Reading diary consisted of 10
What Sort of Ethics Does Technology Require?
Author(syf * H U D O G ' R S S H O t
Source: The Journal of Ethics, Vol. 5, No. 2 (2001yf S S 5
Published by: Springer
Stable URL: http://www.jstor.org/stable/25115687
Accessed: 10/12/2009 14:23
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=springer.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Springer is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Ethics.
http://www.jstor.org GERALD DOPPELT
WHAT SORT OF ETHICS DOES TECHNOLOGY REQUIRE?*
(Received 21 September 2000; accepted 15 May 2001)
ABSTRACT. This essay critically examines the non-essentialist and anti-deterministic
philosophy of technology developed in the work of Andrew Feenberg. As I interpret the
work, Feenberg achieves an important "demystification" of technology. His analysis peels
away the facade of ironclad efficiency, rationality, and necessity that permeates our exper ience of technology. Through theoretical argument and rich examples, he illuminated the
contingent interests, values, meanings, and voices that are built into specific technologies, often by experts. He shows how technology is transformed by lay actors who challenge its
design on behalf of a wider agenda of interests, values, meanings and voices. My critique focuses on Feenberg's attempt to argue from his demystification of technology to the
full democratization of all technical design and decision-making. I argue that Feenberg's framework lacks the ethical resources required both to (1) justify the democratization of
technical decisions, and more basically, (2) to determine when lay challenges to technology do and when they don't, advance democratic ideals, and why. I trace these problems to
ethical inadequacies in his notions of interests, democratization, and an alternative modern
ity. A sub-theme of my argument is that our society's Lockean morality of property rights and market freedoms poses fundamental ethical objections to his philosophy of technology with which it is ill-equipped to deal.
KEY WORDS: alternative modernity, democratization, ethical argument, Lockean moral
ity, modernity, non-essentialist, philosophy of technology, technology
1. DEMYSTIFYING TECHNOLOGY
In a series of books written over the last decade, Andrew Feenberg takes
up the important task of developing a non-essentialist philosophy of tech
nology, aimed at a more democratic politics of technical decision-making
* An earlier version of this paper was read at the Pacific Division, American Philosoph ical Association meetings (April 3-5, 2000, Albuquerque, New Mexico) in an "Author Meets Critics" session on Andrew Feenberg's Questioning Technology. I am happy to
acknowledge my debt to all of Feenberg's work and his valuable responses to my work on this occasion and many others. In writing this paper, I have also gained illumination
from Robert Pippin, Idealism as Modernism (Cambridge: Cambridge University Press,
1997), pp. 185-206. Many thanks to Mina Pang for her excellent assistance in editing and
preparing this paper.
t4 The Journal of Ethics 5: 155-175,2001.
W^ ? 2001 Kluwer Academic Publishers. Printed in the Netherlands. 156 GERALD DOPPELT
and a more rational design of our built environment.1 To this end, Feenberg
challenges the dominant modes of experiencing and understanding tech
nology. He argues that the dominant essentialist model(s) of technology
imprisons us in a world made by experts who use claims of expertise to
exclude the voices and vital human interests of those lay groups most
affected by it.2
On the dominant essentialist image, technology and the built environ
ment are perceived as determined, or dictated, by necessary imperatives
of efficiency and special bodies of expert professionals who enjoy a
monopoly over knowledge of these imperatives.3 The development of tech
nology is seen to obey an autonomous and value-neutral logic in which
science-based, technical elites (engineers, city-planners, physicians, archi
tects, etc.) realize ever more effective and reliable means to attain the
necessary, incontrovertible goals of modern society. As such, existing tech
nology at any particular moment in time appears to have a self-evident
rationality and necessity which repels the very possibility of authentic
ethical choice and political debate. Of course such debate may arise over
the proper use of technology - including questions of access and distri
bution. But the technology itself is seen as essentially outside all such
political perturbations, marking the necessary framework of all rational
action.4
Feenberg's work mounts a theoretically powerful and empirically well
documented critique of this dominant understanding of technology. On
his analysis, it masks the particularity, historicity, contingency, interest
ladenness, and politics of every specific technology that we confront in
our built environment: buildings, hospitals, clinics, highways, cities, clin
ical trials, machines and devices of all sorts, factories, offices, etc. Each
such technology embodies a design, and underlying that, a technical code
embodying established experts' determination of what is and is not a
relevant factor in designing this or that sort of thing. In turn, Feenberg
shows that every such design and technical code embodies particular
peoples' decisions/power over which among many possible considera
1 In this essay, I will mainly focus upon theoretical formulations in Andrew Feenberg,
Questioning Technology (London and New York: Routledge, 1999), which I hereafter
abbreviate as "QT" and the historical cases in Andrew Feenberg, Alternative Modernity: The Technical Turn in Philosophy and Social Theory (Los Angeles: University of Cali
fornia Press, 1995), which hereafter I abbreviate as "AMV Also see Andrew Feenberg, Critical Theory of Technology (New York: Oxford University Press, 1991). All three works
make important contributions to Feenberg's philosophy of technology. 2 Feenberg, QT, pp. vii-xvi. 3 Feenberg, QT, pp. vii-x, 14-17, 201-216. 4 Feenberg, QT,pp. 16-17. WHAT SORT OF ETHICS DOES TECHNOLOGY REQUIRE? 157
tions, interests, values, costs, functions and voices are to be included and
which excluded in that technology.5
Feenberg thus teaches us to ask certain very specific questions about
any given technology and establishes the social reality of these questions
by illuminating contemporary cases in which lay groups have asked these
questions and transformed technology in the process: "Who determined
how this thing would be made, with whose or what purposes in view
and out of view, at whose expense, in the context of what relations of
power, and through what institutional or social process?" When various
lay actors or users of technology are motivated to raise such questions, the paralyzing experience of an implacable technology or environment is
dissolved and human agency is restored. By this route, Feenberg is able
to explain and justify several real-life cases: how AIDS activists were able
to transform clinical trials and experimental therapies from researchers'
pure devices of medical knowledge to technologies of care and personal
autonomy for people with incurable, fatal illness;6 how movements of
people with disabilities were able to transform the design of streets and
buildings to incorporate ramps and thus their access and agency in relation
to public life;7 how millions of users of the French Teletel system were
able to transform it from an information to a communication technology.8 On Feenberg's non-essentialist view, technology emerges as the
embodiment of a social process in which empowered groups of experts
choose to express certain sets of specific interests and standards in specific
technologies, which in turn are re-experienced, challenged, and redefined
by their users. The users have to live in the world made by others and bring their own meanings and interests to it which can be different than those of
its designers. Feenberg's work thus develops a new conception of tech
nology as the site of contingent political contestations between different
groups of actors over the interests, purposes, and meanings which will be
invested and encoded in the built environment. His work is a powerful
contemporary exemplar of critical theory for it succeeds not simply in
providing a more accurate scholarly representation of technology; but also
one which speaks to the consciousness of lay actors and can enable the
public to challenge and transform the environment designed by others.
He effectively achieves a demystification of technology, which parallels the demystification of authoritarian state power achieved by the clas
5 Feenberg, QT, pp. 87-89, 142-143. 6 Feenberg, AM, pp. 96-123 and Feenberg, QT, pp. 127-128, 141-142. 7 Feenberg, gr, p. 141. 8 Feenberg, AM, pp. 144-169, Feenberg, QT, pp. 126-128. 158 GERALD DOPPELT
sical liberal philosophers, and the demystification of capitalist political
economy/market relations accomplished by Karl Marx and his heirs.
2. DEMOCRATIZING TECHNOLOGY AND SOCIETY
Feenberg's most recent work seeks to go well beyond the demystification
of technology. He raises the stakes considerably for his non-essentialist
philosophy by arguing that it paves the way for a democratization of
technology, and indeed, a radical democratization of society itself.9 Feen
berg looks to his account of the politics of technology as the basis for
unifying a whole disparate array of new social movements of reform. He
sees in his philosophy of technology the basis for an overarching ideal of
radical democracy and a Utopian vision of an "alternative" or "redeemed"
modernity.10 In the rest of this essay, I argue that Feenberg's non-essentialist philos
ophy of technology lacks the ethical resources required by these grander
aims and in any case misconstrues the aims themselves. My argument
revolves around three points of criticism. First, I maintain that Feenberg
lacks a clear and plausible standard of what counts as the democratiza
tion of technology. Secondly, I argue that he needs a more substantive
conception of democratic ideals - especially of democratic equality and
rights. Without that, I argue, he cannot distinguish between challenges to,
or changes in, technology which strengthen democratic ideals, from those
that do not, or have a minimal democratic impact. Thirdly, I claim that
there are powerful normative obstacles to democratizing technology which
cannot be comprehended or criticized in Feenberg's theory. In particular,
non-democratic technology, however it is interpreted, rests on our society's
powerful Lockean moral code of private property, and not simply on the
technocratic ideology of essentialism and value-neutral efficiency.
In sum, I argue that there are large theoretical, ethical, and political
gaps between Feenberg's achievement - the demystification of technology - and the broader vision of democratization and alternative modernity
it is supposed to serve. My point is not that Feenberg's non-essentialist
philosophy is no aid to the broader vision, but rather, that the broader
vision requires, in addition, a different kind of analysis and ethical critique
of modern liberal democracy than what is now central in Feenberg's
approach.
9 Feenberg, QT, pp. vii-xvi, 104-109, 131-147. 10 Feenberg, QT, pp. 222-225. WHAT SORT OF ETHICS DOES TECHNOLOGY REQUIRE? 159
3. "Participant Interests" and Democratizing Technology
At rock bottom, Feenberg's non-essentialist conception of technology is
supposed to show all lay actors who are impacted by technology not just
that they can change it, but that they do or may have good reasons for doing so. This is further accentuated by Feenberg's project of democratizing
technology. After all, there is little point in demonstrating that people can
criticize or change something, democratically or otherwise, if they have no
good reason to do so.
What provides the "good reasons" in Feenberg's model? The answer
is given by his notion of "participant interests."11 Lay actors who have
to use or live with the environment built by others either have or may
develop various "participant interests," some of which are not, but could
be embodied in its design and structure. Intuitively, the notion is clear
enough. "Participant interests" refers to the ways that the personal welfare
of various people - participants, users, third parties, etc., is commonly
impacted, for better or worse, by a technology, making them into a group
bound together by such "participant interests," affirmed or frustrated by that technology, as the case may be.
This notion must bear considerable weight in Feenberg's overall
argument against essentialism. After all, the essentialist who equates
technology with efficiency and rationality sees that its design embodies
persons' interests. The essentialist perceives these interests as typically
uncontroversial, inescapable, and universal in modern life - shared by
users, designers, the public, alike. Owners, designers, workers, and buyers all share an interest in making a car that will sell, run, etc. It is the
collective emergence of marginalized "participant interests" - particular users' interests and voices excluded from the design process and tech
nology - which, in practice, reveals the one-sidedness, contingency, and
politics of technology. In seeing what a technology excludes, what it is not
but could be, we gain a clearer and truer grasp of what it is - the bias of
technology beneath the guise of efficiency and rationality.
Feenberg's notion of participant interests must also bear considerable
weight in his argument for democratization. For, it is certain participant interests which must provide (1) the motivation, (2) the justification or
reasons, and possibly (3) part of the criterion, for the democratization
of technology. The establishment of marginalized participant interests, or
ones once marginalized but now included in the technical code, shows
that technology could be other than what it is. But this falls well short of
defending a conception of what technology ought to or should be - which
11 Feenberg, QT, pp. 140-142; Feenberg, AM, pp. 104-109. 160 GERALD DOPPELT
participant interests should be accommodated within a democratized tech
nology, or alternative modernity. Which technologies or aspects of our
built environment ought to be democratically transformed and in accord
ance with what standard of participant interests, or human well-being? This is the key ethical problem that requires exploration by a democratic
critique of technology.
Feenberg may reply that this is not his problem or project. Having shown that technology can be changed, it is up to the users, the public, to assert their marginalized participant interests and to determine how the
built environment ought to be changed. Isn't that the thrust of democratiza
tion - allowing the affected groups to decide for themselves? Once
technology is demystified, a desirable democratization of technical design
inevitably follows, so the argument might go. I see two large problems for
this strategy, which in any case doesn't seem to be Feenberg's. I will refer
to these two problems as the "Which Interests" problem and the "Private
Property" problem.
4. Which Interests?
The first problem is this: it is clear that not every participant interest, or
challenge to technology is legitimate, morally justified, or a victory for
democratization. There are reactionary challenges to technology - ones
which have led or would lead to a less rational, equitable, or democratic
technology.12 Moreover, there are many ambiguous or ambivalent cases
to which a critical theory of technology should speak. In the fractured
social world of modern life, many challenges or revisions to technology
that speak to the interests of one group of users/participants/affected third
parties may leave other actual or potential users worse off. Without an
ethical standard, how can we determine which trade-offs, whose interests,
what challenges contribute to a more, or less, democratic rationalization of
technology? Finally, there are cases of successful challenges to technology
which are neither reactionary nor ambiguous but don't really democratize
the technical code. To my mind, Feenberg's example of Teletel falls into
this category.13 In general terms, it is a case where large numbers of
customers put a technology to uses alien to its designers' will and design. These new lay adaptations are then exploited by producers for a more
marketable and lucrative technology with these new uses now built into
12 Feenberg acknowledges such cases and the problems they raise. See Feenberg, QT,
pp. 92-94. 13 Feenberg, AM, pp. 144-169; Feenberg, QT, pp. 125-128. WHAT SORT OF ETHICS DOES TECHNOLOGY REQUIRE? 161
the design. While this may be desirable, such changes of technology in
response to consumers' initiatives or preferences follow the logic of market
rationalization, not democratization.
Feenberg presents some intuitively quite attractive contemporary
examples of successful lay challenges to established technology - the
struggle of disabled people for access ramps and that of AIDS activists for
a different technical code governing the design of clinical trials. I examine
these cases below in order to determine whether and on what basis, each
can be characterized as democratizing technology. But the cases do not
provide an explication and defense of a normative standard for judging
which interests and challenges might produce a more rational, demo
cratic, or morally defensible technology, and which do not. This is what
Feenberg's argument for democratizing technology and an "alternative
modernity" requires but lacks.
Presumably, Feenberg does not want to embrace any and all partici
pant interests or challenges which happen to surface or succeed. For one
thing, he implicitly acknowledges that lay actors may have participant interests which they fail to interpret and articulate as claims that can gain wider social recognition and legitimacy.14 For another, he implies that
even when they do get articulated as claims or demands, not all such
claims deserve recognition and legitimacy.15 On the other hand, Feenberg cannot justifiably pick and choose the cases he finds attractive and ignore
others, without some systematic ethical justification. The development of a
standard of justification is not external to Feenberg's project - a gratuitous
appendage he can take or leave. Without it, there is little reason to accept
his claim that the emergence of new participant interests and lay chal
lenges in techno-systems informs a process of democratizing technology and creating a more rational "modernity." Rather, all his argument would
establish, against essentialism, is that technology is not destiny; because
local shifts in relations of power and interest alters technology, in some
cases for the good of particular groups.
But my critique could be pre-empted by Feenberg's conception of
democratization and his analysis of cases of it - either of which might contain the normative standard(s) which, I have argued, he requires. Before
turning to examine these contributions, I turn to the second problem facing his critical, non-essentialist theory of technology - the "Private Property"
problem.
14 Feenberg, QT, pp. 140-142. 15 Feenberg, gr. 162 GERALD DOPPELT
5. The "Private Property" Problem
In many contexts, users' marginalized participant interests fail to emerge or gain social recognition as legitimate claims for technical change because
the established technical code is taken to embody the will, property
rights, and legitimate interests of the owners of the technology. Feenberg's
approach is certainly right in stressing that the problem of technocracy
(the rule of experts) is not limited to capitalism or even market-driven
practices within modern society. Nevertheless, in the modern world, a
great deal of technology is private property. Its designers or experts act
in the name of the owners, and their rights as owners, to determine the
technical code in accordance with their economic interests. In my view
this complicates and destabilizes Feenberg's argument from the critique of essentialism and technocratic ideology to democratization. How so? In
these contexts users' participant interests and challenges quickly confront
the claims of designers and experts that they alone have the right to decide,
to determine technology. Feenberg's theory pictures these conventional
prerogatives of expert authority as always an expression of technocratic
ideology - the very essentialist image of efficiency he demystifies. But
in the common case where technology is private property, the rights and
authority of the designers/experts really rests on the fact that they are
employees, representatives of capital. The rights of the designers to exer
cise authority rest not just on their expertise and the logic of efficiency,
but on the rights of private property, and the Lockean moral code of
ownership and free-market exchange. To this extent, Feeriberg's critique
of essentialist technocratic logic is insufficient to explain, motivate, or
justify the democratization of technology. For, in modern society, it is the
powerful moral code of private property, not just technocratic ideology,
which opposes the translation of users' or workers' participant interests
into legitimate rights to re-shape technology. Do they have such rights?
What rights? How are they ethically justified? Those are the key ethical
issues which derail Feenberg's democratic euphoria. If I am right, in the common case where technology is private property,
the technical code(s) is embedded in the moral code of Lockean ownership
and the rationality of capitalist market relations. The experts who claim and
exercise rights of exclusive control over technology are widely perceived not just as authoritative arbiters of efficiency, but as the designated agents
of the will and rights of owners, and less directly, the will and rights of
consumers who get to "vote" on technical design with their dollars. In these
contexts, challenges to technology based on workers', users', or impacted
third parties' participant interests involve challenges to the rights of private
property and modern society's powerful Lockean moral code. To draw what sort of ethics does technology require? 163
on one of Feenberg's examples, workers (or their unions) may challenge
production technology on the basis of their participant interests "in such
things as health and safety on the job, educational qualifications and skill
levels, and so on."16 When they do so, their claims may be discredited
or undermined not simply by experts' judgment concerning what is and
isn't feasible, efficient, etc. but by owners' or top-managements' "right"
to reject such changes as unprofitable, unnecessary, or incompatible with
company policy as they define it. The dominant technical code may be
seen as necessary or unnecessary, reasonable or unreasonable, but if it is
the will of the owners and embodies company policy, then many see it as
legitimate, and workers' or users' challenge to it as lacking moral force.
In sum, if as I am arguing, many technical codes are grounded in the
Lockean moral code of private property, dominant technology is provided
with powerful rights-based protections quite as basic as technocratic essen
tialism. Rational challenges to technology in these cases pushes beyond the
logic of Feenberg's participant interests to the issue of users', workers' or
participants' basic rights - as citizens, persons, human beings, democratic
agents, etc. If established technology can be reasonably re-interpreted as
a violation of these actors' basic rights or entitlements, this may provide a good reason for transforming dominant understandings of the rights of
private property, not simply the prerogatives of expert authority.
The private property problem forces a re-orientation of Feenberg's
project, moving it beyond the critique of essentialism and the discourse of
participant interests. Rather, his argument needs to develop an account of
the logic through which some participant interests but not others have been
or can be reasonably represented as legitimate claims of right, counter
balancing rights of private property. Whatever else it may involve, a
democratization of technology goes well beyond its demystification. It
involves not just a reinterpretation of technology, but an ethically well
argued reinterpretation or revision of the Lockean and liberal-democratic
moralities of right.17 A good start would be a normatively sensitive account
of the actual historical logic through which established technical codes
and rights of private property (e.g., sanctioning child-labor or polluting
factories) were successfully and rationally re-coded as violations of rights or ideals basic to democracy. But, as I argue in the last section, the aim
16 Feenberg, QT, p. 140. 17 For my own attempts to advance such an argument see G. Doppelt, "Rawls' System of
Justice: A Critique from the Left," Nous 15 (1981), pp. 259-307; "Conflicting Paradigms of Human Freedom and the Problem of Justification," Inquiry 27 (1984); "Rawls' Kantian
Ideal and the Viability of Modern Liberalism," Inquiry 31 (1988), pp. 413-419; and "Is
Rawls' Kantian Liberalism Coherent and Defensible?" Ethics 99 (1989), pp. 815-851. 164 GERALD DOPPELT
is to develop a general ethical standard or conception powerful enough to
establish a link between some but not all challenges to technology (and
rights of private property) and the substance of liberal-democratic ideals,
properly interpreted. Can such a standard be identified in Feenberg's conception of demo
cratization or his examples of it?
6. Changing Technology and Democratizing Society
My aim in the remainder of this essay is to argue that Feenberg's theoret
ical account of democratization and his examples of democratizing tech
nology do not provide an adequate normative standard or set of standards.
When we characterize society or social change as more or less demo
cratic, we may operate with very different standards in mind concerning its institutions, practices, and ideals. It is useful to distinguish standards
concerning democratic models of political agency from ones concerning
democratic models of equality or individual freedom/rights. We may focus on how power is exercised, who can or cannot, does or
does not have a voice in the key decision-making practices of the society at various levels of social life. I'll call various standards that focus here
ideals of democratic political agency, or political agency conceptions. On
the other hand, we may focus on the impact of a society's decision-making
practices on its structure of democratic equality or freedom: the degree to which its citizens and groups enjoy equal individual rights, freedom,
opportunities, essential resources, and statuses in the society. Agency is
involved here, but it is agency in the sense of individual autonomy -
personal control over one's own life and activity - rather than polit ical participation/agency in the above sense. I will call standards which
focus on this second dimension, ideals of democratic equality, or equality
conceptions - though it is often personal agency which equal rights,
opportunities, resources, etc., protect. Clearly both political agency and
equality enter into modern understandings of democracy. Furthermore,
each of these goods typically functions to serve or protect the other. Polit
ical theories disagree on how they interpret the substance of one or both
of these two sorts of democratic ideals - political agency and equality -
and which they see as more normatively important or fundamental to a
democratic society. For example, the liberal tradition from classical theorists to John Rawls
sees democratic political agency, representative government, as having
primarily an instrumental value. It is the only or best arrangement for
protecting individual freedom, rights, equality of opportunity, market WHAT SORT OF ETHICS DOES TECHNOLOGY REQUIRE? 165
freedoms - i.e., equality, which is taken to be the value most fundamental
to modern society. Other theorists (civic republicanism, radical democracy,
etc.) embrace far more ambitious models of democratic political agency
which is seen as having intrinsic value or virtues as, or more, central to
"true democracy" than individual freedom or the equality which protects
it.18 Feenberg does not address these normative disagreements.
Yet, it is clear from his treatment of theorists of democracy such
as Benjamin Barber and Richard Sclove, that he is operating with a
"political agency" conception or ideal of democracy.19 With Barber and
Sclove, Feenberg wants to reject the liberal ideal of democracy I charac
terize above. Following Barber's model of "strong" democracy, Feenberg accords central value to a populist, participatory politics involving local
collective action, direct citizen intervention, and bold social movements,
exemplified by AIDS activists, environmentalists, feminists, and other
groups of lay actors. Following Sclove, Feenberg holds that the key
problem for this participatory model of radical democratic political agency
is how to apply it to technology. Indeed his formulation of this problem and
the way he seeks to resolve it constitute impressive insights. Both standard liberal and radical participatory models of democratic
political agency define the relevant units of agency, the public, and repre
sentation by means of conventional geographic boundaries. Thus, on these
various models, the relevant political actors are variously identified as
the citizens of the nation, the citizens of this or that municipality, the
employees of a hospital or factory, etc. But, as Feenberg argues, the lay
public which might exercise democratic control over technology cannot be
identified by such conventional political boundaries.20 Modern technology
implies the "fragmentation of technical publics" - a proliferation of diverse
subgroups of users, each of whom bear different practical relations to a
techno-system and none of whom necessarily occupies one and the same
conventional geographic or political boundary.21 Who then is supposed to
exercise democratic political control over technology, and on what basis?
Feenberg's intriguing answer is that technical networks create new
political subjects - e.g., ill people seeking access to experimental drugs or
clinical trials, or users of a new technology such as the French Minitel.22
18 For a recent and engaging contribution to this tradition, see I. Shapiro, Democratic
Justice (New Haven: Yale University Press, 1999), and my "Does More Democracy Yield
Greater Justice," Inquiry 44 (2001), forthcoming. 19 Feenberg, QT, pp. 132-137. 20 Feenberg, QT, pp. 134-147. 21 Feenberg, QT, pp. 134-147. 22 Feenberg, QT, pp. 134-147. 166 GERALD DOPPELT
While such subjects defy conventional geographical and political bound
aries, their common practical relation to a technology may give them
common "participant interests" which in turn may become the basis
for democratic political agency concerning that technology's impact on
them. In dialectical terms, a technology, and the experts that control
it, create their own "other," specific groups that develop new common
interests which prepare the way for a dialectical overcoming, and more
harmonious "whole," of technical and human relations. Through such a
dialectic, the users, participants, customers, third parties impacted by their
common practical relation(s) to technology theory become political agents - precisely the agents required by a democratization of technology! A
beautiful theory, indeed.
But, at this point, my earlier critique re-arises. The very fragmentation of technical publics stressed by Feenberg's incisive analysis of modern
technology seems to bring in its wake a fragmentation of participant
interests, agency, and the very meaning or prospect of democratization.
What standard of democratization is operative in his argument? Is he
willing to see democratization whenever any politically marginalized lay
group(s) of users exercises power over technology? Or does it also depend on the substance of their interests and demands? Doesn't it depend on the
relationship of what they do to the broader democratic ideals of political
agency and equality? Doesn't it depend on how the group justifies its
interests, how other groups' interests and political agency are impacted, and how other groups interpret what has been accomplished? The fact that one sub-group of users of technology gains some power over it should not necessarily count as democratization, especially if the
change comes at the price of dis-empowering or excluding other broader
groups of users with basic rights, opportunities, or interests at stake.
Suppose a group of industrial or office workers succeed in gaining a
marginally safer, or more pleasurable, or easier to operate, technology in
the workplace. Imagine that the "cost" is that others who are not consulted
lose their jobs or skills. Or, that thousands or millions of consumers lose
access to a basic good or service as a result. Worse, imagine that those
who benefit and those who are disadvantaged or harmed by the change in
technology are already divided by differences of race, class, ethnicity, or
gender. Without taking such zero-sum conflicts of group interest, power,
and identity into account, we cannot reasonably evaluate whether or not
the change in technology should be, or in fact will be, seen as a demo
cratization of technology. And, we require a defensible ethical standard of
democratization to ground such evaluations, and to transform the politics
of technology. WHAT sort OF ethics does technology require? 167
7. Deep Democratization: Political Agency vs. Equality
Feenberg advances his own political ideal of "deep democratization"
perhaps in order to overcome the sorts of problems I've raised above.23
Deep democratization involves a political process or strategy through
which direct action by subgroups of users (fragmented technical publics)
generates the basis for a new practice or institution of technical decision
making in our society. In this new practice, all participants in major
technical institutions or networks would enjoy a normal political participa
tion and role in shaping the technical code and design of technology. This
practice would shift power from experts and owners to all users whose
interests are impacted by a technology. Fragmented technical publics with
diverse or even opposing participant interests in a given technology would
all become democratic agents who recognize that they have a common
interest in exercising shared rights of political representation in the process
of determining that technology.24
My criticisms of deep democratization are as follows. In principle, it
could function as an ethical standard for evaluating any particular chal
lenge to technology as a case or non-case of democratizing technology,
depending on its relation to the ideal of deep democratization. But is there
any empirical evidence that the cases Feenberg cites are part of a process
of deep democratization? He certainly doesn't present any, because his
central examples don't seem to be cases in which the intentions, conscious
ness, or impact of lay actors embody or advance deep democratization.25
Thus if that is his criterion, it is doubtful that any cases of specific
challenges to technology that he discusses count as democratizing it. Feen
berg's participant interests and fragmented technical publics are centered
on specific negative impacts of a technology. Typically, they don't even
involve an interest of the subgroup of users in exercising wider political control as a purpose in itself, not to mention any interest in enabling other
groups to exercise such control.
Secondly, Feenberg's non-essentialist theory lacks the resources to
motivate or justify his ideal of deep democratization. As I have already
argued, grounding such an ideal in modern society involves much more
than demystifying technology. It requires the elaboration of ethical argu ments aimed at a radically different understanding of Lockean rights and
liberties, on the one side, and democratic equality, the citizen's rights and
freedom, on the other. Deep democratization is a version of socialism,
23 Feenberg, QT, pp. 142-147. 24 Feenberg, QT, pp. 145-147. 25 Feenberg, QT, pp. 140-147. 168 GERALD DOPPELT
which is widely perceived as the enemy of individual rights and freedom,
as they are presently understood. While this is unfortunate and invites
theoretical/political critique, Feenberg's demystifying theory, by itself, is
unequal to the task.
Finally, and most fundamentally, the ideal of deep democratization
remains stuck in the political agency paradigm of democracy, and ignores
the need for a theoretical engagement with the equality paradigm. Deep
democratization, even if fully realized, is not democracy enough. As the
equality paradigm insists, the crucial test of new powers of decision-mak
ing must include the impact on the everyday rights, liberties, opportunities,
resources, personal autonomy, and dignity of individuals, and especially of oppressed and disadvantaged groups. As I would construe them, demo
cratic ideals depend on the moral substance of people's interests, and not
just their power to embody their interests in the built environment.
Indeed, when we examine Feenberg's main examples of democratizing
technology and his negative disappointing cases, they implicitly evoke
the equality (or individual rights/personal autonomy) paradigm, though he
fails to theorize it.
8. DEMOCRATIC EQUALITY
If I am right, Feenberg's vision of democratizing technology and an alter
native modernity must be grounded in an ethical account of the interests
which might define, justify, and motivate this emancipatory project. What
I've called the Lockean moral code does not just provide rights-based
protections of technology as private property. As a modern ethos, it shapes
users' very desires and interests in ways which tend to bind them to the
rationality of established market and technical relations.
Feenberg's work is permeated by an ironic tension or contradiction.
His theory of democratization rests exclusively on the "political agency"
paradigm. But his examples of successful democratization, as well as his
evaluations of absent, failed, or flawed democratization, implicitly rest on
the "equality" paradigm: a view of the sorts of substantive interests, the
specific changes in the technical codes at issue in a genuine democratiza
tion of technology. He observes that a version of his ideal of deep demo
cratization, under the heading of "self-management" or "co-management,"
has been embraced and practiced by unions in Germany and Scandinavia.
Though workers gained some degree of democratic control over tech
nology, for Feenberg, the results are disappointing because it produced "no
major changes in technical codes."26 Workers' new found political agency
26 Feenberg, QT, p. 146. WHAT SORT OF ETHICS DOES TECHNOLOGY REQUIRE? 169
failed to generate any significant impact on the democratization of tech
nology. In this context, Feenberg obviously has some implicit view of the
workers' major or true participant interests - what would have produced a significant enough change in technology to count as its democratization.
In a similar vein, he explains the absence of pressure to democratize tech
nology as a result of: (1) the public's lack of awareness of its participant
interests - "... how profoundly affected it is by technology";27 and (2)
American workers' agreement to exclude "the most important implications
of technology for workers" from technical politics - in favor of focusing on
issues such as "job security."28 In all these cases, for Feenberg, the obstacle
to democratizing technology is not primarily political powerlessness (the
political agency paradigm), but users'/participants' failure to recognize or assert the "right" interests and changes in technology (the equality
paradigm). His own assessment underscores the need for going beyond
his whole discourse of deep democratization in order to define and justify the moral substance of an alternative modernity. When we turn to Feenberg's accounts of successful democratizations
of technology, the same points emerge, with some clues concerning how
to address them. He characterizes the movement of disabled people for
barrier-free design or ramps on streets, public buildings, businesses, etc.,
as the most "compelling" example of a democratic politics of tech
nology.29 The dominant technical code for designing sidewalks could
"rationally" or "efficiently" exclude ramps, as long as disabilities count
as purely personal problems, irrelevant to the design of public spaces. But
disabled people comprise a large population with a powerful participant interest in "mainstream social participation."30 Once they mobilize and
gain public recognition of this interest, many technical features of the built
environment are transformed.
Why is this such a compelling case of democratizing technology?
Indeed, it is qualitatively different from Feenberg's other case of millions
of French users of Minitel transforming it from an information to a
communication technology, which hardly seemed to democratize anything,
by my standards (i.e., democratic equality).31 Arguably, in both cases,
groups of users exercise power previously denied to them in order to
bend technology to their legitimate interests. Feenberg's concepts of users,
power, and participant interests, are too abstract to capture the important
27 Feenberg, gr. 28 Feenberg, QT, p. 140. 29 Feenberg, g?; p. 141. 30 Feenberg, QT. 31 Feenberg, AM, pp. 144-169. 170 GERALD DOPPELT
differences between these cases, and to show why the first is so much more
compelling a case of democratization than the second. We need to attend to
the paradigm of democratic equality and ask who is this group of "users,"
where do they stand in society, what have they been denied, and what is
the ethical significance of the technical change they demand?
Disabled people are suppose to be equal citizens with the same rights,
equality of opportunity, personhood, and dignity that other groups enjoy. In practice, they are victims of prejudice and discrimination, inequality and indignity. Their movement for barrier-free technology was identi
fied by them, and eventually, much of the public, as a struggle for basic
civil liberties, citizenship, the right to enjoy the same access to public
buildings, social life, urban mobility, as other Americans. Thus, disabled
peoples' transformations of technology is democratic first and foremost,
because it involves a victimized groups' gaining (1) the same ability to
exercise fundamental rights and liberties as other citizens, and (2) public
recognition that they can do so, and deserve to.
In sum, the action of the disabled, unlike that of the Minitel users, is
about the ideal of democratic equality. It counts as a democratization of
technology precisely for that reason. But doesn't it also count that the
disabled gained this democratic, moral, and technical change or result
through their own political agency? Doesn't this aspect of "democratiza
tion" make it akin to the case of the French users who "hacked" the Minitel
system to bend it to their purposes?
My answer is "yes and no." Had barrier-free design been the result
of political actors other than the disabled themselves, it would still have
constituted a democratic change in technology. Why? Because it enhances
democratic equality. Nonetheless, the fact that the political agency of the
disabled played a key role in generating this result is part of what makes it
a democratization of technology. Why? Against Feenberg, it is not simply
because it is a case of users exercising power over design, like Minitel's
hackers. Rather, in the case of the disabled, it is their political agency on
behalf of effective rights as citizens, as well as the access they gain (the
ramps) which transforms their recognition and position within the space
of democratic equality. It is the fact that they claim their rights and the
equal worth of their freedom, as much as the technical change they win,
that secures their recognition and citizenship. In the next section, I examine Feenberg's treatment of AIDS activ
ists and seek to determine the criteria of democratization that we should
employ in order to evaluate it. what sort of ethics does technology require? 171
9. Democratizing Experimental Medicine
Feenberg provides his most explicit, detailed normative argumentation
concerning participant interests in his account of the challenge of AIDS
activists to the technical code underlying clinical trials of experimental
drugs.32 On the technical code of medical experts, there is a sharp distinc
tion between research and treatment or care. Clinical trials involving the
testing of experimental drugs fall under the category of scientific research
aimed at gaining knowledge of the drug's medical value. Clinical trials are
not a form of legitimate treatment or care because the medical efficacy of
the drug has not yet been scientifically established.
This technical code shapes the design and regulatory policy governing
clinical trials and access to experimental drugs more generally. Patients
must be protected from false hopes, irrational decisions, scientific exploita
tion, and medical abuse, for their own good. This requires stringent public
regulation of patient access to, and physicians' use of, experimental drugs.
They should be used exclusively as objects of research, not modes of treat
ment, until their medical value is scientifically established and certified.
This provides the rational basis for the design of clinical trials. Because
they constitute research, not treatment, it is entirely efficient and appro
priate to limit patient access to clinical trials, use placebos to determine the
causal efficacy of the drug, and employ elaborate conventions of formal
consent to protect participants from deception and false hopes. On the
essentialist paradigm, the technology of the clinical trial is necessarily dictated by the requirements of reliable scientific inquiry and the medical
knowledge already possessed by expert researchers.
In the story that Feenberg recounts, AIDS patients and activists
demanded a radically liberalized and expanded access to experimental
drugs and clinical trials. These are people with terminal and incurable
illnesses. They have the largest personal stake in finding a cure or bene
ficial treatment. They want to be in a position to try unproven drugs that might help or work and to be participants in the life and death race
for a cure, rather than doing nothing. So they demand changes in the
policies, availability, and design of clinical trials, in order to gain access to
experimental drugs and the research process.
On the dominant technical code of established medicine, the AIDS
activists' demands reflect ignorance, irrationality, false hopes, and
destructive outcomes. These demands would destroy the sharp boundary between research and treatment, and in the process, harm the scientific
effort to discover a genuine cure. In response, Feenberg poses the key
32 Feenberg, AM, pp. 96-120. 172 GERALD DOPPELT
normative question inspired by his non-essential theory of technology: Does the desire and demand of AIDS victims for access to drug trials
deserve public recognition as a legitimate participant interest and a rational
basis for revising the technical code governing the design of experimental medicine?
The way Feenberg poses and answers this question fits well with the
general critique I have made of his argument for democratizing technology. How so? The AIDS activists achieved some success in expanding access
and revising the technical code of clinical trials as pure scientific research.
But, as far as I can tell, Feenberg is not content to take the mere fact of
this marginalized group's empowerment as "the" mark of a democratic
rationalization (improvement) of experimental medicine. Rather he seeks
to provide a moral justification for taking AIDS victims' desire/demand for
access as a good ethical reason for revising the technical code of clinical
trials. That is, in this example though not in his general theory, he recog
nizes that the case for a liberating democratization of technology depends on providing a normative justification of the interests or demands which
inform and motivate it. But what he says in this case may provide clues for
developing a general theory. In effect, Feenberg argues that upon analysis, clinical trials and access
to experimental drugs are forms of medical treatment or care and not
exclusively contexts of pure scientific research.33 He marshals a body of
evidence and theory which tends to show that the well-being of patients is positively impacted by their social engagement in clinical trials, and
in various other practices of medicine - all of which provide medical
care, even in the absence of a physical cure. As a result, AIDS patients'
desire to participate in drug trials embodies a legitimate interest in gaining medical care - one of the few or only forms of care available to people with incurable, fatal illnesses. Modern medicine embraces "care" as one
of its major professional callings, even if it has been somewhat eclipsed by the technical pursuit of "cures." Thus, Feenberg provides a moral justifica tion for the AIDS activists' challenge to established medical technology based on its own internal moral code of providing care to the ill and
suffering. On Feenberg's illuminating interpretation, what the AIDS activ
ists accomplished is the leveraging of modern medicine's moral code to
revise its technical code. As a result, previously well-established policies
surrounding drug trials (e.g., limiting participation to statistical minimums,
the employment of placebos, excluding subjects with prior experience in drug trials, etc.) could now be rationally modified in some cases to
accommodate subjects' now "legitimate" interest in participation.
33 Feenberg, AM, pp. 110-118. WHAT SORT OF ETHICS DOES TECHNOLOGY REQUIRE? 173
Does Feenberg's argument concerning this powerful example establish
that it is a case of democratizing technology? Does it provide implicit clues concerning the normative standard(s) of democratization we should
employ to evaluate and generalize such cases of technical change? To
begin with, it is useful to observe the modesty and ambiguity of what
Feenberg takes himself to have established in this case. Feenberg stops
well short of defending the moral right/entitlement of persons with incur
able, terminal illnesses to have access to experimental drugs and clinical
trials. Today, there are many who lack this access, particularly to the
proven but expensive drug cocktails discovered in the intervening years.
Feenberg's argument does not seek to establish a moral right to care in
the case of persons with terminal illness - so the change in the technical
code he defends does not necessarily morally condemn the situation of
scarce access. Indeed, he continues to think of "the principle purpose of
experimentation" as "the acquisition of new knowledge" - not serving the
interest of the ill in the care or well-being provided by participation.34
Furthermore, he recognizes that there are tensions and trade-offs between
increasing the access of ill people to experimental drugs and securing the
sorts of clinical trials most effective for providing knowledge of a cure.
So, the change in the technical code defended by Feenberg amounts to
the important but essentially modest claim that "a good medicine," when
and where it is feasible, will "design experiments that serve patients while
simultaneously serving science."35 But even this claim goes well beyond what his reading of AIDS activists establishes. This case shows that the
interest in experiment as care is an urgent, legitimate factor in design for
persons with incurable terminal illness; it doesn't establish the same point for any and all persons with any illness whatsoever, which is what some
of Feenberg's formulations suggest. Nothing about the case of people with
AIDS speaks to the status of people who desire to enter experiments which
test new drugs for skin rash and their interest in the care such participation
may bring. These considerations return us to my main question of which
changes in a technology bring about democratization and why. In order to evaluate the impact of any particular case of technical
politics on the democratization of technology and society, we need to ask
who is this group of users challenging technology, where do they stand in
society, what have they been denied, and what is the ethical significance of the technical change they seek, for democratic ideals? These questions call the attention to features of the ADDS example which are relevant to the
standard(s) of democratization but are ignored in Feenberg's treatment.
34 Feenberg, AM, p. 108. 35 Feenberg, QT, p. 141; Feenberg, AM, p. 109. 174 GERALD DOPPELT
WTien we ask who the AIDS activists represent, it is clear that they are not simply a group of citizens with a deadly or incurable illness. It
is a stigmatizing illness culturally associated with a group oppressed by
homophobic fear, hatred, prejudice, and discrimination. ADDS was not
treated as a horrible plague requiring the total mobilization of society's
initiative, resources, and compassion. It was treated as the obscure fate of
a marginal group of "others," or worse, nature's way of punishing repug
nant, unnatural, immoral, and irresponsible sex. The struggle of AIDS
activists for access to treatment, care, agency, choice, etc. challenges the
moral code which sustained homophobia and sanctioned the early record
of sparse funding for AIDS research and care. Their success in revising
the technical code governing clinical trials, coupled with related political
attainments, enhances the credibility, legitimacy, and moral standing of
a despised minority in America - gay men with HIV or gays generally.
They gain a significant measure of human dignity, agency, respect, and
citizenship - even though they do not win a right to care, or a right to
participate in clinical trials, as Feenberg admits. It is this positive impact on the ideal of democratic equality that makes it a powerful case of demo
cratizing technology. As in the case of the struggle of the disabled for
ramps, it is not exclusively or primarily the exercise of political power by
lay actors that provides the criterion of democratization. Rather, we need
to also look at the impact of these struggles over technology both on the
moral and political standing of oppressed groups and the long march to
genuine democratic equality for all.
10. Towards an Ethics of Democratic Equality
Let us return briefly to the example of the struggle of the disabled for
ramps, in order to summarize my argument in this essay. We can employ
this example to highlight the strengths and weaknesses of Feenberg's
theory of technology and the path of future research for democratization.
The movement of disabled people for barrier-free access requires
both a demystification of technology and an ethical reconstruction of the
ideal of democratic equality and the meaning of citizenship/personhood
for the disabled. Following Feenberg, for democratic change to occur,
lay actors/users need to see that existing sidewalks, bridges, building
entrances, restrooms, classrooms, etc. (without entry access for the
disabled) do not embody a timeless necessity and efficiency; and that they
do exclude the interests and voices of some group(s) of users. Feenberg's
theory accomplishes this important task, important because it constitutes a
necessary condition for democratization. But this much does not neces WHAT SORT OF ETHICS DOES TECHNOLOGY REQUIRE? 175
sarily provide public or private parties with good reasons to support a
design change or force an opening up of decision-making - both of which
always involve costs of some sort to somebody, whether it is owners,
consumers, taxpayers, or whomever. For various actors to have such good
reasons, they need to and should be convinced that there are powerful moral rights, entitlements, and ideals at stake, especially where other
rights will have to be abridged or overridden. Feenberg's theory lacks
the resources to accomplish this task. He doesn't develop any framework
of ethical concepts for analyzing democratic ideals or showing how they enter into, or are broadly impacted by, the cases of users' challenges to
technology which he examines. He doesn't mobilize his cases to articulate
any general ethical standard(s) of democratization which would enable us
to evaluate particular challenges to technology or to define a liberating
politics of technology, an alternative modernity. The political philosopher who aims at a general democratization of technology confronts a multi
dimensional project - criticizing and revising, in tandem, dominant ethical
understandings of technology, private property, democratic equality, and
the morality of individual rights. The critical theory of technology cannot
complete its mission without turning into a more thorough-going critical
philosophy of liberal-democratic values. An Alternative Modernity is a
modernity with a different vision of human relationships and the good life.
The democratization of technology awaits the development of an alterna
tive ethical understanding of the ends of modern society. More generally, the influential discourses of anti-essentialism and social constructivism
create a political vacuum which can only be filled by affirmative ethical
argument concerning the right and the good.
Department of Philosophy
University of California, San Diego
LaJolla, CA 92093-0119
USA
E-mail: jdoppelt? helix, ucsd. edu
