Criminal Justice & the Rule of Law

What If J. Edgar Hoover Had Been a Moron?

Benjamin Wittes
Tuesday, August 4, 2020, 9:15 AM

My adventures as an intelligence subject of the Department of Homeland Security.

Tear gas from DHS agents and fires at the Mark O. Hatfield United States Courthouse during the 2020 George Floyd protests in Portland, Ore. (Wikimedia/Tedder; CC BY-SA 4.0,

Published by The Lawfare Institute
in Cooperation With

It was on the ninth day of the Trump presidency, writing in response to the new president’s travel ban executive order, that I coined the phrase “malevolence tempered by incompetence.”

I never imagined in doing so that the phrase might aptly describe the Trump administration’s behavior toward me personally.

Yet contemplating the Washington Post’s revelation that the Department of Homeland Security Office of Intelligence and Analysis (DHS I&A) issued two intelligence reports about tweets I had written, I can’t help but think that this is what J. Edgar Hoover’s abuses of power might have looked like had Twitter existed in Hoover’s time—and had Hoover been a total idiot.

On the one hand, DHS I&A was preparing intelligence reports on American journalists—on me and on Mike Baker of the New York Times—based on activity indisputably protected by the First Amendment: reporting unclassified information about the conduct of government. That’s toxic stuff. And DHS knows it. No sooner had the redoubtable Shane Harris published the story in the Post than DHS declared that Acting Secretary Chad Wolf was stopping the activity in question and initiating an investigation:

Upon learning about the practice, Acting Secretary Wolf directed the DHS Intelligence & Analysis Directorate to immediately discontinue collecting information involving members of the press. In no way does the Acting Secretary condone this practice and he has immediately ordered an inquiry into the matter. The Acting Secretary is committed to ensuring that all DHS personnel uphold the principles of professionalism, impartiality and respect for civil rights and civil liberties, particularly as it relates to the exercise of First Amendment rights.

On the other hand, the collection and reporting on me is so trivial—and so dumb—that it can be hard to stop giggling and see the menace. Consider, DHS issued two intelligence reports, noting the shocking fact that I had tweeted things, a fact evident to all of my Twitter followers. The reports added no analysis of any kind. They didn’t mention what this had to do with anything a law enforcement or intelligence officer might find important. If this is Big Brother, he’s not all that impressive.

So let’s look at both the incompetence, which is simple and easy to understand and genuinely amusing, and then the malevolence beneath it—which is more complicated and is not amusing at all.

Here, to start, are the simple facts of my adventure as a DHS intelligence subject.

On July 20, Steve Vladeck and I published a Lawfare article entitled “DHS Authorizes Domestic Surveillance to Protect Statues and Monuments,” based on a DHS I&A document leaked to me. For reasons related to source protection, we did not publish the document itself but described it in some detail:

A document provided to Lawfare on July 19 from the Department of Homeland Security (DHS) Office of Intelligence & Analysis (I&A) describes personnel as “collecting and reporting on various activities in the context of elevated threats targeting monuments, memorials, and statues”—and it gives legal guidance concerning the “expanded intelligence activities necessary to mitigate the significant threat to homeland security” posed by such activities.

The document, titled “Job Aid: DHS Office of Intelligence & Analysis (I&A) Activities in Furtherance of Protecting American Monuments, Memorials, Statues, and Combatting Recent Criminal Violence,” is not classified. Its three pages each bear the heading “UNCLASSIFIED//FOR OFFICIAL USE ONLY.” But it clearly indicates that at least parts of the intelligence community are being tasked with monitoring and collecting information on some protest activities.

Shane Harris wrote about the same document, a copy of which I gave him, in the Washington Post later that day.

In the days that followed, I received more documents from DHS I&A. Rather than analyzing them on Lawfare, I simply tweeted screenshots of them to make them available to other reporters and the general public as quickly as possible. I figured I would come back and analyze them later if the need arose. These included this tweet, which revealed an email complaining about the leaks to me and Harris:

More significantly, they also included this tweet, which disclosed a memo changing the nomenclature status of the perpetrators of violence in Portland, Oregon, from “Violent Opportunists” to “VIOLENT ANTIFA ANARCHISTS INSPIRED (VAAI).” I described this memo in a subsequent tweet as “responsive to POTUS pressure to ‘designate’ Antifa as a terrorist organization”:

Unbeknownst to me until the matter leaked to Harris last week, both of these tweets turned out to have been the subject of what DHS I&A termed an “OPEN SOURCE INTELLIGENCE REPORT.” Here’s the report on the first tweet. Here’s the report on the second tweet. Reporting by Mike Baker of the New York Times was also the subject of a third, similar intelligence report. I tweeted images of all three reports when I got my hands on them after Harris’s story became public:

The reports in question are odd. Each amounts to little more than a write-up of the tweet it covers dressed up in intelligencese. “This information is provided for intelligence and lead purposes only,” each says. “The product contains U.S. person information that has been deemed necessary for the intended recipient to understand, assess, or act on the information provided.” The summary reads, in one case, “Social media user”—which is to say me—“posts a leaked Department of Homeland Security internal memo that discusses changing terminology used in reports.” In the other report, the summary says, “Social media user posts a leaked Department of Homeland Security internal memo on intelligence activity in Portland, Oregon.”

Bizarrely, the reports describe me as a “source,” as though I am publishing my Twitter feed to provide information to DHS I&A. But no, @benjaminwittes was not meeting with anyone from DHS in a garage. Nor was my Twitter feed specifically providing information to DHS I&A; rather, I was taking information from it and making that information public. Both reports describe me as “a social media user” and “a new source whose information has not been validated.” My name is not redacted or withheld in the report, although at one point, that of the then-head of DHS I&A is masked; instead of including his name, Brian Murphy, it quotes my tweet as saying, “And to (Identified Acting Undersecretary): I have read, and I acknowledge receipt.”

The reports are cleared for dissemination to “All Field Ops” and say they are “releasable to the governments of Australia, Canada, United Kingdom, and New Zealand.” So the document is at least cleared for release to state and local law enforcement and to foreign governments; to what extent it has been actively disseminated I do not know. There are certainly more efficient ways to read my tweets in New Zealand.

The rest of the document is quite literally just the tweet, which is both described and included in a screenshot, along with an image of the document it reported and an image of my Twitter header. Perhaps coincidentally—given that the Mike Baker report also displays Baker’s Twitter header—my Twitter header bears the following image:

And if all that sounds really dumb, well, that’s because it is.

To be clear, there is nothing—at least not in my view—wrong with DHS sharing my tweets internally within the government. I write in order for people to read my thoughts, and that includes law enforcement and intelligence officials, along with hundreds of thousands of others. I doubt that anyone has a reasonable expectation of privacy in a Twitter feed intentionally made available to the public, but I certainly don’t harbor any such illusion about mine. If government officials want to share my tweets among themselves, that’s great. That’s what the feed is for.

And, indeed, had the author of these reports merely sent around an email to colleagues saying, “Hey look, @benjaminwittes just posted an internal document”—which is really all these reports say—I would not be remotely concerned about it. Indeed, the first of the two tweets reports on a document that more or less does exactly that in noting that I had published leaked information. Similarly, had someone written to the DHS inspector general asking for a leak investigation based on the tweets, that would have seemed entirely sensible too. Had people shared the tweets socially or professionally within the government, that also would have been fine.

But the idea that this is useful open source intelligence is just goofy, and the gussying up of a tweet available to hundreds of thousands of people into an intelligence “source” is like an intelligence agency playing dress-up. Intelligence is something that plausibly gives policy makers or operators information and analysis with which to do their jobs and make decisions. It is inconceivable to me that this report might be useful as intelligence to state and local governments, much less to America’s partner governments in the Five Eyes intelligence alliance—none of which are responsible for protecting unclassified internal DHS memos and all of which have access to my Twitter feed with just the click of a “follow” button. These documents serve no useful intelligence purpose. In the wake of Harris’s story, I received any number of communications from intelligence professionals—many of them indignant on my behalf, but all of them befuddled by what kind of clown show DHS I&A was running for someone to think such an intelligence report would be useful.

Some people have suggested that the goal might have been to intimidate me, but if that was the goal the report is particularly silly. I was not supposed to find out about it, after all. And it’s hard to intimidate people using intelligence reporting they do not know has been written and disseminated—particularly about material they are endeavoring to make as widely available as possible.

But if the details of the incidents all render them trivial, even laughable, there is a serious side here too. The government isn’t supposed to be gathering, reporting, and disseminating intelligence on U.S. persons without some clear factual predicate for doing so. It particularly isn’t supposed to be doing this solely based on a subject’s First Amendment protected activities. And DHS is only supposed to be collecting intelligence at all on the basis of a limited set of homeland security missions.

It’s not clear to me that DHS I&A was following any of these rules when it reported on Mike Baker and me. And that fact makes me worried about what other First Amendment protected activity might be the subject of intelligence reporting by DHS. I’m worried here less about journalism than about the equally constitutionally protected activity of protesting, to which the DHS has shifted considerable attention and energy. To put my basic question a little differently, if DHS could issue these reports about my Twitter activity, what else must it be reporting on vis a vis protesters?

Happily, DHS has made the rules governing the department’s surveillance and intelligence activities publicly available. They are spelled out at length in a document entitled, “Department of Homeland Security Office of Intelligence and Analysis Intelligence Oversight Guidelines.”

High up in the document is the stark sentence, “I&A personnel are prohibited under all circumstances from engaging in any intelligence activities ... for the sole purpose of monitoring activities protected by the First Amendment or the lawful exercise of other rights secured by the Constitution or laws of the United States.” This is pretty hard to square with the intelligence reports, one of which reads:

(U//FOUO) SOURCE CONTEXT: A social media user who maintains a professional social media presence. . . .


1. (U//FOUO) Source posted a leaked DHS internal document on intelligence activity in Portland, Oregon. Source posted three pictures (See Attachment) with the caption, “Internal memo from DHS I(ampersand)A complaining about leaks to [USPER][AT] lawfareblog and [USPER][AT] shaneharris (neither by name) and insisting intel activity in Portland is all legal and appropriate. Withholding name of author for privacy reasons.” At time of acquisition, post had 15 comments, 98 shares, and 206 likes.

What purpose is such a report serving other than to monitor the First Amendment activities of a U.S. person?

To see this point more clearly, consider the actual authorized bases for DHS I&A collection. “I&A personnel are authorized to engage in intelligence activities where they have a reasonable belief that the activity supports one or more of the national or departmental missions listed below,” the document states. The national missions listed include:

a. International terrorism threats;

b. The proliferation of weapons of mass destruction;

c. Intelligence activities directed against the United States;

d. International criminal drug activities; and

e. Other hostile activities directed against the United States by foreign powers, organizations, persons, and their agents.

Now, I think my tweets on DHS surveillance policy matters are pretty earth-shattering, but even I have to admit that following them will not help anyone stop terrorist threats, WMD proliferation or hostile intelligence activity. Indeed, there’s no way any of the national missions could justify collecting my tweets.

So maybe one might find a justification for collection in one of the departmental missions, a category that includes:

a. Domestic terrorism threats;

b. Threats to critical infrastructure and key resources;

c. Significant threats to the Nation’s economic security, public health, or public safety, including, but not limited to, local manifestations of national threats (e.g., local outbreaks of diseases reasonably likely to pose the risk of becoming a national pandemic);

d. Major disasters and other catastrophic acts; and

e. Any other threat of such severity and magnitude that effective response would be beyond the capabilities of any affected State and local governments, such that Federal assistance would be necessary.

Any of these would be quite a stretch too—at least as stated.

As Vladeck and I reported, DHS has adopted a new homeland security mission in support of which it has also authorized surveillance: the protection of federal property and monuments. Perhaps, it might be argued, collecting and reporting on my tweet is appropriate because such intelligence helps DHS deal with leaks and thus enables it to carry on this and the other missions listed above.

But that’s still a stretch. In no sense does my tweet threaten federal property or any monument. And anyone doing a leak investigation in connection with the disclosures to me already knows that the material has leaked and been published on my Twitter feed. The supposed intelligence reporting adds literally no new information beyond the simple fact of the tweet.

So if I had Brian Murphy in front of me, I would want to ask the man a simple question: Pursuant to what DHS mission was it appropriate to collect my tweets as intelligence in the first place?

This question is important because if DHS I&A is interpreting its missions so broadly as to justify collecting these tweets, then the limits of its collection authority with respect to public record information concerning First Amendment protected activity seem positively nonexistent. My tweets had nothing to do with any of the missions DHS has articulated as warranting collection. They, rather, revealed information about how DHS I&A is doing its job; and preventing leaks is not to my knowledge an articulated homeland security mission justifying intelligence activity. If it is appropriate to collect such material, why would it not be lawful to collect a full dossier of public record information on, say, a protester?

The same broad point is true not only of collection of U.S. person information but of its retention too: “The retention of intelligence or information, whether collected or otherwise obtained by I&A, is permitted only to the extent there is a reasonable belief that retention furthers one or more of the national or departmental missions listed above ....” So again, I ask: On the basis of what mission did DHS retain my tweets? Yes, in and of itself, it’s a silly question: Of course I want everyone to retain my tweets, to dwell over them, and ponder their brilliance. But for DHS I&A to store my tweets as intelligence, it has to follow certain rules. And it’s worth asking: If DHS is flouting the rules with respect to me in this fashion, how else is it doing so—and with respect to whom? In other words, how much information about how many protesters is being “retained” by DHS I&A?

This brings me to the dissemination of the “intelligence” about my tweets, which strikes me as the most obviously improper act. Even if it was proper to collect the tweets, and even if it was proper to retain them, the rules make it crystal clear that information about U.S. persons may be disseminated to other agencies or governments or components only if “[t]here is a reasonable belief that dissemination would assist the recipient of the [information] in fulfilling one or more of the recipient’s lawful intelligence, counterterrorism, law enforcement, or other homeland security-related functions.”

Pause over that. Because now it is not simply a requirement that DHS have some mission that justifies the collection. To disseminate the tweets, DHS needs to believe that the recipient agency has a legitimate and lawful interest in my tweets that is meaningfully connected to a security function. I simply cannot imagine a theory under which my tweets, even dressed up in the borrowed robes of intelligence reporting, might help a tribal government, a local police department, or the government of, say, New Zealand with any lawful intelligence, counterterrorism, law enforcement, or homeland security-related function. If DHS I&A has any theories as to how my tweet might assist Australian counterintelligence efforts or Canadian law enforcement or the Navajo Nation’s counterterrorism functions, or even Oregon’s efforts to protect federal buildings, I’m all ears. But without such a theory, I just don’t see how my or Baker’s reporting is fair game for dissemination.

And yet again, it’s worth asking: If this is okay, what else is fair game for dissemination to intelligence partners?

I personally love that government officials are sending around my tweets. They should all do it more. But for this intelligence report to get filed, dumb as it is, a lot of things have to go wrong. People have to believe that my tweets on DHS’s internal documents are meaningfully connected to some homeland security mission. They have to believe that they are doing something other than monitoring purely First Amendment protected activity—or, worse, they have to not care that they’re doing exactly that. And they have to believe that their partner agencies and governments have a legitimate interest, one reasonably connected to some lawful mission, in seeing such material—which they plainly do not.

If all this could go wrong with my two tweets, where else are similar abuses taking place less stupidly and more menacingly—and how much more harmful have the abuses been in those other situations?

Benjamin Wittes is editor in chief of Lawfare and a Senior Fellow in Governance Studies at the Brookings Institution. He is the author of several books.

Subscribe to Lawfare