United States v. James

UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA UNITED STATES OF AMERICA ) ) ) v. ) Criminal Case No. 17-184 (RJL) ) ) ALLAN JAMES, ) ) Defendant. ) -h., MEMORANDUM OPINION (June / ?, 2019) Defendant Allan James ("James") pleaded guilty m this case to one count of unlawfully possessing a firearm and ammunition after being convicted of a crime punishable by imprisonment for a term exceeding one year. See 18 U.S.C. § 922(g)(l ). He currently awaits sentencing and has objected to the Sentencing Guidelines range calculation that appears in his Presentence Investigation Report ("PSR"). The calculation in the PSR is based, in part, on conduct that was not alleged in James's indictment and that is evidenced only by social media posts and testimony about social media posts. See PSR ,i,i 1Oa-1 Og [Dkt. # 18]. Indeed, James' s Guidelines range quadrupled as a result of its inclusion. James argues that this social media evidence is not sufficiently reliable to be used as a basis for enhancing his Sentencing Guidelines range. I agree, and his objection · will therefore be SUSTAINED. BACKGROUND The United States ("the Government") indict.ed James in October 2017 for a single count of violating 18 U.S.C. § 922(g)(l). See Indictment at 1 [Dkt. # 5]. The one-count indictment refers in turn to a single firearm-a Glock 22 .40 caliber pistol-and alleges that James possessed the gun, along with some .40 caliber ammunition, on or about July 12, 2017. See id. James, as is his right, pleaded guilty to the conduct charged in the indictment without first negotiating a plea deal with the Government. His Sentencing Guidelines range, if calculated based on the conduct described in the indictment, provides for twenty-one to twenty-seven months of incarceration. See Draft PSR ii 6-10, 7 5 n.4 [Dkt. # 17]. That Guidelines range was reflected in the draft PSR prepared by the United States Probation Office ("Probation"). See id. After reviewing the draft PSR, the Government argued that James' s Guidelines range should be significantly enhanced as a result of certain social media posts, most of which were recovered by executing search warrants for Facebook and Snapchat accounts. In particular, the Government contended that James' s base offense level should be increased from fourteen to twenty because these social media posts involved a semiautomatic firearm capable of accepting a large capacity magazme, see U.S.S.G. § 2K2.l(a)(4)(B); that James should be given an additional four-point enhancement because the posts linked him to at least eight firearms, 1 see U.S.S.G. § 2K2.1 (b )( 1)(B); and finally that James should be given another four~point enhancement because he trafficked 1 Alternatively, the Government contends that the social media evidence links James to between three and seven firearms, which would warrant a two-point enhanc-ement to his offense level. See U.S.S.G. § 2K2.l (b )(1 )(A). 2 firearms, see U.S.S.G. § 2k2.l(b)(5). According to the Government, images, videos, and messages posted to the social media accounts between January 2016 and June 2017 show James in possession of and, in some cases, trying to sell firearms. Ultimately, Probation adopted the Government's position in its final PSR, increasing James's Guidelines range to eighty-four to one hundred and five months of incarceration. See PSR i1 7 5. James, of course, strongly objects to this fourfold increase in his Guidelines range. The parties have filed memoranda setting forth their positions on the correct Guidelines range calculation, and the Government has been given multiple opportunities to identify all evidence that supports the calculation in the final PSR. 2 James's objection is now ripe for resolution. 3 ANALYSIS· Courts are naturally wary when prosecutors seek substantial sentencing enhancements based on conduct the same prosecutors declined to charge. Indeed, judges have "note[ d] ... continuing disagreement with the government's practice of charging relatively minor crimes, while using ... 'relevant [uncharged] conduct' ... to argue for 2 To resolve James's objection, the Cowi has considered the PSR, the Government's Memorandum in Aid of Sentencing ("Gov't Sentencing Mem.") [Dkt. # 22], James's Memorandum in Aid of Sentencing [Dkt. # 23], the Government's Notice of Change in Position [Dkt. # 25], the Government's Status Repo1i on Potential Expeti Witness [Dkt. # 29], the parties' Joint Status Repoti as to Sentencing [Dkt. # 30], the Government's Supplemental Guidelines Calculation [Dkt. # 33], James's Response to the Government's Sentencing Supplement [0kt. # 34], the Government's Supplemental Memorandum in Aid of Sentencing ("Gov't Suppl. Sentencing Mem.") [Dkt. # 38], James's second Response to the Government's Sentencing Supplement [Dkt. # 40], and oral argumenffrom the pa1iies at hearings held on July 6, 2018, February 25, 2019, and March 18, 2019. 3 Given the argument, briefing, affidavits, and expeti repo1is before the Couti, it is not necessary to hold an evidentiary hearing to resolve James's objection. See U.S.S.G. § 6A 1.3, cmt. ("Written statements of counsel or affidavits of witnesses may be adequate [to resolve disputes about sentencing factors] under many circumstances."). 3 significantly enhanced terms of imprisonment." United States v. Magee, 834 F.3d 30, 38 (1st Cir. 2016) (Torruella, J., concurring)(quoting U.S.S.G. § 1Bl.3(a)); see United States I . v. Scheiblich, 346 F. Supp. 3d 1076, 1082-85 (S.D. Ohio 2018) (citing cases in which judges express skepticism about this practice). One Court of Appeals 'has even gone so far as to expressly and repeatedly "rernind prosecutors -·not to indict defendants on relatively minor offenses and then seek enhancement sentences later by asserting that the defendant has committed other mor~ serious crimes for which, for whatever reason, the defendant was not prosecuted and has not been convicted. "' 4 United States v. Spiller, 261 F .3 d 683, 691 (7th Cir. 2001) (quoting United States v. Bacallao, 149 F.3d 717, 721 (7th Cir. 1998)). This judicial skepticism is, of course, rooted in the defendant's constitutionally protected rights. After all, by asking "judges to rely on acquitted or uncharged conduct to impose higher sentences than they otherwise would impose," prosecutors adopt a strategy that "seems a dubious infringement of the rights to due process and to a jury trial." United . States v. Bell, 808 F.3d 926,928 (D.C. Cir. 2015) (Kavanaugh, J., concurring with denial of petition for rehearing en bane). To say the least, "trial by sentencing enhancements" was not what_ Congress had in mind when it blessed the federal Sentencing Guidelines regime! But concerns about this practice extend beyond outright violations of the Constitution. "Many judges and commentators" who have questioned prosecutors' zeal to punish defendants for conduct that has never resulted in a conviction argue that the use of 4 See also United States v. Jordan, 221 F.3d 1339 (table), 2000 WL 874960, at *3 n.4 (7th Cir. 2000); United States v. Fischer, 905 F.2d 140, 142 (7th Cir. 1990). 4 uncharged or "acquitted conduct to increase a defendant's sentence undermines respect for the law and the jury system." United States v. Settles, 530 F.3d 920, 924 (D;C. Cir. 2008). Our Circuit Court has specifically acknowledged that it "understand[ s] why defendants find it unfair for district courts to rely on" such conduct at sentencing-even while reiterating that "long-standing precedents of the Supreme Court and [the Circuit] Court" often allow it. 5 Id. at 923. And so it seems obvious, as pointed out by Judge, now Justice, Kavanaugh, that "district judges would" be wise "to heed ... concern[s ]" about unfairness and, "in appropriate cases," to "avoid basing any part of [a defendant's] ultimate sentence on acquitted or uncharged conduct." Bell, 808 F.3d at 928 (Kavanaugh, J., concurring with denial of petition for rehearing en bane); see also id. ("[F]ederal district judges have power in individual cases to disclaim reliance on acquitted or uncharged conduct" by "vary[ing] the sentence downward to avoid basing any part of the ultimate sentence on acquitted or uncharged conduct."); Settles, 530 F.3d at 924 ("[E]ven though district judges are not required to discount acquitted conduct, the Booker-Rita-Kilnbrough-Gall line of cases may allow district judges to discount acquitted conduct in particular cases .... " ( citing Kimbrough v. United States, 552 U.S. 85, 101 (2007)) (emphasis in original)). Moreover, the Government's requested sentence in this particular case deserves considerable scrutiny because its use of uncharged conduct is compounded by the 5 Facts, other than the fact of a prior conviction, that increase a statutorily imposed maximum penalty or mandatory minimum penalty must be proven beyond a reasonable doubt to a jury. See Alleyne v. United States, 570 U.S. 99, 103 (2013); Apprendi 1i. New Jersey, 530 U.S. 466, 490 (2000). But so long as conduct does not alter the statutorily mandated sentencing range and the "conduct has been proved by a preponderance of the evidence," a sentencing cou1i "may co11sider uncharged or even acquitted conduct in calculating an appropriate sentence." Settles, 530 F.3d at923. 5 weakness of the evidence it offers to prove the conduct. To say the least, social media posts-.with little else-can hardly justify the quadrupling of a defendant's Guidelines range. How so? Courts, and commentators, have explained why determining the accuracy of information posted to social media is far from straightforward and hence of questionable reliability. Facebook records, for example, are "no more sufficient to confirm the accuracy or reliability of' information posted to the website "than a postal receipt would be to attest to the accuracy or reliability of the contents of the enclosed mailed letter." United States v. Browne, 834 F.3d 403, 411 (3d Cir. 2016). Indeed, Facebook can actually make photographic evidence less useful because it "'strips ... metadata as the photograph is uploaded' ... rendering [its] date of creation unknown." United States v. Farrad, 895 F.3d 859, 867 (6th Cir. 2018) (quoting testimony from a police officer trained in social media investigations). Snapchat is similarly problematic as some of its "key function[ s]" give users the ability to edit images before sharing them. Agnieszka McPeak, Disappearing Data, 2018 Wrs. L. REV. 17, 34 (2018) ("A key function within Snapchat is the use of Filters, which allows users to add multiple overlays to their images. Lenses also allow 'real-time special effects and sounds' to be added to images." (footnote omitted)). Even determining who posted a particular message, imag~, or video to Facebook or Snapchat is difficult "because of the great ease with which a social media account may be falsified or a legitimate account may be accessed by an imposter." Browne, 834 F .3d at 412. For these reasons, courts usually "consider[] . . . extrinsic evidence to determine whether the 6 government [can meet] its authentication burden" before permitting the introduction of social media evidence against a defendant at trial. Id. at 413. Not surprisingly, the reliability standard applicable at sentencings, see U.S.S.G. § 6Al.3(a), 6 counsels in favor of requiring extrinsic corroboration before social media evidence is used to resolve a dispute relevant to a sentencing determination. Because social media posts may contain false information and doctored images, and because posters may be using hacked accounts or stolen identities, uncorroborated social media posts will not often establish, with any degree of reliability, that the information in the posts is accurate or that the posts are what they purport to be. The Government here has not offered sufficient extrinsic corroboration to establish the necessary reliability of its social media evidence. While it has pointed to reasons to believe James had access to the Facebook and Snapchat accounts from which it collected evidence, see Declaration of Special Agent Robert Mayo ("Mayo Deel.") at 22 [Dkt. # 3 8- 1] (reporting that photos found on James' s cell phone had been uploaded to the social media accounts), account access, alone, cannot possibly justify the sentence enhancements the Government requests here due to the myriad of facts the Government would need to establish in order to render these photos sufficiently reliable. For example, the Government would need to establish that the alleged firearms depicted in the social media pictures and videos are actual firearms. See 18 U.S.C. § 92l(a)(3); U.S.S.G. § 2K2.1 cmt. n.l (defining "[f]irearm" to "ha[ve] the meaning given 6 At sentencing, "information [must] ha[ve] sufficient indicia ofreliability to suppot1 its probable accuracy," if it is to be used to resolve a dispute between the pa11ies. In re Sealed Case, 246 F.3d 696, 699- 700 (D.C. Cir. 2001) (quoting U.S.S.G. § 6Al.3). 7 that term in 18 U.S.C. § 92l(a)(3)"). However, assuming those photos and videos have not been doctored, the Government's own firearms expert here concedes, "it is not possible to determine if an object depicted in a photograph or video is a functional firearm, a replica firearm, or a toy firearm." Mayo Deel. App. A, Laboratory Report of Bryce A. Ziegler ("Ziegler Report") at 4. In addition, the Government would need to establish that the pictures and videos posted to social media were created around the same time they were uploaded to substantiate its argument that the social media evidence depicts a single, year-and-a-half- long course of conduct or common scheme. See Gov't Sentencing Mem. at 24 (arguing that the social media evidence depicts a single course of conduct or common scheme because James allegedly "possessed between eight and 24 firearms over [a] year-and-a- half period"); United States v. Mahone, 688 F .3d 907, 909 (8th Cir. 2012) ("At least four of our sister circuits have concluded that a defendant's pattern of unlawfully possessing multiple firearms over the course of several months constitutes 'the same course of conduct' for relevant conduct purposes." ( emphasis added)). But when the Government seized and searched James' s cell phone, these allegedly inculpatory pictures and videos could not be found, see Mayo Deel. at 22-23, and, to date, none appear to have been analyzed for a date of creation. To say the least, the Government's theory that James possessed multiple weapons over a year-and-a-half period falls apart if its social media evidence consists of pictures and videos that were years old when uploaded. See Farrad, 895 F.3d at 867, 872 (explaining why evidence from Facebook cannot establish the date a photograph was taken). 8 Moreover,· the Government would also need to establish that James actually possessed the weapons depicted in the images and videos. This is especially so when you consider that several of these social media posts depict only apparent firearms, without showing any individual, let alone James, present. See Mayo Deel. at 12-13. To say the least, an image of a firearm alone does not establish that the photographer, videographer, or person who uploaded the image to social media ever owned or possessed the depicted firearm. See United States v. Winters, 530 F. App'x 390, 395 (5th Cir. 2013) ("A photograph's appearance on a personal webpage does not by itself establish that the owner of the page possessed or controlled the items pictured."). Finally, the Government would need to establish that the images and videos that do include James' s image have not been doctored to give the illusion that James possessed firearms he never actually had. See United States v. Frabizio, 445 F. Supp. 2d 152, 15 8 (D. Mass. 2006) ("[T]he government has not made the threshold showing that a visual observer can reliably evaluate the relevant pictures for signs of manipulation and computer- generation. Rather, the evidence strongly suggests that it is extremely difficult, if not impossible, for a photographic expert, let alone a lay observer, to determine whether the images involved in the instant case are real images or images created or manipulated through digital technology."), clarified on reconsideration, 463 F. Supp. 2d 111 (D. Mass. 2006). Unfortunately, the Government offers no testimony that could assist the Court in distinguishing between images and video that have been virtually created or altered and images and video that were recorded on a camera and never later manipulated. 9 In the final analysis, the Government contends that its interpretation of the photos and videos is bolstered by messages and captions that were posted to the same social media accounts as the images. See, e.g., Gov't Sentencing Mem. at 28 ("The defendant received at least the MAC-I 0-style gun, the AR-15-style rifle, and the distinctive white pistol in the 'Back on Dat Action' video with the intent to sell them to other people, as evidenced by his Facebook discussion of pricing."). But introducing statements posted to social media requires still more antecedent proof: The Government must establish that James authored the statements being attributed to him. Doing so here is difficult because the Government did not thoroughly investigate whether any individuals other than James had access to James' s cell phone or social media accounts. See Status Hr' g Tr. at 13: 10-15 :7 (Mar. 18, 2019) [Dkt. # 36]; Gov't Suppl. Sentencing Mem. at 2 n.2. At bottom, the social media posts that the Government counts as evidence of James' s "possession of the eleven firearms," with intent to traffic three, as "part of ... a common scheme or plan," Gov't Sentencing Mem. at 24, 28, are just as consistent with the possibility that a person with access to James' s social media accounts wanted viewers to incorrectly think James had and could sell 7 firearms. Within the world of social media, the latter possibility is no less conceivable than the former. See Farrad, 895 F .3d at 874 ("[P]eople advertise false images of themselves all the time-putting signs on their doors suggesting that they have a fancy. alarm system, a fierce guard dog, or a high-powered firearm that they do not in fact have. The Facebook poseur['s] ... incentive is not 7 None of the social media messages that the Government offers as evidence of James's intent to traffic firearms suggest that James ever finalized an agreement to sell any firearm. See Gov't Sentencing Mem. at 9, 12. Nor does the Govei'nment contend that any sale was consummated. See id. at 28. 10 necessarily to show a real gun, which is costlier, harder to procure, and, needless to say, more·likely to subject its owner to criminal liability."). This potential for digital deception is precisely why courts require extrinsic corroboration before crediting social 1nedia evidence. See United States v. Vayner, 769 F.3d 125, 132-33 (2d Cir. 2014} (holding that a printout from a social media profile should not have been admitted at trial when "the information contained on the [profile] allegedly tying [it] to [the defendant] was also known by ... others, some of whom may have had reasons to create a profile page falsely attributed to the defendant" and the profile was not otherwise corroborated). And persuasive, extrinsic corroboration is what the Government lacks here. Only one firearm was ever found on James or at his residence. 8 The other alleged firearms shown on social media could have been stored outside of James' s residence, given away, thrown out, or sold-or the social media posts offered by the Government could depict toys, replicas, or digital creations.· Neither the posts themselves, nor any witness offered by the Government, reliably establishes one or the other. Moreover, the Government offers no legal authority that suggests its uncorroborated social media posts are sufficiently reliable to enhance James' s sentence. The Government merely points to three out-of-circuit, unpublished, nonprecedential opinions in which sentencing enhancements based in part on digital images were upheld on clear error review. See Gov't Suppl. Sentencing Mem. at 4 (citing United States v. Villanueva, 315 F. App'x 845, 848-49 (11th Cir. 2009) (per curiam); United States v. Spencer, 685 F. App'x 863, 8 The Government did recover ammunition when it executed a search warrant at James's residence. See Mayo Deel. at 22. The recovered ammunition is the same caliber as the firearm and ammunition referred to in James' s indictment. See Indictment at 1. 11 867 (11th Cir.2017) (per curiam); United States v. Weigant, 448 F. App'x 22, 26-27 (11th Cir. 2011) (per curiam)). Unfortunately for the Government, none of the three opinions addresses the fundamental problem with the Government's evidence here-the lack of offline, or otherwise reliable, corroboration. In two of the Government's cited cases, the District Court heard testimony that the firearms depicted in the digital evidence offered at sentencing were real. See Villanueva, 315 F.. App'x at 849; Weigant, 448 F. App'x at 27. Here, the Government submitted a report from a firearms examiner who cannot even determine whether the apparent firearms are real or not. See Ziegler Report at 4. The Government's third case does not deal with social media evidence at all. In United States v. Spencer, a District Court relied on photographs that, unlike here, were recovered from the defendant's cell phone and that, unlike here, were sufficiently clear to identify specific makes and manufacturers of depicted firearms. See 685 F. App'x at 867. By comparison, the Government's expert in this case cautions, (,·Due to poor image quality or a lack of observable physical characteristics, it may not be possible to determine a specific model or manufacturer of a potential firearm" from the offered social media posts. Ziegler Report at 4. Not exactly comforting that! In short, the Government points to cases in which digital evidence was deemed reliable after it was corroborated through testimony less equivocal than what is offered here or through comparisons to functional firearms more detailed than are possible here. Where prosecutors can bolster social media evidence in a way that establishes "sufficient indicia of reliability to support its probable accuracy," In re Sealed Case, 246 F .3d at 700 (quoting U.S.S.G. § 6Al .3), social media posts may well be an appropriate basis on which 12 to resolve disputes at sentencing. But the Government's submiss!ons in this case make clear that it cannot clear that hurdle. Its proposed witnesses offer little by way of extrinsic corroboration for the social media posts allegedly establishing James 's possession and trafficking of multiple firearms and possession of a semiautomatic firearm capable of accepting a large capacity magazine. The Government will not, therefore, be permitted to rely on its social media evidence at James's sentencing to modify in any way the initial Ciuidelines calculation of the Probc1tion Office. CONCLUSION James's objection to the use of the Government ' s social media evidence in calculating his Sentencing Guidelines range will be SUSTAINED. The sentencing hearing in this case remains set for June 18, 2019 , and the applicable Guidelines range will be twenty-one to twenty-seven months of imprisonrnent. ~QJI .. _/ RICHARDJ. ~ United States District Judge 13