Posted by: Johan Normark | September 21, 2011

Animism and polyagency

While travelling in Malaysia during the summer I read Bennett’s book Vibrant Matter and for several reasons it made me think of my “old” idea of polyagency because it resembles her idea of “thing-power”. Levi Bryant’s recent post about vitalism and animism made me decide to summarize this idea five years after I began to abandon it in favour of DeLanda’s assemblage theory. Despite the theoretical jargon, that may be off-putting to archaeologists, polyagency attempts to explain the same thing as animistic ontologies do.

First of all, animism is usually seen as a relational ontology that focuses on the connection between human and non-human environments. Animistic forces in the Maya area are found in objects, buildings, and caves. The objects are alive and born into the social setting. For example, the och-i k’ahk’ “fire-entering” collocation refers to a ritual that gave buildings a soul. Buildings were not only animate, they were described as named cosmic beings, such as Structure 44 at Yaxchilan, which was the Starry Deer Caiman (Plank 2003). Now, this is not just mythology because Boyer (1996) notes that an entity can possess “intentional psychology” without being alive. Thus, “agency” is not defined from biological attributes, it is relational, and it depends on a network of social relations according to Gell (1998).

Animistic ontologies are usually geocentric (they are after all presignifying and indexed to the Earth according to Deleuze and Guattari). In a geocentric worldview, not only is the Earth in the center but it is also the largest object (Brady and Prufer 2005:366). In the Kekchi area, hill-valleys are given names and personalities, which Brady and Prufer sees as a personification and animation of the landscape. Kekchi says that mountains have a wiinqilal “personhood”, which is only applicable to humans and mountains because these have a spirit (Brady and Prufer 2005:367). Objects carry a power to affect other objects with or without human intervention.

In my licentiate thesis (2004) and dissertation thesis (2006) I saw polyagency as something akin to “material agency” or Latourian actants, the way an object affects its surroundings. Polyagency is largely based on Bergsonian “vitalism”, mixed with my first encounter with Deleuze’s writings and Alfred Gell’s study of agency.

In short, polyagency is a collection of intensive processes that lie in-between the virtual and the actual. It is a distributed agency that lacks an identity of its own, but it affects other identities. It is in the in-between where individuation takes place, where virtuality closes onto itself and forms an actualized boundary to what is external. Polyagency both generates polyagents and is part of polyagents . Simply put, polyagents are actual entities. Because polyagency exists in-between parts of polyagents it lacks a defined temporal and spatial location. It is found everywhere in the object but its capacity to affect extends beyond the actual boundaries of the object. Hence, polyagency is a space without a space. If an object is broken, its polyagency is not broken. The polyagency has just differentiated and relates to two objects or actualizations that each have different tendencies, but that maintain the same original virtuality. This means that the whole Earth has polyagency and all its objects also have portions of this polyagency, pretty much what is argued for animism.

Things, causeways, animals, plants and humans all have polyagency and therefore are polyagents. The only distinction I make is between “normal” polyagents and indexical polyagents. The indexical polyagent as a concept is influenced by Alfred Gell’s (1998) use of the index in his art studies. Seen from an artist’s whole oeuvre, a painting in 1950 is an index/retention of a painting the artist made in 1940 and the 1950 painting is a prototype/protention for a painting made in 1960. At the time of death the artist’s whole oeuvre can be seen as one widely distributed object (perhaps a hyperobject?) where the different works of art are retentions and protentions of one another.

In my dissertation thesis I transferred Gell’s ideas to both lower and higher assemblage scale by exemplifying with the causeways at Ichmul and Yo’okop. A causeway is simply index or prototype of another causeway or platform (a causeway is more or less an extended platform). Here I reduced the human architects and constructors to catalysts for the emergence of causeways at these two sites. Once laid out and constructed in the landscape they affected the way people moved, how water flowed, and agricultural practices centuries later. For example, Sacbe 3 at Yo’okop has accumulated deeper deposits of soil on the side that faces a slope. This makes the soil moister and milperos (slash-and-burn agriculturalists) today receive higher yields from crops grown on that side of the causeway compared to on the other side of the causeway. The causeway’s polyagency is still around but it is actualized into an agricultural assemblage rather than the original ceremonial assemblage that Sacbe 3 most likely was intended for. The causeway is alive and gives life from an animistic perspective.


  1. Hi Johan,

    Fascinating ideas! It’s interesting that I’m working with Gell’s ideas coming from a more traditional social archaeology perspective, yet I find the ideas you outline here resonating very well with my own thinking. Not sure whether I want to go into Bergson, Deleuze etc., I read more Wittgenstein.

  2. Gell’s focus on both time and art make him ideal for a Deleuzean perspective. On the other hand, Wittgenstein influenced Bourdieu and agency theory in general. The reason why I prefer Bergson and Deleuze instead of Wittgenstein is that they focus less on language. I feel that they are more in alignment with the data archaeologists usually work with. However, Göran Aijmer’s use of Wittgenstein has also inspired me, particularly his “iconic order”:


%d bloggers like this: