42. Turning pragmatics around: how to say things with objects

This post is also available in: Catalan

Amadeu Viana
Universitat de Lleida


When Alice complains about too many meanings being given to a word, Humpty Dumpty explains that when he makes a word do so much work, he always pays it extra. That is exactly what happens: words work for us, often too much, we move them into different contexts and make them adopt other meanings, we extend them with prefixes and suffixes, or we save them from oblivion when a new state of things emerges and we want to bedazzle with an old name. Similar things happen with the objects that surround us: we turn around their semantics, we rework them via intensive bricolage, we preserve them by invoking their primary function, or we make them undergo such striking transformations that the old denominations become but a reminder of their origin. The idea that human actors share a universe with nonhuman actions and processes is known as distributed agency. This supposes that human actions are effectively carried out by means of and through the operativity (the action) of different types of intermediations, material objects, networks of objects and artefacts, institutions, cooperation or shared action, symbolic activity and extensive collaboration with nonhuman actors, a battery of processes that elude the predominant Western philosophy’s traditional view of the idea of a single agent.

The idea of distributed agency stems from Bruno Latour’s transdisciplinary research and actor-network theory, a practice from the sociology of science that focuses on the role of the material transmission of scientific objects, cooperation or (conversely) controversy among institutions, the laboratory as the locus of intersection, the dissemination of science and the role of the different “translations” that concepts are subjected to from their birth and as they spread across society. Actor-network theory has contributed to a reassessing of the scientist’s role in the abstract as an autonomous agent in favour of a perspectivist approach and collective action.

It might be useful to shore up the notion of distributed agency with bioanthropological reflection, which reminds us that the manipulation of artefacts and the construction of objects long precedes the emergence of articulated language. The topic here is how growing tactile- and visual-coordination capacities laid the foundations for cognitive evolution (L. Malafouris) in a way that meant that the hand would become the visible part of the soul, as Aristotle would ingeniously point out. Objects and their meanings accompanied human evolution, complexifying and interlinking with our consciousness in such a way that it would soon become difficult to say where one thing began and another finished (A. Leroi-Gourhan), in the intricate combination of homo faber (who makes himself in making) and homo ludens (who unsuspectingly extends the time for play, experimentation and learning). Ornaments, clothing, living conditions, musical instruments (the flute mentioned at the beginning of the Aristotle’s Poetics), the variety of artefacts that have been part of life and societies: these are manifestations of human consciousness and extensions through which cognition and agency have earned a presence and become widespread and enduring.

As Cassirer highlighted, the different linguistic varieties owe a debt to this confusion of bodily projections, visual and manual organization, alienable and nonalienable objects, practical metaphors and occasional metonymies, bodily self-perception and classification of space: a production of the Palaeolithic that is also sufficiently complex as to disassociate ourselves from the immediate context. Those who study embodiment and extensionalism (A. Clark, M. Donald) have highlighted the idea that cognition is constructed and sustained from the bodily senses plus the objects we create and/or use as extensions of those senses—objects that give way to processes (actions) through which we not only intervene in the world but also conceive of and think about it. In some respects, we engage with these objects-and-processes as if they were the external part of the mind.

Language and writing (a practice of manual and visual memory) are powerful manifestations of this. But the variety of artefacts that overlap and intertwine with each other at different levels of complexity and that have historically had a substantial impact with industrial production and are having a second significant impact with the emerging autonomy of cybernetics and so-called artificial intelligence are an even more dazzling case of the mind’s presence and ability to externalize. Here it is necessary to clearly understand in what sense human action forms a continuum with the production of artefacts and how objects and different types of artefacts are true nonhuman mediators and actors in the production and articulation of knowledge, taking us away from the old idea of a stark separation between the natural and the artificial, which is often described in opposition to what is characteristically human.

Many of the ancient paradigms (mainly animist but also totemist ones) understood the existence of different modes of a spiritual continuity (one of interiorities, as Philippe Descola notes) between human life and the world. The truth is that language can often not do without these associations, and until recently vitalist metaphors such as talking about the life and the death of languages were seen as acceptable and natural (and for many people still are). The paradigm of naturalist research in the West, objective science, has often been accompanied by the idea of an autonomous rational agent who is masculine, educated and self-satisfied about his findings, who sometimes regrets being alone in the universe and who has imposed an instrumental view of the world around him, beginning with the very objects and instruments that he denies a soul. This idea of a rational agent has also made it preferable and acceptable to speak of languages as instruments, in terms of uses, in a manner consistent with the development of the natural and social sciences. By contrast, the paradigm of distributed agency does not suppose that the objects of the world have an essential soul, and nor does it dehumanize them by characterizing them as simple instruments. Instead, it understands them as mediation, as part of nonhuman action, which is part of us and at the same time redounds to us, very much in the line with Edgar Morin’s analysis of complexity. This is how languages do things for us, and how we participate in these objects-and-processes, halfway between indiscriminate animation and selfish instrumentalization.

One might consider, for example, commercial objects, which are often classified under brands or logos that do not determine exactly what they are but provide information about their origin and their social value (A. Semprini). There are many interlinked institutions and people located in strategic places that decide the processes of industrial objects; there is production, but there is also design; production hides the distribution of components, which in turn are of different qualities and origins; design hides the collaboration of different actors with different specialized competencies; there is the logo that identifies the product or the by-product, which in turn demands a specific design; and there is the attribution of value to it, a social action that has consequences, which could be defined as “an idea put into practice.” Research on traceability show this activity of the product in different contexts, its capacity for non-human action. We know that brands, somewhat like ancient heroes, live among us insofar as they actively participate in the social network, hence their hybrid nature.

The field of health and illness is another old acquaintance of research on distributed agency, from Sebeok’s old descriptions of immune systems as genuine actors with a capacity for interpretation to the agentive vocabulary of the fight for health, not to mention works on biological risk (F. Tirado) or technology and assistance (M. Domènech). Then there is the common idea that we are ourselves and our pieces, our prostheses, that ordinary pills are like external particles of our organism, that we treat with sympathy or indifference, reducing them to colours or attributing them with often uncertain interventional capabilities. We have not been modern at all: we have not at all abandoned the idea of an animated world, in which we work and which also works on us.

The musician, the performer, forms a unit with his instrument. The worker is identified with work, as our basic vocabulary denotes. Robert Frost, in a beautiful poem, imagines the mower’s scythe whispering to the field. When we misplace an object, what opens up before us is a breach of possibilities that we struggle to grasp, somewhat as if the object’s autonomy led it to travel in some unforeseen direction or to reside in some unsuitable place. Correlatively, we deploy a battery of emotions for certain objects often without an “objective value”: that pen or that cup or that cushion that perhaps we do not use so much but that we end up possessing as a part of our life, an untransferable segment that recalls this person or that situation, and a part of our experience is apparently linked to these things’ vicissitudes and activity. Let alone clothes, pictures or photographs, or the music that accompanied us and is also an untransferable part, one that is now neither visual nor tactile, of the soul’s journey. These networks of meanings, in terms of objects and processes, make up what we call distributed agency, in which we participate and of which we are a resultant part at the same time.

The structure of language unfolds over this material eloquence. It is a question not just of the contemporary passion for cognitive metaphors (“John is a snake”; the symbolic dimension or figuration) but also of the agency and projections of action in ordinary language use, of ways of stating things that we do not know how to render differently. Tools become old, dresses impress, signs advise, cars seduce. We say, “The fire didn’t manage to get into town,” without thinking about a final action, but conferring shared agency on the action of fire. We become angry when a zip in our clothes becomes stuck, as if this state of affairs overstepped the autonomy of action that we attribute to the artefact. In many languages, classifiers carry out this important distributive function of allotting agency, material dimension and shared action in many possible ways.

Curiously enough, the advent of cybernetics has contributed to a more extensive perception of animation, if that is possible, affectionately designating new information processors: Mare Nostrum is a supercomputer, Java is a programming language; Irgo wasn’t once the odd acronym for a local server but the name of a village in the Pyrenees; and Kubrick’s perverse HAL parodied the three letters of IBM, scrupulously reworking the Latin alphabet. In many people’s informal register, the algorithm that retrieves information from the Net has become Uncle Google or Saint Google, in a new, unexpected twist of animated thinking.

If material eloquence has scope over language, then discourse itself can be presented in terms of objects-and-processes, as shared action. This is how we unfurl the activity of the word, a topic that François Cooren has laid on the table again in ethnographic and sociolinguistic terms. We say not only that “a poem is sweet” or that a speech “is aggressive,” but also that “words betray him” or that “the document is spreading across the Net.” As classical rhetoric supposed, and as we have reelaborated now, to the extent that texts are part of action, they do things for us. We delegate to discourse so that it represents us. Res ipsa loquitur, the thing speaks for itself, as the old legal expression for defending an argument states.

Chesterton explained how language was invented by hunters, killers and other such artists, long before science was even dreamed of. Now that we have science and we have cybernetics, we have once again made the old idea of animation visible, as far as possible, through the analysis of distributed agency. The vocabulary of critical mediation highlights the communion or participation: halfway between fact and fetish, as Latour liked to say. The thing is that referring to “magical thinking” solely when talking about logical errors creates problems. Distributed agency helps us to formulate and understand the reenchantment of the world, escaping from former disjunctives (between the natural and artificial, between the rational and irrational) that have proved so pernicious.


Ihde, D. (2010). Embodied Technics. Minneapolis: University of Minnesota Press.
Izutsu, T. (2012). Language and magic. Kuala Lumpur: The Other Press.
Latour, B. (2005). Reassembling the Social. Oxford: Oxford University Press.
Lem, S. (2017). Summa Technologiae. Minneapolis: University of Minnesota Press.