Adult education research: exploring an increasingly f ragmented map

Against the background of internal developments of adult education as a field of study, and new external conditions for research, this article examines how the configuration of adult education research has been evolving, particularly over the last decade. Our analysis draws on a two-pronged approach: a reading of four seminal articles written by adult education scholars who have conducted bibliometric analyses of selected adult education journals; as well as our own review of 75 articles, covering a one-year period (2012–2013), in five adult education journals that were chosen to provide a greater variety of the field of adult education in terms of their thematic orientation and geographical scope than has been the case in previous reviews. Our findings suggest that the field is facing two main challenges. First, the fragmentation of the map of the territory that was noticed at the end of the 1990s, has continued and seems to have intensified. Second, not only practitioners, but also the policy community voice their disappointment with adult education research, and we note a disconnect between academic adult education research and policy-related research. We provide a couple of speculations as to the future map of adult education as a field of study and point to the danger of shifting the research agenda away from classical adult education concerns about democracy and social rights. 1

social structure and institutional power. A review of a scientific field should therefore address a two-fold relation, the broad social world in which the field is embedded and the scientific field itself, with its own rules of functioning, of which it is a part (Bourdieu, 1984). Camic (2011) notes that understanding the production and use of knowledge within a specific scientific field requires looking not only at the field as such, but also at its relation to other fields, disciplines and groups such as practitioners and policy makers.
The social world of adult education research The Bourdieusian perspective suggests that the evolving configuration of adult education research is directly impacted by changes to the internal structures of the field as such as well as by changes to the social context of the field, such as the social and economic role awarded to adult learning and education by the policy community. It is therefore important to note that as adult learning and education has come to the forefront of public policy, adult education research is becoming of vital interest to a broader community than the traditional group of adult education scholars (see e.g., European Commision [EC], 2011; Organization for Economic Co-operation and Development [OECD], 2003). The emergent discussions in policy circles on the relevance of adult education research is part of a broader movement, partly driven by supranational organizations like the OECD and the European Union (EU), to build a tradition of evidence-based policy making anchored in research findings. The central role afforded to the PISA and the recently launched PIAAC 2 programs should be seen as reflections of this shift. The call for policy relevant educational research is inducing countries to introduce changes to their educational research and development (R&D) system. For example in England the government has changed the balance between pure basic research and pure applied research through creating what is being labelled "useinspired basic research" that is carried out at dedicated research centres such as the Centre for Wider Benefits of Learning (OECD & Centre for Educational Research and Innovation [CERI], 2002). The same ambition has been driving the EU's Sixth and Seventh Framework Programme research agendas. Adult education research will likely be rather unaffected by these changes as the field only has had limited success in accessing this kind of research funds. However, adult education research is increasingly being criticised by the policy community who voices its disappointment with its usefulness. This was a dominant theme in the national reports submitted by developing and developed countries in preparation for the Sixth International Conference on Adult Education (CONFINTEA VI). To support an adult learning and education agenda the developing countries point to the need for research to more directly support initiatives focusing on reducing poverty, addressing HIV and strengthening the role of women (Aitchison & Alidou, 2009), while the industrialized countries contend that evidencebased policy-making is influenced by research coming out of governments and supranational institutions, but not academic institutions (Keogh, 2009).
While the concern from the policy community is a new phenomenon, there has been a long-standing criticism of the limited relevance of the research enterprise for the practice of adult education. Sork and Cafarella (1989) suggested that the gap between research and practice was widening rather than shrinking. This could be an outcome of the calls during the late 1970s for the field to become more theoretically sophisticated so that it might gain more respect in the scholarly world (Rubenson, 2011). Thus, in a response to this call university departments of adult education tried to affect the institutional structure of the field by recruiting new faculty into adult education who often had less connection to the field of practice than the outgoing faculty. The merit system for academics was increasingly focused on academic merits (articles in preferably refereed journals, acquiring research projects, etc.), while practice-related, developmental work was less honoured by the university system. The situation is not deemed to have improved since the Sork and Cafarella article (Sork & Cafarella, 1989). Rose (2011), reflecting on the 2010 Handbook of Adult and Continuing Education (Kasworm, Rose, & Ross-Gordon, 2010), of which she was one of the co-editors, notes: "Adult educators have a vibrant and impassioned calling, yet the researchers fail the field." She sees this partly as a result of the fact that researchers "have not been able to move beyond a critique of power and oppression" (p. 44). Her harsh conclusion is that "adult education has eschewed any attempt to bring its research into areas that have implications for the actual practice of the field" (p. 44).
The scientific field of adult education Turning to the maturity and evolution of the field, Rubenson (2011) suggests that since adult education began to emerge as a field of study in the late 1920s, it has undergone three quite distinctive phases. He notes that these phases are most noticeable in the United States, which to a large extent has come to define the nature of the scholarly field, but they are also clearly discernible in Europe and to a lesser extent in some other parts of the world. The first phase starting in the 1920s was a response to the beginning professionalization of adult education. With a small but growing number of adult education programs, faculty started to focus on how to generate a body of knowledge that would help in the growth of the evolving field. In 1961, the Commission of Professors of Adult Education had two dozen members. By 1972 that number had grown to 156. While in 1963 86 adult education dissertations were reported, that number increased to 173 by 1969 (Long & Agyekum, 1974). Dickinson and Rusnell (1971) reported a strong increase of research related to adult education and scholars affiliated to university departments, and Long and Agyekum (1974) observed an "increasing sophistication in adult education research" (p. 106) between 1964 and 1973.
Guided by funding from the W. K. Kellogg Foundation, the Commission of Professors of Adult Education in the US set out to define the conceptual foundations of adult education (Jensen, Liveright, & Hallenbeck, 1964). Officially titled Adult Education: Outlines of an Emerging Field of University Study, this book is popularly called the "Black Book." The "Black Book" can be seen as ushering in the second phase of adult education. We can understand this development as a process by which a field of study begins to emerge as a direct response to the needs emerging in adult education as a field of practice. Seeking solutions to primarily teaching and learning issues the emerging field was closely embedded in the field of educational psychology and strongly influenced by external professional organisations. Between the release of the Black Book and the publication of its follow up, Adult Education: Evolution and Achievements in a Developing Field of Study (Peters & Jarvis, 1991) in 1991, the number of adult education graduate programs in the US and Europe increased rapidly, yearly scholarly conferences were initiated and research journals were launched. Thus, this gradually maturing process of the field of study reflects and is affected by internal shifts of the field, primarily with regard to its location and presence in the broader university structure and are less a result of external forces.
The 1991 review painted a very positive picture and ended on an optimistic outlook and with expectations of continuous growth and solidifying of the field of adult education over the coming 25 years. While there does not exist any comprehensive review of what has happened since the 1991 book, there are several indications that the field of study has not progressed as anticipated and that it has entered into a new phase, the third, in its development. In North America and those parts of Europe where the field had expanded and matured during the second phase, the last two decades have not seen a continuing growth in specialized adult education departments. Instead, the trend has been to amalgamate adult education programs with other fields into larger departments or in some instances to close them down. In a Bourdieusian perspective this could be taken as an indication that the field of study has lost some of its legitimation within the university structure.
Outside North America and parts of Europe the process of developing adult education as a field of study began later. This is the case in several African and Latin American countries. In some instances, like in Brazil, there is an acceleration of programs and departments specializing in adult education (Torres, 2009). In China, the first MA program in adult education was launched at East China Normal University in 1993; a PhD program followed in 2004. The number of universities with graduate programs of adult education has increased from seven in 2003 to 23 in 2008 (Huang & Shi, 2008 It is against these external and internal developments of the field of study, and new external conditions for research, that this article examines how the configuration of adult education research has been evolving, particularly over the last decade. Previous findings regarding the map of the territory Numerous articles examining the scholarly field of adult education have appeared since the 1960s, employing content analyses of adult education journals (Dickinson & Rusnell, 1971;Long & Agyekum, 1974), country comparisons (Brookfield, 1982), and citation analysis (Boshier & Pickard, 1979). Some articles looked at subdisciplines such as adult basic education (Fisher & Martin, 1987) or specific aspects such as the impact of feminism on adult education (Hayes, 1992). Rubenson (1982), among other things, found that there was an overwhelming influence of psychology with the consequence that the territory of adult education research was defined primarily through assumptions of the characteristics of the learner and, thus, teaching was reduced to learning; empiricism and research methodology was emphasized in order to build a discipline of adult education; there was strong scepticism against borrowing from other disciplines and fields of study; and North American scholars dominated the landscape with little international exchange.
In a second study, revisiting the previous one, Rubenson (2000) noticed, not surprisingly, that the map was changing in accordance with the general drift of the social sciences. More specifically he noted a shift to more articles invoking a poststructural tradition with its emphasis on gender and critical race theories and a major impact of the new economic paradigm with a rapidly increasing number of articles focusing on workplace learning which resulted in a broadening of the conceptualisation of learning in adult education. He further noticed a major change in the attitude towards borrowing from other disciplines and that policy-oriented studies were less predominant in North American journals than European. The dominance of North American scholars in defining the map of the territory was seen to be less obvious than it was two decades earlier as European scholars were gaining more visibility.
Against this background it is of interest to look closer at how the configuration of adult education research might have changed over the last decade. In the next section we will present the research we have carried out in this regard.

Method, data and analysis
Our analysis of the changing characteristics of the map of adult education research is based on a two-pronged approach. We draw on a reading of four seminal articles written by adult education scholars who have conducted bibliometric analyses of selected adult education journals: Taylor (2006), who conducted a content analysis of comparative and international adult education in SICE, IJLE, and Convergence. It is important to note that Taylor (2001) looked at all submissions, while the other reviews considered only the published articles. We chose these reviews because they are recent and have, with the exception of St. Clair (2011), focused on articles published in what is commonly seen as the core of scholarly journals in the field of adult education a reading of which is expected to provide an authoritative view of the status of the field. We included St. Clair's analysis of the CJSAE because of its neat focus on a country with a distinct adult education tradition and because of our own interest in this journal given that we are based in Canada. It should be noted that these journals are all housed in Anglo-Saxon countries. The journals reflect the central role scholars from these countries play, and they perpetuate their dominance over the field.
In addition to these already existing reviews, we reviewed 75 articles, covering a one year period (2012-2013) in five adult education journals: Adult Education Quarterly (AEQ), the International Journal of Continuing Education & Lifelong Learning (IJCELL), which is published out of Hong Kong, the International Journal of Lifelong Education (IJLE), the European Journal for Research on the Education and Learning of Adults (RELA), and the International Review of Education (IRE) 3 . We have chosen these journals, as they seem to cover a greater variety of the field of adult education in terms of their thematic orientation and geographical scope than has been the case in previous reviews. While the US-based AEQ and the UK-based IJLE represent core authoritative journals in the field, the ILCELL, published out of Hong Kong, reflects, more than other journals, the developments in Asian countries. The RELA is a rather new open-access journal published by the European Society for Research on the Education of Adults (ESREA), which aims at providing a forum for scholars from Europe and for those whose first language is not English. The IRE, edited by the UNESCO Institute for Lifelong Learning, is not exclusively an adult education journal 4 and has a stronger orientation towards developing countries.
We focus particularly on the following categories: • Authorship • Research focus • Research methodology and theoretical orientation. We chose these three categories in relation to the two Rubenson (1982Rubenson ( , 2000 articles and the four reviews mentioned above. For example, one finding of all four articles was the dominance of the Anglophone authorship; we needed to look at authorship in order to relate to those findings. From a Bourdieusian perspective these categories provide some suggestions regarding the maturity of the field, its relationship to other fields, the institutional configuration of the field as well as the impact of outside forces.

Findings regarding characteristics of adult education research
Despite variations between the various reviews we draw upon in terms of journals covered, time periods and categorizations, some clear trends are noticeable.

Authorship
Taylor (2001), carrying out a close analysis of all submissions to AEQ from 1989 to 1999, noted that despite a dominance of male authors there was a steadily growing submission of women who by the end of the period had overtaken men. Later reviews have confirmed this trend and point to a reversed gender gap. In the three journals covered by Fejes and Nylander (2014) for the period 2005-2011 women wrote almost two thirds of the articles and in our analysis of five journals for the period 2012-2013 women made up 55 per cent of the authors. The results reflect the growing number of female faculty in adult education. However, it might be worth noting that the gender distribution seems to have stabilized during the last decade. Taylor (2001) reported considerable consistency in earlier periods in the dominance of single authorship in the AEQ. In contrast to the common trend in the social sciences towards co-publishing (Sethi & Panda, 2013), single authorship seems, according to our one-year review, to prevail in adult education. In the five journals slightly over half of all articles were single-authored.
Four out of five of the authors writing in these five journals were university professors, one third came from an adult education department, another third worked out of a broader education department while 20 per cent came from a non-educational, but often related discipline such as sociology, anthropology or cultural studies. One remarkable finding is that only five articles out of 75 were co-authored by PhD students (two of them identified as PhD Candidates; only in two cases the student was first author), and about ten per cent were authored by consultants or researchers not based in a university. Interestingly, none of those "non-academic" authors published in the AEQ and RELA, which may speak to the purely academic profile of those journals.
Judging by the recent reviews adult education research is increasingly being governed by a set of regional maps. As Mulenga, Al-Harthi, and Carr-Chellman (2006) and Fejes and Nylander (2014) observe, there continues to be a tendency for authors to publish articles in a journal of their home country. Taylor (2001) found that the submissions in AEQ during the 1990s originated almost exclusively from North America, and predominantly from the US, but a slight increase of articles from Western European countries could be noticed. There does not seem to have been any change as our 2012-2013 review shows that out of fifteen articles published in AEQ during that period ten were by Americans and one by a Canadian. Similarly, St. Clair (2011) noted a very strong dominance of Canadian scholars publishing in the Canadian journal. Given the pattern of "home grown publishing" it is not surprising that a vast majority of authors in the AEQ, IJLE and SICE, all hosted in Anglo Saxon countries, were from those parts of the world (Fejes & Nylander, 2014). Similarly, Larsson (2010) and Mulenga, Al-Harthi, and Carr-Chellman (2006) point to the overwhelming dominance in the core adult education journals of scholars from the Anglo-Saxon world. An interesting finding in Larsson's review is that not only do the authors predominantly come from Australia, Canada, UK and the US but in their work they almost exclusively cite other authors from one of these countries. Moreover he notes that well over half of the references in the AEQ, the only adult education journal that is indexed by ISI, are from articles published in the AEQ, which from a Bourdieusian perspective highlights the reproduction of the dominance of Anglo-American scholars as well as of the AEQ over the field. This pattern also shows that research coming from outside universities is not being cited.
Including a partly different set of journals, we found that almost half of the authors in the five journals we analysed came from European countries and about a third from North America. Of the European authors 12 originated from the UK, only two from Eastern Europe (Bulgaria and Russia) and 11 from Scandinavian countries. Portugal was the most represented country from Mediterranean Europe with four articles. Seven authors were from Asian countries, which tend to publish for the most part in the Hong Kong-based IJCELL. Only three articles came from Australian scholars who tend to publish in their own journal, which we did not include in our review.
Looking at differences between the five journals we found that authorship in the IJCELL, the IJLE and the RELA is by majority European, with the IJCELL having a higher proportion of articles coming out of Asia, especially China. Only the IRE shows a somewhat more balanced geographical distribution of articles although with a strong dominance of authors from Europe and North America 5 . Thus, out of 15 articles, five were from Europe, five from North America, two from Africa, and one from Latin America, Australia and the Arab States respectively.
Another finding worth drawing attention to is that not only are the authors publishing in local or regional journals but the scope in the majority of the articles in the AEQ, the IJLE and the IJCELL is national. We have defined articles as national when they focus on national or local issues or when the database on which they draw was collected in the country they originate from. In accordance with Mulenga, Al-Harthi, and Carr-Chellman (2006), we found that few articles (eight out of 75) qualified as international in that the research and used data involved more than one country. Only four articles were co-authored by researchers originating from two different countries. Seven articles were supranational in scope, insofar as they addressed education strategies or initiatives driven by international organizations, such as the OECD literacy surveys. Ten articles addressed development issues, most of them published in the IRE, and one article (in the AEQ) dealt with indigenous education. These results are consistent with what was reported by Larsson (2010).

Research focus
Overall, regardless of the time period, the various authors have noted considerable consistency in terms of research focus. As suggested by Taylor (2001) the field may not be as pluralistic as we like to believe. In the AEQ 70 per cent of all submissions during the 1990s belonged to 5 of the 12 identified categories: adult learning, participation, gender/diversity, adult education as a movement, and analysis of publication patterns. One noticeable change in the AEQ was a drop in historical and conceptual articles (Taylor, 2001). This trend is being echoed by St. Clair (2011), as well as in our one-year analysis which found that only about ten percent of the papers were of theoretical or conceptual nature. All of the reviews showed a strong dominance of articles addressing adult learning and teaching, so for example 40 per cent of all articles published in the 22 volumes of the Canadian journal addressed this topic (St. Clair, 2011). It is thus not suprising that Fejes and Nylander (2014) reported that 28 per cent of the studies were related to an educational institution setting, while about one in five addressed the workplace and around 16 per cent studied e-learning. Taylor's observation about the centrality of gender studies among North American adult education scholars is also noticed by St. Clair who found that one fifth of all articles were located in the area of feminist studies. This is a finding that may be specific to Canada, which has a strong feminist tradition in adult education. We were surprised to find only one study among the 75 using a feminist approach. Altogether, seven articles specifically addressed the learning of women, but only one of them used a feminist framework, the other four drew on bodies of literature related to the research question of the article, such as retention, literacy, development, but without taking a feminist perspective. Employing Hayes' and Smith's (1994) dominant perspectives on women in adult education publications, five articles examine "women as marginalized" and two (relating to the learning in nursing programs) "women as learners." There was a clear dominance of articles addressing adult learning with 60 per cent of the articles in the AEQ and more than 40 per cent in the RELA dealing with learning. To our surprise, we found very few -four -articles on teaching, one each in all journals we looked at except for RELA. This may be an indication of a change from Taylor's (2001) review which had found that teaching and curriculum was among the major topics in the AEQ. With workplace learning evolving as a separate scholarly field it may not be that surprising that relatively few articles reported on work-and skillsrelated research. We found seven articles related to professional and continuing vocational learning. The absence of articles on online learning (we found only two articles on that topic, both in the IJCELL) was another surprise, such as the absence of papers related to literacy, with the IRE, which has a strong focus on developing countries, being the exception with five articles on that topic. Literacy seems to be an out-dated research topic and associated with development.
Another issue is the blurring between lifelong learning and adult learning and education. The IJCELL, with nine out of 15, contained the highest number of articles that self-contextualized into the broader landscape of lifelong learning. In all of the other journals, there was a more balanced proportion of articles in either of the categories, or in many cases the articles did not situate themselves in any of them. Articles covered all kinds of issues related to the learning and experiences and interactions of adults, be it in informal, non-formal and professional learning settings, including methodological research on autobiographies and life histories. Three articles were related to the learning of students in postsecondary institutions -an indication of a further blurring of what is understood as adult education.

Methodology and theoretical orientation
For the period 1989−1999 Taylor (2001) observed a sharp increase in the share of articles that employed some form of qualitative methods and a corresponding decrease in work using a quantitative methodology. This finding is echoed in all of the more recent reviews. As shown by Fejes and Nylander (2014), and according to our own review, presently adult education scholars are almost exclusively relying on qualitative methodologies, with a few using a mixed method but with a total absence of pure quantitative research. Out of 75 analyzed articles, 64 were qualitative and 11 used a mixed-method such as a survey combined with interviews. Fejes and Nylander (2014) report that the three most common theoretical perspectives were socio-cultural (23 per cent), critical pedagogy (17.5 per cent) and poststructuralism (15.5 per cent). Looking at the disciplinary base, we classified about 40 per cent of the articles in the five journals as broadly sociological and about 33 per cent as psychological. By sociological we mean studies that look at social, organizational or institutional behaviour drawing on theories associated to the discipline of sociology such as neo-institutional theories and/or sociologists (the most frequently cited being Bourdieu with 15 citations in 75 articles) and often employing methodologies such as critical policy sociology. By psychological we mean studies looking at categories such as intentions and self-perception, meaning and experiences, employing theories of behavioural change or transformative learning. There were noticeable differences between journals with the majority of the articles in the IJLE, the RELA and the IRE being sociological, whereas the AEQ had a strong psychological orientation. Six out of 15 articles in the AEQ used transformative learning theory as a theoretical lens, making Mezirow one of the most-cited theorists in the AEQ (with seven citations), equalled by Merriam, followed by Freire (five) and Jarvis (four). Overall, Peter Jarvis is the mostcited scholar with citations in 16 of the 75 articles, followed by Paulo Freire and Pierre Bourdieu (15 each), Sharan Merriam (13) and Jack Mezirow (12).
We only found four papers among the 75 that were drawing at least partly on theorists associated with the post-structural paradigm, such as Foucault and Actornetwork theory, e.g. Latour (one in the AEQ, two in RELA, and one in the IJLE). This was a surprise as Rubenson (2000) had noted an increase of poststructural research. It is difficult to say whether this is just a coincidence, given that our analysis covers only one year, or whether this may be a general trend in adult education or even in the social sciences.
The papers published in the IJCELL dealt more often with lifelong learning and generally had a stronger focus on organizational and institutional matters and a weaker theory base than in the other journals. As Mauch (1999, reporting on a conference held in 1994) observed already 20 years ago, in China and some other Asian countries like South Korea the notion of lifelong learning has been fully accepted as the new educational paradigm and articles tend to discuss the development of lifelong learning systems and policies in those countries.

Discussion
Overall the findings suggest that the scientific field of adult education finds itself in a precarious situation. This is reflected in a continuing regional fragmentation of the field, an accelerating hollowing out of the field and a seeming inability to respond to what can be labelled as a relevance deficit.
Regional fragmentation of the field As noted, the move to develop adult education into a field of study accelerated in the US in the 1950s but by the mid-1990s it had become at least as vibrant in Europe as in North America, with special European research journals, scholarly societies and regional research conferences established. Our findings indicate that this shift has resulted in the creation of two quite distinguishable regional maps, one US or North American map and one European. While the AEQ remains the bastion of North American scholarship in adult education all the other journals are dominated by European authors, with the IRE showing a balanced presence of the two. Scholars keep publishing in their local or regional journals without trying to engage with each other in a discussion of the regional differences. Of course this does not mean that there are no contacts between proponents of the different maps but it seems to suggest that adult education does not possess one authoritative map of its territory, a finding confirmed by Larsson (2010). We should also point out that just because of the difference in the number of countries in the regions the North American map could be seen as more cohesive than the European. However, with the increasingly central role taken by an organisation like ESREA, the contour of the European map may become more fixed.
Applying Bourdieu's understanding of scientific fields, the differences in maps should be understood in the larger context of differences in social and cultural traditions and the impact of these on research practices (Popkewitz, 1984). The US (and Canada, although to a lesser extent) with their decentralized political and economic systems and emphasis on social mobility promote a research focus on the individual. The strong focus on psychologically-oriented perspectives by American adult education researchers, as noted in the review, is in accordance with the dominant tradition in educational research in general. To use Kuhn's (1962Kuhn's ( /1996 concept of paradigm at the meta-level the tradition within adult education research is part of the dominant "Weltanschauung." As Brookfield (1982) noted already back in 1982, the North American literature draws "a clear distinction between an audience interested in research and theory, and one interested in practice" (p. 157), which is why it tends to identify practitioners, instructors and/or administrators as the usual target groups. Consequently, the process by which adult education has become a specialized field of study in North America has been linked to the professionalization of adult education. In Europe, a different "Weltanschauung" governs the research tradition. While European research has also been affected by the professionalization of adult education, it has been more influenced by the broader policy realm, as Rubenson observed in his 2000 review. Thus, the differences in topic and theoretical orientation that we observed between the publications in AEQ and RELA speak to differences in what Bourdieu has labelled the social cosmos of the field. Similarly, the articles in IJCELL suggest the beginning of a newly evolving map that emphasizes a technical-practical perspective and the promotion of adult learning as a tool to adapt to a changing economic and technological environment in the context of free market capitalism. This map reflects yet another social cosmos affecting the specific regional field of adult education in Asia.
The regionalisation of the field suggests a lack of maturity where, in Bourdieu's words, the scientific universe of the field of adult education is rather weak and as a consequence it becomes strongly influenced by the social cosmos in which it is embedded.
Hollowing out of the field Two current processes work in tandem to weaken the field, a fragmentation of adult education research and the changes to the institutional structure of research. Returning to our observation regarding the absence of workplace-and skills-related research, which is unexpected given the dominance of the skills discourse in the policy realm (Elfert & Rubenson, 2013), this absence suggests a fragmentation of the field into subdisciplines, which have become fields of study in of themselves. The trend might be most obvious in the area of workplace learning which has started its own scholarly conferences and research journals, e.g. the Journal of Workplace Learning. Areas formerly associated with management and business studies are being subsumed under adult education, such as human resource development (HRD) and career development, in particular in the United States, where numerous professorships combine adult education and HRD. Several academic journals serve the field of HRD such as the Human Resource Development Quarterly (HRDQ). The lists of members of the editorial board of the HRDQ as well as of the Academy of Human Resource Development (AHRD) contains many adult educators, and the current editor of HRDQ is also on the editorial board of the Adult Education Quarterly. The overlap between adult education and HRD is noteworthy, as these constitute fields that are based on different logics. HRD has a strong focus on organizational and managerial performance and employee training. It is rooted in somewhat different theoretical foundations (Yang, 2004) and has less of a critical tradition than adult education (Fenwick, 2004). Fragmentation is also apparent in other traditional core areas of adult education like adult literacy. These subfields that struggle for their own legitimacy are challenging the field of adult education. In this context it is of interest to refer to Albert (2003) who argues that the field of sociology has fragmented into subdisciplines which all have their own standards of knowledge production and scientific legitimacy: "the various specialties and paradigms are now differentiated to such a point that researchers are little or badly acquainted both with the debates going on and the knowledge produced in other domains than their own" (p. 171).
The general restructuring of university departments into larger structures or closing down of adult education graduate programs in some countries, especially in the US but also in Australia and some European countries, further hampers the building of a field of adult education. Milton, Watkins, Spears Studdard, and Burch's (2003) study shows that adult education departments in the US undergo a changing of perspective, often reflected by name changes indicating a broader perspective of lifelong learning. Field (2005) confirms this trend for the UK, where the "coherent and bounded field of adult education is being displaced by the more open and decentred domain of lifelong learning" (p. 207). This development is in line with our finding that only one third of the authors worked out of adult education departments.

Relevance deficit
In the introduction we alluded to an increasing pressure from the policy community for "policy relevant" adult education research. However, our findings point to a disconnect between the policy discourse and academic adult education research. So for example, the outcome-based perspective that is being promoted by supranational organisations such as OECD, UNESCO and the World Bank is largely absent in the core adult education journals. The lack of statistically sophisticated empirical research in adult education further contributes to its perceived "irrelevance" to the policy community.
In our brief historical review at the beginning of the paper we suggested that the scientific field of adult education initially developed "from within" as a reaction to the needs of practice. Later in its search for legitimation within the academy the field moved away from its initial focus on practice resulting in research carried out "for the sake of research." This speaks to the relatively autonomous scientific "microcosms" of the field as represented by core journals like AEQ. It might also suggest that the field is not mature enough to handle the outside social pressures for "relevant" research.

Concluding note
It is always dangerous to speculate about the contours of the future map/s of adult education as a field of study, but we dare to provide a couple of speculations. The first is that the new subdisciplines will jeopardize the traditional field of adult education as they have more "capital" given that they are more in line with the policy discourse and therefore in a better position to obtain funding. The second is that there are no indications that the fragmentation process will come to a halt. On the contrary, it is more likely that it will intensify. One reason is that an Asian map might start to emerge, given that adult education as a field of study is beginning to take shape in countries like China and the Republic of Korea. Our analysis of the IJCELL, which represents to some extent the developments in Asia with its specific characteristics, already points in this direction. This is by itself not a negative development but it is more likely to increase the number of maps than to contribute to the development of an integrated map. Another reason is that we might see the European map becoming even more dominant due to the institutional changes in the US that further weaken American knowledge production in adult education. The strong emphasis in the EU on the economic and social role of adult learning in combination with the stress on evidence-based policy will fuel research activities within the EU while there are fewer opportunities for this kind of research in the US. This development in European adult education research carries with it some obvious risks. Thus, while the policy-related interest in adult education research may provide some new opportunities for the development of more major research programs, something that has been lacking in the field, it also provides a danger of moving the research agenda away from classical adult education concerns about democracy and social rights and "forcing" the researchers to focus on a narrow politically-defined research agenda.