Cyber-race

Harvard Law Review

Volumn 113, Issue 5, 2000, Pages 1131-1208

  Kang, Jerry ^a  

^a UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0346053614     PISSN: 0017811X     EISSN: None     Source Type: Journal    
DOI: 10.2307/1342340     Document Type: Article

References (405)

1. See KENNETH T. JACKSON, CRABGRASS FRONTIER: THE SUBURBANIZATION OF THE UNITED STATES 249 (1985);
2. CLIFFORD STOLL, SILICON SNAKE OIL: SECOND THOUGHTS ON THE INFORMATION HIGHWAY 3, 49 (1995).
3. See Jerry Frug, The Geography of Community, 48 STAN. L. REV. 1047, 1068 (1996)
4. (noting that the suburban boom could not have happened without the Federal-Aid Highway Act of 1956); see also GEORGE M. SMERK, URBAN TRANSPORTATION: THE FEDERAL ROLE 266 (1965) (describing the loss of urban political power and tax revenue that resulted from the population shift to rural areas).
5. See Frug, supra note 2, at 1065 ("[T]he rapid growth of the suburbs, fostered . . . by government programs, provided a secure place to which whites could flee. As a result, African Americans are segregated today in a manner that no other minority in the United States is now or has ever been segregated."); id. at 1069 (describing how highway construction has led to Whites resettling in the suburbs and Blacks resettling in the ghettos, with the highways themselves often acting as the physical boundary between communities); Florence Wagman Roisman, Intentional Racial Discrimination and Segregation by the Federal Government as a Principal Cause of Concentrated Poverty: A Response to Schill and Wachter, 143 U. PA. L. REV. 1351, 1334 n.11 (1995) (suggesting that highway construction and mass transit policies exacerbated residential segregation).
6. See, e.g., JAMES DAY, THE VANISHING VISION: THE INSIDE STORY OF PUBLIC TELEVISION 23 (1995) (describing how enthusiasts "envisioned an opportunity to propagate the best in teaching by using television to share the best teachers").
7. Newton Minow, Address to the National Association of Broadcasters (May 9, 1961), in NEWTON N. MINOW & CRAIG L. LAMAY, ABANDONED IN THE WASTELAND: CHILDREN, TELEVISION, AND THE FIRST AMENDMENT 185, 188 (1995).
8. See infra pp. 1167-68.
9. For another comparison between the interstate highway system and cyberspace, see Steven G. Jones, Understanding Community in the Information Age, in CYBERSOCIETY: COMPUTER-MEDIATED COMMUNICATION AND COMMUNITY 10, 10-11 (Steven G. Jones ed., 1995).
10. For a related discussion of the automobile, see DOUGLAS SCHULER, NEW COMMUNITY NETWORKS: WIRED FOR CHANGE 178 (1996).
11. See infra pp. 1162-64. See generally Donna L. Hoffman & Thomas P. Novak, Bridging the Racial Divide on the Internet, 280 SCIENCE 390, 390-91 (1998) (investigating differences in computer ownership and Web usage among racial groups).
12. Ian Ayres has demonstrated statistically significant differences in the average offers received by car purchasers as a function of gender and race (White versus Black). See Ian Ayres, Fair Driving: Gender and Race Discrimination in Retail Car Negotiations, 104 HARV. L. REV. 817, 817-19 (1991);
13. Ian Ayres, Further Evidence of Discrimination in New Car Negotiations and Estimates of Its Cause, 94 MICH. L. REV. 109, 109-10 (1995)
14. [hereinafter Ayres, Further Evidence].
15. I immigrated to the United States when I was six and grew up in the Midwest. Like everyone, immigrants and non-immigrants alike, I have an accent. See generally Mari J. Matsuda, Voices of America: Accent, Antidiscrimination Law, and a Jurisprudence for the Last Reconstruction, 100 YALE L.J. 1329 (1991) (discussing accent discrimination).
16. For a description of graphical virtual communities and other forms of cyberspace communications, see infra Part II.
17. Unfortunately, I could not capture the entire log. An unedited transcript, which includes simultaneous conversations with other parties, is available from the Harvard Law School Library upon request.
18. A classic nonfiction example of real-space racial passing appears in JOHN HOWARD GRIFFIN, BLACK LIKE ME (1961), which tells the story of a White man passing as a Black through the deep South in late 1959.
19. See, e.g., Lee Gomes, A Giant Game of "I Spy" on the Web: Netcams Bring Videoconferences Into Bedrooms, WALL ST. J., July 19, 1999, at B1 (describing increasing sales of "netcams," which are fueling video exchanges - many of them X-rated - on the Internet).
20. As Mark Dery puts it, technology is "a Janus machine, an engine of liberation and an instrument of repression." MARK DERY, ESCAPE VELOCITY: CYBERCULTURE AT THE END OF THE CENTURY 14 (1996).
21. One repressive possibility is that cyberspace may become a tool for mass surveillance by the private sector. See generally Jerry Kang, Information Privacy in Cyberspace Transactions, 50 STAN. L. REV. 1193 (1998) (analyzing the increased threat to information privacy caused by cyberspace).
22. For example, the difference in access to technology may increase socioeconomic differences between Whites and non-Whites; the lack of accountability in certain online social spaces may increase hate speech and racist rhetoric; the possibility of transmutation, see infra Part V, may reinscribe racist meanings through blackface-like minstrelsy, see infra p. 1184.
23. Gabriel J. Chin, Sumi Cho, Jerry Kang & Frank Wu, Beyond Self-Interest: Asian Pacific Americans Toward a Community of Justice, A Policy Analysis of Affirmative Action, 4 UCLA ASIAN PAC. AM. L.J. 129, 160 (1996).
24. A fine introduction to Critical Race Theory can be found in CRITICAL RACE THEORY: THE KEY WRITINGS THAT FORMED THE MOVEMENT at xiii-xxxii (Kimberlé Crenshaw, Neil Gotanda, Gary Peller & Kendall Thomas eds., 1995).
25. Within any dynamic intellectual tradition, it is hard to identify standard methods or principles. See, e.g., Richard Delgado & Jean Stefancic, Critical Race Theory: An Annotated Bibliography, A Year of Transition, 66 U. COLO. L. REV. 159, 160-61 (1995) (attempting to identify Critical Race Theory's principal themes). The reader, however, should be aware that some of the method, approach, and argument I present in this paper may be inconsistent with other works within this school of thought For example, certain critical race theorists may object strongly to the idea of implementing a strategic delay in the disclosure of race in social spaces. See infra pp. 1172-73. I claim no orthodoxy.
26. See, e.g., M. ETHAN KATSH, LAW IN A DIGITAL WORLD 3-20, 92-113 (1995) (discussing culture and interaction in cyberspace);
27. James Boyle, Foucault in Cyberspace: Surveillance, Sovereignty, and Hardwired Censors, 66 U. CIN. L. REV. 177, 177 (1997)
28. (critiquing digital libertarianism and the current preference for technological rather than legal solutions); James Boyle, A Politics of Intellectual Property: Environmentalism for the Net?, 47 DUKE L.J. 87, 87 (1997);
29. Julie E. Cohen, Lochner in Cyberspace: The New Economic Orthodoxy of "Rights Management", 97 MICH. L. REV. 462, 462-563 (1998) (discussing economic models for governing cyberspace);
30. Lawrence Lessig, Reading the Constitution in Cyberspace, 45 EMORY L.J. 869, 869-910 (1996)
31. (addressing the problem of applying the Constitution in cyberspace); David G. Post, Anarchy, State, and the Internet: An Essay on Law-Making in Cyberspace, 1995 J. ONLINE L. art. 3 (June 1995) 〈http://www. wm.edu/law/publications/jol/post.html〉 (on file with the Harvard Law School Library).
32. See generally ROBERT C. ELLICKSON, ORDER WITHOUT LAW: HOW NEIGHBORS SETTLE DISPUTES 123-259 (1991) (presenting a theory of norms).
33. In the legal literature, this multifaceted approach is sometimes called the New Chicago School. See Lawrence Lessig, The New Chicago School, 27 J. LEGAL STUD. 661, 662-63 (1998)
34. (describing the four types of constraints that regulate behavior). For an illuminating application of this approach to cyberspace, see Lawrence Lessig, The Law of the Horse: What Cyberlaw Might Teach, 113 HARV. L. REV. 501 (1999)
35. [hereinafter Lessig, The Law of the Horse].
36. Cybersocial theory has given special emphasis to architecture. See Joel R. Reidenberg, Lex Informatica: The Formulation of Information Policy Rules Through Technology, 76 TEX. L. REV. 553, 569 tbl.1 (1998) (comparing and contrasting traditional legal regulation and "Lex Informatica," "law" implemented through technological architecture).
37. I briefly discuss universal service policies, which have strong class components, in Parts IV.A and VI.C.2(a). Gender analogues permeate the footnotes. Both subjects, however, obviously deserve serious separate examination in the context of cyberspace. In the future, race can and should be studied in unison with gender and class. Examples of such unified discussion can be found in fields outside of cyberspace. See, e.g., Richard H. Fallon, Jr., Affirmative Action Based on Economic Disadvantage, 43 UCLA L. REV. 1913, 1947-50 (1996)
38. (discussing the use of economically based affirmative action in place of race-based affirmative action); Cheryl I. Harris, Finding Sojourner's Truth: Race, Gender, and the Institution of Property, 18 CARDOZO L. REV. 309, 312-24 (1996) (discussing how slavery as property structured the social and legal meanings of race and gender).
39. See RALF DAHRENDORF, LIFE CHANCES: APPROACHES TO SOCIAL AND POLITICAL THEORY 28, 31, 34 (1979)
40. (discussing the concept of life-chances); ANDREW HACKER, TWO NATIONS: BLACK AND WHITE, SEPARATE, HOSTILE, UNEQUAL 99-111 (2d ed. 1995) (comparing the earnings of Black Americans to White Americans);
41. Paul Ong & Suzanne J. Hee, Economic Diversity, in THE STATE OF ASIAN PACIFIC AMERICA: ECONOMIC DIVERSITY, ISSUES & POLICIES 31, 31-56 (Paul Ong ed., 1994)
42. (comparing the earnings of Asian Americans to Whites); see also Pat K. Chew, Asian Americans: The "Reticent" Minority and Their Paradoxes, 36 WM. & MARY L. REV. 1, 8-53 (1994) (suggesting that despite popular conceptions, Asian Americans still face discrimination).
43. In 1980, the percentage of Blacks who married nonblacks was 2%. The percentage of intermarriage for Latinos was 13%, for Asians 25%, and for Jews 50%. See KAREN BRODKIN, How JEWS BECAME WHITE FOLKS: AND WHAT THAT SAYS ABOUT RACE IN AMERICA 74-75 (1998).
44. See discussion and sources cited supra note 9 (discussing racial discrimination in car purchases).
45. For example, President Clinton won 84% of the Black vote and 72% of the Hispanic vote in the 1996 election. See Maria Puente, Clinton Sees Gain in Hispanic Vote, Retains Black Vote, USA TODAY, Nov. 7, 1996, at 5A.
46. Asian Americans voted overwhelmingly against California's Proposition 209, which ended affirmative action in state education, contracting, and employment. Exit polls in Los Angeles County revealed that 76.2% voted against the Proposition; exit polls in the San Francisco Bay Area revealed that 84.1% voted against the Proposition. See ASIAN PACIFIC AMERICAN LEGAL CENTER OF SOUTHERN CALIFORNIA & NATIONAL ASIAN PACIFIC AMERICAN LEGAL CONSORTIUM, 1996 SOUTHERN CALIFORNIA ASIAN PACIFIC AMERICAN EXIT POLL REPORT: AN ANALYSIS OF APA VOTER BEHAVIOR AND OPINIONS 13 (1999).
47. See, e.g., CHARLES J. OGLETREE, JR., MARY PROSSER, ABBE SMITH & WILLIAM TALLEY, JR., BEYOND THE RODNEY KING STORY: AN INVESTIGATION OF POLICE CONDUCT IN MINORITY COMMUNITIES 13 (1995) (demonstrating disproportionate African American representation "in every aspect of the criminal system as offenders, victims, prisoners, and arrestees").
48. For a literature review and critique of various traditional theories of race, see MICHAEL OMI & HOWARD WINANT, RACIAL FORMATION IN THE UNITED STATES: FROM THE 1960s TO THE 1990s (2d ed. 1994). See id. at 9-47 (describing ethnicity, class, and nation-based theories of race).
49. See generally ZIVA KUNDA, SOCIAL COGNITION: MAKING SENSE OF PEOPLE (1999)
50. (surveying the social cognition field). Readers new to the race literature should not assume that Critical Race Theory generally adopts social psychological explanations of race. Two prominent pieces that do use social psychological research, however, are Linda Hamilton Krieger, The Content of Our Categories: A Cognitive Bias Approach to Discrimination and Equal Employment Opportunity, 47 STAN. L. REV. 1161 (1995);
51. and Charles R. Lawrence III, The Id, the Ego, and Equal Protection: Reckoning with Unconscious Racism, 39 STAN. L. REV. 317, 336-39 (1987).
52. See Leonard Lieberman, Raymond E. Hampton, Alice Littlefield & Glen Hallead, Race in Biology and Anthropology: A Study of College Texts and Professors, 29 J. RES. IN SCI. TEACHING 301, 315 (1992).
53. See Roger Sanjek, Intermarriage and the Future of Races in the United States, in RACE 103, 109 (Steven Gregory & Roger Sanjek eds., 1994)
54. (noting that "street-level discourse" commonly identifies the racial categories as White, Black, Hispanic, Asian, and American Indian). Historically, anthropologists, sociologists, and biologists have identified anywhere between a handful to hundreds of different races. See GORDON ALLPORT, THE NATURE OF PREJUDICE 110 (1958) (noting that anthropologists have identified between two and 200 races);
55. STEPHEN CORNELL & DOUGLAS HARTMANN, ETHNICITY AND RACE: MAKING IDENTITIES IN A CHANGING WORLD 21 (1998) (noting that scientists have identified between three and 63 races).
56. Genetic differences do not justify racial categories. No racial category features a unique gene or substantially differs genetically from the others. See CORNELL & HARTMANN, supra note 31, at 22-23 ("[T]he extent of genetic variation among individuals within supposed racial groups typically exceeds the variation between the groups."). Moreover, the genetic differences among races are of degree, not kind, changing smoothly as a function of ancestral geography. See Roger Sanjek, The Enduring Inequalities of Race, in RACE, supra note 31, at 1, 7
57. ("Small local populations vary slightly from each other as one proceeds east to west from East Asia to western Europe, or north to south from Scandinavia to the Congo basin. Historical movements of peoples, and intermating of populations, complicate but do not disguise the fundamentals of continuous, clinal distribution."). To be sure, some average genetic differences - in addition to environmental factors that trigger gene expression - account for average phenotypical differences in skin color, hair texture, and facial features, such as the eyes and nose. However, it is scientifically arbitrary to categorize Homo sapiens on the basis of these relatively trivial physical characteristics when myriad other differences, some visible (e.g., eye color), some invisible (e.g., blood type), are ignored. Accordingly, "there no longer exists a consensus on the scientific accuracy and utility of" race. Lieberman, Hampton, Littlefield & Hallead, supra note 30, at 317; see also Thomas A. LaVeist, Why We Should Continue to Study Race . . . but Do a Better Job: An Essay on Race, Racism and Health, 6 ETHNICITY & DISEASE 21, 21 (1996) (conceding that "physical anthropologists currently do not recognize the validity of race"). However, this lack of scientific consensus does not prevent scientists from continuing to use "race," sometimes sloppily. See LaVeist, supra, at 21-22.
58. See KUNDA, supra note 29, at 17-18. "As we encounter objects, people, or behaviors, we classify them as instances of particular concepts. Much of this is done automatically - we do not see an elongated yellow object with black stripes, we see a banana." Thomas A. LaVeist, EId. (describing generally one of the functions of concepts, also called schemas). Schemas (not only racial ones) are elements of a conceptual or cognitive structure necessary to process in real-time the sensory stimuli that constantly flood our minds. As Allport explains, the human tendency to generalize and create categories is natural. "Life is so short, and the demands upon us for practical adjustments so great, that we cannot let our ignorance detain us in our daily transactions." ALLPORT, supra note 31, at 9.
59. See Kang, supra note 15, at 1232-37 (discussing the role of transaction facilitators in cyberspace). Facilitators "help execute [a] transaction but are not the principal drivers of the exchange." Id. at 1232.
60. In both the Late-night Walk and the Personal Ads stories, the mapping is unambiguous, but this is not always the case. For instance, a person's looks may not neatly match a racial archetype, perhaps because she is "multiracial." See, e.g., SHIRLEE TAYLOR HAIZLIP, THE SWEETER THE JUICE 13 (1994)
61. In both the Late-night Walk and the Personal Ads stories, the mapping is unambiguous, but this is not always the case. For instance, a person's looks may not neatly match a racial archetype, perhaps because she is "multiracial." See, e.g., Shirlee Tatlor Haizlip, The Sweeter The Juice 13 (1994)
62. (recounting a personal family history that reveals the complexities of living at the boundaries of socially constructed racial categories). Haizlip writes: I am a black woman, but many of you would never know it. . . . My skin is as light as that of an average white person. . . . But . . . [my] psych[e], soul[], and sensibilities are black. Sociologists would say [my family members] have been "socialized" as black people. Yet, our lives have been deeply colored by our absence of deep color. Id.; see also Lawrence Wright, One Drop of Blood, NEW YORKER, July 25, 1994, at 53
63. (describing a National Center for Health Statistics study that found that 5.8% of people who self-identified as Black were not so identified by a census interviewer; the same went for nearly one-third of those who self-identified as Asian American, and 70% of those who self-identified as American Indian). Also, a person's ethnic group may not fit readily into current racial typologies. For example, many Americans may not know how to categorize Americans of Middle Eastern descent. Are they White? Asian (think Oriental/Persian rugs)? Black (think African American, from Egypt, Libya)? See, e.g., Soheir A. Morsy, Beyond the Honorary "White" Classification of Egyptians: Societal Identity in Historical Context, in RACE, supra note 31, at 175
64. (describing how U.S. Immigration regulations map northern Africans, as well as Middle Easterners, to "White"); see also Doug Daniels, The White Race is Shrinking: Perceptions of Race in Canada and Some Speculations on the Political Economy of Race Classification, 4 ETHN. & RACIAL STUD. 353, 354 (1981) (providing informal survey results based on asking students about the race of various nationalities, such as Italians, Argentinians, and Iranians).
65. For legal discussions of interracial romance, see Randall Kennedy, How Are We Doing with Loving?: Race, Law, and Intermarriage, 77 B.U. L. REV. 815, 816-20 (1997)
66. and Note, Racial Steering in the Romantic Marketplace, 107 HARV. L. REV. 877, 894 (1994).
67. For social aspects of interracial dating, see Viet D. Dinh, Single White Female, 2 RECONSTRUCTION 19, 19-20 (1994).
68. I do not use "meaning" in the way that philosophers of language might use it. Instead, my reference point is Critical Race Theory. See, e.g., Lawrence, supra note 29, at 356-60 (presenting an equal protection theory of cultural meanings); see also Jerry Kang, Negative Action Against Asian Americans: The Internal Instability of Dworkin's Defense of Affirmative Action, 31 HARV. C.R.-C.L. L. Rev. 1, 21-36 (1996) (suggesting a banned meanings approach to equal protection, whereby an individual has the right not to suffer disadvantage from a governmental practice that conveys an objective social meaning of stigma).
69. Cf. KUNDA, supra note 29, at 315 (describing "stereotypes . . . as cognitive structures that contain our knowledge, beliefs, and expectations about a social group. There is also growing recognition that group labels bring to mind feelings as well as thoughts." (internal citations omitted)).
70. See Cynthia Kwei Yung Lee, Race and Self-Defense: Toward a Normative Conception of Reasonableness, 81 MINN. L. REV. 367, 402-23 (1996).
71. On the question whether a racial generalization is "justified" because it is accurate, see JODY DAVID ARMOUR, NEGROPHOBIA AND REASONABLE RACISTS (1997).
72. In thinking about "rational racism," Armour asks us to be mindful of the following points: First, we must be sure to have good data. See id. at 40. Second, we must be sure that we are parsing these data correctly. We often choose wrong or irrelevant baselines, focus on relative measures instead of absolute ones, see id. at 38-39, 41-46, and exaggerate threats posed to ourselves and discount harms to others, see id. at 54. Third, we must appreciate the costs we inflict when we act on inaccurate generalizations about a specific individual. Social morality, not statistics, answers the question whether the benefits we derive from acting on such generalizations justify the costs inflicted on the objects of our beliefs. See id. at 46-58. But cf. DINESH D'SOUZA, THE END OF RACISM: PRINCIPLES FOR A MULTICULTURAL SOCIETY 245-87 (1995) (suggesting that most incidences of racial "discrimination" are rational). Moreover, in my view, it behooves us to ask why a particular generalization is true, whether it is true because of unjust background conditions, and whether an action rationally based on that generalization will further entrench those unjust conditions.
73. In an often quoted speech, Jesse Jackson said: "There is nothing more painful for me at this stage in my life than to walk down the street and hear footsteps and start to think about robbery and then look around and see it's somebody white and feel relieved. How humiliating." Paul Glastris & Jeannyne Thornton, A New Civil Rights Frontier, U.S. NEWS & WORLD REP., Jan. 17, 1994, at 38 (quoting Jackson's speech on November 27, 1993). Jackson was making a passionate plea about Black-on-Black crime to a Black congregation. He did not intend his comments to be shared with a general audience, who might misinterpret the thrust and context of his message. As Armour notes, many journalists jumped on this story as vindication of their "rational" fears of Black people. See ARMOUR, supra note 40, at 34.
74. My impression is also affected by any individuating information I might have about these gentlemen. See Ziva Kunda & Paul Thagard, Forming Impressions from Stereotypes, Traits, and Behaviors: A Parallel-Constraint-Satisfaction Theory, 103 PSYCH. REV. 284, 300 (1996). In most stranger-to-stranger interactions, there are few individuating data.
75. Perhaps this attitudinal difference can be explained by the fact that the very cognitive structures that create racial categories dispose us to favor "in-groups" and disfavor "out-groups." See Krieger, supra note 29, at 1186-87. In addition, we are generally clueless that these processes are in effect See id. at 1217.
76. A feminist critique is obvious. See, e.g., WILLIAM WEI, THE ASIAN AMERICAN MOVEMENT 75, 75-77 (1993) (discussing problems of sexism within the Asian American civil rights movement).
77. In one telling experiment, 41 undergraduate students from New York University completed a mundane computer-delivered test. Throughout this test, two groups were subliminally shown pictures of the face of either a young African American or White man. The computers were designed to crash after the 130th question. Observers, who did not know which group they were observing, measured the hostility of the participants' reactions. Those who were flashed Black images reacted with greater hostility. See John A. Bargh, Mark Chen & Lara Burrows, Automaticity of Social Behavior Direct Effects of Trait Construct and Stereotype Activation on Action, 71 J. PERSONALITY & SOC. PSYCHOL. 230, 238-40 (1996). Numerous studies have confirmed that neutral, subliminal exposure to individuals mapped to certain racial categories evokes associated racial meanings. See KUNDA, supra note 29, at 322 (collecting studies and concluding that "neutral reminders of Black people can automatically trigger in other people negative thoughts, feelings, and behaviors, without any awareness on their part that they have even been reminded of this group").
78. See ARMOUR, supra note 40, at 53-54 (stating that the "relentless, cumulative, dehumanizing" occurrences of Whites avoiding Blacks on the street cause "humiliation and stigmatization [that] must be counted among the most painful costs of race-based suspicions"). Similar stories of insult abound. Roy Johnson, a senior editor at Money magazine, was mistaken for a parking valet rather than a customer at a fancy restaurant in a wealthy New York City suburb. "First the parking valet handed him the keys to his Jaguar instead of fetching the car. Then an elderly white couple came out and handed him the keys to their automobile, with the instruction: 'It's a black Mercedes-Benz.'" Jerelyn Eddings, Jeannye Thornton, Dorian Friedman, Scott Minerbrook, Warren Cohen, Betsy Wagner & Jill Jordan Sieder, The Covert Color War, U.S. NEWS & WORLD REP., Oct. 23, 1995, at 40. Mary Frances Berry, chair of the U.S. Commission on Civil Rights, says she has been kept under surveillance at shopping malls. Andrew Barrett, a commissioner of the Federal Communications Commission, says he must call for taxis because they will not stop when he hails them. See id. at 40-41.
79. See ARMOUR, supra note 40, at 13-14 ("[L]ike a tax, racial discrimination is persistent, pervasive, must be dealt with, cannot be avoided, and is not generally resisted.").
80. Of course, I could have responded differently, by signaling to the dealer that I am not the "average" Asian. That might have meant dressing in a particular way, bargaining more aggressively than I would otherwise, or signaling early my educational credentials and professional status - not as engineer, but as law professor. But then, all of these efforts would amount to an Asian Expenditure, potentially as expensive as the Asian Tax. See Devon Carbado & Mitu Gulati, Working Identity, 85 CORNELL L. REV. (forthcoming July 2000) (explaining how responding to racial stereotypes takes time and effort).
81. In one classic Princeton study, researchers tested two hypotheses: First, black, as compared to white, job applicants will receive less immediate nonverbal communications from white job interviewers; second, recipients of less immediate nonverbal communications, whether black or white, will reciprocate these communications and be judged to perform less adequately in the job interview situation than recipients of more positive nonverbal communications. Carl O. Word, Mark P. Zanna & Joel Cooper, The Nonverbal Mediation of Self-Fulfilling Prophecies in Interracial Interaction, 10 J. EXPERIMENTAL SOC. PSYCHOL. 109, 111-12 (1974). Two experiments were performed. In the first, White and Black confederates applied for the same job. They had similar paper records and were trained to respond similarly in the interview. The interviewers, however, treated the applicants differently. Interviewers addressing Black applicants kept greater physical distance, made more speech errors, and ended the interview more quickly. See id. at 114-15. In the second experiment, using only White applicants, confederate interviewers treated half the applicants as the Black applicants had been treated - keeping applicants at greater physical distance, speaking to them with speech errors, and terminating their interviews quickly. Observers, who reviewed all of the interviews on videotape and were unaware of the point of the experiment, believed that those applicants who were given the "Black treatment" performed significantly worse. See id. at 117.
82. See generally Claude M. Steele, A Threat in the Air: How Stereotypes Shape Intellectual Identity and Performance, 52 AM. PSYCHOLOGIST 613, 614 (1997) (defining stereotype threat as "a situational threat - a threat in the air - that, in general form, can affect the members of any group about whom a negative stereotype exists (e.g., skateboarders, older adults, White men, gang members). Where bad stereotypes about these groups apply, members of these groups can fear being reduced to that stereotype. And for those who identify with the domain to which the stereotype is relevant, this predicament can be self-threatening.").
83. In one revealing study, Asian American female undergraduates with excellent math SAT scores at Harvard University were given a quantitative test of 12 math questions. Before taking the exam, each woman was asked to complete a questionnaire designed to unconsciously trigger a female identity, Asian identity, or no specific identity. The group that had its Asian identity triggered performed best in accuracy; the group that had no identity triggered came in second; and the group that had its female identity triggered ranked last. Interestingly, the questionnaires did not seem to affect conscious motivation or confidence. There were no significant differences in the number of questions attempted, the number of questions that participants reported guessing, or the number of participants who reported enjoying the test; nor was there any meaningful distinction in the participant's self-assessment of their performance, their math skills, or their difficulty with the test. See Margaret Shih, Todd L. Pittinsky & Nalini Ambady, Stereotype Susceptibility: Identity Salience and Shifts in Quantitative Performance, 10 PSYCHOL. SCI. 80, 80-81 (1999). The same experiment was performed in Canada, where the stereotype that Asians are good in math is less robust. In that iteration, the control group scored highest, followed by the Asian identity group, and then the female identity group. See id. at 82.
84. Cf. Kenneth L. Karst, Myths of Identity: Individual and Group Portraits of Race and Sexual Orientation, 43 UCLA L. REV. 263, 294 (1995) (stating that "[i]f particular skin colors and physiognomies seem striking, one reason is that race has been, for a very long time, so important a category in the nation's public, official, legal vocabulary").
85. See, e.g., Neil Gotanda, A Critique of "Our Constitution is Color-Blind", 44 STAN. L. REV. 1, 23-27 (1991) (discussing how rules of hypodescent, which map any person with a trace of African ancestry as Black, connote Black blood as "contaminating" to the "pure White race").
86. See, e.g., MELVIN L. OLIVER & THOMAS M. SHAPIRO, BLACK WEALTH/WHITE WEALTH: A NEW PERSPECTIVE ON RACIAL INEQUALITY 50-52, 151-60 (1995) (arguing that wealth is the best measure of "sedimentation of racial inequality" and examining the intergenerational transmission of wealth).
87. See CORNELL & HARTMANN, supra note 31, at xviii; OMI & WINANT, supra note 28, at 64-65 (explaining that American racial categories are produced by social construction, not natural selection); see also discussion and sources cited supra note 32.
88. The rules of racial mapping are configured by politics, social conventions, and law. For example, as a matter of social convention, both Italians and the Irish were not originally considered "White" in the U.S. See BRODKIN, supra note 24, at 54 (examining Irish ethnic identity formation); Kenneth L. Karst, Does Integration Have a Future?, in CULTURAL PLURALISM, IDENTITY POLITICS, AND THE LAW 139, 148-49 (Austin Sarat & Thomas R. Kearns eds., 1999)
89. (discussing Italian ethnicity). Today, both are. Not surprisingly, there has been great fluidity in racial mapping rules applied by the federal government. See, e.g., BRODKIN, supra note 24, at 74 (identifying Mexicans, Middle Easterners, and Indians as shuttling back and forth between White and non-White census categories); OMI & WINANT, supra note 28, at 3 (noting that Japanese Americans have been variously categorized by the U.S. Census as non-White, Oriental, other, and Asian and Pacific Islander). States have historically also adopted myriad mapping rules. See IAN F. HANEY LÓPEZ, WHITE BY LAW: THE LEGAL CONSTRUCTION OF RACE 118-19 (1996)
90. (cataloging the history of state racial definitions). See generally F. JAMES DAVIS, WHO IS BLACK?: ONE NATION'S DEFINITION (1991)
91. (discussing the changing definitions of Blackness in American history). These rules have been challenged in court. See, e.g., HANEY LÓPEZ, supra, at 42-46 (discussing cases prompted by the federal laws that, until 1952, maintained racial bars on naturalization); Ariela J. Gross, Litigating Whiteness: Trials of Racial Determination in the Nineteenth-Century South, 108 YALE L.J. 109, 111-23 (1998) (discussing the law's role in defining the cultural meaning of racial identities); Karst, supra note 52, at 271-74. These challenges have continued to the present. See Jane Doe v. Department of Health and Human Resources, Office of Vital Statistics, 479 So. 2d 369, 372 (La. Ct. App. 1985) (deciding that Suzy Guillory Phipps is not White).
92. See, e.g., Kimberlé Williams Crenshaw, Race, Reform, and Retrenchment: Transformation and Legitimation in Antidiscrimination Law, 101 HARV. L. REV. 1331, 1370-74 (1988) (describing the hegemonic role of racism, which requires identifying Blacks as the despised "other").
93. See CORNELL & HARTMANN, supra note 31, at 6-9.
94. At its broadest, the term "cyberspace" encompasses the entire web of computing-communication technologies that enmesh the globe. It includes every computing processing unit connected through every type of telecommunications, both wireline and wireless. It includes every telecommunications and mass media industry, such as land line and wireless telephony, broadcast radio and television, satellite, and cable. In this paper, however, I use the term more narrowly to apply to all computer networks that interoperate with the Internet.
95. See Peter Kollock & Marc A. Smith, Communities in Cyberspace, in COMMUNITIES IN CYBERSPACE 3, 4-8 (Marc A. Smith & Peter Kollock eds., 1999) (describing the cyberspace landscape).
96. Many forms of communication have both push and pull aspects. For example, one can configure Web browsers to receive channels of information automatically. After the initial specification of channels (akin to "pull"), data are periodically delivered to the individual without a specific request for that information (akin to "push").
97. For good descriptions of the Usenet, see Marc A. Smith, Invisible Crowds in Cyberspace: Mapping the Social Structure of the Usenet, in COMMUNITIES IN CYBERSPACE, cited above in note 60, at 195, 198-200;
98. and Paul K. Ohm, Comment, On Regulating The Internet: Usenet, A Case Study, 46 UCLA L. REV. 1941, 1945-52 (1999). According to Smith, as of 1999, there were 14,347 newsgroups, on which an average of 20,000 people post 300,000 messages daily, generating six gigabytes of messages. See Smith, supra, at 197.
99. For more information about Internet-based chat, see Nicolas Pioch, A Short IRC Primer (last modified Jan. 1, 1997) 〈http://www.irchelp.org/irchelp/ircprimer.html〉
100. (on file with the Harvard Law School Library); and What is IRC? (visited July 22, 1999) 〈http://www.geocities.com/ ∼mirc/irc.html〉 (on file with the Harvard Law School Library).
101. An extremely popular instant messaging program is Mirabilis ICQ, which was purchased by America Online in 1998. See An Introduction to ICQ: What is ICQ (visited July 22, 1999) 〈http:// www.mirabilis.com/products/whatisicq.html〉 (on file with the Harvard Law School Library).
102. See Leslie Helm, Microsoft Sends AOL a Message, L.A. TIMES, July 22, 1999, at C1
103. ("Instant messaging software is the fastest-growing service on the Web, with about 80 million registered users, many of whom often spend hours a day chatting with online friends."); see also Linda Matchan, Forget the Phone. With I.M., Teens Can Talk All Night, and It's, Like, Instant, BOSTON GLOBE, Apr. 21, 1999, at C1 ("[S]ome 432 million instant messages travel across AOL each day.").
104. "[T]hough typically still text-based, [MUDs] are shared, persistent, navigable virtual environments in which user-created characters and scriptable objects can interact with one another in surprisingly rich and compelling ways." Diane J. Schiano, Convergent Methodologies in Cyber-Psychology: A Case Study, 29 BEHAV. RES. METHODS, INSTRUMENTS & COMPUTERS 270, 270 (1997).
105. See Amy S. Bruckman, Gender Swapping on the Internet, in HIGH NOON ON THE ELECTRONIC FRONTIER: CONCEPTUAL ISSUES IN CYBERSPACE 317, 318-20 (Peter Ludlow ed., 1996) (describing the history of MUD development).
106. Some of these graphical MUDs are wildly popular and extremely profitable. Graphical MUDs using the WorldsAway technology supposedly generated over $10 million of revenues in 12 months through virtual worlds in the United States, Japan, and Korea. See Avaterra.com, Inc. Backgrounder (visited Sept. 24, 1999) 〈http://www.worldsaway.com/aboutus/backgrounder.html〉 (on file with the Harvard Law School Library). One of these communities, called Dreamscape, has approximately 100,000 members from 16 different countries. See id. On average, each member spends more than 10 hours per month on Dreamscape. See id.
107. According to a study by Nielsen Media Research and CommerceNet in June 1999, there are 92 million Internet users in North America, aged 16 and older. See New CommerceNet/Nielsen Media Research Study Also Shows Internet Users Top 92 Million in the U.S. and Canada, BUS. WIRE, June 17, 1999, available in Westlaw, AllNewsPlus, Bwire (on file with the Harvard Law School Library).
108. See Barry Wellman & Milena Gulia, Virtual Communities as Communities: Net Surfers Don't Ride Alone, in COMMUNITIES IN CYBERSPACE, supra note 60, at 167, 172-73 (providing examples).
109. See Schiano, supra note 66, at 271 (referring to interviewees who cited sociality as a primary reason for MUDding).
110. See, e.g., Susan C. Herring, Two Variants of an Electronic Message Schema, in COMPUTER-MEDIATED COMMUNICATION: LINGUISTIC, SOCIAL AND CROSS-CULTURAL PERSPECTIVES 81, 103-05 (Susan C. Herring ed., 1996)
111. (finding that both men and women used two Internet mailing lists more for social interaction than for information exchange); Kollock & Smith, supra note 60, at 6 (describing the popularity of chat rooms on America Online); Robert Kraut, Vicki Landmark, Sara Kiesler, Tridas Mukopadhyay & William Scherlis, Internet Paradox: A Social Technology That Reduces Social Involvement and Psychological Well-Being?, 53 AM. PSYCHOLOGIST 1017, 1029 (1998)
112. (reporting that a major reason people use the Internet is "to keep up with family and friends through electronic mail and on-line chats and to make new acquaintances through MUDs, chats, Usenet newsgroups, and listservs"); Barry Wellman, Janet Salaff, Dimitrina Dimitrova, Laura Garton, Milena Gulia & Caroline Haythornthwaite, Computer Networks as Social Networks: Collaborative Work, Telework, and Virtual Community, 22 ANN. REV. SOC. 213, 220 (1996) ("[W]hile most of the elderly users of the 'SeniorNet' virtual community joined to gain access to information, their most popular on-line activity has been companionable chatting.").
113. Such interfaces are often more compelling and functional. For example, the standard interface for chat rooms includes a window of rapidly scrolling text - the chatter of room inhabitants. But this text is hard to follow because multiple conversations are taking place among multiple people, and all comments are interlineated into one streaming transcript. By contrast, in a graphical chat room, characters having a conversation can simply move their avatars to a different corner of the room. Comments made by each character appear in a "balloon" above the avatar's head. Through visual mediation, multiple conversations in a single room can be spatially separated. See, e.g., Avaterra.com, Inc. Backgrounder, supra note 68 (describing the graphical MUD Dreamscape).
114. Not all user interfaces will be designed this way. Even 15 years from now, it will probably be more efficient to use something like current Web browser interfaces to research text-based information. We will use different applications, with different interfaces, depending on the task.
115. See, e.g., Urban Simulation Team (visited Jan. 22, 2000) 〈http://www.ust.ucla.edu/ustweb/ about_us.html〉 (on file with the Harvard Law School Library).
116. Science fiction writers have imagined such futures. See, e.g., ORSON SCOTT CARD, ENDER'S GAME 66-70 (1985);
117. NEAL STEPHENSON, SNOW CRASH 23-24 (1992).
118. The future may lead us not to larger screens but to head-mounted displays or to images painted directly onto our retinas. See, e.g., Peter Wayner, Wraparound Sight and Sound You Can Wear on Your Head, N.Y. TIMES, July 22, 1999, at G3 (describing a Sony head-mounted display).
119. Sony Corporation's Playstation II reportedly includes a chip significantly faster than Intel's Pentium III microprocessor and more than twice as powerful as a Silicon Graphics workstation, the current benchmark of graphics computing. See John Markoff, Silicon Valley's Awesome Look at New Sony Toy, N.Y. TIMES, Mar. 19, 1999, at C1.
120. Stunning sample videos can be downloaded at Playstation 2 Info Center (visited Jan. 22, 2000) 〈http://www.ps2info.com〉 (on file with the Harvard Law School Library).
121. Computer games are a $10 billion industry globally, $5.6 billion in the United States alone. See DON TAPSCOTT, GROWING UP DIGITAL: THE RISE OF THE NET GENERATION 4 (1998).
122. "Computers doubled in speed every three years at the beginning of the twentieth century, every two years in the 1950s and 1960s, and are now doubling in speed every twelve months." RAY KURZWEIL, THE AGE OF SPIRITUAL MACHINES: WHEN COMPUTERS EXCEED HUMAN INTELLIGENCE 3 (1999). Kurzweil predicts that "[t]his trend will continue, with computers achieving the memory capacity and computing speed of the human brain by around the year 2020."
123. RAY KURZWEIL, THE AGE OF SPIRITUAL MACHINES: WHEN COMPUTERS EXCEED HUMAN INTELLIGENCE 3 (Id. His reasoning is as follows: The human brain has approximately 100 billion neurons. Each neuron has on average 1000 connections with its neighbors. Therefore, 100 trillion connections must be modeled. Each neuron can perform approximately 200 calculations per second. This means that the human brain has the ability to perform approximately 20 million billion calculations per second. See id. at 103. Given the rate at which computer calculation speed has been increasing, Kurzweil concludes: "[I]t is reasonable to estimate that a $1,000 personal computer will match the computing speed and capacity of the human brain by around the year 2020 . . . ." Id. at 105.
124. More than half a million U.S. households have connected to the Internet via high speed broadband connections such as cable modems, digital telephone lines, and satellite links. See Amy Harmon, Some Enter the Fast Lane to Get Access to the Web, N.Y. TIMES, Apr. 28, 1999, at A1. Research firms project that broadband penetration will reach 10 to 16 million households by 2002. See id.
125. See A Little History of the World Wide Web (last modified Oct. 3, 1995) 〈http://www.w3. org/History.html〉 (on file with the Harvard Law School Library).
126. For futuristic visions of virtual reality, see HOWARD RHEINGOLD, VIRTUAL REALITY 345-76 (1991).
127. There is another design philosophy that I do not consider seriously: racial separation. This approach would leverage cyberspace in whatever way possible to promote separation between the races. White supremacists who do not want to consort with racial minorities hold this view; so do certain people of color who view integration as a euphemism for conformity to White norms and colorblindness as hopelessly unrealistic given the pervasive racism of American society. Although I understand the frustration that generates a racial separatist attitude among people of color, I reject separation politics as unnecessary and unhelpful. My thoughts align with those of my colleague, Kenneth Karst. See Karst, supra note 56, at 143.
128. Ayres considers four possible forms of discrimination: two that are animus-based (associational, consequential) and two that are statistical (search-based, revenue-based). See Ayres, Further Evidence, supra note 9, at 111. He concludes that different forms of discrimination explain different behaviors. For instance, "sellers' discrimination against black females may in part stem from a belief that black females are more averse to bargaining than white males, and . . . consequential animus may explain part of sellers' discrimination against black males." Id. at 141. Abolition would address racially disparate treatment caused by either animus or statistical generalizations about which social categories include the biggest "suckers."
129. See discussion and source cited supra note 43.
130. I do not align abolition thinking with either the political left or right. On the one hand, opponents of affirmative action contend that race is no longer a salient factor in society, with the exception of affirmative action programs. On the other hand, cultural critics such as Anthony Appiah argue that the continuing use of race, even acknowledging it as a social construct, is a dangerous reliance upon an illusion. See Anthony Appiah, The Uncompleted Argument: Du Bois and the Illusion of Race, in "RACE," WRITING, AND DIFFERENCE 21, 35-36 (Henry Louis Gates, Jr. ed., 1986).
131. In addition, my discussion of abolition should not be confused with the new abolitionism movement represented, for example, by Noel Ignatiev and John Garvey's magazine, Race Traitor. The goal of this movement is to abolish the White race and its related privilege, not all races. See, e.g., Editorial, Abolish the White Race - By Any Means Necessary, in 1 RACE TRAITOR 1, 1 (John Garvey & Noel Ignatiev eds., 1993).
132. Even in clinical research, these are the basic methods of racial mapping. See LaVeist, supra note 32, at 23 (identifying "visual assessment of the study subject or respondent self-report" as the "usual" method of ascertaining race); cf. Karst, supra note 52, at 272 (identifying physical appearance and verbal self-admissions as methods of judicial fact-finding about race).
133. See Lori Kendall, Meaning and Identity in "Cyberspace": The Performance of Gender, Class, and Race Online, 21 SYMBOLIC INTERACTION 129, 142, 145-47 (1998)
134. (describing MUD norms of not presenting or discussing race); cf. Lisa Nakamura, Race in/for Cyberspace: Identity Tourism and Racial Passing on the Internet, 13 WORKS & DAYS 25/26, at 181, 183 (1995) (stating that in LambdaMOO, a multiuser domain, "[r]ace is not only not a required choice, it is not even on the menu").
135. See Kollock & Smith, supra note 60, at 12; Jodi O'Brien, Writing in the Body: Gender (Re)production in Online Interaction, in COMMUNITIES IN CYBERSPACE, supra note 60, at 76, 86
136. ("Gender is one of the first means by which persons introduce and represent themselves to others in electronic communications."); Elizabeth Reid, Virtual Worlds: Culture and Imagination, in CYBERSOCIETY, supra note 7, at 164, 179 (explaining that among gender, race, class, and age, "gender is the only one always 'hard coded' into MUD programs").
137. See Cynthia Joyce, Race Matters in Cyberspace, Too, SALON MAG. (last modified June 5, 1997) 〈http://www.salon.com/june97/21st/race970605.html〉 (on file with the Harvard Law School Library) (discussing the negative audience reaction to a Web journalist's disclosure that she was both Black and Latina).
138. See Earl Babbie, "We Am a Virtual Community", AM. SOCIOLOGIST, Spring 1996, at 65, 66.
139. Opponents of affirmative action often cite King's speech to support their
140. TAPSCOTT, supra note 79, at 111 (quoting Deanna Perry of Florida) (internal quotation marks omitted).
141. See Michele Willson, Community in the Abstract: A Political and Ethical Dilemma?, in VIRTUAL POLITICS: IDENTITY AND COMMUNITY IN CYBERSPACE 145, 149 (David Holmes ed., 1997) ("Race, gender or physical disability is indiscernible over the Internet. Any basis for enacting embodied discrimination is removed, freeing access to participation and granting each participant equal status within the network.").
142. See Siok Kuan Tambyah, Life on the Net: The Reconstruction of Self and Community, 23 ADVANCED CONSUMER RES. 172, 174 (1996) ("The impersonality and immateriality of the online experience has a liberating and leveling effect; it blanks out race, age, gender, looks, timidity, and handicaps.").
143. See Mari J. Matsuda, Looking to the Bottom: Critical Legal Studies and Reparations, 22 HARV. C.R.-C.L. L. REV. 323, 326 (1987)
144. (arguing that the experiences of victims of racial oppression give them "distinct normative insights"). One commentator has criticized computer games for steadfastly denying the significance of race. See Julian Bleeker, Urban Crisis: Past, Present, and Virtual, 24 SOCIALIST REV. 189, 210 (1995) (criticizing the popular game SimCity 2000, which identifies heat, crime, and unemployment - but not race - as factors that cause riots).
145. Obvious examples include "Chang" or "López." Less information is conveyed by the name "Harris"; although it is most likely not Asian, the name could easily be that of a White or a Black. The accuracy of these assumptions turns on the number of interracial family formations as well as social conventions regarding name changes at marriage.
146. In one study, students were given a list of names and asked to identify which were politicians and which were criminals. Students more often picked African American sounding names (for example, Jamal Johnson) as criminals, not politicians. See Connie T. Wolfe & Steven J. Spencer, Stereotypes and Prejudice: Their Overt and Subtle Influence in the Classroom, 40 AM. BEHAV. SCIENTIST 176, 178 (1996).
147. See, e.g., Byron Burkhalter, Reading Race Online: Discovering Racial Identity in Usenet Discussions, in COMMUNITIES IN CYBERSPACE, supra note 60, at 60, 65 (discussing how the subject line "Sisters please explain" may suggest that the author is Black).
148. Richer forms of media are more efficient and satisfying. We speak much faster than we type. Thus, if we have the technological ability to communicate faster, we will do so if it does not cost more. These richer forms of media are also more satisfying because by invoking additional senses, they better create the illusion that the person with whom we are communicating is physically present. Cf. ERVING GOFFMAN, INTERACTION RITUAL: ESSAYS IN FACE-TO-FACE BEHAVIOR 33 (1967) (describing the wealth of information conveyed in face-to-face meetings).
149. One can, however, adopt human-like avatars that are sufficiently cartoonish that race becomes ambiguous, although gender remains clear.
150. In one experiment, individuals - some with identifiably Black voices - called a wrong number, tried to explain that their car had broken down, then asked for assistance in contacting a garage. The study organized its subjects by political affiliation, and both conservatives and liberals were less likely to help Blacks than they were to help Whites. See Samuel L. Gaertner & John F. Dovidio, The Aversive Form of Racism, in PREJUDICE, DISCRIMINATION, AND RACISM 61, 68-69 (John F. Dovidio & Samuel L. Gaertner eds., 1986).
151. For further discussion of access issues, see discussion and sources cited infra notes 132-33, 136.
152. Many virtual communities overtly target specific racial groups. For African Americans see, for example, Black Voices at 〈http://www.blackvoices.com〉, Netnoir at 〈http://www.netnoir. com〉, and WanOnline at 〈http://www.wanonline.com〉. For Asian Americans see, for example, Asian Avenue at 〈http://www2.CommunityConnect.com/AsianAvenue.html〉 and Lambda Phi Epsilon at 〈http://www.lambdaphiepsilon.com〉. For Latinos, see, for example, LatinoLink at 〈http://www.latinolink.com〉.
153. But cf. Kenji Yoshino, Assimilationist Bias in Equal Protection: The Visibility Presumption and the Case of "Don't Ask, Don't Tell", 108 YALE L.J. 485, 544-57 (1998) (arguing that the "don't ask, don't tell" military policy helps to render gays invisible and politically powerless as a group).
154. See, e.g., Elizabeth M. Reid, Text-Based Virtual Realities: Identity and the Cyborg Body, in HIGH NOON ON THE ELECTRONIC FRONTIER, supra note 67, at 339 (describing Furry-MUCK, a highly popular MUD in which all characters are non-human, and most are furry animals).
155. These technologies would likely develop out of more primitive technologies sought by individuals who prefer "soft focus" and other enhancements to their transmitted images that are common to the Hollywood mass media industries.
156. This technology would not be inconceivable because it would simply be an unusual application of real-time bidirectional language "interpretation," without the shift in languages.
157. Pressure can be applied in complicated ways. For example, Lisa Nakamura relates one attempt in a MUD to enact a hate crime penalty against characters who harassed others on the basis of race. The petition failed, with detractors emphasizing that individuals could simply "hide" their race by removing it from their self-descriptions. One character wrote, "Well, who knows my race unless I tell them? If race isn't important than [sic] why mention it? If you want to get in somebody's face with your race then perhaps you deserve a bit of flak." Nakamura, supra note 89, at 189 (internal quotation marks omitted).
158. Cf. Carbado & Gulati, supra note 48 (describing the burdens placed on racial minorities who must abide by "colorblind" institutional norms).
159. See Kendall, supra note 89, at 145-46 (stating that Whiteness is considered the default or the state of being without a race).
160. A religious analogy might illuminate the nature of these disparate burdens: What if we prohibited all people from identifying their religions in cyberspace because historically, religion has been a source of great conflict and strife? How would that make a Jew feel, especially if most Americans in cyberspace are Christian?
161. See generally CRITICAL WHITE STUDIES: LOOKING BEHIND THE MIRROR (Richard Delgado & Jean Stefancic eds., 1997).
162. In our society, Whites do not think much about how their racial identity benefits their lives; men do not think much about their gender; heterosexuals do not think much about their sexual orientation; and able-bodied people do not think much about their physical abilities. See, e.g., RUTH FRANKENBERG, WHITE WOMEN, RACE MATTERS: THE SOCIAL CONSTRUCTION OF WHITENESS 6 (1993)
163. (discussing Whiteness); Barbara J. Flagg, "Was Blind, but Now I See": White Race Consciousness and the Requirement of Discriminatory Intent, 91 MICH. L. REV. 953, 971 (1993)
164. (noting that Whites' social dominance allows them to "relegate [their] own racist specificity to the realm of the subconscious"); see also Devon W. Carbado, Epilogue: Straight Out of the Closet: Men, Feminism, and Male Heterosexual Privileges, in BLACK MEN ON RACE, GENDER & SEXUALITY: A CRITICAL READER 420, 426 (Devon W. Carbado ed., 1999) (noting how men do not think about gender and heterosexuals do not think about sexual orientation).
165. See infra Part VI.
166. See generally Gotanda, supra note 53, at 17-21, 53-60 (questioning the wisdom and practicality of colorblindness within both the public and private spheres); David A. Strauss, The Myth of Colorblindness, 1986 SUP. CT. REV. 99, 114-15 (arguing that colorblindness, as a public policy, actually amounts to color-consciousness).
167. 347 U.S. 483 (1954). Brown had an equally, if not more, important material strand, which I do not emphasize here. "Separate but equal" was a sham not only symbolically but also in terms of the material resources made available to Black institutions.
168. See, e.g., Norman Miller & Marilynn B. Brewer, The Social Psychology of Desegregation: An Introduction, in GROUPS IN CONTACT: THE PSYCHOLOGY OF DESEGREGATION 1, 2 (Norman Miller & Marilynn B. Brewer eds., 1984).
169. For a summary of studies lending support to the social contact hypothesis, see ALLPORT, cited above in note 31, at 232-60; THOMAS F. PETTIGREW, RACIALLY SEPARATE OR TOGETHER? 274-78 (1971);
170. and Thomas F. Pettigrew, Prejudice, in HARVARD ENCYCLOPEDIA OF AMERICAN ETHNIC GROUPS 820-29 (Stephan Thernstrom ed., 1980).
171. See, e.g., David R. Johnson & David Post, Law and Borders - The Rise of Law in Cyberspace, 48 STAN. L. REV. 1367, 1370-76 (1996) (explaining how cyberspace undermines the relationship between physical location and legal governance).
172. The quality of communication in cyberspace is principally a function of the data transfer rate (the rate of bits per second) that a user enjoys. Speed of communication, however, is not greatly affected by the geographical distance between information source and destination. Although geographical distance often correlates with the number of hops that an IP packet (the basic unit of information that travels through the Internet) must take from destination to source, there are far more important determinants of data throughput rates. Finally, the pricing model for Internet access is insensitive to the distance traveled by a data packet. Most users currently pay either a flat rate or a time-metered rate to their Internet Service Provider.
173. To talk about different sides of a city is provincial. Cyberspace allows for equally easy conversations across the globe. See, e.g., SCHULER, supra note 7, at 55-37 (describing a project that connected thousands of students in schools worldwide, in which students recorded daily journals, which were distributed globally through a listserv); Mary Elaine Meagher & Fernando Castaños, Perceptions of American Culture: The Impact of an Electronically-Mediated Cultural Exchange Program on Mexican High School Students, in COMPUTER-MEDIATED COMMUNICATION, supra note 72, at 187, 187-202
174. (describing an experiment in which high school students in Mexico City interacted via computer-mediated communications with high school students in San Diego); Marc A. Smith, Invisible Crowds in Cyberspace: Mapping the Social Structure of the Usenet, in COMMUNITIES IN CYBERSPACE, supra note 60, at 195, 197 (identifying 59% of Usenet posts as originating outside of the United States). Of course, human language is one potent barrier; however, the popularity of English and advances in real-time translation technologies (especially between Romance languages) may make human language less of a barrier.
175. See generally DOUGLAS S. MASSEY & NANCY A. DENTON, AMERICAN APARTHEID: SEGREGATION AND THE MAKING OF THE UNDERCLASS 81, 109-14. (1993).
176. Segregation is worst for Blacks, one-third of whom live in intensive racial segregation. See id. at 77 ("[Blacks] are unambiguously among the nation's most spatially isolated and geographically secluded people, suffering extreme segregation across multiple dimensions simultaneously."); see also Richard H. Sander, Housing Segregation and Housing Integration: The Diverging Paths of Urban America, 32 U. MIAMI L. REV. 977, 978-79 (1998) (describing the persistence of Black segregation). For a brief description of the federal government's role in creating this segregation, see OLIVER & SHAPIRO, cited above in note 34, at 15-18.
177. See Pavel Curtis, MUDding: Social Phenomena in Text-based Virtual Realities, in HIGH NOON ON THE ELECTRONIC FRONTIER, supra note 67, at 347, 337-38.
178. RAY OLDENBURG, THE GREAT GOOD PLACE: CAFÉS, COFFEE SHOPS, COMMUNITY CENTERS, BEAUTY PARLORS, GENERAL STORES, BARS, HANGOUTS AND How THEY GET YOU THROUGH THE DAY 7 (1989);
179. see also Robert D. Putnam, The Strange Disappearance of Civic America, 24 AM. PROSPECT 34, 34 (1996) (describing the decline of civic and political participation).
180. See, e.g., Alecia Wolf, Exposing the Great Equalizer: Demythologizing Internet Equity, in CYBERGHETTO OR CYBERTOPIA? RACE, CLASS, AND GENDER ON THE INTERNET 15, 25 (Bosah Ebo ed., 1998) (reporting that some women feel safer when confronted with aggressive male behavior online than in real space). One important exception, however, is increased contact between adults and children in unsupervised settings.
181. Nonetheless, people may experience substantial emotional distress in cyberspace. There have been reported cases of "virtual rape," in which an individual with the requisite technical expertise sexually batters other characters within a virtual community, without their consent. See, e.g., Julian Dibbell, A Rape in Cyberspace or How an Evil Clown, a Haitian Trickster Spirit, Two Wizards, and a Cast of Dozens Turned a Database into a Society, 2 ANN. SURV. AM. L. 471, 477-78 (1995);
182. Elizabeth Reid, Hierarchy and Power: Social Control in Cyberspace, in COMMUNITIES IN CYBERSPACE, supra note 60, at 107, 115 (describing an attack in Jenny-MUSH, a virtual help center for people who have been sexually assaulted or abused, in which a user named "Daddy" shouted rape narratives to everyone within the MUD).
183. See Reid, supra note 126, at 113 (identifying both physical and reputational safety as sources of disinhibition in MUDs).
184. For explanation of the differences between cyberspace anonymity and pseudonymity, see A. Michael Froomkin, Flood Control on the Information Ocean: Living With Anonymity, Digital Cash, and Distributed Databases, 15 J.L. & COM. 395, 417-24 (1996).
185. See NAT'L TELECOMMUNICATIONS & INFO. ADMIN. (NTIA), FALLING THROUGH THE NET: DEFINING THE DIGITAL DIVIDE: A REPORT ON THE TELECOMMUNICATIONS AND INFORMATION TECHNOLOGY GAP IN AMERICA at xv-xvi (1999), available at 〈http://www.ntia.doc.gov/ntiahome/fttn99/contents.html〉 (on file with the Harvard Law School Library) (describing its methodology, which involved appending questions to the Census Department's current population survey).
186. See id. at 2.
187. See id. at 5; see also Nick Turner, High Anxiety in Low-Cost PC Market, INVESTOR'S BUS. DAILY, Oct. 5, 1999, at A7 (reporting that computers costing less than $600 represent nearly half of all U.S. retail sales).
188. See NTIA, supra note 129, at 5. NTIA also measured individual (not household) access to the Internet, whether from the home or elsewhere, and found that 32.1% of Americans use the Internet.
189. The Internet only took seven years to hit 30% penetration into U.S. households. By contrast, the personal computer took 13 years, television took 17 years, and the telephone took 38 years. See United States Internet Council, State of the Internet: USIC's Report on Use & Threats in 1999 (visited Jan. 16, 2000) 〈http://www.usic.org/currentsite/usic_state_of_net99.htm〉 (on file with the Harvard Law School Library).
190. See id. at 6.
191. See id. The percentage of households using the Internet, by income and by race/origin is as follows: Under $15,000 $15,000-34,999 $35,000-74,999 $75,000+ White, 8.9 17.0 39.0 60.9 non-Hispanic Black, 1.9 7.9 22.2 53.7 non-Hispanic Other, 16.4 24.7 39.9 64.8 non-Hispanic Hispanic 3.8 7.6 26.8 48.1
192. "Between 1997 and 1998, the gap between White and Black households increased by 53.3% (from a 13.5 percentage point difference to a 20.7 percentage point difference), and by 56.0% (from a 12.5 percentage point difference to a 19.5 percentage point difference) between White and Hispanic households." Id.; see also Donna L. Hoffman & Thomas P. Novak, The Evolution of the Digital Divide: Examining the Relationship of Race to Internet Access and Usage Over Time (May 18, 1999) (unpublished manuscript) (on file with the Harvard Law School Library) ("In fact, the overall gap between whites and African Americans in Internet access and having ever used [the Web] have actually increased over time.").
193. See Hoffman & Novak, supra note 136.
194. Whenever interpreting socioeconomic data that characterize Asian Americans as "model minorities," it is important not to jump to conclusions. See, e.g., Chin, Cho, Kang & Wu, supra note 17, at 149 (explaining why the model minority myth based on household income data is misleading). For example, there is tremendous heterogeneity among the various ethnicities that constitute the Asian/Pacific Islander Census category. Because household access to the Internet is correlated with household income and because certain Asian ethnicities are disproportionately poor, it is highly probable that many Asian ethnic groups have very little access to the Internet. See, e.g., Ong & Hee, supra note 23, at 36-37 (noting that nearly half of all Americans of Southeast Asian descent live in poverty). In addition, surveys that are taken only in English or Spanish, such as the NTIA's, will likely produce skewed results given the substantial percentage of Asian Americans who have limited proficiency in those languages.
195. Some firms are giving away computers to users who promise to view advertisements or sign up for ISP accounts. See, e.g., Laurie J. Flynn, Maker of Inexpensive Computers Reports a Flood of Orders, N.Y. TIMES ON THE WEB (Mar. 29, 1999) 〈http://www.nytimes.com/library/tech/99/ 03/biztech/artides/29free.html〉
196. (on file with the Harvard Law School Library) (discussing FreePC.com's business plans to give away computers to those willing to have ads continuously aimed at them on their PCs). Some firms are giving away ISP accounts in exchange for viewing advertisements and collecting transactional records about what the household does on the Internet. See Gordon Black, This Net Service Comes with a Catch: It's Free, SEATTLE TIMES, Jan. 6, 2000, at C1.
197. See infra Part VI.C.2(a).
198. See Anthony G. Wilhelm, A Resource Model of Computer-Mediated Political Life, 25 POL'Y STUD. J. 519, 531 nn.2, 4 (1997) (emphasizing the importance of human capital and other antecedent resources necessary to engage in computer-mediated political life).
199. See supra note 104. Black Entertainment Television recently announced that it has raised $35 million to upgrade its Internet presence. See Saul Hansell, Big Companies Back a New Web Site, Aimed at Blacks, N.Y. TIMES, Aug. 12, 1999, at C1;
200. cf. Keith L. Alexander, Investors Flock to iVillage Stock, USA TODAY, Mar. 22, 1999, at 11B (noting that the share price of iVillage, which targets women, tripled on the first day after the initial public offering).
201. See generally Karst, supra note 56, at 146-47 (arguing that American national culture is a potpourri, with predominant strands from Europe and Africa intermixed).
202. Hip-hop is wildly popular among Asian youths, both in America and in Asia. See Kimberly Chun, Asian Artists Put New Spin on Hip-Hop; Filipino Rappers Rewrite Ghetto Music to Own Experience, S.F. CHRON., May 21,1999, at s (describing Asian American hip-hop).
203. See Arturo Escobar, Welcome to Cyberia: Notes on the Anthropology of Cyberculture, 35 CURRENT ANTHROPOLOGY 211, 218 (1994) (noting the globalization of Nintendo in youth culture).
204. Cf. OSCAR H. GANDY, JR., COMMUNICATION AND RACE: A STRUCTURAL PERSPECTIVE 241 (1998) (citing statistics demonstrating that in 1996, 60% of the audience for Black Entertainment Television was not Black).
205. See KUNDA, supra note 29, at 314.
206. Substantial experimental evidence supports this view. Studies reveal, for example, that people who have recently received negative feedback are more likely to activate negative racial stereotypes, in order to promote their own self-esteem. See id. at 362-64 (summarizing studies by Steven Fein and Steven Spencer).
207. See ALLPORT, supra note 31, at 348.
208. See id. at 40, 348; Wolfe & Spencer, supra note 98, at 177.
209. For a fascinating discussion of how group hostility can easily be manufactured through competition, see MUZAFER SHERIF, O.J. HARVEY, B. JACK WHITE, WILLIAM R. HOOD & CAROLYN W. SHERIF, INTERGROUP CONFLICT AND COOPERATION: THE ROBBERS CAVE EXPERIMENT 8-10 (1961).
210. See Miller & Brewer, supra note 118, at 2.
211. Id.
212. Indeed, even when we are exposed to data that disconfirms our racial meanings, we may process them in such a way as to avoid revising our assumptions. To begin with, we are generally poor at analyzing statistical information. See, e.g., Daniel Kahneman & Amos Tversky, Subjective Probability: A Judgment of Representativeness, 3 COGNITIVE PSYCHOL. 430, 430-52 (1972)
213. (explaining how, through the heuristic of representativeness, we systematically miscalculate objective probabilities). In addition, we are all adept at selective perception, selective interpretation, and selective recall. See generally Ziva Kunda & Paul Thagard, Forming Impressions from Stereotypes, Traits, and Behaviors: A Parallel-Constraint-Satisfaction Theory, 103 PSYCHOL. REV. 284 (1996) (proposing a theory of how stereotypes affect impressions of individuals); Lee, supra note 39, at 404-06 (summarizing studies that demonstrate that ambiguous behavior is interpreted more often as "violent" if committed by a Black rather than a White person). When we are confronted with evidence that runs contrary to our racial schemas that simply cannot be ignored, interpreted away, or forgotten, we often assimilate that evidence as an exception that proves the rule. See KUNDA, supra note 29, at 384-91 (discussing the process of "subtying counterstereotypic individuals"). Even as to the individual with the anti-stereotypical trait, we may still rely on stereotypes to predict her behavior. See id. at 355-57.
214. See Adeno Addis, "Hell Alan, They Did Invent Us": The Mass Media, Law, and African Americans, 41 BUFF. L. REV. 523, 530 (1993)
215. (explaining that Whites and Blacks know about each other largely through the media, not through direct individual contact); Robert M. Entman, Representation and Reality in the Portrayal of Blacks on Network Television News, 71 JOURNALISM Q. 509, 517 (1994) (suggesting that Whites with limited personal contact with Blacks assume that TV news portrayals of Blacks are representative).
216. See Pettigrew, supra note 118, at 827 ("By early adolescence, American children of various ethnic backgrounds yield a social distance rank-ordering of groups that is essentially the same as that found among adults.").
217. My subsequent comments are not meant to suggest that racism came into being because of the electronic mass media. However, the electronic mass media have become the most potent distributors of culture generally, and racial meanings specifically.
218. In addition to forming racial meanings, the mass media shape our entire view of reality. See DORIS A. GRABER, MASS MEDIA AND AMERICAN POLITICS 206 (1997)
219. (reporting the findings of the Cultural Indicators project from the University of Pennsylvania's Annenberg School of Communications: "[H]eavy viewers of television drama (more than four hours daily) see the world as television paints it and react to that world rather than to reality more than do light viewers of the same demographic background and similar circumstances."); JOHN TWITCHIN, THE BLACK AND WHITE MEDIA BOOK 125 (1988). Regrettably, the mass media present a distorted view of reality. See GANDY, supra note 146, at 23 (arguing that our understanding of risk is much closer to what the media present than to reality); SCHULER, supra note 7, at 217 (observing that crime and violence are 55 times more likely to occur on television than in real life).
220. See ALLPORT, supra note 31, at 195 ("[Stereotypes] are socially supported, continually revived and hammered in, by our media of mass communication - by novels, short stories, newspaper items, movies, stage, radio, and television.").
221. See generally C. Edwin Baker, Giving the Audience What It Wants, 58 OHIO ST. L.J. 311, 320 (1997) (describing the distortions caused by the advertiser-driven model of broadcast); Oscar H. Gandy, Jr., Engaging the Rational Racist (1998) (unpublished manuscript) (on file with the Harvard Law School Library) (arguing that advertisers have a strong influence on content).
222. Popular television shows such as The Cosby Show may have led some to believe that African Americans no longer encounter racism. "By consistently sweeping such major societal problems under the rug, television to a certain extent denies the actuality of racism in America. Real world racism, which is pervasive, subtle and blatant, is commonplace in America but virtually invisible on entertainment television." SALLY STEENLAND, UNEQUAL PICTURE: BLACK, HISPANIC, ASIAN, AND NATIVE AMERICAN CHARACTERS ON TELEVISION 21 (1989); see also GRABER, supra note 158, at 205-06 ("The improved television image of African Americans has tended to foster new misconceptions, namely that they are an economically privileged group that neither needs nor deserves special assistance.").
223. Racial caricatures are also common in advertisements. As Gandy explains, "advertisers seeking to introduce upscale Anglo audiences to their new brand of salsa are unconcerned about the impact their ads have on the reproduction of stereotypical impressions of the Latinos who traditionally included such condiments in their cuisine." GANDY, supra note 146, at 100; see also id. at 186 (discussing the Frito Bandito). It therefore stands to racial reason that Aunt Jemima is Black and mammy-like.
224. Films also contribute to racial stereotyping. See id. at 178-79 (explaining how stereotypes are easy to capture on film and are also easily processed by the audience); see also JESSE ALGERON RHINES, BLACK FILM/WHITE MONEY 70 (1996)
225. (arguing that Black films must include stereotypes in order to generate enough crossover appeal to garner funding). See generally KATHY RUSSELL, MIDGE WILSON & RONALD HALL, THE COLOR COMPLEX: THE POLITICS OF SKIN COLOR AMONG AFRICAN AMERICANS 244 (1992)
226. (describing the "dominant gaze" as "the tendency of mainstream culture to replicate, through narrative and imagery, racial inequalities and biases which exist throughout society"). For a persuasive critique of racial stereotyping in Star Wars Episode I: The Phantom Menace,
227. see Bruce Gottlieb, The Merchant of Menace: Racial Stereotypes in a Galaxy Far, Far Away?, SLATE (May 26, 1999) 〈http://www. slate.com/HeyWait/99-05-26/HeyWait.asp〉 (on file with the Harvard Law School Library).
228. See GANDY, supra note 146, at 114-15; see also Howard Kurtz, Time's 'Sinister' Simpson: Cover Photo Was Computer-Enhanced, WASH. POST, June 22, 1994, at D1
229. (describing Time Magazine's decision to darken O.J. Simpson's skin color); cf. Gandy, supra note 160 (explaining why journalists choose extreme exemplars consistent with stereotypes to provide a more powerful story). Many people have also complained about sports newscasters, who tend to attribute Black athletic success to natural talents and White athletic success to effort or smarts. See CHRISTOPHER P. CAMPBELL, RACE, MYTH AND THE NEWS 60 (1995) (characterizing such responses as examples of "aversive racism").
230. Contrast this with today's most recognizable Asian American character on a television show. The character "Ling" on Ally McBeal is an exotic, meretricious, sexually potent, but coquettish dragon lady. Although racial minorities may intentionally play to stereotypes as one way to navigate an institutional environment successfully, few actually go to such lengths. See Carbado & Gulati, supra note 48. For a general discussion of Hollywood's portrayal of Asians, see GINA MARCHETTI, ROMANCE AND THE "YELLOW PERIL": RACE, SEX, AND DISCURSIVE STRATEGIES IN HOLLYWOOD FICTION (1993).
231. Contrast this with gangsta rap music videos, which are vicarious experiences that enable young Whites to cross the cultural distance and become virtual, visual tourists of the inner city. See GANDY, supra note 146, at 104; see also ROBIN D.G. KELLEY, RACE REBELS: CULTURE, POLITICS, AND THE BLACK WORKING CLASS 191 (1994) (arguing that rap serves the same role as the blaxploitation films of the 1970s and presents the ghetto as a place of adventure, violence, and erotic fantasy).
232. One exception for me is sharing facts that challenge the "model minority" stereotype of Asian Americans. See Note, Racial Violence Against Asian Americans, 106 HARV. L. REV. 1926, 1930-33 (1993) (written by Jerry Kang).
233. See, e.g., GANDY, supra note 146, at 139 (discussing an institutional bias that favors the production of "bad news"); PAUL HARTMANN & CHARLES HUSBAND, RACISM AND THE MASS MEDIA: A STUDY OF THE ROLE OF THE MASS MEDIA IN THE FORMATION OF WHITE BELIEFS AND ATTITUDES IN BRITAIN 134 (1974) ("[T]he more negative the event the more likely it is to become 'news'.").
234. Vicarious experiences in cyberspace may be less stereotypical, however, because of greater racial diversity in cyberspace content providers. In real space, racial minorities frequently identify high barriers to entry as one reason why they have not been able to gain ownership of broadcast stations. By contrast, cyberspace "broadcast" platforms are much cheaper. If a more racially diverse ownership will produce less racially stereotypical images, then even vicarious experiences may improve in quality. My skepticism about such claims arises from the reality that even in cyberspace, production of the sort of content that captures substantial browsing market share will require huge financial resources.
235. Cf. Ringo Ma, Computer-Mediated Conversations as a New Dimension of Intercultural Communication Between East Asian and North American College Students, in COMPUTER-MEDIATED COMMUNICATION, supra note 72, at 173, 177 (concluding that computer-mediated communication between students in Asia and America "tends to demystify some distorted mass media reports," for example, regarding the U.S. educational system).
236. Miller & Brewer, supra note 118, at 2; see also KUNDA, supra note 29, at 381-82 (describing the contact hypothesis).
237. Mark Poster, Cyberdemocracy: The Internet and the Public Sphere, in VIRTUAL POLITICS, supra note 94, at 212, 224.
238. There are exceptions, such as the "Wizards" who moderate various MUDs. The owner of the MUD can delegate specific powers to knock people offline, terminate their accounts, muzzle speakers, etc. See Reid, supra note 126, at 119-20 (discussing challenges to the legitimacy of a MUD hierarchy and reporting complaints about the legitimacy of how certain people became Wizards); see also Curtis, supra note 123, at 358-60 (describing Wizards' powers). I am also troubled by differential bandwidth access, which enables certain individuals to communicate more quickly and more often than others.
239. See Sander, supra note 122, at 984 ("[T]he most affluent blacks experience roughly the same degree of segregation as the least affluent blacks.").
240. To avoid being misread, I should emphasize that I favor increasing social contact between different socioeconomic classes. However, the problems that would be ameliorated by such contact are not the problems I target in this paper. Further, I do not believe that cyberspace will necessarily increase classist segregation, which already takes place in real space.
241. Miller & Brewer, supra note 118, at 2.
242. See PETTIGREW, supra note 118, at 275.
243. Cf. Richard H. McAdams, Cooperation and Conflict: The Economies of Group Status Production and Race Discrimination, 108 HARV. L. REV. 1003, 1080 (1995) ("Laws forbidding race discrimination may increase the occurrence of cross-membership and thereby undermine the effectiveness of racial subordination as a status strategy.").
244. For example, on AsianAvenue, one can search all registered profiles to seek out similar members. Information fields that can be searched include: gender, marital status, lifestyle, college, ethnic background, location, job industry, job function, news, sports, arts, music, and hobbies. See Welcome to AsianAvenue.com (visited Nov. 24, 1999) 〈http://www.communityconnect. com/Preview/Pages (on file with the Harvard Law School Library) (previewing the services available through AsianAvenue.com).
245. As of April 1998, there were over 80,000 different newsgroups. See Wellman & Gulia, supra note 70, at 172.
246. See Yahoo! Chat (visited Sept. 5, 1999) 〈http://chat.yahoo.com/?room=Parenting:: 1600326512〉 (on file with the Harvard Law School Library) (observing 35 people chatting about possible names for children in Yahoo?'s "Parenting" chat room at 8:42 AM PST on Sunday, September 5, 1999).
247. See Mary S. Furlong, An Electronic Community for Older Adults: The SeniorNet Network, 39 J. COMM. 145, 149 (1989) (describing "SeniorNet's online grief counselor, [who] provides support for persons who have lost a spouse or loved one").
248. See Yahoo! Chat (visited Sept. 5, 1999) 〈http://chat.yahoo.com/?room=Cancer%20Chat:: 1600326509〉 (on file with the Harvard Law School Library) (observing patients with cancer providing each other with information and support in Yahool's "Cancer Chat" chat room at 8:32 AM PST on Sunday, September 5, 1999).
249. See TAPSCOTT, supra note 79, at 90-91 (describing Ability Online, a virtual community of over 7500 youths with disabilities or chronic illnesses). One child with a visible disability wrote: "The first time I went into KidsCom everyone was so nice. . . . They're much nicer than in class. People don't judge you based on what you look like." Id. (internal quotation marks omitted); see also HOWARD RHEINGOLD, THE VIRTUAL COMMUNITY: HOMESTEADING ON THE ELECTRONIC FRONTIER 26 (1994) ("People whose physical handicaps make it difficult to form new friendships find that virtual communities treat them as they always wanted to be treated - as thinkers and transmitters of ideas and feeling beings, not carnal vessels with certain appearance and a way of walking and talking (or not walking and not talking).").
250. See Wellman & Gulia, supra note 70, at 172.
251. For example, in one study of the Usenet, those who participated in groups concerned with "marginalized but concealable" identities, such as gays and lesbians, characterized their participation as personally meaningful and valuable. Many gay participants chose to come out to their family and friends in the real world because of their online interactions. See Bruce Bower, Marginal Groups Thrive on the Internet, 154 SCI. NEWS 245, 245 (1998); cf. O'Brien, supra note 90, at 82 (describing an Advocate advertisement claiming that "[t]here are no closets in Cyberspace").
252. One might call this playing the "e-race" ("erase") card.
253. Kimchi, a standard part of the Korean meal, is a spicy, pickled vegetable (typically celery cabbage). I recall strict admonitions from my mother not to have any in my college dorm refrigerator, for fear that I would offend my roommates.
254. Wen Ho Lee, a Chinese-American, was fired from Los Alamos National Laboratories (where I spent my summer after Tennessee) and arrested for allegedly violating national security laws. Much evidence suggests that he was the victim of a witch-hunt for spies who released missile secrets to China. For example, Robert Vrooman, the former chief of security at Los Alamos, has stated publicly that Lee was singled out because of his ethnicity. See Joint Press Statement by Asian Pacific American Organizations Regarding Wen Ho Lee (last modified Dec. 16, 1999) 〈http://www.ocanatl.org/jpr.htm〉 (on file with the Harvard Law School Library).
255. Age is another characteristic cloaked by text-based cyber-interactions. See, e.g., TAP-SCOTT, supra note 79, at 71 (describing a 14-year-old child who prefers communicating with adults by computer rather than by telephone because his age is kept hidden and he is given more respect); see also RHEINGOLD, supra note 184, at 228 (describing an 85-year-old woman who communicates regularly with youth online but not in real space).
256. This strategy of temporary delay reinforces the other environmental characteristics as well. Consider, for instance, how existing racial categories reflect a social hierarchy that might cut against the "equal status" requirement.
257. See, e.g., Simon Wiesenthal Center, Cyberwatch: The Taskforce Against Hate (visited Jan. 17, 2000) 〈http://www.wiesenthal.com/watch/index.html〉 (on file with the Harvard Law School Library) (keeping tabs on "the use of the Internet by hate groups to organize and disseminate their virulent message both in the U.S. and across international borders").
258. "Trolling" is participating in a cyberspace discussion forum with the intent to provoke angry responses from the audience. See Judith S. Donath, Identity and Deception in the Virtual Community, in COMMUNITIES IN CYBERSPACE, supra note 60, at 29, 45.
259. See infra pp. 1207-08.
260. See Peter Kollock & Marc Smith, Managing the Virtual Commons: Cooperation and Conflict in Computer Communities, in COMPUTER-MEDIATED COMMUNICATION, supra note 72, at 109, 116 (defining "flaming" as the posting of a hostile, provocative comment).
261. The Whitney Museum of American Art sponsored a four-month long computer-mediated conference on race. According to an observer and participant, the conversation was "honest" and addressed sensitive topics "that rarely come up in polite conversation." The discussion was "astonishingly naked, real, and instructive, allowing everyone present to better understand the ignorance and fear that fuel our collective racial paranoia and turn us on each other." Maurice Berger, Race in Cyberspace?, WIRED (Dec. 1995) 〈http://www.wired.com/wired/archive/3.12/ berger.if_pr.htm〉 (on file with the Harvard Law School Library).
262. For commentary that comes close to negative technological determinism, see KIRKPATRICK SALE, REBELS AGAINST THE FUTURE: THE LUDDITES AND THEIR WAR ON THE INDUSTRIAL REVOLUTION: LESSONS FOR THE COMPUTER AGE (1995), which notes that technology "comes with an inevitable logic, bearing the purposes and the values of the economic system that spawns it, and obeying an imperative that works that logic to its end, quite heedlessly." Id. at 262.
263. Miller & Brewer, supra note 118, at 2. For convincing evidence that intergroup friendship is highly correlated with decreased prejudice, see Thomas F. Pettigrew, Generalized Intergroup Contact Effects on Prejudice, PERS. & SOC. PSYCH. BULL. 173, 180-81 (1997).
264. See STOLL, supra note I, at 3, 58 (arguing that computer networks isolate us instead of bringing us together); Clay Calvert, Regulating Cyberspace: Metaphor, Rhetoric, Reality, and the Framing of Legal Options, 20 HASTINGS COMM. & ENT. L.J. 541, 562 (1998) (discussing the abandonment of face-to-face communities).
265. The study "examined the social and psychological impact of the Internet on 169 people in 73 [Pittsburgh] households during their first 1 to 2 years on-line." Kraut, Landmark, Kiesler, Mukopadhyay & Scherlis, supra note 72, at 1017. Participants agreed to have their Internet usage automatically tracked and answered periodic questionnaires measuring social involvement and psychological well-being. See id. at 1020-21.
266. Kraut's methodology has been subject to criticism. See, e.g., Autumn De Leon, Robert Kraut & Donna Hoffman, Mass Melancholy: A Carnegie Mellon Study Takes a Look at the Effects of Being Online and Comes Up with This Diagnosis:, TIME, Nov. 2, 1998, available in 1998 WL 21377749
267. (on file with the Harvard Law School Library) ("There are so many flaws in [the Kraut study] that it's very difficult to draw any conclusions about the effects of the Internet on individuals.") (quoting Donna Hoffman, Professor of Management at Vanderbilt University); see also Robert Kraut, Sara Kiesler, Tridas Mukhopadhyay, William Scherlis & Michael Patterson, Social Impact of the Internet: What does it Mean?, COMM. ACM, Dec. 1998, at 21 (offering commentary that qualifies and defends the study's research).
268. For example, a survey by James Katz and Philip Aspden found that, "[f]ar from creating a nation of strangers, the Internet is creating a nation richer in friendships and social relationships." James E. Katz & Philip Aspden, A Nation of Strangers?, COMM. ACM, Dec. 1997, at 81, 85. Interestingly, of the 601 Internet users interviewed, 14% had met at least one person through the Internet whom they considered a "friend." See id. at 85. Moreover, 60% of these people reported meeting at least one Internet friend in real space. See id. at 86.
269. See Reid, supra note 90, at 164, 175 (quoting one poster, who wrote "I don't care how much people say they are, muds are not just games, they are *real*!!! My mud friends are my best friends, they are the people who like me most in the entire world. Maybe the only people who do . . . . They are my family, they are not just some dumb game."). There are numerous supporting testimonials. For example, Sherry Turkle describes "Peter," who in real life had few friends. The MUD he participated in became a part of his everyday life. In that virtual world, he successfully courted a female player, something he could not readily do in real space. See Sherry Turkle, Constructions and Reconstructions of Self in Virtual Reality: Playing in the MUDs, 1 MIND, CULTURE & ACTIVITY 158, 161-62 (1994)
270. [hereinafter Turkle, Constructions]. Another MUD enthusiast, who is a man (in real space) playing a woman who is pretending to be a man (in cyberspace), proclaimed that "this [online life] is more real than my real life." Sherry Turkle, Multiple Subjectivity and Virtual Community at the End of the Freudian Century, 67 SOC. INQUIRY 72, 73 (1997) [hereinafter Turkle, Multiple Subjectivity].
271. See Turkle, Constructions, supra note 203, at 163.
272. Relationships initially forged through Internet Relay Chat, for example, have led to real-space marriage. See Christopher C. Werry, Linguistic and Interactional Features of Internet Relay Chat, in COMPUTER-MEDIATED COMMUNICATION, supra note 72, at 50;
273. see also Reid, supra note 90, at 175-76; cf. Keith F. Durkin & Clifton D. Bryant, "Log On To Sex": Some Notes on the Carnal Computer and Erotic Cyberspace as an Emerging Research Frontier, 16 DEVIANT BEHAV. 179, 186 (1995) (relating an example of one person who used sexual talk in cyberspace to get more excited about sex in real life).
274. See Wellman & Gulia, supra note 70, at 174 (describing the creation of a trust fund in memory of a computer scientist who was murdered, and the sending of books to a virtual community member who suffered a fire).
275. See generally Furlong, supra note 182, at 146-47 (describing a senior electronic community initiated at the University of San Francisco in March 1986). One participant wrote: "SeniorNet has given me the first time in memory to talk with people my own age. It's BEAUTIFUL!" Id. at 152.
276. Sherry Turkle relates an interview with a woman who was about to meet an online love interest face-to-face for the first time. The woman said: I feel very different online. I am a lot more outgoing, less inhibited. I would say I feel more like myself. But that's a contradiction. I feel more like I wish I was. I'm just hoping that face-to-face I can find a way to spend some time being the online me. SHERRY TURKLE, LIFE ON THE SCREEN: IDENTITY IN THE AGE OF THE INTERNET 179 (1995). A 26-year-old clerical worker admitted feeling more like "himself because he was able to present multiple aspects of his identity. See id. at 185.
277. See Kollock & Smith, supra note 60, at 16; Wellman & Gulia, supra note 70, at 187.
278. Wellman & Gulia, supra note 70, at 171.
279. See id. (reporting that except for family and small friend clusters, most members of a person's community network do not "really know each other"); see also id. at 188 (noting that even the French go to cafés less often).
280. Ninety-five percent of the United States population watches some television daily. The television is on nearly eight hours per day in the home. The average American adult watches five hours of television per day. See JERRY MANDER, IN THE ABSENCE OF THE SACRED: THE FAILURE OF TECHNOLOGY AND THE SURVIVAL OF THE INDIAN NATIONS 75-77 (1992) (citing a Nielsen study). But cf. TAPSCOTT, supra note 79, at 30 (reporting that the average America Online household watches 15% less TV).
281. See TAPSCOTT, supra note 79, at 25-26 (arguing that children spend time on the Internet much more actively than while watching television).
282. Miller & Brewer, supra note 118, at 2; see also Yehuda Amir, Contact Hypothesis in Ethnic Relations, PSYCHOL. BULL. 319, 334 (1969) ("The effectiveness of interracial contact is greatly increased if the contact is sanctioned by institutional support. The support may come from the law, a custom, a spokesman for the community or any authority which is accepted by the interacting group. In many cases, institutional support comes simply from a social atmosphere or a general public agreement.").
283. In addition, a well-drafted registration agreement that spells out clearly the rights and responsibilities of community members and contains enforceable arbitration and forum selection clauses will help virtual community hosts to manage private suit liability.
284. Incidentally, this is not what is generally done, even in private, for-profit virtual communities that already have the registration necessary to terminate accounts. For example, after my fighting words attack, I was told simply to put the perpetrator on "mute." Only if the perpetrator started to make public racist comments, rather than private comments directed solely at me, would the corporation intervene.
285. See Reidenberg, supra note 21, at 581 (discussing how code can produce ex ante obedience, which is more effective than ex post enforcement).
286. But see Lessig, The Law of the Horse, supra note 21, at 518-19 (describing the potential dangers of using code to enforce norms).
287. I mentioned earlier that one silver lining to the dark cloud of racism is that it may reveal the persistence of racism. See supra p. 1174. However, systems operators should not keep online users who spew racist bile simply to teach this lesson. The costs clearly outweigh the benefits.
288. Cf. Elaine K. Ginsberg, Introduction: The Politics of Passing, in PASSING
289. To keep my analysis manageable, I am trying to avoid larger debates about identity as performance. The standard citations are to JUDITH BUTLER, BODIES THAT MATTER: ON THE DISCURSIVE LIMITS OF "SEX" (1993)
290. and JUDITH BUTLER, GENDER TROUBLE: FEMINISM AND THE SUBVERSION OF IDENTITY (1990) [hereinafter BUTLER, GENDER TROUBLE].
291. Another option, but one with high costs, is surgery. See, e.g., DERY, supra note 15, at 230 (discussing surgery as a form of morphing and providing the example of Cindy Jackson, who has undergone more than 20 surgeries in order to look like Barbie).
292. Ginsberg, supra note 220, at 16.
293. See Nakamura, supra note 89, at 191 ("A diversification of the roles which get played, which are permitted to be played, can enable a thought provoking detachment of race from the body, and an accompanying questioning of the essentialness of race as a category.").
294. Identity switching is not a radical idea. Each of us, in real space, performs numerous roles in different situations. At times, we are citizens to the state; neighbors to the community, parents to our children; teachers to our students; confidants to our friends; lovers to our partners. We play these different roles by wearing different uniforms (a suit and tie at work; jeans and a t-shirt at home), using different vocabularies (affectionate nicknames at home; fancy discourse in front of professional colleagues) and languages (Korean with parents; English with most everyone else), adopting different gestures and mannerisms (respectful body language during a martial arts class; slouching and slacking with college buddies). We are comfortable with the idea of playing multiple roles, as the circumstances require. In a transmutation environment, cyberspace extends our role-playing technique by making it substantially easier to transmit a different racial signal, and to adopt multiple, simultaneously extant characters. Cf. Turkle, Multiple Subjectivity, supra note 203, at 78 (suggesting that once people accept the idea of a flexible identity, the jump to a multiple identity is easy - "a matter of semantics"). Even in real space there are famous examples of identity switching in the form of racial and gender passing. See Ginsberg, supra note 220, at 1-14 (discussing the cases of Howard Griffin and Grace Calsell, White journalists passing as Black, as well as those of Billy Tipton (jazz musician), Salvador Sanchez (bullfighter), and Deborah Sampson (American Revolutionary War soldier), women passing as men).
295. I focus on more serious acts of transmutation because trivial ones will have little impact. For example, many popular video games entail a form of cyber-passing. However, playing Lara Croft in Tomb Raider teaches nothing about being a White woman; playing one of the Mario Brothers teaches nothing about being Italian.
296. See Turkle, Constructions, supra note 203, at 162-63 (giving examples of people adopting different personalities online, such as men "daring to be passive").
297. Mark Poster makes a similar point about choosing gender. See Mark Poster, Cyberdemocracy: The Internet and the Public Sphere, in VIRTUAL POLITICS, supra note 94, at 222-24.
298. Pseudonyms can build up reputations. Hackers who engage in illegal activities have strong incentives to change their identities repeatedly in order to avoid getting caught by the authorities. This would, however, involve constant destruction of their hacking reputations. Not surprisingly, hackers maintain their pseudonyms. See Wellman & Gulia, supra note 70, at 177. See generally David G. Post, Pooling Intellectual Capital: Thoughts on Anonymity, Pseudonymity, and Limited Liability in Cyberspace, 1996 U. CHI. LEGAL F. 139, 132-67 (exploring the connection between pseudonymity, legal fictions, and limited liability).
299. Services, such as Zero-Knowledge Systems' Freedom 1.0, now enable an individual to operate under multiple pseudonyms. A user can surf the Net, chat, and send and receive e-mail without ever disclosing his or her real-space identity. See Freedom (visited Jan. 17, 2000) 〈http:// www.freedom.net〉 (on file with the Harvard Law School Library);
300. see also Patrick Norton, First Looks: Freedom i.o (last modified Dec. 23, 1999) 〈http://www.zdnet.com/pcmag/stories/first-looks/0,6763,2413285,00. html〉 (on file with the Harvard Law School Library).
301. Cf. Elizabeth M. Reid, Communication and Community on Internet Relay Chat: Constructing Communities, in HIGH NOON ON THE ELECTRONIC FRONTIER, supra note 67, at 336 (describing how a man, by playing a woman, felt liberated from having to act macho); Turkle, Multiple Subjectivity, supra note 203, at 79-80 (describing how a timid man learned to become more assertive in real life by playing an aggressive female in cyberspace).
302. Many commentators have claimed that cyberspace teaches empathy across genders. For example, men who cyber-pass as women realize that women in cyberspace are presented with unwanted sexual attention and unsolicited assistance. See, e.g., Curtis, supra note 123, at 355; cf. TURKLE, supra note 208, at 211 (reporting that when she played a male character in a MUD: "I finally experienced that permission to move freely I had always imagined to be the birthright of men. Not only was I approached less frequently, but I found it easier to respond to an unwanted overture with aplomb . . . .").
303. Here is another mechanism for learning interesting lessons. Imagine that I cyber-pass as a White person. In so doing, I may become privy to comments, jokes, and conversations that would not normally take place in front of me. This too would have pedagogical value. See, e.g., Adrian Piper, Passing for White, Passing for Black, in PASSING AND THE FICTIONS OF IDENTITY, supra note 220, at 234, 264 (describing insulting descriptions of Blacks made in front of the author, a woman who self-identifies as Black but who is generally mapped - by Whites and Blacks - as White).
304. This is not make-believe. One 30-year-old teacher explained that she has three identities on Internet Relay Chat: one engages seriously about Yugoslavia, another chats enthusiastically about Melrose Place, and a third is "very active on sexual channels, always looking for a good time." TURKLE, supra note 208, at 179. This is also not an invitation to multiple personality disorder. Individuals can come to grow comfortable with maintaining simultaneous, parallel identities. For instance, Sherry Turkic has interviewed college students who play multiple characters in multiple MUDs. She found that they could fragment their identities in non-pathological ways and play out different characters meaningfully in the various MUDs. See id. at 13; Turkle, Multiple Subjectivity, supra note 203, at 74. For some of them, real life was simply another window on a metaphorical computer screen. See TURKLE, supra note 208, at 13.
305. See Donath, supra note 193, at 29 (explaining that in real space, the norm is "one body, one identity," but that in cyberspace, the norm is different).
306. In contrast to real space, in cyberspace such identity play - under current laws and architectures - has far fewer negative repercussions. See Turkle, Multiple Subjectivity, supra note 203, at 77 (suggesting that individuals who buy Internet access "may find themselves in virtual communities taking on multiple roles; they may find themselves playing characters of different ages, attitudes, personalities, and genders").
307. Of course, it is hard to know whether transmutation will become sufficiently compelling in tomorrow's technology to be broadly popular. In addition, children may have more time and interest in transmutation environments. Cf. TAPSCOTT, supra note 79, at 86 ("Children regularly take on the personas of others or [different] avatars.").
308. A related objection is that our identity play may be constricted. For instance, in choosing body types or appearances, we seem to fall back to extremely conventional definitions of physical beauty. See, e.g., O'Brien, supra note 90, at 87 (explaining that when people cross genders, they adopt extremes). If so many of us are willing to go under the knife to present the "perfect" body in real space, should it be surprising that we choose such bodies in cyberspace, when the costs are zero? It is important, however, to distinguish between "beauty" and "race." Our aesthetics are clearly racialized. See, e.g., RUSSELL, WILSON & HALL, supra note 163, at 51 (stating that in 1990, Americans spent $44 million on chemicals to lighten their skin). Nevertheless, I do not believe that in a regime of transmutation, every individual will choose to play a White character. Put another way, even if there is great homogeneity in physical appearance, preferring the tall to the short, the thin to the fat, this does not mean that there will be homogeneity in the presentation of race, preferring the White to the non-White. I say this even if people prefer presenting light skin over dark skin. It is also important to recognize that these choices depend on the purpose of the space. In virtual pickup bars, it should not be surprising that people present "beautiful" avatars. It is not so clear that this would be the norm in a graphically mediated community focusing, for example, on computer operating systems.
309. See generally ERIC LOTT, LOVE AND THEFT: BLACKFACE MINSTRELSY AND THE AMERICAN WORKING CLASS (1993). Lott quotes Frederick Douglass for what Lott sees as the conventional wisdom about blackface minstrelsy: It was perpetrated by "the filthy scum of white society, who have stolen from us a complexion denied to them by nature, in which to make money, and pander to the corrupt taste of their white fellow citizens." Id. at 15. Lott challenges and complicates this view throughout his book.
310. Butler would press the provocative question: "How do I know what it means to act Asian?" Cf. BUTLER, GENDER TROUBLE, supra note 221, at 126 ("Note as well that the category of sex and the naturalized institution of heterosexuality are constructs, socially instituted and socially regulated fantasies or 'fetishes'."). For Butler, gender identity in real space is no less a performance than is the cyber-passing I describe.
311. Cf. Kendall, supra note 89, at 133 (expressing comparable skepticism about gender switching and wondering whether "crossing gender boundaries can strengthen those boundaries, rather than dissolve them"). Kendall relates an interesting conversation with a man, who had portrayed himself as a woman. He had claimed that he got much more attention as a "full-out . . . woman." Id. at 138. When Kendall, a woman in real space, explained that she did not get such attention, the man responded that she did not act like a traditional female. See id.
312. This may have been the intent of the cyber-passer in the first place. In the blackface example, the entertainer's desire is not really to deceive the audience at all. Indeed, part of "blackface" requires the audience to know that they are watching a minstrel show. Cf. Ginsberg, supra note 220, at 14 (explaining how, in certain forms of male transvestitism - a form of passing - the pass is meant to be revealed eventually to the audience).
313. Cf. Reid, supra note 231, at 334-36 (describing very negative reactions to gender-passing, including examples in which the gender-passing character in an adventure-style MUD was hunted down and killed).
314. I admit that it is hard to know how much cyber-passing must take place, and how much the audience must know before the racial signal is entirely discredited. It is now common knowledge that passing, across race as well as gender, is not uncommon. See id. at 337. Nevertheless, audiences still continue to credit the signals and to be "surprised" when the "pass" is revealed. Perhaps we willingly suspend disbelief when we log-on to MUDs and similar social spaces. See, e.g., Allucquère Rosanne Stone, In the Language of Vampire Speak: Overhearing Our Own Voices, in THE EIGHT TECHNOLOGIES OF OTHERNESS 38, 65 (Sue Golding ed., 1997) (suggesting that "sophisticated" users are more able to accept that everything is up for grabs).
315. For transcripts of Usenet arguments over the real-space racial identity of various posters, see Burkhalter, cited above in note 99, at 71. Similarly, women in some zones of cyberspace are often required to "prove" that they are women in real space. See Curtis, supra note 123, at 355.
316. The entire point of abolition is to be blind to race, whereas integration requires people to see each other's race and to interact with members of different races. For a concrete example of this conflict, consider that abolition would discourage the inclusion of photographs on personal home pages, but integration might approve of such a practice.
317. Abolition permits individuals to express every facet of their identity save one: race. Thus, the space of identity experimentation is significantly restricted. The transmutation approach finds this counterproductive. Not only does abolition flatten the number of dimensions of identity to explore generally, it removes the one specific dimension - race - that needs to be experimented with the most.
318. See, e.g., Lawrence Lessig, The Zones of Cyberspace, 48 STAN. L. REV. 1403, 1409 (1996) (describing the Communications Decency Act of 1996 and the NII White Paper as attempts to "induc[e] a technology for zoning").
319. A familiar example is the URL field in Web browsers, which identifies the name of the object currently browsed.
320. See Johnson & Post, supra note 119, at 1379 (describing how cyberspace boundaries may be defined by reference to screens and passwords).
321. One commentator rejects spatial analogies for cyberspace and instead speaks of invocation. By invoking or naming something, one calls it up on the computer interface. Thus, computer networks are better likened to speech than to travel. See Chris Chesher, The Ontology of Digital Domains, in VIRTUAL POLITICS, supra note 94, at 79, 83-84.
322. Cf. Lessig, supra note 19, at 889 (observing that zoning is increasing in cyberspace "with an efficiency unmatched in real space").
323. In fact, zoning in cyberspace is easier than zoning in real space because in cyberspace geography, there are fewer problems with negative externalities. In real space, there is a zoning problem when an adult bookstore seeks to set up shop next to an elementary school. In cyberspace, however, one can simply increase the "distance" between bookstore and school by refusing to hyperlink the two areas. Accordingly, visitors to the adult bookstore, whose cyber-borders are regulated through a password scheme, need not traverse the school playground; conversely, students need not walk through a cyberspace red light district to get to class. Of course, creating this "distance" becomes more complicated if the adult bookstore wants to be "closer" to the school. For instance, the adult site, White House (visited Nov. 23, 1999) 〈http://www.white-house.com〉
324. (on file with the Harvard Law School Library), purposefully tries to capture traffic intended for the presidential site, Welcome to the White House (visited Nov. 23, 1999) 〈http://www.white-house.gov〉 (on file with the Harvard Law School Library), by using a similar URL.
325. Cf. Lawrence Lessig, The Path of Cyberlaw, 104 YALE L.J. 1743, 1745 (1995) (advocating the "common law" approach to solving problems in cyberspace "with its partial answers, to repeated if slightly varied questions, in a range of contexts . . . in a temporally spaced dialogue of cases and jurisdictions").
326. See SCHULER, supra note 7, at 23 (noting the stronger and longer-lasting influence of early decisions when designing technological systems).
327. I attempted to do this in an earlier article on information privacy. See Kang, supra note 15, at app. (providing a complete draft of a cyberspace privacy statute).
328. See supra p. 1133.
329. Readers will rightly ask how I define "market" places. For example, is an investment club a market place or something else? I use this term only as shorthand, not as an analytic concept that helps us zone. "Market places" is a conclusory label I attach to those cyber spaces that involve an exchange of property, goods, or services that society has decided should be colorblind. This term, in and of itself, does not help us make that decision.
330. One could think of exceptions. For instance, a program to promote residential integration could offer cheaper mortgage rates to individuals of one race who purchase homes in communities that are predominantly of another race.
331. One could imagine a future technology that would enable workers to provide informational work product in exchange for electronic payment, all mediated without any face-to-face contact. Potentially, this narrow branch of employment could also see the abolition of race.
332. Cf. 42 U.S.C. § 1981(a) (1994) ("All persons within the jurisdiction of the United States shall have the same right in every State and Territory to make and enforce contracts . . . and to the full and equal benefit of all laws and proceedings for the security of persons and property as is enjoyed by white citizens . . . .") (emphasis added).
333. Racial discrimination persists in the housing market. Massey and Denton report a comprehensive 1977 HUD study, which found that Whites "were favored on 48% of transactions in the sales market and on 39% of those in the rental market." MASSEY & DENTON, supra note 122, at 100. They also report a 1987 study by George Galster, who re-analyzed 71 audit studies carried out during the 1980s. He concluded that "racial discrimination continues to be a dominant feature of metropolitan housing markets in the 1980s." Id. at 99.
334. See, e.g., REALTOR.COM (visited Nov. 23, 1999) 〈http://www.realtor.com〉 (on file with the Harvard Law School Library). 268 Of course, in theory, I could have tried to access such data even without the Internet. However, the high transaction costs of locating such information on paper and keeping it constantly updated would have been prohibitive.
335. See id. at 1238-41.
336. Excellent introductions to the legal landscape of information privacy can be found in MARC ROTENBERG, THE PRIVACY LAW SOURCEBOOK 1999: UNITED STATES LAW, INTERNATIONAL LAW, AND RECENT DEVELOPMENTS (1999);
337. PAUL M. SCHWARTZ & JOEL R. REIDENBERG, DATA PRIVACY LAW: A STUDY OF UNITED STATES DATA PROTECTION (1996);
338. and PETER P. SWIRE & ROBERT E. LITAN, NONE OF YOUR BUSINESS: WORLD DATA FLOWS, ELECTRONIC COMMERCE, AND THE EUROPEAN PRIVACY DIRECTIVE (1998).
339. See Kang, supra note 15, at 1246-94.
340. Cf. Julie E. Cohen, A Right to Read Anonymously: A Closer Look at "Copyright Management" in Cyberspace, 28 CONN. L. REV. 981, 1003-19 (1996) (arguing for a constitutional right to read anonymously in the context of copyrighted materials available in cyberspace).
341. Access to mortgages continues to depend on race. For a careful examination of the data, including consideration of nonracial explanations such as financial profiling of homebuyers and neighborhood characteristics, see OLIVER & SHAPIRO, cited above in note 54, at 136-47.
342. A readable introduction to cryptography in general and public key cryptography in particular can be found in SIMSON GARFINKEL, PGP: PRETTY GOOD PRIVACY 33-58 (1995).
343. The bible on the subject is BRUCE SCHNEIER, APPLIED CRYPTOGRAPHY: PROTOCOLS, ALGORITHMS, AND SOURCE CODE IN C (2d ed. 1996). For an introduction to these concepts as applied to electronic commerce, see Froomkin, cited above in note 128, at 449-79.
344. See David Chaum, Achieving Electronic Privacy, SCI. AM., Aug. 1992, at 96
345. (discussing digital signatures). A more technical, but still readable, discussion of how "digital pseudonyms" can both protect privacy and maintain security in automated transactions can be found in David Chaum, Security Without Identification: Transaction Systems to Make Big Brother Obsolete, 28 COMM. ACM 1030 (1985).
346. See, e.g., LendingTree.com (visited Nov. 23, 1999) 〈http://www.lendingtree.com〉 (on file with the Harvard Law School Library) (processing loan applications online).
347. If the lending institution knows the address of the home, then it can practice geographical redlining without knowing the race of the home buyer. We could try to engineer economic systems that take even this bit of information away from lending institutions. However, lenders would actively resist losing so much of their financial discretion, which typically depends on information about the home buyer (removed by the third party credentialing system), the property to be purchased, and the property's neighborhood.
348. Trusted third parties are necessary to any e-commerce system established on public key cryptography. An excellent overview of trusted third parties is A. Michael Froomkin, The Essential Role of Trusted Third Parties in Electronic Commerce, 75 OR. L. REV. 49 (1996).
349. A practical overview appears in Thomas J. Smedinghoff, Electronic Contracts & Digital Signatures: An Overview of Law and Legislation, 1999 THIRD ANN. INTERNET L. INST. 125.
350. For descriptions of what might go wrong with a certificate authority, which is one type of trusted third party, see Lonnie Eldridge, Comment, Internet Commerce and the Meltdown of Certification Authorities: Is the Washington State Solution a Good Model?, 45 UCLA L. REV. 1805, 1820-28 (1998).
351. This proposal would go beyond the Community Reinvestment Act of 1977, which uses federal financial regulatory institutions to monitor lending practices, such as the geographical distribution of loans. Supervisory government agencies are required to consider this information, which is available to the public, in approving any application for a bank charter, deposit insurance, merger or acquisition, or office relocation. See Deanna Caldwell, An Overview of Fair Lending Legislation, 28 J. MARSHALL L. REV. 333, 340-42 (1995).
352. To be more precise, race has not been totally filtered out. Specifically, the trusted credentialing agency knows my identity. Also, after the final execution, the lending institution will learn my identity as well. A truly anonymous system would be extraordinarily difficult to implement even if strong cryptography were widely used. See Peter P. S wire, The Uses and Limits of Financial Cryptography: A Law Professor's Perspective (visited Sept. 21, 1999) 〈http://www.osu.edu/ units/law/swire1/pscrypto.htm〉 (on file with the Harvard Law School Library) (discussing, inter alia, problems of adverse selection, end game defection, and key management).
353. The seller may not be as comfortable selling her home to an Asian American man. Faced with multiple offers at the same price, the seller may prefer to sell her home to a "nice" couple, who happen to be White. Maybe she will wonder, "That Chinese fellow might paint his house pink. I know those Chinese like pink. I couldn't do that to my neighbors."
354. An interesting issue worth exploring is the possibility that even electronic agents will be racialized. For example, one program that converts text into computer synthesized speech features four different voices. One voice is supposed to sound African American. Another voice is supposed to imitate Marilyn Monroe. See ReadPlease 2000 (visited Jan. 17, 2000) 〈http://www.read-please.com/rpindex.htm〉 (on file with the Harvard Law School Library).
355. This comparison between human and electronic agents nicely demonstrates what is and is not unique about cyberspace. One could imagine developing a race-blind economic infrastructure without the use of any computers or telecommunications. Indeed, even public key cryptography could be implemented by an army of human beings scribbling on paper with pencils. However, the transaction costs would be prohibitively expensive. By substantially lowering transaction costs, cyberspace makes race-blind transaction systems economically plausible.
356. Cf. DOUGLAS G. BAIRD, ROBERT H. GERTNER & RANDAL C. PICKER, GAME THEORY AND THE LAW 89-109 (1994)
357. (discussing how parties with "preferable" characteristics will try to distinguish themselves from less advantaged parties); Mitu Gulati, When Corporate Managers Fear a Good Thing Is Coming to an End: The Case of Interim Nondisclosure, 46 UCLA L. REV. 675, 702-03 (1999) (discussing why market forces would theoretically produce "unraveling").
358. One might compare this with Internet auto sites, which have grown broadly popular. See, e.g., Barbara Buell, Virtual Highway Easing the Way for Buying, Selling Cars, STANFORD ONLINE REPORT (last modified Jan. 6, 2000) 〈http://www.stanford.edu/dept/news/report/news/ january6/carbuy16.html〉 (on file with the Harvard Law School Library) (reporting studies that predict that by 2000, 50% of new car buyers will use the Internet for preliminary research and up to 20% will make the actual purchase online).
359. One objection is that it is wrong to strong-arm sellers or lenders into "trusting" buyers without face-to-face interaction. One might imagine either on philosophical or psychological grounds that trust should be granted on the basis of personal intimacy. Think, for instance, of the smalltown banker who gives out loans based not only on the "paper record," but also on "personal" and nonquantifiable judgments. Imposing a racially anonymous structure increases the commodification of these marketplace relations. This is a fair point. But I consider this to be a marginal increase in commodification in already heavily commodified marketplaces. We romanticize the market when we think that we have close, personal relations with the local banker, car dealer, and insurance broker. Also, in my calculus, the benefits of decreasing racial discrimination (which increases fairness to racial minorities) outweigh the costs of increased commodification (which marginally increases general marketplace alienation).
360. For example, even if sellers and lenders do not know the buyer's race, buyers will know the racial makeup of the neighborhood. If Whites prefer to live in a mostly White neighborhood - as opposed to an all-White neighborhood or a predominately Black neighborhood - then White buyers will pay a premium for such neighborhoods. This premium might indirectly price Blacks, who simply do not place as much value on a predominately White neighborhood, out of such communities. See David M. Cutler, Edward L. Glaser & Jacob L. Vigdor, The Rise and Decline of the American Ghetto, 107 J. POL. ECON. 435, 493-96 (1999) ("[W]hites still prefer to live with other Whites more than blacks prefer to live in white areas."); Sander, supra note 122, at 986-88. Moreover, after move-in, the neighbors will obviously discover the race of the new residents. A family may fear harassment, hostility, and even violence and thus may decline to move into a neighborhood that is predominately of another race. See id. at 985; see also id. at 1009-10 (citing additional reasons why reducing discrimination in housing sales may not significantly affect residential segregation).
361. To repeat, "authentication" does not involve examining whether my behaviors and comments are "authentic" according to some essentialized definition of a racial category. Cf. SCHNEIER, supra note 275, at 2 (describing authentication in the cryptography context as preventing an intruder from masquerading as someone else).
362. Nakamura, supra note 89, at 187.
363. I can imagine some interesting scenarios in which the state could purposefully generate transmutation environments. Consider the extensive matrix of websites maintained by the federal government. In some future world, all of these sites may be tightly linked together into a sprawling FedWorld. See FedWorld (visited Jan. 22, 2000) 〈http://www.fedworld.gov〉 (on file with the Harvard Law School Library). Indeed, one could imagine a future in which FedWorld becomes an important way that citizens interact with the federal government, in a three-dimensional graphically articulated space. What should the avatars look like, and who would decide? What if each visitor to FedWorld were assigned a randomly racialized, gendered avatar? Then, a White male partner from a Washington, D.C., regulatory law firm might have to navigate FedWorld as a Black woman. What if we allowed individuals to change their avatars, but not without cost? Would the partner do so? What if FedWorld required registration and whatever avatar you were given would have to be kept for 10 years, like a passport? More interestingly, what if we allowed individuals generally to pick the race that they presented, but for some small percentage of individuals, we randomly altered how they appeared to others? It would have to be a small percentage; otherwise, the racial signal transmitted in FedWorld would be discredited as meaningless. Even though one saw oneself as, say, White, everyone else in FedWorld would see that same avatar as Black. What would happen if the White attorney were dissatisfied with the service of a government bureaucrat? Would he start thinking that maybe he was presenting a Black identity to everyone else in FedWorld? I realize that such a transmutation regime could raise some of the same problems that I mentioned when discussing abolition. For instance, there is the possibility of "psychic harm" if racial minorities are forced to play White characters. On the other hand, I think many racial minorities would play that game if Whites also had to suffer the psychic dislocation of playing a racial minority. I further realize that the federal government would undoubtedly shy away from such experiments as controversial.
364. I do not want to take the status quo for granted. If, for example, every cyberspace interaction disclosed an identifying number - whether an IP address or a microprocessor serial number - both abolition and transmutation environments would be jeopardized.
365. See, e.g., Paul Schwartz, Data Processing and Government Administration: The Failure of the American Legal Response to the Computer, 43 HASTINGS L.J. 1321, 1344-48 (1995).
366. See Wright, supra note 35, at 46 (discussing the multiracial category for the 2000 census).
367. Cf. McIntyre v. Ohio Elections Comm'n, 514 U.S. 334, 341-43 (1995) (invalidating a bar on anonymous campaign literature). For further discussion on the constitutionality of banning anonymous speech, see A. Michael Froomkin, Anonymity and Its Enmities, 1995 J. ONLINE L. art. 4 〈http://www.wm.edu/law/publications/jol/froomkin.html〉 (on file with the Harvard Law School Library).
368. Examples can be found in MUDs as well. See, e.g., Bruckman, supra note 67, at 319-20 (describing the professional MUDs of astrophysicists and media researchers).
369. See, e.g., Nancy K. Baym, The Emergence of Community in Computer-Mediated Communication, in CYBERSOCIETY, supra note 7, at 138, 155 (describing the use of real names in the popular soap opera newsgroup r.a.t.s.). Baym explains that soap operas are constantly assessed for socioemotional realism, which requires critics to disclose something about their own areas of experience and expertise. See id. "The use of real names helps to create an intimate environment in which this kind of disclosure can be voiced." Id. Accordingly, participants in this particular newsgroup actively discourage anonymity and pseudonymity. See id. 298 See Donath, supra note 193, at 30 ("The basic premise is that the users are who they claim to be.").
370. See Donath, supra note 193, at 30 ("The basic premise is that the users are who they claim to be.").
371. See, e.g., Kira Hall, Cyberfeminism, in COMPUTER-MEDIATED COMMUNICATION, supra note 72, at 147, 161-62 (describing a women's community whose members occasionally undertook real-space verification of a new member's gender). Some system operators will make unannounced telephone calls to registrants to listen to their voices to determine whether they are male or female. See, e.g., O'Brien, supra note 90, at 86.
372. It is difficult to provide a bogus e-mail address because a password is often sent to the e-mail address provided during registration. Without the password, an individual cannot participate fully in the virtual community. To get the password, however, there must be a functioning e-mail account. Although it is not currently difficult to create an e-mail account while cloaking one's real-space identity, it does take some effort and knowledge.
373. Commercial e-mail addresses (for example, _@yahoo.com) are easier to create with pseudonyms than institutional e-mail addresses (for example, _@ibm.com). Cf. Donath, supra note 193, at 37.
374. See Johnson & Post, supra note 119, at 1388.
375. Cf. Katie Hafner, Guessing Who Is Online, N.Y. TIMES, July 22, 1999, at G4 (describing a game in which the goal is to ask questions to identify the gender-impostor). Interestingly, Amy Bruckman, a co-designer of the game, stated: "We haven't had the guts to try a race game yet."
376. Katie Hafner, N.Y. TIMES, Id. (internal quotation marks omitted).
377. Judith Donath reports that, in a newsgroup that deals with weightlifting, one person challenged another's strength claims by posting a photograph and requesting that the other do the same. See Donath, supra note 193, at 50. She also adds that a social norm of disclosing personal webpages increases the cost of changing identities given the nontrivial costs of setting up websites. See id. at 56.
378. It would not be very difficult for someone who is determined to maintain another identity to create a fake personal webpage. But every incremental increase in the costs of cyber-passing decreases the likelihood that one will cyber-pass.
379. See Donath, supra note 193, at 32 ("If a signal becomes very unreliable due to excessive cheating it ceases to convey information - it stops being a signal. Yet there are stable systems of deception, where the percentage of deceivers does not overwhelm the population, and a signal remains information-bearing, however imperfectly.").
380. See Hoffman & Novak, supra note 8, at 390; see also NTIA, supra note 129, at 78. These access points may be especially important because Internet use requires not only software and hardware, but also cognitive skills and training. See Wilhelm, supra note 141, at 531.
381. In the future, such high-speed access from the home may be practically necessary to participate fully in social spaces, especially on terms of equal status. However, private corporations in pursuit of self-interest may choose not to build-out in predominantly minority areas, which would shut out even affluent members of racial minorities who live in segregated minority communities. This discussion is speculative because the business models for household high-speed data access are in flux. In addition, this discussion depends on which technologies we are considering. For instance, the economics of providing high-speed data access through direct broadcast satellite differ fundamentally from those of rebuilding cable or telephony networks, line by line.
382. For example, in WorldsAway (now called Avaterra), one must generally walk through adjacent spaces to get from point A to point B. However, there are special teleporters that will transport people to distant locations. See Avaterra.com, Inc. Backgrounder, supra note 68.
383. See Curtis, supra note 123, at 365 (describing the "enormous popularity" of a "food fight" game in the LambdaMOO MUD).
384. Cf. Reid, supra note 126, at 121 (describing adventure-style MUDs, in which individual players must cooperate to destroy monsters that cannot be killed individually).
385. See Pettigrew, supra note 118, at 279 ("[T]he preponderance of social-psychological evidence attests to the greater efficacy of the opposite approach. Behaving differently more often precedes thinking differently [than vice versa].").
386. See, e.g., Dragon Systems, Inc.: Support & Services: Discussion Forum (visited Sept. 24, 1999) 〈http://www.dragonsys.com/support/discforum.html〉 (on file with the Harvard Law School Library).
387. See, e.g., Peter Kollock, The Economies of Online Cooperation: Gifts and Public Goods in Cyberspace, in COMMUNITIES IN CYBERSPACE, supra note 60, at 220, 235 (suggesting that one way to encourage cooperation in virtual communities is to give credit where credit is due).
388. See Bargh, Chen & Burrows, supra note 45, at 230 ("Stereotypes become active automatically on the mere presence of physical features associated with the stereotyped group, and categorizing behavior in terms of personality traits and then making dispositional attributions about the actor's personality have both been shown to occur automatically to some extent.") (citations omitted).
389. See Microsoft Outlook 2000 Products Enhancement Guide (last modified Oct. 1998) 〈http:// microsoft.com/office/enterprise/prodinfo/outlkpeg.htm〉 (on file with the Harvard Law School Library) (discussing the implementation of v-card technology).
390. Increased use of avatars may have the same effect. Of course, it is possible to use avatars that are not human, and therefore not racialized. However, using nonhuman avatars may be inconsistent with the norms of certain virtual communities, such as academic lists. Further research should be done on whether our racial schemas are triggered differently as a function of avatar size and photo realism.
391. This delay technique can be seen as a pragmatic combination of the abolition and integration paradigms. This is an attempt to answer the challenge of modern racism, which has been identified as "[h]ow to promote in the white public a racial understanding that avoids the psychological liabilities of ingroup and outgroup categorization while acknowledging the full sociological implications of the past and continuing color line." Marylee C. Taylor & Thomas F. Pettigrew, Prejudice, in 3 ENCYCLOPEDIA OF SOCIOLOGY 1536, 1539-40 (1992).
392. See, e.g., Peter Kollock, Design Principles for Online Communities (visited Feb. 7, 2000) 〈http://www.sscnet.ucla.edu/soc/faculty/kollock/papers/design.htm〉 (on file with the Harvard Law School Library) (identifying general design principles for successful online communities, which include internal economy, a coherent sense of place, rituals, history of community, identity persistence, moderate risk, and likelihood of unplanned interaction).
393. See, e.g., Charles Piller, E-Bonding Via Voice on the Web, L.A. TIMES, Sept. 29, 1999, at A1.
394. 47 U.S.C. § 230(c)(1) (Supp. III 1998).
395. Id.; Blumenthal v. Drudge, 992 F. Supp. 44, 50-53 (D.D.C. 1998); Zeran v. America Online, Inc., 958 F. Supp. 1124, 1132-33 (E.D. Va. 1997).
396. See Taylor & Pettigrew, supra note 318, at 1537. For a historical essay on changing American racial attitudes, see HOWARD SCHUMAN, CHARLOTTE STEEH, LAWRENCE BOBO & MARIA KRYSAN, RACIAL ATTITUDES IN AMERICA: TRENDS AND INTERPRETATIONS 8-53 (1997).
397. This and related terms, such as "Information Age," are now common. For an insightful, sophisticated analysis of the various definitions and theories of the information society, see FRANK WEBSTER, THEORIES OF THE INFORMATION SOCIETY (1995).
398. On affirmative action's role in creating a racially integrated university, see WILLIAM G. BOWEN & DEREK BOK, THE SHAPE OF THE RIVER: LONG-TERM CONSEQUENCES OF CONSIDERING RACE IN COLLEGE AND UNIVERSITY ADMISSIONS (1998).
399. Perhaps this pragmatic approach toward racial zoning can be imported back into real space. For example, many opponents of race-based affirmative action in higher education argue that the state sends mixed messages when it tells employers, on the one hand, to ignore race, but admissions officers, on the other hand, to consider race. This contradiction can be seen as an example of real-space racial zoning. Certain employment spaces can be zoned "abolition," at the same time that certain education spaces are zoned "integration."
400. This attempt is an exploration of what I call critical race technologies, rooted in what Eric Yamamoto calls "critical pragmatism." ERIC K. YAMAMOTO, INTERRACIAL JUSTICE: CONFLICT & RECONCILIATION IN POST-CIVIL RIGHTS AMERICA 133 (1999).
401. A critical race technology takes into account the power of architecture, digital as well as spatial, in the construction of racial mechanics. For examples of the latter, see Keith Aoki, Race, Space, and Place: The Relation Between Architectural Modernism, Post-Modernism, Urban Planning, and Gentrification, 20 FORDHAM URB. L.J. 699 (1993);
402. and Richard Thompson Ford, Geography and Sovereignty: Jurisdictional Formation and Racial Segregation, 49 STAN. L. REV. 1363 (1997).
403. Critical Race Theorists should also recall C.P. Snow's warning: It is dangerous to have two cultures which can't or don't communicate. In a time when science is determining much of our destiny, that is, whether we live or die, it is dangerous in the most practical terms. Scientists can give bad advice and decisionmakers can't know whether it is good or bad. On the other hand, scientists in a divided culture provide a knowledge of some potentialities which is theirs alone. C.P. SNOW, THE TWO CULTURES: AND A SECOND LOOK: AN EXPANDED VERSION OF THE TWO CULTURES AND THE SCIENTIFIC REVOLUTION 98 (1964) (internal citation omitted).
404. [T]he Internet is more like a social space than a thing so that its effects are more like those of Germany than those of hammers. The effects of Germany on the people within it is [sic] to make them Germans (at least for the most part); the effects of hammers is [sic] not to make people hammers . . . . As long as we understand the Internet as a hammer we will fail to discern the way it is like Germany. Mark Poster, Cyberdemocracy: The Internet and the Public Sphere, in VIRTUAL POLITICS, supra note 94, at 212, 216.
405. According to Forrester Research, youths aged 16 to 22 spend nine hours weekly surfing the web, 38% more time than adults. Also, a stunning 47% of youths are Internet users. See David Plotnikoff, Young Internet Explorers Will Never Know a World Without the Net (last modified Oct. 9, 1999) 〈http://www.mercurycenter.com/svtech/columns/modemdriver/docs/dp101099. htm〉 (on file with the Harvard Law School Library).