ABSTRACT
Once we have a clear understanding of what meaning is – or, rather, once we understand the linguistic function of the words ‘means’ and ‘meaning’ – the phenomenological picture described in Chapter 2 quickly unravels. In particular, the axiom of resident meaning must be abandoned. The idea that ‘there is an underlying meaning inherent in the text itself which is not directly accessible to readers, but which exegetic authority can reveal’ (Widdowson 2004: 129) has to go.1 The reason why ‘resident meaning’ is ‘hidden’ in the text has now become clear. It was never resident in the first place. It was brought in from outside. This is confirmed by an examination of the three methodological texts I have considered. All of them fail to comply with the axiom of resident meaning, even though they explicitly endorse it. Giorgi is the author who makes the biggest effort to not go beyond the text, and to ‘understand the meaning of the description based solely on what is present in the data’. But, lacking the resource of a theory on which to base ‘meaning attribution’ inferences, he is reduced to arbitrary synonyms and trivial adjustments to syntax. Unlike the other authors, he genuinely stays within the text; but, for that very reason, he can find no significant ‘meaning’. Changing ‘in the lunch group that he went out with’ to ‘with whom he went out to lunch’ is about the best he can do; and his claim that the second of these expressions conveys the ‘psychological aspect’ of the experience in a ‘heightened articulation’ stretches credulity to the limits. Giorgi sticks to the text, but engenders no meaning. In contrast, van Manen and SFL engender meaning, but only by going beyond the text. SFL have unacknowledged recourse to a theory of identity derived from the work of Mead, according to which a self ‘comes into being through social interaction with others’, and which implies that ‘no hard-and-fast line can be drawn between our own selves and the selves of others’. They look for confirmation of this theory in every line of the HIV interview, warping the text where necessary in order to accommodate it. Meanwhile, van Manen brings his own theories about parenting to the data. In the ‘being left, being abandoned’ example, he places ‘dropping your child off at school’ and ‘abandoning them permanently’ (as in Sophie’s Choice or Hansel and Gretel) in the same category; and in his analysis of Robert’s mother’s narrative, he takes a tangential reference to ‘hope’, and turns it into a lengthy, quasi-Pauline epistle. In both cases, meaning is certainly
propagated; but only by ditching the axiom of resident meaning, and implicitly appealing to external theories.2 The appeal to theory is not confined to meaning attribution in this primary sense. As was evident in the ‘Diane and Keith’ example in Chapter 6, theory determines the categories according to which items of data are classified. In identifying ‘units of information with similar content’, a theory of some kind must be invoked to delimit what will count as relevantly ‘similar’. This delimitation does not just emerge from the data itself, despite the expression ‘emergent themes’. Data/text does not somehow incorporate the relevant criteria-for-similarity which can be used to analyse it. Such criteria are no more resident in the data/text than meaning is. They have to be specified from some external perspective.3 I take it, then, that recourse to an external theory – ‘external’ in the sense that it is not intrinsic to, and cannot be derived from, the data – is inevitable in qualitative analysis. From this point of view, the most familiar way of understanding the contrast between ‘deductive’ and ‘inductive’ forms of social research does not stand up. If ‘inductive’ means that ‘theory is the outcome of research’ (Bryman 2016: 22), and if the implication of this is that no antecedent theory can be used in an inductive study, then research can never be purely inductive.4
