Abstract
The Burrell and Morgan model for classifying organization theory is revisited through meta-theoretical analysis of the major intellectual movement to emerge in recent decades, post-structuralism and more broadly postmodernism. Proposing a retrospective paradigm for this movement, we suggest that its research can be characterized as ontologically relativist, epistemologically relationist and methodologically reflexive; this also represents research that can be termed deconstructionist in its view of human nature. When this paradigm is explored further, in terms of Burrell and Morgan’s assumptions for the ‘nature of society’, two analytical domains emerge – normative post-structural and critical post-structural. Assessing the types of research developed within them, and focusing on actor-network theory in particular, we describe how post-structural and postmodern thinking can be classified within, rather than outside, or after, the Burrell and Morgan model. Consequently we demonstrate not only that organizational knowledge stands on meta-theoretical grounds, but also how recent intellectual developments rest on a qualitatively different set of meta-theoretical assumptions than established traditions of agency and structure.
Introduction
Gibson Burrell and Gareth Morgan’s book (1979) Sociological Paradigms and Organizational Analysis represents one of the most referenced works in organization theory (OT) of the last half-century (Google Scholar suggests over six thousand citations to date). Their four-paradigm model (viz. functionalist, interpretive, radical humanist and radical structuralist) made an enormous impact on the field through defining the meta-theoretical assumptions underpinning major theoretical and methodological positions (see also Morgan, 1980, 1983, 1986). Although similar paradigm models were produced in social science prior to theirs, as Deetz (1996, p. 191) notes, ‘none have gained the almost hegemonic capacity to define the alternatives in organizational analysis’.
It can be argued, however, that in the period since Burrell and Morgan produced their model the world of OT has changed significantly. In particular, from the 1990s a philosophically ‘different’ order of analysis emerged in the guise of post-structuralism and more broadly postmodernism (see Derrida, 1981; Research in the Sociology of Organizations, 2003, 2011; Weiss, 2000). This was a movement whose theoretical characteristics appeared in stark contrast to those directed at explaining agency and structure, which for social and organizational theory have represented, historically, the two major orders of theoretical attention (Reed, 1997). Rather than seeking enlightenment through meta-theory, the emphasis was now placed on narrative, with the influence of Michel Foucault being seminal. Instead of the discrete philosophies of paradigms, it was discourses that were difficult for sociologists to avoid: they affected our views on all matters and things; they established the boundaries defining what could and could not be said – the limits of acceptable speech (Butler, 1997).
Consequently the rise of this post-structural ‘third-order’ saw the practice of defining intellectual communities as paradigms fall into decline. The historical bastion of paradigm thinking in social and organizational theory, the ‘subjective-objective’ dimension, deployed regularly in explanations of agency and structure, appeared problematic as a basis for defining rhetorical or deconstructionist theories, with this leading some to claim that the subject-object antonym had ‘now been eroded’ (Cunliffe, 2011, p. 261; see also Sveningsson & Alvesson, 2003). This latter view underscored a range of sociological inquiries into, for example, agential realism (Barad, 1998), critical realism (Fleetwood & Ackroyd, 2004), rhetorical institutionalism (Green & Li, 2011) and structuration theory (Barley & Tolbert, 1997).
However, while this trend towards synthetic or reconciliatory thinking has promoted an understanding of ‘the ontological inseparability of intra-acting agencies’ (Barad, 1998, p. 87), and thus challenged ‘individual metaphysics’, it has also been suggested that the reasoning behind it remains tentative or theoretically unresolved (Ahmed, 2008; Fraser, 2010). It can also be argued, more prosaically, that the heralded erosion of ontological and epistemological antonyms, and the associated decline in paradigm modelling, may be forces that have worked to the detriment of basic sociological explanation. We suggest that such developments may have inadvertently deprived scholars of tools for appreciating the philosophical principles upon which social science perspectives are based. While such understanding was common in the wake of Burrell and Morgan, in recent years OT researchers have become less aware of the importance of meta-theoretical distinctions for producing significant scholarship.
The aim of this paper, therefore, is to revisit the Burrell and Morgan (1979) model in order to discern whether it has explanatory value for conceptualizing the ‘post-paradigm times’ of contemporary OT. In contrast to much recent theorizing, we suggest that sociological knowledge can be explained meta-theoretically. Our goal is to reverse the decline in paradigmatic thinking through production of a model capable of conceptualizing, philosophically, the main theory orders of the field. We argue that as ‘disciplinary matrices’, comprising a ‘constellation of commitments’, paradigm communities reproduce assumptions of what represents appropriate professional behaviour, with everyday practice reflecting the influence of accepted ‘exemplars’ of research (Kuhn, 1970). This sees a number of philosophical assumptions underpin what we argue are the three main approaches to organizational theorizing of recent times – structural, anti-structural and post-structural.
We develop this analysis in four main phases. First, we introduce our basic argument and approach, explaining how a third-order paradigm can be accounted for in OT. Second, we define our use of the paradigm concept and the types of relations that exist between paradigms as organizational fields. Third, we develop a framework for conceptualizing the recent evolution of OT, one based on the comparison of meta-theoretical positions that underpin our paradigms. And fourth, we explore those third-order research domains that present themselves as a direct result of this paradigm (re)modelling. The paper culminates with a set of conclusions drawn from the investigation.
Argument and Approach
The primary task, therefore, is to extend the Burrell and Morgan framework to account for a third-order paradigm based on post-structuralism and postmodernism. To this end, we identify characteristics for this order in relation to Burrell and Morgan’s two main analytical dimensions: the ‘nature of social science’ and the ‘nature of society’. For the former, our (re)modelling suggests that much third-order OT reflects ontologically relativist, epistemologically relationist and methodologically reflexive characteristics; this also represents work that, in decentring human agency, can be deemed deconstructionist in its treatment of human nature. Expanding this analysis to consider meta-theories of the ‘nature of society’, we find two third-order research domains emerging in recent decades: normative post-structural and critical post-structural. This part of our study explains a range of contributions to these domains, with a particular focus being placed on actor-network theory (ANT), given its sociological impact in recent times, notably in Europe (Meyer & Boxenbaum, 2010; Sismondo, 2009). When this exercise is complete, we conclude that such theorizing demonstrates how a sociological paradigm of post-structuralism and postmodernism can be classified within, rather than outside, or after, the Burrell and Morgan framework.
Creating a recipe for paradigm soup?
In developing this argument we also note how in recent decades there have been many attempts to explain the rise of an analytical order of post-structural and postmodern OT. Such accounts have appeared in journal articles (e.g. Bechara & Van de Ven, 2011; Clegg & Kornberger, 2003; Donaldson, 2003; Kilduff & Mehra, 1997), research monographs (e.g. Casey, 2002; Chan, 2000; Chia, 1996) and edited volumes (e.g. Boje, Gephart, & Thatchenkery, 1995; Grey & Willmott, 2005; Linstead, 2004). In addition, textbooks have promoted a dedicated post-structuralist perspective for OT (e.g. Hancock & Tyler, 2001; Jackson & Carter, 2007) or else positioned postmodernism as one of the main perspectives to have evolved within the field (e.g. Hatch & Cunliffe, 2006). As debates on post-structuralism and postmodernism were arguably at their height during the late 1990s and early 2000s, writers have also speculated on the future for OT ‘past postmodernism’, although as yet heralding no preferred intellectual ‘heir’ (Calás & Smircich, 1999).
Within attempts to define this theoretical third order, however, one ingredient has been missing – an account whose analytical dimensions are compatible with earlier models of community structure. The widespread appeal and popularity of the Burrell and Morgan model stemmed from it offering a plausible map for exploring intellectual terrain in the social sciences. Principally, it explained the approximate location of contributions to theory and research in two-dimensional space and plotted them accordingly. This led to a proposition that the main intellectual domains of OT could be defined as belief systems, or ‘paradigms’, which interpreted the world of OT in qualitatively different, or ‘incommensurable’, ways. Indeed during the 1980s and 1990s such analysis, emphasizing a plurality of intellectual orientations, gave rise to an academic industry founded on debating issues of community structure – the so-called ‘paradigm wars’ (see Astley & Van de Ven, 1983; Deetz, 1996; Donaldson, 1985; Holland, 1990; Jackson & Carter, 1991; Lincoln, 1985; McKelvey, 1997; Pfeffer, 1997; Vogel, 2012).
With the rise of post-structuralism and postmodernism, however, attempts to define organizational theories so paradigmatically appeared to go out of fashion (see Tsoukas & Chia, 2011). Indeed a marked decline in paradigm modelling occurred from the mid-1990s, notably around the time that Deetz (1996, p. 191) suggested a new trajectory for OT – ‘re-think[ing] the Burrell and Morgan legacy’ by ‘work[ing] out the significance of the “linguistic turn” in modern philosophy for organizational analysis and practice’ (p. 193) (cf. Newton, Deetz, & Reed, 2011). In contrast to Burrell and Morgan, Deetz’s thesis accounted for the representational practices of organizational ‘discourses’, an argument directed at identifying the ‘conflicts and contradictions’ internal to ‘mobile but specifiable relations’ rather than the structure and content of communal ‘paradigms’ (Alvesson & Deetz, 2000, p. 25; cf. Goles & Hirschheim, 2000; Vogel, 2012).
The decline in paradigm thinking appeared to be predicated on an argument at the heart of Deetz’s (1996) analysis: that the research characteristics of post-structuralism could not be accounted for through deploying traditional sociological dualisms. Contributions to post-structuralism and postmodernism appeared to be premised not on the use but on the interrogation of those binary oppositions habitually deployed in paradigm structuring accounts (Tsoukas & Knudsen, 2003). It was suggested, above all, that the subjective-objective dimension (Burrell & Morgan, 1979, pp. 1–9), which accounted for explanations of agency and structure, did not make analytical sense for defining the rhetorical and reflexive theories of the new order, where the emphasis was placed instead on ‘representation’ (Deetz, 1996). Under postmodernism the human subject was neither behaviourally determined by external stimuli, nor existentially thrown into the world alive and kicking, but instead was considered philosophically decentred, or even ‘dead’ (Linstead, 2004). From the mid-1990s, the few attempts to offer paradigm-based explanations in OT, notably for classifying qualitative research (e.g. Gephart, 2004; Guba & Lincoln, 1994), reflected pale imitations of the earlier Burrell and Morgan thesis.
As a result, while philosophical thinking has remained buoyant in OT (see Research in the Sociology of Organizations, 2011), during the last decade or so scholars have referred less frequently to the importance of meta-theoretical assumptions when producing significant scholarship. One reason is that, in the current intellectual climate, defining such values for third-order research appears an antithetical or even counter-cultural enterprise, one that has seen OT researchers appear wary of assessing such theorizing alongside extant models of paradigm structure (see Cunliffe, 2011). Another is the fear of theorizing being deemed intellectually nostalgic, for in OT the rewards for neolexia appear to outweigh those for replication or retrospection (Rhodes & Pullen, 2010): neologisms abound in the field and leading journals feel a recurrent need to request ‘new theories of organization’ (Suddaby, Hardy, & Huy, 2011), rather than reflections on the ‘old’.
Furthermore, whereas OT researchers have frequently specified dedicated post-structural strategies for being, for example, reflexively critical of one’s own intellectual practices (Alvesson, 2003), or demonstrating that attempts to discover the genuine order of things are mistaken (Hardy, Phillips, & Clegg, 2001; Weick, 1999), they have been hesitant to suggest that we can capture their essence within a set of basic meta-theoretical principles. As commentators suggest that the theory structure of OT has now turned into ‘paradigm soup’ (Buchanan & Bryman, 2009, p. 4), the prosecution of a seemingly essentialist exercise has appeared not only complex and challenging but also professionally miscreant, or even ‘heretical’ (Calás & Smircich, 2003).
It is our intention, however, to go against this intellectual grain in order to assist researchers in their understanding of OT and its recent evolution. Although writers such as Calás and Smircich (2003) query the extent to which such an enterprise is sociologically valid – for it appears to engender schematization of a form that many post-structuralists would consider problematical – our view is that this is not an intellectual problem per se. Our position is that it is entirely possible to schematize theoretical positions which eschew schematization, for although certain academic communities may not warm to such an endeavour this is not a logical reason for excluding it.
Specifically we wish to complement previous philosophical analyses of organization theory by providing a new set of meta-theoretical assumptions for the field. In so doing, if not a ‘recipe’, we provide a classificatory framework for OT’s so-called paradigm soup, notably in respect of third-order analysis. In contrast to the many discrete attempts to conceptualize post-structuralism and postmodernism in OT (e.g. Chia, 2003; Hatch & Yanow, 2003; Willmott, 2003), this account sees third-order analysis defined relationally through direct meta-theoretical correspondence and comparison with the two traditional theory orders of the field: agency and structure (see Hatch & Schultz, 2000).
A quasi-essentialist method
The approach we adopt to paradigm explanation therefore is basically essentialist, in that it describes key assumptions underpinning theory orders. However, it can be more accurately described as cautiously or quasi-essentialist, in that our paradigms are defined as permeable rather than hermetic phenomena. Metaphorically they are not sovereign states but fields of influence that operate in relative tension with one another. In short, they are relational ideal-types. 1
Similar to Kilduff, Mehra and Dunn (2011), we outline organizing principles of importance to scientific communities in OT, but also include a third-order of perspectives excluded from their analysis of the production of scientific knowledge. Further, like Deetz (1996) and Cunliffe (2011), our analysis is largely retrospective, in that it draws upon earlier work in OT, specifically theorizing on paradigms, for its orientation. Unlike Deetz and Cunliffe, however, we propose that the paradigm notion remains useful for OT, not only for understanding theoretical traditions but also for generating new research methodologies. 2
We argue, further, that in revisiting the paradigm concept our thesis both respects the importance of ‘foundational’ works (see Adler, 2009) and seeks to utilize them for purposes of sociological clarification (cf. LePine & Wilcox-King, 2010). Through helping researchers to understand a range of theoretical and methodological positions we stimulate consideration of how they may be combined in field investigations (see Newton, 2010). In this respect, ours is a traditional sociological project (cf. Burrell, 1997): one that can be contrasted with those wishing to weaken or even collapse ontological and epistemological distinctions (Cunliffe, 2003, 2011) or else cultivate the more complex logics of ‘problematization’ (Locke & Golden-Biddle, 1997; Sandberg & Alvesson, 2011), thereby rendering much ‘contemporary’ (Jones & Munro, 2005) organizational analysis relatively inaccessible to empirical researchers.
Finally, rather than choose a set of arbitrary or discrete labels for characterizing recent OT paradigms, we will differentiate them by reference to a single corresponding term – structure. Although in our case transcending the thesis of Burrell and Morgan, this is akin to how Gioia and Pitre (1990, p. 588) deployed ‘the general concept of structure as a running theme’ in their analysis of multi-paradigm theory building. Within an ideal-type approach that might otherwise be comparing functionalism, interpretivism and postmodernism, we will describe contributions to structural, anti-structural and post-structural analysis, respectively. In relational terms this sees, for example, concepts of organization and organizing treated in the first paradigm as stable and objective phenomena, in the second as unstable and socially constructed, and in the third as destabilized and decentred. 3
Defining Paradigms and Paradigm Relations
We now define what we mean by the term paradigm. As noted, our analysis takes as its departure-point the original Burrell and Morgan model. However, in attempting to extend it, so as to take account of post-structural and postmodern OT, we diverge from some of the principles on which it is based. It is in this sense that we seek to develop rather than translate Burrell and Morgan.
Rethinking paradigms
Burrell and Morgan classified OT theories through a matrix based on the intersection of two axes, subjective-objective and regulation-radical change; basically a reworking of the traditional sociological dualisms of agency-structure and consensus-conflict. The resulting 2 × 2 matrix reflected four sociological paradigms: functionalist; interpretive; radical humanist; and radical structuralist. These paradigms were founded upon mutually exclusive views of the organizational world – each operated in its own right and generated its own distinctive analyses. Taken together these paradigms offered a map for locating the analytical domains of OT, one which offered a convenient means of identifying similarities among and differences between research contributions and the sociological frames of reference they adopt.
In so doing, Burrell and Morgan invoked a specific use of the term ‘paradigm’, one that functioned primarily to justify the dimensions that underpinned their 2 × 2 matrix. This culminated in a model in which their paradigms ‘define fundamentally different perspectives for the analysis of social phenomena’: they represented incommensurable phenomena which ‘generate quite difference concepts and analytical tools’ (Burrell & Morgan, 1979, p. 23). In the process, however, their thesis did not rehearse, in a direct sense, any of Thomas Kuhn’s (1962) original uses of the term paradigm in The Structure of Scientific Revolutions. Indeed it did not even invoke any of those inferred subsequently by his ‘critics’, such as the twenty-one famously identified by Margaret Masterman (1970; see also Lakatos & Musgrave, 1970).
In contrast, the model developed in this paper adheres more closely to a Kuhnian position. We return to Kuhn to understand the original meaning of the term paradigm and why it has been considered so ‘powerful’ (Vasquez, 1998). In the second edition of The Structure of Scientific Revolutions (1970, with a Postscript) on being pressed over definition, Kuhn outlined two principal uses of the concept: ‘exemplar’ and ‘disciplinary matrix’. In broad terms, the former refers to the fundamental scientific model or law, the latter to the sociological nature of the scientific community. In developing a Kuhnian account, we adopt the latter notion as a generic signifier of intellectual orientations reflected in a range of meta-theoretical positions. The focus is upon scientific fields that mirror communal philosophical assumptions and share exemplars of professional practice. These orders and their related research domains are akin to what Haas (1992, p. 3) described as ‘professional communities’, or networks of professionals possessing ‘recognised expertise and competence in a particular domain’ and an ‘authoritative claim to policy relevant knowledge within that domain or issue-area’ (see also Haas, 1989; Knorr Cetina, 1991, 2010). Such networks share normative beliefs which provide a value-based rationale for action and reflect what Vasquez (1998, p. 20) referred to as ‘the basis of classifying an aggregate of scholars as a community’.
For us, therefore, the forms of knowledge generated by a research community’s ‘normal science’ (Kuhn, 1962, 1970) largely reflect the meta-theoretical assumptions to which its network of academics adhere. We argue that deploying the notion of paradigm to reflect the main disciplinary matrices/epistemic communities of OT is a far richer use of the term than in Burrell and Morgan. Unlike Deetz (1996, 2009), who rejects the term paradigm in favour of discourse, because there are too many ‘internal debates’ within OT, we suggest that major intellectual orders can be identified as ideal-type communities for the simple reason that – ontologically, epistemologically and methodologically – they collectively share far more than they withhold. In our analysis, this sees post-structuralism and postmodernism form a generic paradigm, albeit that we recognize the variety of meanings to which these terms can refer.
Paradigms as organizational fields
In line with earlier theorizing (e.g. Lewis & Grimes, 1999; Schultz & Hatch, 1996) we adopt what might be termed a medial position on community science. Metaphorically we argue that sociological paradigms are neither ephemeral ‘clouds’ (Cunliffe, 2011; Gergen, 1992) nor impenetrable ‘citadels’ (Jackson & Carter, 1991, 2008; Kilduff et al., 2011), but instead analogous to ‘fields’ whose habitats are circumscribed by indeterminate ‘edgelands’ (see Farley & Symmons Roberts, 2011; Rapport, Wainwright, & Elwyn, 2005; Shoard, 2002). The latter represent for us the type of unorthodox intellectual spaces that Pullen & Rhodes (2009, p. 11) suggest exist beyond ‘field’ boundaries; that is, in the ‘borderlands and badlands’ of organization theory, where there is a ‘paucity of agreement’ between theory groups (see also Burrell, 2012). Thanem (2006, p. 165) offers a similar analogy in his analysis of ‘living on the edge’ of organizational analysis, in which he discusses ‘disrupting boundaries’ by inhabiting the ‘margins and peripheries’ of the field (see also Rehn & Borgerson, 2005; Thanem, 2011). In research terms, however, we suggest that metaphorically such areas represent not so much ‘boundaryless plains’ (Pullen & Rhodes, 2009, p. 11) as liminal zones between established fields of inquiry. By extension we argue that while paradigm orders reflect communities with well-established beliefs and values, their academic meta-languages are, for the most part, translatable in the process of doing ‘boundary-work’ at these intellectual margins (see Willmott, 1993; also Bloor, 1991; Laudan, 1984; Sismondo, 2005). For us, paradigms are thus incompatible rather than incommensurable phenomena, reflecting what might be termed a neo-Kuhnian position on community structure (see also Vogel, 2012). By way of a simple Venn diagram, Figure 1 describes the kinds of metaphorical relationships we have in mind.

Relational Metaphors for OT Paradigms: Organizational Fields, Liminal Zones and Tensional Forces.
Of course, based on sets of dualisms, earlier paradigm models (often in a somewhat self-contradictory way: see McKelvey, 2008) suggested that sociological paradigms are insulated phenomena locked in conflict of a dialectical kind. In contrast, our description of a triumvirate of organizational fields sees them relationally, to quote Kuhn (1977), in ‘essential tension’ with one another, rather than engaged in ‘wars’ on two fronts (see Mir & Mir, 2002; Pfeffer, 1997). Paradigm communities are constantly seeking a balance between ‘exploitation and exploration’ (Knudsen, 2003, p. 280); at once working within a framework and trying to transcend it. Thus, while communities within one research paradigm may be philosophically opposed to the theoretical or methodological practices of a second, they may also be relatively disposed to those of a third. To borrow a geological analogy, the tensional forces can be both convergent and divergent. Although the paradigm terrain remains relatively stable, constituent research communities change over time, through physically contracting and expanding, or else intellectually developing their theories and methods (cf. Thompson, 2011).
In our thesis, for example, we would argue that some forms of post-structural analysis can be relatively opposed to the nomothetic methods of a structural order yet relatively disposed to the interpretive practices of anti-structuralism. Indeed the epistemological and methodological characteristics of one paradigm may directly influence the development of another. This might see for example (anti-structural) ethnomethodology influencing (post-structural) actor-network theory. Although primarily we offer a pluralist approach, based on the differentiation of paradigm characteristics, in suggesting that our fields reflect ‘ideal-types’ (Weber, 1930) we also accept there is room for significant ambiguity and contradiction at the intellectual margins and in liminal zones. Our thesis accepts such heterodoxy in that, for example, not all post-structural work is reflexive and relativist, for some post-structuralists identify macro (social and cultural) meta-discourses and thus do not always question or undermine their own position. Similarly, some anti-structural or social constructionist research can display elements of relative, reflexive or relational analysis. The same logic applies to ideology, where arguably (post-structural) autonomism shares many political ideals with (structural) labour process theory. In short, we can identify ideal-type characteristics for a paradigm but we do not have to define that paradigm as intellectually sealed, professionally static or methodologically uniform (see also Vogel, 2012). Paradigms in social science, perceived by Kuhn (1970) as scientifically ‘immature’ phenomena, always contain the types of ‘internal debate’ of which Deetz (1996) speaks. 4
A New Meta-theoretical Framework
Having defined our theoretical position we can now describe our model. In so doing, we classify contemporary paradigms, justify their inclusion in the analysis, and populate them in terms of theory and method. The intention is not to describe the character of traditional paradigms based on agency and structure, for this has been accomplished in a plethora of post-Kuhnian works from the 1970s onwards (see Friedrichs, 1970; Gouldner, 1970; Silverman, 1970, etc.) Instead we will reflect on whether a model originally devised to account for such paradigms – Burrell and Morgan (1979) – can be extended to define the major theoretical developments of recent decades.
Assumptions about the nature of social science
We compare initially the main assumptions of our three paradigms on the criteria originally specified by Burrell and Morgan for understanding ‘assumptions about the nature of social science’; that is, ontology, epistemology, human nature and methodology (see Table 1). As the characteristics of our structural and anti-structural paradigms share common ground, respectively, with the objectivist and subjectivist antonyms originally outlined by Burrell and Morgan (1979, pp. 1–9), we will not rehearse them here. Suffice it to say that apart from some minor changes to the original Burrell and Morgan terminology – namely, replacing nomothetic with deductive, idiographic with interpretive and anti-positivist with constructionist, in order to offer more accessible or contemporary terms – these dimensions remain intact in terms of their definition in the original thesis. Instead our model is devoted to unpacking the meta-theoretical assumptions of the paradigm not accounted for in Burrell and Morgan – post-structuralism. This sees each of Burrell and Morgan’s original criteria explored in relation to the literature on third-order analysis. As noted, for the ‘nature of social science’, our argument is that post-structural OT can be classified, paradigmatically, as: ontologically relativist, epistemologically relationist, and methodologically reflexive; this also represents work that, in decentring agency, can be termed deconstructionist in accounting for human nature.
Meta-theories for OT Paradigms.
To make this argument tangible we offer, as a running example, research within a prominent third-order community: actor-network theory (ANT). This is chosen because ANT reflects contributions both to theory and method in organization studies (see Bruce & Nyland, 2011; Hardy et al., 2001; Newton, 2002; Whittle & Spicer, 2008). ANT is a particularly appropriate case to follow, given the considerable sociological attention it has received in recent years, with one monograph alone, Latour’s (2005) Reassembling the Social, selling over 20,000 copies. ANT is also appropriate in that its definition within our model serves to reflect issues of paradigm liminality. 5
Paradigm meta-theories for post-structuralism and postmodernism
We argue that developments in the field of post-structuralism and postmodernism can be explained in terms of the four meta-theories which largely underpin it as an OT paradigm (see Table 1).
Ontology
The first criterion proposed by Burrell and Morgan for assessing research philosophies is ontology (or the theory of being). In post-structural and postmodern analysis, we argue that the ontological position that is most readily inferred is relativism. Although third-order approaches often express agnosticism about the nature of reality, preferring to make epistemological rather than ontological claims, for us the concept of ontological relativism is of heuristic value (see Abbott, 2004) in that it differentiates the metaphysics of a post-structural order from those realist and nominalist ontologies that characterize, respectively, structural and anti-structural theorizing. In other words, relativism is expedient in that it sunders post-structuralism and postmodernism from absolute notions of objective truth and subjective meaning, with the work of Michel Foucault being an exemplar in this respect. It can be argued, further, that literary deconstruction, another exemplar of the paradigm, is particularly ontologically relativist, for it reflects the argument that the fundamental meaning of texts is located in their reading and appropriation. Here the relativist implication is that there can be ‘no true reading of a text’ and ‘no text apart from its reading’ (see Derrida, 1981).
If we turn to our running example, we find that actor-network theory is ontologically relativist in the assumptions it makes about the nature of social and technical forces. For example, ANT’s suppositions about the transient material-semiotic character of such phenomena see judgements about truth and falsity, good and bad, and right and wrong treated as relative to the context in question. Moreover in its early formulations (e.g. Callon, 1986, 1987) ANT appears to relativize cultural differences in assuming, somewhat controversially, that all elements in a network – human and non-human – can and should be described in similar analytical terms, or the principle of ‘generalized symmetry‘ (McLean & Hassard, 2004). With regard to the study that provided much of the inspiration for ANT, Latour and Woolgar’s (1979) Laboratory Life, we agree with Azevedo (1997) that ‘admitting the existence of reality is alone not enough to avoid relativism’. Given their basic philosophy and method, Azevedo argues ‘it is hard to see in what way Latour and Woolgar can be other than radically relativist, intentions notwithstanding’. In short ‘the “reality” of Latour and Woolgar … is not the reality of the realist’ (Azevedo, 1997, p. 71; see also Restivo & Croissant, 2008).
Furthermore, despite ‘anti-foundational’ (Cordella & Shaikh, 2006) claims that in ANT differences between elements are generated in the network of relations and thus should not be presupposed, we argue that this, itself, is an ontological position. When we consider, for example, the nature of ‘actants’ (human and non-human actors) in an actor-network, we assume that they take the shape they do by virtue of their relative interactions with one another (Latour, 1987). In early writing on ANT there appears no essential difference in the ability of technology, animals or other non-humans to act. Ontologically this assumes that we can only have experience of relative enacted alliances, for the moment an actor engages with an actor-network it too is caught up in a web of existential relationships. The early work of Bruno Latour (see Latour, 1987, 1988), for example, is premised on a form of relativism which suggests that claims are always changed (or ‘translated’) later by others, with arguments thus having no status independent of what is attributed to them relatively.
Epistemology
Although for Burrell and Morgan’s second ‘social science’ criterion, epistemology (or the theory of knowledge), third-order contributions can be accounted for under a similar banner of relativism, we feel a specific form is expedient in this respect: relationism. In providing grounds for knowledge, relationism is the theory that there are only relations between individual entities and no intrinsic properties (Emirbayer, 1997; see also Mutch, Delbridge, & Ventresca, 2006).
Given its concern to define social and technical relations, our running example of ANT reflects many of the qualities of relationism expressed in Karl Mannheim’s pioneering work in the sociology of knowledge. In his most famous book, Ideology and Utopia (1936), Mannheim followed Max Scheler (1924) in arguing for a conception of ideology that accounts for actors’ beliefs, including the social scientist’s, as a relational product of the context in which they are created. For Mannheim, to argue that knowledge is relational is to say that ‘there are spheres of thought in which it is impossible to conceive of absolute truth existing independently of the values and position of the subject and unrelated to the social context’ (Mannheim, 1936, p. 79). Put another way, this relationist epistemology suggests that the recognition of different perspectives appears arbitrary only to an abstract and disembodied theory of knowledge (Barnes, 1977; Zammito, 2007).
We contend that in its various formulations ANT exemplifies many of the assumptions of relationist epistemology. This is reflected in the treatment of multiple material-semiotic actors, or the view that the technical and social co-produce each other, with such analysis being relational both in theoretical and empirical terms. In other words, epistemologically ANT has been used to conceptualize, simultaneously, relations between (material) things and (semiotic) concepts. When such assumptions are reflected practically, in fieldwork, the interactions that researchers examine in an organization involve relations between people, ideas and technologies, which together can be understood to form a network. As Law (1999, p. 3) suggests, ‘entities take their form and acquire their attributes as a result of their relations with other entities’. Under ANT, such actor-networks are always contextual and processual phenomena: as they exist only through continuous making and remaking, it is relations that need to be repeatedly performed for such networks not to dissolve.
Human nature
The third Burrell and Morgan criterion, human nature, is a compelling one for post-structural analysis, for there appears to be no unified or autonomous subject to know and no permanent understanding of what human nature – determined, voluntary or otherwise – actually is (see Butler, 1990, 1997). Instead, central to the anti-humanism of our third-order paradigm is the view that concepts such as human nature or humanity should be rejected as historically relative or metaphysical. From this view, much structural and agentic theorizing is charged with reflecting the ‘infinitist metaphysics’ (Derrida, 1976, p. 71) of logocentric philosophies, in which human agency is founded on a core of awareness, and actions are coordinated from a knowing self, the agent acting within the context of its own dynamic presence. Under logocentrism therefore the human subject is treated as either a (structural) psychological marionette or an (anti-structural) agent of cognitive choice.
For post-structuralism and postmodernism, however, the logocentric subject of modern psychology becomes an image that is not sustainable. The grand isolation of the subject is replaced by the notion of agency as a series of transient relations between networked phenomena. The subject is no longer discretely bounded, but instead a convenient location for the throughput of discourses. Much third-order theorizing suggests that any concept of agentic presence is always mediated by one of post-human absence, and thus that consciousness is never unmediated but comes to us in an indirect way. In this view, agency is an artefact and subjectivity a process of locating identity in the language of the ‘other’ (Cooper, 1983; Spoelstra, 2005). Rather than responding mechanistically to the external environment, as per determinism, or being at the centre stage of free will, as per voluntarism, in much third-order theorizing the human subject is relationally ‘decentred’ as the locus of understanding. To account for such a position, rather than invoke an awkward neologism (such as decentrist), given the Derridean overtones and conceptual ancestry, we characterize such theorizing as reflecting a deconstructionist view on human nature.
Turning to our running example, it can be argued that much ANT research reflects such a decentred view of human agency, notably one in which the social and technical are constituted relationally through simultaneous symbolic and material systems. Under ANT the human subject often appears deprived of the logocentric authority it possessed when analytically present. In Organizing Modernity, for example, Law (1994, pp. 23–4) explains how notions of the ‘decentring of the subject’ and ‘heterogeneous materials’ inform his ‘commitment to relational materialism’, and thus how the study of the scientific laboratory at Daresbury (UK) reflects the distributed or heterogeneous character of agency. Discussing Organizing Modernity elsewhere, he suggests that ‘“an organisation”, a noun, is best not understood as an organisation, a noun, at all, but rather as a verb, that is as a process, a continuing process of movement’ (Law, 2003, p. 1). Organizing Modernity is thus a plea to move from nouns to verbs, from things to processes; specifically processes of ‘ordering’. Instead of the Daresbury laboratory representing an essential phenomenon privileging human existence and intention, Law suggests this ‘organization’ represents ‘a materially heterogeneous set of arrangement processes implicated in and implicating people, to be sure, but also including and producing documents, codes, texts, architectures and physical devices’ (Law, 2003, p. 1; emphasis in original). Law notes, however, that for social scientists such diffuse agency can sometimes be ‘difficult to take on board’, in that ‘the non-human just as much as the human may act’, a corollary being that ‘agency does not necessarily belong to people’ (Law, 2003, p. 1).
Methodology
Finally, the rise of post-structuralism and postmodernism in OT has provided significant impetus to the development of reflexive methodology (Alvesson & Sköldberg, 2000; Cunliffe, 2003). In our third-order paradigm, reflexivity is promoted through there being no ‘one best way’ of conducting theoretical or empirical research (Johnson & Duberley, 2003). The stress placed on the constructive forces of language serves to question the notion that passive, rational and scientifically neutral observers can ever account objectively for real experiences (Denzin & Lincoln, 1994). The linguistic turn of postmodernism emphasizes the often unstable, ambiguous, relational and above all context-dependent nature of language and discourse. Methodological reflexivity not only heightens our awareness of the dependence of sense-data on theory and interpretation but also bolsters the notion that scientific ‘facts’ can arise from constructed processes (see Bourdieu & Wacquant, 1992; Woolgar, 1988). Indeed, a reflexive approach to methodology has become almost de rigueur for investigations in our third-order paradigm, with several studies explaining the role of scientists and scientific institutions in the construction of research accounts (see Bruce & Nyland, 2011; Latour 1987; Law, 1994). 6
Turning to our running ANT example, we find reflexivity expressed both in sociological theorizing and accounts of organizational research. For the former, reflexivity attains perhaps its highest profile in the work of Bruno Latour and signally in his analysis of the production of scientific facts in Science in Action (1987), a work devoted as much to ontological and epistemological concerns as to the empirical study of technology. Elsewhere Latour (2003) discusses reflexivity in his debate with Ulrich Beck on ‘reflexive modernization’ (see Beck, Giddens, & Lash, 1994; Beck, Bonss, & Lau, 2003), a discussion which sees Latour (2003, p. 36) explain the ‘unintended consequences’ and ‘side effects’ of modernization, and how they ‘reverberate throughout the whole of society in such a way that they have become intractable’. Under ANT, such side effects are reflexive in that they ‘propagate through multiple, separate networks’ and in the process ‘finally become reflected … back onto what initially triggered them’ (Latour, 2003, p. 36). The end product is thus the ‘opposite of what was originally planned’, with the propagation of side-effects bringing about the ‘dis-ordering of networks created by an initial ordering action’ (Hanseth, Jacucci, Grison, & Margunn, 2006, pp. 4–5).
Closer to home, in OT, Hardy et al. (2001, p. 531) deployed ANT to ‘reflexively investigate the role of the researcher and the research community in the production of a research subject’, in this case the refugee. Using ANT concepts to ‘move beyond the more conventional institutional and discursive analyses’ adopted in their earlier investigations, this study reveals not only the actions of actors in the refugee system, but also ‘their own activities as researchers, as well as those of the broader research community’. Centrally, the concept of ‘translation’ (Callon, 1986) is deployed to explore the role of actors in the processes of social construction that produced refugees as a subject of academic study. Their project ultimately argues for a re-conceptualization of reflexivity in OT, one which moves ‘beyond the common view of heroic individuals struggling to understand and manage their role in their research’ and towards an ‘understanding of reflexivity as involving the research community as a whole’ (Hardy et al., 2001, p. 531).
Exploring Paradigms as Organizational Fields
Having discussed a paradigm of post-structuralism and postmodernism in relation to Burrell and Morgan’s dimensions for the ‘nature of social science’, we complete our revisiting of their model by considering third-order research in regard to the ‘nature of society’. In defining the political and ideological orientations underpinning OT contributions this part of their thesis reflects the traditional ‘order-conflict’ debate in social theory (see Dawe, 1970). By extending our analysis to consider this second dimension we are now able to put political flesh on the philosophical bones of our paradigms.
Assumptions about the nature of society
For assumptions about the nature of society, Burrell and Morgan suggest that researchers adhere to different perceptions of the social world depending on the often tacit political philosophies they embrace. This debate is an enduring one in social science – it contrasts perspectives illuminating the stabilizing effects of social order with those explaining the destabilizing effects of social change. However, instead of the traditional antonyms used to define this debate, such as order-conflict or consensus-coercion, Burrell and Morgan refer to differences between the ‘sociology of regulation’ and the ‘sociology of radical change’. They suggest that research adopting the former perspective explains society in terms of its underlying cohesion; through explicating for example the nature of the status quo, social order, consensus, integration and solidarity. In contrast, research reflecting the latter, ‘radical’, perspective is concerned with explaining society in terms of underlying division; for example, in relation to structural conflict, deprivation, contradiction and modes of domination. Respectively these positions reflect the ideological assumptions of the political Right and Left.
On developing our model to incorporate this second dimension, description of three paradigm fields now expands to incorporate six analytical domains. Table 2 lists these domains which reflect contributions originally identified by Burrell and Morgan plus others generated from an exercise in which third-order research was similarly assessed and classified. The result is a matrix based on a combination of a priori theoretical reasoning (y axis) and a posteriori textual interpretation (x axis). Unlike Burrell and Morgan’s extensive book-length offering, however, this analysis can only facilitate a cursory appreciation of sample contributions. To this end, we have catalogued, per domain, three examples each for theories, theorists and research. Citing a more extensive set of contributions (pre- and post-Burrell and Morgan) would represent an exercise beyond the scope of a journal article.
Typology of OT Research Domains: Examples of Theories, Theorists and Research.
Finally, to classify research in terms of underlying political and ideological assumptions, rather than deploy Burrell and Morgan’s original bifurcation of the sociologies of regulation and radical change, for contemporary OT we consider it more appropriate to differentiate between normative (Habermas, 1987; Jacobs & Hanrahan, 2005; Peters, 2005) and critical (Alvesson & Willmott, 1992, 2003; Grey & Willmott, 2005; Parker, 2002) contributions. This is perhaps compelling in the case of the latter, given the widespread use of the term as an identifier for ideologically Leftist analysis across a range of OT approaches (see Fournier & Grey, 2000; also Alvesson & Deetz, 2000; Deetz, 1996).
Research domains
Based on their political and ideological characteristics we argue therefore that our three paradigm fields reflect the following research domains: normative structural; critical structural; normative anti-structural; critical anti-structural; normative post-structural; and critical post-structural. In terms of their meta-theoretical assumptions the first four basically reflect the original Burrell and Morgan paradigms – i.e. normative structural qua functionalist; critical structural qua radical structuralist; normative anti-structural qua interpretive; critical anti-structural qua radical humanist. For these domains Table 2 offers examples not only in terms of theories originally accounted for in Burrell and Morgan (e.g. anti-organization theory; contingency theory, ethnomethodology; labour process theory) but also for those emerging in the decades since (e.g. critical discourse; institutional theory; population ecology; process theory). 7 As the political and ideological assumptions underpinning these domains have been explained in a number of Burrell and Morgan-related accounts (see Hearn & Parkin, 1983; Jennings, Perren, & Carter, 2005; Lewis & Grimes, 1999) we will not rehearse them here. 8 Instead we will focus on the (third-order) domains not accounted for in Burrell and Morgan – normative post-structural and critical post-structural.
Normative post-structural domain
Within this domain are located contributions that are recurrently classified as post-structural or postmodern but which for many commentators do not take explicit recourse to a traditional politics of the radical Left. In terms of influences on a third-order paradigm, this domain represents the analytical heartland of theorizing. It includes in particular a line of major French philosophers of the twentieth century, from Bataille to Derrida by way of Foucault. Above all, it is a tradition that wishes to dissolve the subject, attack historicism (notably historical materialism) and promote a critique of meaning. Thinking in this domain emphasizes ‘the interaction of reader and text as a productivity’, or a position where ‘reading has lost its status as a passive consumption of a product to become performance’ (Sarup, 1993, p. 3). As the influence of this tradition has been documented on many occasions – notably in relation to Foucault’s work (see Burrell, 1988; Hodgson, 2000; McKinlay & Starkey, 1998; Rowlinson & Carter, 2002) – suffice it to say that the normative post-structural domain has offered OT a bountiful critique of concepts such as causality, identity, self, subject and truth (see Boje et al., 1995; Chia, 2003; Kilduff & Mehra, 1997; Linstead, 2004; Research in the Sociology of Organizations, 2003).
As a research domain, however, this is a tradition whose exemplars have often formed a natural object for ‘radical’, notably Marxist, political and ideological criticism (see Detmer, 2003). The influential work of Foucault has represented a primary target in this respect. Although we would argue that through examination of social institutions, power and discourse, Foucault (e.g. 1970, 1974, 1976, 1979, 1980) represents an archetypal ‘critical’ thinker, we note also how in ideological terms his work has been famously denounced as ‘crypto-normativist’ by Jürgen Habermas (1984, p. 103; see also Habermas, 1987). In particular Habermas admonishes Foucault for relying on the kind of Enlightenment principles he is presumably attempting to deconstruct (see also Ingram, 1994; Owen, 1995). Habermas (1984, p. 106) argues that Foucault’s ‘stoic gaze’ results in a treatment of human history that is ‘frozen into an iceberg covered with the crystals of arbitrary formulations of discourse’. For his Marxist critics, therefore, Foucault’s work is but a mixture of ‘empirical insights and normative confusions’ (Fraser, 1981, p. 283). Arguably the key exemplar in this domain, the political charge is that his theorizing operates without the kind of substantive Marxist critique that, for Burrell and Morgan, is a cornerstone of being genuinely radical. This has seen Foucault’s work readily contrasted with critical theory, which in classical Marxist fashion attempts both to ‘understand the world and to change it’ (Ashenden & Owen, 1999; Kelly, 1994). As Sartre (1966, p. 88) once argued, this lack of a radical political critique sees Foucault’s work ultimately condemned as ‘the last rampart of the bourgeoisie’. 9
If we extend this analysis to our running example, we note that, despite being regularly described as a critical or alternative method for organizational research, ANT has frequently been denounced as sociologically normative, apolitical and even conservative (see Bowker & Star, 1999; Dolwick, 2009; Gareau, 2005; Rudy & Gareau, 2005). Such criticism has been at its strongest when directed at early ANT writing, and especially work by Michel Callon and Bruno Latour (see Winner, 1993). In OT similarly, ANT has been accused of having relatively little to say about a radical politics of a tangible kind, notably as argued by Whittle and Spicer (2008, p. 611), who wished to ‘question the contribution of ANT to developing a critical theory of organization’.
Critical post-structural domain
In contrast, theoretical research in our critical post-structural domain is oriented directly towards a radical Leftist agenda (see Peters, 2001). In the post-Burrell and Morgan period we can identify for example writing on feminism, patriarchy and the body by a number of socialist post-structuralist feminists (see Weedon, 2004 for a review). In terms of political philosophy, among the most influential writers associated with this domain are Julia Kristeva and Donna Haraway. We note however that the relationship between post-structural feminism and Marxism is a complex and often antagonistic one, as reflected for example in the work of Luce Irigaray and associates (Whitford, 1991; see also Vachhani, 2012).
Similarly in this domain, Jones (2005) has brought to our attention the work of the ‘practical Marxist-feminist-deconstructionist’ Gayatri Spivak. Focusing on her work in post-colonialism, Jones’s praise extends to describing Spivak as a ‘crucially important organization theorist’ (Jones, 2005, p. 228). Through a ‘continuous mobilization of deconstructive moves’ Spivak reminds us ‘how much of even the most critical and apparently sophisticated writing on organization today (poststructuralist or otherwise), has yet to alert itself to the task of dismantling colony’ (Jones, 2005, p. 241). Drawing upon the post-colonialist and post-structuralist writing of Homi Bhabha, Frenkel’s (2008) analysis of the discourses of international management offers a kindred exemplar for this domain.
Another theoretical perspective located here is autonomism, and specifically for OT autonomist writing associated with critical management studies (see Bohm, 2005; Fleming, 2009; Harney, 2006, 2007; Wright, 2007; cf. Rowlinson, 2008). Autonomism represents a tradition of Marxism which places at its centre the self-activity of the working class. Its most developed contemporary expression has evolved from the struggles of Italian workers, students and feminists, as formulated in the work of revolutionary intellectuals such as Franco Beradi, Sergio Bologna, Mariorosa Dalla Costa, Raniero Panzieri, Mario Tronti, and especially Antonio Negri (see Hardt & Negri, 2000). These writers/activists address debates on the prospects for a contemporary revolutionary Left and, in particular, the view that we are encountering a distinctive new era of capitalist development, ‘postmodern capitalism’. This is characterized by capitalism’s extensive deployment of information technologies in order to achieve unprecedented levels of workplace automation, societal surveillance and global mobility (see Lazzarato, 1996). Such analysis is distinctive in the insights it offers into new forms of knowledge and communication, not merely as instruments of domination, but also as potential resources for working-class struggle. In this respect the writings of Negri and his colleagues represent a significant Marxist counter-interpretation of the ‘information society’.
Finally, if we turn to our running example, we can identify recent work promoting a Leftist agenda for ANT. Like critics cited above, Alcadipani and Hassard (2010, p. 419) argue that despite recent popularity ANT appears to lack a substantive political critique, and that this is particularly apparent in its ‘“translations” in management and organization studies’. On reviewing the literature, however, they suggest the potential for such critique can be detected in writing on ‘ANT and After’ (after Law & Hassard, 1999), which ‘de-naturalizes organization(s)’, has the ability to ‘deliver critical performativity’, and at the same time ‘offer[s] a reflexive approach to management and organizational knowledge’ (Alcadipani & Hassard, 2010, p. 419). Citing a range of examples, Alcadipani and Hassard describe how this approach represents a new direction for ANT, primarily through a developing a ‘political ontology of organizing’. Other ANT works within this research domain include, for example, the study of Hinchliffe, Kearnes, Degen and Whatmore (2005) on ‘cosmopolitics’ in urban planning, and Bruce and Nyland’s (2011) analysis of the political networks of Elton Mayo and the ideology of the Human Relations School.
Conclusions
This paper has revisited the Burrell and Morgan model with the aim of re-establishing the paradigm concept for assessing intellectual developments in OT. Developing a quasi-essentialist approach, the objective has been to identify paradigmatic assumptions underpinning the major theory orders of recent decades. This ideal-type analysis has seen a new meta-theoretical framework produced for explaining intellectual relations between sociological paradigms.
In developing this argument we contend that the major paradigm to emerge in the post-Burrell and Morgan period is one based on post-structuralism and more broadly postmodernism. We argue that this paradigm is underpinned by a qualitatively different set of intellectual assumptions to those reflected in traditional sociological perspectives directed at analysing agency and structure. Despite the suggestion that developing a framework for comparing post-structural theorizing with established sociological paradigms would be ‘heresy’ (Calás & Smircich, 2003) we have taken up the challenge. In so doing we have shown how such theorizing can be included within, rather than placed outside or after, the most celebrated model for explaining OT paradigms, Burrell and Morgan.
The core of this analysis suggests that third-order developments in OT can be classified as ontologically relativist, epistemologically relationist and methodologically reflexive; this also reflects work that, in decentring human agency, can be termed deconstructionist in accounting for human nature. When this analysis is extended to consider assumptions about the ‘nature of society’ our three paradigms incorporate six research domains, viz. normative structural, critical structural, normative anti-structural, critical anti-structural, normative post-structural and critical post-structural. Focusing on those third-order domains established predominantly post-Burrell and Morgan (i.e. normative post-structural and critical post-structural), a typology of theories, theorists and research is presented to illustrate the kinds of contribution made and against which future contributions may be compared. In constructing this analysis we have offered as a running example organizational research in the field of actor-network theory, which has allowed us to substantiate our arguments in relation to method as much as theory.
Finally, if we consider not only the recent evolution but also the future of OT, we can suggest that as our field has benefitted previously from marrying pluralistic theorizing to methodology, a practical outcome of this project may be a research strategy based on paradigm triangulation. In contrast to previous methodological innovations, this would see post-structural meta-theories deployed alongside those for agency and structure as the basis for undertaking pluralistic investigations. Evolving from strategies such as multiple paradigm research, paradigm interplay and meta-triangulation, and acknowledging paradigmatically both ‘areas of overlap’ and ‘sites of contradiction’ (Okhuysen & Bonardi, 2011, p. 6), this may offer a more theoretically inclusive strategy for investigating contemporary organizational phenomena. Despite a charge that empirical research based on paradigm plurality can be superficial (or ‘Teflon-coated’: Deetz, 2009, p. 36), such an approach would help answer the call by Bechara and Van de Ven (2011) for researchers to triangulate their philosophies of science in order to understand organizational problems more holistically. This strategy would allow us, at once, to acknowledge essential differences between epistemic communities, highlight the contributions of marginalized research domains, and variously challenge the hegemonies of post-positivist discourse. In the words of Richard Rorty (1981), it would provide for a sense of ‘edifying philosophy’ in organization theory.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
