Examining the doctoral viva: Perspectives from a sample of UK academics

The doctoral viva voce examination has existed in Britain since the PhD (or DPhil) was introduced at Oxford University almost a hundred years ago. However, despite some recent research studies, it seems that the viva remains somewhat under-researched and that viva voce examinations continue to take place according to largely unchanged and unchallenged procedures. This paper presents evidence collected via questionnaire and interview from a sample of UK academics working in education departments at 16 universities in England. It is shown that individuals within the sample hold contrasting views about the purpose, value, and degree of reliability associated with oral assessment at doctoral level. The paper critically examines eight aspects of the viva’s design, purpose, conduct, and outcomes, under such headings as ‘examiner judgement’, ‘examiner behaviour/attitudes’ and ‘the viva as opportunity’. The question of whether examiners should have the power to waive the viva under certain circumstances is also addressed. Rather than attempting to provide definitive answers to such questions, the paper highlights a series of inter-related issues which seem problematic. The author’s intention is to spark further discussion in academia about the viva’s current and possible future content, conduct, and purposes.


Introduction
The doctoral viva voce examination (hereafter 'viva') has existed in Britain since the PhD (or DPhil) was introduced at Oxford University towards the end of the First World War (Simpson, 1983). Yet, as Trafford and Leshem put it, 'As researchers we all know that the doctoral viva is an under-researched phenomenon ' (2008: 201). Doctoral supervisors and examiners are experienced academics but they rarely turn their critical gaze on to doctoral assessment itself, a fact which Trafford and Leshem (2008: 201) attribute to the associated barriers of confidentiality and sensitivity. Despite some relatively recent research studies focusing on various aspects of the viva (e.g. Carter and Whittaker, 2009;Carter 2011;Crossouard, 2011), it seems that viva examinations continue to take place much as they have for decades, because the prevailing perception is that overall, as one academic suggested to Park, the viva 'ain't broke', so there is no need to 'fix it ' (2003: 6). However, even if UK academics spend little time contemplating the viva, contrasting views certainly exist among them, as will be illustrated later, about the purpose, value, and degree of reliability associated with oral assessment at doctoral level. Furthermore, various factors coexist around the viva that, taken together, ought perhaps to provoke some concern. These include largely unchanged and unchallenged procedures associated with the viva; the relative lack of research into doctoral assessment; and (as will be shown) a range of sometimes conflicting positions in academia about the viva's purposes and conduct.
This paper on the doctoral viva in British universities is based primarily on a subset of the questionnaire and interview data collected by Poole (2012) from a sample of UK academics with experience of doctoral education as supervisor, examiner, and candidate. UK university publications aimed at actual, or potential, doctoral candidates (such as handbooks for doctoral students) are also briefly discussed, because this material casts light on how the viva is 'packaged' for that readership. For reasons of space, the paper examines the viva at doctoral level without differentiating between 'research doctorates' and 'professional doctorates' (such as Doctor of Education and Doctor of Business Administration).

Rationale for the study
The purpose of this paper is to contribute to and, if possible, invigorate current discussion about the fitness for purpose of the doctoral viva as it currently operates in Britain. There is no ambition to reach definitive conclusions about the way forward for doctoral assessment; no individual researcher could do such a thing. Rather, the intention is, via scrutiny of the questionnaire and interview data, to identify and then hold up for inspection vexatious issues around the doctoral viva that are worthy of wider debate.

Methodology
In this paper, the doctoral viva is subjected to scrutiny through analysis and discussion of data collected by Poole (2012), who elicited responses on a range of issues relevant to doctoral education from 27 academics (see Appendix) in education departments at 16 UK universities via questionnaires and semi-structured online interviews. Poole's questionnaires included approximately thirty statements about doctoral education and its assessment to be rated on a five-point Likert scale from 'strongly agree' to 'strongly disagree', but also, at half a dozen points in the questionnaire, prompted respondents to add written comments if they wished to explain or amplify their Likert-scale ratings. This paper focuses only on those written comments that concern the viva, along with relevant extracts from transcripts of the follow-up interviews conducted several months later with 7 of the questionnaire respondents. These 7 were selected because, among the 27, they either made most use of 'strongly agree' (2 respondents); 'strongly disagree' (2 respondents); or 'neither agree nor disagree' (3 respondents) in rating the questionnaire statements. Analysis of the Likert-scale data is provided in Poole (2012). The questionnaire respondents were assigned a number (1-27) reflecting the order in which responses were received. The 7 interviewees were assigned a letter (T-Z), so that, for instance, questionnaire respondent 6 is interviewee T. The 7 interviewees represented 6 different UK universities, with interviewees U and Y being from the same university. The identities of the 27 participants, and the universities they work for, are known only to the present writer.
This paper aims to foreground particular issues about the doctoral viva for debate by examining the data with general questions in mind, such as: • What are or should be the purposes of the viva? • Do any of these purposes conflict?
• What factors affect the reliability or fairness of the viva (across candidates and institutions)? • Are any changes needed in the conceptualization and conduct of the viva?
The intention is to make a contribution to, or to catalyse, the debate on the future of the doctoral viva. This debate has implications not only for the UK but for doctoral assessment worldwide.
The approach taken in relation to the data is interpretivist. Quantitative methods are not appropriate for what is a relatively small sample: 27 academics out of the thousands who are involved in doctoral education in the UK, or the many hundreds who, like the 27, work in university departments of education. Readers may find the questions raised in the concluding section to be worthy of further debate if they judge that the interpretations offered, earlier in the paper, of the various textual extracts and quotations are themselves plausible. As Kelliher (2005: 123) notes, in social sciences research interpretivism is recognized as providing contextual depth but can be criticized in that findings may not be generalizable. However, since the main purpose of this paper is not to offer 'solutions' but to provoke further discussion of the issues it raises, an interpretivist approach to the data seems appropriate.
The present paper may appear to have something in common with an action research project, in that it aims to contribute to improved practice in a given professional area (assessment in doctoral education). Proponents of action research (such as Reason and Bradbury, 2008: 9) might argue that my vantage point (as a non-member of the UK academic community) must mean that my interpretations are 'filtered through an outsider's perspective'. However, I argue that this need not trouble the reader, as I am not seeking to identify solutions to a problem (as action research characteristically does) but to stimulate further debate among 'insiders' (members of the academy in the UK). Indeed, it might be argued, conversely, that as an 'outsider' I avoid the biases that arise when researching one's own practices. Of course, in the interpretation of data there is no 'view from nowhere' (Nagel, 1986, cited in Sealey andCarter, 2004). Because my academic background is in the social sciences and the data have been gathered from academics in the field of education, discussion of the viva here may be most relevant to disciplines of that kind. This paper follows the ESRC Research Ethics Framework (2012: 3) in that the anonymity of the respondents is respected. Where a pronoun 'he' or 'she' is used, however, this reflects the gender of the respondent.

Interpretation of the data
I shall provide a commentary on data obtained from the questionnaires and the interviews under eight headings. In my view these capture the key issues, which are in some cases interlinked and overlapping, raised by this sample of academics when asked to comment on the design, purposes, and conduct of the UK doctoral viva.

Examiner judgement
Discussion here centres on comments by respondents about judgement, interpretation, and subjectivity in relation to the task facing examiners before, during, and after the viva. Let us begin with two comments from questionnaire respondent 22: Respondent 22 seems to think that the viva can be unfair because of 'inconsistencies of approach' from examiners. He also argues that passing or failing a viva may not reflect the individual's true ability. In essence, this respondent appears to think that there is too much subjectivity at work, and that some examiners follow their own idiosyncratic 'approach' to the viva. However, this is not supported by all the academics in the sample. For example, respondent 27 takes a robust position that directly opposes that of 22: 27: There are already adequate QA procedures in place. Universities provide staff development for examiners; there are both internal and external examiners; the examiners are selected on the basis of appropriate experience; the external examiner's c.v. is scrutinised by a committee; the supervisor can sit in on the viva to ensure fairness; independent pre-viva reports are written and exchanged by the examiners; the post-viva report is co-written by the examiners; the post-viva report is scrutinised by a committee; etc.
Generally speaking, the academics who were interviewed hold more nuanced views than either 22 or 27. They accept that being an examiner for a doctoral viva involves the exercise of (sometimes fallible) judgement, but they see this as unavoidable. Interviewee T sees the examiner's role as 'a matter of interpretation' and as involving 'a value judgement', but also notes that 'universities have very useful criteria for doctorateness and they are written down'. In other words, he accepts that the exercise of examiner judgement entails the possibility of inconsistency (as mentioned by 22), but sees the guidance provided to examiners by UK universities as constraining this. Interviewee X sees these issues in a similar light, commenting that 'subjectivity is unavoidable in any form of assessment' and admitting that this 'can cause problems'. V also regards the outcome of the viva as 'a judgement call for examiners' but adds that 'on the whole I think the process is reliable'. In the questionnaire data, respondent 11 gives the view that 'it is very challenging to standardize the examination of doctoral theses'. This comment provides a suitable coda for this subsection. If a doctoral viva takes place, it will inevitably require examiners to make a judgement, and such judgements (even when guided by university documentation and made in good faith) may at times be unreliable.
However, some of the data (described below) concern 'unreasonable' examiner behaviour or attitudes, eroding the reliability and fairness of the viva.

Examiner behaviour and attitudes
Some comments in the data suggest that examiner behaviour and attitudes can detract from the fairness of a given viva from the candidate's point of view. Interviewee T mentions cases in which he has seen 'friction' between participants in the viva: T: In about sixty vivas I have only seen friction about four times and this has been due to 'bad behaviour' by one examiner, such as being pompous, being aggressive, or arrogant (as in 'I had a hard time during my viva and I am going to do the same here…').
These comments are supported by others. For instance, Emeritus Professor Ronald Barnett (Institute of Education, University College London) estimates that negative issues caused by 'maverick, quirky examiners' tend to occur in about 5 per cent of vivas (Gibney, 2013: 2). Grabbe (2003: 129) also notes that he has found some vivas 'unsatisfactory'.
Various types of unreasonable examiner behaviour or attitudes are cited by individuals in the sample: Y: What irritates me is when an examiner seems to think the process is about her/him, and not the candidate.
Y: Examiners sometimes seem to feel the need to bring their own agenda to the viva, which I think is very unfair.
Y: I have heard of other individuals being present (as candidate or examiner) where conflicts have arisen as a result of one examiner not particularly liking an aspect of the thesis -when it is not about liking and more to do with academic rigour. 23: They [examiners] are sometimes not willing to look outside their own paradigms to recognise doctoral level work.
Interviewee T introduces the notion of the viva as a 'rite of passage', a description commonly found in relevant scholarly literature. For instance, some academics consulted by Jackson and Tinkler (2001: 360) see the viva voce examination in that way, and the phrase occurs in Sinclair's (2007) title. Interviewee T said: T: I don't understand this rite of passage thing. Is it some kind of public schoolboy thing like being bathed in freezing water or being beaten? I would never choose an external if he saw it this way or if I knew he was going to be pompous or have bees in his bonnet.
In linking the expression 'rite of passage' to the kinds of public-school mistreatment described by writers such as Cyril Connolly  in Enemies of Promise and George Orwell (1903-50) in 'Such, such were the joys', respondent T might perhaps be motivated in part by the disagreeable connotations of the phrase as used in anthropological studies. There the term is often associated with experiences such as male and female circumcision (Womack, 2010), hazing (Maruna, 2011), and scarification (Gennep, 1960). Unpleasant initiation ceremonies or practices, such as 'dedovshchina' (дедовщина) for new recruits in the Russian army, are also sometimes dubbed 'rites of passage'. It would be ridiculous, of course, to suggest that doctoral candidates undergo experiences as harrowing as these, but one might still wonder whether examiners who see the viva as a 'rite of passage' are more likely to approach it in a confrontational spirit than those, such as T, who do not.
Even if unreasonable behaviour by examiners occurs in only 5-7 per cent of vivas, as suggested by T and by Barnett, this figure could be reduced. Questionnaire respondent 1 reports that 'In my University there is an independent Chair for each viva, which helps to take care of fairness.' Relevant University of Cambridge guidelines (2015), for example, state (inter alia) that an independent chair should ensure that: the viva voce examination process is rigorous, fair, reliable and consistent; the candidate has the opportunity to defend the thesis and respond to all questions posed by the examiners; questioning by the examiners is conducted fairly and professionally.
(University of Cambridge, 2015) However, by no means all UK universities, or all faculties within specific universities, appoint an independent chair. This may partly be because of the additional costs involved. An alternative, noted by respondent 27, is that 'the supervisor can sit in on the viva to ensure fairness'. A further potential safeguard sometimes mentioned in relation to inconsistent or unreasonable behaviour by examiners is audiotaping or videotaping of the viva. Respondent 15 notes that 'many universities already do this'. However, respondent 18 argues that: The thing is that the viva is an extremely complex business, and things like pauses, intonation, glances, interaction between people, facial expressions, hesitation and physical gestures are not captured by video/audio. Several respondents also suggest that the presence of video or audio recording equipment in the room might make the viva even more stressful for candidates. This relates to the next subsection, which deals, in part, with affective factors in relation to the viva.

Cultural, affective, and 'cognitive style' factors in the viva
It is sometimes argued that cultural and affective factors can inhibit candidate performance at the viva. How (if at all) are such factors taken into account by examiners and by the guidance provided for them? Respondent 14, a very experienced UK academic, suggests that 'examiners should (and generally do) do their best to put the student at his/her ease and get beyond the personality factor in establishing whether the student understands and can defend their thesis'. However, she also states that 'it would be rash to claim' that cultural and affective factors never impact on candidate performance. Respondent 23 states that 'disinclination to challenge authority figures' is not always a cultural factor but can be a personality factor and argues that this can affect the viva's outcome -'Where the candidate is … not expecting the examiners to give them a grilling.' Interviewee X argues strongly that cultural and affective factors genuinely can skew the outcome of the viva: X: I have experience of a Japanese woman who could not bring herself to challenge the authority of the examiners and hardly spoke. The examiners wrongly concluded that she did not understand the questions. I also have experience of non-native speakers misunderstanding questions and giving the wrong answers or misreading the cultural expectations and irritating the examiner (e.g. lack of deference).
In respondent 14's view, irrespective of the cultural assumptions of the candidate and the nature of his or her personality, the fault lies with the supervisor if the candidate enters the viva unprepared for challenging questions and disinclined to mount a defence.
Cognitive style may also affect how a candidate is perceived by examiners at the viva. Noble (1994: 67) argues that traits or abilities such as 'public speaking skills', confidence, and the ability to think on one's feet should not be regarded as prime criteria for the award of a doctorate. Similarly, Tinkler and Jackson (2002: 89), warn against conflating the ability to 'think on the hoof' with 'intellectual competence', noting that 'some people require time and certain conditions to tease through problems'. The slower thinker (who may not perform particularly well in the cut and thrust of a viva) is not necessarily inferior, or less doctoral in nature, and may actually be capable of offering very perceptive observations if allowed sufficient time to ponder. Park (2003: 48) argues that 'the viva is a real hurdle that the student must negotiate'. However, there is evidence to suggest that this is not always the case: an arts student is quoted by Tinkler and Jackson as saying, 'I walked in and they gave me a bottle of champagne and said "Relax you've passed" ' (2002: 94).

Summative and formative purposes of the viva
In this subsection we focus on whether it is proper for the viva to function, at the discretion of the examiners, as either a sort of coronation ('I walked in and they gave me a bottle of champagne') or as an arena for lengthy and sometimes robust questioning. Is it satisfactory that the candidate enters the room having no inkling of whether to expect immediate congratulations or a prolonged interrogation? Here is another example of the 'champagne' sort: W: I had one candidate recently who attended her viva. The external said 'thank you for a very interesting thesis, I enjoyed reading it'. Then he asked her to turn to page 45. Told her off for missing out a comma. She apologized profusely. He then said 'Congratulations Dr B., apart from that there's nothing else you could do to make it any better. You've passed.' She burst into tearsrelease of tension -and we all then had a chat about the thesis and went home.
Similarly, interviewee X notes that 'there are times when the thesis is so good that the result is announced to the student at the beginning of the viva' although 'even in these situations the student is usually asked questions'. Interviewee Z states that in his experience the viva varies according to whether the thesis is 'brilliant', 'middling', or 'weak': Z: … the 'brilliant thesis' ... you start with 'This is brilliant, basically I just want to establish that you wrote it' … but actually you then engage in a really deep intellectual discussion that almost always gets on to publications. Then there is the 'middling thesis' ... You (as examiner) are looking for 'excellent bits' to put in the report (to justify passing) and probably scrapping away at how to improve sections that are not quite as you want. Then there are weak theses ... these are so varied that I won't begin to generalize.
According to Z, therefore, the viva for the 'brilliant' thesis largely consists of advice on publication. The viva for the 'middling' thesis is formative ('scrapping away … to improve sections'). Regarding 'weak' theses he avoids descriptive comment, but one must assume that some of these (if they are 'weak') fail and/or lead to the award of an MPhil (summative).
Interviewee X takes the view that failure at the viva is a 'tragedy'. In her view students should be advised before they reach the viva if their work is not of doctoral standard: X: The viva is actually a small part of getting a doctorate. The production of the thesis can include many assessments from both supervisors and from independent readers before it gets to the viva. Many students fail and don't get to the viva. It's a tragedy if failure occurs at the viva. Students shouldn't be allowed to get this far and fail.
However, X also makes a statement that appears to conflict with the preceding one: X: I have experience of the viva producing surprising outcomes ... a pass for a student who both examiners thought they would be failing after reading the thesis and the reverse -students failing where the examiners thought that they would be passing the student.
In other words, although X thinks that someone (probably the supervisor) should prevent weak candidates/theses from proceeding to the viva, she also accepts that they do sometimes go forward -and pass. Interviewee V also mentions redemption at the viva: V: I think the viva is very important. I have seen people salvage their PhD because they presented themselves and their work well under questioning.
This raises the question of the balance between the written thesis and the candidate's viva performance in deciding whether or not a doctorate should be awarded. It also brings to the fore the issue of whether doctorateness (Trafford and Leshem, 2009;Wellington, 2013;Poole, 2015) should be seen primarily as a property of the thesis, the candidate, or both (this is discussed further below).

The viva as an opportunity or 'chance to shine'
Many UK university documents addressed to doctoral students (such as student handbooks) characterize the viva as an 'opportunity', and some comments from the questionnaire and interview data echo this. An online search of relevant handbooks in February 2014 produced hits at universities including Bath, Birmingham, Cardiff, Edinburgh, Glamorgan, Oxford, and Queen's University Belfast. Here are four examples: The Viva Voce examination is your opportunity to defend your thesis.
(University of Bath School of Management, 2011: 37) During the viva the examiners will give you the opportunity to defend your thesis.
(University of Edinburgh, School of Literatures, Languages and Cultures: 2015) [The viva] is an oral examination at which the EdD candidate has an opportunity to defend their thesis in discussion with their examiners.
(Queen's University Belfast, 2010: 19-20, bracketed words added) … you will have the opportunity to present and discuss your work with the examiners at an oral examination (a viva voce) … (Institute of Education, 2014: 86) While the viva is characterized as being an 'opportunity', it is also compulsory: A viva voce examination is mandatory once a Doctor of Higher Education Management candidate has submitted a thesis.
(University of Bath School of Management, 2011: 36) All EdD candidates will be required to sit a viva-voce to defend their thesis. (Queen's University Belfast,2010: 19-20) A compulsory opportunity is almost a contradiction in terms, because in everyday life 'opportunities' can be seized or declined according to preference. However, this facet of the viva is commonly spoken about by academics in the sample, even when they do not use the word 'opportunity' itself: 26: The viva can also be seen as a kind of reward … in that two examiners are prepared to give extensive consideration to your work.
T: Everyone in a viva can see each other and have a decent dialogue (the chance to shine, to explain, and justify).
T: the candidate should … be given the chance to shine and also to be challenged in a pleasant but thorough way.
T: The candidate deserves a chance to talk in depth about her or his own work after all the effort they have put in.
Such a view is also endorsed by other experienced academics, such as Professor Louise Morley (University of Sussex) and Emeritus Professor Vernon Trafford (Anglia Ruskin University) (Gibney, 2013: 2). For those academics in concurrence, the viva might be seen predominantly as an event entailing the potential to induct new people into 'the academy'. For them it is more a vehicle for constructive academic community-building than a theatre in which to practise flinty-eyed gate-keeping (Jackson and Tinkler, 2000). Nevertheless, characterizing the viva as an 'opportunity' when, for the candidate, it is compulsory and can lead to failure, might be regarded by some as an example of muddled thinking.

The viva as a context for the judgement of originality or original contribution
This subsection looks at respondents' comments on originality or 'original contribution' in relation to the viva. These notions (originality, original contribution) are often regarded as central to decisions made by examiners about whether or not a doctoral candidate passes (see, for example, QAA, 2011;Wellington, 2013). Respondent 27 apparently sees the issue as being straightforward, arguing that 'We read this stuff for a living. We know when somebody is making an original contribution and when they are not.' However, not all academics in the sample agree with him: Interviewer: Do you think this issue of 'original contribution' is the source of any idiosyncratic decisions by examiners -or could potentially be so?
Y: Yes I do -and often as a result either of poor guidance or of lack of attention to that guidance.
Y then develops this thought a little: Y: Having examiners base decisions on their own interpretations (of originality) is unfair on the candidate, and potentially quite inequitable from one student's viva to another.
W, on the other hand, dismisses originality altogether as a factor in the decisions of doctoral examiners, saying that in 16 years of involvement in doctoral education he has 'never had a conversation with anyone either in a viva or outside about originality'. These divergent views provide initial evidence that academics take on the role of doctoral examiner with a range of assumptions about originality, and therefore perhaps about other concepts that surround the notion of 'doctorateness' (or doctoral quality). Can the viva be seen as a reliable instrument if one academic who has experience as an examiner regards originality as straightforwardly identifiable, another as the source of idiosyncratic examiner decisions, and a third as essentially irrelevant?

The locus of 'doctorateness'
This subsection considers the perceptions of the sample of academics of whether 'doctorateness' is best seen as a property of the candidate, the written thesis, or both. This issue arises because if doctorateness is seen primarily as a characteristic of the thesis, it could be argued that the face-to-face viva might no longer be essential in all cases. Several respondents referred to the fact that doctoral assessment in Australia almost never involves a viva: V: I'd like to see Australia go over to vivas. Z: I examined for an Australian PhD about six years ago … quite understood the procedure (in terms of distance) but, frankly, thought it inferior.
It is sometimes argued that the typical Australian procedure (with no viva) exists for historical reasons -and V puts forward a version of this: V: I believe they started this because the community was too small and intimate to be able to provide the distance an examiner needs, but that's not the case anymore and hasn't been for a couple of decades.
Whatever the reason for the absence of the viva in Australia, one could ask whether this in any way compromises the standing of Australian doctorates. If not, then the doctoral viva might be regarded by some as a costly optional extra (in terms of both examiner time, and fees paid by universities to examiners), rather than as an indispensable part of the assessment process. However, the latter view only obtains if one takes the view that the locus of doctorateness is the written thesis and not the candidate. Along with other interviewees (see also the next subsection), Y argues that the viva is needed: Y: The viva examines the links ... between the thesis and the candidate. As such it is not really possible to separate the two.
In the view of Y, the viva assesses the extent to which the candidate understands and can 'fully defend' the thesis. This also relates to the view that the viva is needed in order to confirm the candidate's authorship ('I just want to establish that you wrote it' -interviewee Z, above), although it could be argued that the candidate's university, and especially the supervisor, should already know this beyond reasonable doubt. Z explicitly states that doctorateness should be understood as residing in both thesis and candidate: Interviewer: Does the viva assess the candidate or the thesis? Is doctorateness a property of the candidate, the thesis, or both?
Z: This is easy. 'Both' is the answer to both questions.
If most UK academics agree that doctorateness is a property of both the thesis and the candidate, this suggests that the present viva (or something similar, involving face-to-face real-time dialogue) remains necessary. Logically it also means that Australian doctorates (lacking a viva) rest on a somewhat different conception of doctorateness.

Is the viva needed at all?
This subsection examines the issue of whether the viva should continue to be compulsory, whether it could be waived in certain well-defined cases (at the discretion of the examiners), or whether it should be abolished altogether.
Gill and Burnard argue that, in the doctoral viva, too much is left to the subjective judgement of the examiners and that the approach taken in doctoral assessment is 'anachronistic' and 'would probably not hold up to public scrutiny ' (2012: 477). In seeing things this way, they are adopting a position similar to that of Noble, who puts the forceful case that the viva is 'an anachronism that can be traced back to the Middle Ages' and suggests that there should be 'more emphasis on the examination of the written thesis' and none on 'oral and visual presentation skills ' (1994: 67-8). Although Noble's arguments predate the use of the term, they relate to the issue of whether 'doctorateness' should be seen as a characteristic of the written thesis and again bring into focus the contrast between normal Australian practice (no viva) and UK practice (compulsory viva).
In the interview and questionnaire data there is, however, overwhelming support for the retention of the viva, although some individuals argue that examiners should be allowed to communicate their decision on the thesis in writing only (without conducting a viva) in certain well-defined circumstances. Here are some statements from respondents who find the viva to be an essential component of doctoral assessment: T: The viva is a vital part of the examination process.
V: I think the viva is very important. I have seen people salvage their PhD because they presented themselves and their work well under questioning. I'd like to see Australia go over to vivas.
18: I believe the ... viva is absolutely essential and should be maintained. Z: I examined for an Australian PhD about 6 years ago ... quite understood the procedure (in terms of distance) but, frankly, thought it inferior. Also examined a PhD once in Europe where they do a full theatric display ... I was so concerned for the student that I was academically 'easy' on him. I actually like the UK system where we can grill the student and also backtrack and say comforting words if it looks like they're freaking.
Nevertheless, even among those who favour retention of the viva there are those who feel it should be waived in certain circumstances. W recounts a case (mentioned above) where the external examiner merely asked the candidate to add a comma to her text and then told her that she had passed. The interview continued as follows: W: Better not to have dragged everyone to the viva! Good fun -overnight stay, expenses paid, but not really any need.
Interviewer: So in that case it should have been waived, do you think? And feedback sent by post or email instead? W: Yes, definitely.
Finally, one individual argues for the abolition of the viva. When he completed the questionnaire (as respondent 22) he recalled his own experience of being a doctoral candidate and stated that 'I quite enjoyed my viva -but am not quite sure if it was a necessary aspect of the process.' A few months later, when interviewed (as interviewee Y), his views seemed to have hardened: Interviewer: Would doing away with the viva significantly weaken the standing of UK doctorates? Y: Not in the slightest … I see no real reason to have a viva … The examiners could quite easily pose their queries/demand amendments without a viva.
In the sample of 27 academics the great majority felt that the viva should be retained, but several argued that the examiners should have the right to waive it when they considered it unnecessary. As we have seen, Y considered that it should be abolished altogether.

Limitations of this study
Before proceeding to considering the implications arising from this study, a note of caution is required. While the sample of academics consulted here is larger than in some other studies (see, for instance, Taylor, 2008) it is certainly not possible to suggest that the amount of data justifies the drawing of firm conclusions about the views of academics currently working in education departments in UK universities. This is not the aim or the intention. Instead, it is hoped that the act of highlighting the comments of individual academics may stimulate further debate about the purposes, design, and conduct of the doctoral viva. The following section draws attention to particular issues, arising from the data, which could provide focuses for that debate.

Implications
Within the sample of 27 academics there is little support for abolishing the viva, but disparate views can be found on its purposes and conduct, and on whether modifications could or should be introduced. I now present a set of questions that, according to my interpretation, spring from the data, and that could usefully be the subject of further research or debate among those who are actively involved in doctoral education in the UK or elsewhere: • Given that doctoral examiners do sometimes exhibit 'bad behaviour' (interviewee T) or show 'quirky, maverick' tendencies (Barnett, in Gibney, 2013: 2), what are the most appropriate safeguards for the candidate? Appointment of an independent chair? Presence of the supervisor? Video or audio recording? • Should there be clearer guidelines for examiners on how to integrate their impressions of the written thesis with their view of viva performance in reaching an overall decision? To what extent should poor viva performance detract from the production of a good written thesis, or good viva performance be allowed to compensate for a flawed dissertation?
• Should a greater percentage of candidates be advised by supervisors that their work is not up to doctoral standard rather than going on to the 'tragedy' (interviewee X) of failure at the viva? • Do doctoral examiners adequately take into account cultural and affective factors at the viva? Do they receive enough (or any) training in such matters? • Should examiners have the right to waive the viva when they consider the thesis to be excellent, instead submitting their comments and requirements (such as minor changes) in writing and (perhaps) congratulating the candidate by phone?

Conclusion
This paper has examined the doctoral viva in the UK as it currently exists and is practised through consideration of a selection of comments made by academics. Rather than providing recommendations, the paper has concluded with a series of interrelated questions, designed to spark further discussion about the viva's current and possible future content, conduct, and purposes.