The use of citations as the determining or
principal basis for assessing the performance and potentialities of
scientists falls into the category of ‘promiscuous and careless use of
quantitative data for ... evaluation.’ That description, in the form of a
forewarning, comes, significantly enough, from...Eugene Garfield
himself.
Robert K. Merton (1977: 53)
Democracy, or rather liberal democracy, is in crises. The economic and political conditions which supported the Keynesian welfare state and the associated possibility for reform and redistributive justice evaporated by the end of the 1970s as a result of the “internationalization of capital, new means of production, and declining national growth”(Teeple, 1995: 1). As Teeple (Teeple, 1995: 1) notes:
The 1980s was a watershed decade, a turning point in the history of capitalism. It was a period that witnessed the beginning of the end of a vast system of collective or state property in the so-called socialist countries, the establishment of computer-aided modes of production and distribution, the arrival of the global economy, and the adoption around the world of neo-liberal policies whose principle was the unrestrained economic power of private property. The decade signified the beginning of what has been called the triumph of capitalism (Teeple, 1995: 1)
The ideological and political superstructure which supports this “triumph” of capitalism that Teeple refers to is known as neoliberalism. Neoliberalism refers essentially to a shift in politics and ideology initiated in response to a crises of accumulation in the advanced capitalist nations in the mid seventies and early eighties. As the shockwave of this political and ideological earthquake rolls through developed and developing nations alike, it leaves in its wake a profoundly reconfigured state, business and public sphere. Outcomes of this reconfiguration include the elimination of the post WWII capital/labour accord, the blotting out of the potential for progressive reform, the erasure of public spaces, the creation of a surveillance state to control the human problems created by the new predatorial system (Schiller, 1989; Gandy, 1993; Armstrong et. al, 1997: Clarke, 1997; McBride and Shields, 1997; Brodie, 1995; McQuaig, 1995) and the creation of a more potent hegemonic domination of the population (Kachur, 1995). Most are aware of the profound human costs of the triumph of capitalism. Yet these seem irrelevant in the current environment where global competitiveness and capital accumulation seems the guiding logic.
One of the more interesting, perhaps, outcomes of the spread of neoliberalism has been the rise of surveillance (or as I prefer panoptic) technologies and the creation of a high-tech panoptic state and economy. There are a number of factors contributing to the emergence of surveillance capitalism. On the one hand, the potential for information technologies to increase competitiveness cannot be ignored within a capitalist dynamic that requires unending competition. We see this clearly when new technologies are used to provide extremely powerful methods of gathering information about consumers habits. With the spread of plastic cash and plastic buyers discount cards, most companies can now surveille their clientele in order to determine patterns of consumption (Gandy, 1993). American Express, no slouch when it comes to making money by surveilling its clients, owns two supercomputers that enable them to “profile” and create saleable lists of consumers and their buying habits. Although this use of IT surveillance may seem benign, it violates the complexity of the social world by abstracting features relevant only to capital and victimises those who cannot participate making them non-entities in the new global system of consumption (Gandy, 1993).
Surveillance technologies are also being developed in order to control the human “waste” left by the neoliberal steamroller. Here information technologies become part of systems of discipline designed to ensure labour force discipline, (Fuller and Smith, 1991; Sewell and Wilkinson, 1992; du Gay and Salaman, 1992) and also to facilitate the state's ability to control “problem” populations (Mainprize, 1996). This surveillance discipline can follow a continuum from direct disciplinary methods (as in direct surveillance of prison populations) through towards the more subtle forms of surveillance and disciplinary control identified by Foucault (1977) in his analysis of the behavioural outcomes of internalised surveillance. Arguably, the panoptic versions of disciple and control are far more effective in controlling the population not only because they encourage a willing modification of behaviour through the internalisation of the watchful eye of authority (the superego), but also because they facilitate actual shifts in thinking. Panoptic systems thus function as systems of behavioural and ideational (hegemonic) manipulation and control.
Like most other arenas, education has not escaped the neoliberal assault unscathed. Rather than seeing universities and other educational institutions as public spaces under the tutelage of public officials and serving the public, in the current environment universities and schools are being “re-imagined” as spaces appropriate for the valorisation of capital. As a result of the neoliberal push, universities are being colonised, both physically and intellectually, by capital, its representatives, and its ideologies. This colonisation includes the importation of market discourse (Sinclair, Ironside, and Seifert, 1996; Firestone, 1994; Ball, 1993), a shift away from public funding towards private contribution, and the imposition of “market discipline” through various forms of tied financing. In all this the state as representative of capital has played a critical role in undermining the autonomy of the university by eliminating state funding. This has essentially forced the universities to move “out into the market.” As Kachur (1995: 127) notes of this marriage of convenience, business has welcomed the moves of the university.
The reality of the corporate agenda has a dual linkage. Universities are attempting to move out into the market in search of profits and corporations are attempting to penetrate the university boundaries in order to utilize university research. The window to ongoing research influences the direction of such research and gives access to marketable products based on that research. The role of the state, as perceived by the New Right and liberal technocrats, is to facilitate this new economic relationship.
As in the wider society, surveillance has become a critical component of the experience of higher education. Students can be tracked across campus with their digitised images emblazoned on plastic cards with their own personal UPC. Instructors, both in the classroom and through their research, can also be watched. Surveillance technologies in the form of class evaluations are linked with new discourses of performativity and a re-imagining of students as consumers to create a system that pushes classroom pedagogy away from critical methodologies and towards a safer, less challenging, and more entertaining variety of university education. This forced turn from critical pedagogy of course impacts on our own ability to conceive of and distribute visions of the social world which challenge neoliberal thinking.
Surveillance may play an even more important role in undermining our ability to resist neoliberalism by destroying not only our ability to create an undergraduate population capable of critically assessing their world, but also by gradually eliminating our ability to create the next generation of critical graduate students and, as time passes, by gradually eliminating the old-guard critical voices in the academy (since they won't be replaced). This is an important area of investigation and the concerns are justified. As Kachur (1995) and others have argued, intellectuals are implicated in structures of domination not only as purveyors of hegemonic domination but also, potentially, as victims of hegemonic manipulation. It is fair to ask, in the current ideological environment, whether intellectual discourse is not also coming under attack by systems of surveillance and discipline designed create and reinforce hegemony.
It is the task of this paper to attempt an answer to questions about the future of critical discourse and critical pedagogy. Our question is simple. In the context of the neoliberal drive to privatise public spaces and spread hegemonic domination over the population, do information technologies provide the potential for increased hegemonic control of the intelligentsia? This question is prefigured by Kachur's undeveloped suggestion that scholars themselves might be victims of strategies to mobilise and direct scholarly inquiry and desire through “Cybernetic feedback systems of information processing...” (Kachur, 1995: 35).
The answer to this question, as suggested below, is yes. New configurations of information technologies can provide a mechanism for control of the intelligentsia. In this paper we will explore the potential of new electronic technologies to create a cybernetic panopticon in the academy capable of extremely subtle means of disciplining scholars and controlling intellectual discourse. This closure, I will argue, will be effected through cybernetic technologies that mobilise scholarly desire (for credentials, for prestige, etc.) at the same time that they help limit avenues whereby these desires might be obtained. The net result is the limitation of scholarly discourse and the creation of a one-dimensional (after Marcuse) intellectual edifice resistant to the development of critical discourse and critical pedagogy.
In the next section we begin our analysis by examining the meaning of cybernetics and cybernetic technologies as originally envisaged by Norbert Weiner. This explication is necessary in order to lay the theoretical and metaphorical groundwork for a closer examination of surveillance technologies and their impact on individual behaviour and consciousness. As I will argue, it is my contention that a useful way to conceptualise the implications of information/surveillance technologies is with the metaphor of cybernetics.
So what are these cybernetic technologies which threaten us with an Orwellian surveillance net? Popular discourse on cybernetics characterises cybernetic technologies as a collection of information gathering, processing, and transmitting tools that empower the user of technology. The assumption is that cybernetics embodies technological prosthetics which extend the capabilities of the human organism while maintaining the individual as the Central Control Unit (CCU). 1 This is a profoundly utopian vision of technology congruent with much popular and even some academic discourses on the future of technological change. For example, Dilys E. Morris (1995: 56) follows this line when he suggests that information technology is about democratisation, enhanced participation, and freedom.
As the amount of information has grown, more and more people have come to work with it ... In most areas decision making has become considerably more participatory because the volume of information available on almost any subject can no longer be digested and understood by the limited number of individuals who once made decisions.
Extending this belief in the benign and empowering impact of technology into the realm of cybernetics, we are offered a vision of a machine enhanced human being - a cyborg (de Kerckhove, 1995) - that has capabilities far beyond the physical limits placed on our species. The vision is one of greater power and control over the environment (a masculine vision to be sure). But the vision goes beyond. Individual power and control are only the first offerings of the new cybernetic technologies. Some offer us a vision of enhanced evolutionary potential and a new bio-mechanical fusion - the next step in human evolution. 2
Like much of the current eulogising about the benefits of information technology, these popular myths about potential beneficent outcomes have it about half right. Cybernetics is about enhanced power and control over the environment. But it is not necessarily about individual democratic control, enhanced power, or even an enhanced evolutionary potential. Arguably, cybernetics is more about centralised control or, in the words of Norbert Weiner who coined the term cybernetics, command and control.
We can correct the image of cybernetics by going back to the founding patriarch of cybernetics - Norbert Weiner. Cybernetics, according to Weiner, is about control through communication. Norbert Weiner had originally been concerned with the development of automated weapons systems. He had hoped that with developing computer technology it would be possible to increase the efficiency and accuracy of military technology through the application of cybernetic principles. The principles are easy enough to understand. Weiner felt that effective control (be that of technology or even humans) required a constant flow of information not only about, as in the case of weapons systems, the target, but also about the current state of the weapon itself. Weiner felt that this information could best be provided by designing mechanisms that would sense both environmental conditions and the state of the weapon. This information would then be collated and transmitted to either a automatic device or a human being charged with destroying enemy personnel.
Weiner's conception is really quite revolutionary when you think about it. Normally information is associated with greater democracy. But here Weiner is suggesting an association between information and centralised control. Weiner himself says it best.
In giving the definition of Cybernetics...I classed communication and control together. Why did I do this? When I communicate with another person, I impart a message to him, and when he communicates back with me he returns a related message which contains information primarily accessible to him and not to me. When I control the actions of another person, I communicate a message to him, and although this message is in the imperative mood, the technique of communication does not differ from that of a message of fact. Furthermore, if my control is to be effective I must take cognizance of any messages from him which may indicate that the order is understood and has been obeyed (Weiner, 1954: 16).
There are four essential parts to a cybernetic system of command and control. There must be something to be controlled. This could be a machine, a weapon, a house (the classic example of cybernetic feedback is the home thermostat), or even another human being. The nature of the thing being controlled is unimportant because theoretically, cybernetic systems can be incorporated into anything that operates on the environment if information exchange mechanisms can be attached. In addition to having and output device, there must also be a central control unit (CCU) responsible for directing the actions of the system. There also needs to be sensing devices that gather the requisite information and finally, there needs to be a technology capable of transmitting information back and forth.
is a simplified illustration of the basic components of a cybernetic
system.
As should be evident from Diagram One above, the entire system of Command and Control is wholly dependent on information exchange. In order to be able to efficiently operate on the environment, the CCU (and not the politically neutral and deceiving CPU) must receive continuous information not only from environmental sensors but also from whatever instrument is being controlled. Weiner (1961, 26) was fairly unspecific about the types of mechanisms that would be suitable as sensors. According to him, pretty much anything that could provide useful and relevant information to a CCU , about the environment or the controlled unit, would be acceptable.
In cybernetic systems, the transfer technology (represented by the arrowed lines) passes information to and from the different components of the system. This allows the CCU to continually monitor conditions and adjust its procedures to account for changes. The transfer mechanisms are any technology which allows communication between the different components of the system. In some cybernetic systems, these would be hard wired connections. In a system that controls the climate in a house, for example, the transfer technology consists of the wires, computer interface cards, and connections that link the system together and provide information flow between the CPU, the sensors, and the actuating units (i.e. the air conditioner and the furnace). In other instances, transfer technologies could be based on radio frequencies and other technologies capable of transmitting information without wire. Its worth repeating here that the exact nature of the transfer and sensing technologies is determined pragmatically. Whatever works fits.
This same pragmatic use of technologies for sensing and communication does not apply to the CCU unit of the cybernetic system. The specifics of the command centre are much more specific. According to Weiner, “the modern ultra-rapid computing machine was in principle an ideal central nervous system to an apparatus for automatic control” (Weiner, 1961: 26-7). However according to Weiner not just any computer would do. He compares the digital computer against early analogue computing devices (never fully developed!) and concludes that for the purposes of command and control, digital computers are much better suited to the task. Weiner explains:
This all-or-nothing machine is called a digital machine. It has great advantages for the most varied problems of communication and control. In particular, the sharpness of the decision between “ yes” and “no” permits it to accumulate information in such a way as to allow us to discriminate very small differences in very large numbers (Weiner, 1954: 64).
Hopefully the passage above convinces the reader of the appropriateness of using CCU instead of CPU. Digital technologies are not neutral devices. They were specifically developed because of their suitableness for certain types of control. Hopefully to the above passage alerts the reader to the control implications of computer/information technologies in general. While many have written about the potentials of IT for enhanced control, I don't believe anyone has argued that these technologies are, at the very core, control technologies. This is significant since if this is the case, our conceptions about the nature and potential of IT for any sort of democratic or progressive reform need to be seriously rethought!
Hopefully this has also convinced the reader of the potential usefulness of developing theories of IT around metaphors of cybernetic technology. As a metaphor for thinking about the implications of technologies, cybernetics allows us to reconceptualise information technologies by allowing space for some needed critical analysis. For example, rather than seeing IT as inherently democratic, using a (corrected) cybernetic metaphor forces us to consider the possibility that information technologies are inherently undemocratic. Rather than pushing authority into the margins, IT is about centralising control. The Cybernetic metaphor also allows us to correct mistaken assumptions about the role of the “I” in Information Technology. A closer look at much current discourse on technology reveals that most people who discuss information technology focus almost exclusively on the cybernetic transfer mechanisms. It is the speed of information transfer which most commentators hold up as exemplifying the revolutionary potential of the technosphere (as if speed was the only thing that mattered!). Yet the speed of the hard and soft connections are only a part of the entire cybernetic system. Much more important to consider is the nature of the sensing devices or the strategies for command and control. A cybernetic metaphor forces our attention away from our fixation with transfer mechanisms of technology and forces a consideration of the other components of a cybernetic system - components largely ignored in most analysis of information technology.
The cybernetic metaphor is useful in other areas. For example, we can extend the metaphor to include what we might want to call social cybernetic systems. This extension moves us away from the control of house temperature and towards a theorisation of how cybernetic technologies might control humans. We define Social Cybernetics as a system of hard and soft technology, including sensing devices, transfer mechanisms, and central control units, that is designed primarily for the purpose of enhancing control over human beings. In as much as the system is about command and control, and as much as the same sorts of principles apply, a social cybernetic system will look identical (diagrammatically) as a mechanical cybernetic system.
However there is an additional component of social cybernetic systems not required of exclusively technological systems. In social cybernetics the devices which operate on the environment are human and given to having ideas of their own. Obviously this adds a complicating wrinkle. So, some mechanisms, or set of mechanisms, must therefore be put in place to enhance the possibility that command statements of the CCU will be obeyed. In order to increase the probability of getting humans to respond to the command and control dictates of the system, the social cybernetic system must also be able to mobilise systems of reward and punishment. Diagram Two below adds visual space for this additional wrinkle in the cybernetic system.
In
, the idea that humans do not automatically respond to command
statements is indicated both the addition of a punish/reward line, and also by the fact that the
nature of the information about the controlled object changes. Instead of seeking information
on the condition of the instrument, a social cybernetic system seeks information on
compliance. Depending on the state of the organism (i.e., whether compliance was achieved
or not), commands may then be supplemented by the mobilisation of reward or punishment.
The nature of the reward/punish loop would presumably be set by the current psychological
wisdom on the best methods of controlling humans and also, perhaps, by the nature of the
individual or social group being controlled. We might assume that for some cybernetic
systems, punishment would be the accepted method of ensuring compliance whereas in others
rewards would be the accepted method.
So too the nature of the measures of compliance will vary between systems. In some cases, compliance may be measured by monitoring keystrokes or phone conversations. In others by the speed of the assembly line and various statistical measures of quality control. In system where control of professional or managerial staff are to be effected, systems of monitoring feedback will likely have to be implemented in a subtle and indirect fashion to avoid the possible repercussions should individuals use to autonomy begin to realise they are being cybernetically controlled.
As noted above, thinking of information technologies as imbedded in systems of cybernetic control helps us get beyond overly simplistic views of technology because it allows us to focus on aspects of IT that are normally excluded from the analysis. A concrete example is possible at this juncture. In the sociology of work and industry, considerable attention has been devoted to examining new technologies for organising the labour process. Many of these technologies, for e.g., computer aided manufacture (CIM), Just in Time (JIT) systems of production control, flexible manufacturing systems, quality control circles, and the like, can be fruitfully examined with the cybernetic metaphor. And in fact, there seems to be a clear tendency to move towards this sort of analysis. Consider the following comments by one management theorist. Pay particular attention to the explication of the requirements of effective employee control outlined below.
What has not received enough attention is the equally dramatic but much less visible impact of the new technology on management control systems... The controlling function consists of actions and decisions managers undertake to ensure that actual results are consistent with desired results. Effective control requires three basic conditions: (1) standards that reflect the ideal outcomes, (2) information that indicated deviation between actual and standard results, and (3) corrective actions for any deviations between actual and standard results. The logic is evident that information technology facilitates effective control. In any organization, managers are concerned that resources are productively deployed, job responsibilities properly stated, and various assignments adequately coordinated. To ensure that resources are used appropriately, managers develop structures and use processes, such as planning, monitoring, and reporting, to maintain control. With the advances in information technology, comprehensive control systems based on a global or corporate view are applied today (Karake, 1992: 10, emphasis added).
The existence of a cybernetic system of control is strongly implied by the above account although here it is called a “comprehensive control system.” As the author notes, the goal of the management control system outlined above is to enhance the efficiency and performance of employees in the organisation through total control of the operational environment. Ensuring compliance with management's goals is facilitated via the powerful information gathering, sensing and collating functions of IT. The CCU (here a manager tied to a Management Information System which interprets data) plans, co-ordinates, and adjusts (disciplines) employees if cybernetic feedback indicates a deviation.
The author of the above extract characterises the new managerial systems as enabling “comprehensive control.” This is a rather broad claim. However there is considerable potential in cybernetic systems for this type of total control. Consider the ongoing elimination of middle management levels in major corporations. One of the biggest mistakes that we can make is to assume that just because organisational hierarchies are being eliminated, decision making power is being devolved to the lower employee levels. With the implementation of cybernetic systems, middle management is simply no longer needed in their traditional role as gatherers and digesters of organisational information for upper management. Management theorists explicitly recognise this development and advocate it as a preferred way of doing business because it offers a purer form of control. Karake (1992: 10) notes of the total control made possible by these new management IT systems:
The result is a wider span of control, fewer levels in the hierarchy, and lower complexity. Information technology may also lead to less formalization in organizations.... Since computer technology can warn top management of the effects of any decision... however, it enables them to take corrective action if the decision is not to their liking....we can conclude that even though information technology helps in the decentralization of the decision making process, it does so with no commensurate loss of control by top management.
Karake is not the only one to theorise systems of control. Others have chosen to extend Foucault analysis of panoptic forms of control to understand how new technologies create electronic panopticons (Burrell, 1988; Townley, 1993; du Gay and Salaman, 1992; Fuller and Smith, 1991; Sewell and Wilkinson, 1992). Yet I believe drawing from Foucault, while usefully pointing our attention at the control implications of technology, does not go far enough towards developing a theory of technology adequate to the task of unpacking the implications of new forms of IT. After all, computers and similar forms of technology where unheard of within the historical frame that Foucault analysed.
Thinking of IT in terms of cybernetic systems of control allows us, I believe, to fruitfully rethink our approach to information technologies. Above we have seen, briefly and tentatively, how such a reconceptualisation could be usefully applied in the sociology of work and industry. Of course, almost every area penetrated by information technologies could be usefully examined through this cybernetic lens. In the next section I'd like to return to the central purpose of this paper and examine the potential for information technologies to cybernate the academy of higher learning. Like the social cybernetics outlined above, the cybernation of the academy is about administrative control over employee functions. However unlike the application of cybernetic technologies in the workplace, the implications of cybernating the academy extend right into the core of our workplace and threaten the long traditions of independent and critical inquiry.
We do acknowledge the concerns of those who view citations as “brightly colored glass beads” which can be “exchanged for promotions and salary increases and intellectual reputations in the academic world.” However, we also believe that citations can be used, along with other performance indicators and measures of esteem...to help shape individual faculty productivity profiles (Cronin and Overfelt, 1994: 70-1, emphasis added).
If cybernetic systems of surveillance and control have been developed as a means for controlling labour in the work force, its seems reasonable to ask if similar systems are not also potentially available to administrators in the academy. Certainly, there has always been some surveillance of scholarly output done with the express intent of ranking scholars. Traditionally, this ranking has been done through, among other things, publication counts. And more most institutions of higher learning, this has been a critical method of deselecting individuals for entrance into tenured faculty positions. As every grad student knows these days, publication is a critical step in winning an academic position. Publications are necessary for tenure, for advancement, and for many of the other rewards available as part of academic life.
The neoliberal emphasis on performativity intersects well with this already existing emphasis on scholarly inequality. In fact, with the colonisation of the academy with market discourse (Fairclough, 1992), there is an even great emphasis placed on measuring performance. As academic institutions re-imagine themselves as business organisations, and as the state puts pressure on them to provide “output” measures, there is a concerted push for “better” ways to measure academic and institutional performance. As A.J Meadows (1990) notes, “better” is often conceived of as “quantitative.” Though his comments do not specifically apply to academic settings, he nevertheless captures the current pressures faced by all organisations and the push to adopt more quantitative measures of output.
In recent years, the pressure to justify library activities has grown. It has been particularly noticeable in the U.K. where it has been part of a wider drive for accountability in public enterprise, but the trend is worldwide. Justification for new ventures in times past has often been provided by the gut feelings of experienced librarians. Such feelings may actually be quite good indicators. Unfortunately, they are becoming increasingly unacceptable to committees. Justification, not only for new developments, but for maintenance of the status quo, is now seen as requiring systematically presented evidence preferably based on quantitative data. (Meadows, 1990).
In the past, publication counts have been seen as suitable. They are, after all, quantitative and they do reflect scholarly output (though whether they assess quality is another matter). However now, publication counts are being seen as an increasingly inadequate measure of scholarly performance. This is so especially since the emergence of the WWW and the explosion of electronic journals has created a wealth of alternative publication outlets. This is causing a “crises” in our justification schemes. As James S. Gardner (1993) Vice-President Academic and Provost of the University of Manitoba notes, because of the extremely low entry cost and the relative ease of publishing electronic material, it is much easier for individual scholars or groups of scholars to publish, without the assistance or imprimatur of traditional presses, their own material. The fear is that the increasingly free and easy distribution of scholarly information on the internet will make assessment of scholarly contribution difficult. How can we assess and individuals potential if all young scholars begin publishing their own material on the WWW either individually or in electronic journals. Without the imprimatur of the quality presses and the guarantee of peer review, how will the typical administrator judge the quality of a piece of work? And more importantly, how will we justify inequality in the academy?
This isn't as amusing as it might sound. It is a significant problem because there are only so many jobs to go around. Limited rewards require a justification for denying some individuals access to these rewards. Fortunately, there is a solution. Gardner (1993) suggests the use of post-publication measures of scholarly impact. Post-publication measures of scholarly output focus on tracking the impact of the scholar's work after the paper had been published. This solves the problem of promiscuous publication since if a measure can be found to track “impact” rather than simple publication, it will matter less were the article has been published. With post-publication measures, even research published on a personal homepage could potentially have a number assigned to it indicating its “impact.” This solves the problem of assessing the quality of material when it is not published with the imprimatur of recognised publishing houses and gives administrators their way out of the dilemma.
Gardner's call for post-publication measures is not a theoretical desiderata - it can be easily met. A well developed method already exists which can assume the role required in the new academic environment. This methodology is known as Citation Analysis (CA). CA is a post-publication measure of the scholarly importance of a published piece (normally a journal article). A Citation “IQ” for an article (and by association the scholar who wrote the article) is determined simply by counting the number of textual “citations” that authors receive for the work they publish. By counting textual citations, the assumption is that we can determine the quantitative impact of scholarly contribution. For example, the standard citation of Harry Braverman's Monopoly Capital in most labour process work would give Braverman a high Citation IQ. Because of this high CIQ, we would be safe in assuming that his work has had a major impact on the sociology of work. On the other hand, we can determine trivial work by its lack of citation (lack of use). Mundane work dies a lonely and ignominious death in the citation record. Clearly, the potential is here to solve the administrators dilemma. We can be safe in assuming that even if Braverman had published Monopoly Capital on the internet (and not through a reputable publisher), we would still be able to assess the quality of his work by assessing its post-publication impact.
We shouldn't get the idea that just because administrators are turning towards post- publication measures, that CA as such a measure is a new instrument. Far from it. CA has a long history. Its first use in the quantification of scholarly output occurred in 1927 when Paul L. K. Gross and E.M. Gross used it to help develop a list of the most important (i.e., core) primary journals in chemistry. Those journals with the most citations in the literature where considered the most important for library purchase. Their rationale for using the method in this manner was that by analysing how much journals were actually used in the formal literature, a journal hierarchy could be identified (Gross and Gross, 1927). This is a classic use of CA.
Since that time the methodology has been extended far beyond its initial use to categorise and create a journal hierarchy. It has been used to map the sciences by, for example, identifying the “cutting edge” or research front (de Solla Price, 1965) and even to separate the harder more rigorous sciences, like high energy physics, from the softer sciences - like sociology (Small and Crane, 1979). It has also been used to rank scholarly output. For example, it has been used to rank university departments (Hattie, Print and Krakowski, 1994; Cox and Catt, 1977; Rushton and Endler, 1977;. Howard, Cole and Maxwell, 1987), the scholarly output of countries (Garfield, 1993), and scholars (Watson, Gouvier and Manikam, 1989; Cole and Maxwell, 1989; Klein and Bloom, 1992).
You can probably guess that CA is a tedious process. However new information technologies promise to transform the process. As scholarly publication moves onto the WWW, the future holds the possibility of automated citation indexes and perhaps even daily citation update services delivered directly to the desktops of administrators and scholars. This indeed will be the final value added function of CA likely to bring a gleam to the eye of embattled administrators and ensure its widespread adoption. As scholarly communication moves online, it will become increasingly easy to develop powerful citation engines that index all the relevant publications on the WWW (i.e., those that are the core journals in a field). Like the WWW search engines that now automatically traverse the web gathering documents for indexing in massive databases, Citation Worms will nightly span the internet gathering citation data. Analysis engines will collate the data and Mail Daemons will send out summary packages allowing subscribers to track the ongoing impact of their own or their colleague's work.
The development of CA on the Internet is not as far off as we might think. There are already calls for the development of online citation databases capable of indexing all publications (Cameron, 1997). So two or three years seems a reasonable prediction. The first requirement for automated citation analysis, that all relevant journals be online, is already well on its way to realisation. The second requirement, that appropriate citation engines be developed, is discussed by Andrew Treloar (1996: 147).
It is possible to imagine some sort of automatic Web- traversing robot that builds up a picture of which links pointed to which documents for the purposes of citation analysis. Whether anyone will undertake this task, and what the bandwidth implications would be, are another matter all together.
Treloar expresses doubt about the technical feasibility of Citation Robots. But his doubts are now misplaced. The bandwidth certainly exists. Current internet capacity already allows for 2 or 3 dozen different WWW robots to continually traverse the web gathering and collating information. Robots performing citation counting will require much less bandwidth because they will be targeted at only two or three thousand scholarly journals, and they will only be dealing with the reference section of the document rather than the entire text. As for the time required to develop a citation robot, this is unlikely to be a difficult or time consuming task. The easiest approach would be to modify currently existing WWW robots. However even if this approach is not taken, there is now sufficient programming expertise to enable an entrepreneur to easily hire the requisite programming skills. Already, people are pointing to the future of web based citation and the potentials of online citation databases for fixing (modifying, altering, making more legitimate, etc.) scholarly discourse (Cameron, 1997).
These factors, i.e., the neoliberal need for surveillance, the push for administrative measures of performance and productivity, the growing need for post-publication measures of scholarly impact, and the growing ease and power of information technology to automate citation indexing, all point in the direction of a growing utilisation of citation analysis in the academy. The question now is, why should this be a concern? While some may criticise the academy for its clear biases and structured inequalities, most would agree with the need to provide some measure of scholarly output. The problem, however, lies not so much with the desire to “measure” output per se as it does with the intersection of neoliberalism, information technology (i.e., cybernetic potentials), and the quality of the measurement. When all these things are considered together, we see the potential for the development of a system of academic surveillance with the potential for very deep hegemonic control over scholarly discourse. In short, we see the potential to develop a cybernetic panopticon where scholars can be made perfectly visible and subsequently become perfectly disciplined.
Before I go on to develop the argument concerning the cybernation of the academy and the panoptic control over scholarly discourse, its is necessary to take a much closer look at the Citation Analysis as the sensing methodology (or device) of a cybernetic system.
As notes above, the problem with using CA an administrative (sensing) device is its poor quality. Indeed, shortly after it was first used to rank journals, its usefulness and scientific merit were questioned and undermined. The first to do so was Estelle Brodman who discounted CA as a valid methodology (Brodman, 1944). Since then it has been subjected to a steady stream of criticism which should have, you'd think, force a rejection of the methodology. Its useful to summarise this rather extensive critical literature because it gives us a good idea of the nature of the bias embedded in CA.
For example, there have been purely statistical critiques that accuse CA of providing nothing more than statistical artefacts reflective of the peculiarities of the academic population. The argument here is, given the nature of academic publication and the distribution of articles in the scholarly corpus, we'd simply expect, as a matter of course, a large number of publications would never get cited. It’s a simple outcome of the distribution of publication and not reflective of ability at all. As Per O. Seglen (1991: 635) notes:
The fact that a large fraction of the scientific literature is uncited does not mean that it is not being read, or that it does not contribute to scientific progress, it means that the total number of citations given ... is simply too small to give room for all published articles.... If nothing else, it is safe to assume that all publications exert some influence upon their authors, an influence which may become manifest in other, more heavily cited articles.
Others have pointed to the immaturity of the statistical procedures used. R. Plomp (1989: 71) admits that the advisability of using CA to evaluate scholarship is questionable because of an inadequate modelling of the underlying variance in citation scores. As Plomp notes, equally significant articles can have markedly different rates of citation. Plomp goes on to demonstrate how some (but it would be fair to suspect many) studies which assume a statistically significant difference in citation IQs are in fact statistically insignificant. At thing point we'd seem to be drawn to the conclusion that the statistical problems with CA condemn it to the status of failed method and ineffective. However the problems go deeper. It not just the measure is unreliable and invalid - though surely this is bad enough. There are biases embedded in the procedure which make it worse that useless as a measure of scholarly contribution or article impact.
Part of the bias which completely invalidates the procedure emerges as a result of the foibles of human emotions and cognition. Given the fact that scientists are human beings (despite our often vocal claims of ontological privilege), we shouldn't be surprised to find, as Mengxiong Liu (1993) did, that motivational and psychological factors play a huge role in the selection of citable sources? Liu finds that when we choose to cite a paper, our decision is often based less on the quality of the work contained in a cited piece than it is on our own desire to enhance our academic capital (Liu, 1993). So for example we might choose to cite a particular author as an indication of our desire to participate in an invisible college or as a sign of deference to our mentors. Of course the actual quality of citation does not enter in to the selection.
Perhaps we should also not be surprised to find that CA and CA scores can be manipulated. Myron Boor notes that greater citation visibility (i.e., higher citation counts) can be engineered through such practices as serial publication of segments of research, multiple publications of the same research either in identical or modified form, self-serving citations, and conspiratorial cross-referencing among authors in the know (Boor, 1982). As a graduate student I was informed by an esteemed teacher that he never read all the books and articles he had cited in his books. Rather, this instructor noted that he used citations to create the impression of authority and expertise and also to increase his own and his friends visibility. It was a simple exchange of citation, a clear manipulation of the system, and such common knowledge, apparently, that graduate students were taught about the “proper” methods of manipulating the system in advanced courses. Like the elements of citing behaviour which have nothing to do with the quality of a cited work, these political moves in the scholarly arena irrevocably confound CA.
If there are motivational and psychological factors that muddy the waters of citation analysis, and if citation counts can be engineered, might not CA also be susceptible to larger structural and political influences? The answer to this question is yes. At an international level, CA privileges North American and European scholarly discourse. That is, there is an evident eurocentric bias in citation patterns. Lea Velho found that things like the language barrier, the difficulty of obtaining access to foreign publications (journals especially), and the perceived low value (in the eyes of many scholars from developed nations) of contributions from other countries (regardless of their objective quality), tended to deflate the number of citations that scientists from developing nations received (Velho, 1986). Velho concluded that scientists in the colonising nations make negligible use of work emanating peripheral nations regardless of quality (Velho, 1986).
It is worth dwelling on this for a moment. Here we see CA reproducing the status quo. But it is a subtle reproduction is it not? Using CA to evaluate international contributions institutionalises eurocentric biases by erasing the processes behind biased citation behaviour and obscuring them with a veneer of quantitative objectivity. As a result of these barriers, any CA analysis conducted to evaluate the “international” impact of scholarly research will systematically devalue the contributions made by peripheral nations. Ironically, this devaluation and erasure mirrors and supports an already stratified and one-way flow of information (Wresch, 1996) and news (Hackett, 1991).
There are other systemic biases embedded in the CA procedures as well. Catherine Lutz has studied CA and its gender bias. She concludes that the contributions of women Anthropologists are partially erased from the citation record by the simple failure of male authors to include important contributions in key “canon setting works.” This erasure also occurs because of the differential citation rates of men's and women's articles (Lutz, 1990). Men bring a bias with them. Simply because they are men, they cite women less often. And when we measure the contribution of women with citation analysis, we get lower indicators of contribution. Women are thus erased from the CA record. The implications are clear - at least in Anthropology. Using CA to measure scholarly output privileges men and institutionalises and objectifies politically informed citation choices.
This erasure occurs in sociology as well. Kathryn B. Ward and Linda Grant (1985). examined 10 major sociology journals between 1974 and 1983. Using a typology that distinguished articles on the basis of there being simple additions to sociology (so called sex as variable studies examining gender differences), modifications (conceptual, theoretical or methodological refinements), or recasts (attempts to recast extant theoretical frames that are clearly androcentric and deficient for understanding women's experiences), the authors found evidence to suggest that the more radical the challenge to existing theoretical orientations, the more likely the contribution is ignored and ghettoised. The authors note:
Recase [sic] articles were even more likely to be ghettoized and ignored in citations of later published works. In suggesting that research on gender and women move away from reliance on males as the norm, these articles raised the most fundamental challenges to traditional sociological approaches. Citations were notably absent from “state of the art” or theoretical synthesis papers published in later volumes of journals in which the recasts appeared. Recast papers were not criticized or refuted by authors working in substantive fields where they were relevant; they simply were ignored (Ward and Grant, 1985: 152).
Whatever the explanation for this phenomenon, the result is predictable. More significant contributions, and especially those which challenge scholarly canon, may be partially or wholly erased from the citation record. There are others ways which CA works to obscure scholarly contribution, reinforce and legitimate scholarly contribution. Again, these are biases structured into the measure. As Fran Collyer notes (1998).
... CA obscures social inequalities by assuming all the work published is 'the best work' but this hides the fact that while women may write prolifically, their work is not given an equal chance for publication. The evidence for this is found in the publication and rejection rates of journals. In the Australian and New Zealand Journal of Sociology (which has just changed its name to Sociology) we have been keeping an eye on the figures for some time. Women have a higher rejection rate than men ... so even though more papers come from women (reflecting the greater number of women sociologists in Australia) more men's papers are published. It was decided (at the national conference when we last discussed it) that this in part reflects the fact that women are generally of lower professional rank than men, but also reflects the fact that the papers are usually reviewed by men and women's papers are of less interest to them because they reflect women's particular interests and because the men are unaware of their work and of the importance of the issues raised (yes I know reviewing is supposed to be anonymous, but that doesn't rule out the value of personal contact with a person or with a subject area and the probability that if you are familiar with either you are more likely to give a positive response).
So the hierarchical structure of the academy, with most women at the bottom ranks means that their papers are less likely to be published than those on the top, who are largely men.
Clearly there are political implications to the use of CA. From its statistical failings to its tendency to erase the contributions of women and minorities, CA would seem to be a largely unreliable and invalid tool for assessing scholarly impact. You would think, given this, that most would gladly put the use of the measure to one side. This however appears hopeless fantasy. Even though proponents of CA often recognize its profound limitations, they defend it. And the contortions and desperate clutches at respectability that have emerged can be both amusing and disturbing.
For example, Stephen Cole and Jonathan R. Cole argue that a better indicator has not been developed and since the measure has proved useful to some, why not use it? (Cole and Cole, 1987: 345) But the question remains: no better indication of what? Of scientific eminence? Of scholarly contribution? Of core journal lists? CA is worse that useless for these purposes because, as numerous critics have pointed out, it does not fairly represent the entire scholarly corpus. CA tends to support a particularly one-sided, reified, and elitist view of scientific contributions that ignores (i.e., fails to make visible) certain groups of scholars and by doing so justifies the highly stratified nature of the academy where certain groups are privileged over others. If the indicator is invalid or far from ideal, then that should be the end of the story. You would not find a technician charged with the calibration of a fine piece of equipment holding so lackadaisical an attitude towards the cannons of validity and reliability. Especially if an error in calibration were going to cost somebody a great deal of time and money. And should an error occur, an “its the best I could do” would not suffice to discharge the technician of responsibility. In any case there are better ways to assess the communication system in science that provide a truer representation of the underlying processes. These are the historiographic methods often pointed to by opponents of CA and that, when considered with an eye not blinded by the sterile application of the cannon that suggests that quantitative methods are always superior to qualitative methods, are clearly better suited to determining the dynamics of the scholarly enterprise.
Still, this isn't enough to put to rest the arguments for CA. The last ditch defence is to claim that CA is valid because it correlates with other measures of scientific eminence like the Nobel Prize, other large awards, e.g. the allocation of research funding, membership in major academies such as Royal Society or National Academy of Science, or the use of a scientist by Government or on advisory committees. This criterion validity is taken by some to be the final and irrefutable indication that the methodology is valid. Laura M. Baird and Charles Oppenheim make this argument (Baird and Oppenheim, 1994: 8):
All of these measures, whether objective ... or subjective ... can be measured. What is embarrassing for the critics of citation counting is this fact: whatever measure you take for the eminence of an individual scientists or of a journal or an institution, citation counts provide strong correlations with that result. This must be very frustrating for the people who criticise citation counting...So, despite the many valid criticisms of the crudity of citation counting, the fact is that they reasonably reflect the esteem that a particular author or paper enjoys.
Suffice it to remind ourselves that correlation does not necessarily indicate causation and that covariance of another, invisible variable may be at the root. A factor that immediately comes to mind as one explanation is the author's location in relation to the scholarly mainstream and orthodoxy. Surely it is not unreasonable to suggest that authors that receive all these wonderful awards are members of the academic orthodoxy and are being rewarded as members of that orthodoxy? Might we not simply suggest that CA measures establishment science as identified by critiques down through the years (Bukharin, 1925; Jacob, 1988; Haraway, 1986; Bleier, 1986)?
I believe we might - and should. In fact, making this suggestion is the key to unlocking the potential threat of CA. CA is less about measuring scholarly impact - even advocates admit it does this poorly - and more about justifying various forms of inequality in the academy. As the reader may or may not be aware, class background and serendipity, rather than innate ability, play a major role in determining how fast and far one may advance scientific advance. Very little empirical evidence is available to support a naturalised view of scientific output/intelligence. For example, in the context of scholarly genius, Dean Keith Simonton (1991) notes that even the highest grades of genius are imbedded in a social matrix which determines ability and eminence. As he notes, the highest grades of genius are the result of the highest quality of social contact. “...outstanding contributors are more likely to form part of a complex network of relations both contemporaneous and cross- generational, both personal and impersonal, both symmetric and asymmetric.” (Simonton, 1991: 461). As Hilary Rose notes of the autobiographical accounts of women who have “made it”, “reading across these autobiographical accounts typically shows a highly privileged class origin and the unusual support and encouragement of a scientist father or husband” (Rose, 1986: 61). This notion that genius (and ability) is a product of external influences and opportunities is echoed by Robert Merton's analysis of institutionalised serendipity and the Mathew Effect (Merton, 1977), by Margaret W. Rossiter's Matilda Effect (Rossiter, 1993) and by Bourdieu's explication of cultural capital (Bourdieu, 1984).
Still, its important for a variety of reasons that these views on stratification and inequality not be widely disseminated. So, in order that we might justify academic hierarchy, the contextual nature of academic eminence needs to be obscured. This is done in two parts. In part one, individuals are decontextualised and stripped of their historicity and embeddedness in a class, gender, and racially stratified system of rewards. In part two, the decontextualised individual is recreated as a “great man” or mediocre peon inside a suitable discursive framework. When the discursive work is done, all we see is a naturalised and justified hierarchy.
Part one of the obsfucation is accomplished via CA and similar measures. As we have seen, CA naturally extracts those individuals most embedded in the scholarly status quo. Because CA is just a number, all the historical and social complexity that went into creating an eminent individual of science is lost. This violent abstraction is of course a convenient by product of the “objectivity” of numerical indicators. Yet it is no great loss. In fact to some, it is an improvement. Not only does CA naturalise the scholarly status quo by decontextualising scholarly eminence, but it does so in a particularly powerful manner. CA participates in the scholarly canon that gives discursive authority to “neutral” and “objective” methodologies, i.e., quantitative indicators. Quantitative indicators, or the facts, perform a duty as arbiters of ultimate reality. And they do so without being subject to constant criticism as, for example, more “qualitative” methodologies might be. This feature of CA is particularly attractive to administrators and others priests of the scholarly hierarchy.
Once CA has performed its decontextualization, the next task is interpretation. Numbers are not explanations, after all, and theory enters both into the creation of the number and into its subsequent application. It follows that an additional part of the process of justifying inequality will be the creation of a theoretical and discursive frame that assists in the “correct” interpretation of the number. The creation of this discursive frame is an absolute requirement. Otherwise we are left with a meaningless number that measure nothing. The number “8” is meaningless without some theoretical frame to assist our interpretation. And even if I say that “8” refers to the number of citations an article gets, you still need to know whether 8 is a high number or a low number. And once you've learned that “8” is high, then you have to know what it means to get a high citation score like this. Does it mean you have more stored ability, or simply that you published your piece in the right journal? All these questions are answered, either implicitly or explicitly, by the rhetorical frame packed around the citation number. If done well and with sophistication, the discursive frame applied around the citation number (or any number for that matter) can reduce the possibility of alternative interpretations and even create a black box around the number which few would bother to pry open.
So what is the discursive frame spun around CA? Simply, the frame is articulated in a “great man” (and I use “man” here intentionally because the articulators of the frame seem to have meant “men”) myth of science. This myth is based on the notion that “great” contributions, and great contributors, to science are few and far between (in other words that ability is distributed in a bell shaped curve and thus unevenly distributed). Contributions to scientific progress can be thus expected to centre around largely unique individuals of genius. Derek de Sola Price's suggestion that most academic journals are mere intellectual fluff (presumably outlets for the mediocre peons of science) is symptomatic of this position (de Sola Price, 1965). But we see it more clearly when authors point directly to the uneven contribution of individuals. For example, there is Ziman who says:
Not only is there too much scientific work being published; there is much too much of it ... the need to get recognition by publication forced each of us to shout a little longer and louder so as to be noticed at all in the gathering, swelling, crowd of voices .... The result has been a proliferation of semi-literate, semi-scientific, half-baked and trivial material which threatens to swamp the whole system. 3
Still, there is more direct evidence of a subtle framing. In order to increase the rhetorical weight of their discursive frame, advocates of CA offer their justifications in the language of rigorous statistical procedures. They offer a null hypothesis and an alternative hypothesis. It is in the discursive work beyond this shift to the language of statistical rigour, the contortions needed to get to this point, where we see the ideological underpinnings of CA, and where the discursive work needed to create the frame for interpreting CA, most clearly.
Jonathan R. Cole and Stephen Cole first explicitly stated this null in a seminal article in 1972 where they draw on the work of Jose Ortega y Gasset in order to frame their position. They argue that Ortega believed that science advanced via the contribution of a great mass of men [sic] of only mediocre talent. Cole and Cole call this the Ortega Hypothesis. We might want to state this in the typical terms used to denote statistical hypothesis testing. The null (or Ortega hypothesis), simply stated, is that all individuals involved in science contribute to the scientific enterprise.
Hortega:We all contribute something to science. And all contributions are important.
The Coles offer this alternative.
Ha: Science only advances via the contributions of great men of unusual talent
and set forth to deny the null hypothesis with, what else, citation data. The Coles conclude, not surprisingly, that because only a few get cited in the citation record, while the vast majority go uncited (the skewness of the CA index), science advances only through the contribution of great men [sic]. Rejecting the null, the Coles suggest that:
...even the scientists who make these 'smaller' discoveries come principally from the top strata of the scientific community. In the proper perspective of the history of science, 'normal science' as Kuhn defines it, is not done by the average scientist, but by the elite scientists. Indeed, in the longer perspective, the work of many of today's outstanding scientists, such as Nobel Laureates and members of the National Academy of Sciences, may turn out to be minor footnotes in the history of science (Cole and Cole, 1972: 369).
It is difficult to underestimate the importance of the hypothesis and the intellectual field which it represents. Still, we can get a sense of the depth of feeling that surrounds this field by considering that a criticism of CA and its use to reject the Ortega hypothesis by M. H. MacRoberts and Barbara R. MacRoberts (1987) prompted a vociferous defence of CA and the original rejection of the Ortega hypothesis. It would appear feelings run high around this issue.
But what of this hypothesis? The work of the Coles has all the elements of a good piece of science and, perhaps, a great contribution to scientific progress. There is a theoretical frame, a null hypothesis, and a set of quantitative indicators. But this is science of the worst kind and digging deeper beneath the surface clearly reveals the ideological underpinnings and the discursive work being carried out. Indeed, it turns out that in order to force the creation of this discursive frame, and justify its use, the Coles rely on creative oversight and omission of key passages in Ortega's original work. In short, the Cole's changed the meaning of what Ortega had said so that they could then use what they said he said to knock down a null hypothesis conjured out of thin air. The created a straw man of epic proportions.
Heidi Lee Hoerman and Carole Elizabeth Nowicke 4 provide a brilliant analysis of the critical error on which CA has been based. Contrary to the characterisation of Ortega's work provided by the Coles, they note that that Ortega had a singularly negative view of the modern specialist scientist. He, they argue, lumped all modern scientists into the category of “learned ignoramus” because of their overly specialised, one-sided, and thin view of the world. Hoerman and Nowicke (1995) go on to demonstrate how Cole and Cole in their 1972 article misunderstand, misrepresented, and misquoted Ortega in their conceptualisation of scientific eminence. This misquotation has subsequently become the unfortunate shibboleth of numerous authors who drew not on original sources but on the faulty interpretation of the Coles'. Ortega's true position is included for the reader's edification.
The reason of this lies in what is at the same time the great advantage and the gravest peril of the new science, and of the civilisation directed and represented by it, namely, mechanisation. A fair amount of the things that have to be done in physics or in biology is mechanical work of the mind which can be done by anyone, or almost anyone. For the purpose of innumerable investigations it is possible to divide science into small sections, to enclose oneself in one of these, and to leave out of consideration all the rest. The solidity and exactitude of the methods allows of this temporary but quite real disarticulation of knowledge. The work is done under one of these methods as with a machine, and on order to obtain quite abundant results it is not even necessary to have rigorous notions of their meaning and foundations (Ortega quoted in Hoerman and Nowicke, 1995).
As Hoerman and Nowicke point out, Ortega y Gasset damns all scientists for their one-dimensional, compartmentalised, and superficial view of the world. He characterises them as the epitome of the mass-man and laments the state of the scientific enterprise which allows science to advance without understanding. This is significant. That the Coles would twist this passage is key evidence that they are constructing a frame and supporting it by creating a positive modality with links to older, more established, literature. This black boxing is a common rhetorical device in the literature so in some ways it is not surprising (Latour, 1987). What is surprising, and more important for our purposes, is that the presence of this error points directly to the discursive work which advocates of CA must engage in to support not only their methodology, but the uses to which it is put, in the face of serious scholarly criticism. This myth provides the critical justification for stratification in the academy.
So where are we left now? We have a bit of history - a partial image of how science can twist reality to support and naturalise systems of class and gender hierarchy. And we shouldn't be surprised. Science has an exemplary history of justifying inequality of all forms (Rose, 1980; Rose, Kamin and Lewontin, 1984). It has provide “male-operated exclusion mechanisms,” (Rose, 1986: 60), allowed masculine intellectual and cultural values to dominate (Bleier, 1986) and provided the hegemonic service for class, gender (Haraway, 1986), and colonial structures of domination (Adas, 1989). Because CA is completely insensitive to all but the most formal contributions to the academy, and even then it tends toward the conservative, it is biased and inadequate. CA does not fairly represent the entire range of contributions to the academy and hence using it as a way of measuring contribution will exacerbate already existing hierarchies and ensure that the vast pool of scientific workers never gets credit for their contribution. As we have seen, CA is a perfect prop for supporting the status quo.
The figures are far from useless. But their value comes into their own for outsiders. They constitute science... as an object ready for discipline. It is grouped, ordered, ranked, graphed and compared. Complex fields are broken down into lists of numbers; sophisticated theories are present only as titles and numerical codings. CA provides a world where some can be rewarded and others punished; a world where funds can be targeted and weak spots identified. In fact, it is a world ready for the workings of the science barons - the Research Councils, the Committees of Vice-Chancellors, the Universities Finance Council, the Department of Education and Science (Hicks and Potter, 1991; 485).
The games that could be played with citation indices are a computer programmer's dream come true! (Boor, 1982)
The reader will recall at this point that neoliberalism is pushing productivity measures in the academy and forcing the importation of discourses alien to academic quality. This, coupled with the emergence of the WWW and the coming administrative push for post- publication measures means that CA will very likely become an increasingly popular method of monitoring scholarly output and controlling scholarly discourse. 5 In fact, this is already happening. Blaise Cronin and Kara Overfelt note that citation counts have diagnostic and “medicinal” value (1994: 70):
From an administrators perspective, the kind of data presented in this study could become the basis, or at least a component part, of a performance indicators database....there are obvious benefits. For example, the data in Table 13 provide an at-a- glance sense of how citation performance relates to, inter alia, faculty status, time-in-field, gender, and remuneration levels. This table has diagnostic value, particularly if one looks at variability across status lines....That kind of information has value in that it raises questions relating to the bases on which tenure is awarded, and may be useful in developing predictors of citation performance for tenure-track faculty. Given that recent research has shown that “there appears to be moderate to high correlations between citations to earlier work and citations to later work” ... it should be possible, with a dynamically updated database, to predict the future citation performance of junior faculty and/or to use such data to facilitate the development of customized research and publication strategies in order to improve an individual's citation profile over time. In that sense, citation-based auditing could be a useful self management tool for academic administrators at both the unit and institutional levels, particularly if the data are used to complement other information, both quantitative and qualitative, on faculty performance and productivity.
The implications of CA are clearly stated here. Citation analysis could be used to assign scholars Citation IQs much like intelligence testing is used to assign IQs to individuals. (as Porter, Chubin and Xiao-Yin note [1988], bibliometric like CA function identically to IQ tests) Much like students are now streamed using SAT or GRE scores, scholars could be streamed into careers most appropriate given their level of “ability.” Those that had lower citation IQs (women, scholars of colour, working class scholars) could be directed into streams that focus on instruction and be encouraged to take their first job in a community college. Alternatively, those who have the highest citation IQs (privileged white men and others who support the status quote) could be directed into high prestige and highly rewarding research streams. No doubt prestigious research institutions will feel most comfortable hiring someone who has demonstrated research potential - or whose “potential” can be “predicted” from early citation data. Finally, older scholars who show creeping senility (i.e., declining CA scores) can be offered early retirement packages or other options that take them out of the scholarly mainstream.
Of course, any form of early streaming is ridiculous because it accentuates - over the long term - any educational disadvantages that might exist for scholars in the early stages of their career. For example, if early citation data from the initial publication efforts of young scholars is used to predict the future performance of the up and coming next generation, then we might reasonably expect that those scholars coming from smaller, less well funded institutions with poorer overall publication records would experience a disadvantage. The unfortunate outcome of tagging scholars with a Citation IQ early in the game is that it sets up the conditions for a self fulfilling prophecy. Scholars may well accept their lower status and not even try to overcome any deficiencies (however minor) and, even if they don't accept the validity of the measure, universities will. In the financially strapped environment that is the academy these days it will make little sense for universities (especially the prestigious ones) to risk hiring someone with a low Citation IQ.
While the potential use of CA in hiring, promotion, and even retirement decisions is certainly something to be struggled against, a much darker implication of CA derives from its potential to direct and control scholarly discourse through a form of self policing. In fact, this is actually the state hope of advocates as we saw above. As you will recall, CA has a tendency to make visible only certain types of scholarship and in this it favours that status quo. It is biased against gender, race, and country. And, most important here, it tends to ignore substantive contributions that challenge scholarly cannon. It favours run of the mill scholarship (what Kuhn might call Normal Science). Because of this, it is not hard to imagine that scholars hoping to increase their Citation Quotient would consciously (and perhaps unconsciously) direct their research and teaching interests towards those disciplines and subdisciplines that have clearly demonstrated their greater ability to garner citations. By doing so, they would steer clear of the more controversial areas. The result would be a conservative shift in scholarly discourse.
The insidious part of all this would be that the reasons for the shift away from critical scholarship would be completely invisible to most scholars. Because the evaluative mechanism is embedded in “objective” technologies, the tendency would be towards becoming disciplinary subjects of a new and invisible discursive order. Individuals would internalise the disciplinary mechanisms and the locus of administrative control would disappear in a puff of electronic smoke. This would fit nicely with technologies already in place for disciplining professionals and creating disciplinary subjects. Grey describes the transformation of overt administrative control towards the internalisation of discipline (Grey, 1994: 489):
The rating procedure is thus transformed. Instead of being constituted as an irksome, intrusive and threatening technique of management control, it becomes a benevolent process for their realisation of this perfection, a technique to assist individuals to become their true selves and to realise their aspirations. Even the act of sacking is reconstituted through the personnel department as 'counselling' out'. A supposedly mutual career decision for the employee to leave the firm.
This phenomenon of self-policing might have the greatest impact on graduate students who, understandably nervous about having the best record possible when they enter the job market, would choose their substantive interests with the intention of increasing their CIQ and post-publication impact. Graduate supervisors might also be expected to get into the act of directing scholarly discourse. The sage advice of a student's master may eventually be concerned less about just getting published and more about choosing the correct areas to publish in order to ensure a good CIQ. We can easily imagine a fatherly supervisor imparting the following pearl of wisdom. “The game has changed, publications are no longer good enough. You have to have citations.” It is even conceivable that the prestigious departments, as part of their professional training seminars, would supply lists of highly cited areas in order to assist their graduate students, who ultimately contribute to the reputation of the department, to choose their intellectual interests wisely.
As noted, the beauty of the system is that it is invisible. Regulation is conducted anonymously (Kachur, 1995). Politics enter into the regulation through the seemingly neutral quantitative measures used to police scholarly output. The potential for anonymous regulation of scholarly output will be even greater when and if CA is utilised with electronic journals on the Internet (as we noted above, there is already a push on to move CA onto the Internet). At that point, it will become possible to plug individual scholars directly into a citation loop and have them monitor themselves on an ongoing basis. A central data collection organisation could be responsible for general citation collection. And perhaps each university could develop or purchase their own web roaming robots that go out onto the internet and retrieve data on faculty. Universities would easily be able to provide monthly or even daily citation updates delivered directly to the email of all tenured faculty, sessionals, and even graduate students. The scholar, worrying about tenure, advancement, or job openings will perform all the behavioural modification necessary to ensure the continued structural integrity and functioning of the scholarly system.
At this point we have all the pieces we need and it is time to return to our discussion of cybernetics. As I noted earlier in this paper, CA may function as part of a cybernetic system of control that could potentially alter scholarly discourse. We have seen that in the current environment dominated by neoliberal ideology, information technologies are used to surveille populations for the purposes of control. I have also argued that scholars do not occupy a privileged position vis surveillance technologies. The threat that IT will be used against us and in the interests of administrators (who are more and more representatives of capital) is real.
We have also seen that technologies are not neutral. This goes for actual hard technologies like computers, but also for softer sensing technologies like CA. Despite the politically colourless designation of computers as CPUs, the much more politically charged CCU seems more accurate and appropriate. At least early writers were more forthright about the deeper implications of control inherent in information (command and control) technologies. Similarly citation analysis, despite its status as neutral/objective/quantitative indicator, is politically charged. Because of the embedded politics of CA, any use of it in surveillance loops may very well impact the nature of scholarly discourse.
We can tie our entire discussion together by summarising the contours of a cybernetic system for controlling scholarly discourse in
. There, we see how CA can fit neatly into a cybernetic system
aimed directly at scholars.
The only difference between this conception of CA and the earlier ones is the way compliance is sensed and how punishment and rewards are apportioned. Here, we monitor compliance with techniques of self evaluation like citation analysis. CA provides the information necessary to make decisions to reward or punish individual scholars. Those who get cited (recall the conservative nature of CA) get jobs. Those who don't lead a life of part- time employment at the margins of our disciplines. And in all this, the regulation system is largely anonymous and invisible. We blame ourselves for our failure - we lack the ability, the talent, the knowledge, the genetic makeup - to make it in the cut-throat world of the academy. But regardless of how we envision our failures, the fact remains that our failures are determined in advance by the position we take vis the scholarly status quo. We increase our probability of success by bypassing critical scholarship and silencing ourselves. We lives are stochastically determined by the cybernetic control loops we find ourselves embedded in.
The invention of print, however, made it easier to manipulate public opinion, and the film and the radio carried the process further. With the development of television, and the technical advance which made is possible to receive and transmit simultaneously on the same instrument, private life came to an end. Every citizen, or at least every citizen important enough to be worth watching, could be kept for twenty-four hours a day under the eyes of the police and in the sound of official propaganda, with all other channels of communication closed. The possibility of enforcing not only complete obedience to the will of the State, but complete uniformity of opinion on all subjects, now existed for the first time. (George Orwell - 1984)
This application of the metaphor of cybernetics to understanding the implications of information technology is only a tentative theorisation. Much work needs to be done to extend both the theorisation and the application of the theory to other areas penetrated by cybernetic technologies. There are a couple of different approaches. One is to extend the theorisation of cybernetics to provide more detail around the individual components of cybernetics systems (e.g., the sensing devices and control loops) and another is to more fully explore the biases inherent in technologies of all sorts, but information technologies in particular.
In terms of extending the analysis of the components of the system, fruitful work can be conducted on determining how different “sensing devices” can invisibly import political and social bias into cybernetic systems - thereby allowing a form of anonymous regulation and determining to an extent the outcomes of cybernetic control. This amounts to a methodological prescription for approaching CA since an extension of our understanding of the nature of sensing devices seems critical for understanding the implications of all forms of cybernetic technologies. For example, we understand the implications of scholarly cybernetics only by unpacking the sensing devices used to gather information. This theoretically is important, of course, but there is also a strategic value to this exercise. If we can unpack the politics of the technologies (as I did with CA in this paper), if becomes very difficult to ignore the politics of technology. And, if we're forced to face the politics of technology, we are more likely to resist and more likely to be able to organise resistance.
In addition to unpacking sensing technologies, more work needs to be done on how populations are controlled in cybernetic loops. Probably most of this work is already available and already being put to use in military and industrial applications. It is up to us to (re)discover the sociological and psychological work which underpins systems of control and expose this work for what it is. Fruitful areas of inquiry can be discerned. Freudian theory on the internalisation of authority (or Foucault's reworking of this in his conceptualisation of panoptic technologies and their ability to force the internalisation of the panoptic gaze) seem particularly fertile areas for theoretical linkage. Cybernetic technologies seemed designed for the type of autonomous regulation best conducted through the individuals superego. How different is it, after all, to internalise the all knowing gaze of the parent/society (parental discipline and the parental/social superego) or to internalise the all knowing gaze of the prison warden (panoptic technologies and the disciplinary superego) or the all knowing gaze of the administrator (CA, scholarly cybernetics and the cybernetic superego)?
Perhaps another fruitful area of theoretical linkage is between cybernetics and behavioural and cognitive psychology. Much more than the Freudian versions of psychology these seem concerned primarily with control (through reward and punishment). And while this form of control is primitive compared to the type provided through the internalisation of authority, it is still effective (as a quick glance at the size of behavioural psychology departments attest). Perhaps different control strategies are used for different populations. Anonymous control for those used to autonomy, more direct forms of control for those not. Even chemical forms of control (Prozac) fit well into cybernetic theory do they not?
As noted above, and additional area where theoretical work would be useful is in developing the notion that technologies in general, and information technologies in particular, are not neutral. This is far from a new notion as we have seen. When Norbert Weiner discussed his cybernetic systems he explicitly stated the political implications of his technologies (he may have actually chosen CCU over the politically neutral CPU). Other to have long tried to push this naïve notion aside as well. The best conceptualisation is perhaps Corlan Bush's idea of valence. As Corlan notes:
However, just as morality is a collective concept, so too are guns. As a class of objects, they comprise a technology that is designed for killing in a way that ice picks, hammers, even knives - all tools that have on occasion been used as weapons - are not. To believe that technologies are neutral tools subject only to the motives and morals of the user is to miss completely their collective significance. Tools and technologies have what I can only describe as valence, a bias or “charge” analogous to that of atoms that have lost or gained electrons through ionization. In other words, particular tools or technologies tend to be favored in certain situations, tend to perform in a predictable manner in these situations, and tend to bend other interactions to them (Bush, 1983: 154-5).
As Bush notes, all technologies have biases built into them, uses to which they are more easily turned. Television, for example, is valenced for individuation. It draws attention to itself, hamstrings social interaction, and denies interactivity. The same can be said of automobiles which daily force individuals to drive a solitary mile to work. And guns do, in fact, kill people. So to is it with information technology. They are valenced towards control. Even though we can use a computer on our desktop to increase our productivity, the fact remains that an administrator somewhere can collate and measure our keystrokes, determine what we're working on, and wire us into a system of feedback that increases the possibility of compliance and reduces the possibility of dissent.
Its important we get on with the business of developing our understanding of cybernetic technologies since the stakes seems quite high. The U.S. department of defense hasn't missed the implications. They will spend 12 billion on information technology and command and control systems in 1998 (http://www.gcn.com/gcn/1998/April6/dod.htm).
Its important we develop our ability to resist before we fall victims to a cybernated academy. Otherwise we will get the one dimensional world predicted some decades ago by Marcuse. Critical voices will not be eliminated, but they will no doubt be seriously weakened - probably first in North America (where much sociology, psychology and other sciences already suffer from various forms of one-dimensionality) and then latter in other countries. And as time passes, and graduate students do what they have to to get hired on, less and less of the essence of critical thinking and critical discourse will be rediscovered each generation. It will be a gradual devastation, but it will be devastation nonetheless. And when we couple the cybernetic control of scholarly discourse with technologies of cybernetic instructor control in the undergraduate classroom which force the educator to turn away from critical pedagogies and entertain and titillate, we will surely bequeath the next generation of graduate students and scholars a level of mediocrity suitable ensuring surviving (and nothing more) in the cybernated academy.
Adas, Micheal (1989). Machines as the Measure of Men: Science, Technology, and Ideologies of Western Dominance. Ithaca: Cornell University Press.
Armstrong, Pat, Armstrong, Hugh, Choiniere, Jacqueline, Mykhalovskiy, Eric, and White, Jerry P. (1997). Medical Alert: New Work Organizations in Health Care. Toronto: Garamond.
Bagby, R. Michael, Parker, James D., & Bury, Alison S. (1990). A Comparative Citation Analysis of Attribution Theory and the Theory of Cognitive Dissonance. Personality and Social Psychology Bulletin, 16: 274-283.
Baird, Laura M., & Oppenheim, Charles. (1994). Do Citations Matter? Journal of Information Science, 20: 2-15.
Bleier, Ruth (1986). Feminist Approaches to Science. Toronto: Pergamon.
Boor, Myron. (1982). The Citation Impact Factor: Another Dubious Index of Journal Quality. American Psychologist, 37: 975-977.
Borgman, Christine L., & Rice, Ronald E. (1992). The Convergence of Information Science and Communication: A Bibliometric Analysis. Journal of the American Society for Information Science, 43: 397-411.
Bott, David M., & Hargens, Lowell L. (1991). Are Sociologists' Publications Uncited? Citation Rates of Journal Articles, Chapters, and Books. The American Sociologist, 22: 147-160.
Bourdieu, Pierre (1988). Homo Academicus. Cambridge: Polity Press.
Bourdieu, Pierre (1984). Distinction: A Social Critique of the Judgement of Taste. Trans. Richard Nice. Cambridge: Harvard University Press.
Brodie, Janine (995). Politics on the Margins: Restructuring and the Canadian Women's Movement. Halifax: Fernwood.
Brodman, Estell. (1944). Choosing Physiology Journals. Bulletin of the Medical Library Association, 32: 479-483.
Bush, C. G. (1983). Women and the Assessment of Technology: to Think, to Be; to Unthink, to Free. In J. Rothschild ((Ed.)), Machina Ex Dea: Feminist Perspectives on Technology (pp. 151-170). New York: Pergamon Press.
Bukharin, Nikolai (1925). Historical Materialism. New York: International Publishers.
Buffardi, Louis C., & Nichols, Julia A. (1981). Citation Impact, Acceptance Rate and APA Journals. American Psychologist, 36: 1453-1456.
Bukharin, Nikolai (1925). Historical Materialism. New York: International Publishers.
Burrell, Gibson (1998). Modernism, Post Modernism and Organizational Analysis 2: The Contribution of Michel Foucault. Organizational Studies, 9: 221-235.
Butler, H. Julene. (1995). Where Does Scholarly Electronic Publishing Get You? Journal of Scholarly Publishing, 26: 234-245.
Cameron, Robert D. (1997). A Universal Citation Database as a Catalyst for Reform in Scholarly Communication. First Monday, 2: http://www.firstmonday.dk/issues/issue2_4/cameron/
Clarke, Tony (1997). Silent Coup: Confronting the Big Business Takeover of Canada. Toronto: James Lorimer.
Cohn, Ellen G., & Farrington, David P. (1990). Differences Between British and American Criminology. British Journal of Criminology, 30: 467- 481.
Cole, Stephen, & Cole, Jonathan R. (1987). Testing the Ortega Hypothesis: Milestone or Millstone? Scientometrics, 12: 345-353.
Cole, Jonathan R., & Cole, Stephen. (1972). The Ortega Hypothesis: Citation Analysis Suggests That Only a Few Scientists Contribute to Scientific Progress. Science, 178: 368-375.
Collyer, Fran (1998). Email conversation with author.
Cox, Miles W., & Catt, Viola. (1977). Productivity Ratings of Graduate Programs in Psychology Based on Publication in the Journals of the American Psychological Association. American Psychologist, 32: 793-813.
Cronin, Blaise, & Overfelt, Kara. (1994). Citation-Based Auditing of Academic Performance. Journal of the American Society for Information Science, 45: 61-72.
du Gay, Paul and Salaman, Graeme (1992). The Cult[ure] of the Customer. Journal of Management Studies, 29: 615-633.
de Kerckhove, Derrick (1995). The Skin of Culture: Investigating the New Electronic Reality. Toronto: Somerville House Publishing.
Diamond, Arthur M. Jr, & Levy, David M. (1994). The Metrics of Style: Adam Smith Teaches Efficient Rhetoric. Economic Inquiry, 32: 138- 145.
Edge, David. (1979). Quantitative Measures of Communication in Science: A Critical Review. History of Science, 17: 102-134.
Everett, James E., & Pecotich, Anthony. (1993). Citation Analysis Mapping of Journals in Applied and Clinical Psychology. Journal of Applied Social Psychology, 23: 750-66.
Friman, P. C., Allen, K. D., & Kerwin, M. L. (1993). Changes in Modern Psychology: A Citation Analysis of the Kuhnian Displacement Thesis. American Psychologist, 48: 658-664.
Fuller, Linda, & Vicki, Smith. (1991). Consumers' Reports: Management by Customers in a Changing Economy. Work, Employment and Society, 5: 1- 16.
Furr, L. Allen. (1995). The Relative Influence of Social Work Journals: Impact Factors vs. Core Influence. Journal of Social Work Education, 31: 38- 45.
Gandy, Oscar H. (1993). The Panoptic Sort: A Political Economy of Personal Information. Boulder, Colo.: Westview.
Garfield, Eugene. (1993). What Citations Tell us About Canadian Research. Canadian Journal of Information and Library Science, 18: 14-35.
Glenn, Norval D. (1971). American Sociologists' Evaluations of Sixty-Three Journals. The American Sociologist, 6: 298-303.
Grey, Christopher (1994). Career as a Project of the Self and Labour Process Discipline. Sociology, 28: 479-497.
Gross, Paul L. K., & Gross, E. M. (1927). College Libraries and Chemical Education. Science, 66: 385-389.
Hackett, Robert A. (1991). News and Dissent: The Press and the Politics of Peace in Canada. New Jersey: Ablex.
Hamilton, David P. (1990). Publishing by - and for? - the Numbers. Science, 250: 1331-2.
Haraway, Donna (1986). Primatology is Politics by Other Means. In Ruth Bleier (Ed.). Feminist Approaches to Science. Toronto: Pergamon.
Harding, S. (1994). After the Neutrality Ideal: Science, Politics, and “Strong Objectivity”. In Margaret C. Jacob (Ed.), The Politics of Western Science, (pp. 81-102). New Jersey: Humanities Press.
Harnad, Steve. (1995). Electronic Scholarly Publication: Quo Vadis? Serials Review, 21: 78-80.
Harnad, Stevan (1995a). “Implementing Peer Review on the Net: Scientific Quality Control in Scholarly Electronic Journals.” In R. Peek & G. Newby (Eds.), Electronic Publishing Confronts Academia: The Agenda for the Year 2000,Cambridge: MIT Press [ftp://ftp.princeton.edu/pub/harnad/harnad95.peer.review.txt].
Harnad, Stevan. (1991). Post-Gutenburg Galaxy: The Fourth Revolution in the Means of Production of Knowledge. The Public-Access Computers Systems Review, 3: 39-53 [gopher://info.lib.uh.edu/00/articles/e- journals/uhlibrary/pacsreview/v2/n1/harnad.2 n1].
Harwood, Jonathan (1979). “Heredity, Environment, and the Legitimation of Social Policy.” In B. Barnes, & S. Shaping (Ed.), Natural Order: Historical Studies of Scientific Culture. London: Sage.
Hattie, John, Print, Murray, & Krakowski, Krzysztof. (1994). The Productivity of Australian Academics in Education. Australian Journal of Education, 38: 201-218.
Hicks, Diana, & Potter, Jonathan. (1991). Sociology of Scientific Knowledge: A Reflexive Citation Analysis or Science Disciplines and Disciplining Science. Social Studies of Science,
Hoerman, Heidi Lee, & Nowicke, Carole Elizabeth. Secondary and Tertiary Citing: A Study of Referencing Behavior in the Literature of Citation Analysis Deriving From the Ortega Hypothesis of Cole and Cole. Library Quarterly, 65: 415-434.
Hoffman, Donna L., & Holbrook, Morris B. (1993). The Intellectual Structure of Consumer Research: A Bibliometric Study of Author Cocitations in the First 15 years of the Journal of Consumer Research. Journal of Consumer Research, 19: 505-517.
Howard, G. S., Cole, D. A., & Maxwell, S. E. (1987). Research Productivity in Psychology Based on Publication in the Journals of the American Psychological Association. American Psychologist, 42: 975-986. 505-517.
Jacob, Margaret C. (1988). The Cultural Meaning of the Scientific Revolution. Philadelphia: Temple University Press.
Kachur, Jerrold L. (1995). Hegemony and Anonymous Intellectual Practice: The Politics of Theoretical Representation in an Age of Scientific Forecasting. PhD Dissertation. Department of Educational Foundations, University of Alberta.
Karake, Zeinab A. (1992). Information Technology and Management: An Agent Theory Perspective. London: Preager.
Klein, Waldo C., & Bloom, Martin. (1992). Studies of Scholarly Productivity in Social Work Using Citation Analysis. Journal of Social Work Education, 28: 291-299.
Krantz, David L. (1971). The Separate Worlds of Operant and Non-Operant Psychology. Journal of Applied Behavior Analysis, 4: 61-70.
Latour, Bruno (1988). “The Politics of Explanation: An Alternative.” In Steve Woolgar (Ed.), Knowledge and Reflexivity: New Frontiers in the Sociology of Knowledge,Beverly Hills: Sage.
Lin, Nan, & Nelson, Carnot E. (1969). Bibliographic Reference Patterns in Core Sociological Journals, 1965-1966. The American Sociologist, 14: 47- 50.
Liu, Mengxiong. (1993). A Study of Citing Motivation of Chinese Scientists. Journal of Information Science, 19: 13-23.
Lutz, Catherine. (1970). The Erasure of Women's Writing in Sociocultural Anthropology. American Ethnologist, 17: 611-27.
Mack, Thura. (1991). A Model Methodology for Selecting Women's Studies Core Jounals. Library and Information Science Research, 13: 131-45.
MacRoberts, M. H., & MacRoberts, Barbara R. (1987). Testing the Ortega Hypothesis: Facts and Artifacts. Scientometrics, 12: 293-295.
Mainprize, Steve (1996). Elective Affinities in the Engineering of Social Control: The Evolution of Electronic Monitoring. Electronic Journal of Sociology, 2: http://www.sociology.org/vol002.002/Mainprize.abstract.1996.html
McBride, Stephen and Shields, John (1997). Dismantling a Nation: The Transition to Corporate Rule in Canada. Halifax: Fernwood.
McCain, Katherine W. (1992). Core Journal Networks and Cocitation Maps in the Marine Sciences: Tools for Information Management in Interdisciplinary Resarch. Proceedings of the ASIS annual Meeting, 29: 2-7.
McCain, Katherine W. (1991). Mapping Economics Through the Journal Literature: An Experiment in Journal Cocitation Analysis. Journal of the American Society for Information Science, 42: 290-96.
McQuaig, Linda (1995). Shooting the Hippo: Death by Deficit and Other Canadian Myths. Toronto: Viking.
Meadows, A. J. (1987). Ortega Hypothesis. Scientometrics, 12: 315-316.
Merton, Robert K. (1977). The Sociology of Science: An Episodic Memoir. Illinois: Southern Illinois University Press.
Meyer, Terry, & Spencer, John. (1996). A Citation Analysis Study of Library Science: Who Cites Librarians? College and Research Libraries, 57: 23-33. Morris, Dilys E. (1990). Electronic Information and Technology: Impact and Potential for Academic Libraries. College and Research Libraries, 50: 56-65. Namenwirth, Marion (1986). Science Seen Through a Feminist Prism. In Ruth Bleier (Ed.), Feminist Approaches to Science. Toronto: Pergamon. Newman, Jerry M., & Cooper, Elizabeth. (1993). Determinants of Academic Recognition: The Case of the _Journal of Applied Psychology_. Journal of Applied Psychology, 78: 518-526. Noble, David (1979). America by Design: Science, Technology and the Rise of Corporate Capitalism. : Oxford University Press. Plomp, R. (1989). Statistical Reliability of Citation Frequency as an Indicator of Scientific Impact. Scientometrics, 17: 71-81. Porter, A. L., Chubin, D. E., & Xiao-Yin, Jin. (1988). Citations and Scientific Progress: Comparing Bibliometric Measures With Scientists Judgments. Scientometrics, 13: 103-124. Postell, William Dosite. (1946). Further Comments on the Mathematical Analysis of Evaluating Scientific Journals. Medical Librarian Association Bulletin, 34: 107-109. Price, Derek J. de Solla. (1965). Networks of Scientific Papers. Science, 149: 510-15.
Roche, Thomas, & Smith, David L. (1978). Frequency of Citations as Criterion for the Ranking of Departments, Journals, and Individuals. Sociological Inquiry, 48: 49-57.
Rose Hilary (1986). Beyond Masculinist Realities: A Feminist Epistemology for the Science. In Ruth Bleier (Ed.) Feminist Approaches to Science (p. 57- 76). Toronto: Pergamon.
Rose, Steven (1980). “Scientific Racism and Ideology: The IQ Racket from Galton to Jensen.” H. Rose, & S. Rose (Ed.), Ideology of/in the Natural Sciences, (pp. 87-115). Boston: G.J. Hall. Rose, Steven P. R., Kamin, Leon J., Lewontin, R.C. (1984). Not in our Genes: Biology, Ideology and Human Nature. Harmondsworth: Penguin, 1984.
Rossiter, Margaret W. (1993). The Matilda Effect in Science. Social Studies of Science, 23: 325-41.
Rushton, J. P. ,. &. Endler, N. S. (1977). The Scholarly Impact and Research Productivity of Departments of Psychology in the United Kingdom. Bulletin of the British Psychological Society, 30: 369-373.
Rushton, J. Philippe, & Roediger, Henry L. (1978). An Evaluation of 80 Psychology Journal Based on the Science Citation Index. American Psychologist, 33: 520-523.
Schiller, Herbert I. (1989). Culture Inc.: The Corporate Takeover of Public Expression. New York: Oxford University Press.
Seglen, Per O. (1992). The Skewness of Science. Journal of the American Society for Information Science, 43: 628-638.
Sewell, G., & Wilkinson, B. (1992). Someone to Watch Over Me?: Surveillance, Discipline and the Just-In-Time Labour Process. Sociology, 26, 271-289.
Shapiro, Fred R. (1992). Origins of Bibliometrics, Citation Indexing, and Citation Analysis: The Neglected Literature. Journal of the American Society for Information Science, 43: 337-339.
Small, H. G., & Crane, D. (1979). Specialities and Disciplines in Science and Social Science: An Examination of Their Structure Using Citation Indexes. Scientometrics, 1: 445-461.
Stevens, Garry. (1990). An Alliance Confirmed: Planning Literature and the Social Science. American Planning Association, 56: 341-349.
Teeple, Gary (1995). Globalization and the Decline of Social Reform. Toronto: Garamond Press.
Townley, Barbara (1993). Foucault, Power/Knowledge, and its Relevance for Human Resource Management. Academy of Management Review, 18: 518- 545.
Treloar, Andrew. (1996). Electronic Scholarly Publishing and the World Wide Web. Journal of Scholarly Publishing, 27: 135-150.
Velho, Lea. (1986). The “Meaning” of Citation in the Context of a Scientifically Peripheral Contry. Scientometrics, 9: 86. Wade, Nicholas. (1975). Citation Analysis: A New Tool for Science Administrators. Science, 188: 429-432.
Ward, Kathryn B., Gast, Julie, & Grant, Linda. (1992). Visibility and Dissemination of Women's and Men's Sociological Scholarship. Social Problems, 39: 291-298.
Ward, Kathryn B., & Grant, Linda. (1985). The Feminist Critique and a Decade of Published Research in Sociology Journals. The Sociological Quarterly, 26: 139-157.
Watson, Johnny L., Gouvier, William Drew, & Manikam, Ramasamy. (1989). Publication Counts and Scholastic Productivity: Comment on Howard, Cole and Maxwell. American Psychologist, 44: 737-739.
Weiner, Norbert (1961). Cybernetics. New York: MIT Press.
Weiner, Norbert (1954). The Human Use of Human Beings. New York: Doubleday Anchor.
White Murray J., & White, K. Geoffrey. (1977). Citation Analysis of Psychology Journals. American Psychologist, 32: 301-305.
Wresch, William (1996). Disconnected: Haves and Have-Nots in the Information Age. New Brunswick: Rutgers.
Wright, Richard A. (1994). The Effect of Editorial Appointments on the Citations of Sociology Journal Editors, 1970-1989. The American Sociologist, 25: 40- 45.
Xhignesse, Louis V., & Osgood, Charles E. (1967). Bibliographic Citation Characteristics of the Psychological Journal. American Psychologist, 22: 778-91.
Ziman, J. M. (1970). Ziman Plays Cassandra. New Scientists, 46: 212-213.
Zinkhan, George M., Roth, Martin S., & Saxton, Mary J. (1992). Knowledge Development and Scientific Status in Consumer-Behaviour Research: A Social Exchange Perspective. Journal of Consumer Research, 19: 282-291.
1. Found in Heylighen Cybernetic Technology. http://pespmc1.vub.ac.be/CYBTECH.html
2. Found in Heylighen Cybernetic Technology. http://pespmc1.vub.ac.be/CYBTECH.html
3. J.M. Ziman quoted in Anon (1970: 212
4. Heidi Lee Hoerman and Carole Elizabeth Nowicke (1995).
5. It is unlikely CA will be the only method of using IT and computers to monitor scholarly output. A number of departments at the University of Pittsburgh School of Medicine are experimenting with a faculty performance appraisal system. Faculty are given a form on computer disk. They are then required to update their CV any time they advance their research, teaching, administrative, or patient care activities (Bickel, 1991). This is a rather mundane example but it does point to the increasing use of technology to monitor and control.
© Radical Pedagogy