__I__t's still magical. Twenty-five years since electronic mail first traversed the Net, we can't explain to outsiders our pleasure in it, so we resort to analogy. E-mail, we say, is like a phone call, only written. It's like a paperless letter, but faster. It's like a casual conversation, or a postcard, or a scribbled note to a schoolmate in the middle of class. It's all of the above, and, considering its ethereal nature, none of them; whoever has printed out a pleasing e-mail message discovers that on the page it is rendered lifeless, like an exquisite tropical fish removed from its tank.
Beguiled by its pro-tean nature, a surprising array of boosters trumpet the social impact of electronic mail. Bruce Redford, a Johnsonian scholar at the University of Chicago (Dr. Samuel Johnson was an 18th-century lexicographer and critic), exults that "the advent of electronic mail is fostering a revival of "the familiar letter." Research-ers proclaim e-mail's democratizing effects, citing substantial evidence that it flattens hierarchies, promotes teamwork, and increases involvement of peripheral workers within organizations. Political organizers herald its usefulness in mobilizing support for campaigns around the globe. Management consultants see it as the "enabling technology" for corporate reform, linking employees and information so efficiently that the ranks of middle managers can be dramatically thinned. E-mail-system designer Nathaniel Borenstein goes so far as to assert that because it enables workers to labor outside their place of employment, e-mail "represents the next quantum step towards human freedom."
The growth of e-mail is undeniable. The Electronic Messaging Association, which represents 400 leading vendors of e-mail technology, estimates that between 30 and 50 million people use e-mail. Walter Ulrich, an EMA co-founder and board member for the last eleven years, says that in the busi-ness sector alone, North American users have increased from one million in 1984 to 16 million in 1993, when they sent the ample sum of five or six billion messages. Assuming that each message was 500 words long - not an unreasonable estimate, since some messages are many times that length - business users sent the equivalent of 10,000 manuscripts the length of War and Peace every day of the year. And that is mere prologue, for Ulrich expects the number of business e-mail messages to quintuple in the next decade. According to International Data Corporation, a Framingham, Massachusetts-based research firm, at least a portion of the workforce at every Fortune 500 company uses e-mail, while only 80 percent of the companies have installed voice-mail systems.
Yet for all its alleged virtues, e-mail is a surprisingly quirky medium, so packed with idiosyncrasies that, despite a decade of effort, academic researchers have yet to devise a theoretical structure for understanding its impact on organizations. E-mail is written, yet its language typically embodies a shift toward oral speech patterns. It is the most ephemeral of written mediums, lacking the material form of books or letters and capable of being erased in a keystroke, yet it can be archived and retrieved with unprecedented ease. It has occasioned an astonishing effusion of warm-hearted social interaction outside work settings, yet at work it is the medium of choice for employees who don't like each other and wish to minimize their interactions. It is the glue that holds many virtual communities together, but some corporate managers report that they must supplement e-mail messages to subordinates with occasional phone calls to maintain personal relationships.
And e-mail can just as easily promote workplace centralization as decentralization, for while workers may find that e-mail gives them unaccustomed access to resources throughout their company, their superiors enjoy enhanced means of tracking their performance.
A Multivalent Character
It is no accident that e-mail has had the greatest impact on the business domain, for e-mail has emerged as a useful medium just as many corporations are facing extreme pressure to increase productivity. One result of that pressure has been the popularity of "business process redesign," a formula for using information technology to reconstitute the corporation. The idea, touted in Michael Hammer and James Champy's best-selling Reengineering the Corporation and embraced by corporations from Bell Atlantic to Taco Bell, is essentially that corporations must rearrange themselves around the capabilities of information technology - with e-mail at its core - instead of trying to fit the technology into existing organization structures. Since that usually translates into a prescription for eliminating middle managers, thus reducing costs, many corporations have taken to the concept with enthusiasm.
E-mail, however, need not be part of a grand reorganization scheme for its impact to be felt. David Whitten, a program director for office information systems at the Stamford, Connecticut-based Gartner Group, cites the example of a multinational corporation that received numerous complaints from customers that its salespeople were acting as if they were in competition with one another. When customers asked salespeople for information about products represented by co-workers within the company, the salespeople frequently failed to pass on the queries, sabotaging the interests of both the company and its customers. As it turned out, the failures resulted chiefly from the cumbersome nature of information transmittal within the corporation: The salespeople often didn't know who represented other product lines, and the process of filling out and sending paper memos to co-workers was laborious, so more often than not they didn't bother.
Once e-mail was instituted, however, the problem was solved. After meeting customers, salespeople filled out a simple electronic form, citing whether the customer expressed interest in other products; the form was then automatically sent to co-workers via e-mail. The new process was so easy to carry out and the benefits to salespeople were so obvious that the system "was universally adopted within a couple of months, and the sales force began referring to itself as 'collaborative' instead of 'competitive,'" Whitten says.
Ulrich, the head management consultant in the Los Angeles office of Arthur D. Little, Inc., provides a different kind of example, in which a company faced a suddenly announced hostile takeover attempt while its executives were scattered around the globe. Since the bid was announced on a Friday, the executives had until the opening of financial markets on Monday morning to construct an effective counter-strategy. Lacking time to return to the home office for a meeting, they collaborated by e-mail and succeeded in fending off the bid.
The only trouble with success stories like these is that while they demonstrate the power of e-mail, they do nothing to illuminate its complex impact on corporate climates, and therefore they may be slightly misleading. The distinction is what Lee Sproull and Sara Kiesler, the most prominent of academic e-mail researchers, call the difference between "first-level effects" of technology - the anticipated changes such as the multinational's improved information flow once its salespeople began using e-mail - and "second-level effects" - the unintended consequences, good and bad, of use of the technology. The distinction, which Sproull and Kiesler stress in their influential 1991 book, Connections: New Ways of Working in the Networked Organization, is a vital one, for it points to the areas of controversy that currently dominate research work on e-mail. Most researchers seem to agree on e-mail's first-level effects - indeed, like Sproull and Kiesler, they are e-mail enthusiasts, who are persuaded of its value to organizations and use the medium in their own professional communications - but they are nowhere close to a consensus about its secondary effects. Of course, it's the elusive and variegated nature of those effects that gives e-mail research its spice.
For instance, one thrust of Sproull and Kiesler's findings revolves around the paucity of information e-mail users receive about one another: Unable to see and hear fellow users' responses, they know less about their cohorts than they would in a face-to-face setting and are therefore likely to be more uninhibited. Thus, Sproull and Kiesler argue, e-mail users are more likely to resort to the sort of scurrilous name calling known as "flaming," more likely to be candid, and less likely to be dominated by people with high status, since such evidence of it as expensive clothes or an authoritative speaking manner are absent. Sproull and Kiesler cite studies buttressing all of these assertions. The last point, for example, is borne out in a study showing that when groups of executives met face-to-face, the men in the groups were five times as likely as the women to make the first decision proposal, whereas when the groups met via computer, the women made the first proposal as often as did the men.
What, then, can we make of a study carried out in 1987 by M. Lynne Markus, an associate professor of information science at the Claremont Graduate School? Markus surveyed managers at a large insurance company that had quite purposefully made e-mail its internal medium of choice four years earlier. E-mail addresses were printed on corporate organization charts and telephone directories, and all but one of the senior executives Markus interviewed claimed to be heavy users, exchanging 70 to 250 messages a day. Many managers said they "couldn't live" without e-mail; strong plurality (42 percent) favored e-mail over any other means of communication, including face-to-face (32 percent). The telephone, on the other hand, was widely regarded as a waste of time; some managers even reported that they accepted phone calls from subordinates only if the subordinates first justified the reason for calls in e-mail.
Yet e-mail behavior in the insurance company repeatedly contradicted Sproull and Kiesler's findings. For one thing, flaming was rare - employees clearly understood that in a work setting it was inappropriate. And contrary to Sproull and Kiesler's notion that e-mail promotes candor, managers reported that they were quite conscious of the ease of archiving and forwarding e-mail to superiors, and were therefore careful not to put politically sensitive requests into e-mail. Many managers contended that e-mail was valuable for all work-related communication except personnel issues, precisely because e-mail did not support the level of confidentiality personnel discussions required. The managers even found that e-mail was the medium of choice whenever they had to report an unpleasant fact or communicate with someone they didn't like; the same medium that encouraged candor in Sproull and Kiesler's studies was useful to the managers because it offered the safety of social distance. Finally, Markus found no evidence of diminution in the status of high-ranking executives in e-mail interactions. E-mail, Markus says in an interview, "is as good at preserving the existing social order as it is at breaking down barriers."
Like Sproull and Kiesler, Markus is considered a preeminent researcher, so the conflict almost certainly isn't attributable to poor study design. Instead, the explanation seems to lie in the varying milieus in which the studies took place. Sproull and Kiesler's study subjects were often students in an academic setting, where hierarchy is less pronounced than in a corporation and sanctions for misbehavior may be far less severe; the different norms of the respective milieux seemed to influence the way in which e-mail was used. "E-mail isn't just one thing - it has a multivalent character," says Rob Kling, professor of information and computer science at the University of California, Irvine. "Part of it falls under the norms and behavior of conversation, and another part falls under the norms and behavior of text." It's plausible that Sproull and Kiesler's students assumed they were having a conversation, whereas Markus' managers thought of themselves as exchanging memos for a file.
Fostering a Global Academy
Even within the academic setting, e-mail's impact seems to vary dramatically depending on the discipline. A 1993 study by Sproull, Kiesler, and two other scholars at Carnegie Mellon University describes the benefits of e-mail and computer networks to oceanographers. "Frequent users are the more active, productive scientists," the study says. "Scientists who use the network more also produce more papers, receive more professional recognition from their peers, and know more physical oceanographers."
Many of the reasons for the correlation have to do with the requirements of the discipline. Oceanographers must collect and coordinate data from remote ocean areas, often depending on large-scale projects involving hundreds of scientists from many countries. Terabyte-size databases are common, and supercomputers are needed to do analysis. The utility of e-mail and networks in overseeing data collection and analysis is obvious. Perhaps less obvious is the technology's impact on oceanographers' careers: Because the data is more widely available and oceanographers can more easily contact each other (thanks to telecommunications), a greater number of them can participate in the research effort. The information that previously was available only to oceanographers at prestigious institutions, usually those located on the Atlantic or Pacific coasts, is now available to all.
The result is that "the people who would normally be considered out of the loop, those at smaller or inland institutions and more recent PhDs, are better off to the extent that they use e-mail," says John Walsh, now an assistant professor of sociology at the University of Illinois, Chicago and one of the study's authors. The oceanographers who got their PhDs at major research institutions and then found jobs at less prestigious academies formerly found it impossible to continue doing research; now, Walsh says, they can do research because they have access to their mentors and data via e-mail.
Some other scientific disciplines also favor e-mail, according to Walsh. Experiments in particle physics, for example, are frequently vast, sometimes involving 200 or more geographically dispersed participants, so e-mail is obviously useful. Mathematics is so dominated by specialization that university departments usually hire no more than one scholar per specialty; thus, mathematicians rely on e-mail because their fellow specialists are almost invariably located at other universities. On the other hand, chemists and biologists seldom use e-mail. Their experiments are often small, requiring little monitoring while they are unfolding, and the information they develop may be proprietary. In addition, biological data are often visual, so biologists have grown accustomed to communicating by fax, which can transmit images, instead of e-mail.
Edward Mabry, a communication researcher at the University of Wisconsin, Milwaukee, says that in his field, e-mail has even extended the concept of collegiality. "Historically, the strength of an academic department rested with its resident faculty," he says. "Now it depends on the extent to which each faculty member is interconnected with other professionals - worldwide - pursuing similar interests. And these associations do not rely on face-to-face contact. We now have electronic research teams and electronic water coolers. This dras-tically changes - weakens, in my opinion - indigenous workplace relationships and affects workplace cohesiveness."
E-mail's frequently double-edged impact is perfectly illustrated in the successful 1990 effort to factor a large, theoretically interesting number called the ninth Fermat number. As Rand Corporation researcher Tora Bikson writes, "Mathematicians and logicians had long believed [the number] was calculable because it could not be demonstrated to be noncalculable, but it required so much in the way of human and computer resources that it had never been attempted." Scientists at Bell Communications Research then developed a program to split up the project and put out a call on the Internet to recruit researchers. E-mail was used both to distribute the various parts of the problem and to send back the solutions. The pieces were fitted together, and the problem was solved. However, the paper announcing the results contained a fascinating admission: We'd like to thank everyone who contributed computing cycles to this project, but we can't: We only have records of the person at each site who installed and managed the code. If you helped us, we'd be delighted to hear from you; please send us your name as you would like it to appear in the final version of the paper.
Political Punch
If outsiders have not fully appreciated the impact of e-mail in the academic world, the converse is probably true in the political realm, where boosters have made grandiose claims for e-mail's value during crises like the 1989 Tiananmen Square uprising in Beijing and the 1990 Soviet coup attempt. Yet while e-mail did play a modest role in enabling worldwide dissemination of some eyewitness accounts, no evidence exists that it altered the course of either event. At the same time, e-mail has proved to be a valuable tool in marshaling interest groups for sharply focused political campaigns.
For example, just after the Tiananmen Square massacre, Chinese students in the United States realized they might be in danger if forced to return to China when their American visas expired. Using e-mail and a newsgroup on the international computer conferencing system known as Usenet, they quickly organized to gain Congressional passage of protective legislation. The clearest demonstration of e-mail's power occurred when the students' lobbying committee decided to present a survey of thousands of Chinese students' opinions to a House hearing scheduled four days later. The committee sent a questionnaire via e-mail to coordinators of local student organizations at universities throughout the country. The coordinators promptly held meetings at which students filled out the questionnaires, and the results were tabulated and e-mailed back to the committee within four days, in time for the hearing. Tiger Li, a Michigan State University graduate student who wrote a scholarly article about the Chinese students' use of computer communications, concluded that without e-mail and networks, the students "could not have been organized to engage successfully" in the lobbying campaign and related activities.
An even more dramatic example of e-mail's usefulness took place in Moscow last fall, after Russian President Boris Yeltsin suspended parliament. On the night of October 3, Yeltsin's opponents responded with a disastrous nighttime attempt to storm Moscow's main television station, and rumors circulated that they might also mount an attack against a smaller TV station. As later reported on a Russian computer network called GlasNet, three prominent Russian labor leaders, though opposed to Yeltsin, feared that a second assault would be as bloody and futile as the first; they rushed to a meeting of their comrades to try to head them off. Moscow police arrested them soon after they left the meeting and charged them with stealing an automobile.
Over the next day they were repeatedly kicked and beaten with rubber batons and steel rifle butts. When word of their fate reached the wife of one of the men, she called Vassily Balog, the deputy head of Russia's major labor confederation and an active e-mail user. Balog immediately sent e-mail to friends in the United States and United Kingdom and posted an appeal on GlasNet that was quickly distributed to other networks around the world. The appeal listed the phone number of the police station where the three men were being held and urged readers to call it to protest the arrests.
Renfrey Clarke, an Australian journalist who filed stories about GlasNet from Moscow, estimates that the first phone calls reached the police station within 30 minutes of Balog's appeal. Clarke reported that one of the arrested men said afterwards, "We were watching from the cell as the phone calls came in. One of the first was from Japan. The police didn't seem able to believe it. After that the phone ran hot. The calls seemed to be coming from everywhere - there were quite a few from the San Francisco Bay Area in the US. In the end, the police started saying we'd already been released, but we were shouting through the bars, 'No! No! We're still here!' " Made nervous by the international spotlight suddenly cast upon the station, within a few hours the police released the three men and many others detained in the same incident.
Balog believes that postings on computer networks around the world about the events in October saved many lives by "bringing strong pressure from international public opinion on the Russian authorities." With an information blockade in effect, he said, and "unconditional support from the official West for the actions of Mr. Yeltsin's government, e-mail provided a vital means of communication that was beyond official censorship and control." Of course, occasionally the town crier may spread false news or precipitate a response other than the intended one. E-mail may have had precisely that effect when it was used to disseminate news of the October 1993 kidnapping of Polly Klaas, a 12-year-old Petaluma, California girl who was later found murdered. Images of Klaas and her abductor were distributed around the world on the Internet and hundreds of computer bulletin boards, and Petaluma police were deluged with thousands of tips. Before Klaas and her alleged killer were located, Kling, the University of California, Irvine professor, said presciently, "To the extent that taking the time to sort out bogus leads from way outside of the area overwhelms or distracts the Petaluma police, it should make it harder to find this girl." The leads didn't help, for when the suspected murderer was located two months after the abduction, he was within 30 miles of Petaluma.
New Oral Culture: Permanent and Permeable?
One consequence of the computer age is that most of us have experienced the truth of the dictum that all new technologies appear magical at first. As Brenda Denat, a sociologist at The Hebrew University of Jerusalem, explains in a brilliant unpublished essay about computers' impact on text, "The metaphor of 'magic' is summoned to give expression to new, unsettling experiences which eventually become routin-ized." Yet for all its newness, its apparent magic, in some ways e-mail reaches backwards. Denat points out that in mid-19th century London, for example, postal deliveries occurred every hour for twelve hours a day, and correspondents could exchange several messages a day, just as e-mail users do now. Or consider that the language of e-mail - so laden with abbreviations and constructions like "gonna" and "gotta," so tolerant of spelling and grammatical errors - draws heavily on spoken usage. In that respect e-mail takes us back before the dawn of writing itself, to a time when knowledge was transmitted orally - it's as if e-mail reflects a yearning to trade in the formality of writing for the spontaneity of speech. As Denat says, the impulse that led to writing was partly a desire to make permanent what in oral language was permeable and ephemeral. Now, in e-mail, we've devised a written medium that partly undermines the assumptions of writing, that evokes the uncertainties of oral culture. Getting used to e-mail, it turns out, may mean accepting its vast capacity to sow ambiguity.
