By Sheldon M. Stern and Max Holland
Though it has been slow to develop and achieve recognition, it is now becoming apparent that scholarly works based on the extraordinary cache of presidential recordings from the Kennedy, Johnson, and Nixon administrations actually constitute a new and distinct genre of historical investigation.
The history profession is familiar with books that exploit new primary sources, or interpret old primary sources in a fresh way, along with works that are syntheses of primary and secondary sources. There is also an honored place in the canon for books that annotate the private papers of such prominent figures as Woodrow Wilson. Books based on audio recordings, however, are arguably distinct from these traditional categories. The main reason is that the historian shoulders an even larger burden in this new genre. He or she is obviously selecting, deciphering, and making judgments about a primary source, much like the editor of a documentary collection. But, in the process of transcribing a tape recording, the historian is also creating a facsimile—while still endeavoring to produce a reliable, “original” source. In essence, the historian/editor unavoidably becomes the author of a “new” source because even a transcript alleged to be “verbatim” is irreducibly subjective at some level. As a result, the historian’s responsibility in this genre is a very unusual one, and requires the most careful scholarship imaginable. No other task of discovery and/or interpretation in the historical canon is quite comparable.
As the audio recordings from the Kennedy, Johnson, and Nixon presidencies have become available, historians have eagerly taken up this unprecedented challenge—and understandably so. The attraction of being able to convey even a fraction of what actually transpired in the Oval Office and Cabinet Room from 1962 to 1969, and 1971 to 1973, is irresistible, the ultimate fantasy for many historians. The recordings offer the tantalizing prospect of history “wie es eigentlich gewesen ist”—as it really was—in the famous formulation of the 19th-century German historian Leopold von Ranke. Or as Columbia University professor Alan Brinkley put it in 1997, “No collection of manuscripts, no after-the-fact oral history, no contemporary account by a journalist will ever have the immediacy or the revelatory power of these conversations.” Almost a dozen “tapes-based” books have been published since 1997, more are in the offing, and this does not count the growing number of conventional histories and/or biographies and articles in newspapers, magazines, and scholarly journals that selectively but increasingly rely on the presidential recordings for substantive insights, anecdotal asides, or simply colorful quotes.
One issue must be acknowledged before leveling any criticism: transcribing these tapes is far more difficult than it appears on the surface for subjective and objective reasons. There are knotty stylistic issues, for example, that have substantive consequences. When is a hesitation or “uh-huh” significant? Something as seemingly minor as eliminating a pause, or “ironing out a cadence,” as one historian put it, can shift emphasis and even change meaning. Not being able to fully render tone or intonation runs the same risk. If LBJ lapses into his most Southern dialect, and that is reflected in the transcript, does he risk being portrayed as some character out of a Mark Twain novel?
Alternately, does it misrepresent LBJ to render him speaking the King’s English when he demonstrably does not? Reading even the most faithful transcript will never substitute for actually listening to the recordings themselves; as one historian put it, “transcripts are not interchangeable with the original tapes.” A transcript, or a narrative that attempts to capture both the verbal and affectual dimensions of the tapes, can further refine our understanding. But only the tapes can be legitimately cited as the genuine primary source.
These questions aside, the most daunting issue is the frequently poor audio quality of the tape recordings. It can be exasperating to try to decipher something as fundamental as who is speaking, particularly on the tapes from the Kennedy and Nixon administrations, which include many recordings of meetings. Even the most painstaking effort to transcribe the recordings is bound to result in some errors, the present authors’ own attempts included. Accordingly, all three presidential libraries have desisted from producing official transcripts—although without transcriptions the recordings are not user-friendly, despite the libraries’ best efforts to index and catalog them. The libraries have decided (wisely, in our view) to regard the recordings as the original document and everything after that as a facsimile or interpretation, almost a translation, so to speak, and one that must be vouched for by the scholar(s) who produced it.
Another factor, of course, is the enormous commitment of time and resources it would take for the presidential libraries to produce high-quality transcripts. When the Kennedy Library estimated, nearly 20 years ago, that it would take about one hundred hours of listening to produce an accurate transcript of a one-hour recording, it was widely suspected of manufacturing an excuse to avoid the work. But time and experience have proven that ratio to be right on target. Producing books based on the tape recordings is (or ought to be) an extremely labor-intensive endeavor. It is microhistory, and presenting it accurately demands that the scholar be steeped in the subject matter. Often he or she must know the events of a given day, and sometimes a given hour. There is, in other words, a direct correlation between the time one invests in listening to the tapes, and in researching their context, and the sense one is able to make of them. Regardless of the difficulty in rendering accurate transcripts, the onus remains on those scholars of the Kennedy, Johnson, and Nixon presidencies who willingly assume the burden and claim in print to have carried it off.
The remainder of this essay will examine several of the most acclaimed works in this evolving historical genre, namely, the volumes based on the Kennedy recordings by Ernest May, Philip Zelikow, and now the Miller Center of Public Affairs at the University of Virginia, and the books on the Johnson recordings produced by Michael Beschloss.
Stanley Kutler’s 1997 book, Abuse of Power, was published in the same year as Ernest May and Philip Zelikow’s first effort, The Kennedy Tapes, and Michael Beschloss’sTaking Charge, and on that basis alone Kutler’s work would seem to merit inclusion.Kutler’s book, moreover, has been criticized along some of the lines that will be enumerated below. In 1998, John Taylor, director of the Richard M. Nixon Library, charged that Kutler had intentionally truncated some transcripts to create the misleading impression that Nixon had foreknowledge of the break-in at the office of Daniel Ellsberg’s psychiatrist. And historian Joan Hoff, who also culled the recordings carefully for her book on Richard Nixon, has seconded Taylor’s criticism.Kutler recently declared, “Did I make mistakes? I’m sure I did. But I never knowingly changed an affirmative to a negative or vice-versa. I never added or subtracted words to alter the meaning. But the tapes are difficult . . . and the human ear is not perfect. Besides, I would rather be remembered as the guy who made sure that everyone could get their hands on the Nixon Tapes [by suing to secure their release].” In any case, because the authors of this article are not experts on the Nixon recordings and presidency, we are not in a position to write about Kutler’s transcriptions, editing, and annotations.
The Kennedy Tapes
In 1981, the John F. Kennedy Library (JFKL) began processing the recordings of the ExComm meetings during the 1962 Cuban missile crisis, and in 1983 started releasing them to the public. It took years to review and declassify all 22+ hours of recordings largely because the audio quality was frequently very poor. Consequently, more than 17 hours (77 percent) of the ExComm recordings were not released until 24 October 1996 (15+ hours) and 17 February 1997 (2 hours). Some of these last releases were among the hardest to transcribe.
Notwithstanding the JFKL’s difficulties and its imposing estimates of the time required to do this work carefully, in October 1997, Harvard University Press (HUP) published Ernest May and Philip Zelikow’s The Kennedy Tapes: Inside the White House During the Cuban Missile Crisis. Given that the physical process of copy-editing and manufacturing a book normally takes seven or eight months, this would, to say the least, have been a notable feat even if the recordings had been crystal clear. Nonetheless, HUP declared that “These are the full and authenticated transcripts of those audio recordings.” The stunning achievement was explained by stressing the editors’ “monumental efforts over the past months” to transform “these crackling, rumbling, and hissing tapes . . . into readable transcriptions.” The editors themselves supplied specific details:
we commissioned a team of court reporters . . . [to prepare] draft transcripts from the recordings released by the JFK Library. We then asked an expert in audio forensics to improve the sound quality of most of the tapes. . . . The two of us then worked with the tapes and the court reporters’ drafts to produce the transcripts. . . . The laboriousness of this process would be hard to exaggerate. Each of us listened over and over to every sentence in the recordings. Even after a dozen replays at varying speeds, significant passages remained only partly comprehensible. . . . Notwithstanding the high professionalism of the court reporters, we had to amend and rewrite almost all their texts. For several especially difficult sessions, we prepared transcriptions ourselves from scratch. In a final stage, we asked some veterans of the Kennedy administration to review the tapes and our transcripts in order to clear up as many as possible of the remaining puzzles. The reader has here the best text that we can produce, but it is certainly not perfect.
Reviewers of The Kennedy Tapes took the assurances of these recognized Harvard scholars at face value. The “effort was herculean,” wrote the critic for The New York Times Book Review.” The Economist reviewer declared that the editors had produced “the most accurate, lucid transcript that is at present possible ” [emphasis added].The Wall Street Journal review enthused that the “verbatim” transcripts are not fictionalized or reconstructed dialogue: “this is the real thing.” A reviewer in the St. Louis Post-Dispatch concluded, “With the aid of court reporters and some alert ears, remarkable texts emerged. . . .” “Painstakingly recovered from poor recordings,” wrote Henry Graff in the New Leader. Alexander George summed up the scholarly consensus when he wrote, “Professors May and Zelikow’s meticulous work in transcribing the imperfect recording of President Kennedy’s secret tapes is a remarkable achievement.” Another scholar called the transcripts the “jewel in the documentary crown” of material that had become available about the missile crisis.
Apparently, not a single reviewer of The Kennedy Tapes thought it necessary to listen to any of these tapes to confirm that the transcriptions were, in fact, reliable, and/or the best that could reasonably be expected. This was not surprising with regard to mass-circulation reviews, of course, but it extended to reviews in scholarly publications as well, despite the often nonsensical, fractured syntax of the transcribed conversations. The authority and status of the editors, the reputation of their publisher, and the massive scope and detail of the work seemed proof enough.
In 1998, Zelikow left Harvard’s Kennedy School to become director of the Miller Center of Public Affairs at the University of Virginia. Zelikow soon announced that he intended to dedicate a substantial portion of the Miller Center’s resources to the presidential recordings from the Kennedy, Johnson, and Nixon administrations. “These audiotapes will do for the study of government what the discovery of Pompeii did for the study of Rome,” Zelikow declared in announcing the Miller Center’s new project. “The books and studies that emerge from this project will help replace the Hollywood image of White House decision-making with a real world understanding of how government actually works.” In every respect, including their ambitious publication schedule, the model for this impressive enterprise was, of course, the universally-praised volume that Zelikow and May had edited with conspicuous speed.
In fact, utilizing court reporters unfamiliar with the history and the cast of characters had resulted in transcripts riddled with errors, many of which went uncorrected in the final editing. Speakers were repeatedly misidentified or not identified at all, and scores of passages that had been labeled “[unclear]” were, in fact, audible and discernable. These mistranscriptions, large and small, help explain why the transcripts often seemed nonsensical or studded with remarks that were nonsequiturs. This was not a matter of inevitable and incidental mistakes sprinkled throughout the transcripts, but pervasive errors that significantly distorted the reliability of a unique source and undermined the very purpose behind publishing these transcripts.
The planned chronological series of Miller Center volumes on the Kennedy recordings provided an opportunity to correct these problems but, unexpectedly, one of the authors of this article, Sheldon Stern, first publicly exposed the inadequacies of The Kennedy Tapes in The Atlantic Monthly. In their response to the Atlantic article, May and Zelikow wrote, “We were bemused (though also wryly gratified)” to read about alleged errors in the transcriptions. “None of these amendments is very important. None of it changes 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 Cabinet Room.” Besides being plainly incorrect, this was an odd and contradictory argument, given their earlier claim that The Kennedy Tapes had forced a drastic revision of prior accounts precisely because of the new and telling detail it contained. Indeed, in presidential recordings, one might easily claim (as May and Zelikow sometimes explicitly did) that nuance is everything. If it is not important, the tapes lose much of their historical power.
Before Stern’s article, the Miller Center could have published substantially revised transcripts while simultaneously asserting that new technological wizardry was primarily responsible for the dramatically different transcripts. But Stern’s corrections were based on what can be described as technically pedestrian methods and the aural equivalent of elbow grease. Consequently, Stern’s articles (he published a second, scholarly version in Presidential Studies Quarterly in September 2000) had the effect of causing the Miller Center to circle the wagons rather than embrace open and constructive criticism. The response was to dispute as many errors as possible, to publicly deny the significance of errors that were undeniable, and in general, to pretend that nothing untoward had happened—and besides, these problems would all be corrected in what the editors said would become the deserved “focal point for scholarly attention,” i.e., the “authoritative reference works” to be published by W.W. Norton. Indeed, the Miller Center would soon claim that 120 minutes of listening were necessary to render a single minute of transcript in these new volumes, a ratio that exceeded even the Kennedy Library's initial estimate.
When the Miller Center’s reference volumes were published by Norton in October 2001, academic reviewers again found it difficult to critique the new edition. The claim that a team of expert scholars, using state-of-the-art technology, were responsible for these revised transcripts (about 35% of the 1,800 pages in these three volumes covered the missile crisis) continued to insulate the work from peer criticism. Historians were predictably reluctant to listen to the tapes themselves and, given the cutting-edge methods allegedly required for transcribing, assumed for a second time that the transcripts must be the best that could be done.
A good part of the new missile crisis transcripts were, in fact, significantly more accurate—indeed, sometimes unrecognizable—when compared to the HUP edition. As such, it was certainly a step in the right direction. Nonetheless, upon closer review, Stern discovered that the ExComm retranscriptions still contained many significant errors, some of which modified or changed the intent and meaning of speakers’ remarks. And while the number of “[unclear]s” was drastically diminished, there were still many passages so marked that were, in fact, discernable to other listeners. In addition, words, phrases, sentences, or speakers were sometimes still missing altogether in these new transcripts. Some errors were particularly perplexing because they made no sense in the historical context of the conversations, and for that reason alone should have been flagged during the elaborate Miller Center editing process overseen by Tim Naftali, the director of the Presidential Recordings Program.
Another problem common to the HUP and Norton versions pertained to what May and Zelikow once called “verbal debris.” “What we omit are the noncommunicative fragments that we believe those present would have filtered out for themselves,” the editors wrote in the preface to the HUP edition. “We believe that this gives the reader a truer sense of the actual dialogue as the participants themselves understood it.” But this kind of editing is not as neutral as it sounds. And while it may be appropriate in a commercial book intended for a wide audience, it becomes very problematic in what purport to be scholarly reference works meant to be consulted for decades. As Joan Hoff pointed out in her 2003 AHA commentary on the presidential recordings,
How does [excising ‘debris’] serve or preserve history when it is actually distorting history? . . . Presidential tapes constitute the raw and unpleasant way people communicate or, as often happens in conversations with presidents, do not communicate very well. By eliminating ‘verbal debris’ or ‘noncommunicative fragments’ the participants sound smarter and more rational than they usually are in reaching decisions. Future generations of historians need such untidy, unsanitized material, not neat transcripts which happen to fit mainstream historical interpretations at a certain point in time.
Although sustained criticism of an admittedly difficult task might seem niggling, it must be remembered that the Miller Center set the bar high for itself. Shortly after deciding to undertake the formidable task of making available the recordings from the Kennedy, Johnson, and Nixon presidencies, the Miller Center sought funding from the National Historical Publications and Records Commission (NHPRC), the grant-making division of the National Archives. The NHPRC, established in 1934, underwrites projects that preserve and disseminate archival-quality historical records. Its subsidies are the gold standard; it has underwritten the publication of superbly edited and annotated volumes of the Founding Fathers and other leading political figures. And from 2000 to 2004, the Miller Center received $505,000 in NHPRC funds to support a program that is supposed to reflect only the highest excellence and scholarship. When the three Norton volumes, described by the Miller Center as “perhaps the most reliable record of the Kennedy presidency ever published” were released in 2001, the Center acknowledged that their work “was made possible in part by a generous grant from the National Historical Publications and Records Commission (NHPRC).
The Johnson Tapes
In response to the President John F. Kennedy Assassination Records Collection Act of 1992, the Lyndon B. Johnson Library (LBJL) began releasing recordings in April 1994 related to President Kennedy’s assassination and its aftermath. Soon Harry Middleton, then the LBJL director, prevailed upon Lady Bird Johnson to have all of Lyndon Johnson’s tape recordings processed and released in chronological order. The result has been a justifiable upsurge of interest in this controversial president, and a new round of scholarship as historians have sought to integrate information from the tapes with the standard wisdom.
In comparison to the Kennedy and Nixon tapes, the Johnson recordings are relatively easy to transcribe, despite Beschloss’s assertion that “none of the tapes are easy to decipher.” About ninety percent of the time there is no question as to whose voice was being recorded, since the vast majority of the recordings are of bilateral telephone conversations. There is a problem with overlapping voices, and to be sure, not all of the recordings’ audio quality is good, largely owing to the technological state-of-the-art at the time. Many recordings are only fair, and several are poor. Some key Johnson aides and advisers, such as Walter Jenkins and Abe Fortas, had an unfortunate inclination to mumble and speak in barely audible voices. Still, compared to the tapes made by LBJ’s predecessor and successor, it is a relatively easier task to produce reasonably accurate transcripts of the Johnson recordings, though as noted before, any rendering is at best a facsimile of the bona fide source and never a substitute for it.
In 1997, Michael Beschloss brought out the first volume of what he eventually announced would be a trilogy on the Johnson recordings. Beschloss’s first LBJ book appears to have been completed as rapidly as the May and Zelikow HUP volume of the same year. The Johnson tape recordings for April through June 1964 were not released until 14 February 1997, and the July and August 1964 tapes were not released until 18 July 1997. That means, in short, that 54 percent of the tapes used by Beschloss in Taking Charge were not released until the year of publication—23 percent in February and 31 percent in July (the latter just three months before publication of the book on 17 October 1997). In Beschloss’s second volume, Reaching for Glory, 40 percent of the tapes were not released until 2001 (28 percent on January 12 and 12 percent on June 8). The book was published on 1 November 2001.
Beschloss’s project differed from the initial May and Zelikow volume, and Kutler’s book as well, in that he did not restrict his selection to a particular subject. Rather, Beschloss presented a cross-section of select conversations from a defined chronological period, November 1963 to August 1964 in his first volume, Taking Charge, and September 1964 to August 1965 in his second volume, Reaching for Glory. Although the publisher of both volumes, Simon & Schuster, is a commercial trade press, Beschloss cast his project in scholarly terms:
I have conceived this trilogy in the style of an edited and annotated anthology of private letters written by a public figure in the days when leaders did business on paper, revealing their private purposes, methods and obsessions. . . . My chief standard in deciding which conversations to include in the book is whether they add something of historical importance . . . .
The editor of a volume of new primary source material, like this one, has a different responsibility [than a historian writing a book of his own]—not to drown out the subject’s voice. . . . His task instead is preeminently to explain what the new material means and what it tells us beyond what we know already.
Beschloss also explained, in language reminiscent of May and Zelikow, that his transcripts were the product of a scholar’s meticulousness and attention to detail. To a degree, he had the benefit of earlier transcriptions rendered by Johnson’s secretarial pool. But these, Beschloss correctly noted, were “fragmentary, inaccurate, and unreliable for the historian.” He described his own methodology and editorial conventions this way:
In creating this book, I have listened to virtually every Johnson White House tape . . . often many times—and have personally transcribed most of the conversations that appear here. The main reason for this is accuracy. . . .
[The] only way to make these tapes a reliable source is for a historian to be steeped in the daily history of LBJ’s presidency, armed with names, issues, and context, and to listen hard to every syllable—sometimes over and over again . . . .
I have edited each conversation to exclude extraneous material and repetition, but not where that might change the meaning. Ellipses appear where shorter parts of conversations have been pared; a larger break is used for longer deletions. The only words to be eliminated [emphasis added] without some kind of indication are “uhs,” “wells,” and similar interjections, but only in cases where they do not add meaning.
These statements resulted in the same presumption accorded the May and Zelikow volumes. Beschloss’s transcriptions of the LBJ recordings were greeted with uncritical enthusiasm, although again, no one apparently bothered to listen. Taking Charge was reviewed in both the Sunday and the daily New York Times. Indeed, the newspaper thought so highly of the book that it ran an editorial, declaring the “publication of Lyndon Johnson’s secretly recorded tapes is an important event [emphasis added].” Reviews in scholarly publications also took for granted the accuracy and fidelity of the “brilliantly edited” or “superbly edited” transcripts. Beschloss “had to decipher impenetrable accents and ignore the background blare of chattering aides, TV commercials, and shuffling papers,” wrote one reviewer.
A closer, more learned look, however, would have uncovered numerous shortcomings in the transcripts, and frequent misrepresentations of the conversations in Beschloss’s annotations and footnotes. In addition, explanations were frequently lacking when they were sorely needed, which can be just as misleading as the wrong contextual information. In Taking Charge, both the prologue (where Beschloss actually melds and expurgates five separate LBJ conversations about William Manchester’s book on the JFK assassination into one), and the appendix are conspicuous examples in which the context is minimal or not provided at all. With respect to the Manchester-related conversations, for example, Johnson’s remarks about The Death of a President appear to be petulant—which they most certainly were not.
The sole published criticism of the transcripts from Beschloss’s first volume was a 1999critique in the Los Angeles Times by one of the authors of this article, Max Holland. It pointed out what was easily the most egregious mistake in the transcripts (Beschloss’s rendering the phrase “thread of it” as the word “threat”), which resulted in a significant interpretive error about the work of the Warren Commission. Putting aside mistakes in transcription, misinterpretations in the annotations because of these mistakes, and other errors in Beschloss’s microhistory, the nub of the problem with his volumes is very similar to the one that has cast a shadow over the May/Zelikow and now the Miller Center effort. The critical element, indeed the raison d’etre of the project—reliable, accurate transcripts—is lacking.
In Beschloss’s case, however, the problem is exacerbated by the editorial conventions he adopted to produce his “trilogy of snippets”—as Zelikow once described Beschloss’s project in a brief review. Despite Beschloss’s pledge not to eliminate anything meaningful without leaving an indication, his marked and unmarked deletions are so liberal as to seem indiscriminate. Readers simply have no way of knowing (unless they listen to the conversations themselves) whether a word, phrase, sentence, paragraph, or much longer portion of the conversation has been omitted. If a reader were to prepare a full transcript of any redacted conversation that appears in the Beschloss volumes, he or she would immediately discern major differences in wording, sequence, organization, completeness, interpretation, and impact. Put another way, even when Beschloss gets right all the words he chooses to put down, his conventions frequently produce transcripts that beggar any definition of the word. These transcripts simply cannot be cited or quoted with reasonable confidence by scholars—any more than scholars would rely on the selected and redacted documents in a college reader.
If the presidential recordings, as May and Zelikow once wrote, are “historical treasures in a class with the papers of the Founding Fathers,” they ought to be treated as such by historians producing books that they claim are scholarly and authoritative. These recordings are public records held in trust by the National Archives for the American people, not a private trove to be cornered and exploited without public accountability.
“We obviously are devoted,” May and Zelikow have written, “to producing the best possible transcripts.” Scholarly critiques, they claimed, “spur on such work, letting us know that peers are reviewing it with care and offering constructive criticism.” We agree completely. And that is why the Miller Center, which is an academic institution receiving NHPRC funds, has a special responsibility to see that errors are corrected openly and for the record—as historians have been assured they will be. In February 2003, Philip Zelikow spoke at an major conference on presidential recordings at the Kennedy Library. He announced that the Miller Center was establishing a new website that would allow scholars to submit corrections to the Center’s printed transcripts, with full attribution to the scholars making those corrections. Scholars, he explained, “should invite further comments and criticism and…try to welcome them. … I want to announce today that as of this morning, we have put out on the Internet a new website. . . to enable scholars to download our publications of corrected transcripts . . . [thus] providing a multi-media errata sheet. . . . That’s the way that scholars work.”
Two years later, the website exists (www.whitehousetapes.org) but there are still no corrections and no process for submitting them. Indeed, the Miller Center website currently describes whitehousetapes.org as the Presidential Recordings Program’s “signature website” and as “a clearinghouse for research on the tapes.” However, it does not mention Zelikow’s 2003 public commitment to establish an online errata system—which he reiterated in March 2004. Scholars working independently on presidential recordings deserve open and unambiguous information from the Miller Center about ongoing corrections and revisions.
It is especially incumbent upon Zelikow, Naftali, May, and the Miller Center to clarify the record because four distinct versions of “authoritative” missile crisis transcripts now exist (two published by HUP: hardcover in 1997 and paperback in 1998 and two published by Norton: hardcover in 2001 and paperback in 2002)—in addition to amendments and corrections that were incorporated in various printings without any notation or explanation. Scholars working with the transcripts have to sort out this muddle in order to decide which “authoritative” version to use. The 2002 Norton paperback, for example, uses the identical title and identifies the same two editors, May and Zelikow, as the HUP 1997 hardcover and 1998 paperback. In fact, it is actually a concise version of the substantially different 2001 Norton hardcover edition in which Naftali was an editor as well.
The need for an open and public process for making corrections at the Miller Center is all the more acute because of what business historians call “barriers to market entry.” As Joan Hoff has pointed out, the first books derived from the tapes tend to be regarded as “authoritative” or “the Bible” regardless of serious transcription or editing mistakes. “Their very publication discourages others from undertaking their own comprehensive, literal transcriptions of the presidential tapes,” Hoff warned.
For his part Beschloss, whose work was not supported by taxpayers, has less of a public obligation. He has no more or less a duty than any historian has to correct errors in his work. But if he is going to continue his trilogy, some adjustments are clearly in order, perhaps a change in his editorial conventions so that readers can clearly understand the extent to which the conversations are redacted. In fact, simply adhering to his declared editorial conventions would be a marked improvement.
[Postscript: In February 2006, the Miller Center finally erected a feature that permitted scholars to submit suggested corrections to its published transcipts.]
Sheldon M. Stern served as historian at the John F. Kennedy Library in Boston from 1977 through 1999 and worked extensively with the Kennedy tape recordings as they were being processed for release from 1983 to 1997. Holland was a research fellow at the Miller Center of Public Affairs from 1998 to 2003. A dispute between the Miller Center and Holland ended in a mutually-agreeable settlement.
 It has been suggested repeatedly that because the Nixon recordings were voice-activated, while Kennedy and Johnson had to select which conversations were taped, the former are more likely than the latter to be candid records of a president in action. Such an assertion does not withstand close scrutiny. Even presidents have no idea beforehand what direction a conversation or meeting will take, and hours of listening to the Kennedy or Johnson recordings will persuade even the most hardened cynic that the tapes are revealing of both men, warts and all. They were not posturing for posterity. In addition, and perhaps more to the point, none of these presidents made these tape recordings with the faintest thought that they would ever be released to the public within their lifetimes, indeed, if ever at all. But for the Watergate scandal, they might well have remained as secret as they were fully intended to be.
 Alan Brinkley, “D.C. Confidential,” The New York Times (hereafter NYT) Book Review, 19 October 1997. Not all historians, to be sure, are so enamored of the tapes. Nelson Lichtenstein, a labor historian at the University of California at Santa Barbara, has warned that “a relentless focus on the elite decision-making process tends to fetishize and decontextualize it, stripping away its relationship to larger cultural, ideological, and social currents.” Paul Mitchinson, "All the Presidents' Tapes, Lingua Franca, February 2002, p. 58. Another scholar also warned against a wholesale embrace of “unmediated” history, maintaining that historians would find transcripts less valuable than their authors’ claimed. The “new material merely buttresses prevailing interpretations. . . . In no way does it alter historians’ fundamental understanding of the Kennedy, Johnson, and Nixon presidencies.” Bruce J. Schulman, “Taping History,” The Journal of American History (hereafter JAH), September 1998, p. 576; other views on the recordings’ limitations have been expressed by Philip Terzian, “Real History vs. Reel History,” Wall Street Journal, 20 November 1997; Robert A. Divine, “Tale of the Presidential Tapes: A Review Essay,” Political Science Quarterly, Summer 1998); and Robert Dallek, “Tales of the Tapes,” Reviews in American History (hereafter RAH), June 1998.
 “Tapes-based” books run the gamut from surveys (e.g., Doyle) to annotated conversations of a defined chronological period or subject (e.g., Beschloss) to narratives on a specific subject constructed on the basis of the tapes (e.g., Stern). Since 1997, the following books fall within this definition: Michael Beschloss, ed., Taking Charge: The Johnson White House Tapes, 1963-1964 (New York: Simon & Schuster, 1997); Stanley Kutler, ed., Abuse of Power: The New Nixon Tapes (New York: Free Press, 1997); Ernest May and Philip Zelikow, eds., The Kennedy Tapes: Inside the White House During the Cuban Missile Crisis (Cambridge, MA: Harvard University Press, 1997); William Doyle,Inside the Oval Office: The White House Tapes from FDR to Clinton (New York: Kodansha International, 1999); Michael Beschloss, ed., Reaching for Glory: Lyndon Johnson’s Secret White House Tapes, 1964-1965 (New York: Simon & Schuster, 2001); Philip Zelikow, Ernest May, and Tim Naftali, eds., The Presidential Recordings: John F. Kennedy, The Great Crises, Volumes 1-3 (New York: Norton, 2001); John Dean, The Rehnquist Choice: The Untold Story of the Nixon Appointment That Redefined the Supreme Court (New York: Free Press, 2001); John Prados, ed., The White House Tapes: Eavesdropping on a President (New York: New Press, 2003); Jonathan Rosenberg and Zachary Karabell, eds., Kennedy, Johnson, and the Quest for Justice: The Civil Rights Tapes (New York: Norton, 20003); Sheldon M. Stern, Averting the ‘Final Failure’: John F. Kennedy and the Secret Cuban Missile Crisis Meetings (Stanford, CA: Stanford University Press, 2003); Max Holland, The Kennedy Assassination Tapes: The White House Conversations of Lyndon B. Johnson Regarding the Assassination, the Warren Commission, and theAftermath (New York: Knopf, 2004). Prominent historians who have used excerpts from the presidential recordings include Robert Dallek, Joan Hoff, and Taylor Branch, as references to recorded conversations are becoming increasingly prevalent (and necessary) in new histories, biographies, and documentary series (likeForeign Relations of the United States) that cover the period from 1962 to 1973. A recent example of a book that draws from the tapes is Nick Kotz, Judgment Days: Lyndon Baines Johnson, Martin Luther King, Jr. and the Laws That Changed America (Boston: Houghton Mifflin, 2005).
 In a paper presented at the 2003 American Historical Association (AHA) meeting, Joan Hoff made the case for transcripts that reflect “the good, the bad, and the ugly. This . . . means the uhs, the uh-huhs, the hums, the yeahs, the pauses, the talk-overs, the grunts, the expletives, the ethnic slurs, and the ungrammatical, unintelligible exchanges. These cannot be glossed over as extraneous to the historical value of the tapes. . . . When these guttural expressions are left out of any presidential tape transcript a misleading impression is conveyed that these presidents and their advisers were communicating with each other precisely, decisively, and efficiently. Nothing could be further from the truth . . . ” Joan Hoff, “Comments on the Presidential Tapes Session, 3 January 2003, AHA annual meeting (hereafter AHA Comments), p. 6.
 Schulman, “Taping History,” JAH, September 1998, p. 577.
 As Hoff has suggested, however, archivists familiar with the voices on the tapes and thoroughly grounded in the history and context of the conversations would probably produce reliable transcripts. Hoff, AHA Comments, p. 17.
 Archivists at the National Archives estimated the same ratio was necessary to render a reasonably accurate transcript of one hour from the Nixon recordings. Hoff, AHA Comments, pp. 11-12.
 Kutler utilized 201 hours of Nixon tape recordings that were released in 1996 for his 1997 book. Kutler, Abuse of Power, xiv.
 John Taylor, “Cutting the Nixon Tapes,” The American Spectator, March 1998, pp. 49-50. In the paperback edition of Abuse of Power, Kutler changed his editorial commentary to correct this misleading impression.
 Hoff examined 2,700 pages of transcripts from the first 63 hours of the Watergate (a.k.a. “abuse of power”) tapes for her book, Nixon Reconsidered (New York: Basic Books, 1994). She criticized Kutler for not correcting a single transcript after errors had been pointed out to him, such as when he “cut and pasted two conversations between Nixon and [John] Dean for the morning and evening of March 16, 1973 in an egregiously arbitrary and misleading fashion.” Hoff, AHA Comments, p. 4.
 Kutler message to Stern, 1 January 2005
 Zelikow was already steeped in the subject matter. While teaching at Harvard’s Kennedy School in the mid 1990s, he had begun revising Graham Allison’s book on the crisis, Essence of Decision (1971), in collaboration with Allison. Zelikow’s research led him to the tape recordings, and eventually a second collaboration with a Harvard professor, Ernest May. Paul Mitchinson, Lingua Franca, February 2000, p. 58.
 Press Release, Harvard University Press Publicity Department, August 1997. The editors claimed that “our work transcribing the missile-crisis tapes took the better part of a year.” May and Zelikow, “White House Tapes: Extraordinary Treasures for Historical Research,” The Chronicle of Higher Education (hereafter Chronicle), 28 November 1997, p. B5.
 May and Zelikow, The Kennedy Tapes, xiii.
 Barry Gewen, “Profile in Caution, NYT Book Review, 19 October 1997.
 “The Kennedy Tapes,” The Economist, 18 October 1997.
 Richard Tofel, “Inside the Missile Crisis,” Wall Street Journal, 23 September 1997.
 Joseph Losos, “Tapes of a Superpower in a Supercrisis,” St. Louis Post-Dispatch, 16 November 1997.
 Henry Graff, “Fundamental Questions Remain, New Leader, 29 December 1997.
 Timothy McKeown, “The Cuban Missile Crisis and Politics as Usual,” The Journal of Politics, February 2000, p. 70.
 Lawrence Freedman, “The Kennedy Tapes,” International Affairs, April 1998 (an “enormous effort”); Dallek, “Tales of the Tapes,” RAH, June 1998 (“skillfully edited and annotated”); Gil Troy, “JFK: Celebrity-In-Chief or Commander-In-Chief?” RAH, September 1998 (“transcripts are sickeningly accurate”); Schulman, “Taping History,”JAH, September 1998 (“through herculean efforts); Mark White, “The Kennedy Tapes,” The History Teacher, November 1998 (transcripts of “almost undecipherable tapes” are “great contribution”); Philip Brenner, “The Kennedy Tapes,” The New England Quarterly, December 1998 (editors have “painstakingly deciphered the often jumbled discussions . . . . an enormous public service . . . ”); Mark White, “Revisiting the Cuban Missile Crisis,” Diplomatic History, Summer 1999 (editors “have converted what were muffled recordings into lucid, readable transcripts. . . . sterling work . . . .”) Freedman was the sole critic of the transcripts, writing that the “tapes themselves read like a rather bad radio script,” but he apparently thought they were a true rendering of the conversations. On the fractured nature of the transcripts, see, for example, Terry Sullivan, “Confronting the Kennedy Tapes: The May-Zelikow Transcripts and the Stern Assessments,” Presidential Studies Quarterly (hereafter PSQ), September 2000, pp. 595-596.
 Prior to becoming head of the Miller Center, Zelikow, together with May, had argued that the federal government should underwrite complete transcription of all the presidential tapes, an effort they estimated would take a decade and cost $1 million annually. May and Zelikow, “White House Tapes,” Chronicle, p. B5.
 Judith Miller, “A Trove of Telltale Tapes,” NYT, 27 June 1998. This comment is terribly ironic. May and Zelikow eventually sold the film rights to their 1997 book, and it became the basis of the 2000 production, Thirteen Days. The movie turned out to be a quintessentially misleading and fictionalized Hollywood treatment of the missile crisis, in which, among other things, the existence of the taping system, U.S. subversion of Castro’s regime, and Khrushchev’s complex motives for deploying nuclear missiles in Cuba were never mentioned. Ernest May explained his association with the movie in an article, “Thirteen Days in 145 Minutes,” The American Prospect, January 2001.
 A typical example of an uncorrected error likely created by a court reporter occurred on pages 637-638 of The Kennedy Tapes.
McCone: —the Chinese passed this note to the Cuban ambassador, on page 8 [?], implying that the U.S.S.R. was an untrustworthy ally . . . .
Unidentified: Could you repeat that, John?
McCone: Yes, sir. Peiping, the Chinese Communists, sent a note to the Cuban ambassador in Peiping implying that the U.S.S.R. was an untrustworthy ally . . . .
The second rendering of McCone's remarks is accurate. Since he was essentially repeating himself, however, careful editing should have corrected “on page 8 [?]” to read as “in Peiping.”
 A typical example of an audible portion that was transcribed nonsensically in The Kennedy Tapes, from page 133:
JFK: . . . I would think you [the Soviets would] have to go on the defensive, are not going to commit nuclear weapons to be used against the United States from Cuba . . . .
JFK: I would think you have to go on the assumption that they’re not going to permit nuclear weapons to be used against the United States from Cuba.
 Stern, “What JFK Really Said,” Atlantic Monthly, May 2000, pp. 122-128.
 May and Zelikow, "Letter to the Editor: What JFK Really Said," Atlantic Monthly, August 2000, p. 13.
 “Details drive debate,” wrote May and Zelikow on page 698 of their conclusion to The Kennedy Tapes. That perceptive insight would seem to underscore the desirability of getting details right in transcribing the recordings.
 While describing the historical importance of recordings versus recorded minutes, the editors wrote, the “Kennedy and Johnson tape recordings catch people’s verbal emphases, hesitations, and shifts in voice that written minutes do not reflect.” May and Zelikow, “White House Tapes,” Chronicle, p. B5.
 Stern used a home tape player and low-tech, analog audio cassettes from the JFKL.
 Stern, “Source Material: The 1997 Published Transcripts of the JFK Cuban Missile Crisis Tapes: Too Good to Be True?” PSQ, September 2000.
 Zelikow and May, “Source Material,” PSQ, December 2000, pp. 794, 796.
 “JFK’s ‘Great Crises’ in Bookstores This Month,” Spectrum, a publication of the Miller Center of Public Affairs, Fall 2001.
 A review in Diplomatic History took note of Stern’s criticisms of the HUP edition, but then presumed that the new, revised transcripts were bound to be the best possible. James Giglio, “Kennedy on Tape,” Diplomatic History, November 2003, p. 748.
 May and Zelikow, in their response to Stern’s PSQ article, had assured users of The Kennedy Tapes that the reference volumes would not render the HUP edition unusable. “We think few will find the many amendments in our retranscriptions to be very important,” they asserted in December 2000. According to the editors, they nonetheless did urge HUP to issue a more fully revised work after the Miller Center volumes were published. “But the press disappointed us by deciding that, in their judgment, the amendments and the additions were not significant or interesting enough to justify the cost.” Zelikow and May, “Source Material,” PSQ, December 2000, pp. 793, 794.
 On the first day of Excomm meetings, for example, JFK observed, “If you go into Cuba in the way we’re talking about, and taking out all the planes and all the rest, then you really haven’t got much of an argument against invading.” The Norton version omitted (as did the HUP edition) the single word “out.” Aside from muddling the president’s observation, it could also leave the false impression that JFK was talking about air cover for a U.S. ground invasion when he was actually referring to destroying Soviet and Cuban aircraft on the island’s airfields. Naftali and Zelikow, Great Crises, vol. 2, p. 449; May and Zelikow, The Kennedy Tapes, p. 98.
Another example of a flawed retranscription that changed the speakers’ meaning can be found in the rendition of an exchange from Saturday, October 27, easily the pivotal day of the entire crisis. From pages 492 and 493 of Great Crises, volume three, edited by Zelikow and May:
McNamara: I would say only that we ought to keep some kind of pressure on tonight and tomorrow night that indicates we’re firm. Now if we call off these air strikes tonight, I think that settles that—
JFK: I [unclear] want to do that, I think—
This rendition (similar to the one on page 612 of the HUP edition) should have been recognized as making no sense because there were no air strikes scheduled for Saturday night. McNamara was actually recommending that the Pentagon call up 24 Air Force reserve squadrons.
McNamara: I would say only that we ought to keep some kind of pressure on tonight and tomorrow night that indicates we’re firm. Now if we call up these air squadrons tonight, I think that settles that.
JFK: That’s right. We’re gonna do that, aren’t we?
 For additional examples of errors, see Stern, “The JFK Tapes: Round Two,” RAH, December 2002, pp. 680-688, and “The Published Cuban Missile Crisis Transcripts: Rounds One, Two and Beyond,” in Averting the ‘Final Failure,’ pp. 427-440.
The Miller Center’s process for producing transcripts, as distinguished from the May-Zelikow methodology, was described in detail in Zelikow and May, “Source Material,”PSQ, December 2000, pp. 795-796. “First, the work is done by trained professional historians who have done deep research on the period covered by the tapes and on some of the central themes of the meetings and conversations. . . . Second, each volume uses the team method. . . . Usually one or two scholars painstakingly produces a primary draft. . . .Two or more scholars then carefully go over that transcript, individually or sometimes two listening at the same time. . . . In the case of often-difficult meeting tapes, like the Kennedy recordings, every transcript has benefited from at least four listeners. The volume editors remain accountable for checking the quality and accuracy of all the work in their volume, knitting together the whole. All of this work is then reviewed by the general editors [Zelikow and May]. . . . Third, we use the best technology that the project can afford.”
 May and Zelikow, Kennedy Tapes, p. xiii.
 Ibid., pp. xiii-xiv.
 See also footnote 4.
 The Miller Center volumes explicitly state that its Presidential Recordings Program was undertaken “on the premise that these recordings will remain important historical sources for centuries to come.” Zelikow and May, Great Crises, p. xiii.
 Hoff, AHA Comments, p. 7.
 Spectrum, Fall 2001.
 Johnson’s original instructions were that the tapes would be opened no sooner than 2023, fifty years after his death.
 Beschloss, Taking Charge, p. 551. The fact that the Johnson recordings are played weekly on C-SPAN is good evidence that the audibility of these tapes is often excellent—despite lingering issues of punctuation.
 In addition, Taking Charge included several assassination-related conversations from late 1966 and early 1967.
 Beschloss, Taking Charge, p. 551; Reaching for Glory, pp. 429-430. The description in Reaching for Glory is only slightly different.
 Beschloss, Taking Charge, p. 551.
 Beschloss does not explain who else transcribed the balance of the conversations, or what percentage was represented by “most,” an issue also raised by Joan Hoff in her AHA Comments, p. 4. In Reaching for Glory, Beschloss described his methodology slightly differently. “Some of the tapes I have transcribed from scratch. For others I have started with rough transcripts created by LBJ’s secretaries or by a professional transcriber and listened repeatedly to the tapes, making substantial corrections, adding emphasis, punctuation, and transliteration of words.” Beschloss, Reaching for Glory, p. 431.
 Beschloss, Taking Charge, pp. 551-552.
 Brinkley, “D.C. Confidential,” NYT Book Review, 19 October 1997 (“superbly edited and annotated”); Michiko Kakutani, “Overhearing History: Johnson’s Secret Tapes,”NYT, 10 October 1997 (“expertly selected, edited, and footnoted”).
 Editorial, “LBJ, From the Inside,” NYT, 11 October 1997. Newsweek also published excerpts of Beschloss’s transcripts, and ABC News devoted several Nightline shows to readings of the Johnson (and Nixon) tapes, with commentary by Beschloss and other historians.
 Keith Kyle, “Taking Charge,” International Affairs, April 1999, p. 421; Dallek, “Tales of the Tapes,” RAH, p. 335. In a review of Beschloss’s second volume for Diplomatic History, John Prados chided the author for not telling the reader “much about his criteria for inclusion or exclusion” of a conversation. But questions about the nature of the transcripts themselves were not raised. Prados, “Looking for the Real Lyndon,”Diplomatic History, November 2003, pp. 752, 755-756.
 Schulman, “Taping History,” JAH, p. 574.
 For some examples of meaningful mistakes in Beschloss’s transcriptions and/or presentation of the conversations, see Holland, Kennedy Assassination Tapes, pp. 60, 61, 95, 240, 248, 251.
 One British reviewer specifically faulted Beschloss for the lack of context. Anthony Howard, “He Had It Taped,” Sunday Times ( London), 5 April 1998.
 By contrast, there was some criticism of Beschloss’s second volume for his interpretation of Johnson’s conversations on Vietnam. Jack Valenti, “LBJ’s Unwinnable War,” Washington Post, 28 November 2001, and in a rejoinder, Beschloss, “LBJ’s Secret War,” Washington Post, 1 December 2001. But again, the transcripts themselves were presumed to be above reproach. Michiko Kakutani, “Johnson Tapes Show a Man Full of Doubt, Even as Victor,” NYT, 13 November 2001 (“expertly edited and annotated”).
 Holland, “Tapes: Hearing a Wrong Leaning, er, Meaning,” Los Angeles Times, 1 August 1999. On 18 September 1964, the day the Warren Commission met to deliberate for the last time, Senator Richard Russell had a conversation with President Johnson. When the subject of the Commission arose, Russell said, “I tried my best to get in a dis-sent, but they’d come ‘round and trade me out of it by givin’ me a little old thread of it.” Russell was explaining that the other commissioners massaged the language until Russell had no grounds for a published dissent. In Beschloss’s rendition, Russell said, “I tried my best to get in a dis-sent, but they’d come ‘round and trade me out of it by giving me a little old threat.” The implication created is that pressure was exerted on Russell and that he signed the unanimous report under duress, which was quite untrue.
This error was striking for two reasons. First, it led Beschloss to misrepresent Russell’s perspective on the assassination, if not the Warren Report itself. And second, it was an inexplicable error to make, if only because had Earl Warren (or anyone else on the Commission) even vaguely threatened Russell, all hell would have broken loose.
In the revised “Editor’s Note” in Reaching for Glory, Beschloss pledged to correct errors in future printings of his Johnson volumes. Although there have not been, to the authors’ knowledge, subsequent printings of Taking Charge, Beschloss could have easily used the publication of Reaching for Glory in 2001 to correct this mistake.
 Zelikow, “Recent Books,” Foreign Affairs, January/February 2002.
 The editorial conventions were also not applied consistently, i.e., sometimes larger deletions were not marked off and/or ellipses not employed where they should have been in theory.
 See the comparison between the Holland and Beschloss treatments of a 22 November 1963 conversation between Johnson and Associate Justice Arthur Goldberg in Stern,“Presidential Tapes and Historical Interpretations,” RAH, December 2004, pp. 586-589.
 May and Zelikow, “White House Tapes,” Chronicle, p. B5.
 Zelikow and May, “Source Material,” PSQ, December 2000, p. 793.
 Zelikow presentation at the Presidential Tapes Conference, John F. Kennedy Library, Boston, 16-17 February 2003.
 “Response of Philip Zelikow,” to Stern, “Errors Still Afflict the Transcripts of the Kennedy Tape Recordings,” History News Network, 15 March 2004.
 Hoff, AHA Comments, p. 9.
This essay first appeared on History News Network, 21 February 2005
© 2007 by Sheldon M. Stern and Max Holland
© 2007 by Sheldon M. Stern and Max Holland