Background

How we perceive the past is influenced by the present and our present positions and perspective. Furthermore, perceptions of the past influence how we conceive societies today, which in turn influences the possibilities, or indeed the power, we have to act today (identified as the double hermeneutic of archaeology already in postprocessual archaeology; see Shanks and Tilley Reference Shanks and Tilley1987a; Johnson and Olsen Reference Johnson and Olsen1992). Therefore we aim to re-examine the foundations of the notions of power as a means to nuance our understanding of social processes of the past.

Power as a phenomenon is very often introduced in the interpretations of specific finds, be they graves or settlements, as a one-factor explanation of social change and social processes. In spite of changing paradigms, from cultural-historical to processual, postprocessual and beyond, the concept of power has been used in surprisingly similar ways (see also Hansen Reference Hansen, Kienlin and Zimmerman2012). Overall, power is presented as something which may be achieved and maintained through the use of violence. In addition, since the introduction of structural Marxist perspectives in the early 1980s, power has also been presented in archaeological analyses as something which may be gained and kept through the use of status symbols. This perspective includes the idea of a deception, in which people are lured by a leader, who keeps social control through the possession and display of specific objects. This interpretational framework includes material culture as a means of manipulation by the elite. Consequently, this understanding also presupposes that the actors in power did not themselves believe in the deception. These interpretations of power strategies thus built upon the premise that a scheming elite managed to cheat the rest of the (clueless) population through a constructed link to past monuments in order to remain in their elite position. In addition, in this stereotypical comprehension of power, it is often assumed that the maintainer of power is a single individual, and in most analyses a male. This one-dimensional understanding of power is probably undercut by the implicit idea that male dominance behaviour would be part of what is called ‘human nature’. Such crude ideas – such as the alpha-male trope – are rooted in decades-old animal-behaviour studies that have been long debunked, including by their own creators (Mech Reference Mech1999).

We may ask, what are the alternatives to these predominant models of power? In the following, we will first explore how power as a phenomenon has been approached in the social sciences, and highlight the possibilities that the analysis of Hannah Arendt (Reference Arendt1970) can have on our archaeological perspective on power. Furthermore, we will explore how power has been incorporated in the agency debate. Subsequently, we will examine some of the alternatives that have been introduced in archaeology in recent years: collective-action theory, anarchism theory and the affective turn. Finally, we will identify avenues to follow in order to get a better understanding of power in our interpretations of prehistory.

Power versus violence

A typical feature in archaeological interpretations of power relations is that power is based on the control of violence. In research on later prehistory and early history the chief, magnate or king is often presented as the possessor of weapons, the leader of warfare and the individual in control of the means of coercion. The understanding of power in the highly influential works of Voltaire and Weber are clearly based on the premise of dominance behaviour, as their definitions start from individuals who impose their individual will by breaking others’ resistance to it. For Voltaire, ‘Power consists in making others act as I choose’, and ‘the ultimate kind of power is violence’ (cited in Arendt Reference Arendt1970, 36). Nietzsche’s ‘will to power’ partly echoes this way of thinking, while his understanding of power as a phenomenon is more complex, and has been interpreted in diverse ways by different authors (e.g. Heidegger Reference Heidegger1981; Gerhardt Reference Gerhardt2000). For Weber, power is ‘the chance to assert my own will against the resistance’ of others (Weber Reference Weber1922, 38; Arendt Reference Arendt1970, 36). In her study of power and violence, Arendt (Reference Arendt1970) sets these quotes into the context of Bertrand de Jouvenel’s (Reference Jouvenel1945, 96) discussion of power, where he describes power in terms of command and obedience. ‘To command and to be obeyed: without that, there is no power – with it no other attribute is needed for it to be’. It is not a coincidence that with that definition in mind, Jouvenel emphasizes war as the source of power, and war as ‘essence of the state’ (ibid.). It is crucial that all these authors, as diverse as their views on power actually might be, come from a place where power is ultimately associated with and derives from the state as the ultimate wielder of coercive violence. This makes these definitions problematic for an analysis of stateless societies. But it also skews the picture for an analysis of modern state societies.

The idea of command and obedience as a kind of universal mode of social interaction is strongly undermined by anthropological studies from different parts of the world, where such a manner of interaction is unknown, or at least highly unusual (Amborn Reference Amborn2019; Clastres Reference Clastres1994; Evans-Pritchard Reference Evans-Pritchard1940; Graeber Reference Graeber2004; Sigrist Reference Sigrist1967). The idea of command and obedience is, as Fabian Scheidler (Reference Scheidler2020) has argued, the imposition of a cause–effect mode of thinking onto social relations, and is only thinkable in a context of institutions directly backed by violence. While institutions of such kinds do exist in most modern societies – the military, police, prisons, the capitalistic workplace, the traditional school – it is fair to say that it does not well represent the whole spectrum of how power works in any society, and the model appears particularly irrelevant to non-state societies. This is where Arendt’s notion of power provides an alternative and more relevant perspective. She offers an understanding of power which is, we argue, most useful for an analysis not only of our current, modern world, but also of prehistoric societies. What sets her apart from other thinkers is her distinction between power and violence. This is essential as it provides a solid basis on which to differentiate between power and coercion. In her own words, ‘Power corresponds to the human ability not just to act but to act in concert. Power is never the property of an individual; it belongs to a group and remains in existence only so long as the group keeps together’ (Arendt Reference Arendt1970, 44).

Arendt thus defines power as a force based on collective consensus, a consensus that is the prerequisite for the phenomenon of society. This consensus is based, practically, on an identification with a set of values and codes, or rules of conduct (Amborn Reference Amborn2019). Arendt’s most famous twist is to define violence as the opposite of power. As she points out, ‘Power and violence are opposites; where the one rules absolutely, the other is absent. Violence appears where power is in jeopardy, but left to its own course it ends in power’s disappearance’ (Arendt Reference Arendt1970, 55). It is, however, important that she also stresses that, in practice, in the real world, power and violence usually occur together. Yet the stronger power, the less place there is for violence, while violence destroys the consensus that is the basis of power. Power and violence, then, have a kind of complementary relationship (figure 1). As violence is in essence instrumental, it depends on the means available. The more uneven the distribution of the possibilities to inflict bodily harm, the more possibilities individuals, or groups of individuals, will have to create a situation where violence can actually replace or destroy power. An extreme scenario would be one man with a nuclear bomb terrorizing everybody else into obedience. Yet this example clearly shows that this would not work without a group of people willing to participate in this kind of regime, which would already constitute a form of power. Even the worst tyrannies, Arendt argues, needed the consensus of at least a part of the population, but the more effective and unequally distributed the weaponry available, the less power would be needed. The moment someone decides to disobey an order, power is being strengthened. At the other end of the spectrum, even the most democratic and/or consensus-based community will need some kind of last-resort violence to defend itself against criminal behaviour or violent disruptors.

Figure 1. Sketch of the complementary relationship between power and violence, following Arendt (Reference Arendt1970, 55): ‘where the one rules absolutely, the other is absent. Violence appears where power is in jeopardy, but left to its own course it ends in power’s disappearance’.

Arendt’s perspective on power has parallels with the thinking of other theoreticians. Most famous is perhaps Pierre Clastres’s (Reference Clastres1989; Reference Clastres1994) separation of coercive and non-coercive power, and Michel Foucault’s (Reference Foucault1982) approach. Refining his different earlier ideas, during his last years, Foucault formulated a view on power as a productive collective force, as the very possibility of practice. Similar to Arendt’s position, he points out that violence has the opposite effect of power. While the latter is discursive, and allows for an exchange of actions and reactions, the former is destructive; it takes away the possibilities to act: ‘A relationship of violence acts upon a body or upon things; it forces, it bends, it breaks on the wheel, it destroys, or it closes the door on all possibilities. Its opposite pole can only be passivity’ (ibid., 789). According to Foucault, where there is power, there is also the acceptance of power. Thus power is created, maintained or changed not through violence, but as part of an ongoing negotiation. In line with Arendt, with the notions of power in Foucault’s writing in 1982, the usual individualized perception of power was defied, as Foucault demonstrated that power only occurs when it is accepted by the part who is lowest in the hierarchical relationship (ibid.). Foucault’s notions of power have, however, other challenges. In Arendt’s thinking, power is not an individual’s faculty, but rests on collective consensus. Furthermore, as the archaeologist Lynn Meskell (Reference Meskell1996, 8–9) points out, power may be perceived as inscribed on bodies through discursive discipline. Thereby, Foucault undervalues the active corporeality of the body. Meskell points out the limitations of the models of domination (and resistance) in which the focus has been on power strategies (or rather strategies of domination), which take for granted that all-consuming energy and labour is invested to maintain a position of power (i.e. domination). With such a simplification of the social dynamics, the relevance of studying the life experiences of the individuals involved is overlooked and underestimated (ibid., 8–9). By examining social changes and social processes in a specific context of time and space, we may gain insights which take us beyond power balances. In any context these social processes include the creation of relationships between various social groups and within groups.

The relation between power in the sense of Arendt and domination (Herrschaft) in the sense of Weber is not that of opposites, nor are they mutually exclusive. On the contrary, what Arendt’s work helps us understand is that any form of power, even coercive domination, will fundamentally depend to a varying degree on the collective form of power. Her research thus enriches our understanding of power rather than replacing Weber’s masculine, top-down, state-backed domination model with a purely collective, consensus-based one. Yet we are proposing that the collective component of power is much stronger than is usually accounted for, especially in non-state contexts.

In prehistoric societies, where governments are largely absent, the relation between power and violence was likely much more dynamic, but we have to assume that also here power, and not violence, governs the majority of social interaction. While Keeley (Reference Keeley1996), Bowles (Reference Bowles2009) and Pinker (Reference Pinker2012) have tried to evoke a Hobbesian view of prehistory as an extremely violent, warlike place, Ferguson (Reference Ferguson and Fry2013a) has picked those accounts apart and shown in a systematic survey of the record of prehistoric Europe and the Middle East that indications of extensive violence only occur very sporadically for most of the time, largely associated with periods of crisis, and only increase in relation to early state formation processes (Ferguson Reference Ferguson and Fry2013b). While of course inter-community warfare is something other than intra-community coercion, it is still likely that an overall less violent world will result in a world where collective power is necessarily a much stronger component in social relations than is coercion. However, we do not want to get into a discussion about the scale of violence present in different periods of prehistory; rather, we want to show that power as it is understood in the works of Arendt, Foucault and Meskell opens the door to a richer understanding of social systems in all their variability in space and time, and to a reformed understanding of phenomena like social inequality, prestige and status formation, centralization processes, and core–periphery models, which all are often largely portrayed using the reductive dominance-behaviour notions of power. Building on Arendt’s and Foucault’s understanding of power strongly benefits archaeological discourse, from discussions about the location and configuration of agency to considerations of past social and political organization. In the following we will briefly discuss the conceptualization of power on different levels of archaeological research, looking at the history of archaeological thought as well as its current potentials.

Power and agency in archaeology

In the course of the last three decades agency has been implemented in archaeological discourse in a highly influential manner, focusing on the capacity of humans and others to act. However, it also seems that in the postprocessual wake the eagerness to move away from individual male power asserters, particularly in the early days of postprocessualism’s critique of determinism, was in several cases combined with a misguided extreme individualism. This, we argue, disregards important collective components of power.

With the publication of Social theory and archaeology by Shanks and Tilley in 1987, archaeology took an important step towards the implementation of agency in archaeological interpretation of the past, and a critique of more deterministic models of power assertion was presented. Particularly, Shanks and Tilley countered the processual idea that individuals do not have any real possibilities to change the situation they are located in, always structured by overarching social systems and ecological forces (Shanks and Tilley Reference Shanks and Tilley1987b). Instead, individuals are acknowledged to have their own determined goals and can alter and break out of their structured dispositions through action. However, this effectively underestimates the collective nature of power. Shanks and Tilley based their ideas on concepts first described by Pierre Bourdieu (Reference Bourdieu1977) as ‘structuring structures and structured structures’, or habitus. Although extended on concepts first presented by Marx (Reference Marx1909), Durkheim (Reference Durkheim1966), and Mauss (Reference Mauss2016), the theoretical framework of Bourdieu is relevant for archaeological inquiry on power because it is able to showcase how individuals are prime movers and yet their actions are also a result of past practices. Thus power assertion over others is not simply a deterministic force caused by overarching structures or specific modes of production (cf. Marx Reference Marx1909); it is also historical, dynamic, and ever-changing. For Bourdieu, in studying symbolic power it became increasingly important to understand why societies act beyond economic rationale, but also in order to understand how certain individuals are able to determine others’ position in a hierarchy (Bourdieu Reference Bourdieu1984). Although used to understand modern social systems, the idea of symbolic power is one that has been frequently used in archaeology to explain how power, guised as material manifestations of someone influential, is used and manipulated to gain acceptance from the general commune. It also manages to skew the economic emphasis behind power manifestation towards a cultural emphasis, paving the way for ideological and ritual manifestations in the archaeological record to have a defining role in shaping and situating power in different societies (e.g. Bourdieu and Passeron Reference Bourdieu and Passeron1977). However, in order for symbolic power to work there needs to be an acceptance or awareness of that power by the general community (see above). Part of the appeal of Bourdieu’s concepts, and the reason they are useful for archaeology, is that they create a frame of reference where power assertion within past societies is seen as constantly changing, yet situated, which enables us to see processes of power negotiation. This contradicts the one-way street of traditional coercive power. Similar notions of power can be identified in the works of Anthony Giddens (e.g. Reference Giddens1987, 11) and are discussed in an archaeological context, for instance, by Barrett (Reference Barrett1988).

In contrast to Bourdieu’s views on agency that still hold a firm grip on structures as a cause behind group and individual action, within archaeology at the beginning of the 21st century the debate shifted its focus away from the collectiveness of power and gave way to the agency debate (see, for instance, Barrett Reference Barrett, Robb and Dobres2000; Dobres Reference Dobres2000; Robb Reference Robb, Renfrew and Bahn2005). Rather than mentioning power explicitly, it is incorporated into the debate. This was, for example, seen in the edited volume by Dobres and Robb (Reference Dobres and Robb2000), with many influential contributions that often tend to be oriented toward individual agency, which in many ways harkens back to, and reproduces, many of the same problems seen in more functionalistic narratives on power (ibid. 2000). The potential agency of objects entered the forefront of the archaeological discourse (with perhaps Gosden Reference Gosden2005 as the prime example). Though this led to objects and non-human beings now being included as actors and agents, power was mainly articulated in relation to the potential influence or affect that objects would have had and still have in the world. A debate of primary and secondary agency, based on the scholarship of Gell (Reference Gell1998), was particularly influential in these debates. Indirectly, this also relates to the power debate. Though agency has been added to the analyses, power is still perceived and presented as the product of an individual agent, whether it be a human individual, an object or another type of being. Thereby the remaining actors within a given society are more or less presented as opportunistic sheep, following their leader without real purpose, preferences or even in fact an agency of their own.

In the last 25 years, graves in particular have been utilized in explorations of social identity (see, for instance, Brück Reference Brück2004; Fowler Reference Fowler, Beaudry and Hicks2010; Jones Reference Jones1997; Reference Jones and Insoll2007). Emphasis has been on the social groups that were previously underrepresented, including women, slaves and children. Yet, perhaps surprisingly, though the agency of various social groups is brought to the fore, power discourses have not been explicitly included in these studies. Indirectly, this has also left these groups more or less stripped of power within these interpretational schemes. Thus, rather than challenging the ways in which power was perceived, within these interpretations the lonely warrior groups imagined in the prehistoric villages (be they fortified or not) have been supplemented with other groups, while the power remains, within this approach, in the hands of a male, warrior elite.

The affective turn

Recently, agency in archaeology has moved toward the concept of material agency and the affective turn (Jones and Boivin Reference Jones, Boivin, Hicks and Beaudry2010). The question of how power is perceived within archaeology also opens up towards the study of affects. These studies ask, in any given historical context, what is acting upon what. The debate of agency has, however, also turned into a game of Old Maid, in the sense that focus is on ‘who has it’ rather than on the effects of agency and on how the game moves around the table. The current affective turn (Back Danielsson Reference Back Danielsson, Alberti, Jones and Pollard2013; Brady and Bradley Reference Brady and Bradley2016; DeMarrais Reference DeMarrais2013; Hamilakis Reference Hamilakis2013; Jones and Cochrane Reference Jones and Cochrane2018; Lund Reference Lund2021; Stewart Reference Stewart2007; Tarlow Reference Tarlow2012) may have the potential to direct the attention from the actants and from the cards to the game, so to speak. When discussing the ordinary affects that we as humans are exposed to and expose, Kathleen Stewart describes them as a scene of immanent powers. These affects also include the different, seeking capacities to effect and to be effected and affected (Stewart Reference Stewart2007; for an archaeological example, see Brady and Bradley Reference Brady and Bradley2016). A focus on affects, or what in Germanic languages would be described as Einwirkungskraft, allows us to look into how things work on each other and with which forces. Rather than asking what the agency of a phenomenon is, we may question how material culture works on humans in the past and what activates it. DeMarrais (Reference DeMarrais2013, 103) points out that by acknowledging that affects are embodied states, and through studying material objects’ affecting presence in the present, we may identify traces of emotional experience in the past. Furthermore, these affects do not only deal with emotions and sensuous aspects of past lives; affects contain the influences that objects may have had on humans in the past. In some instances, they may even be prime or relevant movers of actions.

Thus the affective turn has increased the focus on how things work on humans, and how human beings are bound to material culture and to landscapes and places. In essence, these perspectives put their emphasis on relationality. This is not simply about emotions, but also about how things work on each other and on humans and other beings and how forces are brought into this. If the research of agency resulted in a blame game of passing the buck – who is responsible for the actions? – then the affective turn has turned the focus away from the cardholder and onto the game. Yet these perspectives do not fully respond to the question of how power may be approached and perceived in a prehistoric setting.

The affective-turn discourse thus targets on a fundamental, ontological level how power is constituted and is ongoing among scholars interested in the realm of theory of cultural processes, in more socially oriented perspectives. These notions are not directly related to the archaeological debates on social organization in prehistory, but they may prove useful in future perspectives which aim at understanding social processes. On a more practical level, collective-action theory and anarchist theory have been increasingly considered as a tool to highlight the role of individual and collective agencies in the establishment and dynamic use of power relations. In contrast to the affective turn, these theories engage directly with questions of social organization and may therefore be useful tools in reassessing our understanding of power.

Responses to power: collective-action theory and anarchism theory

There is a complex of connected approaches to agency, politics and social organization in the archaeological literature, which are highly compatible with our aim to highlight a new understanding of collective power in the past. Collective-action theory (Blanton et al. Reference Blanton, Feinman, Kowalewski and Peregrine1996) as used in archaeology introduces a voluntaristic perspective that explicitly strengthens the cooperative and collective aspect of political negotiations in the past. Several studies have applied this approach in investigations of state formation or early state political systems (Blanton and Fargher Reference Blanton and Fargher2008; Carballo and Feinman Reference Carballo and Feinman2016), where it highlights the (varying) collective components of many early state institutions. This theoretical approach explicitly seeks to better represent the agency of non-elite actors, by highlighting the collective bargaining power of different groups within society, to balance the overall research focus on elite agency or chiefly power (DeMarrais and Earle Reference DeMarrais and Earle2017; Furholt et al. Reference Furholt, Grier, Spriggs and Earle2020). This corresponds with the concept of heterarchy, introduced into the archaeological discourse by Crumley (Reference Crumley1995), which directs the attention away from a one-dimensional top-down perspective on hierarchies, and instead points to the multiplicity of different, often unranked, forms of social organization. McGuire and Saitta’s (Reference McGuire and Saitta1996) idea of ‘dialectics’ in social organization points in the same direction, as they describe the dynamics and tensions between exclusionary and more cooperative forms of organization, and argue for forms of power which are based in social relations and are not reducible to hierarchies.

Anarchism theory has recently come to the fore in archaeological interpretation of past societies (e.g. Angelbeck and Grier Reference Angelbeck and Grier2012; Grier Reference Grier, Finlayson and Warren2017; Borake Reference Borake2019; Wunderlich Reference Wunderlich2019; Furholt et al. Reference Furholt, Grier, Spriggs and Earle2020). The theory, deriving from scholars including Bakunin (Reference Bakunin1950 [1872]) and Kropotkin (Reference Kropotkin1972) and later used by anthropologists such as Sigrist (Reference Sigrist1967) and Graeber (Reference Graeber2004), specifically focuses on how local societies through cooperative means can withstand or challenge the emergence of coercive power and centralization. It was specifically outlined for archaeological inquiry by Angelbeck and Grier (Reference Angelbeck and Grier2012), who studied social organization of the Coast Salish in the region of the Pacific Northwest. They argue that the complex organizational practices of the Salish, which included status differences and slaves, are nevertheless best understood in terms of decentralized and collective forms of power, a system set up to actively prevent centralized coercive power emerging. The leaders of the Salish were dependent on specific social settings and had to relate to autonomous yet interconnected individuals and local groups (Angelbeck and Grier Reference Angelbeck and Grier2012). Here, power is represented in the form of justified authority, which has to be negotiated in relation to autonomous individuals, flexible and fluid residence groups and decentralized and cooperative networks.

Through an anarchistic perspective, individuals will always seek autonomy instead of oppression, if they have the possibility. Therefore the traditional view of power as residing mainly in a top-down and coercive strategy may refer to specific social configurations only, if it is actually applicable at all. Anarchist theory emphasizes the relational and socially embedded nature of power, and points out that top-down attempts of power concentration must necessarily be seen in relation to a bottom-up response. Within an anarchistic theoretical perspective, leadership is not formed by coercive power. Rather, leaders have access to power based on the collective, mutual understanding of their ‘justified authority’, which they have been granted due to their specific set of knowledge or skill. As opposed to a traditional perspective on power, leadership within this framework is built on power to do something and not power over another part (Angelbeck and Grier Reference Angelbeck and Grier2012, 547–52; Borake Reference Borake2019, 66). Anarchism, as for example it has been characterized by David Graeber (Reference Graeber2013, 187), can be described as the idea that society could be solely built upon power in Arendt’s sense, without an inbuilt last resort to violent coercion. While this might be seen as a utopian vision, the anarchistic perspective helps us to take more seriously the collective, socially embedded nature of power, and differentiate it from forms of violent coercion in our analyses of current social issues, and likewise when studying the past. It counters the mainstream reduction of power to individualized (male) domination, and instead focuses on the collective and affective nature of social practice, which constitutes power in every social context. Anarchism is thus a useful thought model for societies in the past who, to a much lesser degree than our own, are characterized by centralized systems of domination. Referring to the affective turn, we also want to move the focus away from identifying individual actors (be they human or not) towards examining power in terms of how forces and practices work on each other in any social encounter.

Some archaeological cases

The authors work with the Neolithic, the Bronze Age, and the Iron Age societies of Central and Northern Europe respectively. It is revealing to find the interpretations relating to these very different societies to be so strikingly similar in the manner in which power is integrated into the perception of social organization, social change and social processes. Our perspective on power, as a relational, collective and socially embedded force, which is accepted and enacted by all parts of the power relation, has consequences for many fields of archaeological interpretation. Furthermore, these relations may stretch beyond relationships between humans exclusively. As identified in the affective turn, matters, substances and their capacity may influence and affect human and other beings in manners which include power relations. These perspectives on power may have radical consequences for research on prehistory.

In the field of Neolithic research, a popular topic is the idea that this would be the period during which social stratification first occurred. This assertion already implies a teleologic, social-evolutionist scheme, a development from egalitarian to hierarchical, which is problematic as such (e.g. Wengrow and Graeber Reference Wengrow and Graeber2015; Graeber and Wengrow Reference Graeber and Wengrow2021). And more basically, archaeologists have had a hard time actually pointing out empirical cases in which such stratification shows up during the Neolithic period. The kind of argument for social stratification in specific contexts that is applied mostly rests, we would argue, on an understanding of the power concept that is exactly the kind of individualized and top-down domination approach we criticize. For example, all kinds of logistically or artistically elaborate constructions from the Neolithic – be it Göbekli Tepe (9600–7000 B.C., e.g. Price and Bar-Yosef Reference Price, Bar-Yosef, Price and Feinman2010), Central European LBK, Lengyel or Michelsberg enclosure systems (5350–4500 B.C., e.g. Řídký et al. Reference Řídký, Květina, Limburský, Končelová, Burgert and Šumberová2019; Gronenborn et al. Reference Gronenborn, Strien, van Dick, Turchin, Meller, Gronenborn and Risch2018), the Western and Northern European early long barrows and megaliths (4500–3200 B.C., e.g. Midgley Reference Midgley2005; Chambon/Thomas Reference Chambon, Thomas, Furholt, Lüth and Müller2011; Gebauer et al. Reference Gebauer, Sørensen, Teather and Valera2020), to name only a few – are often interpreted as an expression of early social hierarchies. The main argument used is that such complex organizational tasks and the impressive workload could not have otherwise been pulled off – relying on the narrow concept of power as a command–obedience relation, where individual command givers would direct the particularized obeying workers. This whip-wielding overseer is someone that might work in early pharaonic Egypt, with state institutions, a monopoly of violence and an organized military to back it up. But in a Neolithic context, a chief cannot effectively give commands, threaten punishment for non-obedience or prevent people from defecting, because of the absence of the kind of imbalance in the possibility to muster violence, which is a prerequisite for coercive domination.

During most of the Neolithic period in Europe, specialized weapons and the display of warriorhood are almost non-existent (Chapman Reference Chapman, Carman and Harding1999; Schulting Reference Schulting and Ralph2013; Furholt Reference Furholt2021). Instead, Neolithic construction projects would largely have to rely on cooperation, on a collective consensus and a common goal. In cases of disagreement, conflict resolution and decision making must have been based on collectively shared values, norms and rules (Amborn Reference Amborn2019), and will have included collective negotiations in order to achieve consensus accepted by all sides involved. This is not to negate the possibility of hierarchical relations in the organization of the work, but this dominance-based version of hierarchy is just one possible form of organizing collective decision making. Anarchist theory provides the principle of justified authority, and in the context of Neolithic construction projects – be it long barrows, megalithic monuments, enclosure ditches or post-built longhouses – it is conceivable that skilled or experienced individuals would have been granted more prominent roles in the organization and direction of work. Such authority would then have been subject to constant reassessment of collective approval – power in Arendt’s sense. In such a context, the establishment of a sustained hierarchical social structure – a stratified or politically centralized society – would be very hard to establish, if it were attempted at all.

For example, the southern Scandinavian Funnel Beaker period megalithic burials constructed between 3600 and 3100 B.C., as well as the causewayed enclosures of the same period, show a structure that is well compatible with, if not indicative of, such a collective power dynamic, backed by justified authority (Wunderlich Reference Wunderlich2019). These monuments are spread out in clusters over the whole post-glacial landscape (Fritsch et al. Reference Fritsch, Furholt, Hinz, Lorenz, Nelson, Schafferer, Schiesberg and Sjögren2010), usually modest in size and workload, and while early megaliths would often be used for burials of individual deceased, over time burial was collectivized. Larger and more complex monuments turn out, upon closer examination (Mischka Reference Mischka2011), to be made up of a sequence of a number of smaller construction events, which very well could represent a series of independent teams of workers, which would evolve in rather unexpected ways (Furholt, Hinz and Mischka Reference Furholt, Hinz, Mischka, Haug, Käppel and Müller2018). Such segmental organization is also clearly visible in causewayed enclosures such as Sarup (Andersen Reference Andersen1997). The aim and desire to construct a ditched enclosure system, or a megalithic grave barrow, may have been influenced by a wish to identify with other areas and regions, or with construction in neighbouring areas. Thus the constructions themselves may have affected the decision making (Furholt, Hinz and Mischka Reference Furholt, Hinz, Mischka, Haug, Käppel and Müller2018), something which is visible in a remarkable uniformity of forms and building techniques over the whole southern Scandinavian area.

Those few larger monuments which are laid out following a preconceived overall building plan – such as the larger passage graves – are then again used for collective burials, thus decidedly highlighting the group instead of some potential individual persons of authority. Most importantly, and moving beyond the Scandinavian example, while more outstanding monuments, such as Göbekli Tepe in Turkey (9600–7000 B.C.; see Schmidt Reference Schmidt2012), Mont St Michel in Brittany (4500 B.C.; see Scarre Reference Scarre2011) or Newgrange (3200–3000 B.C.; see Cunliffe Reference Cunliffe2004), could be seen as more convincing examples displaying top-down coercive domination as an organizational principle, there is no additional evidence backing up this claim. These monuments are not connected to any signal of more stratified political systems that would be visible outside the monuments themselves, such as in the settlement record. Some authors use the presence of exotic or valuable objects – e.g. jadeite axes in the context of early megaliths in Brittany – as indicators of wealth inequality (e.g. Gronenborn et al. Reference Gronenborn, Strien, van Dick, Turchin, Meller, Gronenborn and Risch2018), but this does not tell us much about power structures, or the nature of power involved. In other words, it is mainly the content of the notion of power on which the interpretational models are based that predetermines the outcome. If power is mainly seen as top-down and coercive, so is the resulting picture of social organization. Therefore a reconsideration of this concept, a broadening towards collective and consensus-based forms of power, will result in a richer account of prehistoric social relations.

The same can be said for the most obvious candidates for social stratification – large-scale population agglomerations such as Çatalhöyük (Hodder Reference Hodder and Hodder2013) or the Tripolje Megasites (Hofmann et al. Reference Hofmann, Müller, Shatilo, Videiko, Ohlrau, Rud, Burdo, Corso, Dreibrodt and Kirleis2019). Here, thousands of people lived in highly concentrated settlements, showing no signs of substantial social inequalities of material wealth or political power between their inhabitants (see also Green Reference Green2021). In these cases, archaeologists have proposed interpretational models which are well compatible with, or might even be said to necessarily rely on, collectively shared power to make sense of the apparent egalitarianism and decentralization (in the case of Tripolje sites, see Hofmann et al. Reference Hofmann, Müller, Shatilo, Videiko, Ohlrau, Rud, Burdo, Corso, Dreibrodt and Kirleis2019), or the lack of material inequality in the face of unequal access to symbolic capital (in the case of Çatalhöyük, see Hodder Reference Hodder and Hodder2013). In other words, when studying Neolithic construction projects or population agglomerations such as the ones mentioned, the evidence calls for a move away from the focus on individual leadership, towards the relational nature of those contexts and the effects and affects which the practices in question – the construction of a monument, the organization of daily life in a megasite – have in relation to the human and non-human actants involved.

During the South Eastern European Copper Age (4700–3800 B.C.), and more widespread since the late 4th millennium B.C., the situation concerning the possibilities for coercion and domination changed markedly. In the late 4th millennium B.C. in the Middle East, an organized military emerges (Hamblin Reference Hamblin2006), wielded by warrior kings. In Europe, the specialized warrior, equipped with specialized weapons, becomes a central social persona we find in the burial record and in iconographic presentations (Vandkilde Reference Vandkilde, Otto, Thrane and Vandkilde2006; Schulting Reference Schulting and Ralph2013). Nevertheless, we want to argue against the idea that the possibility of domination rules out the centrality of collective power. Continuing from the field of Neolithic research, the discussion of an increased degree of hierarchical organization is similarly a key topic within the Northern European Bronze Age (ca 1700–500 B.C.), where it is often seen as a continuation of the warrior ethos that emerges in the 3rd millennium B.C. (e.g. Vandkilde Reference Vandkilde2014). We do not deny the possibility of individualized asserters of power in this period. For example, richly furnished cists in monumental burial mounds are interpreted along these lines (Holst Reference Holst, Holst and Rasmussen2013), as is the introduction of the sword (Kristiansen Reference Kristiansen2002), organized warfare (Jantzen et al. Reference Jantzen, Brinker, Orschiedt, Heinemeier, Piek, Hauenstein, Krüger, Lidke, Lübke, Lampe, Lorenz, Schult and Terberger2011), and detailed iconography of individualized warriors on rock art and metal objects (Horn Reference Horn2019). Still, it is important to bear in mind that the display of wealth in burials primarily reflects what people involved in the burial ceremony were willing, and found appropriate, to give away, and the degree to which this willingness was coerced or based on collective power is what should interest us here. In any case, collective power will most certainly have played an important role, and will have been in accord with ethical principles, social doxa, moral principles and world view (Amborn Reference Amborn2019), while possibilities to exercise coercive power must be seen in relation to it and as being responsive to it. In addition, a model based on fear and control, or top-down coercion, in which the chief is perceived as the individual male possessor of control over violence, who deals out commands to his underlings, ignores the material realities of the Northern European Bronze Age, consisting mostly of dispersed settlements made up of single farmsteads. Moreover, tangible evidence of individuals’ movement patterns, such as the Egtved Girl and the Skrydstrup Woman (Frei, Mannering et al. Reference Frei, Mannering, Kristiansen, Allentoft, Wilson, Skals, Tridico, Nosch, Willerslev, Clarke and Frei2015; Frei, Villa et al. Reference Frei, Villa, Lørkov, Allentoft, Kaul, Ethelberg, Reiter, Wilson, Taube, Olsen, Lynnerup, Willerslev, Kristiansen and Frei2017), forces us to rethink top-down androcentric power (e.g. Reiter and Frei Reference Reiter and Frei2019), and look for alternatives in social organization. While it seems to be more or less consensus now to identify and acknowledge multiple agencies and social identities within a community, oddly enough, whenever decision-making processes are envisaged, agency is taken away from those multiple actors (e.g. Blanton and Fargher Reference Blanton and Fargher2008; Lund and Sindbæk Reference Lund and Sindbæk2021). In Bronze Age societies we identify varied organizational strategies that shift back and forth between coercive and more cooperative strategies (e.g. Skoglund Reference Skoglund2009; Austvoll Reference Austvoll2021, 17–22), while simultaneously the Bronze Age world is now populated with many different groups of social identities, as the studies of the Egtved and Skrydstrup Women testify, and which follow the same line of interpretation as the Beaker-associated Amesbury Archer (Fitzpatrick Reference Fitzpatrick2011) finds, which identify the presence of prospectors and mobile actors in the Bronze Age. That is why the classic models of coercive power operating in clear-cut hierarchical manners have always been challenging to neatly fit into the Bronze Age of Northern Europe. Here, we mostly deal with spread-out settlements and resources that are difficult to control by a single few. This has made several researchers argue for a ‘decentralized chiefdom’ or network system (e.g. Friedman and Rowlands Reference Friedman, Rowlands, Friedman and Rowlands1978; Kristiansen Reference Kristiansen1998; Earle Reference Earle2002; Kristiansen Reference Kristiansen, Price and Feinman2010). What is central for such a system, and often becomes overlooked, is that power does not really effectively reside with a powerful chief or within a selected elite. Because of their reliance on the other members of the community, agencies and in effect power are to a much greater extent collective in nature, resulting in less centralized forms of government (e.g. Blanton and Fargher Reference Blanton and Fargher2008; Carballo and Feinman Reference Carballo and Feinman2016; Feinman Reference Feinman, Chacon and Mendoza2017). The same can be said of the depositional practices in the Bronze Age, which are generally viewed as a contentious issue in terms of interpretation as they imply collective action, which does not match the individualistic interpretation of the graves. Nevertheless, the depositions are frequently analysed through the lense of organized elite rituals, which again contain individualistic top-down perspectives (cf. Fontijn Reference Fontijn2008; Garrow Reference Garrow2012). Yet in areas such as the Netherlands bronze swords are mostly found as depositions in wet areas (e.g. Brück and Fontijn Reference Brück, Fontijn, Fokkens and Harding2013), and in northwest Scandinavia, regions with limited amounts of individualized burial mounds often have rich metal depositions and organizational structures that are more cooperative in nature (Austvoll Reference Austvoll2021, 177–79). This refers back to the discussion of the Coast Salish, for which anarchistic theory was used to argue that the chief’s lack of coercive power, the decentralized political system, were a result of the possibilities, and the mutual interaction and support, of inner social groups rather than of the doings of the chief (Angelbeck and Grier Reference Angelbeck and Grier2012).

Within Late Iron Age studies, the last decade has shown an increase in the analyses of social identities. This has enabled an in-depth understanding of the number of varying social roles and positions within pre-state societies, which are now in the frameworks of interpretation populated not only with magnates, but also with various specialists within trades and production, including women with different social positions, and with a higher degree of sensitivity towards the various elements of hybridization in the relations between different social and cultural groups (see, for instance, Bergstøl Reference Bergstøl2004; Fredriksen Reference Fredriksen2006; Gardeła Reference Gardeła, Bedyński and Povedak2014; Hedenstierna-Jonson et al. Reference Hedenstierna-Jonson, Kjellström, Zachrisson, Krzewińska, Sobrado, Price, Günther, Jakobsson, Götherström and Storå2017; Lund and Moen Reference Lund and Moen2019; Myrberg Reference Myrberg, von Kaenel and Kemmers2009; Naum Reference Naum2008; Price et al. Reference Price, Hedenstierna-Jonson, Zachrisson, Kjellström, Storå, Krzewińska and Göther2019; Røstad Reference Røstad2016; Skogstrand Reference Skogstrand2016; Spangen Reference Spangen2005 for Scandinavian Iron Age and early medieval cases). Yet whereas the population in these societies is now perceived as consisting of a number of different groups, the social dynamics are still presented as top-down, based on the control of violence. Basing our considerations on a broadened notion of power, those social subgroups should be seen – with reference to collective-action theory, or anarchism theory – as members of society with varying degrees of collectively held power and differentially strong negotiation positions, and in general as taking part in the dynamics of social and political negotiations.

Similarly, whether we study the Roman period in areas outside the Roman Empire, the Migration Period, the Merovingian period or Viking Age Scandinavia, the presence of weapons in graves is almost exclusively interpreted as representing the personal belonging of the deceased (though these notions are challenged when the weapons appear in a female grave, as seen with the debate in archaeological journals following Hedenstierna-Jonson et al. Reference Hedenstierna-Jonson, Kjellström, Zachrisson, Krzewińska, Sobrado, Price, Günther, Jakobsson, Götherström and Storå2017), and thus as warrior burials. These weapon graves are furthermore generally interpreted as being of persons who were one of the personal armed companions of a leader, respectively a Roman period chief or a Late Iron Age magnate or king. Fundamentally, these societal models are still founded on the neo-evolutionist thinking of the 1960s (Service Reference Service1962; Fried Reference Fried1967). Within this interpretational framework, the power of the buried individual is manifested in the presence of the weapons as a means of control through the capacity to commit violence. As an example, the Roman period centre of power, Himlingøje on Zealand, Denmark, is interpreted as a maintainer of control, where nine weapon-bearing burials found in two belts scattered over the island of Zealand are interpreted as a defence system for the centre of Himlingsøje (Lund Hansen Reference Lund Hansen1995). Whereas a defence formed by nine dead warriors might give associations with a Lord of the Rings scenario in praxis, this interpretational framework can be challenged.

Overall, within mortuary archaeology the direct interpretation of grave goods as the personal possessions of the deceased has long been contested by focusing on the funerary context, where performance and display may have been important aspects (Halsall Reference Halsall1995; Parker Pearson Reference Parker Pearson1999; Schülke Reference Schülke1999). Acknowledging that the weapons were part of the staging of the funeral, in this context the possessions of a family used in the funeral could have been only part of the negotiation of social identities of the bereaved. Additionally, that the ritual actions performed by the participants of the burial determined the placement of the grave goods during the burial means that a reading of the grave goods as an index of the social status of the deceased must be disregarded. Indirectly, as in the Bronze Age discussed above, what such interpretations presuppose is that if another individual refused to follow orders, a warrior would use the sword to make the wish of the magnate take place. This scenario thus displays a view of these pre-state societies as communities built on fear. Granted that these buried individuals were indeed warriors, the focus on them as the wielders of the executive power – the executive power that uses violence as its main means of getting, possessing and keeping power – totally dismisses the collective nature of power, and is undermined by Arendt’s demonstration of the contrast and complementary relationship between collective power and violence. Most decisive, we would argue, is not so much the will of an individual person, as the collective consensus about who would have the right to put what items into a grave during a funeral ceremony, irrespective of what the actual outcome would then be in each individual case.

Similarly, if we examine the settlement structures of Iron Age Scandinavia, the field systems documented through the accumulated fences of the farmsteads demonstrate that a fixed, modularized field system may be identified in Iron Age Jutland, Denmark, in the first millennium A.D. This has convincingly been interpreted as the emergence of inheritance of land and property during the late Roman Iron Age (A.D. 180–400) (Holst Reference Holst and Bittman2010). Inheritance is by definition based on collective consensus in order to function. If inheritance had to be based on violence or on threats of violence, it would be extremely difficult to maintain across generations. The long continuity documented in the Jutlandic field systems indicates that they were maintained through consensus and are most likely the result of potential negotiations and acceptance, and thus of decisions made within a collective of people. Furthermore, the materiality of the fences and their deep-time history, maintained over centuries, affected these negotiations. Once established and maintained, their physicality and their capacity to live longer than humans affected the upholding of the field system. The settlement structures of other parts of Scandinavia, such as eastern Norway, do not show the same type of continuity (Gjerpe Reference Gjerpe2017). Here, the lack of continuity does not, however, appear to give any indication of a violence-based power. The settlement structures just did not evolve following one clear pattern, but show arbitrariness and diversity, and thus rather point to decentralized and unregulated societies. Furthermore, within an interpretational framework that puts emphasis on collective actions, on justified leadership and on the material world as influencing and affecting human actions, these interpretations of settlement structures are far better understood in terms of long-term changes within structures made by many different human beings. Also, they can be understood as actions that were influenced by structures of fences and fields that created ideas of inheritance rather than the result of individual decision making based on violence or threats of violence.

As we have pointed out, one major challenge within archaeological studies of social organization, including studies of power relations, is that it regularly builds on the modern national state, as either an ideal or a scheme. Yet ethnographic studies have demonstrated that social organization, particularly decision making, may have been structured in highly differentiated manners in various societies. For example, councils of different scope (village councils, elders’ councils, general assemblies) are widely documented in the anthropological record (Amborn Reference Amborn2019), and it is likely that they played an important role in many, if not most, prehistoric societies too (Graeber Reference Graeber2013). Some form of collective negotiation relying largely on collective power is just as likely to be the baseline of human decision making, often found in combination with the presence of chiefs or other individual positions of authority, and some degree of capacity for coercion. Collective and democratic forms of decision making appear to be a transhistorical constant, appearing in different social settings, reaching from villages and megasites in the Neolithic (e.g. Hofmann et al. Reference Hofmann, Müller, Shatilo, Videiko, Ohlrau, Rud, Burdo, Corso, Dreibrodt and Kirleis2019), to urban settings on all continents (Graeber and Wengrow Reference Graeber and Wengrow2020; Reference Graeber and Wengrow2021), to self-organized pirate crews in the 17th and 18th centuries (Markoff Reference Markoff1999; Linebaugh and Rediker Reference Linebaugh and Rediker2013), and to indigenous communities in the Americas (Calloway Reference Calloway1997; Graeber and Wengrow Reference Graeber and Wengrow2021), and of course as elements of modern Western state governments (Arendt Reference Arendt1970). Thus it may also be a relevant thought model for other prehistoric societies.

Concluding thoughts

It might be worth reiterating that our point is not to deny the fact that coercive power and individual rulers played important roles in prehistoric societies. What we want to emphasize is that the discourse on power in archaeology has a strong tendency to narrow down the understanding of power as individualized, male, coercive power, neglecting the role of collective, socially and materially embedded power. The latter, we want to argue, plays much more fundamental roles in every society than is usually acknowledged. It is indeed also a necessary foundation of all forms of political arrangement, even those involving individualized top-down coercion. On a more fundamental level, the emphasis on collective power corresponds with the notion of agency decentralized from the individual human agent, instead collectively distributed among human and non-human entities, as is currently much discussed.

This re-evaluation of the way in which power is interpreted in archaeological contexts may further be understood as part of the criticism of the fundamentally evolutionistic elements which still encapsulate studies of prehistory (see Graeber and Wengrow Reference Graeber and Wengrow2021). As long as power is treated as one type of phenomenon, which includes a movement towards power being placed in fewer hands who monopolize violence, culminating in state formation, the huge number of variations within societal structures and constellations throughout prehistory are overlooked – those very variations which enable us today to understand that there are other ways to live than those currently dominating. Better acknowledging the role of collective, embedded power, agency and affects will help move the focus of social archaeological studies away from the flawed fixation on chiefs and elite agency towards a broader picture of societal agencies, material affects and their mutual interactions. To make this more explicit and visible will, we believe, result in a very different, but much more interesting, account of prehistoric social and political organization than is usually promoted.