Telling Non-Human Stories
Writing and reading narratives show us what other possible ways of encountering the world are possible, thereby potentially transfiguring our own ways of engaging with the world around us. Implicated in narrative, therefore, we find a transformative potential.
Few problems faced by humanity require such a drastic change in engaging with the world as climate change. As a true multitude of authors from all sorts of disciplines now assert, climate change is only resolvable when we venture out of the confines of the purely human, into the breadth of an ecology that both encompasses and transcends us (Chakrabarty 2018, 6; Ghosh 2016, 82; John 2018; Kohn 2013; Morton 2012; Trexler 2015, 13-14; Ustler 2009, 173). Perhaps we can capitalise on the transformative potential of narrative to change the way we deal with climate change as well? Can we use narrative to escape our all-too human way of interacting with the world? To answer this question, we first need to have a clear conception of what narrative can do and how it works as a transformative instrument.
In his Time and Narrative, Paul Ricoeur (1913-2005) offers precisely such an analysis. To understand the transformative aspect of narrative, he introduces the term ‘mimetic circle’. This philosophical tool describes the way in which the reading of a novel – for simplicity’s sake I use the example of written narrative in this text, but Ricoeur’s theory is applicable to various other ways of experiencing narratives – presents a narrative world to the reader, which then potentially influences the reader’s own world, thereby transforming the way the reader encounters the world around them.
So, does that mean we can simply take Ricoeur’s mimetic circle and start creating narratives that effectuate more climate friendly behaviour? Sadly, no. For all its philosophical depth, the mimetic circle always centres on human actions and narratives – it is quite literally anthropo-(human)-centric. This is a problem, because these narratives were supposed to help us overcome the anthropocentrism that led to climate change in the first place. We therefore need narratives that capture not only human, but also non-human, material forms of action, communication, and temporality. Only then can we expect of narratives the transformative potential to help us change our behaviour as drastically as is needed now.
Hence, if we want to use Ricoeur’s mimetic circle to help combat climate change, his philosophy needs to be ‘stretched out’ to also encompass the non-human. In this paper, I attempt such a broadening by consulting various concepts proposed in Levi Bryant’s Onto-Cartography (2014). More precisely, the primary question this paper poses is whether Bryant’s concept of non-human action can be used to widen the scope of Ricoeur’s narrative theory. I tackle this question in the following way: first, I examine Ricoeur’s theory of narrative by discussing his mimetic circle. Second, I introduce Bryant’s alien phenomenology as a conceptual framework for conceiving non-human action. Third, I incorporate Bryant’s non-human action into Ricoeur’s mimetic circle, thereby establishing a framework for the transformative potential of a non-anthropocentric narrative.
Narrative Time and Anthropocentrism
Ricoeur’s theory of narrative and the way it influences human consciousness is, for a large part, carried by his idea of mimesis. To understand what form this influence takes, and how it displays Ricoeur’s anthropocentrism, the mimetic circle itself needs to be elucidated. Afterwards, its anthropocentric elements can be distinguished.
Mimetic Circling
The mimetic circle is constituted by a “single continuous process” that is the rotation of mimesis. It is made up of three parts: mimesis1, mimesis2, and mimesis3. A rotation through these three sections describes three respective situations: a person before reading a book, the process of the person actually reading the book, and then the person changed by this reading. To understand the expanse but also the limitations of Ricoeur’s argument, the three steps are discussed separately.
Mimesis1 refers to the fact that we are always already capable of understanding actions. We have a so-called ‘preunderstanding’ of action. This preunderstanding becomes more well-defined in its three different elements: structure, symbol, and temporality. Structurally, mimesis1 means that we can differentiate between action and mere physical movement (Ricoeur 1984, 55). We do this using a certain “conceptual network of action” that is given to us through a combination of factors hailing from both the nature and the nurture register (56). Symbolically, mimesis1 refers to the fact that actions have an inherent readability, that is given through action always being “symbolically mediated” (57). This means that action is never done outside a sphere of “signs, rules, and norms” (57) – an action always already means something. Temporally, mimesis1 is largely defined by the Heideggerian idea of within-time-ness (62). Within-time-ness is a kind of preoccupied experience of time; time as the ‘thing’ in which our goals and plans are formulated (63). Mimesis1 is, then, “first to preunderstand what human acting is, in its semantics, its symbolic system, its temporality” (64).
Upon this preunderstanding of action, Ricoeur says, a narrative is constructed by means of emplotment. The process of emplotment – to which mimesis2 refers – is the organisation of the as of yet unorganised events distinguished in mimesis1 (64). Emplotment mediates between the preunderstanding of action of mimesis1 and the ‘postunderstanding’ of mimesis3 (65). This mediation occurs in three ways: first, as already discussed, it mediates between the individual “incidents and a story taken as a whole” (65). Second, it mediates between the discordance of a heterogeneous set of elements of “agents, goals, means, interactions, circumstances, [and] unexpected results” (65), and the concordance of the plot (66). This mediation operates, third, on a temporal plane: a temporally ordered configuration is extracted from a temporally meaningless succession (66).
This threefold mediation of mimesis2 allows for the creation of a narrative from the preunderstanding of action registered in mimesis1. This creation of plot, however, does not come forth solely from the individual human being’s preunderstanding. It also bases itself on a certain tradition, on the one hand characterised by novelistic paradigms that have sedimented into the tradition (68), on the other hand made up of the innovative potential of each and every unique narrative work (69). Mimesis2, therefore, is the fact of the production of a story by the mediation between incident and story, discordance and concordance, and succession and temporal order, using the input and rules inherited by a tradition.
The narrative created in mimesis2 does not exist in a vacuum, of course; it is read. The encounter between reader and the work Ricoeur calls mimesis3 (Dowling 2011, 14). This confrontation is a conflict of the actual, practical world of the reader, and the formal, fictive world of the text. The interaction between these worlds is constituted by the fact that both worlds are, by genesis, built on the same preunderstanding of action Ricoeur argues is mimesis1 (15). A preunderstanding of action is needed to construct our own world of experiences but also to construct any narrative in general.
With mimesis3 the world of the reader collides with that of the narrative, the latter exhibiting its different possible ways of experiencing the world to the former. What then occurs in the reader – if she is influenced by the text, of course – is a change of consciousness (16). This process thus refigures the prefigured consciousness of the reader through the mediation of a configured narrative work.
Non-Human Action
Now that Ricoeur’s mimetic circle has been expanded upon, we are ready to understand in what ways the non-human does not fit into the constraints set on narratives by Ricoeur. The problem of anthropocentrism discussed in the introduction resurfaces here: it is hard to conceive of a climate-friendly refiguration – an eco-refiguration, if you will – while only considering the human perspective (Ghosh 2016, 82; Trexler 2015, 13-14; Ustler 2009, 173).
The threat of anthropocentrism is carried most urgently by the first stage of mimesis. This is because mimesis1 dictates which events in general are (pre)understood as an action. When the non-human is allowed into this category, it can be taken up in the process of emplotment that is mimesis2 and subsequently influence the reader’s world by letting it collide with the world of the narrative.
The categorisation of non-human ‘happenings’ or “mere physical movement” (Ricoeur 1984, 55), as action is no self-evident manoeuvre for Ricoeur. His way of conceptualising action is implicitly and explicitly anthropocentric: structurally, he distinguishes between human action and its “goals … motives … [and] agents” and ‘mere’ physical movement (55); symbolically, actions are always mediated by an implicitly human plane of expression that includes “signs, rules and norms” (57); and finally, mimesis1 literally refers to a preunderstanding of “what human acting is” (64, my italics). Ricoeur clearly is not interested in non-human action.
At the same time, I share Utsler’s (2009) optimism for a Ricoeurian eco-philosophy – provided it succeeds in shedding its anthropocentric limitations. To make eco-refiguration possible, the reader needs to be confronted with a configured narrative world that is composed of non-human agents and the accompanying non-human action. To incorporate these elements in mimesis1, I now turn to Bryant’s Onto-Cartography.
Machines All the Way Down
In Onto-Cartography, Bryant puts forward a world view in which all entities – whether they be bureaucracies, volcanoes, human beings, symphonies, or forests – are seen as machines. To differentiate between all these machines, Bryant introduces two theoretical tools: the dual face of machines, and alien phenomenology. In the following, I discuss both tools and explain their role in imagining non-human agency and action.
Dimensions of the Machine
In contrast to traditional philosophy, machinic philosophy is hardly interested in the qualities an entity has. It is mostly focused on what the machine does. Fundamentally, “[a] machine is a system of operations that perform transformations on inputs thereby producing outputs” (Bryant 2014, 38). Every entity that exists, therefore, has a certain set of transformations it operates on (flows of input), and a set of outputs constituted by these operations.
This structure sets up a dichotomy between a machine’s operations and its outputs (40). The capacity a machine has for its specific array of operations is called a ‘power’ (a leaf’s power to photosynthesise, for instance). The outputs created by the machine’s actual operations are called ‘manifestations’ (42). These manifestations do all sorts of things for the machine: they can be the machine’s qualitative appearance (the leaf’s greenness, for example); they can transform the behaviour of the machine itself (its shrinking when there’s little water); and they can be a material output flow (the oxygen produced by photosynthesis) (44-45). The powers a machine has dictate which operations it undertakes, which subsequently generates the entity’s observable facts.
Machines are not open to all inputs, neither do they possess an endless array of powers. To illustrate this, Bryant talks of how machines are respectively ‘structurally open,’ and ‘operationally closed’ (55). The leaf acts as an example again: in terms of the inputs it is receptive to, it is structurally open to the specific part of the electromagnetic spectrum leaves use for photosynthesis (Heliospectra 2012). A machine’s operational closure refers to the finite selection of powers a machine has (Bryant 2014, 56). It also comprises the fact that an input flow is never encountered by a machine ‘as it is’, it is always encountered “in terms of how its operations transform it” (57). According to Bryant, the input flow ‘as it is’, that enters the leaf is a photon with a certain vibration length. By contrast, the leaf encounters it as fuel for internal energy production.
By combining the abovementioned facts – machines have a ‘power-face’ and a ‘manifestation-face,’ and are open to a selection of flows and work on these flows in specific ways – Bryant argues it is possible to infer how a machine encounters and experiences its world (62). His method for this procedure is discussed below.
Alien Phenomenology and Action
The inference of such an encountered world first entails questioning to which flows a machine is structurally open and how the machine structures these flows; second, how the machine operates on these flows and what functions the transformed flow has in the machine; and finally, how the machine encounters the world around it (62).
Bryant’s aspirations must remind readers familiar with phenomenology or philosophy of mind of Thomas Nagel’s seminal essay ‘What is it like to be a Bat?’ (1974). In this paper, Nagel argues that we cannot reduce the “subjective character of experience” to simple physical terms – something Bryant definitely seems to be doing when he wants to describe the phenomenology of all machines (Nagel 1974, 436-37). To speak with Nagel: ‘we cannot possible know what it is like to be a leaf!’ Bryant readily accepts this critique; indeed, his kind of “second-order phenomenology” does not enjoy the same level of epistemic reliability that traditional ‘first-order’ phenomenology does – the fact that we cannot ‘move’ into the point of view of other machines makes all statements about the content of these other points of view merely inferential. But, Bryant argues, “[o]ur alien phenomenologies will always be imperfect, but as we will see, as imperfect as they are, they are nonetheless preferable to the epistemic closure of humanism that approaches all of being in terms of what it is for us” (64). Put more simply: it is better to have an imperfect understanding of the world of plants, couches, and bureaucracies, than to have none at all.
The Nagelian critique can perhaps be further suspended with a short example of the alien phenomenology of one unassuming protagonist: a patch of grass. When a gardener mows down half of this patch, the mown down grass produces a plant stress hormone called GLV (Green Leaf Volatiles) (Dombrowksi et al. 2019). These GLV-molecules enter the rest of the grass patch machine. The GLV-flow is canalised into the receptor kinase of the grass leaf cells (Dombrowski et al. 2020). There, the transformed GLV-flow primes certain “wound stress pathways” in the cell – putting the plant in a sort of ‘defence mode’ (Dombrowski et al. 2020, 780). Put in Bryantian terms: the grass patch is selectively open to GLV-flows, through the receptors in the grass’ cell membranes the flow is structured into the right cellular mechanisms. The grass cells have a multitude of powers for operating on this flow – together constituting the operational closure – that are far too complex to be discussed here, the eventual result of which is that the grass ‘readies itself’ for being wounded – the GLV-flow has become a warning signal to the grass. Finally, we may now carefully conclude that the patch of grass encounters a world populated by all sorts of plant hormones, one of which is GLV, to which it can react defensively.
How does this conclusion imply non-human, plant-action? The answer is quite simple with Bryant: the grass patch machine, like every other machine, operates in accordance with a combination of environmental input flows it takes in, and the specific powers it has – thereby producing a multitude of manifestations. The production of a manifestation by the exercise of powers on flows of input – that is action in both a human and non-human sense.
Non-Human Mimesis
The problem this text has come to is the following: is the description Ricoeur gives of action in mimesis1 compatible with – stretchable to – Bryant’s machine-action? If it is, non-human actions can be emplotted into narratives, these narratives eventually eco-refiguring the minds of their readers. This final section discusses this problem.
As became clear in the section on Ricoeur, mimesis1 grasps those events as actions that have structural “goals … motives … [and] agents” (Ricoeur, 1984, 55) and that are symbolically mediated by a plane of meaningful communication of “signs, rules and norms” (57). The problematic bifurcates, consequently, into a problem of non-human ‘goals’ and ‘motives,’ and a separate problem of ‘meaningful communication.’[1]
To start off with the second problem, the idea that only humans are capable of meaningful communication is wholly alien to Bryant. He claims that “it is likely that planes of expression [fields of meaningful communication] are to be found throughout the animal kingdom” (133). Bryant can say this because a machine never encounters an input flow ‘as it is,’ but always “in terms of how its operations transform it” (57). Especially in highly complex machines like organisms and groups of organisms, input flows can – by repeated reinterpretation by the machine – become meaningful signs. A signal that is utterly meaningless to other machines can thereby become full of meaning for other machines configured to operate on it – for grass plants, the GLVs are a sign of danger, for humans, they just smell good (Ford 2022). Examples of non-human, meaningful communication are virtually endless, moreover: “Acacias alert other acacias to prowling giraffes. Willows, poplars, alders: all are caught warning each other of insect invasion across the open air” (Powers 2018, 176-77).
Bryant’ plant-action thus fulfils the symbolic exigency of action in mimesis1. This redirects us to the first problem: the structural requirement for goals and motives behind an action. Alien phenomenology shows that any given machine can be seen as a whole that encounters the world in a certain way. This tremendous broadening of ‘subjecthood’ is accompanied by Bryant’s insistence that organic machines do have goals: ranging from the goal to “preserve their organization” to the complexity of being “intentionally goal-directed” (Bryant 2014, 30).
These new, strange subjects have their own goals and motives: according to Bryant we may intelligibly speak of the goal of the patch of grass to mitigate the damage done by the grass mower; we may intelligibly think that the plants in our home have the goal to capture as much light as is internally useful to the plant. Hence, it is very possible to imagine complex motives behind non-human action. The only problem that remains, and seems insurmountable, is the fact that we may never know about these motives with the same certainty as we know about our own and other humans’ motives – that is the upshot of Nagel’s barrier. Whether we are convinced of Bryant’s insistence that accepting this limitation and prolonging our efforts to imagine these non-human worlds is better than closing ourselves off to them, in the end, seems a question of intuition, more than one of philosophical argumentation.
Conclusion
Lastly, to return to the problem that spawned this text, is it possible to perform a non-anthropocentric Ricoeurian mimetic circle – to create non-anthropocentric narrative worlds that influence people into changing their stance towards climate change? I started this investigation by discussing Ricoeur’s mimetic circle. The ontological understory of mimesis consisted of first localising the anthropocentrism in the prefigurative part of mimesis. This mimesis1 excluded non-human action on both structural and symbolic grounds. Bryant’s concept of non-human action was a structurally open and operationally closed production of manifestations by the exercise of powers on flows of input. With this, the symbolic grounds for dismissing non-human action could be rejected. The structural grounds proved harder to counter for Bryant – plant-machines can indeed be thought to have motives and goals, but due to the epistemic uncertainty of alien phenomenology we may never know these for sure.
Overcoming this problem can perhaps only be done by making it practical – something we as philosophers have a tendency to refrain from. What are the practical implications of refusing to believe that non-human machines may have goals and motivations – or put more technically: to go along with Bryant’s epistemic uncertainty regarding non-human phenomenology? Most importantly for this paper, they are that non-human ‘actions’ cannot be categorised as real actions because of Ricoeur’s structural requirements. This would, furthermore, mean non-humans cannot be taken up in the mimetic circle, meaning non-anthropocentric narratives that influence human ways of engaging with their world cannot be conceptualised in any serious way. Finally, our own anthropocentrism is therefore made inescapable, we cannot flee from it through telling stories and creating narratives – climate change becomes insurmountable.
So perhaps, indeed, we must remain epistemologically pure and reject non-human actions; with such dire practical consequences, however, this epistemological purity comes at a very serious ethical cost: the evaporation of our ability to create real change in real people. It may come as no surprise that I myself look forward to new, inventive non-human narratives.
Bibliography
Bryant, Levi R. 2014. Onto-Cartography: An Ontology of Machines and Media. Edinburgh: Edinburgh University Press Ltd.
Chakrabarty, Dipesh. 2018. “Anthropocene Time.” History and Theory 57, 1: 5-32. https://doi.org/10.1111/hith.12044.
Dauenhauer, Bernard P. 2013. “Ricoeur and agent causation.” Philosophy and Social Criticism 39, 6: 523-537. https://doi.org/10.1177/0191453713485720.
Dombrowski, J.E., Kronmiller, B.A., Hollenbeck, V.G. et al. 2019. “Transcriptome analysis of the model grass Lolium temulentum exposed to green leaf volatiles.” BMC Plant Biology 19, 222. https://doi.org/10.1186/s12870-019-1799-6.
Dombrowski, James E., Brent A. Kronmiller, Vicky Hollenbeck, and Ruth C. Martin. 2020. “Transcriptome Analysis of Wounding in the Model Grass Lolium temulentum” Plants 9, 6: 780. https://doi.org/10.3390/plants9060780.
Dowling, William C. 2011. Ricoeur on time and narrative: an introduction to Temps et récit. Notre Dame: University of Notre Dame Press.
Ford, Jez. 2022. “The Smell of Fresh-Cut Grass is an Attack Warning.” Science Illustrated. Retrieved on the 10th of June 2023. https://scienceillustrated.com.au/blog/ask-us/the-smell-of-fresh-cut-grass-is-an-attack-warning/.
Ghosh, Amitav. 2016 The Great Derangement: Climate Change and the Unthinkable. Chicago: The University of Chicago Press.
Heliospectra. 2012. Which regions of the electromagnetic spectrum do plants use to drive photosynthesis? Retrieved from heliospectra.com on 5th of June 2023. http://www.esalq.usp.br/lepse/imgs/conteudo_thumb/Which-regions-of-the-electromagnetic-spectrum-do-plants-use-to-drive-photosynthesis.pdf.
Kohn, Eduardo. 2013. How Forests Think: Toward an Anthropology beyond the Human. California: University of California Press.
Morton, Timothy. 2012. The Ecological Thought. Harvard: Harvard University Press.
Nagel, Thomas. 1974. “What is it like to be a Bat?” The Philosophical Review 83, 4: 435-50. https://doi.org/10.2307/2183914.
Potter, Henry D. & Kevin J. Mitchell. 2022. “Naturalising Agent Causation.” Entropy 24, 472: 1-18. https://doi.org/10.3390/e24040472.
Powers, Richard. The Overstory. Dublin: Vintage Classics.
Ricoeur, Paul. 1984. Time and Narrative: Volume 1. Translated by Kathleen MclLaughlin & David Pellauer. Chicago: University of Chicago Press.
Trexler, Adam. 2015. Anthropocene Fictions: The Novel in a Time of Climate Change. Charlottesville: University of Virginia Press.
Utsler, David. 2009. “Paul Ricoeur’s Hermeneutics as a Model for Environmental Philosophy.” Philosophy Today 53, no. 2: 173-78.
[1] The question of agency is a separate and incredibly complex problem that can and should be discussed in a more spacious text than this one. The problem is, however, not at all insurmountable: as we can read in Dauenhauer (2013, 523), Ricoeur in principle uses ‘agent causation’ instead of the ‘distributive agency’ of Bryant (2014, 180). Although these two positions, when taken to their extreme forms, do most definitely contradict one another – extreme agent causation places the source of action fully inside the entity, while extreme distributive agency places next to no source of action inside the entity – I believe they can be synthesized quite elegantly. This should be done in two steps: first Bryant’s insistence on the unity of a machine (while at the same time being a part of a world of other machines, and being made up of endless machines itself) and his subsequent alien phenomenology proofs the distribution of agency implied by his theory is not boundless – the source of a machine’s actions is always also internal. Second, Ricoeur’s agent causation is not boundless either: the source of a person’s actions is never fully internal (as is clear from the effect sedimented tradition has on the types of novels people can and do write). Moreover, as is clear from Potter & Mitchell (2022, 3) for agent causation to be naturalised, it needs, among other things, to take account of both endogenous, and exogenous activity of the organism. A synthesis, hence, is possible, but we naturally need more space to develop this point in its entirety. For now, a discussion of non-human action will have to suffice.