This paper has two purposes. First, it attempts to focus attention on a neglected but important field of research in the social sciences: the study of educative propaganda. Some useful concepts for theory formation in this research area are explored and three competing models for the study of the relevant phenomena are tested. The models are tested with secondary data obtained from a comparative content analytical study. The other purpose is to draw attention to the practical ideological implications of the phenomena studied. This is done by pointing out the technocratic ideology inherent in our own analysis. This technocratic ideology is then contrasted with a suggested alternative: emancipatory ideology for educative propaganda.

1. Educative Propaganda: The Scope of the Field

Educative propaganda as a specific field of research can be delineated best by
indicating its relation to the more "basic" research areas in the social sciences.

First, the term "propaganda" brings us close to communication research. Obviously we are concerned here with the process, where - according to the Lasswellian cliche - somebody tries to transmit a message (information) somehow to somebody with some assumed consequence. This process is unfortunately too often seen to take place in a narrow dyadic relationship: actor x communicates to actor y and causes a change in y's behavior. It is too often forgotten that the communication system of x and y can be a subsystem of several other systems, and the potential behavioral change of y is determined by, and has a feedback to, not only the relationship between x and y, but on the supersystems it is a part of. The neglect of the broader approach has no doubt been due to the specific application orientation of the researchers: the focus has been on the power relation between x and y (e.g. the impact of political propaganda), the efficiency of the communication (e.g.

the communication studies in formal organizations) or the dysfunctions of the content
of communication (the study of the impacts of mass media).

The idea of "education" in educative propaganda necessarily leads us to these broader considerations. It is obvious that here the communicator does not communicate for his own benefit, but the goal of communication is to change the state of a specific supersystem via changing the behavior of y. Usually, although not always, this means a close bond between the political sector (decision-makers) of the supersystem and the communicator (x).

However, the "educative" emphasis in the concept of "educative propaganda" usually also implies a certain neutrality in respect to politics. The message of educative propaganda is not assumed to advocate the legitimacy of the power relation in the supersystem, or the differential manning of the power positions. This is the core of the distinction between political and educative propaganda. This distinction is naturally rather unsharp.

First, the changes advocated in educative propaganda may have a feedback to the political sector, and consequently the decision-makers may try to control and influence the content of educative propaganda. Thus, on a societal level for example, education for temperance may (via moral indignation) favor conservative parties. Secondly, certain strategic educative propaganda cannot be politically neutral. The best example on the societal level is probably "civic education" about the functioning of the political system. By taking existing political systems for granted it is also politically committed to its preservation. However, even in these two respects one can speak about more or less "pure" educative propaganda especially if the broader ideological implications that we will discuss later are neglected.

Educative propaganda research also has close connections with the socialization studies of political science and sociology. Like socialization studies, educative propaganda research is also interested in the transference of some social values from those who adhere to them to those who do not. There is, however, a definite difference in perspectives. Socialization studies emphasize primary socialization that takes place in the family, peer groups and educational system. In the study of propaganda in general and educative propaganda in particular the values and behavioral patterns obtained by these "automatic" methods are assumed to have been changed. Thus, especially in the case of educative propaganda, we are concerned with a process that is auxiliary and corrective to the automatic processes of primary socialization. Educative propaganda begins with the already established value system of its recipients, selects some limited areas thereof, and tries either to introduce new ideas and objectives to be evaluated or change the rank ordering of the old ones.

From the point of view of political science, the power difference involved in any kind of communication is focal. The initiator is always potentially more powerful than the recipient: the latter cannot but receive. Of course, in symmetric communication the power potentials in the x-y dyad oscillate, depending on who at any given time happens to be the communicator and who the recipient. Also, any communication starting as a symmetric one may, and most often will, end as asymmetric,

e.g. in the case of scientific dialogues or political discussions. Thus most
communication could be labeled as a battle of power of varying intensity.

Extreme types of communication that aim at establishing new values are psychoanalysis and Pavlovian conditioning. These are also examples of extreme asymmetric communication where the power position of the communicator permits the use of extremely powerful negative sanctions.

In the case of educative propaganda a certain degree of asymmetry in communication is already inherent in the very term "educative": someone is assumed to be in a legitimate position to initiate the communication aimed at value changes in a supersystem as a whole or in a sector thereof. At the same time, asymmetry is also played down: no actual norms are dictated and no extra-communicational negative sanctions are used. The message is exhortative and aims at certain value changes, but the changes are assumed to be attained through rational persuasion on the part of the communicator and voluntary compliance of the recipients.

The subject matter of educative communication can now be roughly defined: in any social system, any asymmetric, legitimate communication that is politically neutral (does not aim at changes in the power relations) and aims at value changes that are corrective and complementary to the value system produced by primary socialization can be labeled "educative communication".

2. Administrative Setting, Legitimation and the Degree of Suggested Change in the Dissemination of Educative Propaganda

As stated above, in the study of educative propaganda - as in any communication research - the focus may either be on the communicator, the message, the channel or the recipient or any combination of these. Also, the research may focus to varying degrees on the "supersystem" where the communication takes place. This supersystem of communication can be considered either as the environment of the communication system, and thus be treated either as a set of independent variables that determine the characteristics and functioning of the communication system, or interpreted as a set of dependent variables that are determined by the effects of the existence and functioning of a certain type of communication system. Both approaches can be connected by considering the total chain of causal impacts from the causational environmental characteristics of the supersystem to the existence and functioning of the communication system and its feedback into the original environmental characteristics. Our present concern dictates that we begin with some concepts that focus on the characteristics of the communicator and the message, return to the supersystem determinants of these characteristics and then consider the different communication strategies from the ideological perspective they imply.

We cannot here discuss all the aspects of the communicator and the message, but we can first pick up some basic structures of both, then suggest some general variables that characterize these structures and then consider those variables representative enough for testing the models we are interested in.₁

We can take one structure of communication, i.e. the administrative setting, and
two structures of message, i.e. the type of its legitimation and the degree of change
aimed at in it.

Concerning the administrative setting, two variables seem especially relevant. One of these has been already implicitly discussed when we noted the importance of educative propaganda for the decision-makers of the supersystem. Thus the first variable denoting the administrative setting is the closeness of the communicator to the "political" sector of the supersystem and its decision-makers. Here of course the range may go from total dependence to total independence. In the former case the decision-makers themselves or their immediate subordinates in the hierarchy (e.g. an administrative organ of the decision-makers) are the communicator. In the case of total independence, the relation is that of tolerance, neutral approval or total indifference from the point of view of the decision-makers.

The second variable to be considered here is the hierarchical unity of the administrative setting of the communicator (the degree of centralization). Here the variation may range from a strict monolithic communication system through functionally or regionally decentralized, to a totally unorganized, system where individuals and groups can legitimately disseminate propaganda but no cooperation between the diverse groups of communicators exists.

The legitimation of the message in appeals for change can be classified in terms of Weberian types of social action: affectual appeals, traditional appeals, appeals to absolute (moral) values and appeals to efficiency for achieving specific concrete goals.² All these categories can be considered in the analysis of a message as separate variables and they can be operationalized as the number of appeals to emotions, traditions, moral values and knowledge of goals for changing the values or behavior. Weber was discussing these categories in his typological approach as mutually independent theoretical dimensions, but if they are used as empirical variables and operationalized as indicated above they will no doubt correlate to a certain extent.

In the case of the degree of change desired by the communication, the traditional classification of the process of acceptance and adoption of innovations suggested by studies of interpersonal influence can be used. For a value and behavioral change, the recipient must first become aware of his possibilities to change, next he must evaluate the potential change, then he must adopt it and finally become "routinized" in acceptance of the change. Any communication advocating a change can focus on some or all of these phases. Educative propaganda can thus emphasize informing the recipient about possibilities to change his values and behavior in a certain area, it can emphasize the need for assessing the positive values in the change (or negative values of not changing), it can emphasize that the recipient should actually try to change his values and behavior and finally, it can emphasize the necessity of maintaining the already accepted new values and behavioral patterns. All these categories can be operationalized in terms of the content of the message: the frequencies of statements aiming at awareness by factual information about potential new values and patterns of behavior, exhorting their evaluation, e.g.

in terms of their assumed consequences and suggesting their adoption via trying or finally advocating the implementation through continued use. Again this classification can be used in each category as a separate variable (counting presence or absence of statements in them) although they may also intercorrelate.

These variables define operationally the main characteristics of three important structures of the communicator and message. They were selected more or less randomly (as were the structures themselves) and obviously many other variables — as well as structures - could have been chosen for testing our models. They were chosen mainly because they were already coded into the secondary content analytical data. On the other hand - as our analysis will indicate - they are sufficient and representative in the sense that other similar variables denoting either these or other structures of the communicator and message can be assumed to behave in the same way in connection with the models suggested in the following section.

3. Supersystem Change and Formal and Substantive Rationality in Educative Propaganda

The above static discussion of three structures and variables characterizing them may give the impression that variations along the variables depend solely on the rational choice of the communicator. However, taking the original definition of educative propaganda into account, it can be assumed that the communicator himself reacts to some "needs" of the supersystem undergoing or aiming at change. Thus it is natural to investigate the variations in the variables as conscious reactions to changes in the supersystern. For that purpose the process of change in the supersystem must be defined first in general terms and then the variables of educative propaganda can be connected with it by the model(s).

From the point of view of the empirical data, it is no use trying to keep the concept of supersystem change either general or continuous; it must be looked at from an evolutionary and typologizing point of view. As a starting point, on the basis of aspiration theories, the rate of planned change can be considered as the function of achievement in the field where the change takes place. This can be crudely stated in terms of four stages: 1) at the pre-take-off stage, the low level of achievement prevents the change, 2) at the take-off stage, slight improvement in achievement speeds up the rate of change, 3) in the active stage, high rate of change has led to high achievement and 4) at the cool-off stage, the saturation in achievement leads to slackening down the rate of change.

It is not necessary here to spell out in detail the psychological assumptions usually associated with aspiration theories. Supersystem change can be here perceived as a more or less planned attempt (project) to change some central aspects of the supersystem. What is more important is to try to outline the role and strategies of educative propaganda at any of the stages of development.

The "corrective" nature of educative propaganda in respect to system change

has already been indicated. This obviously means that educative propaganda focuses on some unplanned element in the total change and tries to control or eliminate that aspect by aiming to change the values and patterns of behavior in the supersystem. The strategies affecting these changes can best be analyzed in terms of the dichtomy "formal/substantive rationality". Formal rationality in the communication of educative propaganda means the arrangement of administrative setting and formulating the message in terms of immediate goal achievement (i.e. attainment of value and behavior change of the recipient (s)), while substantive rationality pays attention to the unintended consequences of the planned change of values and behavior patterns and tries to establish stable conditions for change by connecting it to those social mechanisms of old that determine the values and behavior patterns of the recipient (s).

This abstract formulation becomes more concrete by relating it to the earlier
conceptualization of the characteristics of the communicator and the message.

In terms of the administrative setting, the attempts at tightening the relationship between decision-makers and the communicator and the centralization of the organization of the communicator indicate an emphasis on fast goal attainment by increasing the rationality of the communicator itself. Decentralization of the administrative setting on the other hand indicates a willingness to fortify the relationship and tighten the ties of the communicator with the recipients by letting the administrative setting associate with the existing control mechanisms of values and behavior patterns. Thus close control and centralization in the administrative setting is an indicator of formal rationality in the communication of educative propaganda, while attempts at an organized but decentralized administrative setting can be perceived as indications of attempts at substantive rationality. In the case of legitimation of the message, appeals to rational grounds (e.g. experts as authority, demands for realization of plans) can be considered as indicators of formal rationality, while appeals to traditional and affiliative grounds can be considered as indicators of substantive rationality. In case of degree of change, suggestions for awareness and adoption can be considered as indicators of formal rationality, while suggestions for implementation can be considered as indicators of substantive rationality.

Now these indicators of formal or substantive rationality in educative propaganda can be expected to vary according to the needs of the supersystem. Obviously, a high rate of supersystem change leads to an acutely felt need for corrective actions and its fast achievement, and thus to a great emphasis on formal rationality in educative propaganda. Similarly, achieved supersystem change leads to emphasis on stabilization of achievements and corrections of the potentially disruptive side effects, and thus to a greater emphasis on substantive rationality.

Formal rationality in educative propaganda can thus be assumed to be a function of rate of change, and substantive rationality a function of the level of achievement effected by the change. This assumption, however, is not_tyet enough, because it does not take the time perspective into account. Educative propaganda can be oriented either to the past, present or future state of the supersystem. Thus we will have

three different models. The first can be called a retroactive model, where formal and/or substantive rationality at any stage of development is the function of the change and development of the previous stage of the supersystem. The other model can be called reactive, where the formal and substantive rationality of educative propaganda is the function of the change and development of the present stage of the supersystem. The third model can be called anticipatory, i.e. the formal and substantive rationality of the educative propaganda is the function of change and development of the future anticipated stage of the supersystem. Supersystem changes can be classified according to level of achievement, and three sets of competing hypotheses concerning the appearance of formal and substantive rationality in the organization and message of educative propaganda can be derived.

4. Societal Change and the Formal and Substantive Rationality of Societal Educative Propaganda

The actual meaning of the above rather abstract analytical discussion can be illustrated by examining the relationship between societal change and level of achievement (development) and the formal and substantive rationality with data on educative propaganda aiming at controlling and correcting the change. The data used here are taken, with some changes, from a cross-cultural content analysis of written birth-control material carried out in 1960.3 Because of classificatory difficulties, the present analysis is based on materials from 23 countries (in the original research, 32) and includes 159 analyzed pieces of written material. The countries were classified according to rate of change between 1935 and 1962, and the level of development in 1957 into four groups: low development - low change, low development - high change, high development - high change and high development - low change.⁴ Since the concepts of structure of administrative setting, type of legitimation and type of suggested change are used here in the same way as in the original content analysis, the tables are presented by combining the categories to correspond to the dichotomy of formal and substantive rationality.

On the basis of the theoretical assumption above, three different sets of hypotheses can be derived. Because of the heterogeneity of the combined categories we focus on the highest percentages of formal and substantive rationality in each category.

First, on the basis of the retroactive model, we could predict that formal rationality is highest at the high change - high development stage and low change - high development stage, and substantive rationality is highest at the low change - high development stage.

Secondly, on the basis of the reactive model, we could predict that formal rationality is highest at the high change - low development stage and high change - high development stage, and substantive rationality is highest at the high and low change — high development stages.

Thirdly, on the basis of the anticipatory model, we could predict that formal rationality is high at the low change - low development and low change - high development stages and substantive rationality is highest at the high change — low development stage.

Table I presents information on the organization of administrative setting. Here, control and centralization can be interpreted as an indicator of formal rationality, and decentralization as an indicator of substantive rationality. The data definitely support the reactive model in respect of formal rationality, and the retroactive model in respect of substantive rationality.

We can turn to the legitimation of the message (Table II). Here legitimation

by experts can be interpreted as an indicator of formal rationality, and legitimation by tradition or majority affiliation as an indicator of substantive rationality. The data support the reactive model both in the case of formal and substantive rationality.

In Table 111 we can examine the type of suggestion for change. Here, awareness and adoption can be classified as indicators of goal-oriented formal rationality, and exhortation for implementation (for continued use) as indicators of substantive rationality. Here too, we can detect a reactive pattern for formal rationality and a retroactive pattern for substantive rationality, although they are somewhat dimmed at the low change - low development stage.

All in all it seems that among the three possible models there is evidence that formal rationality in educative propaganda is closely bound with the rate of change reactively, and substantive rationality is bound with development retroactively (also probably assuming low change). These results are of course by no means conclusive because of the fragmentary nature of the data and primitive level of measurement. The analysis, however, does indicate in what direction the study of educative propaganda could evolve.

5. Practical Considerations and Ideological Commitments: Technocratic and Emancipatory Perspectives of Educative Propaganda

The nature of educative propaganda also raises some practical concerns and
problems of wider ideological perspective in respect of its dissemination.

The practical concerns can be readily understood by the question of the efficiency
of the detected patterns of dissemination of societal educative propaganda. The

problem of efficiency is twofold. First, one can ask to what extent formal and substantive rationality should be emphasized at different stages of the supersystem in order to maximize both the desired value and behavior change and still maintain stability in change. Secondly, one could ask whether educative propaganda should be retroactive (i.e. react to past stages of supersystem), reactive (react to present stage of supersystem) or anticipatory (react to anticipated future stage of supersystem) in order to realize both change in values and behavior and stability in the change.

Needless to say, the present data do not yield an answer to these questions. The analysis would demand a systematic study of the effects of propaganda using different strategies at different stages of societal change and development. But the very fact that this type of question can be asked is indicative of both the phenomenon studied and the models used for studying.

In order to understand the broader ideological commitment underlying the type of educative propaganda we have been discussing here, we can return to our original dyad of x (communicator) and y (recipient). In defining educative propaganda we have freed the communicator from the concern of politics proper. At the same time he became closely bound with the idea of "rational" system change. However, whatever model of communication proves to be valid, the asymmetry between x and y remains, not only in respect of who can legitimately communicate with whom, but also in respect of who defines the rationality in system change.

It has been pointed out above that the legitimacy of the communicator in his "corrective" educative communication often rests on the assumption of his superior information and expertness in respect of the needs for value and behavior change that the supersystem requires. But even if both partners in the communication agree upon this, there still remains a deeper kind of asymmetry in the differential possession of information. Besides having more information, the communicator also has (or at least thinks he has) a specific type of information that makes the real difference. He has information about the recipient and the factors that make his values and behavior relevant from the point of view of system change. And it is the manipulation of these factors (e.g. traditions, moral values, personal preferences) that the communicator aims at in order to effect the needed changes in values and behavior. This information asset is even more pronounced if we look at it from the recipient's point of view. He may be conscious of the system change and its implications for himself, but he may be unaware of the relationship between his individual values and behavior or of the factors that determine them. And he is definitely not conscious of the (more or less) scientific strategies that the communicator may use to change his values and behavior nor about the potential "deeper" motives that may determine the choice of strategies.

This idea of qualitatively differential possession of information makes the originally assumed "legitimate" asymmetry of the communicator and the recipient more problematic than assumed. But what is more important, it also discloses two potentially different ideologies behind the dissemination of educative propaganda. One of these could be called technocratic: the focus is on effective manipulation of

the recipient on the basis of the communicator's superior knowledge about him in order to "correct" his values and behavior and thereby correct the system change. As a counter-ideology from the same differential of information, another ideology of communication can be derived that could be called "emancipatory".5 This means that the educative propaganda transmits the recipient information about the relationship between his values and behavior and the system change, and also about the factors that determine the values and behavior relevant for the system change. Then the corrective change in values and behavior (if such would actually take place) would be based on the recipient's new definition of his relation to the system change and a subsequent "voluntary" change in values and behavior.

Of course one can question the efficiency of educative communication based on "voluntary self-correction". What if the recipient does not actually change his behavior at all, or corrects it in the "wrong" direction? The answer is that we can, in general, question the communicator's right to define the "right" change in values and behavior, and we can question it still better if the differences in the levels of possessed information are still reduced via emancipatory communication.

Analysis of any type of educative propaganda will no doubt show the predominance of technocratic rather than emancipatory ideology. For instance, in the content analysis data used here, no category of information could be found that could be labeled emancipatory, although the researcher has extensively coded a vast array of different types of information contained in the materials. Obviously the emancipatory idea is either very little known or not relied upon.

The two ideologies of educative communication also have implications for the research focusing on it. The predominance of technocratic, manipulatively oriented educative propaganda tends to tempt the researchers to use such concepts and models as "goal attainment", "maintenance of stability" and "level of aspiration". Within such conceptual frameworks the; recipient will inevitably be viewed as an object for manipulation. Our own analysis above is a good example of this effect. This type of research in turn feeds back to practice and reinforces the technocratic ideology in communication.

But are there alternative concepts and models that could be used? Without deeper analysis one could suggest that maybe one should shift the research focus on to the recipient, analyze his level of information, its social determinants and his ability to use new information,, first independent of the idea of educative communication, and then analyze the manipulative or emancipatory effects of educative propaganda against this background. This kind of information could then be expected to have an "emancipatory" effect on the communicators of educative propaganda and result in switching their operations to less technocratic bases.

With respect to our own "technocratic" analysis above, it is obvious that the concepts and models were chosen partly in order to give a good starting point for the analysis of the ideological bifurcation underlying the educative communication. We cannot attempt to change it toward a more emancipatory direction, but we hope that the corrections on the meta-analytical level of this last section may be sufficient.

---

NOTES

1 The variables suggested here have been picked up and elaborated from the study from which the empirical data are utilized, see Veronica Stolte Heiskanen, "A Cross-Cultural Content Analysis of Family Planning Publications," in Donald J. Boque (ed.), Sociological Contributions of Family Planning Research, Community and Family Study Center, Chicago, 1967, pp. 81, 83, 90.

2 See e.g. James March and Herbert Simon, Organizations, New York: John Wiley & Sons, 1958, pp. 48-52.

3 For a discussion of data collection, coding and preliminary analysis, see Heiskanen, op. cit., pp. 77-80.

4 Classification of the countries in respect of change and development was carried out according to criteria of Feierabend, Feierabend and Boroviak. See Ivo K. Feierabend, Rosalind L. Feierabend and Darlene L. Boroviak, "Empirical Typologies of Political Systems: Aggressive Prototypes", paper read at Seventh World Congress of International Political Science Association, Brussels, 1967, pp. 10, 11, 54-55.

6 The term has naturally been borrowed from Jiirgen Habermas, see e.g. Jiirgen Habermas, "Knowledge and Interest," Inquiry, 3, 1966, pp. 285-299.

NOTES

1 The variables suggested here have been picked up and elaborated from the study from which the empirical data are utilized, see Veronica Stolte Heiskanen, "A Cross-Cultural Content Analysis of Family Planning Publications," in Donald J. Boque (ed.), Sociological Contributions of Family Planning Research, Community and Family Study Center, Chicago, 1967, pp. 81, 83, 90.

2 See e.g. James March and Herbert Simon, Organizations, New York: John Wiley & Sons, 1958, pp. 48-52.

3 For a discussion of data collection, coding and preliminary analysis, see Heiskanen, op. cit., pp. 77-80.

4 Classification of the countries in respect of change and development was carried out according to criteria of Feierabend, Feierabend and Boroviak. See Ivo K. Feierabend, Rosalind L. Feierabend and Darlene L. Boroviak, "Empirical Typologies of Political Systems: Aggressive Prototypes", paper read at Seventh World Congress of International Political Science Association, Brussels, 1967, pp. 10, 11, 54-55.

6 The term has naturally been borrowed from Jiirgen Habermas, see e.g. Jiirgen Habermas, "Knowledge and Interest," Inquiry, 3, 1966, pp. 285-299.