Posts

Merrick Garland Hasn’t Done the Specific Thing You Want because DOJ Has Been Busy Doing Things They Have to Do First

The passage of the election has set off the Merrick Garland whingers again, people who like displaying their ignorance by claiming there has been no sign of progress on the investigations into Trump when (often as not) there were signs of progress that the whingers are ignoring in the last few days.

Yes. It has been almost a week since the close of polls last Tuesday. No. Merrick Garland has not carted Trump away in a paddy wagon yet (nor would the FBI, if and when they ever did arrest him).

Yes. We actually know why Garland hasn’t done so — and it’s not for want of actions that might lead there.

There are still known steps that have to or probably will happen before Trump would be indicted in any of the known criminal investigations into him. For those demanding proof of life from the DOJ investigations into Trump, you need look no further than the public record to find that proof of life. The public record easily explains both what DOJ has been doing in the Trump investigations, and why there is likely to be at least a several month delay before any charges can be brought.

The reason is that DOJ is still pursuing the evidence they would need before charging a former President.

Here’s an update on the various investigations into Trump (I’ve bolded the two appellate deadlines below).

Stolen documents

The reason I’m particularly crabby about the Merrick Garland whinging is because people were accusing DOJ of inaction hours after DOJ’s most recent step in the investigation into Trump’s stolen documents. On November 3, for example, DOJ compelled Kash Patel to testify before a grand jury under grant of use immunity, testimony that would be necessary, one way or another, before charging Trump, because DOJ would need to rule out or at least account for any claim that Trump mass-declassified the documents he stole.

DOJ continues to fight to ensure it can keep the documents it seized on August 8, and to be permitted to use the unclassified documents it seized in the investigation. The most recent filings in that fight, as I wrote up here, were filings about the disputes Trump and DOJ have about the seized documents, which Special Master Raymond Dearie will use to rule on those designations by December 16. After Dearie does that, Trump will dispute some of Dearie’s decisions, and Judge Aileen Cannon will make her own decision de novo. She has not set her own deadline for how long that decision would take. But if the Special Master process is the means by which DOJ guarantees its access to the evidence against Trump, it won’t be resolved until after the New Year, even assuming DOJ won’t have to appeal some ridiculous Cannon ruling.

Short of doing a search on another Trump property, preferably in Virginia but possibly in New Jersey or New York, this case cannot be charged until DOJ can present documents the custody of which it has guaranteed to a grand jury. DOJ has to make sure they have the evidence they would use to charge Trump (though adjudicating these disputes now might make any prosecution quicker on the back end).

That said, DOJ may guarantee custody of the documents it seized in August more quickly, via its challenge to Cannon’s decision to appoint a Special Master in the first place, in the 11th Circuit. Trump’s response to that appeal, which he submitted on November 10, seemed desultory, as if Chris Kice knows they will lose this appeal (indeed, that seems likely given that both the 11th Circuit and SCOTUS have already declined to see the case in the way Trump would prefer). DOJ’s response is due on November 17. Because of the way the 11th Circuit has scheduled this appeal, the panel reviewing it will be prepared for oral argument on rather quick turnaround. Even so, DOJ is not likely to guarantee access to these documents via any favorable 11th Circuit decision (which Trump will undoubtedly appeal) before December 1, and it would take about a week to present any case to the grand jury. So the very earliest that DOJ could indict this case would be early- to mid- December.

Update: In a filing submitted on November 8 but only unsealed today, DOJ asked Raymond Dearie to recommend that Judge Cannon lift the injunction on the 2,794 out of 2,916 documents over which Trump is making no privilege claim.

Update: The 11th Circuit has set a hearing for November 22, so DOJ may actually have access to those files sooner than December 1, though not all that sooner.

January 6 investigation(s)

There are at least four ways that Trump might be charged in conjunction with January 6:

  • For asking Mike Pence to illegally overturn legal votes and then threatening him, including with violence, when he refused
  • For setting up fake electors to contest the election
  • For fundraising off false claims of voter fraud and using the money to benefit those who helped the attack
  • Via people like Roger Stone, in a networked conspiracy with those who attacked the Capitol

DOJ sent out subpoenas in the first three prongs of this just before the pre-election pause. This post summarizes who was included.

These are all (and have been) intersecting conspiracies (this CNN story describes how many areas the subpoenas cover). For example, since January, it has been clear that the top-down investigation most visible in the January 6 Committee work and the crime-scene investigation visible in ongoing prosecutions had converged on the pressure both Trump and the mob focused on Mike Pence. It’s unclear how DOJ will treat the intersection of these investigations, and whether DOJ will wait for all prongs to converge before charging.

The Mike Pence prong is where DOJ made its most obvious progress during the pre-election pause. On October 6, Mike Pence Counsel Greg Jacob testified before a grand jury. October 14, Pence’s Chief of Staff Marc Short testified. Also in October, DOJ asked Beryl Howell to compel Trump’s White House Counsels Pat Cipollone and Pat Philbin as well. I’m not aware of the status of appeals on that (or whether Judge Howell compelled testimony from the two Pats in the meantime). We know that all four men would describe the debates over the extent of Pence’s authority to reject lawful electors, including the recognition from people like John Eastman that their legal theories were unsupported by law. The two Pats would also testify about Trump’s reaction to the mob, as he watched the attack on the Capitol from inside the White House dining room, including the tweet that specifically targeted Pence. These are all very credible first-hand witnesses to Trump’s words and actions both in advance of and during the attack. Obtaining their testimony would be necessary before charging a former President. But DOJ’s efforts (and success) at obtaining their testimony reflects the seriousness of the investigation.

The publication of Pence’s book, which relays his version about exchanges with Trump, would seem to invite a demand from DOJ that he testify about the same topics to the grand jury as well, particularly given the way he spun the story in ways that might help Trump. If I were a prosecutor contemplating charging the former President, I would want that potentially exculpatory (to Trump) locked in under oath. And any claim from Pence that he can’t share these details because of Executive Privilege seem ridiculous in the face of a book tour. But if DOJ decided they needed Pence’s testimony it might result in delay.

It’s unclear how much progress DOJ has made on the subpoenas issued before the pause. None of those subpoenaed have been spotted at grand jury appearances at Prettyman (though that may change this week). In particular, there are a bunch of senior Republicans involved in the fake elector plots from whom I expect DOJ to try to lock in testimony.

But two things may cause delay in any case. First, as I wrote here, subpoenas (generally served on people who might be expected to comply) are easy, because they require the person who received the subpoena to do the search for the subpoenaed materials. But it takes time to exploit phones, all the more so if the phone was seized without some way to open it. Here’s how long the communications of various high profile people have taken to exploit:

This is not indolence. It is physics and due process: it just takes time to crack phones, to filter the content, and to scope what is responsive to a warrant.

Among the steps taken before the pause, in early September, DOJ seized the phones of Boris Epshteyn and Mike Roman. While it’s possible DOJ will be able to accelerate the process of exploiting these phones (they have done so with Oath Keeper lawyer Kellye SoRelle’s phone, as last week DOJ submitted material that had gone through a filter review from the phone seized from her in early September in the sedition case), you should not assume they can fully exploit these phones (with whatever Signal content is on them) in less than six months, so March. In Epshteyn’s case, his claims to be playing a legal role in the stolen document case may cause further delays because of a filter review.

As someone involved in vote fraud efforts, Latinos for Trump, and the Oath Keepers, SoRelle is one of the pivots from the White House and Willard focused activities to the crime scene. DOJ seems closer to moving against others at that pivot point. Roger Stone, for example, has been mentioned over and over in the Oath Keeper trial. But that’s probably several months off. Alex Jones sidekick Owen Shroyer has been given until the end of the month to decide whether he wants to plead or take his chances on further charges. And I expect DOJ will wait until the verdict at least in the Oath Keeper case (they might not even get through all the defense witnesses this week), and possibly in the more complex Proud Boy case (which would be February barring likely unforeseen changes), before going too much further.

There’s one more thing that may delay any more spectacular charges in January 6. The oral argument for DOJ’s appeal of Carl Nichols’ outlier decision on the application of 18 USC 1512(c)(2) to the insurrection won’t happen until December 12. It drew a pretty unfavorable panel for that hearing (listed as Joseph Fischer here): Trump appointees Greg Katsas (like Nichols, a former Clarence Thomas clerk, who also worked as Deputy White House Counsel in 2017) and Justin Walker (who is close to Mitch McConnell), and Biden appointee Florence Pan (who presided over January 6 cases before being promoted to the Circuit Court). It’s possible, but by no means certain, that the Trump appointees will do something nutty, in which case, DOJ would surely appeal first to the full DC Circuit panel; if they overturn Nichols, Garret Miller and the other January 6 defendants who got their obstruction charges thrown out will presumably appeal to SCOTUS.

Nichols’ decision, which ruled that January 6 did count as an official proceeding but ruled that any obstruction had to involve some kind of documents, probably wouldn’t stall any charges relating to the fake electors, which were after all about using fraudulent documents to overturn the vote certification. But it might lead DOJ to pause for other charges until the legal application is unquestioned. 18 USC 1512 is the charge on which DOJ has built its set of interlocking conspiracy charges, and so this decision is pretty important going forward.

Unlike the stolen document case, I can’t give you a date that would be the soonest possible date to expect indictments. But for a variety of reasons laid out here, unless DOJ were to indict on charges specifically focused on Mike Pence (with the possibility of superseding later), it probably would not be until March or April at the earliest.

Georgia investigation

The Georgia investigation, like the Federal one, was paused for a period leading up to the election (it’s unclear whether the run-off between Raphael Warnock and Herschel Walker will further delay things). But during the pre-election period, DA Fani Willis won decisions for testimony from Lindsey Graham and Newt Gingrich. Those grand jury appearances were scheduled for the end of this month (though may be pushed back). In any case, Willis has indicated that any charges from this investigation may come before the end of the year.

To be clear, none of this is a guarantee that DOJ (or Willis) will indict Trump and/or his closest aides. It is, however, a summary of the reasons that are public that all these investigations have been taking steps that would have to happen before they could charge Trump, and that most have additional steps that would have to happen before prosecutors could even make a prosecutorial decision.

The Rebellion Rorschach: The Many Faces of the January 6 Investigation

Four different things happened yesterday to demonstrate how differently judges presiding over the January 6 trial view it, and how little they seem to understand the intersecting nature of this investigation.

DC Circuit ignores its own language about co-conspirators and abettors

The final event was the reversal, by a per curiam panel including Karen Henderson, Judith Rogers, and Justin Walker, of Thomas Hogan’s decision to hold George Tanios pretrial.

As a reminder, Tanios is accused of both conspiring and abetting in Julian Khater’s attack on three cops, including Brian Sicknick, with some toxic substance.

I’m not going to complain about Tanios’ release. By way of comparison, Josiah Colt has never been detained, and he pled out of a conspiracy with Ronnie Sandlin and Nate DeGrave in which they, like Tanios and Khater, planned to arm themselves before traveling to DC together, and in which Sandlin and DeGrave, like Khater, are accused of assaulting cops that played a key role in successfully breaching the Capitol. The main difference is that Khater’s attack injured the three officers he targeted using a toxic spray purchased by Tanios.

It’s how the DC Circuit got there that’s of interest. Tanios had argued that Hogan had used the same language from the Munchel decision everyone else does, distinguishing those who assault or abet in assaulting police which the DC Circuit has returned to in upholding detention decisions since, and in so doing had applied a presumption of detention for those accused of assault and abetting assault.

In assessing Tanios’s risk of danger, the District Court placed too much emphasis on this sentence from Munchel: “In our view, those who actually assaulted police officers and broke through windows, doors, and barricades, and those who aided, conspired with, planned, or coordinated such actions, are in a different category of dangerousness than those who cheered on the violence or entered the Capitol after others cleared the way.” Id. at 1284.

This is only one line in a ten-page opinion written by Judge Wilkins. It is dicta. It was not quoted or adopted by Judge Katsas’s separate opinion. This line does not create a new approach for evaluating detention issues in this Circuit. It does not mandate that defendants be placed in two separate categories. It does not require a separate, harsher treatment for defendants accused of specific violent offenses. Critically, it does not create a presumption of future dangerousness and should not create a presumption of detention. Rather, it seems that the line is merely intended to remind district court judges that violence is one factor to consider in making a determination about dangerousness. [my emphasis]

The DC Circuit specifically ruled against Tanios on his claim that Hogan had misapplied Munchel.

[A]ppellant has not shown that the district court applied a presumption of detention in contravention of the Bail Reform Act and precedent, see United States v. Khater, No. 21-3033, Judgment at *2 (D.C. Cir. July 27, 2021)

They had to! As their citation makes clear, just two weeks ago, a per curiam panel of Patricia Millet, Robert Wilkins, and Ketanji Brown Jackson upheld the very same detention order (which covered both defendants), holding that the same line of the Hogan statement that Tanios pointed to did not do what both Tanios and Khater claimed it had, presume that assault defendants must be detained.

Appellant contends that the district court misapplied our decision in United States v. Munchel, 991 F.3d 1273 (D.C. Cir. 2021), by making a categorical finding, based solely on the nature of the offense charged (assaultive conduct on January 6), that no conditions of release could ever mitigate the per se prospective threat that such a defendant poses. If the district court had proceeded in that fashion and applied some sort of non-rebuttable presumption of future dangerousness in favor of detention, it would have been legal error. See id. at 1283 (“Detention determinations must be made individually and, in the final analysis, must be based on the evidence which is before the court regarding the particular defendant. The inquiry is factbound.”) (quoting United States v. Tortora, 922 F.2d 880, 888 (1st Cir. 1990)). However, while the district court stated, “Munchel delineates an elevated category of dangerousness applied [to] those that fall into the category that necessarily impose a concrete prospective threat,” the district court also explained, “I think Munchel does not set a hard-line rule. I don’t think that the categories are solely determinative, but it creates something like a guideline for the Court to follow . . . .” Detention Hr’g Tr. at 42:21-24; 43:11-13, ECF No. 26 (emphasis added). In making its ruling, the district court discussed at length the facts of this case, and expressly noted that “we have to decide whether the defendant is too dangerous based upon that conduct to be released or is not,” “every circumstance is different in every case, and you have to look at individual cases,” and that “the government may well not overcome the concrete and clear and convincing evidence requirement.” Id. at 43:8-10, 43:16-18, 43:20-21. Based on our careful review of the record, we find that the district court made an individualized assessment of future dangerousness as required by the Bail Reform Act and that appellant has not shown that the district court applied an irrefutable presumption of mandatory detention in contravention of the statute and our precedent.

Yesterday’s panel cited the earlier affirmation of the very same opinion that detained Tanios.

It’s in distinguishing Tanios where the panel got crazy. The panel could have argued that the evidence that Tanios conspired with or abetted Khater’s assault was too weak to hold him — Tanios made a non-frivolous argument that in refusing to give Khater one of the two canisters of bear spray he carried, he specifically refused to join in Khater’s attack on the cops. But they don’t mention conspiracy or abetting charges.

Instead, the DC Circuit argued that Hogan clearly erred in finding Khater’s accused co-conspirator to be dangerous.

[T]he district court clearly erred in its individualized assessment of appellant’s dangerousness. The record reflects that Tanios has no past felony convictions, no ties to any extremist organizations, and no post-January 6 criminal behavior that would otherwise show him to pose a danger to the community within the meaning of the Bail Reform Act. Cf. Munchel, 991 F.3d at 1282-84 (remanding pretrial detention orders where the district court did not demonstrate it adequately considered whether the defendants present an articulable threat to the community in light of the absence of record evidence that defendants committed violence or were involved in planning or coordinating the events of January 6).

Munchel isn’t actually a precedent here, because that decision remanded for further consideration. The DC Circuit ordered Hogan to release Tanios. Crazier still, in citing the same passage from Munchel everyone else does, the DC Circuit edited out the language referring to those who abetted or conspired with those who assaulted cops, the language used to hold Tanios. It simply ignores the basis Hogan used to hold Tanios entirely, his liability in a premeditated attack he allegedly helped to make possible, and in so doing argues the very same attack presents a danger to the community for one but not the other of the guys charged in it.

If this were a published opinion, it would do all kinds of havoc to precedent on conspiracy and abetting liability. But with two short paragraphs that don’t, at all, address the basis for Tanios’ detention, the DC Circuit dodges those issues.

Beryl Howell has no reasonable doubt about January 6

Earlier in the day, DC Chief Judge Beryl Howell grew exasperated with another plea hearing.

This time, it was Glenn Wes Lee Croy, another guy pleading guilty to a misdemeanor “parading” charge. The plea colloquy stumbled on whether Croy should have known he wasn’t permitted on the Capitol steps — he claimed, in part, that because this was his first trip to DC, he didn’t know he shouldn’t have been on the steps, even in spite of the barricades. Croy was fine admitting he shouldn’t have been in the building, though.

Things really heated up when Howell started asking Croy why he was parading (Josh Gerstein has a more detailed description of this colloquy here).

Under oath, pleading to a misdemeanor as part of a deal that prohibits DOJ from charging Croy with anything further for his actions on January 6, he made some kind of admission that Howell took to mean he was there to support Trump’s challenge to the election, an admission that his intent was the same as the intent required to charge obstruction of the vote count.

When she quizzed AUSA Clayton O’Connor why Croy hadn’t been charged with felony obstruction for his efforts to obstruct the vote certification, the prosecutor explained that while the government agreed that contextually that’s what Croy had been doing, the government didn’t find direct evidence that would allow him to prove obstruction beyond a reasonable doubt, a sound prosecutorial decision.

O’Connor is what (with no disrespect intended) might be deemed a journeyman prosecutor on the January 6 cases. He has seven cases, five of which charge two buddies or family members. Of those, just Kevin Cordon was charged with the obstruction charge Howell seems to think most defendants should face, in Cordon’s case for explicitly laying out his intent in an interview the day of the riot.

We’re here to take back our democratic republic. It’s clear that this election is stolen, there’s just so much overwhelming evidence and the establishment, the media, big tech are just completely ignoring all of it. And we’re here to show them we’re not having it. We’re not- we’re not just gonna take this laying down. We’re standing up and we’re taking our country back. This is just the beginning.

O’Connor is prosecuting Clifford Mackrell and Jamie Buteau for assault and civil disorder. But otherwise, all his cases are trespass cases like Croy’s (including that of Croy’s codefendant Terry Lindsey).

This was the guy who, with no warning, had the task of explaining to the Chief Judge DOJ’s logic in distinguishing misdemeanor cases from felonies. Unsurprisingly, it’s all about what the government thinks they can prove beyond a reasonable doubt, based on evidence like that which Cordon shared with a journalist or, just as often, what people write in their social media accounts. This process has made sense to the few of us who have covered all these cases, but like O’Connor, Howell is dealing primarily with the misdemeanor cases and my not see how DOJ appears to be making the distinction.

Howell also demanded an explanation from O’Connor in Croy’s sentencing memo why DOJ is not including the cost of the National Guard deployment in the restitution payments required of January 6 defendants.

Both according to its own prosecutorial guidelines and the practical limitations of prosecuting 560 defendants, DOJ can’t use a novel application of the obstruction statute to charge everyone arrested in conjunction with January 6 with a felony. It’s a reality that deserves a better, more formal explanation than the one O’Connor offered the Chief Judge extemporaneously.

Trevor McFadden believes a conspiracy to overthrow democracy is not a complex case

Meanwhile, the Discovery Coordinator for the entire investigation, Emily Miller, missed an opportunity to explain to Trevor McFadden the logic behind ongoing January 6 arrests.

In advance of a hearing for Cowboys for Trump founder Couy Griffin, prosecutor Janani Iyengar submitted a motion for a 60-day continuance to allow for the government to work through discovery. She brought Miller along to a status hearing to explain those discovery challenges to McFadden, who had complained about them in the past and refused to toll the Speedy Trial Act in this case. Because Iyengar recently offered Griffin a plea deal, his attorney Nick Smith was fairly amenable to whatever McFadden decided.

Not so the judge. He expressed a sentiment he has in this and other cases, that the government made a decision to start arresting immediately after the attack and continues to do so. “There seems to be no end in sight,” McFadden complained, suggesting that if DOJ arrested someone in three months who offered up exculpatory evidence that affected hundreds of cases, those would have to be delayed again. In spite of the fact that several prosecutors have explained that the bulk of the evidence was created on January 6, McFadden persists in the belief that the trouble with discovery is the ingestion of new evidence with each new arrest.

Miller noted that the government could start trials based on the Brady obligation of turning over all exculpatory evidence in their possession, so future arrests wouldn’t prohibit trials. The problem is in making the universe of video evidence available to all defense attorneys so they have the opportunity of finding evidence to support theories of defense (such as that the cops actually welcomed the rioters) that would require such broad review of the video.

McFadden then suggested that because Griffin is one of the rare January 6 defendants who never entered the Capitol, Miller’s team ought to be able to segregate out an imagined smaller body of evidence collected outside. “Were that it were so, your honor,” Miller responded, pointing out that there were thousands of hours of surveillance cameras collected from outside, the police moved in and outside as they took breaks or cleaned the bear spray from their eyes so their Body Worn Cameras couldn’t be segregated, and the Geofence warrant includes the perimeter of the Capitol where Griffin stood.

McFadden then said two things that suggested he doesn’t understand this investigation, and certainly doesn’t regard the attack as a threat to democracy (he has, in other hearings, noted that the government hasn’t charged insurrection so it must not have been one). First, he complained that, “In other cases,” the government had dealt with a large number of defendants by giving many deferred prosecutions or focusing just on the worst of the worst, a clear comparison to Portland that right wingers like to make. But that’s an inapt comparison. After noting the data somersaults one has to do to even make this comparison, a filing submitted to Judge Carl Nichols in response to a selective prosecution claim from Garret Miller explained the real differences between Portland and January 6: There was far less evidence in the Portland cases, meaning prosecutions often came down to the word of a cop against that of a defendant and so resulted in a deferred prosecution.

This comparison fails, first and foremost, because the government actually charged nearly all defendants in the listed Oregon cases with civil-disorder or assault offenses. See Doc. 32-1 (Attachments 2-31). Miller has accordingly shown no disparate treatment in the government’s charging approaches. He instead focuses on the manner in which the government ultimately resolved the Oregon cases, and contrasts it with, in his opinion, the “one-sided and draconian plea agreement offer” that the government recently transmitted to him. Doc. 32, at 6. This presentation—which compares the government’s initial plea offer to him with the government’s final resolution in 45 hand-picked Oregon cases—“falls woefully short of demonstrating a consistent pattern of unequal administration of the law.”3 United States v. Bernal-Rojas, 933 F.2d 97, 99 (1st Cir. 1991). In fact, the government’s initial plea offer here rebuts any inference that that it has “refused to plea bargain with [Miller], yet regularly reached agreements with otherwise similarly situated defendants.” Ibid.

More fundamentally, the 45 Oregon cases serve as improper “comparator[s]” because those defendants and Miller are not similarly situated. Stone, 394 F. Supp. 3d at 31. Miller unlawfully entered the U.S. Capitol and resisted the law enforcement officers who tried to move him. Doc. 16, at 4. He did so while elected lawmakers and the Vice President of the United States were present in the building and attempting to certify the results of the 2020 Presidential Election in accordance with Article II of the Constitution. Id. at 2-3. And he committed a host of federal offenses attendant to this riot, including threatening to kill a Congresswoman and a USCP officer. Id. at 5-6. All this was captured on video and Miller’s social-media posts. See 4/1/21 Hr’g Tr. 19:14-15 (“[T]he evidence against Mr. Miller is strong.”). Contrast that with the 45 Oregon defendants, who—despite committing serious offenses—never entered the federal courthouse structure, impeded a congressional proceeding, or targeted a specific federal official or officer for assassination. Additionally, the government’s evidence in those cases often relied on officer recollections (e.g., identifying the particular offender on a darkened plaza with throngs of people) that could be challenged at trial—rather than video and well-documented incriminating statements available in this case. These situational and evidentiary differences represent “distinguishable legitimate prosecutorial factors that might justify making different prosecutorial decisions” in Miller’s case. Branch Ministries, 211 F.3d at 145 (quoting United States v. Hastings, 126 F.3d 310, 315 (4th Cir. 1997)); see also Price v. U.S. Dep’t of Justice, 865 F.3d 676, 681 (D.C. Cir. 2017) (observing that a prosecutor may legitimately consider “concerns such as rehabilitation, allocation of criminal justice resources, the strength of the evidence against the defendant, and the extent of a defendant’s cooperation” in plea negotiations) (brackets and citation omitted).

3 Miller’s motion notably omits reference to the remaining 29 Oregon cases in his survey, presumably because the government’s litigation decisions in those cases do not conform to his inference of selective treatment. [my emphasis]

McFadden ended with one of his most alarming comments. He said something to the effect of, he doesn’t feel that the January 6 investigation was a complex type of case akin to those (often white collar cases) where a year delay before trial was not that unusual.

This was a fairly breathtaking comment, because it suggests that McFadden sees this event as the magical convergence of thousands of criminals at the Capitol rather than the result of a sustained conspiracy to get a mass of bodies to the building, a conspiracy that started at least as early as the days after the election. While McFadden’s highest profile January 6 case is a sprawling assault case against Patrick McCaughey and others (the one that trapped Officer Daniel Hodges in the Capitol door), this view seems not to appreciate some larger investigative questions pertinent to some of his other defendants. For example, what happened to the laptops stolen from various offices, including the theft that Brandon Fellows may have witnessed in Jeff Merkley’s office. Did America First engaged in a conspiracy to gets its members, including Christian Secor, to the Capitol (and did a huge foreign windfall that Nick Fuentes got days before the insurrection have anything to do with that). What kind of coordination, if any, led a bunch of Marines to successfully open a second front to the attack by opening the East Doors also implicates Secor’s case. One of the delays in Griffin’s own case probably pertained to whether he was among the Trump speakers, as members of the 3-Percenter conspiracy allegedly were, who tied their public speaking role to the recruitment of violent, armed rioters (given that he has been given a plea offer, I assume the government has answered that in the negative).

It has become increasingly clear that one of the visible ways that DOJ is attempting to answer these and other, even bigger questions, is to collect selected pieces of evidence from identifiable trespassers with their arrest. For example, Anthony Puma likely got arrested when he did because he captured images of the Golf Cart Conspiracy with his GoPro. He has since been charged with obstruction — unsurprisingly, since he spoke in detailed terms about preventing the vote certification in advance. But his prosecution will be an important step in validating and prosecuting the larger conspiracy, one that may implicate the former President’s closest associates.

This is white collar and complex conspiracy investigation floating on top of a riot prosecution, one on which the fate of our democracy rests.

Melody Steele-Smith evaded the surveillance cameras

A report filed yesterday helps to explain the import of all this. Melody Steele-Smith was arrested within weeks of the riot on trespass charges, then indicted on trespass and obstruction charges. She’s of particular interest in the larger investigation because — per photos she posted on Facebook — she was in Nancy Pelosi’s office and might be a witness to things that happened there, including the theft of Pelosi’s laptop.

At a hearing last week, the second attorney who has represented her in this case, Elizabeth Mullin, said she had received no discovery, particularly as compared to other January 6 defendants. So the judge in that case, Randolph Moss, ordered a status report and disclosure of discovery by this Friday.

That status report admits that there hasn’t been much discovery, in particular because, aside from the surveillance photos used in her arrest warrant, the government hasn’t found many images of Steele-Smith in surveillance footage.

The United States files this memorandum for the purpose of describing the status of discovery. As an initial matter, the government has provided preliminary discovery in this case. On or about June 4, 2021, the government provided counsel for defendant preliminary discovery in this matter. This production had been made previously to the defendant’s initial counsel of record. Counsel for defendant received the preliminary production that had been provided to previous counsel. This preliminary production included the FBI 302 of defendant’s sole interview, the recorded interview of defendant which formed the basis of the aforementioned FBI 302, over one thousand pages of content extracted from defendant’s Facebook account, and thirty-nine photographs confiscated from defendant’s telephone.

The government is prepared to produce an additional discovery production no later than August 13, 2021. The production will include additional items that have been obtained by the government from the FBI. These items include, additional FBI investigative reports and the Facebook search warrant dated January 21, 2021. The FBI has provided the government with the full extent of the materials in its possession. While these items are few in number, the government is continuing to review body worn camera footage in an attempt to locate the defendant. Camera footage will be provided if it is located. The government has been diligent in its efforts to obtain all discoverable items in possession of the FBI.

That still leaves a thousand Facebook pages and 39 photos, some of them taken at a key scene in the Capitol a scene that — given the evidence against Steele-Smith and in other cases — is a relative blind spot in the surveillance of the Capitol. The interview described here is not reflected in her arrest warrant, and so may include non-public information used to support the obstruction case.

Beryl Howell might argue this is sufficient evidence to prove the government’s obstruction case. Trevor McFadden might argue that this case can’t wait for more video evidence obtained from future arrestees of what Steele-Smith did while “storm[ing] the castle” (in her own words), including the office of the Speaker of the House. But the theft of the Pelosi laptop — including whether Groypers like Riley Williams were involved — remains unsolved.

If a single terrorist with suspect ties to foreign entities broke into the office of the Speaker of the House and stole one of her laptops, no one would even think twice if DOJ were still investigating seven months later. But here, because the specific means of investigation include prosecuting the 1,000 people who made that break-in possible, there’s a push to curtail the investigation.

I don’t know what the answer is because the Speedy Trial issues are very real, particularly for people who are detained. But I do know it’s very hard for anyone to get their mind around this investigation.