[00:00.000 --> 00:01.460]  Not really a panel, but a talk together.
[00:02.160 --> 00:07.280]  So we are at the 50th anniversary of the Freedom of Information Act.
[00:07.400 --> 00:13.560]  It's hard to believe that it's 1966 that that was a...what president was that?
[00:14.360 --> 00:14.940]  Was that Nick?
[00:15.040 --> 00:16.260]  That was before Nixon, wasn't it?
[00:16.400 --> 00:16.760]  LBJ.
[00:16.920 --> 00:17.640]  It was LBJ.
[00:17.900 --> 00:21.200]  Believe it or not, signed the Freedom of Information Act into law.
[00:21.840 --> 00:23.560]  And the Supreme Court has stated that...
[00:23.560 --> 00:24.260]  Can you hear me?
[00:24.660 --> 00:25.160]  Thanks, Ralph.
[00:28.120 --> 00:28.520]  Yes.
[00:28.520 --> 00:33.580]  That the Freedom of Information Act, quote, defines a structural necessity in a real democracy, end quote.
[00:33.660 --> 00:34.440]  And I couldn't agree more.
[00:34.720 --> 00:40.740]  So I just want to say who these folks are on our panel today.
[00:42.000 --> 00:46.300]  Jameel Jaffer is from the ACLU, and David Pozen with the Columbia Law School.
[00:46.540 --> 00:52.220]  And they're going to talk about the history of the Freedom of Information Act, where it is now and where it's going, and how it all impacts us.
[00:52.680 --> 00:53.240]  Thank you.
[00:59.110 --> 00:59.770]  Thank you.
[00:59.770 --> 00:59.870]  Thank you.
[01:00.090 --> 01:01.110]  I'm David Pozen.
[01:01.330 --> 01:02.490]  I'm honored to be here.
[01:03.250 --> 01:04.310]  Just a quick note.
[01:04.490 --> 01:06.510]  LBJ didn't sign FOIA into law.
[01:06.690 --> 01:08.330]  It was enacted over his veto.
[01:11.430 --> 01:19.970]  But we've been asked to inform you that there's a hack FOIA workshop tomorrow, one at seven and one at eight in the Paris room on the sixth floor.
[01:20.550 --> 01:20.870]  Tonight.
[01:21.630 --> 01:21.950]  Tonight.
[01:22.110 --> 01:22.150]  Tonight.
[01:22.350 --> 01:22.790]  Today is Saturday.
[01:22.950 --> 01:23.110]  Thanks.
[01:23.910 --> 01:27.910]  We are not going to talk about how to use FOIA, as I gather those sessions will.
[01:28.030 --> 01:30.710]  We're going to talk about bigger picture issues with the law.
[01:31.010 --> 01:34.510]  And we're going to structure it as a kind of rolling, pretty informal conversation.
[01:35.910 --> 01:40.250]  So, to get things started, I'll just say a quick word about what the FOIA is and how it works.
[01:40.370 --> 01:43.530]  Some of you may know this all too well, but in case some of you don't.
[01:44.390 --> 01:46.510]  Quickly, it was enacted in 1966.
[01:46.530 --> 01:49.250]  It took effect in 1967.
[01:49.250 --> 01:55.970]  The engine of FOIA is the request for a government record.
[01:55.970 --> 01:58.510]  And that's the term record in the statute.
[01:59.090 --> 02:05.030]  The federal FOIA law applies only to executive branch agencies.
[02:05.610 --> 02:16.950]  Unlike some of the state level FOI laws, it does not apply to the legislature, to the Congress, or to private parties, or to the president's inner circle.
[02:17.910 --> 02:26.830]  After agencies receive a written request for a record, it then turns over to them to...
[02:26.830 --> 02:30.790]  And the request, I might say, has to reasonably describe the records it seeks.
[02:31.330 --> 02:37.010]  It turns over to them to have 20 days, absent unusual circumstances, to respond to your request.
[02:37.290 --> 02:42.150]  In practice, agencies routinely miss that deadline and complain that it's wildly unrealistic.
[02:43.330 --> 02:55.370]  They have to turn over... the presumption is in favor of disclosure of responsive records, unless they fall into one of nine enumerated exemptions, the first and most famous of which is the national security exemption.
[02:55.910 --> 02:58.990]  But they also cover trade secrets, privacy, and other matters.
[03:00.350 --> 03:08.730]  If you are denied in your request, you have a right to administrative appeal, and then judicial review in the D.C.
[03:08.990 --> 03:09.750]  federal courts.
[03:10.910 --> 03:19.890]  And you have no obligation in submitting a request to explain why you want the information, what you'll do with it, why you have some special relationship to it.
[03:20.010 --> 03:23.010]  That actually was the model under the Administrative Procedure Act in 1946.
[03:24.030 --> 03:25.070]  FOIA did away with that.
[03:25.070 --> 03:30.110]  So it's any person that gets to submit a request for any reason or no reason at all to any agency.
[03:30.770 --> 03:35.930]  And it therefore kind of frames government transparency as an individual right that we all have.
[03:37.530 --> 03:39.030]  And it's held by the requester.
[03:39.810 --> 03:43.910]  Okay, so enough on how... we'll come back to more about how FOIA works, I'm sure.
[03:44.770 --> 03:46.330]  That's just a nutshell summary.
[03:47.030 --> 03:52.430]  In Jamil, we have one of the leading national security litigators in the FOIA area.
[03:53.130 --> 03:56.230]  And that is one of the most controversial areas of FOIA practice.
[03:56.890 --> 04:03.870]  So I want to turn it over to Jamil to start talking about how do you see FOIA working in the national security context?
[04:04.110 --> 04:07.290]  It's routinely claimed that that's an area where FOIA breaks down.
[04:08.030 --> 04:08.750]  Do you agree?
[04:12.390 --> 04:14.750]  So first, it's great to be here.
[04:14.950 --> 04:18.930]  I would say it's great to see all of you, but we can't actually see any of you because of the light.
[04:18.930 --> 04:21.690]  The transparency here is unidirectional.
[04:21.850 --> 04:23.390]  You guys can see us, but we can't see you.
[04:24.630 --> 04:30.250]  So, no, I don't have a very high opinion of FOIA in the national security context.
[04:31.330 --> 04:32.390]  You know, it's true.
[04:32.390 --> 04:39.410]  The government sometimes releases information in response to FOIA requests in the national security context.
[04:39.410 --> 04:48.410]  But it does so selectively and strategically, and usually it regards those disclosures as a matter of discretion rather than legal obligation.
[04:49.710 --> 04:58.650]  It's very rare, almost unheard of, that a court will require the government to release national security information that it doesn't want to release.
[04:59.890 --> 05:02.510]  It does very, very occasionally happen.
[05:02.790 --> 05:10.970]  But those victories in the national security context tend to be not just rare, but procedural rather than substantive.
[05:11.150 --> 05:13.830]  They have to do with their technical victories.
[05:14.010 --> 05:16.930]  They're not meaningful transparency victories.
[05:17.190 --> 05:19.010]  And I thought I'd just give you one example.
[05:19.770 --> 05:27.970]  For the last, I guess, six years, I've been involved in litigation, FOIA litigation relating to the drone campaign, the CIA's drone campaign.
[05:28.170 --> 05:34.930]  The use of remotely piloted aircraft to carry out so-called targeted killings abroad.
[05:35.750 --> 05:42.890]  And we understood from the outset that it would be very difficult to get information from the CIA on this particular topic.
[05:42.930 --> 05:44.650]  But we nonetheless went ahead.
[05:44.650 --> 05:54.590]  We thought that the public had a right to know what legal authority the CIA was relying on and how that authority was actually being used.
[05:54.850 --> 05:56.410]  So how many drone strikes had there been?
[05:56.630 --> 05:57.810]  Where were those drone strikes?
[05:57.950 --> 05:59.190]  How many casualties had there been?
[05:59.490 --> 06:05.530]  Basically, we thought the public had a right to know who the CIA was killing and why it was killing them.
[06:06.350 --> 06:10.470]  The CIA, probably needless to say, didn't share our view.
[06:11.910 --> 06:18.650]  And it responded to the request by invoking something called the Glomar Doctrine.
[06:18.830 --> 06:31.970]  Which is essentially a doctrine that allows an agency to say that even to confirm or deny that it possesses records relating to the request would jeopardize national security.
[06:31.970 --> 06:40.730]  So the CIA essentially said that everything about the drone campaign, even the purported existence of the agency's drone campaign, was a secret.
[06:41.290 --> 06:46.030]  And it took the position that our FOIA request shouldn't be processed in the ordinary course.
[06:46.150 --> 06:50.450]  That it couldn't process our FOIA request in the ordinary course because to do so would jeopardize national security.
[06:50.450 --> 07:05.130]  And even as CIA officials leaked information to the media on an almost daily basis about the CIA's targeted killing practice, practice of targeted killing, the agency maintained in court that nothing at all could be released.
[07:05.430 --> 07:14.410]  So that was sort of the landscape in 2010, 2011 when we first started, when we first filed our request and first started litigating the request.
[07:14.410 --> 07:19.830]  We won two significant victories in this case, in this set of cases.
[07:20.010 --> 07:23.410]  The first was in 2012, the D.C. Circuit, the Appeals Court in D.C., in a decision written by Judge Merrick Garland who, as you know, President Obama has since nominated to fill the late Justice Scalia slot on the Supreme Court.
[07:38.630 --> 07:45.810]  The D.C. Circuit ruled that the CIA could no longer pretend that its interest in the drone campaign was a secret.
[07:46.050 --> 07:57.690]  And it ordered the agency to process our request in the ordinary course, meaning tell us what documents it had that were responsive to the request and explain why those documents were being withheld.
[07:57.690 --> 07:58.970]  So that was 2012.
[07:59.310 --> 08:17.010]  And then in 2014, the Second Circuit here in New York, another federal appeals court, ordered the Justice Department to release a legal memo in which Justice Department lawyers had concluded that the CIA could carry out the targeted killing of an American overseas.
[08:17.010 --> 08:24.970]  So those two rulings were, you know, by, I think by most people's measure, victories.
[08:25.230 --> 08:26.870]  They were extraordinary rulings.
[08:27.050 --> 08:32.050]  It's, like I said, it's very rare that courts side with requesters in the national security context.
[08:32.050 --> 08:48.510]  And in fact, if you look at Judge Garland's congressional disclosure form, you know, filed after he'd been nominated to the Supreme Court, he lists the cases that he has decided that he thinks of as the most significant.
[08:48.690 --> 08:55.950]  And one of the ten cases he lists is this case from 2012 in which he held that the CIA had to acknowledge its interest in the drone campaign.
[08:56.710 --> 09:11.790]  And I think if you talk to litigators about the most significant national security victories, FOIA victories in the post 9-11 era, everybody would put these cases on their list and maybe near the top of their list.
[09:12.510 --> 09:26.270]  But the amazing thing, I think, is that in both of those cases, the courts ordered the release of information only after having concluded that the government had officially acknowledged the information we were asking for.
[09:26.550 --> 09:32.710]  So in other words, the courts didn't conclude that the government's secrets should be disclosed.
[09:33.130 --> 09:37.950]  In both cases, they concluded that the purported secrets weren't secrets at all.
[09:38.110 --> 09:39.830]  And that's a very different thing, right?
[09:39.830 --> 09:57.050]  So, to put it, I guess, a little more sharply, both courts, both appeals courts, in these ostensibly, you know, very significant FOIA victories, were willing to rule for us only after concluding that we already had the information we were asking for.
[09:57.590 --> 10:09.150]  And I think that that kind of victory, those kinds of transparency victories, are reflective of the kinds of victories that are possible to achieve in this context.
[10:09.150 --> 10:20.030]  And, you know, that's sort of, those are the height of, that's the extent of the possibility in this particular context.
[10:20.350 --> 10:27.490]  And, which is why I say I have a relatively low opinion of FOIA's effectiveness here.
[10:27.490 --> 10:32.610]  So I guess I'd, you know, at this point, turn it back to David and just ask, you know, do you share, do you share that assessment?
[10:33.930 --> 10:36.910]  And, you know, as you said at the beginning, I'm a national security litigator.
[10:37.050 --> 10:41.090]  Is that assessment, even if you do share it with respect to national security, what about everything else?
[10:41.090 --> 10:43.670]  So I basically share that view.
[10:43.810 --> 10:44.890]  I just add a few points.
[10:45.510 --> 10:50.290]  The, I remember looking as of 2005 at the case law.
[10:50.610 --> 10:56.430]  Had there been any cases in which the government asserted the national security exemption and lost?
[10:56.970 --> 11:01.370]  Got a definitive judicial resolution that the exemption was improperly invoked.
[11:01.510 --> 11:03.650]  And I could only find one case at that time.
[11:03.810 --> 11:06.490]  Arguably, the number's up a bit since then.
[11:06.490 --> 11:07.970]  But clearly, the U.S.
[11:08.230 --> 11:12.350]  government tends to win when it asserts the national security exemption from FOIA.
[11:12.530 --> 11:14.170]  I should say, the U.S.
[11:14.450 --> 11:17.470]  government tends to win when it asserts FOIA exemptions, period.
[11:18.270 --> 11:21.710]  The statutory standard is what's called de novo review.
[11:21.830 --> 11:26.530]  The court's supposed to give no deference to the agency that withholds information.
[11:26.810 --> 11:28.330]  That's very rare in U.S.
[11:28.390 --> 11:32.010]  administrative law that you get such an undeferential standard.
[11:32.010 --> 11:40.050]  But nonetheless, the government wins, by most estimates, 90% of the time when it asserts any exemption under FOIA.
[11:40.150 --> 11:41.710]  It's probably higher than the national security area.
[11:42.750 --> 11:51.190]  So courts just seem not to want to get into the business of interfering with government assertions of a need for secrecy under FOIA, period.
[11:52.450 --> 11:56.570]  I think the best case for FOIA in the national security area would probably have to be a little more subtle.
[11:57.090 --> 12:00.430]  It would have to concede that the government tends to win in these cases.
[12:00.870 --> 12:05.170]  But nonetheless, in the shadow of FOIA, the government settles more often than it would.
[12:05.270 --> 12:10.810]  Or releases some documents so as to get out ahead of a lawsuit it doesn't want to suffer in court.
[12:10.890 --> 12:15.350]  And I think that happens, and I'd be interested in your perspective, Jamil, to some extent.
[12:15.350 --> 12:28.970]  But that even so, FOIA has to be seen as basically a superficial response to the rise of the classification system after World War II and during the Cold War and beyond.
[12:29.530 --> 12:42.130]  FOIA, the great landmark statute affirming the people's right to know, develops basically in parallel with the rise of the classification system within the executive branch for national security information.
[12:42.130 --> 12:45.550]  And the latter comes to swamp the former, I would say.
[12:45.650 --> 12:50.630]  The classification system is where all the secrecy resides, or the bulk of the secrecy resides in government now.
[12:50.850 --> 13:02.490]  And it's not much touched by FOIA due to the national security exemption, also Exemption 3, which says if another statute says you don't need to be bothered by FOIA, then you don't need to be bothered by FOIA.
[13:02.770 --> 13:08.690]  And NSA relies heavily on Exemption 3 for all of its operations to avoid FOIA.
[13:10.390 --> 13:25.330]  Interestingly, in the congressional debates, which took many years, leading up to FOIA's enactment in 1966, initially, the national security agencies seemed quite threatened by the law and were very worried about it.
[13:25.490 --> 13:37.450]  By the time FOIA's actually enacted, however, in 1966, they had dropped a lot of their opposition and no longer seemed to be as threatened by it, I think making a realistic assessment of how they would fare under it.
[13:37.450 --> 13:48.650]  And ironically, you might say, Congress had never, arguably never, officially recognized the legitimacy of the executive classification system at all prior to FOIA.
[13:49.170 --> 14:00.870]  In writing in Exemption 1, we have Congress actually, in some sense, validating the classification system, even as it, you know, stands up this right to know in other areas.
[14:00.870 --> 14:13.890]  So that would be my take on the national security field, that FOIA is doing some work in the shadows that we shouldn't discount entirely, but it has to be seen as a mismatch and overmatch as compared to the classification system.
[14:14.710 --> 14:21.970]  I want to, I have thoughts about how it's faring elsewhere, too, which I'll get to, but I want to see if you want to riff on any of that.
[14:22.250 --> 14:26.470]  Well, you know, I'm thinking about FOIA, in the shadows of FOIA.
[14:26.570 --> 14:41.410]  I mean, I think you're right that there are circumstances in which the threat of FOIA leads the government, even the small possibility that the government might lose in court, results in the government releasing information that it might not otherwise release.
[14:41.410 --> 14:49.910]  There are also situations in which FOIA forces the government to confront a decision or to make a decision that it would otherwise defer.
[14:51.710 --> 14:57.270]  You know, the government may be of two minds about whether to release a certain piece of information.
[14:57.950 --> 15:12.510]  If a FOIA request is filed and a lawsuit is presented, and that lawsuit is, you know, sharply presented to an appeals court, then maybe the government, maybe, you know, an administration that is ambivalent about the release of the information might be forced to make a decision.
[15:13.110 --> 15:14.890]  So I think that there are, you know, I think you're right.
[15:15.010 --> 15:27.970]  You can't, you know, you can't say that FOIA is never at all effective in compelling the release of information or persuading the administration, the government to release information.
[15:27.970 --> 15:32.810]  But I think you're also right to say that those are sort of, that's at the margins.
[15:33.170 --> 15:47.990]  And the overall, you know, the promise of FOIA, as stated in 1966 and stated many, many times since by the federal courts, has certainly not been fulfilled in this particular sphere, at least.
[15:49.170 --> 15:58.430]  So maybe now it's time to pivot to where FOIA is doing more work, for better or worse, as against the federal government.
[15:58.710 --> 16:09.190]  And I think it's telling to frame that discussion by noting which agency was most worried about FOIA by the time it was actually enacted in 1966.
[16:09.410 --> 16:13.110]  As I said, the national security agencies more or less make their peace with it.
[16:13.110 --> 16:25.090]  The holdout agency that was most worried about FOIA by the time it was clear it was going to pass was HEW, Health, Education and Welfare, now split up into different agencies.
[16:25.890 --> 16:33.150]  And HEW's concern in a nutshell was that FOIA would constrain its ability to regulate economic power.
[16:33.790 --> 16:42.830]  That FOIA would hinder its efforts against big business to try to promote public health, safety, welfare.
[16:42.830 --> 16:44.070]  And so forth.
[16:44.950 --> 16:52.550]  And there I'd say that HEW's concerns were somewhat prescient about how things have played out.
[16:53.010 --> 16:55.130]  And I might put it like this.
[16:55.250 --> 17:04.290]  I have a historian colleague at Columbia named Ira Katznelson who wrote this great book called Fear Itself about the legacy of the New Deal for American governance today.
[17:04.290 --> 17:09.910]  And he says what the New Deal left us with is this schizophrenic or Janus-faced state.
[17:11.090 --> 17:25.410]  Inwardly, when the state is regulating domestically, it looks often weak, obsessed with procedure over substance, and very deferential to and protective of business interests.
[17:26.730 --> 17:36.950]  Externally, however, when facing outward to the rest of the world, the state, the American state looks purpose-driven, powerful, interventionist, eager to take the fight to enemies.
[17:36.950 --> 17:39.250]  So we have this schizophrenia.
[17:39.490 --> 17:47.450]  We have the so-called procedural state in which public institutions are largely too weak to check private economic power.
[17:47.470 --> 17:56.810]  And we have what he calls the crusading state where public institutions of this country dole out overwhelming violence with little democratic oversight.
[17:56.810 --> 18:05.390]  So I would say FOIA largely leaves intact, maybe even reinforces the crusading state for the reasons we just discussed.
[18:05.830 --> 18:16.570]  But actually, and maybe more surprisingly, it feeds into the procedural state too and is a problem for some things agencies try to do that we might like.
[18:17.330 --> 18:22.530]  So a place to start in thinking about those issues is corporate use of FOIA.
[18:23.410 --> 18:28.450]  There have been a lot of studies now, empirical studies, of who actually makes requests under FOIA.
[18:28.630 --> 18:33.730]  They all confirm that commercial requesters dominate the FOIA landscape.
[18:33.950 --> 18:44.730]  And some agencies, like the Food and Drug Administration or the Environmental Protection Agency, they are not just a majority, but a huge supermajority, 90% plus, of the requesters.
[18:44.970 --> 18:50.210]  They have the deepest pockets, and they tend to have lawyers, and they're the most likely to litigate.
[18:50.210 --> 19:02.290]  So there's some evidence too that agencies prioritize commercial requests over other requests, like from the media or the general public, maybe for fear of what will happen if they don't.
[19:02.830 --> 19:09.550]  So commercial requesters are crowding out other requesters in a lot of contexts.
[19:09.550 --> 19:18.250]  They pay somewhat more in fees than members of nonprofit organizations or reporters or academics.
[19:18.690 --> 19:21.330]  However, they don't pay nearly at cost.
[19:21.550 --> 19:30.630]  The government recoups less than 1%, the latest figures from last fiscal year say, of the cost of administering FOIA from fees.
[19:30.630 --> 19:43.410]  So we have this... and there's a cottage industry that's developed, I might add, of companies whose whole business model is requesting information under FOIA and then reselling it often to institutional investors like hedge funds.
[19:44.130 --> 19:52.030]  And some agencies, data brokers or information resellers are actually the main themselves, the main requesters in aggregate.
[19:53.050 --> 20:07.290]  So you have this kind of hidden corporate subsidy, I think, in FOIA, which we don't tend to talk about at such, but it amounts to a wealth transfer from ordinary taxpayers to certain types of corporations.
[20:08.310 --> 20:20.870]  At the same time, in agencies like HEW's successors, there's some evidence that it diminishes regulatory capacity, makes it harder to get things done, and distracts agencies.
[20:21.050 --> 20:26.230]  Congress has never funded FOIA at remotely the level it would take to comply with its deadlines.
[20:26.870 --> 20:36.530]  So agencies end up roping in non-FOIA personnel often to work on FOIA matters, and they complain, at least, that this hinders their ability to do their regulatory mission.
[20:37.530 --> 20:55.250]  Last point I'll make in this regard, and then I'll turn over to Jamil to respond, but also to start thinking about what we might do if we see any of this as a concern, is the type of journalism that FOIA produces, I think, is very interesting and in some ways distinctive.
[20:56.510 --> 20:59.690]  FOIA was passed in part to help journalists.
[21:00.010 --> 21:06.050]  There are a lot of statements by members of Congress about how it was meant to facilitate investigative reporting.
[21:06.230 --> 21:13.450]  And it does that to some extent, although with the economic decline of the news media, it's doing it a lot less these days than it used to.
[21:13.710 --> 21:16.890]  And the delays are so long that it never really did it for a lot of reporting.
[21:17.110 --> 21:24.250]  It really only works for investigative reporters who are willing to stay with the story for a long time, maybe even litigate if they don't get all the records they seek.
[21:24.750 --> 21:29.650]  In any event, there have now been some empirical work about the journalism that comes out of FOIA.
[21:29.750 --> 21:35.190]  What do news stories that say they got records from FOIA tend to talk about?
[21:35.190 --> 21:41.210]  And the main thing they tend to talk about is waste, fraud, and abuse within government agencies.
[21:42.190 --> 21:47.710]  Arguably, the second biggest type of FOIA story now is a story just about how FOIA is broken.
[21:47.950 --> 21:54.010]  So there's like this kind of meta journalism on FOIA's dysfunctionality.
[21:54.690 --> 22:00.470]  Stories on waste, fraud, and abuse are obviously important and an important part of democratic accountability.
[22:00.470 --> 22:08.110]  I would just suggest though that FOIA over-represents the extent to which that is a problem or the problem in government today.
[22:09.250 --> 22:16.050]  Because of other legal mandates, agencies generally have to produce reports on that kind of behavior when they find it.
[22:16.670 --> 22:23.230]  There are therefore already written records out there if you want to go hunt for them on those subjects.
[22:23.510 --> 22:26.150]  They're not exempt like a lot of other things are.
[22:26.230 --> 22:28.250]  They're not exempt under any of the nine exemptions.
[22:28.250 --> 22:33.330]  So the structure of FOIA makes kind of the sitting ducks.
[22:33.970 --> 22:36.990]  Records that tend to talk about waste, fraud, and abuse.
[22:39.470 --> 22:43.050]  And journals aren't getting a lot of other things they might want through FOIA.
[22:44.150 --> 22:51.170]  I think it seems likely to me therefore that you get the systematic over-representation of that as the problem we should all be concerned about.
[22:51.270 --> 22:52.570]  As against other issues in government.
[22:52.770 --> 22:55.150]  Again, not to say that that stuff doesn't matter greatly.
[22:55.790 --> 23:08.030]  And that FOIA therefore kind of produces reporting based on that kind of distrust and hostility to what the government is doing beyond what might be warranted.
[23:08.130 --> 23:13.070]  And moreover in the fact that everyone who uses FOIA finds that the delays are wildly not complied with.
[23:13.210 --> 23:17.950]  FOIA also reproduces that distrust and dysfunctionality in its very experience.
[23:17.950 --> 23:28.250]  So at two levels, FOIA is harming the agencies, again, like HEW, who more than maybe the designers of it would have intended.
[23:28.710 --> 23:38.190]  So that, Jamil, this kind of FOIA harms the procedural state or feeds the procedural state in ways that at least a certain type of liberal or progressive might not like.
[23:39.350 --> 23:40.770]  Whereas it helps the crusading state.
[23:40.830 --> 23:43.050]  Is this too bleak an assessment?
[23:45.190 --> 23:45.530]  No.
[23:45.790 --> 23:46.490]  I don't think it's too bleak.
[23:46.650 --> 23:49.410]  Actually, as you were talking, I was thinking about this.
[23:50.530 --> 23:56.170]  I was trying to remember that joke whose punchline is, you know, the food is so bad and the portions are so small.
[23:56.170 --> 23:58.890]  Because there's something to that here, too, you know.
[24:01.670 --> 24:07.970]  But, you know, one thing, and you may have mentioned this in passing, David, but I think it's important to highlight it.
[24:08.810 --> 24:20.590]  You know, even if FOIA were working exactly as it was supposed to work, it had a very relatively modest ambition in some ways.
[24:20.590 --> 24:30.070]  You know, it reaches only executive branch records and only records, which is a word that's, you know, narrowly defined.
[24:31.110 --> 24:34.750]  And just to give you an idea of the significance of those limitations.
[24:35.010 --> 24:40.030]  So FOIA doesn't reach judicial documents, judicial decisions.
[24:40.050 --> 24:48.910]  And we have a court, as many of you probably know, that oversees foreign intelligence surveillance or certain aspects of foreign intelligence surveillance.
[24:48.910 --> 24:57.410]  And for 30 years has issued decisions relating to the lawfulness of the National Security Agency's practices.
[24:58.570 --> 25:06.690]  And post 9-11 has started issuing these very significant decisions about the constitutionality of the NSA's activities.
[25:07.010 --> 25:09.850]  And none of those decisions were ever made public.
[25:09.850 --> 25:24.970]  And you can imagine that if we had a broader statute, a broader statutory right of access that encompassed judicial decisions, some of that stuff might have come out before Ed Snowden released it in 2013.
[25:25.190 --> 25:29.270]  And I think we would all be better off if that had happened.
[25:29.490 --> 25:34.390]  But that's a result of the narrowness of the FOIA.
[25:34.990 --> 25:37.190]  FOIA doesn't reach congressional records either.
[25:37.190 --> 25:47.750]  The Senate Intelligence Committee has put together this 6,000-page study of the CIA's detention, interrogation, and rendition program post 9-11.
[25:48.030 --> 25:56.430]  And the Intelligence Committee, as a matter of discretion, released an executive summary of that document.
[25:56.690 --> 26:04.550]  But FOIA doesn't reach the report itself insofar as that report is a congressional document.
[26:04.550 --> 26:13.090]  Now, there's a dispute, which the ACLU is involved in, about whether that document should be seen as a congressional record or an agency record.
[26:13.430 --> 26:17.050]  But there's no dispute that if it is a congressional record, the FOIA doesn't reach it.
[26:17.270 --> 26:28.970]  So, you know, it's important to remember that even if you fix the FOIA, you know, in the sense of making the FOIA more effective in the sphere in which it was supposed to operate, it's not actually going to solve every problem.
[26:29.410 --> 26:33.570]  And, you know, this is actually a point that David has made to me before much more effectively than I just made it.
[26:33.650 --> 26:34.890]  But I think it's an important point.
[26:37.130 --> 26:46.570]  So, you know, that's all sort of preamble to how do you fix the FOIA even, you know, even sort of taking the FOIA as it, you know, as it is in some sense.
[26:48.070 --> 27:00.270]  On my list of things that I would do, starting with the least ambitious and moving to the most ambitious, I would start by strengthening the affirmative disclosure requirements.
[27:00.590 --> 27:11.030]  So, just to give one example, there's a part of... FOIA is really a bifurcated statute.
[27:11.030 --> 27:16.450]  Part of it requires the agencies to release information on their own and the other part is reactive.
[27:16.690 --> 27:18.850]  When people file requests, the agency has to respond.
[27:19.030 --> 27:24.070]  So, but the first part is very weak and has been interpreted, almost interpreted away by the courts.
[27:25.930 --> 27:32.570]  And those first affirmative provisions were meant to require the agencies to disclose, among other things, their own law.
[27:32.770 --> 27:44.330]  So, when an agency issues a decision or promulgates a policy that has an effect on individual rights, they're supposed to affirmatively release that document to that policy.
[27:44.890 --> 27:47.070]  And that doesn't always happen.
[27:47.210 --> 27:47.910]  It rarely happens.
[27:48.450 --> 28:05.810]  And some of the most significant documents relating to national security, like the opinions of the Office of Legal Counsel, for example, in my view, should have been released, affirmatively, you know, redacted for intelligence sources and methods, but nonetheless,
[28:06.090 --> 28:08.730]  you know, presumptively released, but weren't.
[28:08.950 --> 28:19.470]  And the result is that we ended up in a fight about the release of those documents in response to, you know, after somebody filed a FOIA request.
[28:20.330 --> 28:31.470]  I think that one fix, you know, one fix I would make if I were in charge, and obviously I'm not, but one fix I would make is strengthening the affirmative disclosure provisions.
[28:31.470 --> 28:43.870]  A second fix would be to prevent or prohibit the agencies from withholding legal analysis under the national security exemption.
[28:44.110 --> 28:57.430]  So, in my mind, it's one thing to say, well, the agency can't release the names of the people who provided it with information, you know, its intelligence sources, or can't release the specifications for the predator drone.
[28:57.430 --> 29:00.550]  That seems legitimate to me, that kind of withholding.
[29:01.170 --> 29:12.310]  But it's something else to say, well, we can't release, we're going to use the national security exemption to withhold the legal analysis that we rely on in order to carry out targeted killings.
[29:12.390 --> 29:19.810]  To me, it seems like the public has a right to know, or should have a right to know, the agency's interpretation of the law.
[29:19.970 --> 29:26.330]  And insofar as the interpretation of the law is entangled with classified facts, well, you can redact the classified facts.
[29:26.510 --> 29:28.370]  So, that's a second fix.
[29:28.550 --> 29:37.850]  And then finally, and this is probably the most ambitious one, I would add a public interest balancing test of some kind.
[29:38.070 --> 29:49.270]  So that, you know, right now, once the agency establishes that a particular piece of information falls within one of the national security exemptions, that's the end of the game.
[29:49.270 --> 29:55.650]  You know, the information is then ordered, you know, ordered properly withheld.
[29:57.010 --> 29:59.910]  But security is one, you know, one value.
[30:00.290 --> 30:01.470]  It's not the only value.
[30:01.490 --> 30:09.310]  And in some context, you would want to, it seems to me, weigh the security concern against other, you know, other interests or other factors.
[30:10.750 --> 30:27.750]  You know, it may be true that, you know, let's assume it's true that the disclosure of Justice Department memos relating to interrogation or surveillance or the drone program would disclose some intelligence sources and methods.
[30:27.790 --> 30:30.590]  I don't think it's actually true, but let's assume it's true for now.
[30:31.290 --> 30:35.490]  That there's no way to disclose those memos without disclosing intelligence sources and methods.
[30:35.490 --> 30:58.770]  I think we should then consider, well, should we nonetheless disclose those things because they are so important to the public debate or because these policies are so consequential or because, you know, you can imagine a million different reasons to disclose those things in spite of a real national security interest in withholding.
[30:58.770 --> 31:01.590]  We don't have that kind of public interest balancing now.
[31:02.150 --> 31:16.130]  It's really just a, you know, a single factor decision that's made in the first instance by the agency and then, in theory, reviewed by the court, but in practice, the courts defer to the agency's initial determination.
[31:16.570 --> 31:23.990]  And there's no one who asks the question, well, you know, in spite of the, you know, notwithstanding, the security concerns should we release anyway?
[31:24.290 --> 31:26.370]  And that doesn't make any sense to me.
[31:26.370 --> 31:27.430]  So those are my three fixes.
[31:27.630 --> 31:28.850]  Do you have your own?
[31:30.150 --> 31:32.830]  I like them and I'll largely just gloss them.
[31:33.670 --> 31:37.410]  I'll note that public interest balancing is increasingly used abroad.
[31:38.210 --> 31:42.490]  FOIA is one of the country's leading legal exports in recent times.
[31:42.510 --> 31:48.170]  Now over 100 countries have some version of a FOIA law, as do all 50 of the U.S.
[31:48.370 --> 31:48.590]  states.
[31:49.110 --> 31:50.930]  But some countries have departed from the U.S.
[31:50.970 --> 31:52.330]  model and gone beyond it in some ways.
[31:52.330 --> 31:55.410]  One of them is this incorporation of public interest balancing.
[31:55.970 --> 32:00.650]  So even if an exemption applies, it's not just a kill switch for disclosure.
[32:00.950 --> 32:05.530]  There's still a judicial inquiry into whether, nonetheless, it might be appropriate to release some of the information.
[32:06.330 --> 32:19.970]  I would just single out the first of Jamil's prescriptions as, I think, the biggest one, the broadest, potentially, which is the move to affirmative disclosure, as it's sometimes called proactive disclosure.
[32:20.270 --> 32:37.030]  And that is just Congress, instead of waiting for a request to come, and there were over 700,000 requests filed with federal agencies last year, Congress just tells agencies, on this timeline, this sort of information has to be publicized in this sort of way.
[32:37.670 --> 32:38.710]  Just do it.
[32:38.830 --> 32:44.090]  Make it available, you know, now presumably posted on the Internet, and you don't need to wait for a request.
[32:44.650 --> 32:49.390]  I see that as the big road not taken in 1966.
[32:50.530 --> 33:00.330]  I think the reason that Congress didn't go with affirmative disclosure and instead went with this request and respond model was twofold, largely.
[33:00.330 --> 33:05.430]  One, just technically, practically, it was unclear how you would do affirmative disclosure.
[33:05.730 --> 33:09.210]  You're just going to put everything in, like, a physical reading room, you know, in some...
[33:10.430 --> 33:12.210]  in DC, in some agency office.
[33:12.330 --> 33:13.730]  That wouldn't be very helpful to most people.
[33:14.490 --> 33:20.470]  So just physically and technically, how do you disseminate printed materials at that kind of scale was unclear in 1966.
[33:21.270 --> 33:27.830]  And then second, I don't think there was faith that the executive branch would enforce an affirmative disclosure requirement reliably.
[33:27.830 --> 33:31.670]  You tell agencies every month, you've got to put out this kind of data.
[33:32.510 --> 33:34.210]  Who's to say that they're actually going to do that?
[33:34.310 --> 33:40.310]  Better to empower a small army of citizen enforcers through FOIA's method.
[33:41.330 --> 33:45.350]  Since 19... that concern will never totally go away and is, I think, a valid one.
[33:45.470 --> 33:50.070]  But since 1966, the executive branch's internal structure has changed.
[33:50.170 --> 33:56.370]  So I think we can have more faith that, at least in large part, an affirmative disclosure mandate would be complied with.
[33:56.370 --> 34:04.490]  There have been the development of offices like inspectors general, ombuds people, auditors within the executive...
[34:04.490 --> 34:06.370]  and whistleblower protection statutes.
[34:06.530 --> 34:11.570]  All of which try to hold the executive, you know, to the fire on laws it's supposed to be enforcing.
[34:12.910 --> 34:18.390]  So I think affirmative disclosure... and of course the technological barriers to affirmative disclosure have fallen as well.
[34:18.530 --> 34:20.810]  So I think it's much more realistic than it was in 1966.
[34:21.910 --> 34:27.070]  And it wouldn't necessarily have the same distributional consequences that I was highlighting that FOIA has.
[34:27.450 --> 34:32.870]  FOIA sounds very attractively democratic and participatory.
[34:33.150 --> 34:39.990]  Anyone, any of us can submit a request on equal terms with the most powerful other person in society.
[34:39.990 --> 34:42.470]  And the agency equally has to respond.
[34:43.250 --> 34:47.830]  Congress kind of washed its hands of deciding who should be prioritized and just said anyone can request.
[34:47.830 --> 34:57.790]  In practice, that has yielded to corporate domination, I was suggesting, of FOIA, and therefore has proven kind of regressive.
[34:58.930 --> 35:09.030]  It sounds nice that Congress made no allocative decision, but, you know, in practice that's yielded a certain arrangement that I don't think people would have wanted beforehand.
[35:09.030 --> 35:21.910]  So again, so for that reason too, I think affirmative disclosure post everything within certain limits would be a way, not just to improve FOIA, but to get us kind of beyond FOIA in the next 50 years.
[35:22.230 --> 35:29.110]  David, what do you think of the argument that some people have made that relying on...
[35:29.110 --> 35:37.770]  So right now, reporters sometimes file these requests, and they file the request in part because they want a scoop, right?
[35:37.850 --> 35:39.610]  They get the information before anyone else.
[35:39.730 --> 35:40.770]  They rely on the...
[35:40.770 --> 35:43.330]  They used to get this information in hard copy.
[35:43.530 --> 35:45.770]  They would rely on that information writing their stories.
[35:45.910 --> 35:50.070]  Their story would go on the front page of the New York Times, and they would get promoted, right?
[35:51.030 --> 35:59.570]  And part of the incentive for individual reporters is to get exclusive access to this information.
[35:59.910 --> 36:13.550]  What do you think of the argument that it would be a mistake to go to a system in which everything was sort of released to everybody at once and affirmatively because it would get rid of that kind of incentive?
[36:13.790 --> 36:19.250]  I think it's a concern that could be alleviated with some design choices.
[36:19.250 --> 36:20.410]  So I think it's a real concern.
[36:20.630 --> 36:27.150]  Journalists seem to feel, in some cases, like they have a quasi-property right in the information they get out of FOIA.
[36:27.250 --> 36:27.750]  It's their scoop.
[36:27.910 --> 36:37.770]  And they went through the effort, and they had the kind of wherewithal to come up with the request, and maybe wait a long time, maybe file a bunch of papers in furtherance of their request.
[36:37.970 --> 36:46.790]  And then when they get the records, they should be entitled to get a scoop out of it and get a story, which, by the way, serves the goal of public debate about what the government is doing.
[36:48.470 --> 36:56.410]  So if you move totally to affirmative disclosure, do you kill that journalistic incentive to kind of ferret out information in a way that we might not want?
[36:56.530 --> 37:02.690]  I think the Obama administration has an interesting pilot program that suggests one partial solution.
[37:02.910 --> 37:06.370]  So it's called Release to One is Release to All.
[37:07.190 --> 37:14.350]  A handful of agencies have tried it out now since last year, and there's a debate about whether to make it the norm going forward.
[37:15.490 --> 37:20.090]  Historically, under FOIA, if you request records and you get them, they're yours.
[37:20.310 --> 37:23.310]  And the government doesn't necessarily make them available to anyone else.
[37:23.450 --> 37:28.610]  And this was a boon to some journalists and academics who, you know, felt like they had almost an exclusive.
[37:30.070 --> 37:31.410]  Release to One is Release to All.
[37:31.410 --> 37:37.090]  The way it's working is once any requester gets anything from FOIA, it goes up on the agency website.
[37:37.590 --> 37:46.090]  And journalists, at least a number of them, have been quite opposed to Release to One is Release to All as undercutting this incentive structure that we were just discussing.
[37:46.770 --> 37:50.410]  I think that there, I think there's an easy solution, which is just a time lag.
[37:51.410 --> 37:55.970]  Do release it to the public, but, you know, maybe lag it, whatever seems appropriate, a week, a month.
[37:56.330 --> 37:57.210]  So the journalist has some...
[37:57.210 --> 38:06.230]  I think there, you could satisfy both sides with Release to One is Release to All just with a time lag.
[38:06.810 --> 38:25.630]  And then I'll just say, on affirmative disclosure, just putting stuff up on websites and seeing what happens, I guess I'm not so pessimistic that an industry wouldn't develop, just as other industries have developed around FOIA, that could, you know, regularly monitor agency websites,
[38:26.450 --> 38:29.490]  figure out how to make sense of the data they put up there.
[38:30.650 --> 38:45.530]  I just imagine a new breed of journalism, it's already starting to develop, but would develop even more fully, that would make use of affirmative disclosure, because, you know, data is going to go up on the FDA website in huge amounts, and the general public won't know what to make of it,
[38:45.670 --> 38:53.690]  and we'll need mediators who can make sense of that, whether it's old school journalists or bloggers or experts in the field.
[38:53.830 --> 39:05.330]  So I imagine just affirm... if we really moved in a big way to affirmative disclosure, we would generate a new kind of journalism and maybe even industries around making that intelligible to all of us.
[39:05.330 --> 39:14.170]  So I'm less pessimistic than I think some of the journalistic attacks are on getting away from the request-driven model.
[39:14.330 --> 39:35.650]  I should also add, in case the reason that we should stick with FOIA as it is is because it's helping journalism so much, I think it bears keeping in mind that journalists are now less than 5% of the requester pool at most agencies, and routinely say that the delays under FOIA just make impossible a lot of the stories they want to write.
[39:36.590 --> 39:42.070]  Still, some of the important stories get written under FOIA, but we shouldn't paint too rosy a picture of the status quo.
[39:42.790 --> 39:48.210]  You know, I think that there's... this is sort of a broader point.
[39:49.210 --> 39:52.870]  There's the rhetoric of transparency, and then there's transparency in practice, right?
[39:52.870 --> 39:59.190]  And transparency in practice is often messier and uglier than the rhetoric would, you know, have you believe.
[39:59.990 --> 40:02.350]  This is one respect we've just been talking about.
[40:02.510 --> 40:07.190]  You know, journalists actually want exclusive access, at least for a time, to this kind of information.
[40:07.350 --> 40:08.290]  NGOs do, too.
[40:08.490 --> 40:14.630]  You know, certainly the ACLU sometimes files FOIA requests because it wants information before anyone else has it.
[40:17.730 --> 40:23.930]  And, you know, but I also... I've also noticed a different phenomenon that's sort of related.
[40:26.070 --> 40:39.630]  So when we would get exclusive access to information, we would sometimes turn it over to a media organization, say the New York Times or the Washington Post or the Intercept or whatever it was, and say, you know, we have this information that we got through FOIA,
[40:39.770 --> 40:53.070]  here's what we think it means, but, you know, you guys do your own investigation, and they would take time, because they had an exclusive through us, they would take time to do a deep story about that particular topic.
[40:54.310 --> 41:00.370]  If everybody gets the information at once, nobody has the same incentive in taking the time.
[41:00.550 --> 41:05.630]  The incentive is to write something really quick, to be the first to write about it.
[41:07.050 --> 41:15.810]  But the incentives change in a way that, you know, on balance, it's hard to know, you know, how that's going to affect the quality of the information that we ultimately get.
[41:17.450 --> 41:20.830]  So that's not to say that, you know, you're wrong.
[41:20.850 --> 41:25.410]  I just think that it's hard to predict how it plays out.
[41:25.530 --> 41:29.670]  Yeah, I mean, I think that's useful to maybe push against some of what I've been saying.
[41:30.270 --> 41:36.890]  You may have heard the distinction between transparency and knowledge that some theorists have introduced.
[41:37.210 --> 41:50.090]  And I think that the argument for FOIA, at least in part as it operates now, would be, it's not so great on transparency, just production of information in the public sphere, as we might think or want.
[41:50.390 --> 42:04.610]  But maybe it's not so bad for knowledge, in that those people get stuff under FOIA, at least the journalists in the civil society requesters, are highly motivated to let you know what they found and to, you know, kind of maybe even get credit for finding it.
[42:05.010 --> 42:06.970]  And that conduces to public knowledge.
[42:06.970 --> 42:12.990]  That is like actual comprehension and ability to act on information, not just the raw data itself.
[42:13.070 --> 42:20.510]  Although, again, I think the pathway from transparency to knowledge can be closed even without some of the way FOIA operates now.
[42:21.010 --> 42:22.950]  Two more minutes and then four minutes Q&A.
[42:22.950 --> 42:23.270]  Great.
[42:23.670 --> 42:41.790]  So let me just, another point is sometimes made about FOIA, which I don't think really helps its case, I'll know quickly, which is there's been another kind of use of FOIA that's really blossomed in recent years, which is its use in removal proceedings for people facing deportation,
[42:41.850 --> 42:43.330]  immigrants facing deportation.
[42:44.050 --> 42:49.910]  They don't have a right of civil discovery when they're trying to fight deportation.
[42:49.910 --> 42:55.950]  And lawyers in that area have figured out you can request through FOIA what does the government know about you or your situation.
[42:56.350 --> 43:01.550]  And so there's now a very large volume of such requests being filed with the Department of Homeland Security.
[43:01.870 --> 43:07.910]  And so at least, you know, certain people with certain views on immigration say this is something to celebrate about FOIA.
[43:07.910 --> 43:11.110]  And I do think FOIA is serving an important purpose there.
[43:11.190 --> 43:12.850]  I would just say, I don't think it's an argument for FOIA.
[43:13.290 --> 43:20.370]  That's an argument for discovery rights in those proceedings if we think we should, we make that decision as a society.
[43:20.530 --> 43:27.730]  So FOIA is like patching up little other parts of society where the laws aren't necessarily there.
[43:27.730 --> 43:33.710]  And I don't think that's necessarily speaks to FOIA's virtues as much as to other areas we might think about.
[43:33.950 --> 43:36.410]  And maybe I'll end with this before we go to Q&A.
[43:36.550 --> 43:41.310]  Jamil and I have talked a bunch about FOIA, what we might call FOIA triumphalism.
[43:42.330 --> 43:54.290]  There are a lot of judicial opinions that lionize FOIA as Bernie said, the Supreme Court once said, a structural necessity in a real democracy.
[43:54.290 --> 44:01.910]  And you can find a lot of other quotes like that from Congress and from journalists and from the courts.
[44:02.190 --> 44:07.350]  And I would just suggest that FOIA is not a structural necessity in a real democracy.
[44:07.630 --> 44:10.830]  Some kind of transparency forcing mechanisms are.
[44:11.430 --> 44:12.250]  But there are many.
[44:12.470 --> 44:13.250]  There are leaks.
[44:13.490 --> 44:15.390]  There are whistleblower protection statutes.
[44:15.570 --> 44:17.630]  There's old-fashioned congressional oversight.
[44:17.750 --> 44:18.750]  There's affirmative disclosure.
[44:19.050 --> 44:21.210]  There are a lot of other options out there as well.
[44:21.990 --> 44:24.050]  FOIA is just one of many mechanisms.
[44:24.050 --> 44:27.670]  That can get at this goal that is so widely shared.
[44:28.130 --> 44:32.210]  So I worry a little bit about the kind of naturalization of the FOIA approach.
[44:32.390 --> 44:38.090]  Like this is the way to do transparency is through the FOIA requester based model.
[44:38.290 --> 44:39.970]  That's what we've exported to the rest of the world.
[44:40.130 --> 44:44.230]  And I think it has some predictable pathologies that the other models might not have.
[44:44.370 --> 44:48.030]  So I would suggest that we should celebrate some of what FOIA has done at year 50.
[44:48.030 --> 44:56.150]  But the triumphalist rhetoric about being a necessity and indispensable to our democracy obscures a lot of things that aren't working so well with it.
[44:58.270 --> 44:59.490]  Yeah, let's go to Q&A.
[45:00.350 --> 45:02.850]  Hello, my name is not on the microphone.
[45:03.110 --> 45:05.510]  Caitlin Henry and I am a FOIA attorney.
[45:05.810 --> 45:06.870]  Can you repeat that again?
[45:06.910 --> 45:08.250]  Oh, my question.
[45:08.510 --> 45:09.450]  That was a little intro.
[45:09.630 --> 45:09.890]  Your name.
[45:10.050 --> 45:10.150]  Your name.
[45:10.330 --> 45:11.230]  Caitlin Henry.
[45:11.390 --> 45:12.430]  I'm a FOIA attorney.
[45:12.870 --> 45:19.730]  And I'm wondering how you think the FOIA Improvement Act of 2016 will impact practices for requesters and for litigators.
[45:23.250 --> 45:29.750]  So, in its early versions, I think this act was great.
[45:30.030 --> 45:32.490]  It was going to make a real difference.
[45:32.690 --> 45:48.450]  There was a provision that would have strengthened the affirmative disclosure requirements, would have made clearer that Office of Legal Counsel memos, that our final memos are documents that were documents that had to be released under the affirmative disclosure provisions.
[45:48.450 --> 45:49.710]  All of that got stripped away.
[45:50.630 --> 45:58.330]  So, my impression now is that it's still a net positive, but it's not going to make the kind of dramatic difference that we hoped it would make.
[45:58.690 --> 46:07.490]  I don't know the details, but my vague recollection is that the changes it makes with respect to older documents are much more significant.
[46:08.970 --> 46:14.590]  I've never litigated a case involving documents that are 25 years old, but those are important requests too.
[46:14.750 --> 46:16.450]  And if it does make a difference there, that's great.
[46:16.910 --> 46:17.630]  I would agree.
[46:17.710 --> 46:20.970]  I mean, it cuts away the deliberative process kind of privilege after 25 years.
[46:21.070 --> 46:21.750]  I think that's useful.
[46:22.450 --> 46:28.350]  It codifies the rule that if three requests are on the same subject, the agency should make that available to everyone.
[46:28.790 --> 46:32.250]  That kind of already was the law, as I read the E-FOIA amendments of 96.
[46:32.430 --> 46:36.010]  I think it's useful to say it more clearly and again, but not a huge change.
[46:36.170 --> 46:44.650]  And it powers modestly this organization, the Office of Government and Information Services, OGIS, which is like a new mediator since 2007 for FOIA disputes in the executive branch.
[46:44.650 --> 46:48.090]  I think that's useful too, but also a modest reform.
[46:48.250 --> 46:56.010]  Maybe the most significant piece of legislation was at the very end, it says, and no funding will follow.
[46:56.570 --> 47:03.090]  And that being a huge problem for FOIA implementation, I think it won't make a huge difference, although maybe do some modest good.
[47:06.470 --> 47:06.990]  Hi there.
[47:07.330 --> 47:10.310]  Thank you both for the work you do and for coming and talking to us.
[47:10.550 --> 47:11.370]  It's really fabulous.
[47:12.070 --> 47:14.890]  This question I have might be out of scope, and if so, we can let it go, but...
[47:17.430 --> 47:22.150]  So Edward Snowden found, delivered to the world a lot of documents.
[47:24.890 --> 47:32.430]  And they were delivered in such a way that, I guess, even if he were dispatched, they would still be released on some timeline.
[47:33.230 --> 47:37.070]  They were encrypted and distributed, and certain people had privilege to see them.
[47:37.490 --> 47:42.390]  My question is, how much of that document stash have we seen?
[47:45.490 --> 47:47.770]  And how much is left to be revealed?
[47:48.010 --> 47:54.050]  And insofar as you're able to talk about it for yourselves, to what degree have you been able to touch those documents yourselves directly?
[47:55.170 --> 47:56.130]  If, you know...
[47:56.130 --> 47:57.770]  I don't know in zero.
[47:58.270 --> 48:00.690]  Yeah, I don't know the answer to that either.
[48:00.850 --> 48:02.510]  I mean, I know that there's more to be mined.
[48:02.610 --> 48:08.730]  I know that some of the media organizations have records that they haven't published or, you know, published about.
[48:08.730 --> 48:11.930]  But I don't have a sense of the scale of it.
[48:15.000 --> 48:19.460]  Where might we go to learn more about that scope?
[48:20.270 --> 48:22.750]  And maybe this is a question for Edward Snowden.
[48:22.800 --> 48:23.290]  I have a question.
[48:23.620 --> 48:24.580]  Okay, great, thanks.
[48:28.040 --> 48:28.520]  Hi.
[48:29.060 --> 48:31.100]  I'm somebody who's studying, like, archival sciences.
[48:31.440 --> 48:35.140]  And I feel like a lot of the conversation is always amongst ourselves.
[48:35.500 --> 48:38.960]  And we're not really tapping into the networks where...
[48:38.960 --> 48:42.060]  I'm demanding transparency and really getting that work done.
[48:42.580 --> 48:47.020]  Do you guys, like, work with archivists a lot to demand transparency?
[48:47.220 --> 48:48.580]  Because we are, like, the records managers.
[48:48.580 --> 48:52.140]  We are the people who, you know, have a lot of say in records collections.
[48:52.140 --> 48:54.820]  I don't know what kind of work you guys do with that.
[48:56.280 --> 49:00.180]  I don't personally, really.
[49:01.700 --> 49:06.540]  I do think it's a critical piece of not just FOIA but the declassification system.
[49:06.800 --> 49:11.440]  The National Archives and Records Administration administers declassification in large part.
[49:11.900 --> 49:16.120]  And there was this recent study that it's the unhappiest place to work in the federal government.
[49:16.940 --> 49:20.240]  Funding levels are really pitiful compared to the scope of the job they have to do.
[49:20.560 --> 49:23.240]  I don't think it's given nearly enough attention or resources.
[49:23.240 --> 49:35.820]  So, in the FOIA bureaucracy, the people who are the professionals implementing it traditionally didn't even have a recognized job category by the Office of Personnel Management.
[49:35.880 --> 49:37.780]  Only in 2012 did they get that.
[49:37.940 --> 49:41.600]  So, I think it's a bad reflection of our commitment to transparency.
[49:41.600 --> 49:44.840]  We don't treat well the frontline people who are administering these laws.
[49:44.960 --> 49:47.700]  And we haven't, particularly in declassification.
[49:49.020 --> 49:52.320]  Yeah, I've never had the chance to work with archivists.
[49:52.460 --> 49:53.920]  But that wasn't a decision.
[49:54.100 --> 49:54.860]  That was just...
[49:54.860 --> 49:57.160]  I feel like it's like an untapped allied resource.
[49:57.540 --> 49:57.640]  Yeah.
[49:57.800 --> 49:58.340]  That sounds right.
[49:58.960 --> 49:59.920]  Just two minutes, guys.
[50:00.060 --> 50:00.260]  Okay.
[50:00.360 --> 50:00.460]  Okay.
[50:01.480 --> 50:03.080]  It's probably our last question, I think.
[50:03.240 --> 50:03.560]  Oh, sorry.
[50:04.080 --> 50:05.640]  Question about balancing tests.
[50:07.540 --> 50:15.460]  So, I'm very much interested in how lenient or withholding the balancing test would be in practice.
[50:15.460 --> 50:16.820]  And I think...
[50:16.820 --> 50:23.740]  And I think Jamil, who brought this up, also implied that the courts tend to be very differential.
[50:24.180 --> 50:29.100]  And I would think even the most independent part of the executive branch would still be very differential.
[50:29.660 --> 50:38.220]  On the other hand, I don't think we want it applied in that kind of very withholding way.
[50:38.220 --> 50:57.600]  On the other hand, even though I love Jamil's work, but I think perhaps you would be inclined to be a lot more generous in giving out information than maybe a lot of the general public would be willing to.
[50:58.180 --> 51:05.460]  Although, of course, your work would very much benefit from a standard that was more in the middle as compared to the current reality.
[51:06.080 --> 51:25.220]  So, I'm wondering, since I don't see any branch of government being inclined to be too generous, I'm wondering about the possibility of setting a precedent by having people outside government unofficially model a more in the middle balancing test.
[51:25.840 --> 51:40.240]  So, like, and one place where that could be done is, like, in whistleblower disclosures, sample public opinion or deliberative public opinion to see, was this worth disclosing or was it not?
[51:40.740 --> 51:42.220]  Classic example, Chelsea Manning...
[51:43.100 --> 51:44.200]  We're at the end.
[51:44.360 --> 51:44.900]  You have to stop.
[51:45.000 --> 51:45.160]  Okay.
[51:45.300 --> 51:45.520]  Okay.
[51:45.720 --> 52:04.580]  So, the question is, would you be willing, as somebody who would benefit from this, would you be willing to cede some of the moral authority on what should be disclosed to people who might...to something closer to public opinion that might be more towards the withholding end than you would be,
[52:04.700 --> 52:06.420]  but less so than the current reality?
[52:07.220 --> 52:07.980]  There's a lot there.
[52:08.080 --> 52:21.220]  I'll say quickly, I do think that there's a story in the way FOIA has evolved about judicial incentives that no matter how...what you say in the law about what they should do, no deference, balancing.
[52:21.840 --> 52:24.660]  We can't have a ton of confidence that they will want to disclose things.
[52:24.760 --> 52:25.400]  The executive doesn't.
[52:25.700 --> 52:27.660]  And that counsel is moving to some other models.
[52:28.040 --> 52:31.540]  I do think it would help, at least at the margins, to have the balancing, but I think you're right.
[52:31.720 --> 52:38.300]  I would just say that on the...what about the public interest as proxied by what the people think in polls or what Congress thinks?
[52:38.900 --> 52:56.820]  I would just point to, in closing here, this great paper by Yochai Benkler where he argued recently that if people are prosecuted like, say, a Snowden for violations of the Espionage Act or other statutes for leaking, they could have as an affirmative defense a public interest kind of standard.
[52:56.960 --> 53:00.760]  And he proposed specific indicators of serving the public interest.
[53:00.860 --> 53:02.020]  Did a congressional hearing result?
[53:02.200 --> 53:05.120]  How much journalism resulted in other...I forget his exact list.
[53:05.220 --> 53:11.260]  But he tried to operationalize this idea of getting the public interest into court, in that case through a defense.
[53:11.400 --> 53:12.380]  But I think it's a promising idea.
[53:13.340 --> 53:13.660]  Okay.
[53:13.820 --> 53:14.420]  Thank you very much.
[53:14.600 --> 53:15.080]  Thank you.