2.1 Introduction

Mapping the meaning of a word and, therefore, the meaning of a related identity is a daunting task. Indeed, the plasticity of language, and the liminal categories which can result from its use, has fascinated humans across the ages. One form in which the multivalency of words has garnered sustained attention is that of the riddle. A riddle, though typically crafted with one intended solution, often elicits a variety of responses. The fluidity of meaning proves to be the very crux of the contest. The challenge of unravelling an unexpected description of a subject proved a popular source of entertainment in early medieval England, as evidenced by the sizeable number of riddles which survive from the period. Even under the scrutiny of generations of scholars, many Old English riddles have refused to yield, with no consensus reached and the many solutions offered seeming as plausible as each other. For example, riddle seventy-four of the Exeter Book, only five short lines in length, has prompted at least thirteen wildly dissimilar answers: barnacle goose, boat, cuttlefish, figurehead, oak, quill pen, sea eagle, shadows, siren, soul, sun, swan, and water.Footnote ³⁵ For the purposes of this section, this diversity of solutions highlights a fundamental problem which pervades our attempts to name the world around us: concepts, and the words associated with them, are highly plastic, or multivalent. The same bundle of words, arranged in the same order, can signify different – and sometimes irreconcilable – things to diverse people. For the historian seeking to impose order on the worlds of the past and to convey their understanding to others in the field, this offers a significant stumbling block.

This section will, therefore, build upon the three observations made by Christine Senecal (see Introduction) and problematise in greater detail the social boundary which lay between ceorl and thegn during the tenth and eleventh centuries. The scholarship which has emerged in the past two decades since Senecal’s thoughts were put to paper is used to inform a new survey of the early English material that seemingly indicates what constituted a thegn. What is revealed is that there are almost no constants across our surviving sources which served to define who and what a thegn might be. Instead, a thegn seems to have connoted many things, but denoted little. Bringing this discussion of the primary sources into conversation with sociological and philosophical models which detail how concepts are constructed and communicated, a new approach to conceptualising what the rank of thegn signified to contemporaries is therefore posited. This produces a body of representational components which served to allow those in early medieval England (and scholars today) to identify who could conceivably claim thegnly rank. Having identified a more holistic and detailed understanding of how the rank of thegn worked, the latter half of the section considers two case studies of how scholars often describe the inhabitants of early medieval England and offers a potential solution for how we may better describe them and improve communication in the field.

2.2 Naming the Past

The philosophical works of Eleanor Rosch and her successors provide a different point of entry into thinking about the contradiction which arises when we, as scholars, attempt to define terms in our historical studies. Lexical concepts are complex entities associated with individual words, composed of other concepts or representational components.Footnote ³⁶ For clarity, let us begin with a simple example: a chair. We may define the lexical concept of the chair via the representational components of having four legs, a seat, and a back support. This traditional, definitional approach to constructing categories is remarkably old, stretching all the way back to Aristotelian philosophy. It is also, however, quite limited. There are many items of furniture that might come to mind which we could classify as a chair that do not fit such a strict definition. Try as we might, there is no one perfect definition. Even the OED’s best and most ambiguous effort – ‘a seat for one person, typically movable, supported by four legs or feet, and having a rest for the back’ – is not foolproof: first and foremost, how is an atypical chair different from a stool?Footnote ³⁷ The definitional approach to lexical concepts is underpinned by an inherent and troublesome inflexibility.Footnote ³⁸ The constituting factors named are either universal or they are not.Footnote ³⁹ This implies that the answers we develop are either right or they are wrong.

Although elegant, the classical theory overlooks that, among philosophers, there is little agreement in defining even fundamental words. We might, therefore, turn to a probabilistic approach or prototype model, which allows that a ‘bundle’ of representational components defines a concept and that an object or person still aligns with the perception as long it demonstrates some of the components.Footnote ⁴⁰ This allows us to build a wider base of representational components which constituted the concept of the chair, for instance. So, one might list that a chair is likely to be a seat for one person, have four legs, feature a back support, include armrests, and be made of a rigid material. There are more items that one could add to the list, but the key is that the probabilistic model allows that no item needs to fulfil all these aspects to be socially conceived as being a chair. If an item displays enough of these representational components, then we are able to identify it as a chair and, importantly, feel confident that another individual – even if their criteria differ somewhat – will still share a similar understanding. This is a fundamental aspect of human communication. As has been noted by George Lakoff, ‘Without the ability to categorize, we could not function at all, either in the physical world or in our social and intellectual lives [and] an understanding of how we categorize is central to any understanding of how we think and how we function.’Footnote ⁴¹ Similarly, if we are to understand the social hierarchy of early medieval England, then we need to develop an approach that identifies the prototypes and/or exemplars (for which read the ‘ideal’) of key sociopolitical categories and that also accommodates the variation that permeates any such groupings.

Terms of representations such as ‘thegn’ and ‘ceorl’ first appear to denote specific social ranks. Indeed, the ‘ideal’ form of a thegn or ceorl can be found in certain cases. However, when we examine evidence from early medieval England more closely, what is revealed is a complex set of potential connotations, or opportunities of representation which are not fixed or constant. Thus, the problem just outlined affects fundamental objects of study more than one might think. The multivalency of social identifiers was not merely relegated to the margins of society and obscure terms – such as fotsetla – which appear only once in the extant Old English corpus, but also affected prevalent social terms – including thegn and ceorl.

Anyone who has studied early medieval England in any depth will have an idea of what constituted a thegn. It may look something like this: a noble-born warrior who was a retainer of a lord, held a manor of some five hides or so which was worked by freemen and slaves, and who was legally distinct from other ranks of society based upon their wergild (the compensation due for injury or murder).Footnote ⁴² This is, certainly, a popular conception, and there are reasonable grounds for such an image. Not only, as discussed in what follows, does the model of a five-hide thegn appear in the works of commentator Archbishop Wulfstan II of York, but there are archaeological examples of such estates, including the division of the sixty-hide estate of Shapwick, Somerset, into smaller five-hide parcels.Footnote ⁴³ However, the problem with this definition quickly becomes apparent. For example, the meaning of the term thegn changed over time, and before the tenth century it denoted a person who supplied service, while high rank was denoted by the word gesith. Thus, one might well argue that I should have described the thegn as a ‘servant’ instead of ‘retainer’. After all, the word thegn can be translated either way and, even though the term gesith had largely fallen by the wayside in the tenth and eleventh centuries, service was still a fundamental part of thegnly identity.Footnote ⁴⁴

So, too, it is problematic to include the legal distinction of wergild. It would seem obvious to include the concept of a thegn being a twelfhynde (1,200 shillings) man given the many historians and students who have recited this as fact.Footnote ⁴⁵ Yet, as Ann Williams points out, this legal division was less secure than one might think. Things were certainly unclear in this regard under King Cnut, who spoke to ‘ealle mine þegnas twelfhynde 7 twihynde’ (all my thegns, twelve-hundreders and two-hundreders).Footnote ⁴⁶ In this instance, the legally separated categories of thegn and ceorl – allegedly rated at 1,200 and 200 shillings respectively – were brought together under the same designation of thegnly status. It may even be that such legal divisions were never explicitly tied to the rank of thegn except within a few minimally circulated documents.Footnote ⁴⁷ Certainly, though the ninth-century Laws of Alfred divided the population into those with a wergild of twihynde (200), syxhynde (600), and twelfhynde (1,200) shillings, it only draws a clear connection between the rank of ceorl and the lowest level of legal compensation. The middling category disappeared from the written record in the tenth century, and it remains unclear to whom it referred.Footnote ⁴⁸ Whatever the case may be, thegnliness was not necessarily directly equated with legal status. It may have operated alongside such legal divisions, and, in many cases, the rank of thegn might have indicated an individual’s expected legal rights, but the relationship between these two was fluid.

Even as early as the first half of the tenth century, the term thegn seems to be used in a rather nebulous manner. Preserved as a Latin translation in the twelfth-century text known as Quadripartitus, a Kentish ratification of the laws of tenth-century King Æthelstan addresses ‘omnes Cantescyrae thaini, comites et villani’.Footnote ⁴⁹ Given that comes could be interpreted to denote a thegn, Williams understandably translates this as ‘all the thegns of Kent, thegns and ceorls’.Footnote ⁵⁰ Still, the framing is remarkably reminiscent of the recurrent Old English rhetorical device of contrasting eorl with ceorl.Footnote ⁵¹ In the Laws of Alfred, treason was discussed in relation to ‘be eallum hadum, ge ceorle ge eorle’ (all ranks, either ceorl or eorl).Footnote ⁵² Another text issued during Æthelstan’s reign and similarly preserved in Quadripartitus, the ordinances of a London peace-guild (frið-gyld), notes that the membership was comprised of ‘ægðen ge eorlisce ge ceorlisce’ (both eorlish and ceorlish) individuals.Footnote ⁵³ From King Edmund’s reign, we find the Latin formulation appear once again: ‘comes uel uillanus’.Footnote ⁵⁴ However, here it is juxtaposed with ‘prepositus uel tainus’ (reeve or thegn). Comes, therefore, does not seem to have indicated thegnliness but, rather, a sense of nobility, at least in the middle of the tenth century, while the title of thegn denoted a particular form of service.Footnote ⁵⁵

At the beginning of the eleventh century, Archbishop Wulfstan remarked that each rank, ‘ge eorl ge ceorl, ge þegen ge þeoden’ (whether eorl or ceorl, thegn or lord), was entitled to the appropriate honour.Footnote ⁵⁶ It may be, as Stefan Jurasinski and Lisi Oliver have suggested, that the juxtaposition of eorl and ceorl is simply ‘a frozen formulaic expression here meaning “all free men”’.Footnote ⁵⁷ Such a reading, while plausible for some instances where it is deployed, would see this turn of phrase rendered somewhat redundant in Æthelstan’s early tenth-century address to the thegns of Kent. If holding the rank of thegn, surely it would be needless to note that they were free from bondage. Whether rendered as comites et villani or eorl ge ceorl, this phrase points towards a sense of social stratification along the lines of something other than legal distinction. That something is probably the division between noble and non-noble.Footnote ⁵⁸ Thus, the royal address to both noble and non-noble Kentish thegns provides an earlier counterpart to that of Cnut’s greeting to his twelve-hundreder and two-hundreder thegns. The relationship between rank, function, legal status, and noble condition had, therefore, long been far from linear in tenth- and eleventh-century England.

To further illustrate the complicated relationship between these factors, it is worth interrogating the idea of nobility in more detail. Discussing medieval Europe, Timothy Reuter defines the nobility as a group whose status was defined legally. For those who exercised ‘power as a result of being well-born in a socially rather than legally defined sense’, he favours the term aristocrat.Footnote ⁵⁹ In light of this, one might argue that the address made by Æthelstan to his thegns, comites and villani could be better rendered as ‘aristocrat and non-aristocrat’.Footnote ⁶⁰ The term aristocrat or aristocracy, however, comes accompanied with its own baggage from early modernity and the Latin word nobilis was, at least, used in early English documents. Rachel Stone raises some further problems that arise from Reuter’s neat division, observing that it ‘leaves unanswered both what counts as exercising power, and the position of particular women, monks or children who did not directly wield power, but whom the sources nevertheless call nobilis’.Footnote ⁶¹ I will, therefore, not be using the term aristocrat here. For the purposes of this discussion, I will continue to use the word noble (and, conversely, non-noble) to describe thegns who are well-born (or not), the distinctions between which will be teased out later in this Element. This allows us to uncouple the term thegn, in line with the evidence from the period, from both legal and lineage-derived standing, opening the door to a more nuanced interpretation of the various ways in which thegnliness could be performed and deployed.

Two mainstays of the characteristics which were used in the Middle Ages to qualify the nobility of an individual were conduct and descent.Footnote ⁶² This appears to be true in early medieval England. Surviving ecclesiastical sources do seem to have sought to instil qualities such as camaraderie, loyalty to companions, urbane eloquence, and gentleness (mansuetudo) among the ruling elites.Footnote ⁶³ The oppositional vices of wrathfulness and vengefulness had the potential to embroil local societies and social groups in destructive networks of conflict. Thus, elites were encouraged to be slow to anger, not to seek revenge, and to tolerate immediate wrongs for the sake of a more distant goal. For a group whose serried ranks dominated the judicial sphere, wisdom and honesty were necessarily paramount virtues. Concerning issues of descent, the Old English corpus provides various words in which the nobility of someone’s heritage is indicated, such as æþelboren (noble-born) and gesiðcundes (gesith-born).Footnote ⁶⁴

The link between nobility and thegnliness is rather complex. A little-discussed late tenth-century law code concerning grave robbery – titled by modern scholars as the Walreaf – illustrates the social importance of familial descent for thegns: ‘Walre is niðinges dæde. Gif hwa of sacenwille, do þæt mid eahta 7 feowertig fulborenra þegena’ (Corpse robbery is an outlaw’s deed. If someone should wish to be acquitted, [they should] do so with forty-eight full-born thegns).Footnote ⁶⁵ Regardless of how widely such an unusual law was ever put into practice, the code suggests that it was possible to conceptualise difference within the thegnhood based on ancestry. A full-born thegn – presumably someone whose parents (or perhaps parents and grandparents) were thegns – was more prestigious than a person who achieved the rank of thegn but did not have the correct familial past. The Ordinance concerning the Dunsæte, arguably issued in the late tenth or early eleventh century, centres the hereditary distinction between thegns and ceorls, contrasting ‘þegenboren’ (thegn-born) with ‘ceorlboren’ (ceorl-born).Footnote ⁶⁶ Wulfstan’s Norðleoda laga reinforces the notion that the attainment of thegnly rank was envisaged as ennobling an individual’s lineage. Still, the use of gesiðcundes to describe the noble lineage they will acquire after several generations perhaps implies that these two qualities, while related, are not synonymous. Drawing these sources together, one could perhaps say that there were: those who were ceorl-born and still ceorls; those who were ceorl-born and gained the prestige of being thegnly yet remained non-noble; and those who were born thegns and were, therefore, noble. In short, there could be a hereditary element to thegnliness yet there was also another division within the thegnage based upon lineage. This might explain why Æthelstan referred to his noble and non-noble thegns.

This survey of nobility in the period is not comprehensive, but it does highlight a variable relationship between the ideas of nobleness and thegnliness. A simpler explanation may simply be that these inconsistencies result from the differing understandings of the various authors. The often-cited example of a seventh-century king’s thegn called Imma draws out the importance of comportment as a means of social differentiation. So ingrained were Imma’s noble qualities, according to Bede, that, even when he tried to hide his rank by pretending to be a peasant during his captivity, his ‘appearance, his bearing, and his speech’ revealed his standing.Footnote ⁶⁷ This seems to draw a firmer connection between thegnliness and noble conduct. Of course, this account, though reworked in the tenth century by Ælfric, may not tell us much about the practicalities of being ‘noble’ in late tenth- and eleventh-century England. We do know that thegns were often expected to be pious and, ideally, generous patrons of the Church. Aspects of these thegnly qualities were often demonstrated through pastimes, such as hunting, or through the sponsoring of a local religious institution.Footnote ⁶⁸ These sorts of activities are certainly what we would expect to see from those seeking to demonstrate their nobility. However, the two, while often linked, do not seem to have been necessarily synonymous. All nobles may have been thegns, but not all thegns were noble. Moreover, a thegn or ceorl in the heartlands of Wessex may well have exhibited nobility somewhat differently than their counterparts in Cornwall, the Welsh Marches, the East Midlands, or Northumbria. It is also worth bearing in mind the caveat that accompanies any such exclusionary behaviours: anyone, if armed with the right resources, can adopt such behaviours and make use of them as part of their social repertoire.

Returning to how we might define the term thegn as a term of representation, a thegn might be labelled a ‘warrior’, a ‘retainer’, a ‘servant’, ‘free’, a ‘1,200’ man’, and a ‘juror’. To this list we might add ‘noble’, ‘powerful’, and ‘lord’. Ownership of physical objects could also form part of these representational components: ‘land’ or to be ‘landed’; military accoutrements, including a ‘coat (of mail)’, ‘shield’, ‘spear’, ‘sword’, and a ‘horse’; and to be ‘wealthy’. We might also include what has been termed the trinoda necessitas, the three obligations of military service, maintenance of fortresses, and work on bridges. Many more aspects might still be added to this list, such as scipe or five hides, or arguably removed. Regardless, even with this relatively modest list of representational components it would not be hard to find examples of thegns who did not fulfil these criteria. Issues arise when we envisage that each thegn fulfilled all of these criteria. At the loosest end of the definitional spectrum, Jake Stattel offers the possibility that the term thegn ‘was simply used as a legal term for any landowner of a certain social and possibly military calibre, even relatively poor individuals’.Footnote ⁶⁹ Moreover, the term thegn was perhaps applied more liberally in the Danelaw than elsewhere, so much so that free Scandinavian newcomers who had settled in the region could call themselves thegns without causing controversy.Footnote ⁷⁰ Speaking of what comprised an English noble after the Conquest, David Crouch provides a rather neat definition: ‘a man who dressed and acted like a nobleman and was not laughed at’.Footnote ⁷¹ This works just as well, I should think, for the early medieval period and the task of defining a thegn. Anyone who could claim the rank of thegn without facing contestation was treated as a thegn. As John Blair succinctly frames the matter: ‘the best way to become a thegn may well have been to behave like one: to be brave and generous, to display conspicuous wealth, to spend it lavishly’.Footnote ⁷² Perhaps the most productive path forward is to consider the representational components already outlined and to assume that an individual considered thegnly was likely able to perform several of these attributes, both displaying various material acquisitions and participating in commensurate social functions.

It should, at this juncture, be quite evident that if one turns to the term ceorl, such issues continue to dog our efforts for terminological clarity. Williams highlights such difficulties in her contemplation of the problems which manifest through translation:

One might argue, given the slew of conflicting evidence relating to the legal standing of thegns in the period, that the word legally might be better replaced by the word socially in Williams’s definition. We return to the idea that the term ceorl indicated little more about the person to whom it was applied than that they were not of noble stock.Footnote ⁷⁴ In short, ceorls could be many things in many circumstances to many people. They could be subsistence farmers, swineherds, artisans, itinerant labourers, officers, and even lords. While some ceorls might be denied the right to possess a weapon, for others, such as the geneat or the cnihtas, it seems to have been necessary to the completion of their activities. We could even make a good argument that they could not necessarily be defined by their freedom. As Dominique Barthélemy and Alice Rio have persuasively argued, freedom was, in practice, a spectrum ranging from the unfree, to the semi-free, through to the free. Of the unfree, the late Paul Hyams stated that, even within this legal category, experiences ranged from ‘the unbearably harsh … to a milder subordination’.Footnote ⁷⁵ In some ways, the task of defining a ceorl presents an even greater mountain to climb. One could compile a list of representational components associated with the concept of ceorlishness that dwarf those associated with the rank of thegn. Thus, there is a suite of components that gives shape to an idea of a generic ceorl which allows (and allowed) a loose identification of such individuals, but it is always contextual and ill-defined when taken as a whole. The plasticity of these social identifiers produces a situation in which we can see representational components that stretch across both the social categories of thegn and ceorl.

2.3 Alternative Models of Hierarchy

Terms such as ceorl and thegn provide a general sense of someone’s position in society and, often, the way they sustained their lifestyle, but they do not necessarily best convey a sense of the dynamics of the power wielded within vertical relations. In short, the labels of ‘ceorl’ and ‘thegn’ are not necessarily as useful or precise as one might wish. Approaching the problem of studying medieval hierarchies and to capture a sense of the forces which produced and maintained structures of inequality, historians have developed other ways of describing the inhabitants of the Middle Ages, including through models derived from the works of late nineteenth- and early twentieth-century thinkers Karl Marx and Max Weber. Marx divided society into classes, viewing vertical relations through the lens of labour. Though dominating early analyses, this was largely superseded by the ‘liberal’ stratification theory in the latter half of the twentieth century. Channelling the Weberian-inspired ‘trinity of inequalities’ – wealth, power, and status – American stratification theorists, such as Robert Nisbet, embraced status, rather than class, as the key vector by which society is divided.Footnote ⁷⁶ Thus, the role of social prestige, rather than economic relations, became much more important in such readings. Though these two approaches have spawned a myriad of ways in which scholars conceptualise and describe the hierarchy of English society in the period, for the sake of brevity, I will detail just two. The first approach frames early medieval society in terms of peasants and landholders or lords, and the second divides the inhabitants of England into lords and their dependants.Footnote ⁷⁷

The word peasant has long been deployed by students of the past to refer to those who worked the land. Though the term is nowhere to be found within the period in question, it does originate in the Middle Ages. Derived from the Old French païsant, it came in Middle English (paisaunt) to denote someone who was not noble and laboured in the countryside.Footnote ⁷⁸ Although a word which has been used in many different scholarly paradigms, it has widely been used to frame modes of extraction through the lens of labour. For those writing in a Marxian or historical materialist framework (and even those writing in response), peasant and lord are terms which speak to individuals’ economic roles. Therefore, these terms represent different ‘classes’ or, to be more faithful to the work of Marx and Engels, ‘estates’.Footnote ⁷⁹ Given the centrality of the desire to maximise income in historical materialism, peasants and lords were seen as inherently and perpetually existing in a state of conflict, the division marked along (theoretically) clear economic lines. Here, the economic base (defined by relations and means of production) informs the superstructure (in short, everything else). The idea of the ‘community’ of peasant society and the tendency to treat the peasantry as a homogenous entity has rightly received sustained critique.Footnote ⁸⁰ Historians of later medieval society, such as Miri Rubin and Stephen Rigby, have pointed out the heterogeneity of such societies.Footnote ⁸¹ Even the eminent Marxist historian Rodney Hilton acknowledged that tension and inequality existed in the peasant strata.Footnote ⁸² Today, few remain wedded to such a rigid, economic definition of the peasant or peasantry, and the label ‘peasant’ is widely utilised in a generous manner by many scholars, accompanied by an implicit acceptance that such imprecise use of the word is largely necessary.

Still, it is hard to get away from the sense that the word peasant indicates a form of labour, one that was commonly performed at a lord’s behest. This, of course, poses some problems when looking to early medieval England. We might struggle to justify calling the very top of the ceorlish stratum peasants. As noted, some were landholders in their own right with a sizeable body of workers who laboured on their land, thus blurring the boundary between ‘peasant’ and ‘lord’. Moreover, those who were of ceorlish rank but earned their livelihood through trade or through an occupation (such as smithing or carpentry) might also, arguably, not be best understood as either peasants or lords. Thus, the lord–peasant framework, though serving to uncover informative aspects about the economic relations (and some aspects of social relations) of this world, often unintentionally elides key social differences within these local landscapes. It is perhaps at its most useful when interrogating the creation of social hierarchy through the lens of economic activity and thus capturing a sense of potential flashpoints which arise through the exploitation of labour, though even in such instances it is sensible to proceed with caution.

By comparison, the lord–dependant model attempts to represent vertical relationships through the cultivation of dependency, particularly along legal and tenurial vectors. Though the word dependant may be applied to anyone, of any rank, who was obligated to serve a lord, it has become synonymous with non-noble laypeople. Even some of the most careful and nuanced readings of the lordly exercise of power and subjugation of the lower ranks of society have framed these relations according to this binary. It is not hard to see why: the less powerful might depend upon the lord in certain ways, such as the right to rent land, to leave his service, or to access justice. Seeing how lords coerced and shaped the lives of those within their reach, scholars have come to commonly depict these relations as between ‘lords’ and ‘dependants’. A key reason for the adoption of the word dependant to describe subordinated members of society might be, as Alice Rio notes, that ‘Since free/unfree corresponded to an earlier system of description which had originated in a very different society from that of the early middle ages, social historians have tended to place greater trust in the couple lord/dependant as representing early medieval relations of power and production, and to abandon free/unfree to historians of law.’Footnote ⁸³ In a bid to ensure that they did not ‘anachronistically’ use a method of social delineation from the late antique period, historians of early medieval Europe embraced an alternative framework which ‘represented society as a sliding scale of hierarchical, bilateral ties of lordship and dependence’.Footnote ⁸⁴ Given that relationality and intersubjectivity are key when trying to understand the manifestation and constitution of status, centring the dependence that a lord can inculcate among his followers can be beneficial.Footnote ⁸⁵

Still, an example taken from the oeuvre of the eminent Paul Hyams highlights some of the risks which arise when using this framework: ‘I use the word “lord” throughout for anyone who enjoys lordship over dependents, including possible slaves.’Footnote ⁸⁶ Hyams’s work commonly displays a nuanced understanding of the application of power in early medieval England and frames subordination as a ‘spectrum’.Footnote ⁸⁷ His use of dependant – which operates in this context as an homogenising term – thus seems to lie at odds with his purposes. After the fashion of Rio – who suggests that the free–unfree binary has misled historians and that ‘grey-area’ statuses were not only present but also, occasionally desired – I, too, argue that the lord–dependant framework presents an often unhelpful binary.Footnote ⁸⁸ It captures a sense of the relationality of power but flattens the degrees of difference between individuals’ experience, especially in cases where dependant is used to refer to anyone from the lower strata of society. I have made the case elsewhere that this is neither a productive nor an ethical way of framing the lives of those who dwelt in the landscape of early medieval England. However, that is all I will say on this matter, for this is not the venue for such discussion.

For the purposes of this section, what is vital is that a lord or dependent in this model could theoretically occupy an array of different social positions. The slave, swineherd, baker, trader, carpenter, smith, wet nurse, wealthy geneat, and thegn could, according to such a model, be seen as dependants. All would be brought under the homogenising term dependant, despite each having a different capacity to influence their personal circumstances. The geneat, merchant, thegn, king’s thegn, and ealdorman might all be identifiable as a lord within this framework. Yet the gulf in power and self-determination enjoyed by geneat and ealdorman was large. Thus, though it can help capture the relationality of a form of power and, when discussing specific instances, may offer some utility, in discussing the broader swathes of social relations it often omits important social gradations.

Neither model represents a ‘silver bullet’ to solving the complexities of these social entanglements nor supersedes usage of terminology from the period. Rather, all can be used in conjunction with each other to generate a more specific image of the period. When treated as part of a larger toolkit, the possibility of a more nuanced understanding of early English society – especially the lower strata – becomes increasingly realisable. There is, of course, the tension that arises from wishing to engage deeply with the granular data which survives from the period and attempting to draw out wider narratives. Hilton, writing in 1974, summed this issue up well, stating that ‘any serious historian has to classify and generalise social phenomena’, but that historians’ ‘oversimplifications of the stages of history’ present a persistent danger.Footnote ⁸⁹ It seems to me that the most productive path is to embrace the plasticity of social boundaries and pinpoint social actors along relevant communal and political vectors. Thus, we might identify an individual as a member of the ceorlish peasantry, the dependant thegnhood, the lordly ceorls, or the unfree peasantry. Even this approach homogenises these people more than one would like, and, in many circumstances, it would be best practice to frame the social position of historic inhabitants of early medieval England in relation to their rank, class, occupation, legal status, tenurial situation, ability to access resources, and, among others, affect change in the local discourse of power. This is not always a practicable option in many studies. Academic writing, while necessitating specificity, demands both clarity and brevity. Nonetheless, it is imperative that we remain aware of the multivalency of the terms used by those who documented the world around them in the past and the ‘flattening’ nature of the tools we use as historians. Even when we look to the period, we find not just one way of thinking about hierarchy, but rather, as seen in Section 3, the signs of an early medieval sociological ‘toolkit’.

3.1 Introduction

It should come as no surprise that our scholarly attempts to model the hierarchy of pre-Conquest England are best served by using the compound method posited earlier in this Element, especially since, as this section will show, even early medieval commentators did not favour a one-size-fits-all approach. Vertical relations were represented using a variety of different lenses in the period, each chosen according to the specific need of the author. Given the need for brevity, Archbishop Wulfstan II of York (r. 1002–d. 1023) will serve as the case study for this section. This choice is, of course, by no means arbitrary. Wulfstan authored or edited many texts which describe the hierarchy of tenth- and eleventh-century England and, given his role in the successive regimes of Kings Æthelred II and Cnut, was well positioned to be one of the foremost political commentators of the period. While this section cannot provide a comprehensive analysis of Wulfstan’s entire corpus (other scholars – such as Dorothy Bethurum and Andrew Rabin – have already admirably performed such a task), four key texts/body of texts will be discussed: the Geþyncðu corpus and the so-called promotion law; his discussion of the Three Orders model in The Institutes of Polity; his previously undiscussed articulation of a binary between the maga (strong) and unmaga (weak) in the law code of Æthelred; and, finally, his amendments to the Rectitudines singularum personarum, a treatise discussing estate management.Footnote ⁹⁰ All these texts bear the traces of Wulfstan’s influence and his desire to forge a ‘holy society’. Each text seeks to regulate and remould the behaviour of the populace of England in a more ‘moral’ fashion, but Wulfstan utilised distinctly differing lenses by which to describe the society which surrounded him. In short, the ideas of thegn and ceorl served to reflect one type of relationship and a way of conceptualising early English society. They were not all-encompassing terms and Wulfstan was forced to make use of a wider lexicon to better capture the complexities of the society of early medieval England. This, I suggest, should inform our own endeavours to describe the hierarchy of the polity in question and, hence, utilise a compound methodology.

3.2 Wulfstan and Geþyncðu

Two texts composed by Wulfstan in the early eleventh century, Geþyncðu (Promotion Law) and Norðleoda laga (Laws of the Northumbrians), have long lain at the heart of academic conceptions of what defined thegnliness and ceorlishness.Footnote ⁹¹ It is not hard to see why, as the texts provide the only explicit descriptions of how someone might move from one rank to the other. Geþyncðu famously describes how each rank was entitled to the appropriate honour:

Featuring a particularly striking caveat, Norðleoda laga presents perhaps one of the most quoted passages on Anglo-Saxon society:

Despite the seemingly plausible regulations and pathway to social advancement laid out by Wulfstan, academic treatments of these texts have grown increasingly sceptical.Footnote ⁹⁷ Whereas once they were taken to be representative of normative social boundaries, the nostalgic tone and the lack of any corroboration for these practices elsewhere has prompted historians to handle them with care.

Forming part of a larger compilation on society and social order, including Mircna laga (Laws of the Mercians), Að (Oath), and Hadbot (Compensation for the Ordained), the Geþyncðu group were not official documents in the same way that we may think of early English law codes or writs. Rather, they seem to have enjoyed only a limited degree of circulation and were recorded in manuscripts which contained material that Wulfstan referenced when drafting later law codes. The collection, at all points, retains its focus on dividing ranks across society and ensuring conformity. Mircna laga describes the wergild of a ceorl, a thegn, and a king according to an alleged Mercian tradition, while Að rounds out the picture of the relative worth of a ceorl to a thegn, stipulating that the wergild of a thegn was six times that of a ceorl.Footnote ⁹⁸ Furthermore, Að distinctly draws out that the ranks of a mass-priest and a thegn were equivalent in the eyes of the compiler. Lastly, the Hadbot, as a means to protect and safeguard the clergy, lays out the levels of compensation to be paid if anyone transgressed against a member of the Church.

Though it has often been held that there was some ninth-century core to these texts (in particular, the first half of Norðleoda laga, Mircna laga, and the first clause of Að), it may be that even this was not the case.Footnote ⁹⁹ Thus, the corpus seems little more than Wulfstan’s personal attempt to conjure order during a period which was witness to an increasingly fraught and confused system of day-to-day social classification. At the very least, it is hard to imagine how the practicalities of these texts would ever have been implemented. How did those ceorls who were temporarily elevated operate within society? Before finally granting permanent elevation, who checked aspirants’ assets to ensure they complied with the requisite qualifications? And what happened to those thegns who fell below the five-hide threshold? Unlike the practice of manumission – a ceremony which saw the emancipation of a slave – elevation to thegnly status was not marked by charter. Presumably this was because the act of manumitting involved the de-commodification of a person, arguably a more profound transition than moving up through society. Without a document to mark such movements, we must conclude it would have relied, as with many things in the period, upon communal remembering. In early medieval England, communities gathered together, acted as witnesses to important events and ceremonies, and remembered. Nevertheless, memory is not static, and social actors rely heavily on signifiers that indicate their place in society, as discussed in Section 2. Even those freed from slavery, whose new status was presumably marked by a charter which laid plain their rights, do not appear to have been safe from challenges to their status.Footnote ¹⁰⁰ In short, the Geþyncðu group offers us limited insight into the actual workings of hierarchy in the period. As Rabin eloquently puts it, the corpus does ‘not so much record defunct regional legal practices as it conjures a nostalgic image of an idealised past’.Footnote ¹⁰¹ Still, it does permit a compelling window into one of the ways Wulfstan chose to envision the society around him.

Unlike the Three Orders model, discussed later in this Element, which centres on the social function of certain groups, the Geþyncðu text focuses upon issues of rank. The rank of the thegn, in this accounting, is not justified by martial prowess. It is a thegn’s tenurial status that affects their position in society. The ceorl who acquired and retained control over five hides was worthy of thegnly rank. Neither their liquid capital nor any martial service they performed, in Wulfstan’s framework, was enough to warrant the promotion of a ceorl to thegnhood. It is in this attempt to outline what defined a thegn that Wulfstan sought to tie landed assets and rank more closely together, re-establishing what we may presume was a fading linkage. This whole tract can be read as an attempt to shore up certain social distinctions. What we are left with is a compelling image of an ordered English society, a text brimming with verisimilitude. Yet it was a corpus of ambitious prescriptive texts, seeking to refashion the social hierarchy in a more structured form. At best, it was loosely inspired by earlier texts and did not make it much beyond his private collection. The key, however, is not how representative it was, but rather that it was one method that Wulfstan utilised to make sense of the world which he inhabited. And, for the purposes of these tracts, it was the binary of ceorl and thegn that served his authorial needs.

3.3 Revisiting the Three Orders

It may seem somewhat odd to revisit ground as well trodden as the ternary or tripartite model of society known as the Three Orders. Its use across Europe has been treated in detail and, even concerning the deployment of the model in early medieval England alone, both Timothy Powell and Inka Moilanen have given comprehensive analyses.Footnote ¹⁰² Yet, for all the nuances present in these earlier studies, one rather fundamental aspect is generally overlooked: that the rank of thegn and ceorl is not always mapped directly onto the roles of bellatores and laboratores. Powell took this model, at its most basic level, as one that divided society into the clergy, nobility, and peasantry. The first appearance of the Three Orders model in England is in an Old English translation of Boethius’s Consolation of Philosophy (880 × 950) attributed to King Alfred. In this translation, it is declared that a king ‘sceal habban gebedmen and fyrdmen and weorcmen’ (must have praying men, fighting men, and working men). These are, according to the text, ‘þrim geferscipum biwiste’ (the three orders/associations of men).Footnote ¹⁰³ They are defined by functionality rather than rank. Indeed, Powell observed that it was ‘more than a synthesis of the bipartite differentiation of clergy and laity with the bipartite stratification of nobles and peasants’.Footnote ¹⁰⁴ This assertion was prompted because the model did not place these orders in a hierarchical structure as did the ‘idea of the Three Estates of Clergy, Aristocracy and Commoners seen in France in 1789’.Footnote ¹⁰⁵ For Powell, while the building blocks were similar, the manner of their depiction was more nuanced in this antecedent formulation. Inka Moilanen is more explicit in her inference that the bellatores were the ‘secular aristocracy’, whose purpose included ‘military defence, the assurance of justice and the order of society’.Footnote ¹⁰⁶ There is good reason for Moilanen’s conclusion if one focuses on Ælfric’s rendition of the ternary model. Expanding upon the tripartite divisions of society, Ælfric critiques the bellatores and states that a warrior who punishes evildoers is ‘Godes þe[g]n’.Footnote ¹⁰⁷ Thus, a connection is drawn between bellatores, thegnliness, and, in particular, ruling. That the bellatores were often read as referring to the thegnhood is thus a reasonable assumption. Nonetheless, the decision to focus upon the functionality of these groups demonstrates an awareness that these facets of status were divisible. Neither the rank nor relative dependency of the groups in question is stressed in these models. Rather, it is what we might describe as their class: the laboratores produce food and can be securely identified as peasants; the bellatores and oratores, given that their labour is directed to other ends and thus must benefit from the surplus produced by the laboratores, are best understood as lords rather than being solely comprised of nobles. It must be acknowledged that even this approach incurs its own problems. After all, not every fighter was necessarily a lord in the period and, while the great religious institutions acted as lords, most ecclesiasts were far from performing this role themselves. Still, this approach moves us further in the right direction and helps illustrate that, even in the period, ecclesiastical scholars made use of overlapping approaches to understand hierarchy.

Returning to the literary giant that is Wulfstan, we encounter once again the topic of the Three Orders in a letter sent to him by Ælfric. This letter, and perhaps even Ælfric’s letter to Sigeweard, appear to have left a lasting impression on Wulfstan, for he later made use of the model in his own meditations on the correct ordering of a Christian society.Footnote ¹⁰⁸ Surviving in a series of successive drafts, now known as the Institutes of Polity, these meditations consider the responsibilities of various sections of society, including the king, counsellors, bishops, nobles, reeves, priests, abbots, monks, nuns, laymen, widows, the Church, and even God.Footnote ¹⁰⁹ In a section titled Be cynestole (Concerning the Throne), Wulfstan makes use of the Three Orders model and redeploys Ælfric’s metaphor of three supports on which the throne stands:

Wulfstan changes the supports of the throne from stela (legs) to stapol (pillars) and, instead of Ælfric’s emphasis upon the broken nature of the society surrounding him and the need for a remedy, he frames the structure as one which needs maintenance but remains strong. Wulfstan’s rendition of the Three Orders model depicts each order like so: the bellatores are wigmen (war-men) who must defend the kingdom in a wiglice (warlike) manner; the oratores are gebedmenn (prayer-men) who þingian (plead) with God; and the laboratores are weorcmen (workmen) who tilian (toil) to ensure that everybody may live. Wulfstan’s version of the model thus returns to the ambiguity of its earliest iteration in Alfred’s translation of Boethius. There is no clear indication from which ranks these warriors and workers are drawn. Quite unlike the version penned by Ælfric and Wulfstan’s own explicit treatment of rank in the Geþyncðu corpus, this represents quite a departure. Here, rank plays a minimal role and, instead, the function of each group in society is what is of paramount importance. For the purposes of stressing the economic unity of society and the need for harmony, Wulfstan chose to downplay rank-based forms of differentiation and, in its place, focus upon a different way of presenting the structures of society, one that stressed function.

3.4 Maga, Unmaga, and the Early English Legal Realm

Another way in which Wulfstan chose to conceptualise hierarchical relations in the period is expressed through the Old English words maga and unmaga. These words survive in the pre-Conquest literature of Alfred, Ælfric, and, most importantly for our purposes, Wulfstan. Yet to receive sustained scholarly attention, the word unmaga refers to someone who is either without means or helpless, such as a child or someone who is ill.Footnote ¹¹⁰ King Alfred’s late ninth-century law code thus declares that: ‘Gif hwa oðrum his unmagan oðfæste, 7 he hine on ðære fæstinge forferie, getriowe hine facnes se ðe hine fede, gif hine hwa hwelces teo’ (If anyone entrusts a helpless (unmagan) individual to another, and they (the new guardian) cause or permit the death of those entrusted, they who previously supported the entrusted shall clear themself of suspicion, if anyone brings an accusation against them.)Footnote ¹¹¹ Alfred’s law indicates that the individual – the unmaga – in question is either a child or someone faced with serious physical or mental challenges. The former is strongly suggested in Ælfric’s Old English translation of the story of Esther: ‘ … 7 hæfde hi for dohtor, forðan þe hire dead wæs ge fæder ge modor, þa þa heo unmagu wæs’ (… and he [Mordechai] took for a daughter [Esther], since she was an orphan/helpless, for her father and mother were dead).Footnote ¹¹² When used in such contexts, unmaga did not connote subjugation or subordination, but rather a state of being unable to care for oneself due to minority or disability.

As an expression of relative political capability, however, unmaga can be found in the late tenth- and early eleventh-century laws of Æthelred and Cnut. As with many of their respective law codes, those in question appear to have been produced under the oversight of Wulfstan.Footnote ¹¹³ These texts portray a particular understanding of the differential in power between those of higher and lower rank and how such difference should be taken into consideration in the process of legal judgment:

This is a call to those who would fulfil the role of legal judge, that they might exercise restraint in deference to the relative social power of those involved in a case such as the maga and unmaga. If those in positions of higher influence were to perform illegal acts (or, more precisely in this context, sinful acts), they would be expected to pay a greater price to make amends. The inequality which ran throughout early medieval English society meant that, in practice, petitioners from lower ranks found the cards stacked against them in a court of law. Thus legal practice dictated that a higher-ranking individual was worth more to a defendant, for their word was given more weight as a compurgator (oath-helper).Footnote ¹¹⁶ Still, legal judgments in early medieval England often featured a healthy dose of compromise. In the laws of Æthelred, we find traces of the systemic preference for cases resolved by compromise (or ‘love’): ‘where a thegn has two choices, love (lufe) or law (lage), and he chooses love, that is to remain as binding as a judgment (dom)’.Footnote ¹¹⁷ Any agreement which reduced the need for the involvement and imposition of ‘state’ apparatus was attractive to the king and his representatives. As Andrew Rabin has observed, the ‘priority was not to determine which claimant had the best legal case, but rather to find a compromise that would preserve the peace of the community’.Footnote ¹¹⁸ Although the leverage of relative social position most likely played a significant, if untraceable role in deciding the course of these early medieval equivalents of out-of-court settlements, there is a possibility that Wulfstan’s exhortation for proportionate judgment was centred upon a pragmatic understanding that many cases were amorphous affairs which used the laws of the land as a guide rather than a strict rule.Footnote ¹¹⁹

The pairing that Wulfstan favours, se maga 7 se unmaga (the strong and the weak), is, in the context of judicial pragmatism, used to stress the dissimilarity of the powerful and the powerless, with the intent to emphasise that each must be accountable in proportion to their situation. A more precise rendering of maga and unmaga acknowledges that these terms characterise an individual’s strength or weakness in relation to their social or political power, or lack thereof.Footnote ¹²⁰ So different are the politically strong and the politically weak from each other that it is comparable to the difference between one who is healthy (hal) and one is sick (unhal). It appears that, at least in this context, Wulfstan is linking the politically strong with the healthy and the politically weak with the sick. While Wulfstan sees the less powerful as dependent in this legal context, just as one might see a sick patient as dependent upon or vulnerable to their healthy caregivers, his linguistic choices seem to stress the need for ethical social practice and care. The ordering of the qualities listed by Wulfstan respectively links political and social strength (maga) with wealth (wela), maturity (ylde), and health (hal), while political and social weakness (unmaga) is associated with poverty (wædl), youth (geoguþ), and sickness (unhal). There is certainly no reason to entertain that the youth of this list are envisaged as lacking vitality or occupying a position of vulnerability in the same sense as the ill. Though ylde can refer to old age, given that it often refers to senior members of a community, in this context it most likely refers to maturity, seniority, and, thus, influence within a community. Likewise, geoguþ indicates the state of being young or, simply, youth, which arguably implies immaturity, juniority, and, thus, less established ties within a community.Footnote ¹²¹ That the politically weak lack wisdom is, perhaps, a viable interpretation.Footnote ¹²² In short, the unmaga were understood to be unwise.

At the centre of Wulfstan’s list of attributes is, nonetheless, a simple awareness that those with less power are vulnerable to predation from those who wield more, with power being measured in terms with which we may be more familiar: wealth, political influence, and social status. That those who are sick, poor, and young are vulnerable to mistreatment from those who are healthy, wealthy, and older is the message that unites these elements. Such an interpretation is reinforced by Wulfstan’s concerns that those bound to a superordinate were vulnerable to coercion: ‘if … anyone commits any misdeed involuntarily or unintentionally, that is not the same as he who offends of his own free will; so also he who does wrong under compulsion is always worthy of protection and of better judgment.’Footnote ¹²³ It is the coercive dimensions of the ‘vertical’ power differential within a legal context that is the focus of Wulfstan’s addition to the law codes of Æthelred and Cnut. Wulfstan’s desire for proportionate judgments and an awareness that each defendant found guilty should bear a punishment proportional to their rank is borne from a pragmatic assessment of the inequity at work in early English society. A judge, in Wulfstan’s eyes, was someone who could distinguish between the forces at play in a complex legal case and accommodate the needs of those who were most vulnerable.

Wulfstan’s use of the words maga and unmaga reveals a reading of power within early medieval England which may be considered alternative or supplemental to the Three Orders model. It evokes the continental framing of pauper (poor) and potens (powerful), a stark rendering of the exploitation of the poor by the powerful.Footnote ¹²⁴ One could, perhaps, infer that these unmaga individuals are ceorls, people who could not represent and protect themselves without their lord’s efforts or, indeed, from their lord. Yet it is telling that Wulfstan does not expressly link the ranks of ceorl and thegn to the weak and the powerful. Wulfstan makes it quite evident that a judge must be able to distinguish carefully between the various duties and responsibilities exercised by those with and without power and to apply the law respectively. Moreover, he exhorts that ecclesiasts must work to uphold what was just and be alert to the possibility of coercion. Therefore, the dependency depicted by the word unmaga is not one strongly linked to the labouring ranks specifically, but rather serves to denote a particular formation of vertical relations. This is, perhaps, the reason we do not find sources discussing a hlaford (lord) and his unmagan (weak men), but rather a hlaford and his mannum (men), ceorlas (peasants), or þeowas (slaves). Wulfstan sought to describe a problem which could develop within vertical relations, particularly as it pertained to the application of the law. It should, therefore, be no surprise that unmaga is never used as a label to categorise ceorls within law codes. Wulfstan saw that in order to best describe the issues which manifested within the legal realm, the terms ceorl and thegn would not serve him effectively. Rather, he needed to turn to the rarely used terms maga and unmaga to best capture a feature of vertical power relations.

3.5 Rectitudines singularum personarum

Rectitudines singularum personarum (hereafter Rectitudines), a treatise describing the obligations and duties of the inhabitants of an estate, provides a snapshot of the ‘ideal’ ordering of the late tenth- or early eleventh-century agrarian manor.Footnote ¹²⁵ Beginning the document is a description of the role of the thegn on an estate:

The thegn will, it is assumed, invariably hold land and, so as to justify his position, perform three services: provide military service, repair town walls, and maintain bridges. Of particular importance, the text stipulates that these services are not just to be performed to substantiate the thegn’s rank but, more precisely, validate their continued ‘boc rihtes’, something often presumed to refer to the ownership of bookland (land that was held by charter and was alienable, though see Section 4 for issues with such a definition). One might see a certain correspondence between this entry in Rectitudines and the image of thegnliness sketched out in the Geþyncðu corpus. In those texts, it was the ability to control five hides across several generations that acted as the marker of thegnly rank, something which – in turn – was linked to clearly demarcated levels of compensation or wergild. Together, these texts produce a rather neat conceptual package, with a well-ordered division between thegn and ceorl. This correspondence could be read as providing compelling evidence of the centrality of bookland and five-hide estates in the construction of thegnly rank. However, the shadowy hand of Wulfstan, which likely links both the Rectitudines and the Geþyncðu texts, should give us reason to pause.

Rectitudines has long been mined for details regarding the administration and relationships on pre-Conquest estates. Indeed, given how few sources we have that depict the minutiae of the manor, it is very tempting to voraciously dissect this account in pursuit of details. Yet its current form serves to offer a rendering of the ‘ideal’ ordering of English society and the ‘reciprocity’ between inhabitants of the estate rather than being bound by the finer points of veracity. The eminent Dorothy Bethurum, a specialist on Wulfstanian literature, identifies several stylistic devices which, like a criminal’s fingerprints at the scene of a crime, indicate his presence in the construction of this text.Footnote ¹²⁷ Rectitudines is often taken to be a companion piece to Gerefa, a text that discusses the duties of the reeve on an estate. Both pieces were likely not authored by Wulfstan in the first instance but were compiled and heavily edited by him at a later date. For this and other reasons, there are few who take the Gerefa as a practical instruction manual for running an estate, especially given the numerous and sizeable omissions within the text regarding key farming and administrative matters. Thus, as Paul Harvey notes, we should see ‘Gerefa as a literary exercise rather than as a didactic or administrative text’.Footnote ¹²⁸ Rectitudines should be seen in a similar manner.Footnote ¹²⁹

While the Rectitudines in its earliest incarnations may have served as a more prosaic handbook of sorts to those who managed the ecclesiastical estates of St Peter’s in Bath, the version which survives to us has been heavily curated by an ecclesiast invested in righting the social disorder left by Æthelred’s reign.Footnote ¹³⁰ The most telling sign of the text’s reworking under Wulfstan and the broader purpose to which it was turned is the inclusion of a section on the thegn. Opening with the phrase ‘ĐEGEN LAGV’ (the law of the thegn), the first line of both this section and the entire document uses the only Scandinavian word in the text (a term which appears regularly elsewhere in Wulfstan’s oeuvre).Footnote ¹³¹ All other sections make use of the Old English word ‘riht’ to describe the obligations of each role. Moreover, the duties of the thegn, as described earlier in this Element, largely look outwards, beyond the confines of the estate and, consequently, the inclusion of this figure seems to run counter to the primary focus of the document. The social order depicted in this reworked version of Rectitudines is, therefore, one that no longer just details the finer workings of an estate, but rather stretches up into the upper echelons of society and details the obligations of a thegn to his king.

Below the rank of thegn, we are confronted with a plethora of specialised roles: geneat (tenant), kotsetla (cottager), gebur (peasant), swane (swineherd), beoceorle (beekeeper), folgere (follower (of the plough)), sædere (sower), oxanhyrde (oxherd), cuhyrde (cowherd), sceaphyrdan (shepherd), gathyrde (goatherd), cyswyrhte (cheese-wright), berebrytte (barley-keeper), bydele (beadle), wudewarde (wood-warden), the hæigwerde (hedge-warden), mylewerde (miller), sutere (shoemaker), and leodgotan (plumber).Footnote ¹³² The role of land suffuses this account and helps delineate between positions: the thegn has bookland, some pay rent (which appears to act as a marker of status), and others are allowed to occupy land solely through their labour. A useful survey referred to as the Feudal Book, which details the inhabitants of estates belonging to the Abbey of Bury St Edmunds, Suffolk, in the late 1080s, provides a useful point of comparison. In the lists surveying the hundreds of Blackbourn, Cosford, and Thedwestry, we find a treasure trove of occupational bynames, in (a) Latin: aurifaber (goldsmith), bercarius (shepherd), clericus (cleric), diaconus (deacon), equarius (groom), faber (smith), hægweard (hayward), mango (monger), mercator (merchant), molendinarius (miller), pelliciarius (perlterer), pistor (baker), porcarius (swineherd), prepositus (reeve), presbiter (priest), sutor (shoemaker), textor (weaver), and (b) Old English: blodlætere (blood-letter), croppere (tree-pruner), dæge (dairymaid), heallemann (hall-man), horsthegn (horse thegn), hweolwyrhta (wheelwright), inngerefa (inn-reeve), and teper (tapper/beer-seller).Footnote ¹³³ In this way, we find some consistencies across both sources in the types of roles listed, such as shepherds, shoemakers, swineherds, and foresters.

Where differences emerge is that the survey from Bury St Edmunds provides a precise indication of how much land each tenant ‘teneo’ (holds) and how much they ‘reddo’ (render) to the institution: Uluric the shepherd held twenty acres and paid twenty denarii; Ulfuine the shoemaker held six acres and paid five denarii; Æluric the swineherd held one acre and paid one denarius; and Lemer the tree-pruner held three acres and paid one denarius and one obulum.Footnote ¹³⁴ By comparison, Rectitudines provides a rather general sense of how much land such individuals may have held, stating that a gebur was to be given seven acres to settle the land and that a ploughman would earn two acres across the year.Footnote ¹³⁵ What is particularly striking is that there is a large variety across these values, and that certain professions, such as shepherding, had the potential to hold more land than other occupations. For example, the Feudal Book records that Godman the smith held only one acre, as did both the mongers Eduine and Goduuine; Goduine the miller shared five acres with Ælric and Uluric; Ælfuine the goldsmith held seven acres; and Ælfstan the baker possessed ten acres.Footnote ¹³⁶ Perhaps most surprising is that Godric the reeve held only three acres and Uluric the merchant held only seven acres.Footnote ¹³⁷ Wulfstan’s Rectitudines depicts a series of serried ranks, each of whom has explicit obligations to his superordinate and can expect certain things in return. More prosaic accounts from later in the eleventh century, such as the Feudal Book, seem to suggest that no such simple division existed.Footnote ¹³⁸ Thus, less ideologically driven descriptions of the local social landscape of the manorial estate show a highly complex web of relationships.

The breadth of ambition which underpins Wulfstan’s revisions to the text – to describe the fundamental ordering of estates and their overall place in the social hierarchy – contrasts starkly with the author’s arguably myopic vision of the local landscape. The use of the word æhtemann, found almost exclusively in ecclesiastical sources, seems to suggest that the types of estates described in Rectitudines were those run by ecclesiastical institutions or, at least, that these were the types of estates with which the author was most familiar.Footnote ¹³⁹ Indeed, the text has been linked to the Tidenham charter, which surveyed an estate belonging to Bath, and which features some similar phrases.Footnote ¹⁴⁰ As an author who did not let the finer points of accuracy get in the way of making an important moral point, it would be entirely befitting Wulfstan’s known rhetorical techniques that he repurposed the text of Rectitudines (and Gerefa) to suit his needs. The addition of a caveat that notes that these duties and obligations differed from estate to estate would seem a rather belated acknowledgement that these details were not entirely representative.Footnote ¹⁴¹ In short, Wulfstan’s extension of his moral ideas into his attempts to detail the workings of local estates was, as Rosamond Faith articulates, a bid to exemplify the ‘“holiness of society”: the structure of society was not simply a matter of social function but of moral order’.Footnote ¹⁴²

A section in Gerefa reveals the importance of maintaining a just and ordered existence on the estate:

In this account, it is needful that a reeve manung (admonish) labourers to do their duty, and they are to employ both rewards and punishments as circumstances dictated. The phrase ‘he sceal his hyrmen scyr pan mid manunge’ is more precisely translated as ‘he (the reeve) must sharpen his subordinates with admonishment’. While acting as a metaphor for the act of inspiring labourers, the image of a reeve honing labourers as his tools is unmistakeable.Footnote ¹⁴⁵ It may not be a coincidence that Alfred the Great likened the three orders to a king’s ‘tol’ (tools).Footnote ¹⁴⁶ Regardless, it is particularly telling that that the ‘hlafordes neode’ (lord’s need) was the ultimate arbiter of what was necessary, and that the expectation of obedience permeates the text. There is an evident concern that the inhabitants of the estate might refuse to yield to the commands of the reeve. To resolve this, the reeve must invoke the lord’s cræft – in this case meaning power, and further evoking a metaphor of trade or skill in the management of hyrmen – and folcriht, or folkright. The threat of a lord’s power was, presumably, not always enough to ensure compliance among the labourers and calls to common custom remained a needed device of persuasion. This sense of justice and duty are emblematic of the ethical code that Wulfstan tried to convey throughout his works. Unlike the simpler ternary model or the binary models discussed earlier in this Element, this account offered a more detailed understanding of local society, one centred upon the language of reciprocal service and rewards. Still, it intersects quite neatly with his use of the binary and ternary models. Echoing the ternary model, all members of the estate must fulfil their function in order to keep the enterprise moving forward. The lord – the maga – must not abuse his power over the unmaga of his estate, but rather reward them fairly for their labour. It perhaps sits less comfortably alongside the Geþyncðu corpus, but there is a similar sense of order. Advancement can only come if an individual uses the correct routes, and the ownership of land remains a key quality of thegnliness. In this way, Rectitudines offers a fourth compelling model of early English society but one that is suffused with ideologically shaped abstractions.

For an individual who sought to draw clear demarcations between ceorls and thegns in his Geþyncðu treatise, it is telling that Wulfstan deployed different models according to the needs of his work. Sometimes the binary of ceorl and thegn sufficed. Elsewhere, he needed to stress the tripartite relationship between the gebedmen (prayer-men), wigmen (war-men), and weorcmen (workmen). When discussing improper judicial practice, he turned to binaries like maga and unmaga. Locating the moral order in the local landscape of estate management required a picture of many serried ranks, all of whom knew their place and received just rewards. In short, the end shaped the means. When it comes to our attempts to conceptualise status in the period, we could do worse than take inspiration from this pre-eminent social commentator of early medieval England, wielding an array of terms and deploying them as appropriate to the ends we seek.

4.1 Introduction

Having examined the proclamations of kings concerning their thegns and the meditations of an archbishop bound up in the world of court politics, I wish now to turn to evidence rooted in the localities of England. That is not, of course, to say that we will be looking at experiences of hierarchy through the eyes of locals – invariably, non-nobles remain the objects of documentation rather than the subject producing such descriptions. Still, this section focuses upon granular forms of documentation which allow for a more nuanced window into the day-to-day ‘realities’ of how the hierarchy of early medieval England functioned. Given the need for brevity, this section concentrates upon three types of sources – charters, manumissions, and guild records – and, wherever possible, supplements the findings with archaeological data (including excavations of manor-house sites and zooarchaeological finds).

To speak of charters is to refer to a remarkably broad category, which includes everything from diplomas issued by kings to the wills of ecclesiasts and laypeople. For the purposes of this section, the charters selected provide us a window into the important roles that subordinates could play in dispute settlement and permit the consideration that leased land, or loanland, may have served in the construction of social status and the acquisition of higher rank. Building upon this initial picture of the inconsistent relationship between power, rank, and tenurial status, an examination of manumissions and status-documents reveals that the lower strata of society in South West England were busily gaining influence throughout the tenth and eleventh centuries. Ecclesiastical institutions at Bodmin, Cornwall, and Exeter, Devon, began to record the names of witnesses who lay beyond the nobility, allowing for the inference that networks of local notables were beginning to emerge. Usually undetected as active participants in legal proceedings elsewhere in the corpus of material from early medieval England, these manumissions show non-nobles as influential actors on the stage of local politics. Rounding out the section, an interrogation of guild statutes shows these to be ideal venues for the intense types of social jockeying discussed by Robin Fleming and Christine Senecal. As seen in the statutes of the Thegns’ Guild from Cambridgeshire, these associations could present opportunities for those threatened by social climbing to enact forms of social closure; they often also seem to have invited intense forms of networking, evidenced in the guild lists from the hinterlands of Exeter.

Altogether these sources show a much more complicated and messier picture than is revealed in the many of the oft-consulted source types. The blurred boundaries postulated by Senecal are thus better documented. More than this, these case-studies allow us to identify both moments and vectors by which inconsistencies across the values of rank and power could emerge. In short, this section illustrates that an individual’s overall status in early English society was constructed via a variety of factors, and it was through the spaces discussed in what follows that the potential meaning(s) and value(s) of social signifiers, such as the rank of thegn, were renegotiated, reshaped, and redeployed across the course of the tenth and eleventh centuries.

4.2 Charters

The lower strata of society are poorly represented in surviving legal material, especially those documents which record active engagement within legal processes. Early English charters documented the conveyance of property, usually bookland, from one party to another, which rarely involved people beyond spheres of thegnhood and nobility. Rather than standing solely on their own merit, these documents worked in tandem with an oral agreement reached in front of a wider meeting, perhaps in front of an assembly at the royal, shire, or local hundred court (shire and hundreds being territorial units of administration).Footnote ¹⁴⁷ The majority of surviving charters record important named witnesses who were willing to testify to the terms of the deal. In almost all instances, those considered to be suitable named witnesses were bishops, ealdormen, ecclesiasts, king’s thegns, and thegns.Footnote ¹⁴⁸ Even noblemen were sometimes referenced via collective descriptions. In a fierce land dispute between a mother and son in early eleventh-century Herefordshire, it is recorded that she bid that ‘eallum þam godan mannum’ (all the good men) attending the moot, or meeting, would bear witness to her decision not to leave anything to her son. In some instances, who comprised the ‘good men’ witnessing certain transactions may have extended beyond the thegnhood, but in many cases, it seems clear that it was intended to denote only the upper echelons of society. Lower-ranking individuals may be glimpsed in the record as part of the several hundreds who were in attendance along with ‘all the leading men’, but this is largely the extent of their visible activities in charters.Footnote ¹⁴⁹ This should be of no surprise when being of higher rank meant that your word was worth more in a court of law.Footnote ¹⁵⁰ Patrick Wormald argued that judicial appeal was made not to written law in the main, but to ‘collective judgement in an atmosphere of public witness’, a group to which ‘local standing was the major qualification for membership’.Footnote ¹⁵¹ Such power formed an imposing force which made the law courts a largely forbidding and inaccessible location for the peasant. It was this phenomenon, as discussed in Section 3, which vexed Wulfstan so greatly. The social and political power of the maga over the unmaga was never so starkly realised as in the domain of law.

It must be noted, however, that power did not necessarily flow in one direction and those who lay outside of the thegnhood or nobility could exert a profound influence on some legal proceedings. In a late eleventh-century legal dispute between Bishop Wulfstan II of Worcester (c.1008–1095) and Abbot Walter of Evesham (r. 1077/8–d. 1104) – which centred upon rights over sake-and-soke, burial dues, church-scot, customs, and military service across the estates of Hampton and Bengeworth in Oswaldslow, Worcester – a steersman named Eadric played a decisive role in deciding the outcome.Footnote ¹⁵² Both plaintiff and defendant came prepared to plead for their cause on the agreed day. Wulfstan put his faith in a cadre of ‘legitimos testes … qui tempore regis Edwardi hoc viderant’ (law-worthy witnesses … who knew the situation before 1066), while Walter preferred to mobilise the corpse of St Egwin to bolster his claim.Footnote ¹⁵³ Evidently, the presence of St Egwin’s body did not prove enough to sway those assembled nor the judges. Rather, it was Wulfstan’s witnesses who would take the day and secure his rights in the region. The importance of witnesses who knew how things had been before the Norman Conquest of 1066 offers a slim but useful window into earlier legal practice.

Wulfstan’s key witness was Eadric, ‘qui fuit tempore regis Edwardi stermannus navis episcopi, et ductor exercitus eiusdem ad servitium regis’ (who had been the steersman of the bishop’s ship in King Edward’s day and who had led the bishop’s army on the king’s service) but, at the time of the court hearing, held land from Bishop Robert of Hereford (d. 1095), also known as Robert the Lotharingian. During a probatio (examination) of the evidence, Eadric’s impartiality appears to have been stressed: ‘et hic erat homo Rodberti Herefordensis episcopi, ea die qua sacramentum optulit et nichil de episcopo W(lstano) tenebat’ (he was the man of Bishop Robert of Hereford, and on the day that the oath was taken, held nothing of Bishop Wulfstan).Footnote ¹⁵⁴ This legal dispute, therefore, likely began after 1078, as Walter of Evesham took office in 1077/8, and, after several meetings, was completely resolved by 1086, as indicated in the records of the Greater Domesday Book.Footnote ¹⁵⁵ Though Eadric cannot be located in records from 1086 onwards, there was an individual of this name listed in the Greater Domesday Book, who held a five-hide estate worth only one-and-a-half pounds in Hindlip and Offerton, near Worcester, in the time of King Edward.Footnote ¹⁵⁶ This identification seems even more certain given that the tenant-in-chief of this estate in 1086 was Bishop Wulfstan and the final settlement of this long-running dispute featured a witness called Edric of Hindlip.Footnote ¹⁵⁷ Given the location of his holdings and that Eadric no longer held the estate in 1086 nor can be identified as holding any land afterwards, this case played out between these two dates. It may even have taken place as late as 1085, given that a charter from Hereford records that a certain Edricus de uuendloc (Eadric of Wenlock, Gloucestershire) acted as a witness to a payment of twenty solidi by Roger De Lacy to Bishop Robert for a lifetime lease of Holme Lacy on St Martin’s Day of that year.Footnote ¹⁵⁸ The legality of the manoeuvrings made by Wulfstan and Robert behind the scenes may be somewhat questionable; after all, they appear to have been close friends.Footnote ¹⁵⁹ Regardless, Eadric was no longer the steersman of Wulfstan’s ship nor holding what I suspect was leased land, given its reversion to his former lord, and thus became the lynchpin of his case against Walter.

So important was Eadric to the clinching of this fraught case that he was singled out above ‘multi alii seniores et nobiles’ (many other elderly and noble men), some of whom we know were either very wealthy or had held socially influential positions, such as the sheriff of Worcestershire.Footnote ¹⁶⁰ Though Eadric seemingly held much less property than the other named witnesses, his testimony held particular weight in a system designed to promote the authority of those of higher rank. Those in higher positions of privilege required, from time to time, the support of those ranked much lower down the standard sociopolitical hierarchy. In this case, Eadric the steersman was a highly experienced individual in charge of a very expensive piece of technology, and the combination of his skill set and service afforded a certain degree of prestige. He managed to leverage his knowledge and labour into such a level of social cachet that he was more valuable to Wulfstan in the legal realm than his more wealthy associates. This may, of course, be an imperfect example for the purposes of this section, given the possibility that Eadric held, at least at one point, five hides and potentially could have ranked as a thegn. Still, even if he held the rank of thegn, it is telling that Eadric does not seem to have been afforded the non-institutional honorific in the documents which survive. His other accomplishments were more pertinent to the demonstration of his position in society. Moreover, what matters most is that he enjoyed significant prestige above and beyond what one might expect for someone of his social standing precisely because of the service he had rendered. A significant proportion of society pursued such a route, putting their skills to work in service to a lord who would offer them patronage and/or commendation.Footnote ¹⁶¹ This, in turn, may have led to the kind of social currency enjoyed by Eadric.

It is very easy to erect in our minds a clear division between ceorl and thegn, particularly one based on tenurial status. As discussed earlier, the work of Archbishop Wulfstan II of York is a tricky beast to handle and clauses 2 and 9 of Geþyncðu and Norðleoda laga respectively are perhaps the trickiest: both refer to five hides as a requisite amount needed for a ceorl to attain the wergild of a thegn. Ann Williams has argued that the stipulation in Geþyncðu – that a ceorl’s agenes lands (own lands) had to be held fullice (fully) – indicates that they were required to be in possession of bookland (bocland) – that is, land often taken by scholars to have been held by charter and which could be freely alienated.Footnote ¹⁶² It is entirely possible that fullice serves in this context to indicate bookland, though I do find it curious that Wulfstan did not simply state that. Rather, I propose that an alternative reading would imply that Norðleoda laga does not necessarily corroborate Williams’s claims on this matter. The word used in this text, hæbban, can simply mean ‘to hold’. Rabin’s translation, quoted earlier, deftly captures the ambiguity of Norðleoda laga, and what seems to have mattered most was not necessarily that one must have held bookland, but rather that one had the ability to hold land in such a way that the three services – the aforementioned trinoda necessitas – most associated with bookland could be discharged effectively.Footnote ¹⁶³ If this were the case, it is possible that leased land might have served to exceed the five-hide threshold, assuming that it actually held any weight in day-to-day social practice. An example from the mid eleventh century, recorded in the Liber Eliensis, provides an account which may support that – although there was a legal and economic distinction – there was little social distinction between leased land and land held by charter. For example, Guthmund, brother of Abbot Wulfric of Ely, was allegedly forced to lease land from his brother because his offer of marriage had been rejected by the daughter of a powerful nobleman. According to this account, despite his noble status (he is described as nobilis), Guthmund did not hold the required ‘lordship of forty hides of land’, which would rank him among the upper echelons of society.Footnote ¹⁶⁴ To make up this shortfall, he leased land from the abbey (though only with the knowledge of Wulfric, who kept the deal a secret and made no ‘written witness’). Soon after, the marriage contract was agreed.Footnote ¹⁶⁵ The would-be bride and her family may not have chosen to distinguish between the ways in which he held parts of his property portfolio – something that may have been characteristic of the different approaches to record-keeping and landholding in eastern England.Footnote ¹⁶⁶ Or else, quite simply, Guthmund kept the truth of the matter hidden, just as his brother Wulfric kept the lease secret from his monks. One can imagine that discretion may have been a preferable option for many would-be lords living on leased land.

Certainly, the place of leased land in the property market may have been more mainstream than the relatively small number of surviving leases suggests. No small amount of work has been undertaken upon the corpus of 147 leases which survive from pre-Conquest England. Stephen Baxter and John Blair have concluded that royal lænland (loanland) – land granted to the king’s men and then potentially reclaimed after the termination of service – was an important facet of the early ‘aristocratic’ experience.Footnote ¹⁶⁷ These lords may not have paid rent in the same sense as manorial tenants on gafolland, but payment or service was provided to secure the right to temporarily inhabit the land. The acquisition of loanland may even have offered an alternative route to ‘intensive lordship’.Footnote ¹⁶⁸ Loanland could, perhaps, eventually be transformed into bookland. As much is suggested by King Alfred’s version of St Augustine’s Soliloquies, in which he describes a situation where loanland could turn into bookland:

Although this passage very much served as a metaphor for salvation, it may not be entirely removed from daily practice. A lease issued by Bishop Oswald of Worcester in the tenth century granted two and a half hides to ‘cynelme his þegne to boc londe swa he hit him ær hæfde toforlætan to læn londe’ (to his thegn Cynelm as bookland just as he had granted it before as loanland).Footnote ¹⁷¹ From the body of surviving leases, we find examples of cneohtas (sing. cniht, non-noble retainers) receiving loanland, perhaps acting as staging posts on the way to bookland.Footnote ¹⁷² Still, these examples from Worcester may be less clear-cut than previously thought. Blair has recently argued that these are not cases of loanland becoming transmuted into bookland, but rather that the booking process merely meant that the ‘three-life’ loan was recorded in ‘some kind of document’.Footnote ¹⁷³ The idea of bookland and perpetual inheritance were not necessarily as closely coupled as they have been traditionally taken to be by scholars. If this is the case, then the bocland of Wulfstan’s Geþyncðu and Norðleoda laga may either have referred to land that was held by some variety of means (bookland, loanland, or otherwise) and recorded by boc or described bookland in the traditional sense in an attempt to reset both blurred social and tenurial boundaries.

Some landholders did not always take the legal path towards trying to retain loanland. A certain Beahmund of Holland, who had held a marsh and fishery from St Æthelthryth’s in Ely, Cambridgeshire, for more than fifteen years in return for a rent of two thousand eels, used this as a means to claim yet more land. After six years of failed suits, Bishop Æthewold eventually managed to reclaim all of this land, but evidently reclaiming loaned (and even un-loaned) land was no small feat.Footnote ¹⁷⁴ In the case discussed previously, Guthmund refused to return the land following the marriage ceremony and even after Wulfric’s death. Having no legal document to prove the land was leased, the abbey was forced to bargain. Guthmund used his possession of the land to reach an agreement that he might hold the land for the rest of his life.Footnote ¹⁷⁵ This problem may well have been much more widespread. As Stephen Baxter has noted, Worcester struggled to reclaim some of its three-life leases during the tribulations faced by the Æthelredian regime, apparently resorting to reluctantly extending them for a fourth life.Footnote ¹⁷⁶

I suspect that loanland played a significant role in the thriving of ceorls and would-be thegns and potentially served to blur locals’ sense of lords’ relative property portfolios. It cannot be contested that the intimate knowledge of the land possessed by locals (termed dense local knowledge by Julio Escalona) was vital to producing stable boundaries between estates.Footnote ¹⁷⁷ Similarly, the testimonies provided by juries of thegns after the Conquest of 1066 proved useful for the Domesday commissioners in creating their infamous record of tenurial holdings.Footnote ¹⁷⁸ Still, one might wonder how accurate these accounts were in trying to establish how absent members of the community had previously held land, especially if Blair is correct in his assertion that there was ‘a significant blurring of boundaries’ between bookland and loanland.Footnote ¹⁷⁹ Together, such uncertainties, when taken alongside attempts of lessees to wrest their land from the hands of the lessors, likely produced a rather complex tenurial landscape that made judgments about precedence based on landholding patterns far less accurate in person than one may tempted to conclude from looking at such records today. Some of those in possession of a charter that ensured their unequivocal ownership over land likely advertised it far and wide, but it was not the only way to forge a claim, and others still may have chosen diverse routes to cement their place as a landowner in the local community.Footnote ¹⁸⁰

Even once an aspiring family had managed to secure bookland, it was not the sunlit fields of freedom one might assume. As David Pratt observes: ‘The landholder’s lifestyle would not be transformed by the gift of bookland, but rather secured with greater permanence.’Footnote ¹⁸¹ The gravity of gaining permanent control over land was undoubtedly significant, else it would not have been so sought after. Still, as long as the land and its appurtenances were returned in their original condition, an aspiring lord could manage his loaned land as intensely as he might. Indeed, we even find evidence of wealthy individuals giving up bookland for access to loanland, presumably because the latter would ultimately prove more profitable.Footnote ¹⁸² The path to greater wealth and greater rank was not always the straight line of an individual moving along the modernist, idealised trajectory from loanland to bookland. Moreover, bookland did not render one free from obligation and obedience. Whatever the finer points of the land market may have been, leasing land was likely a useful stepping stone for aspirants who wished to invest in potentially lucrative land and act in a lordly manner. This is perhaps why a charter of 984 records that King Æthelred gave to his thegn Brihtric eight hides that had formerly been held by Ætheric, ‘quidam rusticus’ (who was a peasant).Footnote ¹⁸³ The concepts of bookland and thegn did not necessarily always align.

Thus, the kinds of concerns about social mobility that one can detect in the Wulfstanian corpus were likely exacerbated by the blurred distinctions between the various ways land might be ‘held’. Alongside the potential confusion around tenurial precedence, archaeological evidence points to a heightened atmosphere of social competition in the tenth and eleventh centuries. Before the tenth century the archaeological remains of elite spaces are often hard to distinguish from other domestic environments. It may be that, in the earlier period, those with higher rank were not so driven as their later counterparts to separate themselves from others. As Blair has shown, lordly halls begin to make their mark on the landscape soon after the tenth century began in earnest. From 900 onwards, a new ‘angle-sided form’ of hall had become popular among what are presumed to be even lower-status thegnly circles. In the third quarter of the tenth century, evidence from the East Midlands suggests that a new preference for halls to be designed in the ‘long-range’ style became common. This fashionable form, with its many private chambers and an elevated room at one end, seems to have swept in from Flanders, an important trading partner. By the 980s, this trend had already lost force, to be replaced by a preference for ‘traditional’ aisled halls.Footnote ¹⁸⁴ Alongside the remains of these buildings, textual evidence points to the importance that an elite was expected to maintain on his estate a burhgeat (enclosure-gate), circan (church), and kycenan (kitchen).Footnote ¹⁸⁵ While we cannot in most cases match excavated estates to the manner in which they were held (i.e., as loanland or bookland), all of this seems to suggest that those with wealth were spending it lavishly to ensure their compounds reinforced social divisions as much as they might.

Architectural changes were, as Robin Fleming and Naomi Sykes have argued, matched by changes in consumption patterns. Osteoarchaeological evidence suggests that the rich and powerful came to dominate the consumption of and dictate access to various animals, including game animals and various seafoods.Footnote ¹⁸⁶ Moreover, the year 1000 seems to have marked the intensification of chicken farming practices and the rise of commercial fishing, presumably with the intent to support a burgeoning demand for such wares among those who could afford them.Footnote ¹⁸⁷ Contemporaneously, a wide range of forces, including marriage, deaths, and the acquisition of new offices, all served to increase land circulation among laymen, of whom some small proportion may well have been ambitious up-and-comers.Footnote ¹⁸⁸ It is no coincidence that the resonances of the Old English word rice – a word which Malcolm Godden has shown had once overwhelmingly been used to denote power – began to take on the additional association of wealth around the first millennium.Footnote ¹⁸⁹ All of this data, when taken together, indicates that the appearance of wealth and the performativity of the status had become ever more pressing. Presumably other social delineators, including tenurial status, were less effective than they previously had been.

4.3 Manumissions

From the rural landscape surrounding Bodmin, Cornwall, and the hinterlands of Exeter, Devon, this potential for prosperity is manifested in a body of material unmatched elsewhere in England: manumissions. Approximately 120 survive from England, with the bulk originating from the south-west. A type of document that records when a slave is freed from their bondage, manumissions mark a profound moment of social transformation – one in which a slave moves ‘up’ through society to a less unfree rank. To be accurate, not all documents in this corpus are, in truth, manumissions, recording instead the sale and transfer of a slave from one owner to another. For this reason, the term status document has been preferred by some scholars.Footnote ¹⁹⁰ The transition of slave to freeperson marked in the documents has drawn the attention of scholars such as Oliver Padel, David Pelteret, Duncan Probert, Frances Rose-Troup, and Alice Rio.Footnote ¹⁹¹ Nevertheless, all of these and other such studies primarily focus on the ‘change of status’ experienced by those manumitted. For the purposes of this section, however, it is the witnesses of these events on whom the focus now falls.

As with charters, manumissions listed the names of key figures who would be able and willing to attest that the act took place, thus ideally ensuring the enduring legitimacy of the ceremony. These ceremonies took place in a variety of locations, but usually in a highly visible venue that had some spiritual resonances. For instance, those manumissions enacted in Bodmin between circa 939 and circa 1075 took place either ‘æt þere cirican dura æt Bodmine’ (at the church door of Bodmin) or ‘super altare sancti Petroci’ (on the altar of St Petrock).Footnote ¹⁹² The manumissions which took place within Exeter and its hinterlands were enacted beyond the direct confines of ecclesiastical spaces, with some even being performed on feower wegas (at the four roads, i.e., crossroads).Footnote ¹⁹³ As the tenth century waxed and waned, those freeing slaves and witnessing their elevation to free status began to include those beyond the clergy and nobility.

Early examples from Bodmin note a pool of witnesses comprised of priests, mass-priests, deacons, and unnamed collectives, referred to via suitably vague formulations – for example, ‘idoneis testibus’ (suitable witnesses), ‘godera manna’ (good men), or ‘alii quam-plurimi de bonis hominibus’ (some many other good people).Footnote ¹⁹⁴ The witness lists of manumissions taking place after the turning of the first millennium began to include a much larger pool of laymen, naming locally important characters. Witnessing the manumission of Ӕlfgyth, freed by a certain Ӕthelflæd, was a range of individuals, including priests, a thegn, a scribe, a reeve, a ‘consul’, and several men who were identified only by their forename and a byname. Many decades later, after the Norman conquest of England, this pattern of a broader range of witnesses continued. A man, named Maccos, both freed a woman named Codgiuo, and took the toll in two different manumissions. Described as a hundredman – someone who represented the administrative unit nominally comprising a hundred hides – this locally influential individual may not have held the rank of thegn. Indeed, Oliver Padel even speculated that Maccos was an imported slave who, having been freed, worked his way up the administrative system.Footnote ¹⁹⁵ Whatever the case may be in this particular instance, scribes were not always taking the opportunity to stress a witness’s thegnly rank and were choosing, or being instructed, to stress different attributes, such as official positions or functions or even distinctive attributes.

Turning to the manumissions which were enacted in Exeter, those witnessing such locally important legal processes included a batsswegen (boatswain), faber (craftsman), webba (weaver), and coc (cook).Footnote ¹⁹⁶ A good example of the type of manumissions executed in Exeter is to be found in the transaction between Ӕlfgifu and Manegot, the former purchasing and freeing Hig, Dunna, and their children from the latter’s control:

The last-named witness to this exchange was Sweinn the scyldwirhrta (shield-maker). While the rank held by Einulf and Godric – who took the toll – or Manleof, Leofweard, and Ӕlfric – who witnessed alongside Sweinn is unclear, it seems likely that Sweinn, described by his trade, was not a member of the nobility. After all, a whole street in York – Skeldergate – was named for the numerous shield-makers who worked there. Compared to earlier examples, where almost all witnesses were drawn from the ecclesiastical community, the witness lists of manumissions had slowly become dominated by secular individuals, some of whom lay somewhere at the juncture between ceorls and minor thegns. That is not to say that the ecclesiastical world had abrogated its duties in recording these proceedings by the late eleventh century – if it had, we would not have such texts to consult today – but it was no longer the main force orchestrating these events. The impetus behind the creation of these status documents were people lower down the social spectrum, individuals who were developing the wealth to engage in such activity and accruing the social influence to warrant appearing in, or drawing together, an assembly of locally important social actors recorded in the witness lists.

At the intersection between a widening group of non-noble individuals who were able to free slaves and those witnessing such transactions sits a manumission in which ‘liueger se bacestere on excestre alysde an wifman Ediþ hatte Godrices doht’ cocraca ut of clist lande at Gosfreige bisceope to xxx p’ (Leofgar (or Liveger) the Baker of Exeter released a certain woman called Eadgyth (or Edith), daughter of Godric ‘Cock’s-throat’, out of Clyst-land from Bishop Geoffrey (of Coutances) for 30 pence).Footnote ¹⁹⁸ Leofgar, a baker, was capable of accumulating enough capital to purchase and subsequently free someone held in bondage.Footnote ¹⁹⁹ Such transactions were now drawing a crowd of witnesses, including those who may have sat beyond the traditional boundaries of nobility, such as Eadric ‘se cipa’ (the merchant).Footnote ²⁰⁰ Narrowing down the actual rank of a merchant remains a particularly challenging task, given that both thegns and ceorls appear to have operated as merchants, but this problem highlights the very plasticity of such social boundaries.Footnote ²⁰¹ A subsequent account documents that a certain Hubert of Clyst sought to overturn the manumission of Eadgyth, highlighting that Leofgar freed her so that any children that they had or might have would have been born outside of bondage. This desire to free a loved one or oneself resulted in a situation whereby individuals managed to free themselves from bondage for significant sums, such as a certain Godwin the Black who paid fifteen shillings (three quarters of a pound) for his own freedom. While examples from Exeter suggest a normative range between two and five shillings (one tenth to one quarter of a pound), a manumission from Bath records that a certain Leofnoth freed himself and his offspring with a sizeable payment of fifty ores, a value three times greater than the pound of silver a thegn holding five hides could expect to make in profit annually.Footnote ²⁰² Despite lives of hardship, shaped by relations underpinned by exploitation, some labourers at the lower end of the social spectrum were able to accrue their own surpluses and even become significant figures within the landscape of local politics.

4.4 Guilds

At the confluence between the importance of community, networking, and the potential to promote and inhibit social advancement sits the guild and the guildhall.Footnote ²⁰³ Guilds took many forms both before and after the Conquest. Unlike the archetypal merchant’s guild of the later medieval period, the guilds of the ninth to eleventh centuries (often referred to as gilds to distinguish them from later iterations) were flexible institutions attending to social, economic, and spiritual affairs. Some, such as peace-guilds, were concerned with maintaining law and order while others were strongly linked to churches and emphasised retaining access to burial rites.Footnote ²⁰⁴ Socialisation and feasting provided the focal point for many such institutions.Footnote ²⁰⁵ They also played a pivotal role in connecting members of the community, with guilds often acting as a form of insurance. Members of the guild were duty bound to protect each other and to pay dues which could act as a method of relief in times of need.Footnote ²⁰⁶

Despite their importance to the functioning of local society, guild statutes rarely form the backbone of scholarly studies. In the seminal collections of ‘Anglo-Saxon’ primary sources compiled by Benjamin Thorpe and Dorothy Whitelock, guild statutes were afforded their own, if small, section.Footnote ²⁰⁷ Elsewhere, guilds are often namechecked as a phenomenon but largely left out of the conversation.Footnote ²⁰⁸ In 1993, Patrick Conner provided a handy study of the physical remains that records the Exeter guilds (i.e., a codicological and palaeographical examination of the Exeter Book).Footnote ²⁰⁹ More recently, Tom Lambert’s monograph considered guilds in greater depth, using them to aid his analysis of the implementation of local order and the application of judicial violence.Footnote ²¹⁰ Studies of this ilk represent the most common scholarly approaches to these records, as a window into social regulation. The work of Duncan Probert has been a notable exception to this pattern of academic research. In 2003, Probert gave a compelling paper – a version of which was published posthumously in 2022 – which unravelled the ‘gildship’ records of Exeter and suggested that the membership of the Exeter guilds included a significant number of ‘peasant farmers’.Footnote ²¹¹ However, Probert’s untimely passing stopped this work in its tracks. Studies by Ann Williams and Rory Naismith in 2016 and 2020 respectively have helped round out our contextual understanding of this corpus of guild records, the latter especially drawing out comparisons between early English and Carolingian associations and their corresponding relationships with state structures.Footnote ²¹² Still, the space that guilds allowed for social competition and the negotiation of social position remains little discussed.Footnote ²¹³ This section, therefore, seeks to extend the focus of Probert’s work to better establish that the membership of guilds included a wide cross section of low-ranking lay society and consider how they facilitated the construction of rank and status in the period.

The Thegns’ Guild at Cambridge is perhaps the most often cited example of these early medieval societies. Given that few guilds clearly indicate who comprised the membership, it is hardly a surprise that this example, which provides some of our clearest data regarding social hierarchy, should draw such attention. The opening of the statutes declares, ‘ðæt æle oþrum aþ on haligdome sealde soþre heldrædenne for Gode 7 for worulde. 7 eal geferræden þæm a fylste þe rihtost hæfd’ (that each [member] was to give to the others an oath of true loyalty, in regard to religious and secular affairs, on the relics; and all the fellowship was ever to aid him who had most right).Footnote ²¹⁴ The guild guaranteed that members would: receive help if a crime was committed against them or if they justly slew someone outside the guild; provide aid if a blood-feud ensued; attend the body of a dying or deceased guild member; retrieve a guild member or their corpse if they had fallen ill or passed away some distance from their own lands; and ensure that fallen guild members would be remembered via a large funeral feast. This seemingly forged a strong horizontal connection between guild members, who could easily focus their thegnly identity through this collective and thus effectively enact a form of social closure. Still, things may be a little more complicated than they initially appear.

While the source carefully identifies that each guild brother was to contribute two ores towards the compensation for the killing of a ceorlish non-member, it also notes that each guild brother was to provide half a mark if the victim was a man with a wergild of 1,200 shillings. Thus, we are presented with two possibilities: (a) that the man with a wergild of 1,200 shillings was a thegnly non-member or (b) that he was neither defined by his thegnliness nor ceorlishness, but solely by his wergild. The former would imply that the Thegns’ Guild at Cambridge, which expressly describes itself in these terms, was not some entity of which all thegns in the region were members. Although the association may have operated as a way for some thegns to stress their collective rank and exclude those without, it is even possible that non-thegnly individuals were allowed entry as long as they were someone’s retainer. Even more pressing is the likelihood that that there were numerous thegns who did not partake in this attempt to separate themselves from other ranks in this way. Given the location of the guild in the Danelaw, it may also be that the thegnliness held relatively little weight and that it was a title that could be adopted by anyone who was free.Footnote ²¹⁵ This prospect perhaps lends credence to the latter possibility, that the man with a 1,200 shilling wergild is not necessarily a thegn or a ceorl. Returning to the apparent discrepancies in the application of wergilds highlighted in Section 2, it may be that the term ceorl is functioning here as a shorthand for a certain wergild and a legal rank rather than a social rank.Footnote ²¹⁶ By comparison, the term thegn may be operating as a social rank rather than defining their legal position. Whatever the case may be, the members of the Thegns’ Guild at Cambridge evidently sought to separate themselves from other lay members of society, stressing the fidelity to one another and excluding others from enjoying their ‘geferes 7 freondscipes’ (society and friendship).Footnote ²¹⁷

Other elites, to borrow a turn of phrase from Patrick Wormald, did not merely write themselves into the activities of a guild but sought to reorchestrate the entire symphony of guild behaviour in a key set by the founder.Footnote ²¹⁸ The Guild of Urk/Ork (also known as the Guild at Abbotsbury) offers an example of one such individual:

Crucially, it is possible to locate this individual in other documents and identify his rank. Urk was a locally important individual in mid eleventh-century Dorset, who was the beneficiary of a writ issued by Edward the Confessor between 1053 and 1058:

Urk was not only a huscarl to royalty but was also in possession of lucrative rights to control the shoreline and sea abutting his property. Urk held additional property at Portesham and Abbott’s Wootton, granted to him by Cnut in 1024 and Edward in 1044 respectively.Footnote ²²¹ The sizes of his smaller estates are readily identifiable: Portesham was held at eleven and three quarter hides; Abbott’s Wootton was assessed at only two and a half hides. Abbotsbury, which appears to have been Urk’s primary residence, was a large estate measured at twenty-two hides.Footnote ²²² Urk’s wife, Tole, likely held another eighteen hides at Tolpuddle in her own right.Footnote ²²³ There are more variables that may have affected the size of his property portfolio than can be addressed here. Nonetheless, it is not controversial to say that Urk and his wife controlled close to forty hides. This would have placed him among some of the wealthiest individuals in the kingdom and might even have numbered among the proceres (chief men) or optimates (best men) described in some sources.Footnote ²²⁴

It is hard to precisely date the gift of the guildhall and the recording of its statutes, but we may presume that the guildhall was gifted sometime after Urk began to acquire serious property in 1024 and before 1066, when a charter records that Tole, now described as his widow, alienated land at Abbotsbury.Footnote ²²⁵ It would appear he was the architect of the guild and the location of the hall in his land allowed the guild to serve as a vehicle for the expression of his local importance. Unlike other examples, we do not know much about the membership of the guild itself. However, we can perhaps speculate regarding the benefits that this guild provided Urk. All guild members had to travel to Urk’s own estate – entering his space at his invitation – and bear witness to the wealth displayed within his guildhall. Existing at the apex of the hierarchy of the guild, Urk’s later gift functioned as an act of memorialisation ‘in lasting memory of himself and his wife’.Footnote ²²⁶ He was both master of the space in which the guild met and its lasting benefactor. All who trod the earth of his guildhall were made debtors to his generosity. The life cycle of this guild and its guildhall was shaped by the legacy of Urk and his status within local society.

Beyond memorialisation, the guildhall may have helped to increase Urk’s social prestige by cementing him as a potential lord. In the latter half of the eleventh century, there appears to have been an increase in the level of competition among lords. Analyses of surveys and the like, such as the Domesday Book, reveal ‘that it was usual for several different lords to have interests in any particular locality’ and thus a lord would find it challenging to monopolize judicial proceedings in the area.Footnote ²²⁷ Consequently, though lords wished to shape vertical relationships in their favour, there were some practical limits. Indeed, the record of commendation practices (the swearing of an oath of loyalty to a lord in exchange for protection) in the Domesday Book ‘demonstrates the existence of competition between lords for commendations in many localities.Footnote ²²⁸ In short, while a good number of freemen were subservient to a lord in a tenurial sense, they also seem – at least at the end of our period – to have sought out a lord to whom they might owe allegiance and who they believed could advocate for them effectively at the local court. At the same time, some lords were building up a base of social influence by seeking to establish as many commendations as possible. Urk’s position as the patron of the guild would have certainly allowed him to construct something of a pipeline, permitting him to acquire as many commendations as he wished. Given the late date of this gift of the guildhall, this may well not have entered Urk’s mind, but it is possible to speculate that such activities occurred elsewhere. Accordingly, guilds may have allowed aspirants to rapidly increase their social standing.

Though guilds could be dominated by thegnly elites, as in the previous examples, many guilds, or at least those in the Devon, had diverse memberships. Only those who were unfree seem to have been wholly excluded from holding membership of these guilds, and even this may have been more down to means than social stigma.Footnote ²²⁹ Indeed, membership was something that one had to maintain, both fiscally and socially. The guild statutes of the Thegns’ Guild at Cambridge and the Guild of Urk make it quite clear that annual dues were an essential component of retaining membership and access to these associations, regardless of any other barriers.Footnote ²³⁰ Through remarkable fortune, the Exeter Book records the names of 304 members of fourteen guilds in the hinterlands surrounding the city and provides a window into the composition of the guilds of the South West. Though what remains preserves only traces of post-Conquest guilds, several of the members can be identified as being active landholders during King Harold’s ill-fated reign.Footnote ²³¹ What emerges when the guild lists are cross-referenced with the Domesday surveys is a notable similarity between the two records, with the population of the vills (an administrative unit comparable to a parish) in the Domesday survey, from villein to bordar, aligning in many instances with the members listed in the guild records. Some of these members can also be confidently identified as lords of the related vill, such as Ailsi (likely a Latinised form of the Old English name Ælfsige) who held in total about two and a half hides in Doddiscombsleigh and Lowley.Footnote ²³² Lords were invested in these localised guilds and seemingly partook in their activities. The guild lists of the Exeter Book were, presumably, carefully updated to reflect who the cathedral could expect to receive renders from and to whom they owed their spiritual services. Establishing precedence may not, however, have been as straightforward as one might think. Returning to Ælfsige, lord of Doddiscombsleigh, it is worth noting that he was not placed at the head of the records for the associated guild. Instead, that honour went to individuals who cannot be securely tied to immediate area.Footnote ²³³ Where numbers exceed those recorded in the Domesday data, it may benefit us to acknowledge the possibility that guild members were members of multiple such associations. If so, this would represent numerous interlocking local communities, predominantly comprised by non-nobles. In short, some aspiring and well-to-do families of the region may have used their surpluses to finance the maintenance of several memberships in different guilds, viewing the networking possibilities as a worthwhile investment.

The presence of these kinds of associations in the everyday fabric of rural society and the upwardly mobile who they may well have served is supported by the distribution of so-called prestige objects (objects which served to indicate status). These items do not offer an entirely conclusive image, especially given that artefacts surviving from tenth-century rural settlements are hard to utilise as indicators of the status of sites.Footnote ²³⁴ Moreover, earlier evidence from the eighth and early ninth centuries shows the presence of a range of prestige objects in non-elite rural settlements, while later survivals in rural settlements commonly include only coins and equestrian equipment.Footnote ²³⁵ This latter category of equipment – such as the copper-alloy elements which comprised part of key equestrian equipment, including the stirrups, harness, and bridle – are widely found into the eleventh century and may well be part of the ‘standard’ equipment of non-noble elites such as rædmaen or geneats.Footnote ²³⁶ This is, perhaps, the material legacy of the kinds of aspirational social actors who jockeyed for position in local guilds.

The importance of local guilds in the tenth and eleventh centuries appears to reflect other demographic changes in the period, such as the increasing urbanisation of the region. To be sure, rural activities overwhelmingly shaped the majority of the lives of those dwelling in England. Yet towns and cities saw an increase in their populations and associated professions, such as merchants. The guilds and associated practices discussed in this section, especially those enacted in the environs of Exeter, seem to have penetrated quite far into the hinterlands of the city. The connectedness of cities to their surrounding populations should be of no surprise at this point in time, especially given that many thegns who lived outside of urban areas held sizeable interests in urban plots.Footnote ²³⁷ If we turn to a quite different area of England, the East Midlands, we find that lords were not only invested in holding profitable land inside the city of Lincoln, but also founded churches to suit the needs of their tenants.Footnote ²³⁸ Moreover, trade was becoming so lucrative in some cities that churches were founded on market squares, presumably not only to provide spiritual services but also to offer other amenities related to mercantile activity, including the swearing of oaths, the scribing and witnessing of documents, the storing of money, and a place to gather.Footnote ²³⁹ Not content to just display their wealth in life, there are even signs that mercantile elites were using their newfound riches to purchase opulent grave markers, thus signalling their social status.Footnote ²⁴⁰ Of course, the East Midlands, featuring a strong Scandinavian influence, are quite some distance from the South West, and it may well be that certain regions were experiencing change in divergent ways and differing degrees. The heartlands of Wessex may well have exhibited a vertical hierarchy that was more resilient compared to other regions. Still, the overall trend across England appears to have been one of increased social competition at the middle and lower points of the social spectrum, one that continued well into the fading of the eleventh century. Hence, Christopher Loveluck’s observation that ‘it becomes very difficult to distinguish a town-house of a landed aristocrat from a town-house of a rich merchant-patrician’ by the early twelfth century.Footnote ²⁴¹

Guilds acted as a vector for various forms of social exclusion and social mobility. One can well imagine that, if faced with a financial surplus, someone might choose to invest in cementing their membership across multiple guilds, thus broadening their networking horizons. Although allegedly taking place towards the end of the eleventh century, the events recorded in the Life of Christina of Markyate are suggestive of the ways that aspirational (mercantile) freemen may have been tempted to utilise such spaces to improve the fortunes of their families, namely brokering favourable marriage agreements.Footnote ²⁴² Moreover, there are also strong indications that bringing retainers or associates to such events was an expected part of a guild members’ behaviour.Footnote ²⁴³ This was done perhaps to not only demonstrate to other members the impressive retinue that one might be able to muster, but also as a reward, both in terms of the material experience and the networking opportunities it might offer. A crude comparison may be to think of the space as something akin to a highly exclusive golf-club, where lavish functions are held, social jostling takes place, and business deals are hashed out.Footnote ²⁴⁴ Of course, just as such avenues may have helped improve a family’s standing, guilds were also potential venues for preventing the rise of the nouveaux riches and practising forms of social exclusion. Even the Exeter guild lists, which hint at diverse membership patterns, objectivised the membership of these associations on parchment, expressly defining who was ‘in’ and who was ‘out’, especially if they could not afford their annual dues. (The golf-club analogy perhaps lends itself neatly here too.) Thus guilds had the potential to both temporarily solidify what was often a plastic social hierarchy and permit its remoulding.