In this article, Richard Carter outlines an ongoing critical and creative engagement in electronic literature, digital sensing, and ecological concerns. Like many who are now publishing critical and creative works together (particularly in The Digital Review), Carter situates his practices in a set of entangled disciplines, and then discusses his developing project Landform.
The global scale of digital infrastructure is now crucial to the charting and modelling of a rapidly deteriorating planetary ecology—as achieved by satellites, drones, remote buoys, weather stations, datalinks, and supercomputers. The electronic signals generated by these systems must then be parsed into formats that can be usefully perceived and processed by human observers. Parikka (2015a: 12), in his examination of the relationship between electronic media and geophysical processes, summarises that ‘it is through and in media that we grasp earth as an object for cognitive, practical, and affective relations’, with acts of visualisation, sonification, or simulation creating different planetary representations for specific contexts. Nevertheless, Parikka goes on to observe that the gathering and reformatting of signals into data is never conducted in the abstract, despite its supposedly intangible qualities, but necessitates its own costly entanglements with the planet it purportedly registers from afar: ‘Data demand their ecology, one that is not merely a metaphorical technoecology but demonstrates dependence on the climate, the ground, and the energies circulating in the environment. Data feeds [off] the environment both through geology and the energy-demand’ (2015a: 24) . In short, the systems through which the world is measured, mapped, and depicted for human knowledge-making will always generate their own extended environmental impacts, which are registered subsequently in the data yielded. Such impacts manifest most immediately in the form of the energy consumed as part of the daily operations of digital sensors and infrastructures, but, in the longer term, we need to include the material demands, wastage, and pollution generated by their manufacture, maintenance, and eventual disposal.
Beyond the material transformations necessary for the collection and assembly of environmental data, the systems involved, and the representations generated, can also facilitate paradigms that enable Earthly harm, such as by feeding into varied military, geopolitical, and technoscientific imperatives. To illustrate, Cosgrove (2001) has characterised the technologies of terrestrial imaging from an airborne and orbital vantage as enabling visions of a unified, and controllable, global space. In Cosgrove’s conception, the view from above has been encoded with a sense of perceptual omniscience and disconnected objectivity from the vagaries of the surface far below, which has been coupled historically to ‘a universalist, progressive, and mobile discourse in which the image of the globe signifies the potential [equality] of all locations networked across frictionless space … an implicitly imperial spatiality, connecting the ends of the earth to privileged hubs and centers of control’ (Cosgrove 263).
Litfin (1997: 38) has similarly observed these dynamics playing out in the domain of Earth Observation satellites specifically, with their overhead gaze becoming both the privileged depiction and technical enabler of an abstract ‘global view’ of Earthly affairs, which are treated as data points to be simulated and managed using large-scale, technocentric initiatives. Litfin contends that ‘the view from space renders human beings invisible, both as agents and as victims of environmental destruction. It also erases difference, lending itself to a totalizing vision. The “global view” cannot adequately depict environmental problems because the impacts of these problems vary with class, gender, age, and race’ (38).
Gabrys (2016a) traces these technical and imperial imaginaries into the domain of ubiquitous environmental sensing and computing more broadly. Although these systems operate at a multitude of differing scales, crystallising localised environments of monitoring and control for specific ends, they contribute nonetheless to a broader discourse that codifies the planet as a computable entity that can be held within a state of complete sensory capture and managerial enclosure. Gabrys (2016a, 2016b) subsequently connects these ‘planetary’ outlooks to interlinked histories of Western militarism, colonialism, and industrial globalisation, contending that sensory technologies are aligned often with the pursuit of their objectives. It is this combination of technology and discourse that perpetuates an extractivist approach towards the Earthly environment, which is parsed only as a resource to be mined and managed in the service of human needs alone, with its attendant sense of competition against any human or nonhuman actors that resist this.
Parikka, Cosgrove, Litfin, and Gabrys all characterise systems of digital sensing and sense-making as both a product, and a facilitator, of political and economic processes that are driving the signals of calamitous change they render. In considering the critical implications of this, I am concerned especially with how it impacts the role of creative digital practices, such as electronic literature, in aiding our critical sensibilities towards ecological transformation.
In assessing the complexities faced by digital artists and artworks in this regard, an analogy can be drawn with the challenges faced by their earliest antecedents in the 1960s, which attempted to express different possibilities for computing beyond militarised technoscience. McGregor and Parsons (2016) note how the pioneering, computer-generated artworks of Manfred Mohr were the subject of critical disdain and even direct vandalism when they were first exhibited in this period, owing to then-close associations between computing systems and Cold War activities. Contemporary digital art manifests in a very different context, but, as pointed out by Naji (2021: 56-8), digital systems more broadly are often culturally characterised in terms of the disembodying, technocentric impulses of commercial investment and marketing rhetoric. Naji (58) observes subsequently how this technocracy presents a key problematic when it comes to how digital art and literature might inspire substantial ecological critique and reflection, for it necessitates various modes of resistance against the very digital medium that enables them in the first instance. The recent eruption of interest in distributing and monetising digital art as unique, possessable, Non-Fungible Tokens (NFTs), via blockchain registers, has certainly exacerbated these debates. Pipkin (2021) provides a typical instance of the critical reactions generated, focusing on the extractive ideologies behind the creation of artificial digital scarcity, and highlighting the extensive energy-demands of blockchain infrastructure. These very latest developments will not be discussed further here, except to acknowledge that they have brought into sharp relief questions regarding the ecological impacts of generating, distributing, and displaying digital art.
In considering this difficult relationship between the ostensive richness of the digital medium and its profound entanglement with environmentally damaging paradigms, instances can be found of electronic literature that are able to negotiate between these aspects. J.R. Carpenter’s This is A Picture of the Wind (2018) is one such, and I have argued that its expressive value lies in its deploying of data-driven language to resist any notion of digital sensors as being able to wholly characterise a changing world (Carter 2020). To summarise briefly, This is A Picture of the Wind draws live weather data from the UK South West region to remix a personal weather diary kept in the aftermath of a series of exceptionally large storms in 2014. In this piece, two very different modes of registering and recording environmental phenomena are brought into dialogue, exploring the different potentials through which digital data can engage verbal modes of human sensing and knowing. The result is a work that is affectively far richer than the standardised formats of charts, tables, and visualisations that typify a purely datified sensory outlook—using language to express the myriad sensations and phenomena that escape the registers of quantification and abstraction, but which characterise fundamentally the experience of living and coping within a profoundly damaged world (Carter 2020).
The intersection of electronic literature and poetry with environmental concerns is not a new development. Raley (2011) observed a distinct ‘ecological turn’ in then-contemporary digital poetic practices: moving away from self-reflexive considerations of technical possibility to ‘an embedding of humans and computational media within a larger assemblage comprised of human and nonhuman actors, and lively, vibrant, animate matter’ (889). More recently, Pereira (2020) has examined various electronic literary projects that engage themes of the provisional, the imperfect, and the uncertain for capturing both the realities of environmental collapse and imagining what configurations of future being, in such a collapsing world, might resemble. Naji (2021: 59) has subsequently developed an extended critical account of how digital ecopoetry seeks to tell different stories that are wholly marginalised by celebratory, technocentric discourses, such as the many forms of underprivileged labour that uphold digital environments and activities in the first instance.
Following my interest in works such as Carpenter’s This is a Picture of the Wind, my own creative and critical practices have charted the capacities of digital writing to realise different speculative forms and futures of environmental sensing and sense-making. I have been pursuing this especially by devising unconventional assemblages of digital systems and other materials—entangling cameras, satellites, drones, web graphics, esoteric code, academic writing, and the printed codex, exploring what their contingent exchanges can reveal about the structures, dynamics, and possibilities of sensing across the contemporary environment. The hybrid art-texts generated by these activities are thus better understood in light of their complex origins, deriving their creative and critical force as much by encouraging reflection on these varied aspects and processes, as the actual markings left behind.
An artistic gesture that I am presently exploring is the use of image generating technologies for producing creative textual outcomes. Specifically, I am employing imaging sensors as a means of gathering and analysing different kinds of visual data about ecologically provoking scenes. This data is then used to actualise textual data-structures into various ‘human-readable’ configurations, which, nevertheless, rarely adhere to the conventions of established literary forms, but resonate instead with various modes of experimental writing and poetry.
The chief project I have been working on in this regard is entitled Landform. This project represents a developing artistic exploration of the speculative entanglements between electronic literature, digital sensing, and ecological thought. In this project, satellite and drone imagery of terrestrial scenes have their brightness signatures plotted into diagrammatic visual algorithms, which are then executed to compose tabulated list poems that are derived from works of environmental scholarship. This elaborate process, founded on a series of unconventional dialogues between disparate elements, is intended to enact a speculative media assemblage: a ‘platform’ constituted from a diverse array of technologies, practices, vocabularies, and environments, whose varied juxtapositions afford insights into their functioning and significance.
At the time of writing, Landform is still being subject to further development, and so what is outlined below reflects its configuration as of late 2021. The functional basis of Landform involves gathering overhead landscape imagery from the vantage point of either an orbiting satellite or a low-flying camera drone, before sequencing these through a function that parses their constituent brightness values into diagrammatic visual structures. This is achieved by reading the source image as a 2D array of brightness samples, and then registering the value shifts along an extending vector from a starting coordinate at the centre of this array, adjusting the trajectory of this vector depending on the magnitude of the shifts encountered (see Fig. 1). Sequences of similar brightness values further extend the length of this vector, while significant differentials result in it being segmented and then adjusted to proceed along a new cardinal direction. An analogy here might be found in the functioning of ‘turtle graphics’ programmes, which use similar principles.
The system tracks the image areas encountered at every step, preventing the growing vector sequence from overlapping itself, and halting any further extensions when it (invariably) enters a cul-de-sac of its own making. This constitutes the mechanism by which its length, and that of the instructions it encodes, are kept within reasonable dimensions, and does not grow indefinitely across the source image. The configuration of this system was inspired by that of the ‘Piet’ programming language developed by Morgan-Mar (2018), which abandons the usual linear chains of symbols and commands in favour of arranging blocks of colour on a grid, which are then sequenced along varying distances, in varying directions, depending on the colour shifts encoded.
The circuitous trajectory that is plotted across the source image is then read by an interpreter function to compose the final textual output. At first this process involves a separate routine that compiles dictionary lists of the unique constituent verbs and nouns within a source corpus. These dictionaries are then mapped onto a multi-dimensional array structure, one that matches the dimensions of the source image array. The vector coordinates plotted across the source image then act as lookup indices for retrieving specific sequences of words from this dictionary array. More precisely, different cardinal directions correspond to specific dictionary lists, and the segment lengths determine the number of words retrieved from a given set of coordinates within these lists.
In the latest configuration of Landform, the generated JSON object is printed as the program output alongside a copy of the source image, which has the plotted visual algorithm overlaid. The resulting diptych presents this plot as the executable code from which the final text, the tabular poem, is compiled and outputted (see Fig. 2).
Concerning the source imagery that drives this entire process, it can be noted that Landform was prototyped initially using photography from a series of drone overflights of North Cornwall, in the United Kingdom—encompassing a selection of moorland regions and coastal areas. What is captured by this footage are sites inscribed by the marks of human activity in various guises, from quarries, to world-war runways, to the ghosts of iron age communities. Despite its official status as an area of ‘Outstanding Natural Beauty’, with its idyllic implications, there is no area of North Cornwall that does not bear the signs of human settlement, cultivation, and extraction. Consequently, it provides an apposite opening setting for depicting the ecological entanglements driving Landform. As the project evolves, it is planned to visit other such areas across the United Kingdom, and so examine their complex status as sites celebrating the ‘natural’ world while being firmly marked and enclosed by human activity.
Other visual sources being considered for Landform include satellite captures of areas designated on a ‘Red List’ of imperilled ecosystems by the International Union for the Conservation of Nature (IUCN). These domains do not necessarily exist as discrete regions, as implied by an individual satellite scene, and many of the intricate processes they encompass, across a multitude of scales, are not necessarily registered from an orbital vantage. By contrast, however, the forces imperilling these ecologies are much more readily apparent, in the form of deforestation, pollution, and human settlement. As noted above, satellite imaging concurrently identifies but also facilitates such threatening activities, as part of the planetary-scale acts of measurement and mapping through which the Earth managed as a resource for human needs. Such points of tension make them intriguing sources for considering how speculative sensory practices might negotiate and express these very aspects.
The source texts behind the computed outputs of Landform are predominantly a mixture of academic works on topics critically relevant to the project’s key themes, encompassing ecological writing (e.g. Engelmann 2021), practices of scientific measurement (e.g. Barad 2003), and media materialities (e.g. Parikka 2015b). To illustrate these texts briefly, Engelmann (2021) provides a critical outline of the ‘geopoetics’ of writer Eric Magrane, focusing on how his writing conveys the more-than-human energies and materials that characterise the Earth’s elemental processes. Barad (2003) provides an influential account of the relations between matter and meaning, as carried through her work on the contingencies of scientific measurement, and discusses how acts of mark-making, whether scientific or artistic, shape the very world they depict. Parikka (2015b) discusses the embedding of all media substrates in the deep time of geological activity, before highlighting the role of speculative media archaeologies in exploring how media technologies evolve out of these conditions.
At a practical level, it should be noted that none of these sources can be readily discerned or reconstructed by examining the final JSON object in which they are arrayed—relying instead on citations in their accompanying artist’s statements. However, the core gesture here is to explore through practice some of the key themes at work in these intricate, provocative texts, while also rearticulating their critical vocabularies for creative effect. The aim is not to naively ‘illustrate’ these ideas in-action, but to actualise their expressive potentials along a more speculative vector.
This approach has been inspired by McGann and Samuels’ (1999) use of digital tools to speculatively ‘deform’ literary texts for analytical ends, such as by adjusting their word sequencing or even their appearance on the page, and so actualising otherwise imperceptible semantic relations and points of emphasis. Premised on a deconstructive rejection of text as a singular, immutable artefact, with a set array of meanings, the experiments of McGann and Samuels treat it instead as a catalyst for interpretative possibility, which can be legitimately realised as much through material intervention as critical dialogue. In the case of Landform, the remapping of academic works into tabular poetry produces both alternative readings of these texts (as computed lists of their constituent keywords) as well as a generative structure for assembling different meanings concerning them. This is achieved by using the JSON format to create clusters of words that do not conform with their sequential arrangements in the source. The tabular structure affords multiple routes of associative nesting, whereby each keyword imparts its surrounding neighbours with different possible inflections—but with no indication of these being definitive. The result is a contingent lexicon of ecological sense-making that is apposite to the contingent sensory apparatus through which it is composed, with the reader invited to trace their own interpretative pathways and associations. Sensing and sense-making are parsed as emergent, generative practices, rather than an assembly of transparencies through which to observe a pregiven domain.
Appropriating the graphical vocabularies of technoscientific endeavour—diagrammatic imagery and tabular formats—is not without its points of tension. Drucker (2014: 92) traces the history of these devices in a convergence of the empirical sciences and managerial attitudes in the nineteenth century, which were then both in the ascendency. Diagrams, tables, and charts, with their tendency to eliminate outliers and emphasise the overall patterning of data, became the chief means by which information, particularly statistical information, could be ‘abstracted from circumstance’, and thus enable the nuances and complexities of specific contexts, of human lives and sentiments, to be effaced from the task of managing the world (Drucker 2014: 94). Landform cannot undo or overcome these histories, but in rendering generative text through formats associated historically with the stratification of worldly phenomena, the effect sought is to unsettle the expectations of factual fixity and quantitative precision that is typically associated with digital data more broadly, before then crystallising its expressive potentials beyond these notions. That is, by outlining a contingent critical lexicon, the resulting interpretative excess and pluralities of meaning are not depicted as intrusions to be eliminated, but a crucial indicator of a world that exceeds any one act of sensing and knowing, and whose present challenges are in-part founded on the fallacy that such is possible.
Image to Text
The source descriptions above provide some measure of insight into the creative aspirations of Landform, but the chief question remaining concerns the act of parsing images into code and then into text, and what this gesture represents from an ecological standpoint. A key driver in this regard was Farocki’s (2004) conception of the ‘operative image’, capturing how visual data is invariably transformed and manipulated across an array of machinic contexts for the fulfilment of specified operational goals—in which its rendering for human perception may never feature. The martial undertones of ‘operative’ acknowledges the kinds of tracking and targeting conducted by military vision machines especially, such as radars or electro-optical guidance systems, which apprehend the world in ways that are very different to that of human eye.
Farocki himself detailed the concept of the operative image across a number of documentary films. In his Eye/Machine series (2000-3), and War at a Distance (2003), Farocki examines the technical and cultural histories behind the automated visual routines found in missiles, drones, and targeting optics, detailing their subsequent role in shaping modern warfare and, eventually, the wider civilian world. Artist Trevor Paglen, drawing on Farocki’s work, has curated exhibitions and projects that variously expose the contexts, technologies, and imperatives structuring the deployment of visual systems for the martial enclosure and regimentation of contemporary life. Like Farocki, Paglen (2014) maintains an interest in how spectral signatures are framed and bounded into discrete, identifiable entities by algorithmic processes—as achieved by structures and techniques such as tracking gates, optical flow vectors, and edge detection plots. These formations may be rendered subsequently for the benefit of human operators, to observe their evolving logics, but they are most often left unseen—a set of automated processes for the cueing of others, ranging from monitoring and surveillance to the guiding of lethal force. Individual systems scale subsequently into the visual regimes of territorial policing and geopolitical competition that arrest the Earthly environment into a calculated space of conflict and control—with living ecologies having no intrinsic weighting within this calculus, beyond their contribution towards, or hindrance of, such activities.
Landform does not expressly invoke military technologies or imagery, but it does enact processes that resonate with them. The vectors plotted across the source imagery are arranged according to a logic that is inherently very different to that of human vision. Instead of scanning for perceptually intuitive formations, the system reads a sequence of value differentials across a mathematical matrix, with the patterning of this sequence varying solely on whether each step crosses a certain numerical threshold. Accordingly, this process does not generate a mimetic ‘capture’ of the depicted terrain, but encodes an imperative of regimenting the gathered data into an executable structure, ready for further processing. The plotted topography of abstract vectors is therefore concurrently entangled with, but also heavily abstracted from the terrain which constituted the source imagery in the first instance.
Landform makes no attempt to occlude the tensions, problematics, and troubling character of such operations—in line with the usual optimism of much digital technology rhetoric—but rather depicts head-on the structural impositions of digital sensory regimes upon the environments variously captured and striated by them. For some onlookers, this will be sufficient to nullify any benefit of the speculative intent of the gesture, and those it supports thereafter, but my own investigation is premised on acknowledging such complexities and contradictions, shaping the interpretative matrix in which Landform emerges and is received. The driving hypothesis is not that automated visual analysis can be fully decoupled from its troubling deployments, but that the act of algorithmic discretion and enclosure can be opened out into more contingent perspectives. Rather than composing yet more technical visualisations, or precise tables of values, the system drives instead the creative rearticulation of varied critical works. This yields a provisional lexicon that is both ecocritically charged and technically reflexive, whose complex origins, and dataified formatting, expresses the now profoundly technogenic character of the contemporary environment. It draws attention both to the discourse in which digital sensory regimes function, and are critiqued, and, from this, the contingent and affective phenomena their operations fail to register. Landform begins its investigation from within this domain, rather than without, and so engages the provocative task of adapting aspects of a thoroughly entrenched, and often harmful digital world to speculate concretely on what other avenues might be pursued—to contribute to ongoing processes of refiguring human sensibilities towards more-than-human harms.
Barad, K. (2003) ‘Posthumanist Performativity: Toward an Understanding of How Matter Comes to Matter’, Signs: Journal of Women in Culture and Society, 28(3), pp. 801-831. doi: https://doi.org/10.1086/345321.
Carpenter, J.R. (2018) This is a Picture of Wind: A Weather Poem for Phones. Available at: http://luckysoap.com/apictureofwind/ (Accessed: 1 May 2021).
Carter, R. (2020) "Electronic Literature in the Anthropocene", Electronic Book Review. Available at: https://doi.org/10.7273/rt06-ts14.
Cosgrove, D. (2001) Apollo’s Eye. Baltimore, M.A.: The Johns Hopkins University Press.
Drucker, J. (2014) Graphesis - Visual Forms of Knowledge Production. Cambridge, MA.: Harvard University Press.
Engelmann, S. (2021) ‘Geopoetics: On Organising, Mourning, and the Incalculable’, Dialogues in Human Geography, 11(1). doi: https://doi.org/10.1177/2043820620986398.
Farocki, H. (2000) Eye Machine. Available at: https://www.harunfarocki.de/installations/2000s/2000/eye-machine.html (Accessed: 10 Dec. 2021).
Farocki, H. (2003) War at a Distance. Available at: https://www.harunfarocki.de/films/2000s/2003/war-at-a-distance.html (Accessed: 10 Dec. 2021).
Gabrys, J. (2016a) Program Earth: Environmental Sensing Technology and the Making of a Computational Planet. Minneapolis, MN.: University of Minnesota Press.
Gabrys, J. (2016b) Becoming Planetary. e-flux architecture. Available at: https://www.e-flux.com/architecture/accumulation/217051/becoming-planetary/. (Accessed 7 Jan. 2022).
Litfin, K. (1997) A Feminist Perspective on Earth Observation Satellites, Frontiers: A Journal of Women Studies, 18(2), pp. 26-47. Available at: https://www.jstor.org/stable/3346964 (Accessed: 15 Feb. 2022).
McGann, J. and Samuels, L. (1999) Deformance and Interpretation, New Literary History 30(1), pp. 25-56. Available at: https://www.jstor.org/stable/20057521 (Accessed: 15 Feb 2022).
Morgan-Mar, D. (2018) Piet. Available at: https://www.dangermouse.net/esoteric/piet.html (Accessed 10 Dec. 2021).
Naji, J. (2021) Digital Poetry. Palgrave Macmillan.
Paglen, T. (2014) Operational Images. E-Flux. Available at: https://www.e-flux.com/journal/59/61130/operational-images/ (Accessed: 10 Dec. 2021).
Parikka, J. (2015a) A Geology of Media. Minneapolis, MN.: University of Minnesota Press.
Parikka, J. (2015b) ‘Mutating Media Ecologies’, continent, 4(2), pp. 24-32. Available at: https://web.archive.org/web/20200712170215/http://continentcontinent.cc/index.php/continent/article/view/183 (Accessed: 1 May 2021).
Pipkin, E. (2021) Here is the Article you can Send to People when they Say “But the environmental issues with Cryptoart will be solved soon, right?” Medium. Available at: https://everestpipkin.medium.com/but-the-environmental-issues-with-cryptoart-1128ef72e6a3 (Accessed: 10 Dec. 2021).
Raley, R. (2011) “Living Letterforms”: The Ecological Turn in Contemporary Digital Poetics’, Contemporary Literature, 52(4), pp. 883-912. Available at: https://www.jstor.org/stable/41472497 (Accessed: 1 May 2021).