G. 2005a Pottery from excavations in Sheffield; a review and assessment of the resource Paper delivered at the Theoretical Archaeology Group Conference, Sheffield 2005 http://independent. academia.edu/ChrisCumberpatch/Talks Cumberpatch, C.G. 2005b Pottery from excavations in Tenter Street, Sheffield (ARCUS 814a and 814b) Unpublished archive report for ARCUS (University of Sheffield) Cumberpatch, C.G. 2006 Pottery from Ash House Farm, Sheffield (ASH05) Unpublished report for Archaeological Services WYAS Cumberpatch, C.G. 2007a Pottery In: BEHS 2007 36 The Boyle, Barwick-in-Elmet, West Yorkshire: Archaeological investigation and interpretation. Unpublished archive report by JB Archaeological Services on behalf of the Barwick-in-Elmet Historical Society Cumberpatch, C.G. 2007b Pottery from excavations in Eleanor Street and Darnall Road, Sheffield (ARCUS 1008) Unpublished archive report for ARCUS (University of Sheffield) Cumberpatch, C.G. 2007c Pottery from excavations at Scholes Lodge Farm, Leeds (SLF05) Unpublished archive report for Archaeological Services WYAS. Cumberpatch, C.G. 2007d Pottery from excavations in Church Walk, Doncaster ‘Askews print shop’ DCW94 (with contributions by K. Hartley, R.S. Leary, A.G. Vince, D. Williams and J. Young) Unpublished archive report for Archaeological Services WYAS. Cumberpatch, C.G. 2007e Pottery from Garden Street, Sheffield (ARCUS 934c) Unpublished archive report for ARCUS / University of Sheffield Cumberpatch, C.G. 2007f Medieval and later pottery from excavations at the Broadway Centre, Bradford (BCE05) Unpublished archive report for Archaeological Services WYAS. Cumberpatch, C.G. 2007g Pottery from 117 Westgate, Wakefield (AOC Site 20240) Unpublished archive report for AOC Archaeology Cumberpatch, C.G. 2008a Pottery from excavations in Upper Allen Street (ARCUS 851b) Unpublished archive report for ARCUS / University of Sheffield Cumberpatch, C.G. 2008b Pottery from excavations in Blonk Street, Sheffield Unpublished archive report for Archaeological Services WYAS Cumberpatch, C.G. 2009 Pottery from excavations in Wakefield Westgate (BA1534 and BA12181) Unpublished report for Birmingham Archaeology

96

Tradition and change: the production and consumption of early modern pottery in South and West Yorkshire Harlan, D. 2010 Sheffield Museum Excavations 1971 http://www.sheffield.ac.uk/content/1/c6/10/51/56/ Sheffield%20Museum%20Excavations%201971.pdf Hey, D. 1969 A dual economy in South Yorkshire Agricultural History Review 17; 108 – 119. Hey, D. 1972 The rural metalworkers of the Sheffield region: A study of rural industry before the industrial revolution Department of English local history Occasional papers no. 5 Leicester University Press. Higgins, D. 2006 Clay pipes In: D. Dungworth and T. Cromwell 2006 Glass and pottery manufacture at Silkstone, Yorkshire Post-Medieval Archaeology 40/1, 182 – 186 Hurley, S. 2009 The production and consumption of traditional tablewares in 18th century Yorkshire unpublished MSc thesis, University of Sheffield Irving, A. unpublished Cistercian ware pottery in Yorkshire and the East Midlands Unpublished PhD thesis, University of Nottingham Johnson, M. 1996 An archaeology of capitalism Blackwell Kenworthy, J. 1928 The broken earthenware of Midhope Potteries. Privately published Kowaleski-Wallace, E. 1997 Consuming subjects: Women, shopping and business in the eighteenth century Columbia University Press Lawrence, H. 1974 Yorkshire pots and potteries David and Charles. Lightfoot, M.,McClusky, B and Cumberpatch, C. 2008 Archaeological excavations at Scholes Lodge Farm, Scholes, West Yorkshire Archaeological Services WYAS Publications 9. Lonton, T. 2006 William Gough and the Brown Moor Pottery 1739 – c.1777 In: Leeds Archaeological Fieldwork Society Lazencroft Excavation http://www. leedsarchaeology.co.uk/Book02.html Markus, S. nd What can excavation materials from 2010 tell us about 18th century ceramic production at Manor Lodge, Sheffield Unpublished BA Dissertation, University of Sheffield Milefanti, H.C. and Brears, P.C.D. 1971 Pottery kilns at Rawmarsh near Rotherham. Yorkshire Post-Medieval Archaeology 5; 193-195 Moorhouse, S. 1983 The medieval pottery In: P. Mayes and L. Butler (Eds.) Sandal Castle excavations 1964 – 1973 Wakefield Historical Publications. Moorhouse, S and Roberts, I, 1992 Wrenthorpe Potteries. Yorkshire Archaeology 2. West Yorkshire Archaeology Service. Wakefield. Peacock, D.P.S. 1982 Pottery in the Roman World: an ethnoarchaeological approach Longman Archaeology Series Porter, R. 1991 English Society in the 18th century Penguin Books Porter, R. 2000 Enlightenment: Britain and the creation of the modern world Allen Lane / Penguin

Richards, S. 1999 Eighteenth-century ceramics: Products for a civilised society Manchester University Press. Richards, T. 1990 The commodity culture in Victorian England: Advertising and spectacle 1851-1914 Stanford University Press Riden, P. 1984 History of Chesterfield volume II Part 1: Tudor and Stuart Chesterfield Borough of Chesterfield Smail, J. 1992 Manufacturer or artisan? The relationship between economic and cultural change in the early stages of the eighteenth century industrialization Journal of Social History 25 (4); 792-814. Spavold, J and Brown, S. 2005 Ticknall pots and potters Landmark Collectors Library Tarlow, S. 2007 The archaeology of improvement in Britain 1750 – 1850 Cambridge Studies in Archaeology Thirsk, J. 1984 The rural economy of England The Hambledon Press Van der Leeuw, S. 1976 Studies in the technology of ancient pottery PhD Dissertation, University of Amsterdam Van der Leeuw, S. 1984 Dust to Dust: A transformational view of the ceramic cycle Universiteit van Amsterdam Vickery, A. 2009 Behind closed doors: at home in Georgian England Yale University Press Walter, J. 1999 Brampton pots in the kitchen University of Derby Weatherill, L. 1996 Consumer behaviour and material culture in Britain 1660 – 1760 (second edition). Routedge. Wenger, E. nd Communities of practice; a brief introduction http://wenger-trayner.com/wpcontent/uploads/2012/01/06-Brief-introduction-tocommunities-of-practice.pdf White, H. 2007 Silkstone pothouse, South Yorkshire: Chemical and petrographic analysis of eighteenth century glazed wares English Heritage Research Department report series no. 60/2007. White, H. 2012 The problem of provenancing English post-medieval slipwares: a chemical and petrographic approach Post-Medieval Archaeology 46-1, 56-69 Wilson, P.R. 1995 Appendix 10 In: B.Vyner (Ed.) Moorland monuments: studies in the archaeology of North East Yorkshire in honour of Raymond Hayes and Don Spratt Council for British Archaeology Research Report 101 Woodrow, K.J. 1971 Cistercian ware from Silcoates School, near Wakefield Post-Medieval Archaeology 5, 185-188 Yentsch, A. 1991 The symbolic divisions of pottery: sex-related attributes of English and Anglo-American household pots In: R.H. McGuire and R. Paynter (Eds) The Archaeology of Inequality Blackwell

97

Not so Much a Pot, More an Expensive Luxury:  Commercial Archaeology and the Slow Death of Pottery Analysis Paul Blinkhorn Pottery is…the greatest resource of the archaeologist. For variety of form and texture, for decoration, for rapid change, for its quick fall into oblivion, and for its incomparable abundance, it is in every respect the most important material for study and it constitutes the essential alphabet of archaeology in every land. (Flinders Petrie, 1904, 15-6)

Introduction

Heritage. They were maintained by these units, and updated as necessary as new finds and knowledge came to light.

Since the beginnings of commercial archaeology in the late 1980s, the question of ‘standards’, ie the quality of the work carried out, has been a vexed one. Most archaeological excavation now takes place under the auspices of the commercial system, and there have been a number of papers published casting doubt of the usefulness and quality of work carried out, particularly in the area of pottery analysis (eg Blinkhorn and Cumberpatch 1999; Cumberpatch and Blinkhorn 2001). Until now, the major problem has always been measuring ‘quality’, a somewhat subjective concept. Much of the debate has, admittedly, been largely rhetorical and one-sided. Those who claim to maintain standards in archaeology have shown no inclination to make any sort of response to the points raised in the papers cited above, and, indeed, others on the same theme. This paper will present hard, data-supported evidence to show that in the last decade, despite much trumpeting from the self-appointed guardians of archaeological standards, the quality of pottery analysis in the commercial sector has declined, is continuing to do so at an increasing rate, and that a sizeable portion of pottery reporting by commercial archaeologists is in breach of standards that the contractors have been given accreditation for supposedly upholding.

The resulted of this was that in many areas of the country, pottery analysis became thoroughly standardized, with the resources available greatly enhancing the ability of anyone working in a region to be able to describe clearly and concisely the types of pottery found during archaeological work. The type-series are essentially the physical manifestation of decades of practitioners’ accumulated experience and knowledge, meaning anyone coming to an area for the first time is able to grasp, virtually instantly, a clear understanding of the ceramics of the area, and produce work which is completely relatable to that carried out by previous incumbents, and understandable by others working in the same area. To understand the importance of type-series, one only has to look at medieval pottery reports from the 1940s and 1950s. This is not meant to denigrate those who were working at that time and producing those reports; the discipline was then in its infancy, and they were the pioneers of it, effectively working ‘in the dark’. Reading them now, it is often difficult to know exactly what specific type of pottery they were describing, although in some cases the level of detail they used in their reporting is such that the experienced practitioner can identify exactly which pottery type they are describing.

Pottery Type-Series One of the basic tools of pottery analysis is the typeseries. The majority of these comprise a physical archive consisting of samples of all the different known pottery types in the region covered by the resource, with each type having their own individual name and/or code number, usually cross-referenced with a text file describing their petrological composition, chronology, geographical distribution, range of vessel forms, place of manufacture, etc. An excellent example is that for Worcestershire, which is available on-line (http://www.worcestershireceramics. org/), and is free to use. Most of those for the post-Roman period were put together in the last three decades of the 20th century, particularly in the 1970s, 1980s and 1990s, usually by ceramicists working within the old system of county-based ‘units’, and, in the main, funded by public money via city and/or county councils (eg. Vince 1985), development corporations in new towns such as Milton Keynes and Northampton (eg. McCarthy 1979), or English

For example, this description of a pot from a 1947 excavation in Oxford by Jope et al. (ibid. 1950, 47): ‘baggy tripod-pitcher in sandy, rather friable fabric with a yellowish glaze, with tubular spout, small side handle, main strap handle with twisted rope filling, somewhat irregular broadly spaced girth grooves and vertical applied strips, either wavy or finger-pressed, on the body,’ is recognizable to the modern ceramicist with a familiarity of the ceramics of the city as being Medieval Oxford Ware, Oxford type-series fabric OXY (Mellor 1994b), and dateable to the mid/late 11th – 12th century. Someone who has not worked with the pottery of Oxford would doubtless struggle to identify the vessel to which Jope et al refer, and describing each sherd from a large excavation in such a manner is not possible or indeed feasible. This is just one of the many reasons why type-series were created; they enable both brevity and clarity.

99

Paul Blinkhorn Pottery Analysis and Standards During the last 20 years, a number of guidelines for the analysis and publication of medieval pottery have been published. In 1994, the Medieval Pottery Research Group, the MPRG, published Maureen Mellor’s ‘Medieval Ceramic Studies in England: A Review for English Heritage’ (ibid.). It included guidelines to the amount of experience a practitioner requires, and stated that local and regional type-series should be established as a matter of priority, should be consulted by all those working in a region, and be curated, maintained and updated as necessary, and any updates published annually. As a result, mainly thanks to English Heritage and local authority funding, some areas commissioned and now have county type series, The MPRG then followed this up in 1998 with the ‘Guide to the Classification of Medieval Ceramic Forms (ibid.), a manual which standardizes the nomenclature of the different vessels which were produced by medieval potters, and in 2001 with ‘Minimum Standards for the Processing, Recording, Analysis and Publication of PostRoman Ceramics’ (ibid.). This dealt with every aspect of pottery work from its initial excavation though to archiving. The Roman and Prehistoric Pottery Groups have compiled similar standards for their own fields of study (Young 1980; PCRG 1997), and the three groups are currently working towards a combined set of standards (DH Brown, pers. comm.). Thus, anyone working in the field of pottery analysis has a clear and concise set of guidelines in which to work, with a standardized language, leaving no doubt as to what exactly is being discussed. Unfortunately, these standards, despite being created by the people most qualified to do so, and based on decadesworth of experience and expertise, are merely guidelines, and archaeologists are under no obligation to use them; it seems that, along with pottery type-series, an increasingly large proportion are not doing so, even though most field archaeologists are now supposedly working to high ethical and professional standards. The Institute for Archaeologists (IfA) was established in 1982, and claims, amongst other things, to ‘have achieved major progress in the advance of archaeology and allied disciplines by promoting professional standards and ethics’ (IfA 2009). However, when their standards document (http://www.archaeologists.net/codes/ifa) is examined, the reader would be hard-pressed to know, if they did not already, that archaeologists ever find pottery during fieldwork. The words ‘pottery’ or ‘ceramic’ do not appear anywhere in the IfA’s guidelines. Hardly surprising then that they do not insist that the guidelines and standards published by the various pottery research groups discussed above should be followed. Why the IfA, a body which claims to represent archaeology and archaeologists, chooses to completely ignore pottery research, and does not insist that their members adhere to standards and guidelines relating to it, produced by the people most qualified to do so, is beyond rational comprehension.

With regards to type-series, in the IfA Standards document “Standard and guidance for the collection, documentation, conservation and research of archaeological materials” it does state in section 3.7.4, that, during post-excavation analysis, ‘The use of regional reference collections is essential’. The same document also states in section 3.7.3 that ‘suitably qualified and experienced archaeologists should be involved in …undertaking finds recovery and analysis’ , although it does not anywhere state what it considers to be suitable qualifications and experience. The Medieval Pottery Research Group published a suggestion as to the required qualifications and experience appropriate in these areas as long ago as 1994 (Mellor 1994). The IfA Standards document makes no reference to it. Pottery Analysis in the Commercial World; A CaseStudy in Bad Practice The following report used as this case-study is from an excavation in Northamptonshire which was published in 2010 (Havard 2010). The county of Northamptonshire has had a ceramic type-series since 1994, and it is held by MOLA Northampton, formerly the Northamptonshire Archaeology Unit. The type-series is, at the time of writing, freely viewable by appointment, and at no charge. It should be stressed that the following report is not the only example of bad practice available, but it is one which demonstrates with some vividness the problems under discussion here. Towcester is a small but growing market town in Northamptonshire located on what was Roman Watling Street, and the archaeological resources include an Iron Age settlement, the Roman town of Lactodurum, an AngloSaxon burh fortified by Edward the Elder in 917, a Norman motte in the form of Bury Mount, and an important medieval market town, possibly with its own pottery manufactory. The archaeology is complex and regionally important. The excavation in question took place on the north-western edge of the town and revealed a series of boundaries which were probably part of a rectilinear field system. A total of around 50 sherds of pottery were recovered. The pottery report took the form of the table shown in Fig 1, and was summarized in the text of the main report as follows (ibid. 11): Sherds of medieval pottery were recovered …. and were mainly identifiable as shell-tempered and quartz/ limestone-tempered cooking-pot fabrics. This material suggests broad dating from the 12th to 14th centuries. In addition, a sherd from a glazed jug fabric was recorded …. and probably dates to the 13th or 14th centuries. Not only is the county type-series not used, but the terminology used to describe the medieval and postmedieval pottery is utterly nonsensical. There are no pottery types in the county which are ever referred to as ‘cooking pot fabric’. Perhaps most grating is the description

100

Fig. 1

Not so Much a Pot, More an Expensive Luxury

Fig 1. Pottery Report

Table X1: Data

an IfA-approved ceramicist, are a guess at the date by of a pottery type as ‘glazed jug fabric’, and there is also someone who has absolutely no knowledge whatsoever of a reference to ‘Post-medieval pottery: internally greenCounty No. sherds No. they Reports with no total the material are analysing and publishing. glazed fabric’, to which similarTotal comments apply. No. The Reports Beds 5720 77 2 report is essentially gibberish. Describing a pottery type Northants 1 and the Use of Post-Roman as ‘glazed jug fabric’ is no different to 3140 describing a stone 32 Commercial Archaeology Oxon 4913 or a coin as 139Pottery Type-Series18 axe as being made from ‘stone axe stone’, Total‘coin metal’. 13773 21 (8.5%) being manufactured from Would this be 248 Assessing the quality of archaeological work, especially considered acceptable for a flint or coin report? It would that of post-excavation is, as noted above, a generally appear it is for medieval pottery. subjective exercise, which in the past has meant that most critiques were based on rhetoric rather than data. This Writing as someone who has spent much of the last 25 Table X2: Use of county-type in ‘Grey Reports’ on thetoOASIS database study is intended rectify this, by examining the contents years working on the post-Roman pottery series of the county of archaeological reports, and seeing how many of them of Northamptonshire, and the creator of the county-typeY Nform,Total %Y Nthants Total with %Ythe Oxon Y to which N Totalclaim%Y guidelines they to adhere. seriesBeds in its present I have absolutely no ideaYwhichN comply 2004 10 pottery 2 types 12 are being 83.3 referred 2004 to 1in the1 2 50 2004 0 0 0 n/a of the known In 4this case, survey 1was carried out the author above report. clear than 2005 10 It is far 0 less 10 100the description, 2005 3 cited1 75 a 2005 0 1 by 100 using the site reports uploaded to the Archaeology Data above, published by Jope et al in 1950. The modern report 2006 5 2 7 71.4 2006 6 4 10 60 2006 4 1 5 80 Service’s ‘Grey Reports’ archive on the OASIS website is anonymous, but it was clearly written by someone with 2007 10 1 11 90.9 2007 1 3 4 25 2007 9 3 12 75 (http://archaeologydataservice.ac.uk/archives/view/ no knowledge or understanding whatsoever of medieval 2008 9 1 10 90 2008 1 1 2 50 2008 16 12 28 57.1 greylit/), and recording how many of them used the or post-medieval ceramics, and who, despite having local 17 post-Roman pottery county type2009 zero 8 0 8 regarding 100 the2009 1 100of the 2009 4 21 81 apparently knowledge material1 they0 conventions series. Other parameters were recorded, such has how were2010 analyzing, did not consult a freely available type1 2 3 33.3 2010 0 2 2 0 2010 18 8 26 69.2 many sherds of pottery were involved in each case, who series2011 located9 about1 12km 10 from the90site of 2011 the excavation. 1 1 2 50 2011 16 8 24 66.7 the contractor was, and if the analyst was named in the 2012 4 2 6 66.6 2012 2 3 5 40 2012 16 6 72.7 report. Reports from 2004 – 2012 were 22 consulted, and the The most disturbing aspect of this is that this report was not Total 77 Total 32 Total 139 counties involved were Bedfordshire, Northamptonshire produced by a ‘one-man-and-his-dog’ operation, but by a and Oxfordshire. These were chosen as they have all large commercial contractor Y = countyarchaeological type series used in reportwith a national have had county type-series since at least the 1990s, if not countyoftype seriesoffices, not usedand in one report profile,Na =number regional which holds earlier. All the reports consulted claimed that the work had IfA ‘Registered Organization’ status. Needless to say, the been carried out according to IfA guidelines, and all the report trumpets at the beginning that contractors involved are IfA Registered Organizations. In most cases, use of the county type-series was specified in “The fieldwork also followed the Standard and the brief prepared for the contractors. Guidance for Archaeological Field Evaluation issued by the Institute for Archaeologists” Bedfordshire The fieldwork may have followed those standards, but the The county-type series for Bedfordshire was started in the post-excavation work most certainly did not, unless, in the 1970’s, based on finds from excavations in the county town case of post-Roman pottery, those standards, and indeed (Baker and Hassall, 1977). More recently, the type-series the undefined suitable qualifications and experience for

101

Paul Blinkhorn was revised and is maintained by Albion Archaeology in Bedford, formerly the Bedfordshire County Archaeology Service, where it can be viewed by appointment

1) Recording shall be carried out in a manner compatible with existing typological series for the City of Oxford.

At the time of writing, all Briefs issued for the county stated that the county type-series had to be used for pottery of all periods. In the survey period, there were a total of 77 reports from sites which produced medieval or postmedieval pottery, with at least 5720 sherds of pottery. Two reports did not give the total number of sherds which were noted during the excavation (Table X1). Ten different contractors carried out the work, only one of which was based in the county.

2) Contractors should, where relevant, follow the guidelines for handling Post Roman Ceramics produced by the Medieval Pottery Research Group (Slowikowski, Nenk & Pearce, 2001). 

The data for the use and non-use of the county type-series in reports per year is shown in Table X2. Northamptonshire The Northamptonshire County type-series was created in the mid-1990s, based on that for the county town which was created by the now-defunct Northampton Development Corporation excavations in the 1970s and 1980s. It has since been maintained and updated by Northamptonshire Archaeology, where it can be viewed by appointment. It has been widely used since creation (eg. Blinkhorn 2010). All Briefs issued throughout the study period required the Fig.to1be used for post-Roman pottery reports. type-series In the survey period, there were a total of 32 reports from sites which produced medieval or post-medieval pottery, with at least 3140 sherds of pottery. One report did not give the total number of sherds which were noted during the excavation (Table X1). Eleven different contractors carried out the work, only one of which is based in the county. The data for the use and non-use of the county type-series in reports per year is shown in Table X2. Oxfordshire The county type-series was first created during excavations in the city of Oxford in the 1970’s, and has been widely used since then (eg. Mellor 1984; 1994). It is held and maintained by Oxford Archaeology, where it can be viewed by appointment. At the time of writing, Briefs for the city of Oxford stated that:

In the case of Oxfordshire county, which is controlled by a separate authority, the briefs ‘conform to the guidance provided by ALGAO and IfA.’, but do not specify that either the county type-series or the MPRG’s Standards must be adhered to. In the survey period, there were a total of 139 reports from sites which produced medieval or post-medieval pottery, with at least 4913 sherds of pottery. Eighteen reports did not give the total number of sherds which were noted during the excavation. Ten different contractors carried out the work, two of which are based in the county (Table X1). The data for the use and non-use of the county type-series in reports per year is shown in Table X2. Data In the survey period, there were a total of 248 reports from sites which produced medieval or post-medieval pottery, with the sites in question producing at least 13,773 sherds of pottery. Twenty-one reports (8.5%) of the total did not state the total number of sherds which occurred during the excavation (Table X1). The data in Table X2 shows the number of Grey Reports on the OASIS website for each county per year since 2004 which mention that medieval and/or post-medieval pottery occurred, and with the total number which used the county type-series (‘Y’) and the total number which did not (‘N’). The percentage which did use it for each year is also given (‘%Y’) The combined totals for three-year periods for the three counties are shown in table X3. The data in Table X3 is unequivocal evidence that, in the counties under consideration, the use of type-series, and hence the quality of reporting, has not only declined, but and also that the rate of decline is increasing. From 2004 to 2006, 78.4% of the reports used the conventions of the

Table X1: Data County Total No. sherds No. Reports No. Reports with no total Beds 5720 77 2 Northants 3140 32 1 Oxon 4913 139 18 Total 13773 248 21 (8.5%) Table X1: Data

Table X2: Use of county-type series in ‘Grey Reports’ on the OASIS database Beds

Y

N

Total

%Y

Nthants

Y

N

2004

10

2

12

83.3

2004

1

2005

10

0

10

100

2005

3

2006

5

2

7

71.4

2006

6

102

Total

%Y

Oxon

Y

N

Total

%Y

1

2

50

2004

0

0

0

n/a

1

4

75

2005

1

0

1

100

4

10

60

2006

4

1

5

80

Not so Much a Pot, More an Expensive Luxury Beds

Y

N

Total

%Y

Nthants

Y

N

Total

%Y

Oxon

Y

N

Total

%Y

2004

10

2

12

83.3

2004

1

1

2

50

2004

0

0

0

n/a

2005

10

0

10

100

2005

3

1

4

75

2005

1

0

1

100

2006

5

2

7

71.4

2006

6

4

10

60

2006

4

1

5

80

2007

10

1

11

90.9

2007

1

3

4

25

2007

9

3

12

75

2008

9

1

10

90

2008

1

1

2

50

2008

16

12

28

57.1

2009

8

0

8

100

2009

1

0

1

100

2009

17

4

21

81

2010

1

2

3

33.3

2010

0

2

2

0

2010

18

8

26

69.2

2011

9

1

10

90

2011

1

1

2

50

2011

16

8

24

66.7

2012

4

2

6

66.6

2012

2

3

5

40

2012

16

6

22

72.7

Total

77

Total

32

Total

139

Y = county type series used in report N = county type series not used in report

Table X3:Table Use X2: of county-type in ‘Grey Reports’, totals Use of countyseries -type series in ‘Grey Reportscombined ’ on the OASIS database %Y Y N Total 2004 78.6 11 3 14 2005 93.3 14 1 15 2006 68.2 15 7 22 Average, 2004-6 78.4 40 11 51 2007 74 20 7 27 Table X3: Use of county-type 2008 series in ‘Grey Reports’, combined totals 65 26 14 40 2009 86.6 26 4 30 %Y Y N Total Average, 2007-9 74.2 72 25 97 2004 78.6 11 3 14 2010 61.3 19 12 31 2005 93.3 14 1 15 2011 72.2 26 10 36 2006 68.2 15 7 22 2012 66.6 22 11 33 Average, 2004-6 78.4 40 11 51 Average, 2010-12 67.0 67 33 100 2007 74 20 7 27 Total 179 72 248 2008 65 26 14 40 2009 -type series 86.6in ‘G26 Table X3: Use of county rey R4 eports30 ’, combined totals Average, 2007-9 74.2 72 25 97 post-Roman pottery type-series for the counties. From61.3 is another indication 2010 19 12 31 that the quality of work has declined 2007 to 2009, it was 74.2%, and from 20102011 to 2012 it was72.2 over26the 10 period36under examination, and by the period 67.0%. Table The useX4: of type-series in grey reports series has declined 2010-2012, incombined the three counties examined here, barely Use of county-type in ‘Grey totals, from reports 2012 66.6 Reports’, 22 11 33 from nearly four-fifths to barely two-thirds over a nine half the contractors working outside their home counties produced by contractorsAverage, not based in the 67.0 county67 in which 100 they are working 2010-12 year period. used the 33 local pottery type-series in the production of Totaltheir 179 72 Lack 248 local expertise amongst contractors reports. %Y All %Y ‘Outsiders’ Y N of Total The data in Table X4 is also worthy of comment. It shows working outside the local area is something that has been 13 8 21 that, generally, when commercial2004-6 contractors78.4 are working 61.9 highlighted as a critique of commercial archaeology in the 2007-9 74.2 62.5 15 9 24 in a county outside that in which their main office is past (eg. Cumberpatch and Blinkhorn 1999). This data 2010-12 67.0 22 is 45 located, they are less likely to use the pottery type-series 51.1 shows that23 not only it is a view which is correct, but also of the county in which they were working 72.2 than are the 56.7 Average 39culprits 90 are making no attempt to rectify that many51 of the Tablearchaeologists. X4: Use of county-type series in ‘Greythe Reports’, reports locally-based Thus, the local contractors situation,combined and that the totals, situationfrom is getting worse. produced by contractors not based in the county in which they are working are likely to produce a report to a higher standard. This 2004-6 2007-9 2010-12 Average

%Y All 78.4 74.2 67.0 72.2

%Y ‘Outsiders’ 61.9 62.5 51.1 56.7

Y 13 15 23 51

N 8 9 22 39

Total 21 24 45 90

Table X4: Use of county-type series in ‘Grey Reports’, combined totals, from reports produced by contractors not based in the county in which they are working

103

Paul Blinkhorn Discussion

Bibliography

It is often argued that it is the responsibility of the local curators to ensure that a brief is adhered to and that the county-type series is used in pottery reports, but in the real world, this is largely unfeasible. As one curator told me, off the record, “if I attempted to refuse to “sign off” a report and stop a development because the correct codes had not been used in the pottery report, the council planning department and the developer’s lawyers would come down on me like a ton of bricks, and my post would probably be scrapped in the next round of council cuts”. At a time when we have a government obsessed with reducing ‘planning red tape’, this is all too believable, and the archaeological curators position is, in many counties, simply too weak to enable them to enforce anything other than the most basic of standards. The IfA does not have this excuse.

Baker, E and Hassall, E, 1979 The Pottery in D Baker, E Baker, J Hassall and A Simco Excavations in Bedford 1967-1977 Bedfordshire Archaeological Journal 13, 147 – 239 Blinkhorn, P, 2010 The Post-Roman Pottery in A Chapman West Cotton: A Study in Settlement Dynamics. Excavations at West Cotton, Raunds, Northamptonshire, 1985-9 Oxbow, Oxford, 259-333 Blinkhorn, P, 2003 The Pottery – Conventual Buildings in M Rylatt and P Mason The Archaeology of the Medieval Cathedral and Priory of St. Mary, Coventry City Development Directorate Coventry County Council, 96 - 107 Blinkhorn, PW and Cumberpatch, CG, 1999 Archaeology in England, 1999 World Archaeological Bulletin 9; 45-55 Cumberpatch, CG, and Blinkhorn, PW, 2001 Clients, Contractor, Curators and Archaeology: who Owns the Past? In M Pluciennick (ed.) The Responsibilities of Archaeologists Oxford Britsh Archaeolgicla Reports International Series 981 Cumberpatch, CG and Blinkhorn, P (eds) 1997 Not So Much a Pot, More a Way of Life Oxbow Monog 83 Flinders Petrie, WM 1904 Methods and Aims in Archaeology London. MacMillan Havard, T (2010) Belle Baulk, Towcester, Northamptonshire: Archaeological Evaluation. Cotswold Archaeology Rep 10100 IfA 2009 The Institute for Archaeologists. a professional institute for the study and care of the historic environment IfA Jope, EM, Jope HM, and Rigold, SE, 1950 Pottery from a late 12th Century Well-filling and other Medieval Finds from St. John’s College, Oxford, 1947 Oxoniensia 15, 44-62 McCarthy, M, 1979 The Pottery in JH Williams St Peter’s St, Northampton. Excavations 1973-76 Northampton Development Corporation Monog Ser 2, 151-242 Mellor, M, 1994 Medieval Ceramic Studies in England: A Review for English Heritage English Heritage and MPRG Mellor, M, 1994b Oxford Pottery: A Synthesis of middle and late Saxon, medieval and early post-medieval pottery in the Oxford Region Oxoniensia 59, 17-217 MPRG, 1998 Guide to the Classification of Medieval Ceramic Forms MPRG Occasional Paper 1 MPRG, 2001 Minimum Standards for the Processing, Recording, Analysis and Publication of post-roman Ceramics Medieval Pottery Res Group Occ Paper 2 PCRG, 1997 The Study of Prehistoric Pottery: general policies and guidelines for analysis and publication PCRG Occasional Papers 1 and 2 Vince, AG, 1985 The Saxon and Medieval Pottery of London: A review Medieval Archaeology 29, 25-93 Young, CJ, 1980 Guidelines for the Processing and Publication of Roman Pottery from Excavations. Directorate of Ancient Monuments and Historic Buildings Occasional Papers 4.

The data compiled above shows that since 2004, on average, over one-quarter of the grey reports for the study area did not use pottery type-series, despite the fact that all the work was carried out by IfA-registered organizations, and the use of reference collections is mandatory according to the guidelines of the IfA. For units working ‘outside their area’, nearly half now do not use the typeseries of the county in which the site is located. All these contractors are IfA approved, and all claim to adhere to their guidelines, which, to repeat, include the statement in their “Standard and guidance for the collection, documentation, conservation and research of archaeological materials” section 3.7.4, that, during post-excavation analysis, ‘The use of regional reference collections is essential’. These guidelines are not only not being adhered to, but, under the IfA’s watch, the situation is getting worse, and the selfappointed upholders of archaeological standards are doing nothing about it, despite the evidence of it being freely available to them via the OASIS website. The MPRG and other pottery research groups have extremely highquality sets of standards and guidelines which have been produced over the last 20 years, yet the IfA refuse to insist that their members use them, despite them having none of their own of any worth, and this at a time when the IfA has just obtained chartered status. In the light of the data presented here, one has to ask what can they offer, other than rhetoric, to show that their claim that they are upholding and improving standards in archaeology has any credibility, as the evidence presented in this paper indicates that, clearly, they are not.

104

The Organisation of Late Bronze Age–Early Iron Age Society in the Peak District National Park Kevin Cootes Background The late 2nd/early 1st millennium BC is perhaps the most enigmatic period of late prehistoric Britain. This is a time of radical upheaval in all aspects of life and death. The impressive burial mounds of the Early Bronze Age cease to be used, ceramic form and decoration become simpler, and settlement patterns change, with a new focus revolving around the great hillforts (Champion 2002:95). The traditional consensus is that these changes were at least partly the result of climatic deterioration (Hicks 1971). Fundamentally, there is still much debate regarding the organisation of prehistoric society during this period. The inter-related topics of landscape utilisation, movement, and the interactions of the prehistoric groups who created them, in both local and trans-regional contexts, is a recurring focus (Barnatt 1987a, 1987b; Fleming 2008). Within the context of societal organisation, the Uplands of Britain offer an exceptional opportunity for the study of late prehistoric society. Such topographically isolated regions have undergone little modern development in comparison to their lower altitude counterparts, often preserving extensive areas of agricultural and ritual remains. The Peak District National Park is one such region, covering an area of 555 sq miles at the southern limits of the Pennines (Bevan 2007), as shown in Figure 1. This area is notable for containing an extensive record of Late BronzeEarly Iron Age land use, with remains characterised by cairnfields, field boundaries, hillforts, lowland settlements and associated ceremonial monuments. Geology, Landscape Use and Archaeological Survival The current boundary of the Peak District National Park is not an arbitrary line, but is governed to the north, east and west by clear geological limits, which encompass a heterogeneous and well studied Carboniferous geology, illustrated in Figure 2. Dominating the centre of the region is a Lower Carboniferous Limestone Plateau commonly referred to as the White Peak. A series of volcanic outcrops comprising exposures of basaltic lavas, basic tuffs, agglomerates, intrusive sills and dolerite dykes are intermittently exposed across the surface, especially in the north, south-east and south of the region (Aitkenhead 2002:34; Cope 1998:7-43; Fearnsides et al 1932:34-38). The majority of these exposures comprise olivine rich basalts and dolerites with high iron content, the latter often resulting in extensive chemical weathering of surface exposures (Simpson 1982:9; Stevenson and Gaunt 1971:120). The present landscape is characterised by rolling hills cut by steep dales, with the region being the

traditional focus for farming and settlement (Cope 1998:743) from the Early Neolithic onwards. This has resulted in the destruction of the majority of evidence for prehistoric land use, with the notable exception of Neolithic and Bronze Age burial mounds, monuments, and hillforts, mostly spread across the higher ground of the region (Barnatt 1990:13). Surrounding the White Peak to the north, west, and east, is the higher country of the Gritstone Uplands. Commonly referred to as the Dark Peak, this area formed during the Upper Carboniferous, and is characterised by a series of interlocking sandstones and shales, reaching a maximum height of 600 mOD in the north at Kinder Scout. These uplands divide into two flanks with their southern extremities near Derby and the Potteries, namely the Eastern Moors to the east, and the Westphalian Coal Measures to the west (Aitkenhead and Waters 2002:37; Cope 1998:39-74; Fearnsides et al 1932:6). The current landscape is dominated by peat deposits, heather moorland, and coarse grasses (Barnatt and Smith 1997:1114; Bevan 2007:11). There is a more complete picture of late prehistoric land use on the Dark Peak than its White Peak counterpart, especially on the Eastern Moors above Sheffield. It is within this region of sheltered plateaus that a combination of grazing and the use of the area during the 19th century for grouse shooting have preserved a series of extensive prehistoric field systems. Dating evidence has broadly indicated activity and settlement during the 2nd and 1st millennia BC (Barnatt 1990:22; Barnatt and Smith 1997:11; Bevan 2007:16). The White and Dark Peaks are separated by a series of dales, fissures, and river valleys created by erosion of the softer shale beds (Cope 1998:46; Smith et al 1967:228). The River Wye, the River Derwent, and its tributaries the Alport and Noe form the east and north boundaries, with the Rivers Dove and Manifold to the west (Cope 1998:46; Smith et al 1967:228). Theoretical Background Archaeological research into the nature of late prehistoric society in the Peak District has a long and distinguished history, with often conflicting interpretations into the nature of human activity which took place in the region. Traditionally, researchers between the 1960s and 1980s, for instance Hicks (1971), Hawke-Smith (1979, 1981), Vine (1982) and Bradley and Hart (1983), saw the Dark Peak and its inhabitants as being of poorer status to their White Peak counterparts. The latter were envisioned as inhabiting the most favourable agricultural land.

105

Kevin Cootes

Figure 1 – Location of the Peak District National Park within the British Isles, shown in red (© Crown rightD2012. An Ordnance Survey/EDINA supplied service) Figure 1.Copyright/database Location of the Peak istrict National Park within the British Isles , shown in red (© Crown Copyright/database right 2012. An Ordnance Survey/EDINA supplied service)

106

The Organisation of Late Bronze Age–Early Iron Age Society in the Peak District National Park

Figure mapofofthe thePPeak (afterBBarnatt andSSmith 1997:13) Figure22.- Geological Geological map eak DDistrict istrict (after arnatt and mith 1997:13)

107

Kevin Cootes The debate over the nature of habitation, human interaction, and the status of populations has continued into the 21st century, with two opposing models dominating theoretical research. John Barnatt (1987a, 1987b, 1990, 1999; Barnatt and Collis 1996; Barnatt and Smith 1997) has produced an extensive theoretic model supported by fieldwork, postulating the populations of both landscape zones were of equal status from the Early Bronze Age onwards. Barnatt suggests, as others had before him, that the White Peak was central for habitation during the Early Bronze Age, but proposes a fundamental change in the Late Bronze Age. In his model, what followed was a decline in the White Peak as the focus of settlement (Barnatt 1999:33-49; Barnatt and Smith 1997:16). The region was now dominated by the rise of the defended hillforts, which overlooked the shale valleys to take advantage of complimentary resources. The large hillforts were accompanied by a series of smaller enclosures dispersed through the landscape, complimented by undefended farmsteads, which effectively divided the Peak District into territories (Barnatt 2000:2, 1999:31-49). Barnatt supported his theory of a divided Peak District by postulating a minimum of 16 separate sub-populations on the Eastern Moors, with an overall regional identity. He observed that viable agricultural land was sometimes ignored, being used to separate field systems, and hence different groups (Barnatt 2000:64-78). John Barnatt’s interpretations have been directly challenged by the research of Willy Kitchen (2000). He presented the dataset within a model of wide-scale mobility and multiple land-use strategies for the Early-Middle Bronze Age, rejecting any suggestion of a White Peak Core. He instead postulated that people were travelling across a far greater area than Barnatt envisioned, with the extensive field remains on the Eastern Moors not requiring permanent occupation. Instead, he proposed intermittent development of fields through a regime of shifting forest or bush fallow cultivation. He suggested that as large areas were undeveloped, there was no premium placed on good soils. Instead, a flexible approach to animal or crop husbandry under mobile land use strategies was practiced (Kitchen 2000:217-313). Kitchen envisioned a fundamental restructuring of societal organisation during the Late Bronze Age, the biggest indication being through the establishment of the hillforts. This change was thought to be the result of climatic deterioration and the maintenance of pasture in a transhumant society. This large scale territorial land division had the effect of creating a sedentary society, with areas around cairnfields used for uncontrolled grazing. Kitchen recommended, however, that the Peak District should not be viewed as a discrete entity, but incorporated into a wider area encompassing neighbouring regions (Kitchen 2000:310 -319). The models which have been constructed to explain the nature of the settlement record in the Peak District have approached this topic from a wide variety of research avenues. There has often been a strong emphasis on certain aspects of the archaeological record, for instance environmental

analysis, upstanding structures, settlement distributions, burial patterns, spatial relationships, and proximity to natural resources (Barnatt and Collis 1996; Champion 2002; HawkeSmith 1979). When material culture has been investigated, it has often been in the context of geographical distribution and/or stylistic traits. In contrast, a widespread compositional study of ceramic paste recipes has yet to be attempted. Aims, Methodology, and Sampling Strategy With the lack of a widespread compositional investigation of ceramic paste recipes in the Peak District, the case study presented in this chapter utilised ceramic petrography in the investigation of six assemblages from sites spread across both main landscape zones. Such an approach is ideal for the region on account of the well studied, heterogeneous geology. This presents the opportunity to identify patterns in raw material procurement which can be used to reconstruct the movements of people through the pottery they created. With such an approach in mind, ceramic petrography has long been recognised as a reliable form of compositional analysis, enabling the identification of tempering agents and naturally occurring inclusions in prehistoric pottery (Woodward 2008; Hodges 1962). This analysis was conducted using an adapted version of Whitbread’s (1995) classification. Ceramic petrography has been utilised for the attainment of three interrelated aims which have been established as central issues in Peak District prehistoric studies. The first of these is to investigate the relationship between settlements across the two main landscape zones, namely the White and Dark Peaks. The second is to identify relationships between the large hillforts, smaller defended enclosures, and open farming areas. The final aim is to identify if evidence exists which could be used to postulate a definable Peak District Society during later prehistory. The dataset comprised 194 sherds from six archaeological sites, representing the total number of Late Bronze-Early Iron Age excavations to have produced substantial ceramic assemblages, as shown in Table 1 and Figure 3. The sampling strategy satisfied some of the recommendations of Prudence Rice (Rice 2005:321-324), covering the full range of domestic upland and lowland habitations in the south and north of the region. In contrast, the limited repertoire of forms, mixed nature of deposits, and poor recording of early 20th century excavations, have made accurate estimates of vessel number per site problematic. It has therefore been necessary to assume that each sherd represents a single pot, although this is probably not always the case. At Ball Cross, Harborough Rocks, Fin Cop, Gardom’s Edge and Roystone Grange, the assemblages were dominated by undecorated body fragments. In contrast, the 50 sherds sampled from the hillfort of Mam Tor originated from a series of habitations which have enabled a more refined selection representing individual pots. This assemblage demonstrated the full complement of vessels for this period, with examples illustrated in Figure 4. Sampling of ceramics within individual sites was not even, but instead governed by the limited range of forms which 108

Figure 9 - Angular, fresh basic igneous rock temper in ceramics from A - Sample B3 and B -Sample B4 from Ball Cross. All micrographs taken in XP with a field of view of 3.8mm.

The Organisation of Late Bronze Age–Early Iron Age Society in the Peak District National Park Site name

Region

Period

Site Type

Number of sherds

Ball Cross

White Peak

LBA-EIA

20

Fin Cop

White Peak

LBA-EIA

Defended hilltop settlement Hillfort

Gardom’s Edge

Gritstone Uplands White Peak

LBA-EIA

Gritstone Uplands White Peak

LBA-EIA

Harborough Rocks Mam Tor Roystone Grange

LBA-EIA

LBA-EIA

12

Farming settlement Hillside settlement Hillfort

96

Undefended settlement

1

Total sherd count

7 50

194 Table 1 – Late Bronze-Early Iron Age sites

Table 1. Late Bronze-Early Iron Age sites

Figure 3. Distribution of Late Bronze-Early Iron Age sites sampled.

(© C3rown Copyright right 2012. n Ordnance Survey supplied service) Figure – Distribution of/database Late Bronze-Early IronAAge sites sampled. (©/EDINA Crown Copyright/database right 2012. An Ordnance Survey/EDINA supplied service) 109

Kevin Cootes

Figure 4 – Rim sherds from Mam Tor, Hut Platform 4, with the lower example containing a line of Figure 4. Rim sherds from Mam Tor, H ut Platform 4, firing with the lower and example containing a line of holes punched holes punched through vessel wall after (Coombs Thompson 1979:36, 39) through vessel wall after firing (Coombs and Thompson 1979:36, 39) dominate the Late Bronze-Early Iron Age. These consisted of a series of simpler and undecorated domestic bucket and barrel shaped urns, and tall, closed form jars which defy firm classification. These vessels are commonly identified as ‘Flat-Rimmed Ware’ (Gibson 2002:109-115, 1997:72), although this is known to be an inaccurate label. Regional variations account for the dominance of bucket and barrel shaped vessels in the Midlands and northern England, with jars and shouldered pots occurring to a lesser extent (Gibson 2002:114-115). Such vessels have a wide and poorly defined chronology, a factor which is reflected in the limited dating of the majority of sites in this study. The petrographic analysis of ceramic paste recipes was complemented by a large-scale sampling strategy, employed to enable the identification of raw materials. Clay sampling was mostly restricted to the general vicinity of each site. In contrast, hard rock sampling was determined by the main tempering agents employed. Geological maps were used to identify significant areas of drift geology, in addition to exposed sources. Where practical, several samples were taken from each identified source to establish localised variation, using a hand-held GPS to record the position. To replicate the conditions of late prehistoric pottery, briquettes were fired at 800oc. Petrographic Analysis: Paste Recipes Analysis of the ceramic assemblages from the Peak District produced striking and unexpected results, manifest in the preferred use of a distinct paste recipe across both landscape zones. Although nine separate fabrics and five sub-fabrics were identified (as illustrated in Table 2) 178 of the 194 sherds (92%) contained basic igneous rocks of basalt or dolerite in non-calcareous clay.

Fabric 1 dominates five of the six sites, being characterised by the addition of basic igneous rocks of predominantly olivine rich basalt and dolerite temper in non-calcareous clay, as shown in Figure 5. The majority of basic igneous inclusions have suffered from varying degrees of weathering. Vesicular characteristics occur in some examples. Considerable variation was observed between sherds, in a pattern which mostly reflected the site of origin, indicating the use of multiple raw material sources. Characteristically, a wide range of minor inclusions were observed in some but not all examples, for instance angular quartz, quartz rich sandstone, shale, and weathered and un-weathered chalcedonic, iron rich and radiolarian chert. Differences in the clay matrices and fine fraction support this trend, although there is no easy sub-division of samples. The four sub-fabrics for Fabric 1 were defined by variations such as rhyolite from a single sample at Gardom’s Edge, or the absence of quartz inclusions in the fine fraction of five sherds from the defended enclosure of Ball Cross. Sub-Fabric 1C, from Mam Tor was characterised by large inclusions of ironstone. Fabric 1D comprised 7 sherds from Gardom’s Edge, Harborough Rocks and Roystone Grange, separated by the sub-angular shape of the quartz inclusions. Fabric 2 and its single sub-fabric were limited to the hillfort of Fin Cop, being characterised by un-weathered, homogeneous augite rich dolerite and rounded quartz in non-calcareous clay. The sub-fabric is only differentiated by the presence of acid igneous inclusions of probable rhyolite, possibly representing natural variation. Fabric 3 comprised inclusions of dolerite and grog temper in non-calcareous clay. Grog inclusions were variable in colour and shape between the two sites where they

110

The Organisation of Late Bronze Age–Early Iron Age Society in the Peak District National Park Fabric 1

2 3 4 5 6 7 8

Tempering Agent Basic Igneous rock

Number of Sherds 162

Overall Percentage 83.50%

12

Sites represented with vessel numbers Ball Cross (13), Gardom's Edge (90), Harborough Rocks (2), Mam Tor (26), Roystone Grange (1) Fin Cop (12)

Basic Igneous rock and sand Basic Igneous rock and grog Polycrystalline Quartz Course Sandstone Weathered Acid Igneous rock Rounded Quartz Acid Igneous rock

3

Ball Cross (1) Mam Tor (2)

1.50%

3

Ball Cross (3)

1.50%

5

Mam Tor (5)

2.60%

6

Ball Cross (1), Mam Tor (5)

3.10%

2

Roystone Grange (2)

1%

1

Gardom's Edge (1)

0.50%

6%

TTable able 2. Fabrics 2 –Petrographic Petrographic Fabrics

occurred, although all inclusions contained basic igneous fragments. Polycrystalline quartz and textural features occur, with sandstone and chert occurring rarely.

Raw Material Samples

Fabric 4 was restricted to Ball Cross, comprising a homogenous group characterised by polycrystalline quartz temper in non-calcareous clay, in addition to argillaceous inclusions and clay pellets. Chert, nepheline, quartz arrenite, granite and shale also occur. Similarity in inclusions and voids indicated a common source.

Ball Cross, Harborough Rocks, and Roystone Grange failed to produce viable samples, but clays were obtained from the environs of Fin Cop, Mam Tor and within the settlement of Gardom’s Edge. Other samples were taken from glacial areas, and weathered basic igneous sources at Cave Dale near Castleton and Ible, in the north and south of the White Peak respectively.

Fabric 5 contained coarse fragments of sub-arkrose in noncalcareous clay. Shale, feldspars, phyllite, basic igneous rock, polycrystalline quartz, micrite and chert also occur, with this fabric being restricted to Mam Tor. Fabric 6, originating from Ball Cross and Mam Tor, contained inclusions of weathered igneous rock in noncalcareous clay. Variability was demonstrated within and between inclusions, being mostly silica abundant but also iron rich. Polycrystalline quartz, shale, chert, basic igneous rock, sanadine and rhyolite also occurred. Fabric 7 contained two samples from Roystone Grange, characterised by sub-angular to rounded quartz inclusions in non-calcareous clay. Chert, feldspar, polycrystalline quartz, basic igneous rock and sub-arkrose also occur. This fabric bore striking similarities to the rounded quartz inclusions from other samples from Roystone Grange in Fabric 1D. Fabric 8 comprised the only loner in this study, originating from Gardom’s Edge. The sherd was characterised by fragments of weathered acid igneous rock of probable granite in non-calcareous clay.

Clays

Four of the clay samples produced compositional matches with ceramics in a pattern which was generally site specific. All sherds from Fin Cop produced strong petrographic matches with fluvial clay from the Wye Valley beneath the hillfort. The matrices were similar in the colour, size, distribution, and shape of the quartz inclusions in the coarse and fine fractions, as shown in Figure 6. The presence of weathered acid igneous rocks in the clay of the sub-group could be due to natural variation. The settlement of Gardom’s Edge on the Eastern Moors of the Dark Peak produced strong compositional similarities between raw material samples and ceramics. Six separate samples were taken by gauge core from a primary clay source which crossed the site. Colour banding observed in these samples was paralleled with nine ceramics from Gardom’s Edge, with examples illustrated in Figure 7. Argillaceous inclusions found in the primary clay were paralleled in 37 vessels. Weathered acid igneous rocks present in one clay sample were paralleled in the single sherd from Fabric 8. Additionally, four vessels produced strong comparative matches with degraded basic igneous

111

Kevin Cootes

A

B

C

D

E

F

Figure 5 - Basic Igneous temper. –C Ball Cross. Gardom’s C am – Mam Darborough – Harborough Figure 5. Basic Igneous temper . A – BA all ross . B- GBardom ’s EdgeEdge. .C–M Tor. Tor. D–H Rocks. ERocks. G E-range Roystone Cop. All micrographs infield XP with a field of mm. Roystone . F – FGrange. abric 2,FF–inFabric Cop. A2,llFin micrographs were taken were in XPtaken with a of view of 3.8 view of 3.8 mm.

A

B

6 - Comparison clay Sample from theVWye beneath Fin Cop, illustrated A, FigureFigure 6. Comparison betweenbetween clay Sample from the Wye alleyValley beneath Fin Cop , illustrated in A,in compared with the matrices from abric 2, from illustrated All micrographs taken in XP taken with ainfield of view of 3.8 compared with theFmatrices Fabric in 2, B. illustrated in B. All micrographs XP with a field of mm. view of 3.8 mm.

A

112

B

Figure 6 - Comparison between clay Sample from the Wye Valley beneath Fin Cop, illustrated in A, compared with the matrices from Fabric 2, illustrated in B. All micrographs taken in XP with a field of The O rganisation of Late Bronze Age–Early Iron Age Society in the Peak District National Park view of 3.8 mm.

A

B

A

B

Figure 7. Examples of similarities between the clay samples from Gardom’s Edge and the ceramics from the site. A Figure 7 - Examples of similarities between the clay samples from Gardom’s Edge and the ceramics RMS15S5. C - Sample D - G25.CA-ll micrographs taken in XP with a field of view of 3.8mm. from B the- RMS15S7. site. A - RMS15S5. B -G8. RMS15S7. Sample G8. D - G25. All micrographs taken in XP

with a field of view of 3.8mm.

clays taken from a columnar basalt outcrop in Cave Dave through the landscape. A total of 24 separate exposures in Castleton, a distance of c.17km from the site. were sampled which represented this distribution. Four Figure 6 - Comparison between clay Sample from the Wye Valley beneath Fin with Cop,the illustrated A, a distinct produced positive matches temper, in with compared theexposures matrices from Fabric illustratedpattern in in XP with a field of Clay samples from around the2,hillfort in micrographs raw material taken procurement emerging. A takenwith B B. All view of 3.8 mm. of Mam Tor demonstrated variability which was directly reflected in the ceramics from this site. Several samples Five basic igneous samples taken from the disused quarry taken from fluvial channels were characterised by naturally of Tideswell produced strong petrographic similarities occurring inclusions of monoand polycrystalline with A B the ceramics of Ball Cross, an area approximately quartz, coarse sandstone, and siltstone. These samples 7km north-west of the site. The randomly orientated were broadly comparable to several vessels, as shown feldspars encompassed within augite crystals, and in Figure 8. Other clays were taken from exposed shale weathered brown colour of the olivines was common to beds, producing matches with several sherds from the both raw material source and the temper in six separate site, indicating that similar, local sources were utilised. vessels from the hilltop settlement. Similarly, marked Additionally, the matrices of four vessels demonstrated petrographic similarities were noted between the temper marked petrographic similarities with degraded basic in all 12 ceramics from the hillfort of Fin Cop, and the igneous clays from the columnar basalt outcrop at Cave prominent basic igneous outcrop of Black Rock Corner, Figure 8of– less A – than Clay2km. sample from fluvial channel below Mam Tor in 1km A, demonstrating similarities to a vesicular Dale, a distance approximately from the site. The olivine rich vessel from the site illustrated in B. All micrographs taken in XP with a field of view of 3.8mm. basalt, amygdaloidal features and high iron content were a Basic Igneous Rock Samples distinct combination. The overwhelming igneous tempered possible temper sources for ceramics the southern Peak Figure 7 -dominance Examples of of basic similarities between the clayAnalysis samplesof from Gardom’s Edge and the ceramics dictated the sampling strategy regarding possible District sites of Harborough Rocks and from the site. A - RMS15S5. B - RMS15S7. C - Sample G8. D - G25. All micrographs takenRoystone in XP Grange temper sources. Such rocks are limited to,with but awidely produced similar results to Ball Cross and Fin Cop. field of view of 3.8mm. distributed across, the White Peak, exposed sporadically The ceramics from both sites contained basic igneous

A

B

– Asample – Clayfrom sample fromchannel fluvial channel below Mam Tordemonstrating in A, demonstrating similarities to a from Figure Figure 8. A – C8lay fluvial below M am Tor in A, similarities to a vessel vessel from the site illustrated in B. All micrographs taken in XP with a field of view of 3.8mm. the site illustrated in B. All micrographs taken in XP with a field of view of 3.8mm

113

Kevin Cootes temper which was homogeneous; being characterised by weathered olivines surrounded by augite crystals and chlorite filled channels. There were strong petrographic comparisons between the temper in the ceramics with locally available resources in the area of Ible, a distance of less than 5km from both sites. Ceramics from Mam Tor produced comparative petrographic matches with several possible sources, demonstrating a wider raw material procurement catchment area than other sites. Two samples were taken from the basaltic lava flow at Cave Dale above Castleton, less than 2km from the hillfort. The basic igneous temper from four vessels also produced strong comparisons to the basic igneous outcrop at Black Rock Corner, whilst the temper from six vessels were petrographically similar to the dolerite quarry of Ible, a site c.20km to the south. Discussion of the Petrographic Analysis The analysis of ceramics from the six Late Bronze-Early Iron Age sites in the study area has revealed a striking dominance in the use basic igneous temper in non-calcareous clay, comprising Fabrics 1, 2 and 3. An initial appraisal of the results may lead to the seemingly logical conclusion that this overall uniformity in paste recipe is the product of a centralised industry. Such an interpretation has previously been utilised for other wide-scale studies of prehistoric ceramics with homogeneous paste recipes. An example is found in Peacock’s investigation of the distribution of Neolithic pottery constructed from gabbroic clays of the Lizard Peninsula (Peacock 1988:302-3). The interpretation of the petrographic data in this study requires more than just an appraisal of broad trends. It is in the consideration of the minor detail that a more accurate interpretation is achieved. Analysis of the basalt and dolerite inclusions on a site by site basis has demonstrated the existence of minor but important textural variations, supported by the hard rock sampling strategy. When these two avenues of research are combined, compelling evidence emerges that indicates specific basic igneous exposures were favoured for ceramic construction by the inhabitants of

separate sites. In the case of Harborough Rocks and Roystone Grange, both sites seem to have utilised one basic igneous source, at Ible. Technological preferences also play a role in this interpretation. The dominance of angular, un-weathered temper at Ball Cross stands in contrast to the use of weathered inclusions at the other sites, indicating the hand of a different potter. Examples of manually crushed temper from Ball Cross are shown in Figure 9. The main question that needs to be asked is regarding the degree of certainty that multiple raw material locations were being utilised, rather than a single, variable source. Variation in crystal size, constituent minerals, degree of weathering, presence of voids, etc, is far greater in the ceramics tempered with this material than in any single basic igneous exposure which has been analysed. Each basalt and dolerite source has proved to be relatively homogeneous, with the only exceptions being those which demonstrate extensive mineral replacement and/or weathering. The widespread nature of the raw material sampling strategy indicates that it is unlikely such a variable source exists within the Peak District. The presence of numerically minor, site specific paste recipes supports the hypotheses of a non-centralised mode of ceramic production, for instance the three polycrystalline quartz tempered pots from Ball Cross. Other examples include the coarse sandstone inclusions which characterise Fabric 5 from Mam Tor, and the rounded quartz temper present in two sherds from Roystone Grange. In comparison to the analysis of temper, differences in base clays within and between sites, most notably Gardom’s Edge, Fin Cop, and Mam Tor indicates the dominant use of locally available raw materials. Although Ball Cross, Harborough Rocks and Roystone Grange failed to produce comparative matches, analysis of the clay matrices from each site indicates the use of separate resources. In contrast, limited evidence for the sharing of clays and their long distance movement as raw materials or completed pots occurs at the columnar basalt outcrop at Cave Dale in Castleton. Analysis indicates that the inhabitants of both Gardom’s Edge and Mam Tor utilised this raw material source, indicating a possible link between both populations.

A

B

A

Figure Figure 9. Angular , fresh basic rock temper in ceramics from Afrom - Sample B3 andB3 B -S ample B4 from Ball 9 - Angular, freshigneous basic igneous rock temper in ceramics A - Sample and B -Sample B4 from Ball. Cross. All micrographs XP awith a field of view 3.8mm. Cross All micrographs taken taken in XPin with field of view of 3.8of mm . 114

The Organisation of Late Bronze Age–Early Iron Age Society in the Peak District National Park The Structure and Organisation of Late Prehistoric Society in the Peak District The petrographic analysis of six Late Bronze-Early Iron Age settlements in the Peak District National Park has produced new evidence which has added greatly to our understanding of societal organisation in this region. The overwhelming dominance of a single paste recipe across both landscape zones provides initial support for Barnatt’s model of a distinct society governed by the geological limits of the region. Moreover, the restriction of basalt and dolerite to sporadic exposures on the White Peak would appear to validate Barnatt’s theory that this area formed a cultural centre, albeit far later in prehistory than he postulated. In this model, the population of the study area would have envisioned their place within an overall cultural identity. There would have been no need to differentiate by the use of distinct raw materials. Instead, there appears to have been wide agreement on the suitability of this dark coloured temper. The assertion of a distinct Peak District population is supported by petrographic studies from outside the geological boundaries of the region. Two sites have been examined directly to the west in Cheshire, namely the defended settlement of Mellor, and the hillfort of Beeston Castle. The Mellor assemblage reflected paste recipes from the Peak District and Cheshire Plains (Ixer 2009:1112), indicating influence from both landscape zones. In comparison ceramics from Beeston Castle revealed a strong preference for locally available crushed rhyolite and granite, obtained from locally available resources (Royle and Woodward 1993:63-66; Microfiche 1:D7-D11; Ellis 1993). Alan Vince’s Petrographic examination of two sites east of the Peak District demonstrated that a wider variety of locally available raw materials were preferred. At Stainton in South Yorkshire, petrographic analysis revealed bivalve shell voids derived from limestone east of the Trent were procured as temper (Vince 2005:1-8), whilst pottery from Kexby, York, contained crushed quartz, dolerite, granite, sandstone, and slag, procured from boulders and pebbles (Vince 1995:1-2). Similarly, Wardle’s (1992) study of Later Bronze Age paste recipes from Thwing in East Yorkshire, demonstrated that imported dolerite temper was limited to the creation of burnished fine ware drinking cups (Wardle 1992:115-116). When taken in combination, petrographic studies from areas adjacent to the Peak District National Park demonstrate that a series of different tempers were utilised. Such a pattern can be interpreted as supporting evidence that the population of the study region was distinct.

Within the model of a unified Peak District society, the data presented in this chapter can be interpreted as the product of a region which was organised into sub-groups. The utilisation of site specific clays in conjunction with distinct basalt and dolerite outcrops indicates the presence of different but related populations, possibly inhabiting more than one site. Such a hypothesis is supported by several aspects of the petrographic analysis. Firstly, the utilisation of basic igneous sources at Ible by the inhabitants of Harborough Rocks and Roystone Grange may indicate a close relationship, or one sub-group inhabiting two sites. Similarly, a link between the population of Gardom’s Edge and Mam Tor is tentatively indicated by the common use of basic igneous clays from Cave Dale in Castleton. Such a relationship is interesting as these sites are separated by c.17km. It must be recognised, however, that within an overall Peak District society, related subgroups may have had open access to the same basic igneous outcrops. Secondly, the preference for freshly crushed basic igneous rocks at Ball Cross indicates potters with slightly different technology. This variation could easily be the result of temporal factors, or experimentation by potters, but may equally indicate a sub-group who worked an accepted paste recipe to a slightly different technique. In conclusion, the distinct pattern of raw material procurement across the Peak District National Park during the Late BronzeEarly Iron Age indicates a closely linked population who were at least partially defined by the geological boundaries of the region. This initial study has indicated that within this society, distinct sub-groups may have existed. With no need to differentiate themselves from other sub-groups, there was a common focus on the White Peak, revealed by their common preference for using dark coloured, basic igneous rocks as ceramic temper. With the wealth of late prehistoric archaeology in the region, there is great potential to expand on these initial findings and test this landscape model. Acknowledgments This study represents a portion of the doctoral research of the author, and as such thanks are due to my supervisors Patrick Quinn, Paul Halstead and Mike Parker Pearson. Gratitude is also extended to my advisors John Barnatt and Pauline Beswick. The Peak District National Park Authority (PDNPA) and the Duke of Devonshire are thanked for granting permission to conduct field sampling. Gill Dunn, curator of Sheffield Museum, is thanked for permitting access to the prehistoric collections for sampling. This research was supported by a three year grant from the Hossein Farmy Scholarship fund.

115

Kevin Cootes Bibliography Aitkenhead, N. 2002. Dinantian. In, Aitkenhead, N., Barclay, W.J., Brandon, A., Chadwick, R.A., Chisholm, J.I., Cooper, A.H., and Johnson, E.W. (eds) The Pennines and Adjacent Areas. Fourth Edition. Nottingham. British Geological Survey. pp. 21-36. Aitkenhead, N. and Waters, C.N. 2002. Dinantian. In, Aitkenhead, N., Barclay, W.J., Brandon, A., Chadwick, R.A., Chisholm, J.I., Cooper, A.H., and Johnson, E.W. (eds) The Pennines and Adjacent Areas. Fourth Edition. Nottingham. British Geological Survey, pp. 37-57. Barnatt, J. 1987a. Bronze Age Settlement on the East Moors of the Peak District of Derbyshire and South Yorkshire. Proceedings of the Prehistoric Society 53: 393-418. Barnatt, J. 1987b. Bronze Age Remains on the Eastern Gritstone Moors. Derbyshire Archaeological Journal 106: 18-101. Barnatt, J. 1990. The Henges, Stone Circles and Ringcairns of the Peak District. Sheffield Archaeological Monographs 1. Sheffield. Sheffield University Press. Barnatt, J. 1999. Taming the Land: Peak District Farming and Ritual in the Bronze Age. Derbyshire Archaeological Journal 119: 19-78. Barnatt, J. 2000. To Each Their Own: Later Prehistoric Farming Communities and their Monuments in the Peak. Derbyshire Archaeological Journal 120: 1-86. Barnatt, J. and Collis, J. 1996. Barrows in the Peak District. Recent Research. Somerset. Bookcraft Ltd. Barnatt, J. and Smith, K. 1997. Peak District. London. B.T. Batsford Ltd. Bevan, B. 2007. Ancient Peakland. Wellington. Halsgrove. Bradley, R. and Hart, P. 1983. Prehistoric Settlement in the Peak District during the Third and Second Millennia BC: A Prehistoric Analysis in Light of Recent Fieldwork. Proceedings of the Prehistoric Society 49: 177-193. Champion, T. 2002. The Later Bronze Age. In Hunter, J. and Ralston, I. 2002. The Archaeology of Britain, 4th Edition. London. Routledge, pp. 95-112. Cope, W.V. 1998. Geology Explained in the Peak District. Cromford. Scarthin Books. Coombs, D.G., and Thompson, F.H. 1979. Excavation of the Hillfort of Mam Tor, Derbyshire 1965-69. Derbyshire Archaeological Journal 99:7-51. Ellis, P. 1993. Beeston Castle, Cheshire. A Report on the Excavations 1968-85 by Lawrence Keen and Peter Hough. London. English Heritage Archaeological Report Number 23. Fearnsides, W.G. Biset, W.S., Edwards, W., Lewis, H.P. and Wilcockson, W.H. 1932. In, Sweeting, G.S. (ed). The Geology of the Eastern Part of the Peak District. London. Stanford. Fleming, A. 2008. The Dartmoor Reaves. Oxford. Oxbow Books Ltd. Gibson, A. 1997. Pottery. In, Britnell, W.J., Silvester, R.J., Gibson, A.M. Caseldine, A.E., Hunter, K.L., Johnson, S., Hamilton-Dyer, S. and Vince, A. A Middle Bronze Age Round-House at Glanfeinion, near Llandinam, Powys. Proceedings of the Prehistoric Society 63: 188.

Gibson, A. 2002. Prehistoric Pottery in Britain and Ireland. Gloucestershire. Tempus Publishing. Hawke-Smith, C.F. 1979. Man-land Relations in Prehistoric Britain: the Dove-Derwent Interfluve, Derbyshire. A Study in Human Ecology. Oxford. BAR British Series 64. Hawke-Smith, C.F. 1981. Land Use, Burial Practice and Territories in the Peak District c. 2000-1000 BC. In, Barker, G. (ed). Prehistoric Communities in Northern England. Sheffield. Department of Prehistory and Archaeology, University of Sheffield, pp. 57-72. Hicks, S.P. 1971. Pollen Analytical Evidence for the Effect of Prehistoric Agriculture on the Vegetation of North Derbyshire. New Phytologist 70 (4): 647-667. Hodges, H.W.M. 1962. Thin Sections of Prehistoric Pottery: An Empirical study. Bulletin of the Archaeology of London 3: 58-69. Ixer, R. 2009. Petrographic Analysis of the Iron Age Pottery from Mellor. In, Cumberpatch, C.G. Hand Made Pottery from Mellor. Unpublished. Kitchen, W.H. 2000. Later Neolithic and Bronze Age Land Use and Settlement in the Derbyshire Peak District: Cairnfields in Context. PhD Thesis. Department of Archaeology. University of Sheffield. Peacock, D.P.S. 1988. The Gabbroic Pottery of Cornwall. Antiquity 62: 302-304. Rice, P.M. 2005. Pottery Analysis: A Sourcebook. London. University of Chicago Press. Royle, C., and Woodward, A. 1993. The Prehistoric Pottery. In, Ellis, P. Beeston Castle, Cheshire. A Report on the Excavations 1968-85 by Lawrence Keen and Peter Hough. London. English Heritage Archaeological Report Number 23, pp. 63-77. Simpson, I.M. 1982. The Peak District. London. Unwin Paperbacks. Smith, E.G., Rhys, G.H., and Eden, R.A. 1967. Geology of the Country around Chesterfield, Matlock and Mansfield. London. Her Majesty’s Stationary Office. Stevenson, I.P. and Gaunt, G.D. 1971. Geology of the Country around Chapel-en-le-Frith. London. Her Majesty’s Stationary Office. Vince, A, 1995. No.5. Pottery and Ceramic Building Material from Site 1997.60/61. Water Pipeline, Kexby, York. Unpublished. Vince, A. 2005. No.71. A Characterisation Study of Some Iron Age and Romano-British Pottery from Stainton, South Yorkshire. Unpublished. Vine, P.M. 1982. The Neolithic and Bronze Age Cultures of the Middle and Upper Trent Basin. Oxford. BAR British Series 105. Wardle, P. 1992. Earlier Prehistoric Pottery Production and Ceramic Petrology in Britain. Oxford. BAR Series 225. Whitbread, I. K. 1995. Greek Transport Amphorae: A Petrological and Archaeological Study, Appendix 3: The Collection, Processing and Interpretation of Petrographic Data. Athens: British School at Athens, pp. 365-396. Woodward, A. 1989. The Prehistoric Pottery. In, Ellis, P. Norton Fitzwarren Hillfort: A Report on the Excavations by Nancy and Philip Langmaid between 1968 and 1971. Somerset Archaeological Journal 133: 39-53.

116