The Networked Image in Post-Digital Culture

Andrew Dewdney, Katrina Sluis, Andrew Dewdney, Katrina Sluis

• Verlag: Routledge
• Jahr: 2022
• ISBN:
  9781000603941
• Auflage: 1
• Thema: Social Sciences
• Thema: Media Studies

This collection examines how the networked image establishes new social practices for the user and presents new challenges for cultural practitioners engaged in making, curating, teaching, exhibiting, archiving and preserving born-digital objects.

The mode of vision and imaging, established through photography over the previous two centuries, has and continues to be radically reconfigured by a hybrid of algorithms, computing, programmed capture and display devices, and an array of online platforms. The image under these new conditions is filtered, fluid, fleeting, permeable, mobile and distributed and is changing our ways of seeing. The chapters in this volume are the outcome of research conducted at the Centre for the Study of the Networked Image (CSNI) and its collaboration with The Photographers' Gallery over the last ten years. The book's contributors investigate radical changes in the meanings and values of hybridised media in socio-technical networks and speak to the creeping automation of culture through applications of AI, social media platforms and the financialisation of data.

This interdisciplinary collection draws upon media and cultural studies, art history, art practice, photographic theory, user design, animation, museology and computer science as a way of making sense of the specific cultural consequences of the rapid succession of changes in image technologies and to bring the story up to date. It will be of particular interest to scholars and students of visual culture, media studies and photography.

Part IThe Condition of the Networked Image

1The Politics of the Networked ImageRepresentation and Reproduction

Andrew Dewdney

DOI: 10.4324/9781003095019-3

More than Visual

This is the time of the networked image, which is transforming settled ways of seeing, cemented over the course of the 20th century by the analogue means of visual reproduction in photography, film and television. The socio-historic code of the analogue technical mode of reproduction remains that of representation, a system of symbolic equivalences, based upon inscription, resemblance and recognition, a mode which ensures a naturalised, if not ideological reality, epitomised visually by the photographic image. With computing, the surface of the screen simulates the image of representation, as a necessary interface to the operations of data and signalling. In the first period of digitisation, across the late 1980s and 1990s, the digital image was initially understood in the analogue mode, as a static electronic version of the photographic image. But over the last two decades, through increased capacity of computing and scale of data, the new default of the image is its position in the processual relays of the network of networked computers. This change from digital to networked image has occurred because, while the front end of computing, what Lev Manovich (2001, p. 45) termed the legible cultural layer is still realised as representation, the backend of computing, the illegible computer layer, has developed an infrastructure on an industrial scale, serving an information economy. The near-limitless production of data in networks drives towards ever greater automation in which computers communicate without the intervention of humans. From the perspective offered here, the operations of computational networks, the black boxing of neural networks, the configuration of algorithms in software applications, proprietary platforms and mobile devices, call for a radically new understanding of what images are doing to us and what we are doing to and with images.

For the academic disciplines in which the visual image is studied and practiced, the computational networked image presents both new problems and new opportunities in revising theories and practices of visuality. The widespread use of computational ‘tools’ demands a re-evaluation of established institutional, historical and ontological categories of the visual in fine art, media, architecture, design, graphics, illustration, cartography, film, photography and animation. The perspective offered here suggests that visuality has entered a decisive era of the more than visual and nonrepresentational in which an ocular-centric worldview, which previously devolved upon the mechanical eye, has been overturned by the operations of data and signalling. The ways of seeing, established by the European Enlightenment and its humanist tradition, have been rewired by computation and now demand new ways of thinking about visuality, temporality and the human sensorium. This is a situation in which Hito Steyerl (2013) can describe images as crossing the screen and acting in reality. It is a situation of simulation argued by Jean Baudrillard (2012, p. 20) in which images no longer operate as mirrors to reality, or as reflections of inner mental states, but are projected directly, without mediation into reality. It is a situation in which the semiotics of the sign, the signifier and signified no longer easily apply. Or, as Deleuze and Guattari (2004, p. 163) think about the same problem, a situation in which language no longer signifies something which must be believed. The assumption of a transparent representational relationship between the image and reality is assailed on all sides, by the technical apparatus of nonrepresentational mathematical code, by intellectual doubt about the status of universal scientific truths, and by right-wing populism and fake news (Latour 2018). As a consequence of such perspectives, the argument pursued here is that it is no longer politically progressive, nor useful, to study media images, in whatever form, as separated from the world historical events, materiality, epistemologies, politics and experiences they participate in.

As this chapter recognises, developing a politically motivated analysis of the new default image of computation and its relation to the world necessarily involves more than one strategy, but where a synthesis is not easily achieved. What follows frames the networked image in three related ways, in the hope that they are consistent with the current problems of rethinking visuality, as well as contributing to other attempts to map out a thorough-going overview of what is happening to ways of seeing. Firstly, it considers the image as a relational assemblage, combining human and machine behaviours, in which the image is thought of as a transaction in a data network. Secondly, building upon the idea of the network as a financialised data system, it considers the image as a form of the social relations of the reproduction of labour. Finally, it considers the image in relationship to heritage, temporality and how the programmable image and the globalised network lead into the state of trans-hypermedia.

There has been a noticeable quickening of academic knowledge concerning the status of the image over the last decade, such that understanding is finally keeping pace with the magnitude of image capture, its aggregation in datasets, cloud storage, platform interfaces, circulation and deployment in machine vision. This chapter sets out, albeit briefly, to sketch what it considers to be some of the most cogently argued understandings of the operations of the network image. But its deeper purpose is to consider the political implications of this knowledge in everyday life, including the conditions of its own production, as well as its opening out on to a revised version of the material state of the world. If understandings of media are changing dramatically, so too are understandings of the world. It is increasingly recognised, certainly to readers of this volume, that individually and collectively, we, the global human population, are living through a profound disturbance of the planetary ecosystem. This situation confronts everyone in the attempt to understand the human impact on the planet and, closer to the interests explored here, confronts media and cultural scholars with how the humanly constructed world is sensorially mediated. We might also add that all living species are in one way or another experiencing precarity, and for some, the more immediate threat of extinction.

With the uncertain state of planetary futures, established knowledge paradigms, models of thinking, the system of representation and habitual action are now more uncertain and provisional, insofar as they were fashioned by and addressed to an older world order. Much of life is still lived out, paradoxically, in relationship to local and regional customs, patterns of work, family ties, traditional social bonds and a solid, if misplaced, sense of the permanence and abundance of nature. But life on earth has changed dramatically, both intimately and externally, by relations and functions of global capitalist economics and sophisticated networked technologies. We are living through nothing short of a cultural revolution, a term offered by Fredric Jameson in the 1980s to account for unevenness in historical changes in the mode of production and its determinations upon social life (2002, p. 81). Media theory can no longer treat the irrevocable harm capital accumulation is wreaking on the planet, nor the extreme reactionary violence, rape, murder and genocide that proxy wars necessary to capital extraction bring about, as simply media content. The global, capitalist mode of production has to be part of a theory of networked media technology. Any account of contemporary visual culture, media and communication which ignores these realities will fail to understand how and why the networked image operates as it does.

There is a very real danger, at what is a crucial moment of the reconsideration of the politics of the image, that critical understandings will quickly become contained by institutional academic orthodoxies, turning into yet another silo in the intensified churn of commodified knowledge. The commodification of knowledge works its way through research and scholarship, such that critique is neutralised and outstripped by instrumentalised research funding demanding market-ready knowledge. This can be seen in the ways in which even progressive knowledge, which seeks to understand the image under conditions of capitalism, is parsed through its publication, much as this book and this chapter will be. Alternatively, it could also be the case, as explored in what follows, that new knowledge of image technologies might contribute to the goal of a new literacy and become more widely disseminated and incorporated into democratic and public life.

The Networked Image

The term networked image is primarily an abstract concept, of a provisional nature, to denote a specific computational organisation of knowledge, but it is also a pointer to and descriptor of a concrete and complex set of material operations. The networked image has an infrastructure that requires labour and capital to produce a constant energy source, the mining of raw materials, the manufacture of electronic devices, the launching of space rockets, the construction of server farms, the laying of cables, and the deployment of transmitters and receivers. On current predictions, the Internet will use 20% of the total world consumption of electricity by 2025. The global network constitutes an entire mode of production, what the neoliberal World Economic Forum has defined as the fourth industrial revolution¹ and what Jonathan Beller has termed computational capitalism (2018, p. 12). The network image requires the unpaid labour of billions of people who spend increasing amounts of their daily lives online and in doing so provide data for corporate platforms, what Nick Srnicek (2017) and others have termed platform capitalism and the attention economy.

Critically, Benjamin Bratton has termed planetary scale computing as ‘The Stack’, an accidental megastructure, which he argues is ‘changing not only how governments govern, but what governance even is in the first place’. Bratton conceptualises The Stack as a complex platform, which can be modelled vertically as six layers: earth, cloud, city, address, interface and user (2015, p. 9). In the architecture of The Stack, the image can be understood to operate specifically between or as the interface and user layers. As Ingrid Hoelzl and Rémie Marie take up, a new definition of the image can be found in computational networks. In Softimage: Towards a New Theory of the Digital Image (2015), they argue that the video signal, which constantly refreshes the screen, removes the difference between the still and moving image and that this new temporality challenges the photographic paradigm. Importantly they recognise a central paradox in that while the algorithmic ‘digital image’ would appear to erode the photographic geometry of perspective and its projection, as well as the philosophy of truth that went along with it, at the level of visual perception, the photographic image continues to occupy the entire field of representation.

How to make sense of the apparent paradox of the algorithmic simulation of the photographic image was also the subject of Daniel Rubinstein and Katrina Sluis’s important essay, ‘The Digital Image in Photographic Culture; Algorithmic Photography and the Crisis of Representation’ (2013). They describe the collision of software and image, not only as a different process of image production but also, in line with the analysis of Hoelzl and Marie, as a cultural paradigm shift, with implications for ‘the ontological link between representation, memory, time and identity’ (p. 25). Rubinstein and Sluis liken the paradoxical nature of the digital photographic image to the two-faced Roman god, Janus, with one side facing appearances and the other facing towards the repetition and serial reproduction of the image. They go on to say that the paradox of the networked image lies in two incompatible logics, a representational and rational system of thought, in which the image refers to the existence of something real, and a recursive and viral logic, in which the image refers only to itself. Hoelzl and Marie reinforce this view of the double face of the image, although reach a somewhat different conclusion, in amplifying the key characteristics of the algorithmic image as its programmability and operability, which they describe as the continuous actualisation of networked data. The networked image no longer has a fixed representational form but a ‘signaletic temporality’ capable of transfer across digital networks in real time.

Between them, Rubinstein and Sluis and Hoelzl and Marie, lay out some of the defining characteristics and behaviours of the algorithmic, networked image, as well as identifying new problems and issues of how meaning is constituted. Rubinstein and Sluis propose that undecidability is the fundamental property of the networked image, as opposed to the immaterial or indeterminate. For them the meaning of the image is no longer fixed, but rather meaning develops through the progressive accumulation of what they term a ‘data shadow’, which determines its visibility and currency. Meaning is ongoing in tagging, commenting and syndication. One direction for future critical analysis contained in the notion of undecidability is to pursue meaning and hence power within the algorithm, to reverse engineer software in order to reveal the agency of computation itself. This project has been taken up by Nicolas Malevé who examines the social ontology of machine vision, and is discussed in this volume (Chapter 4). Another direction for critical studies, hinted at by Rubinstein and Sluis, is one that reconsiders the paradoxical, representational face of the image, and follows the work of Giorgio Agamben, in suggesting that the image might be rethought in terms of the ways in which technologies of the network inform identity and subjectivity. Collectively they emphasise the dialogic nature of interaction in which attention should be paid, not to the appearance of the image on screen, ‘but to the rhythms of repetition and recurrence, to the time signatures, to ...

Inhaltsverzeichnis

• Cover
• Half-Title Page
• Title Page
• Copyright Page
• Contents
• List of Illustrations
• Acknowledgements
• List of Contributors
• Introduction
• Part I The Condition of the Networked Image
• Part II Computation, Software, Learning
• Part III Curating the Networked Image
• Part IV Digitisation and the Reconfiguration of the Archive
• Index