United States v. Rhine

Opinion

UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA

UNITED STATES OF AMERICA :

:

v. : Criminal Action No.: 21-0687 (RC)

:

DAVID CHARLES RHINE, : Re Document Nos.: 42, 43, 46, 47 :

Defendant. :

:

MEMORANDUM OPINION

DENYING DEFENDANT’S MOTION TO TRANSFER VENUE, GRANTING IN PART AND DENYING IN PART DEFENDANT’S MOTION FOR EXPANDED VOIR DIRE; DENYING DEFENDANT’S MOTIONS TO DISMISS COUNTS 1–4, DENYING DEFENDANT’S MOTION TO SUPPRESS GEOFENCE EVIDENCE

I. INTRODUCTION

Defendant David Charles Rhine is charged with four misdemeanor counts arising out of his alleged participation in the riot at the Capitol on January 6, 2021. Specifically, the Government charged Defendant by information with (1) entering or remaining in a restricted building or grounds in violation of 18 U.S.C. § 1752(a)(1); (2) disorderly or disruptive conduct in a restricted building or grounds in violation of 18 U.S.C. § 1752(a)(2); (3) disorderly conduct in a Capitol building in violation of 40 U.S.C. § 5104(e)(2)(D); and (4) parading, demonstrating or picketing in a Capitol building in violation of 40 U.S.C. § 5104(e)(2)(G). See Information, ECF No. 8. Defendant has filed a motion to transfer venue and for expanded voir dire (ECF No. 42), motions to dismiss the charged counts (ECF Nos. 46, 47), and a motion to suppress evidence obtained pursuant to a “geofence warrant” (ECF No. 43). The motions are ripe for consideration. For the reasons stated below, the Court denies Defendant’s motion to transfer venue, grants in

2

part and denies in part Defendant’s motion for expanded voir dire, denies Defendant’s motions to dismiss the charged counts, and denies Defendant’s motion to suppress.

II. FACTUAL BACKGROUND

At approximately 1:00 p.m. on January 6, 2021, Congress convened to count the votes of the Electoral College and certify the results of the 2020 presidential election. Vice President Mike Pence was present to preside over the session in his role as President of the Senate. About an hour later, at approximately 2:00 p.m., the crowd that had gathered outside the Capitol building began to force its way inside. The Government alleges that Defendant, who resides in Bremerton, Washington, was among that crowd. Specifically, the Government alleges that Defendant entered the capitol at approximately 2:42 p.m. wearing a dark blue hooded jacket, a red hat, and a backpack, and carrying a blue flag with white stars and white cow bells. Gov’t’s Statement of Facts, ECF No. 1-1 at 4. Defendant allegedly proceeded to walk through the Capitol until he encountered a U.S. Capitol Police (“USCP”) officer at approximately 2:57 p.m. Id. at 6. The officer allegedly detained Defendant and conducted a search that yielded two knives and pepper spray, which USCP officers seized before placing Defendant in flex cuffs with his hands behind his back. Id. After escorting Defendant through the hallways for a few minutes, at approximately 3:02 p.m. the USCP officer that detained Defendant allegedly released him, still in flex cuffs, to attend to other responsibilities after Defendant told the officer that he would leave the building. Id. at 8. The Government alleges that one minute later an unidentified individual cut the flex cuffs from Defendant’s hands, and one minute after that, at approximately 3:04 p.m., Defendant left the building. Id. at 8–9.

3

III. ANALYSIS

The Court first considers Defendant’s motion to transfer venue and for expanded voir dire, followed by Defendant’s motions to dismiss and motion to suppress.

A. Defendant’s Motion to Transfer Venue and for Expanded Voir Dire Criminal defendants have a constitutional right to trial by “an impartial jury of the State and district wherein the crime [was allegedly] committed.” U.S. CONST. amend. VI.; see also id. Art. III (“The Trial of all Crimes . . . shall be by Jury; and such Trial shall be held in the State where the said Crimes shall have been committed.”). The Federal Rules of Criminal Procedure reflect the requirement to “prosecute an offense in a district where the offense was committed,” Fed. R. Crim. P. 18, but also permit defendants to move to transfer venue either due to local prejudice or for convenience, Fed. R. Crim P. 21(a)–(b). Where a defendant moves to transfer venue due to local prejudice, the “court must transfer the proceeding . . . to another district if the court is satisfied that so great a prejudice against the defendant exists in the transferring district that the defendant cannot obtain a fair and impartial trial there.” Id.

The Supreme Court has recognized the principle that transfer of venue is a “basic requirement of due process” where “extraordinary local prejudice will prevent a fair trial,” but emphasized that a pre-voir dire “presumption of prejudice . . . attends only the extreme case.” Skilling v. United States, 561 U.S. 358, 378, 381 (2010) (internal citation omitted). Because “juror impartiality . . . does not require ignorance,” even “pervasive, adverse publicity” does not necessarily compel a presumption of prejudice, unless the press coverage is so intense as to “utterly corrupt[]” the trial. Id. at 380–81 (emphasis in original). Accordingly, the “default practice of this jurisdiction [is] to conduct voir dire in order to determine whether a fair and impartial jury can be seated.” United States v. Eicher, No. 22-cr-0038, 2022 WL 11737926, at

4

*2 (D.D.C. Oct. 20, 2022) (citing United States v. Haldemann, 599 F.2d 31, 41 (D.C. Cir. 1976). “‘[A]dequate voir dire to identify unqualified jurors’ is the primary safeguard against jury prejudice.” United States v. Ballenger, No. 21-cr-0719, 2022 WL 16533872, at *1 (D.D.C. Oct. 28, 2022) (quoting Morgan v. Illinois, 504 U.S. 719, 729 (1992)).

In Skilling, the Supreme Court identified three main factors to guide the inquiry into whether prejudice should be presumed before voir dire: (1) the “size and characteristics of the community in which the crime occurred;” (2) whether press coverage of the crime “contain[s] a confession or other blatantly prejudicial information of the type readers or viewers could not reasonably be expected to shut from sight;” and (3) the time that elapsed between the crime and the trial. 561 U.S. at 382–83. Courts in this district have considered a large number of motions to transfer venue similar to that submitted by Defendant in this case. In every case, the court has denied the motion after evaluating the Skilling factors, finding that the defendant failed to establish extraordinary local prejudice that would prevent a fair trial. See Eicher, 2022 WL 11737926, at *1 (denying motion to transfer by defendant charged in connection with January 6, 2021 “[l]ike every other court of this jurisdiction to consider the same argument”); Gov’t’s Opp’n to Mot. Transfer Venue (“Gov’t’s Transfer Opp’n”) at 1 n.1, ECF No. 55 (explaining that “[e]very judge on this Court to have ruled on a motion for change of venue in a January 6 prosecution has denied the motion” and listing cases). After thorough review of Defendant’s motion, the Government’s opposition, and Defendant’s reply, the Court is convinced that the same result should obtain here.

1. Size and Characteristics of the Community Defendant argues that the District of Columbia’s size and characteristics weigh in favor of a presumption of prejudice. Regarding size, Defendant explains that the population of D.C. is

5

approximately 700,000, with 550,000 of voting age, and points out that this number is “far smaller” than the 4.5 million-strong juror pool in the Houston area that the Skilling court held was “large and diverse” such that “the suggestion that 12 impartial individuals could not be empaneled [was] hard to sustain.” Def.’s Mot. Transfer Venue at 8–9 (“Def.’s Mot. Transfer”), ECF No. 42; Skilling, 561 U.S. at 382. But as the Government argues in response, the “relevant question is not whether the District of Columbia is as populous as the Southern District of Texas . . but whether it is large enough that an impartial jury can be found.” Gov’t’s Transfer Opp’n at 6. As other courts in this district have pointed out, the Skilling court itself “recognized a ‘reduced likelihood of prejudice where [the] venire was drawn from a pool of over 600,000 individuals.” United States v. Brock, No. 21-cr-0140, 2022 WL 3910549, at *6 (D.D.C. Aug. 31, 2022) (quoting Skilling, 561 U.S. at 382). Moreover, “the District’s population is greater in size than those few cases in which the Court has found that transfer to a different jurisdiction was constitutionally required” and it is “larger than population sizes that the Supreme Court has found reduced the likelihood of prejudice.” United States v. Rhodes, No. 22-cr-0015, 2022 WL 2315554, at *21 (D.D.C. June 28, 2022) (listing examples, including Mu’Min v. Virginia, 500 U.S. 415, 429 (1991), in which the Supreme Court found no presumption of prejudice despite a jury pool of only approximately 182,000). As every other court to consider the question in connection with a January 6 prosecution has held, the size of D.C. does not weigh in favor of a presumption of prejudice.

Defendant next argues that the “events of January 6 have impacted D.C. residents much more directly than persons outside the District,” such that “the aftershocks of January 6 continue to reverberate.” Def.’s Mot. Transfer at 9. This is unpersuasive, as surely most crimes “more directly” impact the local area than elsewhere, and courts have held that a fair trial is possible

6

even where that impact is a result of particularly heinous crimes. See, e.g., In re Tsarnaev, 780 F.3d 14 (1st Cir. 2015) (affirming denial of motion to transfer venue in Boston Marathon bombing case); Ballenger, 2022 WL 16533872, at *3 (listing additional examples and explaining, in the context of a January 6 prosecution, that a “fair trial is possible even if an event had a significant impact on a community”). Defendant also contends that prejudice is likely because D.C. has a lot of residents employed by government agencies impacted by January 6. See Def.’s Mot. Transfer at 10. However, many of these employees likely were not directly affected by the events of January 6 and, regardless, “[v]ague insinuations that federal employees are biased by their employment represent ‘exactly the kind of conjecture that is insufficient to warrant transfer prior to jury selection.’” Ballenger, 2022 WL 16533872, at *2 (quoting United States v Bochene, 579 F. Supp. 3d 177, 181 (D.D.C. 2022)). Finally, Defendant argues that because the “allegations in this case, by their nature, stoke partisan passions,” the fact that the vast majority of D.C. residents voted for the Democratic candidate for president supports a finding of prejudice. Def.’s Mot. Transfer at 10–11. But such reasoning has been soundly rejected by the D.C. Circuit and district courts considering similar arguments in January 6 cases. See Haldeman, 559 F.2d at 64 n.43 (explaining that “a community’s voting patterns” are not “at all pertinent to venue”); see also Brock, 2022 WL 3910549, at *6 (rejecting January 6 defendant’s argument that the voting pattens of D.C. residents demonstrate a likelihood of prejudice); Eicher, 2022 WL 11737926, at *3 (same); Ballenger, 2022 WL 16533872, at *3 (same).

In sum, the size and characteristics of D.C. weigh against a presumption of prejudice.

The Court is confident that thorough voir dire will be sufficient to root out any prejudice along the lines suggested by Defendant that calls into question a potential juror’s ability to be impartial.

7

See Haldeman, 559 F.2d at 63 (“[I]f an impartial jury actually cannot be selected, that fact should become evident at the voir dire.”).

2. Pretrial Publicity

Defendant emphasizes the “extent and the negative tenor of media coverage of the events that Mr. Rhine’s charges link him to.” Def.’s Mot. Transfer at 11. However, “prominence does not necessarily produce prejudice,” and even “pervasive, adverse publicity” does not necessarily compel a presumption of prejudice. Skilling, 561 U.S. at 381–384; see also Halderman, 559 F.2d at 61 (finding no prejudice from pretrial press coverage despite the presence of articles “hostile in tone and accusatory in content”). While there has certainly been significant media coverage of January 6, as other courts in this district have pointed out, much of it has consisted only of “straightforward, unemotional factual accounts of events and of the progress of official and unofficial investigations.” Ballenger, 2022 WL 16533872, at *4. Moreover, “much of the coverage of January 6 has been national, not local, in nature,” such that the “influence of that coverage would be present wherever trial is held.” Id. (quoting Bochene, 579 F. Supp. 3d at 182); see also United States v. Chapin, 515 F.2d 1274, 1288 (D.C. Cir. 1975) (“[P]recedent demands that the court take into account whether the publicity is sufficiently localized that potential jurors in another area would be free of any taint from exposure to the press, enabling the change to serve its purpose.”); Eicher, 2022 WL 11737926, at *3 (“[M]ost communities throughout the country have been exposed to the exact same coverage [of January 6] as Washingtonians”).

Defendant argues that media coverage of January 6 is analogous to that in Rideau v.

Louisiana, 373 U.S. 723 (1963), in which the Supreme Court reversed a conviction based on a finding that pretrial publicity made a fair trial impossible. See Def.’s Mot. Transfer at 11

8

(quoting the Skilling court’s discussion of Rideau). But Rideau involved pretrial publication of the defendant’s own “dramatically staged admission of guilt.” Skilling, 561 U.S. at 382–83 (describing Rideau). By contrast, here, Defendant points to just three articles that refer to Defendant specifically. See Def.’s Mot. Transfer at 5–6. All of those articles are from local or regional publications in Washington state and therefore do not suggest that the D.C. jury pool has heard anything about this particular prosecution. See Skilling, 561 U.S. at 384 n.17 (“[W]hen publicity is about the event, rather than directed at an individual, this may lessen the prejudicial impact.”); Eicher, 2022 WL 11737926, at *3 (“Defendant identifies no pretrial publicity that identifies her specifically.”); Garcia, 2022 WL 2904352, at *9 (“[W]hile the court recognizes that the events of January 6 are receiving substantial attention in the media at this time, and a rigorous voir dire will be needed to ferret out potential biases, this particular case has not been subject of attention, and this fact also does not weigh in favor of transferring the case.”). In addition, all three articles are objective news pieces that simply describe Defendant’s charged conduct; this is not the kind of “vivid, unforgettable information” that the Skilling court identified as “particularly likely to produce prejudice.” Skilling, 561 U.S. at 384; see also Haldeman, 559 F.2d at 61 (explaining, regarding the Watergate trial, that “[w]e have carefully reviewed the ‘Watergate’ articles submitted by appellants, and we find that the pretrial publicity in this case, although massive, was neither as inherently prejudicial nor as unforgettable as the spectacle of Rideau’s dramatically staged and broadcast confession”); see, e.g., Lewis Kamb, Where the Washington Residents Charged with Breaching the Capitol Are Now, Seattle Times (Jan. 6, 2022, 5:37 PM), https://www.seattletimes.com/seattle-news/law-justice/a-year-later- where-are-the-washington-residents-charged-with-breaching-the-capitol.

9

Finally, Defendant points to the results of a survey and media analysis by Select Litigation commissioned by the defense to bolster his argument that press coverage has rendered D.C. residents incapable of reaching an impartial verdict. See Def.’s Mot. Transfer at 6. First, the D.C. Circuit has explained that “comprehensive voir dire examination conducted by the judge in the presence of all parties and their counsel pursuant to procedures, practices, and principles developed by the common law since the reign of Henry II” is favored over “a poll taken in private by private pollsters and paid for by one side.” Haldeman, 559 F.2d at 64 n.43. Second, courts in this district presented with the same Select Litigation analysis have found that “the surveys are flawed, and none supply persuasive evidence that would support a decision to transfer the case without trying the voir dire process first.” Garcia, 2022 WL 2904352, at *10; see also Rhodes, 2022 WL 2315554, at *21 (“Having considered all of the survey evidence presented by Defendants, the court holds that this is not an ‘extreme case’ in which juror prejudice can be presumed and mandatory transfer is warranted.”); Ballenger, 2022 WL 16533872, at *4 (“Because of the general presumption against supplanting voir dire with polling evidence and because the poll submitted by Defendants fails to establish prejudice even if taken at face value, the Court need not reach the various potential methodological problems with the survey that the Government discusses.”).

The Court agrees that the questionable methodology and unpersuasive results of the Select Litigation survey here do nothing to overcome the D.C. Circuit’s preference for voir dire over a privately commissioned survey. The survey includes tendentious question phrasing that calls its results into question. For example, Defendant emphasizes that 85% of D.C. residents described the actions of “people who forced their way into the U.S. Capitol on January 6, 2021” with the phrase “Trying to overturn the election to keep Donald Trump in Power,” versus 63%

10

nationally. Def.’s Mot. Transfer at 12; Select Litig. Surv. at 15, ECF No. 42-1. 1 But, as the Government points out, the use of the term “forced” “suggests a higher degree of culpability than simply entering the Capitol.” Gov’t’s Transfer Opp’n at 21. Even taking the results at face value, they do not compel a presumption of prejudice. For example, Question 5 of the survey asks respondents, if they were on a jury in a case in which the defendant was charged with “crimes for their activities on January 6th,” whether they would be “more likely to vote that the person is guilty or not.” Select Litig. Surv. at 14. Again, this question’s phrasing is at best ambiguous—“more likely” than what? More likely to vote a person charged with a crime guilty than a person not charged with a crime? Regardless, and despite the fact that the survey only presented respondents with the options “Would” and “Would not,” fully 46% of D.C. respondents volunteered answers of “Depends” or “Don’t know/Refused”—a higher percentage of respondents than the 45% in the control jurisdiction (Atlanta, Georgia). See id. Moreover, the 52% of D.C. residents who responded “Would” is not meaningfully higher than the 45% who said the same in the control jurisdiction, especially considering the margin of error of plus-or- minus 4.9% in this poll. See id. at 13–14. Furthermore, as the Government points out, the poll does not ask the key question that would certainly be probed at voir dire: whether the respondent could lay aside any prior impressions and render an impartial verdict based on the testimony and evidence admitted at trial. See Gov’t’s Transfer Opp’n at 21. Indeed, the closest the poll got was Question 7, which asks if the respondent “think[s] the defendants currently charged with crimes for their activities on January 6th will or will not get a fair trial in the District of

1 Defendant’s brief actually appears to refer to the results of a CBS/YouGov poll considered as part of the Select Litigation media analysis, see Select Litig. Surv. at 4–5, however, a similarly worded question was included in the Select Litigation poll as well, see id. at 15.

11

Columbia,” to which fully 80% of D.C. respondents said they will. 2 Select Litig. Surv. at 14. In short, the Select Litigation survey falls well short of showing that this case is among the extreme cases where a presumption of prejudice compels a transfer of venue.

3. Time Elapsed

More than two years have elapsed since January 6, 2021. It may be that the “decibel level of publicity about the crimes” has lowered in that time, Tsarnaev, 780 F.3d at 22, but the Court acknowledges that recent events, including Congressional hearings and reports and ongoing and potential high-profile criminal prosecutions and civil suits arising out of the events of that day, have likely kept January 6 more toward the top of the public mind than it would be otherwise. However, as Defendant has not suggested that his case in particular has been the subject of any D.C. or national press attention, any residual press coverage of January 6 is merely “a factor that must be taken into consideration in jury selection.” Garcia, 2022 WL 2904352, at *9.

For the reasons stated above, the Court finds that rigorous voir dire will be sufficient to guarantee an impartial jury in the District of Columbia and therefore denies Defendant’s motion to transfer venue. To guarantee a searching voir dire, and in line with the approach taken by other courts in this district, the Court grants Defendant’s alternative request for individual questioning during voir dire, but denies as unnecessary Defendant’s request for a written juror questionnaire. See Gov’t’s Transfer Opp’n at 29 n.8 (noting that only one judge in this district has granted a request to use a juror questionnaire in a January 6 trial); Nassif, 2022 WL 4130841, at *11 (declining to adopt any special voir dire procedures, including individual questioning, and

2 The poll does not appear to have asked this question to respondents in the control jurisdiction.

12

explaining that most courts “have empaneled January 6 cases without resorting to enhanced protocols”); see also Mu’Min v. Virginia, 500 U.S. 415, 431–32 (1991) (affirming murder conviction over defendant’s challenge to trial judge’s denial of motion to individually question jurors or to ask specific questions about news coverage of the crime, explaining that precedent “stress[es] the wide discretion granted to the trial court in conducting voir dire in the area of pretrial publicity and in other areas of inquiry that might tend to show juror bias”).

B. Defendant’s Motion to Dismiss Counts 1 and 2 Defendant moves to dismiss Counts 1 and 2 of the Information, which charge Defendant with entering and remaining in a restricted building or grounds in violation of 18 U.S.C. § 1752(a)(1), and disorderly and disruptive conduct in a restricted building or grounds in violation of 18 U.S.C. § 1752(a)(2). Information at 1–2. Criminal defendants “may raise by pretrial motion any defense, objection, or request that the court can determine without a trial on the meris,” including a motion to dismiss an information because it fails to state an offense. Fed. R. Crim. P. 12(b)(1), 12(b)(3)(B)(v). The Court “must decide every pretrial motion before trial” except on a showing of “good cause.” Fed. R. Crim. P. (12)(d). An information must contain “a plain, concise, and definite written statement of the essential facts constituting the offense charged.” Fed. R. Crim. P. 7(c). When considering a motion to dismiss for failure to state an offense, “the court is limited to reviewing the face of the [charging instrument].” United States v. Lewis, No. 19-cr-0307, 2021 WL 2809819 at *3 (D.D.C. July 6, 2021) (citation omitted). “The operative question is whether the[ ] allegations, if proven, are sufficient to permit a jury to find that the crimes charged were committed.” United States v. Payne, 382 F. Supp. 3d 71, 74 (D.D.C. 2019) (cleaned up).

13

Defendant moves to dismiss Counts 1 and 2 for failure to state an offense, for violation of the non-delegation doctrine, and on grounds that the statute under which Defendant was charged is unconstitutionally vague and overbroad and is an unconstitutional content-based restriction on speech. Def.’s Mot. Dismiss Counts 1 and 2 (“Def.’s 1st MTD”) at 3, ECF No. 46. The Court addresses these arguments in turn.

1. Failure to State an Offense Counts 1 and 2 of the Information charge Defendant under 18 U.S.C. § 1752(a)(1) and § 1752(a)(2), which read:

(a) Whoever--

(1) knowingly enters or remains in any restricted building or grounds without lawful authority to do so;

(2) knowingly, and with intent to impede or disrupt the orderly conduct of Government business or official functions, engages in disorderly or disruptive conduct in, or within such proximity to, any restricted building or grounds when, or so that, such conduct, in fact, impedes or disrupts the orderly conduct of Government business or official functions;

...

[shall be punished as provided in the statute.]

§ 1752(c) defines the terms “restricted buildings or grounds” and “other person protected by the Secret Service” 3 as follows:

(c) In this section--

(1) the term “restricted buildings or grounds” means any posted, cordoned off, or otherwise restricted area--

(A) of the White House or its grounds, or the Vice President's official residence or its grounds;

(B) of a building or grounds where the President or other person protected by the Secret Service is or will be temporarily visiting; or (C) of a building or grounds so restricted in conjunction with an event designated as a special event of national significance; and (2) the term “other person protected by the Secret Service” means any person whom the United States Secret Service is authorized to protect under section 3056

3 The below cross-reference to 18 U.S.C. § 3056 makes clear that the Vice President is an “other person protected by the Secret Service” as that term is used in 18 U.S.C. § 1752.

14

of this title or by Presidential memorandum, when such person has not declined such protection.

Defendant makes two arguments as to why Counts 1 and 2 fail to state an offense. First, he contends that the Information does not allege that the U.S. Secret Service (“USSS”) in fact restricted the area Defendant allegedly encroached upon. Def.’s 1st MTD at 3–7. Second, he argues that Vice President Pence was not “temporarily visiting” the restricted area. Id. at 7–9. Despite the fact that the Court rejected nearly identical arguments in United States v. Andries, No. 21-cr-0093, 2022 WL 768684, at *12–17 (D.D.C. March 14, 2022), Defendant makes no attempt to distinguish the present case. In fact, notwithstanding the Government’s repeated references to Andries in its opposition, Defendant only references the case a single time in its motion or reply—in a footnote for an unrelated proposition. See Def.’s 1st MTD at 18 n.4. Nor does Defendant engage with any of the several other opinions from courts in this district similarly rejecting such arguments in January 6 cases. See, e.g., United States v. Puma, 596 F. Supp. 3d 90, 109–114 (D.D.C. 2022); United States v. Bingert, No. 21-cr-0091, 2022 WL 1659163, at *14–15 (D.D.C. May 25, 2022); United States v. McHugh, 583 F. Supp. 3d 1, 31–35 (D.D.C. 2022); United States v. Anthony Williams, No. 21-cr-0377, ECF No. 88 (D.D.C. June 8, 2022); United States v. Riley Williams, No. 21-cr-0618, 2022 WL 2237301, at *18–20 (D.D.C. June 22, 2022); United States v. Mostofsky, 579 F. Supp. 3d 9, 27–28 (D.D.C. 2021).

Accordingly, the Court sees no basis on which to depart from the reasoning Andries or the other substantially similar cases. Like in these other cases, Defendant first argues that § 1752 requires that the Secret Service be the agency to create a restricted area. Def.’s 1st MTD at 4–7. But, as the other courts in this district have held, § 1752 requires no such thing. § 1752(c) defines “restricted buildings or grounds” as “any posted, cordoned off, or otherwise restricted area” but contains no limiting language concerning which agency must have taken the action to

15

restrict the area. Defendant asks the Court to infer such a limitation based on an argument that the legislative history suggests that the “purpose” of the statute was to “designate the [USSS] to restrict areas.” Def.’s 1st MTD at 4. But, as another Court considering the same argument explained, it is improper to “invoke the statute’s supposed purpose or legislative history to create ambiguity where none exists.” United States v. Griffin, 549 F. Supp. 3d 49, 55 (D.D.C. 2021) (“When, as here, ‘the words of a statute are unambiguous, the judicial inquiry is complete.’” (quoting Babb v. Wilkie, 140 S. Ct. 1168, 1177 (2020)); see also Mostofsky, 579 F. Supp. at 28 (“The text plainly does not require that the Secret Service be the entity to restrict or cordon off a particular area.”); Bingert, 2022 WL 1659163, at *14 (“Nothing in the text indicates that the Secret Service is the only agency that can designate a restricted area.”). Additionally, contrary to Defendant’s argument that the interpretation adopted by every court in this district to consider the issue would yield an “absurd result,” it is Defendant’s proposed reading that would “mean that the exact conduct targeted by the statute is subject to a potentially massive procedural loophole created by silence.” 4 McHugh, 583 F. Supp. 3d at 31. The Court declines to ignore the plain text of the statute in favor of such a “counter-intuitive[]” result. Id.

Defendant’s second argument is that the Vice President was not “temporarily visiting”

the restricted area because “he had a permanent office” at the Capitol in his role as President of the Senate. Def.’s 1st MTD at 7–8. Again, other courts in this district have rejected this

4 The “absurd result” Defendant imagines is that “anyone claiming to be a part of law enforcement could post a sign designating an area as restricted and a person could be prosecuted federally for trespassing because they ‘willfully’ ignored the sign if a Secret Service protectee planned to visit the area.” Def.’s 1st MTD at 7. But as another court facing a nearly identical hypothetical explained, “there is nothing absurd about criminalizing the breach of any barrier around a Secret Service protectee, and the Court will not create its own atextual absurdity based on a fringe hypothetical that does not even remotely resemble the facts before the court.” McHugh, 583 F. Supp. 3d at 32.

16

argument, which is based on “cherry-pick[ed]” dictionary definitions. Bingert, 2022 WL 1659163, at *15; see also United States v. Bronstein, 849 F.3d 1101, 1108 (D.C. Cir. 2017) (“[W]e are interpreting a statute, not restating a dictionary. Our search here is not for every facet of [the terms] but their meaning within the statute.”). As this Court explained in Andries, “[l]ike a President who maintains an office at his home-state residence, and like [a] CEO who maintains a reserve office at her firm’s satellite location, Vice President Pence held an office at the Capitol, but did not use that office as his primary, regular workspace.” Andries, 2022 WL 768684, at *17. Accordingly, “[c]ommon sense easily resolves this debate” against Defendant’s proposed construction. Anthony Williams, No. 21-cr-377, ECF No. 88 at 5 (D.D.C. June 8, 2022); Puma, 596 F. Supp. 3d at 114 (explaining that defendant’s “proposed construction would leave an arbitrary gap in the [statute’s] application”); Riley Williams, 2022 WL 2237301, at *19 (rejecting defendant’s “strained interpretation” as “inconsistent with both the text and structure of the statute,” as well as with common dictionary definitions which “obviously encompass[] Vice President Pence’s actions on January 6, 2021”).

2. Non-Delegation Doctrine Defendant argues that § 1752 violates the non-delegation doctrine because it “delegates to the executive branch the power to define a crime” but “provides no meaningful intelligible principle in this delegation.” Def.’s 1st MTD at 9. Broadly, the non-delegation doctrine protects the separation of powers by prohibiting Congress from delegating “its legislative power to another branch of Government.” Touby v. United States, 500 U.S. 160, 165 (1991). However, “Congress does not violate the Constitution merely because it legislates in broad terms.” Id. “So long as Congress lays down by legislative act an intelligible principle to which the person or body authorized to act is directed to conform, such legislative action is not a forbidden

17

delegation of legislative power.” Id. (cleaned up). Accordingly, the Supreme Court has “almost never felt qualified to second-guess Congress regarding the permissible degree of policy judgment that can be left to those executing or applying the law.” Whitman v. Am. Trucking Assocs., 531 U.S. 457, 474 (2001) (summarizing that, “[i]n the history of the Court we have found the requisite ‘intelligible principle’ lacking in only two statutes, one of which provided literally no guidance for the exercise of discretion, and the other of which conferred authority to regulate the entire economy on the basis of no more precise a standard than stimulating the economy by assuming ‘fair competition’” (citations omitted)).

Defendant claims that Congress impermissibly delegated legislative authority because the statute “does not provide any parameters, purposes, or other guidance to the Secret Service in deciding the spatial area to restrict or the length of time to so restrict it” and because “Congress did not specify what methods should be used to restrict access, whether it be by creating barriers, staffing security, etc.” Def.’s 1st MTD at 15. The Government contends that Defendant’s argument fails at the threshold, as § 1752 does not delegate authority to USSS or another agency to restrict certain areas in the first place. See Gov’t’s Opp’n to Def.’s 1st MTD (“Gov’t’s 1st MTD Opp’n”) at 17, ECF No. 56. The Government points to the fact that § 1752(a) simply provides that “[w]hoever” takes the prohibited actions “shall be punished” as provided in subsection (b), while separate statutes “grant[] the Capitol Police and the Secret Service the authority to define a restricted area within the Capitol Grounds on January 6, 2021.” Id. at 18.

A conceptual distinction clarifies the disagreement between the parties: criminal delegations can be either explicit or implicit. An explicit delegation occurs where Congress passes a law granting an agency or government official some authority to define a crime, while an implicit delegation attaches to every criminal statute as a “necessary byproduct of

18

prosecutors’ charging power.” F. Andrew Hessick & Carissa Byrne Hessick, Nondelegation and Criminal Law, 107 V.A. L. REV. 281, 330 (2021); cf. Guedes v. Bureau of Alcohol, Tobacco, Firearms, and Explosives, 920 F.3d 1, 22 (D.C. Cir. 2019) (“If a statute contains ambiguity, Chevron [U.S.A. v. Nat’l Resources Def. Council, 467 U.S. 837 (1984)] directs courts to construe the ambiguity as an implicit delegation from Congress to the agency to fill in the statutory gaps.” (internal quotation omitted)). Within this framework, the Court interprets the Government as arguing that, because § 1752 involves merely an implicit delegation that amounts to nothing more than ordinary prosecutorial discretion, it does not run afoul of the nondelegation doctrine. Defendant’s argument, on the other hand, is that the implied delegation of authority to define a “restricted area” under § 1752 is so broad that it amounts to the legislating from the executive branch.

The Court need not reach the questions of whether § 1752 in fact constitutes an implicit criminal delegation, or whether implicit criminal delegations are entitled to a more forgiving nondelegation test, as § 1752 would clearly pass muster even under the traditional “intelligible principle” test the Supreme Court has applied to explicit criminal delegations. See Touby, 500 U.S. at 165 (applying the “intelligible principle” test to a statute permitting the Attorney General to temporarily designate drugs as Schedule I controlled substances under the Controlled Substances Act); Gundy v. United States, 139 S. Ct. 2116, 2123 (2019) (applying the “intelligible principle” test to statute giving Attorney General authority to require sex offenders convicted before its passage to register pursuant to its requirements, which authorize prosecution for failure to register).

Recall that § 1752(c) defines “restricted buildings or grounds” as follows:

(c) In this section--

19

(1) the term “restricted buildings or grounds” means any posted, cordoned off, or otherwise restricted area--

(A) of the White House or its grounds, or the Vice President's official residence or its grounds;

(B) of a building or grounds where the President or other person protected by the Secret Service is or will be temporarily visiting; or (C) of a building or grounds so restricted in conjunction with an event designated as a special event of national significance; and (2) the term “other person protected by the Secret Service” means any person whom the United States Secret Service is authorized to protect under section 3056 of this title or by Presidential memorandum, when such person has not declined such protection.

Now consider a hypothetical in which § 1752(c) contained a third subparagraph reading, “(3) the Director of the United States Secret Service shall have authority to define the term ‘otherwise restricted area’ as used in paragraph (1) of this subsection.” The long history of Supreme Court precedent makes abundantly clear that this hypothetical explicit delegation would easily satisfy the intelligible principle test and would not be an unconstitutional delegation of legislative authority. For example, in Touby the Supreme Court found it so clear that one “cannot plausibly argue” that an intelligible principle was not embodied in the spare requirement that the Attorney General only temporarily designate Schedule I drugs upon finding it “necessary to avoid an imminent hazard to the public safety.” Touby, 500 U.S. at 165–66. As the Gundy plurality explained, in upholding the delegation to the Attorney General of authority to require those with pre-enactment sex offense convictions to register as sex offenders,

[W]e have over and over upheld even very broad delegations. Here is a sample: We have approved delegations to various agencies to regulate in the “public interest.” We have sustained authorizations for agencies to set “fair and equitable” prices and “just and reasonable” rates. We more recently affirmed a delegation to an agency to issue whatever air quality standards are “requisite to protect the public health.” And so forth.

Gundy, 139 S. Ct. 2116, 2129 (citations omitted). Accordingly, the method-based (e.g., “posted, cordoned off”), place-based (e.g., “a building or grounds where the President or other person protected by the Secret Service is or will be temporarily visiting”), and purpose-based (“a

20

building or grounds so restricted in conjunction with an event designated as a special event of national significance”) guidance contained in § 1752(c) easily satisfies the intelligible principle test. 5 3. Vagueness and Overbreadth Defendant argues that “section 1752 is so broad and its parameters so unclear that an ordinary person could not discern what conduct is criminalized by the statute.” Def.’s 1st MTD at 17–18. Defendant treats the doctrines of vagueness and overbreadth together. While overlapping, they are meaningfully distinct, so the Court analyzes Defendant’s arguments under the applicable legal framework.

With respect to Defendant’s facial vagueness challenge, due process requires that a criminal statute not be “so vague that it fails to give ordinary people fair notice of the conduct it punishes, or so standardless that it invites arbitrary enforcement.” Johnson v. United States, 576 U.S. 591, 595 (2015). “[T]he touchstone is whether the statute, either standing alone or as construed, made it reasonably clear at the relevant time that the defendant's conduct was

5

Defendant also argues that § 1752 violates the “major questions doctrine.” Def.’s 1st MTD at 15. In W. Va. v. Env’t Prot. Agency, 142 S. Ct. 2587, 2608 (2022), the Supreme Court held that the question of “whether Congress in fact meant to confer the power the agency has asserted” may be relevant to the nondelegation analysis in “extraordinary cases” in which the “history and breadth of the authority that the agency has asserted and the economic and political significance of that assertion, provide a reason to hesitate before concluding that Congress meant to confer such authority.” There is no plausible argument that the question of whether a supposed delegation to “otherwise restrict[]” an area to protect the President, Vice President, or other Secret Service protectee rises to this level. § 1752(c); see, e.g., W. Va., 142 S. Ct. at 2608 (identifying, as a possible example of a claimed delegation rising to the level of an “extraordinary case,” the Centers for Disease Control and Prevention adopting a nationwide eviction moratorium during the COVID-19 pandemic). Moreover, it is abundantly clear that if Congress intended to delegate anything in § 1752, what was done here—restricting access to the Capitol while Congress, with the Vice President presiding, carried out its constitutional and statutory duty to count the votes of the Electoral College—would fall squarely within the scope of that delegation.

21

criminal.” United States v. Lanier, 520 U.S. 259, 267 (1997). This relatively low bar is cleared where an “imprecise” statute “whose satisfaction may vary depending upon whom you ask” nonetheless provides a “comprehensive normative standard.” United States v. Bronstein, 849 F.3d 1011, 1107 (D.C. Cir. 2017) (internal quotation omitted). “Rather, a statute is unconstitutionally vague if, applying the rules for interpreting legal texts, its meaning specifies no standard of conduct at all.” Id. (cleaned up).

With respect to Defendant’s facial overbreadth challenge, a statute is facially overbroad under the First Amendment if it “punishes a substantial amount of protected free speech, judged in relation to the statute’s plainly legitimate sweep.” Virginia v. Hicks, 539 U.S. 113, 119 (2003) (internal quotations omitted). 6 The overbreadth analysis requires the Court first to “construe the challenged statute” to determine “what [it] covers,” and then to determine if it “criminalizes a substantial amount of protected expressive activity.” United States v. Williams, 553 U.S. 285, 293, 297 (2008). The doctrine guards against the chilling effect that an overbroad law can have

6 The “legitimate sweep” of a regulation on expressive activity depends on the type of public property in which it occurred, so forum analysis—an inquiry into whether the expressive activity occurred in a traditional public forum, a designated public forum, or a nonpublic forum—is a predicate step in the overbreadth analysis. See Initiative and Referendum Inst. v. U.S. Postal Serv., 417 F.3d 1299, 1313 (D.C. Cir. 2005) (“Given our conclusion that [a statute] is unconstitutional when applied to a public forum, one way in which the regulation would be overbroad is if a substantial number of exterior postal properties constitute public forums”). However, the Court need not wade into forum analysis here because it finds that § 1752 is content-neutral and directed at conduct, not speech, and therefore would not prohibit a substantial amount of protected expression in relation to its legitimate sweep regardless of the forum. See Virginia v. Hicks, 539 U.S. 113, 124 (2003) (“Rarely, if ever, will an overbreadth challenge succeed against a law or regulation that is not specifically addressed to speech or to conduct necessarily associated with speech (such as picketing or demonstrating).”; Mahoney v. United States Capitol Police Bd., 566 F. Supp. 3d 1, 10 (D.D.C. 2022) (upholding regulations on demonstration activity on Capitol grounds put in place after the events of January 6 in part on grounds that September 11, 2001 and January 6 created a “very different security posture” that makes it “eminently reasonable for the Government to submit that greater restrictions and security measures are now warranted to serve the admittedly significant interests at stake”).

22

on constitutionally protected speech, “especially when the overbroad statute imposes criminal sanctions.” Hicks, 539 U.S. at 118–19. However, the Supreme Court has emphasized that declaring a statute facially overbroad is “strong medicine,” to be employed “sparingly and only as a last resort” where no “limiting construction” is possible. Broadrick v. Oklahoma, 413 U.S. 601, 613 (1973).

Defendant makes a few combined arguments as to why the Court should find § 1752(a)(1) and (a)(2) unconstitutionally vague and overbroad. He contends that “the government’s interpretation makes criminal any encroachment past the restriction of any agency” and that the statute does not specify which “visits” by a Secret Service protectee “may occasion qualifying restrictions.” 7 Def.’s 1st MTD at 22–23 (emphasis in original). But regardless of which government agency restricted the area or where a qualifying “visit” may take place, § 1752(a)(1) and (a)(2), as narrowed by the definitions in § 1752(c), clearly pass the low bar to provide “fair notice of the conduct [they] punish[].” Johnson v. United States, 576 U.S. 591, 595 (2015); United States v. Bozell, No. 21-cr-0216, 2022 WL 474144, at *9 (D.D.C. Feb. 16, 2022) (holding that “§1752 ‘is clear, gives fair notice of the conduct it punishes, and [does not] invite arbitrary enforcement’” (citation omitted)). 8 Similarly, Defendant also argues that “the statute provides scant guidance on what ‘restriction’ renders an area off limits.” Def.’s 1st MTD at 23. It is true that § 1752(c) defines “restricted buildings or grounds” to “mean[] any

7 Defendant also makes a three-sentence argument that the statute “includes no causal nexus between restriction of the area and the visit, or anticipated visit, of a Secret Service protectee,” Def.’s 1st MTD at 24. But this is precisely the function of § 1752(c), which defines “restricted buildings or grounds” to include a restricted area of “a building or grounds where [a Secret Service protectee] is or will be temporarily visiting.” § 1752(c)(1)(B).

8 While Defendant merges his arguments regarding vagueness and overbreadth, his arguments about which agency restricts the area and which visits by Secret Service protectees qualify sound only in vagueness.

23

posted, cordoned off, or otherwise restricted area.” § 1752(c)(1) (emphasis added). But this “just means that the statute does not require any particular method for restricting a building or grounds.” McHugh, 583 F. Supp. 3d at 31 (explaining that § 1752(c) is not “ambiguous or unclear”); see also Griffin, 549 F. Supp. 3d at 55, 57 (explaining that “[j]ust because Congress left this part of the statute open-ended does not mean any word or phrase is ambiguous” and rejecting defendant’s vagueness claim as “rehash[ed] . . . complaints about the Government’s reading of the statute”). It is therefore unmysterious that multiple courts in this district have rejected similar arguments that § 1752 is unconstitutionally vague. See, e.g., Nordean, 579 F. Supp. 3d 28, 60 (D.D.C. 2021) (holding that § 1752 “is not unconstitutionally vague”); Griffin, 549 F. Supp. 3d at 57.

Next, Defendant argues that § 1752 is vague and overbroad because it “lack[s] temporal or spatial limits” in § 1752(c)’s definition of “restricted buildings or grounds.” Def.’s 1st MTD at 22. Defendant cites a single case from 1926, Connally v. Gen’l Const. Co., 269 U.S. 385 (1926), for the proposition that the “Supreme Court has similarly found vagueness in statutes that rest on the fuzzy boundary standards of ‘neighborhood’ and ‘locality.’” Id. But the statute in Connally created penalties for violation of a requirement to pay wages “not less than the current rate of per diem wages in the locality” without defining what qualifies as a “locality.” Connally, 269 U.S. at 388. By contrast, § 1752(a)(1) and (a)(2) both require an individual to “knowingly” perform the prohibited conduct, and (a)(2) further requires an individual to have the “intent to impede or disrupt the orderly conduct of Government business or other official functions” together with a result element requiring that such business or functions were in fact impeded or disrupted. And, in stark contrast to the undefined term “locality” in the Connally statute, § 1752(c) provides a three-part definition of the term “restricted buildings or grounds.” In this

24

way, § 1752 in fact does contain spatial limitations. Compare, e.g., Griffin, 549 F. Supp. 3d at 57 (“[§ 1752] is no trap awaiting the unwary.”) with Coates v. City of Cincinnati, 402 U.S. 611, 614 (1971) (striking down as overbroad a statute prohibiting groups of three or more people to assemble on a sidewalk and “conduct themselves in a manner annoying to persons passing by” because “no standard of conduct is specified at all”). Moreover, there is good reason for flexibility in where and how long an area can be restricted under § 1752, as Secret Service protectees tend to have busy and unpredictable schedules.

The alleged lack of temporal or spatial limitations also does not render § 1752 overbroad.

A threshold question in the overbreadth analysis is whether either § 1752(a)(1) or (a)(2) punish a “substantial amount of protected speech.” Hicks, 539 U.S. at 118. The Court finds that they do not. Beginning with § 1752(a)(1), this subparagraph merely makes it unlawful to “enter or remain” in a restricted building or grounds. As another court in this district held in rejecting an as-applied overbreadth challenge to this subsection, “the statute is not related to the suppression of free expression.” United States v. Caputo, 201 F. Supp. 3d 65, 71 (D.D.C. 2016). In other January 6 cases, courts in this district also rejected overbreadth challenges to a different statute with a broader prohibition on “any act to obstruct, impede or interfere” with law enforcement in the performance of official duties. 18 U.S.C. § 231(a)(3) (emphasis added); see, e.g., McHugh, 583 F. Supp. 3d at 28–29; Nordean, 579 F. Supp. 3d at 58 n.15. “Rarely, if ever, will an overbreadth challenge succeed against a law or regulation that is not specifically addressed to speech or to conduct necessarily associated with speech (such as picketing or demonstrating).” Hicks, 539 U.S. at 124.

With respect to § 1752(a)(2), this subparagraph’s prohibition on “disorderly or disruptive conduct” could potentially reach protected expressive activity. However, by its plain text, the

25

statute also is directed toward conduct, not speech. See United States v. Williams, 553 U.S. 285, 293 (2008) (“The first step in overbreadth analysis is to construe the challenged statute.”); cf. United States v. Phomma, 561 F. Supp. 3d 1059, 1067–68 (D.D.C. 2021) (finding that 18 U.S.C. § 231(a)(3)’s prohibition on “any act to obstruct, impede, or interfere” with law enforcement duties is “directed toward conduct rather than speech”). In addition, § 1752(a)(2) includes several limiting provisions that narrow its applicability. First, the subparagraph includes a requirement that the individual have the “intent to impede or disrupt the orderly conduct of Government business or official functions.” § 1752(a)(2); see Williams, 553 U.S. at 293–94 (finding it “important” to its analysis rejecting an overbreadth challenge that the statute contained a “scienter requirement” that the prohibited conduct be done “knowingly”). Importantly, § 1752(a)(2) also contains a result element requiring that such government business or official functions in fact be impeded. This substantially limits the amount of potentially protected expression that could theoretically be caught up in the statute’s “plainly legitimate sweep.” Hicks, 539 U.S. at 118–19. Finally, as discussed above, the definitions in § 1752(c) further limit its applicability. Accordingly, the Court is satisfied that § 1752(a)(2) does not restrict a substantial amount of protected expressive activity in relation to its “core” of legitimate applications. Smith v. Goguen, 415 U.S. 566, 578 (1974). “[T]he mere fact that one can conceive of some impermissible applications of a statute is not sufficient to render it susceptible to an overbreadth challenge.” Members of the City Council v. Taxpayers for Vincent, 466 U.S. 789, 800 (1984); Hicks, 539 U.S. at 124 (“Rarely, if ever, will an overbreadth challenge succeed against a law or regulation that is not specifically addressed to speech or to conduct necessarily associated with speech (such as picketing or demonstrating).”

26

In summary, the Court holds that § 1752(a)(1) and (a)(2) are neither unconstitutionally vague nor overbroad.

4. Content-Based Speech Regulation Finally, Defendant claims that § 1752 “restricts speech and expressive conduct based on its subject matter and purpose,” and specifically political speech, and therefore that it is “presumptively unconstitutional.” Def.’s 1st MTD at 28. While it is true that content-based restrictions on speech generally must pass strict scrutiny, see Reed v. Town of Gilbert, Ariz., 576 U.S. 155, 163 (2015), no such restriction is at issue here. A speech regulation is content-based if it “targets speech based on its communicative content—that is, if it applies to particular speech because of the topic discussed or the idea or message expressed,” but not if it is “agnostic as to content.” City of Austin, Tex. v. Reagan Nat’l Advert. of Austin, 142 S. Ct. 1464, 1471 (2022) (cleaned up). As explained above, § 1752(a)(1) and (a)(2) are directed at conduct, not speech, but insofar as the targeted conduct entails protected expressive activity, the statute’s restrictions are clearly agnostic as to content. The prohibitions on “enter[ing] or remain[ing]” in a restricted building, § 1752(a)(1), or “engag[ing] in disorderly or disruptive conduct” with the “intent to impede or disrupt the orderly conduct of Government business or official functions,” § 1752(a)(2), simply “do not single out any topic or subject matter for differential treatment,” City of Austin, 142 S. Ct. at 1472. Compare Ward v. Rock Against Racism, 491 U.S. 781, 792 (1989) (finding that a city rule generally limiting sound amplification “has nothing to do with content” (internal quotation omitted)) with Police Dep’t of City of Chi. v. Mosley, 408 U.S. 92, 95 (1972) (explaining that the “central problem” with an ordinance exempting labor picketing from a general prohibition on picketing near schools is that “it describes permissible picketing in terms of its subject matter”). It is telling that the principal case Defendant cites to support his

27

argument is City of Austin, in which the Supreme Court held that a regulation containing differential rules for signs posted on business premises and off business premises was not a content-based regulation of speech because it was “agnostic as to content.” City of Austin, 142 S. Ct. at 1471. So too here.

For the foregoing reasons, Defendant’s Motion to Dismiss Counts 1 and 2 is denied.

C. Defendant’s Motion to Dismiss Counts 3 and 4 Defendant moves to dismiss Counts 3 and 4 of the Information, which charge Defendant with engaging in disorderly and disruptive conduct in the Capitol with the intent to impede, disrupt, and disturb the orderly conduct of a session of Congress, in violation of 40 U.S.C. § 5104(e)(2)(D), and parading, demonstrating, and picketing in the Capitol, in violation of 40 U.S.C. § 5104(e)(2)(G). See Information at 2–3; Def.’s Mot. to Dismiss Counts 3 & 4 (“Def.’s 2d MTD”) at 2–3, ECF No. 47. As laid out above, an information must contain “a plain, concise, and definite written statement of the essential facts constituting the offense charged.” Fed. R. Crim. P. 7(c). When considering a motion to dismiss for failure to state an offense, “the court is limited to reviewing the face of the [charging instrument].” Lewis, 2021 WL 2809819 at *3 (citation omitted). “The operative question is whether the[ ] allegations, if proven, are sufficient to permit a jury to find that the crimes charged were committed.” Payne, 382 F. Supp. 3d at 74 (cleaned up).

Defendant argues that the statute under which these counts are brought is unconstitutionally vague and overbroad, that the statute is an unconstitutional content-based restriction on protected expression, and that the Information fails to state an offense as to these counts.

28

1. Vagueness and Overbreadth The legal framework for evaluating vagueness and overbreadth, as laid out supra Section III.B.3 in relation to Defendant’s Motion to Dismiss Counts 1 and 2, also applies here. With respect to Counts 3 and 4, Defendant argues that 40 U.S.C. § 5104(e)(2)(D) and (e)(2)(G) are vague and overbroad principally because they do not contain detailed definitions of certain terms. See Def.’s 2d MTD at 7, 11. Just as in Defendant’s Motion to Dismiss Counts 1 and 2, in his Motion to Dismiss Counts 3 and 4 Defendant again somewhat conflates the overlapping but analytically distinct doctrines of vagueness and overbreadth, so the Court again has endeavored to fit his arguments to the applicable framework.

§ 5104(e)(2)(D) and (e)(2)(G) make it illegal to “willfully and knowingly,”

(D) utter loud, threatening, or abusive language, or engage in disorderly or disruptive conduct, at any place in the Grounds or in any of the Capitol Buildings with the intent to impede, disrupt, or disturb the orderly conduct of a session of Congress or either House of Congress, or the orderly conduct in that building of a hearing before, or any deliberations of, a committee of Congress or either House of Congress; [or]

...

(G) parade, demonstrate, or picket in any of the Capitol Buildings.

With respect to vagueness, Defendant claims that subparagraph (e)(2)(D) is unconstitutionally vague because it criminalizes “disorderly or disruptive conduct” but “provides no definition of these terms.” Def.’s 2d MTD at 7. Defendant makes substantially similar arguments to support this position as he did with respect to the same term—“disorderly or disruptive conduct”—as used in 18 U.S.C. § 1752(a)(2), which the Court rejected supra Section III.B.3. Defendant cites to a more detailed general disorderly conduct law in the District of Columbia, see Def.’s 2d MTD at 8–9, but the fact that another legislature drew a narrower statute says nothing about whether the broader one is unconstitutionally vague. Moreover, the term “disorderly or disruptive conduct” does not stand alone, but must be read in light of the immediately preceding

29

clause that prohibits “utter[ing] loud, threatening or abusive language.” See United States v. Bronstein, 849 F.3d 1101, 1108 (D.C. Cir. 2017) (holding that the terms “harangue” and “oration” in a statute barring disorderly conduct at the Supreme Court were not unconstitutionally vague in part based on more specific prohibitions included in the statute under the interpretative canon whereby “a word is known by the company it keeps”). In addition, similar to § 1752(a)(2), the prohibition on “disorderly or disruptive conduct” in § 5104(e)(2)(D) ties the prohibited conduct to a particular location and requires that the conduct be done “willfully and knowingly” and also “with the intent to impede, disrupt, or disturb.” This is more than enough to enable “a person of ordinary intelligence [to] read this law and understand” the proscribed conduct. Bronstein, 849 F.3d at 1110; see also United States v. Williams, 553 U.S. 285, 306 (2008) (“Close cases can be imagined under virtually any statute. The problem that poses is addressed, not by the doctrine of vagueness, but by the requirement of proof beyond a reasonable doubt.”). Thus, § 5104(e)(2)(D) is not unconstitutionally vague. Nor is § 5104(e)(2)(G), which is even more specific as to the proscribed conduct: parading, demonstrating, or picketing. See Nassif, 2022 WL 4130841, at *7 (“Section 5104(e)(2)(G) requires that an individual willfully and knowingly parade, picket, or demonstrate inside the Capitol building. This language provides sufficient guidance as to what is prohibited.”). 9 Overbreadth presents a closer question. Taking subparagraph (e)(2)(D) first, unlike 18 U.S.C. § 1752, on its face subparagraph (e)(2)(D) is directed at speech—while the Information only charges Defendant with engaging in “disorderly and disruptive conduct,” the

9 Defendant makes a series of arguments based on the legislative history of § 5104(e)(2)(G), but, as the Nassif court well explained in rejecting similar arguments, “reliance on legislative history is misplaced where the plain text of the statute leaves no need to resort to alternative methods of interpretation.” Nassif, 2022 WL 4130841, at *7.

30

statute also makes it illegal to “utter loud, threatening or abusive language.” See Information at 2; 40 U.S.C. § 5104(e)(2)(D). 10 So too subparagraph (e)(2)(G), whose prohibitions on parading, demonstrating, or picketing extend to conduct “necessarily associated with speech.” Hicks, 539 U.S. at 124. The question is whether the statute restricts a “substantial amount” of protected speech in relation to its “plainly legitimate sweep.” Id. at 118–19 (quotation omitted).

The scope of that “legitimate sweep” depends on an inquiry into the type of public property—the “forum”—where the speech regulation applies. See Initiative and Referendum Inst. v. U.S. Postal Serv., 417 F.3d 1299, 1313 (D.C. Cir. 2005) (engaging in forum analysis before explaining that “one way in which the regulation would be overbroad is if a substantial number of [relevant public properties] constitute public forums”). Forum analysis “divides government property into three categories, and the category determines what types of restrictions will be permissible.” Initiative and Referendum Inst. v. U.S. Postal Serv., 685 F.3d 1066, 1070 (D.C. Cir. 2012). The “traditional public forum” includes places like public streets and parks “which by long tradition or by government fiat have been devoted to assembly and debate.” Cornelius v. NAACP Legal Def. and Educ. Fund, 473 U.S. 788, 802 (1985) (quotation omitted). Content-based government regulation of speech in traditional public forums is subject to strict scrutiny, under which the regulation must be “necessary to serve a compelling state interest” and “narrowly drawn to achieve that end,” Perry Educ. Ass’n v. Perry Local Educators’ Ass’n, 460 U.S. 37, 45 (1983). However, content-neutral regulations are subject only to the “less

10 The Government omits the latter language when quoting the statute, but the fact that Defendant is not charged with engaging in the conduct proscribed by that language is not relevant for purpose of a facial overbreadth challenge. See Mass. v. Oakes, 491 U.S. 576, 581 (1989) (“The First Amendment doctrine of substantial overbreadth is an exception to the general rule that a person to whom a statute may be constitutionally applied cannot challenge the statute on the ground that it may be unconstitutionally applied to others.”).

31

demanding time, place, or manner test,” Initiative and Referendum Inst., 685 F.3d at 1070, under which regulations need only be “narrowly tailored to serve a significant government interest” while leaving open “ample alternative channels of communication,” Perry Educ. Ass’n, 460 U.S. at 45. Designated public forums “may be created by government designation of a place or channel of communication for use by the public at large for assembly and speech, for use by certain speakers, or for the discussion of certain subjects.” Cornelius, 473 U.S. at 802. Designated public forums, “[s]o long as the government maintains the public designation of the forum,” are bound by “the same standards as apply in a traditional public forum.” Bynum v. U.S. Capitol Police Bd., 93 F. Supp. 2d 50, 55 (D.D.C. 2000) (citing Perry Educ. Ass’n, 460 U.S. at 46). Nonpublic forums are “all remaining public property.” Int’l Soc’y for Krishna Consciousness v. Lee, 505 U.S. 672, 678–79 (1992). In nonpublic forums, “[i]n addition to time, place, and manner regulations, the state may reserve the forum for its intended purposes, communicative or otherwise,” as long as any regulations are reasonable and viewpoint-neutral. Perry Educ. Ass’n, 460 U.S. at 46.

The Court addresses § 5104(e)(2)(G) first. Multiple courts in this district have found that the interior of the Capitol buildings is a nonpublic forum. See, e.g., Nassif, 2022 WL 4130841, at *4 (adopting the conclusion reached in Bynum, 93 F. Supp. 2d at 56 that the “interior of the Capitol building is a nonpublic forum”); Order at 2, United States v. Ballenger, No. 21-cr-0719 (D.D.C. Oct. 26, 2022), ECF No. 70 (same). The Court agrees with this conclusion, and finds that the legal framework applicable to nonpublic forums, under which the prohibition need only be reasonable and view-point neutral, applies to § 5104(e)(2)(G), because that subparagraph applies only “in any of the Capitol Buildings.”

32

First, § 5104(e)(2)(G) is “reasonable in light of the purpose served by the forum.” Cornelius v. NAACP Legal Def. and Educ. Fund, 473 U.S. 788, 806 (1985). The Capitol buildings serve the purposes to permit “Congress peaceably to carry out its lawmaking responsibilities” and to permit “citizens to bring their concerns to their legislators.” Nassif, 2022 WL 4130841, at *5 (quoting Bynum, 93 F. Supp. 2d at 55). It plainly serves these interests to prohibit “loud, threatening, or abusive language,” “disorderly or disruptive conduct,” or “parad[ing], demonstrate[ing], or picket[ing]” inside the Capitol buildings. See id. Defendant’s best argument concerns the word “demonstrate,” as used in § 5104(e)(2)(G). See Def.’s 2d MTD at 15. In a vacuum, the term could be interpreted broadly to include activity not reasonably related to the goals to permit Congress to do its work and citizens to petition their representatives. But, as noted above, “a word is known by the company it keeps.” Bronstein, 849 F.3d at 1108. The word “demonstrate” is sandwiched between the words “parade” and “picket.” As the Nassif court explained, “[w]hen read in light of its neighbors” § 5104(e)(2)(G) “applies to organized conduct advocating a viewpoint, not to off-handed expressive conduct or remarks.” Nassif, 2022 WL 4130841, at *6–7 & n.9. 11 Thus, the Court reads “demonstrate” as limited to organized activity that could disrupt Congress from carrying out its business, in the vein of parading or picketing. The title of § 5104(e)(2), “Violent Entry and Disorderly Conduct,” also “confirms [the Court’s] construction of the text.” Bronstein, 849 F.3d at 1109. The Court finds that § 5104(e)(2)(G)’s regulation of organized demonstration activity is reasonable in light of the statute’s purposes.

11 For this reason, the Court follows the Nassif court in rejecting identical extreme hypotheticals Defendant asserts based on his inappropriately broad construction of the term “demonstrate.” See Nassif, 2022 WL 4130841, at *7 n.9; Def.’s 2d MTD at 16.

33

The Court also finds that § 5104(e)(2)(G) is clearly viewpoint neutral, as it “contains nothing limiting its application to a particular viewpoint.” Nassif, 2022 WL 4130841, at *5. Accordingly, in light of the legitimate sweep of the statute, while impermissible applications may be theoretically possible, the Court finds that it does not prohibit a substantial amount of protected speech, and therefore that the “strong medicine” of declaring it facially overbroad is unwarranted. Broadrick, 413 U.S. at 613; see Taxpayers for Vincent, 466 U.S. at 800 (“[T]he mere fact that one can conceive of some impermissible applications of a statute is not sufficient to render it susceptible to an overbreadth challenge.”).

Turning to § 5104(e)(2)(D), this subparagraph applies “at any place in the Grounds or in any of the Capitol Buildings.” While its application is limited to activity intended to “impede, disrupt, or disturb the orderly conduct of a session of Congress or either House of Congress, or the orderly conduct in that building,” in theory this subparagraph could apply to expressive activity on the grounds outside the Capitol buildings, which the D.C. Circuit has held is a public forum. See Hodge v. Talkin, 799 F.3d 1145, 1161 (D.C. Cir. 2015) (citing Lederman v. United States, 291 F.3d 36, 41–42 (D.C. Cir. 2002)). However, as explained infra Section III.C.2, § 5104(e)(2)(D) is content-neutral, as its prohibitions on “loud, threatening, or abusive language” and “disorderly or disruptive conduct” are plainly “agnostic as to content.” City of Austin, 142 S. Ct. at 1471. Accordingly, “intermediate scrutiny” applies, under which the statute is constitutional so long as it “furthers an important or substantial government interest” that is “unrelated to the suppression of free expression,” any incidental restriction on protected expressive activity is “no greater than is essential to the furtherance of that interest,” and it “leaves open ample alternative channels for communication.” Edwards v. Dist. of Columbia,

34

755 F.3d 996, 1001–02 (D.C. Cir. 2014) (quoting United States v. O’Brien, 391 U.S. 367, 377 (1968)).

All of these requirements are met here. As another court in this district found in upholding regulations on demonstration activity on the Capitol grounds against a facial First Amendment challenge, “[i]t is well established that ‘ensuring public safety and order’ is a significant government interest,” and “[t]hat interest is amplified near the Capitol . . . where prominent public officials are present and conducting official government business.” Mahoney v. United States Capitol Police Bd., 566 F. Supp. 3d 1, 9 (D.D.C. 2022) (quoting United States v. Mahoney, 247 F.3d 279, 286 (D.C. Cir. 2001)). If there was any doubt, the events of January 6 and subsequent prosecutions of suspected participants under § 5104(e)(2)(D) clearly show that the statute furthers this important government interest. The Court is also convinced that § 5104(e)(2)(D) is narrowly tailored. In stark contrast to the “virtually per se ban on expressive activity” on the Capitol grounds that the D.C. Circuit declared unconstitutional for lack of narrow tailoring in Lederman v. United States, 291 F.3d 36, 45 (D.C. Cir. 2002), § 5104(e)(2)(D)’s prohibition on expressive activity is tethered to a requirement that the individual have the intent to “impede, disrupt, or disturb the orderly conduct of a session of Congress or either House of Congress, or the orderly conduct in that building” of congressional committees. Indeed, the Lederman court explicitly cited the predecessor statute to § 5104(e)(2)(D), which is substantively identical to the present version, as an example of a “substantially less restrictive alternative[] that would equally effectively promote safety.” Lederman, 291 F.3d at 45 (cleaned up); see also Mahoney, 566 F. Supp. 3d at 10–11 (distinguishing Lederman on similar grounds). For the same reason, the Court is satisfied that

35

any incidental restriction on protected expression caused by § 5104(e)(2)(D) is no greater than necessary.

Finally, § 5104(e)(2)(D) leaves open ample alternative channels of communication. The Lederman court suggested that exemptions for “expressive tee-shirts and buttons” amid an otherwise “total restriction” on speech “may establish that it leaves open ample alternative channels of communication.” Lederman, 291 F.3d at 45 (cleaned up). By contrast, here, all speech that is not loud, threatening, or abusive is permissible, as is all conduct that is not disorderly or disruptive. Accordingly, because individuals “remain[] free to engage in a rich variety of expressive activities, ” Mahoney, 566 F. Supp. 3d at 11 (quotation omitted), § 5104(e)(2)(D) is a facially constitutional time, place, or manner regulation.

2. Content-Based Speech Regulation Similar to Defendant’s arguments as to 18 U.S.C. § 1752, he argues that 40 U.S.C.

§ 5104(e)(2)(D) and (e)(2)(G) are content-based restrictions. The Court refers to its earlier discussion supra Section III.B.4 for a fuller explanation of why this is incorrect. Here, it suffices to point out that a speech regulation is content-based only if it “targets speech based on its communicative content—that is, if it applies to particular speech because of the topic discussed or the idea or message expressed,” but not if it is “agnostic as to content.” City of Austin, 142 S. Ct. at 1471 (cleaned up). Because § 5104(e)(2)(D) and (e)(2)(G), by their plain text, are agnostic as to content, they are not content-based restrictions. The only new argument Defendant makes along these lines is that § 5104(e)(3), which exempts government officials from the statute’s prohibitions, transforms the statute into a content-based restriction. Def.’s 2d MTD at 24–26. But this argument has been squarely rejected by the Supreme Court. In McCullen v. Coakley, 573 U.S. 464, 483 (2014), the Supreme Court held that there was “nothing inherently suspect”

36

about an exemption from a buffer zone restriction around abortion clinics for “clinic employees and agents acting within the scope of their employment.” Id. Similarly, here, there is nothing suspect about § 5104(e)(3), which simply clarifies that the listed government officials are not prohibited from “any act performed in the lawful discharge of official duties.”

3. Failure to State an Offense Finally, Defendant’s brief assertion that the Information fails to state an offense as to Counts 3 and 4 largely just rehashes his vagueness and overbreadth arguments. The Court finds that the Information meets the requirement to provide “a plain, concise, and definite written statement of the essential facts constituting the offense charged,” Fed. R. Crim. P. 7(c), such that it “clearly informs the defendant of the precise offense of which he is accused so that he may prepare his defense,” United States v. Conlon, 628 F.2d 150, 155 (D.C. Cir. 1980).

D. Defendant’s Motion to Suppress Defendant moves to suppress Google Location History data obtained by the Government pursuant to a “geofence” warrant (the “Geofence Warrant”). A geofence warrant authorizes the seizure of location data collected from smartphones of individuals within a particular area over a specified range of time. The Geofence Warrant in question here created a multi-step process authorizing the seizure of Google Location History data for individuals in and immediately around the Capitol building between 2:00 p.m. and 6:30 p.m. on January 6, 2021, subject to certain limitations. Defendant argues that the Geofence Warrant was overbroad and lacked particularity. The Government responds that Defendant does not have a reasonable expectation of privacy over his location that day, or over his Google Location History data during the relevant period more generally. In the alternative, the Government contends that the warrant was

37

not overbroad, was sufficiently particular, and regardless that suppression is inappropriate under the good faith exception to the exclusionary rule.

As the relatively few other courts to consider the validity of geofence warrants have noted, technological advances coupled with corporate data collection practices have rapidly expanded law enforcement surveillance capabilities in ways that present new and consequential Fourth Amendment questions, the answers to which are not neatly directed by existing precedent. See, e.g., United States v. Chatrie, 590 F. Supp. 3d 901, 905 (E.D. Va. 2022). Accordingly, after providing relevant legal and factual background, the Court will review the state of the law on this evolving and important topic before turning to consider the merits of the parties’ arguments. Ultimately, the Court finds that, based on the unique facts at issue here, suppression is not warranted in this particular case.

1. Background: Geofences and Location History Data Unlike a warrant authorizing surveillance of a known suspect, geofencing is a technique law enforcement has increasingly utilized when the crime location is known but the identities of suspects is not. 12 At a basic level, a geofence warrant seeks cell phone location data stored by third-party companies like Google, which offers the Android operating system on which millions of smart phones run and offers other applications commonly used on phones running on other operating systems. See Ex. A to Def.’s Mot. Suppress (“Geofence Warrant & Application”) at 21, ECF No. 45-1. The scope of location data captured by a geofence is limited by geographic

12 See Brian L. Owsely, The Best Offense is Good Defense: Fourth Amendment Implications of Geofence Warrants, 50 HOFSTRA L. REV. 829, 834 (2022) (“The government filed its first geofence search warrant in 2016, and by the end of 2019, Google was receiving about 180 search warrant requests per week from law enforcement officials across the country. This number represented a 1,500% increase between 2017 and 2018 and a 500% increase from 2018 to 2019.” (internal quotation omitted)).

38

and temporal parameters, so geofence warrants identify the physical area and the time range in which there is probable cause to believe that criminal activity occurred. See In re Search of Info. That Is Stored at Premises Controlled by Google (“DC”), 579 F. Supp. 3d 62, 69 (D.D.C. 2021)) (Harvey, Mag. J.).

The type of location data at issue here—Google Location History (“LH”)—comes from “a service that Google account holders may choose to use to keep track of locations they have visited while in possession of their compatible mobile devices.” Ex. D. to Def.’s Mot. Suppress, Chatrie Declaration of Marlo McGriff – Google Location History Product Manager (“Decl. of Marlo McGriff”) ¶ 4, ECF No. 43-2. 13 LH is “considerably more precise than other kinds of location data, including cell-site location information” because LH is determined based on “multiple inputs,” including GPS signals, signals from nearby Wi-Fi networks, Bluetooth beacons, and cell towers. Id. ¶ 12; see Chatrie, 590 F. Supp. 3d at 907 (describing LH as “the most sweeping, granular and comprehensive tool—to a significant degree—when it comes to collecting and storing location data”). Google obtains LH data from users with Google accounts who opt in. 14 Decl. of Marlo McGriff ¶ 4. Specifically, after logging into a Google account, a user must enable “Location Reporting,” 15 at which point LH data is sent to Google “for processing and storage” in Google’s “Sensorvault.” Id. ¶ 9. LH “logs a device’s location, on

13 This declaration was filed in conjunction with consideration of a motion to suppress LH data obtained from a geofence warrant in United States v. Chatrie, 590 F. Supp. 3d 901 (E.D. Va. 2022). The Government does not object to and in fact cites to this declaration, as well as to an amicus brief filed by Google in that case, so the Court finds it appropriate to consider those materials here. See e.g., Gov’t’s Opp’n at 5, 16.

14 Nearly every Android user has an associated Google account, and many Google applications running on other devices also require a Google account to enable full usage. Geofence Warrant & Application at 21.

15 This can be done “either at the ‘Settings’ Level, or when installing applications such as Google Assistant, Google Maps, or Google Photos.” Chatrie, 590 F. Supp. 3d at 908.

39

average, every two minutes,” and tracking occurs “across every app and every device associated with a user’s account,” because LH is tied to the user’s account, not any particular application or device. Chatrie, 590 F. Supp. 3d at 908–09. “Once a user opts into Location History, Google is always collecting data and storing all of that data” in the Sensorvault. Id. at 909 (internal quotation omitted). In order to respond to a geofence warrant specifying a timeframe and location, “Google has to compare all the data in the Sensorvault.” Id. at 908. Users can delete their LH data via their Google accounts. See Decl. of Marlo McGriff ¶ 15.

LH location data points, which are reflected in geographic coordinates, represent Google’s “estimate” of the user’s location. Id. ¶ 24. However, the “user’s actual location does not necessarily align perfectly with any one isolated LH data point.” Id. As such, each location data point comes with an error radius (which Google refers to as a “Map Display Radius”)—for example, 100 meters around the specified coordinates—the size of which varies depending on the quality of the data inputs, such as the strength of the GPS signal. See id. Google LH is designed to be correct that a user actually is within the error radius of where they appear to be approximately 68% of the time. See id. Google considers this to be reliable enough for its purposes to allow users to “store and visualize their location and movements in a journal,” and to allow Google to serve location-based advertisements. Id. ¶ 26.

2. Background: The Geofence Warrant On January 13, 2021, the Government applied for and a magistrate judge approved the Geofence Warrant. See Geofence Warrant & Application at 1. The application sought LH data between 2:00 p.m. and 6:30 p.m. on January 6, 2021 for individuals in a target area slightly larger than but roughly tracing the contours of the Capitol building itself, excluding most of the plazas and lawns on both sides of the building and the abutting streets.

40

Geofence Warrant & Application at 5. The warrant approved a three-step process for obtaining the LH data. See Def.’s Mot. Suppress at 6–11; Gov’t’s Opp’n to Mot. Suppress at 5–8, ECF No. 59. At step one, Google was to provide the Government with three anonymized lists of devices—a primary list and two control lists. The primary list consisted of devices that Google “calculated were or could have been (based on the associated margin of error for the estimated latitude/longitude point) within the TARGET LOCATION.” Geofence Warrant & Application at 6. The two control lists were “similar to the [primary] list” for time ranges of 12:00 p.m. to 12:15 p.m. and 9:00 p.m. to 9:15

41

p.m., respectively. Id. At step two, the Government then was to “review these lists in order to identify information, if any, that [was] not evidence of a crime (for example, information pertaining to devices moving through the Target Location(s) in a manner inconsistent with the facts of the underlying case).” Id. That process was to include the Government comparing the primary list to the control lists and “strik[ing] all devices” from the primary list that appear on either of the control lists. 16 Id. At step three, the Government was to “identify to the Court through a supplemental affidavit the devices appearing on the list produced by Google for which it [sought] the Google account identifiers and basic subscriber information.” Id. If ordered by the court after review of that supplemental affidavit, Google would then be required to “disclose to the government the Google account identifier associated with the devices identified by the government to the Court, along with subscriber information for those accounts.” Id. at 7.

The process that played out largely, though not entirely, adhered to process laid out in the warrant. On January 13, 2021, Google produced the three lists required under step one. See Supp. Affidavit, Ex. B to Def.’s Mot. Suppress at 6, ECF No. 45-2. The primary list, which “was based on Google data as it existed on January 13, 2021,” consisted of 5,653 unique devices. Gov’t’s Opp’n to Mot. Suppress at 6; Supp. Affidavit at 6. The control lists included 176 devices for the 12:00–12:15 p.m. time frame and 159 devices for the 9:00–9:15 p.m. timeframe. See id. Two days later, on January 15, 2021, Google also produced two additional versions of the primary list, one “based on data as it existed in the evening of January 6, 2021” that included

16 The application explained that it would use the control lists to “cull” the primary list of people lawfully in the Capitol building, as “there will probably be no tourists or bystanders to be found in any of this data” from the two fifteen-minute periods, due to the “pandemic, the security surrounding the Capitol in preparation for the Inauguration, the security surrounding the Capitol for the protests over the Certification, and the limited scope of the geographic area covered.” Geofence Warrant & Application at 25.

42

5,716 devices, and one that “was based on Google data as it existed in the morning of January 7, 2021” that included 5,721 devices. Gov’t’s Opp’n to Mot. Suppress at 6; Supp. Affidavit at 6. All of the lists “included a unique, anonymous device identifier that was consistent across” the lists, and “included an estimated latitude and longitude location that Google developed through analysis of a number of points of data that it collected about the device,” together with a margin of error for each location point. Id.

Based on the Government’s analysis, the combined primary lists contained a total of 5,723 unique devices. Id. at 7. After culling the devices from the control lists, that number shrank to 5,518. Id. Out of those 5,518 devices, “1,498 of them ha[d] at least one location associated with the device that [was] within the [Capitol] building and the margin of error [fell] entirely within the Geofence.” Id. The Government filed a supplemental affidavit seeking the account identifiers and basic subscriber information for those 1,498 devices. See Gov’t’s Opp’n to Mot. Suppress at 7; Supp. Affidavit at 7. In addition, 70 devices appeared on either of the two versions of the primary list based on data as of the evening of January 6, 2021 and the morning of January 7, 2021, but did not appear on the version of the primary list based on data as of January 13, 2021. Id. The Government suspects that the account data was deleted from those 70 devices in order to cover up the users’ participation in criminal activity on January 6, 2021. Id. Accordingly, the Government also sought account identifiers and subscriber information for a subset of 37 of those devices that had “at least one record that [was] located within the Geofence but some part of their margin of error [fell] outside of the Geofence.” Id. at 8. Based on the Government’s supplemental affidavit, which included a list of all of the anonymized device identifiers for which it sought deanonymized information, on January 18, 2021 the same magistrate judge who approved the initial warrant approved an order requiring Google to

43

produce account identifiers and basic subscriber information for the 1,498 devices showing a location point in within the Capitol building with a margin of error entirely within the geofence and the 37 “deleted devices” showing a location within the geofence but with some part of the margin of error falling outside of the geofence. Id. at 9.

3. Background: Investigation and Arrest of Defendant The affidavit of probable cause attached in support of an application for a warrant to search Defendant, which was submitted and approved on November 5, 2021, summarizes the Government’s investigation as to Defendant in particular and the geofence data’s place in that investigation. According to the affidavit, the Government received two tips on January 10 and January 12, 2021 that Defendant had been inside the Capitol on January 6. See Rhine Search Warrant Affidavit, Ex. M to Def.’s Mot. Suppress at 12, ECF No. 45-6. The FBI also reviewed surveillance footage from inside the Capitol on January 6. See id. at 15. In a March 2021 interview, one of the tipsters provided a text message exchange with Defendant and his wife in which Defendant stated, “I witnessed ZERO violence. I saw no ‘proud boys.’ Capitol police removed barriers and let people in.” Id. at 14. Also in March 2021, investigators received returns from the Geofence Warrant and from another search warrant for cell-site location information (“CSLI”) associated with Defendant’s Verizon cell phone number. See Ex. 1 to Gov’t’s Opp’n to Def.’s Mot. Suppress, ECF 59-1. The Geofence Warrant returns show that Defendant’s cell phone was present in at least 26 points within the geofence, of which 22 were in the Capitol itself, between 2:24 p.m. and 4:37 p.m. on January 6. See Location Map, Ex. H to Def.’s Mot. Suppress, ECF No. 45-4; see also Location Spreadsheet, Ex. G to Def.’s Mot.

44

Suppress, ECF No. 45-3. 17 The CSLI warrant returns indicated that Defendant’s cell phone “utilized a cell site consistent with providing service to a geographic area that included the interior of the United States Capitol building.” Ex. 1 to Gov’t’s Opp’n to Def.’s Mot. Suppress. After an initial review of surveillance footage conducted on June 23, 2021 failed to identify Defendant, Ex. O to Def.’s Mot. Suppress, ECF No. 45-8, a second review conducted on July 26, 2021 identified Defendant in numerous locations throughout the Capitol, Ex. P to Def.’s Mot. Suppress, ECF No. 45-9. In September 2021, the same tipster that provided the text exchange with Defendant and his wife identified Defendant in a screenshot from the surveillance footage taken inside the Capitol on January 6, though could not identify him in other screenshots. See Rhine Search Warrant Affidavit at 14–15.

Based on this evidence, on November 5, 2021, the Government applied for and a magistrate judge approved a warrant to search Defendant and his cell phone(s). See Ex. M to

17 There is some ambiguity concerning the identified location points. A location map created by the Government states that Defendant’s phone was present at 26 locations within the geofence, of which 22 were within the Capitol building, but it notes that it does not reflect all records and refers to an associated spreadsheet for “complete records.” See Location Map. The associated spreadsheet appears to identify 52 locations within the geofence. See Location Spreadsheet. Further, while the location map simply provides a binary “over 100 feet” or “under 100 feet” margin of error for each of the 26 location points, the spreadsheet, though partially cut off, appears to provide the raw number value of the margin of error for each of the 52 location points listed, although it does not contain a unit, so it is unclear whether these numbers refer to feet, meters, or another unit of measurement. Id. (appearing at the column labeled “Maps Disp”). Given the way the margin of error is described in the parties’ other submissions, the Court assumes the unit of measurement is meters. See, e.g., Def.’s Mot. Suppress at 8; Gov’t’s Opp’n to Mot. Suppress at 5; Ex. C. to Def.’s Mot. Suppress, Chatrie Amicus Brief by Google at 13 n.8, ECF No. 73 (“Each set of coordinates saved to a user’s LH includes a value, measured in meters, that reflects Google’s confidence in the reported coordinates.”). Regardless, it bears notice that the raw numbers go as high as 264 and that the location map indicates that at least one of the 26 location points shown had an error radius that extended beyond the boundary of the geofence. See Location Map; Location Spreadsheet. The location map also appears to indicate that Defendant’s was among the “deleted devices.” See Location Map (stating “Yes” under heading “User Deleted Locations”).

45

Def.’s Mot. Suppress; Ex. N. to Def.’s Mot. Suppress, ECF No. 45-7. On November 9, 2021, the Government executed the warrant and seized Defendant’s cell phone. See Ex. Q to Def.’s Mot. Suppress, ECF No. 45-10. The Government arrested Defendant the same day. See Arrest Warrant, ECF No. 5. In his present motion, Defendant argues that, after suppression of the evidence obtained from the Geofence Warrant as fruits of an unconstitutional search, the November 5 search warrant will lack probable cause, and evidence obtained from that warrant should also be suppressed. See Def.’s Mot. Suppress at 32–35.

4. Legal Framework

The Fourth Amendment guarantees that the “right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the person or things to be seized.” U.S. CONST. amend. IV. The Supreme Court has interpreted the constitutional prohibition against unreasonable searches to require that law enforcement obtain a warrant except in a narrow set of special circumstances. See Vernonia Sch. Dist. 47J v. Acton, 515 U.S. 646, 653 (1995). “When an individual ‘seeks to preserve something as private,’ and his expectation of privacy is ‘one that society is prepared to recognize as reasonable,’ . . . official intrusion into that private sphere generally qualifies as a search and requires a warrant supported by probable cause.” Carpenter v. United States, 138 S. Ct. 2206, 2213 (2018) (quoting Smith v. Maryland, 442 U.S. 735, 740 (1979). To issue in compliance with the Fourth Amendment, a warrant requires three things: (1) that it be issued by a “neutral, disinterested magistrate[];” (2) that it be supported by probable cause that the evidence sought will aid in “a particular apprehension or conviction for a particular offense;” and (3) that it “particularly describe the things to be seized, as well as the

46

place to be searched.” Dalia v. United States, 441 U.S. 238, 255 (1979) (cleaned up). Defendant does not dispute that a neutral magistrate issued the Geofence Warrant here, so the Court focuses on the probable cause and particularity requirements.

Assessing probable cause requires a “practical, common-sense decision whether, given all the circumstances set forth in the affidavit . . . there is a fair probability that contraband or evidence of a crime will be found in a particular place.” Illinois v. Gates, 462 U.S. 213, 238 (1983). “Probable cause is more than bare suspicion but is less than beyond a reasonable doubt and, indeed, is less than a preponderance of the evidence.” United States v. Burnett, 827 F.3d 1108, 1114 (D.C. Cir. 2016). As probable cause is a “fluid concept” that turns on “factual and practical considerations of everyday life on which reasonable and prudent [people], not legal technicians, act,” the “duty of a reviewing court is simply to ensure that the magistrate had a substantial basis for concluding that probable cause existed.” Gates, 462 U.S. at 232, 238–39, 241 (internal quotation omitted); see also United States v. Griffith, 867 F.3d 1265, 1271 (D.C. Cir. 2017) (explaining that “great deference to the judge’s initial determination of probable cause” is required (internal quotation omitted)).

The requirement that a warrant state with particularity the place to be searched and the items to be seized serves the “manifest purpose . . . to prevent general searches.” Maryland v. Garrison, 480 U.S. 79, 84 (1987). Accordingly, a warrant must be “no broader than the probable cause on which it is based.” United States v. Hurwitz, 459 F.3d 463, 473 (4th Cir. 2006) (quoting United States v. Zimmerman, 277 F.3d 426, 432 (3d Cir. 2002). “By limiting the authorization to search to the specific areas and things for which there is probable cause to search, the requirement ensures that the search will be carefully tailored to its justifications, and will not take on the character of the wide-ranging exploratory searches the Framers intended to

47

prohibit.” Garrison, 480 U.S. at 84. “In assessing particularity, courts are concerned with realities of administration of criminal justice,” so it suffices if the warrant “is particular enough if read with reasonable effort by the officer executing the warrant.” United States v. Dale, 991 F.2d 819, 846 (D.C. Cir. 1993) (internal quotations omitted). “In other words, a warrant must be ‘sufficiently specific to permit the rational exercise of judgment [by the executing officers] in selecting what items to seize.” DC, 579 F. Supp. 3d at 76 (quoting United States v. LaChance, 788 F.2d 856, 874 (2d Cir. 1986)). While an “indiscriminate sweep is constitutionally intolerable,” a “broader sweep” may be permissible “when a reasonable investigation cannot produce a more particular description.” Griffith, 867 F.3d at 1275–76 (internal quotations omitted).

Violations of the Fourth Amendment’s guarantees are generally subject to the exclusionary rule, which requires courts to suppress evidence obtained through unconstitutional means. See, e.g., United States v. Weaver, 808 F.3d 26, 33 (D.C. Cir. 2015) (citing Mapp v. Ohio, 367 U.S. 643, 655 (1961); Weeks v. United States, 232 U.S. 383, 398 (1914)). This exclusion of evidence includes both “the primary evidence obtained as a direct result of an illegal search or seizure and . . . evidence later discovered and found to be derivative of an illegality, the so-called fruit of the poisonous tree.” Utah v. Strieff, 579 U.S. 232, 237 (2016) (internal quotations omitted). Typically, “[t]he proponent of a motion to suppress has the burden of establishing that his own Fourth Amendment rights were violated by the challenged search or seizure.” Rakas v. Illinois, 439 U.S. 128, 130 n.1 (1978) (citations omitted). In addition, under what has come to be known as the “good faith exception” to the exclusionary rule, “‘evidence seized in reasonable, good faith reliance on a search warrant,’ need not be excluded, even if the warrant turns out to have been unsupported by probable cause.” Griffith, 867 F.3d at 1278

48

(quoting United States v. Leon, 468 U.S. 897, 905 (1984)); see also Mass. v. Sheppard, 468 U.S. 981 (1984) (“[T]he exclusionary rule was adopted to deter unlawful searches by police, not to punish the errors of magistrates and judges.” (citation omitted)).

5. Relevant Precedent

Having set the stage, the Court now turns to existing precedent concerning the validity of geofence warrants. The collection is limited. The Court has identified just one written opinion by a federal district court and one written opinion by a federal magistrate judge reviewing the validity of a search warrant after issuance. 18 In addition, the Court has found five written opinions by federal magistrate judges considering the issue before issuance. As will be revealed by the Court’s summary of these cases below, important factors on which the approval or rejection of geofence warrants has turned are whether the location and time parameters of the geofence in question were appropriately tailored to the scope of probable cause under the facts of each case, and whether the warrant required additional judicial approval before LH data could be deanonymized.

a. United States v. Chatrie The lone district court case to directly consider the validity of a geofence warrant after issuance is United States v. Chatrie, 590 F. Supp. 3d 901 (E.D. Va. 2022) (Lauck, J.). Chatrie involved an armed bank robbery in a suburb near Richmond, Virginia in which law enforcement

18 In addition, a court in this district recently denied a similar motion to suppress a different January 6 defendant’s LH data obtained from the same Geofence Warrant challenged in this case. See Hearing Transcript at 30, United States v. Cruz, Jr., No. 22-cr-0064 (D.D.C. Jan. 13, 2023) (Walton, J.). Ruling from the bench, the Cruz court expressed skepticism that the defendant had a reasonable expectation of privacy over his LH data, but held that, regardless, the warrant was supported by probable cause and, even if it was not, the good faith exception would apply and suppression would not be appropriate. Id. at 15, 19, 27–29. While much of the Cruz court’s reasoning is applicable here and will be referenced in the analysis section below, in this section the Court focuses on written opinions evaluating the validity of geofence warrants.

49

used a geofence warrant to identify a suspect. Unusually, given the novel and important issues presented by the defendant’s motion to suppress LH evidence obtained from the geofence warrant, Google submitted an amicus brief, in addition to four declarations responsive to subpoenas and testimony during an evidentiary hearing on the motion. Chatrie, 590 F. Supp. 3d at 924–25. The Chatrie warrant also used a multi-step framework, but the steps differed meaningfully from those in the warrant at issue here. First, Google would provide anonymized LH data for a geofenced area within a 150-meter radius of the Bank, which included a nearby church, during the one-hour period around the offense. Id. at 918–19. Next, “[l]aw enforcement would return a list of accounts that they had attempted to narrow down,” at which point Google “would then produce contextual data points with points of travel outside of the geographical area.” Id. at 919 (cleaned up). In addition to expanding the geographical area in this way, at this step the time frame was expanded by thirty-minutes in each direction, for a total window of two hours. Id. Finally, after law enforcement reviewed the additional information provided by Google, it would request and Google would provide identifying information for selected accounts. Id.

The court concluded that the “warrant [was] invalid for lack of particularized probable cause,” but that suppression was inappropriate “because the Leon good faith exception applie[d].” Id. at 925. Because the court denied the motion based on the good faith exception, it declined to decide whether defendant had a reasonable expectation of privacy in data obtained through the geofence warrant in the first place. Id. In doing so, however, the court provided insightful commentary on the ways that “Fourth Amendment doctrine may be materially lagging behind technological innovations.” Id. In particular, the court emphasized how the existence of Google’s vast library of “near exact location information for each user who opts in” provides the

50

government with “an almost unlimited pool from which to seek location data,” such that “police need not even know in advance whether they want to follow a particular individual, or when.” Id. (quoting Leaders of a Beautiful Struggle v. Baltimore Police Dep’t, 2 F.4th 330, 342 (4th Cir. 2021) (en banc)). In addition, the court noted that, under current Fourth Amendment standing doctrine, “individuals other than criminal defendants caught within expansive geofences may have no functional way to assert their own privacy rights.” Id. at 926.

Moving to the question of probable cause, the court found that the warrant was not supported by probable cause as to each person whose data was obtained. The court expressed disbelief at the government’s assertion that “law enforcement established probable cause to obtain all information (Steps 1, 2, and 3) from all users within the geofence without any narrowing measures.” Id. at 929. It explained that “the Geofence Warrant is completely devoid of any suggestion that all—or even a substantial number of—the individuals searched had participated in or witnessed the crime.” Id. The court highlighted the “breadth of this warrant, particularly in light of the narrowness of the Government’s probable cause showing,” emphasizing that the geofence was drawn to encompass “the entirety of [a] Church, and the Church’s parking lot” and that the error radius for one user was as large as 387 meters, a radius that included a hotel, a restaurant, a storage facility, an apartment complex, a senior living facility, and multiple public streets. Id. at 930–31. Still, the court found that the good faith exception to the exclusionary rule applied principally because law enforcement’s reliance on the warrant was reasonable in light of the unclear legality of this novel technology and the fact that the detective “sought advice from counsel before applying for the warrant.” Id. at 937–38 (internal quotation omitted).

51

b. Opinions by Magistrate Judges The Court has identified six opinions by magistrate judges that consider the validity of applications for geofence warrants. The only one of these six to consider the subject after issuance of the warrant does not offer a useful comparison to the present case, as the court found that the defendant had no reasonable expectation of privacy over LH data associated with an account that was not owned by him and therefore declined to “journey into the quagmire of geofence search warrants.” United States v. Davis, No. 21-cr-0101, 2022 WL 3009240, at *8–9 (M.D. Ala. July 1, 2022) (Adams, Mag. J.). The other magistrate judges to consider the question have, to varying degrees, all been drawn into the quagmire. The Court reviews those cases in chronological order to place them in conversation with each other and bring out developments in this new area of law over time.

The first three of these are cases from the summer and fall of 2020 in the Northern District of Illinois. The first, which was decided on July 8, 2020, concerned an investigation into the theft and resale of pharmaceuticals. In re Search of Info. Stored at Premises Controlled by Google, as Further Described in Attch. A (“Pharma I”), No. 20 M 297, 2020 WL 5491763 (N.D. Ill. July 8, 2020) (Weisman, Mag. J.). Law enforcement applied for a geofence warrant to obtain Google LH data at two locations during three forty-five-minute periods on different dates. Id. at *1. The first location was where law enforcement believed “the suspect received the stolen pharmaceuticals from a commercial enterprise located within the designated geofence area.” Id. The geofence area encompassed a “100-meter radius . . . in a densely populated city,” an area that included “restaurants, various commercial establishments, and at least one large residential complex.” Id. The second location encompassed a different 100-meter radius “extending from the commercial establishment where the suspect [allegedly] shipped the pharmaceuticals” and

52

“include[d] medical offices and other single and multi-floor commercial establishments that [were] likely to have multiple patrons” during the relevant time periods. Id. Similar to Chatrie, the warrant envisioned a multistep process by which Google would produce anonymized LH data, the government would “review the list to prioritize the devices about which it wishes to obtain associated information,” and then Google would be required to “disclose to the government the information identifying the Google account(s) for those devices about which the government further inquires.” Id.

The court found that the warrant suffered from “two obvious constitutional infirmities:”

overbreadth and lack of particularization. Id. at *3. With respect to overbreadth, the court highlighted that, despite the fact that the “government’s evidence of probable cause is solely focused on one user of a cellular telephone,” the geofence area “is large, and the majority of the area sought encompasses structures and businesses that would necessarily have cell phone users who are not involved in these offenses.” Id. The court rejected the government’s proffered justification that the geofence area would include “possible co-conspirators” on grounds that “[t]here is no evidence in the application’s supporting affidavit that the suspect is conspiring with anyone to commit these offenses.” Id. at *4. The court also rejected the government’s related suggestion that the geofence area would include witnesses on grounds that the only witnesses “are the employees at the targeted businesses who assisted the suspect in the transactions” and because “the notion that individuals in the area would be witnesses to the offense is not mentioned in the government’s affidavit.” Id. at *5. At bottom, the court agreed that “the date and time are sufficiently prescribed,” but held that “the location clearly is not,” emphasizing again that the “congested urban area” included numerous businesses and residences such that the

53

“vast majority of cellular telephones likely to be identified in this geofence will have nothing whatsoever to do with the offenses under investigation.” Id.

With respect to particularity, the court explained that “the warrant application is completely devoid of any meaningful limitation, seeking only “evidence or instrumentalities” of the listed offenses. Id. at *3. In light of the “urban nature of the encompassed area,” the court lamented the lack of any “objective measure that limits the agents’ discretion” such as a limitation that agents could “only seek[] identifying information as to the ‘five phones located closest to the center point of the geofence’ or some similar objective measure of particularity.” Id. at *6. The court noted that, in an unrelated case, a geofence warrant “for an almost empty commercial parking lot where only one vehicle was located” avoided “any overbreadth issue and addressed the particularity requirement necessary for a valid warrant.” Id. at *6 n.8. The court concluded by explaining that the “government could easily have sought a constitutionally valid search warrant” if it “had constrained the geographic size of the geofence and limited the cellular telephone numbers for which agents could seek additional information to those numbers that appear in all three defined geofences.” Id. at *7.

As part of the same investigation, the government tried again about six weeks later. See In re Search of: Info. Stored at Premises Controlled by Google (“Pharma II”), 481 F. Supp. 3d 730, 733 (N.D. Ill. 2020) (Fuentes, Mag. J.) (explaining the denial of the first renewed application). The renewed application changed the geographic boundaries of the geofence, “shrinking the geofences to . . . square or polygon-shaped boundaries around [the two locations].” Id. at 744. Still, a different magistrate judge from Pharma I found that the “modifications the government made to the geofence boundaries do not solve the constitutional problem because although the modifications may well reduce the number of devices Google

54

identifies as having traversed the geofences, the Court still has no idea how many such devices and their users will be identified under the warrant’s authority.” Id. at 744 (quoting from prior order). Moreover, the court took issue with the fact that devices whose data would be seized “fell not only within the delineated coordinates of the three geofences, but also within a ‘margin of error,’” and that the “government had not attempted to quantify the degree to which this inclusion of an ill-defined ‘margin of error’ geographically expanded the geofences,” especially in light of the “busy urban area” covered by the geofence. Id. at 744–45.

About a month after that, the government tried a third time. In the second renewed application the government retained the same geographic boundaries from the first renewed application, but “altered the proposed search protocol to eliminate the third of the three stages proposed in the first two applications.” Id. at 733. That is, the government’s application no longer sought authorization to compel Google to produce “subscriber information identifying the account holders or users” of devices selected by the government off of the anonymized list. The renewed application also “limit[ed] the ‘anonymized’ information to that which ‘identifies individuals who committed or witnessed the offense,’” though it provided “[n]o further methodology or protocol . . . as to how Google would know which of the sought-after anonymized information identifies suspects or witnesses.” Id.

The same magistrate judge that rejected the first renewed application again rejected the second renewed application. After a lengthy and useful review of relevant Fourth Amendment principles, the court first found that the government had “forfeited” any argument that seeking geofence data did not amount to a “search” for Fourth Amendment purposes, though it noted that there is “much to suggest” that it does. Id. at 736–37. Moving to the question of probable cause, the court reiterated all of the issues it identified with the geographic boundaries in its order

55

rejecting the first renewed application, including the “undefined ‘margin of error,’” which were not remediated in the second renewed application. Id. at 745. Accordingly, the court found that while there was probable cause to believe evidence of the alleged crime would be found in the geofence locations, because the geofence locations “will include . . . location information of persons not involved in the crime,” the warrant was overbroad because “the government has not established probable cause to believe that evidence of a crime will be found in the location history and identifying subscriber information of persons other than the Unknown Subject.” Id. at 751.

With respect to particularity, the court again found the list of items to be seized insufficiently particular because it “does not identify any of the persons whose location information the government will obtain from Google.” Id. at 754. The warrant thus gave the government “unbridled discretion as to what device IDs would be used as the basis for” obtaining identifying information. Id. The Court concluded by noting that some geofence warrants could pass muster under the Fourth Amendment, if the government could “establish independently that only the suspected offender(s) would be found in the geofence, or where probable cause to commit an offense could be found as to all present there.” Id. at 756.

Some two months later, in October 2020, another magistrate judge in the same district but presiding over a different case was presented with similar issues. See In re Search Warrant Application for Geofence Location Data Stored at Google Concerning an Arson Investigation (“Arson”), 497 F. Supp. 3d 345 (N.D. Ill. 2020) (Harjani, Mag. J.). In that case, the government applied for a geofence warrant as part of an investigation into approximately 10 arsons, some of which caused significant destruction, in commercial lots in the Chicago area. Id. at 351. Law enforcement sought Google LH data for six locations. The first location was a triangle about “a

56

quarter to a third of the size of the block” covering a commercial lot where the first arson was suspected to have occurred. Id. An event hall, garage, and trailer used by the company owning the lot were included within the triangle geofence. Id. The geofence timeframe for this location was a “24-minute period starting at 2:00 a.m.” on one day. Id. at 351–52. The second location was an “area of roadway” where the suspects allegedly drove through. Id. at 352. This L-shaped geofence location included part of the alley included in the first location, as well as a “street, alley, and grass or landscaping bordering the street or alley.” Id. The timeframe for this location was 17 minutes within the 24-minute boundary designated for the first location. Id. The third location was another commercial lot where an arson was allegedly committed. Id. About a half of a block in size, this location included a “two story mixed use building, and two garages.” Id. The timeframe for this location was a 15-minute window after the time window for the first location. Id. The fourth location was a roadway near the third location where the suspected arsonists allegedly drove though. Id. Approximately 1.25 blocks long, this location “only consist[ed] of street and sidewalk bordering the street.” Id. The timeframe for this location was a 16-minute period overlapping with the timeframe for location three. Id. Location five had the same geographic boundary as location one, but with a timeframe of 37 minutes starting at 12:00 a.m. Id. at 352–53. And location six had the same geographic boundary as location three, but with a timeframe of 30 minutes immediately prior to (and overlapping by one minute with) location five. Id. at 353.

Similar to the warrants in Chatrie, Pharma I, and Pharma II, the Arson warrant contemplated a multi-step process whereby Google would first provide “anonymized lists of devices with corresponding device IDs, timestamps, location coordinates, margins of error, and data sources for the devices that Google calculates were or could have been (i.e. the margin of

57

error) within each target location during the time periods described.” Id. at 353. Then, “the government, at its discretion, [would] identify to Google the devices from the anonymized lists for which the government seeks the Google account identifier and subscriber information,” which “Google [would] then disclose to the government.” Id.

After warning that “it is easy for a geofence warrant, if cast too broadly, to cross the threshold into unconstitutionality because of the lack of probable cause and particularity,” the court approved the warrant on grounds that, “[i]n this particular case, the government has structured the geofence zones to minimize the potential for capturing the location data for uninvolved individuals and maximize the potential for capturing location data for suspects and witnesses.” 19 Id. With respect to overbreadth, the court first found that the government’s time limitations—15–30 minutes in the middle of the night—were “tailored and specific to the time of the arson incidents only.” Id. at 357. Next, the court found the geographic boundaries to be “narrowly crafted to ensure that location data, with a fair probability, will capture evidence of the crime only,” based on the evidence provided by the government. Id. The court elaborated:

Each of these target locations is drawn to capture location data from locations at or closely associated with the arson. In each of these locations, there is a fair probability that the location data of perpetrators, co-conspirators and witnesses to the incidents will be uncovered. More specifically, because of the visible nature of the crime, namely arson, it is likely that individuals that happen to be in the commercial lot at that hour or on the street would have information about the crime.

Id. at 358. Relatedly, the court distinguished the case from Pharma I and Pharma II on grounds that, unlike those cases, in which the geofences had the “potential to capture vast swaths of location data of individuals not connected to the crime,” the Arson warrant was “constructed to

19 As in Chatrie, Pharma I, and Pharma II, the court did not reach the question of whether the defendant had a reasonable expectation of privacy over LH data. See Arson, 497 F. Supp. 3d at 359–60.

58

focus on the arson site” and “[r]esidences and commercial buildings along the streets have been excluded.” Id. The court also noted that the affidavit of probable cause provided additional evidence from interviews and surveillance video showing that uninvolved individuals were unlikely to be present in the target locations. See id. at 358–59. Finally, the court was undisturbed by the margin of error associated with LH data, based on Google’s acknowledgement in Chatrie that it can often be quite precise depending on the strength of the input signals, the fact that the government did “not intentionally seek information outside the geofence zones,” and the legal reality that “the Fourth Amendment deals in probabilities and reasonableness . . . not exactness and pinpoint accuracy.” Id. at 360–61.

The next written opinion concerning the validity of a geofence warrant that the Court identified was issued in June 2021 by a magistrate judge in the District of Kansas. See In re Search of Info. That Is Stored at the Premises Controlled by Google (“Kansas”), 542 F. Supp. 3d 1153 (D. Kan. 2021) (Mitchell, Mag. J.). The court did not provide detailed factual background to protect the ongoing investigation, but noted that the warrant application sought “geofence data from an area surrounding the alleged crime location, which is a sizeable business establishment, during a one-hour period on the relevant date.” Id. at 1155. The subject building also “contain[ed] another business” and the geofence area “encompasse[d] two public streets,” with “residences and other business” just outside the geofence area and potentially within the margin of error. Id. at 1158. On this basis, the court found that the “geofence boundary appears to potentially include the data for cell phone users having nothing to do with the alleged criminal activity.” Id. Moreover, it found the “nexus between the alleged criminal activity and [the] one- hour duration [to be] weak,” as video surveillance showed the suspect at “three discrete times,” while the “geofence’s temporal scope ranges from just before the second sighting to

59

approximately 10 minutes after the suspect fled the scene.” Id. The court noted that “t]here could be a reasonable explanation for this” but that such “explanation [was] not included in the affidavit.” 20 Id.

Finally, a magistrate judge in this district considered a geofence warrant application in December 2021. In re Search of Info. that is Stored at the Premises Controlled by Google (“DC”), 579 F. Supp. 3d 62 (D.D.C. 2021) (Harvey, Mag. J.). 21 Again, the court did not describe the alleged offense to protect the ongoing investigation, but it did provide detail about the warrant application. Specifically, the government requested a geofence covering a building “in an industrial area” which “share[d] a building with another business.” Id. at 72. However, the triangular geofence area covered only “a portion of the front-half of the [building], plus its parking lot,” such that “[n]o other structures [were] included in the geofence area” including “the part of the building . . . share[d] with the other business.” Id. The court estimated the size of the geofence area at 875 square meters, or approximately 30–35 meters on the short sides of the triangle, and 45–50 meters on the long side. Id. This was not the first warrant application presented to the court. A prior version contained a target area in the shape of a circle with a “radius of approximately 35 meters” which “appeared to capture part of the road abutting the building” and “included part of a building behind” the subject building “as well as the business that shares the [subject] building.” Id. at 72 n.12. However, “[f]ollowing discussions with the Court, the government further limited the scope of the geofence to exclude these areas in which it had no evidence that criminal activity occurred.” Id. The geofence timeframe was a total of

20 The court also found that there was not probable cause to believe that the perpetrators even had smartphones on them during commission of the offense. See Kansas, 542 F. Supp. 3d at 1156–57. As explained below, that is not in issue here, as there is ample evidence that the individuals at the Capitol on January 6 possessed and were using smartphones.

21 Magistrate Judge Harvey also approved the Geofence Warrant in this case.

60

185 minutes, split into segments ranging from 2 to 27 minutes on 8 specified days over approximately five-and-a-half months, corresponding to the alleged criminal activity. Id. at 72.

In the initial warrant, the government proposed the familiar steps included in the warrant applications under review in the other cases: Google would produce an anonymized list, the government would identify, at its discretion, a subset for which it wanted identifying information, which Google would then provide. However, “[t]he court had concerns about this protocol, namely the fact that the government could, ‘at its discretion,’ order Google to disclose the identifying information for certain accounts without any guardrails on the exercise of that discretion or further review by the Court.” Id. at 73. Accordingly, “[a]fter discussions with the government regarding the issue, it submitted a revised warrant application” in which, after selecting the subset of devices for which it sought identifying information, the government then had to identify those devices “in additional legal process to the Court.” Id. At that point, the court had discretion to “order Google to disclose” that information. Id. at 74. In the court’s view, this revised process vested “discretion as to what devices falling within the geofence to deanonymize” with the court, not the government. Id.

Turning to the merits, the court found that the warrant was supported by probable cause.

Specifically, the court found that there was probable cause to believe that the suspects were within the geofence during the designated time windows and that the suspects were actually using cell phones during those time windows, based on evidence provided by the government, including surveillance footage. Id. at 77. Regarding particularity, the court found that the government had “appropriately contoured the temporal and geographic windows in which it [was] seeking location data.” Id. at 80. While the court acknowledged that the 185 minutes sought in the application before it was more than the 139 minutes approved in Arson, it found

61

that “the time windows requested by the government [were] closely keyed to the periods during which the suspects were inside the building.” Id. at 81. The court also distinguished Kansas on grounds that, while the court rejected the one-hour time frame requested in that case, that was because the government failed to “tailor[] the warrant to request geofence data for only the approximate times at which the suspect appeared in the [surveillance] footage.” Id. With respect to location, the court, citing Arson, similarly found that the geofence area “encompass[ed] only the location of the suspects . . . and an area closely associated with the location of the suspects.” Id. at 82 (internal quotation omitted).

Accordingly, the court held that the warrant was not overbroad because “the duration and location of the requested geofence closely track[ed] the probable cause presented in the government’s warrant application.” Id. The court acknowledged that the geofence, “when considering its margin of error, will capture the location information for other customers inside the [subject building] or motorists merely driving by the [subject building] on the abutting road or an employee in the adjoining business during the requested time.” Id. at 85. But the court held that this did not make the warrant constitutionally infirm because “constitutionally permissible searches may infringe on the privacy interests of third persons” and “in this case it appears physically impossible for the government to have constructed its geofence to exclude everyone but the suspects.” Id. at 82, 85 (citing cases approving searches that swept in third- party text messages, emails, and business records, among other contexts). Besides, the court reasoned, the “request for location information here does not have the potential of sweeping up the location data of a substantial number of uninvolved persons,” unlike in Pharma I, Pharma II, and Kansas. Id. at 85 (emphasis added). The court continued that, unlike in Pharma I and Pharma II, “the geofence drawn here is located in an industrial area, not a congested urban area,

62

and no residences can be seen within the geofence.” Id. (internal quotation omitted). And the abutting road “is a secondary road” not a “major arterial street” like in those cases. Id. at 86.

The court also emphasized that “any overbreadth concerns raised by the requested geofence are further addressed by the warrant’s two-step search procedure, which ensures identifying information associated with devices found within the geofence will be produced only pursuant to a further directive from the Court.” Id. at 87 (distinguishing this procedure from the procedures at issue in Pharma I and Pharma II, which would have vested discretion to obtain identifying information entirely with the government). In this way, “the ultimate decision as to which subscribers, if any, Google will be compelled to identify lies with the Court.” Id. at 88. Accordingly, the court granted the warrant as based on its finding of particularized probable cause. Id. at 90–91.

6. Analysis

The Court turns now to the merits of Defendant’s motion to suppress. The Court finds that the Geofence Warrant was supported by particularized probable cause, and regardless that its alleged infirmities would fall into the good faith exception to the exclusionary rule, so suppression is unwarranted in this case.

a. Reasonable Expectation of Privacy The Government first argues that Defendant’s motion fails at the threshold because Defendant had no reasonable expectation of privacy over his location within the Capitol building or over his LH data, so no Fourth Amendment search occurred. Gov’t’s Opp’n to Mot. Suppress at 11–25. Because the Court denies Defendant’s motion on other grounds, it follows the approach of the courts in Chatrie, Pharma I, Pharma II, and Arson in declining to reach the issue of Fourth Amendment standing. See United States v. Sheffield, 832 F.3d 296, 304–05 (D.C. Cir.

63

2016) (explaining that Fourth Amendment standing is “non-jurisdictional” and “merely an aspect of the substantive merits of a Fourth Amendment claim”). Still, the Court feels a brief and non- exhaustive review of recent Supreme Court opinions addressed to this question is relevant to the extent that it reveals principles helpful in guiding application of Fourth Amendment doctrine to law enforcement’s increasing use of new technologies.

In Riley v. California, 573 U.S. 373 (2014), the Supreme Court held the search-incident-

to arrest requirement generally inapplicable to cell phones. The Court noted how cell phone location data “can reconstruct someone’s specific movements down to the minute, not only around town but also within a particular building,” and further how “[c]ell phone users often may not know whether particular information is stored on the device or in the cloud.” Riley v. California, 573 U.S. 373, 396–97 (2014). In doing so, it emphasized the central role of the warrant in safeguarding the “privacies of life” contained on modern cell phones, concluding that the “answer to the question of what police must do before searching a cell phone seized incident to an arrest is accordingly simple—get a warrant.” Id. at 403 (citation omitted). In United States v. Jones, 565 U.S. 400 (2012), the Court held that attaching a GPS tracker to a vehicle constituted a Fourth Amendment search generally subject to the warrant requirement. In Carpenter v. United States, 138 S. Ct. 2206 (2018), the Court interpreted Jones to stand for the rule that “individuals have a reasonable expectation of privacy in the whole of their movements.” Id. at 2217. It explained that “[a] person does not surrender all Fourth Amendment protection by venturing into the public sphere.” Id. The Carpenter Court also identified two “basic guideposts” to steer application of the “Fourth Amendment to innovations in surveillance tools:” first, that the Fourth Amendment’s purpose is to “secure the privacies of life against arbitrary power;” and second, that the Framers intended to “place obstacles in the way of a too permeating

64

police surveillance.” Id. at 2214 (cleaned up). On that foundation, the Court held that individuals have a reasonable expectation of privacy over CSLI data, which it noted can “pinpoint a phone’s location within 50 meters.” Id. at 2219. The Court explained that surveillance technologies that provide an “all-encompassing record” of a person’s whereabouts, including the ability to “reconstruct a person’s movements” retrospectively, represent a “seismic shift[]” in surveillance capability that requires something more than “straightforward” application of existing Fourth Amendment doctrine. Id. at 2214–19 (explaining that “digital data” in the form of “personal location information maintained by a third party . . . does not fit neatly under existing precedents”).

While the Court does not decide the question of whether Defendant had a reasonable expectation of privacy over his LH data, it bears in mind the principles reflected in the Supreme Court’s recent opinions as it turns to evaluate the sufficiency of the Geofence Warrant.

b. Overbreadth

Defendant does not dispute that there was probable cause to believe that the geofence area would contain evidence of a crime, but rather argues that the Geofence Warrant was overbroad; that is, that the warrant’s authorization exceeded the scope of probable cause on which it issued. Def.’s Mot. Suppress at 23–26.

Specifically, Defendant first argues that step one, in which Google provided the Government with an anonymized list of devices falling within the geofence’s geographic and temporal parameters, was overbroad because it required Google to query its entire Sensorvault without probable cause “to search untold millions of unknown accounts in a massive fishing expedition.” Id. at 24. But, as the Government points out, the relevant question is not how Google runs searches on its data, but what the warrant authorizes the Government to search and

65

seize. Gov’t’s Opp’n to Def.’s Mot. Suppress at 30. Under Defendant’s theory, no doubt many search warrants and most third-party subpoenas for protected records would be unconstitutionally overbroad because they necessarily would require the third party to search some group of records larger than those specifically requested, whether they reside in a file cabinet or on a server. See Carpenter, 138 S. Ct. at 2221–22 (explaining that the Fourth Amendment applies to third party subpoenas over which an individual has a reasonable expectation of privacy); cf. United States v. Weaver, 808 F.3d 26, 38 (D.C. Cir. 2015) (distinguishing the “more conditional and more circumscribed” authority to enter a home pursuant to an arrest warrant from a search warrant which authorizes law enforcement to “enter a home and search for the items described in the warrant anywhere in the home where those items might be located” (emphasis added)). In addition, it is far from clear that Defendant’s Fourth Amendment rights were implicated by the anonymized list provided at step one. See Brennan v. Dickson, 45 F.4th 48, 64 (D.C. Cir. 2022) (explaining that anonymized location tracking of a drone does not violate the Fourth Amendment in part because the drone’s “unique identifier—the drone’s serial number—does not disclose who is flying the drone”); DC, 579 F. Supp. 3d at 86 n.26 (quoting Sanchez v. Los Angeles Dep’t of Transp., No. CV 20-5044, 2021 WL 1220690, at *3 (C.D. Cal. Feb. 23, 2021) for the proposition that “a person does not have a reasonable expectation of privacy over information that cannot even be connected to her” and listing additional cases); Chatrie, 590 F. Supp. 3d at 933 (explaining the “crucial[]” distinction between DC and the case before the court that the warrant in DC required the “court . . . at its discretion,

66

[to] order Google to disclose to the Government personally identifying information for devices that belonged to likely suspects” (emphasis in original)). 22 Defendant’s challenges to step two are unpersuasive for similar reasons. Defendant first argues that Google should not have disclosed the two additional versions of the primary list from step one on January 15, 2021. See Def.’s Mot. Suppress at 25. These were the lists based on

22 Defendant claims that “[a]lthough Google initially ‘anonymized’ this data, the FBI could have obtained the subscriber information at any time using a subpoena.” Def.’s Mot. Suppress at 16. Defendant cites Pharma II, in which the Court explained that it saw “no practical difference between a warrant that harnesses the technology of the geofence, easily and cheaply, to generate a list of device IDs that the government may easily use to learn the subscriber identities, and a warrant granting the government unbridled discretion to compel Google to disclose some or all of those identities,” and consequently refused to permit the government to “accomplish indirectly what it may not do directly.” Pharma II, 481 F. Supp. 3d at 749 & n.13. In the different factual and procedural setting of this case, the Court has a different perspective. Lawful investigative tactics do not suddenly become unconstitutional simply because they put the government in a position to serve a targeted subpoena for records. It is the seizure pursuant to a subpoena of records subject to a reasonable expectation of privacy without particularized probable cause that would violate the Fourth Amendment. See Carpenter, 138 S. Ct. at 2221–22 (explaining that the Fourth Amendment applies to third party subpoenas over which an individual has a reasonable expectation of privacy). Defendant has made no allegation that his or others’ identity was knowable based on the anonymized list produced at step one, and considering the geographic and temporal limitations on the geofence area, it likely would not be possible to deanonymize the list indirectly by cross-referencing more revealing location points—for example, the location where the device spent the night. Accordingly, on the facts of this case, the Court has no basis on which to find that Defendant’s Fourth Amendment rights were implicated at step one.

That said, the Court acknowledges that the scope of legally obtainable anonymous data made possible by geofencing technology could present potentially significant risks to privacy, even if those privacy interests cannot be expressed through Defendant’s challenge to step one of this particular warrant, on these particular facts, under current law. See Jennifer Valentino- DeVries, Natasha Singer, Michael H. Keller & Aaron Krolik, Your Apps Know Where You Were Last Night, and They’re Not Keeping It Secret, N.Y. Times (Dec. 10, 2018), https://www.nytimes.com/interactive/2018/12/10/business/location-data-privacy-apps.html (explaining various ways that anonymous data can be used to establish identity). The Court aligns itself with the Chatrie court’s impression that “[i]t is not within this Court’s purview to decide” broad questions raised by geofencing technology on the facts of this case, but that these questions’ increasing importance “urges legislative action.” Chatrie, 590 F. Supp. 3d at 926.

67

Google’s data as it existed on the evening of January 6 and the morning of January 7. Defendant also claims that Google violated its own policies with respect to preserving data from the “deleted devices.” Id. These are quarrels with Google, and Defendant makes no allegation that the Government requested or compelled these actions even if they were in excess of the warrant’s authorization. See Burdeau v. McDowell, 256 U.S. 465, 475 (1921) (explaining, where the defendant’s papers were stolen by a third party who turned them over to the government, that the Fourth Amendment “was not intended to be a limitation upon other than governmental agencies” so because “no official from the federal government had anything to do with the wrongful seizure . . . or any knowledge thereof until several months after the property had been taken . . . there was no invasion of the security afforded by the Fourth Amendment”). Besides, the two additional primary lists consisted only of anonymous step one data falling under the warrant’s authorization.

Defendant next takes issue with the control lists, which contained anonymized device information for two fifteen-minute periods at 12:00 p.m. and 9:00 p.m. on January 6, because these windows of time fall outside the “geofence time limit.” Def.’s Mot. Suppress at 25. But again, these lists contained only anonymized device identifiers. 23 And while it is true that these fifteen-minute periods fall outside of the step one timeframe, this proves the opposite of what

23 Defendant maintains that “discovering the likely identities or affiliations of the device IDs was the precise purpose for the control window searches.” Def.’s Mot. Suppress at 25. As Defendant’s own examples show, this is only true if one broadens the definition of identity to mean simply being one among all those lawfully present at the Capitol on January 6, a group too large to permit precise inferences about actual individual identity. See, e.g., INSPECTOR GENERAL, U.S. DEPARTMENT OF DEFENSE, REVIEW OF THE DOD’S ROLE, RESPONSIBILITIES, AND ACTIONS TO PREPARE FOR AND RESPOND TO THE PROTEST AND ITS AFTERMATH AT THE U.S. CAPITOL CAMPUS ON JANUARY 6, 2021 (2021), at 46 (explaining that, as of 6:00 p.m. “approximately one company of [D.C. National Guard] personnel arrived at the Capitol and integrated with Federal law enforcement”).

68

Defendant suggests: the purpose of using control lists from outside the step one timeframe was to narrow the universe of devices to ensure that the supplemental affidavit seeking deanonymization established particularized probable cause. The absence of similar narrowing mechanisms was a significant factor motivating the rejection of the geofence warrants in Chatrie, Pharma I, Pharma II, and Kansas. See Chatrie, 590 F. Supp. at 933 (explaining, after walking through the narrowing procedures employed in DC, which were less stringent than those at issue here, that “[a]lthough the instant warrant is invalid, where law enforcement establishes such narrow, particularized probable cause through a series of steps with a court’s authorization in between, a geofence warrant may be constitutional”); Pharma I, 2020 WL 5491763, at *7 (explaining that the “government could easily have sought a constitutionally valid search warrant,” for example by employing measures to “limit[] the cellular telephone numbers for which agents could seek additional information”); Pharma II, 481 F. Supp. 3d at 756 (rejecting the second renewed geofence warrant application for failure to take the limiting steps described in Pharma I); Kansas, 542 F. Supp. 3d at 1157 (“The application also does not address the anticipated number of individuals likely to be encompassed within the targeted Google location data.”).

Defendant’s overbreadth claim as to step three, in which the court authorized the Government to obtain deanonymized account information for the narrowed list from Google, presents a closer question. 24 At the outset, because a warrant’s authorization may be “no broader than the probable cause on which it is based,” Hurwitz, 459 F.3d at 473 (citation omitted), it is necessary to define the scope of that probable cause. January 6 was a unique event in a

24

The Court finds several of the arguments presented in the section of Defendant’s brief addressed to particularity also relevant, or in some cases more relevant, on the related issue of overbreadth, so it considers them in this section.

69

geographically unusual place such that the scope of probable cause was uncommonly large. Because the Capitol building was not open to the public on January 6 due to the counting of the votes of the Electoral College, the fact of having entered the building during the geofence timeframe itself constitutes evidence of a crime. See, e.g., 18 U.S.C. § 1752(a)(1); see also Transcript of Hearing at 9–10, United States v. Cruz, Jr., No. 22-cr-0064 (D.D.C. Jan. 13, 2023) (explaining, in the process of denying a similar motion to suppress, that “the Capitol was closed on [January 6]” so “anybody who was there who was not authorized to be there was in fact committing a crime, at least based upon a probable cause assessment”). 25 Based on an unusual abundance of surveillance footage, news footage, and photographs and videos taken by the suspects themselves while inside the Capitol building, there is much more than a “fair probability” that the suspects were within the geofence area and were carrying and using smartphones while there, such that their devices’ LH would provide evidence of a crime. See Geofence Warrant & Application at 18; see also United States v. James, 3 F.4th 1102, 1105 (8th Cir. 2021) (“Even if nobody knew for sure whether the robber actually possessed a cell phone, the judges were not required to check their common sense at the door and ignore the fact that most people ‘compulsively carry cell phones with them all the time.’” (quoting Carpenter, 138 S. Ct. at 2218)); DC, 579 F. Supp. 3d at 79 (explaining that Android operates on 74 percent of the world’s smartphones, that “nearly every device using [Android] has an associated Google account,” and that users of phones with other operating systems often have Google accounts, such as Gmail accounts, too (citations omitted)). In addition, as those photos and videos and the volume of individuals charged and convicted in connection with January 6 to date show, the

25 The Government provided a copy of the Cruz hearing transcript to the Court and to Defendant. See Notice of Supplemental Authority at 2 n.1, ECF No. 72.

70

number of suspects is extremely large. See Press Release, U.S. Attorney’s Office for the District of Columbia, 23 Months Since the January 6 Attack on the Capitol (Dec. 8, 2022) (listing over 500 guilty verdicts or pleas and hundreds more pending charges for January 6 defendants).

Having established the unusually broad scope of probable cause that supports the Geofence Warrant based on the unique facts of this case, the Court turns to Defendant’s claim that the warrant’s authorization under step three is nonetheless overbroad. Defendant’s principal argument is that the steps taken to narrow the primary list at step two were insufficient, such that there was “no meaningful showing of probable cause in [the Government’s] follow up warrant affidavit.” Def.’s Mot. Suppress at 26, 30. In support of this argument, Defendant points to the size of the geofence area, particularly in light of the relevant error radius for each given location point. Id. at 29–30.

With respect to the narrowing process at step two, given the broad scope of probable cause, the Court finds the use of control lists to narrow the step three universe to be a reasonable approach that reflected the relevant “factual and practical considerations” under the circumstances—namely, the large volume of suspects and the unusually well-documented timeline of events indicating when they, as opposed to uninvolved bystanders, would have been present within the Geofence area. Gates, 462 U.S. at 241; see, e.g., Ryan Goodman & Justin Hendrix, January 6 Clearinghouse, Just Security (Dec. 22, 2022, https://www.justsecurity.org/77022/january-6-clearinghouse/) (showing, under the “Timelines” drop-down tab, numerous detailed timelines of events on January 6, including multiple from official government sources). Moreover, the Court’s step three deanonymization order was based on further averment by the Government that the 1,498 devices from the primary list for which it sought subscriber information “ha[d] at least one location associated with the device that

71

[was] within the [Capitol] building and the margin of error [fell] entirely within the Geofence.” Supp. Affidavit at 7. This substantially mitigates, albeit does not altogether eliminate, the risk Defendant emphasizes that a device could show a “false positive” location in the Capitol building when in fact it was elsewhere. Similarly, with respect to the 37 “deleted devices,” in addition to the evidence of criminality apparent from the fact that the LH data was deleted, 26 the Government’s supplemental affidavit stated that it sought subscriber information only from those devices for which “at least one record that [was] located within the Geofence but some part of their margin of error [fell] outside of the Geofence.” Id. at 8. Together, these measures substantially reduced the number of devices for which the Government sought deanonymized information from 5,723 down to 1,535—a 73 percent drop.

Similarly effective narrowing measures were not taken in any of the geofence cases discussed above, all of which involved significantly narrower probable cause. As explained above, the absence of such measures was critical in Chartrie, Pharma I, Pharma II, and Kansas, but even the two magistrate judges to approve the geofence warrants insisted on less stringent procedures. See Arson, 497 F. Supp. 3d at 362 (finding that the “government has established probable cause to seize all location and subscriber data within the geofence locations identified” with no required limiting procedures); DC, 579 F. Supp. 3d at 73 (requiring only that the government “review [the] list to identify devices, if any, that it can determine are not likely to be relevant to the investigation”). The Court sees no cause, based on the step two narrowing

26 Defendant cites news reports to suggest that there may have been alternative reasons why people would delete their LH data, such as privacy concerns. Def.’s Mot. Suppress at 10. While it is a possibility that a participant in the events of January 6 would decide to delete his or her LH data shortly thereafter due to privacy or other concerns disconnected from the events of that day, the magistrate judge certainly was not bound to favor that possibility to the exclusion of the possibility that the participant instead deleted the data to conceal evidence of criminal activity.

72

procedures, to question that there was a “substantial basis” for the magistrate judge’s decision to order deanonymization of the devices listed in the supplemental affidavit. Gates, 462 U.S. at 239.

Moving to Defendant’s arguments about the geographic area covered by the geofence, at the outset, the Court reiterates that the geofence area closely, although not perfectly, contours the Capitol building itself, and does not include the vast majority of the plazas or grounds surrounding the building. More importantly, two main factors convince the Court that the geofence area is not overbroad. First, recall that the error radius only extends outside the boundary of the geofence for 37 of the 1,535 devices for which the Government sought subscriber information—the “deleted devices.” The other 1,498 devices “ha[d] at least one location associated with the device that [was] within the [Capitol] building and the margin of error [fell] entirely within the Geofence.” Supp. Affidavit at 7. Recognizing that there is still a roughly 32% chance that any given data point is inaccurate, error radius notwithstanding, Supp. Affidavit at 7, there is still a “substantial basis” for the magistrate judge to have identified a “fair probability” that all of these 1,498 devices were linked to suspects or witnesses, Gates, 462 U.S. at 238–39; Burnett, 827 F.3d at 1114 (“Probable cause is . . . less than a preponderance of the evidence.”).

Second, as relevant to the 37 deleted devices, the area around the Capitol is unusual for its lack of nearby commercial businesses or residences. Indeed, while Defendant does not make any specific allegations about any such nearby buildings, the Court’s best estimate is that the nearest is no less than about a quarter of a mile away, or approximately 400 meters. 27 By Defendant’s own admission, the error radius is not known to exceed 387 meters, Def.’s Mot.

27 The Court, with no intended irony, used Google Maps to make this estimate.

73

Suppress at 8, and the error radius for Defendant’s location points in particular extends only as high as 264 meters. 28 See Location Spreadsheet. Furthermore, while public streets do appear to be somewhat closer to the geofence area, extensive road closures west of the Capitol, in anticipation of the rally on the ellipse on January 6, including on Pennsylvania Avenue, reduce the likelihood that any stray cars would have been picked up in the geofence error radius, rendering them more like the “secondary road[s]” in DC than the “major arterial street[s]” in Pharma I and Pharma II. See Jack Moore, What DC Streets Are Closed for Pro-Trump Rallies and Demonstrations?, Wtop News (Jan. 6, 2021), https://wtop.com/dc/2021/01/downtown-dc- street-closures-planned-for-jan-6-pro-trump-rally/; DC, 579 F. Supp. 3d at 86; Pharma II, 481 F. Supp. 3d at 743. 29 Taken together, the geographic parameters in this case are much closer to those approved in Arson than those rejected in Chatrie, Pharma I, Pharma II, and Kansas. Compare, e.g., Arson, 497 F. Supp. 3d at 357–58 (approving geofence locations mostly in empty commercial lots and streets leading to them as “narrowly crafted” and noting that “[r]esidences and commercial buildings along the streets have been excluded from the geofence zone”) with Chatrie, 590 F. Supp. 3d at 930 (lamenting that “law enforcement simply drew a circle with a 150-meter radius that encompassed . . . the entirety of [a] Church, and the Church’s parking lot”); Pharma I, 2020 WL 5491763, at *1 (“The geofence . . . is in a densely populated city, and the area contains restaurants, various commercial establishments, and at least one large

28 As explained supra note 17, while the unit of measurement is not clear from the location spreadsheet, the Court assumes that the appropriate unit is meters based on the parties’ other submissions. See, e.g., Chatrie Amicus Brief by Google at 13 n.8 (“Each set of coordinates saved to a user’s LH includes a value, measured in meters, that reflects Google’s confidence in the reported coordinates.”).

29 In addition, as of 2:31 p.m. D.C. Mayor Muriel Bowser issued a city-wide curfew, which went into effect at 6 p.m.—an announcement that likely discouraged others from approaching the Capitol. Mayor Muriel Bowser (@MayorBowser), Twitter (Jan. 6, 2021, 2:31 p.m.), https://twitter.com/mayorbowser/status/1346902298044325893?lang=en.

74

residential complex . . . .”); Kansas, 542 F. Supp. 3d at 1158 (taking issue with the fact that the geofence “boundary encompasses two public streets” and “another business”, and that the “area just outside of the perimeter of the geofence includes residences and other businesses that could be implicated by the margin of error”).

With respect to the timeframe, Defendant repeatedly references the “four-and-a-half-hour period” for which the Geofence Warrant authorized seizure of LH data, but does not directly argue that the time period is overbroad. See Def.’s Mot. Suppress at 6, 7, 10, 26, 29. The Court thus has no occasion to second-guess the magistrate judge’s determination that this period was at most co-extensive with the scope of probable cause, a determination that the Court notes is corroborated by the January 6 timelines referenced above. See, e.g. Press Release, Department of Defense, Planning and Execution Timeline for the National Guard’s Involvement in the January 6, 2021 Violent Attack at the U.S. Capitol (Jan. 8, 2021), https://www.defense.gov/News/Releases/Release/Article/2467051/planning-and-execution- timeline-for-the-national-guards-involvement-in-the-janu/ (showing that the Commanding General of the D.C. National Guard received a “request for immediate assistance” from the Chief of the U.S. Capitol Police by 1:49 p.m. and that the Capitol building was not declared secure until 8:00 p.m.). While this period is longer than previous geofence timeframes, this is simply because more criminal activity occurred over a longer period of time than in those cases, and therefore this fact does not undermine the reasonableness of the time parameter used here. See DC, 579 F. Supp. 3d at 81 (finding that “[a]lthough a total of 185 minutes of geofence data is more than the [Arson] court sanctioned, the government’s request in this case [was] reasonable” because the time windows were “closely keyed to the periods during which the” criminal activity occurred) ; see Hearing Transcript at 27, United States v. Cruz, Jr., No. 22-cr-0064 (D.D.C. Jan.

75

13, 2023) (“Here we’re talking about a distinct location where a crime was being committed at a particular time and they sought information to find out who was there at a particular time. I think that’s clearly reasonable.”).

In sum, the Court finds that the Geofence Warrant’s authorization was no greater than the scope of probable cause on which it issued, and therefore that it was not overbroad.

c. Particularity

With respect to particularity, Defendant’s main argument is that the Geofence Warrant vested too much discretion in the Government. Surprisingly, Defendant cites Pharma I and Pharma II for the proposition that “Courts have repeatedly held that the Court must be more involved in narrowing at steps 2 and 3.” Def.’s Mot. Suppress at 30–31. But more involved than what? Pharma I and Pharma II involved geofence warrants that contemplated no role for the Court beyond the issuance of the initial warrant. See, e.g., Pharma I, 2020 WL 5491763, at *1 (“The warrant application includes no criteria or limitations as to which cellular telephones the government agents can seek additional information.”). By contrast, here, the terms of the initial warrant precluded disclosure of deanonymized device information except after separate order of the court based on a supplemental affidavit. This same approach was approved in DC, and the Chatrie court also suggested this approach was constitutionally permissible because, “crucially, the [DC] warrant left ultimate discretion as to which users’ information to disclose to the reviewing court, not to Google or law enforcement.” Chatrie, 590 F. Supp. 3d at 933 (“Although the instant warrant is invalid, where law enforcement establishes such narrow, particularized probable cause through a series of steps with a court's authorization in between, a geofence warrant may be constitutional.”). The Court accordingly finds that the approach taken here did not vest too much authority in the Government. See Dalia v. United States, 441 U.S. 238, 257

76

(1979) (“Nothing in the language of the Constitution or in this Court’s decisions interpreting that language suggests that . . . search warrants also must include a specification of the precise manner in which they are to be executed.”); United States v. Riley, 906 F.2d 841, 844–45 (2d Cir. 1990) (finding that the “particularity requirement is not so exacting” as to “eliminate all discretion of the officers executing the warrant”).

Defendant also takes issue with the list of items to be seized attached to the Geofence Warrant on grounds that it includes the language, “Information that constitutes evidence concerning persons who either (i) collaborated, conspired, or assisted (knowingly or unknowingly) . . . .” Geofence Warrant & Application at 8 (emphasis added). Defendant argues that this permits officers to seize information that is “not evidence of a crime.” Def.’s Mot. Suppress at 31 (alterations omitted). This is a dubious assertion, as surely even unwitting accomplices can provide witness testimony. More importantly, as the Government points out, Defendant misreads the structure of the Warrant. This language appears in the second of twelve subparagraphs in Section II setting out categories of items to be seized. However, the entirety of Section II is limited to “information described in Section I that constitutes evidence of” listed offenses. Geofence Warrant & Application at 8; Gov’t’s Opp’n to Def.’s Mot. Suppress at 36– 37. Section I authorizes search only of LH data and account information for devices with responsive data. Geofence Warrant & Application at 4. In this way, contrary to Defendant’s claim that the “breadth and vagaries of the items to be seized was an invitation to do a general search,” the items to be seized are cabined by (1) Section I; (2) the offenses listed in the umbrella paragraph to Section II; and (3) the description in the twelve subparagraphs in Section II. The Court is satisfied that the terms of the Geofence Warrant did not permit “unbridled rummaging” by the executing officers. See In re Search Warrant Dated July 4, 1977, for Premises at 2125 S

77

St. Northwest Washington, D.C., 572 F.2d 321 (D.C. Cir. 1977) (Robinson III, J., concurring in declining to request rehearing en banc); see also United States v. Vaughn, 830 F.2d 1185, 1186 (D.C. Cir. 1987) (“When judging questions of particularity, we are concerned with realities of administration of criminal justice. It is sufficient if the warrant signed by the judicial officer is particular enough if read with reasonable effort by the officer executing the warrant.” (cleaned up)).

d. Good Faith Exception

Finally, though it need not dwell on the topic, having found the Geofence Warrant to be constitutionally valid, the Court notes that the alleged lack of particularized probable cause would not have been grounds for suppression anyway. Under the good-faith exception, “evidence obtained in objectively reasonable reliance on a subsequently invalidated search warrant” need not be suppressed. Leon, 468 U.S. at 922. Thus, the exclusionary rule only applies where the affidavit of probable cause is “so lacking in indicia of probable cause as to render official belief in its existence entirely unreasonable.” Griffith, 867 F.3d at 1278 (quotation omitted). Defendant reasserts his complaints regarding Google’s search protocols as reasons to infer bad faith on the part of the Government in executing the warrant. Def.’s Reply Mot. Suppress at 9–10, ECF No. 64. But, in line with its rejection of these same arguments in the context of Defendant’s overbreadth claim, the Court does not find that Defendant’s allegations that Google provided an excessive response to the Government’s request sufficient grounds to assume bad faith by the Government. Defendant’s other arguments as to why the good faith exception should not apply simply rehash his arguments regarding overbreadth and particularity. See id. at 11–15. For the reasons stated above, the Court is not persuaded by these arguments, and would not suppress the evidence obtained from the Geofence Warrant under the

78

good faith exception even if it had found that the Warrant lacked probable cause. See Hearing Transcript at 29, United States v. Cruz, Jr., No. 22-cr-0064 (D.D.C. Jan. 13, 2023) (holding, similarly, that the good faith exception would apply even if the Geofence Warrant lacked probable cause). 30 Accordingly, Defendant’s motion to suppress evidence obtained from the Geofence Warrant is denied.

IV. CONCLUSION

For the foregoing reasons, Defendant’s Motion to Transfer Venue (ECF No. 42) is DENIED, Defendant’s Motion for Expanded Voir Dire (ECF No. 42) is GRANTED IN PART and DENIED IN PART, Defendant’s Motion to Dismiss Counts 1 and 2 (ECF No. 46) is DENIED, Defendant’s Motion to Dismiss Counts 3 and 4 (ECF No. 47) is DENIED, and Defendant’s Motion to Suppress (ECF No. 43) is DENIED. An order consistent with this Memorandum Opinion is separately and contemporaneously issued.

Dated: January 24, 2023 RUDOLPH CONTRERAS United States District Judge

30 Because the Court has found the Geofence Warrant to be constitutional, and because it would have applied the good faith exception even if it had not, the Court does not reach Defendant’s argument that the fruits of the Geofence Warrant, including evidence obtained from the search of Defendant on November 9, 2021, should also be suppressed. See Def.’s Mot. Suppress at 32–35.