Posted by: Johan Normark | May 30, 2011

Hodder and relationism

In relation to my previous post I can with confidence state that relationism is a dominant ingredient in archaeology (and anthropology). This leads to several hidden assumptions in the way archaeologists treat objects. In a recent article by Ian Hodder he states that “any notion of ‘the thing itself’ is indefensible given the notion of co-constitution” (p 157). This, he suggests, depends on the anthropocentric accounts of objects. However, “perhaps one consequence of the human-centred approaches […] is that things have appeared to us rather directly as separate, bounded entities. This is indeed how they naïvely appear to us as humans” (p 157).

The assumption hidden in Hodder’s article is that there is no such thing as a thing itself because it is not a discrete bounded entity in space (and in time as well I may add). It depends on its external entanglement with other things and humans. However, from an object-oriented perspective relations can be either domestic (the internal structure of an object) or foreign (the way an object relate to another object). An object cannot change through foreign relations, only through domestic relations. For example, a car that hit another vehicle becomes a temporary limited object where the domestic relations change its component parts. After the crash the vehicles become separate objects again but with new domestic relations making them impossible to drive. This means that objects indeed are discrete bounded entities in both time and space (and it is simply not the effect of our “naïve” mind). Relations do exist but they are not what determine the object. The object is greater than its relations.

Hodder, Ian (2011). Human-thing entanglement: towards an integrated archaeological perspective. Journal of the Royal Anthropological Institute 17:154-177.



  1. Johan,

    Aren’t you and Hodder simply defining things differently? Hodder is making the (relationist) point that things that appear discrete and bounded to observers (like humans) are actually co-constituted in complex processes. That’s not an anthropocentric point at all, though it *could* be if one insisted that humans had to be part of every such relational nexus/assemblage/network/relation (though that would be absurd). There are many theorists (from Buddhists to Whitehead and Deleuze and Latour) who make more or less that point and who aren’t anthropocentrists in principle (though they may be in practice).

    You, on the other hand, seem to both conflate “relationism” with “anthropocentric relationism” (re Hodder’s argument) and conflate object-orientation with *non*anthropocentrism. One could be an object-oriented ontologist and still define “objects” according to human perceptions of what qualifies as such an object. While Harman et al don’t do this in principle, they sometimes do it in practice.

    In any case, it’s all about clear definitions. If an object is what you say it is – discrete and bounded, but complex, changing, and possibly self-motivated or autopoietic (like humans), then what you say about them is true. But that’s not the standard definition, and by the standard definition (which is that objects are objects *of* something – of perception, etc), Hodder is correct to say that objects aren’t really as discrete, bounded, etc as we think they are.

    These debates have, of course, been played out on other blogs. But I do think that definitions must be carefully articulated. If you’ve done this elsewhere in the archaeological context, then forgive my raising the questions unnecessarily.

  2. It appears to me that Hodder sees the idea of the discrete object as being the result of an anthropocentric perspective (well, he says “perhaps”, so he is not sure). I agree with him that objects are engaged in relations and assemblages but I do not believe that the very idea of discrete objects is the result of our human cognitive capabilities. Of course, you can be object-oriented from an anthropocentric perspective as well (Harman after all discusses objects in the works of Husserl and Heidegger), but that is not my goal since I believe that humans now and humans 100,000 years ago are not quite the same. The objects here in our present is only what we can deal with.

    I suspect that Hodder relates to the idea that we only observe objects in the present and therefore lack direct access to the past trajectory of the object, from when it was assembled from other parts and to the future when it will dissolve or disappear. From a Bergsonian/Deleuzian durational perspective the object would simply be the actualization of the virtual, always at the edge of chaos. But that is really not how our everyday objects “behave”. My table in my living room has not changed that much in several years (some scratches). Objects have come and gone on top of it, or the carpet below it, but the object is more or less the same as when I bought it.

    If I have understood Harman’s actualism correctly he suggests that objects should be stripped from their relation to the past as well (that may be the most problematic part for me concerning the object-oriented perspective since I come from the Bergsonian/Deleuzian/DeLandian “camp”). But nowadays (after reading The Prince of Networks) I also have problems with the idea of the object being subdued to its relations, even the temporal relations. From an archaeological perspective we usually have to deal with “snapshots” of the past that we link together through various relations (usually from a meta-narrative called “Culture” that simply fills the voids surrounding these snapshots). Earlier on I attempted to circumscribe this in-filling by focusing on the trajectories of the objects and the way that they form assemblages through relations of exteriority (which led me to the decision to remove the human as much as possible). Yet, now I realize that I am still filling in voids of the history of an object by referring to unactualized potentials of the object. By focusing on the “snapshots” themselves and not the supposed relations between them I can limit the in-filling of the past. The reason why I dislike the filling-in of voids is that it usually falls back on an anthropocentric narrative. Not that I am against humans but the variations of humans make it quite problematic to settle for a generalized image of the human. I prefer to decentralize the human instead.

    That objects are co-constituted in complex processes (or rather in assemblages of various durations as I perhaps would have phrase it before) does not mean that the objects themselves cannot be separated from their relations. I think that is what I wish to keep from Harman (but I am still not fond of just keeping the actual part).

    Regarding the definitions: yes, you are right. Right now I am struggling to see if my former assemblage perspective (a la DeLanda) can be meshed with Harman’s perspective. Relations are, of course, crucial here. Temporality as well. It seems, however, that the similarities are close enough. Maybe Bryant has the solution in his upcoming book?

  3. Isn’t this a question that has to be studied through different kinds of analysis before we can answer “yes” or “no”? Otherwise, do we need archaeologist to solwe the options?

  4. It depends on what you include in these analyses. Hodder exemplifies with several approaches of how clay was an important material. Still, the fundamental point is where this material is placed: is it an object or a process (is the clay the current state of a longer process)? It is of course a metaphysical question, not an archaeological one. I am sure that an archaeologist can make an argument for the “agency” of clay without resorting to metaphysics, but the basic ontological view will affect the whole study. I suspect that most analyses made by a multi-scientific group rely on the basic nature-culture divide. Hence, some study objects from a “cultural” perspective and others from a “natural” perspective . In the synthesis of these analyses one is supposed to bridge a divide that was initially placed on the object.

  5. “My table in my living room has not changed that much in several years (some scratches). Objects have come and gone on top of it, or the carpet below it, but the object is more or less the same as when I bought it.”

    That’s because the table was manufactured by humans to do precisely that – to remain stable over the length of time that humans would want to keep it as a table. Birds build nests, beavers build dams, humans build tables; none of them build themselves (unless you’d want to posit that nests conspire to get birds to build them, etc.). But without putting the table into an appropriate temporal context (a tree was chopped down and carved up and assembled by a tablemaker into a table, which eventually lost its leg and was used as firewood, etc.), you’re not really getting the story of the table *except* for the current human *use* of the table.

    “That objects are co-constituted in complex processes (or rather in assemblages of various durations as I perhaps would have phrase it before) does not mean that the objects themselves cannot be separated from their relations. I think that is what I wish to keep from Harman (but I am still not fond of just keeping the actual part).”

    That’s what I don’t understand. Why would you want to separate objects from their relations? What could an object, alone, explain? Isn’t every object that you are dealing with, by virtue of that fact alone, already in a relation with you? Take away that relation and you can no longer deal with it. What will be left if you subtract all the relations – even just all the *external* relations – of an object? A piece of pottery, traceable (using archaeological dating methods) to several thousand years ago, requires a certain climate, soil, atmosphere, etc. to maintain itself in the shape it’s in when you find it. Take away those conditions and there won’t be that object. All the more so for living organisms, which maintain themselves through interactive (and consumptive) relations with other things. Take those things away, and what’s left?

    I think it makes sense to debate the ontological *priority* of substances (i.e. Harman’s “objects”) versus relations. But to believe that one can exist entirely without the other seems like an untenable position.

  6. Of course the archaeological object is in a relation with me but the object is more than its current relation with me. I am not trying to strip the objects away from their relations, I only try to establish what I should give priority to, the object that I do find in the archaeological context or the entangled relations I assume it once had with other objects in its “systemic context” (the archaeological context depends on those conditions you mentioned). I have not clarified why I separate the relations within these different contexts.

    To me this is not just a matter of ontology but it is also an epistemological question. What can I really know about the object’s systemic context without loading it with qualified assumptions? Most people in the Speculative turn make use of contemporary examples and they take them from here and there. I have not yet seen any greater case study that runs for more than a few pages. Few have tried to operationalize these ideas at book length. I am interested in how these perspectives can replace the outdated culture/material culture concept (and this means more than a few anecdotes).

    Once you try to account for a whole past “society” from just objects as archaeologists always have done you start to run into practical problems if you have a relationist approach. Should I attribute this stone axe to gender, status, trade, fetish, tool, production, distribution, cosmology? All these relations probably existed but we can never know with any certainty. Let the context determine the relation we are told. That is not always the case since there is a whole bundle of black boxed relations that archaeologists project on the object a prior. We simply do not have human informants (and even if we had they would probably tell different stories about the same object).

    To me, then, the archaeological object should be given priority versus relations. That is not the normal procedure in archaeology where objects automatically are subordinated an arborescent typological schema or a cultural identity (a Maya stone axe). Hence, I am decentralizing ancient relations in favor of the object that has remained since those relations were active. If I did the opposite I would have to attribute the object with something I cannot directly observe (indirectly the relations can be accounted for – the form itself, where the axe was quarried, if it was moved a long-distance, if it was found in a burial or a house, etc.). In archaeological contexts that would automatically subdue the object to a non-present human or culture that always will be in a hierarchical/arborescent position above the object. Do we not run the risk of transcending the relations beyond the object rather than see it as immanent to various assemblages, networks or systems (that is my experience from archaeology though)?


%d bloggers like this: