It’s a lot easier to pick out the footprints leading to criminal activity after the activity occurs. We’re awash with footprints — digital, figurative and literal — that might be heading somewhere dangerous, but, happily, they usually don’t. One of the most difficult tasks in law enforcement is differentiating between the dangerous paths and the innocuous ones. Spot the right pattern and you prevent the crime from happening. Otherwise, you’re left going back and reviewing what you missed.

On Tuesday morning, the Senate Rules and Administration and Homeland Security and Governmental Affairs committees released a report reviewing the failure of law enforcement to track, prepare for and prevent the violence at the Capitol on Jan. 6. Over nearly 100 pages, the document details what was known before that day and how that intelligence failed to lead to proper preparations on the part of the U.S. Capitol Police and other organizations.

What is immediately striking about the report is that the footprints leading to that violence extended back nearly a month. It’s hard to determine the extent to which this is simply a function of having the benefit of hindsight; that is, whether the Senate investigators were able to pick out these particular indicators simply because the violence had happened.

But the barely obscured subtext of the document is that the threat from right-wing actors against Congress was simply not treated as particularly robust, thanks to some combination of misunderstanding motivations, outdated expectations and a crying-wolf problem. At the same time, there was heightened concern about the perception of the military activating against civilians, a function both of then-President Donald Trump’s increasingly agitated response to his election loss and the heavy-handed response to left-wing protests last summer. The combination, then, was a leadership not ready to address a threat it didn’t treat as serious in the first place.

The footprints

On Dec. 8, the anti-Trump group the Lincoln Project released an ad drawing attention to the counting of electoral votes on Jan. 6 and then-Vice President Mike Pence’s role in the process.

The intent of the ad was to knock Trump off-kilter and to upset him — an outcome that was reportedly successful.

Within hours, Trump allies were announcing plans to object to the counting of the electoral votes that was the reason for the joint session of Congress that day.

Three days later, on Dec. 11, the director of the Capitol Police’s Intelligence and Interagency Coordination Division (IICD), John Donohue, “anticipated a challenge to the electoral vote from a few Members of Congress and requested a preliminary assessment for the January 6 Joint Session of Congress,” according to the report. He additionally hired an outside company to track social media mentions of “Joint Session of Congress.”

By Dec. 14, the deputy chief of the Capitol Police’s Protective Services Bureau alerted officials that there was an expectation that the Jan. 6 session would “bring some demonstrations, with the potential for some issues on the House floor” — presumably referring to contests from lawmakers. (That quote, as with the others that follow in this section, are from the Senate report.)

On Dec. 16, the IICD released the first of several weekly assessments of what it expected to unfold on Jan. 6. It indicated there were no “social media indications for specific threats” and, more broadly, “no specific known threats” at that time.

After promoting planned objections to the counting of electoral votes that day, Trump on Dec. 19 first called for people to come to D.C. to participate in protests.

“Big protest in D.C. on January 6th,” he wrote on Twitter. “Be there, will be wild!”

That tweet and two other social media messages focusing on the date — including one calling for people to “occupy” D.C. and Congress — were circulated within the IICD.

Two days later, on Dec. 21, D.C. Metropolitan Police flagged the increased attention being paid to the rally on Jan. 6. It compared the likely protest to prior rallies in November and December in which fighting had broken out in the evening between pro- and anti-Trump activists.

That same day, the IICD released a report detailing what it knew about plans for the day. It included a mention that far-right groups such as the Proud Boys and Oath Keepers were expected to attend. The report also shared a number of posts from a pro-Trump website that “promote[d] confronting members of Congress and carrying firearms during the protest.” Among the messages that were highlighted were ones focusing on the tunnel system under the Capitol.

Over the next several days, the group received several outside tips about threats of violence. On Dec. 22, the IICD received emails sharing online posts encouraging an armed response on Jan. 6. Another focused on comments from the head of the Oath Keepers about the possibility of “armed conflict.” On Dec. 23, someone called the IICD to warn about social media posts focused on bringing weapons to the Capitol. A D.C. intelligence analyst on that same day shared information about online posts that included “threats towards the US Congress and elected officials.”

The IICD’s second assessment came out that same day. It stated that there was “no information regarding specific disruptions or acts of civil disobedience targeting this function,” according to the Senate report and, echoing the D.C. police, said the protests would probably be “similar to the previous Million MAGA March rallies in November and December 2020.”

On Dec. 28, someone emailed the Capitol Police to share information about “countless tweets from Trump supporters saying they will be armed on January 6th” and “tweets from people organizing to ‘storm the Capitol’ on January 6th.”

The Dec. 30 assessment of the risks on Jan. 6 again reiterated the expectation that the protests that day would probably “be similar to the previous Million MAGA March rallies in November and December 2020.” It also predicted “the number of people who indicate they are going to the event listed on these social media postings is relatively low,” although other intelligence indicated an uptick in hotel reservations.

It also predicted that, despite multiple planned rallies, “no group is expected to march and all are planning to stay in their designated areas.” Over the next few days — thanks in part to Trump’s explicit endorsement of the day’s events and his plan to address the crowd — those various rallies fused into one coordinated day of events. While there was no permitted march, there was a later expectation from D.C. authorities that there might be movement between the rally at the Ellipse south of the White House and the Capitol over the course of the day.

On Jan. 1, the IICD received a tip that there were “detailed plans to storm federal buildings.” At the same time, Donohue asked for his team to update its intelligence about the Proud Boys. The group “frequently engage[s] in violence against left-wing protestors,” that report stated, adding that the “presence of Proud Boys at a protest increases the likelihood of violence.”

The IICD’s last pre-Jan. 6 report was released on Jan. 3. It included more detailed warnings about the day, including that the crowd would probably include “the Proud Boys (who intend to wear plainclothes and not their traditional yellow and black clothing), white supremacist groups, Antifa, and other extremist groups.”

It also differentiated the day from the November and December rallies in specific ways.

“Supporters of the current president see January 6, 2021, as the last opportunity to overturn the results of the presidential election. This sense of desperation and disappointment may lead to more of an incentive to become violent,” the report stated, with the bolding in the Senate report. “Unlike previous post-election protests, the targets of the pro-Trump supporters are not necessarily the counterprotesters as they were previously, but rather Congress itself is the target on the 6th. As outlined above, there has been a worrisome call for protesters to come to these events armed and there is the possibility that protesters may be inclined to become violent. … This combined with Stop the Steal’s propensity to attract white supremacists, militia members, and others who actively promote violence, may lead to a significantly dangerous situation for law enforcement and the general public alike.

That information, though, was buried deep in the report. The top line remained the same as it had in the prior assessments: “An overview of the expected protests highlighted that they would be similar to previous rallies and that protestors would remain in their designated areas,” the Senate report reads. The IICD released daily reports over the next three days that continued to describe the likelihood of violence at the Jan. 6 events as “improbable.”

The night before the Jan. 6 violence, the FBI National Threat Operations Center received a warning about “a ‘significant uptick’ in new visitors to the website” and a “huge uptick with reporting via open source of the groups intentions of forming a perimeter around the [Capitol] campus.”

This was not elevated to the Capitol Police leadership.

The response

Again, it’s easy to pick out these various flags as significant now that we know what happened. It’s not clear from the Senate report if, for example, these third-party warnings received by the IICD were the entirety of what it got or if they were simply pertinent examples from a flood of thousands. It is clear that the information was not presented in a way that made it stand out to recipients; then-Capitol Police Chief Steven Sund described the intelligence report as being “very similar to the intelligence assessment of the December” rally — an event where acts of violence were sporadic and limited in scale.

But it’s also clear from the report that the people tracking the threat didn’t necessarily consider it as dangerous as they should have.

For one thing, it seems, there are so frequently threats of violence from the far right that it muddies the ability of investigators to pick out the serious ones. A report from the Department of Homeland Security’s Office of Intelligence and Analysis released on Dec. 30 stated “the use of social media to make threats of violence upon which [extremists] often do not act” limits the organization’s ability to disrupt serious plots. Too many footprints.

Melissa Smislova, acting undersecretary of the Office of Intelligence and Analysis, testified to the Senate committees that separating the extremists from the chaff was difficult.

“A lesson learned from the events of January 6th is that distinguishing between those engaged in constitutionally protected activities from those involved in destructive, violent and threat-related behavior is a complex challenge,” she said.

Later, she added an interesting qualifier.

“This is a very difficult threat for us and the intelligence community to understand,” she said, one requiring “more partnerships with nontraditional partners.”

One of the immediately remarked-upon aspects of the violence on Jan. 6 was how the response differed from the reaction to left-wing protests in D.C. the prior summer. Then, during several days of rallies centered on racial justice and the Black Lives Matter movement, there was a heavy-handed response that included a massive deployment of law enforcement. There were no active-duty military involved, despite Trump’s aims, but there were members of the National Guard.

That response spurred a lot of controversy that specifically led to caution on the part of law enforcement as Jan. 6 approached. The Defense Department’s decisions around the deployment of the D.C. National Guard, for example, were “influenced by lessons learned following DCNG deployment during the summer of 2020 to civil disturbances related to the murder of George Floyd,” the Senate report reads.

It’s useful to remember the broader context here. Trump was actively undermining the election results, and there were rumblings that he might seek to use the Insurrection Act to somehow seize power. Again, Trump had wanted to deploy the military the prior year; that he might want the military to intervene on his behalf somehow wasn’t outside the realm of possibility.

The Defense Department had an acting defense secretary at the time, Christopher Miller — who held that position after Trump fired his predecessor for, among other things, reportedly objecting to invoking the Insurrection Act. Miller told aides that among his three goals for his tenure were the outcomes of “no military coup” and “no troops fighting citizens on the streets,” according to reporting from Axios.

On Jan. 5, Miller and Trump spoke on the phone. According to Miller, Trump asked how prepared his team was for the following day. Miller told Trump that his team was ready to support the District’s law enforcement.

“You’re going to need 10,000 people,” Trump told him.

There was an external push in the opposite direction. Ten former defense secretaries (including Miller’s predecessor) released a joint statement expressing concern about involving the military in election disputes. That was published on Jan. 3, the same day as the last IICD assessment.

Put simply, there was a lot of pressure for the government to move cautiously in a way that didn’t exist the previous summer. But there was also an obvious failure to recognize the threat. Consider, for example, a detail from the Senate report spotted by journalist Marcy Wheeler.

The Capitol Police’s Civil Disturbance Unit had a plan for operations on Jan. 6 that required that protective gear be staged near the Capitol and that it be deployed, “depending on the presence of counter-groups and the reported interactions between the opposing groups.” In other words, the perceived threat of violence wasn’t the far-right against the police or Congress but, instead, against elements of the left that might show up. This was, in fact, the pattern in November and December.

The Senate report does not delve into one of the lingering questions that surrounds the day: Was the government response influenced by the fact that the protesters were ostensible advocates of law enforcement? To what extent was the response on Jan. 6 — following weeks of warnings about violence (if not always specific threats) — a function of assuming different patterns of behavior than had been assumed the prior summer?

On this, we’re again left to survey the footprints and draw conclusions.