Go to this issue's Table of Contents.
Carol Gilligan et al. versus Christina Hoff Sommers
Mail response to Christina Hoff Sommers's "The War Against Boys," in the May Atlantic, was swift and substantial. More than a hundred readers offered their opinions of the piece, including what appears to be an entire high school English class from Easton, Connecticut. Admirers of the article outnumbered critics, but by a small margin. What follows is a sample of the correspondence. It begins with a letter from Carol Gilligan, a particular object of Sommers's criticism
Contrary to Christina Hoff Sommers's claim that my research has not been subject to peer review, the findings of all the research studies referred to in In a Different Voice and the subsequent research with adolescent girls and women were published in refereed journals, including Developmental Psychology, the Journal of Personality and Social Psychology, Human Development, and other leading journals in the field; in several volumes of New Directions in Child Development, edited by prominent researchers; and in five peer-reviewed books. Sommers says that she couldn't find out "how subjects were chosen, how interviews were recorded, and the method by which meaning was derived from the data" -- all of which is clearly described in these publications and in monographs that explain the methodology for the studies so that other researchers can replicate them.
Sommers may be confused on this issue because her research assistants (identifying themselves only as students) asked not for published papers or monographs describing the research and methods but for the original data -- meaning, in this case, verbatim transcripts of research interviews. We could not release these transcripts without violating legally and ethically binding agreements to protect the confidentiality of the people we interviewed. Because protecting the anonymity of people often demands more than simply changing their names, my assistant and I have been working with the Henry A. Murray Research Center, at the Radcliffe Institute for Advanced Study, to develop a procedure by which scholars can examine transcripts of in-depth interviews while guaranteeing the privacy of those who participated in the research.
In focusing on questions of whether my research fits her model of standard practice, Sommers not only is inaccurate but also misses the larger point and finds controversy where none exists. The observation that boys are statistically more at risk than girls in early childhood, while girls are more likely to first show symptoms of psychological distress in adolescence, has been in the psychology literature for more than a hundred years and has been widely replicated. The absence of girls and women from major studies of psychological development is readily seen in the original publications, where researchers such as Piaget and Kohlberg clearly describe their use of all-male samples to form theories of moral development. The 1980 Handbook of Adolescent Psychology criticized the omission of girls from studies of adolescence as a major problem in the field. Citing these facts and using the evidence of my research, I have identified new questions about development that reframe the discussion of differences and have offered a way to account for girls' greater resilience in childhood and the heightened vulnerability of young boys and adolescent girls.
There are important unsettled issues in the study of both girls' and boys' development. The results of my research do not lend themselves to simple statements such as "Girls are thriving" and "Girls are at risk." Girls and boys are strong and vulnerable, although in somewhat different ways, and what we are to conclude about what this means for raising our children is a matter for discussion open to many points of view. It's sad that Sommers squanders the opportunity to open a constructive dialogue on our differences of opinion and interpretation and chooses instead to publish an inaccurate, ad hominem attack that can lead nowhere.
Christina Hoff Sommers replies:
In Carol Gilligan's introduction to In a Different Voice she informs readers that her theses in that book are based on research done for three studies she calls "the college student study,""the abortion decision study," and "the rights and responsibilities study." These studies, she says, provide empirical support for her central and most intriguing claim: that women have a distinctive moral voice that expresses "an ethic of care and responsibility," rather than a characteristically male "ethic of justice and right." She has often been challenged to produce that research. Her letter now says that it is out there for anyone to see.
Readers will note, however, that she does not actually direct us to research supporting her hypothesis; instead she refers us to some journals, to volumes "edited by prominent researchers," and to "peer-reviewed books" where, she assures us, such research can be found. In fact it is nowhere to be found.
1. Developmental Psychology, "Moral Development and Reconstructive Memory: Recalling a Decision to Terminate an Unplanned Pregnancy," by Georgia Blackburne-Stover, Mary Field Belenky, and Carol Gilligan, 1982. In this small study (twenty-four women) subjects who were considering terminating a pregnancy were later queried about how they recollected this memorable event in their lives. The article, which was about how one's ethical maturity may affect one's memory of a critical event, reached no conclusions about how men and women differ in making moral decisions (which may explain why it is never mentioned in In a Different Voice).
2. Journal of Personality and Social Psychology, "Images of Violence in Thematic Apperception Test Stories," by Susan Pollak and Carol Gilligan, 1982. This study comes closest to being relevant to Gilligan's contention in In a Different Voice. It analyzes themes of violence in TAT stories and confirms that men's fantasies are more violent. But it does not directly address the question of male-female differences in moral reasoning or perceptions, nor does it arrive at any conclusions about different moral orientations.
3. Human Development, "Moral Development in Late Adolescence and Adulthood," by J. Murphy and Carol Gilligan, 1980. This article is a technical critique of Lawrence Kohlberg's stage theory and a defense of an alternative approach by W. Perry. It actually came to conclusions that contradict Gilligan's thesis: it found no significant statistical differences between men and women in moral orientation. The subjects studied were few in number. After several subjects dropped out, only twenty-one men and five women remained. In their conclusion Murphy and Gilligan acknowledged the negative results but explained them by saying "Failure to find significance [in sex differences] is at least partially due to the small number of subjects."
Why does Gilligan think of these articles as supporting her claims? I believe she considers them to be research for In a Different Voice because she used some of the interview transcripts as data for her three pilot studies. Using interviews and data from previous studies for a different study supporting new conclusions is an acceptable practice -- but the new study must then be peer-reviewed to see whether the data do indeed help to make the case for the new contention. No such review has been attempted, so far as I know.
For more than fifteen years scholars in the fields of moral and developmental psychology have been rebuking Gilligan for the paucity of her research. In 1983 Mary Brabeck, a Boston College psychologist who is now BC's dean of education, succinctly stated the fatal weakness in relying on Gilligan's three unpublished studies: "No quantitative data are reported for any of these three studies. Evidence for Gilligan's theor[ies] ... rests on quoted excerpts from interviews and her interpretations of these selected excerpts."
Three years later, in 1986, at a forum on Gilligan's work organized by the feminist journal Signs, several critics queried Gilligan about her missing evidence. She replied that her work was never intended to be taken as conventional data-based social science to prove an empirical generalization. "The argument [of my book],"she said, "was not statistical -- that is, not based on the representativeness of the women studied or on the generality of the data presented to a larger population of women or men. Rather, the argument was interpretive and hinged on the demonstration that the examples presented illustrated a different way of seeing."
Five years later Susan Faludi quoted Gilligan as saying "I would never want to say this is an exhaustive group of people. [In a Different Voice] was a very small piece of work with three little pilot studies." Faludi observed that Gilligan "does not hold out her studies as scientific research efforts," seeing them rather as "interpretive." "Even so," Faludi said, "Gilligan doesn't give readers the basic data they do need to evaluate her case studies."
When I reviewed Gilligan's published work, I was persuaded that her critics were right. But I wanted access to those "three little pilot studies." My research assistant, Hugh Liebert, e-mailed Gilligan asking her where the studies could be found. Gilligan replied to Liebert through her assistant, Tatiana Bertsch, who wrote, "I am writing on behalf of Professor Gilligan.... None of the In a Different Voice studies have been published.... Sorry that none of what you are interested in is available."
Now, in an astonishing departure from previous explanations, Gilligan claims that "the findings of all the research studies referred to in In a Different Voice ... were published in refereed journals." Moreover, she says that the published research does indeed satisfy the standard social-science protocols for soundly presented replicable research in support of the generalizations about sex differences for which she is famous. That claim cannot be credited.
In the final analysis, the (un)availability and (in)adequacy of Gilligan's research are the concern of her professional colleagues. My Atlantic article touched on In a Different Voice only because that is the book that made Gilligan famous and influential and because I firmly believe that her influence has been harmful -- to boys especially, but to girls as well.
Gilligan went on to successfully promote the sensational but empirically baseless "finding" that America's adolescent girls were being "silenced" and demoralized as they were raised in our "male-voiced" culture, a culture that also harms boys by pressuring them "to internalize a patriarchal voice."
Gilligan's pronouncements on girls and boys are more ideological than scientific. Few academic psychologists take them seriously. In "gender studies" and in many schools of education, however, Gilligan is taken very seriously indeed, and not least because she is thought of as the famed Harvard scholar who did that "landmark research" showing that men and women speak morally in different voices.
In discussing the research undertaken by my late wife, Myra, and me, Christina Hoff Sommers implies that our work is not peer-reviewed. Worse, she claims that our research report "turns out to be missing." As thousands of students and scholars who use our studies know, the "missing" report, several doctoral dissertations, a score of articles in peer-reviewed journals, and several books are available through thousands of libraries in the United States and overseas. The report Sommers describes as "missing" is in fact part of the Educational Research and Information Clearinghouse, Promoting Effectiveness in Classroom Instruction: Year 3 Final Report, National Institute of Education, Washington, D.C. (call number ED 257819).
While ignoring the major findings of our study, and even the existence of the many other studies that document classroom bias, Sommers focuses on the gender call-out gap. We documented that teachers in the classroom allowed boys to call out answers but admonished girls who called out with reminders like "Raise your hand if you want to speak." Individual classrooms differ dramatically in the rate of the male call-out advantage. In our pilot study we found that boys called out eight times as often as girls. In our full study, which involved many more classrooms, we found a two-to-one male advantage. In the 1995 edition of Failing at Fairness we describe it this way: "Our research shows that boys call out significantly more often than girls" (p. 43). Sommers does not acknowledge the problem. In fact she comes out swinging, explaining that if teachers encourage boys to call out twice as often as girls, and then admonish girls who call out, that is no problem at all. That might be Sommers's sense of instructional fairness; it is not mine.
Nor is Sommers disturbed by the gender gap in tests. She offers a partial description of the pool of SAT takers as a way of rationalizing the male advantage in the all-important testing world. In fact girls lag behind boys not only on the SAT but also on the PSAT, the SAT-II, and the ACT. On the Graduate Record Exam the male lead is 129 points (ETS, 1998). Although Sommers cites the fact that more girls are now taking the SAT Advanced Placement Exams, she omits the fact that boys are more likely to get the higher scores needed to gain college credit.
Sommers concludes, "The research commonly cited to support claims of male privilege and male sinfulness is riddled with errors. Almost none of it has been published in peer-reviewed professional journals." Three decades of well over a thousand peer-reviewed studies have documented the disturbing gender gap in schools. By attacking a few researchers, she hopes to taint, even ignore, the work of many.
Beginning in the early 1970s my wife and I consistently wrote about sexism as a "double-edged sword" and called for schools to respond to the very real needs of both boys and girls. Despite Sommers's charges that researchers like me are making a case for "male sinfulness," that phrase is not part of our research lexicon. In her first co-authored book, Sexism in School and Society (1973), Myra included powerful statistics on how bias affects boys, and offered recommendations for creating fairer schools. I wrote one of the first curricula ever designed to help teachers work with boys to counter the limiting impact of the male stereotype (To Be a Man, 1976). Chapter Eight of Failing at Fairness is titled "The Miseducation of Boys." Although it may serve a political purpose to cast our work as "boys against girls," the truth is that Myra and I dedicated our professional lives to battling sexist assumptions that wound both our daughters and our sons. It is bias we oppose, not boys.
Although Christina Hoff Sommers complains that Carol Gilligan's groundbreaking book In a Different Voice devotes only one brief paragraph to the mechanics of the foundational "rights and responsibilities study," this is not the case. Chapter Two lays out the purpose of the study, the basis for selecting the sample, the moral dilemma that was presented to the participants, and the sequence of the interviews. Similarly, Sommers's persistent claim that this and other studies were never published in peer-reviewed journals is simply wrong. The "college student study" appeared in Human Development, the "abortion study" in Developmental Psychology, and the "rights and responsibilities study" in the American Journal of Orthopsychiatry.
Sommers's accusations that Gilligan never used "conventional protocols of social-science research" and that her published research "consists of anecdotes based on a small number of interviews" betray a startling ignorance of research methodology. A quick glance through Child Development or any other peer-reviewed journal will reveal that smaller samples are perfectly acceptable (indeed, the norm) for more-intensive analyses and that qualitative approaches like Gilligan's are common practice. In fact, a detailed account of Gilligan's methodology is documented in several peer-reviewed publications, including Brown and Gilligan's Meeting at the Crossroads, and this is the method of choice for many social scientists.
Jean E. Rhodes
Perhaps the most serious flaw in Christina Hoff Sommers's "The War Against Boys" is its ex post facto perspective, using current conditions to discredit research completed nearly ten years ago. The research conducted by the American Association of University Women Educational Foundation in 1991 and 1992 cited by Sommers focused on shortcomings in the nation's classrooms. Among many other things, it included recommendations to engage girls' interest in -- and access to -- mathematics, science, and school sports.
The problems were real and the solutions appropriate; today girls are pursuing more-challenging math and science classes. Instead of celebrating these achievements, Sommers imagines a battle between the sexes, a zero-sum game in which success for girls comes at the expense of boys. This spurious cause-and-effect reasoning is not supported by empirical evidence or by common sense. Are we to believe that the educational outcomes for African-American males, for example, which have everything to do with urban poverty, are related to the presence of summer science camps for girls in the 1990s or to efforts to broaden girls' access to competitive sports?
Sommers does not acknowledge the findings concerning boys that were included in the 1992 report. How Schools Shortchange Girls reported that boys face higher dropout rates, outnumber girls by "startling percentages" in special-education programs, underperform girls in reading on national tests, and, among socioeconomically low students, are more likely than girls to repeat a grade. In 1998 the AAUW Educational Foundation released an update of the 1992 report. The update, Gender Gaps, featured as its most prominent finding girls' progress, and parity, in mathematics and science course taking, and noted areas of continuing concern, including a gender gap in computer science and technology. We invite readers to return to the 1992 report and to form their own opinions as to its relevance and stature. We are confident that they will find the work less tendentious than Sommers's article would have them believe.
Similarly, studies cited as refutations of the AAUW Educational Foundation's research are, in key respects, inappropriate for comparison. Judith Kleinfeld's review, for example, commits the obvious error of using data from 1996 and other recent years to denounce findings from a report published in 1992 and based largely on 1980s data.
Finally, Sommers criticizes a New York Times reporter for not searching out an opposing viewpoint, yet she herself made no effort to contact the AAUW or the AAUW Educational Foundation to obtain a current response to her outdated arguments. Had she called, she would have learned that months ago the foundation organized a "beyond the (so-called) gender wars" research symposium for the fall. The foundation will convene prominent scholars who have conducted original research on boys' and girls' experiences in and out of school, to share insights and to identify common ground and concerns that emerge from this cumulative body of work. After a decade of research there are many areas where scholars who have looked at all sides of gender and education can come together to improve schooling for all children.
Until I read "The War Against Boys," I had thought maybe I was the only American observing how educators of both sexes are radically feminizing our schools. Sommers marshals an impressive array of supporting evidence, in contrast to the weak evidence cited by those who manufactured the "crisis" of diminished girls. I would add two more citations.
Since the 1970s Deborah Tannen has established that boys and men use everyday conversation to create a stratified pecking order and to establish their place in it, while girls and women use language to replace that pecking order with a community of equals.
And since the 1960s Mary N. Meeker has established that girls' brains are generally stronger in verbal abilities than are boys' brains.
It's beyond me to say whether the male's weaker verbal abilities induce boys and men to prefer conflict to community or, likewise, the female's stronger verbal abilities induce girls and women to prefer community to conflict. But surely Sommers, Tannen, Meeker, and others have established that there is a human blueprint for maleness, and that educators should take advantage of it, not feminize it.
I was disappointed to find Christina Hoff Sommers using the same sort of poorly thought through and inadequately supported arguments that she deplores in Carol Gilligan's work on the treatment of girls in school in order to knock down Gilligan's conclusion that boys would benefit if they were permitted a longer period of attachment to their mothers. I was particularly dismayed by her uncritical acceptance of the assertion, by Daniel Patrick Moynihan and others, that it is the lack of a male role model (and, by implication, the presence of too much mothering) that is responsible for the statistical link between the experience of growing up in a single-parent family and the incidence of violent behavior in boys and young men.
As Sommers takes care to point out and then ignores, the existence of a statistical correlation may suggest there is something to investigate, but it hardly establishes cause. The data certainly do not prove that mothers who are alone cannot raise sons with good values. After all, throughout most of our cultural history the widow's son has been not the villain of the fairy tale but the hero. Nor were the boys raised by women who lost their husbands in World War II notorious for their poor behavior. That distinction was, as I remember, reserved for preachers' sons.
If there is one cause-and-effect that is established as thoroughly as is possible in this dicey field, it is that children who have been abused and neglected are more likely than other children to become violent as they grow older. I believe that Sommers and the experts she cites are wrong. It is not the absence of a male role model that produces violence in the male offspring of the single-parent family. It is that the models they had before the parent became single were bad.
Since I am identified as the source of misinformation about the publication and availability of Professor Gilligan's research, and my e-mail to the Harvard undergraduate Hugh Liebert was published without my knowledge or permission, I am writing to set the record straight and to clarify the meaning of my e-mail.
Contrary to the impression created by Christina Hoff Sommers's quotation of my e-mail to Liebert, in which I said "None of the In a Different Voice studies have been published.... none of what you are interested in is available," the studies have been published and the data are available. Liebert and I had multiple conversations prior to this e-mail correspondence in which we established the word "studies" to mean the verbatim interview transcripts. Since the only specific request that I could discern was for the unpublished data, I assumed that he had already familiarized himself with Professor Gilligan's published work. Indeed, a quick review of the literature would have revealed that the work has been published. For example, the results from the rights and responsibilities study were published in 1988 in "Two Moral Orientations: Gender Differences and Similarities," by C. Gilligan and J. Attanucci, Merrill-Palmer Quarterly, 34 (3), 223-237. The research and design of the abortion study are described in detail in both In a Different Voice and Developmental Psychology, 18 (6), 862-870: "Moral Development and Reconstructive Memory: Recalling a Decision to Terminate an Unplanned Pregnancy." Since Liebert was asking specifically about the raw data, and it is not common practice to publish thousands of pages of raw interview transcripts, I told him that the verbatim transcripts were not published. When the word "studies" in my e-mail to Liebert is read to mean verbatim interview transcripts, my statement is accurate. When Sommers portrays it to mean protocol, method, research design, or results, I am egregiously misunderstood.
The interview data from these studies and from the college-student study have been available to graduate students and scholars over the years. Before I bring specific requests to Professor Gilligan, I gather basic information regarding the nature, scope, and time frame of the proposed research. Because I was unable to get a clear response from Liebert (he never said that he was conducting research for a book, nor did he say he worked with or for Sommers), other than a vague request regarding the procurement of the raw data for an unspecified project, I did not bring this request to Professor Gilligan's attention. Had The Atlantic or Sommers called me, I would have clarified that the e-mail was part of a larger conversation regarding Liebert's request.
Christina Hoff Sommers replies:
David Sadker says he has published "a score of [research] articles in peer-reviewed journals" on classroom bias. But the one source he cites is a report called Promoting Effectiveness in Classroom Instruction: Year 3 Final Report. Such reports are not peer-reviewed and not published -- just filed with the government as an earnest of work performed for an awarded grant. In any case, that government report does not contain the famous missing study about call-outs that I discussed in my article. In their book Failing at Fairness the Sadkers wrote, "Our research shows that boys call out eight times more often than girls.... However, when girls call out, ... the teacher remembers the rule about raising your hand.... And then the girl ... is deftly and swiftly put in her place."
The eight-to-one "call-out gap" became a favorite with journalists and politicians who saw in it a persuasive manifestation of a pervasive gender bias in our nation's schools. The New York Times alone has credited it on three separate occasions. But what evidence is there that it exists?
The Sadkers gave different answers over time, but they never produced the original research. When a reporter from U.S. News & World Report inquired about the research, David Sadker told her that he had presented the findings in an unpublished paper at a conference, and that he no longer had a copy. Now, in his Atlantic letter, he says it was in an unspecified "pilot study" that was superseded by another (also unspecified) "full study." Meanwhile, in a revised version of Failing at Fairness the eightfold claim has been reduced to "boys call out significantly more than girls."
The American Association of University Women is the group that successfully propagated the myth of the shortchanged girl. To question this myth is to risk denunciation. The Spring, 1997, newsletter AAUW Outlook, for example, explicitly compared me and other "gender bias revisionists" to those who deny the Holocaust. The response to the Atlantic article is less acrimonious but scarcely more reasonable.
Sharon Schuster and Sandy Bernard now say that boys' weaknesses were fully acknowledged in the AAUW research. They note, for example, that boys' reading deficits were discussed in How Schools Shortchange Girls. But when boys' deficiencies are mentioned, they are immediately discounted.
In all age groups, girls have consistently received higher test scores in reading and writing since the 1970s. Since 1971, however, boys have made gains relative to girls, particularly in the seventeen-year-old-group.... It has been suggested that even if the small gender difference favoring girls [in reading] is statistically significant, it may not be educationally significant. [Emphasis in original.]
It is in fact not true that boys have made gains relative to girls since 1971. According to the Department of Education, what was true in 1971 is still true today: the average seventeen-year-old boy reads at the level of a typical fifteen-and-a-half-year-old girl. As for writing, the department reports, "male 11th graders scor[e] at about the same level as female 8th graders in 1996."
The AAUW may not regard the reading-and-writing gap as "educationally significant," but it adversely affects the futures of millions of low-achieving boys.
Now, faced with charges of systematic unfairness to boys, the AAUW talks ecumenically about "improv[ing] schooling for all children." Unfortunately, the organization's attitude toward boys has not changed -- and perhaps cannot change.
Indeed, a very recent fundraising letter from the AAUW opens with anonymous quotations from distressed girls: "I kept my hand up for a full five minutes ... and he never called on me!" "Boys will grab whatever they want to." The AAUW is still broadcasting its divisive boy-blighting message: "We can't continue to shortchange more than one half of our population -- one half of our future."
Jean Rhodes, the director of gender studies at Harvard's Graduate School of Education, would assure us that Carol Gilligan actually published the three studies. She says, for example, that the "rights and responsibilities study" appeared in the American Journal of Orthopsychiatry. But Rhodes is mistaken. In a Different Voice describes the rights and responsibilities study as a study of 144 male and female adult subjects. The article in the Journal of Orthopsychiatry discusses comments made by four children; it contains no data and very little, if anything, in the way of research.
Tatiana Bertsch, on the other hand, asserts that the "rights and responsibilities study" was published as a 1988 article in the Merrill-Palmer Quarterly. However, that little article, which Gilligan herself does not cite in her list, reports some highly equivocal findings made six years after the publication of In a Different Voice (1982); it cannot possibly be the research to which Gilligan had referred in the book. Note, too, that Bertsch says that "results" of the rights and responsibilities study were published, not the actual study. That study, as Bertsch told Liebert, has not been published.
Bertsch now asks us to read "study" to mean "raw data." Two of my assistants spoke with Bertsch, and whatever her impression may be, neither ever asked her for "raw data" or for "verbatim interview transcripts." We wanted to know only one thing: are the three studies for In a Different Voice in the public domain? In her e-mail to Liebert, Bertsch was admirably clear and concise, informing him that the studies had not yet been published and were unavailable.
I have one small cavil about Douglas L. Wilson's otherwise excellent piece ("Keeping Lincoln's Secrets," May Atlantic) on William H. Herndon. Wilson writes that Herndon "was conducting the first oral history of a great American hero." This does an injustice to James Parton.
Setting to work a dozen years after the death of Andrew Jackson, Parton read everything written on Jackson and could not make up his mind about him. In his perplexity Parton resorted to oral history. "In Washington I conversed with politicians of the last generation, who have no longer an interest in concealing the truth," he wrote. He visited a third of the states to hear the recollections of men and women who had known Jackson and had supported or opposed him. "Thus it was that contradictions were reconciled, that mysteries were revealed, and that the truth was made apparent." Parton's memorable three-volume biography of Jackson came out in 1859-1960, while Herndon was still collecting material on Lincoln.
Arthur Schlesinger Jr.
The photograph on page 86 of the May, 2000, issue shows William Herndon's crepe-draped law office in the spring of 1865 -- not in 1860, as stated in the caption. The motto above the windows ("He Lives in the Hearts of His People") is not a campaign slogan; it is a lament for the assassinated President, a former partner in the firm of Lincoln & Herndon.
We were bemused (though also wryly gratified) to see in your May issue an article by Sheldon M. Stern ("What JFK Really Said") alleging seventeen transcription errors in our 1997 book, The Kennedy Tapes: Inside the White House During the Cuban Missile Crisis, which published nearly 200,000 words of transcriptions.
Bemused for two reasons. First, the preface of our book explained that most of the taped conversations are very hard to understand. (Some people who were present at the meetings have listened to tapes and confessed that they could scarcely make out a single word.) "The reader has here the best text we can produce," we wrote, "but it is certainly not perfect. We hope that some, perhaps many, will go to the original tapes. If they find an error or make out something we could not, we will enter the corrections and enhancements in subsequent printings of this volume." As Stern acknowledges, we did just that in printings after the first. We personally invited Stern to suggest corrections when the three of us appeared together on a panel at the Kennedy Library in 1997. Yet we never heard a word from him.
Second, as has been widely publicized for more than a year, we have gone back over all the tapes and are still going over them with new colleagues and far better technology. A three-volume reference set with transcriptions of all Kennedy tapes for three months of 1962 is being published by W. W. Norton early next year, the product of the Presidential Recordings Project at the University of Virginia's Miller Center of Public Affairs. These volumes will include the retranscribed missile-crisis records and much, much more (the Mississippi civil-rights crisis, Berlin, tax cuts, and so on), including tapes of telephone conversations which were discovered and made public property only in 1998. Stern mentions none of this. We had also expected (and announced) that the Harvard University Press would issue an entirely revised Kennedy Tapes paperback this coming fall, but Harvard's editors disappointed us by deciding that the amendments and additions were not significant enough to justify the cost of scrapping and replacing the existing stock.
Our wry gratification also has two sources. First, with only two exceptions (Kennedy's saying "final failure" in one sentence instead of "prime failure," and one of his aides' referring to former President Eisenhower offering his opinion to doubting congressmen as a "soldier" instead of a "facilitator"), we had already made the corrections that Stern suggests, either in subsequent printings or for the forthcoming volumes, and we are not finished with them yet. In a few cases our corrections surpass, or correct, his. Indeed, we have made hundreds of other small corrections, have been able to make out a good deal of what was said in passages we had labeled "unclear," and have identified speakers with much greater precision.
Second, none of these amendments are very important. None of them change what a reader of the transcripts takes away concerning the essence or even the minute details of the deliberations that took place in the Oval Office and the Cabinet Room in those terrifying thirteen days of October, 1962.
Finally, we offer a historical perspective. Partial transcriptions for a few missile-crisis meetings were first published in 1983, while Stern was working at the Kennedy Library. Then the National Archives abandoned such efforts as being too costly and interpretive. For fourteen years historians relied on those few initial transcripts. They were right to rely on them, because the people who did that work were conscientious and did the best they could. Though our book significantly improved on those transcripts and transcribed the rest of the meetings, we did not criticize the earlier efforts. We built on them and tried to make them better. We published our book in 1997 because we were eager to offer a fuller portrait of the crisis, and of how Kennedy performed. These revelations and transcriptions have already had a major impact on scholarship, offering a more positive picture of President Kennedy, clarifying what he thought the crisis was about, and illuminating the debates.
Transcriptions are subjective interpretations of the true historical record, the tapes themselves. We're sure that people will keep arguing with this or that rendition of a passage. But the tapes are effectively inaccessible without transcriptions. To bring these treasures to the public, someone must go out on a limb and craft hundreds -- thousands -- of pages of transcriptions. So we keep working.
Ernest R. May
Philip D. Zelikow
Sheldon Stern replies:
Ernest May and Philip Zelikow claim to be "bemused" and "wryly gratified" that I found only seventeen errors in 200,000 words of transcriptions. Plainly, the length of my article permitted only a tiny sampling of errors. The October 18 transcript alone was so flawed that I stopped penciling in corrections and wrote my own transcript. After finding dozens of errors in just the first twenty pages, I spot-checked the other transcripts, with similar results.
May and Zelikow claim that "in a few cases our corrections surpass, or correct" mine. I made clear in the article that revision is a continuing process. I have listened to the tapes again in the six cases in which they claim to have corrected my transcriptions: in four instances I stand by my version; in one case I am less sure; only once, involving one word, are they clearly right.
Perhaps the real purpose of their letter is to divert attention from these errors (suggested by the fact that corrections made in the fourth printing were slipped in without scholarly documentation). My goal as a historian and an educator is not to collaborate with this smokescreen but to inform readers, particularly the history teachers I have urged to use the transcripts in their classrooms.
The assertion, by reputable scholars, that "none of these amendments is very important" is shocking. When the words are wrong, as they are repeatedly, the historical record is wrong. The claim that the transcripts were as good as they could have been, given the technical quality of the tapes, is an affront to thousands of scholars, general readers, schools, and libraries that bought The Kennedy Tapes assuming they would get the best from the brightest.