Suspicious

Images,
Latent
Interfaces

THE ARCHITECTURAL LEAGUE OF NEW YORK

S ituate d T echnologies Pa m phlet S 3
B en j a m in H . B ratton an d N atalie Jere m i j enko

Suspicious
Images,
Latent
Interfaces

Situated Technologies Pamphlets 3: Suspicious Images, Latent Interfaces

Benjamin H. Bratton and Natalie Jeremijenko
(With the editorial assistance of Jordan Kanter)
Series Editors: Omar Khan, Trebor Scholz, and Mark Shepard
www.situatedtechnologies.net
Advisory Committee: Keller Easterling, Anne Galloway,
Malcolm McCullough, Howard Rheingold
Published by: The Architectural League of New York
457 Madison Avenue, New York, NY 10022, 212 753 1722
www.archleague.org
info@archleague.org
Pamphlets Coordinator: Gregory Wessner
Digital Programs and Exhibitions Director, The Architectural League
of New York
Design: Jena Sher

(cc) Benjamin H. Bratton, Natalie Jeremijenko

This work is licensed under the Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License.
To view a copy of this license, visit http://creativecommons.org/licenses/
by-nc-sa/3.0/ or send a letter to Creative Commons, 171 Second Street,
Suite 300, San Francisco, California, 94105, usa.
isbn 978-0-9800994-2-3

Suspicious
Images,
Latent
Interfaces

THE ARCHITECTURAL LEAGUE OF NEW YORK

S ituate d T echnologies Pa m phlet S 3
B en j a m in H . B ratton an d N atalie Jere m i j enko

ABOUT THE SERIES

The Situated Technologies Pamphlet Series extends a discourse initiated

in the summer of 2006 by a three-month-long discussion on the Institute for Distributed Creativity (idc) mailing list that culminated in
the Architecture and Situated Technologies symposium at the Urban
Center and Eyebeam in New York, co-produced by the Center for
Virtual Architecture (cva), the Architectural League of New York and
the idc. The series explores the implications of ubiquitous computing
for architecture and urbanism: how our experience of space and the
choices we make within it are affected by a range of mobile, pervasive,
embedded, or otherwise situated technologies. Published three times
a year over three years, the series is structured as a succession of nine
conversations between researchers, writers, and other practitioners
from architecture, art, philosophy of technology, comparative media
studies, performance studies, and engineering.
www.situatedtechnologies.net

F ro m the E d itors

Advocacy is the act of arguing on behalf of a particular issue, idea or

person, and addresses issues including self-advocacy, environmentalprotection, the rights of women, youth and minorities, social justice,
the re-structured digital divide and political reform. In this specialdouble issuethe result of an open call for submissionswe have invited
contributions from two pairs of authors considering how Situated
Technologies have been mobilized to change and/or influence social
or political policies, practices, and beliefs. In our call for submissions,
we asked: What new forms of advocacy are enabled by contemporary
location-based or context-aware media and information systems? How
might they lend tactical support to the process of managing information
flows and disseminating strategic knowledge that influences individual
behavior or opinion, corporate conduct or public policy and law?
Suspicious Images, Latent Interfaces, by Benjamin H. Bratton and
Natalie Jeremijenko, explores the theoretical implications of new
forms of environmental monitoring enabled by pervasive computing.
Asking what, if anything, current trends in the visualization of environmental data tell us, they introduce the notion of the political
image as a foil by which to explore how networked assemblages of
human and non-human actors might, when considered on a global
scale, initiate a rethinking of how political institutions might work
in environmental governance. Bratton and Jeremijenko suggest that
their contribution can be read as both a design challenge as well as an
experimental political theory for a social ecology configured through
prevasive computational media.

Omar Khan, Trebor Scholz and Mark Shepard

4
5

THE AUTHORS

Benjamin H. Bratton (uc San Diego, sci_Arc) invents systems concepts and translates and transposes them into actual systems. This labor
requires him to wear many hats for different occasions, including sociologist, design strategist, professor, software executive, and historian
of exceptional violence. He has taught architects about double-bind
ironies at sci_Arc, media artists about topologies of logistics at ucla,
and enterprise product strategists about the social specification of
emergent data channels at Yahoo!.
Natalie Jeremijenko (nyu Environmental Health Clinic) is a polymath design technologist and political affectician, working within the
wormholes connecting experimental art and global science policy. She
has taught at the worlds august institutions, been shown in the toniest
arts festivals, received the most competitive awards, and annoyed the
most miscast authoritarians.

CONTENTS

Introduction

10

Is Information Visualization Bad for You?

16

The (Second) Planetary Computer

19

Information and Institution

23

Its OK, Im a Doctor

28

xs Globalization: Interest, Diagnosis, Scale

34

Are Artists the Missing Experts?

40

Force Means Guaranteeing Repetition

46

MetaImages, MetaInterfaces

51

The Death of the User

53

References

6
7

INTRODUCTION

This conversation grew out of a paper that Natalie and I co-wrote for
Pervasive 2008, Sixth International Conference on Pervasive Computing
in Sydney, Notes on the Political Image: Pervasive Computing, Modeling
Assemblage, and Ecological Governance.
It stems from our mutual interest in the potential of pervasive computing to change how ecological systems can be monitored and visualized, and perhaps more importantly, how doing so on a global scale
holds promise for rethinking very deep assumptions about how political institutions can work, and indeed what their very architecture
might be.
What kinds of new territories of participation are opened up when
cycles of everyday action and the representation of collective sovereignty are bound so much more closely within planetary information
networks, now responsive to a molecular level? Does representative
democracy evolve into a democracy of representation, and if so, what
does that mean when every inch of the world comes online, becomes
awake to express its informational existence to us and for us? Would
that expression come as a din of voices we are incapable of listening to,
let alone governing through: a churning cacophony of signals?
Today we are learning to listen by learning to see the data, to render it
visually as colorful diagrams that look like graphical user interfaces,

but usually in fact are not. To that end, Natalie and I start by questioning
the status and ambition of information visualization as a format of the
political image: a potential interface for a potential technology of a
potential networked governance. This is critical. Their power is as an
image of potential assemblage, human and non-human, and in this
they are already political in the possibilities of agency that they
project. But now, how can they be activated? How can monitoring
become redesign of what it monitors? How does the image become
an image-instrument?
In essence, this is both a design challenge and challenge for an experimental political theory for a socio-ecology configurable through pervasive computational media. It is design that becomes, in its expression
of an experimental political theory, also a re-design of what political
forms, spaces, and technologies even are. In ways that we hope are
uniquely productive, one becomes the other.
In the lively spirit of this pamphlet series, the conversation is far from
conclusive, and at best weve contributed a list of to-dos to be taken up
by different projects in different ways. We welcome your feedback and
activation of what weve sketched here.

8
9

Benjamin H. Bratton
Los Angeles, June 2008

I s I nfor m ation V isuali z ation B a d for Y ou ?

BB

Let me start by saying that a lot of the issues that have been
bubbling and cooking in the last ten, fifteen, twenty, even fifty
years around ambient computation are seemingly all coming to a head;
they are mainstreaming very quickly. You could say the future happened this year for pervasive computing.
You cant walk through a design graduate program anywhere in the
first world, whether in architecture, interaction design, or media arts,
without seeing at least half a dozen beautiful data smog projects
modeling ambient urban-environmental information in one way or
another. A lot of the best projects of this type are being published very
quickly and put directly into museums upon final file export.
There is something great about this, but also something troubling. The danger is that in their spectacularization of information, they in fact distance
peoplenow audiences for dataeven further from their abilities and
responsibilities to understand relationships between the multiple ecologies in which they live, and the possibilities for action that they have.
They look like interfaces, but they are not interfaces. They are diagrams
or maps at best. They appear to be interfaces and in this appearance
they imply there must be an expertan expert systemsomewhere
making use of this information in a way that is somehow having some
effect. But mostly there is none. Outside of hanging on a museum wall
or being blogged about, Im not sure what they do.

NJ
BB
NJ

Nothing.
Nothing?

To define what it is they might do, and what they have the
ambition of doing, and what they fail to do, is one of the things
that is particularly interesting for me. Certainly there are these perverse peoplewho may now be the majoritywho see these technologies by themselves as an opportunity for change. With the collision
in the public imagination of the environmental climate destabilization
and environmental concerns more generally, there is suddenly a utopian
idea that we can use these new technologies and sensors and visualization techniques to address pervasive environmental issues with
pervasive computation.

The home page of visual complexity dot com (retrieved August 4, 2008) But why
now? Why and how has InfoViz become the ubiquitous face of ubiquitous computing? Manuel Lima, the editor, notes three factors making this emergence possible
now: Storage (more data than we have time to make sense of), Open Databases
(data extensibility and availability), and Online Social Networks (which make information about human networks available as both source and medium of visualization.) Data Visualization Panel at OFFF, Lisbon. May, 2008. See Kazys Varnelis and
Leah Meisterlin, 2008, for another summary of other exemplary projects.

10
11

On the face of it, I suppose that is a reasonable assumption: weve got

these new technologies and some pressing environmental issues, lets
throw them at each other. But what is really interesting is that these
ubiquitous sensing and inexpensive microprocessing devices suggest
the capacity to ask some fundamental and persistent questions.
And so, do these projects change who is asking the questions? Are
these designers now asking the question of how this pollutant is made,
who made it, where is it coming from, where is it going, what do we
do about it, or not? And what I have observed is that the designers
of these types of projects use extant data sets from the epa, from the
toxic relief inventory, federal databases, and do so without criticism,
without asking how the data is generated, who collected it and under
what conditions. That is, what does the data actually represent? The
criticism of how the data is produced is left out; unless this criticism
is engaged, it doesnt make for a meaningful visualization. If we dont
question how the data is made, we cannot make sense of it. Producing a
nice diagram is not all that is required in making sense of something.
Take the particular example of air quality, which is something Ive
been very interested in for a long time. I have tried to use air quality
data from federal sources. The first thing to understand is that environmental data is mainly collected in response to regulatory compliance
issues. This means the data is being collected by hired engineering
firms or staff, not by people who have a professional reputation invested
in what that data means or why it is being collected. This is different
from academic research, under the scientific regime, where if you dont
collect good data, your reputation is at stake. If you are an environmental engineer working for a nuclear regulatory agency or subcontracted
from the epa or from Con Edison, you just have to do your job. It has
to be done reasonably well, but you just have to follow the legislated
formulas defining what data should be collected and how.
We know this is not cutting edge knowledge production and as a result
cannot drive interesting questions. We get this soup of data, which in
many cases is not even statistically cleaned up, all under this regime
in this country where it is not actually illegal to pollute in any wayit
is illegal to not report that you have polluted. So you report all this
pollution, but the critical question of why you would collect this data
is not asked. Whereas trying to figure out how much you are emitting

because you are interested in how you can change, or because you are
trying to understand the health impacts, is decoupled from the data
collection. If you have that as your starting point, these federal databases
of environmental data, then the visualizations that might be graphically legible are not asking the questions that the data didnt ask in the
first place.

BB

In fact, parameters of facticity are further mystified by the

visualization. Part of the progressive narrative for pervasive
computation and ecological governance is of a world in which every
square inch is in some way constantly outpouring infinitely communicable information about itself, and that this would overwhelm
inherited layers of expert systemscertain people in certain circumstances that collect data from certain instrumental meansenabling
the world to declare itself as a functionally open data continent. From
this new basic infrastructure a new kind of political institutionality
could emerge.

NJ

I think that is the idea, yes, and there is a kind of eco-literacy

that might be increased. So if, instead of having trees growing,
we have trees with precipitation sensors and soil moisture sensors
and particulate matter sensors making explicit some of the environmental variables to which they are exposed, we would therefore somehow be able to make better sense of those trees.

BB

The thinking is that by rendering those variables visible and

transmissible, those trees become things about which decisions can be made more systematically. But of course the possibility
of the decision exists exactly in the natural display of their normal
growth as organisms. Which brings us already back to the problem
of the spectacularization of the information, and refocuses the question of what digitalization actually does and does not accomplish
for us.

NJ

In my OneTrees project, for instance, the trees are there in

the first place, are already self-reporting, and there is a certain
eco-literacy built in. You can look at the growth responses, the particulate layers on the leaves, weather history recorded in tree rings, etc.
and gain a direct understanding of tree growth as a material record.
The tree in itself is a self-reporting dynamic adaptive system of tre-

12
13

Urban Eco-Informatics: A variation on Jeremijenkos OneTrees Project. Here an

analogue sensor displays the relative growth rates of two genetically identical trees,
grown in different locations. See http://www.nyu.edu/projects/xdesign/onetrees

mendous value for certain scientific inquiries. So the idea that putting
sensors around trees would only then open them up to participating in
the political economy, or help us make sense of environmental variability,
further obscures what is already available in the trees themselves.

BB

If we read those trees themselves as interfaces, as your project

suggests we do. Through this we can speak on their behalf,
or even better, they could participate in some new parliament that
would include trees (Latour, 2007).

NJ

Yes, and there is the question of why they havent been included
to begin with. They are visible, they are present, they are active
dynamical systems, and they do have these very visible growth responses.
Why is this not already governance?
Certainly in the OneTrees project there is an invitation to ask people
why the trees look different, and what different environmental exposure they are having. The difficulty there is not in making that data
explicit; the project presents two clones in almost identical environmental contexts. Why is one 50% bigger than the other? This is a very
direct material question in sense-making. The issue is who feels the
permission to speculate that one is closer to the road and might be
exposed to more road borne pollutants? Who actually feels authorized
to ask those questions? I have found that people are tremendously
reluctant to speculate in these terms. They dont feel like they can ask
a scientific question and draw on the material evidence before them.
I question whether or not this reluctance would be reduced if they
received environmental data from these trees via text-message to their
cell phone.
Would they then feel more licensed, given that they dont in the first
place, to ask questions and interrogate and make sense of the situation?

14
15

T he ( S econ d ) P lanetary C o m puter

BB

It is possible that they might. That entails a direct personalization of the information in and of itself, and then perhaps
of that which the information informs us about. Also, the information
presents itself as potentially instrumental. That is, the popular instrumentality of such information is framed by how it is presented.
Perhaps weve all read Bruno Latour, and see the strong parallels
between his notion of a parliament of things and potential realizations of an internet of things. He suggests that a recognition of the
inherently politicized nature of objects and what he calls quasi-objects
forces us to rethink historical distinctions between nature and society,
and therefore the political institutions that have artificially excluded
material and non-human historical actors. If objects come alive with
information in new ways, the possibility of their very public voice
seems not only possible but in some ways inevitable.
I know that you and I both are empathetic to this interest. We are trying to
imagine the possibility of extrapolating new forms of political institutionalization on the basis of computational technologies that we both
discover and invent, including computers that look like trees. I believe
that the narrative of ecologically comprehensive computational media
and the idea of a planetary sensing system, is also tied in a way to the
notion that the world itself is fundamentally already digital, a soft
or hard computational ontology. We learn from thinkers as diverse as
Wolfram and Badiou about the formal discreteness of things. This becomes simplified as the world is a computer and the best way to listen
to that computer is with other computers. For the planetary pervasive
computing narrative, this computational layer that we would smear
across the world is simply a way to get closer to the primordial digital
unfolding of all things. As Friedrich Kittler says, silicon is nature!
Silicon is nature calculating itself. If you leave out the part of engineers
who write little structures on silicon you see one part of matter calculating the rest of matter. (Gane and Sale, 2007)

NJ

But I think that is a contestable claimthat life is computation.

It certainly is a popular idea, that life is in fact formal. There
is a certain amount of history of the formality of life, that it can be
described and even produced with explicit formal algorithms or data
structures. There is a school that believes there is no difference between life in silica, that is virtual or simulated life, and a carbon-based

life, that the same formal rules apply, and that if we tweak the rules we
will get the same thing in a silicon-based life and in the biological life
we inhabit. It is a kind of essentialist claim that life is describable by
algorithms. Of course there is not much to support this claim. It is a tantalizing idea and seductive, but it is just not robustly descriptive. It certainly helps to build explicit formal models of complex systems to help
us to understand that, but in no way do they become the same thing.

BB

But the more sophisticated contemporary exponents of this

position offer more a procedural claim than a formal one, and
politics is itself procedural. In imagining a new political space that
might emerge from listening better to the world, however that might
be achieved, the question for some is whether it suggests some kinds of
ontological moves, some basis of a definition of life that is continuous
or digital, computational or non-computational, if not to ground it then
to move it forward from an over-reliance on social constructivism. (De
Landa, 2006)

NJ

In some senses, it doesnt matter. You dont have to have believed in the project of Artificial Intelligence and the conceits
of that project to have found some of the algorithms that computational linguistics produced, for instance, to be useful and important. They
can be tremendously useful independent of the ontological framing
in a pragmatic world of what works and what is useful. But then it
does matter in terms of the political structure controlling who gets to ask
the questions! Why are some questions thought to be important? What
knowledge gets generated and how?

BB

And what is done with it and how that knowledge becomes

needed, yes, of course. But to risk further essentialism, these
conditions and constructions do not change the processes by which
knowledge knows. This does not, however, resolve how those assemblages themselves, such as science, become broadly established
political forms.

NJ

Right. So of course with Artificial Intelligenceif we take that

as a previous, large paradigm that motivated a great deal of
popular interest and certainly funding dollars, both private and public,
military and non-militarywhat we see is that because Artificial Intelligence was framed in particular ways, it had a sense of bounded

16
17

control in the idea that you could simulate intelligence that had similarities to the interests of control and domination held by the military.
It was an ideology well matched to the military aims through which it
was supported. The consequence was that intelligenceartificial or
notbecame in a sense militarized, even if the algorithms it produced
were not explicitly tied to the military, or even to thinking as such.

I nfor m ation an d I nstitution

BB

Today security is paired with ecology as meta-emergencies to

drive research. But regardless of how research plans for Artificial Intelligence arrived at its discoveries, are they relevant to whether
it works or not? If in some ways the substance of the world is already
computationalbut not exclusively soit was so or was not so before
humans arrived. And for the regimentation of this, bad people do discover good things. That is, military science can be true.

NJ

Sure, bad people can discover good things and good people can
discover bad things. But concerning participatory democracy,
the production of knowledge is the commons we are concerned about:
what kind of knowledge gets supported, for whom, by whom? This really
is the political question. So we can have (and have had) environmental
data collection, and we have a regime where most of the data collection
has been done under these regulatory compliance protocols. And now
we have the capacity to have this collection occur under alternate protocols or through a different institutional framework, operating with
different models of participation.
I think the new collection regime that is so tantalizing consists of citizens actively generating and interpreting environmental data that is
everywhere and available openly, which is very different from, for
instance, the Nuclear Regulatory Commission saying you must collect
data on how many million organisms per week are sacrificed, and then
you must report it. For me or you to take that data and try to understand the impact on an ecosystem becomes very difficult because that
data was gathered for regulatory compliance and not for understanding
complex dynamic adaptive ecology! You would need to collect different
data. So there are a lot of resources and a lot of data for us to use and
interpret in this context, but I question the extent to which it is emblematic of the kind of possibilities the parliament of things suggests.

BB

I presume that there is a there there: that there is a relationship

between the parliament of things, the dingpolitik (Latours
politics of, for, and by things) and this collective of bottom-up nonlinear producers and consumers of information that you call for. It implies a flatter, a less authoritarian, a less anonymous and perhaps even
less designed political space, or perhaps instead another modality and
methodology for its design. It is a further decentralization of a popular
scientific method that itself becomes formalized as a politics.

18
19

Things (and the thing) are temporally-fragile, relational contingencies. Construction Materials of the Exhibition Room, Lara Almarcegui, 2003 . Installation view,
FRAC Bourgogne.

NJ

I think it is worth elaborating on that, because what I see and

in many senses try to instantiate in particular examples is the
capacity to change the structure of participation: who is producing the
data, who is interpreting that data, and who can do something with
that data. So in a participatory democray that means restructuring
participation from the production of scientific or authoritative data
and knowledge to this structured participation.
There are a couple of ways of regarding environmental issues that are
particularly well matched. Environmental issues by definition are in
the shared commons. We all bear the risks and the bodily burdens of pollutants and contaminants and so we are invested as a wide and diverse
citizenry in environmental knowledge production. This is different from
other kinds of scientific knowledge production in which you, I, or
anyone else may not have a direct relationship with the understanding
produced. Our own health may not be directly affected by big science
questions of a different time. General relativity, for instance, is interesting
but does not have a direct, everyday impact on our lives in the same way.
The second thing that I think is characteristic about this whole area of
knowledge that makes it suitable for investigation and interpretation
by diverse, active participants is that ecology is inherently complex.
Environmental systems are by definition multiparametered, unwieldy,
uncontrolled, and contingent. Partially because of this, the science of
urban ecology and urban systems has stayed in these little ecology
departments and has never been a big science. The previous tools of
scientific investigation have not been able to take on these kinds of
questions. In a way its too big and too social and too specific.
The third thing is that we have to do something about them. There is
a popular urgency to re-imagine our relationship to natural systems.
This is not about reading the great books for your intellectual edification.
this is a situation where the urban systems we have created are failing,
and the global climate is changing, and so what are we going to do?
It is critical to have access to the kind of knowledge around complex
adaptive environmental systems and socio-economic systems if there
is going to be a sufficiently effective change. These three aspects make
it suitable and in fact urgent that environmental issues are investigated
by a diverse citizenry, particularly because, not in spite of the fact that
it does not fit into a clear scientific or academic discipline.

20
21

BB

Nor does it fit in a clear political geography. It is not clear at

what scale political action can best be motivated to enact this
kind of change. Lets imagine a situation where the citizen-scientist
is tracing and modeling localized events, thereby coming into a more
reflexive relationship with how those localized events are linked to
non-local events, and is able to make claims for them that are confident and informed, and shared and communitarian. How does it scale?
What and where is the institutionalization necessary for the sufficiently
effective change? Is there an emergent political geography that is as
big as the issue, that is regional, planetary? Or as some would argue, is
such thinking exactly the wrong path to go down, and that this needs
to remain very tactical and liquid, resolutely unglobal? For you when
and where does institutionalization take place? When does the insight
and participation become a new kind of governance? Or economy? Or
supply-chain? Can it scale, or is it a micropolitics only? Can we design
how it scales, or is that impulse also exactly the problem?
I am ambivalent about this, and even about how Ive framed the problem. It is a truism (or clich) on both the political left and right that a
network topology represents the future of political formation, displacing
Enlightenment-era centralized models. But even in assuming this, it is
a parable, not a program. (Sassen, 2008)

 I t s O K , I m a Doctor

NJ

Thats wonderful. No doubt the relationship of design to political reformulation is tied to the question of institutionalization. Designers themselves can produce instances of alternate forms
or structures of participation, snapshots of possibilities, but until they
enmesh into an institutional continuity they remain atomized. In my
case, I have been experimenting with new approaches of institutionalization with the Environmental Health Clinic.
It takes a familiar institutional model of a health clinic and broadens
the idea of health from one that is paradigmatically very closely related
to the medical system and centered with an internal, biological, atomized individual. Health normally is thought of as this individualized
thing, treated by these massive institutional structures of hospitals,
clinics, hmos, etc. on one body at a time. The Environmental Health
Clinic operates in many ways like a health clinic, where people can
make appointments under very familiar regimes. You dont have to
be an environmental activist or a media artist. You bring in particular environmental health concerns and walk out with prescriptions
for design interventions and monitoring protocols that you can do
to understand, interrogate, act on, and improve your local environmental health.
So that is the conceit: that we formulate health not in individual, medicalized, pharmaceutical, and internal terms, but as something external and shared and that we can act on and change. Of course there is
very real public health evidence to show that these external ideas of
health have very real merit. The best example is pediatricians. They
are trained in diseases and nutrition and growth charts, but the top
five things they spend their time doing, in terms of their office hours,
are: 1. asthma; 2. development issues and delays, autism spectrum; 3.
childhood cancers, the occurrence of which has been greatly amplified
in the last fifteen to twenty years; 4. childhood diabetes; 5. other issues
associated with obesity. The environment is heavily implicated in all
of these issues.

BB

So the Environmental Health Clinic has a similar assignment in

certain ways, but instead of looking at the insertion of medicines into subjective bodies, it has to do with changes in adaptive
behavior and in personal microenvironments that would work to eliminate the beginnings of un-health.

22
23

Dr. Natalie Jeremijenko, of the New York University Environmental Health Clinic, in
her lab, examining purposefully the relevant data.

NJ

Yes. Take lead levels or elemental carbon diesel pollution in

the air, for instance. You can pump kids up with asthma medications or you can try to improve the air quality in their schools, parks,
and neighborhoods. These are very different regimes.

BB
NJ

And to do those things is the agency of a clinic?

The useful thing about the clinic and the clinic script is its familiarity and how you can get people to participate. Unlike with
collecting art!

BB

Ha! You are using the clinic to drive participation in this broader
definition of health and healthcare design. Can you expand on
the clinic as a script?

NJ

The reason it is a useful script in terms of political organization

is these one-on-one, indelible meetings. In what other institutional context do you have this kind of direct engagement? It is really
about very local, very personal and particular concerns. It is very much
about your issues.

24
25

It is not Amazonian rain forests, it is not polar bears in the Arctic. It is

about how air quality and water quality is affecting your health. It is about
translating the environmental movement into a very self-interested
one, not a kind of charitable, luxury item, where we have bleeding heart
white liberals who are going to tell you not to cut your trees down. It
is about how does it affect you and your health and what can you do to
improve your local health.
So medicine takes on an objectivity and a sort of translation of authority that is less partisan than politics and therefore more trustworthy
and more personal including assumptions of doctor patient confidentiality, etc.

BB

Im imagining where this leads. You have the clinic at nyu

now, but what we are to imagine is one on every street corner.
All the 7-11s will be replaced with Environmental Health Clinics. They
are everywhere and perhaps there are competing chains of clinics,
competing movements, competing discourses-scripts for these health
design clinics.

Let me just note that you have used the script of a top-down, authoritative medical institution. I assume you would wear a lab coat while
you are in there, youve got a clip board, charts and all the rest of the
accoutrements. You are dragging medical authority, a political/institutional transvestite. Have you unmasked Bruno Latour and found
Judith Butler?!
(laughs)

NJ

Sure! Lets consider this: how do you get something through

the airport security X-ray machine? You cant take toothpaste
through, you cant take robotics through, but you can take medical
equipment through! It has a practical authority.

BB

An objectivity, a disinterestedness. The Red Cross and the Red

Crescent can walk through a political apocalypse of total devastation, but because they are literally waving the flag of medicine they
are totally above, beyond, or to the side of politics or political goals, and
therefore have protected passage.

NJ

Exactly! And that is a very useful thing: Doctors orders.

If you are digging up your road to put in a No-Park project
(for filtering road-borne pollution on your block) you have to be able
to transcend the Department of Transportation and the Community
Board. You have to have some larger claim. And environmental health,
medicine, is what gives you that authority.

One example of a No-Park intervention, providing lung-level clearance of airborne

particulate matter.

26
27

X S G lobali z ation : I nterest, Diagnosis , S cale

NJ

A block association that is implementing a No-Park project

based on the Clinics prescribing it to them can compare their
experiences with other groups doing the same thing. It works differently in different cases, but the institutional context sets up a way for
projects to be compared and contrasted and to become available for
application and adaptation into other contexts. As an academic, my job
is not to do these particular instancesthat is very much the role of
the self-interested partiesbut to try to generalize around and across
instances to make larger truth claims about what works and what
doesnt work, how we can respond, and how we cannot respond. The
academic institutions are useful in this context because the standards
of evidence that exist in these institutions are higher than in political
realms. Ours is a reputation economy.

BB

The clinic is small but officious. It models itself on a system

of information transmission. The clinic is set up not just as a
place to disseminate design information about particular problems, but
also as a channel for academia to gather information. And so your patients are also test subjects . . .

NJ

. . . No, they are impatients, and this is a very important point.

This is where the authoritative model falls down. The impatients are people who are impatient, they have formulated the question, they have come to the clinic, and in this way they are authorized
to act.

BB

The clinic is set up in such a way that people go to receive

personalized prescriptions. Their actions are in some way designed for them, but in fact in the performance of those scripts they
are also generating local sample information that the clinic then reabsorbs, summarizes, and synthesizes into something that might become
knowledge, strategy, or even policy within the larger context of the
academic community.
For you academic institutions become a place where bottom up information is gathered, in what is also a laboratory model as much as
anything else, and this becomes the basis by which this information
is given legitimacy with respect to policy. Yet issues remain, such as
whether clinics can absorb and distribute enough knowledge, and how
it is that this collectivity of actors could concertize this knowledge into

a sufficiently far-reaching re-industrialization of the world that I think

we agree needs to take place. How precisely the knowledge gets turned
into design, in other words, is not necessarily part of the clinic script
per se, or is it?

NJ

It is most definitely part of the script. It is the co-production

and the localization of knowledge, and the situated-ness of
that knowledge. There are six different groups of people working
on the No-Parks projects at the moment, each of them implementing it in really very different ways, funding them in different ways,
invested in different questions. There is similarity between them that
allows comparison, but comparable information is not enough. The
whole Enlightenment concept that knowledge leads to action has
failed drastically!

BB

Or worse, that information leads to action, and still worse is

that a picture of information leads to action or a that picture of
information in a museum leads to action!

NJ

(Laughs) It is through design that we can re-authorize who

can act materially. This is what goes on in the Clinic script.
It is no longer the Department of Transportation who is determining
who digs up the road, it is the block association on Ninth Street, or the
Community Boards from Spring Street. They are changing who can dig
up a road, who owns the so-called public space of the street, who can
determine what the function of it is, and if we put in the No-Park, who
maintains it, who implements it, who invests in it. It is no longer central
government, it is these very different local groups.
This leads to bigger questions regarding the translation into institutionalization, which I am very interested in. Given what weve just said, how
do you take information and translate it into local actions that people can
actually do, feel confident doing, get feedback after doing and continue
to improve and develop as they implement them? That is the nexus that
is not well addressed in the global circulation of Internet knowledge. How
do you translate knowledge into action and then authorize that action?
That is something very different from the Knowledge Society.

BB

Implicit in the conversation we are having around design and

monitoring and the rest of it, is that a diagram of a more appro-

28
29

priate political architecture can be deduced or designed. In the local

aggregations of eco-political knowledge and application, there is a way
for these instances to pluralize, to assemble into networks of different scales, so that they can in fact become more durable forms of both
information and knowledge and are properly empowered to make very
large design gestures and in this way become parliamentary.
In other words, an intensification of governmentality at the local level
does not necessarily mean smallness. There is a wealth presumed for
these network effects, and we are asking the necessary next questions:
how can they create worlds in their image and take on the force of law?

NJ

Yes, because they do relocate the authority, and who has the
authority to act. It is not just writing to your local representative
in the hope that they might do something that improves air quality
around the airport, for instance.

BB

Which is a mediation that is also an abdication. The act of

voting in itself becomes an act of transferring of ones own
sovereignty. You are mailing your capacity to act politically to someone
else. Whereas in the processes of pluralization you describe, the productions of truth become productions of design. It is in these processes
of mediation and translation from one scale to another scale and back
again that the story resides and the action is. What I am trying to get at
is how we might imagine those mediations beyond individual activism,
however self-interested.
The present clich holds that we have, on the one hand, hegemonic
neo-liberal globalization organized around bio-political maintenance of
large and small scale event-systems, and on the other, a more delinked,
yet networked space of heterological, multitudinal agencies working
from the bottom up to bring a new day. Like you, I am dissatisfied with
this framework as an appropriate description of the problem, let alone
a program of action.
Let me ask you this: you said you are not a politician; in what way are
you, through your theory-object engineering, really a political scientist?

NJ

Politics is a very hard word to use. I use this idea of structuring participation, where micropolitics directly engaged in

the material context turns into something that we recognize as social

institutions. This is of course the older question that Latour is responding to with the parliament of things: do things have politics? And we
are now asking this of these ubiquitous computational elements. Do
they have politics, and if so, what are those politics? These are hard
questions to answer and not necessarily useful for practical design
responses. To tell impatients that they are doing political work or
that they are now political activists can be counterproductive. In the
Environmental Health Clinics, we have sidestepped the whole label
of politics by framing the project in terms of health, detaching environmental issues from what people recognize as being big P Politics, in
the same way that technology is politics by another means. I think it is
useful to avoid that association.

BB

That is right. Inherited political-geographic models seem inadequate, and at a point the retrofitting stops, no more layering
new code on top of what is essentially an unscalable architecture.

NJ

Toward this, for me it is critical to appeal to the sense-making

of the everyman. We are trying to translate these technoscientific, industrial and political resource allocation issues to be
self-evident to the everyman, such that they could act as if they were
self-evident.

BB

That is, to an extent, what the information visualization projects

imagine that they do, no? Rendering everything into a visual
common sense.
But let me qualify the political here. What is limiting with the notions of a cosmopolitanization of the global agora is that they presume
eventually that default or ideal political conditions are essentially
consensual, or at the least mutually compatible. But the political image
of consensus covers up or marginalizes the reality of the political as
essentially an antagonistic space, one that is always competitive, one
that is always irreducible, a never ending battle, and even more importantly, one dominated by incommensurate claims, and that is its
purpose, not its problem. It is not just that we will split the middle but
that, quite specifically, we are speaking different languages, and that
there is no middle because we are and should be using different alphabets.
(Mouffe, 2005)

30
31

For me it is this ethos of political irresolvability that must be respected

by any program for the next phase of what global political institutions need to become. But it seems to me that a lot of the political
rhetoric around ecology and the understanding of how environmental
conflicts need to be politicized tend to make use of the same rhetoric of
consensus, whether as a lifeboat ethicsthat we are all in this together
or as a shared space that might as well be rationally communitarian.
Ecology extends older theoretical problems of Totality, especially when
paired with planetary computation.
I am interested in your thoughts on this, because you are talking about
the capacity of people to make truth claims about and through their
participation in this ecology that by necessity, because they are design
prescriptions, involve claims that could be totally incompatible, or
even cancel each other out.

NJ

Well, it certainly is descriptive of what is particular about

environmental monitoring and modeling. The air quality is
shared, the water quality is shared, but how do we describe that quality?
Do we use the dumb epa set standards to organize a consensus about
what we do about air/water quality? Do we use the forcing functions
and co2 measurements as the parameters to model global climate
change? There are problems in all of that. There are many different
ways to think about what counts as air quality. We dont even know
most of the contaminants. Certainly we dont know the health effects
and the interaction effects of the fifty thousand synthetic compounds
that we have released widely into every corner of the world. There are
around two hundred in any given site. It is difficult to imagine being
able to agree on what is air quality, and then to measure it, and then text
it to everybodys phone, assuring them that there is good air quality in
San Francisco today!
What matters is the material context of what counts as a healthy
environment. How do you frame those sorts of questions? The idea
of environmental health is a loose bucket around which you can open
a dialogue but that you dont necessarily need to resolve or produce
consensus. There is no credible checklist to describe (as the epa and
other federal, state, local agencies have tried to do), in a functional
way, what constitutes a healthy environment for everyone, all the
time, in every context.

It is a very difficult realm in which to produce consensus, and that consensus, if ever produced, is very frail. The Feral Robotics projects, for
instance, measures volatile organic compounds on a brownfield that is
also a ballpark and Starlight Park in the Bronx. How you demonstrate
contamination in order to get remediation is tied into how you negotiate
whether public money should be spent on remediation when it could
be spent on a literacy program.
Anything like political action requires persuasive methods. This is
where these technologies of ubiquitous computational devices become
very important actors, because they do have an authority and a persuasiveness. Con Edison engineers that are hired can say there isnt
any contamination in this area, that their data shows this from these
subsoil measurements that were taken fifteen years ago. Who is going
to critique those? The community groups who are concerned about
contamination were not able to critique that data, until, in the case of
the Starlight Park situation, they had produced other evidence that
allowed them to participate in contesting.
The capacity to contest, to be in the position to have an opinion, to
question the evidence, is where these ubiquitous computation devices
can really contribute.

32
33

A re A rtists the Missing E xperts ?

BB

Lets go back to the maximal image of pervasive computation

the 2nd Planetary Computersmearing the planet with an
objective computational film that would construct a public database,
an infinite stream of information about the performance of our shared
socio-natural spaces around which these decisions are being made.
Often pervasive computing is described as a diagnosis model, providing
us with a way to detect what might be thought of as a disease, essentially.
We detect pollution in one way or another and then render it. It is a
way to visualize the invisible evil, drawing a picture of evil, so we can
see where it goes and where it is at and in one way or another prevent
it from happening.

NJ

I am very interested in this missing expert issue that you

brought up earlier. I think it is a really interesting problem.
As you know I have always defended both the amateur and the role of
the artist in complex technical phenomena, in biotech, in distributed
wireless sensing, in ubiquitous computing. The artist stands in as the
everyman, the layman, the non-expert, marking out the capacity of the
civilian to participate in the technological future and its possibilities. I
see a unique role for artists in their very status as non-experts.

BB

Do you think that the scientists you collaborate with have

similar reasons for wanting to bring artists into design collaboration process?

NJ
BB

No! (laughs)

NJ

Let us take a step back and look again at the promise suggested by
ubiquitous computing in relation to the climate crisis and other

As the missing expert, what is the role of the artist for creating
politically legitimate forms of knowledge as minor sciences?
The artist could be understood in terms of a profession, a methodology,
or the economic art-object that would be created, and that any one of
these in and of itself may be understood by itself as the condition for
that participation. Could you talk about how you see the position of the
artist specifically in the creation and governance of what you call the
material public? What do mean by the material public? What do you
see as the position of the artist as politician within that?

environmental concerns, the assumption that these can be solved by

blanketing the world with sensors, and that we would somehow address environmental issues at the largest scale directly and effectively.
I would argue that essentially there is a different kind of politics involved
here, precisely because we are dealing with the environmental commons, with air quality, water quality, and public spaces affected by
these qualities. The thing that has made it so easy to degrade, external
to market conditions, is that it plays out in these commons. It is not
subject, in a sense, to market forces that are geared toward the governance of private property. What characterizes a political engagement
with environmental issues is this idea of acting on or with the stuff
of the world. It is not a discursive engagement, it is not about debating equality or freedoms, etc. For example, the eco-home becomes a
site for political engagement with environmental issues, potentially
privatizing them, whether you have photovoltaics on your roof, or
smart metering, or do your systematic recycling. This is one idea of
the politics of the material public. The interesting role of the artist in
this context, as a kind of politician in the framework of the material
public, is one where the artist stands in as the non-expert, the everyman,
counter to the institutions of expertise (of the scientist, the engineer,
the architect).

BB

So the role of the artist is as a model citizen and a nave scientist

at the same time, who is able to produce public images that
work to construct both the material public and a minor science created
for that material public.
Within some version of a material public that is based upon an infrastructure of ubiquitous computing, the political formations that would
emerge from those relationships would be largely based on the aggregation of these intensely intelligent, hyper-local encounters between
persons, things, and material events, including those in which people
are not necessarily involved. Part of the question, then, for the artist as
missing expert has to do with the status of the image that is created, and
is related directly to the problem of the data cloud or data dashboard,
and the status of the missing expert within those rhetorics as well.
Let me expand on this. To me, ubiquitous computing is understood far
too much as a problem of a dense urban landscape, in the way in which

34
35

the human encounter is always organized around a flneur model,

of a person walking through New York City interacting with parking
meters with his cell phone. It seems to me that with the model of the
material public and the scenario of a kind of second planetary computer,
the real issue has to do not only with non-market interfaces but also
with the interrelationship of non-human actors. (Friere, 2008)
In this sense, the images that the artist has the capacity to produce
has as much to do with giving representational agency to organizations
within a material public matrix that otherwise would not have a parliamentary presentation. I am very interested in the status of the nonhuman in relation to the artist, considered in terms of the emergent
political models we are talking about.

NJ

Thats great. My view of the artist is based on the notion of

the artist as idiosyncratic, as outside of spheres of regulation or authority. Outside of these frameworks, the artist representation becomes the only form of persuasion. In particular this ideal of the
attention that can be paid to environmental phenomena is about this
constancy of, for example, a temperature sensor pinging 24,000 times
a second. And so who is making sense of this system, and what kind of
image-maker are we talking about? And for what? I am very interested
obviously in how the non-human and the non-market intersect.

BB

This also conjoins sensing, sense-making and image-making

into one? The ambient infrastructure itself: sensors, databases, displays, code, etc., that material system itself becomes the
image-maker.

NJ

Right, and in that, the artist is reduced to the role of the illustrator, choosing what four colors to use and how to make the
lines glow just-so in Flash. But there is another even more important
demand of sense-making: who or what is making the sense, who or
what does it make sense to and why? This is what my colleagues and
I and others who are engaging this kind of issue are saying: if we cant
be the ones who make sense of all this, then we as artists, citizens and
scientists are all in real trouble. So again, there is a unique stupidity
about us as artists that can be incredibly useful in this context, that
makes the artist a sort of lowest common denominator in relation to
the institutions of science and technical fields. But as you say, the real

focus is on non-human communication and agency, the actants and

agents, of responsive environmental systems.

BB

As for me, once again, I see the data-cloud model as an image

that employs the rhetoric and authority of the interface to
produce a diagram of some projective potential relationship between
agents within this interfacial system. In this system, a publics ability
to recognize and consume the image, and thereby the projection of a
potential set of governable relationships between the variables within
that interface, takes on an iconic and indexical value in its affective
intensity. But in this iconicity or intensity, there is another kind of projection or proposition being made. It asks how it is that we may sense
the world, or how the sensibility of the world might be distributed (in
Jacques Rancires terms) or organized, instrumentalized, and activated
to become a part of the way the commons understands and narrates
itself. It is not only an image, like a propaganda poster, it is a tool for a
politics that doesnt yet exist. (Rancire, 2007)

NJ

And for me, where, and how, and if the artist intervenes becomes the question, given as weve said that now the system
can make its own images and diagrams. Also important is how these
images do become an interface. That is where this issue of agency
comes in. And I like to speak to how we might understand agency in
this context.
There is a crisis of agency that may or may not be resolved by and/
or attributed to the technologies of ubiquitous computing and sensing.
Certainly we have to acknowledge that in mainstream environmental
activism, agency is conflated with consumer dollars, with the responsibility of your consumption, organic, fair trade, energy star, etc. Within this
rubric, human agency is understood as equivalent to purchasing power.

BB

And in that scenario, the image of the future is for data clouds
to render on the sides of cereal boxes, where, for example, the
transparency of their conditions of assemblage and carbon footprints
become the discourse through which these objects display themselves
to us as interfaces into a vast supply-chain.

NJ

But obviously there are problems with that view of agency. It

becomes tied to wealth, such that those with less purchasing

36
37

power have less agency to affect the public space and act upon environmental issues that are immediately important to them. Income
disparities being what they are, the environmental commons is then
unchanged. The deeper issue is how and by whom the information on the
side of that cereal box is produced, whether it is a marketing agency,
trade organization, or even the environmental organization. This is a
question of the transparency of the institutions behind the production
of this imagery.
It is not necessarily an opportunity for social change if you can get
more information about a product by waving your cell phone near it.
The same marketing firms will produce more information of the same
type, even if it is produced by ubiquitous computing. If the marketing
company is putting the co2 sensor on the chimney, this is very different
from the local community group or environmental justice organization
placing sensors near the outflow of the buildings. The sensors are the
same, but the information is of a different value.

BB

And that is part of the work that the popular idea of spime
accomplishes.

But when the disclosure of transparency of production and consumption goes all the way down the supply and demand chain to the sourcing points, it does not become socially instrumentalized until there is
some process of narration enabling that information to feedback into
the system itself. This is not unlike your How Things Are Made project.
(See http://howstuffismade.org.)
In the spime parable, as with How Things Are Made, there is an opening
up of the biography of assembly of the objects in the world and giving
these biographies an agency and a voice within this process, so that the
possibility of interaction with those biographies extends through the
supply chain to points of origin and back again, such that the socioecological relations that exist between a consumer and a producer or
between a landscape and an objectwhich are already therebecome
part of that objects public representation. The spime model in its
most optimistic sense is a way in which first and foremost the system
describes itself to itself. If access to the raw data of its sensing capacity is
open, then capacity to narrate those chains of connection is also open.
(Sterling, 2004)

NJ

Certainly, but it is important to contrast the kind of agency that

comes from independent individuals or groups placing sensors
and monitoring regimes with the more traditional forms of environmental
agency, which are dependent on data from the epa derived from existing
sensors. If I do a story using my own independent data, it is a different
story, open to a different sort of interpretation.
It is essential that the individual have access to these sensors and be
able to independently deploy them. In New York, for instance, there
was an attempt at legislation that would prevent individuals and community groups from doing environmental monitoring. We think we
have that defeated, but there remains this dangerous idea that people
cant handle this information, that it creates hysteria, and that this
hysteria, which must be managed, is a problem over and above individual agency.
Because in truth there is not a missing expert. My students and I are by
no means experts, but in so far as we have been able to trace through
the production cycle of one contemporary good, there is sufficient information to suggest a viable innovation. That moves the image towards
the interface. This information is being generated in order to change
the conditions of that production.

38
39

F orce Means G uaranteeing R epetition

BB

If I can, let me restate and summarize these ideas on the table.

By the disclosure and display of events that assemble themselves in the production of the material public and the organization
of the chains of connection between those events into diagrams, the
production of the image of those connections in and of itself suggests
a potential reorganization or reconfiguration of those relations themselves. By employing the visual rhetoric of the interface, these images
invite a counter-deployment of the variables that they are mapping.

There is a difference between some of the data smog projects that involve
throwing up balloons and looking at air quality and producing maps
of cities resulting in little more than wow, cool wonderment. They
produce the effect of a missing expert, the implicit presumption that
somewhere along the line, whether inside a mountain in Wyoming, at
the epa, or through an activist on her bicycle, somewhere someone
must be using this interface to actually modulate these things. In this
they further distance people from their own ability to look, hypothesize
and act.
But in their own way, they are projective images of a parliament of the
material public that doesnt yet exist. The question is, as such do they
provoke the possible emergence of this new form of political engagement, or do they provide an alibi for it never happening because they
make it appear that the parliament is already there?
I wonder whether or not that problem exists when we are dealing with
interfaces that dont use the rhetoric of the gui. Take for example the
Hudson Glow Fish project or the One Tree project, which are both
ways in which the events of assemblage themselves are organized and
offered as interfacial information, but dont use a dashboard or the traditional rhetoric of the gui. But I wonder if you see the same problem at
work in this type of project or if they point in a different direction?

NJ

They are different in that they are a critique of existing interface models. The Glow Fish Interface is a 1010 series of buoys
in the Hudson River that sense fish as they swim by and light up and
wiggle. It essentially amounts to a low resolution screen display on the
river. To whom the information is displayed and what reactions can
be taken from that is scripted, but it is scripted differently, and that
difference matters.

The fish eventually learn when the lights go on food is likely to appear,
and people learn that when the lights go on, fish are likely to be near
and so start throwingas they do everywhere elsestale Wonder bread
or bagels in that direction. And there is the opportunity to change that
interaction: the fish food I have developed and deployed at the site
has embedded chelating agents that, when ingested by the fish, will
remove the body burden of heavy metals, helping to take it out of the
ecological system of the river. That loop is enabled by producing an
image of what is present (the lights indicating fish proximity), but the
actions that can be taken and the individual stratum are scripted so
they can be aggregated to achieve significant mediation.
This taps into our persistent desire to interactthe kids who cant resist tapping on the glass at the aquariumbut it scripts this interaction in
a way that produces real, measurable environmental mediation. It closes
the mediation loop and changes who is responsible for that mediation.
It is a very different model from the traditional remediation approach
of doing a massive study of fish populations and the hydrodynamics and then putting out an rfp to the Army Core of Engineers or a
multinational environmental engineering firm who comes along and
dredges the Hudson River and ships off barge loads of toxic sludge
to Pennsylvania or the nearest location that will accept it . . . It is a
different kind of loopand the speed and scale of it matters: it took
thirty years to get an agreement to dredge the Hudson River, through
dedicated lobbying in the traditional model of political agency, a lot
of really good work.
And yet there is a real immediacy to these kinds of issues. I find it
incredibly haunting, to give just one example, that the pcm value of
the river is about the same as the pcm value of the breast milk of the
people in New York City. The challenge and opportunity of ubiquitous
computing is the way in which it can make doing something, making
change possible, through the direct action of people, and in this way
give that capacity to act an immediacy not possible in the traditional
models of political agency. However, there is so much emphasis on the
data fetishism, how much data can we get from everywhere, etc. In contrast, in the Hudson Glow Fish project, we have a very low resolution,
1010 pixel display that I would argue can have a real meaningful effect on the situation.

40
41

BB

Absolutely. An emphasis on high degrees of visual resolution is

always suspicious. I would add that the compunction to produce information visualization diagrams or images of the conditions of
interrelations between events of assemblage is also a desire to interact.
It tries to understand the processes of the world as they come to us, its
constituent parts and how they can be worked with, even and especially
if we cant see them at first.
I want to talk a bit more about governance. You talked about aggregation. There is a desire as you say to aggregate the most information about
which a portraiture could be drawn, in order to produce the highest
resolution image, whereas the real issue of aggregation has less to do
with the resolution of the portrait at any given instance than it has to do
with the aggregation of actions taken in relationship to those images and
the aggregation of those relationships to each other.
In other words, how do every one of these micro-political acts of agency
constitute in themselves, as they themselves aggregate, the possibility to
institutionalize, congregate, or become more systemic in such a way that
they form the basis of a new kind of political and social mediation?

As you said quite clearly, the traditional models of agency around these
issues are dependent upon models of representation that are incredibly
slow in comparison to the capacity and necessity of action. Consider
for example the representation of public will through a vote in a parliamentary system that produces itself in the representation of a policy or
a law, which in turn replicates itself in an action, etc. and etc. down the
line. These chains of representation are too slow, and part of closing
the loop is the disintermediation of these so as to allow for an acceleration of possible reconfigurations. But again, is this, by itself, a model
that is too bottom up, one that doesnt provide a direct way for macroinstitutional organizations to be formed or re-formed?
One way we might frame this question is through the idea of law. What
becomes law in the context of bottom-up micro-political aggregation,
particularly when we consider law as guaranteeing the replication of
certain desirable behaviors? We might consider this in terms of protocol,
how a force of law becomes embedded in inflexible, even stupid systems.
It is actually the inflexibility of such systems that provide the most
social flexibility to bloom. (Galloway, 2004) So how can we consider

the model beyond a nave, volunteerist micro-politics? What forms of

institutionalization or what Latour calls parliaments emerge from
this? How do these closed loops aggregate themselves into new kinds
of enforcements, if not laws?

NJ

That certainly is an open question, but its the right one. What
is underprivileged then is the capacity to automate, as you say
to guarantee repetition. To set up local instances that immediately
invite comparison: if this is the case here, what is the case there?
It seems to me that finding those opportunities for comparison or
aggregation is not easy or obvious. Do we aggregate these things
about nested community structures that we traditionally use to
scale up to a macro-politics? Or do we work through comparison and
have an aggregation of comparison that might lead to a politically
effective coupling?
In fact, I am not sure that you could pre-architect those relationships,
because then you might lose the responsiveness of these systems or
their local integrity. To be able to design at a macro-political level is
where much of the political energy has been directed, and wasted, like
the kid on the street collecting signatures for Greenpeace, etc.

BB

But if the tactical volunteers are satisfied to be always working

at a hyper-local scale, when and at what point are they able to
change the protocols of a system, such that we can go beyond remediation (reactive efforts) and actually begin to restructure the larger
systems, to be sufficiently effective as you said, such that the sludge
doesnt show up in the first place?

NJ

This is the important question, but non-governmental material politics have a durability and a constancy that is surprisingly less than the laws and the traditional political structures. A zoning law, for example, is often more durable than the bricks and mortar
it contains. But nonetheless, in this idea of the material public there
is a durability in ongoing problem-solving that constantly invites or
demands monitoring and sense-making. If we design these ubiquitous
distributed sensing systems so that they are accountable to an ongoing
problem-solving, maybe we can affect the emergence of systemic
reconfiguration. Given that we have a context in which the initial
design infrastructure is set up, in which there is an ongoing process of

42
43

innovation, and sense-making is possible and is privileged, then it can

and will happen.
But if we design these distributed sensing systems in the image of militarized paranoia post-September 11thmonitoring bioterrorism, the
paternalist we are looking after this posture, which is of course the
dominant modethere will be much less chance of innovation and of
drawing on the diverse intelligence of many people to make sense of
very complex interacting systems.
Designers who work in this material realm have this enormous opportunity to demonstrate a new form of knowledge production. The law
that you are looking for is in this knowledge production and its capacity
to reach a point of consensus, a truth point, becoming a persuasive
enough representation to act on again and again.

BB

As you just articulated it, the model would rely on the rationality and good will and communicative rationality of activistactors: presented with reasonable information, reasonable means to
gather more information, share it, and to act reasonably on behalf of it,
reasonable people will do so. Its far from clear then how continentalscale industrial pollution, for example, is engaged systematically.

Would this work as a model for the polis under ubiquitous computing
and sensing more than it would for any other technical regime?
The other cautionary issue for the overemphasis on local activism and
local mutability comes from the way in which the logic of the protocol
as a control system works precisely because protocols are fixed. The
protocols are stupid or immutablethink of tcp/ip, the Apache kernel,
or the grid system of New York City. Or if they are mutable, it is
only by incremental variation within them. Again, it is precisely the
rigidity or stupidity of it that allows for the free flow of information
within them.
We want the aggregation of local action to become law, in the sense
that there is a concretization of these bottom-up closed circuits, such
that they are able to enforce themselves and that they become force,
that they become the force of law, if not formally codified. For local
action within the 2nd Planetary Computer scenario to come to operate

as larger-scale protocols, they have to have a similar kind of capacity

to become rigid and therefore allow flexibility to work in relation to
them. If they remain too liquid, they are less able to in fact close
circuits in any kind of pluralistic way, they remain singular events, and
they dont aggregate and become plural. Do you disagree?

NJ

Actually, I couldnt agree more.

44
45

Meta I m ages , Meta I nterfaces

BB

From Durkheims early work on social representation to

Rancires on the political aesthetic in art, the affective intensity of the image of the collective is not just a narrative of a political body
but is in fact constitutive of that body. Whereas Durkheim saw the collective representation as a central, singular form that represented a contiguous body called the society, our contemporary condition has been
characterized by many as a post-social condition, where the borders of
contiguous social forms dissipate, where flows move in and out of geographies, where territories are occupied by multiple collectives at once,
and where the procedures, networks, and assemblages of objects and
things are vastly distanced from our own capacities to perceive them.
It is precisely for this reasonthe dissolution, dematerialization or deterritorialization of the experience of social connectionthat images or
diagrams of dynamic assembly such as the information visualization
projects we began with become politically important because they are
also affectively resonant. They are political images, and forms of postsocial collective representation. They provide diagrams of possible
connections between things and people and forms and information,
which because of our post-social condition are otherwise unimaginable
or unsurveyable, and therefore unaccountable.

NJ

I agree very much, but the map is not the territory! I would
hope that they can eventually provide a way to explore and
intervene. The post-social also implies the activity of non-human
actors. Post-social implies an evacuated condition, doesnt it?

BB

Lets examine this a bit further. Any kind of participation or

action, even at the most local level, in some way reflexively
imagines and images itself to be operating within a particular geography
and landscape. In that this landscape may be what is diagrammed and
presented to the actor in the form of the interface, the role of the interface is partially to specify and constrain the terms of that geography and
the terms of that landscape into a particular, actionable set of discursive
operators. It enframes the horizon of action.
So in what ways does this interface-image, and the production of this
image, give entry into that geography? Youve suggested that there is
something specific about the artist in her role as image-maker relative
to this question. It seems to me that in the context of our discussion, this

ties directly to the problem of the image in relation to collective representation in a specific way, which we might call the law. Because of course
the law can be thought of as another form of collective representation,
a form by which the sovereign violence of the population is codified into
a fixed protocol that ordinates and organizes behavior and enforces that
will in some way that is both autonomic and representative as well.
So as we talk about the aggregation of these minor sciences and these
micrological interventions, where there are shorter and shorter circuits
between an information referent (within an interface, for example) and
an entry point into acting on that information in an ecology, the cycle
of monitoring, imaging, interfacing, and acting is compressed. It turns
active collective representations into direct collective techniques. These
visualizations start as diagrams for isolated users, but their aggregation
leads toward a plurality. They want to be constitutive of some new
political space, but Im afraid political theory is looking in the wrong
places for emergent institutional media. I hope this can change quickly.

NJ

I hope so too. When you say interface, I say structuring participation. But let me underscore a previous point. Like you said,
many data visualization projects obscure how the data was produced,
by whom, and in so doing they produce an aesthetic of disengagement
and gee whiz. So with respect to distributed sensors in relation to the
issue of politics, technology is always and already exclusive. Most people
dont feel the license to interpret, to rewire, to hack, and so the capacity
to recode the code is vastly limited. People do not yet feel the license to
do so. That is where the political theory needs to focus.

BB

And again, how is it that the micropolitical logic of individual

license does not become one of atomization, of one-off instances,
but actually can become a plurality, and in this plurality can become
replicable and in this replication can become institutionalized and that
institutionalization can have the force of law? To simply refer to the
magic of networks again and again is remedial.

NJ

Right. I dont know how, but I know one condition is giving

people the license to interpret. This becomes the role of the
artist, giving people the capacity to interpret and a license to have an
opinion on a complex technical issue. This is a hard challenge.
In so much as the image provides a license to do that, the permission

46
47

to see the underlying issues, the legibility, the aggregation is possible.

The idea is that the legibility of the data, in terms of its source and
the rationale of its production, and who is responsible to interpret this
data, are the critical elements. It is a precondition of the kind of repetition that has the force of law. It is necessary but not sufficient, but I
think experiments in institutionalization will come out of this.

BB

It is necessary but not sufficient that the conditions of the production of the data are visible and transparent, and therefore
addressable as a condition of the datas truth claims, but what is missing
is the idea that what is produced is a more direct access or circuit to
those conditions of production itself! Such images not only display
what is going on, but the display becomes a means to change what is
happening. So the transparency itself is not sufficient, transparency that
allows for modification is required, but because transparency is always
narrated, always staged to some extent, there is no such thing as a nonspectacularized transparency. Is there?

NJ

No, which is why transparency is an unfortunate term. It

suggests a spontaneity that is unreal. This notion that you are
looking through glass and someone is going to draw back the curtains
and provide you with an immediate transparency.

BB

What supply-chain architects call the glass pipeline. How

would you differentiate that from our need to get into the data,
to see the data and see who is making it and how it is made?

NJ

I think it is just as important to understand the network structure of where the data is coming from, because that network
is what structures any participation in the first place. Structures of
participation is the term I use to understand the network structure
of accountability, the network structure of participation, the network
structure of sense-making (as opposed to sensor networks).
Certain structures of participation project an ideal we call transparency and other structures of participation do not. However, we are living
under an increasing demand for structures of participation that are in
the image of militarized information production, in which we have an authoritative interpreter somewhere, an expert. Your notion of the missing
expert can play out in a non-visual way.

For example, in my Anti-Terror Line project, a distributed data collection system that exploits the fact that many people carry a recording
capacity in their cell phones, which allows people to upload their recording or reports of civil-liberties violations in the name of anti-terrorism. I have this open database that every recording uploads anonymously to a website with a time-stamp which you can annotate or not,
keeping it open to interpretation. If you contrast this project to what
I call the terror linesnysafe or the distributed collection systems
which employ the people as insidious sensors, if you see something,
say somethingyou see a very different structure of participation at
work. All of their data (in the case of New York) goes to the nypd, the
cia, etc. And so the sense-making here is relegated to the 1,944 people
who saw something and said something last year, and some supposed
experts somewhere who filter it.

BB

Your version is an open data capturing system, where the viability of access to the commons of relevant information around
this topic is itself available.

NJ
BB

In which anyone can contribute and anyone can interpret.

And from which an authority could be drawn but has not.

Clearly, each system is its own analogue for the structure of
public participation that each imagines for the public city at large.
Though I wonder if that is too pat. Do you know Jaron Laniers essay
Digital Maoism, his contrarian critique of user-generated culture
and the uncritical presumption that the direct judgment of the masses
will a priori produce better results, better systems? (Lanier, 2006)

NJ

But if not everyone then who does make sense of it? And again,
this is the role of the artist. What different sense can be made
under these different structures of participation and does it matter?
Can one make a different sense of the See Something, Say Something
responses than the cia and the 1,944 people who said something and
called in to their line? Is there a different sense to be made and does it
matter? My intuition is that there is, that there is a better sense to be
made, one that is more robust, one that can be challenged. Not just
better sensors but better sense-making.

48
49

And yet I am struggling with what that means, particularly given the
fact that these open systems, as much as they have the potential to
provide new vehicles of sense-making, can also diffuse the responsibility
and accountability for making sense into nothingness.
To me this is also a problem with many of the information visualization
projects with which we began; they throw up a lot of data and suggest
that the sense is self-evident! Its not!

BB

Yes. In the best cases, in the data-cloud diagram-interface, there

is a sense-making that takes this challenge head on. Sometimes
this is through a strategy of advocacy, like Laura Kurgens maps of
dollar expenditures on prison incarceration rendered Manhattan block
by Manhattan block. There is a way to understand these projects purely
as contemporary images, like political icons, posters, icons or paintings.
(See http://www.arch.columbia.edu/index.php?pageData=46955)
But to repeat, when such projects deploy the visual rhetorics of the
interface, there is for them the latent aspiration that these are not only
images but also technologies through which the condition that is being
represented can also be entered into and reconfigured or remade in some
way. Here and now through these image-instruments, and if not now, soon.
You talk about the militarization of such information. In even the most
apparently progressive of such projects, through the command, control
and communication aesthetic, the biomass of cities is cut, cleaved, and
color coded as in a battlefield operation. This is the bargain to be made
in understanding these conditions. But for the bargain to work, these
images have to function with an instrumentality that is something more
than just iconicity. Its aspiration to become a parliamentary medium
and not just evidence, a declaration of scandal, should be supported.

NJ

This is exactly the crisis of representation that we are facing

with these ubiquitous distributed sensing systems.

T H E Death of the U ser

NJ

How we structure non-human agency in this is an interesting

opportunity. I think it is important to frame it in terms of what
you yourself call the death of the user, how non-humans are at the
instigation point of distributed interactions where the otherwise sacrosanct user used to be.
If we can pay a certain attention to and diversify who is represented,
then with ubiquitous sensing systems we can begin to account for nonhuman agents, whether living or non-living. There is an opportunity
for a more diverse structure of representation. There is a real call to interrogate this technology to enable this kind of representation. Instead
of inherited structures, we can take non-human agents, as we recognize them in agricultural contexts or in urban, suburban contexts, etc.
As some have done, using infrared cameras about the house, we can begin
to account for the non-human actors who coinhabit our spaces with
us. We can reconsider structures of ownership, including private
property and how it extends to these non-human agents in the environmental commons. Are there opportunities in rfid and similar
tagging and sensing systems where the branding and closure methodologies used to manage our relationship to non-human actors can be
reinvented? Absolutely.
For example, can my Feral Dog project, deployed throughout the suburban Northeast, become a model for new relationships? Would or
could ubiquitous distributed sensors enable us to imagine what biodiversity might be here regionally and locally? This whole idea that
biodiversity is some kind of global count, that there is a figure that is
measurable and actable, that there is a worldwide count that ticks down
with every extinct species, is really inadequate. It must be possible to
understand biodiversity as a condition of the backyard, represented as
such, and this might be a better way to understand, narrate and engage
these issues.

BB

This is exemplary of the death of the user, because it puts at

stake not how a human might react ethically or unethically
towards a situation from her central position in local or smaller systems,
but the way in which the ubiquitous, pervasive systems comprise an
ecology where parts of the natural system relate and co-govern directly
with other parts at a distance without requiring the agency of the human

50
51

to intervene or mediate. The techno-ecology works without users. We

are designing our own abdication!

NJ

Yes, and through these systems, we can also begin to recognize

a non-human-centric world. But this is also good for us. I started the conversation by saying that through acting with our own selfinterest, and re-framing environmental issues like global warming as
local environmental health, we can begin to see a non-human point
of view and incorporate that into the ways in which we act upon our
surroundings, and providing better solutions.

BB

Elsewhere you have used the phrase lines of desire to name

the way in which a planned space is used by a public as if it
was an open smooth space, and in doing so, inscribes new striations
onto that landscape. Ive used similar language in describing pervasive computing as a landscape effect: how it striates the smooth, and
smoothes the striated. Can we end with you talking a bit about what you
see as the implications of that sort of back and forth process in terms of
our pervasive computing scenarios weve touched on? It is tempting to
imagine pervasive computing as a deterritorializing process that opens
up the world to new liquidities, more than a new medium of inscription,
naming and fixing, but clearly it is both. (Deleuze and Guattari, 1987)

NJ

Dirt paths worn across a lawn in front of a library is the classic

example. My image of lines of desire are those tracks that
people form across grass that become over time a very visible representation in themselves, and yet also a reference to the existing paths. They
are both in defiance of but in reference to the existing pathways. It is
structured as both an open system, and one that produces emergent
forms of consensus through aggregation, in its reference back to other
adjacent forms of governance and codification, such as the official
paved but less useful walkway.

BB

They are the perfect example of a persistent inscription that

guarantees the repetition of similar action and thereby takes
on the essential force of law.

NJ

Exactly.

REFERENCES

De Landa, Manuel. 2006. A New Philosophy of Society: Assemblage Theory and

Social Complexity. London: Continuum.
Deleuze, Gille and Felix Guattari. 1987. A Thousand Plateaus: Capitalism and Schizophrenia. Trans. Brian Massumi. Minneapolis: University of Minnesota Press.
Friere, Juan. 2007. From the Analogue Commons to the New Hybrid Public
Spaces. http://medialab-prado.es/article/del_procomun_analogico_a_los_nuevos_
espacios_publicos_hibridos (November 22).
Gane, Nicholas and Stephen Sale. 2007. Media Ontology: Interview with Friedrich
Kittler and Mark Hansen. Theory, Culture & Society. Vol. 24, No. 7-8: 323-29.
Galloway, Alexander. 2004. Protocol: How Control Exists after Decentralization.
Cambridge: mit Press.
Lanier, Jaron. Digital Maoism: the Hazards of the New Online Collectivism.
Edge.org (posted May 30, 2006).
Latour, Bruno. 2005. Reassembling the Social: An Introduction to Actor-Network
Theory. Oxford: Oxford University Press.
Moufee, Chantal. 2005. On the Political. New York: Routledge.

52
53

Rancire, Jacques. 2007. The Future of the Image. London: Verso.

Sassen, Saskia. 2008. Territory, Authority, Rights: From Medieval to Global Assemblages. Princeton: Princeton University Press.
Sterling, Bruce. 2004. When Blobjects Rule the Earth. Keynote address to siggraph,
Los Angeles, ca.
Varnelis, Kazys and Leah Meisterlin. 2008. The Invisible City: Design in the Age
of Intelligent Maps. http://www.adobe.com/designcenter/thinktank/tt_varnelis.
html (posted July 15).

S I T U AT E D T E C H N O L O G I E S PA M P H L E T S S E R I E S

also available
Situated Technologies Pamphlets 1:
Urban Computing and Its Discontents
Adam Greenfield and Mark Shepard
The first volume in the Situated Technologies Pamphlets Series, Urban
Computing and Its Discontents is framed as a discussion by the authors
to provide an overview of the key issues, historical precedents, and contemporary approaches surrounding designing situated technologies
and inhabiting cities populated by them.
Situated Technologies Pamphlets 2:
Urban Versioning System 1.0
Matthew Fuller and Usman Haque
What lessons can architecture learn from software development, and
more specifically, from the Free, Libre, and Open Source Software (floss)
movement? Written in the form of a quasi-license, Urban Versioning
System 1.0 posits seven constraints that, if followed, will contribute to
an open source urbanism that radically challenges the conventional
ways in which cities are constructed.
54
55

upcoming
Situated Technologies Pamphlets 4:
Responsive Architecture/ Performative Environments
Philip Beesley and Omar Khan
This pamphlet will examine emerging paradigms for interactive and
responsive architecture. It will frame historical and contemporary arguments for computationally augmented environments, examining how
situated technologies using embedded and mobile devices are affecting
the spatial, social and technical performance of architecture.

ABOUT THE ARCHITECTURAL LEAGUE

The Architectural League of New York is an independent forum for

creative and intellectual work in architecture, urbanism and related
disciplines. Through its lectures, exhibitions, publications and digital
programming, the League fosters discussion and debate of the most
stimulating work and important issues in contemporary architecture
and design.
The Architectural League is supported by public funds from the
National Endowment for the Arts; the New York State Council on the
Arts, a State Agency; and the New York City Department of Cultural
Affairs. Additional support is provided by private contributions from
foundations, corporations, individuals, and by League members. For
information about becoming a member, visit the Leagues web site at
www.archleague.org.

The Architectural League of New York

457 Madison Avenue
New York, ny 10022
212 753 1722
www.archleague.org
info@archleague.org


,
 
/
/
/1,


1 9

,
 
/
/
/1,


1 9
x9
,-

President
Calvin Tsao
Vice Presidents
Annabelle Selldorf
Leo Villareal
Mitch Epstein
Michael Bierut
Ken Smith
Leslie Robertson
Michael Sorkin
Tucker Viemeister
Vicki Goldberg
Secretary
Karen Stein
Treasurer
Gregg Pasquarelli

Directors
Amale Andraos
Vishaan Chakrabarti
Walter Chatham
Arthur Cohen
Lise Anne Couture
Roger Duffy
Leslie Gill
Maxine Griffith
Frances Halsband
Hugh Hardy
Steven Holl
Elise Jaffe
Wendy Evans Joseph
Craig Konyk
Paul Lewis
Frank Lupo
Thom Mayne
Richard Meier
Nat Oppenheimer
Kate Orff
Mahadev Raman
Mark Robbins
Susan Rodriguez
Aby Rosen
Robert A.M. Stern
David Thurm
Billie Tsien

Life Trustees
Christo and
Jeanne-Claude
Ulrich Franzen
Barbara Jakobson
Suzanne Stephens
Massimo Vignelli
Executive Director
Rosalie Genevro

56
57