a witch doctor’s spells are causally efficacious or whether their alleged effects have other causes, there must be a set of events which correspond with the activity of the witch doctor and the individual towards whom their activity is directed.10 Recognising this is a necessary prelude to any kind of interpretive or hermeneutic activity concerned with identifying participants’ understanding of the phenomena that may follow it. Most plausibly, this involves the physical, perceptual reality of unfolding events which are independent of our perception (McNaughton 1988). However, there are also categorial concepts which all members of a society recognise and use in their understanding of their culture and social world, even if they do so implicitly. Thus, two groups may have very different views of the aims and value of their children attending school. However, they do not dispute either that school buildings exist or that there is an officially sanctioned institution of the school. Thus, although schools would not exist if there were no-one to recognise them as schools, the conception-dependence of schools (McNaughton op. cit.,) is not compromised as everyone in the society has some understanding of what a school is. By contrast, if the population disappeared there would be no concept of a school, and hence no schools, but school buildings would still exist.
We can, however, readily admit the following. Participants in practices almost invariably make use of justificatory or assertoric uses of language (Milic and Reining 2017). They talk in terms of truth and falsity of claims (‘this is only training, not education’; ‘A did cast a spell on B; B did not fall ill from natural causes’) and use criteria for determining the truth or falsity of such claims. We cannot begin to understand their practices unless we take into account their understanding of those practices. If we do not, then we run the risk of misunderstanding them. If, for example, witchcraft practices are interpreted as bad science, then we may miss a crucial analogue with practices that we recognise within our own society, namely those connected with religion.11 To some extent, it may be necessary to live within the scope of those criteria, to suspend disbelief and to act as if they were true, in order to appreciate their power and persuasiveness for those who apply them. Neither should we exclude the possibility that such a temporary acceptance of them may give us insights into those practices (and even those of our own culture or subculture) that may otherwise be denied to us. To say, however, that participants in such practices live in different realities does not follow, although it may well follow that their conception of what the world is like differs radically from that of the researcher.12
There remains the difficult question as to whose criteria for establishing truth and falsity should be dominant. We will return to this repeatedly.
PERSPECTIVALISM AS A VARIANT OF REALISM
An alternative approach might be to accept that there are multiple perspectives on a single reality but to maintain that they each offer a distinctive perspective on it, and that the underlying reality may never be known by human investigators. On the face of it, this looks like a similar stance to the one recommended above. However, it is situated in the wrong place, where truth is interpreted as ‘correspondence with reality’ – a doctrine that we have had cause to reject as an explanation of what truth is or what truth-telling practices amount to. It also leads to unacceptable consequences that lead it to collapse back into a multiple realities position.
How can this be? Different perspectives on reality will yield different accounts of that reality. In favourable situations these perspectives will be complementary and will build into a more coherent, multi-dimensional account of the underlying reality. But this benign scenario will, more often than not, fail to apply. To take two examples, A considers a practice to be witchcraft, B considers it to be bad science. C sees as practice to be nothing more than training, D considers it to have an educational element. Neither pair can both be right. If one of each pair is wrong, then perspectivalism does not work. If both of each pair is right, then according to perspectival realism each of both pairs has access to different realities, so we are back once more with multiple realities.
Nevertheless, the insight that there will be different perspectives on the same phenomena cannot be ignored without the danger of losing much of the insight that systematic enquiry can give us. The question is, ‘How to best understand multiple perspectives in a fruitful way?’
Perspectivalism and Salience
Our dealings with the world express our interests, capacities and points of view, together with our cultural inheritance. Unavoidably, out engagement is perspectival, both in the individual sense but also in a cultural one. It is our way of ‘knowing our way around’ (Wittgenstein 1953) or of ‘being in the world’ (Heidegger 1962).13
One of the primary tasks of social researchers is to explore and to understand the perspective of their subjects in order to describe accurately and in detail their dealings with the world of education, a social practice often riven with different perspectives and rival conceptions (Winch, C. 1996). This interpretive or hermeneutic aspect of educational research demands a truthful and accurate account of the perspective of research subjects if any understanding from their perspective is to be gained.
However, acknowledgement of the importance of perspective undoubtedly throws up problems for the educational researcher. These fall into two categories: the competing perspectives of research subjects and the perspective of the researcher. In a sense, the former are relatively unproblematic, since it is the researcher’s task to truthfully and accurately represent the different perspectives of the participants. The position of employers vis a vis their employees gaining qualifications will probably be different from that the employees themselves and different again from that of politicians. These may lead to fundamental disagreements of a conceptual nature: certain practices are thought to be properly educational, whereas others see them as nothing more than training. Such differences of opinion and conceptual framing may be difficult to resolve, but they can be accurately noted, provided the researcher has a good grasp of the relevant concepts as they are used by subjects in different positions.
More difficult is the position of researchers themselves. They will bring their own perspectives to bear on the practices and beliefs under research scrutiny. How do these perspectives mesh with those of the subjects of the research? Very often they will themselves be in tension or even conflict. Researchers will have their own conceptual and normative stances. They will need to be aware of these and even be prepared to revise them in the light of research data. For example, it may be the case that some of the research subjects employ conceptual distinctions which are not recognised by the researcher but which are, nevertheless not only essential for understanding the phenomena under investigation but are also useful in reframing their own understanding of the question.14 In the last analysis however, researchers will offer their own interpretation of phenomena based on their own point of view (however mediated by careful observation and interpretation) that threatens to misunderstand the practices and beliefs under study.15 The tension may well grow more acute when the researcher is offering recommendations to those researched or to those who commissioned the research, whose views may differ from those of the researcher.16
This brings us to an important point. Researchers will not adopt value-neutral stances to the practices and beliefs that they study. The best course is to acknowledge this and any possible concomitant limitations of doing so. One could argue, for example, that the main protagonists of the School Effectiveness Research Programme failed to do this and consequently left themselves open to charges of ideological bias and, perhaps more seriously, to a form of hijacking by policymakers wanting quick results (White 1997).
That said, there is no objection to a researcher adopting a normative stance (for example to what would count as a ‘good teacher’ or an ‘effective school’) and to carry out research on this basis, provided such value commitments are acknowledged.17
Unavoidable is the need for persuasion. This may take many forms (Durante 2016; Perissinotto
