jan angstrom, erik hedenstrom and lars-inge strom

Can small scholarly communities, such as the Swedish political science community, survive in an academic world often said to be dominated not only by American research but also by American concerns? Does the political science profession in Sweden, a small country with a rather odd (!) language, few political science departments, and few members of the national political science association, have a unique independent Swedish identity – or is current Swedish political science a product of international — mostly American —influences? And how have these patterns of influence varied over time? Recently, a Swedish Research Council report claimed that political science in Sweden was heavily influenced by the alleged American hegemony within the discipline (Hydén, et.al., 2002: 15). The report did not, however, produce any data on the extent to which the Swedish political science community is influenced by research from the US or other parts of the world.

The aim of this article is to map the extent of foreign influences on Swedish political science as well as their geographical origins. There are several arguments suggesting that this study is important and late in coming. It has become progressively easier in recent decades to search computerised databases for articles, research papers, and books. Hence, there are reasons to believe that international influences have become more apparent in Swedish political science in recent decades. Furthermore, the extent of international influences has a significant impact on the likelihood of survival of the profession in small political science communities. Naturally, international contacts enrich small scholarly environments. There are, however, dangers for a small community in too great a reliance on international research if it hampers recruitment and the socialisation of young scholars into the community’s norms. Moreover, the increased proliferation of scholarly journals and the increased specialisation of the field in general mean that a small community may run the risk of disintegrating in favour of a greater sub-disciplinary community abroad.

Our study demonstrates that, for good or bad, the impact of international influences on Swedish political science is great indeed. It is, therefore, problematic to claim that Swedish political science has an independent identity. Still, we do not find any major changes from the situation thirty years ago — apart from some variation in the provenance of the foreign influences. It is clear, however, that reliance on foreign work is gradually increasing, and that Swedish political science — according to Swedish political scientists themselves, at least — stands less tall than it did one scholarly generation ago. We also find that US research is not in as prominent a position as is sometimes thought (for example, Hydén, et.al., 2002; Ruin, 1978). Indeed, American influence seems to have decreased over the last thirty years. Instead, it is more accurate to speak of an Anglo-American axis dominating Swedish political science.

Without similar studies in other similar states, it is too early to say whether or not this is a Swedish phenomenon or a well-established pattern of Anglo-American dominance of an entire academic discipline. We have used three different indicators of international influence: numbers of foreign references in doctoral dissertations; numbers of foreign references in the main Swedish political science journal, Statsvetenskaplig tidskrift, and a questionnaire to the political science professors of Sweden’s five established universities.
Swedish political science offers a useful case study of small political science communities since, despite its long history extending back to 1622 when the world’s oldest chair in political science and rhetoric was established at Uppsala University, it is a small player within the global political science community. While the Swedish Political Science Association has roughly 300 members, there are over 15,000 political scientists in the United States (Berndtson, 1987: 85).

The article proceeds as follows. First, we briefly review a number of previous studies of Swedish political science. Second, we describe the methodology informing our own study. Third, we present our findings; and finally, we comment on the overall impression given by the results and discuss whether or not international influences help or hinder the development of national political science communities.

STUDYING SWEDISH POLITICAL SCIENCE

While there have been numerous studies of Swedish political science, their main focus has been the development of the discipline in that country. That is, many (for example, Westerståhl, 1990; Ruin, 2002) have studied the history of Swedish political science – thus mirroring studies of other, related, disciplines (or sub-disciplines) such as international relations (for example, Olson and Onuf, 1985; Smith, 1995) – but few have dealt with the international influences on the profession in Sweden. In one of the few exceptions, Ruin (2002: 42-44) argues that international influences on the Swedish profession have gone through three stages. Prior to an interlude of growing ethnocentrism in the 1940s and 1950s, during the first four decades of the twentieth century the Swedish profession was heavily influenced by German political science – while American dominance, following the behavioural revolution, began in the early 1960s (see also Elvander, 1978).

Perhaps most obviously, it has been suggested that language matters. As Swedish political scientists usually have a knowledge of English but in general are likely to lack further language skills, it is not, perhaps, surprising that American and British research has had the most lasting influence on the Swedish profession. Jönsson (1993) argues that this is particularly relevant for the sub-field of international relations, which has been famously dubbed an ‘American social science’ by many already (Hoffmann, 1977; Waever, 1998; Crawford and Jarvis, 2001).

Another explanation centres around the academic organisation of the discipline, in particular the universities’ relationship with the state. From this perspective, the current hegemony of US political science derives, not only from its greater resources, but from an entrepreneurial and competitive academic system, mainly free from state intervention (Hydén, et al., 2002: 15). The situation in Sweden (and in the other Nordic states), as Wittrock and others have pointed out, is slightly different insofar as Swedish political science has to a greater degree been connected to state power (Wittrock, 1992; Westerståhl, 1990). Similarly, Anckar points out that the Nordic universities ‘were from the beginning established for purposes of the state: it was their main task to produce and educate civil servants and to maintain and legitimise the prevailing style and ideology of the government’ (Anckar, 1991: 243).

Nygren (1996: 431) argues that the political science community in Sweden is heterogeneous and consists of several sub-disciplinary communities, with little or no research co-operation between the fields. Hydén, et al. (2002: 165-167) note the same thing, calling for a greater degree of intra-disciplinary co-operation. In this study, however, we treat Swedish political science as a single entity.

MEASURING INTERNATIONAL INFLUENCE

In order to measure the international influences on Swedish political science we used three different indicators. First, we counted the references per region in the theory and methods sections of the doctoral dissertations in political science presented at the five established universities in Sweden (Gothenburg, Lund, Stockholm, Umea, and Uppsala) in 1971 and 2001. Theory and methods, obviously, are most interesting since they comprise, arguably, the core of any discipline. We separate theory and methods from the empirical sections of the dissertations since the empirical parts depend too much on the substantive areas their authors have chosen for investigation. They may therefore bias the results. This is especially important in Sweden for, as Hanna Larheden (1999) has demonstrated, over half the Swedish political science dissertations presented between 1979 and 1996 dealt with Swedish politics in some form or other. We selected 1971 as our point of comparison as it represents approximately one scholarly generation back in time, but otherwise its choice is arbitrary.

Second, we calculated the geographical distribution of the references contained in the articles appearing in the main Swedish political science journal (Statsvetenskaplig tidskrift) in 1971 and 2000. The journal is published quarterly. The contents of journals are important as an indicator of the nature of debate in political science because of their gate-keeping function: journals ‘determine what will be widely read and hence what kind of research will be socially rewarded’ (Goldmann, 1995: 247). In order to counterbalance our analysis of the dissertations, in this part of the study we did not distinguish between theory, methods and empirical sections. In the analyses of both the dissertations and the journal, the relevant criterion, for the purposes of measuring international influences, is where the texts referred to were published – rather than the author’s nationality, country of residence, or academic affiliation. Admittedly, there may be some bias here insofar as Swedish political scientists could, in theory, publish all their work in other countries – thus suggesting an extreme reliance upon foreign texts in Swedish political science when, in fact, there is none. However, other indicators, such as authors’ nationalities, are also problematic insofar as books by a Swedish scholar conducting her research at, and financed by, an American university can hardly be called Swedish research.

Third, we sent out a questionnaire to all of the political science professors at the five ‘old’ universities. It deals with what professional journals Swedish professors appreciate most, how often and in how many international research projects they are involved, and which professional conferences they value the most.
Applying all three indicators, we have distinguished between the following geographical regions:

• Sweden;
• the other Nordic states – given their geographical proximity, and the similarity, with Swedish, of their languages, it is reasonable to expect the political science communities of the other Nordic states to have a significant influence;
• Germany – as it once heavily influenced Swedish political science, not least in its research on the state;
• the United Kingdom – given its language and the number of political science publishers based there, it is reasonable to assume a significant impact;
• the rest of Europe – given the increasing number of ECPR conferences, workshops, etc., it is reasonable to assume that this region too will have a significant impact;
• the US – allegedly, the current political science hegemon;
• the rest of the world.

The three indicators complement each other and, taken together, they provide a balanced view of the international influences on Swedish political science. Analysis of the doctoral dissertations will give us an idea of how and what young scholars, who are just learning ‘the tricks of the trade’, think of as important. Meanwhile, the questionnaire to the professors will tell us how – by contrast – the already acclaimed work when it comes down to the nitty-gritty of everyday research. Analysing the journal, in turn, will give us an indication of international influences in the principal vehicle for scientific debate within the political science community in Sweden[1].

INTERNATIONAL INFLUENCES IN SWEDISH POLITICAL SCIENCE

It is probably fair to say that Swedish political science has been influenced by the international research – as well as the international and national political – agendas since 1622. In this section we will present a snapshot study of the current situation.

INTERNATIONAL INFLUENCES IN SWEDISH DOCTORAL DISSERTATIONS 1971 AND 2001

Table 1 shows that theory is especially heavily influenced by international research. Only 18 per cent of the references in the theory sections of Swedish doctoral dissertations in 2001 came from research published in Sweden. Together the UK and the US accounted for just over 72 per cent of the references, the US accounting for 42.8 per cent on its own. The score for Germany and the rest of Europe put together was roughly the same as that for the Nordic states, while references from the rest of the world accounted for only 1.8 per cent of all the theory references.
Compared with the results some thirty years ago, we can note two patterns. First, Swedish references have decreased from 25 to 18 per cent. Hence, the theory sections of Swedish doctoral dissertations do show a greater reliance on foreign work now than they did thirty years ago. Second, the international influences show some significant internal variation. In particular, UK references have tripled over the last few decades, while US, Nordic, and German references show a decreasing trend.

When it comes to methodology, a slightly different pattern emerges in 2001. One third of the references are Swedish. Swedish research students are thus more likely to use Swedish texts to develop their methodologies than they are to use them to develop their theories, and it is mainly theoretical influences that are imported into Swedish political science among the junior researchers. Still, however, 36 per cent of the methods references were American and 24 per cent were British. This means that roughly 60 per cent of the methods references were Anglo-American. When it comes to the other regions, the impression from the analysis of the theory sections is further reinforced. Roughly 10 per cent of the references are from the small players — although there is some internal variation. For example, the rest of Europe seems to have scored slightly better and Germany slightly worse than in the analysis of the theory sections.

Compared with the situation in 1971, the results of the analysis of the methods sections are remarkably similar to those deriving from the analysis of the theory sections. Swedish references have decreased but UK references have tripled, while US and Nordic references show a downward trend.

Studying the references contained in Swedish doctoral dissertations has given us important insights into how and what international influences find their way into Swedish political science. The patterns of influence did not change dramatically between 1971 and 2001 – with the exception of UK references, which roughly tripled their share. It is also clear that it is mainly theory that is imported by the research students even though the methods sections too were heavily influenced by international research.

INTERNATIONAL INFLUENCES IN STATSVETENSKAPLIG TIDSKRIFT 1971 AND 2000 [2]

The figures in table 2 largely reinforce the impression given by the analysis of the doctoral dissertations. Swedish political science is heavily influenced by international research, although one should note that in the case of the journal, Swedish references amount to over one third[3]. This suggests that Swedish political science has a stronger position in the scientific debate conducted through the journal than it does in the dissertations.

Influences from Sweden seem stable at just over a third of the references during both periods. Meanwhile, during both periods, the UK and US represent nearly half of the references when taken together. Perhaps most striking is the significant increase in the proportion of references to UK-published sources. In 1971, only 8.4 per cent of the references were from the UK, while in 2000 25.7 per cent of the references were to publications from that country. The trend for the US references runs in the opposite direction. In 1971, 40 per cent of the references were American while in 2000 only 23.8 per cent were. Another significant increase is registered by both German references (which double) and by references from the rest of Europe (which increase three-fold). Nevertheless, influences from all but the Anglo-American axis and Sweden remain at roughly the same level across both periods since the proportion of references from the Nordic states decreases from 8.9 per cent to 2 per cent. The latter decline argues against interpreting the results as indicating an increased overall reliance upon European rather than American political science.

In summary, analysis of the main professional vehicle for Swedish political science, Statsvetenskaplig tidskrift, demonstrates that international political research influences debate within Swedish political science to a considerable degree. However, expectations that international influences would have grown stronger during the past thirty years were unfounded. Just over one third of the references in both 1971 and 2000 were made to a Swedish text. Mirroring the results of the analysis of the dissertations, the results show that UK references have tripled while US and Nordic references have decreased.

INTERNATIONAL INFLUENCES IN SWEDISH POLITICAL SCIENCE: THE PROFESSORIAL VIEW

The results of the questionnaire sent to the professors suggest that Swedish political science is heavily influenced by work produced abroad even when its most prominent members conduct their research[4]. They read Anglo-American more often than Swedish journals; they publish more often in non-Swedish than in Swedish journals; they rate Anglo-American higher than Swedish journals from a scientific point of view, and they value international conferences more highly than they do the national meetings. However, at the same time, the results also show that Swedish professors co-operate more nationally than internationally and that they value this co-operation more highly than co-operation in the international contexts.

Asked what journals they read frequently, Swedish professors, as is shown in table 3, indicated that they mainly read Anglo-American journals. Of the 24 journals that were mentioned by the professors, two thirds were published either in the UK or in the US. Only two Swedish journals were mentioned by the professors.

Table 4 shows the results obtained when the professors were asked to rate, in terms of various criteria, the journals they had mentioned[5]. We have included in the table only those journals that were mentioned by more than one professor. The results are consistent with the others obtained from our study thus far. The only exclusively Swedish journal is dismissed as not particularly original or valuable theoretically. One should note, however, that there are only two journals, the American Political Science Review and the European Journal of Political Research, that are chosen by more than two of the professors. This reflects the high degree of specialisation in the discipline, where different sub-fields have different journals. The only two journals that publish political science ‘in general’ were also the two journals that were selected by more than two of the professors. Our results suggest that not only do Swedish professors read non-Swedish more often than Swedish journals but that they also value them more highly.

Table 5 shows how much Swedish political science professors value, on a scale 0 to 10, a number of professional conferences in terms of various criteria. It shows that, of the conferences that they can choose between, they value the national political science association’s conferences the least highly. Interestingly – something that also supports the findings deriving from analysis of the dissertations as well as those deriving from the professors’ ratings of the journals – new ideas are most often received from the US professional association’s conferences. At the same time, however, the results also show that overall, Swedish professors value the Nordic association’s conferences the highest, not least because of the possibilities they offer for new co-operation. Perhaps the most striking result, though, is the low total value for the national association’s conferences in comparison with the results for the international conferences.

Partly supporting the finding that Swedish professors seek possibilities for joint research projects close to home while seeking to get fresh ideas globally, is the finding that the professors rate the possibilities of new co-operation with their Nordic and European counterparts more highly than the possibilities of new co-operation with Americans. Since, of the conferences, the Swedish Political Science Association’s conference is rated the least highly in terms of the possibilities it offers for new co-operation, the results also indicate that the Swedish political science community may be too small for self-sustained research collaboration.

Table 6 gives a slightly different picture. It shows the extent of international collaboration on the part of Swedish professors, as well as how they rate such collaboration. Clearly, collaboration with fellow Swedish scholars is most common. All of our respondents collaborated in common research projects of some kind with other Swedish scholars. It is also clear that this collaboration is valued as highly as (or more highly than) collaboration with international scholars.

Table 7 shows the regional distribution of the articles published by current Swedish political science professors. The current Swedish professors have published articles worldwide. This is surely a sign of strength. However, the trends are equally interesting. The table clearly shows that the proportion of articles published in Sweden is decreasing, while international publications have risen dramatically as the professors have been increasingly recognised within their respective fields. This indicates that Swedish professors value publication of their work in international outlets more highly than they do publication in national outlets.

In summary, the results of the survey of current Swedish political science professors further reinforce the image of Swedish political science as heavily influenced by international political research. In much the same way as the Swedish research students were influenced by Anglo-American theory, Swedish professors valued international conferences — mainly American — because of the new ideas they generated. At the same time, however, the Swedish professors valued national collaboration as much as (or more than) international collaboration.

SURVIVAL OF THE MOST CITED?

This article has shown that Swedish political science is heavily influenced by international political science. The situation, perhaps contrary to expectations, has not changed dramatically from that of an earlier scholarly generation (even though Swedish references in the theory and methods sections of political science dissertations have decreased slightly). Suggestions that Swedish political science will perish as a result of international competition, seem, therefore, to be unfounded. Perhaps the results should be interpreted in a more positive way. It is not that the proportions of Swedish references in dissertations or Statsvetenskaplig tidskrift are roughly similar to those thirty years ago in spite of international competition but, rather, because of it. As analysis of the Swedish political science professors’ publishing habits has shown, international publications have increased gradually, thereby demonstrating the Swedish political science community’s desire to send its research to journals where competition to get past the gatekeepers is higher than in Swedish journals. Thus, it would be more accurate to say that Swedish political science thrives on the competition. Hence, our study suggests that even small scholarly communities can survive — even if they are not the most cited.

Moreover, it is misleading to claim that the US is in a hegemonic position in Swedish political science. The results demonstrate that it is more accurate to think in terms of an Anglo-American axis dominating Swedish political science – especially since the major change over the past three decades has been the increased reliance upon UK research in both Swedish dissertations and in Statsvetenskaplig Tidskrift. Still, the study also shows that Swedish political science is influenced by work from all over the world.
Finally, we are surprised that roughly 40 per cent of the theory references in Swedish political science dissertations are from America. This suggests the existence of an intellectual deficit within Swedish political science. Naturally, the language barrier is one of the principal reasons for this but the fact remains: Swedish political science in all likelihood imports more than it exports. Perhaps this is a sign of intellectual openness and vigorous academic debate. Perhaps it is a sign of weakness. As this study shows, however, it does not necessarily mean that small scholarly communities will perish: they may with as much likelihood survive and even thrive.

tables:

Table 1: Geographical distribution of references in the theory and methods sections of Swedish political science dissertations, 1971 and 2001 (%)

Table 2: Geographical distribution of references contained in the articles in Statsvetenskaplig tidskrift, 1971 and 2000 (%)

Table 3: Geographical distribution of the journals mentioned by Swedish professors as being ones they read frequently (%)

Table 4: Mean ratings (0-10) of Swedish professors on various criteria, of journals mentioned as ones they read frequently

Table 5: Mean ratings (0-10) of Swedish professors of professional conferences on various criteria

Table 6: Percentages of Swedish professors involved in collaborative research with political scientists from various regions, and their mean ratings (0-10) of the value of such collaboration

Table 7: Geographical distribution of articles published by current Swedish professors, 1960-2002

notes:

1. It is important to note that we did not consider how the references were used. Thus, a dissertation criticising ten American texts on, for example, foreign policy theory while following a theory developed in a Swedish publication will seem to be overwhelmingly influenced by US research. This may seem problematic but we maintain that influences on research work in negative as well as in positive directions. Criticism, after all, is the motor that ensures the continued search for knowledge. Perhaps the best example of influence over research agendas that has not worked in a ‘positive’ direction, but whose impact is nevertheless obvious, is given by Samuel Huntington’s texts on the ‘clash of civilizations’ – which few agree with, but a great many refer to.
2. The number of references in the articles published in 2000 was just over four times as many as in 1971. It is, however, difficult to interpret this. It may suggest that the conditions of political research have changed. Naturally, it may mean that the authors of the articles published in 2000 were more well-read than their colleagues in 1971 but it is perhaps more likely that it simply reflects the fact that it is easier to get hold of relevant literature now than it was in the past, and that the volume of relevant research has increased dramatically over the last thirty years.
3. Though this is significantly higher than in the dissertations, the higher score for the Swedish references could be explained by the fact that we have not separated theoretical from empirical sections. Thus, strong reliance on Swedish material for a case study, for example, may have biased our results. The same may be said of the Nordic states, which scored significantly higher in this analysis than in the analysis of the dissertations. Naturally, case studies of, or comparisons between, the Nordic states should multiply references from the other Nordic states. This is consistent with Anckar’s (1993) results, in which five Nordic political science journals from the 1980s were examined and the overwhelming majority of comparative studies drew on the Nordic states for at least one of the countries studied.
4. The questionnaire was sent to all of the 27 political science professors at the five established universities. We received only 11 answers, however. This relatively low frequency (40.7 per cent) means that the results should be interpreted with considerable caution. For the purposes of international comparison, we should perhaps mention that we did not send the questionnaire to associate or assistant professors.
5. The rating was carried out by means of questionnaire items that largely replicated the analytical tool developed by groups of scholars evaluating the quality of political science journals (Crewe and Norris, 1991; Norris and Crewe, 1993, see also Giles, et.al., 1975; 1989). The professors, thus, could rate the journals from 0 to 10 according to their assessments of the journals’ originality, theoretical value, and quality.

references:

Anckar, D. (1991), ‘Nordic Political Science: Trends, Roles, Approaches’ European Journal of Political Research, 20:3-4, 239-261.

Anckar, D. (1993), ‘Comparative Research in the Nordic Countries: Overcoming Ethnocentrism?’, Scandinavian Political Studies, 16:2, 107-126.

Berndtson, E. (1987), ‘The Rise and Fall of American Political Science: Personalities, Quotations, Speculations’, International Political Science Review, 8:1, 85-100.

Crawford, R. M. A. and D. S. L. Jarvis (eds) (2001), International Relations—Still an American Social Science? Towards Diversity in International Thought, Albany, NY, State University of New York Press.

Crewe, I. and P. Norris (1991), ‘British and American Journal Evaluation: Divergence or Convergence?’ PS: Political Science & Politics, 24, 524-531.

Elvander, N. (1978), ‘The Growth of the Profession 1960-1975: Sweden’ in Scandinavian Political Studies, Yearbook, Volume 12, Oslo, Universitetsforlaget, 75-82.

Giles, M. W., F. Mizell, and D. Patterson. (1989), ‘Political Scientists’ Journal Evaluations Revisited’ PS: Political Science & Politics, 22, 613-626.

Giles, M. W. and G. C. Wright (1975), ‘Political Scientists’ Evaluation of Sixty-Three Journals’ PS: Political Science & Politics, 8, 254-257.

Goldmann, K. (1995), ‘Im Westen nichts Neues: Seven International Relations Journals in 1972 and 1992’ European Journal of International Relations, 1:2, 245-258.

Hoffmann, S. (1977), ‘An American Social Science: International Relations’ Daedalus, 106:3, 41-60.

Hydén, G., E. Immergut, and A. Underdal (2002), Swedish Research in Political Science: An Evaluation, Stockholm, The Swedish Research Council.

Jönsson, C. (1993), ‘International Politics: Scandinavian Identity Amidst American Hegemony?’, Scandinavian Political Studies, 16:2, 149-165.

Larheden, H. (1999), ‘Trender och traditioner inom svensk statsvetenskap’ Statsvetenskaplig tidskrift, 102:1, 25-43.

Norris, P. and I. Crewe, (1993), ‘The Reputation of Political Science Journals: Pluralist and Consensus Views’ Political Studies, 41:1, 5-23.

Nygren, B. (1996), ‘Political Science in Sweden’ in Report on the State of the Discipline, European Political Science Network, 2002-07-17. http://www.epsnet.org/papers/document/sweden.pdf

Olson, W. and N. Onuf (1985), ‘The Growth of a Discipline: Reviewed’ in S. Smith (ed.), International Relations: British and American Perspectives, Oxford, Blackwell.

Ruin, O. (2002), ‘The Development of Swedish Political Science’ in G. Hydén, E. Immergut, and A. Underdal, Swedish Research in Political Science: An Evaluation, Stockholm, The Swedish Research Council.

Ruin, O. (1978), ‘Political Science Research in Sweden 1960-1975: An Overview’, Scandinavian Political Studies, Yearbook, Volume 12, Oslo, Universitetsforlaget, 157-184.

Smith, S. (1995), ‘The Self-Images of a Discipline: A Genealogy of International Relations Theory’ in K. Booth and S. Smith (eds), International Relations Theory Today, Oxford, Polity Press.

Waever, O. (1998), ‘The Sociology of a Not So International Discipline: American and European Developments in International Relations’, International Organization, 52:4, 687-727.

Westerståhl, J. (1990), ‘Om forskning och forskningsresultat i statsvetenskap’ in K. Fridjonsdottir (ed.), Svenska samhällsvetenskaper: Utvecklingslinjer, problem, perspektiv, Stockholm, Carlssons.

Wittrock, B. (1992), ‘Discourse and Discipline: Political Science as Project and Profession’ in M. Direkes and B. Biervert (eds), European Social Science in Transition: Assessment and Outlook, Boulder, CO, Westview Press.