A note on the Manafort texts

This cache of Manafort texts (links at the bottom) is made up of more than 285,000 SMS messages and iMessages (over 7,200 pages) sent and received by one of Paul Manafort’s daughters. Several years ago, the phone was hacked and a database of the messages became available in certain circles. Some outlets have previously reported on some of these messages, such as ones implicating Paul Manafort in the commission of murders, but other messages have gone largely overlooked or have been available without the full context. This is largely due to one thing:

The public haven’t had access to the messages, and the press (including high profile and high budget outlets) haven’t had access to an easily searchable version – instead only to an unwieldy database.

This was nearly not the case, however – WikiLeaks had a copy of the database, but ultimately didn’t publish it, despite its newsworthiness and their willingness to publish unredacted information on the Democrats. Their unexplained decision not to put the database into a searchable format and make it public struck me as questionable at the time, but in light of the disclosure of their preference for the GOP it has become not only questionable, but hypocritical. WikiLeaks’ decision can no longer be trusted prima facie or viewed in terms of presumed good faith.

[WikiLeaks’ allegation that the hack was performed by Ukrainian intelligence should be taken with a grain of salt]

This noteworthy refusal, along with ongoing probes and charges filed against Manafort by the Special Counsel’s office, makes the text messages and their contents undoubtedly newsworthy. Their relevance to the general public may go beyond this, as the personal reality reflected in the messages presents the Manaforts as real people, rather than merely as abstract figures. This same reality is, of course, also relevant to news and current events, and to understanding the character and actions of those involved.

The decision to not redact the text messages was made only after careful consideration and for several reasons, including but not limited to:

  1. The data has already been exposed, and the damage done.
  2. Those involved know that the messages were hacked, and that their phone numbers and email addresses (in the case of some iMessages) have been exposed. They’ve had over a year to change their numbers and take steps to block harassment. Therefore, any harm in this regard is minimal.
  3. Tabloids and trolls have already mined the transcripts and exploited them. While this release may result in a small resurgence of this, the real damage has long been done and mitigated.
  4. The hack was confirmed and some of the messages verified by the Manaforts. While information in the texts may be challenged by various parties, the validity of the texts does not appear to be in doubt.
  5. While some personal details may seem lurid, they are ultimately meaningless. People have bodies which sometimes itch, people have sex and personal lives. The instances that are meaningless will be quickly forgotten (as they already have been by the trolls who were once determined enough to dig through nearly 300,000 messages), while significant pieces of data will be found and preserved. In some instances, the personal messages shine a light on things of more significance (see below).
  6. It’s important for people to have as much information as possible, and it’s not always clear ahead of time what will have significance. For instance, the existence of a personal family issue on a certain date may be correlated to movements and behavior of some of those involved, or a family trip to Trump Tower may be correlated to an significant date.
  7. The participants and subjects of these messages are not not random individual or low level political workers, but individuals, many of them rather wealthy and powerful, who have ties to (or are) someone accused of significant crimes, someone with close ties the White House and politicians around the world.

The database was exported from the original .db file to a series of .csv files. Each of these files is available along with the original database. The .csv files were combined into a single optimized file, that removes extraneous information and presents the messages in chronological order. The handle IDs from the original database have been converted into phone numbers and email addresses, and the Apple Cocoa Core Data timestamps converted into human readable dates. Data on whether the messages were sent or received has also been extracted and incorporated.

In some instances, the field for the exact date and who sent the message was absent in the original database (unsurprisingly). In these instances, a “Data Missing” note is inserted in its place. The order of the text messages is preserved by the database itself even when the date field is missing. In the event of conflicting or corrupted data, the canonical order of messages as stored on the phone (and thus presumably the order in which they were sent/received) is the order they are listed in.

While the .pdf, .txt and .csv versions of the transcript are all searchable, the Optimized.csv version (27MB) may be most useful due to the cell structure allowing additional ways of sorting and filtering the information.

The optimized transcript can be found as part of a searchable wiki of leaked and hacked material on or about WikiLeaks here. The complete set of files, including the original database, can be found here. The database in a compressed format (32MB) can be found here and the original database (208MB) can be found here. The .txt transcript (25MB) can be found here, and the .pdf can be found here (35MB). The database extracted into .csv files can be found here. The complete set is available in a single .zip file here, or as a torrent here (magnet link).