Does the Douglas Cultural Framework Measure Up? A Review of its Utility in Business Research

Recognition of the inherent weaknesses in the Hofstede model of cultural dimensions, and the many critiques of his approach, have led to an increasing diversity of approaches in business research towards comparative analysis of data from diverse nation-states or geo-political jurisdictions. The research question addressed in this paper is: how has the Mary Douglas Cultural Framework (DCF) been used resolve ‘the problem of culture’ in business-related research, and in accounting-related research in particular? To answer these questions requires reviewing some studies in other fields of research using the DCF, and some methodological issues raised in its application. The DCF posits ‘culture as a coercive mechanism’ to ensure sustained continuation of unique norms and beliefs within each of the four solidarities in the DCF. This study offers a review of applications of the Mary Douglas cultural framework in international business research, and the inherent problems in robustly identifying drivers to differences in business activity when a diversity of ethnicities is within the data under study. This study illustrates some applications of the DCF; I suggest in the Discussion section that it remains to be more fully tested. But some may well view the reduction down to ‘the working of human minds’ less attractive than being able to offer grand research designs and conclusions, based on eliciting responses from myriad social groups morphed into nation states, or multi-national enterprises, such as professional bodies. The limitations of the DCF remain; and reflecting on the critiques on Hofstede earlier in this paper these included: i. That culture was treated as static. In DCF the solidarities represent clusters of behavioural preferences connecting people in a specific context, but for each person their affiliation is neither stable nor permanent. Movements in and out of solitaries occur frequently and with relative ease. “In fact, temporary [solidarity] switches are virtually uncontrollable because new [solidarity] affiliations are constantly being created as people move from one context to another” (Patel and Rayner 2012 p. 128, reading ‘solidarity’ for ‘culture’). Proponents of DCF explain conformity and variations in collective action in grid-group terms. They state that any two individuals or entities with similar grid-group preferences will demonstrate similar behaviours in that context (Douglas, 1970). This conformity is neither permanent nor is it a function of people’s national origin. (Patel and Rayner 2012 p. 128) ii. That there is no such thing as national culture, and nation states only very rarely have a single and distinct culture; this critique is well accommodated in the DCF, when the focus in applying the DCF is to a social group below the size of a nation state. For example, the cohort under study could be a group of Big 4 auditors, or a Big 4 organisation itself, or members of one of the professional organisations such as ICAEW, ICAS, AICPA et ors. iii. That the ordinal measures derived from one participant can be treated as numeric and of equal intervals to achieve statistical averaging of what may well be heterogeneous components; this problem with Hofstede/Gray dimensions appears again in applications of the DCF framework as discussed above. I would not concur with Patel’s (2015) claim that “DCF offers a resolution to the unity-infinity dilemma”. It is not a framework for analysing ‘culture’ as used to denote nation states, nor the so-called corporate cultures. Those who have tried to do so have not offered persuasive results, and should not encourage others to do so. This theory or ‘Cultural framework’ needs to be clearly understood with the sense in which Douglas intends us to view cultures i.e. at the level of normative or coercive mechanisms within four solidarities. The DCF offers no explanatory power for such coercive or normative processes. The four solidarities schema is to allow classification and close examination of social relations as experienced by the individual, which then drive audit or accounting behaviours. Business research has long been dominated by the passion with which researchers seek to go further than description, and to extend their research to identify causal powers or explanations for the observed phenomena. However, it is germane to reflect on Douglas’s major contributions in other publications: her book “Purity and Danger” cited more than 20,000 times, examines in great detail the operation of rules about the unclean/clean distinctions in a variety of human social groups. Her book “How Institutions Think” is similarly influential, in which she reviews the means by which a society creates ‘public goods’, those things such as motorways or pollution controls, and worldviews that are produced and used in common, considering arguments ranging from rational-choice theory to self-sacrifice. These have been influential in many branches of anthropology, and in branches of applied anthropology, but it needs to be recalled that many consider the strength of anthropology is that it does not explain, but only describes. Therefore, those business researchers who are automatically driven from description to explanation may do so, but recognise that they may, in so doing, compromise the spirit in which the DCF framework is offered. If, in accounting research, we are to consider the DCF and the four solidarities, it needs to be in conjunction with i. gathering data on ethnic values and norms of respondents, ii. not treating any nation-state, IFRS adopter, or data from stock exchanges, as representing a culture, iii. no expectation that such analysis might offer anything more than description of coercive and normative processes which maintain norms and values, and iv. great care over the use of the term ‘culture’. Often clarity of reasoning is more likely to be gained if that word is avoided completely. The potential for the DCF as a means of identifying causal drives to distinctive accounting behaviours and practices in a variety of settings is yet to be fully realised.