By Robert Reynolds
Once again the National Archives is preparing to release previously redacted documents from the John F. Kennedy Assassination Records Collection (JFK ARC). And once again the idea is becoming rampant that significant and newsworthy information on the assassination of President Kennedy 58 years ago will soon be revealed.
News coverage of the coming releases started out muted this time around compared to the last big releases, which occurred from July 2017 to April 2018. It picked up markedly, however, when President Biden released a memo on October 22 that spelled out short- to long-range plans for the collection.
The memo called for a first release of redacted documents that agencies no longer wish to withhold in mid-December, then another, larger, release by December next year. The memo also mandated a boat-load of paperwork for agencies to fill out if they want to postpone anything past that date. A longer-range element of the plan calls for the National Archives and Records Administration (NARA), the federal agency that houses assassination-related materials, to scan the entire collection and put it on-line. How long that will take is not spelled out in the memo.
Reaction to Biden’s memo ranged from yawns to apoplectic complaints that Biden has sold out to pressure from the Central Intelligence Agency (CIA) and is suppressing incriminating documents.
That a collection of mostly paper documents can stimulate such passions is a tribute to the depth of the national trauma inflicted by the assassination. Given the confusing result of the last releases, it is also understandable that people might think there are still important documents left unreleased. A closer look at NARA’s latest data on the ARC, however, reveals just how unlikely this is.
The JFK Database
The 2017-2018 releases from NARA were confusing, first, because it was unclear how much material was from documents previously withheld in full, and second, because it was equally unclear how much was from documents that had been previously released in part. This should not have been the case.[1]
The primary reason for the uncertainty was outdated information in the JFK database, the computer listing of metadata for most records in the collection. The creation of the database had been mandated by the President John F. Kennedy Assassination Records Collection Act, the 1992 public law that established the JFK archival collection in the first place. The purpose of the database was to provide a detailed index to the contents of the collection, including basic information for every record, such as the title and date of the document, the agency that generated the record, number of pages, and so on.
The main problem that led to the confusion was a failure to update the release status of each record. Why was this data field not better maintained? During the five years the database was being assembled, 1993 to 1998, there were frequent changes in the status of records. Agencies argued with the Assassination Records Review Board (ARRB), the limited-term panel created to oversee the assembly and release of the JFK records, about which documents counted as assassination-related records, and what standards should apply for withholding documents in full or redacting portions of their text. These status changes were not always recorded in the database. With more than three hundred thousand records in the database, that should not be surprising.
In addition, the online version of the JFK database was not updated to reflect the status of records in the collection after the 2017-2018 releases. During this period, NARA posted more than 600,000 pages online, in tens of thousands of pdfs, in seven discrete releases. Many of the documents made available online were posted in multiple versions, or were posted with multiple documents lumped into one pdf file. All of this made it a practical impossibility for the general public to keep track of which record was released in full, and which still had redactions.
This lamentable situation was finally addressed when NARA began an epic review of every record in the collection. The old public version of the JFK database was taken off line in October 2020, and in June of this year, NARA posted an updated version of the JFK database in the form of six large Excel files (available here).
This latest version is a significant update, correcting some widely cited figures. For example, a search of NARA’s own JFK ARC webpages will still turn up the estimate that 88 percent of the ARC documents are open in full. The update shows that following the 2017-2018 releases, approximately 95 percent of the documents are now released in full.
This update, though highly significant, is still not the end of the story. A closer look makes clear that NARA’s top-to- bottom review of the collection was not yet complete by May of this year, although that is the posted online date of the update. No doubt NARA is now even further along in the review process, seven months later, having placed a very high priority on reviewing the ARC (before but especially after President Biden’s memo). The JFK Act mandated a level of indexing and detail, however, that is found in no other NARA collection, and the volume of records is truly huge. If NARA actually finishes its review of the ARC this year, it will be doing very well indeed.
Although a work in progress, the May 2021 update of the JFK database still gives us the most current and accurate view of redactions in, and withholding of, ARC documents. Thanks to the large number of records NARA posted online from 2017-2018, moreover, those who are interested can check the database directly against the documents online.
Unfortunately, this is not the first time NARA’s diligent work on the collection has been largely ignored during the latest flap of publicity. The ARC is once again being touted as a source for information that could “blow the case wide open.” The updated database shows otherwise.
Withheld Records
A look at the updated release status for the 300,000 plus records in the database tells a big part of the story. It distinguishes three types of records in the collection: open, redacted, and withheld.
Withheld means that the entire record is not released. Although database information for these records gives us an idea of what is in them, the actual content is not available for public inspection.
The May 2021 update now shows 597 records withheld in full. Almost all of these are tax returns, including ones from everyone in the family of accused assassin Lee Harvey Oswald; then from everyone in the family of Oswald’s vigilante murderer, Jack Ruby; then from every company that Oswald ever worked for; and one from even Ruby’s primary lawyer, Joe Tonahill.[2]
Why are these withheld in their entirety? Tax returns were specifically exempted from release under section 11 of the 1992 JFK Act. The ARRB tried to find various ways around this loophole, but in the end most of them remain closed under the Internal Revenue Code. It turns out the IRS is even more successful at keeping information secret than the CIA.
Five withheld records are grand jury or court-sealed records. These are withheld under section 10 of the JFK Act. This section actually gave the ARRB power to petition for the release of such records, but with the exception of electronic surveillance on New Orleans mafia figure Carlos Marcello, the ARRB found nothing worth unsealing.
Eleven records are interviews with, and letters by, members of the Kennedy family, donated to the National Archives under deeds of gift that restrict their release. The family did not agree to allow these records to become public. The JFK Act allowed such deeds to close records in the collection to the public, so like the grand jury or court-sealed records these Kennedy documents are not available for inspection.[3]
Adding all these up, 519 records are withheld in full under sections 10 and 11. In April 2018, NARA estimated there were 520 such records, so their original survey was fairly accurate.
It bears pointing out that with one small exception, all of these withheld-in-full records have been examined by at least three sets of investigators: the Warren Commission in 1964, the House Select Committee on Assassinations in the late 1970s, and most recently, the ARRB. But if continuing to keep these documents closed convinces one that JFK’s death must have been a conspiracy, or that one has to be able to see these documents before one can decide whether there was a conspiracy, there was really no reason to wait until now to decide. The decision to withhold these documents was made 29 years ago, when the JFK Act became Public Law 102-526 in 1992.[4]
What about the other 78 records that are “withheld”? These are records where there is an identification aid (metadata), but NARA can’t identify the record that goes with it. There were originally 86 of these, but NARA was able to winnow the number down. The remainder may simply be data inputting mistakes or mislabeled. Time will tell if NARA can turn up a few more.
Redacted Records
In contrast to withheld records, redacted records are available to the public with only a part of the text held back. In April 2018, NARA estimated there were 15,834 of these records. The updated database, however, lists 16,283, an increase of about 400.
Some of this variation is merely book-keeping. An example is the microfilm duplicate of Lee Oswald’s CIA “201” file, registered as 179 individual documents, which have been switched in status from “withheld” to “redacted.” NARA determined that everything in the microfilm duplicate exists as released or redacted records elsewhere in the collection. The remainder of the difference between 2018 and 2021 numbers is probably explained by a corrected count recently.
Besides finally showing which records are still redacted, NARA is also cleaning up the “restrictions” markers in the database. The JFK Act originally provided a complicated set of exemptions that allowed text to be held back. After October 2017, however, most of these loopholes were closed.
For text that does not fall under section 10 and section 11 restrictions, there is now only one basis for withholding it, and the originating agency must appeal to the president directly for this to occur. The JFK Act section that gives the president the authority to hold back text is 5(g)(2)(D), but let’s just call this the 5(g) exemption.
Looking again at the document restrictions column in the updated JFK database, we can see that NARA has done a lot of work here. Most of the records listed as redacted are now marked as either 10(a) (court documents), 11(a) (tax documents), or 5(g), representing the document portions that then President Trump agreed to continue withholding in 2017-2018.
In the May database update, however, there are still a few hundred records marked 6-1, 6-2, etc. These represent documents redacted under the old, pre-2017 withholding exemptions, which means that in May 2021 NARA had not yet finished this part of the update. Once that is done, we will finally have an accurate count of the current universe of records that have text held back and on what grounds.
The May update gives us, in the meantime, a pretty good preview of what is still to come, and who is withholding what. There are still about 1800 records, for example, that have some of their text redacted under sections 10 and 11 of the JFK Act. These are almost all FBI records on organized crime. For records marked with only 5(g) redactions, the CIA holds a commanding lead in this category, with slightly more than 11,000 records listed. The CIA is thus responsible for about 86 percent of all the 5(g) redacted records in the ARC.
Numbers like this are always tentative of course, especially since NARA has not yet finished its review. Even then, one can be sure that errors will remain, as they do in every big text database, from the Library of Congress’s virtual card catalogue to the Oxford English Dictionary. This is what text database users have to live with.
Identifying which records have text redactions, of course, does not tell us how much text in each record was withheld. This obviously matters; a heavily redacted document may not have enough text remaining to make sense. If one has the patience or basic technical knowledge, however, this data is not hard to come by; all of the redacted documents can be downloaded from NARA, which posted them online by April 2018. Downloading is the easy part though; reading through them is a different story.
Finding the little blank boxes that signify redactions is not hard. A careful look shows that there is a wide variation in how much text is held back in each document; it can range from two letters (aka cryptonym digraphs used by the CIA) up to dozens of pages. One can even add up all these blank spaces to discern how much unreleased text remains to be released. The answer is not much.
The 11,000 plus CIA records with 5(g) redactions include 5600-plus documents that are only one page long, or slightly more than half the records. A one page document is not difficult to read (these are mostly short cables and memos), and although counting redaction boxes is boring, it doesn’t take that long. More than 70 percent of these one-page documents have just one or two redactions, often a single word or item. Many are marked with substitute language codes, numbers that the ARRB developed to convey the content of redactions: [03] means a CIA employee; [11] refers to a country; [24] means operational details; and so on.
Where the redaction is not marked with a code, which is often the case in cables and memos, usually it is obvious that the box conceals the name of the sender or recipient. It seems safe to presume, consequently, that most of these single-item redactions are either 1) the names of CIA employees, who wrote or reviewed cables, or sent and received memos; or 2) names of CIA installations abroad.
Documents more than one page sometimes have longer redactions, including runs of whole pages. These are not marked with substitute codes so the contents here are less clear. Yet there are not many of these. Fewer than one hundred CIA documents have whole pages withheld, out of a total of 11,000 plus redacted records. Most of these whole-page redactions are in documents that the ARRB designated NBR: “not believed relevant,” despite being gathered under the rubric of assassination-related. Biden’s memo will set more stringent release standards for these longer redactions. And even if the CIA appeals to keep these deletions, and Biden entertains the agency’s pleas, more details of what is in the documents will be made known.
The most important point to keep in mind, of course, is whether release of any or all of these longer redactions will enhance historical understanding of JFK’s assassination. Is it possible these deletions will significantly change what we know (or think we know) about the event? The answer is a resounding no.
In the first place, the ARRB would have insisted on making the information public during its tenure. All of the board members considered that their prime directive under the JFK Act, even if it wasn’t spelled out precisely in those terms. To believe that the review board consciously sat on documents that fundamentally challenge the official story (i.e., the Warren Report) is to misunderstand completely the animating purpose of not only the five Senate-confirmed board members, but the entire ARRB staff.
A more realistic understanding of how release of more text will affect historical interpretation can be achieved by looking at discrete records, say, for example, JFK ARC 104-10167-10140. This 29 July 1959 cable consists of one sentence: “Request Zamka bring personal letter from Marcos wife to bolster morale.”
The “Marcos” in this cable was Marcos Díaz Lanz, brother of Pedro Díaz Lanz, then head of the Revolutionary Air Force of Cuba under Fidel Castro. Both brothers were strong opponents of the dictator, and by the summer of 1959, they had decided it was time to get out and arranged to be exfiltrated by the CIA. When this cable was sent, Marcos was still in a safe house and his family had not yet left Cuba. Zamka seems to be the pseudonym of a CIA operative (perhaps David Morales), who was running exfiltration efforts. The only redaction now remaining in this document is the name of the “releasing officer” who cleared the transmission of the cable. This is a very common type of redaction.
If this cable was indicative of a conspiracy against Kennedy (at a time when he was still a junior senator from Massachusetts), declassifying the name of the releasing officer might be important. But the cable is clearly unrelated to any conspiracy, and releasing the name will add little to nothing about our understanding of the assassination. There are thousands of such instances in the redacted records. To treat such redactions as evidence of CIA involvement in the assassination is absurd.
The same general principle applies to longer redactions. Take JFK ARC 104-10170-10121. This is an 87-page long CIA file on AMSPELL, the CIA cryptonym for a once pro-Castro student organization known as the Directorio Revolucionario Estudiantil (DRE). It later broke with Castro and became one of his most dedicated organizational opponents. The file was originally withheld in full. It was then released twice, in November 2017 and again in April 2018. The first release held back a big portion of the file, about 60 full pages. In the second release only nine full pages were still held back, along with a sprinkling of individual redactions.
In a November 2021 op-ed for the Miami Herald, Jefferson Morley, a former Washington Post journalist, cited this file as an instance where “If and when they are released, these pages may shed light on the CIA cover-up that followed JFK’s murder.” This assertion makes no sense. The still-withheld pages are part of a 15-page attachment to a dispatch that JMWAVE, the CIA station in Miami, sent to Washington headquarters in November 1965. The letter describing the attachment states, “Forwarded under separate cover is a list of political prisoners in Cuban jails. The list was compiled by the AMSPELL organization and contains the names and addresses of relatives who are living in the United States.”
The attachment itself consists of two parts. The first section lists prisoners with their names redacted but with other information about them intact, including the date they were detained (Detención); what looks to be the legal statutes they were charged with violating (Causa); the length of their sentences (Años); and the provinces where they were arrested or are imprisoned (Provincia). It is a long list, with sentences ranging from one to 30 years.
The other section of the attachment is entirely withheld but the form of the text is preserved, and it is also clearly a list. Based on the cover letter, this list must have provided the names and addresses of the prisoners’ relatives in the United States. It defies credulity to see how “these pages may shed light on the CIA cover-up [sic] that followed JFK’s murder.”
Yet Morley, who has written two books based loosely on ARC documents, is convinced that redactions in the collection conceal powerful evidence of CIA malfeasance in the JFK assassination, while ignoring clear evidence to the contrary. Based on his conviction, he argues that first Trump, and now Biden, are caving to the CIA for reasons of state. Fearful of “alienating his national security partners,” Morley wrote recently, President Biden “has to go with the Agency’s flow.”
As Biden’s memo makes clear, however, he is basically following procedures and guidelines suggested by David S. Ferriero, the tenth archivist of the United States and as such, NARA’s head. Ferriero does not appear to be cowering in fear of the CIA. In a March 2018 letter to Trump on the subject of the JFK records, he wrote that “We have had extensive discussions with CIA and FBI since 26 October 2017 regarding their review methodology and our assessment of prior reviews. During the course of our review, we identified questions or concerns, and adjustments were made by D[epartment of] D[efense] as well as CIA. Based on the results of the further review by agencies, an additional 5,821 records are being released in their entirely.”
Morley essentially argues that proof of the CIA’s complicity in the assassination resides in the mere fact that some CIA records gathered under the JFK Act are still redacted. Not true, and one need not wait for new releases to verify this. The context and nature of current redactions makes it clear just how little historically-significant information is still being withheld. Look for yourself.
Robert Reynolds is a professor of English at National Chi Nan University in Taiwan. This article is his first for Washington Decoded.
[1] For a comprehensive essay on the work that went into the JFK assassination records prior to the 2017-2018 releases, see James Mathis and Martha Wagner Murphy, “Documenting the Death of a President: The John F. Kennedy Assassination Records Collection,” Prologue, Fall 2017, Vol. 49, No. 3.
[2] The update, in fact, shows 602 records as withheld. In a November 2021 email, however, NARA confirmed to me that five of these records were released during 2017-2018.
[3] A deed of gift is not the only obstacle holding some of these records back. Author William Manchester’s interviews with Jacqueline Kennedy and Robert Kennedy, conducted in May 1964 and January 1965, were sealed by a court order until 2065 as the result of a lawsuit the Kennedys brought against Manchester in 1967.
[4] The only documents never reviewed by any federal entity are the six interviews that Jacqueline Kennedy and Robert Kennedy gave to Manchester. The Kennedy family turned down the ARRB’s request to review this material.
©2021 by Robert Reynolds
Comments