I apply a rational choice framework to the problem of revolutionary strategy. A preference ordering is established where ideological recruitment is a cheap recruitment function and organizational recruitment a costly recruitment function. The similar development of revolutionary strategy in disparate cases is explained as result of the cost/benefit calculations made by revolutionary elites with respect to these two possible strategies. Where organizational recruitment strategies are found (Vietnam, Cambodia, Peru, China, Nicaragua) ideological recruitment strategies have preceded them.

Two subsequent dependent variables are then explained as a function of distinct recruitment strategies: (1) divisions within revolutionary groups and (2) in -group revolutionary terrorism. Where revolutionary groups are moving from a principally ideological recruitment strategy to a principally organizational recruitment strategy, I expect to find elite conflict (China, Nicaragua, Peru)(Nolan, 1985; Chang Kuo -t’ao, 1971; Gorritti, 1990).

In -group revolutionary terrorism (terrorism directed against the principal base of support of the revolutionary group) is characteristic exclusively of organizational recruitment strategies. Terror directed at potential recruits is a form of administering negative selective incentives and serves a recruitment function (Peru, Algeria, Cambodia)(Degregori, 1990; Palmer,1992; Tapia, 1997; Crenshaw, 1978).

One fundamental assumption on which a theory of revolutionary strategy might be constructed is the assumption of power in numbers. As DeNardo notes,

One of the first tasks of any theory of revolutionary strategy based on the power in numbers assumption is to attempt to reconcile these various perspectives on the nature of recruitment. As a first cut at clarifying this confusing array of possible recruiting strategies, two approaches are suggested that can serve to simplify the various perspectives. For all their range, perspectives on recruiting can be grouped into two different approaches, ideological recruiting and organizational recruiting (DeNardo, 1985). These two strategies are detailed in the following section.

PART I

RECRUITMENT STRATEGIES

Ideological Recruiting:

Ideological recruiting is characterized by the solicitation of support based on the policy statements of the revolutionary challenger versus those of the current regime.

"People whose political aspirations lie closer to the dissidents’ demand than to the government’s

current policy therefore prefer the demand and join the protest. Others stay home" (ibid).

It is undoubtedly true that this form of recruiting plays an important role in the recruitment strategies of all kinds of regime challengers, and other political movements, not simply revolutionaries (McAdam, McCarthy and Zald, 1996). Consider, for example, the history of the civil rights movement in the United States. In this instance, recruiting was based predominantly on ideology. Civil rights organizers would routinely announce their policy positions, the end to segregation, voting rights etc…, and they would subsequently announce their particular method of protest for the occasion, a march, boycott or some other form of non -violent action (McAdam, 1988). Those individuals whose policy preferences were closer to the protesters’ than to the government’s would join the protesters and those who favored the government’s policy, or those who were indifferent, would abstain (McAdam, 1982).

Sidney Tarrow’s model of collective action also relies on ideological recruiting. Though he notes some variation on reasons for joining social movements on the margin, the most compelling reason he argues is "to mount common claims against opponents, authorities or elites…The most common denominator of social movements is thus interest"(Tarrow, 1994; p.4-5). Material resources play a role only to the extent that they aid in forming a concrete organizational structure that can assume the responsibility of movement direction.

Finally, in the revolutionary literature, it is clear that the movements that recently toppled regimes in Eastern Europe were recruited based on ideology (Lohmann, 1994a; Kuran, 1991; Garton -Ash, 1993). Indeed it is precisely because recruiting in these cases was ideological that models like Lohmann’s and Kuran’s can be constructed that investigate flows of information and how they caused shifts in individual preferences.

There is one serious problem, however, with theories of revolutionary strategy based on ideological recruitment. There is no good reason to believe that recruiting by revolutionaries has been solely, or even predominantly, ideological. In fact many students of revolution who have done careful fieldwork studying various cases of revolutionary organizing have concluded just the opposite. Where revolutionaries did try to recruit ideologically they failed.

Organizational Recruiting:

Organizational recruiting is not a well specified category. As noted earlier, Berman cites at least six different possible recruiting strategies in his example of organizational recruiting in South Vietnam. Olson and Popkin concentrate on a goods based recruiting function which is concerned, above all things, with the provision of public goods by the use of selective incentives. Wickham -Crowley (1991) argues that recruitment among peasants in Latin America is achieved by providing certain kinds of public goods, the kinds states normally provide, while Pike (1966) concentrates on straight coercion in his analysis of recruitment in South Vietnam. The definition, to this point, of organizational recruiting then is that it simply be any recruiting strategy that is not based on the ideology of the movement. DeNardo, for example, argues that an organizational recruitment function is one that

1. Olson on Organization: Olson’s theory of groups is well known to students of organization in most academic disciplines. First and foremost, the famous "free -rider problem" has been frequently invoked even in cases where it does not actually apply, for example when the goods being discussed are not public, or where the group being addressed is not a latent group. In theories of revolutionary organization, however, Olson’s model has stimulated some insightful analyses in particular Popkin’s work on Vietnam.

In order to make this review more succinct, a few critical assumptions are useful with respect to revolutionary collective action. First, revolution is a public good. It is non -divisible and non-rival. Non -divisible means that, once provided, no one can be excluded from consumption of the good. Non -rival simply means that any one person’s consumption of the good does not reduce the amount available to anyone else (Hardin, 1982). Second, revolutionaries seek to organize a constituency, or group, that is latent in Olson’s terms, i.e.,, where each member of the group cannot feasibly confirm that everyone else in the group is participating in providing the public good of revolution and where the potential contribution of each member of the group to the provision of the public good is inconsequential. If both these things are true and if individuals are rational, in other words expected utility maximizers, then no individual will join a revolutionary group. There is rationally no reason to participate in the provision of a good which cannot be divided up and provided only to those who participated in its acquisition because, acting as a "free rider", one can receive the good -if it is provided - regardless.

Since this theory predicts that large revolutionary groups will not form, given again the assumption of revolution as a public good and a revolutionary coalition as a latent group, there is an obvious contradiction between the prediction and the fact that revolutionary groups do form. How can this be reconciled? This is exactly the point Popkin considers in The Rational Peasant.

2. Popkin’s Explanation:

Part I: Popkin argues that revolutionaries overcome the collective action problem in a number of ways but principally by the provision of selective incentives to members who participate in the provision of a public good. He notes, for example, that:

In reality, however, ideology and the act of persuading are found wherever recruitment is present whether it be by communist revolutionaries, religious groups, or other forms of social movements. This does not disappear in rural areas. The observation that ideology is present has frequently been taken to be a "test" of the rational choice approach used by various authors. In fact, it is a poor test. Popkin actually devotes a great deal of attention to exactly this issue, the role for ideology in recruitment. He asks, for example:

Part II: The second part of Popkin’s investigation argues that ideology does play a role in recruitment, just not for the recruitment of peasants. Ideology exists where peasants are "rational egoists" because ideology is the most effective recruitment tool for the elites who then organize peasants. Ideology persuades political entrepreneurs to dedicate themselves to the cause in order that, over the long run, they be rewarded with careers in the new government.

them playing a role in the eventual changing of peasant’s outlooks to the extent that what constitutes a selective incentive could be changed (increasing the value of peer praise, for example) but this clearly plays a subsidiary role to that of ideology as the recruitment mechanism for political entrepreneurs.

The problem with this approach is twofold. First, the approach is one that falls in the methodological individualist tradition that normally begins with some assumption about individuals (Przeworski,1984; Levine, Sober and Wright, 1987). In Popkin’s investigation, and most others, this assumption is a rational actor assumption. In fact, the rational actor assumption drives the entire analysis of peasant participation in revolutionary organizations. However, the rational choice assumption leaves no obvious room for ideology. Ideology is nonetheless present in the Vietnamese case (and all others) and so, in an effort to deal with this problem, he suggests that a movement’s political entrepreneurs are not recruited in the same way peasants are but by ideology (ethos). The problem is that for a whole segment of the revolutionary group the rational actor assumption drops out of the argument, at least over the short and medium run. One is asked to believe that political entrepreneurs are not like peasants in methodologically individualist terms. The rationality assumption, ironically, holds for the less educated peasants but not for their more educated organizers.

The second problem with the position is that it argues that ideological recruitment plays a clearly subservient role to organizational recruitment when peasants are the group being organized. Peasants, as the bulk of the revolutionary group, are recruited organizationally while their leaders are organized ideologically. And yet there are a number of good investigations into peasant recruitment that give an important place to ideological recruiting both for leaders and their constituents. Indeed DeNardo argues that peasant recruitment during the Chinese Revolution was principally ideological.

As an addendum to the above observations about the role of ideology in the recruitment of peasants, a whole host of examples could be presented that reserve a primary role for ideology and its role in persuasion after peasants have joined revolutionary organizations. Ideological ‘conversion’ of organization members is famous throughout revolutionary experiences including Vietnam, Mao’s revolutionary army, the FSLN in Nicaragua, and Sendero Luminoso’s revolutionary schools. In none of these cases are the sole subjects of ideological conversion future political entrepreneurs. If this were the case then every member of the Viet Minh and Chinese Communist armies would have to be considered political entrepreneurs thus watering down the value of the category to nothing.

There is a puzzle then that needs to be explained. First, it is the case that ideological recruiting in revolutionary organizations exists. The East European cases prove this if nothing else. Second, organizational recruiting does exist, most analyses of peasant recruitment in revolutionary organizations find this to be an inescapable conclusion. Popkin’s solution, that ideology is first and foremost a tool of recruitment for political entrepreneurs, is insufficient as an explanation because, in most circumstances of peasant revolutionary organization, ideology plays a large part even though it is smaller during initial efforts to bring individuals into the revolutionary fold. Furthermore, any objection that organizational recruiting is confined to peasants would have to confront numerous examples of this kind of strategy in the history of social movements where peasants no longer exist (the United States for example) or are not involved in the movement (the various European examples of 1989)(Chong, 1991).

The following section is dedicated to resolving the paradox noted above. In so doing a model of recruiting is suggested that is different from the models suggested above. It is different precisely because it reserves an important role for both ideological and organizational recruitment. This is not a simple combination of theoretical approaches, however. Instead, the recruitment model suggested takes a critical step backwards and argues that the best analytical approach is not to choose from ideological or organizational recruitment strategies, as DeNardo and others do, but rather to incorporate these two possible strategies into an opportunity set from which revolutionary elites might choose. What results from the model of recruitment constructed in the following section are a number of interesting dependent variables that are a result of mixes of recruiting strategies, predominantly ideological or predominantly organizational.

PART II

A MODEL OF REVOLUTIONARY RECRUITMENT

The power in numbers assumption demands that any theory of revolutionary strategy begin with the issue of recruitment. But, as the section above made clear, this is not an unambiguous beginning. In fact, careful analysts of revolution have argued persuasively for two different forms of revolutionary recruiting which were labeled either ideological or organizational. How can this be accounted for? The obvious first cut answer is that there are two different forms of revolutionary recruiting that revolutionaries might choose from in the process of forming a revolutionary constituency. This, then, is the shallow answer to the question posed above. Analysts of revolution have identified important recruitment strategies, either ideological or organizational, depending on the cases they were analyzing. There are two problems with this answer, however. The first problem is that it fails to address the most interesting question; why did revolutionaries embrace one strategy over the other? Though authors that have addressed both models of recruitment have sought to provide choice theoretic foundations for the preferences of revolutionary constituents they have not done so for the revolutionary elites that design the recruitment strategies themselves. This void is addressed in the following section of this chapter.

The second problem is that, as will be demonstrated, all recruitment strategies have elements of both ideological and organizational recruitment within them. Obviously there are strategies that are predominantly organizational (Sendero Luminoso after 1984) and predominantly ideological (the Eastern European cases of 1989). However, there are also clear mixes of the two and making a dichotomous variable (either organizational or ideological) out of what is, in reality, a continuous variable creates more problems than it solves. In the following section, in addition to providing the micro-foundations for what motivates revolutionary elites to formulate some mix of recruitment strategies, a model of recruiting is established that postulates strategic mixes of both recruitment strategies. This mixing of strategies is based on a rational choice explanation of revolutionary elite behavior.

Rational Choice

Rational actor assumptions have come under increasing scrutiny (Green and Shapiro, 1994). There are at least two fronts to these attacks. On one front, various authors have questioned whether the assumptions rational choice theory makes about human behavior are correct. Much of this critique is sustained by experimental evidence (Hagel and Roth eds., 1995). On the other front, critics have argued that the explanations themselves rational choice theorists have suggested do not stand up to either a) empirical evidence (like the paradox of voter turnout) or b) theoretical completeness (in particular failures to deal with rival hypotheses and selection bias)(Green and Shapiro, 1994; p.6). With respect to revolutionary studies, rational choice theory has produced some particularly tenuous work (Grossman, 1991). Nonetheless, any theory of revolutionary strategy that did not reserve some role for rational action on the part of revolutionary elites would clearly give one pause. From united front strategies to multi -class alliances with otherwise hostile groups, revolutionary elites have shown tremendous capacity to evaluate the costs and benefits of their potential strategic decisions and to embrace decisions that were "coldly" rational in the most restrictive sense of the term. The real test of whether a rational choice approach to explaining revolutionary strategy is appropriate is the degree to which such a theory can explain the strategies embraced by revolutionary elites given a research design that accounts for questions of bias within the confines of a case based approach. The remainder of this chapter establishes the rational choice foundations for a theory of revolutionary strategy. Subsequent chapters will demonstrate the utility of this approach in explaining various anomalies in the literature on revolution. A defense of the cases selected for the study is also included (see appendix one).

A Rational Choice Model of Revolutionary Recruiting

1. Rationality:

The meaning of rationality can be a hornet's nest in and of itself so it is useful to be specific about the criteria. As Arrow notes,

    1. Asymmetry: P is asymmetric. xPyPx is not permissible (otherwise x would be both preferred and not preferred to y).

    2. Irreflexiveness: P is irreflexive. xPx is not permissible.

    3. Symmetry: I is symmetric. If xIy then yIx.

    4. Reflexiveness: I is reflexive: xIx.

    5. Incompatibility: P and I are incompatible: If xPy then not xIy (x cannot be preferred and indifferent to y at the same time).

    6. Comparability: Any two alternatives are comparable (as is each to itself).

    7. Transitivity: Both P and I are transitive. If xPyPz then xPz. If xIyIz then xIz (Schwartz, 1991).

    The above requirements are often referred to as the core set of assumptions that can lead to some rather dramatic insights on the part of rational choice theory. This is, in reality, rarely the case. These criteria seek above all things to provide some limits to the way in which individuals order their preferences. Most importantly, it eliminates preference cycles of the sort xPyPz but zPx. However, most of the powerful insights of rational choice theory flow from the above assumptions plus the assumption of self -interest on the part of individuals. Self -interest can be (and has been) defined in a number of ways from the highly restrictive material self -interest to the less restrictive expected utility maximization where it is assumed that individuals simply maximize their expected utility with respect to their underlying goals whatever they might be. In the latter case a number of different possible motivations are possible, capitalists might want to maximize profit, politicians their stay in office, children their play time. In this analysis, self-interest is used as an actor motivating assumption where individuals are assumed to be rational egoists. A subsequent assumption can be added to the above then;

2. Elites:

In the following section and throughout the work revolutionary elites are assumed to be rational. Revolutionary elites are defined as those individuals who are in a position to decide on, or directly influence, the policies of the revolutionary group as a whole. As a practical matter, deciding who qualifies as a decision -maker in a revolutionary group can be a difficult matter. Revolutionaries work in an environment of risk in which the ability to maintain a secretive and disciplined core group of decision-makers is critical. However, in certain cases, especially after the fact, it has been possible to identify the decision-making elites in a particular revolutionary group (Gilbert, 1988).

Revolutionaries, it is posited, share at least one common goal, they want to displace the current regime and replace it with one of their own. This is not an uncontroversial assumption. Many models of revolutionary activity assume only policy preferences on the part of regime opponents. They might seek a higher minimum wage, the end to an unpopular war, civil rights reforms, etc… If this is the case, then the issue of replacing the regime with a new one staffed by the revolutionaries is not a requirement. Such a perspective permits reform in exchange for a reduction in protest. However, it is exactly this issue that separates revolutionaries from other regime opponents. Revolutionaries make total claim on the government.

Two important parts of a theory of revolutionary strategy have been established to this point. First, power in numbers operates as the underlying method revolutionary elites use to pursue their goal of regime replacement. Because of this assumption, the issue of recruitment is brought to the fore and two varieties of recruiting have been established, ideological and organizational. Second, revolutionary elites are assumed to be rational. For the principal units of analysis then (revolutionary elites), a motivating assumption has been established. The final task is to apply the motivating assumption to the opportunity set posited and see what kind of results can be deduced.

        3. The Opportunity Set:

Revolutionary elites, as I argued in the first part of this chapter, are assumed to have one route to power, numbers. As a first cut, assume that revolutionary elites can choose between either an ideological recruitment strategy as defined in part one, or an organizational recruitment strategy also as defined in part one. These two possibilities define their opportunity set. Obviously both these recruitment functions can cause the collapse of the current regime. There are numerous examples of successful revolutions that have been predominantly one or the other. However, both strategies are otherwise dissimilar and it is important to carefully establish the differences between these two approaches to revolutionary recruitment. Above all things, because rational choice theory argues that revolutionary elites will act strategically and weigh the costs and benefits of particular strategic decisions, it is important to evaluate these two possible strategies with respect to their potential costs and benefits.

Measurement:

Measurement is one of the most difficult obstacles for an analysis like the one proposed to surmount. There are, unfortunately, no established norms that regulate what might be categorized as a cost of organization, on the one hand, much less any particular scale that might allow one to determine how costly one particular organizing strategy might be relative to any other. On some occasions, as when costs are material in nature, evaluating a particular strategy will be simple. On other occasions, as when the costs being evaluated are non -material in nature, the measurement of costs will be more difficult. What is critical is that one be specific about how the costs, and benefits, of any particular recruitment strategy are measured. The following sections take this challenge seriously. Prior to this, however, this section discusses costs and benefits in a general way. Benefits, it turns out, is not a category that will cause much trouble. It can be easily specified within the power in numbers framework. Costs are different. In an effort to address this issue, two important general measures of costs are discussed which cut across both recruitment strategies and serve to organize the subsequent analysis.

       1.  Benefits:

        2. Costs:

The costs of each of the two recruitment functions to be discussed will vary a great deal. But overarching these varied costs are two general categories that can help to organize the question. For revolutionary elites there are two kinds of costs. Some costs are tangible. In this analysis, and most others, the costs of institution building are the best measure of tangible costs for revolutionary elites. Whether it be the maintenance of man power to provide leadership in various geographic regions, the control of material resources to be administered as private or public incentives, the formalization of chains of command, or the formation of guerrilla armies, all of these must be administered and controlled in an efficient way and this is done through revolutionary institutions. In the following sections the degree to which a particular strategy requires the construction of formal revolutionary institutions is the degree to which it is considered costly, in tangible terms.

Intangible costs can be varied but the most important intangible cost revolutionaries must confront is time. Short the structuralist claim, that revolutions happen but are not made, revolutionaries have a limited amount of time in which to achieve their goals and the time they spend battling the incumbent regime is time at risk. The degree to which they can reduce the length of this battle is a savings, both in terms of material resources, and time not spent at risk of being arrested, injured or killed. In the following sections it will be argued that one of the two recruitment strategies discussed is more likely to be a short term strategy while the other is more likely to be a long term strategy. In intangible terms, one strategy is more costly than the other.

Ideological Recruiting; Costs and Benefits:

On the surface, the costs of ideological recruitment would appear to be minimal. As noted earlier, ideological recruitment is based on the publication of the policy platform of the revolutionary group and its subsequent comparison, by potential constituents, with the policies of the current regime. Revolutionaries can increase recruitment by adjusting their policy platform or by prompting the regime to adjust its policies. Tyrannical regimes, for example, might make the platform of the revolutionary challenger appear more appealing, and policy reform on the part of the regime might similarly make it appear less appealing.

The immediate attraction of this understanding of recruitment is that it is analogous to the positioning political parties go through during electoral competitions. Parties, at least according to Downs, modify their platforms in a search for a larger share of voters. The ideology of each party is a function of the ideological distribution of voters in any particular society (Downs, 1957). In the same way, for example, that political parties in a two party system seek out the median voter where voters are "normally distributed" and worry about the loss of extremist constituents as their platforms become more moderate, so too revolutionaries and regimes seek out larger numbers of constituents by modifying their platforms and policies.

The problem with this analogy is that one knows, just from casual observation of revolutionary ideological proclamations, that revolutionary groups do not routinely modify their policy platforms in order to increase recruiting. Sometimes they even appear to stubbornly resist moderating their policies and in so doing alienate potential constituents who might otherwise be sympathetic to the revolutionary cause. DeNardo deals with this in an interesting way by relaxing some of the materially rational assumptions about revolutionaries and arguing that revolutionary elites make a trade-off between ideological purity and numbers. Because individual revolutionaries might be different in terms of temperament (pragmatists vs. purists, for example) the fascinating role for conflict between revolutionary elites is unveiled. Pragmatists act more like political parties in the Downsian scheme whereas purists sacrifice numbers for the ideological purity of the cause. These separate approaches to revolutionary policy are a source of conflict and can explain conflicts like those one notes between Lenin and those reformers of his day who pursued the electoral path toward socialism in Europe (DeNardo, 1985).

The costs of this strategy are contained in the trade-off between ideological purity and numbers. Numbers grow as the policies of the revolutionary group are moderated. But moderating policies and watering down the ideological purity of the group is a cost many revolutionaries are unwilling to accept as the conflict between pragmatists and purists makes evident. Broadly speaking, however, the costs of ideological recruiting of this variety are not steep. Revolutionaries of course bicker, even brawl, over the policy platform of the group but subsequently the success of recruiting follows along quite naturally. Numbers grow depending on the eventual platform of the revolutionaries and the policies of the regime it challenges.

Many authors have taken issue with this approach especially those who work within the rational choice school because it does not address Olson’s criticism of large groups, most importantly the issue of free riders. In the above model, constituents join the revolutionaries when they prefer the policy platform of that group. But Olson’s position on this matter is clear. If individuals are rational, if revolution is a public good, and if the revolutionary group is latent then, short selective incentives, people will not join the revolutionary cause. DeNardo deals with this problem by squarely confronting all approaches that assume that individuals are expected utility maximizers (ibid, appendix A). Other authors, however, that accept Olson’s critique, have tried to discover the kind of selective incentives that might operate behind the scenes in the organization of a dissenting constituency even where recruiting is ideological. The problem is that many of these authors do not examine instances of revolution specifically but rather social movements. Social movements, because of their reformist goals, cannot easily be grouped with revolutions as members of the same set (Wickham -Crowley, 1991). Nonetheless, writers in this field do hold out the hope that their models will be able to explain revolution as well as social movements. McAdam, for example, in his introduction to the latest edited edition on social movements notes that social movements and revolutions are indeed members of the same closely knit family.

Social Movement Literature and the Costs of Ideological Recruiting:

The three broad sets of factors noted in the above cite are all variables which a synthesis of social movement literature produces as critical in determining the success of these movements. Political opportunities, framing processes, and mobilizing structures organize the different lines of investigation that have proven fruitful. For the purposes of this work, political opportunities can be eliminated as an important category because they are structural constraints on social movements. Movement elites do not determine political opportunities though they might seek to take advantage of them. Mobilizing structures and framing processes, on the other hand, can be manipulated by social movement elites in order to increase the effectiveness of their organizations and hence constitute strategies that have costs associated with them that need to be detailed.

Mobilizing structures are "those collective vehicles, informal as well as formal, through which people mobilize and engage in collective action" (ibid). These things can presumably be manipulated by elites and, more importantly, the structure of organizations can to some degree be changed. Informal organizations can be changed into more formal ones, at a cost, and loosely knit organizations can be changed into more tightly knit ones. However, there are clearly limits to these strategies of organizational manipulation because social movements are characterized by their loose organizational structure. As Oberschall argues;

Framing processes are the "conscious strategic efforts by groups of people to fashion shared understanding of the world and of themselves that legitimate and motivate collective action" (McAdam, McCarthy and Zald, 1996; p.6). Broad definitions like these are notoriously difficult to work with so it might be wise simply to grant the concept wide range and conceive of it as anything ranging from the actual written ideology of a movement all the way to the culture of the movement, however defined. These things can be manipulated by elites in order to increase movement effectiveness. For example, Zald notes that

Boiled down, there are two principal categories of costs associated with an ideological recruitment function. The first of these two costs is best described as "idea" costs and the second as the costs of constructing institutional infrastructure. Idea costs are the costs of deciding on and modifying the ideology of the group whether it be something highly tangible like the written policy platform of the group or something highly intangible like the agreed upon cultural road-markers that bind a group together. Though these costs may at times seem great, they are not when compared to the costs of constructing a formal organization.

The second category of costs associated with ideological recruitment is institutional costs. These are the costs of constructing formal institutions that can lead the social movement or revolutionary group. Again, however, these costs are not high because ideologically recruited revolutionary groups and social movements of all kinds require only skeleton institutions to guide them. These institutions can provide group leadership, they can serve as institutions that confederate constituent groups, but they do not need to be expensive formal structures that keep track of members, control resources used to provide selective incentives for recruitment, structure leadership, or maintain chains of command between leaders and constituents. As will be noted in the following section, organizational forms of recruitment must do all these things and are therefore more costly to build and maintain.

Time:

Much of the more recent literature on revolution has addressed the Eastern European Revolutions of 1989. In these cases recruiting was predominantly ideological. Much of this literature has one unique observation in common. Authors have attempted to detail how it is that these revolutions could have occurred so rapidly when all previous analyses of revolution had seemed to indicate that revolutions were protracted struggles. Emphasizing things like informational cascades (Lohmann, 1994a), bandwagons (Granovetter, 1978), and related phenomena (Kuran, 1991), authors examining these cases have sought to provide foundations for understanding the rapid changes or revelations of individual preferences for regime changes in various different countries in this region.

What is interesting to note is that duration, or time, is not an important cost for the ideological recruitment strategy. Because rapid dissemination of ideas and preference changes are possible with this strategy (though certainly not guaranteed), revolutionaries can hope for, even cause to some degree, extreme shifts in individual preferences and rapidly threaten, at least according to this literature, incumbent regimes. In this sense the strategy can be conceived of as inexpensive when time is the cost measure being considered. In the following sections this conception of the costliness of time will be compared to organizational recruitment. Organizational recruitment, it will be argued, is costly because none of the rapid changes associated with the ideological recruitment strategy are possible. Indeed, no author has noted either cascading information or bandwagoning effects where recruitment has been predominantly organizational and there are no parallels in terms of rapidly shifting individual preferences when this strategy characterizes the case being analyzed.

Organizational Recruiting; Costs and Benefits:

Organizational recruitment, as noted in part one, is any form of recruitment that is not based on the policy platform of the revolutionary group or of the regime it opposes. Instead, constituents are solicited by a host of means, all non -ideological, and therefore the success of recruiting is a strict function of the organizing prowess of the group. Since, again, this is a negative definition, it gives wide range for different strategies of recruitment that could constitute an organizational form. The first task is to break down organizational recruitment into some useful subparts and then, as above, assess their relative costs. The following divisions are a short list of some of the most prominent categories that have been used by proponents of the organizational recruitment approach.

                1. The Use of Positive Selective Incentives:

A number of different authors focus on the issue of selective incentives in their analyses of revolutionary recruitment (Popkin, 1979; Taylor in Taylor ed., 1988; Tullock, 1971 and Scott in Scott ed., 1992). However, these analyses most often include both positive and negative incentives as two sides of the same coin. Revolutionaries can induce individual contributions to a public good through individual rewards or penalties. Migdal, however, concentrates his analysis of peasant participation in revolutionary activity on positive incentives. He states, for example that

Whereas Migdal focuses on positive incentives in general, Popkin argues that revolutionaries and other groups were most effective in organizing peasants in Vietnam when they used selective incentives to induce individuals to contribute to the provision of public goods. These selective incentives could be both positive and negative and it is difficult to determine to what degree one functioned over the other in facilitating peasant recruitment. However, it is clear that positive incentives played an important role in overcoming collective action problems and this variety of organizational recruitment was utilized much more in the beginning of Viet Minh recruitment efforts in the Vietnamese countryside than during later periods of the revolution. Recognizing the problems of free-ridership, Viet Minh recruiters provided material resources, leadership, and security to specific members of villages in order to facilitate the provision of public goods. Popkin argues, for example that political entrepreneurs had the ability to produce collective goods

There are two important things to note here. First, note that those who came to recognize the rebels as the legitimate government in their area did not do so because the dissidents presented them with the most compelling future vision of their country. They did not do so because the existing government’s policies were further away from their ideal policies than the rebel’s and they did not do so because the dissidents argued for land reform or collectivization. In this sense recruiting was clearly organizational.

The second thing to note are the costs involved in this form of recruitment. The recruiting group had to have the resources available to provide the selective incentives necessary to motivate individuals to act to provide important collective goods. The dissident group had to have the resources to place elites who could lead in all areas where it wanted to recruit and to sustain, at least over the short term, these recruiters. Furthermore, because organizational recruiting does not imply the "conversion" of those who benefit from the collective goods provided by the revolutionary challengers, the dissidents had to provide for a constant flow of these same resources in order to maintain the allegiance of recruits.

            2. The Use of Negative Selective Incentives:

Revolutionary recruitment does not rely, as it did for example with social movements, on voluntary membership. Because of this, a host of other recruitment strategies become possible and many of these rely on some form of sanctioning or coercion. Selective negative incentives range from the simple monitoring of potential free riders in a public goods provision environment and the consequent use of social sanctions to punish defectors, to forced draft. In this section three elements of negative selective incentives recruitment are covered that have been commonly dealt with in the literature on revolution; forced draft, terror, and the social sanctioning of free -riders.

Straight coercion is one of the most effective recruitment strategies for revolutionaries during certain periods of revolutionary organizing. Although forced draft has received little serious attention in analyses of revolutionary recruitment, it is common where recruitment has been organizational. In Vietnam, for example, forced recruitment drives of eligible men were routine once the Viet Minh had sufficient numbers and administrative capacity to form an irregular army. Berman, for example, notes that

Finally, most authors who have done extensive fieldwork of different revolutionary cases where recruiting was organizational detail ways in which revolutionaries recruited using various tools including more subtle forms of selective negative incentives. Popkin, for example, discusses the selective negative incentives used by revolutionary elites during the Vietnamese Revolution. By making small groups out of large ones, revolutionaries were able to more efficiently monitor free-riders and apply negative selective incentives against them.

            3. The Provision of Public Goods:

In some more recent work, various authors claim that they have discovered a subtle but new fact about rural recruitment by revolutionary elites. They claim that successful recruitment is dependent on the provision of certain kinds of goods by revolutionary elites (Wickham -Crowley, 1991; Isbell in Scott ed., 1992). The first characteristic of these goods is that they are public and the second is that they are the public goods that states normally supply.

This sounds a great deal like earlier analyses, Popkin’s in particular, that focused on public goods provision. It is, however, distinct. Popkin argued that selective incentives were used by revolutionaries in order to overcome collective action constraints that made the provision of public goods difficult. These authors, on the other hand, are not convinced by Olson’s reasoning and simply argue that revolutionaries recruit by providing the public goods that states normally do in areas where the state lacks presence (ibid). Selective rewards, and sanctions, do not form a significant part of the recruitment function as they did for Popkin. Instead, it is the revolutionaries’ ability to command the resources sufficient to provide these public goods -without recourse to motivating particular individuals- that allows them to attract recruits given various structural constraints like previous organization in the area of recruitment. The following paragraph puts some more meat on the public goods provision variant.

In analyses of revolutionary recruitment in Latin America, Wicham -Crowley notes at least four expensive forms of organizational recruitment he qualifies as elements of the collective goods set (though not all of the goods mentioned are really collective). Revolutionaries recruiting in rural areas first provide material benefits to peasants by purchasing goods at above market rates. They pay villagers for shelter, for transportation assistance and even bribe potential recruits. This part of the organizational strategy is said to resemble the welfare function states normally provide. Subsequently, revolutionary organizers provide for security in the areas they are attempting to control. They can, in some instances, establish a police force that can both defend the local populace against governmental reprisals and sanction community criminals. This part of the strategy is perhaps the most critical of the four. Here revolutionaries, as opposed to the state, provide security for peasants who are beyond either the desired or practical reach of the state. Further down the line, should policing be practical, revolutionaries can provide judicial institutions to support policing. Judicial institutions are the first concrete step revolutionaries take in their eventual efforts to build a truly rival state in a particular region. As Wickham -Crowley again argues, in successful cases "guerrillas may begin to organize formal governments, true counter -states, to control activity in a region" (ibid, p.44).

This variant of organizational recruitment, public goods recruiting, is the most expensive variant of the three categories described. In the "selective positive incentive recruitment" example, recruitment was expensive in terms of the goods that were required for the recruitment of individuals and the infrastructure necessary to monitor their disbursal. But in this last category, all of those things are necessary in addition to the formal institutions of police forces, judiciaries, schools, and in extreme cases even mini -states divided into formal sub -divisions. In Popkin’s scenario, constituents themselves provided the public goods necessary for recruitment though they were motivated by selective incentives (and penalties) as well as by career opportunities within the revolutionary party. But in this variant, it is up to the revolutionary elites themselves to provide the public goods without motivating individual constituents. In sum, two critical attributes of the public goods recruiting strategy make it expensive. First, local constituents do not provide the public goods that entice recruits. Instead, revolutionary elites themselves must provide these goods. And second, the public goods in question are the most expensive possible goods, i.e. the goods states normally provide (roads and schools being two classic examples).

Whether or not the provision of selective incentives and public goods, or the straight provision of public goods, is the critical recruitment strategy of revolutionary elites is not significant for the analysis proposed here. It is probably true that both these forms of recruitment are found in different cases during different time periods. In other words, which of the two forms of recruitment best describe revolutionary recruitment might very well be case dependent. What is important to note is that, in either case, recruitment is organizational and expensive. The institutional infrastructure required to provide selective incentives, or public goods directly, is great. In both cases, for example, there are the material costs of the goods themselves. Should these goods include security, this requires the ability to both monitor and sanction and this presupposes a staff and armaments however unsophisticated. The provision of a legal and administrative order is another example of an expensive institutional requirement for the organizational recruitment model and these costs clearly outweigh those described for the ideological recruitment strategy.

            4. Time:

Organizational recruitment is a long -term strategy. Because formal institutions need to be constructed, because members are solicited selectively, and because the strategy does not rely on the easy transmission of ideas in order to solicit constituents, the strategy must be implemented and nurtured over the long term. Revolutions that have used this strategy, Vietnam, China, Peru, and at different points in there trajectory Cuba and Nicaragua have taken years or decades to triumph. Contrast this with recent revolutionary movements in Eastern Europe, where victory came within months due to the rapid spread of anti-regime sentiment and the breakdown of foreign support for unpopular regimes, and it is easy to see the difference in these two approaches with respect to the duration of conflict. In Peru, the head of Sendero Luminoso warned his constituents that the revolution could take fifty to one hundred years (Tarazona -Sevillano, 1990; p.46).

Ideological vs. Organizational Recruitment: A Cost Based Comparison

The second part of this chapter has been concerned with an evaluation of two stylized forms of revolutionary recruiting, ideological recruitment and organizational recruitment. Because rational choice argues that individuals (revolutionary elites included) will make cost/benefit calculations with respect to their opportunity set, part II of this analysis detailed the costs of both these strategies. The most important results of this comparison were the radically different costs involved in one form of recruitment over the other. Ideological and organizational recruitment vary, above all things, in their differential costs of implementation. In one case, that of ideological recruitment, potential members are recruited by contrasting the policies of the revolutionaries with that of the current regime. The costs of the strategy are found in the contest between revolutionary elites over what the ideological appeal of the revolutionary group should be. There are also subsidiary costs involved in the construction of skeleton institutions that guide the movement but overall these costs are not substantial. In short, the strategy could be called a "cheap strategy."

In the case of organizational recruitment, four categories were examined that helped to organize the wide range of strategies that fall under the organizational rubric. In each case, the costs of the strategy were high. Whether, as in the collective goods variant, the costs of recruitment were found in the construction of formal institutions that could provide public goods without prompting constituent contributions. Or, as in the positive selective incentives variant, the costs of recruitment were found in the need for institutions that could provide resources to be selectively administered, the strategy could be called an "expensive strategy." The strategy is expensive because in each case a formal institution must be built and the building of formal institutions takes time. Pike best summarizes the formidable costs involved in the organizational recruitment strategy in his analysis of the Vietnamese Revolution. Speaking of Viet Minh formal institutions he states that

A RECRUITMENT MODEL FOR RATIONAL REVOLUTIONARIES

The following model provides the basis for subsequent reasoning on revolutionary strategy. It is, nonetheless, a preliminary presentation. Later some modifications will be suggested that make the model resemble reality a bit more. But the basic assumptions of the model and its reasoning will remain unchanged. Though revolutionaries rarely adopt exclusively ideological or organizational recruiting strategies, they do adopt mixes of the two and these are often mixes with dominant characteristics of one variant or the other. For now, however, a more simplified presentation will make the logic of the argument as clear as possible.

Unit of Analysis              Motivating Assumption                 Opportunity Set                  Costs                     Results

Revolutionary Elites                               ------ >                 1. Ideological                              Low                     Preferred
                                                    Rationality                             Recruitment
 

                                                                ------- >                  2. Organizational                     High                   Not Preferred
                                                                                                    Recruitment

From the model two immediate hypotheses can be suggested:

Hypothesis 1: Ideological recruitment strategies are first choice recruitment strategies for revolutionary elites.

Hypothesis 2: Where organizational recruitment strategies are found, they have been preceded by ideological recruitment strategies.

The importance of the above schema should not be understated. The theory argues that ideological recruitment is a first choice strategy for revolutionary elites. This hypothesis, however, would appear to contradict the very large number of revolutionary cases where organizational recruiting was the norm. Given that organizational strategies, at least in some parts of the world, have been the only successful recruitment strategies (Latin America and Asia for example), this is yet more striking.

At this point it is important to briefly set out what would constitute proof of the first two hypotheses. One possible strategy would simply be to examine a large sample of revolutionary recruitment strategies and code these as either ideological or organizational. This, however, would not be very precise. First, it would not constitute evidence for a preference ordering among the two strategies because one might object that one or the other of the two strategies had been tried prior to the observed strategy. Secondly, the likely discovery of a predominance of ideologically based strategies might be misconstrued as evidence for the first hypothesis even though it, again, would not necessarily establish evidence for a clear preference ordering. On the other hand, should it turn out that there be a predominance of cases of organizational recruiting, one would not be able to conclude that hypothesis one was false, or two true for that matter, because again, one strategy might have been tried before the other.

A second possible design would be to use a case based approach that randomly selected cases to be investigated. The problem with this design is that it might unduly confirm the first hypothesis. Because it is not feasible to do historically detailed case investigations on a large number of cases, a randomly selected small N study might end up with only cases of ideological recruiting, again if this strategy is dominant, and falsely conclude that hypothesis one was correct. What then is the best research design for the model proposed?

The solution to this problem is to challenge the hypothesis by selecting the most difficult cases, in other words, those where organizational recruitment strategies were used, and then verify that ideological recruitment strategies were tried before organizational ones. If this can be demonstrated, then there is reason to believe that one strategy is preferred over the other and that organizational strategies, due to their tremendous costs, are less preferred. This design has a dual utility. It can either confirm or disconfirm both of the hypotheses. If it is discovered that, where organizational strategies are found, there are no antecedent ideological strategies then both hypothesis one and two are false. If, on the other hand, where organizational strategies are encountered there are antecedent examples of ideological strategies, then both hypothesis one and hypothesis two are confirmed.

In the following chapter, four examples of organizational recruitment by revolutionary elites are selected. Following a justification of the cases based on the criteria mentioned in chapter one, the question of recruitment strategies is addressed and hypotheses one and two are tested by looking for prior ideological strategies that were tried, but subsequently failed, leading to the implementation of organizational strategies. Before proceeding to this issue, however, it will be useful to further define, to some degree, the model of revolutionary recruiting presented thus far.

Refining a Model of Recruitment

In preceding sections two separate recruitment strategies were detailed and, to this point, this analysis has treated these two strategies as discrete variables. Revolutionaries, it has been argued, might choose between either one of these two strategies and, because one was cheap and the other expensive, it was argued that revolutionaries have a dominant strategy, ideological recruitment. In reality, however, it is difficult to find pure examples of either the ideological or organizational forms. Many authors, for example, that analyze the civil rights movement in the United States are clearly analyzing predominantly ideological recruitment strategies. And yet, even in this case, these authors note that many group members, in particular economic elites, joined the movement in order to protect their status within the larger community. The thrust of this argument is that community status and economic status are connected and that economic elites might suffer, in economic terms, if they fail to take their proper place in the protest movement. Chong, for example, notes that

On the other hand, where organizational recruiting has been predominant, ideological recruiting is also present. In both Vietnam and China, for example, revolutionary recruitment was predominantly organizational relying on the provision of public goods. Potential recruits did not consider policy platforms as much as the potential each competitor had to provide for their physical security, or to threaten it, or to provide other important welfare improvements. Nonetheless, in both these cases ideology played an important role in the recruiting of all kinds of members not just revolutionary elites. Part of the Maoist organizational strategy, for example, used humble theater performances to educate potential recruits about the policy positions of the Communists (Snow, 1961). In Vietnam, recruits were frequently persuaded to become members of the Viet Minh because they preferred the policies of the challenger to those of the government. Pike notes, for example that

Revolutionary Recruiting as a Continuous Variable

There is nothing mutually exclusive about ideological vs. organizational recruitment functions. In fact, a revolutionary group might choose to use one or the other strategy at some point in their development or some mix of the two. Because both strategies can be useful depending on a large number of environmental constraints, what one normally finds is some mix of the two strategies. The most useful way of representing this idea is to place each of the ideal type strategies at the ends of a continuum. All real recruiting strategies fall somewhere between these two end points.

Organizational ____________________________________________________________ Ideological
Recruitment                                                                                                                               Recruitment

The usefulness of this strategy is twofold. First, as a tool of classification, it allows one to place each revolutionary group strategy at some point along the continuum. More importantly, however, where a revolutionary group falls along the continuum can tell a great deal about the strategic choices it will make in a host of other environments. For example, later it will be demonstrated that the thorny issue of revolutionary terrorism can be partly explained with respect to where a group lies along this line.

The most pertinent question for the argument is; what determines the strategic mix chosen by any particular revolutionary group? Here rational choice theory permits an understanding of an issue that has to date not been clearly addressed. Rational revolutionaries must choose the lowest cost recruitment strategy. This implies of course that they always prefer ideological recruitment. However, the simple fact that rational revolutionaries prefer ideological recruitment does not mean that this will inevitably be the most successful strategy. In fact, like most revolutionary strategies, ideological recruitment may fail or simply not yield the kind of results in terms of recruitment that revolutionary elites determine is necessary to defeat their regime opponent. When this happens, revolutionary elites face a choice. The first choice they can make is to continue organizing ideologically and hope for some change in the external environment that will help shift the preferences of individuals toward their position. This may seem implausible but it is actually a frequently embraced option. Revolutionaries all over the world have hoped for, and sometimes tried to cause, the intervention of foreign governments in order to take advantage of nationalist sentiments.

The second option for revolutionary elites is to quit. This might also seem implausible but does happen frequently. Communist parties all over Latin America, not to mention the Philippines, have become involved with militarily active revolutionary groups only to realize that the struggle would be too long, too dangerous, or both to justify the risk involved (Alexander, 1957; Kerkvleit, 1977).

Finally the third option is to embrace a new recruitment strategy in the hopes of obtaining better results. Revolutionary elites who do this move from an ideologically based recruitment strategy towards an organizationally based strategy. Whether it is because their policy positions do not have sufficient resonance among their potential constituents, because their potential constituents are not inclined to be motivated by policy positions, or because an organizational strategy is the only one available given other environmental constraints, this movement is what accounts for the range of recruitment strategies one observes among revolutionary groups.

As a final point, note that the cost/benefit framework presented yields another useful insight. Because ideological forms of revolutionary recruitment are cheap and organizational forms are expensive, revolutionary elites maintain some strategic mix of the two recruitment functions even when they are forced to move along the continuum from ideological to organizational recruiting. This explains the continued presence of patterns of ideological recruiting in revolutionary situations like those analyzed by Popkin. Ideology, though it might not in the end be sufficient to organize a successful revolution, is still preferred to organizational recruiting and is therefore maintained in the recruitment mix to the degree to which it continues to recruit at least some members, be they group entrepreneurs or simple constituents.

The following chapter addresses four revolutionary movements where revolutionary groups have adopted recruitment strategies that were predominantly organizational. For hypotheses one and two to be confirmed, antecedent ideological recruitment strategies that failed to yield sufficient results should be present. Should this turn out to be the case, there will be reason to move along one more step in the argument and suggest the first of the interesting dependent variables that can be deduced from the model of recruitment presented in this chapter.