Art and Agency presents an anthropological theory of art. Gell defines an anthropological theory of art as “social relations in the vicinity of objects mediating social agency” (7).
Gell argues that an anthropological theory of art must focus on the social contexts of art. A social approach is needed in order to examine “the social context of art production, circulation, and reception, rather than the evaluation of particular works of art” (3). Production and circulation are sustained by other social processes like exchange, politics, religion, and kinship. An anthropological theory of art should not elucidate western vs. non-western aesthetic systems. This is a cultural approach to art. Moreover, it assumes that a culture has a universal parameter on which art is produced. To understand art, Gell argues historians must examine the social relationships “between participants in social systems of various kinds” (4). The need for social relationship in art becomes clear when one tries to define art.
Gell rejects the normative definition of art as “whatever is treated as art by members of the institutionally recognized art world” (5). He also rejects the term “art objects” to describe art works because it assumes that “objects are sign-vehicles, conveying ‘meaning,’ or they are objects made in order to provoke a culturally endorsed aesthetic response” (5). The phrases art and art object do not allow objects to act in social relationships. Gell recognizes that objects have agency, intention, causation, and that they are effective and transformative. Thus, Gell seeks to understand the active and mediatory role of objects. He suggests that there is no difference between bodies and artefacts. Gell defines these objects, which are equivalent to persons, as “social agents” (7). Gell argues that to understand social agents, historians must examine the biographical elements, or their life-stages of social agents.
Gell calls a social agent (or art, art object, work of art, etc.) an index. An index is “seen as an outcome, and/or the instrument of, social agency” (16). Gell argues that agency is attributable to persons and indexes. Agency is the attribution of intention to a person or thing. Gell defines things, or indexes as primary and secondary agents. Primary agents are “intentional beings who are categorically distinguished from ‘mere’ things or artefacts” (20). Secondary agents are “artefacts, dolls, cars, works of art, etc. through which primary agents distribute their agency in the causal milieu, and thus render their agency effective” (20). Primary and secondary agents work by distributed personhood.
The idea of distributed personhood allows historians to see the distribution of primary agents’ agency through secondary agents. For example, “as agents, they [soldiers] were not just where their bodies were, but in many different places (and times) simultaneously [as mines]. Those mines were components of their identities as human person, just as much as their fingerprints or the litanies of hate and fear which inspired their action” (21). Calling things secondary agents, according to Gell, does not mean they are not agents. It means they are not primary agents “who initiate happenings through acts of will for which they are morally responsible” (20). Secondary agents are “objective embodiments of the power or capacity to will their use” (21). This “objectification in artefact-form is how social agency manifests and realizes itself, via the proliferation of fragments of ‘primary’ intentional agents in their ‘secondary’ artefactual form” (21).
Agents work in a network of social relations. Agents must have a patient. The patient is the “object which is causally affected by the agent’s action” (22). Primary or secondary agents can act as the patient while the other acts as the agent. Manufactured objects are indexes of their makers, or their artists. Sometimes the artist and the index’s origin are forgotten or concealed. The recipients of indexes are “in a social relationship with the index, either as “patients”…or as ‘agents’ in that, but for them, this index would not have come into existence (they have caused it)” (24). An index must always have some specific reception or recipient. This network of social relationships also includes the prototype. The prototype of an index identifies “the entity which the index represents visually (as an icon, depictions, etc.) or non-visually” (26).
Gell argues that secondary objects have agency in a social network made of humans and other objects. This agency can be witnessed through a biographical examination of objects in relationships. Despite these claim, Gell does not allows secondary agents to have agency. They are always bound to humans or patients. They can only have agency through a primary agent’s agency, or distributed personhood. Moreover, secondary agents are not morally responsible for anything. Gell tries to give objects agency, but takes it back when he makes them rely on humans. Objects are mediators of social agency, not actual agent themselves.