The Volokh Conspiracy

ATF declares that its Final Rule at issue before the Supreme Court in Garland v. VanDerStok "will enhance public safety by helping to ensure that more firearms may be traced by law enforcement to solve crime and arrest the perpetrators."  Radically expanding the definition of "firearm" from what Congress enacted is allegedly justified by the policy argument that the agency will be able to "trace" more firearms.  Whether that will solve more crimes is a big "if."

We're all familiar with the spiel.  A criminal leaves his gun at a "crime scene" (how often does that happen?) but gets away, unidentified.  Police find the gun and ask ATF to trace it.  The gun is engraved with the manufacturer's name and serial number.  ATF starts with the manufacturer and, using the records kept by federal licensees, traces the gun to its retail purchaser.  And voilà, the criminal is identified and arrested.

But now the sky is falling.  ATF insists that its Final Rule is the Ghost Buster for "ghost guns," a propaganda term used to describe privately-made firearms.  Unless the kits from which hobbyists make their own guns are declared to be "firearms," their homemade guns won't be traceable.  Criminals who lose their guns at "crime scenes" won't be caught.

After years of ATF exaggerating the usefulness of tracing, Congress enacted a law in 2013 requiring ATF to "make clear that trace data cannot be used to draw broad conclusions about firearms-related crime" by including in its releases of information the following language: "Law enforcement agencies may request firearms traces for any reason, and those reasons are not necessarily reported to the Federal Government. Not all firearms used in crime are traced and not all firearms traced are used in crime."

Consider the disconnect.  ATF traces all firearms it encounters.  A person is subject to a domestic violence restraining order and ATF learns that he has a very large gun collection.  They raid his house, seize all 200 of his guns, and then trace them.  That goes down as 200 "crime guns" seized at a "crime scene" that have nothing to do with his offense of mere possession while subject to the order.
As explained in my two previous posts (here and here), Congress defines a "firearm" as a weapon "which will or is designed to or may readily be converted to expel a projectile by the action of an explosive" or "the frame or receiver of any such weapon."  ATF's Final Rule expands that definition to include partially-machined raw material, information, jigs, and tools that sufficiently-skilled persons may fabricate into a firearm.  Whether ATF has such authority is the issue before the Court in VanDerStok.

One of the superior amici briefs filed in the case is that of the Citizens Committee for the Right to Keep and Bear Arms, authored by Dan Peterson and C. D. Michel.  I'll cover some of the highlights in that brief and offer some additional material in the following remarks.

Only licensed manufacturers and importers who are "engaged in the business" are required by the Gun Control Act (GCA) to identify and serialize firearms.  18 U.S.C. § 923(i).  Hobbyists are free lawfully to craft their own guns without these requirements.  ATF claims that the resultant "ghost guns" cannot be traced, thus requiring the non-gun materials that hobbyists use to make guns be redefined as guns.

But the GCA, as amended by the Firearm Owners' Protection Act, sharply delineates licensees from private individuals.  While ATF may inspect licensed dealer records "in the course of a bona fide criminal investigation," it is prohibited from establishing "any system of registration of firearms, firearms owners, or firearms transactions."  18 U.S.C. §§ 923(g) & 926(a).

Nevertheless, ATF has been on a crusade to trace all firearms that law enforcement encounters, and its attack on privately-made firearms is only the latest stage in this endeavor.  The Final Rule, ATF urges, is necessary to address an "urgent public safety and law enforcement crisis posed by the exponential rise of untraceable firearms commonly called 'ghost guns.'"

Let's test this claim with reality.  New Jersey is one of the states that traces every firearm it encounters, to include the .22 rifle a widow abandons at a police station.  Not exactly a crime scene.

In 2022, New Jersey criminalized the purchase of a parts kit not made by a licensed manufacturer with a serial number.  ATF trace data for New Jersey that same year shows 5,248 firearm traces, of which 3,824 – 73% – were for "possession of weapon" and "found firearm."  Keep in mind that the Garden State makes possession per se without the right papers a crime.  How many of these were privately-made firearms?  Only 67 traces were for "homicide" and 132 for "aggravated assault."  As to firearms seized from the possessor, how did tracing solve any crime?

The Citizens Committee brief goes on point by point in explaining why tracing isn't what it's cut out to be and how meaningless is the supposed data on "ghost guns."  First, a trace only leads to the first retail purchaser, if that person can be located.  Without evidence, no reason exists to consider that person a "suspect" in whatever the crime is.  And after that first purchase, the gun may have been inherited, given as a gift, sold, lost, or stolen.

Second, criminals don't typically buy guns from a licensed dealer, and thus their acquisitions cannot be traced.  Where do criminals get their guns?  Out of 24,848 prison inmates surveyed, a Bureau of Justice Statistics study Source and Use of Firearms Involved in Crimes (2019) reported:

Off the street/underground market: 43.2%

Obtained from individual: 25.3%

Theft: 6.4%

Purchased/traded at retail source: 10.1% [only 6.9% under one's real name]

Other sources: 17.4%

The study made no mention of any of the firearms being made from kits.  Multiple studies of the sources from which criminals get their guns, going back to the 1980s, report similar results.

Third, evidence does not support the government's argument of an "urgent public safety and law enforcement crisis posed by the exponential rise of untraceable firearms…." Let's compare some numbers.  There are an estimated 500 million firearms in private hands in the United States.  The types of kits that hobbyists most often make into firearms are for AR-15 rifle types and handguns similar to Glocks.  ATF data shows that about two and a half million Glocks were introduced into commerce between 2016 and 2022.  According to the National Shooting Sports Association, there were 24 million+ modern sporting rifles (mostly AR-types) in American civilian circulation as of 2020.

Compare those numbers with the 19,000 privately-made firearms alleged to have been traced in 2021.  That's hardly a drop in the bucket.  And consider this further finding by Congress in the 2013 law cited above: "Firearms selected for tracing are not chosen for purposes of determining which types, makes, or models of firearms are used for illicit purposes. The firearms selected do not constitute a random sample and should not be considered representative of the larger universe of all firearms used by criminals, or any subset of that universe."

Other than the numbers of privately-made firearms traced, no information exists as to why they were traced.  ATF has raided companies that market kits and presumably seized their inventory, which could jack-up the statistics dramatically.  Eleven states and the District of Columbia restrict privately-made firearms, so traces generated in those places may reflect mere possessory offenses.

Based on unverified media accounts, Everytown for Gun Safety Foundation lists 187 alleged "shootings" with "ghost guns" between 2013 and 2024, for an average of about 15 per year.  But the data include accidents and suicides, not just assaults.  In any event, 15 shootings per year are a miniscule fraction of the tens of thousands of traces of "ghost guns" now being reported by ATF annually.

This is not the first time ATF has manipulated trace data for political ends.  In the 1990s, in order to justify a ban on "assault weapons," it was charged with creating the impression that criminals prefer them.  Its Forward Tracing Program entailed getting information from manufacturers on the subject firearms and "tracing" them to the retail dealers.  Then they told the public that the designated firearms were disproportionately used in crime based on them being traced so much.  I document this cooking of the books in America's Rifle, chapter 14.

It goes without saying that the issue before the Supreme Court in VanDerStok is purely legal: does ATF have authority to expand the definition of "firearm" enacted by Congress and thereby to criminalize activity that Congress did not make unlawful?  Contrary to government claims, there is no "urgent public safety and law enforcement crisis posed by the exponential rise of untraceable firearms…." But even if there is, it's a matter for Congress, not the agency, to address.

Today our team submitted an amicus brief in United States v. Trump before the Eleventh Circuit. This case is an appeal from Judge Cannon's decision declaring the appointment of the Special Counsel to be unlawful. Our brief was filed on behalf of Robert Ray, Professor Seth Barrett Tillman, and the Landmark Legal Foundation. We are grateful to Michael A. Sasso for serving as local counsel.

Tillman and Landmark joined our District Court brief. We are honored that Ray joined our effort on appeal. Ray served as one of the last Independent Counsels, replacing Kenneth W. Starr in October 1999, and was in charge of the Whitewater and Monica Lewinsky investigations. He concluded the investigations by March 2002 with the decision not to prosecute President Clinton once he left office.

Our brief makes four primary arguments:

The District Court correctly dismissed the indictment. Amici advance four rationales to support the judgment below.

First, from the 1850s through the 1950s, during six presidential administrations, Attorneys General retained outside lawyers as Special Counsels either: to assist a U.S. Attorney with prosecutions, or to assist the Attorney General with an investigation. Josh Blackman, A Historical Record of Special Counsels Before Watergate (2024), https://papers.ssrn.com/abstract=4970972 (hereinafter"A Historical Record"). And the Watergate Special Prosecutor is a thin reed to stand on. United States v. Nixon expressly and repeatedly recognized that the Watergate Special Prosecutor had "unique authority and tenure." 418 U.S. 683, 694 (1974). Further, in 1973, the Acting Attorney General, with the acquiescence of the President, granted the Special Prosecutor unsurpassed insulation against removal. Apart from those compromises, this insulation would be inconsistent with Bowsher v. Synar. 478 U.S. 714 (1986). Whether the Nixon analysis is holding or dicta, it is not controlling, and it should not be extended to today's context under today's statutory and regulatory framework.

Second, Special Counsel Jack Smith ("Smith") cannot rely on the permanent indefinite appropriation found in a "note" to 28 U.S.C. §591. In 2004, the Government Accountability Office determined that this appropriation can be used for "investigat[ing] and prosecut[ing] high ranking government officials." GAO, Special Counsel and Permanent Indefinite Appropriation, B-302582, 2004 WL 2213560, at *4 (Comp. Gen. Sept. 30, 2004). But Trump was not a "high ranking" official when he was indicted, and all the alleged conduct took place after he was out of office. In these circumstances, the funding mechanism in Section 591's note cannot be used to pay Smith.

Third, Supreme Court precedent distinguishes between officers and employees. An "Officer of the United States" position must have a duration that is continuous. Though Smith's prosecution has already continued for several years, and his duties are regular, his position is not continuous, because his extant position would not continue to a successor. Morrison v. Olson, 487 U.S. 654, 672 (1988). At most, Smith is a mere "employee" who cannot exercise the sweeping powers of a Senate-confirmed U.S. Attorney.

Finally, Amici have properly preserved for review by the Supreme Court the question of whether Morrison v. Olson should be overruled.

The Special Counsel, like the Independent Counsel, still comes as a wolf. Id. at 699 (Scalia, J., dissenting).

We look forward to this litigation proceeding.

I have posted to SSRN a new article titled, A Historical Record of Special Counsels Before Watergate. This Article is based on some extensive research into more than a dozen different archives. This Article is also relevant to the ongoing litigation over the Special Counsel, which I will address shortly.

Here is the abstract:

This Article presents a corpus of primary sources that were written by Presidents, Attorneys General, United States Attorneys, Special Counsels, and others between the 1850s and the 1950s. This corpus reproduces primary sources from more than a dozen archives to present a better legal account showing how Special Counsels were retained by Attorneys General under Presidents Buchanan, Andrew Johnson, Grant, Garfield, Theodore Roosevelt, and Truman.

During these six presidential administrations, Attorneys General retained outside lawyers as Special Counsels either: (1) to assist a U.S. Attorney with prosecutions, or (2) to assist the Attorney General with an investigation. In none of these matters did the Attorney General appoint an outside lawyer as a Special Counsel, and then delegate to him the powers now claimed by modern special counsels: all of the powers of a Senate-confirmed U.S. Attorney.

There was one outlier. In 1924, during the Coolidge Administration, Congress enacted legislation establishing Senate-confirmed special counsels to prosecute Teapot Dome Scandal defendants. These Special Counsels were afforded "total independence." It is doubtful that these positions would be consistent with the Supreme Court's modern separation of powers jurisprudence.

This practice shows that the positions of special counsels in the post-Watergate era are not analogous to the positions of special counsels in the pre-Watergate era. Thus pre-Watergate history does not provide support for the modern, post-Watergate special counsel and the vast powers that they are purportedly vested with.

I welcome comments and feedback!

This is my second installment preceding the upcoming October 8 argument in Garland v. VanDerStok, a challenge to the regulatory redefinition of the term "firearm" in the Gun Control Act.  By expanding the statutory definition, the Bureau of Alcohol, Tobacco, Firearms & Explosives (ATF) in its 2022 Final Rule purports to criminalize numerous innocent acts that Congress never made illegal.

Until the new rule, a kit with partially-machined raw material that can be fabricated into a firearm was not considered to have reached a stage that it is a "firearm."  To prevent Americans from making their own firearms from such material, which has always been and remains lawful, the bugbear term "ghost guns" was recently coined.  In its VanDerStok brief, the government argues that "anyone with basic tools and rudimentary skills" can "assemble a fully functional firearm" from such kits "in as little as twenty minutes."

As explained in my last post, that is refuted by none other than the former Acting Chief of ATF's Firearm Technology Branch, Rick Vasquez, who reviewed and approved hundreds of classifications about whether certain items are "firearms."  As he explained in his amicus brief, fabrication of a firearm from these kits is a complex process requiring skill and special tools beyond the capacity of the average person.

In this post I'll trace the statutory history of the term "firearm" to gain insight into its meaning.  The Gun Control Act defines "firearm" as "(A) 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; (B) the frame or receiver of any such weapon…."  18 U.S.C. § 921(a)(3).  An ATF regulation on the books from 1968 to 2022 defined a "frame or receiver" as "that part of a firearm which provides housing for the hammer, bolt or breechblock and firing mechanism," i.e., the main part of a firearm to which the barrel and stock attach.

ATF's Final Rule stretches these terms to mean parts, material, jigs, tools, and instructions that constitute neither an actual "firearm" nor a "frame or receiver," but may be used by a skilled person with proper tools to fabricate these items.

This new regulatory definition of "firearm" obviously conflicts with the definition enacted by Congress.  Two cases decided by the Supreme Court this year directly apply.  Per Dep't. of Agriculture Rural Dev. Rural Housing Service v. Kirtz: "When Congress takes the trouble to define the terms it uses, a court must respect its definitions as virtually conclusive."  Congress defined "firearm."  And while Congress did not explicitly define "frame or receiver," Snyder v. United States teaches that, after analyzing the statutory text, a court may look at "the statutory history, which reinforces that textual analysis."

Statutory history is a prime focus of the Amicus Curiae Brief of the National Shooting Sports Foundation, which I coauthored with Schaerr Jaffe LLP and NSSF counsel.  As the brief details, the statutory history reinforces the textual analysis.  I have covered the subject further in "Textualism, the Gun Control Act, and ATF's Redefinition of 'Firearm,'" Harvard Journal of Law & Public Policy: Per Curiam, Aug. 27, 2024.

We begin with the Federal Firearms Act of 1938 ("FFA"), 52 Stat. 1250, which defined a firearm as "any weapon, by whatever name known, which is designed to expel a projectile or projectiles by the action of an explosive … or any part or parts of such weapon."  It provided that any person who violated "any of the provisions of this Act or any rules and regulations promulgated hereunder" was subject to fines and imprisonment.  It empowered the Secretary of the Treasury to "prescribe such rules and regulations as he deems necessary to carry out the provisions of this Act."

FFA regulations required licensed manufacturers to record firearms disposed of, including "the serial numbers if such weapons are numbered."  Dealers were required to record acquisitions and dispositions.  Required records included "firearms in an unassembled condition, but not including parts of firearms."  That an "unassembled" firearm constituted a firearm in no way implied that raw material and unfinished parts were considered a firearm.

Revenue Ruling 55-175 (1955) held that "a barrel[ed] action comprised of the barrel …; front and rear stock bands; receiver with complete bolt, trigger action, magazine, etc., is a weapon, complete except for the stock, which is capable of expelling a projectile or projectiles by the action of an explosive."  One can see here the understanding of a "receiver" as the housing that holds the internal parts that would be reflected in the 1968 regulatory definition of "frame or receiver."

Apparently, the only judicial decision on the meaning of "part or parts" in the FFA was United States v. Lauchli (7th Cir. 1966), which mostly concerned dealing in unregistered machine guns.

The court held that "Browning automatic rifle magazines" were "parts" under the FFA because "such weapons could not be fired automatically without the magazines."  These finished parts contained in the machine guns were "serviceable parts, thus bringing them within the scope of the [FFA]."  This statement confirmed that items that were not "serviceable parts" were not considered "parts."

In sum, under the FFA, a "firearm" was a "weapon" designed to expel a projectile, whether assembled or unassembled.  To be a "part or parts," the items had to be serviceable.  A "receiver" housed the bolt, trigger action, and magazine.  This background demonstrates that partially completed material that had not become an actual weapon or useable parts was not considered a "firearm."

Despite recent political jargon about so-called "ghost guns," from the ratification of the Second Amendment in 1791 until 1958, no federal legislation required that anyone—even a firearm manufacturer—mark a firearm with a serial number.  Then in 1958, a regulation required manufacturers and importers to identify each firearm "by stamping … the name of the manufacturer or importer, and the serial number, caliber, and model of the firearm…. However, individual serial numbers and model designation shall not be required on any shotgun or .22 caliber rifle…."

Beginning in 1963, bills were introduced to revise the FFA that would eventually find their way into the Gun Control Act ("GCA") of 1968, the major federal law regulating firearms today.  As reflected in Senate Report No. 90-1097 (1968): "It has been found that it is impractical to have controls over each small part of a firearm.  Thus, the revised definition substitutes only the major parts of the firearm; that is, frame or receiver for the words 'any part or parts.'"

Initially, the GCA bills continued the FFA provision making violation not just of the Act, but also of any rule or regulation, a criminal offense.  In floor debate, Senator Robert Griffin objected that lawmakers "should not delegate our legislative power … in the area of criminal law," and that due process required that "we should spell out in the law what is a crime."    Likewise, Senator Howard Baker rejected "plac[ing] in the hands of an executive branch administrative official the authority to fashion and shape a criminal offense to his own personal liking." 114 Cong. Rec. 14,792 (May 23, 1968).  Making it a crime to violate a regulation was then removed from the bill.

As enacted, the GCA defined "firearm" exactly as it is defined by that statute now.  It required licensed manufacturers and importers to engrave a serial number on each frame or receiver.

Also in 1968, the Treasury Department adopted the same regulatory definition of "frame or receiver" that was retained until 2022: "That part of a firearm which provides housing for the hammer, bolt or breechblock and firing mechanism, and which is usually threaded at its forward portion to receive the barrel."  That reflected the common understanding of the meaning of those terms.  In fact, each of the terms in the definition was defined that same year in Chester Mueller & John Olson, Small Arms Lexicon and Concise Encyclopedia (1968).

Just before adopting its proposed GCA regulations in 1968, Treasury held a public hearing, the only one ever held before or since.  Not a single witness objected to the definition of a frame or receiver.  To the contrary, an industry witness praised the "very clear definition of a … receiver, something we didn't have before[.]"

If the 1968 regulation could talk, it would say: "read my lips – the frame or receiver is the 'part' that 'provides housing' for the internal parts in the present tense, not partially-machined raw material that 'could provide housing' in the future should one perform the required fabrication operations."

In deep-sixing the Chevron deference doctrine in Loper Bright Enters. v. Raimondo, the Supreme Court said that historically "respect was thought especially warranted when an Executive Branch interpretation was issued roughly contemporaneously with enactment of the statute and remained consistent over time."  That applies perfectly to the 1968 regulation, which expressed the common understanding of "frame or receiver" when Congress enacted the GCA, and remained in force for the next fifty-four years until ATF abruptly scrapped it.

In enacting the Firearm Owners' Protection Act ("FOPA") of 1986, Congress found "additional legislation" necessary "to correct existing firearm statutes and enforcement policies."  But it left intact the GCA's definition of "firearm" and expressed no dissatisfaction with ATF's definition of "frame or receiver."  It was the same result in the three subsequent times in which Congress defined certain types of firearms – the Crime Control Act of 1990, defining "semiautomatic rifle"; the Brady Act of 1993, defining "handgun"; and the Public Safety & Recreational Firearms Use Protection Act of 1994, defining "semiautomatic assault weapon" (repealed in 2004).

As the Supreme Court opined in U.S. v. Rutherford (1979), "once an agency's statutory construction has been 'fully brought to the attention of the public and the Congress,' and the latter has not sought to alter that interpretation although it has amended the statute in other respects, then presumably the legislative intent has been correctly discerned."

However, FOPA mandated that the Secretary (now the Attorney General) may prescribe "only such rules and regulations as are necessary to carry out the provisions of this chapter," deleting the prior language that "the Secretary may prescribe such rules and regulations as he deems reasonably necessary."  And yet today, ATF's Final Rule purports to expand the meaning of terms in conflict with the GCA's plain text and thereby to criminalize previously legal conduct through regulations.

In sum, the statutory history reinforces the textual analysis that the term "firearm" is limited to the exact definition that Congress enacted, and does not extend to an open-ended, undefined "parts kit" that flunks that definition.  Further, a "frame or receiver" is the main part of a firearm that provides housing for the internal parts, an understanding that has persisted over a half century. It does not include partially-machined raw material that has not been fabricated into a functional housing.

For much more on the statutory history beginning with the Federal Firearms Act of 1938 and going forward, please see my article "The Meaning of 'Firearm' and 'Frame or Receiver' in the Federal Gun Control Act: ATF's 2022 Final Rule in Light of Text, Precedent, and History."

In debates over whether universities should (or should be allowed to) consider race in admissions, it is often pointed out that many universities give preferences to the children of alumni and university donors. It is also fairly noted that such preferences may have racially disparate effects (particularly at institutions that were formerly segregated).

As a legal matter, however, race-based preferences more suspect than other admissions preferences. Under current 14th Amendment's Equal Protection Clause doctrine, the consideration of race by state institutions is inherently suspect (and this standard has been applied to private universities through federal law). Not so many other characteristics that may be used in university admissions processes -- including relationships to alumni and benefactors.

Earlier today, California Governor Gavin Newsom signed a measure to prohibit legacy and donor preferences in college admissions throughout the state. Of note, the law will apply to both public and private universities, but does not impose meaningful penalties.

From a Politico report:

California's law, which will take effect Sept. 1, 2025, is the nation's fifth legacy admissions ban, but only the second that will apply to private colleges. . . .

Like other states, California won't financially penalize violators, but it will post the names of violators on the state Department of Justice's website.

California will also add to data reporting requirements that it implemented in 2022, when private colleges had to start sharing the percentage of admitted students who were related to donors and alumni. Schools that run afoul of the new law will also have to report more granular demographic information about their incoming classes to the state, including the race and income of enrolled students as well as their participation in athletics. . . .

Public universities in California won't be affected by the change. California State University does not consider legacy or donor ties, and the University of California system stopped doing so in 1998, two years after California voters banned race-conscious admissions through a statewide ballot measure.