the limits to ‘invisuality’ and ‘postrepresentation’?

I read Andrew Dewdney’s chapter in the new book he’s edited with Katrina Sluis called The Networked Image in Post-Digital Culture last week, for a discussion with the fab TRAVIS research team. Dewdney’s chapter is a very useful read – he summarises several of the statements about digital imagery that I’ve returned to repeatedly in my own work, and takes them and their assumptions forward to a specific critical position. The chapter covers a lot of ground, and I’m just going to say something here about how it brought into focus some of the limits of those statements for me.

The key references Dewdney discusses in his chapter include the book Softimage: Towards a New Theory of the Digital Image by Ingrid Hoelzl and Remi Marie, the essay in Theory, Culture and Society by Adrian Mackenzie and Anna Munster on platform seeing, various pieces by Jonathan Beller, a chapter by Daniel Rubinstein and Katrina Sluis in the second edition of Martin Lister’s The Photographic Image in Digital Culture, and a book that I haven’t read but have now added to my list of things to read by John May called Signal Image Architecture.

These are all really important statements about contemporary visual culture. Hoelzl and Marie work through a very productive definition of digital images as images that are co-constituted with software; Rubinstein and Sluis explore the networking of photos; Mackenzie and Munster coin the useful term ‘invisuality’ to refer to the algorithmic analysis of the vast banks of images collated by social media platforms; and Beller points to the political economy driving those same platforms. Dewdney rightly suggests that this work has many significant things to say about “the magnitude of image capture, its aggregation in datasets, cloud storage, platform interfaces, circulation and deployment in machine vision.”

However, Dewdney takes as empirical reality the formations described by Hoelzl and Marie, Beller and Mackenzie and Munster and generalises from them about all digital visuals now, to claim 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.” Representation no longer matters for critique, apparently, because representational imagery (ie photography) has been superceded by algorithmic computation.

All the authors discussed by Dewdney are of course describing important empirical realities. But as various kinds of materialist scholars, or as media archeologists perhaps, they are also theorising from those realities, and Dewdney’s discussion of their analyses shows some of the implications of extending that theoretical position into a universal empirical description of “the politics of the image”.

One issue for me is the continued reference to ‘the image’ in the singular, accompanied by the implicit reference to photographs as a sort of paradigmatic image. I think this shapes the argument in particular ways. Aligned with a critical interest in the political economy of images, for example, it means that much of the argument actually refers to social media photos and platforms and therefore – possibly – only to social media photos.

Another is that Dewdney’s attention to the power of ‘the operations of data and signalling’ produces a number of binary splits in his argument. Splits include those between illegible computational processes and the legible image; or the invisual and the visual; or the representational and the not; or the network and the everyday. Various spatial metaphors underpin this split: a front end and a back, an A-side and a B-side. While in theory all these meet at the interface of a screen that displays a photographic image, the implication is that the image doesn’t matter and indeed is a distraction from the real site of power, which is the algorithm ‘behind’ the screen. This appears to remove users/humans/posthumans from the site of the algorithm – which is odd because the actual sites of algorithmic labour are full of humans (as well as computers).

Another associated binary that emerges is that visual culture is divided into two parts too, between the algorithm – the agential materiality of the image, in this account – and the user of the image. In fact I would say that this is not only a binary, it’s a dualism: the user ends up as simply that which algorithm is not, constituted as immaterial and as desiring and as having an unconscious, in Dewdney’s discussion. This means that what Foucault called the ‘situation’ of visibility is understood in quite limited terms. There’s no attention to the practices or discourses or imaginaries that render certain kinds of images in particular situations recognisable in particular ways, for example.

I am summarising crudely here, but I think not entirely inaccurately. It seems to me that softimages are hugely important now. But I also think that this account underestimates the variety of productive power dynamics in the many different situations of contemporary digital visual culture. Moreover, digital images are not all photographs, or part of profit-driven platforms, or even networked. Embodied users still matter. Not all seeing is platform seeing. And the situations in which images take place are hugely significant for their effects.

For example, at some interfaces with softimages, human agency is itself constituted as calculative – not algorithmic exactly but not just disembodied ‘desire’ either (I’m thinking here of all the apps used for efficiency and convenience in which desires are aligned with calculations about corporeal time, energy, expense). And there are situations in which embodied engagement with images remains absolutely pivotal (as our TRAVIS team discussion emphasised – Kat Tiidenberg is our project lead).

There are also many situations in which enormous care is taken to establish digital images as representational. I’ve been looking at the design of urban digital twins recently, for example. These might be positioned as the next stage of cyber/intelligent/smart cities, but rather than gathering data via city dashboards in smart control centres, they integrate real-time big data about a city into a three-dimensional digital model of that city. (Or that’s the theory anyway – as with cyber/intelligent/smart cities, there’s a lot of hype as well as some serious experimentation going on – experimentation which is not being carried out by social media platforms, nor always for profit…). In this situtation, the relation of data to the actual city matters hugely – the data need to represent the city in ways that enable action in relation to city infrastructure. This is done largely through a number of geometric operations, but photographs continue to be used in various ways in many digital twins as part of this effort to model the ‘real’ city. This is an important example of powerful forms of softimagery being used to manage urban life through representational techniques – and which doesn’t fit the situation Dewdney outlines for ‘the’ image.

Finally, one of the many provocations in Jacob Gaboury’s book Image Objects – which is a history of early computer graphics – is its complete (I think) lack of reference to photographs. The strong implication is that the computational processes that produce computer graphics (may) have nothing to do with the processes of invisuality described in Dewdney’s account, based as it is on social media photos. As Gaboury convincingly argues, computer graphics are mathematical and spatial objects first, that are then made visible; they are not inherently visual. I’m not sure yet how to reconcile these two accounts but given that, as Gaboury points out, computer generated-graphic images are in many situations now visually indistinguishable from photographs, it does seem to me that both arguments and image types need to be considered carefully, in all their diversity, in discussions of contemporary digital visual culture.

My simple point is that I think we need a much more diverse and precise vocabulary for the wide range of uses and effects and distributions of digital images now. We surely need to theorise digital visual culture in ways that could include the huge variety of different kinds of softimages and their diverse effects – like digital twins, for example, or movie VFX – rather continuing to generalise from the invisuality of social media platforms, important as that is.

seeing the city in digital times: a lecture

I gave a keynote lecture at the Neue Kulturgeographie XIV conference a couple of weeks ago, at the University of Bayreuth. My topic was ‘seeing the city in digital times’. I talked about the challenges of keeping cultural geography relevant as a critical project when so much visual culture is now digital, and I shared my recent work looking at how so-called ‘smart cities’are pictured on YouTube and Twitter.   You can hear my talk and see the presentation that accompanied it here.

Listen through til the end if you can (or indeed just skip to about an hour in) because I got some great questions afterwards.  It was a privilege to speak as such an energy-filled event – thankyou to my hosts Matt Hannah, Eberhard Rothfuss and Jan Hutta.

smart cities on Twitter: or, urban / cultural / visual / digital

I’ve been trying to work on a paper about how smart cities look on Twitter over the past few works.  One answer is this:

That’s a trial run I’ve done, working with the 900-odd images attached to a range of smart city-related hashtags, scraped over a week last month by my OU colleague Alistair Willis, and run through Lev Manovich’s ImagePlot software.  Saturation increases closer to the centre of the image, and hue is distributed much like a colour wheel.  Yes, smart cities are mostly either blue or orange!!

This is part of my effort to think about different visual methods that can respond – even if only partially – to the sheer scale of image circulation in digital visual culture now. It doesn’t touch on the dynamics of their circulation, but it does suggest, I think, a possible effect of the speed and numbers of images on social media platforms and the casual way in which they’re often seen: that we might see a certain sort of city as a colour field that enacts smart (for example) rather than a set of images that represent it. So the blue and orange mean almost nothing (though not entirely). What they might suggest, though, is something about the feel of the notion of the smart city, as it’s performed through Twitter.

What I’m now doing is digging a bit deeper into that ‘feeling’: what does a smart city hashtag on Twitter do with both smart cities and with the hashtag followers? What kinds of affect does it intensify?  I think I’m kind of getting towards an answer, but of course I need to do some more reading.  And here is my pile of books that I hope will help me think about what thousands of images of smart are and do, getting me away from smart and Twitter specifically and more towards thinking about the intersection of the urban, visual, cultural and digital. I’m also looking forward to reading a roundtable in the online journal Mediapolis, on the urban as an emergent key concept for media theory.

(I was going to make a snarky comment about it obviously being compulsory to use grey, black, red and white when designing the covers of these sorts of books – but then I realised that my workspace is pretty much the same colours….)

 

 

photographing a smart city

MK:Smart is a large ‘smart city’ project based in Milton Keynes in the UK.  It’s hosted by my home institution, The Open University.  Its core work package when it was set up was the development of a open data hub: an repository of all sorts of big data about Milton Keynes, accessible to anyone.  As the project has developed, though, its efforts to enable local people to engage with such an open data source have increased.  One of these efforts is the website OurMK, and the project team also does lots of outreach in local schools.  You can find about more about their work to facilitate ‘smart citizens’ here and some of their publications are listed here.

As part of this engagement work, the MK:Smart team launched a photography competition in December 2015, with prizes for the best photographs picturing Milton Keynes as a smart city.  You can see the finalists here.

I think these photographs raise some fascinating questions about how a smart city is visualised by its residents (or, more accurately, what the judges thought were the best ways some residents had pictured a smart city).  Nineteen photographs made it to the final stage – not many, so I should be careful about drawing any big interpretive conclusions from them.  On the other hand, as I’ve remarked before, one of the liberating things about writing a blog is that sometimes the robust methodological procedures of the social sciences can be laid to one side and a little more speculative thinking permitted…

Car Park Drama by Suzanna Raymond. Suzanna’s caption read: The way the car park is integrated into the shopping centre looks like a smart design to me, making it an integral part of the layout rather then just a space added on as an afterthought.

So one thing that struck me immediately about these nineteen was how so many of them focus on the landscape of Milton Keynes, and especially on its ‘natural’ landscape: trees, parks, canals, lakes, skies.  There are no pictures of servers or data hubs or smartphones (though there are two photos of electric car charging points, one of a bus charging wirelessly and one of solar panels).

This preference for picturing a smart city as a green city perhaps speaks to the distinctive history of Milton Keynes.  Milton Keynes was designed as a new city in the late 1960s and early 1970s with plenty of experiments in (and symbols of) more sustainable living: houses powered by solar energy, cycle ways paralleling the roads, a dial-a-bus service, a tree cathedral… the whole city is full of trees and parks and is oriented along a ley line! Milton Keynes has a strong sense of itself as green, then, and these photographs might be speaking to that sense of place. The photos perhaps also draw on a rather English preference for rural landscapes, gardens and parks.

The other thing that struck me about the photographs was the way they display the sort of visual aesthetic that seems increasingly common in many digital images, which is a kind of glow against darkness, whether that’s lights gleaming at dusk or (elsewhere) live data feeds pulsating across a black background.  No less than seven are taken around sunset, and one more makes a striking play between a sky darkened by clouds and a golden building.

So, possibly, what we have in this admittedly tiny sample of photographs is an interesting play between what someone like Lev Manovich might suggest is an increasingly widespread visual aesthetic, driven by the extensive use of digital image creation/editing software – even a global visual aesthetic – and something that’s may be much more local, attuned to the specific histories of this particular city and its sense of place.

Now, of course, as Doreen Massey would immediately have pointed out, there’s no clear distinction between the local and the global.  Many of the ideas behind Milton Keynes, and implicit in its first visualisations, were imported by its architects from the west coast of the USA, for example, and I’ve already suggested that a love of rural landscape may be as much English as anything to do with Milton Keynes.  But it’s precisely this play between the new and the old, between existing ways of seeing and of making images with new ways of seeing and making, that I find so fascinating in this small collection of photographs.

 

 

being provoked by an essay on ‘what is 21st century photography’

I commented in passing in my previous post on the freedom that a blogpost offers: to write more loosely and widely than you can in an academic paper.  And along comes an outstanding blog post – well, a blog essay really – that demonstrates that in spades.  The post is Daniel Rubenstein‘s ‘What is 21st Century Photography‘ and it’s on The Photographers’ Gallery website.  It’s racy, provocative, covers several centuries, is stuffed full of quotable aphorisms, and has a clear argument to make.

I think that argument is very interesting.  Its key claim is that:

in a post-Fordist society the locus of political agency and of cultural relevance has shifted from the object – as visually arresting as it might be – to the processes that (re)produce and distribute the object. Processes, however, by their own nature, are less visible and less representational than objects. For that reason, it seems to me that if photography mainly concerns itself with representations of objects in space, it is losing its relevance in a world in which speed, acceleration, distribution and self-replication acquire a significance that overshadows the visual appearance of spaces.

Which leads to probably my favourite line in the whole essay.  If a photograph is now something that is just occasionally assembled from a wave of data that continually shapes all kinds of visual forms – then, says Daniel: “it has little in common with prints in black frames – these coffins of photography”. ‘Coffins’.  Fab.

The post has lots of insights and pleasures then, though I wasn’t sure about the ‘invisible puppet masters’ who are our ‘real rulers’, and I did feel that its recuperation of photography in its very final line – “photography is not a stale sight for sore eyes, but the inquiry into what makes something an image. As such, photography is the most essential task of art in the current time” – was bit of a failure of nerve.  Maybe photography is just past it?

All this interesting provocation is in sharp distinction to what I was going to discuss on this post, which is Nicholas Mirzoeff’s new book on visual culture, modestly called How To See The World.  I was sent this by the publisher so I perhaps shouldn’t look a freebie too directly in the mouth… but.  The book is bit of a mess, I think.  It attempts the sweeping overview but there’s no clear analytical framework, let alone theory, to guide it, and there are also some quite irritating – well, to be frank, just plain wrong – generalisations.  One of which is that images now are all about time. (We know this because an artist made an artwork with lots of clocks in it, apparently.)  As for example Hito Steyerl, and many others, myself included, argue, it is absolutely necessary now to have a sense of the spatiality of (what is better understood as) visual data, as well as its temporality: its form as a swarm, population or wave; its immersivity; its materialisations; and its geometry as a network.  I think one of the difficulties in Mirzoeff’s book is actually that he remains fixed on images as the problematic rather than these sorts of spatialities that articulate their production and circulation and use, so that he flits from place to place, example to example, without thinking about they might (or might not) join up in some way.

But that’s my query to Daniel too: what are the geographies to his account?  Where does the data move, pause and decay?  How is it circulating, and with what effects in different places?  And how are places themselves being reconfigured in this process?

interesting project website on selfies

Thanks to my colleague Rose Capdevila, I’ve just spent a while browsing a site called making selfies/making self, which is both a research project about selfies and about a research project on selfies.  If you see what I mean.

Created by Katie Warfield, it’s an interesting mix. There’s some information about the research project, an online quiz enabling participation in the research, a gallery of images that might be related to selfies (according to the internet, as Katie describes it), links to talks Katie’s done, and links to other networks, including the Selfie Research Network and a Zotero library on selfies.  Katie also curates a related Tumblr feed.  Seems to me to be a really neat example of how to do a research project online, which is both open – sharing resources, inviting comments and contributions – but also offering a scholar’s particular take on an issue.

international urban photography summer school 2015

Just got told about the annual international urban photography summer school, happening this August in London, tutored as usual by the fab visual sociologists at Goldsmiths.  More information and registration is here.

 

the production, composition, audiencing – and circulation – of images

I’ve been spending the first few weeks of this year thinking about how to revise my book on Visual Methodologies for its fourth edition.  Among other things, I’ve been thinking about what difference digital technologies – as both topic and tool – are making to its arguments.  And I’ve decided on at least one significant change for the fourth edition: the three sites through which the book organises its discussion of visual methods are now four.  The site of ‘circulation’ has been added to the sites of the production of an image, the image itself, the site(s) of its audiencing.  ‘Circulation’ is intended to emphasise that all images, to some extent or another, travel.  Images are mobile, and how they travel matters to what effects they have.

This isn’t an insight created by the development in the past few years of massive, extended social networking sites that now carry vast numbers of images between all sorts of different screens.  In fact, the keyword that I’ve attached to it in the book is the idea of a “visual economy”, which comes from anthropologist Deborah Poole‘s book on the way images travelled between the Andes and Europe between the seventeenth and the early twentieth century.

However, I do think that there are some methodological issues involved in looking at – or even thinking about – those huge hoards of online images that require an emphasis on their circulation.  I think it’s important that we pay attention to the work that goes into enabling that circulation, for example, in the workplaces where the labour is done to make those platforms feel so easy to use: the coders and the checkers, the servers and the cables.  Also, images on Instagram or Twitter or Facebook aren’t stored in some ginormous virtual contact sheet, and not every image has an equal chance of appearing on a Google Images search result screen.  Instead, how those images get seen is shaped by algorithmic patterns.  Search results are shaped by your previous searches, by your location, by what other people are also searching; and what you see on Instagram, Facebook and Snapchat also depends on your social network.  I think we need to figure out methods that can show us, somehow, the patterns and processes through which those image collections are structured, just as twenty years ago Foucauldian historians like John Tagg and Alan Sekula showed us how filing systems and labels organised earlier forms of photography archives.

This is a problem with Lev Manovich‘s cultural analytics, I think: it engages with the huge numbers but does so by adding them all up, and creating collages of the total.  This shows us some interesting things – what Tokyo looks like in 50,000 Instagram images is provocative in terms of thinking about what a photograph now is, I think – but it doesn’t engage with the uneven distributions that shape the circulation of social media images at all.

50,000 Instagram photos from Tokyo, organized by brightness mean (radius) and hue mean (perimeter). from phototrails.net

I don’t know where methods might be emerging from that could do that, or what they might look like, though – so please post a comment below if you do!

some early thoughts on the fourth edition of Visual Methodologies

I gave a short talk at The Open University’s Centre for Citizenship, Identity and Governance a few weeks ago, as part of a workshop on methodological challenges for critical social science. As I’m going to start working on the fourth edition of Visual Methodologies this week (for publication in early summer 2016), this was good timing for me; it let me put a few very early thoughts together and explore them with an insightful and lively group. I thought I’d share my notes for the event here.

we’re all cultural studies scholars now?

Rather belatedly, something interesting struck me about the furore in the UK a week or so ago over a tweet sent by Emily Thornberry, now the ex shadow attorney general.  For those of you with short memories or not UK-media-aware, Thornberry’s tweet was criticised for being contemptuous of working-class voters, and she resigned from her shadow cabinet role after a couple of conversations with her party leader.

Here is the tweet (and notice the text as well as the photo):

In the online discussion of the tweet, loads of things were happening, of course, but there was something about the whole dynamic of the discussion that I thought was intriguing: the way the tweet swung in and out of being seen as ‘representing’ something.

On the one hand, there were a lot of claims – including by Emily Thornberry herself – that the image was meaningless.  It meant nothing; the scene had just struck her as something that could be shared on Twitter.  Now, we could discuss the conditions under which certain things become noticeable and photographable, of course, but still, given how so many photos are put onto social media in just that way, not as meaning anything, just as a sort of ‘oh look’ statement, ‘I am here’, ‘that is here’, I think her claim has some credibility.  Perhaps it really wasn’t an image that meant anything, it wasn’t symbolic, it was purely descriptive, just a picture of a van and flags, a pure “Image from #Rochester”.

Its status as pure description also prompted a lot of online discussion about how photographs become meaningful rather than inherently carrying meaning as well.  So on the other hand, huge amounts of online work went into interpreting the meanings the tweet implied.  The flags, the van, the location: all were decoded, re-coded, explicated, interpreted, repeatedly, by very very many people.  And so were the processes through which all that interpretive work was being done, because there was also a lot of discussion about the sort of coverage given to the tweet in and by different media outlets.

Noortje Marres has been interviewed on the excellent LSE Impact blog about digital sociology, and a point she makes very well there is that the tools and techniques of  social analysis are now widely distributed among many kinds of social actors.  As she says, “social actors, practices and events are increasingly and explicitly oriented towards social analysis and are actively involved in it (in collecting and analysing data, applying metrics, eliciting feed-back, and so on)”.  She is particularly referring to the digital tools embedded in social media platforms, the internet of things, online transaction databases and the like.

But one of the things that the Thornberry tweet affair made evident to me was that the same might be said of the tools of cultural interpretation.  Textual and visual analysis, and an understanding of the significance of the role of the media – in the strict sense of the term – are alive and kicking pretty much everywhere, it felt like, as the vigorous debated unfolded.  The furore was a sort of mass cultural studies seminar.  And if cultural studies has gone viral, what are the implications for those of us who do it – possibly more carefully, certainly much more slowly – in the academy?