nd [WWW] edition)



Transfer to Grounding Political Development Table of Contents
Transfer to next chapter
Transfer to previous chapter


Several readers of DPD have asked for a more detailed explanation of the five fundamental theoretical requirements, saying that the FTRs are not as obvious or as widely acceptable as claimed. These readers are particularly concerned with the two FTRs of normative grounding and the micro-macro connection. This chapter addresses these concerns, describing and justifying the five current FTRs, especially the two controversial ones. The first three of the FTRs follow from basic considerations of scientific conceptualization; the final two from more complex considerations.

I. Locus of Development

The term development implies that something changes. Development therefore has to be a change in some thing. This is the brunt of the "locus of development" FTR: that the analyst specify clearly what it is that changes. It is not enough to point to a general sense that "things" are changing, as if one has seen some flicker of movement but not the thing moving. Studying the physiology of development requires commensurate knowledge of its anatomy-

General terms like "society" or "politics itself" (Eckstein, 1982) do not delimit clearly what the domain of our theoretical attention should be. "A society," for example, could be a single hamlet or the entire world; the term leaves unclear how we delimit our attention. We can also imagine one part of "society" developing independently of another, which suggests that we can profitably make a more exact demarcation of development.

The need for focus arises from our sense that different foci result in quite different theories of the developmental process itself. We could study the development of the universe as a whole, but our studies would not be particularly useful in understanding the development of a state. When we lose the boundaries of development, we lose a sense of whatever uniqueness and integrity the process has.

DPD addresses the "locus of development" FTR by defining development as a change in a culture, where culture has a specific demarcation. (See Chapter 6 in this book; DPD's definition of culture appears in its Chapter 2.)

II. Recognition of Development

We also need to be able to recognize development when it occurs, either by specifying the developmental sequence directly (e.g., naming the continuum; describing a sequence of states) or by positing a developmental criterion (e.g., complexity). As Huntington (1971) argues in his attempt to discard the study of' development, development without a means of identifying it cannot be distinguished from any arbitrary change.(1)

I originally termed this FTR the "nature of development," but I have come to recognize that we may not be able to name exact states in advance. The difficulty here arises from the problem of whether development is property conceived of as teleological or experimental in nature, a distinction that appears in biology between the development of an organism and the development of a species. The teleological development of an organism means the expression of the properties wired into it. The organism changes, but along a track predictable in advance. A species, on the other hand, does not develop as an entity. Instead, a certain portion of an existing species breaks with its fellows on the basis of new genes. This sort of development is not the unfolding of a predetermined pattern but an experiment, hopefully an adaptive one.

Is political development teleological or experimental? Both senses are used in political discourse, but only the latter sense seems appropriate to the study of political development. The former sense is used when we refer to political developments, plural, implying that one set of circumstances has naturally unfolded into the new set of circumstances. Circumstances of which we were formerly unaware may have come into view, but we continue to believe, at least retrospectively, that we are still observing the same sort of game.(2) The sense of development as an experiment, however, would seem to be more appropriate for the field of political development, which concerns itself with basic "ruptures" in existing social games. When true development occurs, we are able to look back and recognize a fundamental change in the rules.

The case for regarding development as experimental instead of teleological will grow stronger when, in a later section, I argue that development must be normatively grounded. Teleological development is not susceptible to normative evaluation, while experimental development is.(3)For all these reasons, the FTR is thus better termed "recognition of development."

DPD addresses the "recognition of development" FTR by looking at the cognitive structure of each culture. A culture whose underlying cognitive structure becomes more integrated and comprehensive thin before (resolves existing ambiguities in how people are to relate to one another) is recognized as having become more developed.

III. Exact Specification

Theorists need to define development directly - i.e., to define development itself - instead of defining it only in terms of its causes, consequences, correlates, and constituent elements. Here I am not objecting to our pressing beyond development itself to these associated aspects of it, but rather to our thinking that such indirect references constitute a definition. When the distinction between direct and indirect definitions is stated thus baldly, it is apparent that definitions should be direct. Even if this were unclear as an abstract proposition, it is clear from our practical concerns with development. We find that direct definitions of development are more helpful than indirect ones in understanding developmental dynamics and thus in fostering development. This is most obvious in the case of consequences and correlates. It may be that development results in air pollution or is correlated with air pollution, but knowing this helps us neither understand development nor produce it.

A somewhat harder case is that of definitions of development in terms of causes. Surely our ambitions are served more directly by understanding development's causes than by defining it. Unfortunately, several problems make such a focus on causes less useful than direct definitions of development. First, many causes are not directly manipulable. Religious movements may create development, but they cannot be summoned up on command. Second, many causes are not conscionable. Even if we were able to create civil war, or a plague, or a social class's "loss of status respect" (Hagen, 1962), our conscience would not permit it. Third, causes themselves are contingent on many conditions, including the action of the development agents themselves. Causes that are efficacious in one historical period may not be in another.(4) Causes that work when they happen autonomously may go astray when they are applied by design. For example, suppose enormous, sudden population loss caused development; U.S. nuclear destruction of half a country would probably have rather different effects! Fourth, and probably most important, FTRs should not assume empirical facts. (I discussed this in the previous chapter.) Definitions of development in terms of causes assume that the causal link will remain stable overtime and circumstance, a proposition that can only be empirically defended if we have a theoretically prior, direct definition of development.

Finally, development should not be defined in terms of one of its constituent parts, or even in terms of all of them. Granted, development has numerous constituent parts of no obvious integration: it seems to involve changes in law, in individual psychology, in institutional patterns, in historical memory, in popular culture, in birth rates, and so on.(5) This variety makes it hard to conceptualize. Some people hold that development's multifaceted nature makes this FTR impossible to satisfy and thus theoretically useless; they would suggest that definitions in terms of the many facets should be permitted. The difficulty with this view is that a multifaceted conception of development does not wring as much information from our analyses as we would like. Pye's (1966a) well-known "syndrome" definition of development - whatever its merits otherwise - does not explain why the three facets of development (equality, capacity, and differentiation) occur together in the development process. If they do not, then development becomes only an index score of otherwise unrelated phenomena, a conception to which we may be driven, but one from which it is unhelpful to start. To take a biological example, a definition of someone's development in terms of increasing height plus increasing weight plus an increasing number of neurons, etc., would allow us to miss the fact that these changes are integrated with one another in very precise ways. A complete understanding of development would include an understanding of the connection among development's facets, not just a list of them.

DPD addresses this FTR simply by defining development itself. Discussions of development's causes, consequences, correlates, and/or constituents are always subordinated to the direct definition.

IV. The Micro-Macro Connection

Development must be defined to embrace both the micro level of individual behavior and the macro level of social institutions. Neither individual-centered nor institution-centered definitions can be satisfactory alone. One argument for this claim is, of course, the vast literature on development written from each of these two perspectives. Institutional capacity, legal infrastructure, party systems, and parliamentary institutions - as well as many other macro-level phenomena - have been cited at one time or another as bound up in development. At the same time, development has been attributed at least as often to the various micro-level considerations embodied in people's beliefs, attitudes, and values: culture, partisanship, public-regardingness, and so on. The degree of theoretical interest in both the micro and macro levels in development would suggest that it involves both levels.

This is not a decisive argument, however. It could be that one side (i.e., the micro or macro side) is mistaking some cause (or consequence, or correlate) of development for development itself. If that were the case, then development would "really" lie with the other side. The FTR of exact specification requires that this possibility be addressed.

There is a more telling argument, however, for insisting that a definition must embrace both the micro and micro levels: the argument I advanced in DPD (pp.9-10). Basically, the argument has the form of a proof by contradiction. First, I show that development cannot be defined in terms of changes solely at the micro level. A moment's thought shows that a situation in which more developed people (in whatever sense) interact through exactly the same institutions would not be considered development; the continuation of older, more venal ways of relating would brand their polity an underdeveloped one, contradicting the assumption that we have adequately defined development solely in micro terms. Second, I show on the other hand that development cannot be defined solely at the macro level. Since developed institutions cannot exist without people capable of sustaining them, we would not consider developed a polity in which "developed" institutions were imposed on people unwilling or unable to operate them Properly. (Here I have in mind Particularly U.S. attempts to impose its vision of government on the Vietnamese.) This contradicts our assumption that we have adequately defined development solely in macro terms. It follows, then, that development cannot be defined solely in terms of either micro-level or macro-level considerations, and thus any adequate conception of development must incorporate an understanding of how the two levels are related to each other in the course of development.

DPD addresses this FTR by defining development in terms of changes in culture. Because culture must be publicly common, change in culture means not just a change in individual capacity to employ cultural ways of relating but also a change in the actual use of these ways. This location of culture as intermediate between individuals and institutions is discussed further in Chapter 7.

V. Normative Grounding

The last and most controversial of the FTRs holds that our conception of development must be normatively grounded, so that when a polity (or whatever locus we finally settle on) develops, the new state is morally preferable to the old. "Morally preferable" is here used in a strong sense of preferable according to universal criteria of moral judgment, not in the weak sense of merely conforming to one person's or group's conception of morality.

Given the profound debates over normative evaluation of social systems, this requirement would seem to sink all hope of obtaining any definition of development.(6) Nevertheless, there are important philosophical reasons for insisting on this FTR. Difficulty in satisfying it is no excuse for ignoring these reasons. (In any case I believe it can be satisfied.) Let us look at these reasons in detail.

The argument is in two parts: first, that any complete theory of social change (thus including any complete theory of development) has to incorporate a normatively grounded theory of moral choice; and second, that a conceptualization of development has to align itself with those normative grounds. The first argument hinges on the observation that people are free to act, and in fact sometimes do act, on the basis of moral concerns that are quite distinct from other causal pressures upon them. This freedom to act means that a complete theory of social action must incorporate a theory of their moral judgment. However, because such theorizing is itself potentially a cause of social action, the theory must hold reflexively, that is, people's knowledge of the theory must not affect their adherence to the behavior predicted by that theory. The only such theory is a normatively grounded one, so that the role of moral behavior in the theory is one that the actors themselves would be willing to endorse.(7) Our reluctance to define development in normative terms stems not from our settled conviction that development is not normative but rather from our reluctance to deal with the thorny issues that normative discourse inevitably raises. Our experience in Vietnam (and the underlying imperialistic moralism from which adventures like that continue to flow) gave us good reason to fear normatively involved theory and to suspect its impartiality. But once we agree that any complete theory of social change must deal with the problem of normative grounding, the major objection to defining "development" in normative terms disappears. To understand society we have to grasp normative issues - or so I will argue below - and thus we can give renewed credence to our intuitive sense of development as normative improvement.

The remainder of this chapter, then, is devoted to proving the more difficult part of my claim, namely, that any complete theory of social change (that is, change of any form, whether developmental" or not) must inevitably incorporate a normatively grounded theory of social choice.

A. Introduction

To what extent, or in what way, do we(8) need to pay attention to normative considerations when constructing theories of social change?(9) Three distinct positions are possible. The first form of social change theory does not rely on or incorporate normative considerations in any way. In such theories people act out of necessity or under compulsion: without choice. The forces underlying social action may be historical, psychological, or technological, but they are not normative. Such impersonal theories of social change I will here term change/i theories.

Change/i theories may be represented by Lynn White's classic Medieval Technology and Social Change (1962), which sees social change as arising from technological discoveries like the stirrup and the plough.(10) The stirrup, for example, made possible the use of force in ways not possible earlier in history, enabling certain people - the possessors of the technology - to insist more easily on their own way of relating, inducing acquiescence through the threat of force. The new technology constrained not just the unmounted peasant, however; it also constrained the mounted warrior himself to live in certain ways (White, 1962:28ff.). In White's view, social change comes about not through the imposition of one group's moral preferences on another but instead through technology's broad, unanticipated impact on all groups. Theories of technology-induced change are impersonal in their research focus, even if we retrospectively (and irrelevantly) evaluate these changes morally.

The second form of social change theory admits that normative choices by social actors affect social change to some degree, but explains away these choices by means of some theory of normative choice resting on nonmoral factors. This empirical theory is not itself normative in character, since its purpose is not to make any normative claims about actors' choices but only to predict or explain them. The prediction or explanation may be made on the basis of non-normative causes;(11) in any case the normativeness of the cause is not seen as germane. Theories of social change incorporating such an empirical theory of normative considerations I will here term change/c theories.

Change/e theories include any economic determinism or"vulgar" Marxism that sees social change as arising in part from philosophical, religious, or moral doctrines, but that also sees adherence to such doctrines (as well as their origins) as flowing not from the good reasons supporting the doctrines themselves but rather from their adherents' location in the class structure (and, behind that, from the mode of production creating such a class structure). In this theory normative considerations play a role in social change, but they appear only as derivatives of non-normative considerations.(12)

A non-Marxian example of a change/c theory is L.T. Hobhouse's Morals in Evolution: A Study in Comparative Ethics (1906). Hobhouse surveys the moral codes of many civilizations and finds a steady advancement in moral values.(13) He explains this advancement, however, only in terms of a conjectured "survival of the morally fittest," in much the same manner as Edmund Wilson (1975) explains altruism in terms of its survival value for the altruist's genetic code. Ultimately, that is, Hobhouse explains morality not in its own terms but as a consequence of the non-normative physical and biological forces that determine survival.

The third form of social change theory includes, like change/c theories, an embedded theory of normative choice, with the additional proviso that the embedded theory is normatively grounded-that is, its causal explanations involve to some degree the nature of morality itself. Morality is here taken to be a domain sui generis, so that normative considerations can only be explained in their own terms, not as deriving from some non-normative domain. Such normatively grounded theories of social choice I will here term change/n theories.(14)

I am aware of only two examples of change/n theories: DPD and Radding (1985). Space limitations prevent an exposition here of the complex arguments involved; I will discuss them briefly later.

The burden of my argument is that only change/n theories are theoretically satisfactory.

B. Social Institutions Are Based on Ways of Relating

Meaningful social intercourse (the Weberian-Parsonian "action") depends upon the participants sharing a common framework for interaction - a "way of relating." Such a framework creates meaning by structuring behavior, so that mere physical motion acquires an intellectual status through its location in a context common to all parties. When, encountering Al in a back alley, Bob produces a gun, Al and Bob immediately share a common understanding of the situation, despite the differing roles they happen to play in it. Their interaction is meaningful, if unpleasant; actions and reactions stand in definite relation to one another and are thus imbued with meanings one could not deduce from the physical motions alone.(15)

Such frameworks of meaning often go unnoticed, because in stable social systems people already share them with their associates. When I arrive in my classroom I don't have to explain to my students how I expect to interact with them; we already share that framework. Only rarely do people experience the role conflicts that require a real decision among alternative ways of relating.

The phrase "ways of relating" tends to raise images of family rituals, classrooms, or specific pair relations - the small frame, involving a few people knowing one another well. Such images are too limited, however. Laws, legal systems, governments, states-all represent ways of relating shared among millions of people. The neighbor who just asked to borrow my car relates to me in a way that comprehends laws about car theft, liability, and so on. Even though our knowledge of each other is only casual, our interactions are nonetheless guided by definite understandings.

Social action involves, then, a coordinated choice 6y a set of actors among alternative ways of relating and a subsequent choice of behavior appropriate within the way of relating that has now become normative.(16) Note again that meaningful action requires coordination among actors; all must acquiesce in the framework in order for behavior to have social meaning. But only acquiescence, not support, is required to establish a way of relating as the norm. No assumption is made of a "value consensus" or of an agreed-upon "social contract" - only of a common acquiescence.(17)

The choice among ways of relating is, of course, the choice of interest in political change studies. Political change is not the mere change of actors, with the basic way of relating remaining the same, but a structural political change. Structural political change occurs when people relate to one another through a new set of social categories not homomorphic to categories obtaining previously. The requirement of non-homomorphism eliminates superficial, cosmetic changes such as new titles disguising old positions or new personnel maintaining old power relationships.(18) In structural change, old relationships disappear, displaced or reconfigured by new ways of relating. The linguistic structure of political discourse changes (Foley, 1986).(19) It is with such structural changes, not with superficial or cosmetic ones, that this article deals.

C. It Follows from B that Theories of Social Dynamics Require an Embedded Empirical Theory of How People Choose Among Ways of Relating

Even though every social change arises through people's choice of what ways of relating to adopt, the importance to society of such choice is often obscured. Many of our actions are made in the face of reasons so obvious that to notice them at all seems odd. For example, most people would probably require some time to construct an explanation of why they automatically "choose" to pay their restaurant tab. Even in situations where people do think of themselves as really .'choosing," choice arises only within a taken-for-granted structure of alternatives and considerations. To grade an essay A- or B+; to vote Republican or Democratic; to eat at Wendy's or Burger King-none of these choices changes the system of social categories within which the choices appear. Such choices may interest us as the everyday fabric of social intercourse, but they are not choices to adopt certain structures and ways of relating, and are thus of little interest to students of broad, qualitative social change. The distinction is between "market research," be it commercial or political, and real social theory.

Fundamental choices among alternative ways of relating do exist, however, as revealed most clearly where political alternatives, political conflict, and thus choice among choice structures, are present: in Northern Ireland, where Catholics and Protestants disagree over how to relate to one another; in the civil rights movement, which proposed to U.S. citizens an alternative way races could relate; and in the several current rebellions in Latin America. All of these are clashes over opposing ways people will relate to one another both economically and politically. In such struggles the tension arising from weighty, counterpoised reasons for alternative ways of relating makes apparent the fulcrum of choice creating this tension.

Choice is disguised in social regularity, and revealed in social change. Any theory of social change must accordingly contain at least an empirical theory of why social actors choose one way of relating and not another. Consider once again White's (1962) theory that the impersonal forces of technological change cause social change. Even if the theory were to ignore the action of human choice in the invention and development of new technologies, it must still explain why, faced with new technological possibilities (and the social changes potential in them), people will choose to adopt them. Thus the theory requires an embedded or attached theory of individual choice. In the case of the stick-up, the theory may be that people would always rather live dominated by mounted knights than die opposing them. Such a "theory of choice" is obviously both simple-minded and factually wrong. But the important point for the present argument is that even the most grand theories of powerful, impersonal, economic or geopolitical forces must ultimately rest on a theory of bow people choose.

D. The Choice Among Alternative Ways of Relating Is Made on Both Moral and Practical Grounds

In choosing among ways of relating, the actor finds relevant to the choice two quite distinct considerations. First is the actor's judgment of the moral(20) status of the alternative ways of relating. Ways of relating establish how people are to deal with one another; they specify what interests exist, whether these interests will be satisfied or not, and in what way. Once established as a common framework of interaction, a way of relating is prescriptive and intended to apply to all.(21) A way of relating constitutes a common moral position adopted by the people interacting through it, regardless of whether those actors see it as involving moral issues. Because a way of relating establishes how ethical questions are to be resolved, the choice among ways of relating is an ethical one.(22)

Ethics is not the only consideration relevant to the choice, however. A social actor must also consider how to establish the alternative way of relating as the norm of interaction. One can evaluate the alternatives with reference only to one's personal moral code, but the basis of social life is reciprocal action within the same way of relating.(23) But before action can be social, the actors must first establish which way of relating will be the norm. Each actor's choice of which way of relating to employ will thus depend not only on a moral evaluation, but also on a practical judgment of how feasible it would be to establish any of the alternatives. And feasibility depends on a world of considerations: the distribution of personal moral preferences in the group; the effort required to teach others a way of relating; and the current structural power (Stone, 1980) commanded by the proponents of various views. These varied considerations I call practical, because they exist apart from any single actor's normative preferences. To the extent that distinctively moral discourse is absent among the actors, these practical considerations will prevail, and strategic rather than moral considerations will govern the establishment of a way of relating.

The choice among ways of relating thus involves both moral and practical considerations: moral considerations of what ways of relating are fight, and practical considerations of what ways of relating the actor can establish as normative. This conclusion should be familiar to the reader from the works of Weber (1958) and Mead (1962), as well as the more recent formulations of Parsons (1961). I have taken the reader over this ground again in order to clarify in what way normative evaluations are significant. I now turn to the consequences of that significance.

E . It Follows from D that Two Extreme Forms of Social Change Theories Can Be Distinguished, Differing in Whether They Concentrate on Practical or Normative Considerations as the Source of Social Change

The distinction between practical and moral considerations of choice organizes social change theories along a continuum ranging from those in which social dynamics stem solely from objective social forces, at one extreme, to those in which social dynamics stem solely from the moral choices of social actors, at the other. Theories of objective social change point implicitly to the practical considerations facing people as they make and remake their ways of relating.

White's (1962) theory of social change arising from technological innovation, as cited in Section A of this chapter, is an example of one extreme. A second example can be taken from studies of the impact of communications on political change. Communications theorists (e.g., Lerner, 1958, and the authors in Pye, 1963) see the advent of new communications technologies - prototypically the transistor radio - as creating new mass demands and, in the face of institutional incapacity to satisfy those demands, mass frustration. Control over the media enables emerging national elites to make their preferred ways of relating more prominent (in particular, their desire for mass mobilization in support of the central government). As is also made clear in White's analysis of medieval technological changes, however, the new technology constrains the new elites (e.g., radio carries well only certain types of appeals) and blocks their objectives in unanticipated ways (e.g., through the mass frustration that follows rapid politicization). In these examples the new technologies act impersonally; moral choice plays no role. The national elites just mentioned, for example, had their preferred way of relating well before the advent of the transistor. Technology merely altered the mix of objective forces in the struggle of the new against the traditional elites; it did not create a new morality.

I know of no theories of social change at the other extreme of the continuum discussed here; all theories readily admit that impersonal social forces affect social change to some degree. Some theories do focus on individual moral choice to a greater degree than others, however. Gabriel Almond (1973), for example, sees development as testing on the framework of rational choices made by social actors. Almond does not specifically term these choices "moral," but the choice among different social arrangements is obviously of ethical significance, regardless of the grounds - "rational" or otherwise - on which it is made. James C. Scott (1976), who also focuses on "rational" choices of peasants, refers directly to such choices as moral. Jürgen Habermas provides the clearest example of a theory focusing on moral choice; he views social change, and particularly political development, as the consequence of direct efforts of social actors to establish and maintain a social system allowing for undistorted communication. Because communicative action raises several validity claims, Habermas calls attention to the different methods required to redeem these claims: the scientific method, to redeem the claims of (factual) "truth"; the psychoanalytic method, to redeem the claims of "truthfulness" (authenticity); and, of most concern in this essay, the methods of ethical justification, to redeem the claim of (interpersonal) "rightness."(24) Habermas argues in Legitimation Crisis (1975) and Communication and the Evolution of Society (1979:esp. Ch. 5) that people's normative evaluations of a state's legitimacy claims have a substantial, immediate impact on the stability of that state.

In principle, then, social change theories can incorporate moral choice to a greater or lesser extent. The question I now turn to is whether we can exclude moral choice altogether. In other words, can a change/i theory be complete?

F. The Choice Among Alternative Ways of Relating Can Always Potentially Be Made on Moral Grounds

One cannot exclude the possibility of moral choice from social change theories. To show this, I must deal with two major arguments by which moral choice is excluded: the theoretical argument that moral grounds are merely a post hoc rationalization of nonmoral processes; and the practical argument that the moral aspects of choice are swamped by the practical (nonmoral; impersonal) social forces. The theoretical argument denies the existence of moral considerations; the practical argument denies their force. Note that I need not counter these arguments by taking the opposite position-that, theoretically, all moralizing speech is genuinely moral and that, practically, moral considerations overwhelm all others. I need only argue the limited position that there does exist a domain of moral considerations and that such considerations can determine action.

The theoretical argument holds that moral speech does not indicate the existence of an autonomous moral domain, i.e., a domain of discourse with its own concerns and standards of justification. Instead, moral speech is held to exist merely to disguise from oneself or others the real basis on which one makes one's decisions: moral speech has no authenticity, because the decisions are made on quite another basis. An extreme Freudianism, for example, might hold that all moralizing is merely a papering over of subconscious urges unacceptable to the conscious mind.

This argument obviously has some validity. At some point in our lives, we have all found ourselves, through ignorance of or distaste for the real basis of our actions, justifying them by advancing arguments we didn't really believe. Ordinary observation of others tells us also that people exhibit to different degrees this tendency to rationalization. "Strong" people seem capable of reporting the real bases for their actions, so their explanations maintain the ring of truth over time; "weak" people have difficulty reporting honestly, so their explanations inevitably ring false or eventually come into question.

On the other hand, our very ability to recognize that we do occasionally (or even often) rationalize depends on our recognition of a domain of authentic speech. If we can distinguish rationalization, it must be from truthfulness. We recognize rationalization not as a natural state but as a psychological difficulty or disturbance from which we wish to free ourselves. Thus psychoanalysis is not simply diagnostic but also emancipatory; psychoanalysts do not shake their heads in sorrow at the irremediable falsity of humans but rather seek through psychoanalysis to assist people in recovering their real selves. In sum, the theoretical argument that moral speech indicates no authentic domain of discourse cannot be maintained. Moral speech cannot be taken uncritically, but our very capacity for criticism reveals the presence of a domain of authenticity.

There still remains the practical argument, however, that in the real world, nonmoral considerations of action outweigh any moral sense. This position grants that there is a domain of moral authenticity, but contends that our actual decisions are made, whatever our intentions, on other grounds. According to this position, the domain of moral authenticity may enable us to decry the actions we are forced to make and the pressures on us to make them, but it cannot alter the necessity of the actions.

It is apparent, though, that people are able to advocate and establish ways of relating they believe to be ethically right, even against practical difficulties. One need only cite great reformers like Mohandas Gandhi and Martin Luther King, Jr., to see that the practical considerations of clubs, dogs, jails, and popular antagonism are not the only social forces at work. People do occasionally make "impractical" choices on moral grounds. Although practical considerations may make moral choice painful or even fatal, the possibility and force of moral choice can never be dismissed.

G. Any Complete Theory of Social Change Must Therefore Incorporate at Least an Empirical Theory of Moral Choice

It follows from the above that strictly impersonal theories of change - in my terminology, change/i theories - are not tenable. Any complete theory of social change must be able to account not only for the practical forces affecting change but also for how people's moral choices affect it.

I have not yet distinguished between change/c and change/n theories, however, since the argument thus far makes no special claim about the nature of the required theory of moral choice. If we are to have an empirical/predictive theory of social change, the required, associated theory of moral choice must also be empirical/predictive in nature, but I have not yet established any necessity for it to capture the normativeness of the choice.(25)

H. It Follows from C, D, and F that Any Complete Theory of Social Change Must Incorporate a Theory of How Social Actors Evaluate the Virtue of Alternative Ways of Relating

If an adequate theory of social change requires at least an empirical theory of moral choice, what sort of demands do we make on the litter theory? Such a theory must account for two different types of moral choices: choice within a way of relating and choice between ways of relating. Choice within a way of relating has to do with the ordinary course of meaningful social action; choice between ways of relating has to do with the very structures through which social action acquires meaning.

To describe choice within a way of relating is simply to describe moral reasoning itself. When established as the basis of a relationship, a way of relating is normatively prescriptive. Deciding what the prescription means - and how, therefore, one should satisfy it - is moral reasoning. Thus a theory of moral choice must at a minimum describe the nature and forms of moral reasoning itself, or equivalently, the possible ways of relating.

Not all moral choices arise within a mutually accepted way of relating, however. Occasionally social actors must choose between alternative ways of relating. Such choice may occur when an actor is able to visualize several alternative ways of relating with others in a situation (possibly including completely novel ways of relating), or when different actors simply meet and wish to employ different ways of relating. As noted previously, the choice among ways of relating is itself in part a moral choice, and so a complete theory of moral choice must describe how such choices are made.

I. It Follows from F that Empirical Theories of Actors' Moral Evaluations Must Capture the Normativeness of the Choice

The theory of culture to a large extent disentangles the description of human life from moral judgments. Yet, since moral judgment is integral to human experience, an amoral approach cannot do full justice to this experience and indeed disrupts its phenomenological integrity (Metzger, 1981:ix-xxv).
I claim in this section that a purely empirical theory of moral choice - that is, one that describes moral choice as a consequence of nonmoral factors - is impossible. The argument proceeds by contradiction. Suppose that one were able to predict on exclusively empirical and nonmoral grounds people's moral choices - for example, on the basis of their greediness, their reptilian hindbrain, or their mothers' having dropped them on their heads at birth. That theory must also apply reflexively, that is, to oneself. One is then in a situation in which one claims that one's own moral choices are made on nonmoral grounds. This position is untenable, however, because it is vulnerable to the so-called "open question": "But is it right that I make reptilian moral decisions?" This is a meaningful question because we have assumed that the theory is based on nonmoral considerations.(26) The open question may be posed either by others or by oneself, but in either case it demands a response. One's possible responses are to (a) do nothing, since morality has no autonomy from empirical forces; (b) change one's definition of morality so that it is coterminous with the empirical theory; or (c) alter one's behavior.(27) But none of these responses is satisfactory. I disposed of (a) earlier: morality is not an epiphenomenon or post hoc rationalization of nonmoral considerations. Response (b) violates our initial assumption that the empirical theory is based on nonmoral considerations. Response (c) contradicts our assumption that the empirical theory predicts moral choice, since one has chosen to behave not according to the theory. Thus our original assumption, that an empirical theory of moral choice can be constructed on exclusively nonmoral grounds, is untenable.

J. It Follows from I that Only Change/n Theories Are Tenable

The basic point is that if morality is a quasi-autonomous domain, then a theory of moral choice must take into account to some degree the authentically moral considerations involved in moral choice. A theory of moral choice applies to us as well as our subjects. If we are to treat our subjects as respectfully as ourselves, our theory of moral choice must capture the normativeness of their choices, not just the choices alone.(28) Change/e theories are therefore not tenable, and any complete theory of social change must incorporate a normatively grounded theory of moral choice. Q.E.D.

Of the many possible empirical theories of moral choice, that of Kohlberg (1981, 1984) appears to have the most solid normative grounding. Kohlberg's theory of the development of moral reasoning can be applied (with some care to avoid the fallacy of composition) to groups of people sharing a common way of relating. The result (DPD; Rosenberg, Ward, and Chilton, 1988: Chapter 6; see also Power and Reimer, 1978) is a normatively grounded conception of political development rooted in Kohlberg's empirically well-established theory. A similar approach is taken by Radding (1985), who explains certain historical changes in terms of Piaget's analysis of the development of abstract logical reasoning. Piaget's claims are of logical adequacy, while Kohlberg's are of moral adequacy, but the theoretical issues raised are quite similar.(29)

K. It Follows from I that Conceptions of Development Must Be Normatively Grounded

If social scientists are then to take moral discourse seriously, we must accept as valid our discourse about which social system is best for a society, or, at least, which is better than the present one. (There is no implication here that what is better or best will be the same for every society.) Since any complete theory of social change must incorporate a normatively grounded theory of moral choice, we have to recognize the meaningfulness of the assertion that a society has become better. Our natural language term for a society's change from one configuration to a normatively superior one is development: we say that the society has developed.(30)

L. Implications for Social Science

Almost two decades ago, Huntington (1971) proposed that social scientists abandon the normatively loaded study of development and take up instead the purely empirical study of "change." The thrust of my argument, however, is that normative considerations are inevitably a factor in social change, so that the vision of a strictly empirical theory of social change is an illusion. Every complete theory of social change must also include a theory of the nature and circumstances of development. Certainly some social changes are morally neutral, or arise from nonmoral causes. But to the extent that moral choice does have force in the world, some social change will be developmental in nature.

The examples cited above are of large-scale social movements, but normative choice may have as great or greater an effect on small-scale interactions. James C. Scott (1985) describes the tension between the large-scale structural power of the landlord and the small-scale moral expectations and traditions of the peasants, in which tension the peasants have recourse to the "weapons of the weak." Similarly, Tilly (1978:186, citing E. P. Thompson, 1971) sees the nineteenth-century European social conflicts over the price of staple foods as moral conflicts concerning the proper relationship among producers, consumers, and governmental authorities. Such small-scale conflicts, if pervasive enough, can generate large-scale movements, but it is clear that moral choice is important at all scales of' social activity. Social scientists thus cannot rely upon a "division of labor" argument to avoid studying moral choice: all levels of social activity appear drenched in such choice.

Some social scientists might still take the position that moral choice is not important enough in comparison to nonmoral considerations to warrant attention. If social science can explain (say) 99 percent of all explainable variance by nonmoral factors, why should it enter the mine field of morality to explain the remaining one percent?

An obvious answer is that the relative importance of moral and nonmoral factors cannot be estimated unless both are examined. We will never know what percentage of the variance moral factors explain unless we look at them.

But assume that the figure of one percent is correct. A restriction of social change theories to nonmoral factors would still affect the character of social science-in particular, by making it conservative instead of emancipatory. Let me quickly define those terms: a conservative social science is one that gives no scope to moral intentions. It is conservative not in the sense of preserving the past, because it sees many potential nonmoral sources of change, but rather in the sense of denying the possibility of a willed emancipation. It is conservative in its denial of the possibility of progress, since humans are the helpless victims of forces beyond themselves. An emancipatory social science, on the other hand, specifically recognizes the possibility of moral choice. A normatively grounded theory of moral choice allows social scientists legitimately to adopt a specifically critical posture vis-a-vis alternative social futures. It recognizes and raises to consciousness both the impersonal forces a society experiences and the moral considerations that the choice among futures raises.

In essence, then, social scientists must choose one of two visions of their project: on the one hand, a social science that disempowers people by denying the possibility they can change anything (or, to cast this as this project's adherents might, that liberates people from the dashed hopes attendant on a misguided faith in moral agency); and on the other hand, a social science that recognizes the possibility of moral change and ultimately of human emancipation (at the cost of the frustrations and conflicts of normative discourse).

Two considerations argue that only the latter choice makes sense. First and most direct, people already make, and know deep down they can always already make, moral choices-that they are not victims. A complete social science must recognize that.

Even in the face of the first argument one could continue to maintain that the role of moral choice is so slight, and its threat to our self-conception as scientists so immediate, that we are justified and even obligated to neglect it. We are comfortable with our long-established role as "scientists" and all too aware of the hubris tinging the role of prophet, visionary, or moralist. And yet this argument seems to depend at heart on our willed ignorance of the role of moral choice, not on a realistic assessment of the dangers of the different paths. We can choose to remain "scientists" in the narrow sense only if we are sure that moral choice is really of little consequence; but this is not at all certain. Each side sees what it wants to see: "scientists," ignoring the role of moral choice, devise experiments showing less and less role for it; "emancipators" discover, by contrast, that the more scope they provide for recognizing and making possible moral choice, the more important it seems to become. My second argument is cast in the face of this uncertainty: that when one cannot choose between two alternatives on the basis of clear evidence or logic, then one should choose the ,alternative with the more interesting consequences. If one ultimately decides that the consequences are not really very interesting, or if the evidence really does mount up in the opposite direction, one has lost no more than time and has by recompense acquired a firmer understanding of why the other alternative is superior. But to foreclose options without looking at them seems to me to be a grave crime, a crime only made worse by the excuse that we are afraid of ourselves. This second argument is less direct but I believe more persuasive than the first.

Given our uncertainty of evidence and logic, the only reasonable alternative is to admit and examine the possibility that moral choice is important and that human emancipation is possible, and to look vigorously for how it might appear and be fostered. Obviously we will have to remain conscious of the many pitfalls of such a stance - notably, the possibility that real emancipation will be swamped by charlatanry, moral imperialism, and muddy thinking. Our self-definition as scientists has not protected us from these pitfalls. My own sense is that a determined respect for bottom-up emancipation, however, as found in liberation theology or in Paulo Friere's "pedagogy of the oppressed," will provide sufficient moral compass.

In conclusion, it seems to me that people already know that they can make moral choices - that they are not victims. If social scientists do not take up the challenge of explaining to them where and what their choices are, then other, less thoughtful - or less scrupulous - people will. As long as the emancipation of all people lies ahead of us, social scientists have as clear a charter as anyone, and better than most, to create it. We should welcome the challenge instead of running from it.

This way to the UMD homepage
This way to the Political Science Department home page
Page URL:
Page Author: Stephen Chilton
Link to Home Page:
Last Modified: September 25, 2000
Mail comments, esp. typos and other errors, to:
Proofreaders / Colleagues Honor Roll:

The University of Minnesota is an equal opportunity educator and employer.
Copyright © 1999 Regents of the University of Minnesota. All rights reserved.


1. I agree with Huntington's argument here to the extent that if, as he claims, we cannot reach agreement on the nature of development we should stop talking about it. My claim is, of course, that the conception in DPD is the basis of agreement. 

2. Such a usage is seen in Wellhofer's (1989) article, "The Comparative Method and the Study of Development, Diffusion, and Social Change." Even though the term development is singular in the title, its meaning in the article does not go beyond that of an inevitable process (in particular, of industrialization). 

3. Normative evaluations of teleological developments are like evaluations of the law of gravity. We might prefer that gravity be stronger, or weaker, or more variable, but such evaluations are irrelevant to gravity itself. Normative evaluations of alternative forms of society, on the other hand, are quite relevant to the political-developmental process. 

4. This is the argument of dependency theorists, who argue that development consequent on the rise of capitalism in Western Europe does not carry over to the penetration of capitalism into underdeveloped countries today. 

5. Palmer (1989:Table 2.1) lists an enormous variety of changes that appear to be associated with development. 

6. Indeed, in my opinion, normative grounding has been the reef on which our predecessors' theoretical ships have sunk, either because other theorists would not accept the normative premises (implicit or explicit), or because the societies they purported to describe would not choose as they were supposed to. Huntington (1965) suffered at least the first fate; most development planners have suffered the second. 

7. Edwards (1989:1) similarly argues that because "the theory is a factor" and "the theorist is an actor," we must move away from theory as prediction and toward theory as creation. Edwards asserts (1989:14) that this reflexivity ultimately falsifies all theories, but this is an overstatement: reflexivity falsifies only those theories that are not normatively grounded. 

8. Unless otherwise specified, "we" in this article means all social scientists (political scientists, economists, sociologists, anthropologists, social psychologists, cultural geographers, philologists) working in the area of social change and development. 

9. Any theory of social change is obviously also a theory of social stasis, since a complete explanation of change will also explain the occurrence of stasis as the absence of change. 

10. For the example's sake I am casting White's view of social change in a more extreme way than he probably intends. 

11. For example, an act of moral courage could be explained as the consequence of being raised in a certain social class. 

12. Hudelson (1990) discusses the historical variety of theories termed "scientific Marxism." 

13. I use the term "advancement" advisedly: Hobhouse has no doubt that, in the long run, our moral systems have gotten better and better. 

14. This trichotomy of approaches to social change resembles in part the distinction made by Charles Tilly (1978:6, citing James Coleman, 1973:1-5) between "causal" and "purposive" explanations of social action. While change/i theories are clearly causal, Tilly does not distinguish purposive explanations resting on nonmoral, empirical theories of choice (change/c) from purposive explanations resting on the nature of morality itself (change/n). 

15. This distinction is, of course, that of Clifford Geertz (1973) between"thick"and "thin" descriptions. 

16. The choice may not be of whether to relate, as with Al and Bob above, but it will always be among alternative ways to relate. In a larger sense, the choice of whether or not to relate is itself a choice of a relationship. This is the crux of the issue of how we are to relate to the starving peoples of the world: ignoring their problems is only a more horrible alternative. 

17. I must widen the connotation of my terms here. To agree on a way of relating sounds like an agreement among the parties immediately present in some transaction: e.g., if a candidate buys a citizen's vote, the way of relating established appears to concern only them. But it is evident that their action violates the basic relationship among all voters, whose physical presence is not required (and was not even contemplated) by the general way of relating we call voting. Bribing voters destroys the meaning of all social life predicated on the sharing of that understanding. Thus Gandhian nonviolent resisters accept legal punishment for their actions in order to uphold the ideal of a shared understanding even while they try to change its form (Bondurant, 1971, esp. Ch. 5). 

18. Palmer (1989:111-113), for example, carefully distinguishes coups d'etat and revolutions on precisely this basis. He also notes in this vein (p. 252) Ghana's (quickly suppressed) post-coup song entitled, "The Cars Are the Same, Only the Drivers Have Changed." 

19. Historically, we have seen yesterday's blood feud become today's political and economic competition. Intimate, personal ties of community have become contractual and impersonal. These changes have not been cosmetic or superficial: rather, they have altered radically the way people in Western society look at and relate to one another. 

20. The term moral can mean either "evaluable against standards of morality" or "in accord with standards of morality." The former sense notes the mere presence of a claim; the latter sense requires a redemption of that claim. In this article, the former sense is meant unless the latter sense is specified. 

21. Of course, "all" is limited to the people who the way of relating contemplates as interacting. The Rotary Club's way of relating, for example, binds only Rotarians. 

22. I hasten to add that I am not positing here the empirical existence of any single system of moral choice. Each culture (and, ultimately, reach person) has a unique understanding of what is "right." But as Kohlberg (1981) points out, the empirical observation that people differ in their moral judgment does not imply the philosophical conclusion that their judgments or modes of judgment are of equal merit. To maintain such an implication is a form of the naturalistic fallacy. 

23. If I register to vote, for example, I do so in expectation of others doing the same; voting is a social action because everyone's action contemplates others following the same rules. 

24. The terms in quotes are those employed by Thomas McCarthy, the translator of Habermas (1979). 

25. By "capture the normativeness of the choice" I mean to suggest a theory that explains moral choice by reference to the nature of morality itself, not to nonmoral considerations. (Recall the discussion in Section A of change/n theories.) 

26. It is one thing to claim that one's reptilian moral decisions are moral; it is quite another to claim that one makes reptilian moral decisions even though they aren't moral. 

27. I ignore a fourth alternative: that while nonmoral forces inevitably and irremediably push and pull other people, they do not affect oneself. Aside from its condescension, such a position ultimately abandons the field of moral discourse itself. 

28. The old joke goes: "They have prejudices; you have opinions; I have a philosophy." 

29. Radding (1985) concentrates on specific historical analysis and does not fully confront the composition fallacy and the philosophical establishment of logical adequacy; DPD does the reverse. 

30. The term "progress" could also be used, but to my mind its connotation of unilinear change is too great a burden on its use. Of course the same charge can be made against the term, "development", but we have to employ some term for morally grounded change, and the connotations of "development" are to my mind a lighter burden than those of "progress" or related terms.

Page URL:
Author:  Stephen Chilton [email]  |  Last Modified:  2004-11-30
Honor Roll  |  UMD  |  Pol Sci Department

The University of Minnesota is an equal opportunity educator and employer.
Copyright © 2000-4 Regents of the University of Minnesota. All rights reserved.