In her recent book, Softimage: Towards a new Theory of the Digital Image, co-authored with Remi Marie, Ingrid Hoelzl discusses our new understanding of the image.
Nina Strand: Throughout your book you show how the image isn’t a fixed representational form, but is active and multi-platform, the ‘softimage’ as you call it – the image as programmable software. It no longer functions as a political and iconic representation, but plays a vital role in synchronic data-to-data relationships. Can you tell Objektiv readers more about this?
Ingrid Hoelzl: Through a progressive series of case studies, Softimage traces the dissolution of the image from hard to soft: from print to screen, from still to moving, from geometric to algorithmic, from output to moment of network access. With animation, postproduction, compression, navigation, and wireless access, the held-for-granted concepts like indexicality, reference and frame that characterised the photographic paradigm and built on the assumed solidity of the photographic image as a stable representation of a stable world (hardimage), make way for what we call the softimage. It’s kind of ironic that the publication of this book coincides with the final release of the famous 3D animation software, Softimage® (with product support ending this year). So while Softimage will be dead by the end of this year, the softimage lives on – not only in the multitude of new image software (programmable images), but in the sense of the image having become a programme in itself, and of the concept of image having become soft, of eluding a clear-cut definition.
The cover of the book is meant to highlight this ‛soft’ state of the image. Using one of Clement Valla’s Postcards from Google Earth, a series of Google Earth screen shots begun in 2010, shown as postcards on stacks and on a website, we stenciled out the word softimage from a solid black cover laid on top of one of Valla’s soft highways – a metaphor for the softimage, a bit like like Dalí’s melted pocket watch as a metaphor for memory (or for a camembert melting in the sun as he famously claimed). The original image is part of the Google Earth image-database, which visualizes 3D vistas of the planet in real time. Google Earth images look photographic but are in fact a combination of different data: photo-, carto-, computer graphic. Valla insists on the same thing in his description of the series: his soft highways are not glitches, but mere anomalies of a system that processes/renders heterogeneous data in a photorealistic way.
NS: In the chapter The Operative Image (Google Street View: The World as Database), you take as an example from what the artist Harun Farocki has called ‘the operative image’: that is, ‘images that do not represent an object, but rather are part of an operation’. Can you elaborate on this?
IH: Farocki coined this term in the context of the automated warheads that were used for the first time in the First Gulf War. But the term is employed on a much wider scale in his Eye/Machine trilogy (2001–2003), where he brings together changes in industrial production and warfare: both are operated through the automation of vision. In Chapter 5, we apply the term to Google Street View images, linking the debate on online surveillance with Farocki's (and Virilio's) concerns. Just as ‘seeing bombs’ (as Klaus Theweleit has called them) are aimed at us, Google Street View images are not simply 'interactive images' reacting to our whims (move down a street, zoom in, swirl around, switch back to map mode etc). Rather, they’re part of a navigational and informational ensemble: images, arrows and pancakes, laser point-cloud, GPS data, pop-up location tags within the image and ‘nearby locations’ on the panel next to it, locations that are mostly commercial. Thus, Street View is actually taking us on a shopping tour, while tracking our moves and preferences. These are continually fed back into the database in a cybernetic feedback loop, resulting in what we’ve called a 'reverse operativity': when we operate Google Street View images, Google Street View images are operating us (on the behalf of Google). This commercial exploitation of user data (which is probably the least worrying of Google’s purposes) isn’t specific to the image, it happens in all online services, but it is new to the image.
NS: Another chapter in your book, The Photographic Now, From Sign to Signal, argues that with digital projection and screening there’s no difference between still and moving images. As you explain, the still image is only a loop in the digital video signal; as with a freeze frame in film, the signal doesn’t change with each reactualisation, but instead repeats itself. The result of this is a radical change in the temporal orientation of the photograph; it no longer reactualises the past, but instead actualises a possible present.
IH: Yes, in The Photographic Now, we stress that photography no longer refers to the past (of its production) but to the present (of its display), the present of its own reactualisation on screen: as a looped video signal (Chapter 2), as a signal (Chapter 3). With Thomas Ruff’s JPGs, which we analyse in detail in Chapter 4, it’s the generic temporality that interested us, how indexical images (a nuclear test, an iceberg etc) become generic, floating, anonymous through the process of hypercompression. This generic temporality is also at play in Google Street View: its temporality is both sequential (the result of sequential recording/navigation) and generic (a seamless photographic panorama tied to Google Maps whose temporality is hybrid or composite). In Chapter 6, we leave these questions of display, compression and navigation behind and approach the image’s temporality in terms of its integration with digital networks. We posit the image-screen as the place and time of network access and data transfer, where the image isn’t the result of these processes, but occurs during or at the moment of data exchange.
NS: You and co-author Rémi Marie are currently writing for Fotomuseum Winterthur's blog Still Searching. Here, until the end of April, you’ll be reflecting on the status of the image in digital culture and the the age of autonomous machines, discussing the image as software – the postimage. Fotomuseum Winterthur has also launched an exhibition format called SITUATIONS, a multi-part programme that will respond quickly to ongoing developments in the photographic culture. What are your thoughts on running a museum for photography today?
IH: Winterthur, as a Museum for Photography, follows the migration of photographic images from print to screen and from offline to online, and I think many other museums will follow. It’s a difficult situation, of course, for a hard museum, because a museum contradicts the idea of conversation, but it’s a necessary move to account for the new online image practices and to pursue what the Germans call Bildkritik through the very same practices.
SITUATIONS recently published a call to send in video-game screenshots. This is a great initiative I think, because photography, especially art photography, has always flirted with popular photographic genres and styles (just think of Ruff, or Tillmanns, or the current hype around social media selfies), but has kept a safe distance from video-game imagery, which is judged to be aesthetically poor, either grossly misrepresenting reality or exaggerating its imitation.
NS: Part two of Objectiv’s The Flexible Image, out this October, will look at the image as text: the image as readable sign. Here we’ll hopefully interview W.J.T. Mitchell, who in his book What Do Pictures Want? The Lives and Loves of Images, underlines the fact that pictures are assemblages of images, objects and media. Pictures, as he notes, are ‘understood as complex assemblages of virtual, material, and symbolic elements’ – pictures have ‘lives and loves’. Thus we need to move beyond the idea of images as world mirrors. Do you agree?
IH: Well, Mitchell’s question ‛what do pictures want’ is a rhetorical question, since the basic assumption of his book (and study of visual culture) is that images actually don’t have power but are attributed power by us. He describes a kind of pre-modern relationship to images that grants them a gaze, agency, even life. But this analysis, rooted in Marxist, psychoanalytic and postcolonial theory and a speech-act notion of images (what we do with images), can’t account for a world of computer-computer interaction where images are involved in a multitude of processes that are hidden behind their appearance on screen and their so-called ‛interactivity’. Images embedded in online services have real agency, they’re part of algorithmic processes (of surveillance and control), which is what we stress with our concept of the operative image and of reverse operativity.
Do we need to move from the idea of the image as world mirrors? In Chapter 5 of our book we quoted Google Earth Outreach geo-strategist Karin Tuxen-Bettman, who declared aboard the GSV boat as it made its maiden run along the Rio Negro, ‘We want to create a digital mirror of the world’ (Zeenews Bureau Report 2011). She referred to this desire of Google – not only of GSV – to reflect the entire world. We explain in the book that, ‘in reality, the reverse is the case: it is the world that is a mirror of the GSV database. Put differently, the people of the banks of the Rio Negro claim their existence via their being part of the GSV database. They exist because the database exists. This recalls Baudrillard’s argument of the map preceding the territory, and Latour’s argument of the “real image” of the map creating the “virtual image” of the territory. But in fact, both map and territory, world and database, engender each other in what we could call, drawing on Jacques Lacan, a speculative feedback loop. For Lacan, the infant overcomes, in what he calls the “mirror stage” (1968) occurring between six and eighteen months, its bodily fragmentation only via identifying itself with its mirror image perceived as a whole. If we extend the scope of Lacan’s investigation to the relation between the world and the image of the world, what we have is that the world overcomes its own fragmentation via its identification with its specular image or imaginary counterpart, the database. And the database in turn exists as a whole only via its identification with its specular image or imaginary counterpart, the world.’
Softimage: Towards a new Theory of the Digital Image was published in September 2015 by Intellect/Chicago University Press.