National Security and the Islamic State: Foreign and Domestic

National Security and the Islamic State: Foreign and Domestic
(Dr. John M. Poindexter, February 3, 2016)

Transcript available below

Watch his speaker playlist here

About the speaker

Dr. John M. Poindexter most recently served as Director of the Information Awareness Office (IAO) at the Defense Advanced Research Projects Agency (DARPA).

Dr. Poindexter served as National Security Advisor and Deputy National Security Advisor for President Ronald Reagan from 1983 to 1986, and as Military Assistant in the White House prior to that. Dr. Poindexter served 29 years active duty in the U.S. Navy, rising to the rank of Vice Admiral. He is a graduate of the U.S. Naval Academy.


Robert R. Reilly:

Since we have someone from the Navy speaking tonight, we are going to run a tight ship and start on time. Although, since I know there are some other people who are coming, I was tempted to ask the short people – there is one of them – to ask the short people to sit in front so that people in the back could see the slides. Does it look okay to everyone from where you are you can see the slides? Okay, alright. Excellent.

Well, our speaker tonight really does not need an introduction, but I will make a brief one, nonetheless. As you know, Admiral Poindexter served as National Security Advisor and Deputy National Security Advisor for President Ronald Reagan – remember? – from 1983 to 1986, and as military assistant in the White House prior to that. He was also responsible for improving the command and management systems to support the president and national security crisis management, culminating in the creation of a new high technology crisis management center.

We are also happy to have Mrs. Poindexter with us tonight.

Linda Poindexter:

The late Mrs. Poindexter.

Robert R. Reilly:

No, no, not the late Mrs. Poindexter, you are very much alive.

Admiral Poindexter served for 29 years, active duty in the U.S. Navy, rising to the rank of Vice Admiral. While in the Navy, he specialized in training, new tactics, and battle management procedures, and pioneering use of chip board computers. That is how he is able to do the slide presentation. Something beyond my competence. He holds doctoral and master’s degrees in physics from the California Institute of Technology. He is also a graduate of the U.S. Naval Academy.

Now, most recently Dr. Poindexter served as a director of the Informational Awareness Office at DARPA, the Defense Advanced Research Projects Agency where he developed and demonstrated information technologies systems to counter asymmetric threats by achieving total information awareness useful for preemption, national security, and national security decision making. Please join me in welcoming Admiral Poindexter.

Admiral John Poindexter:

Bob, thank you very much. Good evening, ladies and gentlemen. As Bob so stated, I am not an expert on Islam, and I am a retired naval officer and scientist. But Bob asked me several weeks ago to come and speak and I asked him what he wanted me to talk about and he did not have an exact specification, but I told him what I wanted to talk about. So what I want to talk to you about tonight is national security and how it relates to the Islamic State and what we can do about it from a foreign perspective and the domestic perspective.

And I want to compare it with what we did in the Reagan Administration. Our strategy in those years was based on five pillars: diplomatic, military, economic, public diplomacy – which, by the way, Bob is very big on – and covert operations. It is very important to understand that an integrated strategy is essential because each action in each of the areas provides leverage to actions in the other areas and I strongly believe that we should be using this kind of strategy in the present situation with the Islamic State.

Now, it is true that the administration is maybe doing pieces of this, but it has not been integrated in any meaningful way and explained not only to the domestic audience, but also to our foreign audience. And I think that is one of the reasons that we are in a state right now of great uncertainty.

Geography of the Middle East

Many of you may be aware of the next few slides, but for those that are not, I want to talk a little bit about geography and the geopolitical situation, the geopolitical history. At the end of World War I, British and French diplomats, Sykes and Picot, came up with a scheme for dividing the old Ottoman Empire. It was primarily based on at the time on British and French interests. It was arbitrary in terms of the tribal and ethnic culture of the Middle East. In other words, the Sunni, and the Shia, and the Kurds. In hindsight, it is my belief that that was a huge mistake and we are suffering through the result of that right now.

The area involved is at present time anyway largely desert. Here is a Google Earth image of Syria and Iraq. And note that there are two major rivers that cut through this area, the Euphrates and Tigris Rivers. But all of this area here is essentially desert.

Current Operational Picture

Now, the current operational picture in Iraq is represented here. Note that the yellow dots are the Kurds, the red dots are the Iraqi forces, and the black dots are the Islamic State. Also note that all of the activity occurs basically along the rivers, and, of course, the reason is that is where the people are.

Now moving on to Kurdistan. This is the northern part of Iraq. And again, the scheme is the yellow are the Kurdish forces, the red are the Iraqi forces, and the black dots are the Islamic State. And, of course, Mosul is a very large city presently occupied by the Islamic State.

Then in Syria the Islamic State is this grey-ish area. The Kurds are up here to the north, this area over here. The rebels are in these green areas. And the government controls the red areas. Now one of the things to note here is that this is a border with Turkey, and this area is presently occupied by the Islamic State.

Now, recently there have been reports that the Turks and the Kurds are working to close off this border area, which would be very helpful for the prosecution of the war.

Islamic State Control Expanding

Now the Islamic State control is expanding. It is shown here in December of 2014. And recall they announced the Islamic State in June of 2014. So the yellow-ish color here represents where the Islamic State forces were in December and this red represents the expansion that occurred by May of last year. Now, this slide is not exactly up-to-date because recently, as you heard there have been some advances of the Iraqi forces into Ramadi and in this area here.

Alright, now, this is the published Islamic State 5-Year Plan. Now, recall as I mentioned, the Islamic State is formed in June of 2014, and their stated goal is that the Islamic State will expand within five years to its natural, caliphal borders, in other words the black areas. This is the original Arabic and the English translation here.

Now, granted, one of my colleagues says that this is ridiculous, that they will never do that, and granted, it is a little over the top in terms of what they say they plan to do in five years. However, this kind of published plan within the Islamic State forces becomes a great aid in recruiting people to come from foreign countries into the area to participate in the battle. Just recently, there was a report of a radio calling themselves the Voice of the Caliphate, broadcasting in Afghanistan, and talking about Khurasan, so it is not out of the question. And it is certainly something that we should take seriously, that this is their stated objective. We should take that seriously and anticipate having to resist that movement.

Terror Battlefield Spreads

Here is from an article in The Wall Street Journal in January of this year, and it shows how the terror battlefield has spread, and the black dots here are major events, terror activity. Of course, San Bernardino here, the destruction of the Soviet/Russian aircraft [is] here, but the thing I want to point out is this bottom line, showing how the number of fatalities, which are represented by the size of the circles – the number of fatalities and the frequency of events have been increasing over the past few months.

Now, President Obama has said that the Islamic State is not an existential threat to the United States, but I would argue that it all depends on your definition of existential. Certainly, to the people who are murdered it is an existential threat, and we cannot ignore it, and cannot imply that we are willing to accept other San Bernardinos.

We are at War

So, in my opinion we are at war, and we ought to recognize that. The war is not against Islam, but it is against the Islamic State that holds an old interpretation of Islam, and I think this is a generally accepted definition these days, but Islamism is carried out by the Islamists. This interpretation of Islam, as Bob has described in great detail based on very significant research, this interpretation of the Qur’an that the Islamists are using was radical when it first was proposed centuries ago, and it is certainly radical in terms of the modern world, but it is supported in some of the Qur’anic passages, and that is one of the problems.

And in my opinion, it is an existential threat to the free world way of life. You know they may not be able to destroy the United States physically, but they can certainly have an impact on the way of life in the United States and the free world, and so I consider that an existential threat. The solution, since we are at war, takes a whole-of-government approach to execute. It is not just the Defense Department or the State Department, but it takes all of the elements of the U.S. government and the governments of our allies to prosecute this war.

Diplomatic Actions

So now I want to talk about the areas that I think need to be integrated, and I will start with the diplomatic actions. I think we should formally declare war on the Islamic State. The Islamic State is a little different than Al Qaeda and other terrorist groups in that they have proclaimed that they are a state, so we can hold them accountable. They hold territory. They are providing some social services, besides all the atrocities they commit, and they have a government of sorts. They have an organization chart, and so in in almost all respects, it is a prototype state, and I think that we should formally declare war on it.

Now, the problem is – I hope you are all familiar with the guns and butter concept, and it is my opinion that guns and butter does not work. In fact, I think that is one of the mistakes that President Johnson made in the Vietnam War. I think it is a mistake that we made in the Vietnam War, and I think it is mistake that we made in the wars in Iraq. The problem is that if the American people do not understand that we are at war, then it is going to take sacrifices. They soon tire of the effort, and the Congress usually capitulates, and we wind up starting these wars but not being able to win them in the end.

So I think in my opinion, that a formal declaration of war is a way to address this issue. It brings Congress into the act, and I think that is essential, and it makes it very clear to the domestic audience and the foreign audience that the United States is serious about this and is going to pursue it as an all-out war in order to defeat the forces of the Islamic State.

I have concentrated so far tonight on Iraq and Syria, but obviously, it applies to other areas as well. [For] example, Libya. As you may have noted on the previous Wall Street Journal graph, there have been numerous what I would call terrorist activities in Libya. The Libyan government is in shambles right now, and it becomes a very likely state to go over to the Islamic State.

I think we need to quit dancing around the issue. Leading from behind in my opinion does not work, and it only results in international chaos. The problem is that our allies, especially in the Arab world, if they see the United States, the most powerful country in the world, not out in front and leading the effort, they see no reason that they should put their necks on the line and help.

I think there is an international coalition right now, and I do not think this president is actively leading that coalition. I think we should. We need to make it clear that the Muslims must help in solving the problem. Their countries after all are at risk, and in some of these countries it probably is a physical, existential threat to those countries.

We need to put pressure on Saudi Arabia to stop support for radical elements and modernize their religious interpretation. We need to put extreme pressure on Saudi Arabia, Jordan, Egypt, Pakistan, and Turkey to provide ground troops. I do not think we ought to use large numbers of U.S. troops on the ground, and that is not from the standpoint of trying to minimize U.S. casualties, but it makes sense from the standpoint that if you have Muslim troops on the ground, it takes away [credibility from] one of the Islamic State statements that it is another crusade with the West coming in to defeat Islam.

And since these Muslim countries are threatened, they should be the ones providing the bulk of the ground forces. And I think the reasons they have not been willing to do that is they do not see the United States as serious and they do not see the United States taking a major lead in the operations of the coalition.

I think we also should support Kurdistan. They are some of the most effective fighters in the area, and Iraq and Turkey are both opposed to providing significant support to the Kurds, but I think we should tell Turkey and Iraq that it is for their benefit and we have got to bring the Kurds on board in a much more effective way.

Now, last on the diplomatic list is that we have got to develop a plan with a coalition for governance of the areas recovered, and I would propose – this is my personal opinion – that we need to start thinking about revisiting the Sykes-Picot boundaries that were set up after the end of the First World War, and we obviously cannot do that by ourselves, but we should work with the international community to revisit these boundaries.

Military Actions

So military actions: we are doing some of these things today, but there is an awful lot of micromanagement going on, and we have got some rules of engagement that are absolutely ridiculous in my opinion. So I listed here all of the things that I think the U.S. should do. Provide intelligence to the coalition, including overhead imagery. Provide AWACS aircraft for support to provide a good tactical ground picture. Of course, the overhead imagery is largely strategic. There are some tactical aspects of it, but you cannot station the satellites permanently over some areas and ignore others, so AWACS provides good support for the ground tactic picture.

Provide drone support under the control of the local commanders. Do not micromanage. Today, much of the drone activity is controlled at the White House level, which in my opinion is absolutely ridiculous. We need to continue providing tactical air support from our carriers and Middle East land bases, but we need to increase the sortie rate to support the ground operations.

Now, one of the reasons they claim that the sortie rate is so low is they do not have good target identification on the ground, but to solve that problem, we have got to put more of our people over there, especially for targeting and as military advisors in the field. We have got to provide and operate regional command and control centers for the coalition with authority to conduct the war, again, without micromanagement from Washington.

We need to provide logistics support and financial resources for all of the coalition forces as they need it, and as I mentioned briefly earlier, we need to get serious in changing rules of engagement, and expect some collateral damage. After all, if my prescription is followed, we would be at war, and let me give you an example. I understand that our aircraft were not allowed to strike oil tankers that were moving Islamic State oil out of the oil fields that they have control over because it was thought that the drivers might be civilian, and they would claim civilian casualties. [That is an] absolutely, in my opinion, mindless restriction.

We need to provide the Kurds with effective weapons and supplies, and we are not doing that today directly, although I understand that there have been some better weapons provided at our request from a NATO ally, but of course, the problem here are the Turks and the Iraqis that do not particularly want us to support the Kurds, but as I said earlier, we just have to tell Iraq and Turkey that this is what we are going to do and it will help in ending this terrible scourge.

We need to pressure NATO to participate. The European Union right now is struggling with the migration problem, which threatens a rupture. Europe not only has an economic problem, but they have this huge migration problem, and of course, one of the reasons is that the border between Syria and Turkey is open right now.

Economic Actions

Okay, next, economic actions: we need to attack the Islamic State-controlled oil infrastructure in a serious way. We did stop oil smuggling through Turkey and elsewhere, and [this is] another good reason to close this border between Turkey and Syria.

Stop foreign payments to the Islamic State, and there is some evidence, apparently, that Saudi Arabia is continuing to – not necessarily as a government, but the Saudi Arabian people are providing financial support to the Islamic State. We need to identify their accounts and seize them, and lastly, but most importantly, probably, as the coalition recovers areas, we have got to be willing to provide economic support for development and social services. It is going to be expensive, but it is absolutely necessary for stability.

Public Diplomacy

Public diplomacy is a largely forgotten part of one of the pillars that our strategy should contain. Right now, U.S. public diplomacy is in a shambles. It is controlled by a Board of Governors. The Chairman is a Hollywood executive. The CEO is a TV entertainment executive. And I am sure they are great people, and they probably are very good at what they do in Hollywood and entertainment on TV, but in my opinion, and that of a lot of others, including my good friend, Bob, back there, they do not understand what is required for a strategic public diplomacy program. One of the techniques that one of these entertainers have promoted is taking MTV videos, and broadcasting that to the Muslim world.

Anyway, jumping down to this, the ‘US Secular Culture,’ our strategic public diplomacy program needs to be inclusive and it needs to be representative of the description of our culture. It is not just the extremes on both coasts in the big cities. They nave got to include the fly-over country, what the liberals call the “fly-over country,” and in my opinion, the extremes of our secular culture right now are great Islamist recruiting aids. MTVs do not help, but going back, there is a Bill in the House, HR 2323, titled U.S. International Communications Reform Act. The effort is led by the Chairman, Ed Royce.

And probably one of the most important of our public diplomacy actions has got to be to encourage [a] Muslim Reformation. In other words, [we should encourage Muslims to embrace] a modern interpretation of Qur’an. I mentioned earlier that a radio station in Afghanistan is calling itself the Voice of the Caliphate. So the Islamic State takes public diplomacy very seriously, and they do a very good job of it in terms of the message they want to get out to their followers. We do not have a comparable program.

And then the last leg or pillar of our strategy should be covert actions, and for obvious reasons, I do not want to get into a lot of details, but there must be multiple opportunities in this area. We need to be very active, and the whole advantage of covert operations is it provides us with plausible deniability that we were involved.

Domestic Situation

And now, I want to switch to the domestic situation. Now, although it is not on the slide, it occurred to me earlier in talking about declaring war and the impact that it has on the domestic audience, that is a big aspect of the domestic situation. If the people understand we are at war, then they are much more willing to make sacrifices, and if they see the Congress and the administration all pulling in the same direction, it changes the complexion of the problem in the United States.

But what I want to talk about here are the terrorist attacks, but whether we go to war or not, terrorist attacks are going to be attempted against us and our allies in the name of Islamism, and this has amply been demonstrated by numerous events. To preempt these attacks, I think we have got to use what I call pattern-based searches to identify terrorists planning.

We often do not have the identities of the perpetrators a priori, and the way our system works today, if you are talking about collecting information or recording telephone calls and so forth, the FISA Act and warrants require you to have the name of the individual for which you want to take this action, the surveillance action, but the problem is as amply demonstrated, for example, most recently by the San Bernardino attack is that we were unaware of the identities of these two people.

So right now, the Intelligence Community and the law enforcement community are not using pattern-based searches. They are people specific. We are very good in forensic analysis after the fact, but the problem is to prevent San Bernardinos in the future, we have got to get much better at preempting them. And in my opinion, the way to do that is with what I call pattern-based searches.

Privacy Appliance Concept

I worked on this issue in the time I spent at DARPA after 9/11 in the Information Awareness Office. We were struggling with this very point, as we predicted then, that we were going to have lone wolf attacks and attacks by people that were anonymous to us a priori, and we had to figure out a way with technology to try to solve this problem, so we came up with what I call the ‘privacy appliance concept.’

But there are some problems with it from the U.S. domestic standpoint, and those concern with all of this data that is out there, we have got to figure out some way to gain the security information we need and at the same time will protect the privacy of the individual.

Now, the government agencies in the national security domain worked diligently in accordance with the law to protect the privacy of innocent individuals while protecting the U.S. from various threats. The people want this protection, but they are concerned about privacy, and the problem is the people do not trust the government.

In the early days of the TIA program, I had a contract with the RAND Corporation to think about this whole business of how to achieve security and maintain the privacy of people, and they did a short study, and they compared the thinking about privacy in governments in Europe compared to the United States.

And the interesting thing they found is that in Europe people trust the government, but they do not trust corporations. In the United States, it is just the reverse. People tend to trust the corporations. For instance, all of the personal information that is given out on Twitter, and Facebook, and blogs. And people do not seem to be worried about corporations having this information, but they are concerned about the government having it.

So we tried to figure out how we could use this technology to help, and it is complicated, but I think it is possible. We came up with this concept, and the interesting thing is a video that you probably know. Congress shut down the TIA program. We started it in January of 2002, and Congress shut it down in late 2003, so we were operational for about two years, but the interesting thing is that although the office that I had initiated in DARPA was closed, but the money and all the programs – and we were operating it as an unclassified program.

One of the reasons, by the way, that we wanted to make it a classified [program] was that we knew we did not have all the answers, and we wanted to take advantage of the universities and a broad spectrum of commercial companies. And the trouble is when you make a program classified, especially a highly classified program, you limit the amount of input that you can get from the universities, and so we made a decision very early on that we would make the program unclassified. And we were not secretive about what we were trying to do. We issued a broad agency announcement in early January of 2002, and we were very open in that BAA on what kind of technology we were interested in.

Just as a sidebar, we talked about the need of having to address the problem of how to achieve security with privacy as one section of the BAA. We got almost zero responses from the universities and the commercial companies. People just had not thought about the problem. Eventually, we prompted what used to be called Xerox Park. We prompted them, and they came in with a reasonable proposal.

But anyway, when it came time for Congress to close down [TIA], they moved all of my programs out of the unclassified budget, and they moved it into the classified section of the DoD budget, which supports the Intelligence Community, and the leadership was turned over to what was called at the time ARPA, and eventually became IARPA, the Intelligence Advanced Research Projects Activity.

But the only thing that was not transferred was work on the privacy appliance concept. And one of the reasons is that we had resistance from the Intelligence Community about the concept because they thought that it would slow down their analytical efforts too much. As I said, the people into these agencies like the NSA and so on, the FBI, all think that they are doing things in accordance with the law. And they are trying to protect the privacy of innocent people, and find the information they need for security, and so why should they be bothered with something like all the privacy protection appliance?

But in my opinion what they forget is that the American people do not trust them, and that is a sad state of affairs, but that is the way it is. And one of the purposes of this appliance would be to, as best we can do with technology, to restore that confidence.

So how does this thing work?

Concept for Controlled Data Access

The idea is that you have got databases out there around the world that are relevant to the problems that you are trying to solve, and atop each of those databases, you put this thing called a privacy appliance. [It] sits on top. So you have got transactions that are taking place around the world, and by transaction [I mean like an] email message, [that] would be a transaction, a credit card purchase would be a transaction, the airplane flight would be a transaction, and so on and so forth.

So you have a collaborative, multi-agency, analytical environment over here. And since 2002, the Intelligence Community and law enforcement community have pretty well been able to establish this kind of multi-agency analytical environment. So the idea is that from this environment, pattern-based queries – and I will explain those a bit more in a minute – go in to the cloud of databases out here, and they go to the privacy appliance – and we are going to talk about how it works in a minute – and then out comes the filtered results. There are answers to this query into automated data repositories.

But one of the key points here is that you do not pull all this data into government databases. Obviously, some of the databases are government, some of them are commercial, and largely what has happened with the changes recently within the past year to NSA’s collection is that they have actually moved in this direction. In other words, the telephone companies are going to keep the records, and then queries are made to the telephone companies, and the results are returned. But unless this process is highly automated, the time late in that process is going to be very significant.

So the patterns are very important in that the pattern is the information that you are looking for in the context of terrorist activity. If you do not include a lot of context here in the pattern-based queries, you run into this problem they used to call the six degrees of separation where every single person is connected through six degrees to every other person. But that is only true if you forget the context.

So how do you get these patterns?

Well, my solution was that you have Red Teams, and they are simulating threat organizations that plan attacks and develop the pattern of transactions that they would have to conduct in order to carry out the attack planning. And just to give you an example, from a research and development standpoint, I set up one Red Team, and it was headed by a former director of DARPA.

There are about a dozen people, a very eclectic group. There was a lawyer, there was a Special Forces guy, a retired guy. Special Forces is probably the closest thing that we have to terrorists. And there was a banker. And [it was] just a very eclectic group, covering a large area, a large domain of the kinds of activities that terrorists would have to take in order to plan and carry out an attack.

So how does this privacy appliance work?

And I called it an appliance because I wanted to separate it from a computer or a piece of software, and I wanted you to think about it in terms of something that you could get your hands around and was well-defined, and so that is the reason for this dotted line around it. And that is a transparent, cryptographically protected shell, and it is transparent because we want everybody to understand exactly how it works. It is cryptographically protected so that it cannot be modified by insiders or outsiders.

And we publish the source code for the appliance. Some of you may remember what is called the clipper chip problem, which was a solid-state device that NSA developed a few decades ago for encryption. And the problem was they had a backdoor into the encryption algorithm so that NSA could break the encryption easily. There was a lot of discussion about it. I think there were some lawsuits involved, but anyway, whenever we talk about something like this, we need to avoid the clipper chip problem.

And my way of doing that is to publish the source code for it and have this transparent shell around how it actually works. So there are several complexes here, the policy and business rules and regulations are embedded in it, but they have got to be in machine-readable form. This is a huge problem because frankly – we probably have some lawyers in the room and I hope you will not take offense when I say that lawyers like to write laws and regulations in ways that can be interpreted, and it is very hard to convert that into something that is a black or white and machine readable, but it needs to be done.

Authorization tables about who is authorized to make what queries, the Associative Memory Processing is an index into all those worldwide distributed databases that are out there. Then there is an Inference Control Knowledge Base and Selective Revelation, which I will talk more about in a minute, and an Immutable Audit Trail and Masking.

Now, we talked a little bit about the Search Patterns, about how you come up with them, but in this concept, just as you get a ISA request or a warrant in the law enforcement community to conduct surveillance, you get a judicial authority to authorize these search patterns. So you take the search patterns the Red Team has come up with, and then you go to something like the FISA court and say, okay, we want to make searches on this pattern to see how many instances there are out of this particular pattern.

The judicial authority authorizes the Search Pattern, and at the same time, they also authorize what kind of details can you get back from that query. Initially, the concept is that the only thing you could get back at first would be the number of instances of this search of this pattern that is out there. And for example, if you went out with a pattern, and you got a hundred thousand hits, then what that would tell you is that the patterns are not discriminating enough because in a whole scheme of things, terrorist activity is relatively rare in terms of all of the actions that take place in the world.

So you go back, and you refine the pattern to make it more discriminating and get approval. You go back out again and let us say this time you get a hundred hits. Well, you know, that is a reasonable number, so then you go to the judicial authority and get authorization, let us say to find out the details at some level, maybe the names are still anonymized, but you find out what countries that people are from, how many people are involved, and so on.

But anyway, I hope you get the idea that you go through this process, and at each level of authorization, you would get more and more details until you finally get down to being able to say okay, we want to go after this person and this person. Again, the system that we have got in place now for doing identity searches would take a week.

Now, one of the problems with this, and the reason that it is complicated is that the way I have described it sounds like it would be a very long process, and so the system has to work in a high, automated way. Right now, a FISA request as I understand it takes a stack of paper that is about like this for every FISA request, and, especially with the restrictions that NSA is under now, that is intolerable in terms of making it so time late in order to find the information that you are looking for. So it has to be a highly automated system.

Inference Control has to do with the fact that you have to be careful. If you have access to multiple databases, and you get a little bit of information here, and a little bit here, and a little bit here, if you put all that together, even though the name may be anonymized, you can come down pretty close in identifying who the person is, so that can be handled with a knowledge base that talks about what kind of data is in each of the databases and so forth.

Access Control I think we have talk about that.

In my opinion, the only way that you can deter abuse of the system and satisfy the public’s concern about not trusting the government is that you have to be able to punish the offenders, and it needs to be an ongoing effort, not simply something after the fact [when] some situation occurs, and it is unknown that somebody was doing such and so forth.

So our solution to that was that Immutable Audit Trail. By ‘immutable’ I mean it cannot be modified by the insiders or the outsiders, and the congressional oversight committees have access to the Immutable Audit Trail, and they use the same tools that are used for analysis to find the bad guys to look at the audit trail and find people that are abusing the system. And then they need to be punished. And I would argue there are very few of these, but the point is that to make the people trust the system, you have got to provide for this kind of audit trail and accountability.

There needs to be what I call a Masking function because obviously, with worldwide distributed databases out there, unclassified, some owned by corporations, maybe some owned by foreign governments, you need to figure out a way of masking the query in such a way so that is not obvious what the analysts are looking for. And I mentioned the Associative Memory index, so that is it. I will entertain a few questions.


Audience member:

Yeah, what tool would you suggest for this, like a Palantir type of a tool that would meet all of that criteria?

Admiral John Poindexter:

No, I am not a fan of Palantir. Well, anybody that has used it is not a fan. It is just they have got very good advertising. It is a very slick product, but it is very manual, and this has to be a purpose-built [tool]. Again, I am not going to call it a device but an appliance, and it is not going to be easy to develop, but right now we are not working on it.

Audience member:

Admiral Poindexter, would you recommend to help the Kurds, allowing them to get their independence as a republic?

Admiral John Poindexter:

Well, that was what I was hinting at, and saying that I think in the long term we need to revisit Sykes-Picot. And I would personally be in favor of giving the Kurds their independence, which would be anathema to Turkey.

Audience member:

[It would be] anathema to Turkey, to Syria, to Iran, to just about everybody. Everybody agrees on one thing, they all hate the Kurds. But you kind of made a mistake in that you completely avoided mentioning Iran in your entire presentation. On one level, there is a strong case to say ISIS is irrelevant. Their body count compared to what Al Qaeda and Iran have done is trivial.

Admiral John Poindexter:

So are you suggesting that we want war with Iran? Is that what you are saying?

Audience member:

Absolutely not, but a solution or a proposal that does not address the fact that there are already people playing on the ground there [is ignoring key elements of the situation]. There is already a complicated playing field, and to say that, well, gee, ISIS is bad, we need to take them out [is shortsighted]. Who is going to take their place? We already found out what happened when we took out the Sunni domination in Iraq. It is now an Iranian-aligned state. Syria is going to be an Iranian-aligned state.

Admiral John Poindexter:

That is a weak point in the lecture, I admit.

Audience member:

Since he has opened the issue of Iran, and since you are a Navy man, what can you tell us about the capture of the boats?

Admiral John Poindexter:

I think it is a very peculiar story, and I do not think we have heard all of the details. I cannot imagine how in that environment they could be 50 miles off course, so there is a lot we do not understand.

Audience member:

Admiral [James] Lyons has suggested that they may have been able to jam the GPS.

Admiral John Poindexter:

I read that. Ace is a good friend, and he may be right, but I do not think there is any evidence to that yet, so I think we ought to keep our powder dry for a little bit. And the interesting thing will be if the officer in charge is not court-martialed. If he is court-martialed, then a lot of information will come out. If he is not court-martialed, then it was some kind of contrived incident in my opinion.

Audience member:

The Islamic State has said that the reason why the Middle East and Muslims are down is that they are divided by countries. They want to unify those countries into one nation. This is something that is not just Islamic State is saying but al Qaeda and even the Arab nationalists of Gamal Abdel Nasser said the same thing.

And one problem in the Middle East is there are a lot of people in Saudi Arabia that are Egyptian citizens, Yemen citizens, and they were born and grew up in Saudi Arabia, but they cannot get citizenship. Those people are prime recruits for the Islamic State because they realize that their lives are those of second-class citizens in the country where they have lived all their lives because their parents were not there before 1920. And that is one of the reasons why the Islamic State is doing such a great job with recruiting. [It is because] there are a lot of disparities in those countries.

And I was talking to a guy who grew up in Saudi Arabia, he is from Yemen, he is of Yemeni descent, [and] he says the Saudis are racist. He says, “The Saudis are racist because I can never become a citizen there. I will only be a second-class citizen.” And you wonder why the Islamic State is growing.

Admiral John Poindexter:

Well, there certainly needs to be a lot of work done, and I covered it in one of these points, the whole business of governing these areas needs to be addressed, and hopefully, if we have strong U.S. leadership, we might make some progress because these countries, like Saudi Arabia, are threatened.

Audience member:

I simply want to express my appreciation as an American that you care enough about my freedom and privacy to undertake what you have done.

Audience member:

I had a question about the TIA. You mentioned that it went from being unclassified to classified, so it was handed over, presumably to the NSA and other intelligence agencies. I had a two-part question. So what have they done with it, and have they transformed your system, and how have they applied it? And secondly, to what extent? I know that there was a symbol for TIA that showed sort of this all-seeing eye. So to what extent is it targeting, to what extent is it vacuuming in everything and related to our hegemonic status in the world?

Admiral John Poindexter:

Okay, well, first of all, the symbol is the eye, the all-seeing eye. That is the Great Seal of the United States. After [receiving] the criticism, internally, we changed the logo. We put a blindfold over the eyes. We made the world a smiley yellow face and changed the Latin to ignorantia sit beatitudo, ignorance is bliss.

When Congress closed down TIA, as I said, they moved all the programs into the Intelligence Community, and the funding was placed into the classified DOD budget. And with the exception of the privacy appliance, work continued. TIA was an umbrella over a lot of different programs, and work continued on all the programs that were in that umbrella, except for the privacy appliance. So from my point of view, TIA was a great success, except for the privacy appliance, in that we have got the Intelligence Community to think differently about the whole process of analysis, a much more systematic approach to analysis. I will tell you about a sea story in a minute.

Anyway, that is really one of DARPA’s roles. DARPA is not an operational agency; they are research and development. They are designed, going back to the Eisenhower administration when it was first formed, with creating new ideas, starting research, and then turning it over to the appropriate other agency of the government. And that is essentially what happened, with the exception of the privacy appliance.

Now, the anecdote was that I was involved [in Genova]. A predecessor of TIA was another program called Genova. I had a big private business, and in ’95, I closed that down and started working as a SETA, as an advisor to DARPA. And the first program was called Genova, and Genova transitioned to be part of TIA, but it was not nearly as broad, as all-encompassing. And we had demonstrations. We had a lab and we had demonstrations.

And one day – and I will keep him anonymous, but he was a chairman of the National Intelligence Council, which he had been for several years, and so he came into the demonstration, and sat there, watched. And Genova was all about, as TIA was, taking a more systematic approach to analysis. When we finished, he said to me, guys, that is all very nice, but we do not have time to do all the systematic analysis. To me the only thing that is important is on the day after, who knew what when.

And [this was] a very, very discouraging response. Now, granted, this man had been in office for several years, he was tired, he was worn out, but unfortunately, we had that problem. I do not think we have it as much today, and whether rightfully or wrongfully, I attribute a lot of that to the work that we did at TIA to get the Intelligence Community and law enforcement to think about more systematic approaches. Does that answer your question?

Robert “Bud” McFarlane:

John, thank you for all you have been doing for the past five years in developing the algorithms, the methodology. It seems to me that your work really has laid a foundation that is coming into its own, or certainly could if you are able to continue nurturing it with resources and big data capabilities that are required. I think what you have done and what you could oversee is just essential, and I think here and there people are demonstrating some applications that are derivative, I think, of your work.

I was in Israel about a year ago, and about a half a dozen expatriates from Mossad have begun to look at all the open source databases that are available, everything from air traffic control, and border crossings, and all of these things in a historical context, in short to use both HUMINT, for example, to see all the graduates of terrorist training schools that they find by HUMINT, crank them in, and then track those people for the rest of their lives, and associate them as your system allows with who they associated with, who they talk to, and ultimately, they believe, achieve a predictive capability.

Well, that is phenomenal. I mean, as you know, in intelligence it is one thing to know capabilities, it is another to know intentions.

Robert R. Reilly:

John, if I could just point out to those of you who may not have seen the questioner, we particularly want to welcome tonight another National Security Advisor to President Reagan, who along with Admiral Poindexter was a contributor toward the United States’ victory in the Cold War, and that is Bud McFarlane.

Robert “Bud” McFarlane:

Actually, if you want to give kudos, give them to Bob Reilly.

Admiral John Poindexter:

If you have not read Bob’s book, The Closing of the Muslim Mind, I highly recommend it.

Robert R. Reilly:

I asked John to say that.

Audience member:

You mentioned a declaration of war. What do you think is the likelihood of either a declaration of war or letters of marque and reprisal actually coming out of the United States Congress?

Admiral John Poindexter:

I would prefer the [declaration of] war. I think it is cleaner, and the other smacks a little of mercenary forces.

Audience member:

And Americans have an aversion to that, yeah, a repulsion to mercenaries.

Admiral John Poindexter:

Although it is certainly not uncommon in our history, but the [declaration of] war is the way to go. And I sense that our problems in Vietnam and Iraq were largely caused by the country not ‘going to war,’ and the people not recognizing that it is a significant threat to the United States. Also, it brings Congress into the act.

Audience member:

Well, what are the chances Congress will do that?

Admiral John Poindexter:

It depends on the member, and I do not think any candidate is thinking in those terms.

Audience member:

Is there any connection between ISIS and Boko Haram? And the second question is [about] the President of Chad, Idriss Déby. He has been nominated chairman of the African Union, and he is one of the major allies in the fight against Boko Haram. Do you think that his new function as head of the African Union [will help the fight against Boko Haram]?

Admiral John Poindexter:

Well, I did not mean to exclude Africa from the countries that need to provide troops. I think the countries in Africa there that are fighting the problem today and are willing to continue the fighting ought to be included in a coalition. In that Islamic State five-year plan, I forget what they called that province, but it is on their list.

Audience member:

I want to compliment you on an outstanding presentation. One comment that relates to what this gentleman brought up about Iran, and that is the role of Russia in all of this. They are there. We have to deal with that, and if you were going to add anything to it in addition to Iran, I would add Russia because Russia is part of the triangle with Iran.

Admiral John Poindexter:

Yeah, I agree. I mainly wanted to focus on Muslim forces, but again, in my opinion, Putin does not hold President Obama in high regard. He does not pay attention to him.

Audience member:

But quite seriously they change the whole on ground reality where they are operating.

Admiral John Poindexter:

Yeah, and obviously, we should not draw red lines in the sand if we are going to ignore them.

But I have about talked out people.

Robert R. Reilly:

Ladies and gentlemen, thank you very much for coming tonight. We are so grateful to Admiral Poindexter for coming this evening. If you come back, I promise we will have more chairs. So please go to the Westminster site, and we will soon have our announcements of upcoming lectures, and of course, we will send you emails. Thank you so much for coming.