|
Article Excerpt
Perspectives on technological change
The majority of contemporary perspectives towards understanding technology treat it as having objective material properties which constrain its range of possible uses. This review illustrates this claim by delineating approaches towards understanding technological change according to two theoretical undercurrents running through them; if, and where, they draw a line between the social and technical, and the extent to which they rely on structural or agency explanations. Perspectives including contingency theory (e.g. Woodward, 1965), market determinism (e.g. Bessant, Caffyn, & Gilbert, 1996), Marxist approaches (e.g. Braverman, 1974), transaction-cost economies (e.g. Ciborra, 1993) and network analyses (e.g. Barley, 1990) adopt a realist ontology, taking technology to be rigid and immutable, and social systems as malleable (Saetnan, 1991). They understand the outcomes of change as determined by exogenous structural forces, including technology, market conditions, economics and capitalism.
Although strategic choice (e.g. Child, 1972), social action (e.g. Wilkinson, 1983) and processual perspectives (e.g. Dawson, 1994) acknowledge the importance of external structures, they treat them as 'reference points for the choices, decisions and negotiating activity of key actors and groups within organizations' (McLoughlin, 1999, p. 69). Consequently, they argue to understand the outcomes of change, it is crucial to explore how both global structures and the micro-political context influence the interpretations, goals, values and motives of dominant organizational coalitions.
In contrast to the approaches above, social shaping (e.g. Mackenzie & Wajcman, 1999) and constructivist perspectives, including the social construction of technology (e.g. Pinch & Bijker, 1984), structuration approaches (e.g. DeSanctis & Scott-Poole, 1994; Orlikowski, 2000) and actor-network theory (e.g. Bloomfield & Danieli, 1995) challenge the social-technical divide by attempting to open the black box of technology. McLoughlin (1999) argues that social shaping can be differentiated from constructivist perspectives according to the direction of analysis they prescribe. The former work inwards from outside the technology, arguing it is shaped by, and comes to embody, structures such as class, gender or ethnicity. In contrast, the latter work outwards, emphasizing the interpretative flexibility of technology, and providing explanations of how, over the course of their development, technologies come to embody (DeSanctis & Scott-Poole, 1994) or have inscribed into them, the tastes, competences, motives and political prejudices of their designers (Orlikowski, 2000).
Distinct from these perspectives is the thoroughgoing interpretivist position (e.g. Grint & Woolgar, 1995). The term thoroughgoing emerged out of the authors' dissatisfaction with constructivist accounts to 'thoroughly' apply the methodological relativism of the sociology of scientific knowledge to the study of technology (see Woolgar, 1991, for discussion). Grint and Woolgar (1995) delineate perspectives towards technology into three camps according to their ontological stance: essentialist, anti-essentialist and post-essentialist. As their definition implies essentialist perspectives (e.g. contingency theory, market determinism, Marxist approaches, transaction-cost economies and network analyses) assume technologies have essential material properties which enable them to bring about specific effects in the world. They criticize these approaches for reinforcing the subject-object (or social-technical) dualism, making no attempt to open the black box of technology, and privileging structural explanations. They also take issue with anti-essentialist positions (social shaping and constructivist perspectives). Specifically, they dispute the separation between the subject and object implied by embodiment and inscription metaphors. They argue these assume that objective accounts of a technology remain when these 'evaluative aspects are stripped away from the essential object' (Grint & Woolgar, 1995, p. 289). To illustrate their point, they highlight the interpretative flexibility of the antecedent social conditions (assumptions, values, political prejudices, motives, etc.). They propose the idea that these embodied or inscribed characteristics constrain or enable action replacing technological with social and political determinism.
Instead, they adopt a post-essentialist (or thoroughgoing) understanding centred on the metaphor of technology-as-text. In contrast to essentialist and anti-essentialist perspectives it stresses that 'machines (systems) do not have inherent capacities; rather their capacity and capability is the upshot of users' interaction with the system' (Woolgar, 1994, p. 205). Technological development is 'construed as a struggle to configure (to define, enable and constrain) the user' (Grint & Woolgar 1997, p. 73). However, because technologies are interpretatively flexible, these texts are read (interpreted) differently. Therefore, an important implication of this approach is that outcomes of technological change are contingent on the effectiveness of user configuration. Well-configured users will be disciplined to use the technology as expected by its promogulators while poorly configured users will interpret it differently. As McLoughlin (1999) observes, they may resist using it, not use it at all, criticize those who do or campaign against its use entirely.
The thoroughgoing perspective has stimulated a provocative debate over the ontological status of technology (e.g. Grint & Woolgar, 1992, 1995, 1997; Kling, 1992a, 1992b; McLoughlin & Harris, 1997; Woolgar, 1991). Kling (1992b) argues that it renders technology irrelevant, posing the evocative question of whether guns have a superior capacity to maim than roses. Similarly, McLoughlin and Harris (1997) argue it is tantamount to throwing the technology baby out with the determinist bathwater. Recently, some constructivists have attempted to overcome these difficulties by refraining the subject-object and structure-agency dualisms as dualities (e.g. Harre, 2002; Hutchby, 2001; Orlikowski, 2000). For example, Hutchby (2001)proposes a 'third way' between realism and relativism whereby affordances given by technology's material substratum constrain or enable their range of possible uses. However, as Woolgar (2002) observes, this attempts to reintroduce essentialism by the backdoor and entrenches these dualities rather than challenging them.
In summary, thoroughgoing interpretivism has some significant advantages over other perspectives. First, it highlights the residual essentialism in technology studies, emphasizing how embodiment and inscription metaphors favoured by anti-essentialists reinforce the separation between the social and technical. Second, it rejects realism and relativism, adopting an ontologically agnostic position since it is impossible to transcend language and know the capabilities of technologies objectively: Epistemologically, thoroughgoing interpretivists accept the constructivist metaphor on pragmatic grounds. Methodologically, they are ambivalent towards the truth or falsity of descriptions, a position Potter refers to as 'practical scepticism' (2003, p. 799). This has the advantage that it avoids privileging one version of a technology's capabilities or effects over others. Also, it encourages researchers to be reflexive about their own 'author-texts' (Woolgar, 1991).
Despite furnishing us with a useful metaphor, Grint and Woolgar do not offer systematic methodological guidance to enable researchers to understand the rhetorical techniques used to construct the capabilities of technology and its organizational effects. Moreover, organizational discourse analysis has burgeoned in recent times, making it difficult for researchers interested in thoroughgoing interpretivism to distinguish it from other forms of discourse analysis. The following section locates thoroughgoing interpretivism within the milieu of discourse analytic approaches, and justifies the use of systematic constructionism in this study (Potter, 1996).
Introducing organizational discourse analysis
Despite the recent proliferation of discourse analytic perspectives 'in many texts, there are no definitions or discussions of what discourse means' (Alvesson & Karreman, 2000, p. 1126). Alvesson and Karreman distinguish between varieties of discourse analysis by mapping them along two dimensions. The first dimension, determinism-autonomy, captures the extent to which different perspectives consider meaning as durable, or independent of context. Foucauldian discourse analysis (e.g. Foucault, 1980) is described as lying at the determinism end of this dimension, since it treats objects and subjects as constituted from the top down by historically situated discourses with durable meanings. In contrast, they observe how Potter and Wetberall (1987) understand objects and subjects as constituted from the bottom up and treat meaning as transient and context specific.
The second dimension, close/long range, is delineated into four levels; micro-discourse, meso-discourse, Grand-Discourse and Mega-Discourse. This considers whether discourse is 'best understood as a local, context-dependent phenomenon to be studied in detail or an interest in understanding broader, more generalized vocabularies/ways of structuring the social world' (Alvesson & Karreman, 2000, p. 1129). They argue that constructionist perspectives (e.g. Potter & Wetherall, 1987) are examples of micro-discourse (with a lower case 'd') since they analyse discourse at the micro-level. Foucauldian discourse analysis, with its broader focus, is an example of mega-Discourse (with an upper case 'D').
Thoroughgoing interpretivism is located at the level of micro/meso-discourse, since it emphasizes that technology-texts 'are essentially embedded in (and, at the same time, constituted by) their interpretative contexts' (Grint & Woolgar, 1992, p.370). Attempting to place it on the discourse determinism-autonomy dimension is less helpful since this reinforces the structure-agency duality. Indeed, recent efforts to demonstrate the interplay between micro, meso, Grand and Mega levels have been described as attempting to avoid the determinism-voluntarism trap (Conrad, 2004). As outlined earlier, theorizing this duality is rejected as a false dilemma by thoroughgoing interpretivists.
As outlined above, arguably the main limitation of the technology-as-texts approach is it offers limited methodological guidance for researchers interested in analysing how technologies are constructed. In contrast, the form of discourse analysis used in this study, systematic constructionism (Potter, 1996), provides a comprehensive overview of the rhetorical strategies and linguistic resources used for factual accounting. Similar to the technology-as-texts approach, it is an ontologically agnostic, 'thoroughgoing' form of discourse analysis (Potter, 2004, p. 610). Moreover, it is also located in the range of micro/meso discourse. Potter describes it as broader than conversation analysis but more focused on the processes of fact construction than 'the Foucauldian notion of discourse...
|
