Don’t believe everything you read in an FBI file

In late May, David J. Garrow, author of a well-regarded biography of civil rights leader Martin Luther King, published an upsetting claim: In a newly released cache of FBI documents, he claimed to have found credible evidence that King had “looked on, laughed and offered advice” while a fellow minister committed rape.

Garrow’s claim appeared in Standpoint, a rightwing British magazine, after more than two dozen other publications declined to run it. Soon after, a number of historians raised doubts about Garrow’s interpretation of the evidence, as the New York Times, the Washington Post, and the Guardian reported. Several pointed out that it seemed naive to take an FBI document about King at face value, given the long campaign that the bureau had waged to discredit and demoralize him. Historians also noted that the words “looked on, laughed and offered advice” hadn’t even been typed but appeared in a handwritten annotation by an unidentified writer.

In his Standpoint article, Garrow explained why he gave credit to the words nonetheless. At the end of King’s life, the FBI was wiretapping King’s phone calls and bugging his hotel rooms. Garrow guessed that the handwritten annotation in question was made either by William C. Sullivan, an assistant director at the FBI, or one of his deputies, on the basis of a transcript of audio surveillance. “Without question Sullivan and his aides had both the microphone-transmitted tape-recording, and a subsequent full transcript at hand while they were annotating their existing typescript,” Garrow wrote. Garrow didn’t think it likely that Sullivan and his colleagues would have distorted their evidence because, he argued, “Sullivan could not have imagined that his and his aides’ jottings would ever see the light of day,” since at that point, none of the FBI’s internal files had ever been released to the public.

That claim seems a little far-fetched, given that Garrow himself describes, in his Standpoint article, how hotly the FBI lobbied the President and the attorney general for permission to put King under surveillance and keep him there. Often Sullivan and his colleagues were writing for an audience, though their audience was a select group of politicians rather than the general public. And if the aim of a piece of writing is to persuade, rather than merely to document, a historian has to read it more skeptically. Nor is it at all unreasonable to imagine that the bureau’s hatred of King might have distorted even its internal analysis; the campaign against King extended as far as sending the civil rights leader an anonymous letter advising him to kill himself, which doesn’t suggest that the bureau’s decision-makers were very cool and dispassionate.

Though the historian Barbara Ransby has critiqued Garrow’s claim incisively, the commentator Lance Morrow has found Garrow’s claim credible enough to wrestle with it, suggesting that it’s being taken seriously. I don’t have Ransby’s expert knowledge of King or of the mid-20th-century FBI, but when I followed the links in Garrow’s Standpoint article to his evidence, which is available online in the form of PDFs hosted by the National Archives and Records Administration (NARA), I noticed that the documents themselves provide reasons to doubt Garrow’s description of them. To be clear, the documents do seem to me to make it pretty hard to deny that King was a libertine in sexual matters—and maybe a more adventurous libertine than is popularly understood. But the allegation that King was a complacent bystander to rape is to my mind much more serious, morally speaking. Since I haven’t yet seen anyone else online set out the evidence within the documents that militate against taking the documents at face value, it seemed worthwhile for me to try to do that here.

I’m actually only going to be discussing one document in detail, the one containing the allegation that King witnessed and encouraged rape. Note that the beginning and end of the document are redacted; the earliest legible page is numbered 8 in its handwritten pagination (page 15, in NARA’s pagination of the PDF), and the last legible page is numbered 33 (page 39, in NARA’s pagination). From this point forward, for convenience, I’m going to cite only NARA’s page numbers.

Take a look for yourself; a couple of observations are easy to make. First, this document was a group effort. Even the typing seems to have been done on more than one typewriter. On page 17, for example, the paragraph beginning “In 1964, he gave another friend several hundred dollars . . .” has different margins than the rest of the page and is in italics. It also looks to me as though page 22 was typed on a different machine from the preceding pages, though I’m not that good at recognizing typescripts, and I might be misreading the distortion of photocopying.

More pertinent for our purposes, note that the document is annotated in two distinctive styles of handwriting. On pages 15 to 22, the handwriting of these annotations is a loopy cursive. Here’s a sample:

NARA document 32989551 page 16

In this screenshot, taken from page 16, the handwritten portion reads, “In an argument with Mrs Dolores Evans in New York one night he threatened to commit suicide.” For reference’s sake, let’s name the author of these cursive annotations Loopy.

On page 22, another style handwriting appears: printed letters, in a darker ink. Here’s a sample:

NARA document 32989551 page 27

This screenshot is from page 27. The handwritten additions include the letter “D,” the words “another Negro,” and the label “Sensitive Foreign Intelligence Operation—ONGOING.” Let’s give the author of these hand-printed annotations the name Blocky. And then let’s forget about him, because he’s not really germane to our argument.

A further observation: In the typescript, between pages 15 and 21, references are intermittently provided within parentheses to what seem to be the sources of particular claims. For example, on page 20, the claim that an aide “later confided to a friend that King’s group was running naked, drunk white prostitutes up and down the halls of the hotel” is sourced to “NY airtel, 12/17/64, re Martin Luther King.” In the next paragraph, there’s a claim that “at least five other men in King’s party made the same inquiry, being particularly interested in learning where the ‘Norwegian girls’ could be found,” and this is sourced to “100-106670-825,” which seems to correspond to a numbering system for documents used internally by the FBI.

What can we infer from this group of (rather obvious) observations? It would be reasonable to guess that Loopy and Blocky are different people, though it’s possible that they’re one person with two different handwriting styles. Either of them might also have been one of the typists of the document they annotated, although, again, it seems likelier that neither of them was. In any case, even if one of the annotators was also one of the typists, it’s safe to say that the handwritten annotations were made after the typing. The annotator was returning to the typed document, with an intention to change it. And by means of a few more observations, we will be able to infer something about that intention.

It’s Loopy that we’re particularly interested in, since he’s the one who made the explosive allegation against King. Here’s the relevant passage, which spans pages 17 and 18 in NARA’s pagination:

NARA document 32989551 pages 17-18

The underlying typescript of this passage reads as follows:

Willard Hotel Episode

On January 5, 1964, King and several SCLC officials checked into the Willard Hotel in Washington, D. C. In a room nearby was a Baptist minister from Baltimore, Maryland, who had brought to Washington several women “parishioners” of his church. The group met in his room and discussed which women among the parishioners would be suitable for natural or unnatural sex acts. When one of the women protested that she did not approve of this, the Baptist minister immediately and forcibly raped her. (100-3-116-762)

To the heading “Willard Hotel Episode,” Loopy has added the annotation “more on this.” Underneath the phrase “a Baptist minister from Baltimore,” Loopy has written “his name.” Loopy has written “name” again, above a second instance of the phrase “Baptist minister” later in the paragraph. Another addition of Loopy’s specifies that it’s “unnatural sex acts” that one of the women in the room disapproves of; the way the original typescript reads, it’s possible to think she disapproved of sorting the women by sex act. And then at the close of the paragraph, Loopy adds the incriminating sentence about the alleged rape: “King looked on laughed and offered advise” [sic].

Just from Loopy’s intervention in this paragraph alone, a couple of inferences seem justifiable. First, Loopy seems to be revising the underlying typescript with an eye to presenting the evidence it contains to a new audience, and for that purpose, he wants more details added to it about the Willard Hotel episode. Second, either those details aren’t available to Loopy as he’s revising, or they’re not within easy reach; he seems to be deputizing someone else to look the details up and insert them. In his Standpoint article, Garrow identifies the other minister in the room with King as Logan Kearse, which isn’t a long name. If Loopy had had Kearse’s name on the tip of his tongue, or staring at him from an audio transcript, he would probably have just written “Kearse” twice instead of writing “his name” and “name.”

Nor is this the only evidence of Loopy not knowing, or not having ready access to, factual details. The very next paragraph on page 18 concerns a “sex orgy,” and in the margin, Loopy has written “where?” At the bottom of page 19, a note from Loopy directs someone to “set out statement”, i.e., insert a statement that a prostitute gave to the FBI. In the jargon of journalism, Loopy is doing a top edit, not a line edit. In any case, he’s certainly not a writer doing his own revision. He isn’t looking at the sources; he’s directing someone else to go back to them. It’s simply not true that, as Garrow wrote in Standpoint, “Without question Sullivan and his aides had both the microphone-transmitted tape-recording, and a subsequent full transcript at hand while they were annotating their existing typescript.” To the contrary, Loopy seems to be editing either from a hazy memory of the intel or from hearsay.

Loopy’s annotations also reveal him to be someone with a heavy slant on King. At the top of page 16, for example, he crosses out the word “assignations” and replaces it with “adulterous sex relations,” which is poor editing from a stylistic point of view, since the phrase “adulterous sex relations” is longer than “assignations” and means more or less the same thing. Loopy seems motivated here by a wish to make King’s conduct sound worse—to remind the reader that King was married when he was having this promiscuous sex. In fact, almost all of Loopy’s additions seem aimed at making King look worse than the typescript already makes him sound. Lower on page 16, below the typescript’s claim that King’s romances had come to the attention of his associates, Loopy adds “and also to the employees of motels and hotels.” On the top of page 17, after the typescript notes that “King once told some SCLC associates that he was wasting his money on sex (100-3-116-762),” Loopy adds,

This is true. for years he has [missing text] on females with whom he has sex relations. They stay at the most expensive hotels he pays their travelling expenses and gives them costly gifts. This comes from funds given to him to help the poor negro.

Unlike most of the claims against King in the typescript, Loopy’s addition here is without detail or sourcing. Which women? Which hotels? Which funds? Loopy doesn’t have the receipts. The tone, moreover, is a little ranty. A little lower on the page, after a description in the typescript of King spending more than $600 on phone calls, Loopy feels compelled to add, “merely to carry on sentimental conversations”. Again, Loopy is contributing tendentiousness rather than detail. His addition doesn’t tell the reader anything new about King and his lover; it merely tells us that Loopy deems the expression of their feelings to be of little worth.

Just by inspection of these pages, in other words, we have good reason to infer that the handwritten annotation alleging that King “looked on laughed and offered advise” to a rapist was added by someone who (1) was at a remove from the facts while he was making his annotations, and (2) had an axe to grind against King. From this internal evidence, therefore, it seems likely to me that the annotation about King looking on at rape and laughing is an embellishment on Loopy’s part, grounded in nothing more than his animus.

One last doubt, about the alleged rape itself. In the typescript, the allegation that Kearse committed rape is sourced to a document with the number 100-3-116-762. (Note that the claim that King once said he was “wasting his money on sex,” on the preceding page, is sourced to the same document.) Garrow has suggested that if a document like 100-3-116-762 was the transcript of an audio recording, it should be more persuasive to historians than the write-up of an interview with an informant would be. Speaking to the Washington Post, Garrow argued that it is reasonable to give more credence to the results of “electronic surveillance” than to reports that came “literally third- or fourth-hand from a human informant.” But if document 100-3-116-762 is based on an audio transcript, rather than on the testimony of a victim or an eyewitness, then there’s another problem. Are we willing to trust that Loopy, or the typist whose work he was revising, was able to distinguish the sound of a rape from the sound of normal sexual excitement? And are we willing to trust either of them to be honest about what he heard? A historian who takes the documentary evidence discovered by Garrow at face value is choosing to rely on the perception, judgment, and moral fiber of a nameless FBI officer of the 1960s, who, among other telltale signs of prudishness, seems to have consistently described oral sex as “unnatural.” Unfortunately for Garrow’s argument, if living with the internet for the past couple of decades has taught us anything, it is a distrust of the notion that it’s possible to reach through electronic surveillance a truth that will not have been distorted by human intention and perception.