One of the consequences of the growth of academic interest in Reception has been an increased focus on the thoughts and opinions of theatre practitioners. However, because of the ephemeral nature of theatre production and the general sparseness of production documentation, traditional methodological approaches do not generate the evidence needed for academic analysis. To fill this gap, scholars can look to the accounts of practitioners' intentions in published interviews. These interviews can take the form of either journalistic or academic interviews. In a previous discussion [Burke 2003] it has been seen that journalistic interviews are a useful but problematic source of data collection. Journalistic interviews are useful because they provide access to a restricted world, but problematic because they are circumscribed by a need for mutual promotion that arises as a consequence of the symbiotic relationship in operation between publications and productions, and interviewers and practitioners. As part of the analysis of the importance of interviews, attention now turns to the academic interview. Initially it may seem that the academic interview answers the shortfalls of the journalistic interview. On the surface it could seem as though the academic interview is the more thorough form of interview: freed from the need to promote a publication or production, the academic interview could appear to be a more scholarly, de-personalised and probing inquiry into a practitioner's intentions/motivations. Whether the researcher in Classical performance uses published interviews conducted by other academics or embarks on a field study using the academic interview model, it is important to note that the academic interview is similarly subject to a collection of limitations that need to be factored into the analysis of interview data. In the main, this article explores the strengths and weaknesses of using published academic interviews, but it is also hoped that in studying the form of the academic interview this article will also be of value for academics interested in conducting interviews for research purposes.
Positively, as a consequence of the knowledgeable position of the academic interviewer and theatre interviewee, academic interviews provide an informed discourse about a practitioner's involvement with a production. Indeed, the key strength of academic interviews is that it they are conducted from a position of specialised knowledge, which would be too in-depth for a journalistic interview, which is aimed at a general media audience. Negatively, however, the academic interview is subject to limitations that are both inherent in the methodological process of interviewing and particular to the dynamic that exists between the academic interviewer and the 'Elite' interviewee. In the course of this article, we shall argue that for 'second-hand' data to be used correctly, it needs to be analysed with reference to the respective intentions of the interviewer and interviewee. Moreover, what will also be seen, is that although publication is by means of a written text, interviews are not in fact written texts. Rather, the published piece is an interpretation of a discussion, an interpretation that cannot convey the dynamic of a conversation (including tone, register, accompanying gesture and so forth). Many idiosyncratic factors impinge on the reliability of the interview and these factors must be factored into the analysis of the data. These limitations acknowledged, however, in this article we shall argue that interviews can be a useful tool and productive source of enquiry. They cannot replace or supersede other lines of enquiry, but, as part of an embracing methodology, academic interviews are an essential part of production analysis.
The following discussion divides into four main sections (II-V). In Section II we explore bibliographic issues in order to appreciate the evolution of the interview form and the distinction between academic interviews, which are qualitative research interviews, and other interview models, for example standardised questionnaires. In Section III we focus on the participants. An understanding of the roles and interaction of interviewer and interviewee demonstrates that the status of 'Elite' is both advantageous and problematic. Following on from the advantages/disadvantages of the 'Elite' model, Section IV looks at further limitations generated by the interview process and establishes factors that need to be taken in to account when using information taken from a published interview. In Section V, examples of published academic interviews will be analysed. For ease of comparison two interviews conducted by established academics (Marianne McDonald and John Haffenden) with the poet/playwright Tony Harrison are compared and contrasted with reference to interview context, language and the aims of the interviewer and interviewee.
At the very least, a surface understanding of the dynamics of the interview as a form of social interaction is needed so as to appreciate the form of the academic interview. However, it can be laborious to achieve this understanding simply because the bibliography on interviewing has traditionally gone hand-in-hand with the subject discipline for which the interview is being conducted. Moreover, although the thoughts and intentions of practitioners are an essential part of performance research, we have been unable to locate any bibliographic literature that specifically analyses the form and role of the interview in theatre research. Consequently, in contrast with other disciplines, no methodological framework for theatre interviewing has yet been forged by decades of debate and counter debate. Without a working methodology for interviewing theatre practitioners, it is necessary to look to the methodological debate in other disciplines to understand the complexity of the interview process.
On the whole, however, the relevant literature tends to offer advice about the methodology of interviewing as part of a justification for a particular research interest, which often has a therapeutic element. Social science literature, for example, does not really engage with issues entailed in using published interviews conducted by other researchers. The researcher in theatre, however, might be interested in the methodology of interviews that he/she conducts as part of his/her research (which requires the interviewee to be alive, able and amenable to being interviewed), as well as published interviews that have been conducted by another researcher (usually an established academic). The published interview can serve several functions. One would be that if the target respondent has been interviewed before and is for some reason unable or unwilling to be interviewed again, then previously published interviews can provide evidence for his/her motivations/intentions and so on. Another would be that if the interviewee is able/amenable to being interviewed, then the previously published interview can serve as a source for the questions asked in the interview schema, as well as providing another account with which the results of the new interview can be compared and contrasted. But the lack of theoretical advice and discussion from interviewing specialists about the use of the information gained from a published interview presents a challenge to understanding the theoretical complexities of using 'second-hand information'. A theoretical basis to the use of information is, nonetheless, a vital element of data gathering, and so a methodological framework needs to be extrapolated from the theoretical approach to subject-specific pragmatics advocated by sociological literature.
This essay cannot do full justice to the decades of scholarship concerning the interview debate that has raged in sociology, anthropology, psychoanalysis etc. It would be patronising to such disciplines to suggest that the complex issues that social scientists and social anthropologists have discussed throughout the twentieth century can be faithfully and cogently communicated by non-specialists. Indeed, according to Kvale [1996, p.9] until recently there has been a lack of cross-fertilization between disciplines so that 'isolated researchers have invented small qualitative wheels over and over again.' The diverse literature generated by different disciplines has resulted in a seemingly never-ending flow of literature, yet by virtue of being written in isolation it is tailored to the respective area of enquiry. A theatre researcher is potentially lost in this maze of at once relevant and irrelevant literature. However, the theatre researcher aiming at developing an interview methodology derived from the theoretical premises and practical experience of other disciplines is greatly helped by the publication of Nigel Fielding's  compendium in four volumes of salient articles. Fielding's collection gathers together previously published seminal articles, which discuss areas ranging from the historical contextualisation of interview scholarship; identification of distinct interview types and concomitant methods; explanation of the qualitative/quantitative dialectic; and interview ethics (interviewer behaviour, transcription, confidentiality, and the impact of gender, status and sexuality on the interview process). Also of central bibliographic importance is the collection of specially commissioned essays by Jaber F. Gubrium and James A Holstein , which provides fresh discussions on the same areas as are covered by Fielding's compendium. In addition to these collections, key specialist monographs alert the researcher to the important issues current in the interview debate. A particularly accessible monograph by Bill Gillham  explores research interviewing in straightforward language and provides clarity on the major issues of interviewing. A philosophical exploration of qualitative research interviewing is offered by Steiner Kvale , who explores more complex issues of interviewing in the light of post-modernist, hermeneutic and dialectical positions. Kvale also discusses Hamlet's 'interview' with Polonius (Hamlet Act III Scene 2) and Plato's Symposium to forge a link between interviews and the humanities. A detailed understanding of the complexity of interview issues is available through specialist monographs that examine a particular type of interview, for example Robert K. Merton, Margorie Fiske and Patricia L. Kendall's [1990; (1956)] examination of focused interview methods investigates issues of range, specificity and depth, and Lewis Anthony Dexter  relates those same issues to interviewing 'Elite' respondents.
Interview scholarship has traditionally tended to focus and advise on the respective role, status and interaction between the interviewer and interviewee, and the composition of the interview schema, with particular reference to the nature of the interview questions. Since the 1940s debate has on the whole focused on the extent to which the interview is structured (that is composed of generic questions that are asked of all respondents) or non-structured (that is composed of open questions tailored to the individual respondent). An overview of the evolution of scholarship examining issues of standardisation is compiled by Beatty [2003, passim; (1996)] and the perceived strengths and criticisms of standardised and non-standardised interviews are detailed by Pawson [2003, p.154; (1996)]. Since the 1940s qualitative as opposed to quantitative interviews have been generally considered to require a method that privileged the interviewee and suppressed the role of the interviewer. Thus, for example, Merton and Kendall [2003; (1946)] devised a methodology for their 'Focused interview' that privileged non-direction. Whilst he acknowledges the prior analysis of the interviewer and his/her role in the composition of the interview schema, Merton nevertheless argues for non-direction in the questions asked in order to maximise the importance of the interviewee's thoughts and judgements. Similarly, Gordon's [2003; (1956)] methodology of the 'Depth' interview insists on non-direction on the part of the interviewer. Essentially, these types of interviews aim to reduce the influence and control the role of the interviewer; indeed Dohrenwend and Richarson [2003 p.334; (1956)] for example consider that the use of a pre-planned interview schema will inevitably lead to the interviewer controlling the interview. In modern times, however, the role of the interviewer is no longer seen as a role that can potentially be expunged; rather the act of conversation is seen to involve both participants in the shaping of interview data. With reference to quantitative research methods, this has led Beatty to re-examine the structured/unstructured debate; and with reference to qualitative research methods, modern scholars stress the relationship between the research agenda for which the interview is being conducted and the role of the interviewer. As will be discussed in more detail in Section III, with particular regard to 'Elite' interviewing, the research aims of the interview will be fundamental to the direction of the interview.
The bibliographic literature of other disciplines provides a crucial grounding in the study of interview methods, which can then be transferred to the study of theatre interviewing. It is essential to understand the dynamic of interviewing through an appreciation of theoretical issues when one seeks to analyse a published interview or plan a research study that will be based on interviews. The literature on interviews, although often tailored for therapeutic case studies, does alert the theatre researcher to issues of the control of meaning, circumscribing relevance and determining range and specificity. The roles and agendas of interviewers and interviewees constitute an essential part of understanding the factors that impinge on the published academic interview.
In an academic interview, the theatre practitioner is rightly considered an 'Elite' respondent. Accordingly, the interview schema and the behaviour of the interviewer will be tailored to this 'Elite' status. An 'Elite' can be defined as a high-ranking figure characterised by his/her position. Odendahl and Shaw, for example, define 'Elites' in the following terms:
Elites generally have more knowledge, money, and status and assume a higher position than others in the population. The privileges and responsibilities of elites are often not tangible or transparent, making their world difficult to penetrate.
However, the juxtaposition of 'knowledge, money, and status' in this definition could encourage a limited understanding of 'Elites' derived from a wealth and power based superiority. The danger of characterising 'Elites' through 'hierarchical superiority' led Dexter to define 'Elite' interviewing, not so much by the status of the interviewee, but through the purpose of the interview. For Dexter an 'Elite' interview is one in which the interviewer is looking for instruction, thus the interview is framed with reference to the interviewee's knowledge which the interviewer is trying to access. This relationship encouraged Dexter to posit a three-point definition of 'Elite' interviewing. Firstly, he stresses that the interviewee's definition of the situation is paramount. Secondly, it is the interviewee's structuring of his/her account of the situation under investigation that is paramount. And, thirdly, the interviewee rather than the interviewer determines issues of relevance. Dexter's use of methodology to define the 'Elite interview is slightly problematic. As he notes, in the second and third parts of the definition, he incorporates, to a large extent, ideas from Merton's focused interview. Freedom from a methodologically determined definition of 'Elite', however, is suggested by Odendahl and Shaw , who argue:
The designation of who or what is elite varies according to the area of inquiry, and part of the investigative process is determining the scope of the enquiry. … Those studying elites have developed many different typologies.
With respect to academic interviews with theatre practitioners, 'Elite' is feasibly defined through knowledge. Method aside, Dexter, is right to establish that communication of knowledge is the key point to characterising the 'Elite' interview. What really defines an 'Elite' interviewee is the information that the respondent can provide. An 'Elite' respondent can communicate information that is not available from any other source, from the vantage of his/her personal involvement in the source material. According to this definition, theatre practitioners are certainly 'Elite' respondents, since their experience alone can provide 'authoritative', and hitherto not communicated, information about his/her role in the subject under study or 'inside' understanding of events.
The position of knowledge that the 'Elite' interviewee occupies necessitates a particular interview methodology: the bedrock of 'Elite' interviewing is intense prior research and freedom in the interview schema (to whatever degree), which is composed in appropriate terminology.
An interview with an 'Elite' respondent clearly requires advance research into his/her career. Even so, the prominent writing about 'Elite' interviewing stills sees a need to stress that advance research is a key factor. Zuckerman, for example, as preparation for her interviews with laureate scientists, writes:
Intensive preparation facilitated the process of interviewing in two principal ways. First, it gave evidence of the seriousness of the interviewer and helped to legitimize expenditure of time on the interview. This was an adaptive function designed to provide a firm basis for the temporary laureate-interviewer relation….Second, questions based on materials gathered in preparation often called forth responses that would otherwise not have been elicited particularly if an entirely standardized interview guide had been employed.
In published academic interviews the level of research undertaken in advance is significant, usually in terms of the life history of the interviewee and critical interpretations of their creative ventures. Although in some respects obvious (courtesy and sensible preparation), this advance research does have an interrogative function: simply, the responses of the interviewee can be debated during the interviews if the interviewer is fully versed in the issues [Dexter: 1970, p.14]. One of the major strengths of the published academic interview is the academic's level of knowledge. The specialist academic is uniquely qualified to undertake an interrogative role, and to the 'second-hand' researcher, who may not have the years of academic experience, the established academic's knowledge (exhibited through questions and discussion) can open new lines of enquiry or reach a deeper level of analysis. The reliance of prior critical evaluation, however, should also sound a warning bell for the position of interview: the interview is at the end of a long road of enquiry and, however 'authoritative' the transcript appears to be, its very existence is dependent on the prior research. The interview is rightly placed in a wider methodology of traditional lines of academic enquiry and performance analysis. The key point to remember is that the interview does not override issues raised by alternative sources, but provides another perspective.
The use of prior research to determine the range of the discussion also raises the issue of the range and specificity of the interview schema. 'Elite' interviewing replicates the same debate as in other interviewing literature, that is, to what extent does the interviewer control the range of schema through determining the areas to be discussed, or allow the interviewee to direct the range of the interview. In effect, this is the key issue. Major elements of this debate are issues of whether the interview schema should be composed of structured (variously called standardised/generic/direct) questions or unstructured (not standardised/specific/non-direct/open ended) questions. Dexter, for example, whilst noting the possible constraints of research aims, strongly argues in favour of an unstructured schema that uses entirely unstructured questions. Dexter's preference is derived from his concern with limiting the role of the interviewer and privileging the interviewee. In the sparse scholarship on 'Elite' interviewing, since Dexter there has been a general move towards a semi-structured schema, which uses a variety of structured and unstructured questions. This change possibly arises as a consequence of the changing perceptions of the role of the interviewer and the various disciplines that now use interview research.  The fact that interviewers always participate in the construction of meaning, regardless of any attempts to limit their role (for example, by standardised questionnaires), has created an environment wherein the interviewer has an acknowledged role that can be factored into interview analysis. Kvale, for example, whilst still advocating an interviewee-centred approach, accepts that the direction of the interview ought to respond to the aims of the research project for which the interview is being conducted. Kvale advises those about to undertake research based on interviews in the following way:
Think about how the interviews are to be analyzed before they are conducted. The method of analysis decided on – or at least considered - will then direct the preparation of the interview guide, the interview process, and the transcription of the interviews. Every stage in an interview project involves decisions that offer both possibilities and constraints in later stages of the project.
Furthermore, Odendahl and Shaw [2003, p.310], and Ostrander [2003, p.399; (1993)] both note that the area under investigation circumscribes their respective interviews with 'Elites'. What this means is that the interview schema corresponds to previously determined research aims: although the answers to specific questions might not be known or expected, the general areas of enquiry can be quite specific and can be pre-planned. In this way, research interviewers aim at 'covering their bases': specific questions allow for the areas of research interest to be addressed, whilst the less tightly structured elements allow 'free reign' to the interviewee to explore the areas that are of his/her concern. In this way the formal aspect of interview question and answer is maintained, but similarly so is the conversational transaction between participants; the research agenda is met and potential new unanticipated lines of enquiry are opened up.
The range of the interview schema (derived from the interviewer's research aims) in effect determines the scope of the interview and influences the range and specificity of responses that an interviewee will give. But the causal relationship between research aims, schema and responses can be problematic if the interview is to be used by a researcher who possesses different research aims from the original research agenda that underpinned the schema. By virtue of publication the academic interview establishes itself as a usable source to the readership of the publication in which it appears; even so, using an academic interview can be problematic. For example, if an interviewer was not interested in the political interpretation of a practitioner and did not allow for the political dimension to be raised through the schema, then the lack of discussion could suggest that the political dimension was not a prime-motivating factor for the practitioner being interviewed, which might not necessarily be the case. Thus, the schema can close out potential meanings and interpretation. This is not a fault of the interviewer, who has rightly tailored the interview to the purpose of his/her research interest, but it does mean that the information is filtered through a third person's research priorities. When using information gained from an interview conducted by a third party, however prestigious that third party may be, questions about the way that the information has been generated need to be asked. Thus, the research context and focus of the interview need to be factored into the analysis of any information extrapolated from the interview context.
The 'Elite' status of the interviewer and interviewee is also marked in the phraseology of the interview questions. Echoing the wider structured/unstructured debate in sociology literature, as part of the methodology for interviewing 'Elites', Dexter  and Zuckerman [2003; (1972)] both advocate using unstructured questions. The way that these questions are phrased, however, marks an important difference between the standard qualitative research interview and an 'Elite' interview. On the whole, the advice to a research interviewer is to use language easily understood and divested of academic terminology. In an academic interview, however, the terminology of the interview will correspond to the 'Elite' status of the participants. In published academic interviewsboth the interviewer and the interviewee are designated as 'Elite' by virtue of their knowledge, and their qualification to take part in the interview is constantly reaffirmed through their informed discourse, which is further stressed by their lexical choices. Moreover, the act of publication 'legitimises' their role as informed interlocutors; thus, it is in the interests of both participants to reaffirm their unique qualification through thought and language. The shared frame of reference between the established academic and theatre practitioner allows the interview to be conducted with due reference to specialist theatrical terminology, which in turn maximises the specificity of the interview. But there is always a danger that a closed hermeneutic world is created that excludes those who are not privy to the implied meanings in a coded language. The need to display intellectual credentials, most importantly by the interviewer, can lead a competitive use of language that evidences a struggle to 'control' the interview. Consequently, using data from a published academic interview could require a hermeneutic analysis to appreciate the motivation behind the text.
The information generated by published academic interviews is subject to a wide variety of further limitations that need to be understood and taken into account by the reader. A need for mutual promotion encroaches on the reliability of information generated by journalistic interviews. But the academic interview, whilst seeming more authoritative because of issues of range, language, presentation of the interviewer/interviewee, is still potentially limited: issues of transcription, interview context and idiosyncratic factors all impinge on the interview.
The give and take of a published academic interview might create the illusion that the transcript is a written account of a conversation; therefore, by reading the published transcript the reader is allowed to eavesdrop on the interview. However, the published written text is an imperfect version/interpretation of the actual interview. What appears on the published page is not a verbatim account of what took place in actuality. The published academic interview undergoes an editorial process that allows the interview to become a read text. In conversation, interviewees will hesitate, repeat themselves, not finish sentences, and so forth. The published interview, however, revises the verbatim transcript so that the text appears as a continuous text. Kvale, for example, appreciates the complexities of moving from tape to page:
Transcripts are not copies or representations of some original reality, they are interpretative constructions that are useful tools for given purposes. Transcripts are decontextualised conversations, they are abstractions … The interview is an evolving conversation between two people. The transcriptions are frozen in time and abstracted from their base in a social interaction. The lived face-to-face conversation becomes fixated into transcripts. A transcript is a transgression, a transformation of one narrative mode – oral discourse – into another narrative mode – written discourse. To transscribe means to transform, to change from one form to another.
One of the serious drawbacks of using published interviews is that a reader is unable to enter the 'spirit' of the interview and the interview context. There may be hesitations, repetitions, digressions, translator error and so on in the transcription of words. Due attention to the impact of punctuation on meaning might also be problematic. Hence the reader is not necessarily able to capture the context into which the words are delivered (physical and linguistic), nor the emotion with which a word or sentence is invested. The problem is not resolved by having the interviewer subsequently act in the role of third party narrator, describing the interviewee (as can happen in transcriptions of psychiatry interviews for example) because the evaluation of the interviewee's behaviour, tone, intentions etc. is always subjective to the interviewer. The logic of this would require the interviewer to document his/her own physical and verbal behaviour as part of the interactive creation of meaning. Moreover, the interviewer may not be in possession of all relevant factors embedded in an interviewee's behaviour. Thus, the transcript is akin to a script without stage directions, design, direction and actor's interpretation, requiring the reader to be aware that he/she has not heard the nuances that determine or shape meaning. Caution needs to be exercised when utilising information gained from published interviews.
The temporal relationship between the published transcript, the interview, and the areas discussed in the interview is also a potentially limiting factor. If a published interview does not indicate the date on which it was conducted, it is difficult to know what time gap exists between the areas under investigation and the source material. Time is also important because of the transitory and ephemeral nature of theatrical event. If the interview is conducted some time after the interviewee's involvement with the creative project, this time gap can impinge on the reliability of the information in several ways. Initially memory might be compromised: a practitioner might simply have forgotten his/her experiences, motivation, intentions and so on. Detail of events in sequence might have become hazy and subsequent events might create a romantic or distorted filter through which the past is remembered. Furthermore, continuing theatrical experience can also encourage a standardising (or theorising) of motivation; thus, the past can be interpreted through the interviewee's present motivations. Accordingly, in their study of truth and interviews, Dean and Whyte argue:
The difficulties in interpreting informants' reports of subjective data are seriously increased when the informant is reporting not his present feelings or attitudes but those he recollects from the past. This is because of the widespread tendency we all have to modify a recollection of past feelings in a selective way that fits them more comfortably into our current point of view.
The limitations imposed by time on memory, should not, however, be allowed entirely to dismiss interviews that are conducted at a distance from the event. The limitations imposed by time can be factored into analysis and the potential weaknesses noted and accommodated. Moreover, depending on the research aim of the interviewer (or the researcher reading the published transcript), the way that a practitioner remembers a past theatrical event and interprets it can be instructive. The selection of the memory process could also be used as evidence for what the practitioner held to be important, and, as has been seen, what is important to the interviewee is a key factor in qualitative research. Furthermore, the influence of the past on the present can help the reader appreciate the importance of the past event in the way that it shapes and influences future creative projects. The influence of the present on the memory of the past is also instructive, as it allows the interviewer/reader to understand the evolution of the stance (or emerging style) of a practitioner; seeing the creative event in the light of a wider creative process. Again, as with the direction of the interview schema, the key point is research aim: a diachronic study will look for relationships and interconnections, whereas a synchronic analysis will need to appreciate the potential limitations imposed by time, whilst not allowing time to overshadow the value of the interview data.
In addition to issues of transcription and context, there is a host of idiosyncratic reasons, which might not be immediately obvious to the reader, why the information gained from an interview can be questioned. The mood of the interviewee, his/her response to the interview process, the interaction between interviewer/interviewee, the experiences of the day, can all lead to presentation of events that are only temporally true, but are not factored into the interview process and cannot be accounted for in the subsequently published transcript. Although it may be countered that personal idiosyncratic factors are not as much of an issue in 'Elite' interviews, circumscribed as they are by inquiry into a specific creative venture, factors related to the creative product might still be important. For example, the relationship between the practitioner/interviewee and the performance/production can fundamentally impinge on the nature of the interview data. Hypothetically, if an actor is the interviewee and has had a negative (or positive) experience of working with a particular director/designer/choreographer/fellow cast member[s], then that experience can shape his/her responses. Those experiences are subjectively true (therefore providing a perspective on the production), but are not necessarily abidingly true of the experience of others. The theatre grapevine echoes with stories of infamous fallings out (or conversely especially productive relationships), and these emotions can further determine the perspective through which the data is filtered.
There are two potential responses to these limitations: wider research and an understanding of the value/limitation of subjectivity. With respect to wider research, information gleaned from a published transcript needs to be set in context by being assessed in the light of another published accounts, for example, journalistic interviews, theatre programmes, specialist monographs and so on. The key point is that the academic interview is only one mode of inquiry, which is no more authoritative than other modes. Simply, the comparison of different sources, rather than the privileging of seemingly 'straight from the horse's mouth' data, is necessary for a balanced account; a richer understanding will be achieved by appreciating points of similitude or difference. Setting the interview data in a wider research context, however, should not be taken to mean that the perspective of the interview is only validated when equally corroborative subjective data is forthcoming from another source. According to Kvale, 'intersubjective reliability' could lead to:
tyranny by the lowest possible denominator: that an interpretation is only reliable when it can be followed by everyone, a criterion that could lead to trivialization of the interpretations. This may again involve a consensualist conception of truth: that an observation or an interpretation is only considered valid if it can be repeated by everyone, irrespective of the quality of the observation and the argumentation.
The focus on the interview as a shared construction of meaning in which each participant brings to the interview ideas shaped by their experiences/agendas/intentions is now part of the bedrock of qualitative interviewing which uses an interactionalist method. With respect to the role of the interviewee, this focus on the role of the individual foregrounds the individual's lived experience as central to appreciating the data that qualitative research interviews generate [Kvale, 1996, passim]. The appreciation and application of this premise to theatre research enables the very subjectivity of interviewee's responses and interpretation to become as instructive as a less subjective account. Thus, using the example of a disgruntled actor, the question becomes why does he/she possess such feelings, and what can those feelings tell the researcher about the theatrical event which may not have been evident from witnessed-based performance analysis.
Even so, the comparison between McDonald's interview and that by Haffenden does point towards a literary theoretical implication: that such interview material does have its uses for the performance researcher, but in ways that are circumscribed by the form that is being used. Even as the elite interviewer is trying to elicit important information, his or her own context necessarily impinges on the interview situation. As a form of recorded literary discussion, such an interview needs to be carefully unpacked.
Clearly, the use of academic interviews by the researcher has both benefits and limitations. How the researcher makes practical use of them depends on the research project at hand, but some useful general points can be made. Four areas of enquiry are open to the reader of academic interviews: the context of the interview; the language it uses; the research agenda of the interviewer; and the interaction between the interview schema and the conversation that takes place. In addition to the contextual implications noted previously, the researcher who is using the interview as a primary source needs to work out how the interview being read relates to his or her own research agenda. The language used during the conversation will provide clues to the schema and the interviewer's own agenda, particularly when the respondent begins to go beyond the interviewer's questions.
These general points form the immediate context for our own analysis of the two academic interviews we have chosen. Within this context, we move between two categories, each of which has further implications. On a theoretical level, there is the major issue of authority over meaning, and this in itself breaks down into three parts:
1. the authority of the interviewee
2. that of the interviewer
3. and the relationship between both.
The ways in which authority is interleaved through the interview process is of great importance, because it impinges so much on the crucial terrain over which meanings play out. Our second level is that of the pragmatic purpose of the interview, the purpose it serves.
The two levels we have identified exist in other forms of interviews. Some analysis of journalistic interviews has already been undertaken in Burke :
The aims of this paper are: to set journalistic interviews within a wider methodology; to consider the value of interview evidence and to evaluate the strengths and weaknesses of the interview process through analysis of the relationship of intent (mutual promotion) that characterises the interviewer/ interviewee relationship.
The focus here upon the interview relationship in a journalistic context will be of value for our second set of issues, associated with the pragmatic purposes of the interview. Moreover, the assumptions that underpin this kind of work also have ramifications for the first, or theoretical, level of enquiry:
The value of journalistic interviews is that they can provide a medium by which practitioners can explain their intentions in the performance. The unique feature of interviews (academic and journalistic) is that they provide the opportunity for theatre practitioners to elucidate the creative process.
The linking of the two kinds of interviews via the realm of the production of meaning lays bare a fundamental theoretical problem: should an interview with a cultural practitioner be considered a full statement of meaning? Perhaps the viewpoint even of a writer is inevitably always partial, especially in relation to a dynamic art such as live theatre, in which meanings generated within the same audience can vary so widely. An extreme position could be that the practitioners' views are irrelevant, since there is no guarantee that their intentions are fulfilled in practice. Additionally, these issues are further undercut by the status of the interview itself as a form of publication, which leads back again to pragmatics:
Journalistic interviews, however, are not a neutral method of data collection, as the media interview process is circumscribed by a relationship of intent: a symbiotic relationship exists between the media and the theatre that is predicated on the basis of mutual promotion. 
The production of publicity material lies at the heart of the whole process and leads to a startling clarity of function as well as a clear stratification of roles:
The promotion of the production is achieved through the promotion of the practitioner; because the act of being interviewed assumes superior knowledge on behalf of the practitioner, the respondent is defined as 'Elite' and is, therefore, invested with authority. 
However, a distinction between journalism and academia arises here, because the academic interview operates according to a different set of priorities. The academic is also perceived to be of an 'Elite', with an insider's access to the privileged realms in which meanings are produced and articulated. It would therefore seem to be appropriate to note at this stage that academic interviews need to be analysed very carefully because of the range of interests they serve.
Interestingly, the title of the first of our two chosen examples is Interview With Tony Harrison by John Haffenden , which was published in an academic monograph of collected articles about the poet Tony Harrison. It is not an interview of Tony Harrison by John Haffenden. The connective implies a more equal playing field between two protagonists and indeed the very beginning of the published text reinforces the academic's 'Elite' status in the sense we have outlined:
JH. You are on record as having said elsewhere that 'every poem is a momentary defeat of pessimism', which suggests that you share Robert Frost's sense of a poem being 'a momentary stay against confusion', except for the fact that 'pessimism' is the operative word for you.
The language here may of course have been edited for publication, but it hardly reads like a semi-conversational interview situation. This is appropriate for inclusion in a collection of scholarly articles, particularly given the intertextual reference to Harrison's paraphrasing of Frost. Rhetorically, as the very beginning of the published interview, the paragraph serves the purpose of laying bare not only the parameters of the interview, but also the level of knowledge that has been attained by the academic interviewer. Its purpose is as much to demonstrate the interviewer's own 'Elite' status as it is to elicit a response.
Slightly later in the interview, the academic demonstrates more knowledge: 'You also take on a Marvellian air of lyric grace' [1991, p.228]. Statements such as these reinforce the status of the interviewer in a way that marks out a difference between academic and journalistic interviews. The published product has a very clear sense of relation to the level of pragmatics with which we are concerned. This could be interpreted as a positive element; more negatively, however, is its relationship to the theoretical level of enquiry. There are two reasons why it should be so. The first is the elision of historical difference: Harrison is complimented as being Marvellian, as though two radically different poets separated by 350 years can simply be compared in this way. The implicit underlying assumption is one of unbroken continuity, a Great Tradition, a common enough position and one that does not necessarily detract from the interview in its published form. The second, however, is more fundamental to the very nature of the academic interview, and follows on logically from the issue of relative 'Elite' status. The problem here is one of elitism: such language and such a display of knowledge can in fact close off interpretation from a wider audience. Even other academics will not necessarily have enough elite knowledge to be able to follow what is going on here. The closed circle that is created points to ways in which different kinds of academic interview may be posited on different conceptions of what academics do and how accessible their work should be. In other words the academic interview could tend towards being the opposite of the journalistic one in the crucial level of theoretical enquiry. A closed hermeneutic circle is being constructed, one that is accessible only to those who have enough specialised knowledge to be able to decode the meanings that are being brought into play.
There is therefore a potential contradiction between the two levels of who defines the meaning and the purpose of the interview. Perhaps it should be viewed as a productive tension. After all, this is an interview intended for publication in an academic collection. As it progresses, interestingly, a less elevated tone begins to develop as Harrison starts to relate his poetry to his working-class background [1991, p.234]. It becomes apparent in spite of the question of elitism in the context of the interview that political relevance is crucial for this poet, and the interview moves away from the kind of agenda set up by the interviewer. Haffenden in fact misquotes a line, and Harrison picks up on this to make an important point about the personal consequences of the relationship between education and politics:
I don't feel happy in the world of 'literature', and nor do I feel happy – with my education and my identity as a poet – in my old working-class background: I'm in a way alienated from both, and I have to do justice to that alienation in the poem.
The poet's awareness of what is individually important to him leads on to an engagement with gender, especially in the Oresteia:
TH. Yes, whereas you feel the Indian culture outside contains the image of the destroying female as well as the productive female. It's also something that I tried to bring back into the Oresteia – the idea that the creator and the destroyer are the same person, which is a more ancient idea than the godhead as patriarchal […] One of the points of producing the Oresteia is to clarify the fact that in the end it's a male statement: it's written by men and for men. 
During the course of the interview, it becomes clear that Harrison sees his poetry as a personal method of engaging with political issues. The different interests of the two protagonists create a tension that produces multiple meanings, some of which may not have been intended by either. The ways in which they play off against each other makes this particular interview an interesting one to study. It raises fundamental issues concerning the status of the academic interview itself. What it is for, what purpose it serves: these are, in the end, not entirely clear. To return to the title of the published piece: no purpose is stated. It is simply an interview with Tony Harrison by the academic John Haffenden, as though that in itself gives it sufficient rationale. The implication is that this rationale is sufficient as the interview stands, and that it does have enough value to be included in a collection of essays. More work needs to be done on the underlying issue of what kinds of interviews (by and with whom) are included in 'academic' volumes. Two possibilities might emerge: that the interviews are considered a form of primary source; and that they add celebrity status to the academic voice because of the 'star' quality and influence of the practitioners.
Our second interview, again with Tony Harrison, comes in the form of a chapter in Marianne McDonald's book Ancient Sun, Modern Light: Greek Drama on the Modern Stage . The relationship between theory and practice in McDonald's aims and methods is set out in a Prologue in which she discusses her primary aim of investigating what it is that makes Greek tragedy exciting in the modern world and her belief in the importance of the role of creative writers and directors in reinterpreting the plays for modern audiences. The context is different from the Haffenden interview in that the whole book is written by one person, rather than being a collection of works by various authors. Although the interview (Chapter VIII) is simply entitled Tony Harrison's Interview, it nevertheless carries a certain sense of place as it fits into the overall project of the book, which draws heavily on interviews and talks by practitioners. The context provides the reason for the interview's inclusion, which helps to keep the interview focused, in this case on Harrison's re-interpretations of Greek drama for modern performance. Additionally, the title implicitly gives ownership of the interview to Harrison. This text takes a different direction from Haffenden's, since the underlying strategy is not so much one in which the academic 'Elite' credentials need to be established. Even the style adopted as the piece opens is radically different from Haffenden's: the first paragraph sets the scene, and the next few sentences are represented as conversational short sentences [1992, p.127].
This massive difference in style should not be taken as somehow less rigorous than Haffenden's interview. The relative informality has its own rhetorical purpose, and in its own way this one is at least as well constructed as the other. The sense of purpose supplied by the context helps to keep the piece moving, as it is obvious that the interviewer is seeking elucidation of performance issues. McDonald seems open to popular culture inferences; a reader of Harrison's comments quoted earlier might expect this to chime effectively with the poet's own political interests:
MCDONALD: […] That combination of mind and body relates to Plato's image in thePhaidros, namely, a chariot where the rational element drives and the irrational horses pull the chariot forward. Ancient Greek philosophy saw no isolation here. And many Greek festivals had this model: sports were included besides dramatic performances.
HARRISON: Well, also what inspires me about a place like Delphi, and many places inGreece, is that you get a sense of a festival where the person who would now be the football supporter and the person who would now be the concertgoer is the same person.
MCDONALD: Absolutely! You have large crowds, but there wasn't the same sociological division between the people.
HARRISON: And it's a destructive division. 
Given the flow of the conversation at this point, the reference to Plato seems illustrative of the point rather than potentially exclusionary and Harrison picks up on it in order to make a comparison between ancient Greece and the modern theatre that he feels is important. The style helps to keep things moving, with its informality as well as the use of connectives giving it a sense of spontaneity. The placing of these connectives at the beginning of each person's speeches makes the points flow smoothly from one to the other.
The conversation continues to touch upon the complex interplay of ancient and modern, leading into Harrison's interest in high art versus popular culture:
HARRISON: […] Now, it seems to me, that by making culture a matter of refinement it stops doing the job of culture, which is allow us to unite, even for the brief duration of a play, or a novel, or a poem, those elements of ourselves that we are always encouraged to separate by religion, or refined culture, or by social convention.
The interplay of the two protagonists' interests produces insights into Harrison's critical position, permitting the interview situation to elicit information that might not otherwise come out of secondary analysis. At the very least, the information gained from the interview helps to gloss the position Harrison takes on modern re-interpretations of classical works. The pragmatic needs of the interview do not in this instance conflict with theoretical enquiry, but enhance it:
MCDONALD: It's a beautiful postmodernist gesture to revive a culture that no one knows existed, in a sort of substratum of language that we also don't know, but which you have revived. You've signified the trace.
HARRISON: Yes, right. And my interest as a poet is to use the most 'refined' forms and fill them with a language that has not normally been granted permission to inhabit those forms. 
Harrison is able to pick up on the highly theorised language used by the interviewer to build upon the point he wants to make. So much so, in fact, that he feels able to make contentious statements in very straightforward demotic language: 'Well, Aristotle was wrong from the beginning. I think he's up the pole' [1992, p.138]. Such use of language contrasts with his responses in the Haffenden interview, pointing to something of a paradox: although the McDonald interview is less formal, it is far more revealing. The techniques used are more effective in gaining a response. However, both McDonald and Haffenden display their 'Elite' status in a manner that could be detrimental to the overall effect of their relative interviews. McDonald does make comparisons that are very similar to those in the Haffenden piece:
But of course philosophy has gone between these poles for almost all time. There's Kant and then again there's Heidegger, showing the age-old conflict between the eternal and the ephemeral, the unchanging and the changing.
Such a comment could seem exclusionist, but here the interviewer glosses it by means of a generalising statement and makes it relevant to the flow of the interview.
Published academic interviews are an important source for understanding the intentions of theatre practitioners. Academics are able to interview high profile practitioners, whereas student researchers might not. Moreover, a specialist academic's level of knowledge ensures that pre-planned interview questions and un-planned conversation between the participants is detailed and informed. This level of knowledge can allow for more probing and detailed questions and immediate response to hitherto unexplored areas. The specialist academic is also familiar with the appropriate terminology, and so should be able to ensure depth and specificity in the interview process. As a consequence of these factors, the academic interview is a vital part of performance research.
What this discussion has shown, however, is that the academic interview is also subject to a wide variety of limitations. These need to be factored into the analysis of information gained from the published interview. Chief amongst them is the agenda of the interviewer. Modern interview scholarship stresses the role of the interviewer's research agenda in shaping the range and content of the interview. Published interviews generate their information in response to pre-determined research aims, and these aims may not necessarily correspond to the research aims of the reader of the published interview. Thus the published interview can potentially close down as well as expand possible meanings. Furthermore, the very strength of the academic interviewer's knowledge and ability to communicate can be a potential weakness, if the interview loses its interrogative function and becomes a display of knowledge.
The subjectivity entailed in planning an interview schema is also matched by the subjectivity of the respondent. In addition to subjectivity derived from perspective and experience, other factors such as context (temporal) and issues of transcription can impinge on the reliability of the interview. What has been seen is that, depending on the research aim of the reader/interviewer, subjectivity need not be considered limiting, as the very act of interviewing acknowledges the importance of perspective. Yet, as part of a methodology for performance research, it is necessary to place the interview in relation to other sources of information. The interview cannot be seen as a definitive statement that encapsulates the truth of a performance; rather it is one more perspective in the building picture of a production. It may be tempting to see the interview as providing unassailable insight into the often inaccessible 'behind the scenes' world of theatre, but, in effect, the interview needs to be assessed both in the light its own structure, language and provenance and in the light of other forms of documentation and analysis.
Alison Burke and Paul Innes
P. Beatty, Understanding the Standardized/Non-Standardized Interviewing Controversy', Journal of Official Statistics, Vol. 11, no 2 1995, pp. 147-60, republished in N. Fielding (ed.) Interviewing Vol. I, (London: SAGE Publications, 2003).
E.S. Bogardus ' Methods of Interviewing', Journal of Applied Sociology, 9 (1924) 456-457, republished in N. Fielding (ed.) Interviewing Vol. I, (London: SAGE Publications, 2003).
A. Burke 'Interviews in classical performance research: (1) journalistic interviews', The Reception of the Texts and Images of Ancient Greece in Late Twentieth-Century Drama and Poetry in English, (Milton Keynes: The Open University, 2003).
N.K. Denzin 'The Cinematic Society and the Reflexive Interview', in J.F. Gubrium and J.A. Holstein (edds.) Handbook of interview research: Context & method, (London: Sage Publications, 2002).
L.A. Dexter Elite and specialized interviewing, (Evanston: Northwestern University Press, 1970).
B.S. Dohrenwend & S.A. Richarson 'Analysis of the Interviewer's Behavior', Human Organization 15 no.2 (1956) 29-32, republished in N. Fielding (ed.) Interviewing Vol. III, (London: SAGE Publications, 2003).
N. Fielding Interviewing Volumes I-IV, (London: SAGE Publications, 2003).
B. Gillham The Research Interview, (London and New York: Continuum, 2000).
R.L. Gordon 'Dimensions of the Depth Interview', American Journal of Sociology, 62
(1956) 158-164, republished in N. Fielding (ed.) Interviewing Vol. I, (London: SAGE Publications, 2003).
J.F. Gubrium and J.A. Holstein (edds.) Handbook of interview research: Context & method, (London: Sage Publications, 2002).
J. Haffenden 'Interview With Tony Harrison', in N. Astley, (ed.) Tony Harrison (Newcastle upon Tyne: Bloodaxe, 1991).
S. Kvale InterViews: An introduction to qualitative research interviewing, (London: Sage Publications, Inc., 1996).
P.F. Lazarsfeld 'The Controversy Over Detailed Interviews: An Offer for Negotiation', Public Opinion Quarterly, 8 (1944) 38-60, republished in N. Fielding (ed.) Interviewing Vol. I, (London: SAGE Publications, 2003).
M. McDonald Ancient sun, modern light: Greek drama on the modern stage, (New York: Columbia University Press, 1992).
R.K. Merton, M. Fiske, P.L. Kendall The focused interview: a manual of problems and procedures, (New York: Free Press; London; Collier Macmillan, 1956; second edition 1990).
R.K. Merton and P.L. Kendall 'The Focused Interview', American Journal of Sociology, 51 (1946) 541-57, republished in N. Fielding (ed.) Interviewing Vol. I, (London: SAGE Publications, 2003).
T. Odendahl and A. Shaw 'Interviewing Elites', in J.F. Gubrium and J.A. Holstein (edds.) Handbook of interview research: Context & method, (London: Sage Publications, 2002).
S.A. Ostrander ' 'Surely You're Not in This Just to be Helpful': Access, Rapport, and
Interviews in Three Studies of Elites', Journal of Contemporary Ethnography, 22 (1993) 7-27, republished in N. Fielding (ed.) Interviewing Vol. III, (London: SAGE Publications, 2003).
R. Pawson, 'Theorizing the Interview', British Journal of Sociology, Vol.47, no.2 (1996) 297-314 republished in N. Fielding (ed.) Interviewing Vol. I, (London: SAGE Publications, 2003).
Blake D. Poland 'Transcription Quality', in J.F. Gubrium and J.A. Holstein (edds.) Handbook of interview research: Context & method, (London: Sage Publications:, 2002).
C.A.B. Warren 'Qualitative Interviewing' in J.F. Gubrium and J.A. Holstein (edds.) Handbook of interview research: Context & method, (London: Sage Publications, 2002).
H. Zuckerman 'Interviewing an Ultra-Elite', Public Opinion Quarterly, 36 (1972) 159-175, republished in N. Fielding (ed.) Interviewing Vol. III, (London: SAGE Publications, 2003).
 We would like to thank Prof. Ian Brown and Dr Lizzie Eldridge (QMUC), and we would also like to acknowledge the theatre practitioners who have agreed to be interviewed; their invaluable participation has led to a greater understanding of the theory and ethics of interviewing.
 Odendahl and Shaw (2002) p.299.
 Odendahl and Shaw (2002) p.301.
 Zuckerman (2003 p.377; ); see also Dexter (1970) p.14.
 In comparison to Dexter's 'Elite' interview method, Kvale (1996) p.6 stresses the control of the interviewer in the research interview: 'The research interview is not a conversation between equal partners, because the researchers defines and controls the situation. The topic of the interview is introduced by the researcher, who also critically follows up on the subject's answers to his or her questions.' On the control of the interviewer see also Dohrenwend and Richarson (2003, passim; ).
 Kvale (1996) p.178.
 Zuckerman (2003, p.379; 1972), again drawing on her experience of interviewing laureate scientists, stresses that an 'Elite' respondent needs to feel his/her individuality: 'Members of this top elite and presumably others are accustomed to being treated as individuals who have a mind of their own, following their own bent. They soon detect whether questions are standardized or tailored to their interests and histories. They resent being encased in the straightjacket of standardized questions.' On the requirement not to rigidly stick to a script see also Odendahl and Shaw (2002) p.311.
 For example, Kvale (1996) p.130 argues: 'Dynamically, the questions should promote a positive interaction; keep the flow of the conversation going and motivate the subjects to talk about their experiences and feelings. The questions should be easy to understand, short, and devoid of academic language… The academic research questions need to be translated into an easy-going, colloquial form to generate spontaneous and rich descriptions.'
 Kvale (1996) pp.165-6.
 On the possible mistakes and oversights in transcriptions, see Poland (2002) p.630-633.
 Dean and Whyte in Dexter (1970) p.122. The limitation imposed by retrospection is also briefly noted by Gordon (2003 p178; ), but only to alert the interviewer to the potential difficulty.
[12 Kvale (1996) p.181.
 Burke (2003).
 Burke (2003).
 Burke (2003).
 Haffenden (1991) p. 227.
 Haffenden (1991) p.234.
 Haffenden (1991) p. 241.
 McDonald (1992) p.129.
 McDonald (1992) p.131.
 McDonald (1992) p.133.
 McDonald (1992) p.142.