I. Introduction

“You can print a lethal device. It’s kind of scary, but that’s what we’re aiming to show.”[1]

As gun violence continues to plague the United States, government bodies attempt to legislate and regulate gun technology. These efforts existed long before the first 3D-printed gun. In 1976, Congress passed the Arms Export Control Act (AECA), placing the importing and exporting of defense articles under the President’s authority.[2] At the time, defense articles included any “technical data” relating to weapons, such as the manufacturing blueprints of a firearm.[3] After decades of innovation and improvements, “technical data” in the form of 3D computer-aided design (CAD) files may presently be downloaded by anyone.[4] Advances in 3D printing now allow a person to print a fully functional plastic weapon in a few hours in the comfort of their home.[5]

In 2013, Cody Wilson of Defense Distributed fired the first gun made from nearly all 3D-printed parts and posted the design online, where it was downloaded over “a hundred thousand times” in two days.[6] The State Department requested the model be taken offline, asserting that Wilson needed a license under the International Traffic in Arms Regulation (ITAR).[7] Defense Distributed filed for an injunction, arguing that the State Department infringed on its First and Second Amendment rights.[8] Defense Distributed was denied a preliminary injunction in both the district court and the Fifth Circuit, with the Supreme Court denying certiorari.[9] However, the case settled—with the State Department waiving the prior restraint against Defense Distributed, allowing them to publish the files freely.[10] Immediately, U.S. District Court Judge Robert Lasnik extended a nationwide injunction to block the online distribution of 3D-printed gun files.[11] Other responses to the settlement included a lawsuit to prevent the release of the files and a Presidential tweet.[12]

Unfortunately, the Supreme Court did not grant certiorari to address whether the ITAR qualified as a prior restraint.[13] This Comment answers the inquiry in the affirmative. Furthermore, this Comment aims to identify the weaknesses of the ITAR and of other gun laws so legislatures and agencies may write stronger, enforceable laws—laws that can withstand a First Amendment challenge. First, Part II provides background information about 3D printing, First Amendment principles, and current gun laws. Then, Part III reviews the procedural history of Defense Distributed. In Part IV, this Comment discusses how Defense Distributed would have been heard on First Amendment grounds. Part V of this Comment identifies a need to resolve the constitutional weaknesses in the current law. It also critiques other potential avenues that address the imminent threat of 3D-printed guns.

II. The Backdrop to This Gun Show

“But how can you exact some kind of legal regime that enforces [responsible firearm ownership] without infringing on the rights of countless people[?]”[14]

This Part provides relevant background information for an understanding of the current landscape. First, it covers important First Amendment caselaw and frameworks. Then, it provides a brief overview of 3D printing. Finally, it includes a discussion of current gun laws, foreshadowing the lack of authority over 3D-printed guns.

A. First Amendment

The First Amendment protects free speech, including conventional speech, words not yet spoken, and expressive conduct.[15] While the traditional scope of speech has expanded, courts have carved out a few exceptions to the default rule of applying strict scrutiny to actions that limit free speech.[16] Exceptions must meet a high bar: not only is a strong compelling government interest required, but the restriction must also be narrowly tailored.[17] Below is an overview of how courts typically analyze content-based restrictions, prior restraints, conduct as speech, and (more recently) information and digital files as speech. With this First Amendment caselaw, a court would almost certainly find the ITAR unconstitutional.

1. Content-Based Restrictions

Content-based government restrictions are “subject to strict scrutiny regardless of the government’s benign motive, content-neutral justification, or lack of ‘animus toward the ideas contained’ in the regulated speech.”[18] Although restrictions do not completely prohibit certain kinds of speech, but rather just place a burden on the speech, restrictions based on content still require a strict scrutiny analysis.[19] In contrast, content-neutral restrictions appear neutral on their face but ultimately affect speech when the restriction is enforced.[20] This kind of restriction only needs to survive a time-place-manner analysis, a less rigorous standard of scrutiny.[21]

2. Prior Restraints

Occasionally, the government may try to license speech in advance. Formally, this is known as a prior restraint, an order that “forbid[s] certain communications when issued in advance of the time that such communications are to occur.”[22] Prior restraints are presumed to be unconstitutional and require a strict scrutiny analysis.[23] Prior restraints created by administrative agencies may receive even more scrutiny than court-created prior restraints due to the “agencies['] significant discretion to engage in broad censorship.”[24] The Supreme Court has found assertions of national security insufficient to justify a prior restraint.[25]

Furthermore, content-based prior restraints must meet the following “procedural safeguards designed to obviate the dangers of a censorship system”: (1) the censor bears the burden of proving that the expression is unprotected; (2) the restraint will only be imposed for a brief, specified period of time; and (3) prompt judicial determination must be available.[26]

3. Restricting Speech in the Form of Conduct

Courts will occasionally scrutinize speech as conduct or action. Expressive conduct that is “sufficiently imbued with elements of communication” receives First Amendment protection.[27] When conduct contains both speech and nonspeech elements, “a sufficiently important governmental interest in regulating the nonspeech element can justify incidental limitations on First Amendment freedoms.”[28]

The Supreme Court provided the O’Brien test to assist courts in analyzing restrictions on non-verbal conduct that incidentally impacts speech.[29] According to the test, a government regulation is sufficiently justified if the regulation: (1) is within the government’s power; (2) furthers an important or substantial government interest; (3) that interest is unrelated to the suppression of speech (content-neutral); and (4) prohibits no more speech than is needed to further that interest.[30] Justice Harlan, in his concurrence, added another element: whether the regulation leaves open ample alternative channels to express oneself.[31]

4. First Amendment Challenges to Digital Files.

Censorship is not new in the Internet Age. Courts have recognized that code, instructions, and blueprints are protected under the First Amendment.[32] Bernstein v. United States Department of State was one of the first cases to consider whether source code constitutes speech under the First Amendment, and the district court found that the First Amendment protected the cryptographic computer source code at issue.[33] In Commodity Futures Trading Commission v. Vartuli, the Second Circuit found that computer code or programs that require no human interception did not merit any First Amendment protection.[34] In Universal City Studios, Inc. v. Corley, the Second Circuit considered prior restraint and content-neutral regulation issues regarding an injunction that prohibited a website from posting DVD decryption code.[35] Refining Vartuli, Corley recognized the importance of programmers being able to “communicat[e] ideas to one another,”[36] and concluded that code and software programs “can merit First Amendment protection.”[37]

Courts have also scrutinized policies aimed at restricting internet user conduct. For instance, the court in Mainstream Loudoun v. Board of Trustees of the Loudoun County Library invalidated a county library’s internet policy aimed at preventing access to illegal pornography.[38] The district court found the policy to be over-restrictive because there was no evidence that the government tried to apply the policy in a “less restrictive way” than the written policy, which unconditionally filtered the internet access provided to patrons.[39] Furthermore, the prior restraint lacked procedural safeguards.[40]

B. 3D Printing

In its most basic definition, 3D printing is the process of taking a digital file and printing a three-dimensional form of the file using some sort of deposited material.[41] The digital file is a CAD file that contains the blueprint for the object to be printed.[42] A CAD file is not actually the type of file the printer reads;[43] instead, the CAD file is a source of information that needs to be formatted into a file type compatible with the printer.[44] The printer then deposits the material layer by layer, turning each pixel into the physical equivalent.[45] This process may take several hours before the whole object is completely printed.[46] The creation of CAD models is time-consuming and typically requires some expertise, but the internet has allowed people to easily share their files.

The creative freedom and convenience of 3D printing attracts all kinds of hobbyists, including gun enthusiasts. However, as Adam Thierer and Adam Marcus pointed out, "[t]he legal status of 3D-printed guns and the designs for those guns remains unclear . . . . "[47] But because the CAD files are digital files, they may, as mentioned above, merit First Amendment protection.

C. Current Laws/‌Regulations Pertaining to 3D Printing and Guns

At the time of this Comment, no U.S. law explicitly prohibits the dissemination or printing of CAD gun blueprints.[48] However, an overview of the current laws and regulations reveals the legal shortcomings of controlling 3D-printed guns.

First, the Undetectable Firearms Act of 1988 (UFA) prohibits the manufacture of guns that cannot be detected by metal detectors.[49] For the purposes of 3D printing, 3D-printed guns that do not meet the 3.7-ounce metal requirement are in violation of the UFA and would normally be subject to a $200 tax.[50] However, enforcement of the UFA is nearly impossible due to 3D printer use in private homes.[51] The UFA may have targeted the commercial aspect of plastic gun ownership, but it is insufficient to address private use.[52]

Next, the Arms Export Control Act of 1976 gives the President the authority to regulate the export of defense articles.[53] The International Traffic in Arms Regulations, a result of the President delegating authority to the Secretary of State, prevents the “transfer of certain arms and munitions to foreign nationals.”[54] No one can export an item deemed to be a defense article without being granted a license, including transmitting the defense article “out of the United States in any manner.”[55] The ITAR also covers the export of “technical data,” which is defined as information “which is required for the design, development, production, manufacture, assembly, . . . or modification of defense articles. This includes information in the form of blueprints . . . .”[56] A 3D-printed gun blueprint likely qualifies as a defense article or technical data under the ITAR, thus making this regulation relevant to those disseminating the blueprints online.[57]

Finally, the Gun Control Act of 1968 (GCA) regulates the interstate commerce of firearms, preventing transfers except among registered manufacturers, importers, or dealers.[58] 3D-printed guns might fall into the GCA’s scope under its “firearm” language, which is defined as “any weapon (including a starter gun) which will or is designed to or may readily be converted to expel a projectile by the action of an explosive.”[59] Therefore, under 18 U.S.C. § 922(a)⁠(1), it would be unlawful for anyone who is not licensed as an importer, manufacturer, or dealer to manufacture 3D guns. Unfortunately, home manufacturing and individual design make the government’s task of enforcing these laws against 3D-printed guns nearly impossible.

Furthermore, the Bureau of Alcohol, Tobacco, Firearms and Explosives allows for the unlicensed manufacture of firearms if made for personal use.[60] Because there is ambiguity and a lack of enforceability, the current laws are insufficient to explicitly govern 3D-printed guns.

III. Defense Distributed v. United States Department of State[61]

“This case concerns Defense Distributed’s desire to share all of its 3D printing and CNC milling files online, available without cost to anyone located anywhere in the world, free of regulatory restrictions.”[62]

A. Factual History

Defense Distributed is a non-profit organization created “for the purpose of promoting popular access to arms guaranteed by the [U.S.] Constitution.”[63] In early May 2013, Defense Distributed posted online, for free download by anyone in the world, the files for the Liberator pistol.[64] The State Department requested that Defense Distributed remove the files on the ground that sharing them violated certain laws, including the AECA.[65] The State Department contended: that the Liberator files potentially related to ITAR-controlled “technical data;” that posting ITAR-controlled files on the internet for foreign nationals to download constitutes an “export;” and that Defense Distributed must obtain prior approval from the State Department before exporting these files.[66] Defense Distributed sued the State Department, seeking to enjoin the Department from enforcing the ITAR regulations.[67]

B. District Court Proceedings

The district court denied Defense Distributed’s request for a preliminary injunction because Defense Distributed failed to show that the harm from failing to receive an injunction outweighed the damage the injunction would cause and that the injunction would not harm the public interest.[68] Furthermore, the district court concluded that Defense Distributed did not meet its burden to show a likelihood of success on the merits.[69]

Defense Distributed asserted that the State Department’s interpretation of the AECA violated free speech.[70] Although the district court acknowledged that “[t]he ITAR . . . unquestionably regulates speech concerning a specific topic,” it reasoned the ITAR was not directed toward the message being communicated by the technical data.[71] As a result, the district court found the ITAR to be content-neutral and “thus subject to intermediate scrutiny.”[72] Relying on the Ninth Circuit’s conclusions in a previous AECA case, the district court held that the ITAR advanced governmental interests which outweighed the burden on expression or public speech.[73] The district court did not find Defense Distributed’s harm to be grievous because there were other ways for plaintiffs to post the files; Defense Distributed could still apply for a permit under the ITAR or use domestic forms of communication, such as mail, so long as the medium “does not provide the ability to disseminate the information internationally.”[74] Finally, the district court did not agree with Defense Distributed’s contention that an internet ban would not prevent the distribution of files internationally, and thus the government interest would not actually be promoted.[75]

C. The Fifth Circuit’s Ruling

On appeal, the Fifth Circuit affirmed the district court’s denial of the preliminary injunction on procedural grounds while also declining to address the merits requirement.[76] Although it admitted that 3D gun files themselves were legal, the Fifth Circuit held that the public interest in national security outweighed Defense Distributed’s interest of “constitutional rights.”[77] The Fifth Circuit scolded Defense Distributed for “fail[ing] to give any weight to the public interest in national defense and national security,” and disagreed with the notion that the highest public interest is always in preventing violations to constitutional rights.[78] The Fifth Circuit described the public interest in national security and defense as “the State Department’s stated interest in preventing foreign nationals—including all manner of enemies of this country—from obtaining technical data on how to produce weapons.”[79]

Defense Distributed contested the importance of the stated security interest, considering the files were already available on third-party websites.[80] The Fifth Circuit recognized that the posted files might “remain online essentially forever.”[81] However, the Fifth Circuit emphasized that while Defense Distributed’s rights may be burdened in perpetuity, the national security public interest could be harmed forever.[82] Finally, the Fifth Circuit ordered the district court to address the merits on remand, once it had a fuller record.[83]

D. Judge Jones’ Dissent

As the sole dissenter, Fifth Circuit Judge Edith Jones expressed concerns about the majority’s disregard for free speech protections.[84] Judge Jones saw the ITAR licensing scheme as a content-based speech restriction and as a prior restraint deserving of strict scrutiny.[85]

First, Judge Jones found the ITAR to be purely content-based.[86] Although CAD files of other items, such as jewelry or model airplanes, may be posted on the internet, Defense Distributed’s files were prohibited because the files pertained to firearms.[87] The regulation of the speech here focuses on the “direct impact of speech on its audience” because the State Department wants to prevent a class of listeners—in this case, foreign nationals—from using the files to create firearms.[88] While Judge Jones agreed that the government had a compelling interest in regulating the exportation of arms and technical data, she thought the regulations failed strict scrutiny. According to her dissent, the ITAR was overinclusive, without enough alternative remedies for these speakers.[89]

E. The Unexpected Settlement

Defense Distributed petitioned for a writ of certiorari to the Supreme Court, but was denied.[90] However, before the district court could make a merits determination, Defense Distributed eventually “won” when it accepted a settlement offer from the Department of State.[91] The settlement waived the prior restraint against Defense Distributed, approving Defense Distributed’s files for public release.[92] The State Department agreed to temporarily modify the requirements of the Category I permits, granting any U.S. person the right to access, use, or reproduce the technical data.[93] Defense Distributed also successfully negotiated legal and filing fees.[94]

Shortly after the settlement, several gun safety organizations moved to intervene, seeking a temporary restraining order, but their motion was denied.[95] Additionally, eight states and the District of Columbia claimed the settlement was a violation of both the Administrative Procedure Act and the States’ Tenth Amendment right to regulate firearms.[96]

F. Judicial Orders

Immediately following the settlement, U.S. District Court Judge Robert Lasnik of Seattle issued an order to suspend the settlement and the public release of the CAD files, a win for the aforementioned group of states.[97] But even before the injunction order was issued, Defense Distributed violated the settlement terms by reposting the files before August 1, 2018.[98] Judge Lasnik’s order blocking the distribution of the files was later extended.[99] The State Department, in changing its opinion, determined before settlement that the Liberator files “did not pose a national security risk as they could be bought ‘in any store.’”[100] Additionally, the Department of Justice argued that the states were overreaching their powers by restricting a person from sharing information with others.[101]

Those unsatisfied with the settlement may still have hope because the settlement is not binding, leaving room for a Supreme Court challenge or agency regulation to restrict 3D-printed guns.[102] The next Part discusses how such a challenge (or one with similar circumstances) may have turned out, given the current Supreme Court’s conservative composition.

IV.First Amendment Issues and the Merits

“This opens a lot of doors . . . . Any advance in technology has posed these questions. And it’s not clear cut that this is just a good thing. But liberty and responsibility are scary.”[103]

The Fifth Circuit left for the district court many legal questions to be answered, including: (1) whether the 3D-printing files at issue are considered protected speech under the First Amendment; (2) the degree of scrutiny required for the ITAR; and (3) whether the ITAR regulations are an impermissible prior restraint.[104] Because the case settled, the district court was not afforded the opportunity to address the merits of the case. This Part will exclusively discuss the First Amendment issues by first establishing that the blueprint files and the creation of files should qualify as protected speech.[105] Then, it will analyze the merits of this case using the appropriate scrutiny demanded by caselaw and doctrine.

A. CAD Files as Speech

Although some scholars object, the electronic files themselves should constitute a form of speech, deserving First Amendment protection.[106] As mentioned above, some courts have analyzed code and digital files under the First Amendment. Scholarship has compared the 3D CAD files to blueprints, arguing that the CAD files do not constitute pure speech.[107] Although source files or object code intuitively do not appear to be speech, “information” can be “speech within the meaning of the First Amendment.”[108] The following caselaw supports the idea that the 3D CAD files deserve First Amendment protection.

In Brown v. Entertainment Merchants Association, an industry group challenged a California regulation that imposed restrictions on the sale of violent video games.[109] Justice Scalia applauded California for “declin[ing] to restrict Saturday morning cartoons, the sale of games rated for young children, or the distribution of pictures of guns.”[110] He then explained that California did not meet the strict scrutiny standard because it failed to show a “direct causal link between violent video games and harm to minors.”[111] Notably, the Court dismissed California’s argument that the State “need not produce such proof because the legislature can make a predictive judgment that such a link exists.”[112]

These pictures of firearms are fundamentally similar to 3D-printing blueprints or files.[113] The obvious analogy is that a 3D CAD file, when displayed in a program, conveys a picture at a certain angle to the user. Such files should be considered as expressive conduct because they communicate a designer’s expression, information, and ideas. But more importantly and more apparently, a user sees the message or expression in the form of a printed object.[114]

However, the message communicated by the 3D CAD file is not as explicit as pure speech’s message would be.[115] Thus, a court must check that the CAD file has sufficient “elements of communication,” in order for the file to receive First Amendment protection.[116] In Anderson v. City of Hermosa Beach, the Ninth Circuit analyzed both a tattoo and the tattooing process separately, despite how intertwined the message (in the form of a tattoo) was with the tattooing process.[117] The tattoo itself, as the output of the tattooing process, receives First Amendment protection because it represents a range of messages and functions.[118] The tattooing process is also afforded protection because the tattoo cannot be produced without the process and is “more akin to traditional modes of expression (like writing)” by putting words or a picture onto skin.[119]

More analogously, the Sixth Circuit held in Junger v. Daley that the First Amendment can protect source code because it “is the preferred method of communication among computer programmers . . . [and] is an expressive means for the exchange of information and ideas about computer programming.”[120] Software thus represents a programmer’s thoughts, which is readable by other programmers and computers. Why should speech written in a programming language receive less protection than speech in a natural language?[121]

Similarly, the 3D files and the creation of these files should be afforded First Amendment protection. First, the 3D CAD file is the embodiment of a person’s ideas.[122] The edges, the points, and the information in the file originate from the drafter’s interpretation of the item he is recreating. The drafter has the freedom to completely change or add features to already existing items through his 3D CAD design. Although the files have a strong functional presence, the speech component is sufficient enough to get a First Amendment analysis.[123] Second, the creation of the 3D CAD files would also receive protection. Designing the blueprint resembles traditional modes of expression by putting a picture onto a computer screen and eventually into a digital file.

B. The Intermediate Scrutiny Scenario

Using a less stringent standard, the court in Corley determined functionality to be a limitation on the kind of protection electronic files should receive, despite what the file is expressing.[124] Because the 3D CAD file’s expression and functionality are distinct from each other, intermediate scrutiny may be appropriate.[125] One form of the intermediate scrutiny stems from O’Brien.[126]

The O’Brien test validates a government regulation if it: (1) is within government power; (2) furthers an important government interest that is unrelated to suppressing speech (in other words, is content-neutral); (3) prohibits no more speech than what is essential to further the interest; and (4) leaves open ample alternative channels.[127] First, there is no question that the State Department has been authorized to make regulations through the AECA over defense articles.[128] Although scholarship has argued otherwise, the remaining elements are contestable.[129]

The second prong of the O’Brien test essentially asks if the regulation is content-neutral.[130] Content-neutral analysis may be appropriate for scrutinizing “activity” that combines both speech and nonspeech elements.[131] The first step to determining whether a regulation on code is content-neutral is to ask “whether the regulated activity is sufficiently imbued with elements of communication to fall within the scope of the First . . . Amendment[].”[132] Then, if the code does have a speech component, the court must ask “whether the regulation is justified without reference to the content of regulated speech.”[133]

The ITAR is not content-neutral.[134] Regarding the first question, 3D CAD files contain communication within the scope of the First Amendment.[135] The next inquiry reveals that the restriction is not justified because the ITAR is looking specifically at the contents of a 3D CAD file for gun blueprints. Despite the State Department’s assertion that the ITAR is content-neutral, the ITAR focuses on the “direct impact of speech on its audience” because the State Department wants to prevent specific listeners—foreign nationals—from using Defense Distributed’s speech.[136]

Next, under O’Brien’s third prong, the ITAR prevents more speech than what is essential to further the asserted interest. The State Department wants to prevent “all domestic posting on the [i]nternet of ‘technical data’” regardless of the intent to assist foreign nationals.[137] This regulation is overinclusive because it sweeps the otherwise lawful activity of domestic gun enthusiasts and amateur CAD artists into a regulation that should be reserved for people assisting terrorist organizations.[138] Hence, the third prong fails because the ITAR is not narrowly tailored.[139]

The fourth prong of the O’Brien test looks at alternative channels for the would-be speaker. In Kleindienst v. Mandel, the Court analyzed whether the attorney general could refuse Mandel, “a revolutionary Marxist” invited to an academic conference, entry into the United States.[140] The government in Kleindienst suggested that books, tapes, and telephoning “readily supplant[ed] his physical presence,” and that these alternative means to Mandel’s ideas prevented the assertion of a First Amendment violation.[141] However, the Court criticized this alternative means argument because the listeners would be deprived of debate and discussion.[142]

The methods offered by the State Department raise doubts about the legitimacy of the alternative channels remaining for Defense Distributed. The State Department contended Defense Distributed could still use mail, offer the data for sale at “newsstands and bookstores,” or distribute the files “at a conference . . . [or] seminar . . . in the United States.”[143] Being able to provide the 3D CAD files at a convention or on a thumb drive in the mail undermines the argument being made for the national security interest.[144] What prevents someone, i.e., a foreign national, from purchasing the Liberator design at a convention or through the mail? The illegitimacy of the available alternative channels strengthens the notion that the ITAR restrictions on 3D-printing files for plastic guns cannot even survive intermediate scrutiny.

C. Access and Distribution of Information as Rights

The First Amendment not only protects pure speech or conduct, it also protects the right to receive information.[145] The idea around the First Amendment is to “preserve an uninhibited marketplace of ideas in which truth will ultimately prevail,” for it is the public’s right to access ideas that are unabridged by regulations and laws.[146] For example, the Supreme Court has struck down laws and rules prohibiting the distribution of religious information without a permit[147] or prohibiting private possession of obscene materials[148] because it recognizes that the First Amendment embraces the right to distribute literature and necessarily protects the right to receive it.[149] Specifically, the right to express and the right to access information on the internet has been recognized by the Court.[150]

In Sorrell v. IMS Health, the Supreme Court struck down a Vermont law requiring a physician’s consent before records with their prescribing practices can be sold or used for marketing purposes.[151] Although Vermont argued that the law was merely a commercial regulation, the Court noted that “the creation and dissemination of information are speech within the meaning of the First Amendment.”[152] The Sorrell court appreciated the domain for the prescriber-identifying information, i.e., commercial marketplace, as “a forum where ideas and information flourish.”[153]

Similarly, the forum of the Liberator files, the internet, is essential to the wealth and growth of ideas. Removing the essential means for communicating electronic files, i.e., prohibiting the 3D CAD files from being posted on the internet, would be a disservice to the spread of knowledge. Today’s primary means of communication requires an internet connection. Preventing communication of one type of information means the eventual prevention of another type of information, and such censorship cuts against the First Amendment and the public’s right to access information.[154]

Here, the Liberator was offered as a free download. Like the Vermont law in Sorrell, the ITAR is not merely a commercial transaction. Furthermore, the fear that people may make poor decisions based on this information “cannot justify content-based burdens on speech.”[155] It is up to the reader of the information to choose what to do with this information, whether it is to print, to modify, or to ignore the file.

D. Distribution of CAD Files as Speech by Conduct

As mentioned above, the act of disseminating the 3D CAD file relates to the message being conveyed by the 3D CAD file. As scholars have suggested, the dissemination of the 3D-printing files can be viewed as conduct worthy of receiving First Amendment protection.[156] The Supreme Court has used at least two approaches to addressing regulation on conduct that “speaks.” First, the Court used a chilling effect-like test in Cohen v. California to determine whether the government may prohibit a specific kind of conduct.[157] The Supreme Court addressed the overbreadth of a California statute that prohibited the display of offensive messages, finding them “inherently boundless.”[158] Furthermore, the Court in Cohen found the California statute to be unconstitutional under the First Amendment because the connotation of “fuck” in this case did not directly incite violence and did not relate to sex.[159] The Supreme Court did not use the O’Brien test for the Cohen case but instead applied a “more demanding standard” because the statute was related to expression.[160] Similarly, the distribution of the 3D CAD files does not directly incite violence, but rather the act is deserving of a more demanding standard of review because the ITAR regulation is directed at expression—the contents of the files.

Second, under Spence v. Washington, conduct should be treated as speech protected by the First Amendment if the conduct conveys a particularized or symbolic message that “would be understood by those who viewed it.”[161] Symbolic messages that the Court has recognized include pointed anguish over foreign and domestic affairs,[162] desire to remove unfit judges from the bench,[163] or protest over segregation.[164] One could argue that the particularized message in the act of disseminating the blueprint files reveals one’s stance on the current political debate on gun control; people advocating for gun control may be the ones distributing the blueprints. Thus, the conduct of distributing the blueprints should be protected by the First Amendment if such distribution conveys a message that would be understood by those who viewed it.

E. The ITAR Licensing Scheme as an Impermissible Prior Restraint

Prior restraints are governmental restrictions on speech before the speech occurs, and licensing schemes are a classic example.[165] Generally, prior restraints are presumed to be unconstitutional and should receive strict scrutiny.[166] Despite the Court’s reluctance to label a restriction as a prior restraint,[167] a prior-restraint challenge to the ITAR scheme will likely arise.[168]

Jones’s dissent in Defense Distributed analyzed the question of whether the ITAR scheme is an impermissible prior restraint.[169] Speech licensing schemes may be upheld if they meet these two elements: (1) “the licensors must not exercise excessive discretion;” and (2) the schemes must provide for “adequate procedural protections.”[170] The ITAR licensing scheme fails to meet either element. First, the Department of State’s deliberately ambiguous language in the ITAR creates a “chilling effect” on would-be licensees, causing unpredictability about which internet publications constitute as technical data.[171] As designed, the ITAR scheme provides the State Department sweeping discretion because the Department gives licenses on a case-by-case basis.[172] A presumption of an agency’s good faith is ill-advised when the agency has unbridled discretion.[173]

For the second element, the procedural safety nets are inadequate.[174] A prior restraint will be upheld if procedural protections are in place.[175] This requires demonstrating that: (1) the censor bears the burden of proof to show that the expression is unprotected; (2) the restraint will only be imposed for a specified brief period of time; and (3) prompt judicial determination is available.[176] For the first prong, the 3D files would likely be protected expression.[177] Regarding the second prong, the inability to distribute the files lasted over three years and with potential injunctions, such a ban could have lasted into perpetuity.[178] Finally, prompt judicial determination under the ITAR is nonexistent because Congress explicitly “withheld judicial review of the State Department’s designation of items as defense articles.”[179] Although the Directorate of Defense Trade Controls (DDTC) offers a “case-by-case” determination when there is an issue of whether an export is within the ITAR’s scope, there are problems with the process. Regulated parties are either (1) left hanging because the DDTC apparently ignores its forty-five-day deadline for commodity jurisdiction applications; or (2) scrambling to respond to the DDTC within the time limit of only sixty days.[180] The inability to judicially review the constitutionality of the ITAR licensing scheme should raise flags about the constitutionality of the ITAR licensing scheme.[181] Where the opportunity for judicial review is compromised, the censoring body is not able to “bear its burden” in court.[182] As such, a prior restraint with these inadequate procedural safeguards would not survive strict scrutiny.

F. Content-Based Restriction on CAD Files

On its face, the ITAR seems to be a content-based regulation deserving of strict scrutiny.[183] The Supreme Court has held laws attempting to censor or govern “the topic discussed or the idea or message expressed” as content-based.[184] As such, the ITAR licensing scheme is content-based. Rather than regulating 3D CAD files generally (e.g., blueprints for a plastic model plane), the ITAR specifically regulates the 3D CAD file with blueprints for plastic guns. To pass strict scrutiny, content-based regulations must further a “compelling governmental interest” and be “narrowly tailored” to achieve that interest.[185]

Undoubtedly, national security is an important government interest.[186] However, the State Department specified the interest in question here: “preventing foreign nationals—including all manner of enemies of this country—from obtaining technical data on how to produce weapons and weapon parts.”[187] This stated interest struggles to reach “compelling” for many reasons. The 3D-printing files containing blueprints for guns already exist on the internet, and with the “hundreds of thousands” of downloads of the Liberator file prior to the injunction, the file inevitably is sitting on some foreign national’s computer.[188] The lack of evidence of anything resulting from the already available files suggests “no immediate danger to national security.”[189] If the State Department actually wanted to prevent the future distribution of files, it would not have so many exceptions to the rule. The same technical files that are banned from internet publication are allowed within the United States “at conferences, meetings, trade shows, in domestic print publications and in libraries.”[190] How is the file less dangerous in the hands of a conference attendee than the hands of an internet user? This alleged compelling interest has already lost this battle.[191]

Next is the question of whether the ITAR regulation is narrowly tailored. Speech regulation is narrowly tailored when it does not “burden substantially more speech than is necessary to further the government’s legitimate interests.”[192] The ITAR lacks precision from being overinclusive and underinclusive. First, a “significantly overinclusive” regulation is not narrowly tailored.[193] A law is overinclusive when it severely restricts speech that does not pertain to the asserted interest.[194] Here, the ITAR seeks to require “all domestic posting on the [i]nternet of ‘technical data’ to be pre-approved or licensed by the DDTC,” regardless of whether the would-be posters intend to assist foreign enemies directly.[195] The files of gun hobbyists and CAD creatives then fall under the umbrella of ITAR even if they only intended to share the files among colleagues. Thus, the ITAR is overinclusive.

Moreover, a regulation is not narrowly tailored if the regulation is underinclusive.[196] Being underinclusive could suggest that the interest is not as compelling as asserted.[197] As mentioned above, the exceptions to the ITAR allow for the 3D CAD files to be distributed at other venues besides the internet.[198] In addition, the ITAR does not address the 3D CAD files that are already online.[199] Even if the State Department could show it had a compelling interest, these showings of being both underinclusive and overinclusive cut against the ITAR surviving strict scrutiny.

Perhaps one of the reasons for the reluctance to protect this form of expression is a fear of the fast pace of technology. However, such a fear should not strip 3D blueprint files of protection.[200] In fact, the Supreme Court stresses that just because technology quickly changes, First Amendment protection does not suddenly disappear.[201] Furthermore, the ability to communicate one’s ideas to the public is a deeply-rooted principle in a society that prides itself on freedom of speech.[202]

V. Concerns for the Future

“Look, as a consequence of this technology, the way we do things is altered. We’ve stepped outside of your program. So pass your law.”[203]

At the time of Defense Distributed, 3D-printed guns could frequently only withstand one bullet firing because the firearm would break after the initial shot using traditional ammunition.[204] But since then, 3D-printed gun technology has continued to develop. For instance, there is now the ability to print plastic ammunition compatible with 3D plastic guns, causing even more urgency to regulate 3D printing guns.[205] Also, a blueprint for a multiple use 3D-printed Glock exists.[206]

With Defense Distributed settled, many “coulda-woulda-shouldas” arise. For instance, the Supreme Court should have granted certiorari for this case.[207] The Fifth Circuit specifically noted that the legal questions presented by the parties were novel. The lack of direction from the courts leaves open the door for more litigation. For the time being, it is up to the Legislative and Executive Branches to act.[208] Although the Department of State has already settled the case, the Oval Office supports legislation to make 3D guns illegal.[209] Moreover, laws are already in the works. For example, New York’s governor signed legislation prohibiting the possession, manufacture, sale, transport, or possession of undetectable firearms.[210] New Jersey has also passed a bill making the production of 3D-printed guns or the distribution of designs a crime “punishable by up to five years in prison.”[211] California passed a bill that requires a person making or assembling a gun to apply for a serial number through its Department of Justice, and these guns cannot be transferred to anyone else.[212] However, these laws still have the issue of enforceability.[213] Some gunsmiths and gun enthusiasts simply will not comply if they want to distribute the 3D-printed guns.[214]

A blanket ban on 3D-printed guns is an unrealistic strategy, and even if it did pass, such bans and regulations would be difficult to enforce.[215] Suggestions of banning the materials to make the 3D-printed guns are easily shot down because bans on plastics and metals would unduly burden non-gun designers.[216] This prohibition would fail constitutional arguments because the Court has found unconstitutional instances where someone is denied the materials needed to exercise a right.[217]

Furthermore, attempts to control online file sharing may be fruitless.[218] Regulations on 3D-printed guns may take on the shape of traditional firearm regulations by focusing efforts on manufacturing and domestic distribution.[219] One approach is to make the process of printing out a gun and possessing it more difficult and more expensive, perhaps delaying the technology from becoming widespread. One proposal would require the distributor, e.g., a software company or printer manufacturer, to use software to monitor the 3D files that users wish to print.[220] Computer programs could operate similarly to the programs used to prevent printing counterfeit currency.[221] Admittedly, this approach has its own difficulties, such as the lack of uniformity in printers and printing software and difficulty in identifying guns through code.[222] This also places a burden on the manufacturers to devote resources to a client base, i.e., gun printers, which they did not originally intend to serve in order to be in compliance.[223] Such efforts to protect the software may be fruitless because software could potentially be jailbroken.[224]

Another approach involves regulating the manufacturing using the Undetectable Firearms Act, by requiring a small amount of metal to be in the gun.[225] The UFA prohibits the manufacture, sale, or transfer of a firearm that is undetectable by metal detectors.[226] Proposed amendments to the UFA would have extended the ban to manufacturing 3D-printed firearms for personal consumption.[227] Requiring a small amount of metal would likely be narrowly tailored to survive strict scrutiny because the burden is minimal while serving a government interest in protecting places that are guarded by metal detectors.[228] This approach appeals to those in the camp wanting to regulate guns, rather than regulate 3D printing.[229] Although this proposed amendment was defeated, this approach may be worth revisiting.[230]

The regulation of code to prevent the creation of untraceable firearms is “good policy.”[231] Because the ITAR is concerned about security with foreign nationals, the ITAR is not the means of gun safety and security on the domestic front. With technology outpacing lawmakers, the government should find a way to address this imminent threat to public safety while preserving free speech.[232] By finding current weaknesses in the ITAR, this Comment hopes to strengthen future gun laws and regulations.

VI. Conclusion

“If the government were to regulate this, would we consider that a success?”[233]

With new technology comes new issues about rights, specifically the First Amendment. The facts and procedural history of Defense Distributed left open the door for litigation and unpredictability.[234] However, the settlement gives opportunity to the Legislative and Executive Branches to pass new laws or amend existing ones.

While making these rules, the government should bear in mind the many benefits from 3D printing as a technology. Broadly regulating 3D printing would hinder medical and industry innovation.[235] Although there may always be some user with ill intent, regulation would help dissuade some people from misuse.[236] A lesson from the First Amendment analysis of the ITAR here may assist the Legislative or Executive Branches to create stronger gun laws and regulations. To avoid a strict scrutiny analysis, the government may want to create narrower laws, such that the law is “content-neutral,” rather than “content-specific.” If, however, the government must use a prior restraint or must restrict content or conduct, then it should articulate a compelling governmental interest that would allow such regulation to withstand strict scrutiny.

Khanh Nguyen Leon

  1. Andy Greenberg, ‘Wiki Weapon Project’ Aims to Create a Gun Anyone Can 3D-Print at Home, Forbes (Aug. 23, 2012, 9:00 AM), https://www.forbes.com/sites/andygreenberg/2012/08/23/wiki-weapon-project-aims-to-create-a-gun-anyone-can-3d-print-at-home [https://perma.cc/2KU9-NRTZ] (quoting Cody Wilson).

  2. See 22 U.S.C. § 2778(a)(1) (2012).

  3. See id. § 2778(j)(4)(A).

  4. See Rebecca Morelle, Working Gun Made with 3D Printer, BBC (May 6, 2013), https://www.bbc.com/news/science-environment-22421185 [https://perma.cc/PV78-3QKZ].

  5. Id.

  6. Andy Greenberg, 3D-Printed Gun’s Blueprints Downloaded 100,000 Times in Two Days (with Some Help from Kim Dotcom), Forbes (May 8, 2013, 5:12 PM), https://www.forbes.com/sites/andygreenberg/2013/05/08/3d-printed-guns-blueprints-downloaded-100000-times-in-two-days-with-some-help-from-kim-dotcom [https://perma.cc/475X-RKLN].

  7. Def. Distributed v. U.S. Dep’t of State, 838 F.3d 451, 455–56 (5th Cir. 2016), cert. denied, 138 S. Ct. 638 (2018) (mem.).

  8. See Def. Distributed v. U.S. Dep’t of State, 121 F. Supp. 3d 680, 688 (W.D. Tex. 2015), aff’d, 838 F.3d 451 (5th Cir. 2016), cert. denied, 138 S. Ct. 638 (2018) (mem.).

  9. Def. Distributed, 138 S. Ct. at 638; Def. Distributed, 838 F.3d at 461; Def. Distributed, 121 F. Supp. 3d at 701.

  10. See Second Amendment Found., DOJ, SAF Reach Settlement in Defense Distributed Lawsuit, PR Newswire (July 10, 2018, 3:35 PM), https://www.prnewswire.com/news-releases/doj-saf-reach-settlement-in-defense-distributed-lawsuit-300678872.html [https://perma.cc/W66A-2HRX].

  11. Kelsey Wilbanks, 3D Gun Legality After Defense Distributed Settlement, Ruling, Law360 (Aug. 2, 2018, 2:26 PM), https://www.law360.com/articles/1069142/3d-gun-legality-after-defense-distributed-settlement-ruling.

  12. See id.; Donald Trump (@realDonaldTrump), Twitter (July 31, 2018, 7:03 AM), https://twitter.com/realDonaldTrump/status/1024264286418489345 [https://perma.cc/5LJA-9HGV].

  13. Def. Distributed, 138 S. Ct. at 638.

  14. Clay Dillow, Q+A: Cody Wilson of the Wiki Weapon Project on the 3-D Printed Future of Firearms, Popular Sci. (Dec. 21, 2012), https://www.popsci.com/technology/article/2012-12/qa-cody-wilson-wiki-weapons-project-3-d-printed-future-firearms (quoting Cody Wilson).

  15. U.S. Const. amend. I (“Congress shall make no law . . . abridging the freedom of speech . . . .”); see also, e.g., Texas v. Johnson, 491 U.S. 397, 404–05 (1989) (acknowledging conduct with certain communicative aspects as speech).

  16. See, e.g., New York v. Ferber, 458 U.S. 747, 773 (1982) (upholding a statute prohibiting people from promoting child pornography); Brandenburg v. Ohio, 395 U.S. 444, 447 (1969) (upholding restrictions on speech if it advocates for “imminent lawless action and is likely to incite or produce such action”).

  17. See Sable Commc’ns of Cal., Inc. v. FCC, 492 U.S. 115, 130–31 (1989) (finding that a blanket ban on “dial-a-porn” services was not narrowly tailored enough despite the state’s compelling interest in protecting minors from indecent messages).

  18. Reed v. Town of Gilbert, 135 S. Ct. 2218, 2228 (2015) (quoting Cincinnati v. Discovery Network, Inc., 507 U.S. 410, 429 (1993)).

  19. See, e.g., Boos v. Barry, 485 U.S. 312, 318–19 (1988) (strictly scrutinizing a display clause that prohibited picket signs in front of an embassy if the signs were “critical of the foreign government”).

  20. See Va. State Bd. of Pharmacy v. Va. Citizens Consumer Council, 425 U.S. 748, 771 (1976).

  21. Clark v. Cmty. for Creative Non-Violence, 468 U.S. 288, 295, 297–98 (1984) (finding that the prohibition of sleeping overnight in a park was a reasonable time, place, or manner regulation).

  22. Alexander v. United States, 509 U.S. 544, 550 (1993) (quoting Melville Nimmer, Nimmer on Freedom of Speech 4–14 (1984)) (emphasis omitted).

  23. FW/PBS, Inc. v. City of Dallas, 493 U.S. 215, 225 (1990) (quoting Se. Promotions, Ltd. v. Conrad, 420 U.S. 546, 558 (1975)); see also, e.g., Lovell v. Griffin, 303 U.S. 444, 451–52 (1938).

  24. Ariel L. Bendor, Prior Restraint, Incommensurability, and the Constitutionalism of Means, 68 Fordham L. Rev. 289, 341–42 (1999).

  25. See, e.g., N.Y. Times v. United States, 403 U.S. 713, 714 (1971) (finding that the government failed to meet the burden of “showing justification for the imposition of such a [prior] restraint”) (quoting Org. for a Better Austin v. Keefe, 402 U.S. 415, 419 (1971)); id. at 726–27 (Brennan, J., concurring) (finding a lack of proof that publication of the Pentagon Papers would “inevitably, directly, and immediately” cause a perilous event); United States v. Robel, 389 U.S. 258, 265–66 (1967) (finding mere membership in the Communist Party could not prevent a person from employment in matters important to national security). But see United States v. Progressive, Inc., 467 F. Supp. 990, 1000 (W.D. Wis. 1979) (permitting a prior restraint on a report containing information about the hydrogen bomb).

  26. See Freedman v. Maryland, 380 U.S. 51, 58–59 (1965); see also Near v. Minnesota, 283 U.S. 697 (1931).

  27. Spence v. Washington, 418 U.S. 405, 409 (1974) (per curiam).

  28. United States v. O’Brien, 391 U.S. 367, 376 (1968).

  29. See id. at 376–77.

  30. Id. at 377.

  31. Id. at 388–89 (1968) (Harlan, J., concurring) (introducing the ample alternative channels). Subsequent cases later solidified this element. See, e.g., Renton v. Playtime Theatres, Inc., 475 U.S. 41, 53 (1986); Clark v. Cmty. for Creative Non-Violence, 468 U.S. 288, 293 (1984).

  32. Josh Blackman, The 1st Amendment, 2nd Amendment, and 3D Printed Guns, 81 Tenn. L. Rev. 479, 526–27 (2014).

  33. Bernstein v. U.S. Dep’t of State, 922 F. Supp. 1426, 1428–31, 1436 (N.D. Cal. 1996).

  34. Commodity Futures Trading Comm’n v. Vartuli, 228 F.3d 94, 111 (2d Cir. 2000).

  35. Universal City Studios, Inc. v. Corley, 273 F.3d 429, 448, 456 (2d Cir. 2001) (recognizing the risk of suppressing expression but ultimately holding there to be substantial government interests unrelated to the suppression of free expression).

  36. Id. at 448.

  37. Id. at 449 (emphasis added).

  38. Mainstream Loudoun v. Bd. of Trs. of Loudoun Cty. Library, 24 F. Supp. 2d 552, 570 (E.D. Va. 1998).

  39. Id. at 568–69.

  40. Id. at 570.

  41. See Jasper L. Tran, The Law and 3D Printing, 31 J. Marshall J. Info. Tech. & Privacy L. 505, 508 (2015).

  42. Id.

  43. Blackman, supra note 32, at 484–85.

  44. Id. The STL (short for “stereolithography”) file format is the most common for 3D printers. See Lucas S. Osborn, Regulating Three-Dimensional Printing: The Converging Worlds of Bits and Atoms, 51 San Diego L. Rev. 553, 559–60 (2014) (explaining that there are various printer-compatible file types).

  45. Osborn, supra note 44, at 560.

  46. See Blackman, supra note 32.

  47. See Adam Thierer & Adam Marcus, Gun, Limbs, and Toys: What Future for 3D Printing?, 17 Minn. J.L. Sci. & Tech. 805, 831 (2016).

  48. However, some states have enacted laws regulating the distribution and production of 3D-printed guns and their designs. See infra Part V.

  49. See 18 U.S.C. § 922(p)(1)(A) (2012).

  50. See id. § 922(p)(2)(C).

  51. See Joseph Muha, 3D Printing and Gun Laws, 18 W. Mich. U. Cooley J. Prac. & Clinical L. 67, 78 (2016).

  52. See id. at 72.

  53. See 22 U.S.C. § 2778(a)(1) (2012).

  54. Exec. Order No. 13637, 78 Fed. Reg. 16,127, 16,129 (Mar. 8, 2013); Blackman, supra note 32, at 508.

  55. 22 C.F.R. § 120.17(a)(1) (2018) (emphasis added).

  56. Id. §§ 120.1(c), 120.10(a)(1).

  57. See infra Part III.

  58. 18 U.S.C. § 922(a) (2012).

  59. Id. § 921(a)(3).

  60. Q&As: 3-D Printing Technology of Firearms, ATF (Nov. 13, 2013), https://www.atf.gov/file/4566/download [https://perma.cc/AT4Y-EGXV] (“An individual may generally make a firearm for personal use.”).

  61. Def. Distributed v. U.S. Dep’t of State, 838 F.3d 451 (5th Cir. 2016), cert. denied, 138 S. Ct. 638 (2018) (mem.).

  62. Id. at 455.

  63. Id. at 454.

  64. See Greenberg, supra note 6.

  65. Id.; Philip Bump, State Department Asks Defense Distributed to Take Down Its 3D-Printed Gun Plans, Atlantic (May 9, 2013), https://www.theatlantic.com/national/archive/2013/05/state-department-defense-distributed-3d-printed-gun-plans/315405/.

  66. Def. Distributed, 838 F.3d at 462 (Jones, J., dissenting).

  67. See id.

  68. See Def. Distributed v. U.S. Dep’t of State, 121 F. Supp. 3d 680, 688–90, 701 (W.D. Tex. 2015), aff’d, 838 F.3d 451 (5th Cir. 2016), cert. denied, 138 S. Ct. 638 (2018).

  69. Id. at 689–91.

  70. Id. at 691.

  71. Id. at 694.

  72. Id. (citing United States v. Chi Mak, 683 F.3d 1126, 1135 (9th Cir. 2012) (finding that “the AECA and its implementing regulations are content-neutral”)).

  73. Def. Distributed, 121 F. Supp. 3d at 694–96.

  74. Id. at 695.

  75. Id.

  76. Def. Distributed v. U.S. Dep’t of State, 838 F.3d 451, 461 (5th Cir. 2016), cert. denied, 138 S. Ct. 638 (2018) (mem.).

  77. Id. at 453, 455, 460.

  78. See id. at 458.

  79. Id.

  80. See id. at 459–60.

  81. Id. at 460.

  82. See id. (“Because those files would never go away, a preliminary injunction would function, in effect, as a permanent injunction as to all files released in the interim.”).

  83. See id. at 461.

  84. Id. at 476 (Jones, J., dissenting) (finding the majority’s decision “toward the First Freedom . . . highly regrettable”).

  85. Id. at 470, 472.

  86. Id. at 470.

  87. Id. at 469.

  88. Id. at 470 (citing Boos v. Barry, 485 U.S. 312, 321 (1988) (overturning an ordinance that restricted criticism of foreign governments near their embassies because it “focus[ed] on the direct impact of speech on its audience”)).

  89. See id. at 470–71.

  90. Def. Distributed v. U.S. Dep’t of State, 138 S. Ct. 638 (2018) (mem.).

  91. See Second Amendment Found., supra note 10.

  92. Id.

  93. See Wilbanks, supra note 11.

  94. Id.

  95. David Shortell, Gun Control Advocates Can’t Stop Group from Posting Instructions to 3-D Print a Gun, CNNPolitics (July 27, 2018, 9:23 PM), https://www.cnn.com/2018/07/27/politics/plastic-gun-control-lawsuit/index.html [https://perma.cc/G4S2-ZZWC].

  96. Complaint for Declaratory and Injunctive Relief at 41–47, Washington v. U.S. Dep’t of State, No. 2:2018-cv-01115 (W.D. Wash. July 30, 2018).

  97. Steve Almasy, A Judge Ruled That a Website Has to Suspend Downloads for 3D Gun Plans. But They’re Already Out There, CNN (Aug. 1, 2018, 3:29 PM), https://www.cnn.com/2018/07/31/us/3d-guns-downloaded-plans-states/index.html [https://perma.cc/3AUQ-R39Y].

  98. See Wilbanks, supra note 11.

  99. See Tina Bellon, U.S. Judge Extends Ban of Online 3-D Printed Gun Blueprints, Reuters (Aug. 27, 2018, 11:47 AM), https://www.reuters.com/article/us-usa-court-guns/u-s-judge-extends-ban-of-online-3-d-printed-gun-blueprints-idUSKCN1LC1UF [https://perma.cc/W2CR-H9NQ].

  100. Id.

  101. See Deanna Paul, Federal Judge Blocks Publication of 3-D Printed Gun Blueprints, Wash. Post: Post Nation (Aug. 27, 2018), https://www.washingtonpost.com/news/post-nation/wp/2018/08/21/federal-judge-will-soon-decide-whether-to-block-3-d-printed-gun-blueprints/ [https://perma.cc/UK8Y-N92Z].

  102. See Wilbanks, supra note 11.

  103. Greenberg, supra note 1 (quoting Cody Wilson).

  104. Def. Distributed v. U.S, Dep’t of State, 838 F.3d 451, 461 (5th Cir. 2016), cert. denied, 138 S. Ct. 638 (2018) (mem.).

  105. For other legal issues within the circumstances of Defense Distributed and 3D printing, see generally James B. Jacobs & Alex Haberman, 3D-Printed Firearms, Do-It-Yourself Guns, & the Second Amendment, 80 Law & Contemp. Probs. 129 (2017); Osborn, supra note 44, at 556–57.

  106. See Derk Westermeyer, The State Department Can Gun Down 3-D Printed Firearms, 13 Wash. J.L. Tech. & Arts 201, 214–15 (2018) (arguing that the ITAR would “likely pass a First Amendment challenge”).

  107. See, e.g., Kyle Langvardt, The Doctrinal Toll of "Information as Speech," 47 Loy. U. Chi. L.J. 761, 766 (2016).

  108. Sorrell v. IMS Health, Inc., 564 U.S. 552, 570 (2011) (referring to physician medication prescribing data).

  109. Brown v. Entm’t Merchs. Ass’n, 564 U.S. 786 (2011) (emphasis added) (finding a California regulation of violent video games to be unconstitutional).

  110. Id. at 801–02.

  111. Id. at 799.

  112. Id.

  113. Cf. Blackman, supra note 32, at 501 (comparing CAD files to “pictures of guns”).

  114. Id.; see also *First Amendment—Technology—Fifth Circuit Declines to Enjoin Regulation of Online Publication of 3D-Printing Files.—*Defense Distributed v. United States Department of State, 838 F.3d 451 (5th Cir. 2016), 130 Harv. L. Rev. 1744, 1750 (2017) [hereinafter Fifth Circuit Declines] (“[S]ource code . . . may say something about an object.”).

  115. See, e.g., Spence v. Washington, 418 U.S. 405, 409 (1974) (per curiam) (requiring a determination of whether displaying a flag with a peace symbol included sufficient elements of communication).

  116. See id.

  117. Anderson v. City of Hermosa Beach, 621 F.3d 1051, 1062 (9th Cir. 2010) (finding that the tattooing process receives First Amendment protection like the tattoo because the process itself is expressive and relates to the expressive end product).

  118. Id. at 1061.

  119. Id. at 1062.

  120. Junger v. Daley, 209 F.3d 481, 484–85 (6th Cir. 2000); see also Bernstein v. U.S. Dep’t of State, 922 F. Supp. 1426, 1436 (N.D. Cal. 1996) (holding that source code is speech under the First Amendment).

  121. Lee Tien, Publishing Software as a Speech Act, 15 Berkeley Tech. L.J. 629, 633 (2000).

  122. See Blackman, supra note 32, at 501.

  123. See id.; see also Tenafly Eruv Ass’n v. Borough of Tenafly, 309 F.3d 144, 164 (3d Cir. 2002) (finding purely functional objects not deserving of First Amendment protection).

  124. Universal City Studios, Inc. v. Corley, 273 F.3d 429, 451 (2d Cir. 2001) (using a First Amendment analysis in combination with the function-expression dichotomy).

  125. See Def. Distributed v. U.S. Dep’t of State, 121 F. Supp. 3d 680, 693–94 (W.D. Tex. 2015), aff’d, Def. Distributed v. U.S. Dep’t of State, 838 F.3d 451 (5th Cir. 2016), cert. denied, 138 S. Ct. 638 (2018) (mem.).

  126. See Corley, 272 F.3d at 456 (using the O’Brien test on software).

  127. See United States v. O’Brien, 391 U.S. 367, 377 (1968).

  128. See 22 U.S.C. § 2752(b) (2012).

  129. See Julia Cosans, Comment, Between Firearm Regulation and Information Censorship: Analyzing First Amendment Concerns Facing the World’s First 3-D Printed Plastic Gun, 22 Am. U. J. Gender Soc. Pol’y & L. 915, 937–38 (2014) (arguing that the “substantial risk of imminent harm” of the 3D CAD file for gun blueprints necessitates the application of intermediate scrutiny).

  130. See Corley, 273 F.3d at 454 (whether posting prohibition that regulates decryption software is content-neutral).

  131. See Spence v. Washington, 418 U.S. 405, 410–11 (1974) (per curiam).

  132. See Corley, 273 F.3d at 450 (alteration in original) (quoting Spence, 418 U.S. at 409).

  133. See id. at 450–51 (citing Hill v. Colorado, 530 U.S. 703, 720 (2000)).

  134. Def. Distributed v. U.S. Dep’t of State, 838 F.3d 451, 470, 476 (5th Cir. 2016) (Jones, J., dissenting), cert. denied, 138 S. Ct. 638 (2018) (mem.) (insisting that the ITAR is a content-based restriction). This analysis also requires recognizing the government interest, which will be discussed later in Part V.

  135. See supra Section II.B.

  136. See Boos v. Barry, 485 U.S. 312, 320–21 (1988).

  137. Def. Distributed, 838 F.3d at 470–71 (Jones, J., dissenting) (emphasis added).

  138. Id.

  139. Id. at 472.

  140. Kleindienst v. Mandel, 408 U.S. 753, 756, 770 (1972).

  141. Id. at 765.

  142. Id. (still holding for the government because alternative means is not dispositive to “any constitutional interest”).

  143. 22 C.F.R. § 120.11(a) (2018).

  144. Wilson fulfilled orders for the CAD files for the 3D gun using thumb drives. See Jim Vertuno & Martha Bellisle, Texan Says He’s Selling 3D-Printed Gun Plans After Ruling, AP News (Aug. 28, 2018), https://www.apnews.com/f1238ad5c636429aad9203b6a3a8af55 [https://perma.cc/5TJ7-84VL].

  145. See Bigelow v. Virginia, 421 U.S. 809, 822 (1975) (finding certain commercial information needs to be available to the general public).

  146. Red Lion Broad. Co. v. FCC, 395 U.S. 367, 390 (1969).

  147. See Lovell v. City of Griffin, 303 U.S. 444, 447, 449 (1938).

  148. See generally Stanley v. Georgia, 394 U.S. 557 (1969) (holding that the First Amendment prohibits criminalization of possession of obscene material).

  149. Lovell, 303 U.S. at 452.

  150. Reno v. ACLU, 521 U.S. 844, 874 (1997).

  151. Sorrell v. IMS Health, Inc., 564 U.S. 552, 579–80 (2011).

  152. Id. at 570 (citations omitted) (finding that the statute did not permissibly advance state government interest of decreasing medical costs and promoting public health).

  153. Id. at 577–79.

  154. Blackman, supra note 32, at 503 (asserting that a First Amendment inquiry considers “the individual right to express information and the right of individuals . . . to learn and consume that information”).

  155. Sorrell, 564 U.S. at 577 (citing Thompson v. W. States Med. Ctr., 535 U.S. 357, 374 (2002)).

  156. See, e.g., Blackman, supra note 32, at 502–04 (suggesting that the ITAR’s regulation on the dissemination of CAD files is content-neutral).

  157. See Cohen v. California, 403 U.S. 15, 26 (1971).

  158. Id. at 25.

  159. Id. at 15, 18, 20.

  160. See id. at 18–20; Texas v. Johnson, 491 U.S. 397, 403 (1989) (providing framework for when to use O’Brien test and when to use a more rigorous standard).

  161. Spence v. Washington, 418 U.S. 405, 409–11 (1974) (per curiam) (holding a Washington state statute prohibiting additions or modifications to the U.S. flag to be unconstitutional).

  162. See id. at 410.

  163. See United States v. Grace, 461 U.S. 171, 173, 176 (1983).

  164. See Brown v. Louisiana, 383 U.S. 131, 141–42 (1966).

  165. See supra Section II.A.2.

  166. See FW/PBS, Inc. v. City of Dallas, 493 U.S. 215, 225 (1990).

  167. See Cohen v. California, 403 U.S. 15, 25 (1971).

  168. By definition, a prior restraint is a government restraint on expression before it takes place. See supra Section II.A.2. Although there are limited circumstances for when prior restraints are allowed, an analysis is still required. But see Westermeyer, supra note 106, at 214 (arguing that the licensing scheme would probably avoid the prior restraint label).

  169. See Def. Distributed v. U.S. Dep’t of State, 838 F.3d 451, 472–75 (5th Cir. 2016) (Jones, J., dissenting), cert. denied, 138 S. Ct. 638 (2018) (mem.).

  170. Id. at 472–73.

  171. See supra Section IV.D.

  172. See 22 CFR § 123.1 (1993).

  173. See Def. Distributed, 838 F.3d at 471, 473 (Jones, J., dissenting) (fearing unfettered discretion and the use of prior restraint intimidated regulated parties into censoring their own speech); see also Thornhill v. Alabama, 310 U.S. 88, 97 (1940) (finding a law prohibiting picketing illegal because such a law “constitutes the danger to freedom of discussion”).

  174. See Def. Distributed, 838 F.3d at 473–74 (Jones, J., dissenting).

  175. Freedman v. Maryland, 380 U.S. 51, 58–59 (1965).

  176. Id.

  177. See supra Section IV.A.

  178. Def. Distributed, 838 F.3d at 463 (Jones, J., dissenting).

  179. Id. at 474.

  180. See id. at 473–74.

  181. Id. at 474 (“The withholding of judicial review alone should be fatal . . . .”).

  182. Id.

  183. See id. at 468–69 (citing Reed v. Town of Gilbert, 135 S. Ct. 2218, 2226 (2015)).

  184. See Reed, 135 S. Ct. at 2227 (finding a sign law to be content-based because it applied different rules based on the message being conveyed).

  185. See, e.g., Austin v. Mich. Chamber of Commerce, 494 U.S. 652, 655, 666 (1990); United States v. Grace, 461 U.S. 171, 177 (1983).

  186. See Def. Distributed, 838 F.3d at 458.

  187. Id.

  188. Id. at 461–62 (Jones, J., dissenting).

  189. Id. at 475 (Jones, J., dissenting).

  190. Id. at 476 (Jones, J., dissenting).

  191. A more convincing governmental interest would be domestic gun safety. See infra Part V.

  192. Ward v. Rock Against Racism, 491 U.S. 781, 799 (1989).

  193. Def. Distributed, 838 F.3d at 470 (Jones, J., dissenting) (citing Simon & Schuster, Inc. v. Members of N.Y. State Crime Victims Bd., 502 U.S. 105, 121 (1991)).

  194. Eugene Volokh, Freedom of Speech, Permissible Tailoring and Transcending Strict Scrutiny, 144 U. Pa. L. Rev. 2417, 2422 (1997); see also, e.g., Brown v. Entm’t Merchs. Ass’n, 564 U.S. 786, 804 (2011) (finding a video game law restricting the sale of violent video games to minors to be overinclusive because not all children have parents concerned about such purchases).

  195. Def. Distributed, 838 F.3d at 470–71 (Jones, J., dissenting).

  196. See, e.g., Rutan v. Republican Party of Ill., 497 U.S. 62, 74 (1990); Sable Commc’ns of Cal, Inc. v. FCC, 492 U.S. 115, 126 (1989).

  197. Volokh, supra note 194, at 2420.

  198. See Def. Distributed, 838 F.3d at 476 (Jones, J., dissenting).

  199. See Almasy, supra note 97.

  200. See Steven E. Halpern, Harmonizing the Convergence of Medium, Expression, and Functionality: A Study of the Speech Interest in Computer Software, 14 Harv. J.L. & Tech. 139, 148–49 (2000).

  201. See Brown v. Entm’t Merchs. Ass’n, 564 U.S. 786, 790 (2011) (citing Joseph Burstyn, Inc. v. Wilson, 343 U.S. 495, 503 (1952)).

  202. See Blackman, supra note 32, at 502–03.

  203. Dillow, supra note 14 (quoting Cody Wilson).

  204. Josh Hafner, What Is a 3D Printed Gun, and How Is It Legal? Your Questions, Answered, USA Today (Aug. 2, 2018, 2:58 PM), https://www.usatoday.com/story/tech/nation-now/2018/08/01/3-d-guns-how-3-d-printed-gun-parts-made-and-how-theyre-legal/879349002/ [https://perma.cc/5VL7-XK6M].

  205. See Lance Ulanoff, Now There Are Bullets That Won’t Break Your 3D-Printed Gun, Mashable (Nov. 6, 2014), http://mashable.com/2014/11/06/bullets-3d-printed-gun/ [https://perma.cc/LUA3-WLSE]. Interestingly, the most likely victim from a 3D printed gun is the shooter himself. See Jeremy Straub, 3D-Printing Guns at Home Is Dangerous — Mostly for the Person Shooting It, Fast Company (Jan. 10, 2019), https://www.fastcompany.com/90290217/3d-printing-guns-at-home-is-dangerous-mostly-for-the-person-shooting-it [https://perma.cc/T5UR-CNYU].

  206. Jake Hanrahan, 3D-Printed Guns Are Back, and This Time They Are Unstoppable, Wired (May 20, 2019), https://www.wired.co.uk/article/3d-printed-guns-blueprints [https://perma.cc/57X4-AZMY].

  207. No reason was given for the denial. Def. Distributed v. Dep’t of State, 138 S. Ct. 638 (2018) (mem.).

  208. The Legislative and Executive Branches are certainly getting help from the Judicial Branch. See Paul, supra note 101.

  209. See Ken Thomas & Matthew Daly, White House: DOJ Didn’t Consult Trump on 3D-Printed Guns, AP News (Aug. 1, 2018), https://apnews.com/d13f094c44214406ac5a1ef239dadebf [https://perma.cc/6EQ6-R4EW].

  210. Office of the Governor, Governor Cuomo Signs Legislation Banning Undetectable Guns and Expanding Firearm Safe Storage Laws to Protect Children (July 30, 2019), https://www.governor.ny.gov/news/governor-cuomo-signs-legislation-banning-undetectable-guns-and-expanding-firearm-safe-storage [https://perma.cc/7HN8-DF2F].

  211. S. 2465, 2018 Leg., 218th Sess. (N.J. 2018); see also Nicholas Pugliese & Trenton Bureau, 'Ghost Guns,’ 3-D Guns and Other New-Age Firearms Illegal in NJ, app. (Nov. 8, 2018, 5:30 P.M.), https://www.app.com/story/news/new-jersey/governor/2018/11/08/nj-bans-ghost-guns-3-d-printable-guns-other-weapons-new-laws/1933236002/ [https://perma.cc/FS2C-CA4G]; Moriarty, Schaer & Quijano Bill Banning Ghost Guns, 3-D Plastic Firearms in NJ Advances in Assembly, Insider NJ (Sept. 17, 2018, 4:45 PM), https://www.insidernj.com/press-release/moriarty-schaer-quijano-bill-banning-ghost-guns-3-d-plastic-firearms-nj-advances-assembly-2/ [https://perma.cc/A8D8-9SNH].

  212. See Assemb. 857, 2015–2016 Leg., Reg. Sess. (Cal. 2016).

  213. See Jacobs & Haberman, supra note 105, at 146.

  214. Id.

  215. See Osborn, supra note 44, at 579.

  216. See Blackman, supra note 32, at 512.

  217. See Minneapolis Star & Tribune Co. v. Minn. Comm’r of Revenue, 460 U.S. 575, 591 (1983) (finding a tax on newspaper ink to be a First Amendment violation because it solely affected the press).

  218. Langvardt, supra note 107, at 794.

  219. See Osborn, supra note 44, at 577–79.

  220. Id. at 579.

  221. See Melissa Riofrio, Why You Can’t Print Counterfeit Money on Your Color Laser, PCWorld (June 7, 2011, 6:00 PM), https://www.pcworld.com/article/229647/counterfeit_money_on_color_laser_printers.html [https://perma.cc/74A6-TL4S].

  222. See Osborn, supra note 44, at 578–79.

  223. An interesting approach is to have manufacturers police users and take down unauthorized gun CAD files, similar to websites removing content under the DMCA. See Katie Fleschner McMullen, Comment, Worlds Collide When 3D Printers Reach the Public: Modeling a Digital Gun Control Law After the Digital Millennium Copyright Act, 2014 Mich. St. L. Rev. 187, 219–20.

  224. See German Lopez, The Battle to Stop 3D-Printed Guns, Explained, Vox (Aug. 29, 2018, 12:45 PM), https://www.vox.com/2018/7/31/17634558/3d-printed-guns-trump-cody-wilson-defcad [https://perma.cc/C8BJ-4WMH].

  225. See Blackman, supra note 32, at 509.

  226. 18 U.S.C § 922(p) (2012).

  227. See Blackman, supra note 32, at 510.

  228. Id. at 511.

  229. See Rory K. Little, Guns Don’t Kill People, 3D Printing Does? Why the Technology Is a Distraction from Effective Gun Controls, 65 Hastings L.J. 1505, 1509 (2014) (noting that controlling production, possession, or use of a 3D gun would be better than controlling the mere printing of the guns).

  230. See Blackman, supra note 32, at 510–11.

  231. Fifth Circuit Declines, supra note 114, at 1751.

  232. See Thierer & Marcus, supra note 47, at 825–27.

  233. See Dillow, supra note 14 (quoting Cody Wilson).

  234. See supra Part III.

  235. See Thierer & Marcus, supra note 47, at 853.

  236. See McMullen, supra note 223, at 225.