Moritz Rosenmund is head of the R&D department of the Pestalozzianum, Beckenhofstr. 31, CH--8035 Z¸rich, Switzerland (e-mail: moritz.rosenmund@pestalozzianum.ch). Most recently he has been engaged in projects on both school development as an adaptation to growing multiculturalism in Swiss schools and on participation in curriculum-making. This latter project compares curriculum-making in Switzerland, Germany, and the Nordic countries.

JCS invites comments responding to the views in this paper. Such comments should be addressed via e-mail to westbury@uiuc.edu.

Copyright © 2000 Taylor & Francis Ltd. ISSN 0022-0272. Copies may be made under the normal terms of copyright law.
 

 We have become familiar with international, global comparisons. We compare the GNP of industrialized and developing countries. We look at unemployment rates within the OECD frame and at the relative stability of political regimes, or emissions of CO² per capita on a worldwide scale. Or we compare expenditure for education per capita or lower secondary test scores in mathematics and science. All such comparisons employ elaborate indicators and sophisticated measurement models. These indicators usually allow us to come to grips with significant states and processes -- or even with differences between states and processes.

 We have dreamed of such cross-national comparisons since the 1950s and 1960s when a dominant, over-simple paradigm of social development assumed that modernization would occur in a linear way on a worldwide scale, i.e. as a sequence of necessary steps. We have, of course, become more skeptical about this starting point: we know that societies do not develop according to a universal mechanism; we know that there is no Îsingle best wayÌ. Most importantly, we think we understand that a sequence of initial conditions, observable processes and effects that might occur in a particular society does not necessarily follow a common, universal pattern or logic.

 However, we still feel uncomfortable with the idea that all these sequences of initial societal conditions, trajectories, and impacts are society-specific -- with the implication there can be no basis for comparison across societies. In fact, there are some persuasive arguments against this view: thus John Meyer and colleagues (see, e.g. Benevot et al. 1991, Kamens et al. 1996, Meyer and Baker 1996, Meyer et al. 1997) have shown that even the curriculum processes within highly dissimilar countries have converged in the last few decades, resulting in a few typical curricular patterns.

 Such considerations bring us to the general problem of this paper, i.e. how feasible is it to compare curricular processes internationally and to link -- nationally or internationally -- research across national borders? In what follows, I will explore these question within the frame of a tentative theoretical outline. I shall proceed in three steps. In a first step, I will consider the object of curriculum research itself. I will propose a distinction between the notions of 'curriculum process' and 'curriculum-making processes' and point to a tension which is inherent in such processes. Curriculum process and curriculum-making processes always deal with the relationship between 'the general' (or universal) and 'the particular', between aspects common to any society, and other aspects which are specific to a single case only. In a second step, I will review some typical approaches to curriculum research, trying to allocate them along the lines of the tension I identify. This paper will conclude with some explorations of the methodological problems resulting from arguments for international comparisons.

 Two analytic levels: curriculum process and curriculum-making processes

We must clearly distinguish the level of the curriculum process as a process which refers to a society as a whole, and the level of curriculum-making processes as an institutionalized practice of educational administrations. As Rudolf K¸nzli (1999) has suggested, the curriculum process (see figure 1) is 'the complex social process of selection and weighing of socially available knowledge and know-how [A] aiming at concrete educational practice and expecting possible educational outcomes' [D]. It is also obvious that a society will select and weigh knowledge and know-how in the light of other societal conditions [B]. And we may reasonably assume that a society will examine knowledge and other conditions in considering its own future [C] -- whether in the sense of a future simply to be ÎexpectedÌ, i.e. a future to which one has to adapt more or less fatalistically, or a future conceived as something to be actively shaped -- last but not least through the curriculum-making processes. Put in another way, the curriculum process combines given knowledge and societal conditions [A, B] with the anticipation of future societal states in general and future forms of education in particular [C, D]. The curriculum process is, in this sense, part of the development of a society which moves toward the future, which is itself aware of this movement, and which is seeking to come to grips with given stocks of knowledge and given conditions. Curriculum research analysing the curriculum process on this level is part of research efforts which deal with the development of whole societies.
 

 Figure 1. The curriculum process

But, of course, the two factors we have referred to as 'givens' -- knowledge [A] and other social conditions [B] -- are internally differentiated configurations:

When seen from a given historical and social perspective, socially-available knowledge is divided, generally speaking, in two components. On the one hand, we have general, universally accepted knowledge as it has been accumulated by scholarship. On the other hand, however, there is the particular knowledge which a specific society has accumulated in the course of its history. This knowledge, which may assume the form of traditions or simply of practical experiences with the problems and situations specific to one society, is part of the particular culture of this society.

The same holds for other contextual conditions. Thus we have to make a similar distinction between conditions and developments characterizing world-society -- the often-mentioned globalization process -- and the conditions and developments which are specific to a given society and which hold for that society only.

 Hence, the curriculum process within a particular society always instantiates a tension between a universalistic and a context-specific pole (see figure 2). The curriculum process mediates between available knowledge and significant social conditions, and between present conditions and expectations about the future. As the curriculum making process weighs knowledge and social conditions in the light of a possible future, its outcomes always become decisions about the relative importance to be attributed to the universalistic and the particularistic.

 Moving now to the second level, I suggest that curriculum-making processes are integral parts of the curriculum process described above. They are rooted in societies' awareness of their own development and in their weighing of knowledge and contextual conditions around this development. Societies react to this situation by developing institutionalized forms of regulation, i.e. by developing structures and procedures which channel of the process of weighing the different factors.

 Curriculum research has analysed the structures and procedures of this institutional regulation from several different perspectives. It has sought to explore the black box in which the 'translation' of knowledge-stocks and contextual conditions takes place. Let me summarize four of curriculum research's perspectives:

• administrative rationality (organization theory);

• socio-structural perspective (class theory, stratification theory);

• social cohesion (theory of societal systems);

• world-system approach (world-system theory).

These perspectives support the premise that the term 'regulation' can be understood in narrower or broader senses.

A first perspective, the perspective of administrative rationality, approaches the regulation issue in a very direct way. In fact, historically the institutionalization of education, the family, and -- in part -- the state has created a regulation problem inasmuch as different discourses, initially parts of one unified discourse, have become separate and autonomous. In particular, discourse about the future of society and the knowledge to be passed to the next generation on the one hand and discourse about appropriate contents and forms of educational practice in schools on the other hand have become separated -- but at the same time they are, of course, inevitably linked. As Hopmann (1988) has shown, this double-edged process is what happens in modern societies within the framework of curriculum administrations which perform exactly these two functions. They separate -- at least partially -- the discourses and at the same time couple them within the frame of an administrative curriculum-making process.

 If it is regulatory endeavours which, within the framework of the administrative-rationality perspective, focus autonomous, partial discourses. In the socio-structural perspective, it is the structures of society and of education which must be adjusted to each other. Bernstein (1977), in particular, has pursued this perspective and shown that the borders within the educational system -- between subjects or disciplines -- are closely related to the borders between the structural positions of different categories of teachers, and even to the allocation of students within the larger social structure. Curriculum-making processes, in this second perspective, have to do with the maintenance or change -- mostly the maintenance -- of status-structures. They are a means of reproducing social structure.

 Social reproduction is also one of the key concepts of the social cohesion perspective. This perspective reflects the idea that whole societies are composites of institutional arrangements: structural relationships, institutional arrangements, and their symbolic representations form an integral whole, a totality. Within this perspective curriculum-making processes are related to the prerequisites of societal integration -- specific to a given society -- to be realized on both the levels of material conditions and symbolic representations. The important issue, according to this perspective, is the bargaining that occurs about the shape of education with respect to a society's structure and symbolic representation. Hence, curriculum-making processes treat the 'totality' of a given (single) society; they are processes which concern 'everyone'. They are also processes which accentuate the borders between a society and its surrounding societies. It is obvious that these ideas come close to what we have discussed above as the curriculum process.

 The world-society perspective is, in a sense, the global version of the social cohesion perspective. The empirical point of departure for the world-society perspective is the observation that there is a tendency for curriculum-making processes to converge on a worldwide scale. In other words, curricula as well as the formal structures of education are becoming more similar across countries. This holds in particular for primary and lower secondary education, but also -- with some exceptions -- for upper secondary education. Meyer and his colleagues (1997) explain this phenomenon by invoking a supranational level of social organization. They assume that there is a relatively autonomous world-society exerting an influence on single national units, even in regard to education. The actors at the systemic level in individual nation-states invoke universal values, such as human or civil rights, socio-economic development, or 'education', when they seek to legitimize themselves. The diffusion of these values is not the result of a clearly targeted policy of creating a 'world state' but is a result of a tight network of international organizations and associations with which individual countries are connected by way of the mediation of state agencies or extra-governmental organizations (Meyer et al. 1997).

 I would not claim this selection of four perspectives to be exhaustive. And I do not want to express any preference for one of them; but each one allows us to understand one or another aspects of phenomena within both the curriculum process and curriculum-making processes, although for the explanation of other phenomena another perspective would be more appropriate. For instance, the structural and the social cohesion perspectives may explain the high public interest in curriculum-making better than the administrative-rationality perspective. The latter, however, seems to be more illuminating when applied to the massive involvement of the political system in curricular processes -- in harmony with the political systems of other countries and, as we have seen, with convergent effects.
 
 

 

Figure 3. The four perspectives along the particularism--universalism continuum

 If we seek to associate the perspectives with the axis between a culture-free and a context-bound pole, a picture, as shown in figure 3, may result. If we move to either extreme, we would expect to find the cohesion approach and the administrative rationality perspectives. The location of the former seems obvious if we consider again the accentuation of the borders of a given unit that is related to this perspective. On the other hand, the rationale of administration seems, at first glance, to be inherent to the functional mode of administrative practice, and it is only by the application of this functional mode to a specific context that administrative practice produces a phenomenology which differs between countries.

Methodological problems

The conflict between context-bound and culture-free approaches has important implications for research methodology. Thus in the sociology of work, the culture-free approach has given rise to a Îcross-nationalÌ research (Grootings 1989) which focusses on the general features of organizations and neglects the context in which an organization is located. 'International comparative research' on the other hand argues the other way round: it assumes that the organization of work can be explained by the historical -- cultural and social -- conditions particular to a given societal context -- and only in this context. There is no reason why this distinction from the sociology of work should not be applied to curriculum research.

 The same may be said with respect to Heidenreich's (1991) discussion of the problem of how comparative research might reach generalizable statements. If we assume, according to the culture-free approach, that contextual conditions can be ignored, quasi-experimental research designs similar to those used in the exact sciences would seem reasonable. However there are several arguments against this kind of reasoning. First, it is not clear if and how we can identify 'comparable' cases across countries. Still more important is the question of whether the relevant concepts and variables can be defined and operationalized while ignoring the context. For instance, in the context of the sociology of work, the term 'foreman' has significant differences in meaning across France, England, or Germany. And, finally, the concentration on static parameters impedes the inclusion of dynamic relationships between social actors in the explanatory model. It must be assumed that international comparative research on curricular processes would face the same obstacles and problems as it follows the culture-free model.

 The context-bound methodology -- 'holistic' research design -- seeks to seize observable relationships and to reconstruct the rationale of the strategies, reasoning and regulatory systems behind them. The latter are seen as closely connected with the characteristics of the societal context where they appear. In this sense they cannot be transferred to other contexts.

 In the light of this contrast, then, international comparative research is confronted by a fundamental dilemma: If it follows the first approach it has to expect difficult obstacles and not very differentiated explanations. In the other case it has to limit itself to the highly elaborated and convincing reconstruction of totalities, which it cannot apply in a comparative way.

 As a conclusion, I would like to sketch two possible ways out of the dilemma. A first trace has been prepared by Heidenreich (1991). Recalling Max Weber's concept of ideal types, he would suggest -- transposed to curriculum research -- the reconstruction, as a first step, of the inherent rationale of curricular processes in at least two countries as distinct forms of national curricular processes, and their explanation in terms of the interdependency between context and curricular processes that characterizes the set of countries. Types of regulation and patterns of relevant contextual conditions may be inferred in a second step. The third step would look for the particular historical and institutional factors which would explain how the types and patterns could be varied and combined in other societal contexts. Thus, the ideal types would serve as a kind of Îcontrast substanceÌ to explore the situation of the society where they are applied.

 In addition to this heuristic approach, a second approach, developed in connection with the problem of designing survey instruments for IEA Third International Mathematics and Science Study (TIMSS) -- and termed the 'discourse method' by Schmidt and his colleagues (Schmidt et al. 1996), seems to be promising. In practice, researchers from the different countries documented observations of classroom teaching, interpreted them in terms of their own (culture-specific) concepts, and presented the result to the community of researchers. Here the concepts were questioned, from the viewpoint of the other countries' conceptual tools, until a common understanding of concepts was reached. This procedure allowed the development of appropriate instruments for the international survey study, and the preliminary discourse also provided an understanding of the fundamental differences in classroom teaching in the different countries.

 Acknowledgements

This paper was presented at an International Symposium on curriculum-making processes and curriculum research, Zürich, December, 1999. An extended version of parts 2 and 3 has been published in Rosenmund (1999).

References

Benavot, A., Cha, Y.K., Kamens, D. Meyer, J. W., and Wong, S.-Y. (1991) Knowledge for the masses: world models and national curricula, 1920-1986. American Sociological Review, 56 (1), 85-100.

Bernstein, B. (1977) Beiträge zu einer Theorie des pädagogischen Prozesses (Frankfurt/M.: Suhrkamp).

Grootings, P. (1989) New forms of work organization in Europe: East-West comparisons. In P. Grootings, B. Gustavsen, and G. HÈthy (eds), New Forms of Work Organization in Europe (New Brunswick, NJ and Oxford, UK: Transaction Publishers), 217--245.

Heidenreich, M. (1991) Problems of generalization in cross-national studies of organizations. International Sociology, 6 (2), 181-200,

Hopmann, S. (1988) Lehrplanarbeit als Verwaltungshandeln (Kiel, Germany: Universität Kiel, Institut für die Pädagogik der Naturwissenschaften).

Kamens, D. H., Meyer, J. W. and. Benavot, A. (1996) Worldwide patterns in academic secondary education curricula. Comparative Education Review, 40 (2), 116-138.

Künzli, R. (1999) Lehrplanarbeit -- Steuerung von Schule und Unterricht. In R. Künzli, K. Bähr, A.-V. Fries, G. Ghisla, R. Rosenmund and G. Seliner-Müller (eds), Lehrplanarbeit: Ðber den Nutzen von Lehrplänen für die Schule und ihre Entwicklung (Chur and Zürich, Switzerland: Rügger), 11-30.

Meyer, J. W. and Baker, D. P. (1996) Forming American educational policy with international data: lessons from the sociology of education. Sociology of Education, 69 (extra issue), 123-130.

Meyer, J. W., Boli, J. W., Thomas, G. M., and Ramirez, F. O. (1997) World society and the nation-state. American Journal of Sociology, 103 (1), 144-181.

Rosenmund, M. (1999) Vergleich von Unvergleichbarem? Theoretische und methodologische Zugänge zur international vergleichenden Lehrplanforschung. Bildungsforschung und Bildungspraxis, 21 (2), 173-187.

Schmidt, W. H., Jorde, D., Cogan, L., Barrier, E., Gonzalo, I., Moser, U., Shimizu, K., Sawada, T., Valverde, G. A., McKnight, C. C., Prawat, R. S., Wiley, D. E., Raizen, S. A., Britton, E. D. and Wolfe, R. G. (1996) Characterizing Pedagogical Flow: An Investigation of Mathematics and Science Teaching in Six Countries (Dordrecht, The Netherlands: Kluwer).

---

Publisher's information |  | Other Op-eds and Commentary