Introduction

The MP’s position in the Turkish political system: central, but dominated

The Parliament within the institutional framework

3Since the multiparty system (1946), the Parliament is an institution with a limited autonomy. After the first competitive elections, the Parliament became the centre of the political life during several decades, in spite of the coups d’état in 1960 and 1971. The 1961 constitution set up institutions to counterbalance the weight of the National Assembly, notably the MGK (Milli Güvenlik Kurumu) and even more so the Senate, part of the Senators being installed to oversee that the civilian politicians abided by the constitution. The system remained bicameral until the 1971 coup d’état.

4The 1982 constitution introduced a kind of ‘rationalised parliamentarianism’ in a securitarian regime with the MGK functioning as a parallel government (İnsel 2004; Dorronsoro 2005). For instance, until very recently, the military budget remained de facto outside the control of the Parliament even if the TBMM’s Defence Committee discussed the military budget with the Ministry of Defence. Questions related to the ‘national security’ (defined in a very extensive way) were largely decided by the MGK. Moreover, the relation between the executive and the legislative branch shows a strong and increasing influence of the President, who has the right to veto any passed law and to appeal to the Constitutional Court (Tachau 1980; Heper 1996). This evolution depends of course on the balance of powers. For example, Demirel had less influence than Özal, and the weight of Sezer is linked to the support of the military apparatus.

The MPs

5A weakened Parliament does not necessarily mean that MPs are not individually recognized. It is known that in some circumstances, members of the Parliament benefit from the antiparliamentarianism of the electorate. Polls indicate an important difference between popularity ratings of the Parliament and of the MPs themselves in the United States. However, according to the reports of the European Commission (European Commission 2002: 29), there appears to be a bad image both of the Turkish Parliament (47% tending to trust), of the MPs, and even more so of the parties (16% tending to trust). It is true that the trust rate in the Italian, Spanish, Danish, and the British Parliaments are even lower, but in a 2001 TESEV survey, to be a MP is ranked as the most dishonest job by a large percentage of the population (Pérouse 2005: 226).

6The MPs themselves are aware that their political career could be short. Firstly, it is interesting to note the cyclical trend in the ratio of re-election during the last decades. The number of re-elected MPs is very low after brutal changes in the political offer (1950, 1960, 1980, 2002), but grows afterwards up to more or less 50% (Massicard 2003: 33, table 8). Secondly, the duration of their mandate is precarious, since it depends not only on political reversals but also on the electoral rhythm. In this respect, we should point out that early elections are usual in Turkey: since the beginning of the multiparty system in 1946, ten out of fifteen legislative elections were early ones. Despite the fact that the article 77 of the 1982 constitution brings the term of legislature from four years – as it was since the Ottoman Empire – to five years, all elections have been early ones since then. Thirdly, the volatility of the electorate is amplified since the 1990’s by important swings in the election results, showing the rejection of the parties in power. In the aftermath of the 1982 constitution and the ban on the old political parties, party identification sharply diminished and the coalition governments in the 1990’s have been generally unpopular. On the whole, when the turnover is high, as it is the case since the 1990’s, very few MPs are assured of being re-elected, and the threat of a removal by the electorate, by the party, or that their party doesn’t reach the 10% threshold, is very present.

7Nevertheless, MPs are an important component in the relation between the electorate and the central power. They are able to pass on demands from the population to the centre, even if the impossibility to assume several (local and national) mandates at the same time limits in practice their influence. They also mediate between different interest groups and the centre. However, powerful lobbies such as industrialists or unions can deal directly with the government. Besides, MPs do not control much resources compared to other political actors. Indeed, the Istanbul mayors, as early as the 1960’s, starting with Faruk Ilgaz and Haşim İşçan, got considerable resources form their position in local and national politics. Since metropolitan municipalities were established in 1984, their mayors detain considerable resources. Decentralisation projects might accentuate this trend and largesse out of their city resources that every mayor can distribute, while most MPs fail to control significant resources.

Sources and state of the question

8Primary qualitative sources about MPs and parliamentarianism are relatively numerous, in contrast to the weakness of statistical and quantitative data. Besides minutes of debates in plenary sessions (which have hardly been exploited), primary sources are made up of publications from TBMM, official bulletins as well as the series of Turkey in Statistics. There are also lots of autobiographies or memoirs (Kirişçioğlu 1968; Ağaoğlu 1969; Burçak 1976; Perin 1990; Kocaoğlu 2003). Finally, we find many biographies of famous politicians – mostly party leaders – written by journalists, but which generally do not concentrate on the Parliament.

10Otherwise, it is noticeable that many political scientists have lost interest in the functioning of the Parliament and the office of MP. For instance, Sarıbay’s work of political sociology (1994) doesn’t address the political elite or activities of politicians which reflects both the objectively weak position of the institution and new trends in Turkish political science. However, we can still rely on the works of Gençkaya, Hazama, Kalaycıoğlu, Özbudun and Turan (see bibliography). Turan (1985) and Kalaycıoğlu (1988, 1995, 1999) addressed the behaviour of the MPs in the TBMM, the time spent on ‘case work’ as well as the committee activities of the MPs in the 1980s and the 1990s. Furthermore, there are some local studies, sometimes of great quality, describing the notables – some of them being MPs – in a more anthropological way (Unbehaun 1994).

11Since this thematic issue focuses on relations between the MPs’ social background and their way to fulfil their parliamentary roles (see infra), we would like to address what has been a significant part of the literature about MPs: the relation between the political field and the society. This issue has mostly been addressed in the 1960’s and 1970’s by studies producing and analysing statistics about the social characteristics – mostly the professional background – of MPs. The two dominant paradigms in those studies, Marxism and developmentalism, share the same methodological presuppositions and, most of the time, their results are compatible.

12The study of Ünsal (1982) on the profile of the 1977 MPs falls within the scope of a critical Marxist sociology. Through the analysis of the social background of the MPs, the author notes a difference between the social composition of the TBMM and the society at large, notably the under-representation of the working class within the TBMM. The underlying idea is the denunciation of the confiscation of political power by the elite (Kızır 2002).

13Other authors have tried to show the gradual diversification of the political elite since the beginning of the Republic, in a developmentalist – thus fundamentally diachronic – way: ‘Stages of political development, if not actual distributions of power, may be marked by regularities in the differential participation in formal politics of distinctive social groups’ (Frey 1965: 180). They are based on the hypothesis of a relation between macro-social changes and the sociological profile of the MPs. By using statistical data on the social status of MPs (essentially their occupation, education, sex), researchers explain political developments like the institutionalisation of the political field, the professionalisation of the political elite, or the increasing pluralism among the social groups represented in the parliament (Tachau 1980). Those results can be summed up in two statements: the diversification of the elite, and the opposition between the statist elite and those coming from other social groups.

14Starting from the relation between social profiles and party membership, several authors (Tachau 1980; Heper et al. 1996) notice an opposition between a group of bureaucrats and a political elite coming from the middle-class. Both social groups, through ‘their’ parties, fight for state resources. The (implicit) scheme is often that the bureaucracy might represent long-term interests of the country and has a stabilizing influence on the political system. Then, ‘general interest’ is considered to be objective more than constructed, and bureaucrats as its best advocates, because they are preserved from ‘populist pressure’.

15The second statement is the diversification of the elite. Ünsal starts from the hypothesis of an opposition between a provincial and a national elite. According to Tachau (1980: 220), the Demokrat Parti (DP) represents the interests of provincial middle-classes in the 1950’s, and ‘localism’ –which he defines after Frey as birth in the represented district– is increasing. Later, the rise of new social groups (engineers, etc.) and the diversification of the political elite have been stressed upon. After the 1980 coup d’état, a significant proportion of religious figures –imams or graduated from imam-hatip schools– are to be found in the different parties run by Erbakan. Starting from that statement, some works, drawing on the theory of social coalitions (Tarrow 1977), assume that a political party comes to power when it manages to build an ‘inter-class coalition’, to pose as the representative of one or several social groups.

16We might call into question this paradigm through two main critics. First, data processing seems questionable. On the one hand, the construction of long statistical series while the political system has been disrupted several times, is methodologically problematic, but sometimes not even commented. In some cases, data concerning non-competitive and competitive periods are used as if they were homogenous, although continuity can hardly be established between the one-party system and the multi-party period. Moreover, the coherence of categories used to establish biographies is not called into question. For instance, the status of a profession can change in a significant way in the course of several decades, and the stable presence of a certain amount of MPs coming from one professional sector can hide changing social dynamics. On the other hand, data are insufficient, and some pieces of information are not reliable: for example, the plurality of professional experiences is rarely mentioned, most of the time because of the sources used – the TBMM albums only mention one profession. In the same way, simplified socio-professional data are favoured to the detriment of those linked to practical experiences and social skills. Finally, the selection of relevant elements in biographies is very simplified, because it does not describe personal trajectories but static social positions. Therefore, the evolution of the MP during his mandate is not taken into account, whereas fame and personal enrichment for example can alter his social position during his mandate.

17These statistical constructions are a consequence of the general hypothesis concerning the political and social fields. Developmentalist and Marxist paradigms tend to postulate simple relations between social groups and political representatives. The political system appears as a (relatively) transparent mechanism dealing with the demands of social groups. Even when addressing issues like professionalisation or institutionalisation, the autonomy of the political field and its capacity to transform social resources are put in brackets. It appears that the opposition between the bureaucratic elite and the one coming from the middle-class can be deceptive. The election of bureaucrats can be the result of their bigger capacity to mobilise state resources and to build a clientele in this way, which qualifies the idea that bureaucrats are immune to ‘populist’ drifts. In another way, local elites (provincial notables) are constituted in relation to the state (Meeker 2002). Besides, this duality does not work anymore in the 1970’s, since there is an extreme politicisation within the institutions themselves. As a consequence, the hypothesis of a diversification of the elite led by the political development in Turkey has to be questioned. All those elements challenge the idea of a transparency between society and politics.

Resources and roles

18Choosing MPs as the object of study opens different options. We have decided to focus on the articulation between two paradigms, the mobilization of resources and role perspective. First, we address the issue of the mobilization of resources in the political field. The resources are defined here as what is efficiently mobilized in the political field to win election, a position of power inside a party, etc. Hence, we do not consider the social profile of the MPs per se, but the actualisation of resources in the political field. For example, it is not directly the professional background that is pertinent as such but the way the practice of a profession gives some means to enter politics. So, we have to describe the concrete mobilization of resources, for example in an electoral competition. In this perspective, the analysis of the trajectory of politicians highlights the readjustment of personal resources which permits to enter the political field (Best et al. 2000). We make the hypothesis that for the elections, the value of these resources is to be considered on a local level, since the level of electoral competition is the province. Hence, it is difficult to use data built exclusively on a national level.

19Second, the concern for the effective conditions of the practice of a political mandate leads to an analysis in terms of ‘social roles’ and to stress the learning of behaviours according to norms considered as legitimate (Briquet 1994: 17-18; Müller et al. 1997). In his everyday practice, a MP is constantly compelled to signify his conformity to a set of norms that define his role. Entering the role implicates having the practical knowledge of what is convenient to do when one has to fit into successive interactions. These ascriptions of roles are multiple and heterogeneous since roles are defined by formal and informal ascriptions from different actors, juridical norms, etc. Elected people are interacting in different social spaces, which are not necessarily structured by specifically political principles. The ascriptions of roles are contextual: they depend on the characteristics of the publics to which one is confronted. As such, they are not necessarily homogeneous at the national level. Behaviours expected from a MP are a product of a particular, sometimes localized, configuration, and might appear weird in other parts of the country. ‘The concept of ‘role-set’ is to be distinguished from the concept of ‘multiple roles’, which are associated not with a single social status but with the various statuses in which individual operates’ (Saafeld 1997: 39-40).

20Our endeavour is to link the question of resources with that of roles. In the same perspective, Turan (1985) linked the practise of transfers from one party to another with the resources of the MPs. In a first step, we see how the political field is characterised by mechanisms of selection and transformation of resources. The analysis of the trajectories of MPs, through their initial resources (before their election), the accumulation of specific resources during the mandate, and then their possible retaining afterwards, is intended to test the hypothesis of the ‘autonomisation’ of the political field. In a second step, we deal with the complex role-set of MPs, which can be analysed through three loci: within the party, within the Parliament, within the constituency. From this, we will question the relation between the resources of MPs and the way they interpret and fulfil their roles. Because of the many institutional breakdowns, the temporal scope of the core analysis begins after 1980, even if a broader historical perspective is taken in account in some papers. Let us first present the resources perspective, then the role perspective, keeping in mind that this distinction is only analytical, and does not imply any determinist causal relation.

I. The resources and their transformation

21Since we take into account the transformation of resources as a dynamic and temporal phenomenon, we address first the resources of the MPs before the election and their local value, then the trajectories of MPs during and after their mandate. On this basis, we shall then question the professionalisation of MPs.

Non-political resources

22Non-political resources mobilized in the political field can be listed in a non-exhaustive way as economic capital, proximity to state institutions, professional skills, activism, social capital, local roots, religious charisma, and family membership. In that sense, available data are often insufficient to give a clear view of these resources, and their reduction to the very profession would be deceptive. These resources work in different ways according to periods and places; some can be illegitimate for some time. For instance, if we exclude the first Parliament, it was very exceptional for a man whose main profession involved religious matters, to become an MP (Koçak in this issue). The religious charismatic resource has been de facto forbidden for a candidate before 1946 and after the different coup d’état, but favoured in the 1980’s and 1990’s by the centre-right parties like ANAP, DYP, and Islamist parties such as RF and FP (Dorronsoro in this issue). In a diachronic perspective, we can also ask why certain kinds of capital are more easily transferable in the political field according to socio-economical situations. For instance, the presence of engineers is probably linked, more than to a questionable rise of the middle-class, to a change in the organisation of this profession and to the emergence of a technicist discourse in politics (Göle 1990).

23During the single-party period, professional profiles like high rank army members and bureaucrats who had reached the higher echelons in state administration could be rewarded after their retirement by being made MPs. These civil servants formed the most crowded group in Parliament. According to Frey’s study (1965: 114-122), the rate of this group in the Parliaments of the single-party period ranged from 23% to 57%. The number of parliamentarians who were self-employed, as well as tradesmen and industrialists from the ‘private enterprise’, were much lower (Koçak in this issue). Another resource which was valorised was regime loyalty for the country’s forthcoming artists, writers, poets and journalists. They were often made MPs, but were also supposed to be foremost spokesmen of the regime (Günay in this issue).

24How are the professional resources used in the political field? According to the results of Frey (1965), completed by latter studies, three professional groups are over-represented within the Turkish Parliament: bureaucrats and lawyers in the early Republic, and engineers since the 1970’s. As we have seen, the number of civil servants is important during the one-party period before decreasing regularly, whereas liberal professions experience a parallel rise, notably just before the multi-party system (Frey 1965; Tachau 1980: 207). Within this group, lawyers are the most numerous, except in 1957.

25Frey explains the importance of the lawyers by the capacity to resume one’s initial job and a ‘natural’ link between the occupations of lawyer and of MP (Frey 1965: 111-112). We can challenge these explanations by questioning instead social dispositions and professional habitus gained through accomplishing a profession, and their latter use in the political field. So, we can analyse how specific resources linked to the profession of lawyer are reinvested in the political field. In Turkey, the presence of lawyers can be interpreted in different ways according to the period. Until the 1960’s, the lawyer is seen as a notable. Hence, his election as a MP is linked to his local status. After the 1970’s, we can emphasis the role of the professional chambers as a locus of politicisation (Bora 2000; Dorronsoro in this issue).

26In some cases, the initial profession is the gate to enter politics, but at the same time renders the day-to-day MP’s occupation difficult, since the two professional habitus are deeply contradictory. Günay in her essay underlines that some writers co-opted to the one-party Parliament as prestigious intellectuals or, in later periods, as ordinary competitors in the election process have often disturbed the power holders either as MPs in their parliamentary activities, or as writers in their literary work, and sometimes as both.

The value of resources

27The value of the resources is to be defined in relation to the local political field. The characterization of the local political field is complex; thus, when Tachau underlines the differentiated profile of the MPs according to the level of development of their constituency (1980), the choice of this unique criterion is problematic. Here comes the question of the varying political value of resources in different contexts. For instance, inherited charismatic religious capital is largely confined to some regional (mainly, but not only, South-Eastern) and social (rural or little cities) contexts, while other resources seem really national. In a mostly Kurdish-populated province like Diyarbakır, militants without any notoriety or social capital can be elected to a major city on the list of a Kurdist party in municipal elections (Dorronsoro in this volume). Belonging to a mezhep (Alevi or Sunnite) can constitute a major resource as well in contexts of differentiation or tension like in the late 1970’s (Cumhuriyet 1977). Thus, resources have a specific value depending on the local political field.

28According to Tachau (1980), MPs are increasingly oriented towards ‘localism’, because they are more and more born in the constituency where they are elected. However, this fact is difficult to interpret, because it postulates a natural relation between place of birth and the ability to mobilize local resources. Instead, it seems relevant to reflect upon the local use of national resources and vice-versa. In other words, the ‘local’ can be constituted as a resource in different ways: first, the ‘local’ can be constructed as a legitimising category in political discourses and media; second, party apparatus can establish hemşehri associations, or links can be formed between both (Schüler 1998; Kurtoğlu 2004; ejts 2005). In that way, a pitchforked candidate can take benefit from a strong local presence, if he enjoys the support of a local party apparatus.

29During the single-party period, the ‘local’ was sometimes valorised as a political resource, sometimes not. Some parliamentarians had never ever been to the region they were elected from, and, possibly, were never ever to visit this region as MPs, either. It was not uncommon for those who were to become MPs to hear about this either through the dailies or through radio news. It would be very hard to state that these parliamentarians really did represent ‘their’ regions. From time to time some of them were re-elected from entirely different regions. By contrast, there were also parliamentarians who were born in and had grown up in the region from which they were elected. These had a certain kind of relationship to the region, for they were the ‘wealthy’ of the region. Especially in areas of the country where a feudal agricultural structure prevailed, wealthy landowners were constantly in Parliament as MPs. In particular, the notables, who had positive and close relations with party headquarters, easily entered and were able to remain in Parliament. Therefore, local ties were sometimes valorised by the Kemalist regime (Koçak in this issue), depending however on the political situation in the region. For example, during the military administration of the South-East, most of the Diyarbakır MPs were civil servants with no local ties; but all MPs elected in 1946 –at the beginning of the multiparty period– were local notables (Dorronsoro in this issue).

The accumulation of specific resources during the mandate

30The MP mandate is only a moment in a personal trajectory. How do MPs take advantage of the resources accumulated during mandate afterwards? In a dynamic perspective, the mandate – especially the first of a MP – is a period when the MP social status is changing. From an anecdotic point of view, former MPs’ visiting cards almost always mention their former position, which indicates a willingness to put to the fore the social capital associated with it. Investing in politics represents a risk; resources can increase or, in some cases, decrease. For instance, a notable – with an important initial social capital – whose passage in politics is marked by scandals can lose part of his respectability. As they expose publicly their political preferences during their mandate, professionals can alienate some of their former clients (Massicard in this issue).

31Resources accumulated during the mandate may vary depending on how the roles have been fulfilled. In general, we can assume an accumulation of social capital, because the role of the MP puts in contact individuals and institutions. The MP benefits from a modest wage and numerous advantages in nature (even if the accommodations were recently withdrawn). But, more important, economical capital can be acquired irregularly, for example through public contracts [ihale], a possibility which may not be accessible for all MPs, but only MPs from the party in power or with good connections to decision-making bodies. Indirectly, the relations created with bureaucrats and politicians as an MP can be useful and possibly help to increase one’s economic capital during or after the mandate.

32For the 1920-1946 period, Koçak suggests that some parliamentarians would quit Parliament in order to become ambassador or governor. Those who worked in the public service in deep loyalty to the regime could return to their former duties, if they wished. Being a deputy was just another way of being in the public service. The MPs, who additionally became Ministers, formed an even narrower group, as the turnover was quite low. For the multiparty period, Dorronsoro underlines that a significant part of Diyarbakır MPs tend to settle in Istanbul or Ankara after their mandate, a fact which can be interpreted as a part of the formation of a national elite and the professionalisation of politics.

Professionalisation and autonomy of the political field

33The existence of a specialized political field leads to consider the professionalisation of political agents through a set of questions. The professionalisation of politicians and the rationalization of this occupation should be distinguished. Whereas professionalisation is generally linked to the central rule of parties, notably in Germany (Saafeld 1997), rationalisation as an expertise – the mobilisation of specific techniques requiring important resources showing through notably during election campaigns, political marketing, etc. –may not be necessarily linked to parties. In some cases, candidates benefiting from enough resources can display the same level of rationalization than candidates supported by a party, relying for example on lobbies.

35Second, the level of autonomy of the political field is varying, as well as its ability to influence other social fields (for the relations between the political and the literary field, see Günay in this issue). Periods of crisis and periods of routine should be distinguished, since they affect the autonomy of the political field. Crisis of the political system (declining objectivisation of social roles, institutional breakdowns, etc.) permit a ‘dedifferentiation’ of the fields. In other words, in spite of the tendency toward a professionalisation of political elites, the inverse mechanism can be observed during periods of crisis, with the opening of the field to outsiders. The 2002 legislative elections, with the multiplication of parties, an important turnover of MPs, and the strong presence of MPs without political experience (Massicard in this issue), can be an example. Thus, we might find an important opening of the field in those contexts: the investment of resources in the political field is then facilitated, the entrance ticket is low, even for the creation of parties, and non-political attributes are valorised. Logically, we find a correlation between periods of crisis and a growing presence of non-political capitals (Dorronsoro in this issue).

II. Roles and resources

36Who defines the role of MP? First, there is a juridical definition of the role of MP. Legal norms define institutional roles (regulations of parties about the designation of candidates, poll systems, degree of ‘rationalization’ of the parliamentarian system, etc).

37Second, the parties have very strong role prescriptions concerning MPs. These role prescriptions are even more important since the MPs highly depend on their party for re-election (see infra).

38Third, the ascription of roles also comes from the representations concerning the MPs. What are they in contemporary Turkey? First, there are representations of politics by those who practice it, elected representatives themselves, through their discourses in terms of moral obligation or their believing in legitimacy given by election. But there are also representations among the electorate about efficiency, the place given to service [hizmet], about what is licit and what is illicit. The construction by journalists of the occupation of MP is of great importance in the formation of the ascription of roles, even if representatives themselves are far from being passive. Turkish MPs are relatively absent in the media, except party leaders and when some incidents like scuffles occur in Parliament. Parliamentary debates are broadcast on TRT, but they rarely appear in newspapers. Violence – verbal, but also physical – occurs from time to time, leading to scandals: it is perceived as such a transgression of the proper behaviour of a MP that it is then broadcast in the news bulletin (Günay and Aslan-Akman in this issue). Through those performances, representations of elected members do constrain their behaviour.

39The main tasks recognized to MPs are first, legislative activities which are constituted as follow: work in parliamentary committees; parliamentary debates (when and how do MPs intervene in the proceedings?), and speeches from the desk (written and oral questions to the government) (Bakırcı 2000). Besides the legislative activities, another task recognised to MPs consists of transmitting individual requests – in general, but not exclusively, from their electorate – and answering them. This role is both legitimate (a MP has to be at the service of the people and accessible to its requests) and illegitimate (as far as these practices are often disparaged as being particularistic, flouting the public good and the general interest) (Güneş-Ayata 1994: 62). Despite of this ambiguity, it is quasi institutionalised, since parliamentary sessions start at 15 pm, and mornings are devoted to visits. One may observe queues of unequal size, but sometimes really impressive, in front of MPs’ office. This role of resort is all the more marked since MPs are often seen as being very powerful, although themselves are quite aware of the limits of their power.

40MPs learn on the job (Kocaoğlu 2003); thus, the learning process varies from one individual to another: it depends on the conditions of access to deputation, according to whether they have been active in a party and assumed responsibilities in it, whether they have involved in the associative or trade-union work, or whether their good fortune is due to supports outside the political field. Thus, the apprenticeship of the roles also depends on the social properties of the candidates, because they predispose them more or less to assume the prescribed rules. Meetings – sometimes informal – of elected representatives and what happens behind the scene permit the circulation of practical know-how, warnings and advices. Information and knowledge necessary to fulfil obligations linked to the roles are thus transmitted.

41The process of learning to be a MP is comparable to the apprenticeship of know-how and of knowledge constituting roles, which catch MPs with variable intensity and different kinds of obligations according whether they subscribe to it with conviction, reason or duty. Therefore, ascriptions of roles are constraints for MPs, their personal history having prepared them more or less to accept. There seem to be no strong standardization of role learning in Turkey, neither through the main ways of transmission of political roles (party and elected offices) nor through the professionalisation of legislative activity (Massicard in this issue). The conditions in which MPs learn the political occupation show that the socialisation can only be limited and that habitus are only marginally modified.

42The political occupation leads to interact in several social worlds (Lagroye 1994). Being a representative implicates being able to play different roles, which may also appear to be contradictory (Briquet 1994; Müller et al. 1997). Empirical evidence suggests that the problem of disturbances in role-sets is a very real one for modern Members of Parliament.’ (Saalfeld 1997: 40). These contradictory role ascriptions are perceived as such by the MPs, even if interiorized differently (Massicard in this issue). Whatever the MP’s dominant characteristic (a grassroots candidate or a candidate supported by a party), the role he is said to incarnate or the one he is favouring, he has to take into account the plurality of his roles. One may observe them through the succession of behaviours and ways of being in everyday practices of MPs, obliging them to resort to different legitimacy registers. Thus, MPs use different know-how, practices and beliefs, which they cannot forget without taking the risk to offend one of the publics they are in relation with. In some cases, behaviours of MPs within the Parliament can interfere with the voters’ requests and their responsibilities in their party. Sometimes, parliamentary sessions start late or are cancelled because of the absence of quorum, or because MPs, busy with some other tasks, are not available.

43How do individuals take prescribed roles up? The ways to fulfil these roles depend notably on the resources of MPs, leading them for example to attend or to escape public meetings, to engage more or less in legislative work, etc. (Massicard in this issue). For example, MPs without political experience, put on the list by a party mainly because of their extra-political skills which can be useful for parliamentary activities, will mostly engage more in legislative work than, say, MPs elected with strong personal social support, and who depend less on their party’s backing (Massicard in this issue). Therefore, new social profiles among MPs can lead to changes in the ways to practice this occupation (Norton 1997).

The centrality of the party

44To understand the MPs occupation in the Parliament, the relations between MPs and their parties are central. All the essays in this issue (Aslan-Akman, Günay, Koçak, Massicard) underline the central role of the parties in the day-to-day activities of the MP in the TBMM. It is not supposed here that every party is organised in the same way, the importance of elected members is obviously different for instance in ANAP and MHP and may evolve in every party.

45The centrality of the parties derives both from institutional regulations and political rationality. Firstly, the 1982 constitution and the Rules of Procedure adopted in 1996 aimed to increase the efficiency of the Parliament, thereby limiting the autonomy of the individual MPs, reinforce the place of the party group (Gençkaya 1999). Secondly, the great majority of representatives are in a state of dependence vis-à-vis the parties for their re-election. Because of the electoral system, the major step for the MP is his nomination by the party on an eligible position in the list more than the electoral campaign itself. The fact that the parties determine a candidate’s election more than his personal resources affects the MPs’ behaviour within the Parliament, for example in relation to party discipline.

47Many clues indicate that parties constrain MPs’ behaviour, at least concerning the discipline of vote. We shall now see in details the weight of party in the organization of the MPs’ work and then their level of autonomy. Is the legislative activity of a MP correlated to his place in the party? The first point here is that the party group decides upon the attribution of the different parliamentary positions, i.e. member or chair of a committee etc., which in turn highly constraints the MPs’ role-orientation (Massicard in this issue).

48The second point is that the political configuration can define the MP as an opposition MP with a specific role. Aslan-Akman focuses on CHP MPs with a strong position in the party in the precise context of the 22^nd Assembly characterized by a high level of polarization between the two parties. In her essay, she shows that the behaviour of MPs of the main opposition party is both strongly constraint by rules of procedure and defined by the party, may be leading to the ‘existence of an opposition MP role’.

49A way for MPs to escape party pressure is to change party, a common practice of transfers since the beginnings of multiparty politics. Historically, there have been high levels of inter-party mobility in the Turkish Parliament. The elected representatives are able to move quite easily from a party to another, for instance if they think that they are not positioned at an eligible level on the list, or that their party won’t go through the 10% threshold (Massicard 2003). At least before 1980, party changing appears to improve career prospects (Turan 1985). In his first account of the specific profile of the party changers, Turan (1985) showed that, in the case of individual transfers, party changers tended to come more from small and least developed constituencies.

50As political contexts influence party changing decisions by individual deputies and factions, Turan’s paper in this issue examines the developments of inter-party mobility since the 1980s (Turan et al.). Feeling that party changing not only encouraged political instability and was unethical, and fearing that it might undermine the two-party system they were trying to build, the military makers of the 1982 constitution introduced measures to prevent it. These failed to stop the mobility of deputies who found many ways to circumvent restrictions to move between parties. However, an interesting difference with the period before 1980 is that individual party changers and non-changers do not appear any more to have different attributes. It would be interesting to investigate further if changers and non-changers show differences in terms of resources. Is party-changing correlated to some special way to behave as a MP, for example constituency-service orientated?

MPs and their constituencies

51Another important dimension of the MP’s occupation is his relation to his constituency. It is the service and allocational aspects of responsiveness that compose the most frequent demands of constituents to parliamentarians in Turkey. In terms of daily activities, parliamentarians spend for constituency service the largest part of their time at their own disposal (Hazama in this issue). What do we know about the interactions about MPs and their constituents? Localised studies like those of Güneş-Ayata (1990), and Unbehaun (1994), without focusing on the local role of MPs, set on MPs in a context of interaction among bureaucrats, local politicians and notables. By doing so, they give sometimes sketchy, but interesting indications concerning the role of MPs in their constituency. However, almost all these case studies deal with rural Turkey or small towns. But in a widely urbanised society, it is risky to generalise clientelist relations, which are by definition encompassing and undifferentiated. Modalities of exchange are probably different in urban contexts. Attempts to grasp them there in a more precise way (Erder 1996, Schüler 1998, Kurtoğlu 2004) have not really dealt with MPs and their role in the relation between electors and parties.

52It is then important to clarify the different dimensions of the relation between MPs and their constituents. The term constituency service includes different kinds of activities: mainly personal services to the citizens, activities encouraging infrastructural services or job-creating investments in the constituency or more generally collective goods, which are often related to collective actors like local lobbies. Personal services is what legislative studies usually call ‘case work’ and what is often called in Turkish hizmet: Kalaycıoğlu (1995) asserts that constituents approach MPs for three main reasons: to find jobs, to get promotion or transfers at a certain job, and to seek health care. One also expects MPs to help to go through the bureaucracy.

53Why are MPs doing hizmet? In a proportional representational system, Turkey’s province-size constituencies generate a significant size of personal votes for a candidate, though nominated in the party list. On the base of a questionnaire on the XIX^th Assembly – an Assembly elected with preference system and, for some parties, primaries – Hazama explores how and to what extent parliamentarians are responding to demands from constituents and why they do so. Second, for the purpose of re-election, parliamentarians consider their individual activities to be almost as important as the party popularity. Those activities are publicized through the local media, including the press, the most important of which, however, being the constituents that they meet in person.

54Do the characteristics of the constituency orientate the role of the MP toward constituency service? According to Hazama, the demand for and supply of constituency service depend on different politico-economic structures of the constituency (Hazama in this issue). The smaller the constituency size, the larger will be the demand for and the supply of constituency service. This is because in small constituencies parliamentarians are more recognizable by their constituents and the weight of personal votes is larger than in large constituencies. Thus, even under proportional representation, relatively small constituencies in Turkey make constituency service important for parliamentarians seeking re-election. In addition, since the Turkish parliamentarian’s turnover rate is very high, even re-elected parliamentarians have to take constituency service seriously. The statistical test showed no significant difference in the practices of constituency service between newly elected and re-elected parliamentarians. Massicard suggests that MPs from big cities are faced with less direct representative duties. But the downside is that these MPs have a much harder time than their colleagues from smaller constituencies to get their names and pictures into the newspapers on a regular basis or to answer the demands in a way that can bring sizeable social support. In a constituency like Istanbul, it is very difficult to construct a local base (Massicard in this issue).

55Which kind of MPs tries to increase their local resources, or to take the role of local representatives? Massicard suggests that MPs elected without strong local support hardly entertain any kind of localism, probably because social support is an insecure resource, and enhancing it may appear more time and energy-consuming than, say, demonstrating loyalty to the party. It seems that social support is a resource which is invested by MPs as a secondary choice or as a complement to other resources. However, having a social strategy for a MP is not necessarily outside the party, since some parties tend to encourage their MPs to do constituency service.

Conclusions

56There is a theoretical profit in displacing the analysis from the MPs’ biographies to the resources they mobilize. The downside is that new data have to be produced, since the exiting ones are not tailored to this approach. In order to understand the resources mobilized both in political campaigns and during the MP mandate, more attention should be drawn to the activities in the constituencies. This in turn implies working at the constituency level, which is where electoral competition takes place and the relative value of resources can best be understood. Working on resources also constitutes an empirical way to address the autonomy of the political field which, in the Turkish case, varies heavily as shown by the strong and punctual increase of non-political resources mobilized by MPs in times of crisis.

57There is a relation between the resources mobilized by the MPs and the way they fulfil their roles. This issue indicates that the parties are central not only in the administrative functioning of the Parliament and for the MPs’ career, but also in the very definition of individual MPs’ role-orientation. The centrality of the relation to the party in role-orientation, however, does not necessarily imply increasing professionalisation. As a matter of fact, non-partisan and non-political resources – especially professional ones – can strongly impact the role-orientation, but also be decisive in the very selection of MPs by parties. As a consequence, strong contradictions between previously acquired non-political professional habitus and values, and the MP role-set, are likely to appear, resulting on role-conflict. Finally, constituency service appears to be less central in defining role-orientation than the party, even if it is spectacular and often time-absorbing for the MPs.