Hardy: “Researching Organizational Discourse”
Summary
Discourse refers to a wide variety of texts both verbal and written. Many forms of discourse abound in our society relating to things as wide-ranging as management, disease, war, and immigration (Hardy, 2001, p. 26). In “Researching Organizational Discourse” (2001), Hardy defines discourse as “a system of texts that brings objects into being” (p. 26). Discourse about management can bring into being an organization, discourse about disease can bring into being an empowerment movement, and so on. He states that discourses "are embodied in texts, but exist beyond the individual texts that compose them. Consequently, texts can be considered a discursive 'unit' and a manifestation of discourse" (Chalaby in Hardy, 2001, p. 26). The texts used to bring about these “things” can include “written or spoken language, cultural artifacts, and visual representations” (Hardy, 2001, p. 26). Once the texts are produced, meaning is made between and among them in both formal and informal ways. It is through that meaning-making, or discourse analysis, that “’situations, objects of knowledge, and the social identities of and relationships between people and groups of people’” are brought into being (Fairclough and Wodka in Hardy, 2001, p. 27). Because discourse analysis is so critical to much empirical research, it is vital to understand discourse analysis and the contested areas of discourse analysis.
While there are sevel understandings and aspects of discourse analysis, Hardy provides some key thoughts on it. First, it explains how discourses are made meaningful (Alvesson in Hardy, 2002, p. 27). It also focuses on "how social reality is created through historically situated discursive moves (Alvesson and Karreman in Hardy, 2001, p. 27). Sherzer and van Dijk are cited by Hardy as asserting that we must understand the context in which discourses arise if we are to understand them (in Hardy, 2001, p. 28).
Given the above understanding of discourse analysis, Hardy’s article details four main challenges in using discourse analysis. They are: difficulties in collecting and analyzing data, studying organizational texts within their broader context, the debate between structure and agency, and the demand for more reflexive research (Hardy, 2001, p. 26). After summarizing the challenges, she gives examples of research projects of which she has been a part, and insight into how her teams have dealt with the challenges. Because Hardy and her co-researchers’ methods of dealing with the challenges represent only one view, she also gives suggestions about how other researchers might consider addressing the challenges.
Specifics
The first challenge deals with “difficulties associated with data collection and analysis and the role of theory” (Hardy, 2001, p. 29). The first part of this challenge is in the data collection. The questions a researcher must ask are what data should I collect and from whom. Answers to these questions also raise issues related to sampling (Hardy, 2001, p. 29). Even if that can be over-come, a large amount of data means a large amount of data analysis, which is second part of this challenge. Discourse analysis is “labor-intensive” (Hardy, 2001, p. 29). In the best of worlds time and financing can abate these issues. The third part of this challenge, however, is not easily addressed. The role of theory in research is highly contested. The two sides of the theory argument are “whether to conduct more theoretically informed work or to let the data drive the research” (Hardy, 2001, p. 29). Both options have their critics, and in the end Hardy advocates individual researchers making and defending their own positions.
With relation to her own work, Hardy tries to be both “pragmatic and theoretical” so that she can “make a defensible case to colleagues and reviewers” (Hardy, 2001, p. 36). The best advice from her example is to collect data from far and wide as long as it can be justified, and deal with the inevitable labor to follow. As for the issue of theory vs. data driven research, it seems that middle ground is safe, but that a preference for one form over the other is appropriate if well-defended.
The second challenge Hardy notes deals with the amount of focus placed on “text and context” (Hardy, 2001, p. 30). A heavy focus on text draws criticism that the researcher has “failed to adequately historicize their data” (Hardy, 2001, p. 30). A heavy focus on context draws criticism that the researcher is “at risk of putting words – or meaning – into the mouths of others by privileging their interpretation of the context” (Hardy, 2001, p. 30). While a balance between theory and data-driven research is logical balance, no such balance is as easily achieved here. Because there are levels of analysis with inherent preferences toward text or context, it is very difficult to achieve a logical balance.
In her own research, Hardy makes it clear that both text and context are constructed interpretations. This means that there is no “truth” to be found in either. Given that, Hardy recommends an analysis of both text and context that is wide enough to link to some general theoretical interpretation. This is called discursive pragmatism by Alvesson and Karreman; “researchers, although needing to modest in claiming empirical grounding for ‘reality,’ are interested in discursively produced outcomes in ways that allow wider interpretations” (Hardy, 2001, p. 41). It is through discursive pragmatism that researchers can deal with the criticism of focusing too heavily on text or context.
The third challenge Hardy notes deals with “the role of structure and agency” (Hardy, 2001, p. 31). This is, perhaps, the most difficult challenge to grasp. The issue of structure and agency is whether a researcher feels that there can be “explicit or implicit identification of intentional agents manipulating discourses or engaging in discursive strategies” (Burman and Parker in Hardy, 2001, p. 31). Foucault, for example, rejects the idea of identifiable agents, while Fairclough rejects the idea that there are no identifiable agents. This is basically an argument between “realists and constructivists” (Hardy, 2001, p. 31). If researchers can acknowledge that both have validity, then they can reach a middle ground. If researchers stand rigidly on one side or the other, then they will have to face the inevitable critique of their work.
Hardy gives very little space to her own research work relative to structure and agency. She seems to have a personal bent toward a more realistic, non-Foucauldian, view of structure and agency. However, she says that she and her fellow researchers are “aware of the interplay between structure and agency and would encourage researchers to address this complex issue” (Hardy, 2001, p. 41). Simple discourse analysis of this article shows that this is a challenge with which Hardy is still grappling, thus the limited discussion of the topic.
The final challenge detailed by Hardy (likely there are more not identified here) is between “’reflexive’ research, which involves reflecting on the ways in which research is carried out and understanding how the process of doing research shapes its outcomes” and “pragmatism” (Hardy, 2001, p. 32). Reflexive research is intended to bring to light the “power relations between the researcher and the researched” (Hardy, 2001, p. 32). This, however, runs the risk of focusing too much attention back on the researcher, thus giving the researcher even more power. A more pragmatic approach involves some reflection on the research and the decisions made, but relies more heavily on peer review to point out disparity.
As we’ve seen in other research practice by Hardy, she supports a more pragmatic approach. While she recognizes the importance of reflexivity, she finds it a bit too researcher-centered. For that reason, she supports involving “the broader research community” in the reflexive activity (Hardy, 2001, p. 42).
Conclusions
While these are important challenges to consider in discourse analysis, equally important is personal reflection. While Hardy provides solid examples of decision making in practice, she provides only one perspective. She is clear and we fully concur that researchers must consider these challenges based on their own epistemological and theoretical frameworks. What words for Hardy, a relatively pragmatic and realistic approach, may not work for all researchers. What we can take from this is that there are issues to be considered and criticisms to address regardless of decisions made in relation to each challenge.
Specific Considerations for Beginning "Readers" of Organizations
As we consider discourse analysis as a tool to analyze the relationships that exist in organizations, we should also consider how the “challenges” that Hardy highlights may impact our own efforts at gaining a fuller understanding of our respective organizations. Specifically, we need to consider how these “challenges” will either facilitate or hinder our research efforts and how we might move forward in light of these concerns.
First, as we consider discourse analysis, Hardy tells us that there is a challenge with data collection and analysis. This concern is multi-dimensional. We must determine the type of data we will collect and then, based on our response, determine the amount of data necessary and where we will gather it. In terms of analysis, this is more problematic. Over analyzing the data may lead to conclusions that are not defensible while simply sharing the data and allowing the consumer to reach their own conclusions may result in research that is of little value. It seems reasonable here to share one’s analysis in a manner that goes beyond interpretation and provides the reader with some insights into the data that has been collected. For the researcher, the difficulty is to reach a balance between sharing the data-driven results and providing a theoretical grounded research results. In the end, the researcher must know their audience and their own work in determining their response to this problem.
A second challenge revealed by Hardy relates to what the author refers to as “text and context.” The problem for the researcher is to strike a balance that locates and places the “text” in a perspective that considers the research setting. Conversely, the researcher who tries to share the text in its context must be concerned with going too far in claiming understanding or knowledge. Indeed, this creates the risk that the researcher is inserting their own interpretations into their research. The researcher must be constantly aware and reflective of any analysis that is interpreted by them and question its validity. Hardy recommends, and we concur, the need for balance and peer review. Again, balance and disclosure are critical for the researchers dealing with this problem.
Another challenge is what Hardy refers to as “the role of structure and agency.” As noted above, the concern relates to the manipulation of discourses or engaging in discursive strategies that may alter the researchers understanding of the organization they are researching. The problem lies with whether the researcher can ever identify all the agents that may be involved in this manipulation. As researchers, it seems appropriate to acknowledge our inadequacies at identifying all the agents that may be impacting the organizational discourses. Even when we feel we have identified several or all potential actors, it seems to be a prudent precaution to make clear that there is no way to be fully certain we have uncovered all the agents involved.
Finally, it is important as researchers to understand how our personal epistemological and theoretical frameworks will drive our research, its associated processes, and our understanding of organizations. Realizing this should encourage us to consider our own perspectives, to become more reflective research practitioners, and acknowledging these to users of our research.
In considering the four challenges that Hardy highlights in the article on organizational discourse, it seems that our challenge as researchers will be to avoid extreme perspectives in our consideration of how to approach our research. Even if we choose not to be moderate in our approach, we certainly need to be cognizant of other perspectives. Finally, it is reasonable as a researcher to be transparent about our own biases and points of view so that the consumers of our research are evaluating our findings in a manner that allows them to consider our results in light of our research perspectives.
Comments (0)
You don't have permission to comment on this page.