Twitter — Tweet Authenticated by Twitter Handle, Profile Photo, Contents, Reply Letter Doctrine — N.J. Appeals Court Adopts Circumstantial Approach to Social Media Authentication
State v. Hannah, 2016 N.J. Super. LEXIS 156 (N.J. Super. App. Div. Dec. 20, 2016):
Defendant Terri Hannah appeals her July 10, 2015 conviction for simple assault after a trial de novo in the Law Division, following her conviction in municipal court. She argues that a Twitter posting was improperly admitted into evidence, citing a Maryland case requiring that social media postings must be subjected to a greater level of authentication. We reject that contention, holding that New Jersey's current standards for authentication are adequate to evaluate the admission of social media postings. Under those standards, we find it was not an abuse of discretion to admit the tweet. Finding defendant's remaining claims lack merit, [*2] we affirm.c
The Law Division found the following facts based on the testimony in the Vineland Municipal Court. On September 22, 2012, Arnett Blake and his girlfriend, Cindy Edwards, attended a party at a community center. Defendant, Blake's ex-girlfriend, also attended the party.
While in the bathroom, Edwards encountered defendant "making rude comments about her." While Edwards was still in the bathroom, defendant exited the bathroom, approached Blake, and said "I should F your girlfriend up." Later that night, defendant purposefully bumped into Blake.
As Edwards and Blake were in the lobby trying to leave the party, defendant quickly approached Blake with her closed fist in the air. Blake reacted by pushing defendant away, prompting security to grab him. When Edwards turned to say something, she saw defendant holding a high-heeled shoe, with which defendant struck Edwards in the face. Blake also saw defendant hit Edwards with a shoe as he was being escorted outside. When defendant was brought outside, Edwards saw defendant did not have her shoes on.
Edwards and Blake went to the police station to report the incident and then went to the hospital, where Edwards received nine stitches. [*3] After the assault, defendant and Edwards had communications "back and forth" on Twitter. On December 28, 2012, Edwards saw defendant posted a tweet saying "shoe to ya face bitch."
In municipal court, defendant offered a different version of events. Defendant testified she approached Blake and told him that she heard "hearsay . . . saying that [she] was going to . . . beat his girlfriend up." Defendant told Blake she "wanted to clear the air and let him know that [she was] not going to do anything to [his girlfriend]." Later during the party he "push[ed] [defendant] to the side." Defendant later saw Blake in the lobby and decided to ask him why he pushed her. She became aggressive and started yelling, and a security guard took her "straight out . . . of the party." Defendant testified she never saw Edwards that night and never punched anyone or hit anyone with a shoe.
Defendant called as a witness a security guard at the party, who testified he saw defendant approaching a man "in an aggressive manner" and heard her make hostile remarks. "[B]efore she could do anything," the guard "snatched her up and . . . took her out of the building." He told her she was not permitted to reenter the party. He did not [*4] see Blake or Edwards or see defendant hit anyone with a shoe.
Defendant was charged with aggravated assault, but the charge was downgraded to simple assault, a disorderly persons offense. N.J.S.A. 2C:12-1(a)(1). On January 12, 2015, the municipal court found defendant guilty and imposed a $307 fine plus costs and assessments. Defendant appealed. On June 5, 2015, the Law Division conducted a trial de novo, hearing oral argument. After reserving decision, the Law Division found defendant guilty of simple assault and imposed the same monetary penalties. The Law Division credited Edwards and Blake, found defendant not credible, and found the passage of two years compromised the security guard's recollection of the event.
On appeal to this court, defendant argues:
POINT I - THE COURT'S ADMISSION OF THE TWEET (S-4), CLAIMED BY THE STATE TO HAVE BEEN POSTED BY THE DEFENDANT TO HER TWITTER ACCOUNT, WAS ERROR AS:
(1) THE SUPERIOR COURT JUDGE MISTAKENLY ADOPTED WHAT HE BELIEVED TO BE THE DIFFERENT, MORE LENIENT TEXAS AUTHENTICATION STANDARD [RATHER THAN THE MARYLAND STANDARD] WITHOUT UTILIZING NEW JERSEY'S CIRCUMSTANTIAL EVIDENCE MODE OF AUTHENTICATION, N.J.R.E. 901, AND ASSESSING THE NON-PRODUCTION OF THE OTHER "DIFFERENT" SNAPSHOTS [*5] SUPPOSEDLY TAKEN BY THE ACCUSER IN AN ALLEGED EXCHANGE OF TWEETS BETWEEN ACCUSER AND DEFENDANT SOME THREE MONTHS AFTER THE ALLEGED ASSAULT;
(2) THIS JUDGE IMPROPERLY AUTHENTICATED THE TWEET BY RELYING ON THE ACCUSER'S TESTIMONY AS WELL AS THAT OF THE DEFENDANT, WHO ONLY TESTIFIED AFTER THE STATE HAD RESTED;
(3) WITH THIS JUDGE FINDING [SIC] THAT THE DEFENDANT'S JANUARY 12, 2015 MUNICIPAL COURT TESTIMONY WAS NOT CREDIBLE BECAUSE HE CONTRASTED HER TESTIMONY WITH EXHIBIT D-4 ATTACHED TO DEFENSE COUNSEL'S MAY 8, 2015 APPEAL BRIEF; AND,
(4) THIS JUDGE ADMITTED THE TWEET, WITHOUT ANALYSIS AS TO THE TWEET'S RELEVANCE OR PROBATIVE VALUE.
Defendant argues a message sent on Twitter should not have been admitted [*7] as it was not properly authenticated.1 "[C]onsiderable latitude is afforded a trial court in determining whether to admit evidence, and that determination will be reversed only if it constitutes an abuse of discretion." State v. Kuropchak, 221 N.J. 368, 385-86 (2015) (citation omitted). "Under that standard, an appellate court should not substitute its own judgment for that of the trial court, unless 'the trial court's ruling "was so wide of the mark that a manifest denial of justice resulted."'" Ibid. (citation omitted). We must hew to our standard of review.
1 "Twitter is self-described as 'an information network made up of 140-character messages called Tweets.'" State ex rel. J.F., 446 N.J. Super. 39, 44 n.7 (App. Div. 2016) (citation omitted); accord The Twitter Glossary, Twitter, Inc., https://support.twitter.com/articles/166337# (last visited Dec. 13, 2016) (hereinafter Glossary). "These messages are posted to your profile, sent to your followers, and are searchable on Twitter search." New User FAQs, Twitter, Inc., https://support.twitter.com/articles/13920# (last visited Dec. 6, 2016).
The municipal court and the Law Division each admitted as Exhibit S-4 the following tweet allegedly posted by defendant on December 28, 2012: "No need for me to keep responding to ya stupid unhappy fake mole having ass.. how u cring2 in a corner with a shoe to ya face bitch." The tweet displayed defendant's profile photo and defendant's Twitter handle, "@cirocgirl25."3
2 Edwards interpreted "cring" as "crying." Defendant read "cring" as "cringe."
3 A Twitter "'handle' is used to identify a particular user on Twitter and is formed by placing the @ symbol next to a username." Roca Labs, Inc. v. Consumer Op. Corp., 140 F. Supp. 3d 1311, 1319 n.4 (M.D. Fla. 2015). A Twitter "username is how you're identified on Twitter, and is always preceded immediately by the @ symbol." Glossary. A Twitter "header photo" is "[y]our personal image that you upload, which appears at the top of your profile." Ibid. This "profile photo" "appears next to each of your Tweets." Ibid.
Edwards testified she recognized the tweet as being written by defendant because it displayed defendant's picture. She also was familiar with defendant's Twitter handle, "@cirocgirl25." Moreover, Edwards testified the tweet was posted "in response to things that [*8] [Edwards] was saying" and they were communicating "back and forth." On December 28, 2012, Edwards went onto defendant's Twitter page, saw the posted tweet, and captured it as a screenshot.4
4 "A 'screenshot' is a snapshot image of the information displayed on a computer screen at a given point in time." State v. Ravi, 447 N.J. Super. 261, 270 n.8 (App. Div. 2016).
Defendant testified the Twitter page displayed a picture of her and her Twitter handle. However, she testified she did not author the tweet.
When the State sought to admit the tweet, defense counsel objected, arguing "[t]here's no way anybody besides Twitter can say that this came from [defendant]." In admitting the tweet, the municipal court ruled nothing "requires somebody to be here from Twitter. I think somebody can testify as to it as Ms. Edwards [did] and we go from there."
At the trial de novo, the Law Division classified the methods of authenticating a social media post into two camps: the Maryland approach and the Texas approach, respectively citing Griffin v. State, 19 A.3d 415 (Md. 2010), and Tienda v. State, 358 S.W.3d 633 (Tex. Crim. App. 2012).
In Griffin, the Maryland Court of Appeals considered what the test should be for the authentication of printed pages of a MySpace profile. Griffin, supra, 19 A.3d at 416-17. Citing "[t]he potential for abuse and manipulation of a social networking site by someone other than its purported creator and/or user," Griffin ruled that images from such a site require "greater scrutiny" than [*9] "letters and other paper records." Id. at 423-24 (concluding that "a printout of an image from such a site requires a greater degree of authentication"). The court suggested three possible methods of authentication. Id. at 427.
The first method was "to ask the purported creator if she indeed created the profile and also if she added the posting in question, i.e. '[t]estimony of a witness with knowledge that the offered evidence is what it is claimed to be.'" Ibid. (citation omitted). The second method was "to search the computer of the person who allegedly created the profile and posting and examine the computer's internet history and hard drive to determine whether that computer was used to originate the social networking profile and posting in question." Ibid. The third method was "to obtain information directly from the social networking website that links the establishment of the profile to the person who allegedly created it and also links the posting sought to be introduced to the person who initiated it." Id. at 428.
In Tienda, the Texas Court of Criminal Appeals did not employ any of the three Griffin methods but concluded "there are far more circumstantial indicia of authenticity in this case than in Griffin - enough, [*10] we think, to support a prima facie case that would justify admitting the evidence and submitting the ultimate question of authenticity to the jury." Tienda, supra, 358 S.W.3d at 647. The Texas court found "the internal content of . . . [the] MySpace postings - photographs, comments, and music - was sufficient circumstantial evidence to establish a prima facie case such that a reasonable juror could have found that they were created and maintained by" a particular individual. Id. at 642.
Here, the Law Division found "[t]he Maryland approach is too strict in its authentication requirements," stating that its three methods "are unrealistic for a party to fulfill" and "create a higher bar than originally intended by the Rules." Accordingly, the Law Division "chose to adopt a rule of admissibility more similar to the Texas approach."
Defendant argues that Texas follows the Maryland approach and that we should adopt the Maryland approach with its "three non-exclusive methods" of authentication. Id. at 647. We reject any suggestion that the three methods of authentication suggested in Griffin are the only methods of authenticating social media posts. We also reject Griffin's suggestion that courts should apply greater scrutiny when authenticating [*11] information from social networks. See Parker v. State, 85 A.3d 682, 686-87 (Del. 2014) (rejecting the Griffin "greater scrutiny" approach and "conclud[ing] that social media evidence should be subject to the same authentication requirements under the Delaware Rules of Evidence Rule 901(b) as any other evidence"); see also United States v. Vayner, 769 F.3d 125, 131 n.5 (2d Cir. 2014) (noting that Griffin requires "greater scrutiny" and stating "we are skeptical that such scrutiny is required").
Rather, we agree with Tienda's observation that
[c]ourts and legal commentators have reached a virtual consensus that, although rapidly developing electronic communications technology often presents new and protean issues with respect to the admissibility of electronically generated, transmitted and/or stored information, including information found on social networking web sites, the rules of evidence already in place for determining authenticity are at least generally "adequate to the task."
[Tienda, supra, 358 S.W.3d at 638-39 (citation omitted).]
Indeed, "jurisdictions across the country have recognized that electronic evidence may be authenticated in a number of different ways consistent with Federal Rule 901 and its various state analogs." Id. at 639.
"Despite the seeming novelty of social network-generated documents, courts have applied the existing concepts of authentication under [*12] Federal Rule 901 to them," including "the reply letter doctrine [and] content known only to the participants." 2 McCormick on Evidence § 227, at 108 (Broun ed., 2013).5 N.J.R.E. 901 "generally follows Fed. R. Evid. 901" and incorporates both of those methods for authentication. Biunno, Weissbard & Zegas, Current N.J. Rules of Evidence [Biunno], 1991 Supreme Court Committee Comment & comment 3 on N.J.R.E. 901 (2016).
5 McCormick notes that Griffin "imposed a heavier burden of authentication," but "[a]s with the advent of the telegraph, the computer, and the internet," "the perceived need for this additional burden may dissipate." McCormick on Evidence, supra, § 227, at 109-10.
We need not create a new test for social media postings. Defendant argues a tweet can be easily forged, but so can a letter or any other kind of writing. The simple fact that a tweet is created on the Internet does not set it apart from other writings. Accordingly, we apply our traditional rules of authentication under N.J.R.E. 901.
Though in "electronic" form, a tweet is a "writing." See N.J.R.E. 801(e). "The requirement of authentication of writings . . . and the recognized modes of proving genuineness have been developed by case law over two centuries." Biunno, supra, comment 1 on N.J.R.E. 901 (2016). "Over the years authentication requirements have become more flexible, perhaps because the technology has become more commonplace." Suanez v. Egeland, 330 N.J. Super. 190, 195 (App. Div. 2000).
N.J.R.E. 901 provides: "The requirement of authentication or identification as a condition precedent to admissibility is satisfied by evidence [*13] sufficient to support a finding that the matter is what its proponent claims." Authentication "'does not require absolute certainty or conclusive proof' - only 'a prima facie showing of authenticity' is required." State v. Tormasi, 443 N.J. Super. 146, 155 (App. Div. 2015) (quoting State v. Mays, 321 N.J. Super. 619, 628 (App. Div.), certif. denied, 162 N.J. 132 (1999)). "This burden was not designed to be onerous." State v. Hockett, 443 N.J. Super. 605, 613 (App. Div. 2016).
"'Courts are inclined to assess their role in authentication as that of a screening process[,]' and 'will admit as genuine writings which have been proved prima facie genuine . . . leaving to the jury more intense review of the documents.'" Konop v. Rosen, 425 N.J. Super. 391, 411 (App. Div. 2012) (quoting Biunno, supra, comment 1 on N.J.R.E. 901 (2011)). In a bench trial, as here, "considering the judge's dual role with regard to its admission and weight, the better practice in such a circumstance will often warrant the admission of the document and then a consideration by the judge, as factfinder." Tormasi, supra, 443 N.J. Super. at 156-57.
Authenticity can be established by direct proof - such as testimony by the author admitting authenticity - but direct proof is not required. Biunno, supra, comment 2 on N.J.R.E. 901 (2016); N.J.R.E. 903. "A prima facie showing may be made circumstantially." Konop, supra, 425 N.J. Super. at 411. "Such circumstantial proof may include demonstrating that the statement 'divulged intimate knowledge of information which one would expect only the person alleged [*14] to have been the writer or participant to have.'" Ibid. (quoting Biunno, supra, comment 3(b) on N.J.R.E. 901 (2011)). Here, the tweet contained several such details, including "shoe to ya face," information that one would expect only a participant in the incident to have.6
6 In Konop, we cited with approval Kalola v. Eisenberg, 344 N.J. Super. 198, 200 (Law Div. 2001), which found a threatening phone call to the plaintiff dentist authenticated because the caller "identified himself as the defendant, referenced the plaintiff and described the dental work previously performed." Konop, supra, 425 N.J. Super. at 411-13.
Additionally, under the reply doctrine, a writing "may be authenticated by circumstantial evidence establishing that it was sent in reply to a previous communication." Mays, supra, 321 N.J. Super. at 629; see Biunno, supra, comment 3(c) on N.J.R.E. 901 (2016). Here, Edwards testified that the tweet was posted in response to her communications with defendant, as part of a "back and forth" between them. Moreover, the tweet said there was "[n]o need for me to keep responding to ya," apparently referring to Edwards who received a "shoe to ya face."
Defendant's Twitter handle, her profile photo, the content of the tweet, its nature as a reply, and the testimony presented at trial was sufficient to meet the low burden imposed by our authentication rules. Those facts established a prima facie case "sufficient to support a finding that the matter is what its proponent claims." N.J.R.E. 901. Other courts have admitted tweets applying their similar authentication standard. See Wilson v. State, 30 N.E.3d 1264, 1267-69 (Ind. Ct. App. 2015); Sublet v. State, 113 A.3d 695, 720-21 (Md. 2015); see also 5 Weinstein's Federal Evidence: [*15] Discovering and Admitting Computer-Based Evidence § 900.07[4A] (Joseph M. McLaughlin ed., 2016).
Defendant argues the Law Division cited not only the State's evidence but also defendant's testimony in the municipal court that the tweet bore her picture associated with her Twitter account. However, she cites no authority precluding the Law Division from considering the uncontested fact that the tweet bore defendant's photo and Twitter handle, which was established through the testimony of Edwards as well as defendant.
In the municipal court, defendant testified "[a]nybody can make a fake Twitter page and put your name on it and put something on there." She testified that because she deleted her Twitter account months before, someone could have taken the same Twitter handle and used it. After the municipal court did not credit this claim, defendant tried to bolster her testimony by submitting new evidence to the Law Division, including printouts of Twitter policies showing that Twitter "is currently unable to accommodate individual requests for inactive or suspended usernames." The Law Division cited that policy as one of several reasons for finding that defendant's testimony was not credible and that she [*16] "did not actually delete her Twitter account and that she did, in fact, author and publish the Tweet in question."
Defendant now argues it was improper for the Law Division to rely on evidence that was not before the municipal court. Notably, defendant herself presented the Twitter policies to the Law Division and did not object to the court's consideration of them. Therefore, she must show at least plain error. However, she fails to show the court's consideration of the policies was "clearly capable of producing an unjust result." R. 2:10-2. There was ample other evidence supporting the court's decision not to credit defendant's denial that she wrote and posted the tweet.
The Law Division, like the municipal court, provided sufficient reasons for finding the tweet authentic, relevant, and admissible. Defendant's remaining arguments regarding authentication lack sufficient merit to warrant discussion. R. 2:11-3(e)(2). Accordingly, we find no abuse of discretion in admitting the tweet.
Share this article: