This is a brief note on an issue related to FBI multi-record pdfs (see here for an introduction to this troublesome topic). A much longer, more substantial note on these pdfs is coming up.
The short version
The point of this post is that there is a bad link in the 2017 ARC pdfs posted at NARA. The FBI record 124-10201-10039 was not actually posted online, even though there is a link to it. To see how this happened, read on!
If you want the record, and you don’t care how the FBI and/or NARA screwed up, you can find a copy released in full at MFF here.
The phantom in the excel sheet
The FBI record we are looking at, 124-10201-10039. was supposedly released on 2017-12-15, and appears on NARA’s spreadsheet for the 2017-2018 ARC releases here.
This spreadsheet gives ARC record numbers, FBI file numbers (along with the usual record metadata), and links to pdfs on NARA’s web-server. The usual filename for an ARC pdf is the ARC record number, but in this case, NARA has used a different filename for ARC 124-10201-10039. It uses the name “docid-32289264.pdf”.
The problem is that this same excel sheet also links to “docid-32289264.pdf” for ANOTHER record: ARC 124-10201-10040 (see here). How is this possible?
Perhaps these two records are just duplicates? Let’s check the JFK database metadata for these two records to see. Just enter the ARC record numbers in the MFF JFK Database Explorer here. This shows us that these are two distinct records:
1) 124-10201-10039 has the FBI file number 92-3182-42
2) 124-10201-10040 has the FBI file number 92-3182-43
So these are two sequential serials (documents) in one FBI case file (the file on Chicago mobster Gus Alex). That means they are different records.
Looking at docid-32289264.pdf, we see that this is FBI file number 92-3182-43. An identical file was released at NARA on 2017-11-17. This earlier release was marked on NARA’s spread sheet as just plain 124-10201-10040. So for sure this pdf, docid-32289264.pdf, is ARC 124-10201-10040. That means the excel sheet entry for ARC 124-10201-10039 in December is an error, giving us a link to a record that is not actually there: a phantom.
As a further sign that there is an error, the NARA excel sheet metadata for 124-10201-10040 is complete: all the fields in that row are filled out. The fields in the row for 124-10201-10039, however, are almost all blank. Clearly the FBI has somehow messed up here.
This error propagated from NARA to the MFF website, which has links for the 12-15 release of both 124-10201-10039 and 124-10201-10040. Click on 124-10201-10040 and you get 124-10201-10040. Click on 124-10201-10039 and you get — 124-10201-10040. This is basically NARA’s fault, not MFF’s fault.
The real story
Both of these FBI records were previously released. MFF has the earlier releases and some checking will show you what the story is.
ARC 124-10201-10039 had TWO earlier releases, available at MFF here and here. It turns out this record is an airtel (a kind of telegram) from the Chicago SAC to DIRECTOR FBI. The first version of the pdf santized (blacked out) the source of the information in the airtel. This version of the record was released in 1994. The second pdf restores the name of the source. The second version was released in 1998.
Why two different versions? Because in 1994 the ARRB, the federal board reviewing all these records for release, was not yet up and running. The FBI therefore followed its usual FOIA practice of redacting all source information. Once the ARRB was up and running, they told the FBI to cut that out and release source info for people who weren’t going to get their throats cut for squealing. (The source was a travel agent who booked a hotel for Gus Alex in Montana.)
ARC 124-10201-10040 also had two earlier releases, available at MFF here and here. These are both sanitized as well. The reason for sanitizing is that they deal with the FBI’s attempts to track Gus Alex’s travels in Europe, so they include mentions of FBI liaison with Austria and Switzerland. Liaison information is not supposed to be released, and the ARRB did indeed allow the FBI to hold back this information. So although there were two releases in 1994 and 1998, neither one released the liaison info. That had to wait for 2017.
TWO pdfs of ARC 124-10201-10040 were posted at NARA in 2017, as mentioned above: in November and December. The November pdf released the liaison info from record in full. The December release was thus unnecessary, but this sort of duplication often occurred in the 2017-2018. This is one of the reasons it was so hard to figure out what was released and what was not.
Two cents
How is this phantom record related to the problems of multi-record pdfs? The confusion here was due to using a “docid-xxxx” filename for the pdf, instead of using an ARC record number. Don’t do that, and you will have no problems. This problem comes back to haunt multi-record FBI pdfs with a vengeance, as we will see in an upcoming note.