[Author’s note: After being posted, this article was accepted for publication in the Southern Methodist University, Dedman School of Law Science and Technology Law Review. It is available at my SSRN author page here: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2704630. The full citation is: Daniel A. Horwitz, A Picture’s Worth A Thousand Words: Why Ballot Selfies Are Protected By The First Amendment, 18 SMU Sci. & Tech. L. Rev. __ (2015).]
A Picture’s Worth A Thousand Words: Why Ballot Selfies Are Protected By The First Amendment
On September 9, 2014, Andrew Langlois walked into his polling place disgusted with the candidates who were running for U.S. Senate in the New Hampshire Republican Primary. Rather than voting for one of the ten candidates who appeared on his ballot, Mr. Langlois instead decided to vote for “his recently-deceased dog, ‘Akira,’” whom he selected as a write-in candidate. Thereafter, Mr. Langlois whipped out his smartphone, snapped a photo of his completed ballot, and posted the photo to his Facebook page along with the caption: “Because all of the candidates SUCK, I did a write-in of Akira.”
Unfortunately for Mr. Langlois, the election did not go his way. To start, Akira failed to win the Republican Party’s nomination—instead, that honor went to former Massachusetts Senator Scott Brown, who went on to lose the general election to incumbent New Hampshire Senator Jeanne Shaheen. Adding insult to injury, Mr. Langlois also received a call from the New Hampshire Attorney General’s Office, which “explained that he was being investigated for posting his ballot on social media” in violation of a recently enacted New Hampshire law that prohibited “taking a digital image or photograph of [one’s] marked ballot and distributing or sharing the image via social media or by any other means.” New Hampshire, for its part, is not alone in prohibiting voters from photographing and sharing photographs of their marked ballots—which have since become known as “ballot selfies.” In fact, according to a recent CBS report on the issue, “ballot selfies are against the law” in thirty-five states, and violators are frequently subject to being “punished with fines, invalidated ballots, or even jail time.”
Along with two other voters who had posted photos of their ballots on social media during New Hampshire’s September 2014 primary election, Mr. Langlois took up the torch of New Hampshire’s “live free or die” tradition and sued his state in federal court. Along with State Representative Leon Rideout and first-time candidate Brandon Ross—both of whom had publicly memorialized their own candidacies by taking ballot selfies—Mr. Langlois argued that New Hampshire’s ban on ballot photography abridged his rights to freedom of speech and freedom of expression in violation of the First Amendment to the United States Constitution. In an August 11, 2015, opinion, U.S. District Court Judge Paul Barbadoro agreed, holding that New Hampshire’s ballot selfie ban was “invalid because it is a content-based restriction on speech that cannot survive strict scrutiny.” A similar ruling from the U.S. District Court for the Southern District of Indiana quickly followed suit.
Unexpectedly, Judge Barbadoro’s ruling quickly set off a firestorm in the election law world. Within a week, U.C. Irvine Professor Rick Hasen had penned an article in Reuters entitled: Why the Selfie is a Threat to Democracy, in which he argued that Judge Barbadoro “made a huge mistake because without the ballot selfie ban, we could see the reemergence of the buying and selling of votes—and even potential coercion from employers, union bosses and others.” Of note, Professor Hasen was not alone in expressing these feuters entitled: istrict of Indiana quickly followed suit. of State, et al., ve because they leave open sfy strict scruinty. ars. For example, according to Doug Chapin, director of the program for excellence in election administration at the University of Minnesota, “ballot selfies create a vulnerability in the election process that vastly outweighs any societal or personal benefit the selfie brings.”
Despite the concerns expressed by authorities like Professor Hasen, Mr. Chapin and others, however, Judge Barbadoro was correct in holding that ballot selfies are protected by the First Amendment. To begin, “as a general matter, ‘the First Amendment means that government has no power to restrict expression because of its message, its ideas, its subject matter, or its content.’” Because ballot selfie prohibitions forbid individuals from sharing one type of image—a photograph of a marked election ballot—but not other images, there is little doubt that such laws represent a content-based restriction on speech. As a result, such laws must satisfy an onerous legal standard known as “strict scrutiny.” To uphold a law under this standard, the government must prove that the challenged law is both narrowly tailored and the least-restrictive means available to further a compelling governmental interest. Unfortunately for proponents of ballot selfie bans, however, the Supreme Court has described strict scrutiny as a “well-nigh insurmountable burden,” and vanishingly few laws can overcome it. In particular, ballot selfie bans cannot satisfy strict scrutiny for three reasons.
First, assuming that preventing vote buying qualifies as a compelling governmental interest, ballot selfie bans are not narrowly tailored to achieve this objective. In particular, blanket prohibitions on ballot photography fail to satisfy strict scrutiny because they unnecessarily restrict a substantial amount of protected political speech while simultaneously doing nothing to prevent far simpler methods of vote buying. Second, the “compelling” nature of the government’s interest in enacting sweeping laws to guard against vote buying is subject to considerable doubt, given that vote buying is so rare as to be statistically non-existent even in jurisdictions where it is theoretically easy to accomplish. Third, because in most cases voters have the ability to change their votes even after photographing their ballots, a ballot selfie does not actually provide a would-be vote buyer any assurance that a voter has cast his or her ballot in a particular way—rendering the entire premise behind ballot photography prohibitions completely baseless.
This Article addresses each of these issues in turn. Part I explains why prohibitions on ballot selfies are not narrowly tailored to accomplish their intended goal of prohibiting vote buying. Part II addresses the governmental interests that purportedly justify ballot selfie prohibitions, and it explains why ballot selfie bans do not in fact address an “actual problem” that is in need of solving. Part III expounds on these problems by explaining why ballot selfies do not even provide a would-be vote buyer any assurance that a voter has actually cast his or her ballot in a particular way, rendering the ballot selfie a valueless method of facilitating voter fraud. Taken together, Part IV concludes that ballot selfies are a constitutionally protected form of political speech, and that the widespread efforts to prohibit their distribution cannot withstand constitutional scrutiny.
I. Prohibitions on Ballot Selfies Are Not Narrowly Tailored To Prohibit Vote buying
Adopting the mindset of the many legislators who have supported ballot selfie prohibitions, suppose that vote buying were a huge problem—as it unquestionably was back in the nineteenth and early twentieth centuries. Next, suppose that you were a legislator seeking to fix this problem. Would prohibiting voters from photographing their ballots and distributing the images on social media do the trick?
The answer to this question, as Professor Hasen correctly argues, is yes—but only in part, and not even particularly well at that. Certainly, if vote buying were a problem, then a ballot selfie prohibition could theoretically derail a vote-buying arrangement in one instance: a situation in which a voter is promised money to vote for a specific candidate, and the voter offers proof that he voted for that particular candidate by transmitting a photograph of his completed ballot.
But what about other types of vote-buying arrangements? In particular, what about the simplest vote-buying arrangement of all—absentee ballot vote buying—which can be accomplished at home outside the comparatively well-surveilled setting of a polling place? In the 2004 presidential election, for example, more than one in every seven votes were cast by absentee ballot. Additionally, by 2012, twenty-seven states and Washington, D.C. offered voters “no-excuse” absentee voting, which allows any registered voter to vote by absentee ballot for any reason. Plainly, ballot selfie bans do nothing to curb this far simpler type of vote-buying fraud, which could theoretically affect several million ballots cast each year.
Further, what about the other possible vote-buying arrangements that take place inside a polling place? Professor Hasen correctly notes that in the exceedingly rare instances in which vote buying does occur today, “it usually requires the cooperation of someone in the polling place to verify how people voted.” Ballot selfie bans, however, do nothing to root out such internal corruption. Similarly, in virtually every jurisdiction in the United States, a voter who claims to be disabled is permitted to bring another person into the voting booth with him to help him cast his ballot. In fact, in Professor Hasen’s home state of California, not one but “two persons selected by the voter” can join such a voter inside the voting booth. Certainly, this kind of accommodation can be exploited by eager vote buyers just as easily as a vote-buying arrangement involving ballot selfies. However, a ballot selfie ban accomplishes nothing in the way of preventing this potential avenue for vote buying, either.
Given these problems, ballot selfie prohibitions are substantially under-inclusive because they selectively preclude only one theoretical method of vote buying—ballot photography—while doing nothing to prevent far simpler methods of accomplishing the same result. “Laws that are underinclusive,” however, “cannot be narrowly tailored” in keeping with strict scrutiny. As the Supreme Court has explained, “a law cannot be regarded as protecting an interest of the highest order, and thus as justifying a restriction on [protected] speech, when it leaves appreciable damage to [a] supposedly vital interest unprohibited.” Thus, ballot selfie prohibitions are insufficiently protective to satisfy strict scrutiny’s narrow tailoring requirement.
Conversely, ballot selfie bans fail the constitutional “narrow tailoring” requirement in another way: they are substantially over-inclusive in what they prohibit. Specifically, prohibitions on ballot photography and ballot photo sharing do not simply prevent fraudsters from transmitting photographed ballots for the purpose of vote buying. Instead, they prohibit all people from distributing photographed ballots for any purpose at all—regardless of whether or not they have a legitimate basis for doing so. Thus, ballot selfie bans similarly fail to satisfy strict scrutiny because they sweep within their net a great deal of protected political speech that is completely unrelated to the government’s interest in preventing vote buying.
Without a doubt, by “ensnar[ing] a large number of voters wishing to make a political point or expressing their pride in voting or recording the moment for some innocuous personal reason,” categorical prohibitions on sharing images of one’s ballot cast an unnecessarily wide net over protected and unprotected speech alike, rendering such laws fatal to any serious claim of “narrow tailoring.” As Gilles Bissonnette—the legal director of the ACLU of New Hampshire—has astutely quipped: “The best way to combat vote buying and coercion is to investigate and prosecute cases of vote buying and coercion.”
Furthermore, although election-related selfies have been criticized as a valueless “generational” phenomenon, the reality is that being able to distribute a photograph of one’s ballot truly does carry tremendous social value for several reasons. First, ballot selfies frequently represent personal endorsements of candidates for public office, and political endorsements are indisputably protected by the First Amendment. As such—personal vanity aside—the ballot selfie represents a treasured and quintessentially protected form of core political speech, thus falling within a realm in which the Supreme Court has held repeatedly that “the First Amendment has its fullest and most urgent application,” and where “the importance of First Amendment protections is at its zenith.”
Furthermore, as any number of individuals have observed, “a lot of people like to take pictures of their ballot because they’re excited about voting and participating in the process.” For example, as ACLU of Indiana Legal Director Ken Falk has explained, “[t]aking a picture of one’s ballot and sharing it with family and friends is an expression of pride and enthusiasm about voting.” Consequently, proudly sharing one’s voting experience and political preferences on social media represents “a positive sign of civic engagement” that should be welcomed and perhaps even encouraged in a political climate in which many voters—especially young voters—never vote at all. Similarly, as in the case of Mr. Langlois and others, a ballot selfie can also serve precisely the opposite purpose as well—offering a public and equally valuable expression of disgust with one’s voting options.
Finally, freedom of expression notwithstanding, being able to photograph one’s ballot can also provide valuable information about malfunctioning ballot machines. For example, in 2012, a Perry County, Pennsylvania voter used his smartphone to record a widely disseminated video that demonstrated his voting machine switching his vote from Barack Obama to Mitt Romney. After election officials reviewed the voter’s video, the machine was promptly taken out of service. Notably, a non-trivial number of similar “vote-flipping” claims have also been alleged in recent years, although such claims can often be viewed as self-serving, and actual proof of malfunctioning ballot machines has been rare. Consequently, photographic evidence of a malfunctioning electronic ballot machine offers helpful, substantive proof that a machine is in fact malfunctioning, and it facilitates prompt replacement in the event that one is. In this fashion, giving the public the right to document and expose such problems when they occur plainly enhances voter confidence in the integrity of the election process, rather than diminishing it.
In sum, even if vote buying were commonplace, ballot selfie bans have two substantial problems that render them unable to satisfy the Constitution’s “narrow tailoring” requirement. Specifically, they are simultaneously both under-inclusive and over-inclusive, and thus, they are insufficiently narrowly tailored to achieve the government’s stated goal of preventing vote buying. In particular, ballot selfie bans are substantially under-inclusive in their protection because they do nothing to guard against far simpler means of vote buying, such as absentee vote buying. Simultaneously, ballot selfie bans are substantially over-inclusive in what they prohibit because they criminalize a vast amount of protected speech that is unnecessary to achieve the government’s purportedly compelling goal. For both of these reasons, blanket prohibitions on ballot selfies cannot satisfy strict scrutiny, and they violate the First Amendment as a result.
II. Vote buying Is Statistically Non-Existent
Stepping out of the world in which vote buying was assumed to be a problem that needed fixing, let us return to the real world: one in which vote buying occurs with infinitesimal frequency, rendering the “compelling” nature of the government’s interest in enacting sweeping laws to guard against it subject to considerable doubt. In addition to proving that a law abridging protected speech is narrowly tailored, in order to satisfy strict scrutiny in the context of the First Amendment, the government must also “specifically identify an ‘actual problem’ in need of solving, and the curtailment of free speech must be actually necessary to the solution.” Moreover, “[t]his burden is not satisfied by mere speculation or conjecture; rather, a governmental body seeking to sustain a restriction on . . . speech must demonstrate that the harms it recites are real and that its restriction will in fact alleviate them to a material degree.” With respect to vote buying, however, no “actual problem” exists, and for the reasons explained in Section III, prohibiting ballot selfies would not solve the problem even if one did.
To be sure, as Professor Hasen argues and as Judge Barbadoro acknowledged, “[t]here is no doubt that vote buying and voter coercion were at one time significant problems in the United States.” Voters have not always voted by secret ballot throughout American history, which once made vote buying and voter coercion commonplace. Indeed, around the turn of the twentieth century, political parties printed their own ballots in “flamboyant colors, distinctive designs, and emblems so that they could be recognized at a distance,” and they paraded people to the polls en masse in an effort to provide a coercive, conspicuous and public spectacle of voters’ ballot selections. Further, prior to the implementation of the secret ballot, approaching a polling place was often akin to “entering an open auction,” with rival political parties competing vigorously to bribe voters to accept their pre-printed ballots, and then paying them for their vote immediately after they had cast it. Practices like this resulted in an estimated incidence of vote buying of sixteen percent in Connecticut in 1892 and twenty percent in New York City in 1887, which represents an almost unimaginably high degree of voter fraud. Notably, as Professor Hasen explains, voter turnout also “went down as each state adopted the secret ballot,” lending additional support to the argument that vote buying and voter coercion were to blame.
While there are crucial differences between the historical traditions that took place before the secret ballot was adopted and the voluntarily exposed ballot made possible today by personal ballot photography and social media photo sharing, there is also another historical development to consider: today, vote buying effectively does not exist anywhere in the United States—even in the many jurisdictions where it is easy to accomplish. As noted in Section I, for example, at present, a majority of states offer “no-excuse absentee voting,” meaning that anyone who wishes to do so is eligible to vote by mail. Despite the widespread availability of absentee voting in most U.S. jurisdictions, however, Oregon and Washington stand out in this regard. In these two states, voters are not only offered the option of absentee voting; instead, all Oregon and Washington elections are conducted by mail. Thus, if vote buying were a problem, then one would reasonably expect to find it in the Pacific Northwest.
Lowering expectations, Professor Hasen acknowledges that “[t]hese days in the United States, vote buying is relatively rare.” But exactly how rare? For example, out of the more than 21.6 million combined absentee ballots that have been cast in Oregon primary and general elections since the state adopted universal vote-by-mail balloting in 1998, what proportion have given rise to concerns about vote buying? One percent? One-tenth of one percent? One one-hundredth of one percent?
According to the Oregon Secretary of State’s office, the answer is one. Not one percent, mind you. Instead, out of nearly twenty-two million absentee ballots that Oregon voters have cast in the past two decades, there has been only a single demonstrated instance of attempted vote buying. For the sake of comparison, nearly three-dozen UFO sightings have occurred during the same time period, and Owlcapone—Salem, Oregon’s merciless, hat-stealing attack owl—has terrorized at least four innocent joggers during the past year alone. Thus, as far as Oregon is concerned, “vote buying is relatively rare” indeed.
Oregon is not alone, however. In response to an identical public records request about vote buying in Washington state, for example, the Washington Secretary of State’s Office indicated that it could find no record of any vote-buying prosecution “in the last 15 years or more,” during which time more than twenty-five million absentee ballots were cast. And lest one be tempted to presume that Oregon and Washington are simply delinquent in investigating vote-buying schemes, it is worth noting that their experiences are in no way anomalous. According to Judge Barbadoro’s opinion, for example, New Hampshire’s experience reflects the very same reality, as “there have been no vote buying prosecutions and no complaints of vote buying in the state since at least 1976.” Further, with respect to the specific concern of vote buying facilitated by ballot selfies, Indiana’s Attorney General was “unable to point to a single instance in which digital photography facilitated vote buying or selling, despite the fact that . . . approximately two-thirds of Americans own and/or use a smartphone with a camera and approximately three-quarters of Americans participate in some type of social media website.” The dearth of vote-buying prosecutions anywhere else in the United States further confirms this pattern. Thus, despite the occasional “isolated and anachronistic” instance of vote buying in one jurisdiction or another, statistically speaking, vote buying is non-existent.
III. Ballot Selfies Generally Do not Reflect a Completed Ballot
Finally, setting aside the many concerns expressed above, suppose that vote buying were both a serious problem and that prohibiting voters from photographing their completed ballots would be an effective way to address it. What then? Would the nation’s many ballot selfie bans finally be able to survive strict scrutiny?
Unfortunately for proponents of such measures, the answer is still no, and for a simple reason: as a matter of practical reality, ballot selfies do not actually reflect a completed ballot. Thus, rather than equipping a voter with “the tools to sell [his or her] vote” or making it possible for a voter to “get forced to vote one way or another,” as Professor Hasen contends, in practice, permitting ballot selfies does no such thing.
In most jurisdictions, regardless of whether a ballot is marked electronically or by hand, a voter’s “completed” ballot is still subject to alteration before it is formally cast. In jurisdictions that use electronic ballots, for example, the electronic balloting software generally permits a voter to return to any portion of his or her ballot and to change any prior selection before the ballot is submitted. Similarly, in paper ballot jurisdictions, a voter who snaps a selfie of a marked ballot can generally request a new ballot or intentionally “spoil” the ballot—for example, by selecting two people for the same office—in order to be entitled to a new one.
Given this reality, anybody who is interested in vote buying gains nothing in the way of actual verification by reviewing a ballot selfie—which is self-evidently a prerequisite to the effectiveness of such a scheme. It is true that a ballot selfie reflects that a voter marked her ballot in a certain way at the moment the photo was taken. But was the ballot altered thereafter? Only the voter herself can answer this question, rendering any such vote-buying strategy valueless.
To illustrate this problem in practice, below are a pair of ballot selfies that I snapped while voting in the September 10, 2015, Metropolitan runoff election held in Nashville, Tennessee. Plainly, based on these photos, attempting to figure out whom I supported for Mayor is not possible. The first photo reflects a vote for mayoral candidate David Fox. The second photo reflects a vote for mayoral candidate Megan Barry. The third photo reflects the final page that appears before one’s ballot is actually counted, and it permits voters to select a “Return to ballot” button and to change a selection in any contest before the ballot is formally cast. Notably, this final page does not contain any information about my ballot selections whatsoever.
The same is true of photographed paper ballots. Like snapshots of an electronic ballot, a photograph of a paper ballot reflects that a voter marked her ballot in a certain way only at the moment the photo was taken—rather than documenting a vote that was actually cast. If a voter wants to switch one of her votes or makes a mistake, however, then the voter may simply request a new ballot and begin again. For example, sticking with Tennessee, the applicable section of the state election code provides that “[i]f any voter spoils a paper ballot, the voter may obtain others, one (1) at a time, not exceeding three (3) in all, upon returning each spoiled one.” Thus, with respect to paper ballots, too, a ballot selfie does not provide a would-be vote buyer any assurance that a voter has actually cast his or her ballot in a particular way, rendering any such scheme ineffectual.
In sum, laws that prohibit ballot photography and distribution represent content-based restrictions on core political speech, and states’ widespread efforts to prohibit ballot selfies cannot withstand constitutional scrutiny. The problems with such prohibitions are multifaceted. First, ballot selfie prohibitions are not narrowly tailored because they unnecessarily restrict a substantial amount of protected speech while simultaneously doing nothing to prevent far simpler forms of vote buying. Second, the government’s purportedly compelling need to prohibit personal ballot photography in order to prevent vote buying is highly questionable in light of the fact that vote buying is statistically non-existent even in jurisdictions where it is easy to accomplish. Third, because voters generally have the ability to change their votes after photographing their ballot selections, ballot selfies represent a useless tool for promoting vote buying anyway—rendering the entire premise behind such laws baseless.
. Rideout v. Gardner, No. 14-CV-489-PB, 2015 WL 4743731, at *7 (D.N.H. Aug. 11, 2015).
. Full 2014 primary election results in New Hampshire, WMUR Manchester (Sept. 9, 2014, 12:00 AM), www.wmur.com/politics/full-primary-election-results-in-new-hampshire/27883790; Senate: New Hampshire (Shaheen vs Brown), CNN Politics (Nov. 5, 2014, 3:28 PM), www.cnn.com/election/2014/results/state/NH/senate.
. N.H. Rev. Stat. Ann. § 659:35 (2014); Rideout, 2015 WL 4743731, at *7.
. Illegal voter “ballot selfies” flood social media (CBS This Morning broadcast Nov. 5, 2014), http://www.cbsnews.com/videos/illegal-voter-ballot-selfies-flood-social-media. Notably, however, some states have taken the opposite approach. For example, ballot selfies are expressly permitted in Arizona and Utah. See Ariz. Rev. Stat. Ann. § 16-1018(4) (Supp. 2015) (West) (“A voter who makes available an image of the voter’s own ballot by posting on the internet or in some other electronic medium is deemed to have consented to retransmittal of that image and that retransmittal does not constitute a violation of this section.”); Utah Code Ann. § 20A-3-504(3) (West Supp. 2015) (“Subsection (1) does not prohibit an individual from transferring a photograph of the individual’s own ballot in a manner that allows the photograph to be viewed by the individual or another.”).
. Erik Eckholm, Selfies in Voting Booths Raise Legal Questions on Speech and Secrecy, N. Y. Times (Aug. 24, 2015), www.nytimes.com/2015/08/25/us/selfies-in-voting-booths-raise-legal-questions-on-speech-and-secrecy.html?_r=0.
. Kristen Carosa, Lawsuit challenges law forbidding posting pics of ballots, WMUR Manchester (Oct. 31, 2014, 10:52 PM), http://www.wmur.com/politics/lawsuit-challenges-law-forbidding-posting-pics-of-ballots/29466942
. Rideout, 2015 WL 4743731, at *8.
. Id. at *1.
. Order Granting Preliminary Injunction, Ind. Civil Liberties Union Found. v. Ind. Sec’y of State, et al., No. 1:15-cv-01356-SEB-DML (S.D. Ind. Oct. 19, 2015), http://www.aclu-in.org/images/newsReleases/DECISION_1_15-cv-1356-SEB-DML_ICLU_v_IN_SOS_10-19-2015.pdf.
. Richard L. Hasen, Why the Selfie is a Threat to Democracy, Reuters (Aug. 18, 2015, 8:47 PM), blogs.reuters.com/great-debate/2015/08/17/why-the-selfie-is-a-threat-to-democracy/.
. Eckholm, supra note 7; see also Donna Ballman, Why Ballot Selfies Are A Terrible Idea For Workers, Lexis Nexis Legal Newsroom (Aug. 31. 2015, 11:33 AM), http://www.lexisnexis.com/legalnewsroom/labor-employment/b/labor-employment-top-blogs/archive/2015/08/31/why-ballot-selfies-are-a-terrible-idea-for-workers.aspx.
. Bolger v. Youngs Drug Products Corp., 463 U.S. 60, 65 (1983) (quoting Police Department v. Mosley, 408 U.S. 92, 95 (1972)).
. See, e.g., Reed v. Town of Gilbert, Ariz., 135 S. Ct. 2218, 2227 (2015) (“Government regulation of speech is content based if a law applies to particular speech because of the topic discussed or the idea or message expressed.”); see also Rideout v. Gardner, No. 14-CV-489-PB, 2015 WL 4743731, at *9 (D.N.H. Aug. 11, 2015) (“the law under review is content based on its face because it restricts speech on the basis of its subject matter. The only digital or photographic images that are barred . . . are images of marked ballots that are intended to disclose how a voter has voted. Images of unmarked ballots and facsimile ballots may be shared with others without restriction. In fact, the law does not restrict any person from sharing any other kinds of images with anyone.”); Order Granting Preliminary Injunction, supra note 11, at 6 (holding that, because Indiana’s ballot selfie prohibition “clearly defines the regulated expression according to its subject matter and its purpose, it is properly construed as being content based ‘on its face.’”).
. See Brown v. Entm’t Merchants Ass’n, 131 S. Ct. 2729, 2738 (2011).
. See, e.g., Sable Commc’ns of California, Inc. v. F.C.C., 492 U.S. 115, 126 (1989) (“The Government may . . . regulate the content of constitutionally protected speech in order to promote a compelling interest if it chooses the least restrictive means to further the articulated interest.”); Brown, 131 S. Ct. at 2738 (“Because the Act imposes a restriction on the content of protected speech, it is invalid unless California can demonstrate that it passes strict scrutiny—that is, unless it is justified by a compelling government interest and is narrowly drawn to serve that interest.”).
. Meyer v. Grant, 486 U.S. 414, 425 (1988).
. See Brown, 131 S. Ct. at 2738.
. See id.
. See id.
. See generally, Rideout v. Gardner, No. 14-CV-489-PB, 2015 WL 4743731, at *4–6 (D.N.H. Aug. 11, 2015) (providing an overview on vote buying throughout American history and how it has become close to obsolete in modern voting).
. See e.g., Me. Rev. Stat. tit. 21-A, § 693 (1964); Ark. Code Ann. § 7-5-604 (2011); W. Va. Code § 3-4A-9 (2015).
. See Rideout, 2015 WL 4743731, at *4–5.
. See Hasen, supra note 10.
. Sarah Jane Capper & Michael Ciaglo, The Real Vote-Fraud Opportunity Has Arrived: Casting Your Ballot by Mail, NBC News (Sept. 25, 2012, 7:13 AM) http://investigations.nbcnews.com/_news/2012/09/25/14092264-the-real-vote-fraud-opportunity-has-arrived-casting-your-ballot-by-mail (“[V]ote buying and bribery could occur more easily with mail voting and absentee voting. . . . A person who bribes mail voters could watch as they mark ballots or even mark ballots for them.”).
. John C. Fortier, Absentee and Early Voting: Trends, Promises, and Perils, Am. Enter. Inst. 22–23 (2006), http://www.aei.org/wp-content/uploads/2014/06/-absentee-and-early-voting_155531845547.pdf (“Over 123 million votes were cast in the 2004 general election. Of those, nearly 18 million, or 14.5 percent, were cast absentee.”).
. Capper & Ciaglo, supra note 26.
. Hasen, supra note 10.
. See e.g., Tex. Elec. Ann. § 64.031 (West 2015); Tenn. Code Ann. § 2-7-116 (2014); Mont. Code Ann. § 13-13-119 (2015).
. Cal. Elec. Code § 14282 (West 2015) (“When a voter declares under oath, administered by any member of the precinct board at the time the voter appears at the polling place to vote, that the voter is then unable to mark a ballot, the voter shall receive the assistance of not more than two persons selected by the voter, other than the voter’s employer, an agent of the voter’s employer, or an officer or agent of the union of which the voter is a member.”).
. Rosemond v. Markham, No. CV 13-42-GFVT, 2015 WL 5769091, at *9 (E.D. Ky. Sept. 30, 2015).
. Reed v. Town of Gilbert, Ariz., 135 S. Ct. 2218, 2232 (2015) (quoting Republican Party of Minn. v. White, 536 U.S. 765, 780 (2002)) (internal quotation marks omitted).
. See Rideout v. Gardner, No. 14-CV-489-PB, 2015 WL 4743731, at *14 (D.N.H. Aug. 11, 2015).
. See e.g., N.H. Rev. Stat. Ann. § 659:35 (2014).
. See Rideout, 2015 WL 4743731 at *15.
. Order Granting Preliminary Injunction, supra note 11, at 14.
. Id.; see also Rideout, 2015 WL 4743731 at *14 (“In the present case, the state has an obviously less restrictive way to address any concern that images of completed ballots will be used to facilitate vote buying and voter coercion: it can simply make it unlawful to use an image of a completed ballot in connection with vote buying and voter coercion schemes.”).
. Eckholm, supra note 7.
. See generally Eu v. San Fransisco Cnty. Democratic Cent. Comm., 489 U.S. 214, 223 (1989).
. Id. at 223.
. Id. (quotation omitted).
. Meyer, 486 U.S. at 425 (quotation omitted).
. Zach Pluhacek, No Ballot ‘Selfies’ in Nebraska, Secretary of State Says, Journal Star (Sept. 4, 2015), http://journalstar.com/news/state-and-regional/nebraska/no-ballot-selfies-in-nebraska-secretary-of-state-says/article_a6a74c18-5f98-5794-b852-e1d21de53b4b.html?mobile_touch=true (“A lot of people like to take pictures of their ballot because they’re excited about voting and participating in the process.”).
. ACLU of Indiana Challenges State Law Prohibiting Ballot “Selfies”, Am. Civil Liberties Union (Aug. 27, 2015), https://www.aclu.org/news/aclu-indiana-challenges-state-law-prohibiting-ballot-selfies; see also Pluhacek, supra note 36. See also http://www.indystar.com/story/opinion/2015/10/26/downs-ballot-selfies-unlikely-corrupt-democracy/74631244/ (“Voters who want to take and share photos of their ballots are showing support for candidates and trying to influence people in their social media networks.”).
. Jack Morse, That ‘Ballot Selfie’ You Just Posted? Yeah, That’s Illegal., SFIST (Nov. 3, 2015, 1:05 PM), http://sfist.com/2015/11/03/that_ballot_selfie_you_just_posted.php (“Assemblyman Marc Levine of San Rafael believes that ‘ballot selfies,’ as they’re (I guess?) called, are a positive sign of civic engagement and should be encouraged.”).
. See Youth Voting, Circle, http://www.civicyouth.org/quick-facts/youth-voting/ (last visited Oct. 18, 2015); see also Jack McElroy, Stand up for freedom; take a ballot selfie, Knoxville News Sentinel (Sept. 13, 2015, 3:00 AM), http://www.knoxnews.com/opinion/columnists/jack-mcelroy-stand-up-for-freedom-take-a-ballot-selfie_77697777 (“It is hard to see what the actual or imminent problem is in Tennessee, either, when a first-time teenage voter, or a newly sworn American, or a get-out-the-vote campaigner or any other citizen proudly and publicly documents the fulfillment of their civic duty. What we need these days are more ballot selfies, a lot more—not a ban to stomp out a problem that doesn’t exist.”).
. Rideout v. Gardner, No. 14-CV-489-PB, 2015 WL 4743731, at *7 (D.N.H. Aug. 11, 2015).
. See, e.g., Richard Locker, Bill Nearing Approval Would Ban Cellphone Pictures, Video at Polling Places, The Commercial Appeal (Apr. 7, 2015, 7:55PM), http://www.commercialappeal.com/news/state/bill-nearing-approval-would-ban-cellphone-pictures-video-at-polling-places_04827022 (quoting a state Representative as saying, “In Davidson County two years ago after the election, the state election coordinator put out a 20-page report on all the improprieties that happened in that election and a lot of that we were able to document using phones and video.”).
. 2012 Voting Machines Alerting Votes, YouTube (Nov. 6, 2012), https://www.youtube.com/watch?v=QdpGd74DrBM.
. Jaikumar Vijayan, E-voting machine swaps Obama vote for Romney; taken offline, Computerworld (Nov. 6, 2012, 3:30PM), http://www.computerworld.com/article/2493314/government-it/e-voting-machine-swaps-obama-vote-for-romney–taken-offline.html.
. Voting Machine Switching GOP Votes to Democrat in Illinois, YouTube (Oct. 23, 2014) https://www.youtube.com/watch?v=8Pi0whC2nK4; see also Alanna Autler, Voters Report Issues at Maury County Polls, WSMV-TV Nashville (Nov. 6, 2014, 9:35 PM), http://www.wsmv.com/story/26880035/voters-report-issues-at-maury-county-polls; see also Nick Bechtel, Voting Machine Swaps Romney for Obama, Marion Star (Oct. 31, 2012, 6:51 AM), http://www.marionstar.com/article/20121031/NEWS03/310310009.
. See, e.g., Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520, 578 (1993) (“A State may no more create an underinclusive statute, one that fails truly to promote its purported compelling interest, than it may create an overinclusive statute, one that encompasses more protected conduct than necessary to achieve its goal.”).
. Rideout v. Gardner, No. 14-CV-489-PB, 2015 WL 4743731, at *14 (D.N.H. Aug. 11, 2015).
. Proponents of ballot selfie bans are likely to take issue with this assertion by contending that such bans are necessary to further the government’s interest in preventing even the possibility of vote buying or voter coercion. As noted in Section III, however, the bans do not do that.
. Brown v. Entm’t Merchants Ass’n, 131 S. Ct. 2729, 2738 (2011) (quoting United States v. Playboy Entm’t Grp., Inc., 529 U.S. 803, 822–23 (2000); see also United States v. Alvarez, 132 S. Ct. 2537, 2549 (2012) (“The First Amendment requires that the Government’s chosen restriction on the speech at issue be ‘actually necessary’ to achieve its interest.”); Consol. Edison Co. of N. Y. v. Pub. Serv. Comm’n of N. Y., 447 U.S. 530, 543 (1980) (“Mere speculation of harm does not constitute a compelling state interest.”).
 Edenfield v. Fane, 507 U.S. 761, 771 (1993).
. Hasen, supra note 10 (“These days in the United States, vote buying is relatively rare.”).
. Rideout, 2015 WL 4743731, at *4 (citing Doe v. Reed, 561 U.S. 186, 226 (2010) (Scalia, J., concurring) (citing Burson v. Freeman, 504 U.S. 191, 202 (1992) (plurality opinion))); see also Susan C. Stokes et al., Brokers, Voters, and Clientelism: The Puzzle of Distributive Politics 200 (2013); Richard Hasen, Vote Buying, 88 Cal. L.Rev. 1323, 1327 (2000); Jill Lepore, Rock, Paper, Scissors: How We Used To Vote, New Yorker (Oct. 13, 2008), http:// www.newyorker.com/magazine/ 2008/10/13/rock-paper-scissors.
. See generally Burson, 504 U.S. at 200–01 (“Wishing to gain influence, political parties began to produce their own ballots for voters. These ballots were often printed with flamboyant colors, distinctive designs, and emblems so that they could be recognized at a distance. State attempts to standardize the ballots were easily thwarted—the vote buyer could simply place a ballot in the hands of the bribed voter and watch until he placed it in the polling box. Thus, the evils associated with the earlier viva voce system reinfected the election process; the failure of the law to secure secrecy opened the door to bribery and intimidation.”).
. Id. at 200.
. Id. at 202.
. Rideout v. Gardner, No. 14-CV-489-PB, 2015 WL 4743731, at *4 (D.N.H. Aug. 11, 2015) (citing L.E. Fredman, The Australian Ballot: The Story of an American Reform 22 (1968)).
. Id. (citing Stokes et al., supra note 53, at 227).
. Hasen, supra note 10.
. Capper & Ciaglo, supra note 25.
. See Hasen, supra note 10.
. Voter Turnout History for General Elections, Or. Sec’y of State, http://sos.oregon.gov/elections/Documents/Voter_Turnout_History_General_Election.pdf; Voter Turnout History for Primary Elections, Or. Sec’y of State, http://sos.oregon.gov/elections/Documents/Voter-Turnout-History-Primary.pdf.
. E-mail from Tony Green, Commc’ns Dir., Or. Sec’y of State, to author (Aug. 31, 2015, 6:27 PM) (on file with author).
. Election Statistics, Or. Sec’y of State, http://sos.oregon.gov/elections/Pages/electionsstatistics.aspx.Of note, this number does not even include local elections that are held in May and November of odd-numbered years or Special elections. Id.
. E-mail from Tony Green, supra note 71.
. List of Reported UFO Citings, Wikipedia, https://en.wikipedia.org/wiki/List_of_reported_UFO_sightings (last visited Sept. 1, 2015, at 4:14 PM).
. Hat-Stealing Attack Owl In Oregon Officially Named ‘Owlcapone’, The Huffington Post (Feb. 13, 2015, 9:59AM), http://www.huffingtonpost.com/2015/02/13/owl-attacks-owlcapone-oregon-name_n_6677410.html.
. Hasen, supra note 12.
. E-mail from Brenda Galarza, Pub. Records Officer, Wash. Sec’y of State, to author (Sept. 2, 2015, 11:32 AM) (on file with author); e-mail and attached Excel spreadsheet from Brenda Galarza, Pub. Records Officer, Wash. Sec’y of State, to author (Sept. 3, 2015, 3:29 PM) (on file with author).
. Rideout v. Gardner, No. 14-CV-489-PB, 2015 WL 4743731, at *12 (D.N.H. Aug. 11, 2015).
. Order Granting Preliminary Injunction, supra note 11, at 13–14..
. See id. at *6.
. Id. (citing Stokes et al., supra note 53, at 231); see also United States v. Thomas, 510 F.3d 714, 717 (7th Cir. 2007); United States v. Shatley, 448 F.3d 264, 265 (4th Cir. 2006); United States v. Johnson, No. 5:11–cr–143, 2012 WL 3610254, at *1 (E. D. Ky. Aug. 21, 2012).
. Hasen, supra note 10.
. See, e.g., Ark. Code Ann. § 7-5-604(4) (2011); 10 Ill. Comp. Stat. 5 / 17-16 (2010); Me. Rev. Stat. tit. 21-A, § 693 (1964); W. Va. Code § 3-4A-9(5) (2015).
. See, e.g., Ark. Code Ann. § 7-5-604(4); W. Va. Code § 3-4A-9(5).
. See, e.g., 10 Ill. Comp. Stat. 5 / 17-16; Me. Rev. Stat. tit. 21-A, § 693.
 Andrew Downs, Ballot selfies unlikely to corrupt democracy, IndyStar (Oct. 26, 2015, 1:13 PM), http://www.indystar.com/story/opinion/2015/10/26/downs-ballot-selfies-unlikely-corrupt-democracy/74631244/ (“For vote buying purposes, a photo of a ballot may seem better than a distinguishing mark on a paper ballot, but it is not. A voter could mark the ballot, take the picture and then change the vote. I suspect that there is limited honor among vote buyers.”).
. See, e.g., 10 Ill. Comp. Stat. 5 / 17-16; Me. Rev. Stat. tit. 21-A, § 693.
. Tenn. Code Ann. § 2-7-120 (2014).