Finding Their Way: The Socialisation of Freshmen MEPs in the European Parliament Tim Bale and Paul Taggart Department of International Relations & Politics/Sussex European Institute, University of Sussex Brighton BN1 9SH UNITED KINGDOM new.mep.project@sussex.ac.uk Paper presented to the European Union Studies Association of the USA, Austin, Texas, March 31 to April 2, 2005* * The authors would like to thank the British Academy for the conference grant that enabled Tim Bale to travel to Austin to present this paper. A role's attitudinal components (desires and beliefs) can be thought of as proximate attitudes intertwined with behavior. The reconstruction of these proximate attitudes and behaviors constitutes an interpretative explanation. Now one of the most important ways that this interpretative explanation becomes nested in more-naturalistic macroexplanations is by constructing generalizations that link it back to more-contingent systemic factors that structure the experiences through which the attitudes are learned. Investigations of these origins of roles help us to explain them more fully. We also explain them more fully by investigating their consequences. Donald D. Searing, Westminster's World, 1994: 22. The European Parliament (EP) is important and so are its members, the MEPs. The institution has managed, in less than half a century, 'to develop from a token talking- shop into a significant player', with the result that 'every year, thousands of amendments to draft legislation put forward by ordinary backbench MEPs end up on the statute book' (Corbett, Jacobs and Shackleton, 2003b: 354, 358; see also Judge and Earnshaw, 2003). There is, then, as Scully and Farrell (2003: 271) observe, 'much more that we can and should know' about MEPs as 'individual elected representatives'. One thing, however, we know already: 'A remarkable feature of the European Parliament', write Corbett et al. (2003a: 40), 'is the high turnover in its membership at each election.' It may be that, as research on previous parliaments has shown, between a quarter and a third of all MEPs have national parliamentary experience and approximately one in ten have served in a ministerial capacity 'back home'; but around half of all the elected representatives who gather in Brussels and Strasbourg at the start of each parliamentary term will be new or 'freshmen' MEPs. Whether we know much about them is another matter. The new European Parliament in the new European Union represents a substantially new setting with freshmen members constituting the majority. With 405 new members (out of a total of 732), the EP is numerically more of a new institution than an old one and, as a research object, this provides us with an opportunity to look closely at a range of new European Parliamentarians. The research presented here is based on research, funded by the Nuffield Foundation, covering the period July 2004 to July 2005 entitled The New European Parliament and the New European Parliamentarians. Our focus so far has been on the roles orientations of the parliamentarians and in particular the extent to which their backgrounds and their interests influence the roles they take on. This project is designed to 'capture' the new intake of MEPs so that we can establish a benchmark to measure their subsequent socialisation and gather data on their backgrounds, interests and experiences. In order to do this, we are conducting an intensive round of semi-structured, in-depth interviews with a sample of the new intake. We are also constructing a database containing information on all the new MEPs. This paper is an attempt to provide a 'field report' on this on-going research into the cohort of new MEPs. This means that the paper take an unusual form; we are presenting initial rather than final findings and we are taking the opportunity of this paper to present some hypotheses that we hope to test in subsequent research. The paper has three parts: in the first part we examine the way in which the EP and parliamentary socialisation has been treated in existing literature. In the second part we turn to the research and provide a brief description of our methods. In the third section we present our typology of role orientations based on our research, before concluding with some of the implications of our typology for the study of the EP. The European Parliament and Parliamentary Socialisation Traditionally, much of the research on the EP has concentrated on its external face – on the parliament's relationship with other EU bodies (eg Westlake, 1998) and its seemingly 'second-order' elections (eg van der Eijk and Franklin, 1996). In the last few years, however, researchers have begun to move inside the institution. Most have focused on the EP's transnational party groups and their cohesion (see, for example, Hix, Kreppel and Noury, 2003). But work has also been done on individual MEPs as opposed to parties, looking at variation in their preferences for roles derived from the classical functions of the legislature, such as representation, law-making and oversight, as well as exploring the extent to which they become more supportive of integration as a result of their experience in the EP (see Katz, 1997 and 1999, and Scully and Farrell, 2003). Common to all these studies is an overarching interest in Europeanisation – the extent to which the objects of study show symptoms of, or are contributing to, an ongoing process in which the boundaries between member states and the EU, in as much as they ever existed, are breaking down, institutionally and/or attitudinally. As regards the EP, easily the most thorough study of the attitudinal aspect – Europeanisation as socialisation and vice versa – is the recent work conducted by Roger Scully (2005). Although interested in the relationship between MEPs and their national parties (Scully, 2003: 135-6), Scully's main focus is on the widely-held assumption (rooted in neo-functionalism but boosted by the appearance of the 'new institutionalism') that those 'sent to Europe' end up 'going native' as a result of being in the EP (see Franklin and Scarrow, 1999) in much the same way as their American counterparts sent to Washington (Leal, 2002). Scully – like Jeffrey Checkel, who conducted interview and observation based research on the Council of Europe (see Checkel, 2003) – finds little empirical support for the assumption. He also notes that the presumably core goals of policy, office and re-election may well militate against such an integrationist outcome: most members remain national politicians operating in a European setting. This negative finding, combined with his own caveats regarding the difficulty of tracking individual change in surveys and voting analysis, leads Scully to argue strongly that the Europeanisation focus of EP studies has turned out to be something of a blind alley and suggests the need to move on to an individual-level approach which enriches quantitative roll-call analysis and one-off survey work with qualitative, over-time methods. This is far from saying, of course, that existing modes of analysis of the EP have taught us little of value. Quite the contrary: they have improved our understanding of the institution in myriad ways (see Hix, Raunio and Scully, 2003), diachronically as well as synchronically (see Kreppel, 2002). Moreover, they have also confirmed that the EP, like the EU as a whole (see Hix, and Marks and Hooghe, 2001), can be studied comparatively rather than being treated as a sui generis phenomenon (Judge and Earnshaw, 2003: 8-25) and that '[m]ajor developments in the study of the European Parliament in recent years have come, and are likely to come, from making judicious use of the concepts and advanced methods common to the study of politics across the world' (Scully, 2003: 137-8). The findings and techniques from deductive, quantitative and/or rational choice work on the US Congress have been particularly influential in this regard. Given Scully's call for additional (but hopefully complementary) methods, it seems logical to begin to think about supplementing that work with analytical techniques and frameworks drawn from another, more inductive, observational and qualitative stream of US legislative studies. This stream tends to take as its unit of analysis the individual legislator, with a particular interest in roles and/or 'role cognitions' - how legislators 'conceive and express their norms of behaviour' (Davidson, 1969: 72). The pioneering work in this stream began with the attempt by Wahlke et al. (1962: 32-3, 239-40) to demonstrate that a significant proportion of legislators' behavior is role behavior, that this behavior is substantially congruent with their role concepts, and that insight into the working of legislative bodies can therefore be gained by ascertaining their role concepts. Wahlke and his colleagues focused in particular on roles they labelled 'purposive' (essentially, what's the point of what someone does) and 'representative' (basically, who do they think they're doing it for), and argued for the utility of the concept on the grounds (Wahlke et al, 1962: 8) that it yields a model of the legislator as an acting human individual which is consistent with the basic understandings of individual and group psychology. At the same time, it yields a model of the legislature as an institutionalized human group which logically incorporates the model of the individual legislator and which relates the behavior of legislators to problems of legislative structure and function which are the traditional concern of students in the field. Buried in this rationale, and indeed in many kindred studies, is the assumption that '[t]he real payoff in explaining role orientations comes if they, in turn, explain behaviour' (Katz, 1997). And buried beneath this, is the chicken and egg question of causation: as one student of legislative roles (Davidson, 1969: 106) puts it, Do the member's attributes induce him both to express a given purposive role and to engage in activities relevant to that role? Or does the role itself exert an independent influence upon his choice of activities? The answer, according to probably the most thoroughly elaborated study of legislative roles yet published, is not so much that the behaviour and role feed back on each other - the answer Davidson himself suggested - but that their strict analytical separation is itself misguided. According to Donald Searing's work on Westminster's World (Searing, 1994) 'the roles of politicians are dynamic and adaptive patterns of goals, attitudes and behaviors' and therefore need to be explored from a more interpretive (or interpretative) than positivistic (or naturalistic) stance. But, as the passage quoted at the head of this paper asserts, this need not - indeed should not (ibid: 23) - preclude us from linking an interest in roles back into what in recent years have become the conventional modes of legislative analysis. Nor should it prevent us (ibid) from employing some of the same vocabulary: for instance, roles (or aspects of them) can be treated (though not of course at the same time) as dependent variables and independent variables. Searing's study was intended as contribution to the new institutionalism's project of bringing together, to use his terminology, the rules, norms and roles of Homo sociologicus with the individual preferences and rational calculation of Homo economicus the better to understand Homo politicus. By abandoning what he saw as an inevitably fruitless and counterproductive search for a 'general role theory' and going on 'a series of quests for particular explanations about particular types of roles in particular types of institutional contexts' (ibid.: 7), Searing hoped to rescue an analytical framework that had for some time been out of fashion - partly, in his view, because misguided researchers had wasted their time trying (and failing) to demonstrate a link between far too narrow a range of behaviour (voting) and a series of roles (typically 'trustee' and 'delegate') derived from democratic theory rather than from those 'that exist in the minds of politicians' (ibid: 12). According to Searing, there are essentially three approaches to studying politicians roles, each with its own principal topics of interest and preferred research methods. The 'structural approach' typified by Wahlke et al. was most interested in 'sets of norms linked to performance of institutional functions' that could be uncovered by highly structured, survey interviews of large samples; but it overemphasised the (albeit significant) extent to which roles were about the expectations of others and the degree of conformity with them. The 'interactional approach', typified by Fenno's participant observation studies with smaller numbers, was interested (quite properly) in behaviour in specific settings and in how roles were negotiated and picked up; but it neglected both institutional specifics and exogenous preferences. The 'motivational approach', which Searing himself preferred, is found in less theoretically-self- conscious studies (eg Barber, 1965, Woshinsky, 1973, King, 1974, Payne, Aberbach et al., 1981) based on semi-structured interviews with medium sized samples; it aims principally at describing situationally-specific role orientations that recall Wahlke et al's 'purposive roles', as well as emphasising the influence of individual (career- based, ideological or even emotional) preferences, incentives and rationalities determined not just within but also outside the institutional context. Interestingly, Searing's more interpretivist approach seems to have played only a marginal role, if any, in the handful of recent studies of the EP that do take roles seriously. All focus on roles built on textbook understandings of legislative functions and notions of representation rather than being based on the self-understandings of the subjects. Richard Katz's two studies, for example, (Katz, 1997 and 1999) employ the concept of role orientations but mainly to tap national variation (on things like protecting national interests rather than boosting European cooperation or using ones own trustee-like judgement rather than defer, delegate style to the demands of party) between a sample of candidates to the European Parliament Scully and Farrell (2003) likewise link their study of roles in the EP with both democratic theory's concerns with representation and the classic textbook functions of parliament. Using a survey carried out in September 2000 by Hix and Scully, they are concerned with 'parliamentarians' views of their role as individual representatives: how do they understand this role, what are their priorities within it, and what explains differences between them?' The role of the legislator was subdivided by the authors of the survey into six (legislating, parliamentary oversight, social group representation, representing individual citizens, developing common strategies for EU policies and mediating different social interests), and factor analysis performed to generate four distinct outlooks or 'understanding[s] of the MEP's job' that might explain the importance attached to each of the roles: these were 'party orientation', 'social arbitration', 'interest articulation', and 'parliamentarian'. Next an attempt was made to explain why some MEPs tended toward one outlook while others tended toward others. Interestingly, however, the researchers concluded (Scully and Farrell, 2003: 278-9) that measurable, systematic factors provide only a partial ability to explain differences....Thus, our suspicion [is] that much of the variance in attitudes towards the representative role may arise from individual differences not readily captured in more systematic analysis....[T]his may indicate the extent to which most European parliamentarians have thought deeply about their role, and reached differing, individual interpretations of it. Before hurrying off, however, to find new ways to tease out and explain those 'differing, individual interpretations', however, it is worth reflecting on a couple of the few 'systematic factors' that did seem to make a difference. Two, in particular, are tucked away yet stand out. The first is an observation that 'social arbitration' - 'most strongly associated with attitudes towards social group representation, social mediation and the development of common EU strategies,...suggesting a commonality between representing broad social agendas and seeking to ameliorate social problems' - was 'negatively correlated with experience in the chamber' (Scully and Farrell, 2003: 278). The second is the that 'those with greater length of service in the chamber are more inclined to emphasize the importance of traditional parliamentary activities' (ibid.: 278). These findings are interesting because they recall those of a notable investigation of the US congress (Davidson, 1969), and they are important because like that investigation, they point to the significance of socialisation when it comes to the role orientations, and by inference the behaviour, of legislators. Using over one hundred interviews with US congressmen, Roger H. Davidson, developed a taxonomy of role cognitions which predisposed those holding them to certain activities. These cognitions included 'tribune' (big on representation) 'ritualist' (big on the legislative process), 'inventor' (big on policy ideas), 'broker' (big on balancing and blending interests) and 'opportunist' (big on campaigning), while the activities focused on were legislation, playing 'errand boy' for constituents, campaigning and communicating. In so doing, Davidson came up with a crucial distinction - implicit, too, in Scully and Farrell's work - as well as a useful working definition of role orientation (or more precisely 'role cognition'), which he saw (Davidson, 1969: 97) as perhaps the most revealing single indicator we have of how legislators define their jobs. It affords a snapshot of the member's mental image of himself as a legislator. While presumably adjacent to the choices members make in performing their daily tasks, role cognition is not actual behavior in the commonly accepted usage. As an expressed attitude, it is a predisposition to behave in certain ways. This is why the normative quality of role cognition has been stressed: As a self-assessment of what the legislator is expected to do, his role cognition will pull him in certain directions and not in others. Like Scully and Farrell, and indeed numerous investigators before him, Davidson discovered that, while factors like occupational training, political experience and career patterns, 'help to illuminate the ways in which legislators respond to their environment in selecting and acting out their roles', 'so-called "social background variables" offer scant explanation of legislative role-taking' (Davidson, 1969: ix). But, more perhaps than others, he was led to a logical conclusion that may be the most interesting aspect of his study. Davidson found that, when it came to why a legislator would choose one role more than another, social characteristics were, on balance, less important than length of time and status in the legislature: in particular, freshmen tended toward the role (and associated activity) of tribune and more experienced members toward the role (and associated activity) of ritualist, in much the same way, presumably, as Scully and Farrell's less experienced MEPs emphasised their representational role while their more experienced colleagues placed more stress on their parliamentary role (Davidson, 1969: 92, 103, 106-7). This strongly suggested to him that some kind of socialisation process was taking place, but it was one he admitted was difficult to trace properly - as it would presumably have been for Scully and Farrell. This was because (like their research) his investigation was based on a one-off interview and because the number of 'freshmen' congressmen in the sample was limited (Davidson, 1969: 60). This should come as no surprise. The literature on freshmen legislators is hardly voluminous because the legislature in the country from which one might expect such literature to come - the pioneer in the field (see Barber, 1965) was from the US, after all - has a markedly low turnover: very few congressional seats are genuinely competitive and by far the majority of congressmen and women go on and on. While there has been work done, sometimes more journalistic than academic, done on new legislators in the last ten years, most of it (though see Fenno, 1991 and Docherty, 1997) concerns the fate of the Republican Revolution that was supposed to take place when an unusually large wave of new US congressmen and women were elected on the coattails of Newt Gingrich (see Killian, 1998 and Rae, 1998). Most research on 'freshmen', however, was published or carried out in the 1960s and 1970s and covered the US (eg Asher, 1973) and Canada (eg Clarke and Price. 1977, 1980, 1981 and Price, Clarke and Krause, 1976) or both (Kornberg and Thomas, 1965). Interestingly, while most of Searing's data was collected around the same time as those studies and not published until after them, his study only deals with socialisation implicitly rather than head-on. That is not to say, of course, that he regarded it as unimportant: as he notes (Searing, 1994: 483), Politicians do bring their own goals and incentives to their organizations - but then they reconstruct them there, sometimes consciously, sometimes half-consciously, sometimes unconsciously, as they adapt to these institutional environments. Interestingly, the lack of explicit attention paid to socialisation in Searing's study was picked up on by, and perhaps helped drive, one of the few more recent studies to make it its main focus (Rush and Giddings, 2002). Based mainly on written questionnaires to the new intake into the British House of Commons in 1997, its focus was on making practical, detailed suggestions to improve the institutionalised induction provided to new MPs. This focus necessarily limited its capacity to explore in depth how legislators learned the (legislative) norms and 'folkways' of the legislatures, the way their backgrounds impacted on that learning, and the extent to which they changed over time, even if a follow up survey suggested that almost half of the new intake of MPs in 1997 declared that their experience since becoming an MP had led them to 'change [their] views about what the job entails.' Notwithstanding these limitations, the research revealed, among other things, some interesting variation in the extent to which each party helps its new recruits to learn about their job, as well as the crucial part played by unofficial or informal advice from fellow members. Rush and Giddings, like Clarke and Price (1977) before them, also found that 'newly- elected members of a legislature have already been subject to pre-entry socialisation, particularly in relation of legislative norms and party cohesion', although this was balanced by the fact that 'few new MPs are familiar with the detailed operation of Westminster, not just in the procedural sense, but in a broader sense of what it is actually like to be a Member of Parliament' (Rush and Giddings, (2002: 12-13). The applicability to the EP of what little research there is out there in freshmen may be limited, primarily because it is rooted in mainly in institutions in which length of service - and what Weatherford (1985), in the US context, calls 'the seniority pyramid' - is so much more important than it is in Strasbourg and Brussels. For instance, research from the US suggests that in general freshmen legislators (and women in particular) are less likely to get their preferred committee assignments than their more senior counterparts (Frisch and Kelly, 2003), whereas in Europe committee responsibilities depend not so much on seniority as on a combination of carve-ups between the parties, communication skills and individual activism (see Mamadouh and Raunio, 2003). The purchase of the existing literature on freshmen, and indeed on all legislators, may also be limited by a universal assumption that is routinely made but only recently questioned. The assumption's paradigmatic expression is David R. Mayhew's classic Congress: the Electoral Connection (Mayhew, 1974). Mayhew argued that beneath the myriad activities undertaken by legislators – from 'advertising', through 'credit claiming' to 'position taking' – lay one, overriding goal. This goal, capable of trumping, say, policy interests or the desire to advance within the party was, of course, re-election. It may be obvious that outside the US responsibility for the latter lies more with the party than with the individual – something which helps account for the greater level of what Mayhew calls 'teamsmanship' among European legislators. It may be equally obvious that in the EP, more even than in Congress, 'there is a lack of usable lore among [legislators] on what legislative actions will produce what national electoral effects' (Mayhew, 1974: 31). Yet the presumption that potentially competing goals will be subordinated when it comes to the crunch continues to underpin much recent and otherwise valuable research on the EP (eg Fass, 2003). Accordingly, while some attention is paid to the extent to which different nomination and electoral systems may affect the extent of MEPs' dependence on their parties, virtually none is paid to the distinct possibility that not all of them prioritise re- election. This re-election assumption is a problem because recent research from other legislatures (including, now, the US congress) suggests there is a very good chance that a legislator's role, behaviour and strategy during the parliamentary term is influenced by what he or she intends to do when it finishes (Leoni, Pereira and Rennó, 2003 and Herrick, Moore and Hibbing, 1992). Ambition, or the lack of it, is similarly important (see Herrick and Moore, 1993). If, in the mind of some legislators (even if they are in the minority among their more ambitious or more conventional or simply younger colleagues), their current term is not simply a precursor to another, then we need to question our understanding of what goes on inside the legislature. This point draws our attention to another potential limitation in applying to the EP the findings and assumptions of an existing literature dominated, for the most part, by research on legislatures at the apex of their respective polities. Notwithstanding suggestions that more and more MEPs intend to make a career out of being in the EP (Scarrow, 1997), anyone studying the parliament has to take account of the fact that for many of them it does not represent the summit, either because they have already been there (or as near as they are ever going to get) or because they are still on the way up - all of which is likely to affect their role orientations and their behaviour. To be fair, this tendency to forget this 'two-level game' aspect is by no means limited to those interested in MEPs. A valuable recent volume on political elites covering twenty countries in Western Europe and Australia, Israel, Japan, New Zealand and the USA (Borchert and Zeiss, 2003), for instance, does not contain a chapter on the EP. Nor does it give much thought to how its main focus, the 'political class' is affected (at least in Europe) by multilevel governance in general and the possibility – though not as yet probability (see, for example, Rush, 2001: 14-15) – of two-way, inter- legislative movement. All of the above suggests there are more questions to ask and answer and different ways of doing both - ways which build on an already encouraging trend to eclecticism and synthesis among scholars prepared to adapt and integrate the analytical frameworks of others the better to investigate their own area of interest (see Kreppel, 2002: 213-4). This is as it should be: notwithstanding clear, crucial and sometimes systematic differences between legislatures, after all, legislative behaviour is legislative behaviour (see Mezey, 1993), and whether our primary concern is in that behaviour or in a particular institution, we need to gather as much data and as many ways of gathering it as possible, the better to build our understanding (see Maestas, Neeley and Richardson, 2003). A Searing-style 'motivational approach' to the roles of MEPs has the potential to add value to our understanding of them and the institution in which they work. In tune with the 'interpretive' turn in British political studies (see Bevir and Rhodes, 2003), which argues passionately and persuasively for the importance of researching actors' own views of their behaviour and the institutional context in which they function, it follows (Searing, 1994: 28) a rule that all of us study politicians would do well to stencil on our briefcases: If one wants to know why politicians do something, the most sensible way to begin the investigation is by asking them about their conduct and listening very carefully to what they have to say. Methodology The research on this paper draws on two sources of data . The first is a series of in- depth interviews with new MEPs. So far, thirty seven interviews have been conducted in seven interview trips to Brussels between September 2004 and March 2005. The interviewees are randomly selected from a comprehensive listing of all new MEPs and are approached through letters and follow-up emails. Interviews take between thirty minutes and an hour and are recorded and transcribed. The timing of the interviews has been both importantly intellectually and of great practical utility. The intellectual importance comes through the fact that MEPs are being approached while they are first engaging with the institution. This gives us both a 'first impression' and an important bench-mark on which we hope to gauge subsequent change. The practical utility comes from the fact that early in the process, it appears that MEPs have more open schedules and are open to being interviewed. Interviews have been conducted by both researchers where possible and individually where necessary. All interviews are conducted under the premise of anonymity and are semi-structured with some standard questions being covered. Interviews have generally taken place in English with some interviews being interpreted so allowing us to broaden the sample to a non-Anglophone sample. One problem in gauging which role orientations particular MEPs take is that politicians of any type, if directly asked to choose between various 'good's, will almost invariably tend to say that all the choices are good and that they choose all. This means simply that either an anonymous questionnaire to MEPs or presenting them with the types and asking which one they are, are invalid measures of role orientations. The socialisation of electoral campaigning reinforces the need for politicians to avoid explicitly ruling out the advocacy of any particular 'good'. The importance of interviews and particularly of semi-structured interviews is that they allow the researchers to probe and observe whether and how research subject prioritise certain orientations and downplay, or often not even mention, others. The second source of data is a database being constructed on the new MEPs. This database includes information on the background and experience of the new cohort. The focus in particular is on the previous legislative and ministerial experience of the MEPs. What is surprising is perhaps the difficulty of obtaining consistent and systematic data on the new MEPs. While CVs are on the MEPs' EP webpages these depend on what has been sent by the MEPs and their offices and do not conform to a standard pattern. We have therefore worked through these and supplemented them with material from Eurosource (2004). This research is not yet complete but we present below the findings based on 50% of the population. The MEPs covered are simply those in the first part of the alphabet, so we can expect these to be representative of the group as a whole. Table 1: Previous Political Experience of New MEPS, 2004- (N in brackets) Local Legislative experience or Mayoral experience 44.3% (90) Sub-national/regional legislative experience 13.8% (28) National Legislative experience 46.8% (95) Governmental/Ministerial Experience 23.7% (48) Source: Original research based on 203 of 405 new MEPs. This first cut at the research demonstrates that the new cohort is distinct in a number of ways. There is clearly a high degree of experience in national parliaments compared to other EPs. This may well be a consequence of the post-communist new member states bringing in a large proportion of the new cohort with much experience in them. Similar reasons may explain why the number of MEPs with ministerial experience is roughly twice that of previous parliaments as a whole. The new cohort of MEPs is clearly more experienced that its predecessors. A Typology of Role Orientations of Freshmen MEPs Interestingly, previous research on legislative roles, as we noted above, has found it hard if not impossible to demonstrate any clear links between the prior experience and demographic characteristics of new entrants and the roles they take. We therefore take our cue from the existing literature and take seriously what freshmen say about their attitudes, desires and beliefs. The central concept in our research so far -role orientations - covers these. For us, role orientations comprise patterns of beliefs, perhaps even narratives, and self-perceptions that guide behaviour. We work from an assumption that MEPs, like the rest of us, construct understandings of themselves, their context and their behaviour that lend meaning to and help determine their actions. It is because these constructions are essentially individual that we take as our unit of analysis individual parliamentarians. Our interviews are designed in such a way so as to allow us to probe in depth the different narratives, perceptions and opinions of the individual MEPs. Working inductively from our dataset of thirty-seven in-depth interviews, we can identify four ideal types of role orientations for the new MEPs: policy advocate, constituency representative, European evangelist, and institutionalist. The identification of these types is, we hope, useful in its own right, and also the basis on which to develop further models of legislative behaviour. These four types are truly ideal types, in the sense that very rarely does any particular MEP fit any one type in an unequivocal manner. The identification of these types has three major functions for us. Firstly they function as a thick descriptive device giving insight into both individual MEPs but also to the new cohort of MEPs as a collective entity. The second function of the types is to allow us to chart subsequent change and development on the part of the individual MEPs. This requires obviously further research and specifically the opportunity to reinterview MEPs at different stages in the parliamentary term. The third function of the types is as potential independent variables in charting the trajectories of individual MEPs over the parliamentary term. We hypothesise that role orientations will play a key role in determining patterns of behaviour of individual MEPs and that, in turn behaviour patterns will have a key role in determining how 'successful' MEPs will be seen to be. Policy Advocate Policy advocacy can range from very precise, technical concerns perhaps derived from particular professional (non-political) experience to broader, ideological concerns such as Euroscepticism, the promotion of market liberalism, fighting corruption, or protecting the environment or labour standards. The issue is not narrowness of focus, more an instrumental view of the institution: the EP is seen as a forum, a stage, a means to an end. There is of course an extent to which all MEPs, by standing for election on a party list, are all but obliged to display a focus on a set of policy topics. This is reinforced once in the EP by the system of committee assignments. If prompted, we suspect that all MEPs would be able to identify areas of interest or focus. But for some this commitment is more than cosmetic; it is, to coin a phrase, 'a skin not a sweater'. For these parliamentarians, policy is what they are there for, and advocacy trumps other concerns and activities and is central to their self-image. Policy advocacy has implications beyond the pushing of particular agendas. It spills over into the way those agendas are pushed. It means that the MEP is prepared to work, where necessary, across party group lines: party groups are useful only insofar as they are commensurate with the agenda being pursued - by no means always the case in the EP. The committee, by contrast, is the natural arena for policy advocates. As well as a locus for action, it provides them with a peer or reference group - a network in which they feel at home and which they believe helps them to get things done. When asked about their expectations and objectives for the parliamentary term, policy advocates were more than ready to provide a clear answer. Moreover, during the course of the interview, a policy advocate would typically consistently return with interest, engagement and often expertise to their particular enthusiasm. This was in marked contrast to those who offered more abstract answers to questions expectations and objectives and rarely came back, unprompted, to policy matters. Satisfaction for this type of MEP comes primarily from legislative activity. The policy orientation means that the primary focus of their work is likely to be getting policies through the parliament. This offers them a clear set of goals. But it also presents the difficulty that, given the diversity and size of the EP, actual attainment of these goals may well be particularly difficult Constituency Representative The constituency representative is a common role orientation found among MEPs. In the strict sense of the word, of course, there is a relatively weak link between MEPs and their constituencies. It is clear that they are less subject to constituency demands than either national or local parliamentarians because constituents use the EP less in this way. (Although we should note that there are strong national variations based on the culture and institutional parliamentary structures at domestic levels). The distance for (non-Belgian) MEPs from their constituencies compounds the issue. We take constituency here in the broadest sense, to refer to three types of constituency - local/regional, national or functional. The local/regional case of most obvious and relates to those MEPs that have usually spent a major part of their political careers at local or regional levels. This will often have served as the basis of their electoral appeal in the EP elections and therefore means that, where they have one, MEPs of this type will see their role as representing their specific electoral constituency. The national level constituency type is where the MEPs see their role as representing their country either specifically in and to the EP or to the EU institutions in general. This can take the form of either presenting a positive image of their country by 'performing' well in the EP or it can be seen in more instrumental terms so that MEPs see the EP as a forum for the pursuit of specific goals or redressing of national grievances. The final form of constituency is where the MEP places a primary emphasis on one aspect of their own identity that is not territorially defined (such as an ethnic or religious identity) and sees the EP as a forum to advance the case of that group. In these cases there is a reciprocal relationship between the MEP and that constituency that presumably will be practically reinforced as the MEP comes to publicly identify themself as a spokesperson for that group. What is notable about this type of role orientation is that it more easily provides a clear potential set of objectives for MEPs and has the implication that there are bench- marks by which they can 'measure' their success in the future. These bench-marks are the delivery of selective benefits but perhaps, more importantly for our focus on role orientations, measures of success are likely to come more in the perception of success in representing constituency interests during the parliamentary term from the individual MEPs themselves (have I delivered what I felt I should?), from their constituencies (have they delivered what they said they would?) and from the other MEPs (have they done a good job in representing their group?). A practical implication of the constituency representative role orientation is that it places a strong emphasis on time spent in their constituencies. Substantively and symbolically, there are powerful incentives for such MEPs to spend a lot of time travelling between Brussels/Strasbourg and their constituencies. This would obviously apply to those weeks devoted to constituency business in the EP calendar but it also clearly meant that many constituency representatives were feeling obligated to travel every week so that they could spend a long weekend 'back home'. The personal cost of this sort of working life is clearly very high for such MEPs. Those MEPs that define constituency in national terms are likely to place a greater emphasis on the role of national delegations than those that are more narrow in their constituency orientation. This means that such MEPs commonly see their peer (or reference) group in terms of their national groups and so this has the effect of cutting across party lines. It is notable that this national constituency orientation appears more common in the new member states and especially the smaller new member states. European Evangelist For many MEPs election to the EP is a means by which they can promote the European project to national constituencies. MEPs displaying this role orientation will have a strong ideological commitment to the principle of European integration or will have had previous European or international political experience or both. For some MEPs election to the EP is part of a natural chain of political development for someone who has strong ideological commitment to the European project. For others it was clearly a means by which they could remain at the 'European' level and a transition from other European or Brussels-based institutions. The EP is a means of spreading the message of European integration. The European evangelist role orientation means that MEPs are prepared to work across national and party lines. A focus on the party group or national interest will too often feel at odds with their 'European' orientation. Their attention is drawn to issues that are pan-European which tend to require the broadest coalitions of support. The pro-European aspect of the European evangelist has a particular implication for their relationship to the EP. The EP is clearly a complex and somewhat daunting institutional setting for MEPs but European evangelists have strong rational and instinctive pressures that lead them to downplay the difficulties of the institution. Simply put the pro-European position means that they are predisposed to a pro-EP position. This can have important consequences for these MEPs. One aspect of European evangelism is that it reverses the relationship between MEP and constituency displayed by constituency representatives. For European evangelists the direction of the relationship is downward from Europe towards their constituency rather than bottom-up. This means that it affords the MEPs a great deal of autonomy in defining how and how much they relate to their constituencies. But it does mean that the European evangelists may share a strong emphasis on time spent in constituencies with constituency representatives. The time spent in constituencies for evangelists is perceived as for talking about and promoting European integration, whereas, for constituency representatives, it is perceived that this time is for listening and for establishing connections with constituents and constituency groups. It is in the nature of the role orientation of European evangelists that it does not provide a clear set of objectives by which MEPs may seek to both guide and evaluate their parliamentary behaviour. While this may seem a constraint, it tends to be more of an opportunity for such MEPs. The role orientation allows them both latitude and a high degree of autonomy in determining what they focus on in the EP and how they focus upon it. Where milestones are needed by European evangelists they tend to find them in key symbolic moments of European integration. An example of this would be a strong and explicit identification with the European constitutional treaty. In practice, MEPs have a relatively limited role in securing the ratification of this treaty but it is a frequently mentioned issue and goal for European evangelists. One aspect that we need to be aware of with this type is that it has a limited comparative utility. The unusual nature of the EP and the international nature of the European project itself means that we would not expect to find an equivalent of the European evangelist in national parliaments. There is not the need in national legislatures to 'sell' their nationhood to sub-national constituencies. However, the frequent occurrence of this type of role orientation among our sample of MEPs so far means that we cannot simply ignore what is clearly a prevalent role orientation. Institutionalist The final category of institutionalist is designed to capture those MEPs whose orientation is towards the institution they are in. For these MEPs the institution is not a means to an end but an end itself. The satisfaction of their work comes from the EP itself rather than from what it can deliver. While, in theory, the institution is the EP, in practice, institutionalists tend to be oriented towards the embedded institutions of the party groups. There is strong variation in how far new MEPs feel institutionally 'at home' in the EP and the most obvious agent for integrating the MEPs is the party group. It is clear from the interviews that these are perceived very differently by different MEPs. Of course there is variation according to what party group they are a member of but the differences go deeper than that. Some MEPs clearly feel the party groups are an essential component of their parliamentary lives while others feel them to be, at best, tangential. One thing that has also come up has been the tendency for MEPs to be unaware that there are very different attitudes towards the party groups. MEPs tend, perhaps naturally enough, to generalise from their own relationships to the party groups to other MEPs. Institutionalists focus on the party groups as a central component of their parliamentary lives. Institutionalists feel naturally at home in the party groups and will make them the basis of their reference group and the object of some of their aspirations. One interviewee described the party group as the 'family' and this gives a clear cue as to the role the party group plays. This type divides into two different manifestations. The first is the leadership aspirant. where they have either explicit or implicit leadership aspirations. If the role orientation is towards the operation of the institution, then it makes sense that such MEPs may feel that a desirable goal is to offer leadership to that institution. But some institutionalists may not harbour leadership aspirations and some were clear that their decision to run for the EP was made out of party loyalty - and specifically a sense of duty towards their party and, to use the terminology of Searing (1994), we characterise these as subalterns. In respect to the dispersion of the types of role orientations we can make some preliminary observations. The first observation is that there is a wide variation in the backgrounds of the new MEPs, both in terms of their professional lives and political experience. New MEPs vary from being heavily experienced national politicians including former Prime Ministers and foreign ministers through those with extensive experience in other EU and European/international institutions to those who are essentially 'political virgins' with the campaign of 2004 being their first formal political experiences. The second preliminary observation we can make is that patterns and types of role orientation cut across background characteristics, so that, for example, those without parliamentary experience take on the same range of roles as those with that experience. This leads us to suspect that role orientations, while no doubt related to experience, draw on a deeper level of political orientation. Those with political experience have often commented that the EP is such a new environment and one of such diversity that they have felt previous experience to be only a limited resource in helping them operate in the EP. The third preliminary observation is that patterns and types of role orientation cut across party and nationality. This simply means that we do not see any clear relationship between MEP role orientation and party group membership or between role orientation and nationality. While this is perhaps to be expected with the larger party groups of the centre-left and the centre-right, it is notable that it is true of smaller party groups such as the Green and new left groupings, Group of the Greens/European Free Alliance and Confederal Group of the European United Left - Nordic Green Left. The fourth preliminary observation derives from the fact that we have anecdotal evidence of persistent non-attendance in the EP by a minority of members. In this case, it is likely that our sample of will be systematically biased against the inclusion of such members. If there exists a role orientation that, for whatever reason, results in MEPs essentially opting out from the EP, then it is important that we can gauge how widespread or how limited this group is in size. We should also note that our sample has included MEPs who have a dual mandate and who presumably have less time to devote to EP work and therefore to research projects on the EP. Conclusions Our starting point in this research has been to focus on the parliamentary rather than on the European dimension of the EP. Our reason for doing this is to make our research explicitly comparative. But to do this we need to identify those aspects of the EP which are distinctive. For us there are four of these: first, the EP is an international institution; second, the party groups do not yet provide as strong a set of cues as parties do in national parliaments; third, the scope, size and competencies of the EP are still far less fixed than those of national parliaments; fourth, in the EP there is no clear institutional or party linkage between parliamentarians and the executive as there is in other systems. All this means MEPs operate in a more fluid institutional environment, and therefore one in which roles may be more negotiable. Our work so far has allowed and encouraged us, in line with the literature, to inductively develop a taxonomy of role orientations for the 2004 intake of freshmen MEPs. This provides a benchmark from which we can go on to do two things. First, we can attempt to see whether these role orientations lead MEPs to behave in distinct ways. Second, we can chart whether MEPs roles change over time and in response to being and acting in the EP. We, like some of the MEPs themselves, have been surprised by the relative lack of institutionalised procedures and processes for easing new parliamentarians into what for nearly all of them is an unfamiliar and complex working environment. Travelling up this steep learning curve will almost certainly impact on the roles they take. Finding out what they can and cannot do will have consequences for both their attitudes, desires and beliefs and their behaviour. Different roles will be differently affected by this journey along the learning curve. Beginning with the European evangelist, we expect this role to be easier to maintain and less subject to change than others. The European project is sufficiently diffuse that evidence of its success can always be found by enthusiasts, just as, for Eurosceptics, evidence of failure is always around. MEPs taking this role will generally enjoy the greatest degree of latitude in an institution that already grants its members considerable degree of freedom. In contrast we believe policy advocates may have the greatest difficulty in sustaining their roles. This is because, we assume, the EP's relative lack of purchase on policy militates against an MEP notching up clear policy wins. For those constituency representative types whose constituency is defined in national terms, we expect that they, like the European evangelists, will have a relatively easy time in sustaining their roles. Again, the breadth of an agenda that is about using the EP to defend national interests and the ability (indeed perhaps need) to identify those interests in selective ways, means that frustrations should be relatively rare. For constituency representatives whose constituencies are more specific, we expect that there will be more dissonance within this role as the EP provides a difficult arena for the unequivocally successful delivery of constituency service. Institutionalists of the leadership aspirant type will have difficulty in sustaining these roles if they are unsuccessful in attaining office. Unusually for a parliament, the EP does provide the opportunity for freshmen MEPs to be relatively successful relatively quickly. Conversely, those unlikely to succeed will realise this equally swiftly and may well switch roles accordingly. The EP is the sum of its parts, and therefore the sum of the roles its members take. If, as we hypothesise, those roles change over time, then so, too, will the institution and, quite possibly, its place in the political system (Hix, 2005 ) of the European Union. Identifying the roles orientations of freshmen MEPs as we have attempted to do, albeit in preliminary fashion, is a first step to putting together those orientations with the actions that one might expect to flow from them and then tracking institutionally- constrained and enabled changes in the 'attitudes intertwined with behavior' we call roles. References Aberbach, Joel, Putnam, Robert, D. and Rockman, Bert A. (1981) Bureaucrats and Politicians in Western Democracies (Cambridge, Mass: Harvard University Press). Barber, James David (1965) The Lawmakers: Recruitment and Adaptation to Legislative Life (New Haven: Yale University Press). Bevir Mark and Rhodes, RAW (2003) Interpreting British Politics (London: Routledge). Borchert, Jens and Zeiss, Jürgen (2003) The Political Class in Advanced Democracies: a Comparative Handbook (Oxford: Oxford University Press). Checkel, Jeffrey (2003) '"Going Native in Europe": Theorizing Social Interaction in European Institutions', Comparative Political Studies, 36 (1-2), pp. 209-231. Clarke, Harold D. and Price, Richard G. (1977) 'A note on the pre-nomination role socialisation of freshmen members of Parliament', Canadian Journal of Political Science, 10 (2), pp. 391-406. Clarke, Harold D. and Price, Richard G.(1980) 'Freshmen MPs' Job Images: the Effects of Incumbency, Ambition and Position', Canadian Journal of Political Science, 13 (3), pp. 583-606. Clarke, Harold D. and Price, Richard G. (1981) 'Parliamentary experience and representational role orientations in Canada', Legislative Studies Quarterly, 6 (3), pp. 373-90. Corbett, Richard, Jacobs, Francis and Shackleton (2003a) The European Parliament, Fifth Edition (London: John Harper). Corbett, Richard, Jacobs, Francis and Shackleton (2003b) 'The European Parliament at Fifty: a View from the Inside', Journal of Common Market Studies, 41 (2) pp. 353- 73. Docherty, David C. (1997) Mr Smith goes to Ottawa. Life in the House of Commons (Vancouver: University of British Columbia Press) Eurosource 2005 (2004) Dod's Parliamentary Communications. Fass, Thorsten (2003) 'To Defect or Not to Defect? National, Institutional and Party Group Pressures on MEPs and their Consequences for Party Group Cohesion in the European Parliament', European Journal of Political Research, 42 (6) pp. 841-866. Fenno, Richard F. (1991) Learning to Legislate: the Senate Education of Arlen Specter (Washington DC: Congressional Quarterly Press). Franklin, Mark and Scarrow, Susan (1999) 'Making Europeans: The Socializing Power of the European Parliament', in Richard Katz and Bernard Wessels (eds) The European Parliament, National Parliaments, and European Integration (Oxford: Oxford University Press). Frisch, S.A. and Kelly, S.Q. (2003) 'A Place at the Table: Women's Committee Requests and Women's Committee Assignments in the US House', Women and Politics, 25 (3), pp. 1-26. Herrick, Rebekah and Moore, Michael K. (1993) 'Political Ambition's Effect on Legislative Behavior: Schlesinger's Typology Reconsidered and Revised', Journal of Politics, 55 (3), pp. 765-776. Herrick, Rebekah, Moore, Michael K. and Hibbing, John R. (1994) 'Unfastening the Electoral Connection: the Behavior of US Representatives when Reelection Is No Longer a Factor', Journal of Politics, 56 (1), pp. 214-27. Hix, Simon (2005) The Political System of the European Union (Basingstoke: Macmillan). Hix, Simon (2002) Parliamentary behavior with two principals: preferences, parties, and voting in the European Parliament', American Journal of Political Science, 46 (3), pp. 688-698. Hix, Simon, Kreppel, Amie and Noury, Abdul (2003) 'The Party System in the European Parliament: Collusive or Competitive?', Journal of Common Market Studies, 41 (2), pp. 309-31. Hix, Simon, Raunio, Tapio, and Scully, Roger (2003) 'Fifty years on: research on the European Parliament', Journal of Common Market Studies, 41 (2), pp. 191-202. Hooghe, Liesbet and Marks, Gary (2001) Multilevel Governance and European Integration (Oxford: Rowman and Littlefield). Katz, Richard S. (1997) 'Representational Roles', European Journal of Political Research 32 (2) pp. 211-226. Katz, Richard S. (1999) 'Role Orientations in Parliaments', in Richard S. Katz and Bernard Wessels (eds) The European Parliament, National Parliaments and European Integration (Oxford: Oxford University Press). Killian, Linda (1998) The Freshmen: What Happened to the Republican Revolution? (Westview Press). King, Anthony (1974) British Members of Parliament: a Self-Portrait (London: Macmillan). Kornberg, Allan and Thomas, Norman C. (1965) 'The Political Socialization of National Legislative Elites in the United States and Canada', Journal of Politics, 27, pp. 761-775. Kreppel, Amie (2002) The European Parliament and the Supranational Party System: a Study in Institutional Development (Cambridge: Cambridge University Press). Leal, D.L. (2002) 'Home is Where the Heart is: Congressional Tenure, Retirement, and the Implications for Representation', American Politics Research, 30 (3), pp. 265- 284. Leoni, E, Pereira, C, and Rennó, L (2003) 'Estragégias para sobreviver politicamente: esscolhas de carreiras na câmara de deputados do Brasil [Strategies for Surviving Politically: Career Choices in the Brazilian Chamber of Deputies]', Opinião Pública, 40 (1), pp. 44-67. Mamadouh, Virgine and Raunio, Tapio (2003) 'The Committee System: Powers, Appointments and Report Allocation', Journal of Common Market Studies, 41 (2), pp. 333-51. Masestas, C, Neeley, GW, Richardson, LE (2003) 'The State of Surveying Legislators: Dilemmas and Suggestions', State Politics and Policy Quarterly, 3 (1), pp. 90-108. Mayhew, David R. (1974) Congress: the Electoral Connection (London: Yale University Press). Mezey, Michael (1993) 'Legislatures: Individual Purpose and Institutional Performance' in Ada W. Finifter, ed. Political Science: The State of the Discipline, Revised Edition (Washington, D.C.: American Political Science Association). Noury, Abdul (2002) 'Ideology, Nationality and Euro-Parliamentarians', European Union Politics, 3 (1), pp. 33-58. Payne, James L. (1980) 'Show Horses and Work Horses in the United States House of Representatives', Polity, 12 (3), pp. 428-456. Price, Richard G., Clarke, Harold D. and Krause, R. (1976) 'The socialisation of freshmen legislators: the case of Canadian MPs' in Jon H. Pammett and M.S. Whittington (eds.) Foundations of Political Culture (Toronto: Macmillan) Rae, Nicol C. (1998) Conservative Reformers: the Republican Freshmen and the Lessons of the 104th Congress (ME Sharpe). Rush, Michael (2001) 'Parallel but Separate: Political Recruitment at Three Levels in the UK', Paper Delivered at the ECPR Joint Sessions of Workshops, Grenoble, 6-11 April. Scarrow, Susan (1997) 'Political Career Paths and the European Parliament', Legislative Studies Quarterly, 22 (2), pp. 253-62. Scully, Roger (2003) 'Review Article. External Change, Internal Development: Studying the European Parliament', Government and Opposition. Scully, Roger (2005) Becoming Europeans? Attitudes, Roles and Socialization in the European Parliament (Basingstoke: Palgrave). Searing, Donald D. (1994) Westminster's World: Understanding Political Roles (Cambridge, MA: Harvard University Press). van der Eijk, Cees and Franklin, Mark (1996) Choosing Europe? The European Electorate and National Politics in the Face of Union (Ann Arbor: University of Michigan Press). Wahlke, John, Eulau, Heinz, Buchanan, William and Ferguson, Leroy C. (1962) The Legislative System (New York: John Wiley and Sons, 1962). Weatherford, J. McIver (1985) Tribes on the Hill (South Hadley, Mass: Bergin & Garvey). Westlake, Martin (1998) 'The European Parliament's Emerging Powers of Appointment', Journal of Common Market Studies, 36 (3) pp.431-44. Woshinsky, Oliver H. (1973) The French Deputy (Lexington, Mass: DC Heath and Co.). The research is funded by the Nuffield Foundation (UK). The authors would like to thank Brigid Fowler (University of Birmingham) for her extensive practical help with the interviews. We would also like to thank our Sussex team of Jo Amos and Dan Keith as well as Michael Shackleton (European Parliament) and Daniel Denruyter of the European Parliament Information Service. We also are particularly appreciative of the help and time of all the MEPs who have given us time for interviews and to their assistants for help in making this possible. The anonymity guaranteed to interviewees means that we are at pains not to provide specific illustrations of general points in this paper where this would identify the MEPs. We should perhaps note here that there are cases where MEPs have taken on the interests of a group of which they are not themselves a part. These cases we regard as policy advocates rather than as constituency representatives. 1