e-mail: jmjaffe@research.haifa.ac.il
Presented at the Annual Conference of the Midwest Association for Public Opinion Research (MAPOR), November 1994.
The information contained herein is made available on a strictly non-commercial basis and for the express purpose of furthering academic scholarship and research. Printed, photocopied, or electronic distribution (outside the domain of this particular web-site) of all or parts of this document is expressly forbidden without the tacit consent of the author.
Computer bulletin boards (CBBs) comprise the first "many-to-many"
mass medium due to the potential for many bulletin board messages to be
received by an unlimited number of CBB participants. Also, along the
information superhighway, interactive hypermedia presentations in
"webspace" and in "gopherspace" offer unprecedented opportunities for
information archives, of various symbol sets, to be disseminated and
accessed.
[Note: The terms "webspace" and "gopherspace" refer to CMC information
protocols. webspace refers to the world-wide web (WWW), a multimedia
information protocol which allows for hypermedia links. gopherspace
refers to a protocol which presents information archives indexed within
menus. Both WWW and Gopher employ the internet to connect conceptual
links between information sources.]
This paper concentrates on the use of interpersonal, interactive CMC within the context of popular political participation. The unprecedented involvement of the "audience," if indeed that term is appropriate to the "new media" information consumer, raises questions regarding both media uses and effects. Will a media which suits itself to individual become a force for social unity or separatism? Will social barriers between different sociodemographic and interest groups intensify or soften? Will this new media serve to decentralize media influence on social and political interaction? Can we expect popular adoption of CMC to lead to greater citizen influence in political discourse and policy decisions? Communication and social psychological theory will be utilized to frame some of these questions.
Cognitive dissonance theory, for instance, might be applied towards the question of whether information received through a sequence of active decision-making would project greater credibility and/or salience. Social cognition theory might help predict whether users of interactive media would develop higher levels of self-efficacy in the subsequent utilization of related new knowledge. Social learning theory might be used to prescribe how interactive messages could exploit individual differences influencing attitude and behavior.
Following World War II, a number of technological developments resulted in a proliferation of "personal" electronic media. The term "personal" describes media to which individuals and small groups have meaningful access, and which can be used to acquire, create, store, and/or disseminate information (Ganley, 1992). It also connotes a measure of independence from entities, such as film processing labs, which make the production of information tedious or subject to censorship. Citizens band radio emerged in the 1940s, followed by transistor radios and reel-to-reel audio tape in the 1950s, dry copiers, audiocassettes, and reel-to-reel video in the 1960s, portable video recorders, video cameras, and satellite dishes in the 1970s, and personal computers, personal fax machines, and camcorders in the 1980s. These innovations, which diffused quickly into western society, have arguably widened the ranks of individuals and groups who might spread political messages. It is no coincidence that, along with this development of personal media, we have moved into an economically postindustrial society, in which the information service sector accounts for more than half of our gross national product (Hiltz & Turoff, 1993).
The impact of personal media technologies on the political and social landscape can be both deliberate and accidental. The Zapbruder film, which graphically recorded President Kennedy's assassination, and the more recently videotaped Rodney King beating were non-orchestrated personal media artifacts which had great influence. The deliberate production of politically charged personal media was instrumental in the Iranian uprising of 1979 (Ganley, 1992). During the 1980s and 1990s, the practice of political kidnapping entered a new dimension, using emotionally charged videotaped pleas by victims for fulfilling the demands of the perpetrators. In this fashion, the politically disenfranchised used personal media to co-opt western mass media in disseminating political messages.
Reductions in size and price have made the personal computer practically available to individuals. In its present form, it is also arguable that computers are widely used for purposes of communication as well as mathematical calculation and other data processing. Word-processing and graphic information presentation are among the most common personal computer funcions. Configured with inexpensive laser printers and dry copiers, computers often fill the function of typewriters and offset printing systems, providing easy, convenient ways of composing printed messages. While the personal computer has joined the ranks of "personal" media, computer-mediated communication (CMC) has qualities of interconnectivity and interactivity which provide the user with far greater power in sharing messages on a global scale.
"Interconnectivity" refers to the capability of any entity on a network to send or receive messages from any other such entity. Along with the development of the personal computer, computer communication networks have spun interconnecting webs across the globe. The main conduit for the international computer communication, the Internet, has evolved into a collection of information sites capable of transferring information at ever-increasing speeds. Corporations, universities, governments, institutions, and organizations have established presences on the internet. In a projection of interconnectivity taken to extremes, Hiltz & Turoff (1993, p. 455) have coined the term "superconnectivity" to mean "1. The phenomenon of almost perfect transmission of communication and information throughout the human habitations of the universe, via computers. 2. The interconnections of all social and economic institutions as a result of communication via computer networks." While such a state of affairs is not yet imminent, it is clear to "cyberspace" observers that the degree of CMC interconnectivity is growing exponentially (Engst, 1993).
"Interactivity," as Rafaeli (1988) explains, is a communication characteristic which is greatest when discursive role assignment and turn-taking are nonautomatic or nearly so. This implies a level of conversationality, a blurring in the roles of sender and receiver towards serving some shared information goal. In such a scenario, no single or collective participant has absolute power over the transmission and reception of messages on the part of others. Printed media are essentially non-interactive. TV and radio, which allow for channel control, are slightly more interactive than printed media. Computer aided instruction (CAI), hypermedia, and simulations, including video games, are more interactive due to the user's continuous engagement in controlling the display of information. Computer bulletin boards and e-mail are more interactive than CAI and simulations because of their potential for asynchronous conversational interaction. Face-to-face and telephone are most interactive because of their synchronous conversational character.
Because of its capacity for interconnectivity, CMC can be considered a mass medium (Rafaeli, 1986). CMC interconnectivity enables one to send a message to a multitude of receivers simultaneously. This is often accomplished through the CMC formats of computer bulletin boards (CBBs) and electronic mailing lists. Moreover, the internet is becoming popularly perceived as a mass medium. Originally a channel devoted to military and academic purposes, the internet is now accessible to anyone with a personal computer and modem (available for less than $1,000), and the necessary funds for access fees (about $10 a month minimum). One count puts the growing number of internet users at 20 million (Cole, 1994). The capacity for true interactivity, however, sets CMC apart from other, more audience-passive mass media channels.
The combination of interconnectivity and interactivity within the same media format presents the technical possibility of the first many-to-many mass medium (Rafaeli, 1986), in contrast with the one-to-many sender/receiver configuration of broadcast and print media. This combined attribute is highly significant in the application of CMC within different models of democratic political participation.
In a Limited Representative State, citizens are highly restricted in the extent to which they might participate in determining public policy. Their options are limited to voting for representatives, lobbying them, and campaigning for them.
In the Expanded Representative State, citizen participation is expanded to include such activities as referenda and ballot initiatives, whereby citizens may occasionally be called upon to set policy. This is similar to the notion of a Plebiscitary Democracy (Abramson, Arterton, & Orren, 1988), in which citizens participate by casting their votes on specific issues. Because policy decisions are subject to plebiscites, there is a theoretically tighter fit between public opinion and public policy. Plebiscitary democracy is well-suited to the opinion that more direct forms of democracy, including public deliberation, would prove unwieldy in large populations with fragmented and divisive interests.
In the Participatory Representative State, citizens recognize a shared responsibility and interdependency in formulating policy. Political interaction would include public discussion and debate to which political representatives would be mindful. When the interaction is characterized by uniformly shared interests and concerns as well as a common national identity, it resembles what Abramson et al. (1988) call a Communitarian Democracy. In order to maintain the vitality of a democratic system, citizens are expected to participate in ongoing conversation and debate toward choosing and effecting policy. When the interaction exhibits free competition among groups with differing interests and strategic alliances of complementary interests, this resembles Pluralist Democracy (Abramson et al., 1988)
A Representative Participatory State would assume that all or most citizens would be prepared to serve some representative function at some point in time. This model shares the characteristic of public discussion and debate. However, the determination of representatives would come about from a combined process of election and random selection. This random selection would eliminate structural biases which otherwise limit participation to those with the requisite time, money, and other resources to serve as representatives. Rotation of representatives would be mandatory.
Within each of these systems, CMC could enhance the capability of conducting polls by efficiently distributing background information and by collecting poll data interactively. As Abramson et al. (1988, p. 105) indicate, Interactive systems can improve election polling because of their ability to tailor questions based on prior response. A greater depth of information is obtained and the respondent's experience is more pleasant and less intrusive in nature.
Interactive polling would also address long-standing concerns of Blumer (1948, cited in Converse, 1986) regarding the failure of standard polling practices to tap the complex nuances of public opinion. As Blumer states, public opinion is, "a kind of complex organic whole which mirrored the organization of society into functional groups. The only entity worthy of the name of public opinion is something generated by interactions in such a structure and which is 'effective' in the sense that people in positions of power judge it to be worth taking into account."
The capacity of CMC for group conversation would not be utilized under the limited representative and expanded representative states. The latter two models, participatory representative and representative participatory, would make better use of CMC's capabilities to manipulate greater volumes of information, decentralize media control, and provide interactive communication capability to citizens.
As Slaton (1992, p. 74) describes, the United States is presently at crossroads between an expanded representative and a participatory representative state. The present Clinton administration's involvement with the National Information Infrastructure (NII) would indicate that we are crossing over that line.
It is conceivable that a wider citizen participation within CBBs might resolve whether the character of a participatory representative state is communitarian or pluralist, and under what conditions. The characteristic of decentralization might lead one to surmise the latter. Toward this end, it would be useful to measure several groups of content-analytic indicators, including the following.
a) Homogeneity/heterogeneity of interests expressed by individuals.
b) Organizing tendencies within and across interest groups.
c) The degree to which polarization of opinions encourages and/or inhibits
dialogue.
On a smaller scale, it would be useful to determine the degree to which the CMC context itself influences the nature of participatory democracy toward communitarian or pluralist interaction. This would require either quasi-experimental or field study comparing participatory decision-making patterns in CMC and non-CMC contexts. To some extent this has already been done and useful characteristics of group communication have been mapped out. However, relatively few studies (e.g., the Televote studies outlined in Slaton, 1992) have examined this in a political discussion situation.
It is likely that the incidence of a participatory representative democratic state and adoption of CMC as a popular forum for political discourse would be related in a model of mutual causation.
Fig. 1. Mutual causation of CMC adoption and participatory representation.
Free speech is a major foundation to participatory representative democracy. While the interactive quality of CMC presents the capacity for individuals to be "heard" by all who care to listen, the interconnective quality of CMC, in a sense, safeguards it as a free speech forum. Although, governments could exercise a prerogative to constrain CMC activity within their borders, as Hiltz and Turoff (1993) point out, this would practically mean shutting down a nation's telephone system. Even this would not eliminate the use of packet switched radio communication between computers. As John Gilmore, of the Electronic Frontier Foundation puts it, "The (Inter)Net interprets censorship as damage and routes around it" (Rheingold, 1993, p. 7).
As mentioned above, the interconnective and interactive characteristics of CMC make it uniquely suited to such political participation. Rice & Love (1987) have documented how users of CMC systems can become more connected with people who share common interests but who would otherwise be unknown. It is also worth noting that heuristic computer information technologies may develop to the point of sophisticatedly content analyzing verbal information, providing users with textual digests which might help a political participant cope with what might amount to reams of discursive content. It is also worth noting that protocols and customs on CBBs have developed which serve to compartmentalize CMC discourse according to discrete topics. Examination of the USENET system reveals an intricate breakdown of topics in which interested parties can quickly locate a forum of interest. When a particular topic acquires too much "traffic," subtopics are quickly established. These are some of the ways that CMC can serve as a participatory forum and pragmatically accommodate large political discourses.
It is generally accepted that the text-only character of CMC strips away nonverbal and paraverbal cues which exist in face-to-face (FTF) communication. This reduces the degree to which cues of socioeconomic status differences, norms, physical appearance, and speech behavior might erect social barriers related to group stereotypes and social expectations. This reduced cuesÊsituation would likely have an impact upon sociopolitical discourse. Users tend to participate more equally on CMC systems than in many FTF interactions (Rice & Love, 1987). Furthermore, Hiltz and Turoff (1993) contend that CMC is better thought out, better organized, and richer than natural conversation. The verbal quality of CMC raises interesting questions of how the present mass media preoccupation with political image would be affected when candidates face an audience who read, rather than watch, their responses. Furthermore, in CMC, audience members have the capacity to directly indicate when they feel their concerns are not being seriously addressed. While the type of compromise and compliance which occur in policy-decision would probably not happen during election debates, the research cited above would predict a reduction in the confrontational rhetoric which seems to have intensified in broadcast mass media.
Aside from the text-only charactersitic, there is the matter of asynchronous communication which is afforded by CMC. Synchronous communication refers to the simultaneous presence of communicants on the communication channel, as in FTF, telephone, live radio/TV broadcasting, and teleconferencing situations. In debates and call-in shows, candidates and decision makers must answer on the spot and are often as preoccupied with how they appear in a confrontational situation as they are with how they conceptually address issues. The accent is on image and performance in the immediate sense rather than meaningful discourse.
Asynchronous communication refers to contexts in which communicants "post messages" in order of their reception, affording unlimited time and space to a message's composition. This is in sharp contrast to the exposure of television reporting, which has accelerated its news gathering to the point that a candidate's risk of verbal error is much greater than in the pre-electronic age (Abramson et al., p. 111). In asynchronous communication, there is no need to rely on a posturing, conversational "quick comeback" on the spot. There is "breathing room" to the extent that individuals have time after they receive messages to reference facts and compose statements before responding on the record. This increased breathing room and the stripped social cues in CMC may actually increase decision-making speed and productivity in the midst of popular participation. This projection directly contradicts the presumptions associated with plebiscitary democracy.
Recently, the Minnesota Electronic Democracy Project effected two CMC debates for senatorial and gubernatorial races in that state. Each debate consisted of three electronic discussion groups; one for the exclusive and private use of the candidates and the moderator, a second forum for the candidates and moderator to express themselves with citizens reading the discussion, and a third in which candidates, moderator, and citizens have full read and write access. As in previous electronic participation studies, such as the Televote study (Slaton, 1992), it will be important to note
a) the rate of participation (percentage of citizenry who communicate)
within the community,
b) the extent (volume) of participation of each participant,
c) the nature of political discourse among candidates, and
d) the responsiveness of candidates to citizens' messages.
Studies such as these will likely increase along with the popular adoption of computer-mediated communication. As these studies are conducted, effectively introducing political CMC to limited areas over time, it will also be important to determine the extent to which CMC is adapted to political discourse as well as the degree to which political discourse is actually changed by the adoption of CMC.
Physical access, however, is not the only means issue. Another concerns the ability to use and understand the communication technology. A societal ability to engage in CMC would rely on basic instructional resources and an orchestrated program of socialization for participation within children's and adult education systems. It is arguable that computer interfaces, the methods and practical models with which users manipulate information and make information choices, have become more "user-friendly." They are designed with socially intuitive symbols to make control of computer functions easier to understand and execute. It is also conceivable that, given time, the cost of a basic computer communication system will decrease further, perhaps to the level of today's "smart telephones," whose function computers might also replace. However, at this point in time, issues of physical access and technological ability still combine to pose thorny questions of how far a democratic system must go to satisfy criteria for popular democratic participation.
Diffusion researchers and political scientists may find common interest in examining the degree to which proliferation of interactive media correlates with the character of democratic participation.
A reasonable query would ask how a public agenda would be set within a CMC context, with so many diverse issues being discussed simultaneously, with each interest group organizing and deliberating. How will policy and media agendas be affected? Will we see a decrease in the incluence of media agendas upon the polled public agenda? Will it be appropriate or useful to derive a single public agenda? Sixty newspapers presently provide current events information on the internet. Will the representation of these non-interactive agenda-setting media entities on the internet affect the variety of perspectives expressed? Would the presence of partisan news services, including those belonging to political parties, on the internet influence the undecided or convert the opposition?
These concerns essentially converge on the question of whether the volume of messages and the number of participants drown neutralize the effects of messages because of a human inability to attend to a great portion of the traffic. It will be interesting to note whether fringe and disenfranchised groups end up with greater satisfaction at having a voice in the public forum or greater frustration in the event that they are ignored nonetheless in a din of electronic cacophony.
Both of these styles have different strengths in political discourse. On the one hand, male communication patterns, with their focus on independence, secure a measure of empowerment through assertion. In a communication forum in which a multitude of messages vie for the attention of decision-makers, those which are imperative and unambiguous may receive more attention than those which are conciliatory or otherwise construed as equivocal. On the other hand, in a political situation where there are many fractionated interests, there may be a distinct advantage to exhibiting interdependence which would cull alliances of shared interest. It would be worthwhile to see how men and women express themselves differently in pursuit of political empowerment.
On a related subject, a common practice in CMC is the use of pseudonyms. Pseudonym usage reduces feelings of personal risk when people self-disclose personal information (Matheson & Zanna, 1990). Thus, political participants might be more apt to speak up on issues they deem personally important and feel less danger of embarrassment. Pseudonym usage also allows communicants to further bypass constraints of social expectations, such as those of gender or when names denote ethnic origin. A general observation of CMC discourse is that communicants tend to be less inhibited than when they are in face-to-face situations (Kiesler, Siegel, & McGuire, 1984; Rice & Love, 1987).
In a study which compared mixed gender discourse in both pseudonymous and real-name computer bulletin boards, Jaffe, Lee, Huang & Oshagan (1994) concluded that men tended to exhibit greater social support and socioemotional language in pseudonymous CMC than in real-name CMC contexts. Furthermore, women in the pseudonymous context tended to choose pseudonyms which were either gender-neutral or cross-gender (male) while men exhibited no such cross-gender identification need.
Because CMC is less "media rich" than FTF communication, there is a greater ability for individuals to "manage" their identities. Individuals may reveal what they wish about themselves to serve social goals, including political goals. An individual might exhibit identifying characteristics of other participants for the sake of political influence. This would certainly fit in with persuasive communication research which explores the exploitation of in-group bias on the part of salespeople (Brewer, 1979; Tajfel, 1981). It would be interesting to note if the use of CMC would bring out political "chameleons" in the course of political discussion.
(a) previous enactments of the behavior or similar behaviors,
(b) vicarious experiences with the behavior as communicated through live or
symbolic (i.e., mediated) modeling,
(c) verbal persuasion regarding capabilities to engage in the behavior, and
(d) inferences from physiological states experienced when engaging in or
anticipating the behavior.
It would be intuitive to state that a person who was both accomplished at using CBBs and experienced in political participation would be more likely to adopt CMC as a medium for political communication than one who was not so initiated. This would certainly fit with condition (a) specified above. The same conditional relationship might predict that individuals who are experienced with CBBs in non-political contexts might feel self-efficacious in political computer forums though, perhaps, not in FTF communication. Vice-versa, people who are politically active, but inexperienced with CMC, might willingly and successfully adopt computer conferencing for the sake of political participation, but not otherwise.
The second condition might explain why individuals who take part in political conversation in the virtual forum of a CBB might develop a confidence to communicate politically in a non-virtual forum. Thus, instead of acting as a substitute for FTF political communication and activity, political CMC might actually promote, or supplement, more active face-to-face interaction for political purposes.
At this time, there is no clear-cut remedy for what has been termed "infoglut." Some research programs are underway to develop a process of helping CMC users through automatic prioritizing and cataloguing of messages based on sender and content, but this technology is not yet available. However, the impossibility of an individual, whether a representative or a constituent, to attend to every message on a political forum need not nullify CMC as a political information channel. Depending upon the information purpose of a participant, an electronic forum may be sampled for a limited duration. This might help a legislator or other representative develop estimates of public opinion. If a missed message contains extremely important information, it is likely that key concepts in that message may reverberate through subsequent responses, i.e., generate lengthy discussion or stimulate social organization. These ripple effects are more easily detectable than individual messages.
If CMC-based political participation becomes a reality, there will be exciting challenges for information and political scientists to develop new technologies and methodologies toward effective sampling of continuous political discourse.
An asynchronous, cues-filtered medium, such as a CMC bulletin board, presents an opportunity for participatory representation encompassing candidates, elected officials, and constituents. Whereas the permanent campaign will probably remain an occupational nuisance of the career politician, the conduct of the campaign can follow a more efficient and collaborative communication process. It is possible that wide scale adoption of such a political communication channel, and its effective use, might increase public trust in government.
In summary, CMC follows a parade of personal media have been used to produce political messages. However, the application of computer-mediated communication for political discourse has the potential to serve our present state of representative democracy or to significantly change it towards a more participatory system. For different reasons, the societal adoption of such a system of might significantly alter the national character of different social groups towards either unity or separatism. Much of this might hinge on whether different segments of society have equal access to and understanding of the communication technology involved in an electronic political forum. The conduct of election campaigning might also change as a result of a more direct, verbal, and asynchronous connection with the voting public. Finally, there is the question of how this communication technology might affect some of our basic cognitive, heuristic processes which govern how we interact socially and politically.
Bandura, A. (1986). Social foundations of thought and action. Englewood Cliffs, NJ: Prentice-Hall.
Blumer, H. (1948). Public opinion and public opinion polling. American Sociological Review, 13, 542-554.
Brewer, M. (1979). In-group bias in the minimal intergroup situation: A cognitive-motivational analysis. Psychological Bulletin, 86, 307-324.
Cole, D. (1994). Information evolution: Medium may change, but words remain the same. Quill, January.
Converse, P. J. (1987). Changing conceptions of public opinion in the political Process. Public Opinion Quarterly, Winter, pp. 12-24.
Festinger, L. (1957). A theory of cognitive dissonance. Stanford: Stanford University Press.
Ganley, G. (1992). The exploding political power of personal media. Norwood, NJ: Ablex.
Herring, S. C. (1993). Gender and democracy in computer-mediated communication. Electronic Journal of Communication/La Revue Electronique de Communication, On-line serial, 3:2. Available E-mail: COMSERVE@vm.its.rpi.edu . Message: SEND HERRING V3N293.
Hiltz, S. R. (1984). On-line Communities: A Case Study of the Office of the Future. Norwood, NJ: Ablex.
Hiltz, S. R., & Turoff, M. (1993). The Network Nation: Human Communication via Computer, Revised Edition (pp. 455-514). Cambridge, MA: MIT Press.
Jaffe, J. M., Lee, Y., Huang, L., & Oshagan, H. (1994). Gender, pseudonyms, and CMC: Masking identities and baring souls. Unpublished study currently submitted to the 45th Annual Conference of the International Communication Association (ICA).
Kaplan, N., & Farrell, E. (1994). Weavers of webs: A portrait of young women on the net. The Arachnet Journal on Virtual Culture, 2:3. Available e-mail: listserv@kentvm.kent.edu. Message: GET KAPLAN V2N3 . Available gopher: gopher.cic.net .
Kiesler, S., Siegel, J., & McGuire, T. W. (1984). Social psychological aspects of computer-mediated communication. American Psychologist, 39, 1123-1134.
Matheson, K., & Zanna, M. P. (1992). Computer-mediated communications: The focus is on me. Social Science Computer Review, 8:1, 1-12.
Miller, G. (1956). The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63:2, 81-96.
Rafaeli, S. (1986). The electronic bulletin board: A computer-driven mass medium. Computers and the Social Sciences, 2, 123-136.
Rafaeli, S. (1988). Interactivity: From new media to communication. In R. P. Hawkins, J. M. Wiemann, & S. Pingree (Eds.), Advancing communication research: Merging mass and interpersonal processes (pp. 110-134). Newbury Park, CA: Sage.
Rheingold, H. (1993). The virtual community: Homesteading on the electronic frontier. Reading, MA: Addison-Wesley.
Rice, R. E., & Love, G. (1987). Electronic emotion: Socioemotional content in a computer-mediated network. Communication Research, 14, 85-108.
Rogers, E. M. (1983). The diffusion of innovations. New York: Basic Books.
Selfe, C. A., & Meyer, P. R. (1991). Testing claims for on-line conferences. Written Communication, 8(2), 163-192.
Slaton, C. D. (1992). Televote: Expanding citizen participation in the quantum age. New York: Praeger.
Splichal, S. (1993). Searching for new paradigms: An introduction. In S. Splichal & J. Wasko (Eds.), Communication and democracy. Norwood, NJ: Ablex.
Sproull, L., & Kiesler, S. (1986). Reducing social context cues: Electronic mail in organizational communication. Management Science, 32, 1492-1512.
Tajfel, J. (1981). Human groups and social categories. London: Cambridge University Press.
Tannen, D. (1990). You Just Don't Understand: Woman and Men in Conversation. New York: William Morrow.
The Ottowa Citizen (11/9/94). p. E6
Walther, J. B. (1992). Interpersonal effects in computer-mediated interaction: A relational perspective. Communication Research, 19:1, 52-90.
Walther, J., & Burgoon, J. K. (1992). Relational communication in computer-mediated interaction. Human Communication Research, 19:1, 50-88.
Click here to access J. Michael Jaffe's Home Page