Manos Savvakis and Manolis Tzanakis (2004) 'The Researcher, the Field and the Issue of Entry: Two Cases of Ethnographic Research Concerning Asylums in Greece'
Sociological Research Online, vol. 9, no. 2, <http://www.socresonline.org.uk/9/2/savvakis.html>
Received: 24 Sep 2003 Accepted: 24 May 2004 Published: 31 May 2004
Abstract
The way the researcher enters the research field can constitute a privileged mode of observing the structure and qualities of the research field, particularly in qualitative sociological inquiries. In the process of the initial contact of the researcher with a social place, especially in those cases when his/her physical presence is required, the structural features of the place gradually manifest themselves. Quite often, a strictly 'technical' approach to research-work tends to overlook the potential usefulness of this phase. In this article, we will put forward the hypothesis that by investigating the way research participants observe the researcher, especially during the initial stage of interaction, we can gain useful knowledge regarding particular structural aspects of the research field.Keywords: Bias; Biographical-Narrative Method; Biography; Ethnographic Research; Participant Observation; Research Field
The Question of Reflection and its Relation to Research Practice
1.1 The entry into the research field and the involvement of the researcher with the informants/narrators constitute two significant issues that emerge during qualitatively[1] oriented fieldwork. Many scholars reflect on these issues as vital elements of their inquires, suggesting that they might highly influence the final research outcome. They also propose that these elements of qualitatively oriented empirical research can be potentially methodologically and analytically used in order to gain a more spherical and elaborated picture of the case under examination (Bytheway, 1993; Evans 1993; Humphrey, 1993; Rosie, 1993; Temple, 1997; Collins, 1998; Scott, 1998).
1.2 Besides, some of them recommend that without the researcher's personal involvement and commitment to the informants/narrators, which is both a practical and pedagogic constituent of qualitative inquires, the final research outcome runs the danger of being a standardized account that will not offer 'actual' access to the intrinsic qualities of the social worlds under investigation (Kaufmann, 1996; Scott, 1998; Collins, 1998). In other words, these scholars essentially put forward the issue of the researcher's personal emotions and beliefs as fostering elements of a sophisticated sensibility that can enrich sociological understanding and promote practical action (Wilkins, 1998, pp. 93-100).
1.3 Thus, in every sociological inquiry, the way the researcher is initiated into the social world under examination constitutes an index to the intrinsic qualities and structure of this world. In other words, it comprises an indicator to the formal and informal hierarchies that exist there-namely, all those elements that impart to the research field its particular logic and idiosyncrasy (Angrosino, 1989; Denzin & Lincoln, 1994; Coffey & Atkinson, 1996; Denzin, 1997).
1.4 In qualitative social inquiries, the respect for the informants/narrators, the game of communication, the theoretical sensitivity and the unavoidable involvement of the researcher represent pivotal elements, which seem to strongly influence the final research result (Denzin, 1989, 1996; Riessman, 1993; Hollway & Jefferson, 2000; Chamberlayne et al 2000, pp. 167-180). Researchers, most of the times, bring in their own social world and think grounded in their own social experiences, which are linked to corresponding practices and interpretive schemes.
1.5 These experiences reflect internalized social knowledge and skills that very often betoken distinct social and cultural modes of understanding and explanation. This happens because the researcher very often belongs to different 'lived worlds' (Habermas, 1987, pp. 23-176) from those of the informants/narrators. The different social place the researcher occupies within the social world, the distinct conceptual schemes s/he employs, and the 'symbolic violence' (Bourdieu, 1999) that s/he always seems to exercise during the research process may have partial negative effects and distort the analytic process.
1.6 Despite the difficulties indicated above, the different places the researcher and the informants/narrators occupy almost always mutually shape the 'research object' in a potentially open and communicative way (Kaufmann, 1996, pp. 19-24). The effects of the broader social world-that is of the 'interpretive horizon' (Gadamer, 1975) of the researcher-are impossible to entirely neutralize. Social researchers live, think and act within the frameworks of particular social times and spaces. In addition, they face certain limitations and command particular possibilities as a result of their inherent relation to the social world. Among others things, researchers always seem to be condemned to act as inter- mediators between different perspectives, between different ways of grasping the social and between different ways of 'reading and writing history' (Carr, 1999).
1.7 The research act itself forces the researcher to intermediate through a manifold of logics of the social worlds under investigation in order to mutually shape with the informants/narrators an 'obvious object of scientific study' (Glaser & Strauss, 1967; Travers, 2001). Many times, these social worlds are forced through indirect symbolic violence to expose itself with all its potential breaks and antinomies. Consequently, we can claim that in qualitatively oriented social inquiries, the researcher's involvement is desirable for it reveals some features of the underlying structure of the field (Kaufmann, 1996; Scott, 1998, Collins, 1998). Besides, it is judged as a methodological sine-qua-non. The researcher's detachment in relating to the informants/narrators and his/her exaggerated research formality and conventionality seem to finally result in a corresponding detachment of the informants/narrators. According to Kaufmann, the 'petition of neutrality' quite often results into a disassociated and typified description (Kaufmann, 1996).
1.8 On the other hand, the biographical- narrative method, for example, is often accused of registering social phenomena without managing to grasp their concrete historical dimension (Bourdieu, 1993, 1999, pp. 131-139). Bourdieu, for instance, speaks of the 'biographical illusion' (Bourdieu, 1999, pp. 131-139). In his point of view, biographical narration is a process of construction that is mutually shaped by the narrator and the researcher. This requires of the subject a coherent story and a competence to narrate it. Thus, the narrator and the interviewer reinforce underlying social obligations that presuppose un-fragmented individuals. This, for Bourdieu, is a form of ideological illusion that ultimately overlooks the fundamental division of the social world into partial social fields and the impact of economic, social and cultural inequalities (Bourdieu, 1999, pp. 131-139).
1.9 Without solely focusing on the methodological implications of the previous remarks, we could, paraphrasing Bourdieu, speak of the 'illusion of society as the object of sociology'. Society never presents itself to the subject of research experience as something 'out there' or as an a-historically constituted 'object'. The 'objects' of sociology and sociology as science per se are essentially social artifacts (Combessie, 1996, p. 60; Ion, 1997, pp. 20- 21). In other words, the actors involved maintain a certain amount of social knowledge, employ their own ways of coping with the given situations, interpret the social world and therefore act in order to transform it (Stanley, 1993, pp. 41-52; Wilkins, 1993, pp. 93-100).
1.10Sociological knowledge is a priori involved into its 'object of study' and a priori pre-interpreted by the schemes of its contemporariness and the researchers' particular social, temporal and spatial frameworks. In the social sciences especially, it is rather futile to search for an Archimedean point of leverage for an ultimate sort of truth. This occurs for both the object of sociology as an 'object' and the research gaze that examines it are 'results' and at the same time 'producers' of historical processes (Gadamer, 1975; Ricoeur, 1990, p. 244).
1.11 An elaborated understanding of the historicity of the 'object of sociological study'-what Gadamer describes as 'interpretive reflection' (Gadamer, 1975, pp. 107-135) - does not strictly imply restrictive and negative consequences regarding the research act. Besides, it cannot simply and one-dimensionally be understood as a refusal of the possibility of ultimate grounding (Champagne, 1999). On the contrary, one could posit that fruitful interpretive perspectives can be put forward taking into account the issues raised above. These perspectives concern the exploration of social processes in ways that relativise methodological fetishism and the doctrinaire insistence on positivistic criteria of verification.
1.12 The dogmatic insistence on the 'use of the right method' simultaneously indicates that we are attempting to reconstruct social reality based on ad hoc sociological concepts, an effort that implies normative claims regarding social reality. Thus, we run the risk of miring ourselves and understanding silence (i.e. someone's refusal to give an interview) as a mere 'technical dysfunction', an obstacle to fulfilling the criterion of sample representativity, and not as a structural characteristic of the social place under investigation (Coffey & Atkinson, 1996, pp. 214- 217; Collins, 1998). This implies that the fact that a person has the right to speak publicly (i.e. in an interview) is not a simple issue of methodological significance but also an issue of structurally conferred authority (Temple, 1997; Collins 1998; Scott, 1998).
1.13 The relationship between the researcher and the informants/narrators is complex, shifting and fluid and it is always, even in the more 'loose or hierarchical contexts', inter-subjectively structured. This implies that it is not always crystal clear where the equilibrium of power or authority lies and what are the 'real' practical advantages of the researcher's symbolic capital. In other words, the informants/narrators also hold power and might, if they are forced or challenged to, 'bring into play' it against a pressing situation/question. In this negotiating and communicative process, which actually constitutes selves and transforms identities, we can detect costs, losses and gains. Apart form the previous, we also find out an enormous effort to improve the quality of the data in a reflexive way, for the social word is always an ambivalent and ambiguous universe that holds well-hidden petit secrets, which are not exposed to the researcher without personal involvement (Collins, 1998).
1.14 In this corpus of methodological difficulties-typical of fieldwork-based inquiries-we can identify a 'privileged forum' of analyzing the structure of the research field (Schwartz, 1993, p. 272)-that of the researcher's entry to the field.A strictly technical and instrumental approach to these refusals to speak, which are quite often encountered in fieldwork, their reading as mere 'methodological failures' or as failures to gather a 'representative sample', would distort the structure of the research field as this is crystallized in a particular historical moment, namely the moment of the research act. These 'technical difficulties' constitute integral elements of the field. In these methodological fallacies we can detect the grounding of a general exhortation to take advantage of 'bias' (Temple, 1997).
1.15 This taking advantage of 'bias' should not be overlooked or simply downgraded to the status of 'sampling problem' (Kaufmann, 1996, p. 66; Guibert & Jumel, 1997, pp. 100- 124; Temple, 1997). It should not also be posed to a mere unscientific issue that does not concern or does not influence the final research result, for the social world is neither a ready-made entity waiting 'out there' to be discovered; nor a monolithic and static reality that is irrelevant or indifferent to the interpretative schemes or the actions of the agents involved[2].
1.16 In every case, however, the researcher should reflect on the way s/he enters the field of inquiry (Laplantine, 1996, p. 21; Guilbert & Jumel, 1997, p. 100; Scott, 1998). For instance, the mediator influences the way the researcher is welcomed and perceived. In addition, this reflection on the way the researcher enters the field and is received by the informants/narrators constitutes an integral dimension of his/her analysis for it reveals the way the research field deals with and perceives him/her. According to Ricoeur, the understanding of the other is simultaneously self-reflection and every knowledge of the other is always a dynamic process of self- knowing (Ricoeur, 1990, p. 40).
1.17 Under the light of the previous theoretical thoughts, the present analysis is focused on the entry of the researcher in the empirical research field and on his/her first contact with the informants/narrators. In the present contribution, two cases[3] of empirical research are examined. The first case concerns the psychiatric reform in Greece, especially in Crete, an endeavor that took place the last eighteen years. The second case regards the Leprosarium of Spinalonga in Crete, an institution that was established in 1903 and terminated its function in 1957.
1.18 Both cases are related to asylums of confinement, which functioned in the geographical area of Crete. Based on these two research projects, which seem to hold certain analogies and differentiations, two examples of entering the research field are analyzed. These two empirical examples, which are based on two particular case studies, seem to demonstrate corresponding structural elements of the social worlds under examination (Wengraf, 2000, pp. 140-164; Travers 2001).
1.19 In the first case, which is based on the psychiatric reform in Greece, what is put forward is the weight of the professional status of the researcher. In the second case, which is based on the Leprosarium of Spinalonga, what is highlighted is the importance of the common geographical origin of the researcher and the narrators. In both cases, these elements, which emerge during the process of entering the respective research fields, demonstrate intrinsic characteristics of the cases examined and do not simply concern technical dimensions of the research process. We might even argue that on the spectrum of attitudes towards the researcher, these two cases occupy relatively extreme positions: the first introducing a negative bias and the second a correspondingly positive bias to the research.
1.20As we mentioned earlier, the present paper examines in the first case (psychiatric reform in Greece) the 'professional status' of the researcher and in the second case (Leprosarium of Spinalonga) the 'common origin' of the researcher and the informants as constitutive elements of the research process based on three criteria.
1.21 More particularly, these criteria are strongly related to:
- The acceptance of the researcher by the informants/narrators,
- The expansion of the theoretical sampling and
- The internal division of the empirical research field.
1.22 The core aim is to demonstrate the presuppositions and problems of the entry into the research field as issues of major methodological significance, which is to say that these issues are qualitatively different from the way a possible classification or reduction would understand them- that is as 'simple technical dysfunctions'.
First Case: 'I do not have the right to speak in public ...'
2.1 The first case is based on participant observation in community psychiatry services and concerns psychiatric reform in Greece (1985-2003). As we mentioned earlier, for the past eighteen years a reform of the psychiatric services is being attempted in Greece. For many psychiatrists, this undertaking essentially involves the replacement of the 'asylum' model' with a 'community model' of psychiatric treatment (Tzanakis, 2003, pp. 138- 197). This process is intensely conflictual and therefore results in competitive perspectives of evaluating the process of transformation, particularly on the level of everyday life.
2.2 The three prisms for evaluating these processes of transformation regarding the field of Greek psychiatry schematically are:
- 'Traditional' psychiatry,
- 'Critical' psychiatry and
- 'Modernizing' psychiatry (Tzanakis, 2003, pp. 138-197).
- The very structure of the field as such and
- The relations among the involved parts.
2.16 It is obvious, therefore, that all these potential interpretations regarding the 'identity of the institution' are not equal with respect to the validity claims they put forward. There exists a constantly negotiable division of labor concerning the 'authority of representation' and finally the 'right to speak' within the context of a hierarchically constituted social world (Strauss, 1992).
2.17 The constitution of the 'object' is part of this process of managing the public image and is connected to its strength or fragility. The project of the subject of the research experience is the result of an intermediation, which takes the form of a constant negotiation with the participants in the field, in other words with all these persons of whom s/he asks, even by his/her simple physical presence, to tell him/her 'what they are doing', 'how they understand it', and 'how they interpret it'. The power structure within the field finally affects the constitution of the 'object'. It is through this structure of the field itself that the persons participating in it receive the social researcher, listen, and finally speak or refuse contact with him/her.
Second Case: 'Since you come from the homeland...'
3.1 The second case, as we mentioned before, is based on historical material and biographical-narrative interviews and concerns the community of lepers and the Leprosarium of Spinalonga, Crete (1903- 1957).
3.2 The effort to maintain the fragile communicative and personal balances and to control or at least strive to control personal emotions and opinions sometimes distorts the research process but does not seem to overall cancel the research endeavor[4]. In our example, we encountered cases where the narrators asked the researcher 'what is he doing', 'why is he hanging around' and 'where does he come from', submitting him to the game of 'exchanging good impressions, that is presenting and negotiating oneself in public' (Goffman, 1963, pp. 34- 56; Collins, 1998)-a 'game' that the researcher was called to 'play' in a way that would not interfere with, but respect, the everyday routines of the informants[5].
3.3 Particular factors also facilitated the overcoming of initial difficulties and the entry into the research field. Out of these factors, the fact that the researcher shared a common geographical origin with some of the narrators proved to be one of the more significant elements during the entry phase into the research field. The common origin as a criterion of acceptance of the researcher by the narrators finally seemed to function as an important reason for communication and as a happy coincidence. The research subjects eventually felt more comfortable and relaxed when they realized where the researcher hailed from.
3.4 What is characteristic, in relation to the previous issue, is the reception that the researcher received during his very first research visit, when he announced his name, and place of birth. The reaction of a female narrator was indicative: 'Well, so you also come from Crete. You are a compatriot, our child so to speak. Certainly, certainly, this is the case, it is quite obvious from your name...'. The common geographical origin largely contributed to the acceptance of the researcher as a person that the narrators could potentially trust and confide lived experiences to. The narrators felt that they symbolically communicated with an 'indigenous person' who knew their birthplace and origins, had visited those places, and had-at least on an imaginary level-common references with them. The researcher was also seen as a person who brought information and news from the homeland.
3.5 In addition, the common geographical origin seemed to create a powerful ethical bond between the researcher and the narrators and a moral obligation not to reject the researcher without a sort of negotiation. The common provenance contributed to potential mutual cooperation and to the creation of a communicative atmosphere, elements that are of vital importance in fieldwork involving biographical-narrative interviews or oral histories (Bertaux & Kohli, 1984, pp. 206-217; Corbin & Strauss, 1987, pp. 249-281; Wilkins, 1993, pp. 93-100; Travers, 2001).
3.6 Consequently, we see that the common geographical origin of the researcher and the narrators-a fact that is both 'objective' and a product of multiple personal interpretations-contributed in a material way to his entry to the research field. This mode of entry contributed one of the many others elements for understanding the structure of the research field and the relations among narrators. In other words, we can detect in this case for another time the relevance of the researcher's biographical background in doing qualitatively sociological inquires (Denzin, 1989; Angrosino, 1989; Collins, 1998).
3.7 The common geographical origin as a criterion of expanding the theoretical sampling seemed to function as a 'passport' during contacts with people from other geographical regions at a later stage of the research process. The researcher had successfully passed the 'crash test of trustworthiness'. This implied that the narrators could introduce him to his friends as 'the compatriot that traveled so many kilometers and came from our island to visit us here because he is writing a book on Spinalonga and would like to have a few words with you'. In this case, the initial acceptance of the researcher constituted a 'visa' or a 'gateway' to the interior of the research field, which gradually began to reveal some of its unknown but essential elements to the 'friendly and important stranger', namely the particular researcher.
3.8 Furthermore, this common origin seemed to work also as a starting criterion for expanding the theoretical sampling. This happened because--among other things--the rest of the narrators, who hailed from other places, felt that they could confide personal stories and lived experiences to the researcher without creating problems or exposing persons. This 'gossip' had already taken place with some of their friends without the emergence of serious problems.
3.9 Besides, it seemed that they did not wish to be excluded from a process that deemed their personal accounts as essential with respect to the 'writing of a book[6]'. In addition, the fact that the everyday life of the narrators was minimally distracted contributed to the familiarity with and acceptance of 'the strange man, who was hanging around asking questions regarding the past and especially Spinalonga'.
3.10 In relation to the common geographical origin as a criterion of expanding the theoretical sampling, we might add that narrators who came from other places viewed the fact of this difference in origins as an opportunity to exchange information and opinions with the researcher concerning the situation at their birthplaces in relation to the topic examined. Thus, the researcher gradually began to create a comparative framework, gathering a lot of valuable information that could be employed analytically in order to enrich and cast light upon the issues under investigation.
3.11 In addition, the common geographical origin functioned as a decoding key of certain qualities of the empirical research field. The common origin contributed to an initial operational 'diagnosis' of the divisions and alliances apparent in the research field. As it later turned out one of the most fundamental internal divisions of the narrators was based on geographical origin. In other words, one of the deep and intrinsic divisions of the research field was that between people that came from Crete (Kritikoi) and people that came from other areas in Greece and specifically from the mainland (Palaioelladites).
3.12 Differences in geographical origin influenced the way the narrators understood themselves and the way they judged issues of central importance-for instance, the issue of the continuation or termination of the Leprosarium and the transfer of patients to the Antileprosy Station in Athens. As becomes obvious this is an issue of great importance that is not simply related to 'sampling problems'.
3.13 Differences of geographical origin also affected the way the narrators positioned themselves inside the field and the way they chose to narrate their life stories. Geographical origin became a means of recognition of potential kindred persons inside the community of the narrators. At the same time, origin seemed to constitute the ground for identity. This, in the end, constituted the basis of internal divisions. Finally, origin led people to choose different places for social gatherings and to put forward different interpretations of social reality.
3.14 Provenance does not neutralize the structure of the empirical field. On the contrary, it emphasizes the multiplicity and dynamics intrinsic in every empirical research effort. Origin as a bedrock of differentiation negatively crystallizes, at least on an initial level, the position of the narrators in the social milieu of the empirical research field and strongly influences its social cartography.
3.15 The original dichotomies that are based on the polarity 'Kritikos-Palaioelladitis' shape different definitions and presentations of the self and different evaluations of the situation. They also constitute different ways of understanding social realities and shape different versions of the examined topic. At the same time, the narrators constitute a fragmented 'we', which is divided on the basis of geographical origin, and a community of ex-inmate leper patients, who form a potentially unified collectivity.
Conclusions
4.1 Social researchers and especially sociologists should take into serious consideration the social places of the informants in order to locate their position in the social fields they study. In other words, they have to 'objectify' the point of view of the informants/narrators and the information (Combessie, 1996). For this reason, they could orient their research to this sort of information that is connected to contrasting social viewpoints. The differentiation among informants potentially increases the possibilities of objectifying the findings (Kaufmann, 1996; Ion, 1997).
4.2 This 'bet of differentiation' gains importance when the informant/narrator is also the mediator between the researcher and the subjects who participate in the research. Consequently, the mediator, acting as a connecting figure, influences and advises the researcher during his/her contacts with individuals or groups of the social world under examination. Quite often, this 'accommodation' concerns people that are socially related to the mediator. This network of persons and groups pre-configures the information and the representations the researcher finally accesses (Ion, 1997; Creswell, 1998).
4.3 If research is also a form of art, this is the case precisely because it demands a creative synthesis and an energetic contribution toward the creation of what is called the 'scientific object of study' (Atkinson, 1992; Schwartz, 1993; Kaufmann, 1996). Research is an art of bringing things into being and the 'poetics' of sociology is rather a way of living one's life (Atkinson, 1996). To put it in a different way, it is not always a prompt opening to a series of underlying and obscure social trajectories. It is, in most cases, an intentional surrendering of the self to the unpredictability of an 'unknown and ambiguous world'.
4.4 This world is called upon to become familiar and known and sometimes it is forced to expose itself with all its fragmentations, contradictions and antinomies[7]. A strictly technical and instrumentalised reading of 'bias' runs the danger of analytically distorting the structures and the qualities of the research field. It also runs the danger of overlooking an essential way of understanding the intrinsic characteristics of the 'research object' (Collins, 1998; Scott, 1998).
4.5 This article essentially proposes taking advantage of the positive or negative 'bias' that is expressed by the informants/narrators in relation to the researcher. As it has been demonstrated by the two empirical cases that were analyzed, namely the case of psychiatric reform in Greece and the case of the Leprosarium of Spinalonga, this sort of 'bias'-in other words the structurally produced 'wrong' identification of the researcher's intentions and qualities by the informants/narrators-can constitute a privileged way of examining the structures and the qualities of particular social worlds.
4.6 Especially during the process of entering an ambivalent social world, this 'bias' is more marked since both the mediators and the key persons can direct the research task to a 'biased sampling'. However, this seems to constitute a fruitful way to understand the internal constitution of particular social worlds. The application of strictly positivistic criteria of sampling, especially in qualitatively oriented empirical inquiries, seems to overlook these opportunities of sophisticated understanding and theoretically sensible sociological explanation. [9]
Notes
1 In suggesting that within qualitative work the researcher's 'presence' is desirable or inevitable, we do not imply that the researcher' self is not an energetic or even a constitutive part of quantitative research. For example, conducting a questionnaire or a survey very often presupposes particular theoretical interpretations and methodological assumptions regarding the characteristics of the social world and more particularly the case under examination (i.e. a questionnaire or a survey regarding the attitude of males and females nurses on people suffering from HIV or a statistical based research on immigrant's unemployment rates, etc). These theoretical interpretations and methodological assumptions, which can potentially influence the final research outcome, could be strongly related to the researcher's social background, interests, skills and characteristics (i.e. male or female, European or Asian, working class or middle class background etc). However, it is rather the case that these issues maintain a more crucial part in the qualitatively oriented research for they hold a dimension that can be possibly developed in a methodologically interpretative and reflexive way. This interpretative dimension (reflexivity) of the qualitatively oriented research might contribute to a more elaborated understanding of the case under examination and it is precisely this aspect of empirical inquiry that this article strives to highlight.
2 The call for self-referential methodology that only takes into account as ultimate judges the research community is rather an epistemological fallacy in as much as it bypasses the political and ethical character of every social research project. As feminists, amongst others, have argued the notion of a value-free scientific subject that examines an 'crystal clear object' is rather a way of masquerading unequal distributions of power and preserve particular social hierarchies. The possible solution to this problem is to declare your stand and strive to be as honest and open as possible while maintaining a very explicit methodological approach that specifies its presuppositions and concessions (Evans, 1993, pp. 5-13; Temple, 1997; Collins, 1998; Scott, 1998).
3 Many scholars, who belong to the so-called qualitative social research tradition, point out to the analytical importance of case studies. They argue that the researcher is possible to understand 'bigger' and complicated social realities through a sophisticated analysis of particular cases, which are related to concrete social worlds. Thus, the researcher might generalize to a higher level of abstraction retaining as his/her point of reference empirically appreciable entities. For instance, one might understand what is to be an unemployed immigrant in modern western societies through the analysis of a case study, which would include a specific geographical area or a particular theoretical sample (Wengraf, 2000, pp. 140-164; King, 2000, pp. 305-320; Travers, 2001).
4 The fact that the narrators knew other potential narrators was a significant characteristic of the entry into the research field. In addition, the fact that the informants knew a great number of details of the private life of the rest of the informants aided in comparing and the double-checking the empirical data gathered. In case, as Scott remarks in her article on ritual abuse, the biographical background of the researcher is a crucial factor that can radically influence the whole research endeavor (Scott, 1998). In our case, the researcher's biographical background also maintained another dimension: it assisted the narrators to imaginary communicate with the present of their homeland and to feel that there is still someone that remembers of them and their stories. Besides, an interesting element of our research was the fact the some narrators lied during the interviews in a very sophisticated way. Actually speaking, they did not lie but they overexaggerated issues in such a way that one could think that they might intent to 'deceive'. However, as Rosie remarks, people use fantasy to achieve particular ends and to construct certain versions of their lives and actions (Rosie, 1993, pp. 144-152). In the case of the survivors of the Leprosarium of Spinalonga, the aim was to achieve a 'balanced' life history account that could potentially 'redeem' the biographical disruptures in order to ease social stigma, exclusion and suffering.
5 We also encountered cases where it was impossible to persuade people to narrate unless the researcher proved to them that he was actually from Crete and had some kind of 'real' connection with, what they perceived as the, 'traditional (authentic)' character of a Cretan male. For example, they asked the researcher why he had earrings, since that was a practice that did not precisely match with the majority of the narrators thought to be a 'proper' Cretan man. Through this critical observation, the researcher had the opportunity to detect the narrators' attitudes on morality and religiousness.
6 The researcher legitimated his presence in the research field maintaining the opinion that he would like to write a book regarding the community of lepers of Spinalonga. On the one hand, this occurred because the researcher did not want to be identified with a journalist or a newscaster and, on the other hand, because the participation in writing a book is quite often an attractive proposal that mobilizes narration and participation, particularly when the narrators are relevantly aged (Denzin, 1989; Bytheway, 1993, pp. 153-165; Humphrey, 1993, pp. 16-178; Bytheway, 1993, pp. 153-165; Kleinman, 1994, 1995; Kaufmann, 1996; Bertaux, 1997).
7 At this point, we can possibly detect the potential dangers the use of the snowball technique implies when it is employed without a critical perspective: the snowball's course is determined on the morphology of the ground it finds on its way down. However, this morphology constitutes at the same time a structural version of the examined social field. The starting point of the snowball is of vital importance for it determines what features of the morphology of the ground it will encounter. Similarly, the researcher's mode of entry attributes to him/her a potential label that locates him/her in the internal hierarchy of the social world under investigation and determines how s/he will come to know this world. Thus, the snowball technique can be also understood as a journey in the structure of the field. This is to say that the snowball technique can be also understood as a structure of exclusions and inclusions among the members of the group of reference. The same holds true regarding the key persons that are used as mediators in social places where access is difficult or problematic. The snowball's course always seems to be shaped by the social relations that are formed within the group of reference, the structure of the research field and the data gathered.
Acknowledgements
The authors would like to thank the anonymous referees for their comments and critical remarks.
References
ANGROSINO, M. V. (1989) Documents of Interaction: Biography, Autobiography and Life History in Social Science Perspective.Florida: University of Florida Press.
ATKINSON, P. (1992) Understanding Ethnographic Texts. Thousand Oaks. Sage Publications.
ATKINSON, R. (1998) The Life Story Interview. Thousand Oaks: Sage Publications.
BERTAUX, D. and KOHLI, M. (1984) 'The Life Story Approach: A Continental View', Annual Review of Sociology, Vol. 10, No. 1, pp. 206-217.
BERTAUX, D. (1997) Les Récits de Vie. Paris: Nathan.
BOURDIEU, P. (editor) (1993) La Misère du Monde. Paris: Seuil.
BOURDIEU, P. (1999) Language and Symbolic Power. Cambridge: Cambridge University Press.
BYTHEWAY, B. (1993) 'Ageing and Biography: The Letters of Bernard and Mary Berenson', Sociology, V. 27, No. 1, pp. 153-165.
CARR, E. X. (1999) What is History. London: Routledge.
CHAMBERLAYNE, P., BORNAT, J. & WENGRAF, T. (editors) (2000) The Turn to Biographical Methods in Social Science. Comparative Issues and Examples. London: Routledge.
CHAMPAGNE, P. (1999) Initiation à la Pratique Sociologie. Paris: Dumond.
COFFEY, A. and ATKINSON, P. (1996) Making Sense of Qualitative Data. London: Sage.
COLLINS, P. (1998) 'Negotiating Selves: Reflections on 'Unstructured Interviewing', Sociological Research On Line, V. 3, No. 3, <http://www.socresonline.org.uk/3/3/2.html >.
COMBESSIE, J. C. (1996) La Méthode en Sociologie. Paris: La Découverte.
CORBIN, J. and STRAUSS, A. (1987) 'Accompaniments of Chronic Illness. Changes in Body, Self, Biography and Biographical Time', Research in the Sociology of Health Care, Vol. 6, No. 3, pp. 249-281.
CRESWELL, J. W. (1998) Qualitative Inquiry and Research Design. Choosing Among Five Traditions. Thousand Oaks: Sage Publications.
DENZIN, N. and LINCOLN, Y. (editors) (1994) Handbook of Qualitative Research. Thousand Oaks: Sage, 1994.
DENZIN, N. (1989) Interpretative Biography. London: Sage.
DENZIN, N. (1996) Interpretative Interactionism. London: Sage.
DENZIN, N. (1997) Interpretive Ethnography: Ethnographic Practices for the 21st Century. Thousand Oaks: Sage.
EVANS, E. (1993) 'Reading Lives: How the Personal Might be Social', Sociology, V. 27, No. 1, pp. 5-13.
GADAMER, H. G. (1975) Truth and Method. New York: Routledge.
GLASER, B. G. and STRAUSS, A. (1971) The Discover of Grounded Theory in Practice. Chicago: Aldin.
GUILBERT, J. and JUMEL, G. (1997) Méthodologie des Pratiques de Terrain en Sciences Humaines et Sociales. Paris: Armand Colin.
GOFFMAN, E. (1963) Encounters. Two Studies in the Sociology of Interaction. London: Penguin.
HABERMAS, J. (1987) The Theory of Communicative Action V. 2. A Critique of Functionalist Reason. Cambridge: Polity Press. HAMEL, J. (1993) Case Study Methods. Thousand Oaks: Sage Publications.
HOLLWAY, W. and JEFFERSON, T. (2000) Qualitative Research Differently: Free Association, Narrative and the Interview Method. London: Sage.
HUMPHREY, R. (1993) 'Life Stories and Social Careers: Ageing and Social Life in an Ex-mining Town', Sociology, V. 27, No. 1, pp. 166-178.
ION, J. (1997) La Fin des Militants?. Paris: Les Editions de l' Atelier.
KAUFMANN, J. C. (1996) L' Entretien Compréhensif. Paris: Nathan.
KING, A., (2000) 'Part of the System: The Experience of Home-based Caring in West Germany', in CHAMBERLAYNE, P., BORNAT, J. & WENGRAF, T. (editors), London: Routledge, pp. 305-320.
KLEINMAN, A. (1994) The Illness Narratives. Suffering, Healing and the Human Condition. New York: Basic Books.
KLEINMAN, A. (1995) Writing at the Margin: Discourse Between Anthropology and Medicine. Berkley: University of California Press.
LAPLANTINE, F. (1996) La Description Ethnographique. Paris: Nathan. MILLS, C. W. (1985) The Sociological Imagination. Thousand Oaks: Sage, 1985.
RIESSMAN, C. K. (1993) Narrative Analysis. Newbury Park: Sage.
RICOEUR, P. (1990) The Narrative Function. London: Sage.
ROSIE, A. (1993) 'He's a Liar, I'm Afraid: Truth and Lies in a Narrative Account', Sociology, V. 27, No. 1, pp. 144-152.
SAVVAKIS, M. and TZANAKIS, M. (2002) 'Institutional Social Establishments, Constitution of Social Borders and Processes of Problematisation: The Psychiatric Asylum of Leros and the Leprosarium of Spinalonga, Psychiatry Notebooks, Vol. 79, pp. 62- 77, (in Greek).
SCOTT, S. (1998) 'Here Be Dragons: Researching the Unbelievable, Hearing the Unthinkable. A Feminist Sociologist in Uncharted Territory', Sociological Research on Line, V. 3, No. 3, <http://www.socresonline.org.uk/3/3/1.html >.
SQUIRE, C. (2000) 'Situated Selves, the Coming-out Genre and Equivalent Citizenship in Narratives of HIV' in P. Chamberlayne, J. Bornat, J. and Wengraf, T. (editors) The Turn to Biographical Methods in Social Science. Comparative Issues and Examples. London: Routledge, pp. 196-213.
STANLEY, L. (1993) 'On Auto/Biography in Sociology', Sociology, V. 27, No. 1, pp. 41-52.
STRAUSS, A. (1992) La Trame de la Négociation. Paris: L' Harmattan.
SCHWARTZ, O. (1993) 'L' Empirisme Irréductible', in Anderson, N., Le Hobo. Sociologie du Sans-abri. Paris: Nathan, pp. 265-308.
TEMPLE, B. (1997), ''Collegial Accountability' and Bias: The Solution or the Problem?', Sociological Research On Line, V. 2, No. 4, <http://www.socresonline.org.uk/2/4/8.html >.
TRAVERS, M. (2001) Qualitative Research Through Case Studies. Thousand Oaks: Sage Publications.
TZANAKIS, M. (2003) Transformations of Subjectivity and Everyday Life. A Sociological Approach of the Psychiatric Reform in Greece. PhD Thesis, Department of Sociology, Rethymno: University of Crete, (in Greek).
WENGRAF, T. (2000) 'Uncovering the General from Within the Particular: From Contingencies to Typologies in the Understanding of Cases', in P. Chamberlayne, J. Bornat, J. and Wengraf, T. (editors)London: Routledge, pp. 140-164.
WILKINS, R. (1993) 'Taking it Personally: A Note on Emotion and Autobiography', Sociology, V. 27, No. 1, pp. 93-100.
2.3 While the first perspective of evaluating the field of Greek psychiatry, which is constantly changing, reflects a negative and defensive approach, the other two result in the recognition of the necessity for psychiatric reform. All perspectives, which emerge from the process of psychiatric reform as all-embracing ways of representing the reform enterprise and its organizing principles, offer a predefined place to the 'visitor-sociologist' who enters the everyday routine of the psychiatric institution. Before s/he even asks a question, before s/he even wonders how things are or may be in what s/he conceives as his/her 'area and object of study', a place will be proposed to him/her according to one of these perspectives: The informants will respond either by making a call for coordination of the gazes, displaying a tendency for co- operation or even by expressing a clear dislike.
2.4 Against those who retain the perspective of 'traditional psychiatry', the sociologist will encounter silence and a mood of certain annoyance. At this place, s/he is viewed as the bearer of a discourse that is considered alien and absolutely irrelevant in relation to the 'object' of psychiatry-that is to wit psychopathology. However, apart from being identified as the bearer of a 'non-competent' discourse, the researcher is identified as a potential enemy by those who express 'traditional' psychiatry and is designated as such: s/he, the sociologist, is a strategic ally of their opponents, namely of the representatives of 'critical' psychiatric discourse.
2.5 'Critical' psychiatry, this opposite pole of the 'traditional' discourse, advocates the socialization of both psychiatric 'object' and psychiatric services ('community psychiatry'). In addition, it calls for co-operation with the social sciences in general and sociology in particular. The researcher did not claim that 'role' for him/herself. However, the participants in the research field proposed this 'role' to the sociologist from the first moment s/he entered the field.
2.6 In the 'unfriendly reception' by the representatives of the 'traditional' discourse, we can detect the traces of psychiatric history and the results of the contestations that brought into the field of Greek psychiatry this motley multitude of 'critical-modernizing' discourses advocating a 'social' psychiatry. Thus, what prevails is silence, cautiousness and a refusal to speak, which is at the same time a contestation of the authority of the sociologist and his/her allies: the representatives of the 'critical' and the 'modernizing' psychiatric discourse. In all these polite or symbolically violent refusals to accept the sociologist as a partner in a potential dialogue-which emerged during fieldwork-we would, most of the times, recognize a 'positive' act and a definite step towards de-legitimizing the competence of those who retain a 'traditional psychiatric' perspective with whom they associate the visiting sociologist.
2.7 However, the reverse statement also holds true: a 'friendly reception' becomes evident when the researcher observes the social ceremonies thereby indicating his/her acceptance of the structural transformations the 'socialisation' of psychiatric treatment puts forward. The psychiatrists-fighters welcome even a sociological hint. However, this welcome is not elicited by any supposed technical mastery involved in conducting fieldwork, but arises due the ideological-symbolic conflicts that emerge from the process of reshaping the system of Greek psychiatric services-a project assumed to meet with sociologists' approval.
2.8 What is of great interest in this case is the common articulation of the ideological disputes that emerge from the enterprise of 'socializing' psychiatric care. This effort is strongly connected to the established hierarchies in the field, an issue that implies direct consequences on the 'sampling process' of all sociological fieldwork in the psychiatric institutions. For instance, before attempting a series of taped interviews with the staff of a psychiatric clinic, the researcher's first step was to gain permission from the chief psychiatrist of the department.
2.9 The direct welcome by a 'psychiatrist-fighter', an ally of the 'alternative' social psychiatry, is neither an outcome of the technical ability of the researcher nor of the technical perfection of conducting fieldwork. Thus, the chief psychiatrist, representative of the 'critical' psychiatric discourse, immediately agreed to assist the researcher's task and personally intermediated so that the staff of the clinic would contribute to the research effort.
2.10 However, the visit to the clinic the same day met the refusals of the nurses to provide interviews, especially when they detected the tape recorder. This was not, however, a 'blind refusal': for more than half an hour a male nurse analytically explained to the researcher the reasons why he could not go into details on certain issues, especially if these opinions were to be tape-recorded. Essentially, this male nurse offered an interview in order to make explicit the reasons that did not permit him to give the interview, laying emphasis on his non-competence. This research 'failure' proved to be quite elucidating regarding the issue of 'who has the authority to speak in public'.
2.11 On the other hand, the direct welcome that the researcher received was not so much due to a properly pre-planned methodology, but to the very structure of the research field: the sociologist, what he symbolized and also what he did, was a potential ally of an advocate of psychiatric reform. For this particular psychiatrist, the presence of the sociologist in the examined psychiatric institution practically contributed to the goal of psychiatric reform.
2.12 Based on the previous remarks, the political role of the sociologist becomes prominent. The researcher fundamentally aims to constitute an 'object' and a perspective of reading and analyzing this 'object' (Savvakis & Tzanakis, 2002, pp. 62-77). This perspective, which is gradually constructed during the process of intermediation between theory and research practice is 'objectively' recorded in the 'legitimate vision' (Bourdieu, 1993) of the examined social world.
2.13 The physical presence of the researcher as such verifies, one way or another, the structure of the research field. The recognition of the political significance and the dynamics of the ethnographic research finally result in all these practices of treating the 'research presence'. From this recognition comes the alternation of the role of the researcher as 'intruder' to that of the 'friend', the 'scientist', the 'ally', the 'stranger', the 'outsider' and the 'enemy'.
2.14 Especially, in the case of an organized institution such as a psychiatric asylum, the constitution of the 'object' is 'objectively' recorded in the process of managing the 'public image'. The public image is neither something absolutely homogenous nor the result of a 'discursively produced consensus' (Habermas, 1987). The handling of the public image of the institution as such seems to inscribe a circle. Within this circle, multiple, diverse and contradictory attempts at interpretations arise as potential alternative and competitive definitions of the field.
2.15 The competitive interpretations and definitions of the social world arise from:
