Today, we are presented with the question of whether the annotations contained in the Official Code of Georgia Annotated *1232 (OCGA), authored by the Georgia General Assembly and made an inextricable part of the official codification of Georgia's laws, may be copyrighted by the State of Georgia. Answering this question means confronting profound and difficult issues about the nature of law in our society and the rights of citizens to have unfettered access to the legal edicts that govern their lives. After a thorough review of the law, and an examination of the annotations, we conclude that no valid copyright interest can be asserted in any part of the OCGA.

From the earliest day of the Republic, under federal copyright law, copyright interests have vested in the author of the work. Authorship, therefore, is central to many questions that arise under the Copyright Act, 17 U.S.C. § 101 et seq . This case is no exception. In most states the "official" code is comprised of statutory text alone, and all agree that a state's codification cannot be copyrighted because the authorship is ultimately attributable to the People. Conversely, all agree that annotations created by a private party generally can be copyrighted because the annotations are an original work created by a private publisher. But the annotations in the OCGA are not exactly like either of these two types of works. Rather, they fall somewhere in between -- their legal effect and ultimate authorship more indeterminate. To resolve this question, then, we reason by analogy, and drill down on the core attributes that make the OCGA annotations what they are -- namely an exercise of sovereign power.

The general rule that legislative codifications are uncopyrightable derives from an understanding of the nature of law and the basic idea that the People, as the reservoir of all sovereignty, are the source of our law. For purposes of the Copyright Act, this means that the People are the constructive authors of those official legal promulgations of government that represent an exercise of sovereign authority. And because they are the authors, the People are the owners of these works, meaning that the works are intrinsically public domain material and, therefore, uncopyrightable.

That the law itself, whether it takes the form of a legislative enactment or of a judicial opinion, is subject to the rule is clear and not contested. This is because these works represent the quintessential exercise of sovereign power. When a legislature enacts a law, or a court writes an opinion rendering an official interpretation of the law in a case or controversy, they are undisputedly speaking on behalf of the People, who are properly regarded as the author of the work. The task we face today is whether we should similarly treat Georgia's entire official code, which expressly merges its statutes and their official annotations, as the sovereign expression of the People by their legislature, as public domain material.

To navigate the ambiguities surrounding how to characterize this work, we resort to first principles. Because our ultimate inquiry is whether a work is authored by the People, meaning whether it represents an articulation of the sovereign will, our analysis is guided by a consideration of those characteristics that are the hallmarks of law. In particular, we rely on the identity of the public officials who created the work, the authoritativeness of the work, and the process by which the work was created. These are critical markers. Where all three point in the direction that a work was made in the exercise of sovereign power -- which is to say where the official who created the work is entrusted with delegated sovereign authority, where the work carries authoritative weight, and where the work was created through the procedural channels in which sovereign *1233 power ordinarily flows -- it follows that the work would be attributable to the constructive authorship of the People, and therefore uncopyrightable.

The question is a close one -- and important considerations of public policy are at stake on either side -- but, at the end of the day, we conclude that the annotations in the OCGA are sufficiently law-like so as to be properly regarded as a sovereign work. Like the statutory text itself, the annotations are created by the duly constituted legislative authority of the State of Georgia. Moreover, the annotations clearly have authoritative weight in explicating and establishing the meaning and effect of Georgia's laws. Furthermore, the procedures by which the annotations were incorporated bear the hallmarks of legislative process, namely bicameralism and presentment. In short, the annotations are legislative works created by Georgia's legislators in the exercise of their legislative authority.

As a consequence, we conclude that the People are the ultimate authors of the annotations. As a work of the People the annotations are inherently public domain material and therefore uncopyrightable. Because we conclude that no copyright can be held in the annotations, we have no occasion to address the parties' other arguments regarding originality and fair use.

I.

A.

The Official Code of Georgia Annotated (OCGA or the Code) is an annotated compilation of Georgia statutes that has been published annually since 1982. The statutory text contained in the OCGA has been "enacted and [has] the effect of statutes enacted by the General Assembly of Georgia." O.C.G.A. § 1-1-1. As the Code itself explains, the statutory text in the OCGA is the official published version of Georgia's laws, and when the Georgia General Assembly enacts a new law, the bill typically reads "An Act... To amend... the Official Code of Georgia Annotated."

Appearing alongside the statutory text are various annotations, consisting of history lines, repeal lines, cross references, commentaries, case notations, editor's notes, excerpts from law review articles, summaries of opinions of the Attorney General of Georgia, summaries of advisory opinions of the State Bar, and other research references. The Code itself makes clear that these annotations are a part of the official Code, stating that the statutory portions of the Code "shall be merged with annotations... and [are] published by authority of the state ...and when so published [are to] be known and may be cited as the 'Official Code of Georgia Annotated.' " O.C.G.A. § 1-1-1.

Despite the fact that they are part of the official Code, Georgia law says that the annotations themselves do not have the force of law in the way that the statutory portions of the Code do. One provision of the Code explains that:

Unless otherwise provided in this Code, the descriptive headings or catchlines immediately preceding or within the text of the individual Code sections of this Code, except the Code section numbers included in the headings or catchlines immediately preceding the text of the Code sections, and title and chapter analyses do not constitute part of the law and shall in no manner limit or expand the construction of any Code section. All historical citations, title and chapter analyses, and notes set out in this Code are given for the purpose of convenient reference and do not constitute part of the law.

O.C.G.A. § 1-1-7. Laws passed during each session of the Georgia General Assembly that reenact the OCGA as the state's official code similarly provide that the annotations *1234 "contained in the Official Code of Georgia Annotated are not enacted as statutes by the provisions of this Act." See, e.g. , 2015 Ga. Laws 9 , § 54.

The annotations were initially prepared by Mathew Bender & Co., Inc., an operating division of the LexisNexis Group, (Lexis), pursuant to an agreement it entered into with the State of Georgia. Under the terms of the agreement, Lexis is responsible for the ongoing publication and maintenance of the Code, and all editorial, publication, and distribution costs. In exchange, Lexis was given the exclusive right of publication by Georgia. But, notably, Georgia holds the copyright in the annotations in its own name. The publication agreement also specifies what types of annotations should appear alongside the statutory text, and provides detailed and specific directions as to how Lexis is to generate and arrange this content. The agreement also provides that the Code Revision Commission (the "Commission") supervises the work of Lexis and has final editorial control over the contents of the OCGA.

The Commission is a body established by the Georgia General Assembly in 1977 that was originally tasked with undertaking the recodification of all of Georgia's laws, a project that had not been done since 1933. The Commission is comprised of Georgia officials, including the Lieutenant Governor, four members of the Georgia Senate, the Speaker of the Georgia House of Representatives, four additional members of the Georgia House of Representatives, and five members appointed by the president of the State Bar of Georgia. Following its successful recodification of Georgia law and the publication of the OCGA in 1982, the Commission is now responsible for updating the OCGA and supervising Lexis's editing and publication of the OCGA.

In addition to providing instructions to Lexis about how the annotations should be created, compiled, and arranged, the publication agreement establishes a number of other conditions governing the relationship between Lexis and the State of Georgia. First, the agreement requires that Lexis create a free, unannotated , online version of the Code for use by the general public. Second, the agreement limits the price that Lexis can charge for the OCGA. While other commercial annotations of the Georgia Code can cost as much as $2,570, the price of the OCGA is currently $404. Third, it grants Lexis the exclusive right to produce and sell print, CD-ROM, and online versions of the OCGA. Finally, it provides that the Commission shall receive royalties on the sale of CD-ROM and online versions of the OCGA, but shall not receive royalties from the sale of print volumes.

The publication agreement also provides that "[a]ll the contents of the Code... shall be copyrighted in the name of the State of Georgia... [and] [t]he copyrights shall cover all copyrightable parts of the Code." The Commission asserts a copyright in all portions of the OCGA except for the statutory text, which it recognizes cannot be copyrighted. Despite the copyright and the exclusive publishing rights granted to Lexis, the State of Georgia makes the CD-ROM version of the OCGA available to the general public at over 60 state and county-operated facilities throughout Georgia, such as libraries and universities. In addition, state agencies are granted the right to print and distribute or sell to the public portions of the OCGA that they are responsible for administering.

B.

Public.Resource.Org (PRO) is a non-profit organization with a mission of improving public access to government records and primary legal materials. Thus for *1235 example, PRO has been responsible for the free, online publication of all U.S. Supreme Court opinions and every post-1950 U.S. Court of Appeals opinion. PRO has also been responsible for the online publication of various state statutory codes.

In 2013 PRO purchased all 186 volumes of the print version of the OCGA and its supplements, scanned them, and uploaded them to its website to be freely accessible to the public. It also placed digital copies of the OCGA onto USB drives and mailed them to various Georgia legislators. Additionally, PRO distributed copies of the OCGA to other organizations and on other websites in order to facilitate its further dissemination by other parties.

On multiple occasions the Commission sent letters to PRO demanding that it cease and desist from publishing the OCGA on the grounds that publication infringes on the State of Georgia's copyright in the work. PRO refused to comply, arguing that there was no valid copyright in the OCGA because the law cannot be copyrighted. The Commission, acting on behalf of the Georgia General Assembly and the State of Georgia, sued PRO on July 21, 2015 in the United District Court for the Northern District of Georgia. The complaint sought injunctive relief against PRO's "widespread and unauthorized copying and distribution of the copyrighted annotations in the Official Code of Georgia Annotated through the distribution of thumb drives containing copies of the O.C.G.A. and the posting of the O.C.G.A. on various websites." On September 14, 2015, PRO filed its answer to the complaint, acknowledging its widespread publication of the OCGA, but denying that the State of Georgia holds an enforceable copyright in the Code. PRO also asserted the defense of fair use. Finally, PRO counterclaimed seeking a declaratory judgment that "the State of Georgia has no valid copyright in any portion of the O.C.G.A. because the O.C.G.A. is in the public domain."

Following briefing and argument, the district court granted the Commission's motion for partial summary judgment and denied PRO's motion. The court concluded that because the annotations in the OCGA lack the force of law, they are not public domain material. Also, it rejected PRO's other challenges to the validity of Georgia's copyright as well as its fair use defense. Soon thereafter, the district court entered a permanent injunction against PRO enjoining it "from all unauthorized use, including through reproduction, display, distribution, or creation of derivative works, of the Official Code of Georgia Annotated (O.C.G.A.)." The injunction also ordered PRO to "remove all versions of the O.C.G.A. from its website," and to cease any fundraising activities connected with PRO's publication of the OCGA.

This timely appeal ensued.

II

We review the grant of summary judgment de novo, applying the same legal standards which bound the district court. Whatley v. CNA Ins. Cos , 189 F.3d 1310 , 1313 (11th Cir. 1999). In doing so, we consider "the evidence and all factual inferences therefrom in the light most favorable to the party opposing the motion." Shaw v. Conn. Gen. Life Ins. Co. , 353 F.3d 1276 , 1282 (11th Cir 2003). Summary judgment is proper only where there is no genuine issue of material fact. Anderson v. Liberty Lobby, Inc. , 477 U.S. 242 , 248, 106 S.Ct. 2505 , 91 L.Ed.2d 202 (1986). A genuine issue of material fact exists where the dispute is "over facts that might affect the outcome of the suit under the governing law" and where the "evidence is such that a reasonable jury could return a verdict for the nonmoving party." Id . We also review a district court's decision to grant *1236 equitable relief for abuse of discretion, considering questions of law de novo and findings of fact for clear error. Preferred Sites, LLC v. Troup Cty. , 296 F.3d 1210 , 1220 (11th Cir. 2002).

In order to establish a prima facie case of copyright infringement, "a plaintiff must show that (1) it owns a valid copyright in the [work] and (2) defendants copied protected elements from the [work]." Peter Letterese And Assocs., Inc. v. World Inst. of Scientology Enters. , 533 F.3d 1287 , 1300 (11th Cir. 2008). A valid copyright registration "constitute[s] prima facie evidence of the validity of the copyright." 17 U.S.C. § 410 (c). Once the plaintiff has produced a valid copyright registration, the burden shifts to the defendant to establish that the copyright is invalid. See Latimer v. Roaring Toyz, Inc. , 601 F.3d 1224 , 1233 (11th Cir. 2010). There is no dispute that the State of Georgia has a registered copyright in the OCGA annotations. Nor do the parties contest that PRO copied the OCGA in its entirety. Thus, at the heart of this case is the question whether Georgia's copyright in the OCGA is valid; on this issue PRO carries the burden of proof.

A.

The Constitution grants Congress the power "[t]o promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries." Art. I, Sec. 8, cl. 8. Congress has exercised this power by passing the Copyright Act. 17 U.S.C. § 101 et seq . Under the Copyright Act:

Copyright protection subsists... in original works of authorship fixed in any tangible medium of expression, now known or later developed, from which they can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device.

17 U.S.C. § 102 .

As this provision makes clear, "authorship" is central to the statutory scheme. Only "original works of authorship" are eligible for copyright protection. What's more, authorship generally determines who has a possessory interest in a work. "Copyright in a work... vests initially in the author or authors of the work." 17 U.S.C. § 201 (a). Indeed, authorship allows a person to claim copyright protection regardless of whether the work has been registered with the United States Copyright Office. As we have explained, "[c]opyright inheres in authorship and exists whether or not it is ever registered." Arthur Rutenberg Homes, Inc. v. Drew Homes , Inc., 29 F.3d 1529 , 1531 (11th Cir. 1994). In consequence, to ascertain who holds a copyright in a work, we ordinarily must ascertain the identity of the author.

The meaning of authorship takes on special significance in cases like this where we consider the copyrightability of a government edict. A long line of authority, stretching back more than 180 years, establishes that, with respect to certain governmental works, the term "author" should be construed to mean "the People," so that the general public is treated as the owner of the work. This means that a work subject to the rule is inherently public domain material and thus not eligible for copyright protection. The foundations of the case law establishing this doctrine are far from clear. Few courts have fully explained the basis for this idea and the Supreme Court last addressed the question in 1888. Thus, before explaining why we construe the "author" of the OCGA to mean "the People," it's worth examining the principal cases in some detail in order to understand the considerations that guided them.

*1237 The Supreme Court first addressed whether a government edict can be copyrighted in Wheaton v. Peters , 33 U.S. (8 Pet.) 591, 8 L.Ed. 1055 (1834). The Court unanimously held that "no reporter has or can have any copyright in the written opinions delivered by this Court; and that the judges thereof cannot confer on any reporter any such right." Id . at 668. The Court was interpreting the Copyright Act of 1790, but it did not explain the foundations for the rule that "the law" was excluded from copyright protection. See id . at 593.

The Court revisited the question in Banks v. Manchester , 128 U.S. 244 , 9 S.Ct. 36 , 32 L.Ed. 425 (1888), and held that the opinions of state court judges, just like Supreme Court opinions, were not copyrightable. In Banks the Court considered an infringement suit filed by a publishing firm that had published official reports containing the decisions of the Supreme Court of Ohio against a defendant who had published the same material in the American Law Journal. Id . at 249 , 9 S.Ct. 36 . An Ohio statute provided for the appointment of an official reporter for the Supreme Court of Ohio, and tasked him with compiling the decisions and other materials authored by the judges and securing "for the benefit of the state" a copyright on the compilations. Id . at 245, 249 , 9 S.Ct. 36 . The Ohio statute also required the Secretary of State to contract with a publisher, who would be given the exclusive right to publish the reports compiled by the official court reporter "so far as the state can confer [such right]." Id . at 246 , 9 S.Ct. 36 . The plaintiff publishing firm in Banks was the chosen publisher, and, in suing, was attempting to enforce a copyright interest in the work of the Ohio judges assigned to it by the State of Ohio.

The Court found the copyright invalid. Id . at 252 , 9 S.Ct. 36 . It emphasized that under then-extant copyright law only "authors" could obtain a copyright in their work. The Court determined that the reporter who had created the compilations did not qualify as the author of the opinions or the other materials written by the judges since he had not created the works. Id . Moreover, the Supreme Court explained that "[i]n no proper sense can the judge who, in his judicial capacity, prepares the opinion or decision, the statement of the case, and the syllabus, or head-note, be regarded as their author." Id . at 253 , 9 S.Ct. 36 . Thus, the Court rested its decision on a construction of the statutory term "author" that excluded both the judges and the reporter from qualifying as authors of the material in question, which in turn meant that neither the judges nor the reporter could have conveyed a valid copyright interest to the publishing firm bringing suit.

The Court offered a number of reasons for holding that the judges could not be considered the "authors" of their work. In the first place judges "receive from the public treasury a stated annual salary, fixed by law," and therefore can "have no pecuniary interest or proprietorship, as against the public at large, in the fruits of their judicial labors." Id . Furthermore, although the Court said that it was only construing the statutory meaning of the term "author," it also acknowledged that, fundamentally, "[t]he question is one of public policy." Id . In articulating this public policy interest, the Court explained that "[t]he whole work done by the judges constitutes the authentic exposition and interpretation of the law, which, binding every citizen, is free for publication to all, whether it is a declaration of unwritten law, or an interpretation of a constitution or a statute." Id . Banks expressly relied on a ruling of the Massachusetts Supreme Judicial Court in Nash v. Lathrop , 142 Mass. 29 , 6 N.E. 559 (1886), which had similarly observed that "it needs no argument to *1238 show that justice requires that all should have free access to the opinions, and that it is against sound public policy to prevent this, or to suppress and keep from the earliest knowledge of the public the statutes, or the decisions and opinions of the justices." Id . at 35, 6 N.E. 559 .

The next, and to date last time the Supreme Court considered the rule that government edicts cannot be copyrighted came less than a month after the Court had decided Banks , in Callaghan v. Myers , 128 U.S. 617 , 9 S.Ct. 177 , 32 L.Ed. 547 (1888). There, a publisher of a set of reports containing the opinions of the Supreme Court of Illinois, known as the Illinois Reports, brought suit for copyright infringement against a rival publisher that had copied and published the reports. Id . at 619-22 , 9 S.Ct. 177 . The original publisher had obtained a proprietary interest in the reports from a salaried official of the State of Illinois whose duties, defined by statute, consisted of compiling the Illinois Reports; organizing the cases; writing annotations such as headnotes and syllabi to appear alongside the opinions in the reports; and providing a certain number of copies of the final product to the Secretary of State of Illinois. Id . at 645-46 , 9 S.Ct. 177 . Having fulfilled his statutory duties, the reporter sold whatever proprietary interest he had in the Illinois Reports to the publishing firm. When the firm sued for copyright infringement, the alleged infringer attempted to defend, claiming that the reports were public property because they had been created by a state-employed reporter who could himself have no proprietary interest in the work since he created the reports as part of his public duties and therefore was not their "author." Id . at 645-47 , 9 S.Ct. 177 .

The Court began its analysis by reinforcing the basic rule announced in Banks that "there can be no copyright in the opinions of the judges, or in the work done by them in their official capacity as judges." Id . at 647 , 9 S.Ct. 177 . Nevertheless it rejected the claim that the copyright in the Illinois Reports was invalid. It explained that the underlying rationale of Banks did not apply, observing that "there is no ground of public policy on which a reporter who prepares a volume of law reports, of the character of those in this case, can... be debarred from obtaining a copyright for the volume which will cover the matter which is the result of his intellectual labor." Id . The Court further suggested that, since the court reporter was a "sworn public officer, appointed by the authority of the government... [and] paid a fixed salary for his labors," the state government might have taken any proprietary interest in his work for itself, but the fact that it had not done so suggested that there was "a tacit assent by the government to his exercising such privilege" on his own. Id . The Court thus reasoned that federal copyright law as explicated in Banks did not prevent the reporter from holding a valid copyright in the work and that the state had not reserved the copyright to itself. As a result, the copyright the reporter obtained and conveyed to the publishing firm was valid. The compilation of judicial decisions and other explanatory material like headnotes, tables, and indices, was different from Banks in two ways: first, the reporter, who had been appointed by the Illinois Supreme Court, and not the judges, had written the material accompanying the opinion; and, second, the reporter, and not the State of Illinois, claimed to hold the copyright.

The Supreme Court has not examined the doctrine since it decided Callaghan in 1888. However, since Banks and Callaghan the lower courts have further explored the nature and application of the rule. Thus, for example, the Sixth Circuit, in an opinion authored by Justice Harlan, applied the rule to state statutes.

*1239 Howell v. Miller , 91 F. 129 (6th Cir. 1898). The Fifth Circuit has extended the rule to encompass regulatory materials. Veeck v. S. Bldg. Code Cong. Int'l, Inc. , 293 F.3d 791 (5th Cir. 2002) (en banc). However, other courts have declined to extend the rule in other, related contexts. See, e.g. , CCC Info. Servs., Inc. v. Maclean Hunter Mkt. Reports, Inc. , 44 F.3d 61 (2d Cir. 1994) (declining to apply the rule to a privately prepared listing of automobile values that several states required insurance companies to use in calculating insurance payouts); Practice Mgmt. Info. Corp. v. Am. Med. Ass'n , 121 F.3d 516 (9th Cir. 1997), amended, 133 F.3d 1140 (9th Cir. 1998) (declining to apply the rule to a privately authored coding system that was incorporated into a government reimbursement scheme through publication in the Federal Register); Cty. of Suffolk v. First Am. Real Estate Sols. , 261 F.3d 179 , 193 (2d Cir. 2001) (declining to apply the rule to tax maps created by a county assessor's office); John G. Danielson, Inc. v. Winchester-Conant Properties, Inc. , 322 F.3d 26 (1st Cir. 2003) (declining to apply the rule to the terms of a restrictive covenant a town entered into as part of a zoning scheme).

It is also worth observing that Congress has partially codified the rule announced in Banks . Specifically, the 1909 version of the Copyright Act provided that "no copyright shall subsist in the original text of any work which is in the public domain... or in any publication of the United States Government, or any reprint, in whole or in part, thereof." 17 U.S.C. § 8 (repealed 1976). This prohibition persists under current copyright law, enacted in 1976, which, in turn, provides that "[c]opyright protection under this title is not available for any work of the United States Government." 17 U.S.C. § 105 .This partial codification of Banks for works created by the federal government leaves unmodified the rule as it applies to works created by the states. As the Copyright Office's 1961 Register's Report stated, even though Congress enacted a prohibition that only applies to the federal government, "the judicially established rule [ ] still prevent[s] copyright in the text of state laws, municipal ordinances, court decisions, and similar official documents." 1961 Register's Report, at 129-30.

Although case precedent and congressional enactments have long established the rule that government works are not copyrightable, the foundations of the rule are generally implicit and unstated. Since the Court in Banks was not especially clear about the legal source of the rule it had announced and since the issue has not been raised before in our Court, we start with a relatively clean canvas. What is clear, however, is that the rule enunciated in Banks was grounded on the Court's interpretation of the term "author" in the Copyright Act of 1790, that works created by courts in the performance of their official duties did not belong to the judges, and that public policy compelled the conclusion that these works were in the public domain and uncopyrightable.

Thus, we understand the rule in Banks to derive from first principles about the nature of law in our democracy. Under democratic rule, the People are sovereign, they govern themselves through their legislative and judicial representatives, and they are ultimately the source of our law. Under this arrangement, lawmakers and judges are draftsmen of the law, exercising delegated authority, and acting as servants of the People, and whatever they produce the People are the true authors. When the legislative or judicial chords are plucked it is in fact the People's voice that is heard. Not surprisingly, then, for purposes of copyright law, this means that the People, as the constructive authors are also the owners of the law. And in this way, any *1240 work of which the People are the constructive authors is intrinsically public domain material and is freely accessible to all so that no valid copyright can ever be held in it.

The concept of popular sovereignty is deeply rooted in our politics, our law, and our history. The seminal statement of America's political creed boldly proclaims that "[g]overnments ... deriv[e] their just powers from the consent of the governed." THE DECLARATION OF INDEPENDENCE para. 2 (U.S. 1776). During the ratification debates that followed the Revolution, James Madison similarly began with the foundational idea that the People were sovereign, and that under the proposed form of government "the public voice" was "pronounced by the representatives of the people." THE FEDERALIST No. 10 at 77 (James Madison) (Clinton Rossiter ed., 1961). Still again, in the midst of the Civil War, President Lincoln etched an indelible description of this form of government in the national memory, describing ours as a "government of the people, by the people, for the people." Abraham Lincoln, Gettysburg Address (November 19, 1863).

In fact, the United States Reports are filled with invocations of the sovereignty of the People. As Chief Justice Marshall expressed the fundamental idea many years ago: "[t]he government proceeds directly from the people; is 'ordained and established,' in the name of the people... [and] is emphatically and truly, a government of the people. In form, and in substance, it emanates from them. Its powers are granted by them, and are to be exercised directly on them, and for their benefit." M'Culloch v. Maryland , 17 U.S. (4 Wheat.) 316, 403-05, 4 L.Ed. 579 (1819) ; see also Chisholm v. Georgia , 2 U.S. (2 Dall.) 419, 1 L.Ed. 440 (1793) ; Luther v. Borden , 48 U.S. (7 How.) 1, 12 L.Ed. 581 (1849). See also TOCQUEVILLE, DEMOCRACY IN AMERICA 53 (Mansfield ed. 2002). ("In America, the principle of the sovereignty of the people... is recognized by mores, proclaimed by laws; [ ] spreads with freedom and reaches its final consequences without obstacle... when one wants to speak of the political laws of the United States, it is always with the dogma of the sovereignty of the people that one must begin.").

While Banks is not explicit in grounding its holding in this conception of sovereignty, other federal courts have ruled that government works are intrinsically public domain material precisely because the People are sovereign and are therefore the authors and owners of the law. Thus, for example, in Banks & Bros. v. W. Pub. Co. , 27 F. 50 (C.C.D. Minn. 1886), the court justified the rule on the grounds that "[e]ach citizen is a ruler,- a law-maker,- and as such has the right of access to the laws he joins in making and to any official interpretation thereof. If the right of property enters into the question, he is a part owner, and as such cannot be deprived of equal access by his co-owners." Id . at 57 .

In the same vein, and more recently, several courts have applied the rule announced in Banks and understood the rule to rest on foundational principles about the nature of law in a democratic society. Thus, in Veeck , the Fifth Circuit, sitting en banc, confronted the question of whether a model building code, once adopted by two municipalities, lost its copyright protection. Veeck , 293 F.3d at 796 . In concluding that the work was uncopyrightable, the court asserted as a basic principle that the law is in "the public domain and thus not amenable to copyright," and that cases like Wheaton and Banks evince a "broad understanding of what constitutes 'the law' " so as to make judicial opinions in addition to statutes ineligible for copyright protection. Id . at 795-96 . On this basis, the court held that, "[a]s governing law," the municipal building codes also could not be copyrighted. Id . at 796 .

*1241 The court went on to explain that its holding rested on a deeper principle, a "metaphorical concept of citizen authorship." Id . at 799 . As the court reasoned, "[l]awmaking bodies in this country enact rules and regulations only with the consent of the governed. The very process of lawmaking demands and incorporates contributions by 'the people,' in an infinite variety of individual and organizational capacities... In performing their function, the lawmakers represent the public will, and the public are the final 'authors' of the law." Id . The court discerned that there are strong public policy interests in giving the public unfettered access to the law. "[P]ublic ownership of the law means precisely that 'the law' is in the 'public domain' for whatever use the citizens choose to make of it. Citizens may reproduce copies of the law for many purposes, not only to guide their actions but to influence future legislation, educate their neighborhood association, or simply to amuse." Id . Thus, the "metaphorical concept of citizen authorship together with the need for citizens to have free access to the laws are the ultimate holding of Banks ." Id . (quotation omitted).

The First Circuit has also emphasized popular sovereignty as being foundational to its understanding of the rule announced in Banks . In BuildingOfficials & Code Administrators v. Code Technology, Inc. , 628 F.2d 730 (1st Cir. 1980), the court considered, on an interlocutory appeal challenging the issue of a preliminary injunction, a copyright infringement suit brought by the private sector author of a model building code against a publisher of the Massachusetts building code, which the Massachusetts legislature had based in large measure on the model code. The court ruled that the inclusion of the otherwise copyrightable model building code in the official Massachusetts building code likely rendered those materials, just like the rest of the materials in the Massachusetts building code, "freely available for copying by anyone." Id . at 732 .

After reviewing case precedent going as far back as Wheaton , a panel of the First Circuit asserted that "[t]he law thus seems clear that judicial opinions and statutes are in the public domain and are not subject to copyright." Id . at 734 . The court reasoned that this principle extends to regulatory codes as much as it does to statutes and judicial opinions. While acknowledging that cases like Banks and Wheaton seemed to rest in part on the identity of the creators of the works in question, namely salaried public officials performing official duties, it explained that a more fundamental principle was at work. In particular, "citizens are the authors of the law, and therefore its owners, regardless of who actually drafts the provisions, because the law derives its authority from the consent of the public, expressed through the democratic process." Id . The reason why judges and legislators cannot copyright works they create, was not because they are working for the government rather than for themselves, but rather because of a "metaphorical concept of citizen authorship," which means that, once it adopts a text as law, the body politic becomes the author of the work in question, leaving the original drafter with no proprietary interest. Id . The court reasoned that this was true even where the original creator of the work was a private sector actor. 1

*1242 III.

The ultimate inquiry posed by the rule in Banks is thus whether a work is attributable to the constructive authorship of the People, which is to say whether it was created by an agent of the People in the direct exercise of sovereign authority. Statutes and judicial opinions are the most obvious examples of what falls within the ambit of the rule. See Veeck , 293 F.3d at 796 (" Banks represents a continuous understanding that 'the law,' whether articulated in judicial opinions or legislative acts or ordinances, is in the public domain and thus not amenable to copyright.")

This does not mean that statutes, judicial opinions, and other texts that carry the clear force of law are the only works that may be subject to the rule. For one thing, relying, as the district court did, on a bright line distinction between edicts that have the force of law and those that do not to apply the Banks rule simply does not work in some cases. This is one of them. It is clear to us that there exists a zone of indeterminacy at the frontier between edicts that carry the force of law and those that do not. See Jean v. Nelson , 711 F.2d 1455 , 1480-83 (11th Cir. 1983), on reh'g , 727 F.2d 957 (11th Cir. 1984), aff'd, 472 U.S. 846 , 105 S.Ct. 2992 , 86 L.Ed.2d 664 (1985). In this small band of cases a government work may not be characterized as law, and yet still be so sufficiently law-like as to implicate the core policy interests undergirding Banks .

Statutory texts are the kinds of works most obviously subject to the rule announced in Banks . Because statutes are the prototypical works to which the rule applies, we rely on the statutory example as the lodestar for our inquiry. Whether or not a work is subject to the rule is dependent on whether the work is the law, or sufficiently like the law, so as to be deemed the product of the direct exercise of sovereign authority, and therefore attributable to the constructive authorship of the People. Basing the inquiry on whether a work is similar enough to the law so as to be attributable to the People, of course, does little to diminish the difficulty of applying the Banks rule in the unique circumstances presented here. See John G. Danielson, Inc. v. Winchester-Conant Properties, Inc. , 322 F.3d 26 , 38 (1st Cir. 2003) ("[The] straightforward general rule [of Banks ] has proven difficult to apply when the material in question does not fall neatly into the categories of statutes or judicial opinions."). But it does point us toward the right way of structuring our analysis.

Put simply, there are certain things that make the law what it is. The law is written by particular public officials who are entrusted with the exercise of legislative power; the law is, by nature, authoritative; and the law is created through certain, prescribed processes, the deviation from which would deprive it of legal effect. Each of these attributes is a hallmark of law. These characteristics distinguish written works that carry the force of law from all other works. Since we are concerned here with whether a work is attributable to the *1243 constructive authorship of the People, these factors guide our inquiry into whether a work is law or sufficiently law-like so as to be subject to the rule in Banks .

An analysis of these factors yields the conclusion that the annotations in the OCGA, while not having the force of law, are part and parcel of the law. They are so enmeshed with Georgia's law as to be inextricable. The annotations are themselves law-like insofar as we examine who made them, how they were made, and the role they play in the legislative and jurisprudential spheres of Georgia's public life. In consequence, they too represent a work, like the statutes themselves, that is constructively authored by the People. They are therefore uncopyrightable.

A.

First, and of critical importance to our analysis is that the Georgia General Assembly is the driving force behind their creation. The Code Revision Commission exerts authoritative influence over the creation of the annotations and the Commission indisputably is an arm of the General Assembly. Thus, just as the uncopyrightable works in Banks were created by the Ohio Supreme Court, the annotations are, in a powerful sense, a work created by the Georgia state legislature.

While it is true that the annotations were initially prepared by a private party, in this case Lexis, it is also the case that Lexis drafts the annotations pursuant to highly detailed instructions contained in the contract it entered into with the Code Revision Commission. In particular, the publication agreement not only lists the types of materials that Lexis must include in the OCGA, but also provides punctiliously specific instructions on how these materials are to be prepared. Thus, by way of example, in addition to instructing Lexis to include annotations summarizing court decisions that are relevant to various statutory provisions in the OCGA, the publication contract tells Lexis which court decisions to include. Moreover, the contract specifies the content of these summaries, instructing Lexis to include discussion of those portions of judicial opinions that involve "direct constructions" of a statute, including "constructions concerning constitutionality, purpose, intent, and the meaning of words and phrases as well as illustrations as to what a particular provision applies and to what a particular provision does not apply." Leaving even less to Lexis's independent judgment, the contract also instructs Lexis what not to include in the judicial summaries, ordering Lexis's editors to "avoid long factual annotations where they do not bear directly upon the statute involved." Further, the agreement tells Lexis the order in which the various case annotations are to be arranged.

The annotations containing summaries of judicial opinions are not the only ones for which the publication contract provides highly specific directions. The agreement also requires Lexis to include research references in the annotations, and names the specific reference sources that must be included. Similarly, the contract directs Lexis to include annotations dealing with legislative history and specifies just how far back into a statutory provision's history the annotations may go.

In addition to providing detailed instructions that guide the creation of the OCGA annotations, the Commission acts in a supervisory capacity as well, monitoring Lexis's work throughout the process. The contract says that the annotations are prepared under the "direct supervision" of the Commission. The contract spells out in some detail what this supervision means. In addition to including the research references listed in the publication agreement, Lexis is required to "include any new [references]... as required by the Commission." Sections of the agreement dealing *1244 with other annotations similarly allow the Commission to direct the inclusion of new material. Indeed, the very first section of the agreement states that the OCGA shall include, in addition to the various, specified annotations, "other material related to or included in such Code at the direction of the Commission."

Finally, the publication agreement describes in detail how the Commission is to give its final assent to the annotations. First, as for each type of annotation, the agreement affirms the Commission's role in approving Lexis's work. Thus, with respect to the summaries of judicial opinions, the agreement provides that "the form of the annotations shall be subject to the approval of the Commission." The agreement contains similar provisions with respect to the other annotations. More generally, the agreement provides that the "ultimate right of editorial control over all material contained in the Code shall be in the Commission, and in the event of any disagreement between the Commission and the Publisher over the material to be included, the decision of the Commission shall control." A separate provision of the agreement similarly provides that in the event of any disagreement "the Commission shall prevail." Moreover, the agreement requires that the Commission have an opportunity to conduct pre-publication review of all subsequent supplements, replacement volumes, and other updates to the OCGA.

In short, the Commission exercises direct, authoritative control over the creation of the OCGA annotations at every stage of their preparation. The Commission provides initial instructions to Lexis, directly supervises Lexis's work throughout the preparation process, and must give its final editorial assent to the annotations before they can become part of the OCGA. In this way, the Commission undeniably controls the creation of the OCGA annotations.

The Commission's intimate involvement in the creation of the annotations is of great significance. This is because a close examination of the nature of the Commission confirms that it is for all intents and purposes an arm of the Georgia General Assembly. As we've noted, the Commission is composed of fifteen members, nine of whom are sitting members of the Georgia General Assembly, along with the Lieutenant Governor of the State. Further, funding for the Commission comes directly from appropriations "provided for the legislative branch of state government." O.C.G.A. § 28-9-2. In addition, Georgia law provides that "[t]he Office of Legislative Counsel shall serve as staff for the commission." O.C.G.A. § 28-9-4. This is notable because, under Georgia law, the Office of Legislative Counsel is tasked with providing various advisory and legal services "for the legislative branch of government" and is therefore properly seen as an adjunct to the General Assembly. O.C.G.A. § 28-4-3. Thus, not only is the Commission funded by legislative branch appropriations, but its staff is drawn from an office that is itself an agency of the Georgia General Assembly.

Further confirming the Commission's deep connection to the Georgia General Assembly, the Georgia Supreme Court has held that the Commission's work is properly characterized as "legislative" in nature, and that it is therefore proper for the Commission to be largely composed of officials from the legislative branch. Harrison Co. v. Code Revision Comm'n , 244 Ga. 325 , 260 S.E.2d 30 (1979). Thus, in light of how it is funded and staffed, and since its work is legislative in nature, it is abundantly clear that the Commission is a creation and an agent of the Georgia General Assembly.

Indeed, the connection between the Commission and the elected legislators *1245 who make up the General Assembly is so close that the Commission may be properly regarded as one in the same with the legislators for our purposes. As the Supreme Court has explained in another context, "it is literally impossible, in view of the complexities of the modern legislative process... for [legislators] to perform their legislative tasks without the help of aides and assistants...the day-to-day work of such aides is so critical to the Members' performance that they must be treated as the latter's alter egos." Gravel v. United States , 408 U.S. 606 , 616-17, 92 S.Ct. 2614 , 33 L.Ed.2d 583 (1972). In consequence, the Court has held that legislative immunity "applies not only to a Member but also to his aides insofar as the conduct of the latter would be a protected legislative act if performed by the Member himself." Id . at 618 , 92 S.Ct. 2614 ; see also Ellis v. Coffee Cty. Bd. of Registrars , 981 F.2d 1185 , 1192 (11th Cir. 1993) ("To the extent that a legislator is cloaked with legislative immunity, an adjunct to that legislative body possesses the same immunity."). "The test for applicability of this derivative legislative immunity is whether the legislator, counsel or aide was engaged within a legitimate sphere of legislative activity." Id .

The basic intuition underlying cases applying the Speech and Debate Clause seems to us equally instructive in identifying which entity in the Georgia state government is the creative force behind the OCGA annotations. While the Commission's staff and six of its fifteen members are not Georgia legislators, the Commission is plainly an adjunct of the General Assembly. As we have detailed, its staff, funding, and responsibilities all fall under the legislative umbrella. The Commission is therefore, in a real sense, the "alter ego" of the General Assembly, meaning that the creative force behind the annotations are Georgia's elected legislators. Acting through the Commission, the legislators closely supervise and direct the production of the annotations.

Moreover, and of even greater importance to our analysis, the OCGA annotations, once completed, are subject to the approval not only of the Commission, but also to the approval of the Georgia General Assembly. The General Assembly actually votes (and must vote) to make the OCGA the official codification of Georgia's laws and, in doing so, also votes to incorporate the annotations as part of the OCGA. O.C.G.A. § 1-1-1 ("The statutory portion of such codification shall be merged with annotations, captions, catchlines, history lines, editorial notes, cross-references, indices, title and chapter analyses, and other materials pursuant to the contract and shall be published by authority of the state pursuant to such contract and when so published shall be known and may be cited as the 'Official Code of Georgia Annotated.' "). In other words, the OCGA annotations are not only authored at the direction and under the close supervision of the Georgia General Assembly, but they also obtain their peculiar status as official annotations because they are adopted annually by the General Assembly.

That Georgia's legislators are in a very real way the creators of the annotations is a powerful indication that the annotations are subject to the Banks rule. To begin, it is apparent that the rule established by Banks that government edicts cannot be copyrighted, as applied to the works of state governments, is more limited than the statutory prohibition on copyright protection for works of the federal government. As we have explained, § 105 states that "[c]opyright protection... is not available for any work of the United States Government," and § 101 defines a "work of the United States Government" as "a work prepared by an officer or employee of the United States Government *1246 as part of that person's official duties." Thus, under this prohibition, the work of any federal employee, made in his capacity as a government employee, is uncopyrightable. See, e.g. , Scherr v. Universal Match Corp. , 417 F.2d 497 (2d Cir. 1969) ; United States v. First Tr. Co. of St. Paul , 251 F.2d 686 , 690 (8th Cir. 1958) ; Pub. Affairs Assocs., Inc. v. Rickover , 268 F.Supp. 444 , 448 (D.D.C 1967). By contrast, the rule in Banks is more circumscribed, applying to a limited subclass of government works. Thus, some works made by state employees, that would be subject to § 105 if made by a federal employee, are nevertheless copyrightable under Banks . See, e.g. , Callaghan , 128 U.S. at 645-46 , 9 S.Ct. 177 (upholding the validity of a copyright in the work created by a state employee that was created pursuant to his statutorily imposed duties); County of Suffolk , 261 F.3d at 193 (declining to apply the rule in Banks to tax maps created by a county assessor's office).

The reasoning of Banks points to why the rule it has announced is applicable to a more limited class of public officials than those governed by § 105's prohibition. The Court in Banks explained, "[i]n no proper sense can the judge who, in his judicial capacity, prepares the opinion or decision, the statement of the case, and the syllabus, or head-note, be regarded as their author or their proprietor...Judges, as is well understood, receive from the public treasury a stated annual salary, fixed by law, and can themselves have no pecuniary interest or proprietorship, as against the public at large, in the fruits of their judicial labors... The whole work done by the judges constitutes the authentic exposition and interpretation of the law, which, binding every citizen, is free for publication to all." Banks , 128 U.S. at 253, 9 S.Ct. 36 . Thus, like § 105, the Banks decision emphasizes the fact that judges are producing works in their capacity as employees, but it also goes further than § 105 and emphasizes that judges are unique among government employees. In addition to receiving "from the public treasury a stated annual salary," judges are empowered to create "authentic exposition[s] and interpretation[s] of the law, which[ ] bind[ ] every citizen." Id .

As a result, the mere fact that a work was created by a state-paid employee in his capacity as an employee is not enough to trigger the rule in Banks . Something more is needed. Specifically, the government official must be entrusted with unique powers beyond those possessed by the typical government employee, such as the power to pronounce official interpretations of the law.

In short, it is clear that the rule in Banks is not concerned, as § 105 is, with the works of all government employees, but rather only with the works of certain government employees, which is to say government employees who are possessed of particular powers, namely the ability to promulgate official, binding edicts. This distinction between the rules is no doubt attributable to the difference in their underlying rationales. Section 105's prohibition is justified on the grounds that the public paid for the work and is therefore entitled to access it, and because wide dissemination of federal government materials strengthens democratic discourse. See Scherr , 297 F.Supp. at 110 ("[The]fundamental purpose underlying the prohibition [ ] is based on the necessity of wide public dissemination of the contents of materials produced by and relating to issues and problems of national interest, which policy is unquestionably a desirable one in a democracy, much of whose success is dependent on a well-informed public.") (quotations omitted and alterations adopted); Hearings on H.R. 4347, H.R. 5680, H.R. 6831, H.R. 6835, before Subcomm. No. 3 of the House Comm. on the Judiciary, 89th Cong., 1st Sess. 1924 (1965)

*1247 (statement of Sen. Russell Long) ("The original and continuing purpose of this prohibition is to assure maximum availability and dissemination of informational material prepared by or for the Government at the expense of the public.").

On the other hand, the rule in Banks derives more directly from the concept of popular sovereignty. As a result, while § 105 is concerned with any work created by a federal employee, since all government works are paid for by the taxpayer and, as a policy matter, are potentially useful to conscientious and informed citizens, the rule in Banks is concerned with works created by a select group of government employees, because only certain public officials are empowered with the direct exercise of the sovereign power. 2

This explains why the state-paid court reporter acting pursuant to his statutory duties in Callaghan did not run afoul of the rule in Banks and could hold a valid copyright in his work even though the work he created likely would fall within § 105's prohibition if he had been a federal employee. See Callaghan , 128 U.S. at 645-47 , 9 S.Ct. 177 . Though paid by the state, and acting pursuant to his official duties, the court reporter was tasked with essentially administrative and clerical responsibilities, to wit compiling and summarizing judicial decisions, rather than the promulgation of binding legal edicts. Id . at 646 , 9 S.Ct. 177 . There was therefore "no ground of public policy" standing in the way of his works' copyrightability. Id . at 647 , 9 S.Ct. 177 .

In contrast, the judges in Banks , when considered in their relationship to the sovereignty of the People, fulfill a different function than the court reporter in Callaghan . Legislators and judges, unlike other government workers, are peculiarly entrusted with the exercise of sovereign power to write or officially interpret the law. Since the power to make law rests ultimately and exclusively with the People, the primary, official duty of lawmakers and judges is therefore to act as agents of the People. While government workers like the reporter in Callaghan might also be said to be engaged in conducting the People's business, their relation to the exercise of sovereign power is more attenuated. As a result, if a government work is created by a public official who is so empowered, it is substantially more likely that the work is constructively authored by the people. 3

In light of these considerations, that the Georgia General Assembly is the driving *1248 force behind and ultimately adopts the OCGA annotations is significant. Like the Ohio Supreme Court in Banks , the Georgia General Assembly is not simply composed of ordinary government employees but rather of public officials whose official duties peculiarly include the direct exercise of sovereign power. See Ga. Const. Art. III, § I, Para. I ("The legislative power of the state shall be vested in a General Assembly which shall consist of a Senate and a House of Representatives."). Of the many government workers employed by the state of Georgia, the creators of the OCGA annotations are unique insofar as they are entrusted by the sovereign with legislative power.

This is not to say that every work produced by a legislative body is automatically uncopyrightable. As we detail below, still more is necessary to demonstrate that the OCGA annotations are the kind of work that is attributable to the constructive authorship of the People. However, because the OCGA annotations were created by public officials entrusted with sovereign, legislative authority, just like the opinions in Banks were created by justices on the Ohio Supreme Court entrusted with sovereign, judicial authority, this weighs in favor of a determination that the OCGA annotations belong in the public domain.

B.

We are also persuaded because, while not carrying the force of law in the way that the statutory portions of the OCGA do, the annotations are "law-like" in the sense that they are "authoritative" sources on the meaning of Georgia statutes. Having been merged by the General Assembly with the statutory text into a single, unified edict, stamped with the state's imprimatur, and created and embraced by the same body that wrote the text that they explicate, the annotations have been suffused with powerful indicia of legal significance that is impossible to ignore. The annotations cast an undeniable, official shadow over how Georgia laws are interpreted and understood. Indeed, Georgia's courts have cited to the annotations as authoritative sources on statutory meaning and legislative intent. The annotations' authoritativeness makes them closely analogous to the types of works that ordinarily represent an exercise of sovereign authority. The nature of the work, like the identity of its creator, therefore impels us further toward the conclusion that these annotations are attributable to the constructive authorship of the People.

The nature of the OCGA annotations is spelled out in some detail by Georgia's General Assembly. While disclaiming any legal effect in the annotations, the Georgia law providing for the creation of the OCGA also states that the "statutory portion of such codification shall be merged with annotations, captions, catchlines, history lines, editorial notes, cross-references, indices, title and chapter analyses, and other *1249 materials." O.C.G.A. § 1-1-1 (emphasis added). This language is telling. In various dictionaries, the word "merge" is defined as meaning to combine or unite, often in such a way that the constituent elements of the merger lose their distinct identity or characteristics and become one. The Random House Dictionary of the English Language defines "merge" as "to lose or cause to lose identity by uniting or blending" and "to combine or unite into a single unit." RANDOM HOUSE DICTIONARY OF THE ENGLISH LANGUAGE 550 (1980). Similarly, Webster's Third New International Dictionary defines "merge" as "to become combined into one" and to "lose identity by absorption or intermingling." WEBSTER'S THIRD NEW INTERNATIONAL DICTIONARY 1414 (1981). And the Oxford English Dictionary variously defines "merge" as "to be absorbed and disappear, to lose character or identity by absorption into something else; to join or blend," and "to combine to form a single entity." OXFORD ENGLISH DICTIONARY (3rd ed. 2001). The use of the word "merge" thus carries with it strong connotations of unification or combination of disparate elements into a single whole in which the previously distinct attributes of each element become intermingled and shared.

The question then becomes, what is the nature of the new thing created when the Georgia General Assembly explicitly chose to merge the annotations with statutory text? Here too Georgia law supplies an answer. In particular, Georgia law provides that the merged text "shall be published by authority of the state ... and when so published shall be known and may be cited as the 'Official Code of Georgia Annotated.' " O.C.G.A. § 1-1-1. Thus, the product of the merger is an official state publication, labelled and cited as the authoritative embodiment of the laws of the State of Georgia.

It of course remains true that portions of the OCGA clearly carry the force of law while O.C.G.A. § 1-1-7 disclaims any legal effect in the annotations. Yet the significance of the legislature's decision to "merge" these two things into a single edict remains. The Georgia legislature was not required to merge the annotations with the statutes in order to create the OCGA, which it then stamped with the imprimatur of the State. But the bicameral legislature chose to do so. By combining these two components into a unified whole, their attributes have been intermingled and their distinct character altered. While this does not mean that the annotations, by virtue of appearing alongside statutory text, are suddenly possessed of binding legal effect, it does mean that their combination with the statutory text imbues them with an official, legislative quality.

The statutory text, having been merged with these legislatively authored expositions on the meaning of Georgia law, must be read in pari materia with them. The annotations' combination with the statutes means that any understanding of the statutory text arrived at without reference to the annotations is axiomatically incomplete. Because Georgia law tells us that the official codification of Georgia statutes contains not only statutory text but also annotations that have been combined and unified with the statutory text into a single edict, a full understanding of the laws of Georgia necessarily includes an understanding of the contents of the annotations. In this way, the annotations are clearly laden with legal significance.

Their significance is strengthened further by the legislature's decision to label the unified whole "Official." The OCGA is not simply one of a number of competing annotated codifications of Georgia laws. It does not stand on equal footing with West's annotated Georgia code. Rather, it is the official codification of Georgia laws, stamped with the imprimatur of the state.

*1250 This status necessarily causes the annotations to cast a long shadow over how the statutory portions of the OCGA are understood. Because these are the official comments to the Code, they are to be read as authoritative in a way that annotations ordinarily are not.

Indeed, demonstrating the importance of the state's decision to stamp the OCGA with its imprimatur, the very first annotation in the very first section of the OCGA favorably cites to a court case that warns that "[a]ttorneys who cite unofficial publication of 1981 Code do so at their peril." O.C.G.A. § 1-1-1 (citing State of Ga. ex rel. General Assembly of Ga. v. Harrison Co. , 548 F.Supp. 110 (N.D. Ga. 1982) ). Similarly, the importance the Georgia legislature attached to its branding of the Code as "Official" is further demonstrated by its enactment of a law allowing the publisher of the "official Code... to use the state emblem on the cover of the publication," whereas all other private parties are prohibited from using the state emblem in any context. O.C.G.A. § 50-3-8. Thus, while stamping the annotations with the state's imprimatur and labelling it official does not suddenly elevate the annotations to the status of binding law, it too enhances their already potent cachet in a way that is undeniable and also impossible to ignore.

Moreover, as we have already noted, the annotations are not simply adopted by the legislature as an official reference work, but also, in a very meaningful sense, are written by the General Assembly -- a fact that further accentuates their legal significance. The annotations are not merely expositions on the meaning of statutes, but rather are official comments authored by the same body that also wrote the statutes. Thus, it would be only natural for the citizens of Georgia to consider the annotations as containing special insight into the meaning of the statutory text, and to therefore confer upon the annotations a special status. Cf . Stigars v. State , 674 A.2d 477 , 483 (Del. 1996) ("In the search for legislative intent, considerable weight is given to an official commentary written by the drafters of the statute."); Horenkamp v. Van Winkle And Co. , 402 F.3d 1129 , 1132 (11th Cir. 2005) ("Although not binding, the interpretations in the Advisory Committee Notes [in the Federal Rules of Civil Procedure] are nearly universally accorded great weight in interpreting federal rules."); Tome v. United States , 513 U.S. 150 , 167, 115 S.Ct. 696 , 130 L.Ed.2d 574 (1995) (Scalia, J., dissenting) ("Having been prepared by a body of experts, the [official Notes to the Federal Rules of Evidence] are assuredly persuasive scholarly commentaries-ordinarily the most persuasive-concerning the meaning of the Rules."); Schiavone v. Fortune , 477 U.S. 21 , 31, 106 S.Ct. 2379 , 91 L.Ed.2d 18 (1986) ("Although the Advisory Committee's comments [to the Federal Rules of Civil Procedure] do not foreclose judicial consideration of the Rule's validity and meaning, the construction given by the Committee is 'of weight.' "); Auer v. Robbins , 519 U.S. 452 , 461, 117 S.Ct. 905 , 137 L.Ed.2d 79 (1997) (giving substantial deference to an agency's interpretation of a regulation that the agency itself authored).

Our view is reinforced by an examination of how the annotations have been treated by Georgia's courts. In particular, the state courts frequently have characterized OCGA comments as conclusive statements about statutory meaning and legislative intent. See, e.g. , Jackson v. S. Pan & Shoring Co. , 258 Ga. 401 , 369 S.E.2d 239 (1988) (explaining that "[t]he express intent of [the statutory provision] ... is set out in the Comment to O.C.G.A. § 14-2-86"); Cox v. Fowler , 279 Ga. 501 , 614 S.E.2d 59 (2005) (citing OCGA comments as showing the "legal effect" of and "the General Assembly's intention" with respect to a statutory provision);

*1251 Prodigy Centers/Atlanta No. 1 L.P. v. T-C Assocs., Ltd. , 269 Ga. 522 , 501 S.E.2d 209 (1998) (citing OCGA comment as establishing the scope of a statutory definition); Quinn v. Cardiovascular Physicians, P.C. , 254 Ga. 216 , 326 S.E.2d 460 (1985) (citing OCGA comment as stating "the purpose" of a statutory provision); Chaney v. Burdett , 274 Ga. 805 , 560 S.E.2d 21 (2002) (citing OCGA comment as stating the purpose behind a revision to a statutory provision); Grace Bros. v. Farley Indus., Inc. , 264 Ga. 817 , 450 S.E.2d 814 (1994) (citing OCGA comment as defining the nature of a statutory remedy); Magner v. One Sec. Corp. , 258 Ga. App. 520 , 574 S.E.2d 555 (2002) (citing OCGA comment as giving the definition of a statutory term); VSI Enterprises, Inc. v. Edwards , 238 Ga. App. 369 , 518 S.E.2d 765 (1999) (citing OCGA comment as stating the "intent of the legislature" and what the "legislature expected" when enacting a statutory provision); Leventhal v. Post Properties, Inc. , 276 Ga. App. 742 , 624 S.E.2d 223 (2005) (citing OCGA comment as showing the meaning of statutory provision); Rosenfeld v. Rosenfeld , 286 Ga. App. 61 , 648 S.E.2d 399 (2007) (citing OCGA comments as establishing the burden of proof that a party must carry under a statutory provision); Weir v. Kirby Const. Co. , 213 Ga. App. 832 , 446 S.E.2d 186 (1994) (citing OCGA comment as stating the purpose of a statutory provision).

The nature and authoritativeness of the work, like the identity of the author, are material in determining whether the work is attributable to the constructive authorship of the People. After all, the decision in Banks not only emphasized the identity of the creator of the work but also the nature of the work, reasoning that the work was uncopyrightable precisely because it was an "authentic exposition and interpretation of the law [ ] binding [on] every citizen." Banks , 128 U.S. at 253, 9 S.Ct. 36 .

Many other courts applying the rule in Banks , or a rule like it, have emphasized that the law, as an authoritative work that governs people's lives, is uncopyrightable. See, e.g. , Nash v. Lathrop , 142 Mass. 29 , 6 N.E. 559 (1886) ("The decisions and opinions of the justices are the authorized expositions and interpretations of the laws, which are binding upon all the citizens... justice requires that all should have free access to the opinions, and [ ] it is against sound public policy to prevent this, or to suppress and keep from the earliest knowledge of the public the statutes, or the decisions and opinions of the justices."); West Publishing , 27 F. at 57 ("But it is a maxim of universal application that every man is presumed to know the law, and it would seem inherent that freedom of access to the laws, or the official interpretation of those laws, should be co-extensive with the sweep of the maxim. Knowledge is the only just condition of obedience."); State of Connecticut v. Gould , 34 F. 319 , 319 (C.C.N.D.N.Y. 1888) ("[C]onsiderations of public policy which, it is said, demand, in a country where every person is presumed and required to know the law, that the fullest and earliest opportunity of access to the expositions of the judicial tribunals should be afforded to all.").

By way of contrast, a judge might create a work in his capacity as an employee of the government that bears little relation to his role as an official expositor of the law. A speech delivered by a judge, depending on the circumstances of the address, may or may not count as a work created by a government employee. See Pub. Affairs Assocs., Inc. v. Rickover , 268 F.Supp. 444 (D.D.C 1967). But such a work assuredly does not count as a work made in the exercise of the sovereign power to make or interpret the law. A judicial speech is assigned no authoritative weight -- it binds no one and has no official effect on the law or on how it is understood. Only those *1252 works that derive from the legitimate exercise of sovereign power, such as official interpretations of the law and the law itself, are assigned authoritative weight.

Put another way, whether or not a work is assigned the authoritative weight associated with law is deeply intertwined with the question of whether the work was made by the agents of the People in the legitimate exercise of delegated, sovereign power. As Hamilton explained during the ratification debates, "[n]o legislative act [ ] contrary to the Constitution, can be valid. To deny this, would be to affirm, that the deputy is greater than his principal; that the servant is above his master; that the representatives of the people are superior to the people themselves; that men acting by virtue of powers, may do not only what their powers do not authorize, but what they forbid." THE FEDERALIST No. 78 at 466 (Alexander Hamilton) (Clinton Rossiter ed., 1961). As a result, the authoritativeness of a work is probative on the question of whether a work is created in an exercise of sovereign power, and is also probative on the question of whether a work falls within the scope of the rule in Banks . Thus, in addition to whether the work was prepared by a judicial or legislative body, an examination of the nature of the work, which is another way of asking whether it carries authoritative weight, may indicate whether the work is uncopyrightable.

These annotations carry authoritative weight and therefore make it more likely that the work is attributable to the constructive authorship of the People. Quite simply, they are much closer to resembling the judicially authored materials found in Banks than other works produced by state employees, such as the materials produced by the Court reporter in Callaghan .

C.

The final factor we consider is the process by which the annotations were created. While the process by which the annotations were made into an official edict of the State of Georgia is not identical to the process by which the statutory provisions were made into binding law, they are very closely related. As a result, like the identity of the work's creator and the nature of the work, the process also weighs in favor of the conclusion that the work is uncopyrightable.

Both parties acknowledge that the Georgia General Assembly does not individually enact each separate annotation as part of the ordinary legislative process. In this respect the annotations are different than the statutory portions of the OCGA. The statutory portions of the Code are introduced as bills in the Georgia legislature, generally pass through the committee process where legislators can directly influence the text of the bill, are voted on by both Houses, and are signed by the Governor. See Tracking a Bill Through the General Assembly, http://www.legis.ga.gov/Legislation/en-US/default.aspx.

The enacted laws of a session of the legislature are then "published in Georgia Laws as a collection of session laws, representing all of the acts and resolutions passed during that particular legislative session." Austin Williams, "Researching Georgia Law," 34 Ga. St. U.L. Rev. 741 , 761 (2015). Later, the laws are incorporated into the OCGA. Id . Each year, the Georgia legislature then votes to "reenact the statutory portion of [the] Code as amended, in furtherance of the work of the Code Revision Commission," thereby voting on the statutory text in the form in which it has been incorporated into the OCGA. See, e.g. , 2017 Ga. Laws 275 , § 54; 2016 Ga. Laws 625 , § 54; 2015 Ga. Laws 9 , § 54.

Further, under Georgia law, it is the responsibility of the Code Revision Commission to "prepare and have introduced at *1253 each regular session of the General Assembly one or more bills to reenact and make corrections in the Official Code of Georgia Annotated." O.C.G.A. § 28-9-5. In this way, the statutory portions of the OCGA are voted on at least twice, once when they are voted on as individual bills after having gone through the regular legislative process, and once as part of the Georgia legislature's vote to reenact the updated OCGA as prepared by the Commission. By contrast, the annotations are prepared by the Commission outside of the normal channels of the legislative process in the manner we have detailed, and are not voted on individually in the way that Georgia session laws are.

However, it is also the case that the Georgia General Assembly voted to adopt the annotations as prepared by the Commission as an integral part of the official Code. See O.C.G.A. § 1-1-1. Further, it did so through a legislative act that necessarily passed both Houses of the legislature and was signed into law by the Governor. Moreover, and significant for our purposes, the General Assembly votes each year to amend the OCGA and reaffirm its status as the official codification of Georgia's laws.

Under the American system of government, the essential hallmarks of legislative process are bicameralism and presentment. See I.N.S. v. Chadha , 462 U.S. 919 , 103 S.Ct. 2764 , 77 L.Ed.2d 317 (1983) ; see also Ga. Const. Art. V, § II, Para. IV ; Ga. Const. Art. III, § V, Para. V. While legislative processes may ordinarily include the introduction of an individual bill and its passage through the relevant committee before it receives a vote of the full House, those are not the essential steps that endow the bill with its legal status. Rather, the vote of both Houses of the legislature, and presentment to an executive are the defining moments in an exercise of the sovereign authority. This is so even when the legislature adopts as its own a work authored outside the normal channels of the legislative process. See Veeck , 293 F.3d at 799 ("Even when a governmental body consciously decides to enact proposed model building codes, it does so based on various legislative considerations, the sum of which produce its version of 'the law.' In performing their function, the lawmakers represent the public will, and the public are the final 'authors' of the law.").

That the process by which the OCGA annotations were created is similar to the ordinary process by which laws are enacted also is relevant to our inquiry. The importance of this consideration is apparent from well settled procedural mechanisms by which the power to make and interpret the law is exercised, and from the observation that deviating from the process may deprive the edict of its legal effect. As we've noted, bicameral passage of a bill and its presentment to the executive are the ordinary means by which a legislative body exercises the sovereign power entrusted to it. See Chadha , 462 U.S. at 957 , 103 S.Ct. 2764 (invalidating a purported exercise of the legislative power that failed to adhere to "the standards prescribed in Article I" for the exercise of such power); U.S. Const. art. I, § 7, cl. 2. Similarly, the judicial power to propound the meaning of the law must be exercised according to established procedures. In particular, judges issue official interpretations of the law as part of deciding a case or controversy, after considering the arguments made by both parties to the case. See Hayburn's Case , 2 U.S. (2 Dall.) 408, 1 L.Ed. 436 (1792). An exposition on the meaning of a law, even if written by a judge, would obviously not qualify as an exercise of the sovereign power to interpret law if it were written outside the ordinary procedural channels by which that power is exercised. See Correspondence of the Justices (1793) (found in 3 *1254 Johnston , Correspondence and Public Papers of John Jay 486-89 (1891) ).

In short, as is the case with the identity of the creator of the work and the nature of the work, fundamental principles that govern how sovereign power is exercised under a republican form of government suggest that the process by which an edict is promulgated is probative as well on the question of whether a work was created through the exercise of such power. Cf . Clinton v. City of New York , 524 U.S. 417 , 118 S.Ct. 2091 , 141 L.Ed.2d 393 (1998) (invalidating the Line Item Veto Act on the grounds that it impermissibly deviated from the "finely wrought" constitutional processes established for the exercise of legislative power). Just as an action is not deemed a legitimate exercise of sovereign power if it is undertaken by the wrong official, so too it may be invalid if undertaken outside the proper procedural channels. The converse follows naturally: if an action is undertaken through the ordinary procedural channels by which the sovereign power is exercised, it is more likely that the action represents an exercise of sovereign power.

The importance of process was suggested long ago in Banks when the Supreme Court emphasized that only those works created by judges in "the discharge of their judicial duties" are uncopyrigthable. Banks , 128 U.S. at 253 , 9 S.Ct. 36 . In other words, a work made by a judge outside the normal channels by which judicial action is taken would not be subject to the rule in Banks . See Veeck , 293 F.3d at 799 ("The very process of lawmaking demands and incorporates contributions by 'the people.' "). It is therefore fair to say that, just as the Court in Banks emphasized that the justices of the Supreme Court of Ohio had authored the work in question "in the discharge of their judicial duties," the Georgia legislature's use of bicameralism and presentment to adopt the annotations as their own and merge them with statutory text indicates that the work was created by the legislators in the discharge of their official duties. This too bolsters our conclusion

IV.

Our inquiry has focused on whether the official annotations represent a direct exercise of sovereign power, and are therefore attributable to the constructive authorship of the People. In making this determination, we have compared the work in question to works that represent the prototypical exercise of sovereign power, which is to say statutes and official interpretations of the law. We have been guided by three factors that may be regarded as the defining characteristics of law -- the identity of the public official who created the work; the nature of the work; and the process by which the work was produced.

When the wrong public official exercises a power delegated in the law, when the power exercised is of a type not contemplated by the law, or when the power is exercised outside the procedural channels prescribed by the law, the act cannot be considered a valid exercise of the sovereign power. From these principles, the corollary logically follows: when the action taken is of the type entrusted by the People to their agents, when it is wielded by a public official whose assigned duties include the exercise of sovereign power, and when it is exercised pursuant to constitutionally designated processes, it more likely represents an exercise of the sovereign authority. The reasoning found in Banks also suggests the importance of these factors.

All of them point strongly toward the conclusion that the OCGA annotations are not copyrightable. The OCGA annotations are created by Georgia's legislative body, which has been entrusted with exercising *1255 sovereign power on behalf of the people of Georgia. While the annotations do not carry the force of law in the way that statutes or judicial opinions do, they are expressly given legal significance so that, while not "law," the annotations undeniably are authoritative sources on the meaning of Georgia statutes. The legislature has stamped them "official" and has chosen to make them an integral part of the official codification of Georgia's laws. By wrapping the annotations and the statutory text into a single unified edict, the Georgia General Assembly has made the connection between the two inextricable and, thereby, ensured that obtaining a full understanding of the laws of Georgia requires having unfettered access to the annotations. Finally, the General Assembly's annual adoption of the annotations as part of the laws of Georgia is effected by the legislative process -- namely bicameralism and presentment -- that is ordinarily reserved for the exercise of sovereign power.

Thus, we conclude that the annotations in the OCGA are attributable to the constructive authorship of the People. To advance the interests and effect the will of the People, their agents in the General Assembly have chosen to create an official exposition on the meaning of the laws of Georgia. In creating the annotations, the legislators have acted as draftsmen giving voice to the sovereign's will. The resulting work is intrinsically public domain material, belonging to the People, and, as such, must be free for publication by all.

As a result, no valid copyright can subsist in these works. We, therefore, reverse the judgment of the district court, direct that judgment be entered for appellant PRO, vacate the district court's order granting the State of Georgia injunctive relief, and remand for further proceedings consistent with this opinion.

REVERSED IN PART, VACATED IN PART AND REMANDED