The Lawfare Podcast: Government Use of Open-Source Information
Published by The Lawfare Institute
in Cooperation With
In front of a live audience at the Knight Foundation's INFORMED conference in Miami, Florida, Lawfare Editor-in-Chief Benjamin Wittes spoke with Hon. Kenneth L. Wainstein, Under Secretary for Intelligence and Analysis at the Department of Homeland Security; Jameel Jaffer, Executive Director of the Knight First Amendment Institute at Columbia University; and Lawfare Senior Editor Quinta Jurecic about government surveillance of open source social media.
Click the button below to view a transcript of this podcast. Please note that the transcript was auto-generated and may contain errors.
Transcript
[Introduction]
Kenneth Wainstein: Here's the problem with that
authority. I mean, do I want it, in many ways, no, I don't want it because it's
a hot potato. It's incredibly difficult, but it has to be done and, at the end
of the day, it's our job. And so, we just have to do it effectively. Effectively,
in terms of preventing violence and threats against our people, but, also,
effectively in a way that protects people's privacy and civil liberties. And
here's the problem, when you think about it, our mission, in fact, our limited
mission, is to look online for social media that might relate to threats to the
homeland, which is our intelligence space. And we can only either go to
publicly available, or we have to be overt about it. So, we can't lie about who
we are and get into a private chat room. We can't do that, right? We're very
constrained in where we can go. And the ironic thing is that each one of you
can be much more effective in doing this than we can because you don't have
those constraints. Private individuals, private organizations can do a lot more
than we can in the open environment.
Benjamin Wittes: I'm
Benjamin Wittes, and this is the Lawfare Podcast. We are to you from the
Knight Foundation's INFORMED Conference in Miami, Florida. We are on the stage
in front of a live audience, as you can hear. I am here with Ken Wainstein,
Under Secretary of DHS for Intelligence and Analysis, Jameel Jaffer of the
Knight Institute at Columbia, and of course, Quinta Jurecic of Lawfare
and Brookings. And we're here to talk about government surveillance of open-source
social media.
[Main Podcast]
Ken, I want to get right into it. You guys, the agency that you
head, has a very particular and unusual authority and mission, and it relates
to everybody's social media, and I want you to explain to us what it is, and
what you are and aren't allowed to do.
Kenneth L. Wainstein:
Okay, thanks Ben. Good to see everybody. It's a real pleasure to be here. Good
to be with this panel of distinguished thinkers in this issue. So look, the
Office of Intelligence and Analysis at DHS was stood up with the department 20
years ago. And it was stood up to address some of the deficiencies that the
intelligence and law enforcement communities had pre-9/11 that were laid bare
by 9/11. And one of them was the fact that there was inadequate information-
and intelligence-sharing between the federal authorities, intelligence
community, and federal law enforcement authorities, and the state, local,
territorial, tribal, and private sector partners out in the country. So, INA
was stood up to help to bridge that gap and share information, but, also to be
an intelligence agency.
We are one of the agencies within the intelligence community.
So we are responsible for all parts of the intelligence cycle: that's
collecting intelligence that's relevant to intelligence requirements, analyzing
that information, producing intelligence products, and disseminating it to our
partners. In this case, it's largely to our state and local, territorial,
tribal, and private sector partners. In the course of doing that, of course, it
is our job to be meeting the needs of our partners. And the needs of our
partners are broadly to protect the homeland, and to protect against threats to
the homeland that manifest out in the states, in the counties, in the cities
and jurisdictions where our partners have responsibility to protect their
people.
So, we have to collect, analyze, and disseminate the
intelligence that helps them. In terms of collecting, we get a lot of that
intelligence from our fellow intelligence community partners by way of
reporting from the CIA, reporting from the FBI about threats that might
manifest here in the homeland. We also collect some of our own intelligence.
And that's where we get to the issue that you're talking about. We collect that
in a variety of different ways. We can do interviews with people throughout the
country. But, also, we can look on the internet and we can look in social
media. And I've often likened what we do, and others like the FBI do, on social
media, sort of, to the police officer 50 years ago who is patrolling a town and
walking along looking for open doors or windows or anomalies that might suggest
something is wrong. That was pre-internet. Now, when you're trying to help
people protect against and anticipate threats to the homeland, oftentimes
you're not walking down a street looking for open doors; you're looking for
indications in social media about a coming threat.
And there's no greater, probably example, no example I think
that's sort of most vivid in its demonstration of this than January 6th. We all
know that January 6th-- prior to January 6th--there was a lot of discussion in
the open-source space and social media. And a lot of it related to
anticipation, preparations, and discussion of actual steps to perpetuate
violence up on Capitol Hill. And those threat indicators that were online were
viewed and the question was sort of how to get them out and give sufficient
alert to the people who could possibly have prevented January 6th from
happening. That didn't happen. And so, in the aftermath of January 6th, there
had been a lot of discussion about what collection should have happened, what
alarm or what alert should have been given. And we are one of the organizations
that looked at that open-source area, the FBI was one of the organizations that
did. And I think it's important to look at that episode because if nothing
else, January 6th demonstrates the need to do this kind of surveillance, to do
this kind of review, because we can't let the next January 6th happen if those
indicators are out there in social media.
That is what we do, and we do it in a very constrained way. In
fact, I've often said that we are the kinder, gentler intelligence agency. And
I say that because we have the most constraints on us, the most limitations on
us, the most reporting requirements on us, which is exactly as it should be.
Because our job is to do intelligence here in the homeland, where we have
constitutional rights and privacy are paramount. And so we have to do this
mission and we look at the social media environment with those constraints in
place. So that's the challenge--that's the mission we have, but also the
challenge we have.
Benjamin Wittes: All
right, so Jameel, you have been in one way or another thinking about this issue
since the creation of the department after 9/11. A lot of people react to this
as like, "Hey, it's available for anybody. Why shouldn't an intelligence
agency take a look at it?" A lot of other people look at it and say,
"Oh my God, the government tracking social media. That's really
scary." What is your instinctive reaction and, for that matter, your
considered reaction to the idea that some government agency, whether DHS, INA,
or the FBI, is watching for "Storm the Capitol" hashtags?
Jameel Jaffer: So,
what is my instinctive reaction? My instinctive reaction is I don't like
anything about it. I don't even like the word "homeland." My
considered reaction is--let's just take a step back first. Okay, so, I think
there are very good reasons why--and I'm sure Ken would agree with me--very
good reasons why a democratic society should place strict limits on government
surveillance.
If you want a democracy to work, you need people to feel
comfortable communicating with each other, privately and publicly. You need
them to feel comfortable organizing into political groups, taking political
action, collective action, protesting. And people who think or know that the
government is tracking their expressive or associational activities, are going
to be chilled from doing all of those things. That's just Civil Liberties 101,
right? People whose activities, expressive and associational activities, are
tracked, will be much less likely to participate in the activities that are
necessary for any democracy to function and every authoritarian regime knows
this. Every authoritarian regime knows that surveillance can be a powerful tool
of censorship.
You don't actually have to look to authoritarian regimes
overseas to see how all this can go wrong. It's gone wrong in the United
States, in the homeland, if you like, many, many times, right? In the 1950s and
'60s, intelligence agencies abuse their surveillance powers to kneecap civil
rights organizations. After 9/11, the NYPD tracked countless Muslim groups in
New York City, including Muslim student groups. Not because they'd done
anything wrong, but because the NYPD just wanted to map out what Muslims were
doing in in New York City. There are lots of other abuses over the last 50
years. You don't, again, have to look overseas to see the ways this kind of
thing can go wrong.
Now, all of that said, there are circumstances in which I think
we need to accept that, even in an open society, the government is going to
have a legitimate reason to track what citizens are doing, including their
expressive and associational activities. But I think we should approach that
with an awareness of past abuses, and approach this with the view that these
kinds of surveillance activities need to be surrounded by safeguards. It can't
be that executive branch officials have broad discretion, unsupervised by
another branch, unsupervised by the, courts to engage in this kind of
surveillance. We need to think about not just what they're collecting, but what
they're doing with the information, how long they're retaining it, who it gets
disseminated to. So, I guess, that's sort of the way I come at the set of
questions.
I also think that this is the wrong moment to be talking about
whether the government's surveillance power should be expanded. I think we
haven't yet fully responded to the abuses that were disclosed by Edward
Snowden. I think that those abuses, which were written about endlessly by
American newspapers, those documents and those stories showed that the
safeguards we have in place right now are insufficient to keep these
surveillance powers in check. That surveillance powers we thought were granted
for relatively limited purposes were ultimately used for much broader purposes.
The institutional checks and balances that were supposed to keep all of this in
within limits, failed over and over again.
And when I say failed, one way to measure the failure is just
to look at what happened after those surveillance programs were disclosed. And
one of the things that happened is that Congress, once the public knew about,
for example, the call records program, which was the program, that involved
collecting metadata from hundreds of millions of Americans' phone calls. Once
that was public, Congress felt obliged to rein in those surveillance
activities. So, they were allowed to persist only because, only while they were
secret. Once they were public Congress felt obliged to rein them in. So, I
think we haven't addressed those failures.
I also think that, as many of us are thinking about, what life
might be like under a different administration, a new administration, or the
same administration we had a few years ago, one of the questions we should be
asking is, what kinds of surveillance, even if you trust this administration
with these surveillance powers, what kinds of surveillance powers do you want
in the hands of the next administration?
So, I guess, if I were framing this conversation, I would frame
it in exactly the opposite way. I think the right question to be asking is not
what additional surveillance powers should the government have right now, but,
rather, what new safeguards do we need to ensure that these powers are used
consistent with the public interest and consistent with our values?
Benjamin Wittes: So,
fair enough. And to be clear, I don't think anybody is actually talking about
expanding these authorities. The DHS just got through fending off an amendment
from Marco Rubio, a local senator, to substantially curtail these powers. I
actually think the question, as currently, framed is very much in the direction
that you would want it to be.
But, I fear that the de facto compromise that has been reached
over the last few years is to have these powers but then not use them. And I
want, Quinta, you to sort of--this creates a kind of a weird, almost like
pincer action, where on the one hand, Congress has given this set of
authorities to INA. On the other hand--and to the FBI. On the other hand the
INA, as I will ask Ken to develop, is not really staffed to actually use them
in an active way, and then, the FBI more or less chooses not to. And so, you
end up de facto with very much the environment that, Jameel, you're describing,
which Congress then gets upset about when January 6th happens.
So, Quinta, walk us through the actual--who in the government
is reading whose social media and under what circumstances?
Quinta Jurecic: So,
let me focus specifically on the story of January 6th because I do think that
that is very telling in a number of ways. We now have a pretty enormous body of
reporting from Congress, from various inspectors general, less so actually from
the House Select Committee on January 6th, which kind of set this part of the
story aside. But, through a lot of this investigative work, we have a sense of
what was going on at DHS INA and at the FBI in the run up to January 6th. And
what you see essentially is an environment where, as you say Ben, these are
agencies that pretty clearly have some level, that is constrained, of authority
to monitor public social media posts, so, your posts on Twitter. The problem is
that, as you hinted, these agencies appear to have been really unwilling to
actually use that authority. And to the extent that analysts were using that
authority to review material, they weren't recording it; to the extent that
they weren't recording it, they weren't passing it along the chain; to the
extent that it was passed along the chain, it wasn't shared inside or amongst
agencies.
The reasons for why that might be are complicated, and I'll get
to that in a little bit, to the extent that we have a sense of that. But, what
you really do see if you look through these Inspector General reports, these
reports from Congress, is people who are aware that something bad is going to
happen because they have an internet connection, just like everybody else. And
like a lot of us in the run up to January 6, we're looking at Twitter and
looking at Facebook and looking at Reddit and could see that something was
going to happen. And yet, are very gun shy, for lack of a better term, about
moving forward. And turning that material into something that those agencies
could actually do something with. And as an end result, you reach the situation
where DHS and the FBI are, kind of, frankly, caught with their pants down. You
would have expected that these agencies who are very much tasked with
protecting the U.S. Capitol would have seen something like this coming, but
we've all seen the footage of people rushing into the building, and it took
hours and hours before there was an adequate response.
So, then let me move a little bit to the kind of after-action
report. What you see, and I'll focus a little more on the, on the Bureau here,
just because I've focused a little bit more on it in my own work--is that FBI
Director Christopher Wray, and then Assistant FBI Director for Counterterrorism
Jill Sanborn go in front of Congress after January 6th and are asked, "Do
you have the authority to look at public social media posts? And if so, why
didn't you?" And their answers, I would argue, are pretty misleading. They
kind of suggest that there's a higher level of authority that is required than
the Bureau actually had in those circumstances. But, if you actually look at
the very, very in depth manual that was written in response to the abuses of
the FBI in the '50s and '60s that Jameel is pointing to, it's pretty clear that
they explicitly do have the authority and the ability to look at, again,
publicly available social media posts without jumping through the particular
set of hoops that Director Wray seemed to suggest before Congress. And there's
also, in addition to that, there's some reporting from the Senate Homeland
Security and Government Affairs Committee that suggested that the Bureau had
actually begun some of the investigative steps that would have allowed it to
have even more authorities.
So, it's not a question of not having the authority, it's a
question of having the authority and then not doing anything with it. This
leads us to the question, why not do anything with it? I think this is actually
something that we have much less of a sense of. It's much more speculative. I
will say, there's a book that recently came out from Ryan Reilly at NBC
News--he's been doing a lot of reporting on the fallout from January
6th--that's called Sedition Hunters, that makes the argument that
essentially the Bureau was scared about what would happen if it went after far-right
extremists. Specifically, and if you look too closely at supporters of the
president, who is, after all in charge of the FBI ultimately, and there was a
level of anxiety about what would happen if that moved too far forward. You
also see a level of anxiety, again, according to Reilly's book and according to
Inspector General reports, from INA about perceived overreach by that agency in
response to the protests in Portland earlier that year, which I think gets
exactly to some of the abuses that Jameel was touching on.
And so, you end up, as you say, Ben, with this kind of pressure
from both sides, where the agencies nominally have this authority. In the case
of Portland and other George Floyd protests, they did use it to some extent in
ways that I would argue were actually abusive, and then become, sort of pull
back and are unwilling to take that step going forward.
Benjamin Wittes: All
right. So, can--Jameel doesn't want anybody to have this authority. Chris Wray
wants anybody but him to have this authority. I'm caricaturing both's point of
view. Is this an authority you actually want?
Jameel Jaffer: But
before you answer, Ken, can I just ask a question about, like, what authority
exactly are we are we talking about? Because obviously there are many versions
of this.
Benjamin Wittes:
Sure.
Jameel Jaffer: And it
might matter which one we're talking about.
Benjamin Wittes:
Right. So, so right. So, let's say that the authority in question is the
authority to review open-source social media material for violent threats, not
to compile dossiers on any individual, not to track networks, but to, you see
in the newspaper that there's a "Storm the Capitol" hashtag trending.
What can you do with that? Right? And, and I, I think Quinta's account of,
first of all, if you're, if you disagree with Quinta's account, that's, love to
know that, but I'm, my impression is that it's quite accurate that this was
something that the Bureau sort of disclaimed. How many people do you have doing
this sort of thing and, and what is, is this a mission that DHS actually wants
to have?
Kenneth Wainstein: Okay, you got a few questions in
there.
Benjamin Wittes: Yeah, yeah. Take such time as
the gentleman may require.
Kenneth Wainstein: Thank you. Quinta's article, I read
that just last night, it's excellent, excellent sort of dissection and
description of what happened and how that square, didn't square with the
relevant, guidelines under which the FBI operated, and the same thing could be
done with our conduct throughout the January 6th. Let me just step back and try
to generally answer this. The reason I started with January 6th is not because
it's one that, well, partly because it's one that everybody knows, but also
because I think it so clearly demonstrates the need for this work to be done. I
mean, as Jameel said, there are circumstances where the government needs to be
able to do this. And so, I think that's just an example, a good starting point,
okay, that human smugglers online and them trying to promote their, their
operations online. Obviously, there are areas where I think we'd all agree that
government authorities should have the ability to see that information, act on
it in appropriate circumstances. So let's start with that. Here's the problem
with that authority. I mean, do I want it? In many ways, no, I don't want it
because it's a hot potato. It's incredibly difficult, but it has to be done.
And at the end of the day, it's our job. And so we just have to do it
effectively, effectively in terms of preventing violence and threats against
our people, but also effectively in a way that protects people's privacy and
civil liberties.
And here's the problem. I mean, when you think about it, our
mission, in fact, our limited mission is to look online for social media that
might relate to threats to the homeland, which is our intelligence space. And
we can only either go to publicly available information, or we have to be overt
about it. So we can't lie about who we are and get into a private chat room. We
can't do that. Right? We're very constrained in where we can go. And the ironic
thing is that each one of you can be much more effective in doing this, then we
can, because you don't have those constraints, private individuals, private
organizations can do a lot more than we can in in the open environment. Which
is ironic of course, since we're the ones with the mission for to protect the
homeland, but understandable for all the reasons that Jameel laid out about and-
Benjamin Wittes: And to be clear, do you object to that?
Are you asking for authorities that you don’t have?
Kenneth Wainstein: No, we're not asking for further
authority.
Benjamin Wittes: Okay, but, but, but hang on, so I don't
want to, I don't want to make the case for authorities you don't want. But it
seems to me that we are in a situation that's a little bit absurd, which is to
say that you have one agency that doesn't want to do this work at all, which is
the Bureau, which is has 35,000 people and it's a very serious manpower
operation. It has field offices all over the country. And then you have one
agency that is tiny. How many people do you have on this?
Kenneth Wainstein: We have like a couple, three dozen, I
think, all together.
Benjamin Wittes: All right, so three dozen people in a,
with a highly constrained set of Attorney General's guidelines. And so it seems
like the, we've set up a situation in which it is essentially impossible for
this to be done effectively. Whether the answer to that is more people, or to
let the agency that's bigger do it, or maybe to get out of the business of
doing it at all, which is Jameel's, I'm, I'm sort of agnostic, but it seems
like there's, it seems like we've built a situation that can't be successful.
Kenneth Wainstein: Yeah, I agree with you. You know I'm
saying, we're not, we're not urging, we're not asking Congress for INA to have
additional authorities at this moment. I do agree, though, that we need to look
across the whole enterprise, the Bureau and us, see that there is a need to
deal with conditions of violence in particular in the open source arena, and
then make sure that we, that we, the government, have the capability to see
that, detect plots before they happen, and then get the intelligence out to
those actors who can prevent those plots from becoming reality.
We, you're right, we don't have the size, we don't have the
staffing, we don't have the authorities, frankly, to effectively do that. Now,
in our case, it's a balancing act, right? We were set up against the backdrop
of the Bureau and other agencies that had authorities. And as Quinta said the
Bureau does have the ability to look in social media in the first instance. They
then have the, once they establish that there is the predicate for an
investigation or there are different degrees of investigation, each of which
allows more intrusive tools, investigative tools. So they can do that. We don't
have those investigative tools. We don't have, we can't get search warrants. We
can't do 702 surveillance. We can't do, you know we can't arrest people. So, we
are limited to just reviewing what we see in social media and then reporting on
it. And keep in mind, when we look in social media, we can't just track Ben
Wittes. We can't just say, let's just say-
Benjamin Wittes: Well, you did once.
Kenneth Wainstein: There was that. We'll talk about that
later. But we can't, we can't just put a person in and say, okay, let's just
follow that person. We have to look only in areas where we have a reasonable
belief there's going to be information that relates to a threat, that relates
to one of our missions, either a departmental mission or a national mission. And
look, at the end of the day, let me just step back for a quick second, then
I'll wind up on this. This sounds like it should be easy, but it's incredibly
difficult, because just look at the domestic terrorism area. I think that's the
best example. We say, we don't look in the, in social media anywhere where we
don't have a reasonable belief that there's discussion of, coordination of,
what have you, relating to violence.
The problem with domestic terrorism is that the vast majority
of it emanates from political belief, political rhetoric and expression. That
is the most core protected type of speech we have. So, what we have to do is,
we might go to a website where we know that there are domestic violent
extremists, and there will be discussion that sounds quite extremist, but
that's fully protected unless it crosses that line into discussion of violence.
And that's what makes this such a difficult, sort of constitutionally perilous
undertaking. But it has to be done, and the fact that it's perilous, yet it has
to be done is the reason why we have such tight constraints on us.
Benjamin Wittes: All right, so I want to give Jameel a
chance to respond to this, but before I do, I actually want to take the
temperature of the audience here, because it seems to me a reasonable person
listening to this could have one of, one of two reactions, and I'm just
interested from a show of noise which one how many of you have.
One is the government shouldn't be reading public social media
without a criminal predicate. How many of you react in that way?
Jameel Jaffer: Ben, Ben, hold on.
Benjamin Wittes: Sorry?
Jameel Jaffer: Okay, so, it matters what the
surveillance looks like, okay? We're talking at a very high level of
generality. At some level, I mean, who's going to disagree with the proposition
that the government should be able to go on social media and see if anything
somebody's planning something terrible?
Quinta Jurecic: Well, Christopher Wray.
Benjamin Wittes: Well, yeah, I mean, the FBI director.
Jameel Jaffer: Christopher Wray may misunderstand First
Amendment limits in this context, I don't know. But at some level of
generality, I think I certainly would agree, I think most civil libertarians
would agree, that the government needs to know what's going on in the world,
and looking at public social media posts can be a way of learning what's going
on in the world.
But there's a difference between, there are different steps
that the government can take, right? One is there's a difference between
looking at social media posts generally and then following a particular person.
There might be a difference between following a particular person disclosing
that you're a government agent and following a particular person pretending
you're somebody other than a government agent, right? Infiltrating a political
group might raise additional concerns. And I think that you have to have a
graduated system here where the safeguards are commensurate to the intrusion
into speech and privacy, right?
I suspect that most people at that level of abstraction were
probably all on the same page. The question is what, what level of intrusion
can be justified on the basis of, you know, what kind of predicate do you need
to justify any particular level of intrusion?
Benjamin Wittes: Okay? So let, let me formulate. And let
me formulate it a little differently: One reaction possible reaction to the
events of the last several years is to say there should be less public social
media monitoring by law enforcement outside the context of predicated
investigations. Another possible reaction would be to say there should be more
monitoring of social media for intelligence purposes outside of the context of
predicated investigations. Is that a fair?
Jameel Jaffer: I still don't like it, but I've
interrupted so much.
Benjamin Wittes: Well, let's, let's see who here thinks
there should be less public monitoring of social media outside of the context
of predicated criminal investigations.
Audience: [Scattered applause.]
Benjamin Wittess: Alright, who here thinks there should,
the surveillance crew, thinks there should be more public monitoring of social
media outside, for intelligence purposes, outside of the context of predicated
investigations?
Audience: [Silence]
Benjamin Wittes: Wow, Ken.
Kenneth Wainsten: Oh, I'm taking it personally. Thank
you. Thank you.
Benjamin Wittes: The smattering of applause for Jameel's
position carries the day.
Jameel Jaffer: Well, okay, so, so can I say a couple of
things here?
Benjamin Wittes: Yeah, please.
Jameel Jaffer: So, we have a case at the Knight
Institute where we're challenging a State Department rule that requires visa
applicants to register their social media handles with the State Department.
And this has entirely predictable results. Some people who would otherwise
apply for visas to come to the United States now think twice about doing it.
Some people don't apply at all, especially if you're, if you're a, say a Saudi
dissident who uses a pseudonym on social media, you might be hesitant to turn
over your real identity to associate yourself with your pseudonym in your
application to the State Department.
And further consequences, people in the United States who would
otherwise have had the opportunity to engage with people from other countries
maybe lose out on that, you know opportunity sometimes. So we were concerned
about the First Amendment implications of this, surveillance and urged the, the
Trump administration not to go down this road, urged the Biden administration
to reject the rule after the Trump administration had adopted it, were
unsuccessful in these advocacy efforts. And then so finally filed a FOIA
lawsuit to try to get internal government reports about the effectiveness of
this surveillance. And our FOIA lawsuit was unsuccessful at unearthing the
actual report, but we did get email exchanges in which government officials.
exchanged views about this social media surveillance and the underlying report,
which we didn't get.
And documents from the office of the
Director of National Intelligence, email exchanges from that, that, that
office, characterized the value, characterized this social media surveillance
program as having, quote, no value, no intelligence value. This is a program
that has been in place now for six years or something like that. And the
government itself, the ODNI, characterizes the program as having no
intelligence value. Now, I don't want to suggest that this means that no form
of social media surveillance would be effective or useful, but I think that
there is this instinct after any intelligence failure to think that the problem
was, we didn't have the information we needed, when often the problem is, we
didn't share the information, or we didn't know what the info, we didn't
actually look at the information, or we didn't analyze it in the way that we,
we should have. And I started this whole conversation by saying we should
approach proposals for expansion of surveillance authority, and I understand
that that's not that's not perhaps what's being proposed here, but we should
approach those with a degree of skepticism, and this is just another reason to
approach those those kinds of proposals with, with skepticism.
Benjamin Wittes: All right, we are going to go to
audience questions momentarily. But before we do, Ken, I just want you to
address the question, was there, was there information in the case of January
6th that INA had available that didn't get shared, or was this an information
deficit problem? And Quinta in the case of the Bureau, I think the answer to
that question is a little bit more florid than in the case of INA, right?
Kenneth Wainstein: Well, look, I, I've just had the
benefit of going back and looking at the various reports, and I think bottom
line is, for both organizations, the FBI and DHS, and you can read this in the
recent HSGAC report, is that they did have access to information that would
have suggested that there was a possible looming threat, and that that
information did not get promptly conveyed to the people who could act on it.
And look, so that's been a major focus of ours since I got
there, and major focus of the Secretary and the Deputy Secretary, has been to
make sure that we're doing everything pursuant to the guidelines, but we're
also making sure to get the information out. And so, for example, and this is a
much different situation, when the Dobbs decision came out, the abortion
decision, came out by surprise one morning, it was a Friday morning, and there
was great concern that that could foment potential violence, and there have
been indications that there, there were plans to possibly do something when
that came down on both sides, right. And so by that afternoon, we had corralled
whatever our intelligence we had about, both from our partners, but also what
we saw in open-source media, and put something out to all our partners saying
here's what we're seeing and what we're not seeing. Fortunately, as you recall,
things were quite peaceful and protests were dignified and respectful.
That being said, we were able to get out, this is the
intelligence picture, that process is now much, much more in place than it was.
And I think there were some very difficult circumstances that you all have
alluded to in January 6th that hopefully won't replicate.
Benjamin Wittes: Quinta?
Quinta Jurecic: I think that on the part of the Bureau
in advance of January 6th, it's both. It's a failure to process information
that was sitting around waiting for someone to pick it up, and it was a failure
to do anything with the information once it was presented.
I also think it's, it's worth connecting what we're talking
about here to some of the dynamics that we've alluded to over the course of the
day in terms of political pressure on people who are looking at questions of far-right
extremism online. I think if, if you look at the dynamics, and within the
Bureau as has been reported in these documents from Congress before and then
after the 6th, what you see is an organization that essentially was
successfully bullied into not doing anything. I don't think I'm getting out
over my skis too much in saying that, and an organization that has been,
continued to be bullied into not doing anything, or not exercising its
authorities within the appropriate First Amendment limits because of concern
that if anything is done that the director will be held before Congress. And so
I think that that is an additional dynamic that is worth keeping in mind here,
and adds an additional and important layer of complication to everything that
we've said.
Benjamin Wittes: All right, a question from over there.
Audience Member: Justin Henderson, Tech Policy Press. So
not to, kind of redirect the conversation away from social media, but another
thing that Christopher Wray is mealy mouthed about in front of Congress is the
FBI's use of commercially available information from data brokers and the like,
which arguably, much larger ingestion going on potentially in the federal
government of that information than perhaps social media. And I know we have a
partially declassified ODNI report on this last year, et cetera. I just put
that to the panel, anything to say about commercially available information
from data brokers and the extent to which, some of the topics that you've
discussed here perhaps should apply to that question?
Benjamin Wittes: Yeah, so super interesting question.
Jameel, first of all any thoughts on ingestion by the Bureau and Ken, what are
the rules affecting you guys for commercially available information? Jameel
first, or?
Jameel Jaffer: I mean, I guess I would just say that I
think this is another area where the law hasn't kept up to new technology. And
if the FBI searches, wants to search somebody's home, they need to get a
warrant to do it. But if they want the same information, if they access the
same information by purchasing it from a data broker there's no, no judicial
oversight, no probable cause requirement.
Even the issues we were talking about earlier, I would say that
the, the, part of the problem here is that the law has not kept the, the First
Amendment hasn't kept up to, kept up with new technology. And this kind of
social media surveillance looks so different from the kinds of surveillance
that the Supreme Court was thinking about, in cases decided 50 years ago, like Laird
v. Tatum which involved an army surveillance program in which army
officials were, were taking news clippings, clippings from the newspaper about
a particular set of individuals and the court said well, this kind of
collection of news clippings doesn't amount to a First Amendment injury.
And now we're talking about these massive dragnet programs
where the government officials are tracking individuals over long periods of
time, and your, your information is kept indefinitely in these huge government
databases. And the same cases that the court decided 50 years ago, those are
the cases that have sort of created the legal, that, that still form the legal
architecture in which government officials are making these decisions. And
there has been, I mean this would be a long digression, but there has been a
kind of updating of the Fourth Amendment over the last few years with cases
like Jones and Carpenter, I think we're still waiting for the
court to do that with the First Amendment.
Benjamin Wittes: Ken, what are you, what, are you
allowed to buy all the stuff that you're not allowed to collect yourself?
Kenneth Wainstein: Look, there, the rules are taking
shape in terms of what, what publicly available and commercially available
information can just be purchased and adjusted. And, and there are a lot of
programs, both in our department and elsewhere, that rely on that for
legitimate national security and law enforcement reasons. But I do want to just
sort of add on, or, or, carry on Jameel's point, he's right that the law in
terms of the court decisions and the analysis of the constitutional
implications of these new, new technologies and new issues hasn't kept up with
today's reality. But the same with legislation, and I think Congress needs to
wrestle these issues to the ground, and I actually applaud them doing it,
whether it's commercially available information, whether it's our access to
social media, whether it's our old friend 702, the authority which is up for
reauthorization.
Benjamin Wittes: Pushed back till April.
Kenneth Wainstein: Pushed back till April, but it was, I
mean, it's a classic example of technology outpacing a law that was passed in
the late 1970s based on the technology of the time. In 2008, Congress, in a
really admirable show of what I think was good, strong bipartisan national
security legislation, rolled up their sleeves, hammered out a reform of the
FISA statute that brought it up to date, put in place a lot of good protections
and safeguards that weren't in place before. And now it's been reauthorized a
number of times and it's back up and they're looking at it again. I applaud
that. I applaud the fact that they're looking at it again. But as you know, in
the 702 context, it's absolutely critical that it, for national security, that
it be continued. I think that same congressional interest, as much as it, as
much as I'd prefer not to be up in Capitol Hill having to deal with the
situation you talked about last year where there was an attempt to curtail our
collection efforts, at the end of the day that was a healthy exercise.
Congress scrutinized what we were doing, scrutinized the
safeguards we had in place, whether they were sufficient, considered
legislation, ultimately backed off of that legislation, but gave us a package
of safeguards that really enhanced the oversight of our operations. That's good
legislation. I think we need more of it.
Benjamin Wittes: Last question over here.
Audience Question: Hi there, I am Bea Covello. I'm the
director of emerging technologies at Aspen Digital, program of the Aspen
Institute. And I have a two part question for you, and feel free to let me know
if there's not enough time for that. But I am not a constitutional scholar, I
don't know a whole lot about intelligence agencies. So I'd love if you could
break down for me two things. One is, you talked about, monitoring the
potentially monitoring the social media and, looking for expressions of
violence or intent for violence and I was wondering if you could give a little
more clarity on what, what qualifies, what does that mean? Where, where's the
threshold?
And then two, in the situation, perhaps for good reason or
wrong, in the situation where agencies are constrained or afraid to act, what
alternative pathways are there. You talked about, kind of, in our perhaps,
unpopular poll, should there be more or less monitoring, there was a caveat
there, which was, like, there was a caveat whether or not there was sort of a,
a cause, an instigating circumstance, and I'm wondering if you could share what
is the alternative path? Are we, the citizens of social media, meant to report
when we see expressions of intent to violence? One, what is intent of violence?
When do we cross that line? And then two, what is the alternative to government
monitoring?
Benjamin Wittes: Great. Please.
Kenneth Wainstein: Okay. Excellent set of questions. And
before I answer the first question, I think you've hit on a really important
point, and I don't know exactly how everything played out with January 6th, but
one thing I've seen over the years is paralysis occurs by government workers
when they feel they don't have clear guidance, clear leadership, and clear
support for what they're doing. So they don't really understand exactly where
they should go. And I think that's what happened with January 6th. There's a
lack of, of understanding of exactly where the right and left boundaries were.
And when they hear that, when that, that's the case. People tend, as a matter
of just human nature, to say, okay, I'm just going to back off from doing that
work because I'm scared, right?
And so we saw some of that, which leads back to my last point,
which is incumbent on Congress to give us the rules, but it's incumbent on
people like me to make sure that those rules are enunciated, trained, clarified
among the workforce so that people understand where the, the guidelines are,
but also understand and can move confidently forward and do their job for the
American people. So that's a really important thing.
In terms of what is that guideline as it applies to violence
here, that's a very good point. What we, we cannot, we're prohibited by the
Attorney General guidelines from doing any, accessing any post for the purposes
of just monitoring somebody's First Amendment activity. We can't do it just
solely for that reason. We can't do it to try to stifle dissent. We can't do it
in any way to try to prevent political views being expressed. So we're
prohibited from doing that. We can only go and look at a post if we have a
reasonable belief, and that's defined in the guidelines, in other words, not a
hunch, but a reasonable belief based on training, expertise, understanding of
maybe that particular website and the extremists who visit that website, that
there might be a post that, that is related to domestic violence and extremism.
And we do it by using search terms that relate to that possibility and go into
the places where we have a reasonable belief we might find that discussion.
So that's how it's done and that's, so roughly speaking, how we
will zero in on something and how frankly prior to January 6th, they did zero
in on some conversations that suggested that some sort of attack was coming.
And so once you understand it's carefully circumscribed and I think that's,
people need to take comfort from that. We cannot just monitor people unless we
have a basis for a reasonable belief that they're engaged in some kind of
threatening behavior.
Your second question is what the alternative is. I mean, it's
interesting I don't think there really is. I don't think the American people
want the government to delegate national security to private actors. But you
could see a world where that is what happens, right? I mean, as we discussed,
every one of you has more authority than I do and my people do in the open
source environment. You guys can go a lot of places we can't go and see more.
There are private organizations out there that can do that and do do that day
in and day out to protect their schools, or their faith institutions, or what
have you. We cannot make them agents of ours, but there is a world where
they're the ones who are generating all the threat indicia, and then putting
them out, and the government's on the sidelines.
I'm not sure, at the end of the day, that addresses the
constitutional concerns and the privacy concerns, because those become almost
quasi government actors. So I guess, if I'm trying to think of like an
alternative world, that's it, and I guess if we don't sort of iron out this
conundrum of, okay, this needs to happen. Right now you have agencies that are
not fully equipped to do it, either resource wise or authorities wise, how is
this, how should this be reconfigured to be effective, but also to be suitably
constrained if we can't get that worked out, who knows? It might be that we,
you move to that alternative world, but I just don't see that as a viable
possibility.
Benjamin Wittes: Jameel, do you have final thoughts?
Jameel Jaffer: I, I guess the only thing I'd say in
response to that is it's, it's not an unusual thing that private citizens would
have more access to a lot of these spaces than government officials do and it's
I don't think it's a problematic thing. Like for example if you know, let's say,
you go to church and somebody brings a gun to church I don't think it would be
reassuring to know that there's a government agent sitting in the back pew who
will, you know, see that that person has a gun and then take whatever action is
required.
No you assume that in those kinds of circumstances private
citizens will, yes, report to the government what's, what's going on. And I
don't think people would find it reassuring to know that there are government
agents, and I don't actually think this is what you're proposing, but
government agents in private political organizations just in case somebody
happens to make a threat of, of, of violence. So it's not a problem in my view
that the government doesn't have visibility into all of these spaces, that's
just a function of living in a democratic society in which the government has
limited, limited powers.
I don't want to pretend that these questions though are easier
than they are. I do think this question of when the government should have
access to even publicly available information is a complicated question. It's a
complicated question because the government is engaged in the same kinds of
collection activities that journalists are engaged in, that Clearview AI is
engaged in when it puts together its facial recognition app, right? And we
don't have a legal framework right now that draws clear distinctions, that sort
of distinguishes between these activities. Our legal framework right now kind
of squints at all these activities and says these look like the same thing. And
that doesn't seem like a workable framework, as we go forward. We need, we need
a legal framework that kind of distinguishes all of these collection activities
that has a, has different legal rules for each of these collection activities.
And we don't have that right now and it's not an easy thing to figure out what
that framework should look like.
Benjamin Wittes: Quinta, bring us home.
Quinta Jurecic: I think, actually, to some extent, we
are already seeing this kind of collaboration between government agencies and
private researchers and, again, I'll point to Ryan Reily’s book, Sedition
Hunters. He's not paying me, but you should read it, it's a great book. And
the ‘sedition hunters’ he's referring to there in the title are a group of
private citizens who turned out to be much better at sleuthing social media then
the FBI. And who were able to look at the enormous amount of material that
folks posted on the day of January 6th selfies, videos of themselves in the
Capitol and use open source investigative techniques among themselves to help
identify those people. And you see to this day, there are still people being
arrested who entered the Capitol who have been identified through those
techniques. And those are people who are working with the Bureau, to help the
Bureau identify those people. And so I do think that raises a lot of questions,
whether you think that is a good or a bad thing, but whatever you make of it,
it has been extraordinarily effective.
The second thing, going back to your, your first question about
how, how we think about what constitutes a level of violence that merits some
kind of review or collection. I think that's a really important issue, particularly
because, again, if you look at congressional reports about what's going on in
these agencies in advance of January 6th, what you see is a lot of missing the
forest for the trees.
Where if you, if you zoom back and look at the whole body of
posts that people are making on social media saying I'm bringing my gun, let's
go to the Capitol, right, or hang Mike Pence. In the aggregate, there's just an
extraordinary amount of material there that is deeply, deeply concerning. But
if you look at any individual posts, you might say, okay, this person's saying
hang Mike Pence, but he's in Montana, right? This person says, let's all go to
the Capitol. But like, are they really going to the Capitol? Probably not. And
that might not rise to a level where an FBI analyst would have previously
thought this is something that is actually concerning.
And I think that this dynamic in terms of how we move from a
world of looking at sort of more individualized communication as a potentially
concerning threat to understanding aggregate information, and sort of mob
movement and harassment as something that could potentially be threatening and,
and worthy of concern, whether from private actors or from government actors is
something that we're still working out. I think it, it maybe touches on the
point that Jameel made about the difficulty of thinking through First Amendment
implications of different kinds of, of work right now in a world that looks
very different than a lot of that original jurisprudence was, was decided in. But
I, I think that you, you see these dynamics not only in terms of how government
agencies are trying to think through these things, but how judges now are
trying to think through the question of what constitutes incitement to
violence, and also in terms of how private researchers are, are understanding
the material that they're seeing on these platforms.
Benjamin Wittes: We are going to leave it there. Please
join me in thanking our panelists.
The Lawfare Podcast is produced in cooperation with the
Brookings Institution. You need to do your part to become a material supporter
of Lawfare. Hey, you don't like the ads on this? Go become a material
supporter and get access to our ad-free podcast. We need your help to bring you
everything that you want from us this year.
The Lawfare Podcast is edited by Jen Patja Howell. Our
music is performed by Sophia Yan. And as always, thanks for listening.
