The archaeological problematic

The separation of the ‘social’ from the ‘material’ is implicit for an archaeological methodology that treats material residues as the static evidence for past dynamics and distinguishes social/cultural dynamics from taphonomic processes (Schiffer Reference Schiffer1976). The material evidence has been used in three ways to bring into view extinct social categories of humanity situated within certain systemic contexts. The first is to signal particular social categories (e.g. by reference to ‘status’ items), the second is to demonstrate how the social world articulated with its environment (e.g. via the functional attributes of tools), and the third is to establish how certain social distinctions developed and were maintained (e.g. by mapping patterns of exchange and redistribution, and by identifying monuments and deposits resulting from competitive display). Thus have emerged the various images of social systems, operating through the use of material culture, and extracting resources from various environmental conditions (e.g. Binford Reference Binford2001).

All this has long seemed to be uncontentious. The separation between an active social world and passive material residue is enshrined in the concept of the archaeological record with the well-worn formula that a human social agency must have acted on, and thus moulded, the things that have survived. Archaeological materials have, as a consequence, come to be treated as the representations of a human agency's motivations (cf. Knappett Reference Knappett2005, 3–4).

The question of agency

Much of the debate that was prompted by the New Archaeology arose from the unresolved need to establish the motivation for the actions that are supposedly represented archaeologically. Given that we do not normally have testimony regarding such motivations, archaeology has fallen back upon various a priori assertions (cf. Bintliff Reference Bintliff, Bintliff and Pearce2011). Processual archaeology assumed that human motivations were self-evidently represented in their outcomes (the functionalist notion that things were made to do what they did), whilst postprocessual archaeology laid emphasis initially upon actions as expressive of a cognized order (things were the ‘symbolic’ representations of that cognitive order). Both processual and postprocessual archaeology have therefore founded their analyses upon a priori assumptions about the kinds of force and motivation that determine human behaviour, and it was into this debate that the concept of agency was introduced.

Agencies are mechanisms that have a material effect. One possible way we might identify an agency, therefore, is as a quality or mechanism that directs action, in the way a signpost or a manual directs a person's behaviour. This is similar to the way Gell treats ‘art’ objects as instigating an emotional and behavioural response in the observer (Gell Reference Gell1998). However, both Gell and Robb (Reference Robb, DeMarrais, Gosden and Renfrew2004) distinguish this kind of effect as secondary to the mobilization and application of the sign itself. In other words, the text that you are currently reading is a secondary mechanism mobilized by my own agency that desires to have an effect on the way you practice your archaeology. Obviously my agency can only have an effect through this text (if indeed it does) in virtue of the way that it is distributed across my typing, the work of commentators and editors, and the medium through which you are currently reading, as well as your own disposition as a reader. However, if the status of a ‘secondary’ or ‘extended’ agency were to be accepted, then the definition of agency sensu stricto would seem to be that agencies do work to mobilize the mechanisms that change material conditions, and it follows that agencies have the ability to do work in virtue of their ability to direct the expenditure of energy. Consequently, analysis should be very interested in distinguishing the different mechanisms by which such energy is managed and its expenditure compensated for (if indeed it is).

Postprocessual archaeology seized upon the concept of agency to establish the case that human agency operates at a level of autonomy independent of various environmental determinates. The implication was that the work undertaken by human agency was not secondarily orientated on behalf of some externally derived stimulus, but was internally motivated by the primacy of its own desires. Nonetheless, those desires were necessarily expressed through the strategic use of existing conditions, in the same way that to say something original requires the strategic use of an existing grammar. This last point was captured in the idea of a duality uniting an existing structure (grammar) with an agency's practice (talking) that could, over time, modify that structure. This is the cyclical renewal of the structure–agency duality then being developed in the social theory of Anthony Giddens (Reference Giddens1979; Reference Giddens1981). The potential problem is that human agency becomes just that – a quality seemingly possessed by humanity in general which is mobilized contingently by the use of available structures of rules and resources (to use Giddens's terminology). Thus an essential human quality (agency) is supposedly expressed with reference to specific structural conditions (cf. Casper Reference Casper1994). This perspective returns us to the mechanisms of a secondary agency where the human reach is extended, or distributed, by the use of various technologies and forms of material expression. Postprocesssual archaeology therefore claimed that the fundamental quality of human agency was expressed in historically specific ways, and that these were represented by the surviving arrangement of archaeological residues (Hodder Reference Hodder1982).

Processual and postprocessual archaeology made radically different claims as to whether human agency was a secondary product of certain external stimuli, or primarily motivated by internally formulated desires. Both start from conflicting uniformitarian assumptions, defining humanity as the possessor of different but nonetheless essential qualities. But is the claim that humanity is reducible to some essential quality not in need of a more critical evaluation? How might such an essential ‘human nature’, if it were to exist, have come into being, let alone have been maintained across the millennia? There is a growing suspicion that the humanity being studied in these different ways is all too easily taken as a given: it is what most archaeologists assume it to be, without considering how it might have been constituted historically. Perhaps we need to be more radical in what we hope to achieve archaeologically, making the ongoing creation of different kinds of humanness the historical problem that we need to investigate, rather than use archaeology to account for what a taken-for-granted humanity once did. The need for such an investigation is prefigured in calls for an archaeology of personhood (Fowler Reference Fowler2004), although this call can itself all too easily be reduced to treating personhood as a kind of social identity that clothed, rather than transformed, a form of life.

Archaeologies of life

The notion that humanity is not a given, but has to be brought into being, reorientates our use of the concept of agency away from being the possession of humanity to being the systemic context within which humanness was created. This finds parallels in the desire to question the dualism that distinguishes human agency as the already existing ‘social’ maker of history from the inert fabric of materials upon which that agency supposedly worked. This questioning has given rise to the demand to treat ‘symmetrically’ the contribution of material and biological things (‘actants’) to the rise of the various ‘entities’ that have emerged in history (Olsen Reference Olsen2003; Reference Olsen2007; Webmoor and Witmore Reference Webmoor and Witmore2008). The definition of such entities is, however, never entirely clear. John Law (Reference Law, Law and Hassard1999, 3), for example, suggests that they ‘take their form and acquire their attributes as a result of their relations with other entities. In this scheme of things entities have no inherent qualities’. However, if entities are entirely relational then things must exist that mobilize those relationships. This level of things must presumably possess some kind of primitive quality, for otherwise ‘that would deny any form of stability of different ontological kinds’ (Johannsen Reference Johannsen, Johannsen, Jessen and Jensen2012, 326). Entities are therefore emergent possibilities, born out of the networks created when one thing is brought into relation with another and, from this perspective, possible forms of life might be expected to emerge out of the networking of biological and inorganic materials.

If entities emerge from relationships between things, then we might accept that to treat humanness as an entity substantiates the view that, rather than being a given quality, it instead emerges as a product of the intersecting and changing relationships afforded between numerous organic and inorganic conditions. For Latour, this has meant that the social can no longer be isolated as a system of purely human relationships that are extracted from a background noise of objects and an ‘externalized’ natural world (Latour Reference Latour1993; Reference Latour2005). If social entities, whether presumably expressed as forms of human identity or in terms of ‘personhood’, have existed then they only emerged in the work undertaken by the entangled relationships of bodies, nature and things (cf. Hodder Reference Hodder2012). This would certainly explain why various abstractly formulated sequences of social types have never fulfilled their promise of revealing the mechanisms that drove the transformation of one type into another. The evolutionary process did not work in abstraction; it only worked through the reproduction of actually existing (i.e. historically specific) material conditions.

The recent argument that things achieve an agency, in as much as they can make a causal contribution to the process of creating entities, has been taken to imply that the agency of things is made manifest when one thing works upon another (Olsen Reference Olsen, Shanks, Webmoor and Witmore2012). Malafouris (Reference Malafouris, Knappett and Malafouris2008, 34) characterizes this as an agency that is distributed: ‘Agency is a property or possession neither of humans nor of nonhumans. Agency is the relational and emergent product of material engagement’. However, this claim only renders agency as a force or quality identified as a ‘product of material engagement’, which neither is specific enough for our purposes, nor clarifies what, if anything, instigates such an engagement (another agency?) (Johannsen Reference Johannsen, Johannsen, Jessen and Jensen2012, 333 ff.).

If we follow Malafouris's argument that agency is not a property of things but is afforded by the ways assemblages of things operate, then humans achieve an agency that is afforded by the various ways their bodies, materials and natural resources work with each other. Humans might thereby be regarded as developing and learning their competences as actors (their humanness) by the practical and embodied security that comes by simply getting on with growing and living amongst things. But humans also become self-reflexive; they think about their own subjectivities and act accordingly. Humanness is therefore defined by embodied actions relative to things and by the consequent practical and discursive knowledge of knowing how to go on, which arises from experience.

Language enabled modern humans to objectify their practical awareness of knowing how to proceed (Noble and Davidson Reference Noble and Davidson1996), but its possession, and thus self-consciousness, do not mean that consciousness itself is restricted to humanity. The process of consciousness that is able to objectify that which is known practically may be understood with reference to the threefold system of categories presented in the work of Charles Sanders Peirce. Whilst few are likely to claim that the writings of this philosopher are easy to access (Misak Reference Misak and Misak2004), and whilst Peirce's commitment to triadic classifications can appear to have verged upon the obsessive (Burch Reference Burch and Zalta2010), his system of categories works well for our purposes. It is important to recognize that Peirce's system is abstractly formulated as a system of logic (Hookway Reference Hookway1985), being heavily influenced by his familiarity with the work of Kant (Peirce Reference Peirce and Bucher1955, 2). Peirce is not concerned with the description of different types of thing (it is not a typology), but is instead concerned with the logical development of conscious dispositions that arise from the relationships between things. His work is thus committed to a phenomenological understanding of the way the world is given to an embodied being. In this scheme Firsts are things in and of themselves, Seconds are relations of impact or the clash between things, and Thirds arise in the practices that recognize the qualities made manifest in that clash. It is therefore the direct impact of one thing upon another that facilitates a conscious entity's ability to respond to the existence of certain qualities. Thirdness is emergent in the ways that categories of quality, revealed by the hybrid relationships of organic and inorganic things, are given to consciousness and made manifest in practice. Different forms of consciousness might be distinguished in the ways that they clump together qualities that seem similar or appear to be of the same kind despite arising from different material relationships (Seconds). This hierarchical scheme therefore clearly distinguishes between consciousness as a general and emergent quality arising in all forms of life (for example, animals identify the qualities of danger, security, sustenance, sexuality, solidarity, and so forth) from assemblages of non-living, non-conscious things, such as chemical reactions (contra Knappett Reference Knappett2005, 12 ff.).

Life in all its forms is thus a particular level of organization within the wider assemblages of things with which it engages. Organisms are metabolic systems that make themselves, facilitating growth, cellular renewal, and the self-production of their internal order. Organisms and machines certainly both do work, but the work of the organism, unlike that of the machine, is directed towards its self-affirmation and renewal (Thompson Reference Thompson2007, 140–49; cf. Johannsen Reference Johannsen, Johannsen, Jessen and Jensen2012, 328–31). It is difficult to understand why this distinction appears to be so easily missed (cf. Olsen et al. Reference Olsen and Hodder2012). Maturana and Varela (Reference Maturana and Varela1980; Reference Maturana and Varela1987) defined autopoiesis as the condition expressed by the simplest form of organism operating as an autocatalytic system with a semi-permeable boundary whose internal structure, workings and boundary conditions are self-produced. The boundary of an organism is necessarily semi-permeable because the organism can only exist relative to sources of energy whilst maintaining its internal organizational integrity. This means that, while necessary, the organism's genetic inheritance is not sufficient for the ontological process of growth. The information encoded within the genome has to be utilized to build the proteins of the cell by means of the complex internal process of translation. The latter depend upon, and are guided with reference to, a number of external, environmental factors (Oyama Reference Oyama2000; Oyama, Griffiths and Gray Reference Oyama, Griffiths and Gray2001). Forms of life can therefore only exist, grow and develop the particular qualities that they display in their agency by being intimately engaged within the particular conditions provided by their environments. ‘[A]utopoiesis always has to be ecologically embedded. “Self-producing” refers to the kind of circular organization that makes the cell an individual: it does not mean that the cell makes itself apart from its environment’ (Thompson Reference Thompson2007, 118). Thus the development of a living organism arises from an internally organized process that requires the organism to orientate itself towards the environment in ways that provide it with the necessary energy and security for its own development. It follows that all life is conscious by virtue of the ways in which it is practised as a subjective awareness, displaying intentionality in the phenomenological sense of being directed towards the conditions that will sustain it (Thompson Reference Thompson2007). As a consequence, forms of life will act in ways that sustain and enhance the niche conditions necessary for their survival (Odling-Smee, Laland and Feldman Reference Odling-Smee, Laland and Feldman2003); their actions in doing this work (expressive of their agency) are a form of ‘sense-making’ (Thompson Reference Thompson2007, 152 ff.). Hominin life forms are subsumed within this general characterization of life, although it has been through the evolution of their particular physical characteristics, and their inhabitation of the material debris of an increasingly enculturated world, that different kinds of humanness have become possible.