This is a depiction of what a kill-list "baseball card" looks like:
[caption id="attachment_14245" align="aligncenter" width="519"] This is a mock-up/depiction of what an actual kill-list baseball card looks like. It is not a real baseball card and is not based on any classified information.[/caption]
What is shown above is a mock-up of what one of the last documents the President of the United States or his designee (a.k.a. strike approval authority) will likely see before signing off on the addition of a name to a kill-list. (It's also the last thing a strike approval authority may see in a pre-planned strike). If you're on Capitol Hill and you want to know the when, where, and why of a particular targeted killing, a document like this is the place to start (note, that image is a mock-up, it's not the real thing).
As I mentioned in my previous post, a lot of analysis goes into adding a name to a kill-list, thus the baseball card is the end of a long process. It is an executive summary of intelligence information and analysis, provided to decision makers who are sufficiently down the line in the targeting process that they can rely on the vetting and validation that preceded their decision, acting on the assumption that the process weeded out questionable target nominations. Those decision makers always have the authority and obligation to ask for more information should questions arise (although as I'll discuss below, they may not do so). So, the baseball card is merely what comes at the end of the process of documentation and analysis, therefore to understand the kill-list process and how to hold persons in it accountable, we need to know something about what happens before a baseball card is presented to a decision-maker and how that predicate information is documented.
TARGET FOLDERS AND THE DISPOSITION MATRIX
As yesterday's posts highlighted, there are multiple incremental questions that must be asked to determine whether an individual can be lawfully targeted, and whether that lawful target should be added to a kill-list. That process involves the creation of an extensive paper and electronic trail. As analysts develop lists of targets, they create target folders containing detailed analysis on the target (intelligence reports, modeling, simulation products). Bureaucrats identify requirements for additional intelligence, they file requests for legal opinions or have legal counsel sign off on lists, and they create briefing materials about targets, uploading this information to secure databases. The process is formalized and documented, which lends itself to various mechanisms of accountability.
In current practice, the analytical steps I have described are documented in target folders, those folders are part of the process for creating kill-lists and the information in them is available right up through the execution of a strike. The folders contain target information, such as data about how the target was validated, who approved the target and at what step in the process, along with any identified potential collateral damage concerns associated with the target. Contrary to the claims of critics who worry about stale or out of date intelligence, target folders are continuously updated to reflect the most recent information regarding a target's status and the compiled data is independently reviewed by personnel not responsible for its collection. The independent review is designed to ensure mistakes do not proliferate throughout the targeting process. Across government, the targeting folders have now been reduced to a database, and the information is now maintained in Electronic Targeting Folders (ETF)’s within that database. According to some reports, the database has evolved and is colloquially referred to as the “disposition matrix.” The information in a related set of procedures, sometimes referred to as “the playbook” explains how to proceed in selecting and targeting suspects drawn from the disposition matrix. The disposition matrix seems to be a multidimensional database that goes well beyond just names and corresponding intelligence to include many of the variables discussed in my prior posts.
While the system is still “a database in development” it represents a tool that allows analysts to consider additional factors beyond merely adding a name to a kill-list, however it also allows for tracking of decisions through the bureaucracy, meaning that the possibility of accountability exists. The ETF’s contain a record of the approvals, changes in intelligence, collateral concerns, anticipated benefits of attacking the target, and other information as it becomes available. That information includes human intelligence reports referencing the target, signals intelligence referencing the target, imagery and floor plans of likely locations of the target, a diagram showing the social and communications links of the target as derived from human and signals intelligence, and previous operations against the target. Also documented are intelligence gaps that will form the basis of additional intelligence requirements. Analysts who identify needs for more information can request additional pieces of information that they believe are needed to complete target development, and those requests will also be documented.
VETTING AND VALIDATING TARGETS
The United States government has developed a formal vetting process which allows members of agencies from across the government to comment on the validity of the target intelligence and any concerns related to targeting an individual. At a minimum, the vetting considers the following factors: target identification, significance, collateral damage estimates, location issues, impact on the enemy, environmental concerns, intelligence gain/loss concerns, and issues of legality. An important part of the analysis also includes assessing the impact of not conducting operations against the target. Vetting occurs at multiple points in the kill-list creation process, as targets are progressively refined within particular agencies and at interagency meetings.
A validation step follows the vetting step, it is intended to ensure that all proposed targets meet the objectives and criteria outlined in strategic guidance. The term strategic is a reference to national level objectives --the assessment is not just whether the strike will succeed tactically (i.e. will it eliminate the targeted individual) but also asks whether the strike will advance broader national policy goals. Accordingly, at this stage there is also a reassessment of whether the killing will comport with domestic legal authorities such as the AUMF or a particular covert action finding. At this stage, participants will also resolve whether the agency that will be tasked with the strike has the authority to do so. Individuals participating at this stage focus their analysis on a mix of military, political, diplomatic, informational, and economic consequences that flow from killing an individual. Other questions addressed at this stage are whether killing an individual will comply with the law of armed conflict, and rules of engagement (including theater specific rules of engagement).
Importantly, many of the concerns that critics say should be weighed in the targeted killing process in fact are considered prior to nominating a target for inclusion on a kill-list. For example, analysts in the kill-list development process will weigh whether striking a particular individual will improve world standing and whether the strike is worth it in terms of weakening the adversary's power. They will analyze the possibility a strike will adversely affect diplomatic relations, and they will consider whether there would be an intelligence loss that outweighs the value of the target. During this process, the intelligence community may also make an estimate regarding the likely success of achieving objectives (e.g. degraded enemy leadership, diminished capacity to conduct certain types of attacks, etc.) associated with the strike. Importantly, they will also consider the risk of blowback (e.g. creating more terrorists as a result of the killing, contributing to enemy recruitment efforts, etc.).
VOTING ON TARGETS
Following the vetting and validation of targets, participants in the kill-list creation process also have an opportunity to vote on whether a name should be nominated for inclusion on a kill-list. In the Obama administration, Dan Klaidman, Jo Becker and Scott Shane have reported that this this debate and voting process was colloquially referred to as Terror Tuesdays. (Interestingly, during the Vietnam War targets were approved by the President at what become known as Tuesday luncheons, it seems Tuesdays are a bad day for America’s enemies). At this stage, information from the ETF’s is reduced to more manageable summaries of information -- the baseball cards. Those baseball cards differ by agency, but they generally look like the image depicted above. They are Powerpoint slides that display a color picture of the target and physical characteristics (such as height and weight.) The slide lists information such as the individual’s rank in the organization, professional expertise, family ties and links to individual attacks. Also included is specific intelligence to support the individual’s nomination with an explanation of the source of the intelligence. Other data may include a map of the area where the target has been operating, a personal history of the target, patterns of life for the target, cell phone number of the target and even what vehicle the target is known to travel in. If the person reviewing a baseball card wants more information, they can dig into the ETF to see what intelligence supports the information on the baseball card.
The formal process has constantly evolved throughout the Bush and Obama administrations however the expectation that senior bureaucrats will vote on whether names should be formally approved for killing has remained an enduring feature. Participants in the process may vote to concur, concur with comment, nonconcur or abstain from voting. Unanimity is not always required, rather abstentions, nonconcurrence and concurrence with comments are indicators of greater operational and strategic risk which the President or other approval authority will take note of when reviewing the target for final approval.
As this post has begun to highlight, the kill-list creation process rests on multiple incremental steps of analysis that are documented in various ways. Dozens of people have their hands in the process, which has significant implications. On the one hand, many hands means many eyes, and impropriety may be discovered or reported by whistleblowers. However, many hands also means distributed responsibility that may allow responsible persons to avoid accountability by claiming to have relied on the information and analysis conducted by others.
In my first post I explained how law creates categories of targets, and how bureaucrats begin to create lists of targets. In my second post I explained how network analysis contributes to the kill-list creation process. This post described the extensive paper and electronic trail of approvals and intelligence associated with the kill-list creation process. My next post will focus on who is likely involved in the voting process, and the implications of what I've described for the accountability debate.
(As in previous posts, I have omitted internal references, they can be found in the paper once it is posted).
This post is Part 3 of a 7 Part series based on the article Kill-Lists and Accountability
- Post 1: How to Make a Kill-List
- Post 2: Kill-Lists and Network Analysis
- Post 3: Kill-List Baseball Cards and the Targeting Paper Trail
- Post 4: The Politics of Accountability for Targeted Killings
- Post 5: Presidential Politics, International Affairs and (a bit on) Pakistani Sovereignty
- Post 6: It’s Time for a White Paper on Congressional Oversight of Targeted Killings
- Post 7: Five Ways to Reform the Targeted Killing Program