ÿÿ Diane Thompson Institute of Distance Education ÿ Deakin University ÿÿ ÿÿ ABSTRACT ÿ2ÿ ÿÿ ÿÿ Twenty years ago Parlett and Hamilton wrote of organising ÿDÿ@research heuristically, with the researchers "progressively ÿ@ÿtheory, formulation and critique shaped and re-shaped the ÿ>ÿ=research as it evolved and freed the researcher from the ÿ=ÿ1constraints of the traditional linear model. ÿ1ÿ ÿÿ ÿFÿ ÿÿnThe word `research' comes from the Latin r e - c i r c e r e, `to go round ÿnÿFagain'. This encapsulates a critical aspect of the methodology I ÿFÿhave espoused. ÿÿ ÿÿ>In earlier studies within the Faculty of Education, I was ÿ>ÿEparticularly responsive to Scriven's (1971) writing on goal-free ÿEÿ@evaluation where the significance of unintended and unknown ÿ@ÿEobjectives is recognised. Hence, when more than a decade later, ÿEÿ?as a researcher rather than a practitioner, I came to read ÿ?ÿ?Parlett and Hamilton's (1972) seminal work on illuminative ÿ?ÿDevaluation where they outlined a `social anthropology paradigm' ÿDÿ>that accommodated such outcomes, I was both reassured and ÿ>ÿDchallenged. They wrote of organising their work heuristically, ÿDÿDwith the researchers "progressively focusing and redefining the ÿDÿFareas of study as the study unfolds, in the light of accumulating ÿFÿ>experience and as the crucial issues-to-be studied become ÿ>ÿAuncovered". Eclectic, adaptive and responsive to individual ÿAÿFresearch milieu, there is no methodological package premised. In ÿFÿframework, most research on distance learning attempts to ÿ>ÿ>find solutions to perceived problems. While approval for ÿ>ÿAthis approach from administrators and sponsors may be easier ÿAÿ5to obtain, the results are rarely generalizable. ÿ5ÿ ÿÿÿÿDMoore (1991) explicitly connected knowledge with theory when he ÿDÿFcontended that theory is the summarizing and synthesis of what is ÿFÿEknown about a field; it is the reduction of our knowledge to the ÿEÿ@basic ideas, presented in a way that shows their underlying ÿ@ÿ+patterns and relationships. He wrote: ÿ+ÿ ÿÿC Research that is not grounded in theory is wasteful. It ÿCÿAmight solve an immediate problem, but it doesn't fulfill its ÿAÿus to analyze the particular instance more effectively... ÿ>ÿ ÿÿ ÿÿCFor this study, the ultimate interest is in the generalizable, ÿCÿconsciousness and thereby enlarge their chances of a more ÿ>ÿ?reflected-upon and better choice of action. (Marton, 1979, ÿ?ÿ p13) ÿ ÿ ÿÿ ÿÿFI believe that, in ways such as this, a bridge can be constructed ÿFÿ:between theory and practice - between scholarship and ÿ:ÿDapplication. `The problem' unifies theory and practice. Where ÿDÿFenquiry is motivated by a problem, there is no question about the ÿFÿArelevance of its product - knowledge or theory - to practice ÿAÿB(Sherman and Webb, 1988, p16). `The problem' studied here is ÿBÿDhow, and how much, tertiary learning is effected by technologic-ÿDÿ@ally mediated formal spoken discourse. Substantively, `the ÿ@ÿCproblem' has been considered in relation to distance education ÿCÿEstudents who engage in oral interaction by telephone, in what we ÿEÿhave termed teletutorials. ÿÿ ÿÿ ÿáÿ ÿÿBFrequently equated with interpretive (see, for example, Howe, ÿBÿB1992, p240), hermeneutics has grown from a subsidiary role in ÿBÿFtheology covering the study of methods for the authentication and ÿFÿCinterpretation of texts to "a general name for current studies ÿCÿEfavoring the hypothesis of something special in the epistemology ÿEÿFof `understanding'" (Howard, 1982, pp1,3). Significantly for the ÿFÿ@predilections and beliefs of this researcher, hermeneutics: ÿ@ÿ ÿÿ? ... accommodates an approach which is exploratory in ÿ?ÿ>intention rather than goal-oriented; which gives positive ÿ>ÿ@recognition to the subjective prejudice and tacit knowledge ÿ@ÿAof the researcher; and which acknowledges that the direction ÿAÿBof the research is propelled by a dynamic interaction between ÿBÿ@the researcher and the research setting. describe ÿ?ÿ@it, ... illustrates ... the concept that the part is always ÿ@ÿBto be understood in relation to the whole. (Grace, 1990, p23) ÿBÿ ÿÿ ÿ(ÿDAlthough seen as a hallmark of any qualitative research (Edson, ÿDÿE1988, p46) and not confined to hermeneutics, appreciation of the ÿEÿCsignificance of context is an integral tenet of hermeneutics. ÿCÿ?Although there are exceptions, one criticism of positivist ÿ?ÿEmethodology is that it tends to diminish the significance of the ÿEÿFexistential condition in which problems arise with its propensity ÿFÿEto fragment and isolate inquiries (Giarelli and Chambliss, 1988, ÿEÿDp34). In contrast, qualitative inquiry is neither context-free ÿDÿ>nor context-independent. Rather, it is context-specific, ÿ>ÿCpositing that ideas, people and events cannot be understood if ÿCÿ5isolated from their contexts (Edson, 1988, p46). ÿ5ÿ ÿÿCAs the hermeneutic paradigm became more pervasive, there was a ÿCÿAnecessity to confront the issue of context-specificity. The ÿAÿFnotion of discrete episodes that had the comfort and assurance of ÿFÿ@bounded events gave way to a realisation that no reading is ÿ@ÿFcontext-free and each session, whether teletutorial or interview, ÿFÿFis a step in a sequence. As Kahn and Cannell (1983, p104) wrote: ÿFÿE"The interview takes place in a context or matrix of such steps, ÿEÿDand it can be understood and successfully practiced only if the ÿDÿ+interviewer is aware of this context". ÿ+ÿ ÿÿDSuch an approach necessitates awareness of both the literal and ÿDÿClinguistic context in which the spoken utterance is embedded. ÿCÿFContextual description is now recognised as a requirement for the ÿFÿFanalysis of discourse (Brown and Yule, 1983, p27). Stubbs (1988, ÿFÿ@p66) has provided a useful summation of the main arguments: ÿ@ÿ ÿÿF ... a failure to study context reifies the object of study, ÿFÿBby neglecting the interpretative procedures by which situated ÿBÿ9meanings are constructed, and by failing to treat as ÿ9ÿ?problematic the ways in which social order is successfully ÿ?ÿ:accomplished by members. ... Context-free studies of ÿ:ÿ>language are reductionist: they reduce praxis to process; ÿ>ÿBthey reduce the study of meaningful behaviour to the study of ÿBÿ:ideal-typical structures, taking for granted how such ÿ:ÿ=structures are interpreted and used in context; and they ÿ=ÿ:refuse to study the essential meaningfulness of human ÿ:ÿ@behaviour, and how people make sense of social interaction. ÿ@ÿ ÿÿ ÿÿ=Several methodological considerations were surfaced as a ÿ=ÿCconsequence of the decision to pay close attention to context. ÿCÿ ÿÿ@Conventionally, the context that provides the frame for the ÿ@ÿ/discourse is interpreted as the situation: ÿ/ÿ ÿÿD We do not experience language in isolation - if we did we ÿDÿEwould not recognise it as language - but always in relation ÿEÿ>to a scenario, some background of persons and actions and ÿ>ÿ=events from which the things which are said derive their ÿ=ÿBmeaning. This is referred to as the `situation', so language ÿBÿ;is said to function in `contexts of situation' and any ÿ;ÿAaccount of language which fails to build in the situation as ÿAÿ;an essential ingredient is likely to be artificial and ÿ;ÿ+unrewarding. (Halliday, 1978, pp28-29) ÿ+ÿ ÿÿ ÿÿEReferring to classroom research, Furlong and Edwards (1988, p57) ÿEÿFwarned that there is a temptation to let "common-sense notions of ÿFÿFwhat the situation is like" do much of the explanatory work where ÿFÿBthe context is familiar. One way in which the context is not ÿBÿB`familiar' but warranted particular investigation is that the ÿBÿFsetting in which the telephone calls are taken is external to the ÿFÿDuniversity. A significant problem here is that this context is ÿDÿFvirtually limitless as it potentially includes the environment of ÿFÿCeach participant. Although the environment in which calls are ÿCÿ?taken was addressed in both questionnaires and interviews, ÿ?ÿ>practical and ethical issues denied the researcher direct ÿ>ÿ5knowledge to shape her perceptions in this area. ÿ5ÿ ÿÿFAlthough, then, the teletutorial has a less familiar setting, the ÿFÿ(Speier, 1973, p34). Once this was recognised, it became ÿ>ÿDimportant to interview participants before they had experienced ÿDÿFany teletutorials to determine what their preconceptions were and ÿFÿ@to set these alongside what they said of their experiences. ÿ@ÿ ÿÿBA further contextual matter of concern to Furlong and Edwards ÿBÿB(1988, p57) is that classroom research has frequently assumed ÿBÿDthat pedagogy varies little from teacher to teacher. Awareness ÿDÿCof this issue shaped the earlier research to include different ÿCÿEteachers and it was rapidly discovered that other factors needed ÿEÿCto be considered. Institution, faculty, discipline, level and ÿCÿFgender are just some of the significant issues to explore and the ÿFÿCselection of case studies has been strongly influenced by this ÿCÿappreciation. ÿÿ ÿÿFTo study context necessitates, however, going further than paying ÿFÿBattention to background factors. Edwards and Westgate (1987, ÿBÿBp178) wrote that `severe problems' occur "once it is accepted ÿBÿFthat contexts are not fixed frames of reference within which talk ÿFÿDtakes place and has its meaning, but are themselves talked into ÿDÿDbeing, renewed or challenged". The collection of papers Tannen ÿDÿD(1982) edited demonstrates that complexities found in discourse ÿDÿCin context reflect not only its spoken or written mode but its ÿCÿEinteractive goals and structures with genre, register and speech ÿEÿ@event all playing significant roles. Thus distinctions are ÿ@ÿ>`foggier'. The situation is made more complex when it is ÿ>ÿ@appreciated that "individuals and societies (are) caught in ÿ@ÿFchanging traditions of orality and literacy intertwined with each ÿFÿDother and with chirography, print and technology" (pxii). This ÿDÿ;research, therefore, has increasingly sought to deepen ÿ;ÿ@understanding of the complexity of the context in which the ÿ@ÿBdiscourse is set as one means of more fully understanding the ÿBÿ@meaning of the interviews and the spoken interaction of the ÿ@ÿ?teletutorial. Here the mediation of the technology of the ÿ?ÿ1telephone has demanded particular attention. ÿ1ÿ ÿÿDI am indebted to Idhe (1982, 1990) for strengthening my resolve ÿDÿ>to make the `transparency' of the medium of the telephone ÿ>ÿ=problematic. His explicit connection of the medium to a ÿ=ÿFhermeneutic role "by being situated between the direct expressive ÿFÿDactivity and recipient" (1982, p62); "by being mediated" (1982, ÿDÿEp70), as well as his writing on hermeneutic technics (1990, pp80-ÿEÿF97) seemed particularly apposite. As the tape-recorded data were ÿFÿDlistened to, a further technical mediation became apparent: the ÿDÿBtape recorder, itself. Thus this study has been increasingly ÿBÿDconcerned to discern the transformations the technology implies ÿDÿ?and reflect on the ways these influence the communication. ÿ?ÿ ÿÿDAs far as context is concerned, a critical factor is awareness: ÿDÿAof the need to explore a chosen issue or issues from a multi-ÿAÿCdimensional approach; of the need to be sensitive to the other ÿCÿAelements that are begun to be perceived as understanding and ÿAÿBsharing of ideas develops; of the need to seek a network that ÿBÿAcollaboratively and synthetically merges different fields of ÿAÿ@scholastic endeavour. Qualitative research helps to expand ÿ@ÿCframes of reference and fosters an appreciation for complexity ÿCÿD(Edson, 1988, p45); it moves us closer to totality but helps us ÿDÿClive with the awareness that an exhaustive study is beyond us. ÿCÿ ÿÿEA tension remains. If the aim is to portray as comprehensively, ÿEÿBas substantively, as possible, to appreciate all nuances, but ÿBÿBthere is a compulsion to take the data beyond description, to ÿBÿ@generate concepts - even theory, there is competition. One ÿ@ÿFapproach seeks to mirror; the other seeks to explain, to account. ÿFÿ ÿÿF To opt for the portrait or the mirror reflection is to want ÿFÿ@to capture the intimate flavour, the feeling of uniqueness, ÿ@ÿ?complexity and idiosyncrasy, a portion of life in its most ÿ?ÿ@`real' form. To lean towards interpretation arises through ÿ@ÿAwanting, at some level, to generalise or to seek patterns of ÿAÿessence of what is being studied. It is necessarily more ÿ>ÿ/selective. (Hamilton et al, 1977, pp235-6) ÿ/ÿ ÿÿ8The approach I have taken leans towards the latter. ÿ8ÿ ÿÿ Selectivity and Subjectivity One consequence of that decision is the need to give particular ÿDÿFattention to the issue of selectivity. One chooses various modes ÿFÿ?of operation for certain reasons and for certain perceived ÿ?ÿDeffects. This does not mean that the context has been ignored; ÿDÿDrather, that boundaries have been set for pragmatic reasons and ÿDÿCthe implications of this for the study have been recognised as ÿCÿCproblematic and limiting. Such boundaries cannot, however, be ÿCÿDdrawn and enforced without privileging beliefs and interests of ÿDÿDone individual or group at the expense of others (Cherryholmes, ÿDÿE1988, p449). Increasingly I came to appreciate that integral to ÿEÿBany selection of theory, process and data was an exposure and ÿBÿCrecognition of subjective prejudice and tacit knowledge of the ÿCÿ1researcher and their influence on selection. ÿ1ÿ ÿÿ@In their first paper, where they discussed the problems and ÿ@ÿCpossibilities of illuminative evaluation, Parlett and Hamilton ÿCÿC(1972, p18) noted that concern over the `subjective' nature of ÿCÿ@the approach "is usually foremost". Although conceding the ÿ@ÿEpossibility of `gross partiality' on the part of the researcher, ÿEÿEthey contended that any research is vulnerable, with none immune ÿEÿ(to prejudice, bias and human error. ÿ(ÿ ÿÿEHabermas (1970, pp295-6), writing explicitly about hermeneutics, ÿEÿcontended: ÿÿ ÿÿC Hermeneutical understanding cannot enter into a question ÿCÿAwithout prejudice; on the contrary, it is unavoidably biased ÿAÿ@by the context in which the understanding subject has first ÿ@ÿ-acquired his schemata of interpretation. ÿ-ÿ ÿÿ ÿÿäIn his opening chapter, E p i s t e m o l o g i c a l p a r a d i g m s h i f t s : a g a m e ÿäÿf o f c h e s s, Notturno (1985) demarcated the conceptual essentials of ÿfÿEthe old and new paradigms by "a quest for objective and absolute ÿEÿCcertainty" being replaced by acceptance of fallibility. While ÿCÿCobjectivity and certainty may be aimed for, they are no longer ÿCÿ=considered to be criteria of knowledge (p3). With human ÿ=ÿ?investigations into natural phenomena, "whatever objective ÿ?ÿFknowledge is gained is thought to be both temporal and uncertain, ÿFÿDby its very nature evolutionary and subject to revision" (p4). ÿDÿBSuch a view fits comfortably with a sense of the emergent and ÿBÿ>evolutionary nature of the research. Such a decline from ÿ>ÿDabsolutism to fallibility and a recognition of subjectivity is, ÿDÿhowever, problematic. ÿÿ ÿÿCLather (1986, p66) relegated efforts to set `subjective, tacit ÿCÿ@knowledge' apart from the context of verification as `naive ÿ@ÿ?empiricism', and claimed that what she called the `rampant ÿ?ÿ@subjectivity' inherent in the more phenomenologically based ÿ@ÿAparadigms would prove "the nemesis of new paradigm research" ÿAÿ?(p68). Her concern is that research in the postpositivist ÿ?ÿEcontext "mandates a self-corrective element to prevent phenomena ÿEÿFbeing forced into preconceived interpretive schemes" (p65). Thus: ÿFÿ ÿÿE .. if we want illuminating and resonant theory grounded in ÿEÿ=trustworthy data, we must formulate self-corrective ÿ=ÿDtechniques that will check the credibility of our data and ÿDÿBminimize the distorting effect of personal bias upon the ÿBÿ"logic of evidence (p65). ÿ"ÿ ÿÿ ÿÿESo empirical validation requires a critical stance regarding the ÿEÿoccasions to the many voices that have contributed to the ÿ>ÿDresearch, searching for disconfirmation and counter-patterns as ÿDÿ>well as convergence ... these have intrinsic merit. They ÿ>ÿ@contribute to what Parlett (1977, p40) termed `recognizable ÿ@ÿBreality' which he considered to be "a major means of validity ÿBÿ>testing in illuminative studies". This does not preclude ÿ>ÿArecognition that I, as the researcher and the writer, am the ÿAÿforegrounded figure. ÿÿ ÿÿEWhen something is known, there is no such thing as a `detached', ÿEÿ=`neutral' or `objective' place to stand; we speak from a ÿ=ÿC`prejudiced' (in the sense of pre-judgement), `interested' and ÿCÿD`evaluative' posture (Bowles, 1984, p187). The anthropologist, ÿDÿFMargaret Mead (1975, p155), appreciated that "to clear one's mind ÿFÿFof pre-suppositions is a very hard thing to do and, without years ÿFÿEof practice, all but impossible when one is working in one's own ÿEÿBculture or in another that is very close to it". I could not ÿBÿ=have taught for nearly twenty years without gaining some ÿ=ÿCunderstanding of what teaching and learning entails. To strip ÿCÿEthis away, and come to the research clear from my pre-judgments, ÿEÿFwould have been impossible and would have impoverished the study. ÿFÿ ÿÿDIt is important, however, that such values do not remain covert ÿDÿDor are imposed arbitrarily. As the investigation developed, it ÿDÿ@became critical not to disclaim `subjective bias', which is ÿ@ÿEprobably inevitable, but to own it, identify it - be aware of it ÿEÿDand of its constraints and limitations. Our own viewpoints are ÿDÿEseductive - we judge favourably what represents our own sense of ÿEÿEwhat is right. I have striven to keep this in mind consistently ÿEÿDthrough this research. To provide an instance. I believe that ÿDÿ@one indicator of a successful tutorial is consistent spoken ÿ@ÿCinteraction, yet I have spoken with lecturers and students who ÿCÿ@have judged a session as highly satisfactory when that same ÿ@ÿAexperience was lecture-dominated with minimal direct student ÿAÿDspoken input. Here data and theory were not of accord. It was ÿDÿEhelpful to recognise that I held this `bias', for this led me to ÿEÿEinterrogate it; to seek perceptions of others and to explore the ÿEÿ@bases for contentment with such an approach to teaching and ÿ@ÿElearning. From this basis, the dialogue between data and theory ÿEÿbecame more productive. ÿÿ ÿÿBThus I am concerned by - and need to continue pondering - the ÿBÿCissue of validity of interpretation; the interface between the ÿCÿ?findings of the study and the reality from which they were ÿ?ÿEextracted (Walker, 1980, p45). As the person who undertakes the ÿEÿEresearch and attempts to understand and interpret, the extent to ÿEÿAwhich I understand myself, my own values and assumptions, is ÿAÿ?basic. Ricoeur (1971, pxv) wrote of his opposition to any ÿ?ÿDphilosophy based on "the transparency of the ego to itself". I ÿDÿEhave come to see that a developing self-awareness is a necessary ÿEÿCconsequence of the study and this will parallel the developing ÿCÿEawareness of the perceptions of the participants in the research ÿEÿ+and the technologies being considered. ÿ+ÿ ÿÿ Empathy ÿ(ÿ@If, as in this instance, the research is rooted in a direct ÿ@ÿDexploration of teachers' and students' experiences of learning, ÿDÿ=to explain the behaviour from the outside as a detached, ÿ=ÿCobjective observer seems far less satisfactory than to seek an ÿCÿEempathetic understanding of what is involved. When I first came ÿEÿFto appreciate that empathy with respondents was more critical for ÿFÿAthis research than to retain detachment, I felt I had made a ÿAÿFsignificant break through. I had, on a personal level, for I had ÿFÿAcome to that understanding by myself and for myself. That I ÿAÿCthought for a moment it was `new' simply bore testimony to the ÿCÿFpaucity of my reading. Much later I see how my sense of what was ÿFÿDappropriate is a discovery that many have made before me and is ÿDÿ:part of a long-standing debate. I now have a greater ÿ:ÿDappreciation of the philosophical context but the fact that the ÿDÿAinitial idea was not derivative, but grew from the research, ÿAÿEgives it a peculiar value for me. As I set it in the context of ÿEÿEhermeneutics, the idea gains substance and takes on new contours ÿEÿof richness. ÿÿ ÿÿinterviewer is an empathetic individual, a person who can ÿ>ÿunderstand him (sic). ÿÿ ÿÿ ÿÿprocess of the study under the traditional research model ÿ>ÿ ÿÿH ... defines the interaction of researched and researcher in a ÿHÿErather formal act of role-playing. This formal interaction ÿEÿFseverely constrains not only what the researcher can see but ÿFÿ;what can be effected. (Goodson and Walker, 1988, p112) ÿ;ÿ ÿÿESimilarly, Oakley (1981, p58) contended that the `mystification' ÿEÿ?of the researcher and interpretation of the researched `as ÿ?ÿDobjective instruments of data production' should be replaced by ÿDÿFthe recognition that "personal involvement is more than dangerous ÿFÿDbias - it is the condition under which people come to know each ÿDÿ1other and to admit others into their lives". ÿ1ÿ ÿÿ Empowerment ÿ<ÿEDynamic interaction between the researcher and the researched is ÿEÿAintegrally linked for writers such as Lather (1986, 1991) to ÿAÿFconcerns with research being `empowering' and `emancipatory'. To ÿFÿFMishler (1986, p119), who believes that research should be guided ÿFÿCby the aim of empowering respondents, "... to be empowered is ÿCÿFnot only to speak in one's own voice and to tell one own's story, ÿFÿDbut to apply the understanding arrived at to act in accord with ÿDÿ%one's own interests". He wrote: ÿ%ÿ ÿÿE The central question is whether and how different research ÿEÿBpractices and forms of interviewing may function to hinder or ÿBÿAto facilitate respondents' efforts to construct meaning from ÿAÿ9their experience, develop a fuller and more adequate ÿ9ÿ7understanding of their own interests, and act more ÿ7ÿ9effectively to achieve their purposes. The proposed ÿ9ÿAtransformations in research practice are intended to empower ÿAÿ?respondents by facilitating their efforts to learn and act ÿ?ÿ (p135). ÿ ÿ ÿÿ ÿÿtheir response to a number of set questions. It would be ÿ>ÿ=encouraging to think that some of the issues raised were ÿ=ÿCinfluential on their own thinking about learning and teaching, ÿCÿAbut this is simply not known. If the research had been more ÿAÿDsecurely grounded in an empowering, emancipatory model from the ÿDÿFoutset it would be, but it cannot be denied that there would have ÿFÿEbeen a cost: practical constraints would have made it impossible ÿEÿDto enter a meaningful dialogue with the several hundred partici- ÿDÿ@pants whose contribution represents the groundswell of this ÿ@ÿFstudy. Thus the contribution on that level would have been lost. ÿFÿ ÿÿ Spoken Language ÿPÿ?A "remarkable feature of twentieth century thought" is its ÿ?ÿA"convergence on the phenomenon of language" (Thompson, 1981, ÿAÿAp214). Hemeneutics has always been associated not only with ÿAÿ>interpretation, but with the interpretation of text, with ÿ>ÿ