Tuesday, October 18, 2011

LTG Michael Flynn IC Reform Challenge

SAN ANTONIO, Oct 18, 2011

Newly promoted LTG Mike Flynn USA, famous for his Fixing Intel paper, continued to provoke the IC today at the GEOINT2011 Conference. Flynn's opening statement was a clear shot across the bow of the techno-centric GEOINT crowd: "the major barriers to INT integration are cultural, not technological" he said. LTG Flynn spoke of the world being in a perpetual war with a related requirement for perpetual high quality intelligence, not just in combat but in anticipation of future crises. He went on to highlight one of the biggest issues facing the IC - how to get useful predictive strategic analysis in Phase 0 - in other words, before getting involved in international disputes in the first place. Notably no one had a response to his challenge. Questions by MIL INT on this issue were not selected for discussion. (See also the Integrating Intel Panel with future G2 MG (P) Legere).

This is significant because US INT is strong at collection, experiences problems in distribution and integration (the conference focus) and is often patchy at analysis. How we improve all of these categories in a coherent whole is a key challenge and it really isnt about technology - although it is supported by technology. But perhaps GEOINT was not the best place to have that debate. Which raises the question - where might that debate take place? What industry is going to fund a 5 star gala event like GEOINT to talk about improving analytic standards and predictive analysis?

Because of this, MIL INT could not help but feel LTG Flynn was a bit of a fish out of water. The room was packed when the IC leaders were talking budget cuts yesterday, but was notably much lighter during LTG Flynn's panel today. The problem Flynn identified can't be fixed with a rapidly deployable COTS software package or The Cloud. It if were only that easy. We are not in the find, fix, finish world anymore, we are in the individually tailored intelligence world (look out for a near future blog on crowdsourcing a fascinating subject here at GEOINT) where how we think about global events and create the context for decision making is more important than which pipe a piece of the mosaic is delivered through. 

GEOINT 2011 is a major industry conference and the IT sector is heavily represented. One of the trends of the conference is the emphasis placed on getting INT down to the  tactical level. Typically this is represented as a bandwidth, IT systems, or classification access challenge. These are critical challenges and a number of speakers from DNI Clapper down have stressed that the extraordinary proliferation of stovepipes and contractors serving those pipes, has to be urgently cut. Whether Clapper can deliver a 50% saving over the entire national and military INT enterprises is another matter. But there is little doubt that information management is in crisis, so any efforts to simplify and resolve those problems is a very important task. The many senior IC leaders here in San Antonio should be applauded for their focus on this problem set.

LTG Flynn is the Assistant Director of National Intelligence for Partnership Engagement at ODNI. DNI Clapper first mentioned Flynn was going to join his team back in November of last year. Clapper was obviously impressed with Flynn's Fixing Intel report offering Flynn this challenge: “So, hey buddy, you can come back here and help me fix it.” Flynn will act as a “kind of the bully pulpit for collaboration and sharing,” Clapper said. LTG Flynn did not fail to deliver today, noting that his top three priorities will be:

1. Military liaison - moving around the Joint Forces deployed and CONUS ensuring they are getting what they need from the IC - a laudable initiative.
2. Domestic liaison - essentially the same focus for law enforcement and related users of IC products CONUS.
3. Making sure the IT reforms highlighted by Clapper will deliver results with partners at home and internationally - something about which he is clearly passionate. "We will only win wars with partners, we will not win unilaterally" Flynn said today, "get over it".

Flynn's frustrations surfaced again when he spoke about the challenges of military leadership and working in an inflexible bureaucracy. "More people say no the further you go up the chain of command". He said "we need leaders to say yes and figure out how to do things" rather than taking the easy path and "avoiding work" by just saying no. Clearly LTG Flynn is going to have an interesting tour at ODNI!

Brigadier NR Davies from the UK, one of Flynn's fellow panelists, had some interesting things to say about intelligence sharing. Clearly a capable thinker, Davies key take away was all coalition partners should drop nationally classified information and instead swtich to operation based intelligence classifications. Thus US NOFORN would pass into history to be replaced by TS OPERATION XYZ - where access is decided on the basis of involvement in a mission or operation vice which nation was supplying the intelligence. His Canadian and Australian counterparts nodded approvingly. Its a great idea but the chances of it happening are very slim... I'd like to see LTG Flynn say "Yes" and see how he could make it happen! 

MG Vance from Canada was very frank on a range of issues. MIL INT asked the coalition partners how dependent they were on US INT and whether their own agencies provided better analysis or support. MG Vance had clearly got "opinions" from the Canadian intelligence system with which he disagreed. His intel folks "were off the mark" on a number of issues in Afghanistan. He did stress the role of coalition partners to validate or challenge assumptions and analysis. LTG Flynn was adamant in rejecting reach back intelligence. He said "reach back does not work, forward deployed SMEs do" but they must have access to the top leaders and they must "be in the same time zone as the combat zone". All of the partners, mostly by ommission, implied that they were quite dependent on US INT (except for their troops supporting fusion centers etc). Only the Australian, Brig Collin Gillian, noted that while Australia was quite dependent on the US, Australia returned the favor with a degree of expertize on particularly SE Asia that the US "just does not have" according to Gillian. But what he means here, again, is analytical expertize - not technical expertize and in this he is completely right. It will be interesting to see Australian intelligence enhance its capacity to deliver on the Indo-Pacific AO construct discussed by Secretary Clinton in her comments about valued Australian counsel on the region. 


MIL INT asked the panel a number of questions. Those that were not addressed are listed below. This is not a criticism, this was not a 1:1 interview with the panel, but it is an indicator of the panel's thinking. Often what is not said is as interesting as what is said.

1. Will your COIN INT design work for major conventional war in maritime Asia? Or will we just drift back to Red INT focus?
2. Much of the conference has focused on technology and mostly at the tactical level. The big missing piece seems to be strategic predictive analysis. A computer can't do that for you. How are you improving your INT enterprises to enhance analysis and engage in predictive strategic analysis support to national decision makers? 
3. The target set seems to be IEDs and HVIs - what else should be targetted?

If any of the panel members would like to write a response to these questions or alternatively write a guest column for publication by MIL INT, they would be most welcome to do so.

A great panel and MIL INT wishes LTG Flynn well in his important new endeavor.


  1. Good insights. I can say that quite a bit has been done to look into improving analysis throughout the IC and within the USMC in particular. That said, a major focus is looking at improving "predictive analysis" which I think is a laudible goal, but necessarily achievable (us humans just aren't good at predicting things).

    As far as technology goes...well, some of the things we can do are absolutely amazing. One hopes that we don't rely so much on these systems that we can't function if they are taken away. To me, this implies that we need just as heavy focus on people. Perhaps we can address these predictive issues with a greater regionalization focus and less a jack of all trades way of doing things. Conversely, there is something to be said about being a generalist as well.

  2. There are two issues at play here: policy and military intelligence culture. Gaining intelligence in phase 0, regardless of the location on the globe, requires shifts in policy to permit the implementation of ISR assets to collect in those regions out of the declared theatres of active armed conflict. Today's USG mechanisms to change policy are overly bureacratic, making it difficult to garner the necessary permissions to collect in a timely manner.

    Culturally, there are two aspects that prevent the application of predictive analysis. First, the larger intelligence community is still fixated on the antiquated methodology of "static" intel analysis - looking at the enemy's order of battle by counting tanks, positions, ships etc; instead of applying a network analysis methodolgy ("COIN INT"). By incorporating traditional analysis into a network model we can expect to have better predictive awareness. Just because you are looking at a N Korea or Iran doesn't mean you cant apply a network analysis model to assess the EOOB.

    Secondly, MILINT analysts ask "what did the enemy do yesterday or today" and less on making those predicitve assessments ..."what will the enemy do tomorrow?" This a larger military culture issue; predictions carry risk and bias. In order to better utilize intelligence to "predict" the future, the military must change it culture from looking predominately at "lagging" information and spend more time on making those future assessments.

  3. Ultimately General Flynn's piece is inconsequential - great for talk, little to actually show. Unfortunately this is part of a greater trend of misplaced logic and poor reasoning. To elaborate, how much time, effort, and money is being spent pushing female engagement teams into combat? The assumption is that we are clearly missing 50% of the population and therefore if we only had females we would gain the upper hand. This is a gross oversimplification on both inductive and deductive fronts. But, it was mentioned in Fixing Intel so it must be correct...

    For the comment on October 22nd: intelligence is a requirements driven process, nothing more. If the requirements are not thought out and vague, so too is the analysis. A bit more time fleshing out what the problem is and how best to approach it would go a long way in helping keep intelligence relevant. Thus, DA SOF missions are relatively easy - find the bad dudes and kill/capture. On the other hand, everyone else is told to collect on the price of bread at the market or find out who local power brokers are...just a wee bit more difficult I say. To me, this rests on the commander and staff to clearly articulate what the mission is, what the objectives are, and what he hopes to achieve. Framing the initial problem(s) is critical - otherwise all the future assessments in the world will not matter.