Fred Eidlin ..... from the "friendly-hostile cooperation of many scientists". (Popper, 1965b: .... "[W]e cannot foresee," Popper writes, "which parts of our scientific ...
SOCIAL SCIENCE, SOCIAL ENGINEERING, AND PUBLIC POLICY
Fred Eidlin Department of Political Studies University of Guelph and Richard Appelbaum Department of Sociology University of California at Santa Barbara
ABSTRACT This paper attempts to take stock of problems bound up with the application of reason and social scientific knowledge to public problems to ask what role reason and social science can and should play in the formulation and execution of public policy, and to inquire into the nature and sources of disillusionment with rational debate and social science in contemporary society.
Paper prepared for presentation at the 4th Conference on Public Policy and Administrative Studies, University of Guelph, 10-11 April 1987.
Electronic copy available at: http://ssrn.com/abstract=1884399
2 INTRODUCTION: Santa Barbara, California, is a community which prides itself on its rational (and pluralist) approach to policy issues of general concern. No decisions of consequence are made by the City or its agencies without a lengthy process involving expert studies, public hearings, and extensive public debate. The incumbents with whom we are concerned in the present paper (council, mayor) for the most part began their public careers as planning commissioners or administrators, and have repeatedly shown a willingness to subordinate particular interests (for example, those of businessmen, developers, or neighborhood groups, to what they believe serves the common interest of the community as a whole. In keeping with this good government approach--and consistent with State policies governing local planning--the City relies heavily on rational analysis and debate before acting. A few years ago, City Council was involved in a year-long debate concerning the use of public funds to attract one or more major department stores in what was to be the City's largest redevelopment project. Some expert studies argued that the expenditure was necessary to assure downtown retail health, and that it would more than pay for itself through its long-term beneficial effect on the tax base; others argued quite to the contrary that the expenditure represented a wasteful and unnecessary use of public funds that would never be repaid. The issues were complex and technical--so typical of debates about public policy--involving assumptions about future regional income and population growth, leakage in retail spending to other areas, the growth trajectory for sales in the proposed department store(s), spillover effects on neighboring businesses, and the future return on alternative investments. How is a policy analyst to weigh such indeterminate factors in his/her calculus? Growth trends in local retail spending depend first on local population and income growth, themselves the product of inter-regional job shifts, national and regional economic growth, and politically-determined local growth control policies. All of these factors are outside the purview of local retail economic analysis, reflecting national and regional economic trends which from a purely local standpoint must be regarded as indeterminate. Moreover, many of these factors may be regarded as indeterminate in principle as well. This is partly due to theoretical insufficiency, and partly to the political character of many seemingly economic forces. Concerning the first, economic theory presently does not permit accurate long-range forecasting; such basic factors as interest rates, monetary growth, inflation, consumer spending, and productivity cannot be foretold with any accuracy. All of these factors will influence local patterns of
Electronic copy available at: http://ssrn.com/abstract=1884399
3 retail spending. Concerning the second, insofar as the local retail market is driven by growth in local population and income, political decisions made in Washington or locally are crucial. Will the Space Shuttle program at neighboring Vandenberg Air Force Base be expanded, with a major effect on local jobs and spending? Will a military contract be given one of the local electronics firms, since the U.S. President now resides in the Congressional District? Will water be imported from the California State Water Project, permitting substantially increased local growth? Will local growth controls be strengthened or weakened?). These larger (i.e., non-local and/or non-economic) issues are not trivial for the policy analyst attempting to conduct a rational cost-benefit study of the proposed retail expansion project for City Redevelopment officials. On the contrary, future sales projections--on which all evaluations of net fiscal impact necessarily turn--require answers to just such questions. Yet it is the inability to satisfactorily address these issues that partially explains the deep disagreement among experts, all of whom espouse contending versions of the truth. 1 Throughout history, man has always tried to use his reason and knowledge, if not to better his condition, at least to provide some kind of guide to action. Many different things have been meant by the words "knowledge" and "reason," and there has been profound disagreement about what role knowledge and reason should play in human affairs. For some who have believed, for example, that their knowledge came from God through revelation, the proper use of reason has been to discover and understand God's will so that knowledge of it could serve to guide to action. For others, who have believed in laws of nature and/or laws of history that govern the affairs of men, but which are themselves outside human control, the task has been to discover these laws to understand what is happening and to predict the future, even though nothing can be done to change what is destined to happen. More interesting and obviously more relevant for students of politics and public policy are views that consider reason and knowledge as possibly influencing, even determining the course of human affairs. There would, after all, be little if any room for politically determined change and social engineering if human affairs were wholly or even mainly determined by something like an "invisible hand of the market" or "inexorable laws of history." Even traditional statecraft deals with the problem of how reason and knowledge can effect change. The modern era has been characterized by bold optimism regarding possibilities for scientific understanding of society and for improvement of society through policies informed by rational principles and social scientific knowledge. An unstated but essential assumption underlying most public policy research, indeed perhaps most social science, is that the knowledge gained
4 through such research can better the human condition--liberating people from bondage to false beliefs which keep them from realizing a better life, diagnosing social ills, and discovering ways of changing society for the better. Social science, like science more generally (and liberalism and socialism, among other things), are products of the Enlightenment, and central to the Enlightenment tradition has been a rejection the view that man must simply accept the conditions into which he is born. It is a central belief and ideal of the Enlightenment that knowledge and reason can be not only an instrument of liberation breaking the bonds restrain man, but also a means for bringing about beneficial changes in nature and society. And yet, there has, in recent years, been a growing skepticism about the possibility of knowing and understanding society, and basing action aiming at social change on rational discourse and social scientific knowledge. Gone is the boundless optimism of many of the 19th century founders of social science and even of many of the 20th century proponents of social reform based on scientific political science and scientific public administration that social science and social engineering can cure all the ills of society. The grandiose hopes and expectations of social science and social engineering have been dealt hard blows by the experiences of advanced industrial societies in this century. Few contemporary marxists still believe in inexorable laws of history and not many free-market liberal theorists can be found nowadays who believe seriously in an "invisible hand" of the market which can be relied upon to produce public good automatically. Furthermore, it hardly needs pointing out how little faith there is today in the possibilities for rational planning and social engineering. Quite the contrary, (and not without good reason) there is broad awareness of the horrors and inefficiencies of "rational" planning. Despite widespread agreement that contemporary society remains deeply flawed and could much stand improvement, there does not exist any theory which would be generally accepted as a basis for wide-scale planning and social engineering. In fact, despite the reasonable amount of respect enjoyed by some kinds of "expert" social scientists, anyone arguing seriously that social or policy science can help bring about the "good society" would most likely be regarded as a naive utopian. The present paper attempts to inquire into the sources of this disillusionment, to take stock of problems bound up with the application of reason and social scientific knowledge to public problems, and to ask what role reason can and should play in the formulation and execution of public policy. The first part of the paper explores four categories of problems inherent in the use of reason, as applied to public policy--problems of knowledge, problems of rational discourse, ethical problems, and practical problems. The policy analyst
5 studies problems in light of prevailing theory (economic, political, sociological), formulates alternative courses of action, and attempts to spell out the consequences of each. In theory this appears straightforward. In practice, however, the rational approach comes up against limitations which call into question the use of reason itself. The response--on the part of critics of a particular line of policy--is often intellectual luddism or anti-rationalism, calling for the rejection of reason and reliance instead on deeper intuitions, prejudices, or past practices. Our aim in this section is to bring out the full force of the case against reason. We shall illustrate our arguments with reference to certain policy issues with which we have had concrete experience as urban policy analysts. In principle, however, the arguments apply to any level of public policy formulation. The remainder of our discussion deals with the question of what role rational debate and social science can and should play in the formulation and execution of public policy. We will try to defend the ideal of a theory-guided approach to public policy that takes account of the kinds of problems identified up to this point, while avoiding the pessimistic, anti-science, antirationalist, anti-theoretical conclusions often drawn in response to these problems. I. LIMITS INHERENT IN SOCIAL SCIENTIFIC KNOWLEDGE: A. Partial Knowledge about Open Systems: Society is complex and interdependent, with new, unforeseen, unrecognized variables always coming into play. Our information is always incomplete, bad, or just unavailable. It is difficult, if not impossible to control the settings we study. Controlled experiments are rarely possible. Information is affected by such factors as being gathered by bureaucratic agencies who gather it for non-scientific purposes (including their own systemmaintenance--covering their asses). Information is often gathered under severe constraints of time and money. Most significantly, perhaps, we never know whether we have isolated the relevant variables, or whether problems in our knowledge are attributable to flaws in our information or to changes in the system. Social scientists now recognize that social systems are complex and interdependent. Systems theory, from Parsons to Wallerstein, has sought to demonstrate that particular objects of analysis (the polity, the nation-state, the city) are always interconnected in larger webs of relationships. Thus, for example, political economists such as Wallerstein, Frank, and Cardoso argue that one cannot understand the operation of national economies without locating them in relation to a world economy with particular local manifestations. The proper unit of
6 analysis, according to this line of reasoning, is the most inclusive totality that can be understood as relatively closed from the standpoint of the central problems that are of scientific concern. While the various levels of a system may be analytically distinct, they can be grasped only as part of this larger totality. It is this fundamental insight that unites such otherwise divergent approaches as functionalism (from Comte and Durkheim to Parsons and Buckley), structural Marxism (Althusser, Poulantzsas), and even neo-Hegelian philosophy (Habermas, Schroyer). To take another example, such urban theorists as Castells have gone so far as to argue that there is no such thing as urban theory--that cities are properly conceptualized as a reflecting the national (or international) interplay of labor and capital, and cannot be analyzed in isolation from such flows. It is not our purpose here to evaluate these very diverse theories, but simply to draw attention to their common insight: that social processes are best theorized in terms of systematic relationships that are both vertical (e.g., world economy, national economy, local economy) and horizontal (the United States and Japan, New York and Washington). From the viewpoint of the policy analyst, the problem can be stated as follows: the unit of concern, from a policy standpoint, is in fact an open system with theoretically significant connections to other systems (vertical, horizontal) whose influence on the course of events may prove decisive, yet whose operation is beyond the purview of the analysis in question. This restriction arises from two sources, a practical and a theoretical one. In practice, (as will be discussed further below) the analyst usually lacks the time, resources, and mandate to take into account the necessary systemic relationships. Even were this not so, however, from a theoretical standpoint our knowledge of social/economic/political systems remains sufficiently limited that the necessary relationships can seldom be adequately specified. This deficiency is only partly to the current poverty of social theory. It also results from the fact that many things which decisively impact a particular governmental/administrative unit are the result of decisions taken elsewhere--decisions which could not in principle have been foreseen by local policy-makers, whatever their degree of theoretical acuity. Let us illustrate these points with a concrete example from our experience in public policy analysis.
B. Reflexivity and the Anti-political Character of Social Scientific Knowledge: Unlike our knowledge of nature, knowledge of social laws can result in changing the conditions which give rise to those laws, and hence the laws themselves. By adopting deterministic models,
7 social science may simply be the ideological reflex of a society in which human choice is restricted within the narrowest possible bounds. Prediction often confuses trend with destiny, (Daly, 1976: 9) becoming self-fulfilling prophecy. It is thus inherently reificatory. In fact, politics often makes reality what has previously been dogmatically declared to be impossible. Seeing public policy problems as technical rather than as political tends to take public problems out of the realm of political discourse and to turn them over to the "experts," who see them as informational and technical rather than as political. Attention is drawn away from root causes and turned towards that which can be dealt with; narrow technicism may result. Focusing on the "right knowledge" (the "best studies" etc.) diverts attention from the imbalances in power that are often the root causes of many social problems. The natural sciences, as positivistically constructed, tend to identify explanation with prediction: to accurately account for the future behavior of a system is to be able to derive a consequence of the present system state which is subsequently confirmed as true. The social sciences, especially the hard (i.e., quantifiable) policy sciences, by-and-large adhere to such a model of reality. This model is, of course, often inaccurate for the kinds of reasons just noted--partial knowledge about open systems, and changability and plasticity of social regularities. Such limitations, however, seldom discourage policy-analysts from making medium and long range (10-20 year) predictions. Decisions made on the basis of such forecasted trends may subsequently confirm those trends, thus becoming self-fulfilling prophecies. When the Redevelopment Agencies invests public funds to support private retail expansion, under the expectation that future economic growth will sustain that expansion, a range of subsequent actions are likely to follow to secure that very outcome. Initially, the very act of spending large sums on a redevelopment project may itself encourage ancillary growth and in general help contribute to the conditions for its success. Should the initial prediction prove inaccurate, however, actions will quickly be taken to secure the initial investment. It may be decided, for example, that the original expenditure was simply insufficient to trigger the necessary growth in shopping, requiring additional expenditures on (e.g.) other stores in order to really make a difference and achieve the predicted result. Insofar as policy studies are blueprints for future action, their hidden assumptions help to shape the very future they forecast. Prediction always assumes ceterus world in which behavior is a known relatively unchanging. This entails a of social change. Insofar as human delimited bounds, predictions are more
paribus; it presupposes a quantity, measurable and highly conservative model choice is restricted to likely to be realized. It
8 follows that accurate forecasting flourishes to the extent that political discourse is limited, with debate and choice supplanted by purely technical analysis. We shall have more to say on this topic in the following section. The problem of predictability is related to that of reflexivity: people, unlike atoms and molecules, can talk back. People are capable of reflecting on their conditions, analyzing the laws which appear to govern them, and then acting in such a fashion as to modify those conditions that gave rise to the laws in the first place. The ability to do so is, of course, highly variable in fact, depending on a host of particular circumstances. We merely note what we regard to be two salient features of such reflexivity. First, to act reflexively presupposes a theoretically adequate understanding of the processes in question on the part of some strategically located group(s). Second, understanding must be linked with a willingness to act. Informed, concerted action is capable of altering the conditions which give rise to theory, thereby altering the theory itself. To return to our example, numerous expert studies had indicated that a significant shift was occurring in Santa Barbara, away from the downtown core and towards peripheral locations outside the City limits. Expert economic consultants, drawing upon local market research as well as general retail trade theory, concluded that regional demand was sufficient to sustain a large regional shopping mall--one which, they believed, would in all likelihood locate in the suburbs, where land was both available and inexpensive. This, in turn, would further hasten the decline of the downtown, eroding the City's tax base and contributing to a downward spiral of revenues, services, and blight. These conditions were largely regarded as inevitable, given shoppers' presumed preference for large malls, and the inability of the market--operating freely--to provide a large, cheap parcel of land downtown to potential mall developers. To alter this undesired scenario, local merchants and created a parking district, with powers of condemnation and taxation. This district, in turn, assembled land for centrally located parking facilities, using powers of eminent domain. Inexpensive parking (free for 90 minutes), accessible to downtown stores, did much to forestall the anticipated economic decline. Additionally, the City created a redevelopment Agency, with substantial powers to acquire parcels of private land, to be leased to large mall developers at prices competitive with more suburban locations, using property tax revenues. By these selfconscious actions it was hoped that a regional shopping center could be attracted to the City itself, rather than the periphery, thereby altering the conditions which otherwise would result in downtown economic decline. To re-emphasize the central point, an understanding of retail economics permitted local businessmen and policy-makers to intervene in the operation of the marketplace, in a concerted effort to circumvent the predicted outcomes.
9 C. Knowledge and interest: In order for there to be a genuine social science or policy science, one would have to be able to conceive of some objective reality (such as a "public interest" or "public good"), independent of the interests, values, and other biases of social/policy scientists and those affected by policies based on such knowledge. But does such an objective reality actually exist? What is observed and what are judged to count as "the facts" always depends on interests, values, and other biases. In other words, there seems to be an unavoidable tension between "truth" and "interest" (i.e. the "right" policy in an objective sense versus the fact that we are faced with different versions of the "right" policy depending on one's interests). For example, is the "problem of poverty" to be described as a syndrome of lazy, inferior people, or a syndrome of unjust systemic structures? Is the "black problem" a problem to be described in terms of characteristics of black Americans or in terms of attitudes of white Americans?) How can there be a science without the possibility of objective descriptions? Some would want to treat this as an epistemological problem-to argue that since there is no knowledge apart from interest, there is no such thing as uninterested "truth" or "rightness." This is why we are introducing the problem of knowledge and interest in this section even though we do not regard it as an epistemological problem. We recognize that all knowledge, in a sense, "depends" on the knower's framework. We agree with Kuhn, Popper, and many others who have argued that there is no such thing as unprejudiced inquiry, even in the natural sciences. As Popper puts it: Everyone who has an inkling of the history of the natural sciences is aware of the passionate tenacity which characterizes many of its quarrels. No amount of political partiality can influence political theories more strongly than the partiality shown by some natural scientists in favour of their intellectual offspring. If scientific objectivity were founded . . . upon the individual scientist's impartiality or objectivity, then we should have to say good-bye to it. (Popper, 1965b: 217) Clearly, the views of participants in public debates can be influenced--often very strongly--by their personal and social situations, which include their interests among other things. It is not surprising that blacks, women, and jews, for example, tend to be more aware of problems of discrimination in society than white male protestants tend to be. The experiences and angles on the world provided by their social situations may literally cause them to see facts which people in other social situations may not
10 see. It does not follow from this, however, that the truth and justice of arguments made in public debates depend on whether or not people understand or believe them. White male protestants who have never experienced discrimination personally but who are willing to listen to argument are quite capable of recognizing truth and justice in the arguments of people who have experienced it. They may even be rationally persuaded to do something against their interests (e.g. be persuaded by wife or girlfriend to spend their valuable free time doing unpleasant housework). People are sometimes quite capable of recognizing some proposal as objectively good and yet opposing it on grounds of self-interest. Some professional negotiators and diplomats, for example, may come to understand very well the justice of the opposing side's position, as well as public interest arguments against the positions they are defending in negotiations. Their personal interest will, of course, be to defend the narrow interests of their employer to the best of their abilities. Since most people don't like to think of themselves as hypocrites or as doing the wrong thing, many negotiators and diplomats naturally tend to rationalize the positions they must defend as the right ones. But none of the psychological and sociological arguments of which we are aware, stressing the great difficulties people often have in recognizing as true and just claims which conflict with their interests establishes that truth and justice are inextricably bound up with interest in an epistemological sense. This denial of epistemological status to the problem of knowledge and interest does not, of course, solve the problem. It merely shifts it over into the next three categories of problems to be discussed. Nevertheless, we believe it essential to the ideal of rational debate over public policy for notions such as objective truth and justice to be preserved, at least as regulative principles, even though we readily concede that it will often be very difficult to reach agreement as to what, in fact, is true and just in matters of public policy and that, in many cases, even where there is agreement, it doesn't necessarily mean that the true and just has been found. II. PROBLEMS OF RATIONAL DISCOURSE: Even where planners or social engineers manage to deal satisfactorily with the kinds of problems inherent in social scientific knowledge, this is only the beginning of their troubles. It isn't sufficient merely to have such knowledge. Others involved in the policy process (from ordinary citizens to the political leaders who will bear responsibility for the policies based on this knowledge) have to be convinced that this social scientific knowledge is sound, and that public policy should be based on it. And, as everyone familiar with public policy debates knows very well, the good argument, theory, policy,
11 etc., often loses, and the bad often wins. In other words, it is not enough for knowledge of society to be objectively true, and policy to be objectively good in order for it to prevail in policy debates. In this section we address some of the reasons for this. What is usually regarded as the "scientific" model of policy making assumes an ideal setting--one in which discourse can take place free of all institutional constraints and distortions. Habermas (1970, 1975) has characterized this kind of situation as one of unconstrained discourse, and has identified the conditions he believes essential to the achievement of communicative competence. Popper extols the scientific objectivity resulting from the "friendly-hostile cooperation of many scientists" (Popper, 1965b: 217) and, as we have noted above, stresses that fruitful discourse is possible among scientists having vastly different frameworks (Popper, 1974). This idealized situation is, of course, seldom realized fully even in the natural sciences. Thomas Kuhn (1970), among others, has shown that since scientific paradigms are like different world views, debate across paradigms is extremely difficult if it is possible at all. In Kuhn's "normal science," free and unconstrained discourse has ceased. Popper also stresses that serious critical discussions are always difficult. Normative human elements always enter. Many participants in a rational, that is, a critical discussion, find it particularly difficult that they have to unlearn what everybody is taught in a debating society, for they have to learn that victory in a debate is nothing, while even the slightest clarification of one's problem, even the smallest contribution made towards a clearer understanding of one's own position or that of one's opponent is a great success. In the advanced sciences, however, the fact that political, sociological, and psychological factors may affect the practice of scientific inquiry, should not obscure the fact that objectivity and unconstrained discourse remain ideals. And these ideals are at least partly secured through the procedures and ethics prevailing in scientific communities. In the policy sciences, however, a very different set of conditions prevail. There are many different kinds of constraints on discourse, some practical and some more fundamental, that often prevent debate from running its full course. A. Truth and rhetoric: This is an old and perennial problem of politics. The philosopher and the scientist are engaged in a quest for the truth about the universe--including the social and political universe.
12 The political community partly shares this aim and partly is deeply hostile towards it. On the one hand, objective knowledge and rational argument plays an important role in politics everywhere. Even Hitler and Stalin, for example, were interested in knowledge and rational argument bearing on at least some questions of what (in an objective sense) might be the best policies for them to pursue. On the other hand, the political community does not always value knowledge of what is true and just very highly. Sometimes, as Socrates, Jesus, and others who have inquired into such matters and tried to instruct their respective communities about them have discovered, it may meet with incomprehension and generate intense anger. Says Plato's Socrates: . . . if I am ever dragged into court and exposed to any of these risks, it will be, as you say, some vicious fellow that brings me there (for no honest man could ever so deal with the innocent); and, indeed, it would not be surprising if I were put to death. . . . Being unwilling to follow your advice as to the employment of rhetorical tricks, I shall have nothing to say in the court room. . . . I shall be like a physician tried before a jury of children on the accusation of a cook. (Gorgias: 521) B. Problem of competence and mystification: In the Athenian version of democracy (by the time of the Constitution of Solon), every citizen was assumed to be competent to make judgments about all matters of policy facing the city. And this assumption remains part of the contemporary ideal of democracy, despite the replacement of direct democracy with representative democracy. Given the size and complexity of modern society and the institutions of the modern state and the amount of specialized knowledge often required to understand many of the problems of contemporary society, how is possible the have any kind of rational consultation with the entire citizenry? Public opinion (however it may be read) is literally ignorant even of the existence of most of the issues with which public policy must deal, and especially about the complex interrelationships of various public problems and policies. Even legislators and members of the political executive are nowadays severely limited in their possibilities even for keeping track of the policies which emerge from the bureaucracies for which they are responsible, let alone understanding their implications in terms of the public interest. C. Rational discourse and crisis situations: Many people are quite willing to pay attention to evidence
13 and listen to argument as long the situation is conducive to this. When under pressure, or in situations of danger; when they become angry, anxious, and frightened, they fall back onto other modes of behavior. The 20 knots problem: When the wind is blowing 5 knots, I let you take the mainsail; when it is blowing 10, I let you take the jib; when it is blowing 20, I sail the boat. When stakes are high and pressures to act great, willingness to continue with rational debate decreases markedly. For the guerilla leader in El Salvador or Nicaragua, slogans like "you can't make an omelette without first breaking eggs" are likely to be more compelling than "I may be wrong and you may be right and by a common effort we may get nearer to the truth." This line of argument, of course, has no epistemological significance whatsoever. The interests of the guerilla leader and the ideas he is fighting for may, in fact, be best served by stubborn adherence to the rationalist attitude of "I may be wrong and you may be right," even in the face of high stakes and great pressures. The revolutionary's argument that he is too busy changing the world to worry about debating the theory on which his actions are based (an argument which, incidentally, can also be heard from many "pragmatic" policy analysts) is nonsense from an intellectual point of view, since the program and gut-level theories on which his actions are based may be mistaken, and few situations are so bad that they cannot get any worse. Unless the revolutionary (or "pragmatic" policy analyst) can be sure that his actions will actually have the results anticipated, all the sacrifices required to carry out the revolution may be completely in vain or worse (Eidlin, 1983c). Openness to critical debate may not only prevent disastrous consequences; it may also increase chances of success by improvement of the theory on which action is based. But how many revolutionaries, people who have to act under conditions of danger and stress, people who are angry, frightened, and frustrated with rational debate and politics as usual are willing to listen to arguments of this kind? D. Emotional stress of rational debate: Critical discussion, because of the kinds of problems we are striving to identify in the present discussion, does not often lead to decisive conclusions. Not only does rational debate require faith, hope, courage, and persistance, but there is no guarantee of ultimate success in it. This leads many people to despair of a rational approach to public policy. Assaults on the experts may lead to radical questioning of all knowledge claims and of rationality in general. Anti-intellectualism. If my experts are wrong, then you (who criticize them, and offer counter-expertise) may prove wrong too. In the even of bickering among the experts, I will fall back on intuition, conventional
14 wisdom--meaning the prevailing ideology, which favors the status quo. The dogmatism of reason is thus replaced by the dogmatism of tradition and prejudice. III. ETHICAL CONSTRAINTS: Even where knowledge of adequate quality is available upon which policy could be based, and even where those who have to be convinced have been convinced to go ahead with the policy, this is still not enough. Questions may still remain as to whether it is a good thing that we have gotten so far. Assuming that we are benevolent social engineers, there are several ethical problems yet to be confronted. A. Whose values?
Whose interests?
We have already touched on the problem of knowledge and interest, arguing that there is such a thing as knowledge distinct from, even opposed to interest, fully recognizing, however, that it can be exceedingly difficult to persuade people to accept such knowledge when they see it as inconsistent with their interests. Nevertheless, there still remains the problem of determining where planners get the right to legislate in the interest of the collective? Once the problems of the "good of the collective," have been seriously thought through, what if the planners are faced with ignorant, unreflective, even evil opposition from all quarters of the community? If the planners take the ideal of democracy at all seriously, what right do they have to impose their version of the "good of the collective" against the will of the people they supposedly serve. We will deal below with some of the practical problems of democratic control and accountability. Our concern here has been with the problem in political ethics that has to be faced by any planner or social engineer seriously interested in acting in the public interest. B. The elements of harm and risk: "[W]e cannot foresee," Popper writes, "which parts of our scientific knowledge may come to grief one day. Thus, the belief in scientific certainty and in the authority of science is just wishful thinking: science is fallible because science is human" (Popper, 1965b: 375). In theoretical science, the fallibility of human knowledge gives little cause for concern. Since erroneous hypotheses and theories can be eliminated without anyone being harmed, the main reason for being preoccupied with the avoidance
15 of mistakes disappears. In fact, the more mistakes one makes, the better chance there is of improving one's theories. Thus, in Popper's view, in theoretical science there is nothing wrong with being wrong. "The history of science, like the history of all human ideas, is a history of irresponsible dreams, of obstinacy, and of error" (Popper, 1965c: 216). Yet, while "the task of science is the search for truth, that is for true theories," we want more than "mere truth," we want "interesting truth." (Popper, 1965c: 229). A bold conjecture which attempts to solve an interesting problem is much preferable to "any recital of true but uninteresting assertions." Such a bold conjecture is preferable even if it soon turns out to be false, for it is in this way that we can learn from our mistakes. In finding out that our conjecture was false, we learn much about the truth (Popper, 1965c: 230-231). In contrast to this, whenever a decision must be made whether or not to apply some supposed knowledge in ways that might have undesirable consequences for people's lives, the question necessarily arises of how certain one can be of this supposed knowledge. Is it certain enough to justify the risks involved? Since human knowledge is notoriously fallible, risk is inevitable. The social engineer's interventions may harm or risk harming people. Furthermore, because of the complexity and interdependence of society, all interventions are bound to have unforeseen, unintended consequences which can even result in worse problems than the one solved by the intervention. Thus, for application of knowledge to planning and social engineering, degree of certainty of the knowledge on which action is to be based may be very important. Furthermore, any decision to accept a given level of risk will necessarily have an irreducible moral component which cannot be resolved by social science. In conclusion of this section on ethical problems, we would like to make note in passing of a popular anti-rationalist argument often made by conservatives and reactionaries, which we believe rationalists cannot completely dismiss, namely that knowledge of society is potentially subversive of the social order. This kind of argument can be traced back at least as far as Plato's "noble lie." It is the view that there may some necessary untruths, which serve as the foundation of society and which, if exposed, can no longer serve their foundational purpose. This is the flip-side of the argument against mystification. It is an argument that the right kind of mystification is a good thing, even where the mystification covers up realities which most people would conisider unjust. To the extent this argument is true, it becomes dangerous to society and hence wrong to expose the truth. While on balance, we reject this kind of argument, it is not central to the basic concerns of this paper, and space precludes our giving it the attention it would require.
16 IV. PRACTICAL CONSTRAINTS: A. Constraints of time, money, and other resources: The principal practical limitations, of course, are time and money: policy studies are usually done under pressure, with deadlines that are politically motivated rather than derived from the internal logic of the research agenda. In pure scientific research the concern in principle is with "truth," and funding agencies are often willing to extend deadlines and resources if it appears that reasonable progress is being made towards this ideal. In the policy sciences, on the other hand, there is a "bottomline" emphasis on providing timely results: analysis occurs in discrete packages, according to timetables set by bureaucratic agencies and public bodies, often without reference to methodological rationality. B. Institutional Inertia and Resistance: A second practical problem lies in the institutional setting of most policy research. Institutions value their own survival over most other considerations, including the pursuit of truth in research studies they might commission. Decisionmaking within organizations is often opportunistic rather than committed to rationality, and there is a great deal of decisional inertia within organizations. Moreover, the safeguards ordinarily built into scientific research are usually absent from policy-related research. To return to our example of urban redevelopment in Santa Barbara, the agencies concerned with downtown renewal are created on the assumption that large-scale public expenditures are necessary to secure "retail vitality" in "dying" central cities. The community of renewal experts, administrators, and officials accept this assumption without question. The studies they commission assume public expenditures of certain sorts (hotels, shopping centers, convention centers, highways, upper income housing) and seek to assess the returns on different levels or mixes of such investment. Any research which questions these assumptions is correctly perceived as threatening the survival of agencies predicated on the basic assumptions. It is not surprising, therefore, that such research is dismissed as being either inaccurate or (worse) inexpert. It will go unreported in the chief journals of the field, will be largely ignored in public testimony, and will result in a loss of credibility on the part of the researcher. It is not surprising that there are few incentives to challenge the dominant frameworks in most areas of public policies. The practical considerations are simply too strong. Even where no clearcut preference exists for a particular
17 research outcome, a concern with feasibility often outweighs the desire for rationality. Public agencies are seldom interested in studies which indicate the need for changes that are unlikely to be implemented. Studies which "go to the root of the problem" are dismissed as irrelevant academic exercises. This creates a conservative bias in much policy research. Even studies which call for comprehensive reforms--rather than more fundamental changes--are likely to be shelved for want of the necessary resources for implementation. This produces a "realism" or "practical mindedness" among policy analysts which discourages the asking of probing questions, and therefore precludes obtaining results which are likely to challenge the status quo. C. Political Definition of Policy Problems: Related to our discussion in the preceding section, but more fundamental in nature, is the definition of "problem" in policy research. In pure science, problems are determined with reference to theory: they acquire significance only within theory, and are addressed with the ultimate hope of refining theory and thereby advancing knowledge. In the policy sciences, problems are given in a more narrow, political sense: sales are declining, school test scores are lower, unemployment is higher. Here the concern is not with advancing understanding of urban economics (or cognitive development, or labor economics) per se,but rather with eliminating the more practical problem. Scientific understanding is a secondary objective, and policy analysts often feel the tension between making a "contribution to knowledge" and problemsolving in the narrow sense. In other words, policy problems are given politically; their formulation grows out of a political process, and not an intellectual one. By way of illustration, rising rents in the City of Santa Barbara, along with pressures from tenants' organizations, led the City to create a Rental Housing Task Force, mandated to study the "housing problem" and recommend policy solutions. The Task Force was balanced between tenant and landlord representatives, with a "neutral" group of "citizens" to provide a buffer. The tenant members argued that real estate speculation and price-fixing were largely responsible for the rent increases. The landlord members denied the fundamental premisses implicit in this formulation of the problem. The concept of speculation was denied; rapid resale of property was instead viewed as "investment activity" producing "market liquidity." Price-fixing was viewed as impossible, since so long as two or more landlords existed competition must necessarily bring rents down to "market" levels. When the Task Force eventually issued its request for proposals, it was clear that funding was not likely to be available for studies that addressed either of these two issues. Moreover, even were such studies funded, their results would clearly be interpreted differently by the landlord and tenant members. Since it is unlikely that any
18 possible set of results would have been perceived by the landlord members as evidence of phenomena whose existence they denied, it made no sense to fund such studies. This example brings us back once again to the problem of frameworks, particularly salient in the policy sciences. As we have argued above, debate across frameworks is not impossible in principle (Popper, 1974). It may requires time, patience, and a commitment to truth, but frameworks need not be insuperable obstacles to rational discourse. In the policy sciences, however, there are few incentives to such communication. Typically, each side hires its own experts, determines a research agenda based on widely differing frameworks, addresses different problems, and comes up with different and conflicting answers. Once the research issues have been joined in political debate, lines characteristically harden as the various experts are called to give testimony in a charged and sometimes acrimonious atmosphere. The analyst who admits that "I may be wrong and you may be right and by an effort, we may get nearer to the truth"--the very definition of the rationalist attitude according to Popper (1975b: 225)--courts political defeat for his/her client. In some cases, the model is adversarial rather than scientific: like the courtroom lawyer, the analyst is constrained to "win" the argument for his/her client, regardless of the truth of the matter. This need not imply outright distortions or even selective reporting of findings, although these occur often enough: even the scrupulous analyst can be trapped by research strategy which prejudges the results by selective posing of questions. D. Fragmentation of the polity: Closely related to the problems discussed in the preceding section is the question of who today is in a position to look at the interests of the community as a whole? The fragmentation of the polity into specialized interests in contemporary North American society, represents a practical problem distinct from the more general problem of interest bias. Even where people can readily see a public interest as distinct from the special interests with which they are personally associated, it may be impractical to argue for it. The institutions of government are colonized by special interests. Even "public interest groups" tend to take a narrow, specialized view of the public interest, their positions often predetermined by fixed prejudices and programmatic approaches which are themselves not particularly open to rational debate. Furthermore, the success of "public interest groups," like that of other participants in the pluralist balancing process seems more likely to depend upon such irrational (or at least extra-rational) factors as the quality of their bureaucracies and the rhetorical and political skills of the agents they hire to promote their policies. Furthermore, in contemporary society, people are locked into jobs and social
19 situations which force them to view the public interest in a narrow, specialized way whether they want to or not. Think, for example, of the labor union lobbyist who becomes convinced that job sharing (fewer hours and less pay for union members) is the best way to solve the unemployment problem, but is prevented from representing such a "rational" view of the public interest by the logic of his situation. Not only lobbyists, but citizens, bureaucrats, and others tend to see public problems from the logic of their particular situations. How can the "God's eye" policy analyst possibly hope to argue public interest when private interest is practically dictated to his discussion partners? V. CONSEQUENCES: REASON
DISILLUSIONMENT
AND
CYNICISM
WITH
REGARD
TO
These limitations on the policy sciences are sensed, if not articulated, by many politicians and administrators. Anyone charged with responsibility for rational planning--whether for a large corporation or governmental body--has experienced the best of analyses belied by subsequent events. This produces a certain degree of skepticism among policy makers--which, in our experience, seldom appears to transfer to the experts and consultants they hire. Humility sells neither cars nor research proposals, and the myth of scientificity is little penetrated within the community of policy analysts. This myth is also pervasive throughout our culture generally. Almost all significant public actions, from local zoning changes to the conduct of war, are justified on the basis of rational analysis and technical study. Whether it be City Hall or the Oval Office, the image of charts and graphs and bound technical reports is essential to legitimating political decisions. There appear to us to be two opposing consequences of such mandatory reliance on technical expertise in our culture. Political apathy and the technicizing of politics go hand-inhand. The result is a failure to delve into the underlying causes of social problems, as political issues are transmuted into technical ones. The policy arena is left to those who can afford to hire the experts and market the results of their studies. At the same time there exists a significant mistrust of reason that lies not far below the surface of our culture--a sort of intellectual luddism that has been reinforced in recent years by four notable failures of foreign and domestic policy: the Vietnam War, the War on Poverty, Civil Rights (especially bussing), and management of the national economy. In all of these cases policy was or is based on expert analysis, and its failure has reinforced apathy with a cynical mistrust of all reason. When contending experts offer widely differing interpretations of truth, citizens as well as policy makers face two alternative
20 courses of action: to let the experts "have it out," taking as long as necessary until firm conclusions can be drawn; or to deny all experts and rely instead on one's own preconceptions and intuitions. If "you may be right and I may be wrong," then it is possible that all knowledge is wrong. In the absence of compelling reasons to the contrary, the temptation is strong to fall back on one's native beliefs and prejudices. This, also, reinforces the status quo. This discussion may seem to end on a very pessimistic note. But this is not at all our intention. We believe that in order to tackle the problems we have discussed, they first must be clearly understood. Although reason is not an infallible instrument, although social science can only provide limited assistance in the solution of public problems, it goes without saying that reason, argument and the kinds of knowledge social science research can provide can be extremely useful in the policy process. We believe that the kinds of problems we have identified, though often posing enormous difficulties, are far from intractable. If better understood, perhaps they can be better dealt with. Part of the disillusionment with reason and social science derives from the tendency to expect too much of them. Our aim in the present paper has been to establish a program of problems as a prolegomenon to examining existing solutions and hopefully improving upon them.
REFERENCES Agassi, Joseph, 1975. Reidel).
Science in Flux (Dordrecht-Holland: D.
Appelbaum, Richard, 1976. Natural for whom? Studying Growth in Santa Barbara. Environmental Comment, (February). _____, 1977. The future is made, not predicted: technocratic planners vs. public interests. Society, 14:4 (May/June), pp. 49-53. _____, 1978. Planning as Technique: Some Consequences of the Rational-Comprehensive Model, in Harvey A. Goldstein and Sara Rosenberry (eds.), The Structural Crisis of the 1970's and Beyond: The Need for a New Planning Theory (Blacksburg, VA: Division of Environmental and Urban Systems, College of Architecture and Urban Studies, Virginia Polytechnic & State University). Braybrooke, David and Charles E. Lindblom, 1973. A Strategy of Decision: Policy Evaluation as a Social Process (New York: The
21 Free Press) Dahl, Robert A., and Charles E. Lindblom, 1963. Economics, and Welfare (New York: Harper Torchbooks).
Politics,
Daly, H. E., 1976. Energy Demand Forecasting: Prediction or Planning? .Journal of the American Institute of Planners, 19:1 (February), pp. 19-27. Dye, Thomas, 1972. Understanding Public Policy (Englewood Cliffs, N.J.: Prentice-Hall). Eidlin, Fred, 1979. Soviet Studies and 'Scientific' Political Science, .Studies in Comparative Communism, 12 (2/3) (Summer/Autumn): 133-143. _____, 1983a. Popper's Fact-Standard Dualism Contra Value-Free Social Science, 64 (1) (March): 1-18. _____, 1983b. Area Studies and/or Social Science: ContextuallyLimited Generalizations Versus General Laws, in Fred Eidlin (ed.), Constitutional Democracy: Essays in Comparative Politics (Boulder, CO). _____, 1985d. "The Radical, Revolutionary Strain in Popper's Social and Political Theory," et cetera, Journal of the International Society for General Semantics, 42(3): 283-298. Habermas, Juergen, 1970. Beacon). _____, 1975.
Towards a Rational Society, (Boston:
Legitimation Crisis, (Boston: Beacon).
Jarvie, I. C., 1964. The Revolution in Anthropology, (London: Routledge and Kegan Paul). _____, 1979. The Notion of a Social Science, in H. K. Betz (ed.), .Recent Approaches to the Social Sciences, (Calgary, Alberta: Faculty of Social Sciences, The University of Calgary: pp. 76-88. Popper, Karl R., 1965a. The Open Society and It Enemies I: The Spell of Plato, (Princeton, N.J.: Princeton University Press). _____, 1965b. The Open Society and Its Enemies II: The High Tide of Prophesy, Hegel and Marx (Princeton, N.J.: Princeton University Press). _____, 1965c. Conjectures and Refutations: The Scientific Knowledge (New York: Harper Torchbooks). _____, 1967. rationalite,
La in
rationalite et le statut Emil M. Claasen (ed.),
Growth
of
du principe de Les fondements
22 philosophiques des systemes economiques, (Paris: Payot). _____, 1969. The Poverty of Historicism, Second Edition (London: Routledge and Kegan Paul). _____, 1972. Objective Knowledge: An Evolutionary Approach, (Oxford: Oxford University Press). _____, 1975. Reason or Revolution, in Theodor W. Adorno, et. al., translated by Glyn Adey and David Frisby, The Positivist Dispute in German Sociology, (London: Heineman). _____, 1976. The Myth of the Framework, in Eugene Freeman (ed.), The Abdication of Philosophy, (La Salle, IL: Open Court). _____, 1979. Die beiden Grundprobleme der Erkenntnistheorie, (Tuebingen: J.C.B. Mohr). _____, 1982. The Open Universe: An Argument for Indeterminism, (Totowa, N.J. Rowman and Littlefield). Schroyer, T. 1973. The critique of domination, (New York: George Braziller). 1. There are, of course, even deeper assumptions that almost nobody questions, which are nonetheless crucial to the analysis. For example, will department stores be obsolete in ten or fifteen years? Will people shop largely by computerized telecommunications systems?
