«Andrés Gregor Zelman The University of Amsterdam 2002 ii Mediated Communication and the Evolving Science System: Mapping the Network Architecture of ...»
The scientific articles written by SOEIS members and participants for the years 1996, 1997, 1998, 1999, and 2000 were collected from the Science Citation Index (SCI), the Social Science Citation Index (SSCI), and the Arts & Humanities Index (AHI). A database containing all relevant information concerning authors, addresses, journals, and keywords was created. Secondarily the SOEIS relative articles were then used as search terms on the same databases for work that had cited them. In this way two distinct architectural dimensions of the SOEIS publication environment were defined and related. The frequency of cited and citing articles for each year of the analysis were compared, and then the most frequently citing and cited journals were compared.
In this way the architectures of each dimension of the publication domain were delimited. In addition, the cited journals were compared by country, for both the SOEIS core and its associates, and for the density of co-authorship relations.
9 By active participants I mean the percentage of the whole group which are active in the discussion.
48 The network analysis of the publication environment of the SOEIS collected and compared the referenced journals for both the cited and citing databases. By comparing the databases in a similar way to the architecture analysis above, the larger scientific domains of the cited and citing dimensions were compared for their most frequently occurring journals or indeed, disciplines. The citing and cited databases were compared for the frequency of referenced articles for 1996 through 2000, and the references cited by the core SOEIS members were delineated. Two journals arose as central to the research project and these are used in the subsequent system analysis.
The analysis of the systemic dimension explores the journal-journal relationships using the two most central journals of the SOEIS publication environment as the seed journals for the analysis, normalized over the Social Science Citation Index (SSCI).
The citing and cited dimensions were compared, paying special attention to the interface between the two. Comparison at the journal level enabled an analysis of the field level of Science, Technology & Innovation Studies. The procedure examined journal-journal citations to further specify the field level. The approach thereby provides this study with a richer perspective of the role of the SOEIS project in the larger context of the science system. Chapter VI: Analysis of Journal Publication examines publication dynamics relevant to the SOEIS research group for the years prior, during and immediately following the research project to show its architectural, network and systemic communicative patterns.
Mailing List Environment The mailing list environment refers to the use of mailing lists for enhancing academic research. Of particular interest is the EuroCon-Knowflow and Self Organization of the Information Society (SOIS) mailing lists which collectively housed the electronic communication of the Self-Organization of the European Information Society (SOEIS) research group. In addition to the Eurocon-knowflow list internal to the project, 10 other Science & Technology Studies and Self-Organization Theory oriented mailing lists were analyzed. Chapter VII: Analysis of Mailing List Environment illustrates the network distribution of these eleven STS & SelfOrganization oriented mailing lists in terms of threaded messaging behaviour. Here a distinctly Mode II type of knowledge production is addressed.
As indicated, selected Science & Technology Studies and Self-Organization Theory oriented Internet mailing lists were selected as they reflected the theoretical priorities of this study. In addition, the lists used for this study were selected on the basis of whether the message archives were publicly available. Using this criteria, eleven mailing lists were selected for the analysis: Autopoiesis, CyberUrbanity, Deukalion, ETK, EuroCon-Knowflow, Luhmann, Principia Cybernetica, Sci-Tech Studies, SimSoc, SOIS, and Xaos.10 Among these lists are project oriented lists, intermediate lists and field level lists.
The only list which functioned solely at the project level was the EuroCon-Knowflow.
However, the SOIS list can also be considered as project level as it originated as an 10 Please see Appendix D: Archive of Mailing List Analysis for the subscription data, list-server data, and archive location for each list employed in this analysis.
49 extension of the EuroCon-Knowflow list; the Autopoiesis and SimSoc lists similarly have their roots as project lists but have acquired the status of field level lists. The intermediary lists examined in the analysis are all of Greek origin; they include CyberUrbanity, Deukalion, ETK, and Xaos.11 Finally, the field level lists include the Luhmann list, Principia Cybernetica, and Sci-Tech Studies. The eleven mailing lists were observed from their dates of inception up to and including November 6, 1998.12 The general architectural features of the SOEIS mailing list environment were thereby identified.
The network dimensions of the eleven mailing lists were identified and isolated and described as levels of list participation and thread participation. The lists were then compared on the basis of whether they fall into one of three categories: project, intermediary (national), and field. The analysis then entailed a categorization of thread topics, thereby highlighting content biases. Importantly this level of the analysis reduced the amount of lists relevant for the subsequent systemic analysis.
The measure for systemic properties entailed the application of tests for selforganized criticality, thereby indicating the importance of threaded messaging to processes of knowledge production.
The theoretical lens outlined in Chapter II: Theoretical Grounding described a triadic framework upon which the empirical analyses would be contextualized, performed and interpreted. This section has served to integrate the empirical analyses and the theoretical lens through the shared interface of the Architecture – Network – System triad. The bibliometric, scientometric and cybermetric approaches outlined above each overlap in a number of unique ways. It is expected that the integration of theoretical approaches and metric methods can benefit both Science & Technology Studies and Communications Studies.
Here it is important to unpack the theoretical lens comprised of the architectural, network and systemic categorization. Media enable and constrain human communication; one can thereby assume certain biases specific to each medium. The Information Network metaphor borrowed from Medium Theory is salient here. It provides a metaphor for the architectural dimension of the SOEIS print or electronic communications as particular modes of information or processes of mediation. This is the key – the architectures revealed in the following analyses are best interpreted in light of the Medium Theory information network, which is by definition both synchronic and diachronic.
Actor Network Theory enables a complexification of this architecture by theorizing the medium as used, and mediation as a process. Observing word use is understood as integral to assessing modes of knowledge production. Importantly, the symbolic 11 This research was performed in Patras, Greece, enabling access to a several Greek Science & Technology Studies and Self-Organization Theory oriented mailing lists.
12 The Mailing List Analysis began on this date and served as the cut-off date for all lists under analysis.
50 aspects of mediated communication as theorized by Structuralist and Poststructuralist theorists and in Giddens’ Structuration Theory provide metaphors to aid in understanding the meaning of fluctuations in patterns of keyword use. The hermeneutic units of the word and it’s collocates, of journal publications, and of threaded messages are precisely where these theoretical bodies may intersect. Finally, Self-Organization Theory enables a macro view of the systemic properties of SOEIS communications.
Research Question & Expectations
Several key expectations can be identified, corresponding to the framework for the analysis as outlined above. The first expectation is that Information and Communication Technologies (ICT) affect the ways that scientists communicate, perform their research, and contribute to the production of knowledge. The reader is reminded that communication implies mediation. This position was argued by Poster (1990) with respect to changes in our dominant modes of communication, by Gibbons (et al.; 1994) as a shift from Mode I to Mode II knowledge production, and was emphasized by the OECD reports (1996, 1997) that suggested a causal relation between the changing nature of knowledge production and the advent of electronic media. This position was further established in Chapter II: Theoretical Grounding.
Bibliometric, scientometric and cybermetric analyses enable a means to assess the impact of electronic media by providing analyses that can compare print and electronic communications. A second expectation, then, is that metric analyses and mapping techniques can enhance our understanding of the overlap, and distinction between, print and electronic communication technologies in academic environments, and therefore provide insight into the dynamic character of the evolving research practice.
Finally, it is expected that the theoretical orientation of the study can be used to contextualize and interpret the results of the metric analyses, despite the different phenomenological issues raised by the difference between modelling and symbolic approaches. The theoretical lens comprised of Medium Theory, Structuralism, Poststructuralism, Structuration Theory, Actor Network Theory, and SelfOrganization Theory provides a well rounded context to frame the entire analysis. It spans a large and unwieldy intellectual tradition, yet is able to simplify their interrelation by narrowing the field of reference to mediated communication, and knowledge production, and to the triadic relationship between the architectural, network, and systemic dimensions of knowledge production.
In the context of these expectations an overarching research question can be stated:
Given that mediated communication and processes of knowledge production are mutually implicated phenomena, and that the changing information environment in academic contexts can be theorized with respect to differences between print and electronic media, can the application of metric techniques to academic communications in tandem with applied theorizing show biases particular to each medium and thereby reveal the nature of said changes?
51 The following section Part II – Analyses contains the four empirical chapters. In each
of the following chapters: Chapter IV: Analysis of Print Communication, Chapter V:
Analysis of Electronic Communication, Chapter VI: Analysis of Journal Publication, and Chapter VII: Analysis of Mailing List Environment three additional research questions are posed. The structure of each chapter follows the logic of the model: the Architectural – Network – System theoretic triad; for each additional research question posed there are corresponding expectations provided. This approach will help contain the diverse range of analyses herein, and should prove advantageous when the results are integrated in Part III – Reflection.
52 PART II – ANALYSIS
Introduction The chapter describes the computer assisted textual analysis performed upon the print communications of the Self-Organization of the European Information Society (SOEIS) research group. Key features are identified to provide an initial architecture to the document set, networks of keyword and associate relation are then revealed, and finally the systemic properties are described – each of which reflect biases of the print medium. The question of how this mode of communication differs from that of electronic writing will be addressed in the next chapter where the results of the analysis of the electronic dimension of the SOEIS communications are compared with the results of this analysis. The different roles media play in relation to each other in processes of knowledge production are thereby identified.
The textual analyses were performed upon the print output of the participants of the SOEIS research project as outlined in Chapter III: Materials & Methodologies. Word usage in processes of communication can be shown to exhibit architectural (Ong, 1982), network (Callon, 1986) and systemic properties (Leysdesdorff, 1997) through different types of analyses. Here, the distribution of word use in the print documents is examined to determine the architectural parameters of the database, the relationships between keywords for each respective time period and over the whole project for the network analysis, and the overall pattern of word use as an indicator of the systemic properties of the whole of the SOEIS print communications. In this way we identify elements particular to this mode of knowledge production. Thus, the primary aim in this chapter is to identify the dominant features of SOEIS print communication, and the secondary aim is to locate those features that may be fruitfully compared with the results of the analysis of the SOEIS electronic communications in the next chapter.
Given the centrality of the print medium to the functioning of the SOEIS research project, the analysis aimed to determine if print provided a decidedly different mode of communication than electronic media, and if this difference had any measurable impact on the general make-up of SOEIS communications. The basic hypothesis in this chapter is that an analysis of the architectural, network and systemic features of print media will reveal biases particular to the print medium itself. Only when compared with the results of the analysis the electronic dataset in the next chapter can this hypothesis be confirmed.
Several significant elements should be considered when measuring features of print communication. Importantly, the objects of this study are texts written by individuals but written in communication as collective pieces. Note that this differs from the electronic communications database analyzed in the next chapter where the collated texts are individual emails; this difference slightly changes the priorities of the