Actor-network theory claims to be just description. Is this the case, and if so, is that good or bad?
To theorise is to is to build on a supposition, to develop a system of ideas with the intention of explaining something. A theory at once builds from a set of accepted beliefs and creates new ones. Theories thus often echo, answer, challenge each other. The issue of escaping the confines of theory is of note. Actor-Network Theory (henceforth ANT) interestingly claims to not be a theory at all, claiming to describe not explain, thus venturing beyond commonly accepted beliefs.
This essay will assess ANT’s claim to be ‘just a description’ and evaluate, if this is the case, whether that is good or bad. Firstly, it will look at the ways in which ANT can rightly claim to be purely descriptive. Then, it will focus on how nothing can ever be simply ‘described’, thus trapping ANT, despite its best efforts, within the confines of theory. Lastly, it will briefly address the issues behind ANT’s claim of being ‘just description’.
There are a variety of reasons why ANT can assuredly claim to be just description. For instance, by its somewhat adamant rejection of explanation. Indeed, proponents of ANT express the desire to look beyond established social categories, to abandon classification. Latour (2006) brings to light the inherent flaw of social categories, which is to accelerate description, thus overshadowing the processual nature of cause and effect and embedding observations within larger narratives in which they might not even find a place. He argues that ‘sociologists of the social [pronouncing] the words ‘society’, ‘power’, ‘structure’ and ‘context’ [often jump] straight ahead to connect vast arrays of life and history … to detect dramatic patterns emerging out of confusing interactions, to see everywhere in the cases at hand yet more examples of well-known types’. The danger of this acceleration is that it can do nothing other than further strengthen pre-existing narratives, casting a shadow on what does not fit within these readily available classifications. The value of being purely descriptive is to avoid casting such a shadow. To describe enables processes to be revealed, in their own time, and on their own terms. Proponents of ANT thus tend to pride themselves on their impartiality. In this, they edge away from theory insofar as instead of ‘taking a reasonable position and imposing some order beforehand, ANT claims to be able to find order much better after having let the actors deploy the full range of controversies in which they are immersed’ (Latour, 2006). They do not see their task as being one of intercepting facts in order to decide where, within the available theoretical knowledge, they best fit; rather they simply aim to transcribe the facts. Further, in line with this strive towards impartiality, actor-network ‘theorists’ avoid prescribing fixed roles or identities and avoid the presupposition of hierarchy or relevance. Callon (1986) argues that the observer not only must be impartial towards the scientific and technological arguments used by the protagonists of the controversy but ‘also [abstain] from censoring the actors when they speak about themselves or the social environment. He refrains from judging the way in which the actors analyse the society which surrounds them. No point of view is privileged and no interpretation is censored’. Hence, humility goes hand in hand with impartiality. Indeed, to claim one’s aim as purely descriptive entails a degree of evening of the playing field between actors and observers. If the aim is to describe, then one must not feel superior to those observed. Latour (2006) highlights this in saying that to observe as an actor-network ‘theorist’ is to say to the actors ‘we won’t try to discipline you, to make you fit into our categories; we will let you deploy your own worlds, and only later we will ask you to explain how you came about settling them’. In short, Latour posits that ‘the task of defining and ordering the social should be left to the actors themselves, not taken up by the analyst’. Importantly, vocabulary is taken as a marker of success in this task insofar as in ANT vocabulary is chosen to specifically conform with the descriptive aim rather than try to explain, alter, speak for, or silence actors. Latour suggests that the ideal employment of vocabulary in ANT is one that is ‘banal’ so as to eclipse the risk of confusing the ‘actor’s own prolific idioms’. He explicitly cites his opposition to social theorists in this respect by arguing that they are ‘keen to produce precise, well chosen, sophisticated terms for what they say the actors say. But then they might run the risk of confusing the two meta-languages – since actors, too, have their own elaborate and fully reflexive language’. He adds, ‘if they practice critical sociology, then there is an even greater risk to render actors mute altogether’. Latour explains that what is needed is the development of an ‘infra-language’, and that one must at all times try to be in the shadow of the actors, not take centre stage, and ask ‘are the concepts of actors allowed to be stronger than that of the analysts, or is it the analyst who is doing all the talking?’.
Thus, by letting ‘actors deploy the full range of controversies in which they are immersed’, actor-network ‘theorists’ do not impose a binary, a field of study, whose relevance lays in previously subscribed to beliefs and categories of thought; rather, complexity is embraced. Therefore, ANT tries to reveal the process, from cause to consequence, as faithfully reported as possible, instead of just making assumptions based on the consequence. Callon (1986) reports that what is important is that actor-network ‘theorists’ systematically force themselves to ‘judge neither the positions taken by the actors nor to reduce them to a particular ‘sociological’ interpretation’. Additionally, he adds that conclusions can only be drawn at the end, based on thorough observation and validated in and by itself in the faithful description of what was observed. Thus in his particular context of observation, that is the domestication of scallops in St Brieuc Bay, ‘the existence or the non-existence of the anchorage or of this social group may only be determined at the end of the course which was followed and it is the three researchers who reveal this through their different endeavours’. Accordingly, what ANT attempts to do is render visible what is commonly cast to the shadows; the description can be likened to a translation of sorts.
Yet, this is precisely the problem. No matter how good a translation, it will always present an innate bias. The act of translation is unique, contingent on the translator and the choices they make. They might choose one word over another, one turn of phrase they believe best encapsulates what was said, instead of translating word for word what was said, insofar as word for word translation does not always make sense. This idea can be extrapolated and applied to actor-network theorists. Despite their best efforts to be untainted by sociological preoccupations and guided by pre-established categories, actor-network theorists must make choices. Indeed, ANT claims to be just description, unaffected by pre-conditioned expectancies of relevance. Latour (2006) explains that ANT is unlike other theories because it does not limit its own scope of inquiry. He highlights that many sociological enquires begin by ‘setting up one – or several – type of groupings, before apologizing profusely for this somewhat arbitrary limitation made necessary … by the ‘obligation to limit one’s scope’ or ‘by the right of a scientist to define one’s object’. ANT characterises its difference by suggesting that it does not feel the need to ‘stabilise – whether at the beginning for clarity, for convenience, or to look reasonable – the list of groupings making up the social’. However, while this may be an obvious point, even actor-network theorists must start somewhere; and although their starting point might not be the conventional one, it is still a choice made. This creates a sort of paradox of ANT being, on the one hand, against the idea of limiting one’s scope, and on the other, faced with the reality of the impossibility of the omnipotence they seek in their description. Similarly, as Shapin (1988) notes, ‘since ‘actor networks’ are heterogeneous, they may extend anywhere in nature or in society’. Indeed, ANT sees one of its qualities as being able to go off the beaten track, exploring new associations usually left unexplored by restrictive social categorisation and strict binarism. Nonetheless, Strathern (1996) outlines that ‘the power of such analytical networks is also their problem: theoretically they are without limit’. She continues by underlining that while ‘analysis appears able to take in account, and thus create any number of forms [and] one can always discover networks within networks [this being] the fractal logic that renders any length a multiple of other lengths, or a link in a chain of further links’; ‘analysis, like interpretation must have a point; it must be enacted as a stopping place’. As such, networks, despite their seemingly infinite scope, must be cut at some point, and this is always a choice. Yet, while it may appear that choosing where to start and stop a network does not detract from it being ‘just description’ (indeed, descriptions begin and end somewhere), ANT stops being purely descriptive because of its latent biases and inherent power.
It is impossible to throw all categories of thought out the window. Amsterdamska (1990) underlines that the inescapable shortcoming of ANT is that while is presents itself as rejecting the constricting binarism and limiting categories, they can never really be escaped. She argues that history has taught us that no one ever succeeds in creating a ‘naïve outsider’ and contends that ‘Latour’s ‘dissenter’ is no exception. Even though he has had to ‘abandon knowledge about knowledge’, he certainly has not forgotten what war and politics are all about; and equipped with this Machiavellian view of the world around him, he attempts to interpret science in the only terms he knows’. Moreover, it can be argued that perhaps such ingrained beliefs and categories of thought which shape individuals regardless, can never really be rejected, only opposed. Yet, to be opposed to something is to, in some way, subscribe to the narrative; it is a response, a realisation that something is lacking and must be bettered. Thus, in this regard, ANT is not a disinterested description, but rather subtly points to where other theories are lacking.
Further, the question of power is of note. Latour (2006) seemingly contradicts himself by suggesting that while the actors should take centre stage, the observers should also take centre stage. In fact, he claims that ‘in ANT, if you stop making and remaking groups, you stop having groups’. As such he posits that ANT is not descriptive but rather active in contrast to traditional social analysis and that this is ‘one of the essential differences between the two schools of thought’; ‘for the sociologists of the social, sociology should strive to become a science in the traditional disinterested sense of a gaze directed to a world outside, allowing for a description that is somewhat independent of the groups being materialised by the actors. For the sociologist of the associations, any study of any group by any social scientist is part and parcel of what makes the group exist, last, decay, or disappear’. This insistence on the performative offers stark contrast to the ideal of objective description. Exactly because ANT takes on this performative and active tinge, where boundaries of what should or should not be studied, what is relevant or not, are constantly negotiated, it ultimately becomes transformative in much the same way a theory could claim to be. It creates a narrative, extending from previous ways, offering new perspectives, new ways to think with and to think of things, conferring it a form of power that a simple description would not have. For example, Candea’s account of fire and identity as matters of concern in Corsica (2008) leads to a breakdown of usual classifications, thus enabling new perspectives to be acknowledged. He suggests that relationship to land varies not inherently along the binary of ‘insider’ and ‘outsider’ but rather ‘relationship to land varies substantively according to which connections are made, and to what. In a context in which spectatorship is always-already participation, insiderness and outsiderness are not stable states that can be found sitting quietly at opposite ends of a continuum. Each person will find something slightly different as the connections accumulate’. Further, as Marcus (1995) notes, multi-sited ethnography, which follows the process instead of just focusing on the outcome, ‘investigates and ethnographically constructs lifeworlds of variously situated subject, it also ethnographically constructs aspects of the system itself through the associations and connections it suggests among sites’, thus suggesting a form of mutual dependency between description and explanation.
Additionally, it is interesting to think of how power is implicit not only in exclusion but also in inclusion. Indeed, to include what is usually excluded from analysis is powerful and transcends mere description. Shapin (1988) shows that ‘Latour’s erosion of the conventional boundaries separating politics from science is predicated upon the insistence that objects and non-human entities as well as people are political beings. Things belong to the study of the political order as much as human agents’. It is in claiming this that ANT cannot be just description – in fact Latour (2006) argues that, in a way, it is precisely because it presents as a description that it gains political significance. In fact, Latour suggests that ANT is perhaps more radical than traditional sociology in saying that raising a ‘political question often means to reveal behind a given state of affairs the presence of forces hitherto hidden. But then you risk falling into the same trap of providing social explanations … and end up doing exactly the opposite of what I mean here by politics. You use the same old repertoire of already gathered social ties to ‘explain’ new associations. Although you seem to speak about politics, you don’t speak politically. What you are doing is simply extending one step further the same small repertoire of already standardised forces’. Yet, in criticising social science for adding to the mass of readily available knowledge about ‘what’ the social ‘is’, which actually consists also in deciding what the social ‘should be’, Latour suggests that ANT breaks free from these restraints, yet perhaps it simply deviates from the conventional idea of what the social is or should be, but still informs it.
To address whether ANT claiming to be just description is a good or bad thing, there are a few things to consider. Firstly, why does ANT feel the need to claim this status? Indeed, the fact that it actively seeks out the descriptive quality as characteristic of itself is, in itself, interesting. There is a certain innocence, a form of purity, associated with the idea of simply ‘describing’ something, instead of explaining it. However, what is the use of pure description? Surely, behind describing something there is an aim, a purpose. In this case, it can be argued that the aim is to ‘reject’ social theories in place and make way for something new. Yet, this can be misleading because in a way, this is nothing more than a disguised ‘third way’. As Bloor (1999) notes, by explicitly rejecting what is currently in place, Latour no more and no less presents a vision, his idea being that ‘we must not try to explain nature in terms of society, or society in terms of nature, nor should we explain knowledge as a mixture: we must explain both society and nature at once, in terms of a third thing or process’. Thus, while the methodology might be heavily based on description, in a way which breaks from convention, there is, however latent, a desire to explain, reveal something, otherwise description for the sake of description wouldn’t be taken up.
Further, there is the issue of the distance produced by claiming to be ‘just description’. Amsterdamska (1990) raises this issue in addressing Latour’s rejection of explanation. She claims that Latour assertion that ‘the ideal of explanation … is not a desirable goal’ and that what should be aimed for instead is ‘telling stories’ would lead to an abandonment of ‘all responsibility for what we are saying’. Further, she asks whether any stories can truly be ‘so ‘innocent’ that they could not be regarded as stratagems in a struggle for power and control? To her, ANT is inherently limited, and unambitious, if it claims the way forward for social science to be telling ‘inconsistent, false, and incoherent stories about nothing in particular’.
Shapin moreover questions the utility of something aiming to be ‘just description’. He argues that the world ANT is creating is the world as a seamless web, ‘a world in which everything is connected to everything, in which even the discrete existence of things and the categorisation of processes cannot be used to interpret or explain the actions of those who are said to produce them’ (1988). He further suggests that ‘there is much to be said in favour of the monistic impulses and the close inspections of seams, but there is little to be said from within a seamless web’.
In the same vein, Bloor (1999) questions the value of ANT as ‘just description’. Arguing that Latour’s attempt to ‘get to metaphysical bedrock doesn’t work’ because he cannot ‘get away from a pre-existing nature and a pre-existing society’, Bloor suggests that to want to simply and impartially describe processes might not be the way forward, and perhaps, instead, ‘we have to begin our investigations into the nature of knowledge from where we are standing. Our feet are on the ground of nature, and our position is in the midst of an existing culture, our own culture. This is, perhaps, not quite as limiting as it may seem, or as Latour paints it, because we don’t have to take our culture entirely at face value or respond to it uncritically’.
To conclude, for ANT to be just ‘description’ begs the question of what a ‘pure’ description would entail. Yet, one quickly realises that such pure descriptions might be constrained to a conceptual existence. Indeed, very rarely does anyone describe something just to describe it; behind it usually lies the aim of understanding, rendering visible. As such, for ANT to claim itself as just description is thus misleading insofar as achieving that would mean being impartial, unbiased, disinterested, distant, and unmotivated in a manner quite impossible. Beyond asking whether it is good or bad lies the question of value; what would be the point of something purely descriptive, why would someone go to the trouble of describing something if not to challenge current views, offer alternatives.
Bibliography
Olga Amsterdamska (1990) ‘Surely You are Joking, Monsieur Latour!’ (Review of Science in Action). Science, Technology, and Human Values, 15: 495-504.
David Bloor (1999) ‘Anti-Latour’. Studies in the History and Philosophy of Science, 30: 81- 112.
Michel Callon (1986), ‘Some Elements in the Sociology of Translation: Domestication of the Scallops and the Fishermen of St Brieuc Bay’. In John Law (ed.) Power, Action and Belief: A New Sociology of Knowledge? London.
Candea, M. (2008) ‘Fire and Identity as Matters of Concern in Corsica’. Anthropological Theory, 8: 201-16.
Candea, M. 2018. No Actor, no Network, No Theory: Bruno Latour’s Anthropology of the Moderns. In Schools and Styles of Anthropological Theory (ed) M. Candea. London: Routledge.
Bruno Latour (2006) Reassembling the Social: An Introduction to Actor-Network- Theory. Oxford: OUP
Emily Martin (1998) ‘Anthropology and the Cultural Study of Science’. Science, Technology, and Human Values, 23: 24-44.
Marcus, G. E. (1995). Ethnography in/of the World System: The Emergence of Multi-Sited Ethnography. Annual Review of Anthropology, 24, 95-117.
Mitchell, T. (2002) Rule of Experts: Egypt, Techno-Politics, Modernity. Berkeley: University of California Press.
Steven Shapin (1988) ‘Following Scientists Around’. (Review of Science in Action). Social Studies of Science, 18: 533-50.
Strathern, M. (1996). Cutting the Network. The Journal of the Royal Anthropological Institute, 2(3), 517-535.