DA has been criticised for culminating in a nihilistic relativism (Raskin, 2001; Fletcher, 1996; Raskin, 2002). This is, if no reading or interpretation is said to be right or valid the identification of discourse in texts and language is no more than a mere academic exercise (Dixon & Durrheim, 2000). Or if all ‘readings’ and interpretations are equally valid discourse analysts cannot claim that people are ‘really’ being oppressed (see Burr, 1995). So, to this extent, on what grounds can discourse analysts promote their own versions of ‘truth’ regarding psychological phenomena such as CD? Who is to say that their version is any superior to that held by positivist psychologists? Nonetheless, Harper (2004) argued that one can still identify discourses that are ‘better’ than others despite the fact that none of them can be said to be more ‘real’ than the other. It is also interesting to notice that, sometimes DA erroneously assumes that the world can be changed simply by writing it differently (Crowe, 2005). Such assumptions often lead DA to deny the existence of a world independent from language (Burr, 2003; Willig, 2000). In this way, discourse analysts tend to reduce diagnostic labels such as ‘Conduct Disorder’ to a linguistic construction and therefore denying its real symptoms (see Parker, 1999c). Hence, this extreme relativist position makes it difficult to sustain the project of DA (Wetherell, 1998). This leads us to a problem which discourse analysts have termed the ‘So what?’ factor (see, Allen & Hardin, 2001). This refers to the difficulty in applying DA for practical purposes such as social policy (Ingleby, 2004). Nonetheless, there is a way in which DA’s problem of relativism can be solved. For instance, Ian Parker (1999a), a distinguished scholar in this field, argued that although symptoms are real, the problem arises when diagnostic labels, whilst possessing an epistemological status, acquire an ontological status (becomes linked to biological phenomena) and justifies oppressive and unjust status-quo. In this respect, we must conclude that one of DA’s strengths as a social constructionst methodology is its awareness of the socio-political and moral implications of psychological research (Parker, 1999b).
Another common charge made against discourse analysts is their strategic and political choice about which texts to analyse (Edwards, 1991a). In this way, they often held a priori assumptions that foster particular interpretations of a given text (Cheek, 2000). In this way, one may criticise discourse analysts for obscuring alternative interpretations and thus contributing towards creating another ideology (see Fairclough, 1992; Edwards, 1991b). Therefore, presenting only one of a number of interpretations is particularly problematic, given that DA challenges dominant ideologies (see Becker, 1995). Despite such criticisms, DA provides rigorous and systematic means of addressing social problems and formulating solutions for political change (Mather, 2000). Additionally, it places emphasis in the equal status between researchers and respondents (Burr, 1995). In contrast, positivist psychologists retain a relatively powerful position since that ‘the researcher’s version of events has greater warrant and is given more voice than that of the subject whose experience is interpreted and given (sometimes quite different) meanings by the researcher’ (Burr, 1995: 161). In this way, DA emphasizes the importance of giving voice to respondents and bridging the gap between researcher and those being studied. In this way the research process becomes a dialogue rather than the social scientist simply imposing an authoritative voice upon the object in question. However, this view is not shared by Sherrard (1991) who asserted that discourse analysts sometimes fail to adequately address their role in the research process when taking part in the interaction with interviewers. She insisted that ‘discourse analysts typically fail to examine explicitly their role in the production of the discourse they are analysing’ (p. 181).
It is also interesting to notice that Abrams and Hogg (1990) criticised DA regarding its political intentions. They argued that there is an implicit assumption that discourse analysts are the most qualified researchers to identify disempowered social groups. In this respect, it becomes difficult to justify discourse analysts privilege over positivist psychologists in terms of their underpinning methodological assumptions (Antaki, 2006). In this way discourse analysts, just like mainstream psychologists, risk presenting their accounts as ‘truth’ (Burr, 2003; Gough & McFadden, 2001). In this way, some argue that reflexivity may falsely ‘democratize’ the relationship between researchers and participants, since that discourse analysts ‘readings’ are the ones who tend to carry weight (Marks, 1993). Another common charge made against DA is the way in which the ‘the identification of discourses has a tendency to become little more than the labelling of everyday common-sense categories’ (Burr, 1995: 174). Thus, from this vantage point it seems that discourse analysts tend to reproduce the same categories that they are trying to challenge in the process of identification of discourses (Hare-Mustin &Marecek, 2001). Moreover, we could also criticise DA on the grounds of overfocusing on texts and disregarding other forms of discourse that are manifested in ways other than words (Widdowson, 2001; Crowe, 2005). This is, DA does not provide guidance on how to study ‘private manifestations of discourse such as thought and self-awareness.’ (Willing, 2001: 101). In this way, discourse analysts tend to privilege the role of language and texts as opposed mental states or subjectivity (as if these did not exist [Parker, 1998]). However, despite some reservations DA is widely regarded as a tool for challenging the moral and socio-political implications of positivist Psychology’s research, theory and practice (Kvale, 1995; Potter, 1992; Billing, 1989; Fox & Prilleltensky, 2001; Wetherell, 1998; Sloan, 2000)