[applause] we are going to have conversation amongst ourselves for the first hour or so, and then we will open it up to questions from you all. there are two microphones, i see one in the middle and one on the side over there. it seems to me that all of the fundamental questions on the nsa and privacy rotate around balancing security and privacy. how have we done it so far? how should we do it? there is a further question of what privacy even means in a digital age. where every time you go to the store or use your easy pass,

people are keeping track of you, let alone every time you use an e-mail on massive accounts. to make an informed judgment about the nsa programs, we have to know what they are. i want to spend a lot of the time during our our talking about what those particular programs are, how they work, etc. but first, since we are hosting this at a political theory institute, i would like to start by asking a few normative questions around which i think all the others will revolve. so what i want to do, i will ask my first question to each of you now, so you will have a few seconds to think while i'm asking the other. first, what right do you have to publish secrets determined to be such by our duly constituted political authorities?

how can you justify supplanting your own judgment for that of the people's government? and for general hayden, there is a parallel question. what right does the nsa or any other governmental body have to keep secrets in the first place? democracy is based on consent and if we the people don't know what you are doing, doesn't that necessarily undermine democracy? >> i'm going to give an answer that starts with the proposition that my history and politics grades are no longer subject to change. i will give a little bit of a theoretical or structural answer to that question. it has to do with the structure of self-government in a democracy. if the government is working for us, it is usually thought to have two elements, authorization and accountability.

in elections, we say you get to be in charge for a while, and then we hold them accountable for their performance, not only for purposes of reelection but throughout a constant, civil society, political conversation in which the people get their say through a variety of mechanisms between elections. to me, the principal conflict here may not be between security and privacy. that is an important set of questions, but to me, it is between self-government and self-defense. both of which are core values. in fact they are both touched on by the time you get through the preamble to our constitution. defense is fourth among the six principle purposes of the creation of this form of government. the first words are "we the people." so you have this conflict which is to say we have secrets that

are important to keep inherently for purposes of protecting ourselves. but information is essential to govern ourselves. the model i have in mind is that nobody gets to maximize one interest or the other. they have to be balanced. it would be cleaner and neater if you had a model in which there was some trustworthy principal balance or who could who could do that balance for you, but there is not. i'm not elected. i'm not accountable to the people for decisions i make about what to publish. but it is also the case that for the president and everyone working for him, they are not competent to decide what we need to know in order to hold them accountable. we cannot give them that power. we know information is power. secretly, the power to withhold -- secrecy is the power to withhold and obscure information and is a very great power, especially when combined with

surveillance. it is sort of a one-way mirror in which we become more transparent and they become more and more opiate with the growth -- oh pay -- they become more us with theque to growth of secrecy that surrounds it. so the somewhat unsatisfactory answer we are left with, and i can go into this at much greater length, once you rule out all the other possibilities, is that there has to be a process of competition in which governments try to keep secrets, sometimes for good reasons and sometimes not. there is nobody who works in the intelligence field -- and mike would agree with this -- that would not accept that there is this massive overclassification. there are all kinds of things that are stamped with secrecy that need to be. -- that do not need to be. we can get into examples of that if you like.

the ideas the government tries to keep secrets and would try to find them out. no one exerts coercive power at the margin. i guess i will just put it this way. there are trade-offs. we have studied one society very closely which maximized either security or the entrenchment of the state authority against all other interests. when i was a tourist in the soviet union, and moscow in 1984, i was getting lost all the time. wandering the streets of moscow. i usually do not need that much help with that. but it turned out that they deliberately drew the tourist maps incorrectly, because in case napoleon came by, he would take a wrong turn. it's not literally irrational to say were going to publish and in accurate map. all information in principle is useful to the enemy.

but that is not the model we have. the model we have is not -- for example, jfk in his inaugural speech saying we will pay no price, we will bear no burden to secure the blessings of liberty. we are prepared to accept some risk in order to govern ourselves. after the competition -- and i will close here -- i find something out, or i get a big pile of documents, and i now know secrets that i'm not supposed to know, according to the government. there is a cooperative process in which i then go and talk to people, like mike hayden when he held that old job. we discuss the equities. the washington post and every journalist covering this stuff a great deal that they

know if the public interest of the subject is outweighed by the danger of disclosure. >> bart and i agree on an awful lot of this question. the most important agreement is that this is a conflict of values. this is not the forces of light versus the forces of darkness. these are the things that free peoples have to balance all the time. frankly, where that line is is dependent on the totality of the situation in which we find ourselves. that line can move back and forth, depending on realities. when my public affairs officer or someone else would be aware that someone in bart's profession would have a story we felt really problematic, they would come to me and lay it out and i would say all right, and i generally would call the editor. my secretary would make the call and generally speaking, the editor would take the call from the director of the cia. the editor would come on the line, i really want to thank you

for taking the call, and invariably, i would then begin the conversation with look, we both have responsibilities for the welfare and safety of this republic, but how you are about to exercise your responsibility, i believe, is going to make it much harder for me to exercise mind. so we really do need to talk. and we would have that dialogue that bart suggests. this is really tricky. he's got to have freedom of the press. there is a reason that is in the first amendment. on the other hand, the government has legitimate secrets. no fooling things are over classified, and human beings in government will make decisions that perhaps are less noble than we all want to be. as bart also suggests, when the post or the times the classifies -- declassify information on

their judgment, they are performing an inherently governmental act without the system of checks and balances and responsibility that normally go over here on this side of the equation. this is getting much, much more complicated in our society. bart mentioned the old question of accountability and how do you hold a government accountable without a great deal of transparency? let me walk you through the arc of american history when it comes to american espionage. the american intelligence community isn't the only one that keeps secrets. the federal reserve -- they keep secrets. there are a lot of organs of government where the general welfare is improved by the government being closed and secret. at the beginning of the republic, and in most western democracies still, this espionage thing is pretty much contained in the executive

branch. george washington was the nation's first spymaster. when he became president after running the spy ring for the continental army him he insisted on a secret overt action budget. with very little, if any, congressional oversight. so espionage has generally been viewed as being a province of the executive, and the way the nation changed espionage policy was changed the executive. by the mid-1970's, there was broad agreement after the church commission that having this responsibility, this knowledge, this oversight only within the executive branch was insufficient. and the great compromise of the 1970's was to create two bodies in congress and a rather unusual court to spread this transparency and accountability over all three branches of government.

so we end up with the house permanent select committee and the senate select committee, and by intelligence we end up with the fisa court. the theory was that that was sufficient transparency to legitimately get consent of the government in a representative democracy. after all, this isn't ancient athens. that consensus in the grand compromise of the 1970's was adequate, has eroded. i'm sure we will talk about the 215 program here very soon, and people are going to say wait a minute, has been authorized by two presidents, the congress legislated it, it is the madisonian trifecta. i got it all. we've got all three branches of government. and great swing in american popular culture right now, and i'm not just talking about out here on the wings, i mean sustained among thoughtful

people, is that that no longer constitutes sufficient consent of the governed. ok, ok. you told dianne feinstein, mike rogers, dutch ruppersberger, but you didn't tell me. we are kind of at that cusp in our political culture. i was director of the cia and i had a wonderful team of outside civilians. one guy got permission to use -- one of them i got permission iorina.her name, carly f i gave her a tough job. another subcommittee was doing i.t. and another doing recruiting. i gave carly a tough job. study this problem and come out and tell me, will america be able to conduct espionage in the future inside a broader political culture that everyday demands for transparency and more public accountability from every aspect of national life?

they studied it for about four months and came back and gave me a firm, hardly speaking "we are not sure." which is a very telling answer. so here we are. with the national enterprise that i'm just going to assume we all believe is legitimate and necessary, american espionage, and a broader political culture that demands more transparency of an enterprise content demand secrecy for its very nature and its very success. that is why this is now so difficult for all of us. >> thank you. following up, bart, general hayden described what you are doing or the release of this information as being inherently governmental function. we will go into all the various checks that exist on the nsa and whether they work or not. but when you get that

information, i guess there is no checks on you as a reporter, or are there? what is the process you go through in determining this fell on my lap, should i make it public or not? >> are there checks and what is the process? some of the same checks on us, not the defective checks, are not so different which is to say . which is to say that a lot of what the government does or does not do is not because it doesn't have the power according to some legal interpretation of the words on the page, but because it believes that there would be great public consequences, that public opinion would not support that. they would suffer reputational damage or political harm. the washington post cares right a bit about its reputation for

telling the truth and telling it responsibly. and nobody wants to be responsible for some terrible thing happening if we publish story x and the next day there are 10 people strung up in the square of some foreign country because we disclose their identities. that would be a really bad thing for the washington post, in addition to being a bad thing for the country. so there are checks there. the process is as follows. >> before we go on -- on the checks point, for any of the nsa stories that you have run, are you aware of any bad consequences? >> there is a bit of a conundrum, which is to say that the secret has built and the -- has spilled and the washington post has published something. suppose you're inside the intelligence community and you know very specifically what some consequence was. disclosing that knowledge itself, there are constraints from telling you about it.

on the other hand, there are consequences that we would know about. it's also the case that james clapper, the director of national intelligence, in a closed-door briefing used language quite different from what he usually said in public, that they went back and did a study and they found that over the years, there have been breaches and secrets lots of time, sometimes in the context of new stories or a foreign service disclosing something, or sometimes presidents disclose really sensitive secrets because they have a political reason for doing so. for example, when ronald reagan played a recording or quoted a recording of a conversation that took place in libya in order to persuade european allies to allow american planes to fly over their airspace so they could bomb libya. he wanted to prove that we had the case nailed, and very much against the recommendations of

the intelligence community, he revealed that capability. they do a study, what is the damage assessment? clapper said that the damage assessments totally overestimate damage. it is not unnatural or abnormal for the nsa to gain and lose capabilities. technology is changing all the time. the channel that is working to listen in on a target stops working tomorrow because microsoft changes their network architecture, or the other country chooses a different vendor, or the target gets tired of that and switches to a different company or whatever. or targets are trying to practice communication security, so they switch some things. the other thing that clapper said is that people will change their behavior, but they will

make mistakes, and we will exploit them. the nsa is not in immediate danger of losing its global preeminence in signals intelligence. i concede that it is possible that some of the stories published by some of these organizations sooner or later will have degraded the ability to listen in on something. i am prepared to accept that and i believe that you should be prepared to accept that, because at the same time, we are finding out things that have been doing that are completely contrary to what the public understanding was of what their job is and how they do it. if there is a program going on and the executive branch has blessed it and a select few people in congress have blessed it and the fisa court has blessed it, then when it becomes public american people are shocked by it, and many people are angry about it, and they

start telling members of congress and they start introducing bills to stop it, and it gets brought to federal court in the first federal judge to assess its constitutionality says it is orwellian and very likely unconstitutional, then there is another way to tell the story of whether we had already considered as a society to this -- already consented as a society to this happening. i will stop there. >> general hayden, i know you're not not in government now, but are you aware of specific negative consequences from any of the nsa stories? >> three tiers of consequences -- all of them damaging. no, i'm not in government, so i don't know. >> are you being briefed on it? >> no, so that i can speak freely at events like this, i do not go to get updated. >> if i knew this was outdated information, i would have chosen someone else.

[laughter] >> i know how to read a newspaper. there are three tiers of damage. one is operational. by the way if i were in , government, let me rephrase the question you just asked me. would you give me a list of all the communications you did not intercept yesterday? i mean that is really the , question you want me to give you the answer to. and well, i don't know. -- i agreen erosion with bart that intelligence is constantly moving. all advantage is transient. you have penetration, you just download the next whatever, you don't have penetration anymore. somebody buys the next level of encryption you don't have that you don't have that frequency , anymore. people with the communications path change patterns -- not running away from you -- just because people who build a communication path have a smarter way to spend a buck.

we have got to go after it. we have simply accelerated the rate at which we will lose our current accesses. right. to my mind, that is not arguable. that is one level. will we recover it? sure. how much? $8 billion to $10 billion in order to recoup. the second level of loss is -- let me just stay with that for a minute. because bart makes a good point that these things have informed the necessary national debate. but they have done more than inform. they have also misshaped. let's take the one program i know we are going to dwell on which is 215. this was not bart's work, his flagship piece on this was 702. the 215 was pushed out the door with an incredibly detailed coverage of the front and, which -- the front end, which is they

got all your phone records. only over a period of several weeks did the back end story begin to get out there into the public domain. so what is it they do with them once they have them? people let nsa would make the -- people at nsa would make the argument you may disagree even after you've gotten the back end story, but you really can't make an informed judgment until you have the whole context. i saw a recent poll that said over 30% of americans, because of the way this has been portrayed, over 30% of americans think the 215 program allows nsa to listen to the content of your phone calls. all right, so operational loss. second loss is in relation to foreign governments. here i'm not talking about about angela merkel being really upset. what i am talking about is orela merkel's spy service

any other intelligence service around the world to whom we would go to for help in a matter of great sensitivity and a matter that would require great discretion, in which we say this is lawful, a little politically edgy, but it is the right thing to do for both our countries. come on, let's cooperate. we can keep a secret. there isn't another intelligence service in the world i should take that last sentence to the bank. our ability to cooperate has been eroded by foreign -- hasntally fundamentally been eroded by foreign intelligence services having no faith in our ability to keep things where they ought to be. the third level of damage has to do with american industry, who has been unfairly pilloried by this whole thing. now you have got google and microsoft. you have young mr. zuckerberg calling the president and complaining about how his

company is being treated in the global digital domain. american companies do nothing more for the united states than other countries do for their sovereigns around the world. microsoft, at&t, etc., do the same thing that deutsche telekom does for german security services, etc., etc. but it's only american information that has been put out there, it has been put out there in a way that suggests, almost declares, that only the americans are doing this. i would suggest to you that practically the entire continent of europe, including their elected representatives, now know more about american signals intelligence than they do about their own country's signals intelligence. and that's not a balanced discussion.

>> there are lots of things that america might want to be for -- want to compete for global preeminence on, but i don't think one of them is going to be that we are the most closed government and we are the best at keeping secrets. and we don't need it, because our security, we actually have a little bit of room, the luxury of allowing democrat debate over -- allowing democratic debate over big questions like -- should the nsa be collecting all of the records of all the calls that you and i make, are taking -- or, taking it out of the , thist of section 215 court-authorized telephone program what do we think about , something that is new that the nsa never did in its history on anything remotely like this scale, because the capabilities did not exist? the nsa now wants and has taken many steps to achieve a kind of comprehensive situational

awareness of the world. it wants to be able to collect every electron and photon that passes over communications channels in case they might be needed. it is not just metadata, although i think that is enormously important for privacy. but it is also what they do overseas. the debate has been channeled somewhat deliberately in the current political context for the statutory kinds of surveillance. that is to say, under section 702 under the fisa amendments part.he prism under section 215 of the patriot act, they collect all your call records. overseas they are not bound by , statute, supervised at all by the intelligence community or supervised by any court including the intelligence

court. they operate solely under the authority of a presidential executive order. since we had a little bit of a historical suite before about the 1970's, that me give you an additional point of view on that. the 1970's, after a bunch of gross abuses that today's programs don't at all resemble, there was a deliberate misuse of surveillance for domestic, political, corrupt purposes. they created the structure of oversight in the foreign intelligence surveillance act. the idea was you can do what you like overseas, spy on the foreigners, but you cannot do it at home. and you cannot do it from u.s. territories.

so around the time that general hayden became nsa director, the u.s. government started making the argument and executive started making the argument to congress, you need to change this because the structure of global telecommunications has changed fundamentally. it's not american phone calls or e-mails, is just the nature of the distributed networks that you could be in germany and calling brazil and the call passes and transits through the united states, for example. so we need you to take the leash off. we need to have a lot more freedom to spy from u.s. territories.

what they didn't say, and congress change that law. what congress did not address is that lots of u.s. communications are now floating across foreign services. if you're sitting here in washington and you send an e-mail to cleveland, it is literally quite possible unlikely that your e-mail is going to cross ireland or a data center in hong kong. the global internet balances its load by distributing in microseconds the communications flow and the path of communications everywhere. where it is allowed to resume that the people it is intercepting our foreign, it does things like this. the nsa with its british counterpart is breaking into the privately owned links that google has between its many data centers around the world. so a data center in hong kong, one in latin america, they have private fiber that links them directly. nsa was breaking into that and intercepting the traffic that went through them, which is to say ultimately is able to reassemble the entire contents of the data center if it wants to. googling yahoo! won't say what percentage of their users are

americans. let's say it is half or one third. you are talking about ultimately not only the metadata but also the content of hundreds of millions of people that are crossing into these flows that the u.s. feels free to dip into. -- that the nsa feels free to dip into. the legal term for that is incidental flexion. you are targeting foreigners of some kind or another, and as it happens, incidentally, foreseeably, but not your actual purpose, is getting a lot of americans with it. once you have them in your storage tank, as was acknowledged in writing, the nsa feels free to search through it in its own data repositories. if i say to you collectively in this room that i have a foreign target, i'm just going to collect everybody smart phone and download everything from it, all your appointments, all your

text messages, all the little confidence as you have exchanged -- confidences you have exchanged with one another, your bank records, and i promise not to look at it unless i have a really good reason, and it turns out that two thirds or one third of you are americans, turns out that was incidental. it's perfectly legal, and you should be fine. it is not surveillance unless we're doing something with it. don't try that argument if you want to download movies. do not say i have three terabytes of pirated movies and music on my hard drive, but i haven't listened to them yet, so it doesn't count. that's not our ordinary understanding either of the law or of what it means to be understood violence. -- under surveillance. were never debated publicly because we did not know about them.

what snowden allowed us to do is say are we ok with that? >> one of the recent articles -- magic? >> mistake. >> mystic. thank you. it describes that effort. in the clapper letter that reveal this, he pointed out that the committee actually debated whether or not they want to reauthorize the mystic approach, and they agreed that it had. so we are not putting the cart on the bottom of the deck here. bart is right. it was known and consciously, specifically debated in committee whether or not that should continue. wayhe way, that is an edgy to do that. it gets close to what we used to call reverse targeting. decent argument, but don't think this was a renegade nsa. fully debated. i'll try to be really efficient about this. because an awful lot of these snowden regulations are pretty

-- putting things out there, i frankly think some of them are mischaracterized or in accurately characterized. this is really complicated stuff. >> i would be over polite about sitting up there on the stage. >> ok, he screwed this thing up so badly. [laughter] thank you for that. for anybody who really knows how this stuff works. i think a lot of the story -- >> you cannot entirely blame him because he only knows what he has been given. >> right. >> that is obviously not the case. [laughter] >> all right, then it is your fault. >> i'm given a piece of paper. i don't assume that it is authentic or that the facts contained therein are accurate. it turns out there are authentic documents in the world that are wrong.

>> i used to get some of those too. pubic >> i and my i used to get some of those briefings, too. >> i and my colleagues do a lot of reporting around the material. a lot of the reporting, certainly a big part of it is to talk to the peoples who are the stakeholders in government who are running the program. mystic is the story the washington post wrote a couple of weeks ago and said the u.s. is intercepting and recording all of the telephone calls of an entire foreign country. and storing them for 30 days. and then of the billions of calls they go into this 30 day holding tank, millions get sent back to fort meade to beast processed ined and their repositories. >> it's got to be a little country, right? >> we did not name the country and we did not give any hints about the country. because we had already decided before we went to the u.s. government that we were not

going to name the country. i cannot say exactly why. i am in this bizarre position as a reporter of saying i have to sort of hold things back from you. i've always done that from time to time in stories over the years. i could give some examples of that, but here it happens all the time. but we were persuaded, even without yet talking to the u.s. government, that there were specific, concrete, foreseeable harms that would come from naming the country or saying what kind of country it was what or saying what kind of country it was or what kind of intelligence targets required this approach. what we can say is, this raises a big public policy question. if you live in that country or work in that country or you have relatives or friends in that country and you make calls there, that your call is recorded. that would amount to millions of americans in any country in the world. so now they are not just collecting metadata, they are collecting content.

they are listening to your words. let's say they are recording your words and they feel free to keep whichever ones they want and search whichever ones they want because it was incidentally collected in the course of surveilling a foreign country. which is a legitimate foreign target. something the nsa could have ever unimaginably done before in recent years. it is a capability that the nsa did without and kept us safe throughout the cold war, and it is as the general says it is quite edgy and worthy of debate. there is one last point to be made about oversight. judge walton, the chief judge of the fisa court, the foreign intelligence surveillance court, said in his first ever interview with any news organization that the court knows nothing unless

the nsa or the u.s. government tells it. it doesn't have any independent fact-finding ability, that it cannot go out and invest gate. -- and investigate. it doesn't have a former nsa person like that. when the court, in 2009, found that the nsa had been unlawfully running wannabes big row graham bigunning one of these programs and had been consistently misrepresenting the facts of what it was doing to the court, the nsa response to the justice department was, you could make fun of it as these computers are so darn complicated. who understands what they are doing? the argument actually was there was no one at the nsa fully understood the workings of the program and therefore was inadvertent that they were describing it incorrectly to the judge. so when you're saying a judge is supervising it or that congress is overseeing in, the nsa is classifying almost all of this stuff at the top-secret codeword

level. there are fewer than 10% of members of congress who have a staff member cleared to read anything at that level. members of congress do not go into a 10,000 page intelligence budget or sit and studied dance se hearingsstudy den and reports. they just don't. they rely on staff for that. most of them don't get to talk to a staff member who is cleared to read what they are talking about. so the level of supervision in some cases is pretty theoretical. >> i am falling way behind on my accusation countering count, so my turn. [laughter] >> go ahead. >> as i was saying, bart does this have to well and half of half thehis half well

time. even the best of folks are going to get some of this wrong. is he just said, the nsa is so complicated it is hard to explain. not everyone is as conscientious on reporting these stories. a lot of people take the stories and run as quickly as they can to the darkest corner of the room. let's put that aside. let's just say that we can reduce all the stories that are out there to hard-core facts, that we have pulled a hyperbole and the misrepresentation out, and we can get it done. i'm not even going to argue about that's not what we're are doing. let's assume we can all get it there. you realize you're coming in late in the third real of a complex movieof a and you are looking at the third reel. i'm assuming we have all the chaff out of the way. you're looking at the third reel and making the decision, i know the butler did it. and you really cannot make that decision until you walk back to

the first reel, and where did these things come from? why were they developed? this requires a lot more time, but i'll try to be very efficient about this. i became director of nsa in 1999. we were being overwhelmed by the volume of modern communications. because we were approaching communications away we had approached all of our previous life going after communications, which was burrowed down, get the right frequency and listen to the very specifically targeted call. now there we were with a tsunami of global data flying at us in a way that was beyond our ability to control. look at your -- you are all pretty young, you don't know what it was like communicating in 1999. trust me, a lot more photons and electrons out there now. we decided before 9/11 that the

only way we could deal with volume was light pushing back a -- was too quick -- to quit acting like we on the beach trying to keep back the tsunami wave and saying let's try harder to keep the way from engulfing us. we decided turn around and swim with the wave and make the power of the wave your tool. that is collection and metadata. that is why we did it, it was in response to very specific challenge that if we had not dealt with, we would have gone deaf. technology change number two, the nsa spent most of his time watching the soviet union. there was not anyone regardless of your political or suasion -- persuasion that was going to raise that much of a finger in civil liberty concerns, with with nsa intercept doing -- intercepting communications coming out of moscow, in soviet siberia while we are looking for interesting words to pop up on the net like "launch."

[laughter] the 21st century equivalent of that isolated signal on a dedicated network being run by an oligarchic superpower, the 21st century movement of that signal are proliferator, terrorist, money laundering, child trafficker, narcotraffickers, etc. e-mails coexisting with your e-mails in a single, unified, global communication structure. there is no way nsa continues to do what it used to do for you if it cannot go out there and be in the flow where your communications and mine are cold o-mingled with legitimately targeted communications. that is just the reality. the third change, and bart mentioned this already, that is the fact that the global

communications grid changed in such a way that most global communication either resided or transited the united states just because of american dominance in global telecommunications. yet, without the fisa amendment act, and you go from a bad man -- an e-mail from a bad man in pakistan to another bad man in yemen, i don't think it was the intent of the founders and that is why the change in the fisa act. the fourth change in reality, and this is 9/11, and that is simply the fact that the enemy was inside the gate, and that legitimate foreign intelligence targets were here in america. after the 9/11 there was the joint inquiry commission that was made up of both the house and senate intelligence committees.

one of their findings was that nsa, that would be my nsa, the one i was running before 9/11, was entirely too cautious in dealing with the al qaeda communications, and that would be al qaeda communications, one end of which was in this country. and the metadata program was designed to be, frankly, the lightest possible touch we could devise to know which of those communications entering or leaving this land would be the ones we want to drill down on for content collection. after having seen the whole movie now, you may still be pretty confident the butler did it. you may still have kept any or all of those four things i just suggested, but they did not develop out of an nsa director like me or keith alexander, acting like mr. burns on the simpsons going "excellent." [laughter]

i've been waiting to do this for decades. [laughter] they were logical responses to technological and operational challenges. one more thought and i will yield the floor. picture life as a triangle for the subject. there are three sides of a triangle. a happy triangle is an equilateral triangle. it is in harmony. and the three sides are threat, technology, law and policy. you got it? what you want is that if is responsive to the threat and used within the confines of agreed law and policy. those three things never changed at the same rate. but they are changing at amazingly different rates today. and the problem that bart points out in his writing, and one that i will freely admit is that the other two sides change must

-- change much faster than law and that equilateral triangle turns into a violent looking isosceles thing because these things have changed while law and policy are still struggling to catch up. that is kind of where we are in the debate. >> i got a little bit lost with the triangles. >> i could do it again. [laughter] draw diagrams like this. i am used to the curves. i actually want to come back to the question and the point he has not yet actually supported, which is the idea that we are getting it wrong a lot are we are getting out of context. -- or we are getting it out of context. that is certainly possible. it is possible there are things we don't know or that we are mischaracterizing something. there was one quite good story, i thought, that we were getting ready to publish and i was persuaded that it was wrong. i'm talking to sources and i'm talking to the government and i

have very specific reasons -- i received very specific reasons to believe that i had the context entirely wrong and we simply killed the story. but what we are actually doing is trying to said light on the dash to shed light on the subject. we are trying to get it all the way that is not the case when it comes to the public debate on the part of the government. the government wants to keep the secrets and in a few outlying cases, people come out and say flatly false things to the public. everyone knows about example of director clapper and the question about whether the nsa collects any kind of data at all about tens of millions of americans, and he said no. but much more often, there is this radically misleading

information. information is provided that if a friend or family member did that to you, you would consider them to have betrayed you. to have deliberately lead you down the wrong path. here is an example. this 215 program the collection , of all the call records, it does keep coming back to that and not the overseas collection. which in some ways is much more intrusive. but for a long time, the government said we won't tell you at all hammy times we have used this power to collect business records. it is the business records provision. it is top secret which means it would do great harm to the security of the united states if we told you the number of times we used it. then congress passed a law requiring them to disclose how many times they used it. so they came out one year and said it turns out there's nothing to worry about. i think this was 2009. they said we have only used it 24 times last year. don't worry about it. it turned out that they needed

four of those times, let's say four quarterly applications of this power to get trillions of records. trillions of call records. substantially all the call records. the way the program works, it may be that it is a large fraction of that but not the entirety of it. they say 24, it turns out it is trillions of records. they say we only take it when it is relevant to a terrorist investigation. it turns out their meaning of the word relevant is all of them. every one of them could possibly be relevant. if they had been straight with the american people and said we would like to be able to collect all the call records and we will regulate our use of them as follows, if they had done that around the time they actually started collecting them, soon after 9/11, or even if they had done it in 2007, they might well have won public support for it.

but the doing it behind our misleadinghe highly statements about it have led to a considerable degree of shock when people find out what they are doing. >> i would like to go systematically through a few things that we both referred to a few times. if i can just take three or five minutes on each of the three things. the fisa courts and how that first, works. second, the 215 program. the 702. starting with the fisa courts. my understanding that they were created in the 1970's, coming out of a recommendation which uncovered all sorts of abuses in the american intelligence program before that. the spying on martin luther king and all of that. to prevent that, they said they're going to create the special fisa court. i question -- how exactly does the fisa court work? can someone who is an intern or relatively young person like edward snowden or one of the

american university students who are interning at nsa or cia, can they send an e-mail to the fisa court and say hey, i would like some data on this person? or what is the process to have a fisa request? how is the request made and what is the process? >> lots of things to be said about it. the first thing said about the fisa court is incredibly odd. let me tell you why. it exists. [laughter] we are the only western democracy that takes these questions to the third branch of government. other countries, last time i checked that were democratic, like the united kingdom, issued these warrants within the ministries. it is a grand american experiment to involve the third what isf government and essentially the exercise of executive prerogative in terms of conducting espionage.

what is unusual is that we do it this way. in order to do it this way, we put certain restrictions on this court because it cannot nearly be as open as other courts in our system. first of all, it is not adversarial. on the other hand, tony soprano does not get a lawyer either when the fbi comes up with a warrant on him. unfortunately, when we set up the fisa court, we used the language of law which we use warrants in criminal proceedings in order to go up on wiretaps in fbi and other investigations. there is a sharp distinction that is not often made in the public discourse about this. there's an incredibly sharp distinction between doing surveillance for law enforcement and for foreign intelligence purposes. i read commentary very often of suspicionless surveillance. that word has absolutely no meaning to me as a foreign intelligence officer.

i don't collect information on foreign nationals because they are bad. i collect information on foreign nationals because they are interesting. i collect information on foreign nationals because their communications contain data that will keep americans more free and more safe. they do not have to be bad people. do not confuse the two. so now the court is designed to warrant collection against protected persons. so we have reason to believe that a, b, or c is now the agent of a foreign power. even though they enjoy protection under the fourth amendment -- >> by protected persons you mean americans? >> no, american citizens all over the world and anyone in the united states of america, including diplomats, and any organizations composed of the above.

to collect against them for foreign intelligence purposes requires a warrant. in addition, because the peculiarities of the way the law was written -- remember i told you it echoed the law enforcement statute? it put specific restrictions on collecting on a wire in the united states that did not exist for collecting out of the air in the united states. i am not talking about cell phones. i am talking about satellite bounces. >> when you want a warrant on a protected person, what do you have to give to the fisa court? >> the fisa court gets a request for warrant that looks like about the size of a phone book in the city the size of cincinnati. it's about that thick. it is heavily laden with a lot of legal jargon. it is not all boilerplate. it requires a great deal of specifics. in essence, what you are trying to prove to the judge is that the individual you want to target is the agent of a foreign power.

i recall, in one instance -- >> not anymore. >> that is what it does. that changed in the fisa amendment, i think 2008. >> it also changed under the patriot act, which is a lot older. >> and the protect america act which came between the two. >> you are describing this old model that is irrelevant to every single story that has been written. true, bart.is not in order to collect on a wire inside the united states, you also need the court's permission because of the peculiarities of the way the law was written. >> the big packet of information you were talking about, is that submitted by anybody at nsa or cia or department of justice? or does it have to go through a department of justice lawyer? >> it will be crafted at the agency that wants the coverage. it will go through a whole series of bureaucratic stops. it ends up that the department of justice.

all the ones i am aware of when to the white house for approval. we hearf the things about the fisa courts is that they have approved almost every request they have had. this gives the impression it is just a rubber stamp for an irrelevant process. that is why i was pushing for some of the details. >> if the judge is not happy with the fisa application, it is withdrawn and redone. all right? there is a dialogue there. no one who has been through the process would describe it as a rubberstamp. >> is this your understanding of the fisa process? >> almost everything that mike is talking about still happens. and is accurately described, with one exception, i think. but that is not the stuff that people are writing stories about and what the public is debating. that is the old model. they still use the old model for specific targets. we have something short of

probable cause. it could be a reasonable articulable suspicion under the law. but we have reason to think that this person is not -- it used to be that they have to be an agent of a foreign power. now it has to be relevant to the investigation of a foreign power and they have to demonstrate relevance. but this is an individual target. they want my records. they want your records and my records, with a name on them. and it is in ones and twos and threes, and there are a few thousand of them a year. >> let's make that shift to the 215 program. >> ok. there are four kinds of data. george bush ordered michael hayden to start collecting these after 9/11 without the benefit of a statute or court supervision for several years. until the justice department

rebelled and people were about to resign because they deemed parts of the government to be illegal. this is the famous "warrantless surveillance" that was first disclosed at the end of 2005. there is metadata. it is data about data. it is who talks to who when and what equipment they are using. that is metadata. there is content. they were collecting metadata and content for telephone and internet communications. internet communications are not only e-mails. they are skype chats, documents you have stored in a cloud, video, the entire universe of content that travels the internet. and the telephony is your phone calls. what happened after the justice department rebellion and some public knowledge began to lead

the debate, the fisa court was asked to switch gears and did switch gears. no longer was it only issuing warrants for individuals, john and tom and mary. it was now asked to authorize on a quarterly or annual basis huge programs of surveillance. the court would now say, once a year, that you may collect anything that you like from google, yahoo!, microsoft, provided that you certify that it fits targeting rules and it is treated and handled to minimization procedures that are intended to protect the privacy of u.s. persons, consistent with the operational needs. there is some protection. i will not deny that. not all that is as airtight as one may be led to believe. in any case, the court has a

program with standing authority. you do not have to tell me the names. you do not tell the court the names of any of the people that they are collecting. you do not have to tell -- you know, there is not going to be an individual warrant for that person. it used to be that, in the intelligence world, it was not wrongdoing necessarily. there had to be a specific showing that the individual is likely to be a valid foreign surveillance target. that is gone now. the fisa court has accepted and interpreted the law that nobody in the public and, frankly, anyone in the unclassified world who studied this closely, and i count myself among them. several chapters of my last book was about this. i never suspected that the executive branch had told the court and the court had agreed that collecting every single

person's call records was a definition of relevant to a terrorism investigation. >> why was that done, all the metadata? >> the description of what went on under the terrorist surveillance program when i was director of nsa is not accurate. that is about all i can say about that. he is right about the court and the executive branch going to the court for warrants that were that were broad, generalized warrants, rather than specific targeted warrants against individuals. the fisa act put restrictions on who you can target and the techniques of targeting inside the united states. most, if not all, of the generalized warrants that are being described are because a lot of foreign conversations are in the united states.

we approach the court for approval. the american-ness was created by temporary geography, not the communicants. bart is right. in any information collection, there is collection of other individuals. the nsa has procedures to protect u.s. privacy. your question was, why metadata? >> yes. >> as i said, the first warrant is in 2002. the chief judge of the fisa court. it allows us to resume things that were the subject of heated discussions only the previous march. we did it by not relying on the presidential power to conduct foreign intelligence but, by involving an additional branch

of government, the judicial branch, to determine whether what nsa was proposing met the reasonableness standard of the fourth amendment. you and i are not protected against all searches. you and i are protected against unreasonable searches. and it was felt that the best way to do this within the constitutional structure was to go to a court that is familiar with surveillance and ask that question of a judge. in this case, the judge did issue a warrant. >> basically, the collection of metadata is the collection of all the phone call records. the nsa would store this in their warehouses. all of these records. they would not access or read them, except when they had process from a fisa judge. >> no. >> ok.

>> let us take the 215 program, the one we are talking about. after snowden's revelation, but before the president's speech, how it existed, because it changes. the nsa, under the fisa court, get metadata. from the phone company, these are billing records. the justification is, those are records you have already shared with the phone company. they do not belong to you. the nsa get these records. you are not identified. all that is in there is fact of call. this number called that number at that time for that long. and, you know, how many phone calls do 350 million people make in a day. you get the idea of how much data is coming in. actually, what was happening was, this program was set up by me when most people are making phone calls through land lines. it is much more difficult and legally complex.

the fact that cell phones do not have individual call billing -- it was more difficult as people went to wireless communication to get the same kind of information in. as time went by, the percentage of american phone calls it showed up on a daily basis in the database was shrinking to about 25%. the database is sealed off. it can only be approached by two dozen people at nsa, all specifically trained. the machines that they use are keystroke-monitor. the nsa takes what they call a seed number. almost always a foreign number, a number that shows up in a safe house in yemen. we find literature on an individual that shows that he has affiliations with our qaeda and has a cell phone. nsa goes to the database, and we

ask, does the phone number show up in here? if someone in the bronx says, i talked to that guy once a week, the nsa asks that phone number in the bronx who they talk to. that is it. that is the metadata program. the change the president made in his speech is that the nsa used to be able to do three hops. the seed number to the bronx to who he called, to the next one. the president made it two hops. seed number, bad guy, to the bronx, and one hopped out. the president also said the nsa has to go to the court every time they make that query. this is unusual. keep in mind what the court is judging. the court is judging whether or not nsa has a reasonable suspicion that the foreign number collected through foreign

intelligence is affiliated with al qaeda. it is not you that the court is checking on or your rights that it is double-booking. it is looking over the shoulder of the intelligence professional to see if their judgment was correct. the last full year in which the nsa has records, the yelling through the transom thing happened 288 times. >> people's eyes start to glaze over. you are talking about nuances in law and operations. this is complicated and difficult. let's take it down to fundamentals. i am not advocating a policy position and i do not know where the line should be drawn. i know that there are questions that belong in the public domain, boundaries we collectively ought to defend.

that is my ideology as a reporter. transparency enables the checking power of civil society, of the courts, of lobbying, of voting, and of market power. people get to decide why they do business with a company as it is not as safe as it used to seem. companies respond by competing on privacy. let's look at the big picture. a wallet goes missing. what is the most efficient way to search for it? everybody gets searched on the way out the door. lots of societies do that. the british did that in the 18th century. that, generically, is the idea of a general warrant and that was a principal cause of the american revolution and one of the main grievances. the nsa wants to be able to have that kind of efficiency, because global communications flows are enormous, and because it can.

they have never been able to do that before until recent years. most of these programs could not have happened before the year 2000 or so. the amount of data that exists in the world that has been produced and transmitted in the world since 2000 vastly eclipses all the data that was ever produced in the world up until 2000. there is a lot of it. and they want access to all of it. they are not just collecting metadata. the programs that no one ever wants to talk about, overseas, they are collecting content. lots and lots of content. there is an access point called "muscular." one strand of cable transmits 14 times the print in the library of congress every second.

it is huge data flows. they are sitting on that. anything that looks like an e-mail address book, and inbox, or a buddy list, they say "we want that." they suck that in. that is considered content under law. all the transmissions between google and yahoo! is content. that is being collected in bulk. the question is, are we comfortable with that? i understand why -- >> there are not people reading it. it is being collected and sent to a server somewhere. >> obviously, humans cannot lay eyes on all of it. they collect it. there is assumed to be billions of communications that are incidentally american. they feel free to store as much of it as they rationally believe they can use. for example, address books. they don't want all address books.

they do not want ownerless address books. they want address books of people where the formatting that gets transmitted lets them know what it is. useless forms of data, they throw out. they keep it and they search it. >> we should open up to questions. we have 40 minutes left. we have two microphones, one here, and one over there. i will ask people to line up at the microphones, please. finish your answer. >> that is unusual. i'm glad we have an hour for this. >> ok. all right. so, what i will ask -- >> sounds good to me. >> i will ask you to identify yourself and ask a brief question. save speeches for the dorm.

>> my name is ari. i am a graduating senior. my question is for both of you. at what point do you want to start collecting on a terror subject inside of the united states who is an american citizen? >> i will go first. >> that is beyond my competence. >> you mean like the tsarnaev kids? we did not collect on them. >> i know that. >> and why didn't we? because they are u.s. persons, and they had a first amendment right to visit any websites they wanted to visit. their web surfing was not monitored and would not be monitored, and there is not a program that allows us to monitor it. >> hang on. that is not why the fbi and others did not monitor them. it is because when they got a tip from the russian government

that that the older one might be up to no good, they did a preliminary investigation. they said, we do not see anything here and they stopped. if they had wanted to -- they had a tip from the foreign government. that gave the months of authority to do the surveillance. it was not because they were protected persons. >> i am sorry. i was not clear. that was not the point i wanted to make. given the coverage of the snowden revelation, there is a presumption that the nsa is sucking everything up. the fact of the matter is, they are not. bart is exactly right. in order to get that, they would have to have had a certain level of suspicion and go through the processes that allow them to be targeted. after boston, i get questions like, how did you not know they were going to those websites? because we do not routinely look at that.

>> but the overseas collection of americans is much more interesting? >> sure. >> ok. here is the google data center. here is another google data center. two different countries. here is the line between them. nsa goes between them and take in anything and everything it wants from that. even though hundreds of billions of those communications will be u.s. communications. incidentally collected. why is that ok? >> let me make sure i understand the premise you are suggesting. you believe that gmail should be a safe haven for legitimate foreign intelligence targets of the united states. >> is that my question? >> that is how i translate your question. if i am not on that wire, i am not collecting gmail users. bart said, and this hurts your head, we have spent the whole evening here and have not gotten off of section 215. >> we are now. >> the approach you use in

collecting foreign intelligence looks like a funnel. you have access. the next step is collect. then you process. then you read or listen. you analyze. you draw up a report. and you disseminate. you get the picture? the funnel gets smaller. very often, people use the number out here that describes the potential access and move that number up the chain to the more sophisticated and narrowly focused activities. that is not true. i go back to my premise that gmail is not a safe haven. google is an international company. a lot of people -- >> google is a u.s. person. >> google is a u.s. person in certain circumstances.

by the way, that is not their cable. it is a virtual cable that they use. i asked you the question, hotmail, gmail, safe havens? these are global e-mail providers that are used by everyone. is your theory that you cannot touch it because eric schmidt will be very mad? >> i want to allow another question, but, in one line, there is a long distance to go from saying that gmail should be a safe haven for terrorists and saying, on the other hand, that you should be, as the u.s. government, siphoning in all of the content that goes across that table. including all the american communications. there is a room for gradation. >> there is. i just gave you the steps. not everything you collect is processed.

not everything you process is reviewed. not everything you report is disseminated. >> a thing to point out that exists here -- the prism program, the first story i wrote about the nsa, gives the nsa more power than it ever has had to get gmail, all of the google content, from the providers, under an annual authorization from the court. it had quite a bit of latitude. it was not allowed to sift through everything in the server to see if there was anything that interested us. that is what it is doing in the overseas collection. >> bart is right. binary is too strong of a word. there is a manichean view of the world when it comes to american intelligence. there is protected

communications and everything else. nsa has a lot more freedom of action going after communications. some europeans do not attach privacy to a particular sovereign. it is a more general human rights. i got it. that is a subject of discussion. the american approach to this is that it is really a binary world. protected persons and persons who are not protected. nsa possibility to work over here, under 12333, but not without congressional oversight, is really quite powerful. >> let's take a question over here. >> this is something i heard about, reading the newspapers, about various governments trying to segregate their internets. i know china is doing this and germany is thinking about doing this. what effect would this have on the nsa possibility to collect information or, is such a thing even possible?

can you cut off from the world internet? >> this is beyond my expertise. i have a few thoughts. first, there are a lot more people in countries talking hyperbolically about their desire to cut off the global internet and keep all their data at home than are actually taking steps to do it or will possibly do so. you are cutting yourself off from the world and it is damaging. it is not practical. there are countries that say google, yahoo!, all of the data produced by germans have to stay in german data storage and you cannot go anywhere else in the world. certainly not the u.s. the fundamental structure of the internet does not allow that. there are countries like china or iran. or, at the extreme end, north korea.

they want to cut themselves off from free debate and free information and are using the nsa as an excuse to strengthen that argument. in some ways, if it was possible, it would be cleaner. it would take us back to world war ii-era. you could break the german or japanese military code and it was a no-brainer. if you could do it, nobody would see any downside to it at all. this is a code being used exclusively on networks being used by the enemy, in a war. now, if you try to break encryption, everybody -- almost everybody uses the same stuff. and by the way, when there is stuff in the file that shows the nsa knows about this or that entity that has its own homebrew, we are not writing about that. we are writing about stuff that

has general public interest. if you are breaking the encryption standard that run the internet because we want to get the bad guys, you are breaking everybody else's encryption. that is just the way, fundamentally, it works. >> a question here. >> thank you for coming. i am edmund sweeney. i am a freshman in the school of international service. i am a native of pittsburgh. >> philadelphia. we need a visa to go to philadelphia. >> i have one question in two parts. one for mr. gelman and one for general hayden. we talked about current and past. i want to talk about the future. we live in a world of wikileaks and edward snowden now, and i do not think that will be going away. my question is, in the future, the next 10, 30, 50 years, the increasing demand for transparency and this general

sentiment and people like julian assange and edward snowden, what -- with the release of information, what implications does this have for the future of organizations like the nsa and cia and how they will deal with those problems, and how they might change as a result? for mr. gellman, how will the media respond and interact with this information? >> i will go first and go quick. we have to recruit for the generation that produced bradley manning and edward snowden. that is just reality. that generation has a different compass to transparency than my generation. that is a reality. that is something my old community is going to have to

deal with, particularly since one of the major muscle movements in my community has been to generalized sharing. we used to have a principle called "need to know." for 10 or 11 years, it has been "need to share." then you have snowden. this is hard. the other impact of this is that, when you undertake an espionage activity, you put what you can gain over here and what you can lose over there. let's talk about signal intelligence. what can i gain? i could hear the head of state. i could hear the head of state's personal views. that is really cool. what could i lose? you could be embarrassed if that is public. you make that judgment. by the way, this hand is not a moral hand. it is a political and public-relations hand. this is what i would lose if it

was made public. that actually gets multiplied by the likelihood it gets made public. and the general factor of the likelihood of signals intelligence being made public, particularly when it comes to grabbing a signal out of the air, it is zero. it is undetectable. therefore, although the gain over here might not have been massive, the potential loss is not measurable because you cannot figure out a way in which it becomes public. with edward snowden, that number over here will never be zero again. you have to factor in insider threat that makes operations known that the target nations are never going to detect. that might be the strongest governor, brake on aggressive signals intelligence collection against some countries. not american guilt or

friendship. it is that calculus. that raw "national interest" calculus. there is another factor over here. >> i will also try to be brief. in july, a month after the revelations started, keith alexander was the director of the nsa and he gave a talk. he said that he wished he could bring every american into his huddle and tell them the plays he is calling. if he did, the bad guys would know. and so that is too dangerous. that is true. it is true that if he reveals his capabilities, everyone can hear it. there is no encapsulated walled off communication in global affairs. it is also true that when we did find out what was happening inside of the huddle, a lot of

americans do not like the place being called. that is among the reasons why the existence of a classified stamp does not end the inquiry for me when i am deciding that something should be public. there are all kinds of stuff that is classified that is self evidently trivial. i have in my position a navy laundry manual for how you wash clothes on warships that is stamped secret. i also have noticed that there are interesting variations on how classification works, depending on the forum. there is an internal nsa report. i wrote about it. general hayden did not like my story at all. i talked about the number of compliance incidences, the nsa calls it, errors or violations of law. there are statistics. how many times do we have this compliance problem or that

compliance problem. the internal report classified that as secret, which by definition means that it would do harm to national security if we knew the number of times they screwed up. that is hard to defend. when they report to congress, it is classified, top secret, special intelligence. therefore, 91% of all congressional staffers could not read it and it greatly restricted the oversight possibilities. so, the american public reacts when there is something that the u.s. government is doing and something they did not know they were doing. something they might have accepted if they had known in advance. or if it had been openly debated. >> thank you. >> i want to return to the question of call-chaining. i want to ask general hayden is my understanding of multiplication is correct in the program. you mentioned 288 inquiries.

>> yes. >> each of those, which until recently was three hops, over the course of a year, 1000 is the first hop. the second would be a million. there is some overlap. i am extrapolating. not by orders of magnitude. i just wanted, if that understanding is correct -- if you are talking about three hops, you are not talking about 288 things. you are talking about many, many millions. that is close to the totality of u.s. communications. >> i'm relying on statistics that the former deputy director put out. in chris's words, and those inquiries, 600,000 numbers. i agree with you that that is not the right number of hops

out. but, you know, this is an investigation. by the way, it is just numbers. no other identifying data is there. what they are looking for are patterns that may or may not -- so, you have a spider chart. you have a number in buffalo. it is interesting when the second hop out of buffalo hits a second hop out of schenectady. that is why it is done. it is done as an investigative process. you are looking for linkages that might merit further investigation. >> chris inglis did not explain what he meant by 6000. it is almost -- there is a saying in the newsroom that there is a danger when a reporter does arithmetic. but it is hard to imagine you can go even one hop, with on the order of 300 numbers, and touch only 600,000. that means you are multiplying by only 200. that means that each number

called an average of less than one person per year. this is what chaining is, from a non-computer science person. we have all heard of six degrees of separation. six degrees of separation on average. it is not just a play on a novel. it is important computer science theoretical work that was done. it said that the mean number of hops you have to do to touch every single person in the world, to touch from any one person to any other person, is six. let's suppose, more conservatively, the 300 people call 100 people a year and you multiply that by a hundred and that by 100 and that by 100. i don't do math. i count on my fingers. that is close to one million.

>> i do math. it is a lot more than 6000. >> it is. that does not mean they did chain 100 number from each of the 300 numbers. they also said that it was fewer than 300 numbers in one year and the numbers have varied from year to year. you can be confident that the numbers have not varied downward. the statements are perfectly consistent with the possibility that some other year they had 5000 seeds. they would not have lied if they said it this way, but that is the exact kind of massaging of data in public statements that they have done and in other demonstrable ways. >> my name is harry. my question is less technical. it is for both of you or neither of you. it seems obvious to the casual observer that, every time there is a national tragedy, there is a reaction within the populace

and government where we give up liberty for security. the patriot act, the internment of japanese americans. obvious examples. it takes some time and society scales this back. is it concerning to both or either of you that, given multiple 9/11's in a short time, we could end up giving up a lot of liberty for security. whether or not you agree with the programs. >> thank you. right behind you? >> my name is sarah harvard. i am an aspiring journalist. i go to school at the school of international service. my question has to do with journalism in light of edward snowden. we have seen glenn greenwald and james rosen accused of conspiring with whistleblowers.

congressional members have question for them to be put to trial. i was wondering what your thoughts were on investigative journalism. and, do you think that press freedoms are under attack by the obama administration? >> i would jump in on 9/11. here is where i do my moral calculus. despite my professional role, we are all responsible for our actions as human beings. my great fear is that we did not use the authorities available to us right now. i used the metaphor that, this is the box you want us to play in legally. articles raise issues about national debate. the nsa will say that they have two houses of congress, a president, a court that says it

is good to go. until otherwise, they will play section 215 out to the edges. we are going to take the hops however many they take us. the background ethical argument for this is that, if we do not play with in this box and we fail and we have another catastrophic attack on the homeland, the box you will demand, not the one you will accept, the one you will demand will be out here. in a reverse psychology way, playing this aggressively with in the law of now is actually protection against dangers to civil liberties in the future. >> do you share that? the concern that if any of these revelations make it easier for another terrorist attack on the u.s., that in the future it will perversely lead to demands for

much greater surveillance? >> everybody wants to talk about terrorism. that does not account for nearly anything close to the majority of intelligence gathering. there are a lot of purposes. it is an emotional trigger. you can talk about all of this stuff as counterterrorism. some of it is and some of it is not. our society has always had pendulum swings. it is a cliché. the difference is, what happens under the regime of extreme secrecy? when you intern every japanese citizen of the united states, they are in big interment camps, and everybody knows it. you have announced it. and they were behind barbed wire and you can take pictures of them and look at your history books and there are the pictures. we know when we overreached and come to a decision after a certain span of time.

we say, i do not like that picture and that was not the right thing to do. the supreme court reverses. when you are doing everything in secret or large parts of it in secret, it is hard to decide if you have gone too far. we are still now, all these years later, we are trying to resolve the torture debate. the senate intelligence committee has now voted to recommend for declassification hundreds of pages of the executive summary of a 6000 page report. they cannot publish their report under the current understandings of the way these things work and there is no law that says that only the president gets to decide what is classified. classification is the artifact of an executive order. congress could pass a statute on it. it just has not done so. in any case, we still do not know what happened. very fully, there are 6000 more pages of conversation that they

want to have. as of this date, they are not allowed to publish. when you have taken steps like, let's collect all the content that flows across this cable or all of this data about the phone calls that you make or, allowing us to put communications in our repository. we will decide what we can look at and when. we have good motives, trying to protect you. we do not get to debate those things. we would not be having these conversations it not -- if not for edward snowden. if it were not for stories that disclosed things the government intended never to disclose. >> i have one minor thought. those americans in the room, you are citizens of the nation with the most transparent intelligence community. there is no one near us. it does not make barton's argument that there needs to be more transparency.

i agree that it does. you need to understand the baseline that you are working at. you have more information about your country's espionage than any other citizenry on the planet. >> let's take two questions over here and close that microphone. >> my name is james manning. my question piggybacks off of that well. we have heard about an overclassification of data and a shock reaction about sources and methods. my question is, how do we go about having that debate without compromising intelligence undertakings with the methods being used while avoiding shocks? particularly as mentioned here. i will let general hayden answer. >> yeah. >> actually, the two speakers before me cover the questions i would ask. if someone else, from this side -- >> you did not answer my question. >> the one about the

journalistic -- i would be happy to answer. i got distracted. on the journalistic one, there are some alarming signs. there have been more probes from prosecutions under obama. everyone has read this. more than all previous presidencies combined. they have used aggressive means of surveillance against the associated press and fox news. their legal construct is that a leak of classified information is a counterintelligence problem. counterintelligence is a legitimate foreign intelligence surveillance target. therefore, that makes reporters doing reporting legitimate targets of intrusive methods of electronic surveillance. i spent a lot of my time trying to protect the security of my communications. not only from the u.s.

government. that is been in my threat model as a reporter for a long time. but also from the foreign governments that want documents that i have that should not be public. they might want to come get them. all of that being said, i am sitting on a stage, a free person and not in handcuffs having a civil conversation with general hayden. you can read the espionage act or various other statutes. there are ways to read them. there are available theories of prosecution under which i could be charged. it is not because we do not have laws to prosecute journalists. it is because we have a legal and political norm in this country that says that this is a line that we do not want to cross without reasons very different from what we are seeing right now. and, so, by the way, when you ask about sanctions or controls

on me, i am well aware that if i published a story that clearly and obviously led to a security disaster, that could change the legal and political norm according to which i am not charged with a crime. >> i wrote an op-ed for cnn.com on the investigations. i do not come out quite where bart is, but i am closer than you might think. i thought that the prosecution of the nsa collapse under its own weight because of overreach. some things deserved response from the government. not that. i was troubled by the scattershot subpoena on the associated press and the identification of rosen as a co-conspirator. that is someone of my background. transparency, how do you make that work? it is very hard. i do not have a good answer.

we are going to have to -- the intelligence community has to be more transparent than it ever has before. if we are not, you are not going to let us do this stuff in the first place. there is one caveat. by doing this, we will make you more comfortable. but it is inevitable that we will make you less safe. that is the deal. >> thank you for coming. both the president and the congress, in the usa freedom act and several other bills, have introduced the special advocate. this is a privacy and civil liberties advocate that would be placed within the fisc court. and some proposals of this, you would be granted powers to appeal decisions of the fifth court of appeals to the federal courts, potentially ending up in

the supreme court. i was open to get your general thoughts and comments on this position, in general. the potential effectiveness and, also, whether you think the granting of standing was going too far. >> one more. >> hi. thank you for coming. my name is alejandro alvarez. i am a sophomore in the school for communication. dick cheney spoke at our campus. he said that he believed that 9/11 could have been prevented if the nsa had capabilities back then that it currently has. do you agree with this statement? why or why not? >> there has been enormous amounts of scholarship on whether 9/11 could have been prevented with the technology of today. it has been fairly well established and is the view of the 9/11 commission that 9/11 could have been prevented with what they had then if they had

used it as designed and had the coordination between the fbi and cia had gone as it should have. but that is sort of beyond my qualification to speak. the issue of the fisc advocate is also beyond my competence. i follow and have read a lot of opinions now. i have read some fisa requests in the past. any proceeding before a neutral arbiter is going to be, in the long run, working better and better if there is a checking power and an advocate. we know that the fisa court has blessed a number of big decisions for the first time in its history that said certain very large-scale programs were lawful. before, it was only individual

warrants. we know that the first time that that a federal judge considered that program in a contested legal proceeding, it was found to be unconstitutional. another one considered it shortly after and said that it was constitutional. that is the whole point. a contested proceeding tends to bring out the best evidence. in principle, i would be for that. >> advocate for individualized warrants. we do that in the criminal system. number one, it happens less frequently. it has implications. you want the benefit of the adversarial process. with regard to 9/11, what vice president cheney was referring to was two guys in san diego. we, at nsa, intercepted seven or

eight phone calls from san diego to a safe house in yemen. clearly, the selector, the target, was the safe house in yemen. we detected the call. there was nothing in the physics of the intercept or in the content of the call that told us that these guys were making the call from san diego. we have people speaking arabic calling to a safe house in yemen. i think one of them, his wife was pregnant and he was asking about her welfare. we intercepted seven calls and pushed up reports on intelligence. again, the content of the call in san diego.

with the 215 program, when you go through the transom, the phone number in san diego raises its hand. that is what the vice president was referring to. barton is right. there are other ways we could have done better prior to 9/11. saying that we could have done otherwise is like saying, all i want to do is win games close. no coach has that as a game plan. i would like to have the additional capacity. >> when you say the number would have raised its hands under the 215 program -- >> the metadata would have recorded the connection between the u.s. number and the yemen number. >> thank you. we have two last questions. one here, and then here. >> my name is chris ayres. i am a sophomore in the school of public affairs. my question is for mr. gellman.

you mentioned the data that you have to protect because you are a target for the u.s. and outside countries. what do you do with the data that you consider to be too important to publish? you delete the data? do you keep that? do you know what the other main receivers of this data have done? >> go ahead. >> for the final question, thank you very much for getting into the details of this program. to return more to where we started talking about, we talked about how radically data collection has expanded in the last decade. mr. gellman, you pointed us to this intercepted between google centers and bulk data collection. since data collection has changed so much, should our concept of surveillance and what it means to be surveilled change? we were talking about data going from a private company. we, as americans, have enough

our data to private companies. should that mean the same thing? >> security, for reasons that are obvious, i'm not going into all of the details. [laughter] as a general proposition, what i have done, with the benefit of the expertise of some of the best people in the world on this because they have been available, i have kept the material physically secure. there are physical barriers as to where it is. it is on encrypted computers with the best modern encryption. it is now, when you get to if frontal attack on encryption,

the encryption still stands against any country in the world and it never touches a network. it is on a computer that never touches a network. there are other things. i'm not going to go any further than that. i have, actually, the ironic thing is, i have been asking since the first day of last year, for the more secure way to communicate with the government when i tell them what i have each time i write a story. i say, i would like to tell you about this story. i'm telling you every single fact that i am contemplating publishing. i want to understand whether it is correct, in the right context, if you want to tell me anything about the damage that might be done. i have said, it is crazy to have this conversation on an open phone line or e-mail. and, nine months later, they have not provided me with a secure means of having those conversations.

it is not bureaucratically or legally simple for them. and so we have these odd conversations about page 17, and a squiggly looking shape in the bottom left and words that rhyme with each other. i'm -- literally, words that rhyme with each other on the page. so as to avoid putting an all out on open phone lines. there are ways that this could be better approached. >> let the record show that bart retains the data. just like nsa. [laughter] >> if you think that is the same, we have more to talk about. >> on the last question, on the meaning of privacy in the digital age, in an age where every time anyone uses easy pass or goes shopping, giving all sorts of information to

companies voluntarily. we are comfortable giving away tons of metadata and content of data as everyone who has ever gotten a pop-up ad after they have made an inquiry knows. does this change how we think about these topics at all? >> of course it does. privacy is the line we continually negotiate between ourselves as unique creatures of god and as social animals. and that line changes. i think it is rapidly changing now with the digital, interconnected age. it is a nightmare for security services in a democracy who are sworn to protect reasonable

expectations of privacy when the definition of reasonable is a movable feast within a broader culture. >> there are two things going on legally and conceptually, in terms of privacy. there is a third-party doctrine. i voluntarily told verizon, who was transmitting my voice over vacations that cross all sorts of other companies, therefore, i have no lawful or moral reason to believe that that is confidential. privacy is relational. i may be happy to tell you something and not all these other guys. that is not the way it works. the third-party doctrine began as an analog, pencil and paper thing.

you voluntarily told the bank something and you waved your privacy. i'm going to cut myself off entirely from civilization. there is no way for us to function in the city or society or be employed or students without these communication methods. if we allow google to know that we are sending a love letter to our love interest, we are happy to let anyone read it. that is preposterous. the other is the reasonable expectation of privacy. that has been interpreted that, when you know the means exist, then you have no reasonable expectation to believe it is not happening to you. if i tell you that we have "through the wall" thermal

imaging, you now know that your sex life is not immune. that technology does exist. you are all warned. the concepts have been stretched. there is a lot of work being done to try to bring that back into reason. >> thank you for an illuminating conversation. we will surveil your further thoughts on this. [applause] [captioning performed by national captioning institute] [captions copyright national cable satellite corp. 2013]

>> the united nations security council is meeting this afternoon to discuss ukraine will stop that's coming up in just a few minutes here on c-span. pro-russian insurgents commandeered six ukrainian armored vehicles along with their crews today. meanwhile, nato announced its moving troops and planes to the region. we will have live coverage of the human meeting starting shortly at 4:00 eastern here on c-span. congress is on break this week. we are bringing you some of the supreme court's notable oral arguments this term. as evening, a case dealing with free speech and discrimination. secrets look at whether service agents can be sued for removing protesters from an where george w. bush was dining while supporters were allowed to stay. that's at 7:00 eastern. at 8:00 eastern, the fifth annual women in the world summit at the lincoln center. hillary clinton joined a pop

singer leading protests in ukraine. also, and opposition leader says what she describes in the aftermath of chemical attack tonight on c-span. u.n.e look from headquarters in new york city. the un security council scheduled to meet at 4:00 eastern today, just a few minutes from now to discuss ukraine. we will have that free live shortly on c-span as soon as it gets underway. a portion ofme, today's "washington journal" looking at the role russian gas is playing in the tensions between russia and europe. >> the energy weapon here is the gas flowing from russia to

europe. been doing with its natural gas supplies since the start of this crisis we continue to your about in ukraine? >> in terms of supplies, russia hasn't done anything yet. what they have done in the past since the fall of the soviet union, russia has used a gas weapon on 50 occasions, either an overt threat or cutting off supplies and gas. this is straight out of the playbook of the post-soviet russian state, to use a factor they are you just supplier for natural gas as a weapon to influence what other states do. since the start of this crisis, they have not shopped supplies. to, but thereatened main thing they have done is jacked up surprise -- jacked up prices severely, so that adds to the stress of the new government and kiev. this matter to the greater eu? guest: the european union gets a big russian of its gas from

russia. the biggest single supplier for the european union is russia. and about half of that goes across ukraine. depending on what happens in the dynamics between russia and ukraine, you could physically disrupt the amount of gas that get into europe. it met us for heating and for electricity. it matters a little bit less now that we are in the springtime. in the past, russian austerity in the winter, you can feel it a little -- it is a bit more acute. april and may, not as much. but this is something everyone in europe is on tender hooks about. let's talk supply and demand. the first is the chart for energy. the top is russian gas reserves followed by iran, qatar, and the united states on down. the russian gas supply, how big is that in the world market, the