Science, Technology and Democracy: Distinctions and Connections

Andrew Feenberg

Abstract:

This paper argues that despite considerable overlap, science and technology must be distinguished. Research aimed at understanding of nature is controlled by the community of researchers. This distinguishes it from activities aimed at the production of products under the control of organizations such as corporations and government agencies. Even where one and the same activity aims at both truth and utility, it is controlled in these two different contexts. This distinction is traced in the paper through the post-War history of science and society in America, through direct comparison of several cases and their implications, and through a discussion of the paradoxical structure of technology-society relations. These relations constitute an “entangled hierarchy” because social groups form around technical mediations which they in turn mediate and transform. The politics of science and technology differ in that the contribution of social groups to scientific change is far less direct than to technological change.

Prologue: The Cold Fusion Fiasco

On March 23, 1989 Martin Fleischman and Stanley Pons appeared at a press conference at the University of Utah where they announced the discovery of cold fusion. The President of the university and several other officials were also present and spoke to the press. The unaccustomed involvement of the press and these officials signalled that cold fusion was more than a scientific advance. Soon the University announced the formation of a research institute with funding from the state. Its goal was not only to produce knowledge of the phenomenon but also to prepare large scale commercial applications. It seemed possible at first that cold fusion would revolutionize electricity production and transform the world economy.

We know the end of the story. Within a short time cold fusion was discredited and most researchers lost interest in it. The institute at the University of Utah closed in 1991 and support for further work in this field quickly evaporated (Simon, 2002). These events provide a particularly clear illustration of the complexity of the relation between science and technology today.

The classic but generally discredited account of these relationships holds that science is a body of truths about nature and technology an application of these truths in the production of useful devices. Truth and utility belong to different worlds linked only by the subordination of the latter to the former. But historians have shown that few technologies arose as applications of science until quite recently. Most were developed independent of science and, indeed, in cases such as optics had more impact on science than vice versa. Science is even more dependent on technology today than in the past. It is true that the 20^th century saw a dramatic increase in practical applications of scientific knowledge, but this new situation does not reveal the essence of the science-technology relationship. Rather, it confounds the common sense distinction by establishing the productive character of science itself.

In any case, the classic model does not describe cold fusion. Fleischman and Pons did not apply any existing science in their work but made an empirical discovery of the sort that we associate with invention. They were not seeking to confirm or invalidate a theory with experiment as philosophical accounts of scientific method would have it, but rather aimed to produce an unexplained (and ultimately unexplainable) effect. Their discovery employed a technical device that was both an experimental apparatus and a commercial  prototype. Accordingly, the two pronged launch of their discovery aimed at both the scientific and the business communities.

Cases such as this one proliferate in the biological sciences, where scientific techniques are deployed in the search for results of interest not only to researchers but also to pharmaceutical houses. Products and knowledge emerge from the laboratory together. The pursuit of knowledge and the making of money are joined in a single labor. The distinction between science and technology appears to break down. Hence the widespread use of the term “technoscience.”

Distinguishing Science and Technology

Postmodern scholars and many researchers in Science and Technology Studies no longer believe there is any distinction of principle between science and technology. Certainly the boundaries are much fuzzier than in the past. But if we conclude that they are no longer distinguishable at all, what becomes of the associated distinctions between theory and practice, research and application, scholarship and business, truth and utility? Must they be given up too?

The old distinction between science and technology and all these associated distinctions implied a value hierarchy. Science, theory, research, scholarship and truth were considered nobler than technology, practice, application, business and utility, in accordance with the ancient preference for disinterested contemplation over worldly activity. This hierarchy grounded the demand for the complete autonomy of science. In 1948 P.W. Bridgman expressed this “ivory tower” indifference when he said “The assumption of the right of society to impose a responsibility on the scientist which he does not desire obviously involves the acceptance of the right of the stupid to exploit the bright” (Bridgman, 1948: 70).

As the distinction between science and technology blurs the value hierarchy that justified such outrageous snobbery loses its persuasive force. A basic change has occurred in the relationship between science and society. There is growing openness on the part of science to various forms of political and economic control and in some cases what I will call “democratic intervention” by lay members of the public. But what exactly do we mean by this?

Certainly not eliminating the laboratory, obliging scientists to work with the public looking over their shoulders, and relying on government for epistemic decisions. Democratization and political and economic intervention into science are more modest in their objectives for many reasons. But the struggle for social control of technology is hardly modest. It intensifies constantly and it often leads to direct intervention by citizens and governments into technological decisions and even into the decision-making criteria employed to select technologies.

The old value hierarchy has certainly been scrambled in recent years as more and more scientific work aims directly at producing marketable goods. We live in a two dimensional flatland, not a three dimensional universe with vertical coordinates. But despite the changes, we cannot do without the old distinctions. They correspond to vital strategic divisions within the  world of politics. The question is, how can we reconstruct the distinction between science and technology without falling back into an outmoded valuative framework? That is what I will attempt here.

In the remainder of this presentation I want to offer a new framework for discussing the relationship between science, technology and democracy. I will discuss four issues in the time allowed. First, I want to introduce some basic criteria for making the distinction that concerns us here. Second, I will propose a historical sketch of the evolving cognitive relation of science and society. Third, I will argue that democratization has a specific normative significance for technology it does not have for science. Fourth, I will present some philosophical reflections on the paradoxical structure of the emerging technical public sphere.

Two Criteria

Even if it is sometimes difficult to distinguish the pursuit of truth from the pursuit of utility, other criteria enable us to the make a usable distinction between science and technology. I am not concerned here with the obvious cases such as the difference between theoretical physics and road work. The difficult cases are more interesting. They arise in the expanding zone of activities that appear to cross the line between science and technology. In that zone criteria can be developed from study of scientific and technological practice, for example, the subtle differences in the roles of knowledge and technical work in experimentation and science based technology (Radder, 2009). Here I will focus on criteria reflecting significant differences in governance and procedures because they are directly relevant to the politics of science and technology.

The science/technology distinction used to be associated with the distinction between academic and corporate research. But there are obvious counter-instances such as Bell Labs where high quality scientific world has been done under corporate auspices. Nevertheless, there is a difference between the kind of research done in universities and Bell Labs and most product development, including development that employs laboratory methods but which is conducted in secret or used to promote specific products. This suggests a first criterion for distinguishing science and technology: the difference in decision procedures in the two cases.

Scientific controversies are decided by the scientific community, or rather, by what sociologists of science designate as a “core set” of researchers engaged in debating the relevant scientific issues. Social, cultural and economic constraints play only indirect roles in these debates, for example, empowering some participants to carry out expensive experiments or influencing the initial response to the announcement of results. But in the final analysis epistemic tests carried out by individuals or small groups in conferences, articles, and laboratories are the principal measure of competing ideas.

I do not mean to imply that scientists arrive at absolute truth, but they do achieve credible knowledge of nature and this is their primary aim, the make-or-break factor in their work, even if that work also involves them in commercial activity. Technology too involves knowledge of nature but many of the most important decisions in this case are not about knowledge. Social and economic criteria are relevant to technological choices and intervene through the mediation of organizations such as corporations or government agencies that employ technical workers. These workers, who may be scientists, are usually situated in a chain of administrative command leading up to individuals in non-technical roles with wide responsibilities that have nothing to do with knowledge of nature. Where those individuals determine outcomes, we can be fairly certain we are dealing with a primarily technical activity, even if scientific knowledge is ultimately generated as a by-product.

This difference is clearly illustrated by the cold fusion affair. The pursuit of commercial cold fusion depended on the willingness of the state of Utah to invest in a likely money maker. The research was to be oriented toward this goal. Within the institute the existence of cold fusion was not in question and the experiments were conducted in secret. But the very same effect which the organization was created to exploit was also exposed to scientific evaluation and this proved to be decisive. There the potential profits to be made on commercial electricity production were attention-getting but less significant. Scientific criteria were brought to bear on the effect, so far as knowledge of its production was available, and it was rapidly discredited, primarily by two epistemically significant factors: failures to reproduce the effect in the laboratory, and lack of a plausible connection between the effect and existing theory. Clearly, truth and utility still belong to distinguishable worlds, even if they refer to aspects of one and the same phenomenon and often cross boundaries in pursuit of their separate goals. The point of intersection, where scientific and technological criteria must both be aligned, corresponds to the proper application of the term “technoscience.”

This blurring of boundaries has had an unfortunate influence on the evolution of research funding. In recent years neo-liberal ideologists have convinced governments that the responsiveness of science to society is measured by the commercial success of its applications. An ever tighter bond between business interests and funded research programs has increasingly harmful impacts on the research community. Public support for basic research in a wide variety of fields, including many with no immediate prospect of commercial payoffs, is the basis of long term scientific advance. It is also essential that science have the means to serve the public interest even where business prospects are poor, as in the case of medicines for “orphan” diseases. This new system reduces science to a handmaiden of technology, with disastrous consequences because not all of science is “techno-”.

The second criterion useful for distinguishing science and technology is the different role of underdetermination in the two cases. The concept of underdetermination was introduced by the French historian Pierre Duhem to explain the fact that scientific theories are not uniquely determined by observation and experiment. The interpretation of these tests of theory always depends on other theories and so the whole edifice of knowledge is implicated in the evaluation of any particular branch of it. In practice, this means that no logically decisive experiment can relieve the researcher of the need to make a personal decision about the truth or falsity of the tested theory. Such decisions, Duhem claimed, are based on “good sense.” They are rational, but not possessed of the certainty often claimed for science.

Cold fusion illustrates this conclusion, if not Duhem’s precise point, since failures to reproduce the effect were interpreted by Pons and Fleischman as technical breakdowns and by their opponents as proving the non-existence of the effect. The decision between these two interpretations could not be made on the basis of experiment alone since the competence of the experimenters was in question.

Variations on this theme have been discussed in philosophy of science for a century. No doubt there is something to it. But Pons and Fleischman discovered that ad hoc explanations are weak defences for anomalous and conflicting experimental results such as characterized the cold fusion case. The only effective move in such cases is the production of new theory that encompasses old and new observations alike. But the production of plausible alternatives is extraordinarily difficult. Advocates of cold fusion were unable to supply one. Their failure is not unusual. Although Einstein objected to quantum mechanical uncertainty, he found it impossible to come up with something better. Creating new scientific theory requires rare originality and a special kind of critical insight into existing theory.

The case with technology is quite different once again, not least because alternatives are usually easy to invent. The concept of underdetermination can be adapted to signify this difference. It is obvious to engineers and other technical workers that no “technological determinism” or “technological rationality” dictates a single design of each device. The  technical equivalent of Duhem’s “underdetermination” of scientific observation and experiment is the proliferation of alternative designs of roughly similar devices. Just as observation and experiment can have different meanings in different theoretical contexts, so devices can be designed differently and have different meanings in the larger framework of existing technology.  

There are of course hard problems such as the AIDS vaccine. We will be lucky to find a single successful design, much less a multiplicity among which to choose. But most technical problems are not so hard and alternatives are available. The question then is how choices are made among them. Technical underdetermination leaves a wide opening for social, cultural and economic criteria to weigh on the final decision between alternatives. The equivalent of scientists’ “good sense” in this case is supplied by management sending orders down the chain of command to technical workers whose advice they may or may not have taken into consideration.

Democratizing Science

With these distinctions in mind, I want to introduce some historical considerations on the concept of the democratization of science. Science was always marginal to national politics until the Second World War. The Manhattan Project and radar research actually changed the course of the War and thereafter the union of science, government, and eventually business became one of the driving forces of social and economic development. But science was exposed to new forms of public intervention as a result. I will sketch this history very briefly in the American context.

The Manhattan Project played a special role in this transformation of the relationship between science and society. The scientists involved were sworn to secrecy throughout the War. They acted as agents of the federal government under military command. But they realized toward the end, when it came time to decide whether or not to use the bomb, that they were not simply government employees. Because of the secrecy of the project, they were also the only citizens able to understand the issues and express an opinion.

Under the leadership of Leo Szilard and James Frank they attempted to enact their role as citizens by petitions and reports advocating non-use. They were unsuccessful but after the War, when they were no longer bound by military secrecy to the same degree, a number of them committed themselves to informing public opinion. The famous Bulletin of the Atomic Scientists was the semi-official organ of this “scientists’ movement.” It had wide influence but it took many years for its advocacy of test bans and disarmament treaties to have an effect on public policy.

There was a strong element of technocratic paternalism in this movement. In the immediate post-War period, up until the middle 1960s, technocratic notions were widely believed to chart the course for the future of modern societies. Politics was increasingly guided by technical experts of one sort or another. But the problem of what to do about public opinion remained once its input was devalued relative to expert advice. One solution consisted in refining the techniques of persuasion. Scientists chose a more respectful alternative and attempted to educate the public. Their efforts were motivated by the sense that an uninformed public might obstruct essential government decisions based on scientific knowledge.

This experience influenced the attitude of scientists in the 1960s and ‘70s as the environmental movement began to take shape. Biologists saw themselves in the role of the atomic scientists of the post-War period, possessed of knowledge of critical importance to the public. They too attempted to inform the public, advocating science-based solutions to problems most people could barely understand.

But technocratic paternalism soon gave way to a new pattern. Disagreements arose among environmentalists in the early 1970s and weakened the authority of science. True, some physicists disagreed over issues such as civil defense but the vast majority of the articulate scientific community favored the policies embodied in the treaties that still falteringly regulate nuclear affairs. No such consensus emerged in the environmental movement. In fact there were open conflicts over the causes of pollution, some blaming over-population and others blaming faulty technology, some calling for involuntary controls on births, others more vigorous regulation of industry, still others a return to nature or at least to “voluntary simplicity” (Feenberg, 1999: chap. 3).

The appearance of politically significant splits in the environmental movement meant scientists could no longer occupy the role of teacher to an ignorant public, but that they were obliged instead to play politics in the search for public support. For a population that made little distinction between science and technology, the loss of authority that resulted from these controversies was amplified by a series of technological disasters. The Vietnam debacle testified to the limits of the kinds of knowledge and power the technocratic state had at its disposal. The Three Mile Island nuclear accident in 1979 refuted the standard measures of risk put forward with such misplaced confidence by the scientific and engineering community. The Challenger accident in 1986 was a rebuke to the hubris of a nation that was proud of having put a man on the Moon. Many other incidents contributed to a gradual shift in sentiment and by the end of the millennium few young people were choosing scientific careers and strong fundamentalist movements were increasingly effective in opposing the teaching of science in schools.

Against this background a new configuration gradually emerged. By the 1970s we were beginning to see more public awareness of medical and environmental issues that affected individuals directly in their everyday experience. These issues were not confined to the realm of public discourse as had been nuclear issues in an earlier period. Now individuals found themselves involved in scientific-technical controversies as victims or potential victims of risky technical activities. In cases such as these ordinary people may well possess part of the truth before scientists interest themselves in their problems. That is a reason for scientists to listen as well as speak, to accept the role of learners as well as the role of teachers. In this context small groups of scientists, technologists and citizens began to explore an entirely new relationship between science and society. This relationship took the form not of paternalistic education but of a true collaboration with community activists.

A signal instance was the Love Canal struggle in the late 1970s. Residents of this community organized to demand government help dealing with the nearby toxic waste site that was sickening them and their children. They worked closely with volunteer scientists to document the extent of the problem and eventually won reparations. In this case lay informants brought a problematic situation to the awareness of scientists and collected useful epidemiological data for them to analyze.

Another similar movement among AIDS activists in the 1980s started out with considerable conflict and distrust between patients and the scientific-medical community. Patients objected to restrictions on the distribution of experimental medicines and the design of clinical trials. But the struggle eventually died down as the leaders of patient organizations were invited to advise scientists and physicians on a more humane organization of research (Epstein, 1996). This lay intervention added a new ethical dimension to scientific practices that were not well conceived from the standpoint of current values. The changes were also cognitively significant since they made it easier to recruit human subjects and to insure that they cooperated in supplying researchers with the desired information.

These are American examples but other cases and other institutional procedures in other countries confirm the general pattern: from indifference to paternalism to signs of democratic engagement between science and society. If this trend develops widely, it promises to make a lasting contribution to democracy in technologically advanced societies.

Technology and Society

I have left an ambiguity in the above history. My examples included a weapon, a toxic waste site, and a disease. Scientists were involved in all three. But is “science” the right word to describe their activities in these three cases? Clearly, the making of a bomb differs from the work of the physicists whose theories made it possible. In fact the making of the bomb involved many industrial crafts and aimed primarily at producing a weapon, not better understanding of nature. The other cases are similar. Chemists and microbiologists were involved (and still are in the case of AIDS). But their activities were organized by an elaborate industrial apparatus to produce goods, not to contribute to our understanding of nature, although they will certainly do that too.

In my view it is a mistake to focus exclusively on the relationship between science and society in discussing cases such as these. That approach tends to place the emphasis on the cognitive aspect of the relationship. But when science leaves the laboratory and enters society as technology, it must serve many other interests besides the interest in knowledge. As we have seen, technology is a field of activity in its own right. It does not cancel out of the equation as a mere application of science. Industrial organizations intervene between the work of scientists and their technoscientific products. These organizations are independent mediations with their own logic and procedures. For reasons I will explore in this section, technical creation is far less protected from lay intervention than is science in its cognitive role.

In those fields properly described as technosciences the situation is complicated by the ambiguity of the various activities involved in research and commercialization. When the actors seek more autonomy, they claim to be doing science; when they seek financial support they claim to be engaged in technology. Jessika Kammen describes an interesting case where researchers working on a contraceptive vaccine attempted to offload all the difficulties onto complementary “technologies” while reserving the title of “science” for their work. The distinction enabled them to continue pursuing the vaccine without worrying about the practical obstacles to its actual deployment (Kammen, 2003). Here the distinctions we are working with become political resources, but this should not blind us to what is really at stake, namely, the welfare of millions of women and their families.

The reason for the difference between the role of the public in science and technology is simple. While scientific theories are abstractions and experiments confined to the lab, technologies supply environments within which ordinary people live. Experience with these environments is a potential source of knowledge as we have seen, and everyday attitudes toward risk and benefit prevail there. All this distinguishes lay publics from scientists and technologists whose knowledge is formalized and who evaluate risks and benefits with mathematical tools.

Bridgman simply dismissed the public as “stupid,” but this is no longer possible. All too often lay observers have turned out to be the canaries in the mine, alerting scientists to overlooked dangers. And scientific and technical disciplines contain many traditional elements introduced during an earlier state of the society and its culture. In the case of technology the persistence of these elements past their time sometimes causes harm and motivates challenges from below that bring the tradition up to date.

Consider the huge variations in obstetrics from one time and place to another. Not so long ago husbands paced back and forth in waiting rooms while their wives gave birth under anesthesia. Today husbands are invited into labor and delivery rooms and women encouraged to rely less on anesthetics. The result of scientific discoveries? Hardly. But in both cases the system is medically prescribed and the feminist and natural childbirth movements of the 1970s that brought about the change forgotten. A technological unconscious covers over the interaction between reason and experience.

There is a further distinction between the relation of science and technology to society. Even when they employ scientists and scientific knowledge, corporations and government agencies should not enjoy the relative autonomy of science. Their products give rise to controversy not about ideas but about potential harm. Those in the best position to know are usually associated with the very organizations responsible for the problems. But these organizations cannot be trusted to tell the truth or to act on it. Of course many corporations and agencies are honest, have the public welfare at heart and act accordingly, but it would be imprudent to generalize from such instances to the conclusion that vigilance and regulation is unnecessary.

The dominant feature of this relationship is the potential for conflict of interest. Familiar examples are the manipulation of information and the manufacture of artificial controversy by the tobacco industry with respect to lung cancer and energy companies with respect to climate change (Michaels, 2008). Conflicts of interest in such cases give rise to political struggles over regulation and, unlike scientific controversies, we do hope democratic procedures will decide the outcome rather than a “core set” of actors, namely, the corporations and agencies involved.

There is thus an enormous strategic difference between the science-society and the technology-society relationships. No matter how extensive the many interdependencies of scientific research and technology, no matter how blurred the boundary between them may sometimes be, there remains a fundamental difference with real consequences. In the case of scientific research we seek public interaction and mutual engagement but leave scientists to draw their own conclusions. We may suspect particular scientists of incompetence or chicanery and ask for second opinions, but in the end we must rely on the scientific community. We do not have a similar confidence in corporations and governments. When they order up “truths” on command the results are disastrous. Nothing has changed in this respect from Lysenko to HIV denial in South Africa.

As public institutions corporations and government agencies, including those that employ scientists, must submit to democratic control of their activities. That control is often extensive and detailed and needs to be where their products circulate widely with significant public impacts. Thus we do not want an oil company rather than scientists to decide if climate change is real, but we are not worried when the government orders a medicine off the market or bans a pesticide. Such decisions are a normal exercise of governmental authority and easily implemented by technical workers because, as noted above, so many viable alternatives are generally available.

The danger in confusing the cases is that when we demand democratic intervention into “technoscience,” we will be understood to blur the line between cognitive and regulatory issues. Unless we keep these issues clearly separate we will appear to be irrationalists rejecting science when in fact we need it precisely in order to control the activities of technological actors such as corporations.

The Entangled Hierarchy of Technology and Society

I want to conclude this talk by considering the paradoxical structure of the relationship between technology and society. The paradox tells us something important about what it means to be a human being in a technological society.

While ordinary people often play an important role in alerting scientists to problems and sometimes in collecting data as well, for them the new relationship is not primarily about knowledge but rather about experience. It concerns how people understand the world in which they live, the lived world of everyday experience.

The inhabitants of the Love Canal neighborhood recognized a new element in their world, a toxic element boiling up from the waste dump on which their houses were situated. Their experienced world turned out to be more complicated than they had realized. This discovery about the world was also a self-discovery: these neighbors suddenly became actors in new relationships to scientists, doctors, the government and the corporate author of their misfortune. Understanding of the world, identity and group formation go hand in hand. All are fluid in modern societies and intertwined with technology.

This is the result of a historic change. In traditional societies the specialized knowledge of craft workers and the lessons of everyday experience shared by all members of society merged together in a tradition passed down through the generations. Social identities were also stable since disruptions of the sort occasioned by rapid technological change were rare. But as capitalism develops, control of design is restricted to a small dominant class and its technical servants. They are not restrained by the lessons of experience and accelerate technological change to the point where society is in constant turmoil.

This change has consequences for the structure of knowledge. Scientific and technical disciplines are freed to become specialized formal systems. It is in this context that the idea of a pure rationality that would be independent of experience arises. Although expressed in secular forms, this idea is essentially theological. One imagines a hypothetical infinite being capable of acting on His objects without reciprocity. God is at the top of the ultimate practical hierarchy in a one way relation to His objects, not involved with things and exposed to their independent power. He creates the world without suffering any recoil, side effects, or blowback. He has nothing like what we call experience.

Modern thought takes this imaginary relation as the model of rationality and objectivity, the point at which humanity transcends itself in pure theory. But in reality we are not gods. Human beings can only act on a system to which they themselves belong. This is the practical significance of embodiment and implies participation in a world of meanings and causal powers we do not control. Finitude shows up as the reciprocity of action and reaction. Every one of our interventions returns to us in some form as feedback from our objects. This is obvious in everyday communication where anger usually evokes anger, kindness kindness, and so on.

The technical subject is finite too, but the reciprocity of finite action is dissipated or deferred in such a way as to create a necessary illusion of transcendence. We call an action "technical" when the actor's impact on the object is out of all proportion to the return feedback affecting the actor. We hammer in nails, transforming a stack of lumber into a table, but we are not transformed. All we experience is a little fatigue. This typical instance of technical action is narrowly framed to highlight the apparent independence of actor from object. In the larger scheme of things, the actor is in fact at stake in his action if not in the same way as the lumber. His action has an impact on his identity: he becomes a carpenter or at the very least a hobbyist. But that impact is not visible in the immediate technical situation where big changes occur in the wood while it seems that the person wielding the hammer is unaffected.

This example may seem trivial but from a systems point of view there is no difference of principle between making a table and making an atom bomb. When J. Robert Oppenheimer exploded the first bomb at the Trinity test site, a passage from the Bhagavad-Gita flashed through his mind: "I am become death, the shatterer of worlds." In this case the similarity between technical labor and divine action is all too clear. Technique appears to represent a partial escape from the human condition. But it did not take Oppenheimer long to realize that the destroyer is also exposed to destruction, and to call for international control of nuclear weapons. Unlike Oppenheimer, Shiva, the god of Death, did not have to worry about the Russians.

Without wishing to return to traditional arrangements, we can nevertheless appreciate their wisdom, based as they were on a longer term view of the wider context of technology than we are accustomed to today. Tradition was overthrown in modern times and society exposed to the full consequences of rapid and unrestrained technical advance, with both good and bad results. The good results were celebrated as progress, while the unintended and undesirable consequences of technology were ignored as long as it was possible to isolate and suppress the victims and their complaints. The dissipated and deferred feedback from technical activity, such unfortunate side effects as pollution and the deskilling of industrial work, were dismissed as the price of progress. The illusion of technique became the dominant ideology.

These side effects are consequences of a technology cut off to a considerable extent from the experience of those who live with it and use it. As it grows more powerful and pervasive, it has become more and more difficult to insulate technology from feedback from the underlying population. The experience of users and victims of technology eventually influences the technical codes that preside over design. Early examples emerge in the labor movement around health and safety at work. Later, such issues as food safety and environmental pollution signal the widening circle of affected publics. Today, as we have seen, such interactions are becoming routine and new groups emerge frequently as “worlds” change in response to technological change.

In the technology studies literature, this is called the “co-construction” of technology and society. The examples cited here show this “co-construction” resulting in ever tighter feedback loops, like the “Drawing Hands” in M. C. Escher’s famous print of that name. I want to use this image to discuss the underlying structure of the technology-society relationship.

Escher's self-drawing hands are emblematic of the concept of the "strange loop" or "entangled hierarchy" introduced by Douglas Hofstadter in his book Gödel, Escher, Bach (1979). The strange loop arises when moving up or down a logical hierarchy leads paradoxically back to the starting point. A logical hierarchy in this sense can include a relationship between actors and their objects, such as seeing and being seen or talking and listening. The active side stands at the top and the passive side at the bottom of these hierarchies.

With this in mind the famous liar's paradox can be analyzed as an example of a strange loop in which top and bottom trade places. Like all statements, the statement "This sentence is false" refers to an object. The statement itself is the actor at the top of the hierarchy. But the object to which it refers is also itself and in describing itself as false it reverses the direction of action. When one claims that something is false that claim is the actor and what it describes as false is the object. But that object is itself. Now the sentence is only true if it is false and false if it is true. A strange loop indeed!

In the Escher print, the paradox is illustrated in a visible form. The hierarchy of "drawing subject" and "drawn object" is "entangled" by the fact that each hand plays both functions with respect to the other. If we say the hand on the right is at the top of the hierarchy, drawing the hand on the left, we come up against the fact that the hand on the left draws the hand on the right and so is also located at the top level. Thus neither hand is at the top or both are, which is contradictory.

On Hofstadter's terms, the relation between technology and society is an entangled hierarchy. Social groups form around the technologies that mediate their relations, make possible their common identity and shape their experience. We all belong to many such groups. Some are defined social categories and the salience of technology to their experience is obvious. Such is the case with factory or hospital workers, whose organization and employment depends on the technology they use. Other groups are latent, unconscious of their commonalities until disaster strikes. The inhabitants of Love Canal may have been indifferent neighbors, but when toxic waste was discovered in the land they inhabited they were alerted to a shared danger. They quickly acquired an experiential expertise that enabled them to give meaning to their plight and protest it effectively (Collins and Evans, 2002). Their world was transformed and, as a conscious collective, they recruited scientists to help them understand it and made demands on the government. Such encounters between the individuals and the technologies that bind them in groups proliferate with consequences of all sorts. Social identities and worlds emerge together and form the backbone of a modern society (Callon et al., 2001).

Once formed and conscious of their identity, technologically mediated groups influence technical design through their choices and protests. This feedback from society to technology is paradoxical. Insofar as the group is constituted by the technical links that associate its members, its status is that of the "drawn" object in Escher's scheme. But it reacts back on those links in terms of its experience, "drawing" that which draws it. Neither society nor technology, reason nor experience can be understood in isolation from each other because neither has a stable identity or form. This paradox is endemic to democracy in general. Self-rule is an entangled hierarchy.

Hofstadter's scheme has a limitation that does not apply in the case of technology. The strange loop is never more than a partial subsystem in a consistent, objectively conceived universe. Hofstadter evades ultimate paradox by positing an "inviolate level" of strictly hierarchical relations above the strange loop that makes it possible. He calls this level "inviolate" because it is not logically entangled with the entangled hierarchy it creates. The person who says "This sentence is false" is not entangled in the paradox she announces. In the case of the Escher drawing, the paradox only exists because of the unparadoxical activity of the actual printmaker Escher who drew it in the ordinary way without himself being drawn by anyone.

But there is no equivalent of this "Escher" in the real world of co-construction, no inviolate god creating technology and society from the outside. All the creative activity takes place in a world that is itself created by that activity. Only in our fantasies do we transcend the strange loops of technology and experience. In the real world there is no escape from the logic of finitude.

References

Bridgman, P.W. (1948). “Scientists and Social Responsibility,” in Bulletin of the Atomic Scientists, vol. 4, no. 3.

Callon, Michel, Pierre Lascoumbes, Yannick Barthe (2001). Agir dans un Monde Incertain. Paris: Seuil.

Collins, H. M. and Robert Evans, “The Third Wave of Science Studies: Studies of Expertise and Experience,” Social Studies of Science 32/2(April 2002) 235–296.

Epstein, Steven (1996). Impure Science. Berkeley, University of California Press.

Feenberg, Andrew (1999). Questioning Technology. New York: Routledge.

Hofstadter, Douglas (1979). Gödel, Escher, Bach. New York: Basic Books.

Kammen, Jessika (2003). "Who Represents the Users? Critical Encounters between Women's Health Advocates and Scientists in Contraceptive R&D," in N. Oudshoorn and R. Pinch, eds., How Users Matter: The Co-Construction of Users

and Technology, Cambridge, Mass.:MIT Press, pp. 151-171.

Michaels, David (2008). Doubt Is Their Product: How Industry’s Assault on Science Threatens Your Health. Oxford: Oxford University Press.

Radder, Hans (2009). Handbook Philosophy of Technology and Engineering Sciences, ed. A.Meijers,. Amsterdam: Elsevier, pp. 71-87.

.Simon, Bart (2002). Undead Science: Science Studies and the Afterlife of Cold Fusion. New Brunswick: Rutgers University Press.