Mass Image, Anthropocene Image, Image Commons

Sean Cubitt
Photography Off the Scale, edited by Tomas Dvorak and Jussi Parikka

The Mass Image hypothesis.

Let us start where I would like to end, with the mass image: not the image of apocalypse but
the apocalypse of images, as Baudrillard might say. Absurdly, given the immense proliferation
of images, the mass image is, strictly speaking, invisible (MacKenzie and Munster 2019). It is a
name for the huge aggregation of images and their metadata amassed in the vast corporate
databases of Google-YouTube, Facebook- Instagram-WhatsApp, Twitter-Vine, IAC-Vimeo-
Tinder and in repositories of medical and scientific imaging, by CCTV, with its supports of
facial recognition and biometrics collected not only by police forces but at airports and
shopping malls, and in rather different ways and under severe financial constraints in cultural
archives. To exist, an image must be discoverable, and for that it must be incorporated in a
database. We retrieve some few images for care and attention – the care of attention
characteristic of cultural archives – but there are limits. At a conservative estimate, in 2018
thirty plus million images were uploaded to Twitter, fifty-two million to Instagram, 350
million to Facebook – daily. If everyone on Earth spent eight hours a day, we could not
possibly look at all of them with any kind of seriousness.

In many respects, even though they retain significance for their uploaders and their
immediate circle, the images themselves are insignificant. From the perspective of the
database, the image is not only meaningless (the machine does not recognise semantic or
affective values): it exists only as a point of transit for what databases really treasure:
relations. Relational databases like these are more interested in metadata - geo-location,
personal identifiers, device identifiers, date stamps, facial recognition, distances and
dimensions recorded by autofocus functions, upload addresses and the kinds of measurement
prized by astronomical and meteorological databases. Commercially-operated databases
combine such data with data from searches and likes, swipes, shares, tags and other
interactions, to construe from the relations between images a mass image of the world in
constantly evolving organisational diagrams that only machines can read.

1
Run an image search for any popular tourist spot. If it isn't already tagged by the user or her
device, algorithms associate upload GPS data with data from Google Maps and Street View to
fill in information on cameras' orientations; temporal data provides a fourth dimension. The
mass image of your site appears. It is an aggregate portrait tending towards a total image,
especially of popular sites like Times Square, or the Charles Bridge, extending in time (during
Spring, at dawn, in 1945). Your search results will include links to video content, whose
moving-image and audio elements also become components of the account the machine can
now give itself of the place. Google is busy training artificial intelligences to recognise the
world they depend on but which their status as representational media excludes them from.
On one hand, the mass image aspires to become the most ecological image possible: complete,
formed of intricate webs of connections, and permanently evolving. On the other, it is already
the most environmentalising of images, in the sense that it constructs a total image of the
world so complex and complete that it excludes the world it depicts, turning it into pure
context, pure environment, that which environs what is truly valued. The mass image
inscribes the world as data.

At its first level, the one we can at least glimpse if not really see, the mass image undertakes to
compose a total photographic and videographic image of the world. At a second level this
image is invisible. The individual images we see arrayed in the archetypal modernist grid are
only an epiphenomenon of the mass image, whose actuality is constant processing in
topologies we might vaguely intuit, and whose properties we could derive from the
algorithms if they were public, but which even then we would not be able to see in action
because the mass image is only truly perceptible to the machines that house it. What you saw
in an image search yesterday will not repeat tomorrow, and you will get a different result
from me, because the display responds to your data history and its position in the evolving
global environment of searches, clicks and links. The display page of an image search is itself a
representation, not of Times Square or the Charles Bridge but of the activity of the database
collating the images and the behaviours of its viewers. It looks like a wall of pictures but is in
fact a data visualisation.

Archaeology of the Mass Image: To understand how this might constitute an apocalypse of
images, let me try a very short history of photography. The earliest photographs concentrate
the duration of exposure into a single artefact, the plate, in this way imitating drawing, a

2
temporal practice that results in an immobile object which, however, because it was made
over time and of time invites us to spend time contemplating it. Long-exposure glass plate
photography represented time, not only by recording and making endure, but as Benjamin
wrote, by producing 'the inconspicuous spot where in the immediacy of that long-forgotten
moment., the future nests so eloquently' (Benjamin 1999: 510). To look into one of those
early photos is to recognise, curled within it, the viewer's present, as potential and as an
obligation to honour that lost moment: there is a continuum between the sitter and the
viewer. Here re-presentation, while never denying the existence of the sitter, focuses on what
connects the sitter and the viewer: the continuity of history that relates one to the other. The
photographic snapshot invented in the mass take-up of the Box Brownie created a rift in time
by tearing the stilled instant out of the continuum; both the continuum of sitter and viewer
and the ecological continuum of time. The snapshot's capture of reflected light fails to unify
image, appearance and being (in the way drawings could aspire to). And yet, snapshots have a
saving grace, deriving from their exile from the continuum: they are negatives not only in the
technical sense of the physical intermediary between exposure and print, but because they
negate what they represent – the unhappiness of the world that confronts them as referent.
What we look at and prize in a snapshot is not the truth of the moment but its non-identity:
the capacity of the moment to be otherwise than it is.

The movies came almost immediately to repair both trauma and non-identity by adding
another image and another and another in emulation of the flow of time that the snapshot had
interrupted. They created a new art whose raw material was time. Scanned and interlaced
video introduce time into the individual frame, and overcame the ending problem that dogged
cinema, replacing the interruption that structured films as narrative units with unending
broadcast flow. Today however there is a new cure for the trauma of images: the mass image
construed in relational databases. In place of the temporal arts of film – montage, mobility,
under- and over-cranking and all – the mass image agglutinates every image, still and moving,
into a single vast, inter-connected artifice. The mass image heals the rift by negating the
negation of the real, so losing the capacity to picture the world as otherwise than it is. The
mass image is relentlessly positive, and therefore incapable of change. Thus in its way the
mass image fulfils Fox Talbot's claim to allow reality to write itself in the photograph. In place
of perpetual change, establishing networks connecting each item to any other is more
complete, more complex emulation of the planetary ecology than film or broadcast flow could

3
ever provide.

At the same time, the mass image not only negates the world as potential but the image. The
mass image is no longer interested in the image as image, that is as artefact in and of time, nor
even in the tragic seriality of the moving image, whose striving for impossible completion so
accurately sums up the alienation of humanity as such from nature as nonesuch. The mass
database collects images not for themselves but as instances of behaviours and the
distributions that behaviours form. Potential as the presence of futurity now is annihilated in
the simulation of the world accomplished in the mass image, which no longer distinguishes
between real and probabilistic.

The ubiquity of exchange value has its equivalent in the equivalence of all nodes and all
relations in a database. Not only is every image exchangeable for any other, as is clear from the
unique results of every given image search, but the various forms of relation between images
are allocated numerical values to render the relations machine-readable These numerical
relations distinguish the database from the ecologies, social or physical, that it still tries not
only to replicate (as total image) but to displace (the represented world-as-data is analysable
as messy reality is not). The total image of the world risks not only emulating the futureless
world of waste and debt, and not only healing the rift in time by ending time altogether, but of
making it appear that the world is truly accounted for in this architecture without potential,
where action can no longer occur, and no future other than the present can emerge. As Flusser
(2013: 57) says, only slightly over-stating the case, 'The aim of networked dialogues is not the
production of new information but feedback'.

The ecology (in the singular), which includes the natural, technological and human worlds, is
not the same as the internal ecology of a database. This is so because, firstly, the planetary
ecology, dependent on lunar cycles and sunshine and bathed in cosmic radiation, is borderless,
but there is a definitive outside-of – an outside that defines – the database, which creates and
is defined by its externalities and environments. Secondly, the planetary ecology has no goal,
while the goal of proprietary databases is profit. The planetary ecology has an eschatology, the
database has a teleology. The teleology of the mass image is no longer human, because the
profit that it calculates is entirely financial and no longer anchored in even residual lip-service
to the welfare of humanity or planet; and because the viewer for whom, phenomenologically,

4
the mass image exists, is no longer human but mechanical. The apocalypse of images in
database logic heals the rift in time by constructing a timeless, homeostatic universe which,
and this is its tragedy, is also the model of the ecology seized on by deep ecology and other
environmentalist movements. Such cybernetic models of ecology posit stasis as the only
alternative to disaster.

Jennifer Gabrys calls the social effect of this 'Programmable Earth': 'Programmability, the
programming of Earth, yields processes for making new environments not necessarily as
extensions of humans, but rather as new configurations or “techno-geographies” that
concretize across technologies, people, practices, and nonhuman entities' (Gabrys 2016: 4).
Her analysis not only suggests that the dynamic structuring typical of the database economy is
the pinnacle, to date, of the logistics of empire. It also makes a critical distinction between
environments as techno-geographies and as extensions of humans. Those early photographs
still presumed an individual human viewer. Photography and cinematography in their use as
scientific instruments moved towards a collective subject, Science, the assembled One who
knows. But with the mass image, the phenomenological centrality of the human, individual or
collective, in technical media, reformulates the decay of individualism, not into a new social
formation like science, but into a distributed cloud of behaviours. Perhaps nothing
demonstrates this phenomenological crisis more than the ubiquitous selfie, a photo we
capture so repeatedly precisely because the self is in crisis, an action vainly attempting to
anchor its presence as present in the moment of its dissolution.

The Anthropocene Image:

The eternal present of homeostatic systems, among which we must now count the mass
image, is the enemy of history which, as Dipesh Chakrabarty (2009) has argued, is now
indistinguishable from natural history. The mass image, however, does not exclude the
Anthropocene but attempts to assimilate it. To understand this attempt we have to look at the
genesis and practice of data visualisation so crucial to contemporary understanding of climate
change.

Though it has a precursor in maps (Harley 2001, Kitchin et al 2017), data visualisation can be
traced back to the late eighteenth century, when William Playfair's Statistical Breviary
(Wainer and Spence 2005) introduced the use of pie charts, bar charts and line diagrams and

5
standardised the use of the horizontal t-axis to include time in spatial representation. Like
earlier map-makers, Playfair assembled information garnered from various sources,
reconciled them to his satisfaction, and produced a visual account. This first stage involved
turning already existing narratives and numerical tables into graphical form by an act of
abstraction. The second phase arrived with the design of instruments tailored to the
production of new data in forms ready for abstraction, specifically instruments that not only
tracked vectors, like a compass, but that returned numerical accounts of the phenomena they
observed, like thermometers and seismometers. In the third stage, networks of standardised
measuring instruments provided large, and eventually huge quantities of usable readings
(Adams 2003, Edwards 2010), processing so as to make them amenable to the abstracting
processes of data visualisation.

One of the instruments tailored for this informational processing was the photographic
camera. The advent of CCD and CMOS chips brought the principle of sampling into
photography. Each discrete pixel on the chip reacted to incoming photons by averaging their
wavelength and frequency to produce a single figure for the pixel over the duration of
exposure, expressed as charge. Digitising the charge (and the addresses of the pixels it was
associated with) produced an account of the light reaching the whole chip in an array typically
arranged in the same order as the receiving pixels, thus delivering an image. But as anyone
who has used photographic software knows, there are many other ways of reading image files
through histograms and other visualisation tools. Digital images are in this sense
visualisations, specific orderings of data in a form perceptible to humans. They become truly
data visualisations when the recording instrument not only administers light but the
metadata associated with the exposure (time-stamp and location for example, corresponding
to the address already associated with a particular colour and luminance value for individual
pixels). The mass image completes the final step in processing the data produced from
photons and their spatio-temporal placing when it organises all images in its purview under a
single topological regime. Structured not only by the unit-counting of discrete devices and the
assimilation of time (starting with date stamps, rising to the ability to sort vast numbers of by
date), this topology also conforms the relations between images to the commodity form.

The mass image is in this sense an already functioning post-human hybrid of work and labour
performed by natural human and technical agencies. But it is also structured in an ascending

6
hierarchy of enclosure, the production of units, and the conversion of units into tradable
entities. The isolated unit is subsumed into relational topologies where its individual
characteristics are of diminishing value, but only because they can be treated as counters in
an exchange whose capacity to produce value is greater than that of the units traded. The
depth of this organisation of what once was given but is now captured, stored and processed
for profit can be seen in the cultural ubiquity of grids, from interfaces to street maps, and their
functional universality in workplace media.

It is no longer a question of conforming, or attempting to conform, human perception to the

media, because human perception is no longer central to their functioning. Reduced to their
behaviours, humans need only perceive in order to respond, and their responses have no
agency because they are raw materials for processing as data commodities. This has its
positive side: the individualism that plagued Western civilisation from Luther to Freud is
overcome by a simultaneous dissolution of the self and the compulsory collection of its
fragments into supra-individual formations that are no longer exclusively human. Becoming
human is an unfinished project. We are in the middle of a sea-change in its history. Since we
are no longer senders and receivers, it is clear that all of us are channels, although not entirely
returned to the primal state in which we were all media. Channels are after all communicative
rather than mediating: they connect pre-existing senders and receivers, but in this instance
the sender and the receiver are the same: the enclosed feedback loop of a cyborg
communication system. Our dreams of identity are symptoms of our estrangement from this
cycle which however subsumes us under its own dream of totality.

Individual photographic images are still shaped by an older system, centred on the camera
body imitative of the human eye socket and lenses designed to reproduce 'perspective as
symbolic form'. That system persists in promoting the idea that the sovereign individual
(historically denoted by the gender-presumptive 'Man') is the focal point and fixed centre of
the universe. The new network system has no centre, its dynamic topology has no fixity, and it
has no focus because it is not primarily visual (or in any humanly recognisable sense
perceptual). At the interface between these two systems, the human that disappears is only
the 'data subject' articulated by the European Union's General Data Protection Regulation
(GDPR), already necessarily performing its commodifying and biopolitical tasks in the mass

7
image. The human as data is wholly subsumed, but the human as code, and therefore
performative, also persists as the ghost of work in a world of labour.

The human code is performative, then, but in great measure only for the network. Its value
lies in being an intermediary between the network (which in its quest for universality must
exclude what environs it) and the planetary ecology it externalises. It is because humans still
have bodies and senses that they can act as intermediaries. At the same time, because they are
both embodied and networked, emblematised in the mass image as the mismatch between
perspective and diagram, humans in the wealthy sectors of the world – those closest to and
most integrated in networks – are suffering an epidemic of mental illness. Torn between
(physical and therefore ecological) body and (photographic and data-oriented, app-mediated)
body-image, the contemporary human condition is that of the schiz once proposed by post-
structuralism as the panacea for bourgeois individualism. Yet this dividual condition is indeed
the grounds for a more generative, post-digital, post-network collectivity. As intermediary
between ecology and technology, the human mass is no longer an aggregate of individuals, a
single population acting as an efficient agent in the market model of economists like von Mises
(1998, originally published in 1949), but an affective storm, whose capabilities are
unexplored and dangerous.

So too may prove to be the forces of technology, notably of artificial intelligences at work
analysing the behaviours collected in the mass image. In the Grundrisse , Marx (1973: 690-
711) describes technologies as ‘dead labour’, the aggregation into concrete form of all the
skills and knowledge gathered by our forerunners, owned and enslave din the form of ‘fixed
capital’. Where traditional cultures respect and converse with their ancestors, we keep ours
locked in black boxes. Liberating computers (Nelson 1974) is an admirable strategy towards
freedom of all humans (‘Even the dead will not be safe’, as Benjamin wrote, if fascism
triumphed); but the risk of auto-orienting AIs is that the ancestors, after their centuries of
isolation and indentured labour, may be mad. The same may be the case, we are beginning to
understand, with the symptoms of the Anthropocene, a savage return of repressed nature. All
three phyla are potentially at the brink of catastrophic breakdown, although it is the technical
sphere that seems to be triumphant in the ubiquity of programming and its embedding not
only in machinery (my computer does not play certain files, not because it doesn't want to but
because Apple doesn't want it to) but in, for example, the audit culture that everybody hates

8
and few people escape or rebel against. Affective irrationality is however only the obverse of
techno-rationality; and mental illness is not a solution because it hurts. Politically, in the day
to day conduct of public affairs, the Anthropocene features only as a terrain for antagonism
between the credibility of science and neo-populist table-thumping. Like radicals discovering
that they needed to defend the FBI against Trump, environmentalists rally to the cause of
science, even though science's credibility and politically progressive credentials are by no
means assured. On the negative side, science rested on a set of protocols designed to produce
certainty which however have moved towards the production of probabilities and the
uncertainty principle. More self-reflexive but more prone to self-doubt, science no longer
claims to know but to ascertain likelihood. Science no longer lays claim to Truth as it did in its
triumphal era from Newton to Humboldt. On the positive side, science's planet-spanning
networks have undone an older faith in the reliability of witnessing, of the self-evident, of the
eye-witness, of the testimony of my own senses when laid in the scale against the witnessing
of others, human and non-human. Dr Johnson could not refute Berkeley today by kicking a
stone. The value of this photograph of a departed friend does not prove that the mass image
doesn't exist, and vice versa1.

Bereft of universality, truth fails. It cannot form the basis of politics in its second sense, the
pursuit of happiness for all of us, because even science admits that no truth is universal. The
suspiciously coherent rage of demagogues may truthfully encapsulate the wrath of electorates
but cannot articulate the differences within them, since it is bound to the logic of
representation. The mass image can claim universality but only by environmentalising the
ecology, including much of what we cling to in our residual and ecologically-embedded bodies.
Its truth as representation rests on this curtailed universality, and so is not truth if by truth is
meant a statement that both names and enacts what is the case. The central problem,
especially as expressed in photography and the mass image, is that the modes of truth they
appear to claim rest on either the identity of the image with the object-world it names; or the
self-identity of the image. In the first case, no flat, rectangular image is the world it pictures;
in the second, both the gap between represented and representation, and the distinction
between the image as image and the image as print or screen display, demonstrates that there

1
This does not prove that truth is relative, only that (phenomenologically) it appears
differently to different observers. However, if truth is not manifest in itself but only in
appearance, then it is no longer singular and self-sufficient, and therefore is not Truth as it
was defined in the classical scientific era.

9
is not even a self-identity (the image, qua object in the world, does not contain in itself the
reason why it is so and not otherwise). Images are inherently non-identical, and to that extent
cannot be bearers of truth. The unique capacity of photographic images, however, is to picture
not what is the case but the potential of the represented to become other than it was, since
that has already happened in the moment when the scene in front of the lens became an
image, that is something utterly different. Photographs do not image the world as it is but in
the moment of its becoming otherwise. They show the world as it might or could or should
have been or be or become. Photographs are subjunctive.

Or perhaps photographs were subjunctive, and retain the power of the subjunctive as a
residue of their older role that distinguished them from drawing. Drawing produced a
complete and self-sufficient work: photographs, as tears in the continuum, were insufficient,
until the mass image constructed a framework for producing a total image of the world, an
image which by definition is complete and certainly aims to be self-sufficient, specifically in
the sense of no longer requiring the world to complement it. Without that congruence and its
inconsistency, the mass image can no longer reveal the non-identity and therefore the
potential of the world to be otherwise. This could not have come at a worse time. The
subjunctive is the precise mode we need for a humanity defined by its massed behaviours and
a nature in crisis. It is the one mode that can embrace the pure act of giving that erases itself
in the moment of the gift, and the great chain of giving that modulates and evolves whatever
passes along it. Can it be possible that the subjunctive retains its utopian power when we
consider the mass image as the activity of ancestral labour made concrete in its apparatus?

The Image Commons:

When humans donate their work to a system that abstracts it in the form of data, property
and exchange value, they act exactly as their ancestors did. Their free gift of knowledge, skill
and creativity was concretised in the dead labour of machinery. So to is our living work
concretised in the vast accumulations of the mass image. As we have seen, this must be
understood as an enclosure of free creativity, and a result of the real subsumption of
consumption, but it is also accumulation in the economic sense of investment in land,
machinery and money itself, from which capital expects to derive profit. Investment implies
both accumulation and the extraction of profit. In theory, corporations invest in human capital
and expect to accumulate and exploit it. In practice, they invest as little as possible,

10
demanding that other agents, formerly then state, now the individual, invest in themselves so
that capital can exploit them. The importance of the mass image is that it represents a novel
way to accumulate human capital in a form that can be exploited, like technology, without
paying wages. The problem is the risk of over-accumulation.

There appears to be a generalised belief that the accumulation in the digital industries,
especially among the FAANG companies (Facebook, Apple, Amazon, Netflix and Google), is
infinite; that there will never come a moment when the accumulation of capital and its
reinvestment fails to find new consumers. There is a degree of truth in this belief. The systems
of image capital are premised on a feedback loop whose major channel is human behaviours
which can be constantly monitored to track evolving desires and meet them with evolving
products. This solution succeeds earlier ones – continuing enclosure of new sources of
physical or informational wealth (primitive accumulation seen as a permanent resort of
capital since Rosa Luxemburg's 1913 analysis [Luxemburg 1951]), moving fixed capital to
new territories and the creation of new markets. Each of those however eventually comes up
against the limits of growth, marked in our instance by the finite resources of the planet
marked most distinctively by the Anthropocene. Culturally we are caught between faith in
infinite continuation of the present regime of accumulation and fear of its sudden and
catastrophic demise as a result of external factors. But these factors only appear external
because they have been environmentalised, excluded as externalities from the internal
working of a total system that recognises nothing beyond itself. This contradiction, rather
than the contradiction between over-accumulation and under-consumption that instigated
the debt crisis of 2007, sets the new terms for a new crisis. We have grown used to the idea
that capital never lets a good crisis go to waste: that there are always factions within capital
that make a profit from disaster. We have also a historical record that tells us that economic
crises have a tendency to favour despotism. The exceptions have been aspects of
Keynesianism and welfarism, state infrastructural investment providing employment to
enable new consumption and state investment in social well-being to create new consumption
and new opportunities for capital to contract for state business. There is every reason to ask
why the oppressed have not similarly made the most of crises.

For historical reasons – basically the privatisation of telecoms in the Reagan-Thatcher era
just before the internet and mobile communication booms – the state has little traction in the

11
infrastructure or regulation of digital networks. In the period since the welfarist epoch in the
wake of World War Two, the historical trend has been towards state capture (Hellman et al
2000), to the point that even the most powerful political factions are constrained to deliver
the requirements of corporate and other capital – so bitterly clear in the incapacity of states to
rein in the fossil fuel industries. Yet it is also the case that the state acts as a major buffer
between their populations and the worst predations of capital (du Gay 2005). Regulation is
correctly seen as the operation of power, but in many instances it is also the site of operations
of resistance, assaulted by capital as 'bureaucracy' and 'red tape' opposed to the
untrammelled working of the profit motive. This resistance has emerged historically as noise
in the technical system, both environmental noise from beyond the system and system-
generated noise, deriving from insufficiently disciplined uploaders and the internal frictions
of operating protocols and the hardware they run on. Since the system reduces significance to
a minimum in order to prioritise pure exchange, meaning functions as noise and vicde versa:
noise is meaningful. Where desire cannot be directly expressed but only assimilated indirectly
in the form of behaviours, desire operates as meaningful noise; and we should take
technically-generated and ecological noise as the expression of the excluded and
unassimilated desires of natural and technical phyla.

Affective images – images designed to shock, images of excess – do not work in this way. They
provide a reward for the beleaguered self in the form of feedback rather than new
information, and reinforce the illusion that meaning is generated by and for a self rather than
in a channel operating within a total system. If instead we take affect as the condition of
primal mediation, then its mobilisation as meaningful noise, that is to say desire, matches far
better the new condition of distributed behaviours. Thus on the one hand humans become
more intensely organised as channels and intermediaries through their dissolution into
behaviours optimised for data mining, but on the other beyond the behavioural schiz they
become less exclusively human, approaching more and more the condition of ancestors and,
as affect, the condition of nature.

It is time to reverse Stallman's motto, 'Free as in speech, not free as in beer'. Today free
speech is not the right of free citizens that proves they are free, but compulsory and unpaid
participation in generating content (Dean 2005, again). Nor does Stuart Brand's slogan
'Information wants to be free' foot the bill, not because information wants to be shared or has

12
no volition of its own but because, as information, it has already lost the freedom of the gift.
Instead, we could be forgiven for arguing, the visual wants to be free as in free beer: without
cost, and prone to evoke jollity and mayhem rather than efficiency or, least likely, equilibrium.

The atomisation of the self has also shown that there is no single happiness that binds us: if
there is to be a 'we', then not a single one can be left behind. If there is to be a 'we' adequate to
the intermediary status of humans, it must be inclusive, and therefore bring to an end the
political exclusion of ecology and technology, domains that are governed without a role in
their own governing. The primordial soup of mediation is symmetrical: the same in every
direction, forwards and backwards in time. Communication, with its structural division of
sender, channel and receiver, introduced time as the condition of the message and its
interruptions. At root belonging to the indicative mood that separates the speaker from the
spoken ('There is a tree'), communication has always been vulnerable to takeover by the
imperative ('Let the tree exist'). In its seizure of distribution, network communication not
only assumes the imperative as the formal mood of software design but arrests the forward
trajectory of time implicit in the communicative model, producing not timeless symmetry but
the eternity of perpetual feedback. Noise – glitches, breakdowns, crises of over-production,
over-accumulation and, more recently, of over-distribution – reintroduces history as the zone
of conflict within and between material and informational levels of the network.

The greatest of these crises is ecological, and on it depends the specific operation of older
culturally-specific binaries (religious, ethnic, gendered). The eternity of distributed networks
has taken to itself a second mood, adding to the imperative the conditional, which allows it to
simulate the future as projection of present data. Ironically, many simulations point towards
catastrophe. Humans are caught between image regimes, on one side the Intergovernmental
Panel on Climate Change (IPCC)'s infamous hockey stick diagram of vertiginously rising
temperatures, on the other emblematic images of a polar bear on an ice floe, shattering
glaciers, or roof-stranded citizens of Dhaka. Any photo or video short outdoors today bears
evidence of air pollution. Sharing and liking only assimilates these futures into the planned
regime of networks, governed now more by concerns for security (as in climate change
security for the wealthy masked as anti-migrant and anti-poor security and the 'war on terror'
[Wemmel and Femia 2018]) than for the welfare of humans, planet or even the technological
infrastructure of the system itself.

13
The great skill of revolution is to precipitate a crisis that the oppressor cannot benefit from
but the oppressed can. One possibility is an image strike: to refuse to make, upload or interact
with images. For that to work beyond the individualist ethics of a consumer boycott requires
organisation on the scale of the industrial era trades unions which, however, can always be
criticised for adopting the same disciplinary structures as the factory discipline they aimed to
end. More powerful are those movements which have aimed at new organisational principles
and new tasks, like the shop stewards' movement at Lucas Aerospace in the UK in the 1970s
that proposed to turn the factories over from arms to socially useful products (Salisbury
2019). Now, in the era of distribution, it is not only production but distribution that needs to
be remade. Practices aimed at developing tactics and strategies that might precipitate the
crisis for the planetary oppressed are everywhere. Their major characteristic is their claim to
the commons, from collaborative art and social projects to the Peer-to-Peer movement, and
their openness to a future which, unlike the conditional simulations of the network, are
entirely subjunctive. The lesson of the mass image for this aesthetic politics is that the future
need not and should not be planned; and that it can only emerge in a working (but not
labouring) alliance of humans, nature and technology. Already necessarily exceeding the
prison-house of networks and the limits of human chauvinism, the means of revolution must
be as joyful as the end.

CODA:
Labour in the mass image is everywhere freely given and everywhere in chains. It
encompasses
1. the work of the planetary ecosystem and the labour of capturing the free gift of light and
natural processes;
2. the work of uploaders and viewers, freely donated, and the labour of converting it into
copyright-protected and exchangeable units; and
3. the work of ancestors to devise skills and knowledges, an unforced legacy purloined as
forced labour in the form of machinery and constrained to convert the use-value of units into
exchange value.
In all three cases, natural, human and technical work initially forms a commons which is then
enclosed and exploited in the interests of capital. The initial work involves sharing; the labour
involves ownership: converting into units, commodifying, and reducing to exchange-value. It

14
is clear that the labour component of these three layers of activity is entirely unnecessary,
inefficient and absurd, since its only effect is to add a price tag to free beer. The mass image is
novel because it completes the circuit of enclosure, objectification and conversion to
exchange-value. In the mass image, the real subsumption of consumption completes the cycle
by converting viewing from playful reprocessing into labour. This happens in two phases.
First, consumption becomes purposeful, a trait exploited by the excessive efficiency of search
engines, which always deliver what you look for. The purpose of search engines is to provide
gratification, in much the same way that classical cinema does: create a desire in Shot A to see
something, then magically provide it in Shot B. Gratification becomes self-fulfilling when we
associate searching and finding as a single moment. Second, purposes become readable in the
form of clicks, links and likes, and so become tradable data. Desires and gratifications, now
identical, can form new objects, whose organisation into larger aggregations in database
topologies provides the end-user – no longer the viewer but the database itself – with an
image of gratified desire on which to build new self-enclosed micro-circuits of aspiration and
fulfilment whose efficiency can be instantly read off from the next set of viewer behaviours.
The result of this system is stasis, in this imitating homeostatic models of the free market and
environmentalism, with their belief that, if only aberrant agents like states and humans
stopped interfering, the system would be perfectly self-equilibrating. Work is time, labour is
eternity.

Unifying the collectivity of the three phyla in strength and hope at the crossroads of
resistance, desire and meaning, the name of the commons is beauty.

REFERENCES
Adams, R.D. (2003). International Seismology. International Handbook of Earthquake &
Engineering Seismology Part 1. Edited by William H.K. Lee, Paul Jennings, Carl
Kisslinger, Hiroo Kanamori. Amsterdam: Elsevier. 29-37
Benjamin, Walter (1999), 'Little History of Photography'. Translated by Edmund Jephcott and
KIngsley Shorter. Selected Writings, vol 2, part 2, 1931-1934. Edited by Michael W
Jennings, Howard Eiland and Gary Smith. Bellknap Press / Harvard University Press,
Cambridge MA. 507-530.
Chakrabarty, Dipesh (2009). 'The Climate of History: Four Theses'. Critical Inquiry 35. Winter.
197-222.

15
Dean, Jodi (2005). 'Communicative Capitalism: Circulation and the Foreclosure of Politics'.
Cultural Politics 1(1). 51-74.
du Gay, Paul (2005). The Values of Bureaucracy. Oxford: Oxford University Press.
Edwards, Paul N. (2010). A Vast Machine: Computer Models, Climate Data, and the Politics of
Global Warming. Cambridge MA: MIT Press.
Fine, Ben (2013). Consumption matters Ephemera: Theory and Politics in Organisation 13(2):
217-248. http://www.ephemerajournal.org/contribution/consumption-matters
Flusser, Vilém (2013). Post-History. Translated by Rodrigo Maltez Novaes. Edited by Siegfried
Zielinski. Minneapolis: Univocal.
Gabrys, Jennifer (2016). Program Earth: Environmental Sensing Technology and the Making
of a Computational Planet. Minneapolis: University of Minnesota Press.
Harley, J.B. (2001). The New Nature of Maps: Essays in the History of Cartography .Edited by
Paul Laxton, Introduction by J.H. Andrews. Baltimore: Johns Hopkins University Press.
Hellman, Joel S., Geraint Jones Daniel Kaufmann (2000). 'Seize the State, Seize the Day': State
Capture, Corruption, and Influence in Transition. Policy Research Working Paper 2444.
September. Washington DC: The World Bank.
Kitchin, Rob, Tracey P. Lauriault and Matthew H Wilson (eds) (2017). Understanding Spatial
Media. London: Sage.
Luxemburg, Rosa (1951). The Accumulation of Capital. Translated by Agnes Schwarzschild.
London: Routledge & Kegan Paul.

Mackenzie, Adrian and Anna Munster (2019). Platform Seeing: Image Ensembles and the
Invisualities. Theory Culture and Society 36(5). 3–22.

Nelson, Theodore H. (‘Ted’) (1974). Computer Lib/Dream Machines. Self-published.

Salisbury, Brian (2019). 'Story of the Lucas Plan'. A New Lucas Plan
https://lucasplan.org.uk/story-of-the-lucas-plan/

von Mises, Ludwig (1998). Human Action: A Treatise on Economics . Edited by Bettina Bien
Greaves. Auburn, Alabama: Ludwig Von Mises Institute.

Wainer, Howard and Ian Spence (2005). Playfair’s Commercial and Political Atlas and
Statistical Breviary. New York, NY: Cambridge University Press.
Werrell, Caitlin E. and Francesco Femia (2018). Climate change raises conflict concerns. The
UNESCO Courier 2018-2. https://en.unesco.org/courier/2018-2/climate-change-raises-
conflict-concerns

16
World Meteorological Organization (2019). WMO Statement on the state of the global climate
in 2018. Geneva: World Meteorological Organization.
https://library.wmo.int/index.php?lvl=notice_display&id=20799#.XJ9Wtuv7Q6h

17